1 2011-05-05 00:00:22 inductor has quit (Quit: leaving)
2 2011-05-05 00:00:53 inductor has joined
3 2011-05-05 00:05:20 Teslah has quit (Ping timeout: 252 seconds)
4 2011-05-05 00:08:18 theorb has joined
5 2011-05-05 00:08:34 theorbtwo has quit (Ping timeout: 240 seconds)
6 2011-05-05 00:08:41 theorb is now known as theorbtwo
7 2011-05-05 00:09:52 skyewm has joined
8 2011-05-05 00:10:04 brooss has quit (Read error: Connection reset by peer)
9 2011-05-05 00:10:19 brooss has joined
10 2011-05-05 00:12:00 metonymous has joined
11 2011-05-05 00:12:28 <metonymous> php json rpc question: "intval($bitcoin->getbalance($uid, 4) * 100000000) > 0"; is this a solid expression for lossless conversion?
12 2011-05-05 00:12:38 grbgout has quit (Ping timeout: 246 seconds)
13 2011-05-05 00:12:38 sshc_ has quit (Ping timeout: 246 seconds)
14 2011-05-05 00:12:39 <molecular> ArtForz, the PID fan controller ADL code you once pastebinned... that's public domain?
15 2011-05-05 00:12:43 grbgout has joined
16 2011-05-05 00:13:14 sshc has joined
17 2011-05-05 00:13:36 steve____ has left ()
18 2011-05-05 00:13:39 <molecular> ArtForz, wanna give modified version of it it to xelister so he can set his mem clock
19 2011-05-05 00:14:02 <ArtForz> molecular: yeah, it's PD
20 2011-05-05 00:14:13 noagendamarket has quit (Ping timeout: 276 seconds)
21 2011-05-05 00:14:35 <luke-jr> metonymous: erm, you just want to check if it's >0
22 2011-05-05 00:14:40 <genjix> metonymous: use my branch: https://github.com/genjix/bitcoin/tree/strrpc
23 2011-05-05 00:14:51 <luke-jr> metonymous: https://en.bitcoin.it/wiki/Proper_Money_Handling_(JSON-RPC)#PHP
24 2011-05-05 00:14:57 <metonymous> ooh, a wiki page
25 2011-05-05 00:15:08 <metonymous> when i use gmp_cmp it throws an error
26 2011-05-05 00:15:10 <metonymous> if given a float
27 2011-05-05 00:15:23 <genjix> dont use floats, use integers
28 2011-05-05 00:15:36 <metonymous> how do i get the json rpc to return an integer
29 2011-05-05 00:15:40 <luke-jr> metonymous: https://en.bitcoin.it/wiki/Proper_Money_Handling_(JSON-RPC)#PHP
30 2011-05-05 00:15:47 <metonymous> ta for wiki link
31 2011-05-05 00:16:07 <metonymous> aaah nicness
32 2011-05-05 00:16:13 <metonymous> i wondered if there was a power function
33 2011-05-05 00:16:15 <luke-jr> round($bitcoin->getbalance($uid, 4) * 1e8)
34 2011-05-05 00:16:23 <luke-jr> always gets the right integer
35 2011-05-05 00:16:27 <metonymous> thankyou very much luke-jr
36 2011-05-05 00:16:32 <metonymous> what's ur bitcoin address
37 2011-05-05 00:16:34 <metonymous> i'll send u cents
38 2011-05-05 00:16:40 <luke-jr> 1MjDfFLkfVquBc8kphTJDETeTr7vFmRLXu
39 2011-05-05 00:16:43 noagendamarket has joined
40 2011-05-05 00:17:23 <metonymous> thankyou all
41 2011-05-05 00:17:26 metonymous has left ("Leaving")
42 2011-05-05 00:17:52 <molecular> ArtForz, thx
43 2011-05-05 00:17:53 DukeOfURL has joined
44 2011-05-05 00:18:34 metonymous has joined
45 2011-05-05 00:18:52 <metonymous> wait a secondoozie, genjix, u the britcoin dev?
46 2011-05-05 00:19:55 <luke-jr> lol
47 2011-05-05 00:20:04 <luke-jr> he is
48 2011-05-05 00:20:31 <metonymous> line ~122 of util.php, I've changed mine to "if (round($balance * 1e8) > 0) {"
49 2011-05-05 00:20:46 <metonymous> i presume ur php is different or supressing the error, anyway, back to code i go
50 2011-05-05 00:21:13 <luke-jr> metonymous: I would check that $balance is straight from JSON-RPC, or if it has already been processed in some way
51 2011-05-05 00:21:35 threepio has quit (Ping timeout: 240 seconds)
52 2011-05-05 00:21:37 <metonymous> the json rpc doesn't touch it, unless it does it with meta coding, which is possible.
53 2011-05-05 00:21:41 <metonymous> but ur right, i should check
54 2011-05-05 00:21:43 RenaKunisaki has quit (Ping timeout: 246 seconds)
55 2011-05-05 00:21:47 <metonymous> i've checked with "0"
56 2011-05-05 00:21:50 <metonymous> and that works :P
57 2011-05-05 00:22:24 x6763 has quit (Ping timeout: 246 seconds)
58 2011-05-05 00:22:26 witten has quit (Ping timeout: 246 seconds)
59 2011-05-05 00:22:32 <metonymous> luke-jr, can you test this url for me? http://dev.btc.metonymous.com/
60 2011-05-05 00:23:32 TheKid has left ()
61 2011-05-05 00:23:36 subpar has quit (Quit: Leaving)
62 2011-05-05 00:23:50 x6763 has joined
63 2011-05-05 00:24:05 threepio has joined
64 2011-05-05 00:24:29 <genjix> metonymous: hey if you're using that then you need my branch of bitcoin
65 2011-05-05 00:24:34 <metonymous> oooh
66 2011-05-05 00:24:34 <genjix> https://github.com/genjix/bitcoin/tree/strrpc
67 2011-05-05 00:24:38 <metonymous> that's a good thing to know
68 2011-05-05 00:24:39 <metonymous> ta
69 2011-05-05 00:24:41 witten has joined
70 2011-05-05 00:24:49 <genjix> metonymous: #bitcoinconsultancy
71 2011-05-05 00:24:49 <metonymous> what's strrpc?
72 2011-05-05 00:25:02 <genjix> it returns all the numbers as strings not floats
73 2011-05-05 00:25:12 <metonymous> ur branch does?
74 2011-05-05 00:25:13 <metonymous> ahhh
75 2011-05-05 00:25:14 <metonymous> makes sense
76 2011-05-05 00:25:19 <metonymous> i'll switch channels if u like
77 2011-05-05 00:25:19 <genjix> i wrote this: https://en.bitcoin.it/w/index.php?title=PHP_developer_intro&oldid=5737
78 2011-05-05 00:25:30 <metonymous> i hope i haven't trodden on any toes... i just wanted to dive in
79 2011-05-05 00:25:46 <genjix> but the devs think hacks are much better
80 2011-05-05 00:25:49 <luke-jr> metonymous: it's an ugly hack that passes amounts as strings
81 2011-05-05 00:26:19 <luke-jr> metonymous: also totally incompatible with everything
82 2011-05-05 00:26:32 <metonymous> if i use my round code, can i use normal bitcoin client? or is ur branch even more different
83 2011-05-05 00:26:55 <luke-jr> metonymous: you'd need to edit every place he gets amounts from JSON-RPC
84 2011-05-05 00:27:04 <metonymous> i think i'll do that for now
85 2011-05-05 00:27:06 <metonymous> i'm windows
86 2011-05-05 00:27:14 <luke-jr> â¦
87 2011-05-05 00:27:15 <metonymous> and getting it to build bitcoin has proved apain
88 2011-05-05 00:27:19 <metonymous> i'll do it eventually
89 2011-05-05 00:27:24 <luke-jr> Windows is a pain in general
90 2011-05-05 00:27:27 ragga has joined
91 2011-05-05 00:27:32 <metonymous> not for gaming :P
92 2011-05-05 00:27:38 <luke-jr> yes, even for gaming
93 2011-05-05 00:27:50 <metonymous> but for now... i'll meta hack the json rpc to convert all string numbers to ints at that point or something
94 2011-05-05 00:27:57 <metonymous> it makes sense to push it up the chain
95 2011-05-05 00:27:59 <genjix> metonymous: ok well if you want to use floats then you can just change the parts in util.php
96 2011-05-05 00:28:06 <metonymous> ta genjix
97 2011-05-05 00:28:15 <metonymous> is there any other differences i should be aware of?
98 2011-05-05 00:29:01 <metonymous> genjix, i wanna give u cents too for ur help and project, even if i take it no-where, btc address?
99 2011-05-05 00:29:29 <genjix> dont worry about it, im glad people are trying to setup more exchanges :p
100 2011-05-05 00:29:35 <genjix> bitcoin needs them to survive
101 2011-05-05 00:30:35 <luke-jr> integer != floats
102 2011-05-05 00:37:21 <metonymous> agreed! :D
103 2011-05-05 00:37:26 <metonymous> i'm too used to ruby
104 2011-05-05 00:37:45 <metonymous> was gonna port britcoin to ruby... but grrr
105 2011-05-05 00:37:48 eao has joined
106 2011-05-05 00:37:52 <metonymous> it'd be splitting dev effort up
107 2011-05-05 00:38:03 * genjix is more a python person
108 2011-05-05 00:38:12 <metonymous> i love that one too
109 2011-05-05 00:38:14 <genjix> but yeah php is standard in webdev
110 2011-05-05 00:38:22 <metonymous> in "old" webdev lol
111 2011-05-05 00:38:23 <metonymous> hehe
112 2011-05-05 00:38:30 <metonymous> i've contracted for startups the last 5 yars
113 2011-05-05 00:38:37 <metonymous> only php we encountered was wikipedia
114 2011-05-05 00:38:39 <metonymous> too big to port
115 2011-05-05 00:38:40 <metonymous> :P
116 2011-05-05 00:38:50 <metonymous> god i sound pretentious
117 2011-05-05 00:38:53 <metonymous> python is the shiz
118 2011-05-05 00:39:00 <metonymous> so is ruby
119 2011-05-05 00:39:03 <metonymous> new php is ok
120 2011-05-05 00:39:19 <metonymous> i'll shutup now
121 2011-05-05 00:40:26 <genjix> http://exchange.surething.biz/
122 2011-05-05 00:40:29 <genjix> can't stop laughing
123 2011-05-05 00:46:19 <sirius> :D
124 2011-05-05 00:48:00 <molecular> ArtForz, Diablo-D3: I'm confused now. xelister says he can set his mem clock using "aticonfig --od-setclocks x,y". this doesn't work for me, the mem clock doesn't change, the core clock does. does settint mem-clock using aticonfig work for you guys?
125 2011-05-05 00:48:07 Beremat has quit (Read error: Connection reset by peer)
126 2011-05-05 00:48:29 <Diablo-D3> molecular: he can... he just cant lower it to 300.
127 2011-05-05 00:48:37 <xelister> molecular: worked for me on 2 x 5970's and 1 x 5770
128 2011-05-05 00:48:45 <molecular> damn, I can't at all
129 2011-05-05 00:48:52 <xelister> Diablo-D3: so again what utlity and HOW EXACTLY use to allow lowering to say 300 ???
130 2011-05-05 00:48:55 <xelister> just RBE ?
131 2011-05-05 00:48:57 Beremat has joined
132 2011-05-05 00:48:59 <xelister> or soemthing better u recommend
133 2011-05-05 00:49:06 <Diablo-D3> xelister: amdoverdrivectrl
134 2011-05-05 00:49:10 <Diablo-D3> you've been told this like ten times
135 2011-05-05 00:49:33 <xelister> k..
136 2011-05-05 00:49:50 <xelister> is this linux program enough? or do I need to reprogram bios too?
137 2011-05-05 00:50:36 <xelister> molecular: what you send us is that program, amdoverdrivectrl ? or some speciall version of it patched by Artz?
138 2011-05-05 00:51:17 <xelister> Im confused. Which things can be NOT be done in linux and need BIOS reprogramming?
139 2011-05-05 00:51:49 <xelister> incrementing-Vcore, decrementing-Vmem, decrementing-MinMemClock, incrementing-MaxCoreClock ?
140 2011-05-05 00:52:27 donchate has joined
141 2011-05-05 00:52:53 <molecular> xelister, no, I didn't know about amdoverdrivectrl, checking it out
142 2011-05-05 00:52:58 <molecular> xelister, mine is just a hack
143 2011-05-05 00:54:07 xxx_xxx has left ()
144 2011-05-05 00:55:11 subpar has joined
145 2011-05-05 00:55:47 <sipa> xelister: amdoverdrivectrl can reuse existing profiles in bios and modify them to some extent, as i understand it
146 2011-05-05 00:56:08 <xelister> blargh
147 2011-05-05 00:56:09 <xelister> ;)
148 2011-05-05 00:56:14 <xelister> ok lets sort it out
149 2011-05-05 00:56:18 <sipa> so it can be used to change the memory clock to its normal lowest value (for 5xxx and 6xxx that is 300 MHz, i believe)
150 2011-05-05 00:56:30 <xelister> who here is using lower-then-normally-available RAM clock?
151 2011-05-05 00:57:15 <molecular> what's normally available? looks to me 150Mhz is available normally
152 2011-05-05 00:57:45 <molecular> at least my aticonfig reports range 150-1500 for mem clock and I sure didn't change that in BIOS
153 2011-05-05 00:58:11 <molecular> (5970)
154 2011-05-05 00:58:28 BitterTea has joined
155 2011-05-05 00:58:35 <xelister> molecular: ???
156 2011-05-05 00:58:47 <xelister> molecular: I have range 1000-1500 in r5970
157 2011-05-05 00:58:47 <Diablo-D3> molecular: huh
158 2011-05-05 00:58:53 <Diablo-D3> cant be stock bios
159 2011-05-05 00:59:00 <xelister> did youu bought used radeom from some miner, molecular? :P
160 2011-05-05 00:59:50 <molecular> no, new one, a sapphire
161 2011-05-05 01:00:13 <molecular> it's not stock, but all I changed with RBE was register 0x18 (to my knowledge)
162 2011-05-05 01:00:53 <luke-jr> sirius: are you MyBitcoinâ
163 2011-05-05 01:02:46 <xelister> Diablo-D3: are you sure amdoverdrivectrl actually allows to go OUTSIDE of normal memory clock limits? it does reprogram BIOS or what?
164 2011-05-05 01:03:23 <molecular> Diablo-D3, is it possible that the lower limit of memclock is the actual memclock of the lower performance level? that would make sense, because I set the mem clock of the medium perf level to 150. HA, that's got to be it!
165 2011-05-05 01:03:52 <sipa> xelister: before using amdoverdrivectrl the range shown by aticonfig was 1000-1500, now it is 300-1500
166 2011-05-05 01:04:20 <molecular> sipa: and doe the memclock of medium perf-level happen to be set to 300?
167 2011-05-05 01:04:44 <molecular> *does
168 2011-05-05 01:04:51 <sipa> yes, there already needed to be some profile with 300 MHz mem speed
169 2011-05-05 01:05:24 <xelister> sipa: and wher in amdoverdr.. is that low speed 300 configurable? or is it just the effect of launching this app
170 2011-05-05 01:05:26 <molecular> so we got the ranges figured out. they're given by the values from the perf level below/above respectively I assume
171 2011-05-05 01:06:08 <molecular> perflevel[1].memclock == perflevel[2].min_memclock
172 2011-05-05 01:06:30 <xelister> not really... the ranges... still not sure do I have to just use amdover, or change bios with RBE and then run dos to flash it
173 2011-05-05 01:06:36 <sipa> xelister: in the configfile i have it set to 300 MHz obviously, as i want 300 MHz :)
174 2011-05-05 01:06:47 <sipa> haven't tried just running it without any profile
175 2011-05-05 01:06:49 <molecular> xelister, just try amdover, I think it will work
176 2011-05-05 01:07:13 <xelister> sipa: config file of what?
177 2011-05-05 01:07:36 * xelister bbl
178 2011-05-05 01:07:36 <sipa> of amdoverdrivectrl
179 2011-05-05 01:08:27 Beremat has quit (Read error: Connection reset by peer)
180 2011-05-05 01:08:47 Beremat has joined
181 2011-05-05 01:09:12 ezl has quit (Ping timeout: 240 seconds)
182 2011-05-05 01:09:22 <molecular> gn8 guys, tonight I can sleep well since my miner has new hd and system ;)
183 2011-05-05 01:09:59 <CIA-30> bitcoin: droidbird exper * r1001306d5662 intersango/util.php: comment some functions in util.php. http://tinyurl.com/3t7a9lb
184 2011-05-05 01:10:33 <Diablo-D3> [08:57:52] <xelister> Diablo-D3: are you sure amdoverdrivectrl actually allows to go OUTSIDE of normal memory clock limits? it does reprogram BIOS or what?
185 2011-05-05 01:10:34 <Diablo-D3> not outside
186 2011-05-05 01:10:46 <Diablo-D3> like, my card does 500mhz memory on idle
187 2011-05-05 01:10:51 <Diablo-D3> so I can safely go down to 500
188 2011-05-05 01:10:53 <xelister> so what the fuck are you wasting my time with
189 2011-05-05 01:11:03 <Diablo-D3> because 58xx cards go down to 300
190 2011-05-05 01:11:11 <Diablo-D3> 300 on high profile.
191 2011-05-05 01:11:14 <xelister> I can alraedy since like half year use stupid amdconfig to get to lowest mem range
192 2011-05-05 01:11:19 <xelister> jesus.
193 2011-05-05 01:11:27 <Diablo-D3> xelister: nope, you cant.
194 2011-05-05 01:11:30 <sirius> luke-jr: no
195 2011-05-05 01:11:36 <xelister> how to CHANGE THE LOWEST LIMIT (like to get 300 not 1000 lowest speed on r5970)
196 2011-05-05 01:11:47 <xelister> Diablo-D3: I can. the r5970 limit is 1000-1500 on mem clk
197 2011-05-05 01:11:47 <Diablo-D3> xelister: I just told you moron
198 2011-05-05 01:11:54 <Diablo-D3> thats 1000-1500
199 2011-05-05 01:11:57 <Diablo-D3> not 300-1500
200 2011-05-05 01:12:07 <xelister> wtf
201 2011-05-05 01:12:12 <Diablo-D3> amdoverdrivecontrol can set down to the _idle_ limit
202 2011-05-05 01:12:12 <xelister> so why amdconfig says 1000-1500
203 2011-05-05 01:12:16 <xelister> are there like 2 limits?
204 2011-05-05 01:12:22 <xelister> oh Ati you troll you.
205 2011-05-05 01:12:24 <Diablo-D3> there are multiple limits
206 2011-05-05 01:12:28 * xelister headdesks
207 2011-05-05 01:12:33 <Diablo-D3> aticonfig only sets the high profile range
208 2011-05-05 01:12:44 <Diablo-D3> amdoverdrivecontrol can set even idle speeds in high profile
209 2011-05-05 01:12:59 <xelister> ok, so this allowes to get 300 MHz at r5970 for mining ?
210 2011-05-05 01:13:02 <Diablo-D3> yes.
211 2011-05-05 01:13:02 <xelister> without fucking with bios?
212 2011-05-05 01:13:04 <Diablo-D3> YES
213 2011-05-05 01:13:05 <xelister> neat.
214 2011-05-05 01:13:07 <Diablo-D3> PAY FUCKING ATTENTION
215 2011-05-05 01:13:12 <xelister> and what about lowering voltage?
216 2011-05-05 01:13:21 <xelister> sorry was too bussy paying ursis instead >_>
217 2011-05-05 01:13:23 <Diablo-D3> it can set voltages in your cards profiles
218 2011-05-05 01:13:27 <Diablo-D3> its not very flexible
219 2011-05-05 01:13:29 <xelister> well lowering memV and rising coreV
220 2011-05-05 01:13:34 <Diablo-D3> it cant set memv
221 2011-05-05 01:13:46 <Diablo-D3> memory voltage is very tightly controlled because it has to be
222 2011-05-05 01:13:54 <Diablo-D3> HOWEVER
223 2011-05-05 01:14:01 <Diablo-D3> when you set that 300mhz speed, its using the voltage for 300mhz
224 2011-05-05 01:14:05 sanesmith has quit (Ping timeout: 240 seconds)
225 2011-05-05 01:14:07 <xelister> so can I increase to 1.10 using hist amdoverctrl...?
226 2011-05-05 01:14:17 <Diablo-D3> xelister: if your card offers it in the profile
227 2011-05-05 01:14:35 <xelister> and if not?
228 2011-05-05 01:14:41 <xelister> then I have to fuck with bios flashing?
229 2011-05-05 01:15:02 <Diablo-D3> yes.
230 2011-05-05 01:15:10 <xelister> now its finally clear =)
231 2011-05-05 01:15:26 <Diablo-D3> thats the ONE thing it cant do without flashing
232 2011-05-05 01:15:30 <xelister> so which one of million settings in RBE is the "Vcore in idle profile" ?
233 2011-05-05 01:15:37 <Diablo-D3> what the fuck is RBE anyhow
234 2011-05-05 01:15:44 <xelister> radeon bios editor
235 2011-05-05 01:15:58 <xelister> http://www.techpowerup.com/rbe/
236 2011-05-05 01:16:06 <xelister> windoze tool to edit .rom's of radeons
237 2011-05-05 01:16:09 <Diablo-D3> oh, windows shit, dont care.
238 2011-05-05 01:16:16 <xelister> (edit them offline, then you flash separatelly liek from dos)
239 2011-05-05 01:16:24 <xelister> well I would prefer linux editor but dont know any
240 2011-05-05 01:16:40 <xelister> can you just edit the .rom in hexedit? is that the magic.. uhm... 0x18 byte? or smth?
241 2011-05-05 01:16:41 * Diablo-D3 shrugs
242 2011-05-05 01:16:50 <Diablo-D3> well, thats what its doing
243 2011-05-05 01:16:57 <Diablo-D3> its essentially byte editing it
244 2011-05-05 01:17:01 <Diablo-D3> it just happens to know all the right indexes
245 2011-05-05 01:17:01 <xelister> no shit sherlock
246 2011-05-05 01:17:17 <xelister> yeah, so Im asking, do you guys know the right index to change idle level Vcore
247 2011-05-05 01:18:17 <Diablo-D3> nope.
248 2011-05-05 01:18:17 <xelister> so when I will force 300 MHz mem clock then card will auto siwthc to idle profile... that makes the problem with Vcore... any other bad result of this>
249 2011-05-05 01:19:22 <xelister> any other bad result?
250 2011-05-05 01:20:02 metonymous has quit (Quit: Leaving)
251 2011-05-05 01:20:56 <Diablo-D3> xelister: no, it'll switch to the right voltage on the memory
252 2011-05-05 01:21:01 <Diablo-D3> which most likely has not changed.
253 2011-05-05 01:21:14 <Diablo-D3> quit looking free 10c degrees in the mouth
254 2011-05-05 01:21:21 <xelister> hm?
255 2011-05-05 01:21:40 <Diablo-D3> 1200mhz -> 300mhz == gpu temps drop 10c.
256 2011-05-05 01:21:45 <xelister> ok
257 2011-05-05 01:21:53 <xelister> but it also changes Vcore back to default
258 2011-05-05 01:22:10 <xelister> and this can be remedied by BIOS editing, in order to allow higher oc of core while remaining stable?
259 2011-05-05 01:22:27 <Diablo-D3> xelister: except
260 2011-05-05 01:22:30 <Diablo-D3> try overclocking first
261 2011-05-05 01:22:35 <Diablo-D3> before fucking with voltages.
262 2011-05-05 01:22:40 <xelister> I go to 810-stable
263 2011-05-05 01:22:43 <xelister> or rather
264 2011-05-05 01:23:07 <xelister> 800-rather stable 810-video locks up often 820-calculations are starting dying
265 2011-05-05 01:23:26 <xelister> on other card, on 830, all works stable almost ideally
266 2011-05-05 01:23:33 <xelister> so I suppose higher V on first card could fix it
267 2011-05-05 01:23:53 <xelister> ArtForz: when you edited bios in RBE, what positions you exactly change to set 1.10 Vcore for all profiles?
268 2011-05-05 01:23:57 <Diablo-D3> xelister: try the fucking mem hack first
269 2011-05-05 01:24:03 <Diablo-D3> you'll have more stable VRM usage
270 2011-05-05 01:24:08 <xelister> oh? ok good idea.
271 2011-05-05 01:24:43 <xelister> btw I wonder can all this be detected when warrantin'g time one day
272 2011-05-05 01:25:16 <Diablo-D3> yes.
273 2011-05-05 01:25:39 <xelister> how?
274 2011-05-05 01:26:06 <Diablo-D3> magic.
275 2011-05-05 01:26:42 <xelister> and?
276 2011-05-05 01:26:52 <Diablo-D3> =|
277 2011-05-05 01:26:58 <Diablo-D3> fucking blackbox recorder dude
278 2011-05-05 01:27:02 <xelister> orly?
279 2011-05-05 01:27:12 <xelister> no seriously. can they or not and how to avoid ;)
280 2011-05-05 01:27:22 <Diablo-D3> they can, and you cant.
281 2011-05-05 01:27:55 <xelister> bummer
282 2011-05-05 01:29:36 Zarutian has quit (Quit: Zarutian)
283 2011-05-05 01:30:24 <luke-jr> ;;bc,stats [bc,eligius]
284 2011-05-05 01:30:27 <gribble> Current Blocks: 121908 | Current Difficulty: 109670.13329248 | Next Difficulty At Block: 122975 | Next Difficulty In: 1067 blocks | Next Difficulty In About: 5 days, 13 hours, 22 minutes, and 30 seconds | Next Difficulty Estimate: 146345.76449282
285 2011-05-05 01:30:34 <luke-jr> ;;bc,calc [bc,eligius]
286 2011-05-05 01:30:35 <gribble> The average time to generate a block at 15748213.4187 Khps, given current difficulty of 109670.13329248 , is 8 hours, 18 minutes, and 30 seconds
287 2011-05-05 01:30:41 dreamer_ has joined
288 2011-05-05 01:31:18 wolfspraul has joined
289 2011-05-05 01:33:41 DukeOfURL has quit (Ping timeout: 240 seconds)
290 2011-05-05 01:36:00 gasteve_ has joined
291 2011-05-05 01:36:38 DukeOfURL has joined
292 2011-05-05 01:37:30 johnlockwood has quit (Ping timeout: 250 seconds)
293 2011-05-05 01:39:42 johnlockwood has joined
294 2011-05-05 01:47:32 niekie has quit (Quit: No Ping reply in 180 seconds.)
295 2011-05-05 01:47:47 idnar has quit (Remote host closed the connection)
296 2011-05-05 01:48:06 niekie has joined
297 2011-05-05 01:48:06 idnar has joined
298 2011-05-05 01:48:13 KBme has quit (Quit: KBme kthxbye)
299 2011-05-05 01:48:51 luke-jr has quit (otg!~luke-jr@ishibashi.dashjr.org|Quit: ZNC - http://znc.sourceforge.net)
300 2011-05-05 01:49:14 KBme has joined
301 2011-05-05 01:49:20 luke-jr has joined
302 2011-05-05 01:50:39 <_ape> fucking nextag/mwave
303 2011-05-05 01:50:53 <_ape> are incompetent. i ordered 2 5870s and they actually have 0 in stock (site still says they do after the email)
304 2011-05-05 01:51:33 gat3way has quit (Read error: Operation timed out)
305 2011-05-05 01:51:49 gat3way has joined
306 2011-05-05 01:51:56 donchate has quit (Read error: Operation timed out)
307 2011-05-05 01:52:20 donchate has joined
308 2011-05-05 01:52:57 Cusipzzz has quit (Quit: KVIrc 4.0.2 Insomnia http://www.kvirc.net/)
309 2011-05-05 01:54:28 <vorlov> does anyone experience really loud fan noise from the 5870?
310 2011-05-05 01:54:42 <[Tycho]> Everyone does.
311 2011-05-05 01:54:47 <vorlov> i got a used one and its making quite the noise.... not sure if its normal
312 2011-05-05 01:56:49 <xelister> vorlov: when you start mining?
313 2011-05-05 01:57:10 <vorlov> soon i hope... but my friend who was kind enough to let me use his place doesn't really like the noise...
314 2011-05-05 01:57:12 <vorlov> so im not sure
315 2011-05-05 01:57:23 <xelister> vorlov: a SUPER loud noise (airplain taking off - can't talk normally in 3 meter radious from the computer - on 5970 at least)
316 2011-05-05 01:57:27 <vorlov> is there anything that can be done to mitigate the noise other than shoving it in a closet...?
317 2011-05-05 01:57:32 <xelister> happens when the extra power supply is not connected
318 2011-05-05 01:57:44 <xelister> then the card goes to 100% fan power, which on r5970 is overkill (normally its limited to 30%)
319 2011-05-05 01:58:20 DukeOfURL has quit (Ping timeout: 246 seconds)
320 2011-05-05 01:58:30 kermit has quit (Quit: Leaving.)
321 2011-05-05 01:58:38 <[Tycho]> "<vorlov> is there anything that can be done to mitigate the noise" - put your PC outside in winter
322 2011-05-05 01:58:43 kermit has joined
323 2011-05-05 01:58:53 wolfspraul has quit (Ping timeout: 276 seconds)
324 2011-05-05 01:59:06 <xelister> in soviet russia
325 2011-05-05 01:59:11 <xelister> the computer puts YOU out in the winder
326 2011-05-05 01:59:13 <vorlov> xelister: does the 5870 go down from 100% when all power connectors are connected?
327 2011-05-05 01:59:26 <xelister> vorlov: not sure, but I think that yea it should.
328 2011-05-05 01:59:34 wolfspraul has joined
329 2011-05-05 01:59:37 <vorlov> okay
330 2011-05-05 01:59:44 <vorlov> ill try it out
331 2011-05-05 01:59:54 <vorlov> cuz right now its only connected to the PCI-E
332 2011-05-05 02:00:00 <vorlov> without a good power supply
333 2011-05-05 02:02:12 <xelister> red led near PCI-E on the card, mean that power is bad
334 2011-05-05 02:04:34 <vorlov> i see
335 2011-05-05 02:04:49 <vorlov> im patching the psu so i could turn it on w/o the mobo
336 2011-05-05 02:04:57 <vorlov> and see if that solves the noise problem at all
337 2011-05-05 02:05:03 <vorlov> or atleast mitigate it....
338 2011-05-05 02:05:49 <xelister> vorlov: lol
339 2011-05-05 02:05:52 <xelister> I did same thing
340 2011-05-05 02:05:55 <xelister> its the green wire.
341 2011-05-05 02:06:10 <vorlov> green and any black right?
342 2011-05-05 02:07:02 dvide has quit (Read error: Connection reset by peer)
343 2011-05-05 02:07:04 <xelister> yea
344 2011-05-05 02:07:05 genjix has quit (Ping timeout: 240 seconds)
345 2011-05-05 02:07:16 dvide has joined
346 2011-05-05 02:10:04 orbiting has joined
347 2011-05-05 02:11:17 Beremat has quit (Read error: Connection reset by peer)
348 2011-05-05 02:11:49 Beremat has joined
349 2011-05-05 02:14:42 jmpespxoreax has joined
350 2011-05-05 02:15:15 <jmpespxoreax> There is a comment in main.h stating that private keys have an expiration date: why is this?
351 2011-05-05 02:15:25 <vorlov> btw can anyone recomend a pool to join?
352 2011-05-05 02:17:41 <forrestv> what tool do you use for overclocking ati cards on windows..?
353 2011-05-05 02:18:33 dreamer_ has quit (Quit: Page closed)
354 2011-05-05 02:18:44 <vorlov> also what do you guys use in terms of miner? what produces more mhash (pocl/pocl-mod/diablo ?)?
355 2011-05-05 02:19:52 <Diablo-D3> well
356 2011-05-05 02:20:22 <vorlov> Diablo-D3: i gotta tell you that on my macbook, your miner took pocl ;-)
357 2011-05-05 02:20:43 <Diablo-D3> outside of bfi_int, mine is.
358 2011-05-05 02:21:20 <vorlov> BFI_INT is phoenix?
359 2011-05-05 02:21:32 <vorlov> what did they do exactly to go up there? what does that BFI_INT do?
360 2011-05-05 02:21:39 <Diablo-D3> bfi_int refers to a specific radeon 5xxx instruction
361 2011-05-05 02:21:52 <Diablo-D3> pheonix is not the first to implement the use of it
362 2011-05-05 02:23:41 <vorlov> i see
363 2011-05-05 02:23:48 <vorlov> question
364 2011-05-05 02:23:58 <vorlov> my bios i can tell to keep using the integrated video card on the mobo
365 2011-05-05 02:24:09 <vorlov> but windows ensists on using the my new 5870... is there a way to force window
366 2011-05-05 02:24:12 <vorlov> windows
367 2011-05-05 02:24:15 <vorlov> to use the mobo one
368 2011-05-05 02:24:30 Cheeseman has joined
369 2011-05-05 02:29:52 <luke-jr> vorlov: poclbm
370 2011-05-05 02:30:40 <Diablo-D3> vorlov: plug your monitor into the mobo one
371 2011-05-05 02:31:12 Cheeseman has quit (Ping timeout: 258 seconds)
372 2011-05-05 02:31:25 <vorlov> Diablo-D3: i did.... but when windows loads up it defaults to trying to use the new 5870
373 2011-05-05 02:31:41 <vorlov> Diablo-D3: so i changed the setting in the bios to allow me to keep using the onboard one
374 2011-05-05 02:32:06 <vorlov> Diablo-D3: and when the computer loads it indeed shows stuff up while the monitor is connected to the mobo card... however when windows loads up it goes dark
375 2011-05-05 02:32:20 <Diablo-D3> vorlov: hrm
376 2011-05-05 02:32:24 <Diablo-D3> weird
377 2011-05-05 02:32:36 <Diablo-D3> Im glad I dont use windows
378 2011-05-05 02:33:44 <luke-jr> lol
379 2011-05-05 02:33:51 Beremat has quit (Read error: Connection reset by peer)
380 2011-05-05 02:34:13 <Sthebig> vorlov: Try safe mode with it set up to use the mobo.
381 2011-05-05 02:34:14 Beremat has joined
382 2011-05-05 02:34:37 <vorlov> Diablo-D3: to be quite honest with u im in unfamiliar teritorry myself.... last time i used windows was 13+ years ago... been a command prompt junky for all those years :)
383 2011-05-05 02:34:53 <vorlov> Sthebig: ill try that!
384 2011-05-05 02:35:22 <Diablo-D3> Ive ran nothing but linux for the past ten years
385 2011-05-05 02:35:23 <Sthebig> Configure it to use the mobo while in safe mode
386 2011-05-05 02:35:36 <Sthebig> Then reboot and hopefully it will work
387 2011-05-05 02:35:39 <vorlov> Sthebig: do you happen to know where in windows 7 i can find that?
388 2011-05-05 02:36:25 glassresistor has joined
389 2011-05-05 02:36:25 glassresistor has quit (Changing host)
390 2011-05-05 02:36:25 glassresistor has joined
391 2011-05-05 02:37:01 <Sthebig> Not off the top of my head.
392 2011-05-05 02:37:17 <Sthebig> Try Display settings (same place where you choose resolution and such)
393 2011-05-05 02:37:23 <Sthebig> Or Device manager
394 2011-05-05 02:38:12 IncitatusOnWater has quit (Ping timeout: 240 seconds)
395 2011-05-05 02:40:18 <vorlov> okay i will
396 2011-05-05 02:40:19 <vorlov> ill report back
397 2011-05-05 02:40:42 wally__ has joined
398 2011-05-05 02:41:27 <Sthebig> good luck
399 2011-05-05 02:41:46 <xelister> 74829bdb7c0c2a3a0450905a54ce776ffb33f3047202fce29619402b61e179a0 amdoverdrivectrl_1.1.4_amd64.deb
400 2011-05-05 02:42:41 <xelister> uh wrong window
401 2011-05-05 02:47:25 LightRider is now known as afk!~LightRide@unaffiliated/lightrider|LightRIder
402 2011-05-05 02:51:48 <vorlov> i think i got it under control
403 2011-05-05 02:52:00 <vorlov> in the mean time, im wondering... have anyone tried to run a miner as a "Service" in windows?
404 2011-05-05 02:52:15 <Diablo-D3> dpesnt seem worth it
405 2011-05-05 02:58:12 <Diablo-D3> alright Im going to bed
406 2011-05-05 02:58:13 <Diablo-D3> night
407 2011-05-05 02:59:16 <xelister> Diablo-D3:
408 2011-05-05 02:59:34 <xelister> an idea how to gt r5970 mem to 300? I got it to 500 so far
409 2011-05-05 02:59:45 gasteve_ has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
410 2011-05-05 03:00:54 Diablo-D3 has quit (Read error: Operation timed out)
411 2011-05-05 03:04:49 wally__ has quit (Ping timeout: 252 seconds)
412 2011-05-05 03:05:55 xelister has quit (Remote host closed the connection)
413 2011-05-05 03:08:40 sc8nt4u has quit (Ping timeout: 246 seconds)
414 2011-05-05 03:26:01 bryan has joined
415 2011-05-05 03:26:24 threepio_ has joined
416 2011-05-05 03:26:24 threepio has quit (Read error: Connection reset by peer)
417 2011-05-05 03:26:24 threepio_ is now known as threepio
418 2011-05-05 03:26:27 bryan is now known as Guest14071
419 2011-05-05 03:29:19 sc8nt4u has joined
420 2011-05-05 03:30:58 mmoya has joined
421 2011-05-05 03:33:14 coconnor has joined
422 2011-05-05 03:33:33 kika has joined
423 2011-05-05 03:34:08 <kika> jgarzik
424 2011-05-05 03:36:37 coconnor has left ()
425 2011-05-05 03:38:04 wolfspraul has quit (Quit: leaving)
426 2011-05-05 03:40:09 peck has quit (Read error: Connection reset by peer)
427 2011-05-05 03:40:25 coconnor has joined
428 2011-05-05 03:40:54 peck has joined
429 2011-05-05 03:41:17 peck has quit (Read error: Connection reset by peer)
430 2011-05-05 03:44:01 peck has joined
431 2011-05-05 03:46:00 Kiba has joined
432 2011-05-05 03:46:15 fimp has joined
433 2011-05-05 03:50:25 <davex__> molecular
434 2011-05-05 03:55:03 bk128pc has joined
435 2011-05-05 04:05:50 <kika> so basically each node has a list of txs in ram, when a node wants to create a txt it notifies every other node about a new tx and then every other node adds the txt to his ram?
436 2011-05-05 04:06:33 Guest14071 has quit (Remote host closed the connection)
437 2011-05-05 04:07:54 TheKid has joined
438 2011-05-05 04:07:54 TheKid has quit (Changing host)
439 2011-05-05 04:07:54 TheKid has joined
440 2011-05-05 04:12:02 fimp has quit (Quit: This computer has gone to sleep)
441 2011-05-05 04:12:04 jarly_ has joined
442 2011-05-05 04:14:24 jarly_ has left ()
443 2011-05-05 04:18:40 redMBA has quit (Ping timeout: 248 seconds)
444 2011-05-05 04:20:04 x6763 has quit (Ping timeout: 246 seconds)
445 2011-05-05 04:20:46 tippenein has joined
446 2011-05-05 04:22:37 <kika> anyone here?
447 2011-05-05 04:23:03 <kika> anyone can help me test my pool?
448 2011-05-05 04:26:23 phantomcircuit has quit (Remote host closed the connection)
449 2011-05-05 04:28:10 zyb has joined
450 2011-05-05 04:29:34 x6763 has joined
451 2011-05-05 04:34:14 phantomcircuit has joined
452 2011-05-05 04:34:38 <kika> jgarzik: are you there?
453 2011-05-05 04:34:38 <kika> luke-jr|otg: are you there?
454 2011-05-05 04:34:39 x5x`brb is now known as x5x
455 2011-05-05 04:34:42 <kika> sipa: ?
456 2011-05-05 04:34:50 <kika> ArtForz: ?
457 2011-05-05 04:34:58 <kika> sacarlson: ?
458 2011-05-05 04:35:14 <sacarlson> kika who me?
459 2011-05-05 04:35:30 <kika> sacarlson: yes
460 2011-05-05 04:35:34 <kika> sacarlson: when a node wants to create tx
461 2011-05-05 04:35:46 <sacarlson> kika: what do I need to run to test your pool?
462 2011-05-05 04:35:56 <kika> sacarlson: it notifies all other nodes about the tx right?
463 2011-05-05 04:35:56 <kika> sacarlson: its a pushpoold
464 2011-05-05 04:36:04 <kika> sacarlson: which miners do you have?
465 2011-05-05 04:36:22 x5x is now known as x5x`brb
466 2011-05-05 04:36:24 x5x`brb is now known as x5x
467 2011-05-05 04:36:26 <sacarlson> kika I just run the one built into bitcoind to drive my net
468 2011-05-05 04:37:01 <kika> sacarlson: that one is pundippop one? its very slow
469 2011-05-05 04:37:01 <[Tycho]> Oh, another new pool ? :)
470 2011-05-05 04:37:07 <sacarlson> kika: will it run on ubuntu ?
471 2011-05-05 04:37:16 <kika> sacarlson: which one?
472 2011-05-05 04:37:43 <kika> sacarlson: which miner?
473 2011-05-05 04:37:43 <kika> sacarlson: do you have a GPU?
474 2011-05-05 04:37:43 <kika> sacarlson: do you have diablominer?
475 2011-05-05 04:38:09 <sacarlson> kika no it just drives 1/2 of my dual amd procesor
476 2011-05-05 04:38:20 <sacarlson> no gpu
477 2011-05-05 04:38:37 <kika> okay you can try using
478 2011-05-05 04:38:40 <sacarlson> you don't need alot of power to drive a small net
479 2011-05-05 04:39:11 <kika> sacarlson: but the main bitcoin network is very big
480 2011-05-05 04:39:28 <sacarlson> kika yes it is
481 2011-05-05 04:40:01 <sacarlson> another reason to have multi net I think it will require less power to run
482 2011-05-05 04:40:08 <kika> sacarlson: im testing this on the main network
483 2011-05-05 04:40:58 <sacarlson> kika: that's fine if you need help on main I would be glad to help just choose a minner for me to run if you just want to verify comunication
484 2011-05-05 04:41:36 <kika> you dont have a gpu so you wont be able to test gpu miners :(
485 2011-05-05 04:42:17 <sacarlson> kika: oh IC, well I have a nvida gpu but I don't think I've seen miners that suport it
486 2011-05-05 04:42:32 <kika> sacarlson: thxs for your help im falling asleep now, ill get back tomorrow so maybe we can test
487 2011-05-05 04:42:48 <sacarlson> but I thought minners ran on anything just slower with less power cpu gpu
488 2011-05-05 04:43:15 <kika> sacarlson: what do you mean?
489 2011-05-05 04:44:21 <sacarlson> kika: I'm not sure what protocol your minners use it should work with any engin to share the resources available on each node
490 2011-05-05 04:45:47 <tippenein> what do I have to do on linux besides put bitcoin bin in /usr/bin ?
491 2011-05-05 04:45:59 <sacarlson> I guess the job of the pool is to detect the amount of work being performed by each node and determine the payoff distibution to the amount of work done by each
492 2011-05-05 04:46:24 <sacarlson> tippenein: ubuntu?
493 2011-05-05 04:46:34 <tippenein> basically
494 2011-05-05 04:46:40 <tippenein> gnome
495 2011-05-05 04:46:53 <sacarlson> tippenein: there is a package that you can get from ppa
496 2011-05-05 04:47:17 <sacarlson> tippenein: it will only work on main net not -testnet
497 2011-05-05 04:47:23 phantomcircuit has quit (Read error: Connection reset by peer)
498 2011-05-05 04:49:14 <tippenein> thanks, i'm adding repo now
499 2011-05-05 04:49:17 <sacarlson> tippenein: I've personaly tested this package on ubuntu 10.04 https://launchpad.net/~stretch/+archive/bitcoin on main
500 2011-05-05 04:49:51 ragga is now known as GRIBBLEZ
501 2011-05-05 04:50:50 GRIBBLEZ is now known as ragga
502 2011-05-05 04:50:51 <kika> what would happen if i have a mother wiht 4 pcie slots
503 2011-05-05 04:50:51 <kika> and i connect 4 ati cards to it?
504 2011-05-05 04:50:55 <kika> without crossfire?
505 2011-05-05 04:51:20 <JFK911> nothing bad
506 2011-05-05 04:51:28 <kika> how it would choose which graphic card to use as main output?
507 2011-05-05 04:51:32 <tippenein> ty, that repository works perfect
508 2011-05-05 04:52:16 <sacarlson> tippenein: cool remember it only works on main it's the old version so the -testnet is no longer supported
509 2011-05-05 04:52:24 <kika> anyone here can test using diablominer
510 2011-05-05 04:52:24 <kika> to connect to my pool ?
511 2011-05-05 04:55:24 <sacarlson> kika this one https://github.com/Diablo-D3/DiabloMiner ?
512 2011-05-05 04:55:44 <kika> sacarlson: yes
513 2011-05-05 04:55:44 <kika> sacarlson: it only works for GPU
514 2011-05-05 04:55:57 <sacarlson> kika it seems to suport my nvidia
515 2011-05-05 04:55:59 <sacarlson> gpu
516 2011-05-05 04:56:19 <sacarlson> but like I said I think it will still run without just slower
517 2011-05-05 04:56:27 <sacarlson> I'll see if it compiles
518 2011-05-05 04:57:00 <kika> sacarlson: you dont need to compile it just run it using java
519 2011-05-05 04:57:50 <sacarlson> kika it must have binaries then?
520 2011-05-05 04:58:12 <sacarlson> I'll download it and take a look
521 2011-05-05 04:58:56 <kika> sacarlson: you can get the command line for diablominer from deepbit.net
522 2011-05-05 04:58:59 <kika> sacarlson: the diablominer you download it from github
523 2011-05-05 04:59:10 diki has quit (Read error: Connection reset by peer)
524 2011-05-05 04:59:32 diki has joined
525 2011-05-05 05:00:18 <tippenein> what does it mean if it won't connect and start generating the blocks?
526 2011-05-05 05:00:31 <kika> sacarlson: guiminer should work with your card i think
527 2011-05-05 05:00:32 <kika> sacarlson: i need to go to sleep
528 2011-05-05 05:00:37 <kika> sacarlson: we continue tomorro
529 2011-05-05 05:00:38 <kika> wthanks
530 2011-05-05 05:00:40 c0ldfront has joined
531 2011-05-05 05:00:44 coldfront has quit (Read error: Connection reset by peer)
532 2011-05-05 05:00:51 <sacarlson> kika ok night
533 2011-05-05 05:01:29 <sacarlson> tippenein: won't connect? you running in -testnet?
534 2011-05-05 05:01:59 <tippenein> i did no arguments after bitcoin cmd
535 2011-05-05 05:02:45 <sacarlson> tippenein: when it first connects in main as you started it, it will need to download the chain so you should see the block count increasing
536 2011-05-05 05:02:59 <tippenein> that's what I'm not seeing
537 2011-05-05 05:04:03 <sacarlson> yes I guess first you should see a connection count that should go up to about 8, so maybe comunication problems? does your isp filter IRC?
538 2011-05-05 05:04:26 <sacarlson> if your here now on the same system than that must not be it
539 2011-05-05 05:04:39 <tippenein> never had a problem before. I used to have bitcoin on this same computer running windows
540 2011-05-05 05:04:52 phantomcircuit has joined
541 2011-05-05 05:05:45 <sacarlson> tippenein: user privige problem? I guess to start try sudo netstat -pant and see if you see connections on port 8333 and port 8332
542 2011-05-05 05:05:50 kika has quit (Ping timeout: 252 seconds)
543 2011-05-05 05:06:03 <tippenein> howbout sudo bitcoin?
544 2011-05-05 05:06:17 <sacarlson> tippenein: that shouldn't be required
545 2011-05-05 05:06:39 <sacarlson> if those ports are open you won't need sudo to run bitcoin
546 2011-05-05 05:07:01 <tippenein> 8333 and 8332 are not shown in netstat -pant
547 2011-05-05 05:07:24 <sacarlson> are there any 18333 or 18332 ?
548 2011-05-05 05:07:40 <tippenein> nothing about 5145
549 2011-05-05 05:07:58 <sacarlson> you did run sudo netstat -pant since netstat require sudo
550 2011-05-05 05:08:50 gjs278 has quit (Read error: Connection reset by peer)
551 2011-05-05 05:09:04 <tippenein> yes
552 2011-05-05 05:09:18 gjs278 has joined
553 2011-05-05 05:09:40 <sacarlson> you can take a look at the log file I think it's in /home/youruser/.bitcoin and see what error it must have passed to not be able to open those ports
554 2011-05-05 05:09:44 vorlov has quit (Quit: vorlov)
555 2011-05-05 05:11:03 donchate has quit (Quit: in /dev/null no one can hear you scream...)
556 2011-05-05 05:11:09 amiller has quit (Remote host closed the connection)
557 2011-05-05 05:11:27 TheKid has quit (Quit: TheKid)
558 2011-05-05 05:11:34 <sacarlson> you might check the user priv in System>administration>user and groups>advanced
559 2011-05-05 05:12:41 <tippenein> that's the ticket
560 2011-05-05 05:12:41 <sacarlson> and make sure allow connect to wireless and ethernet networks is checked
561 2011-05-05 05:12:43 zyb_ has joined
562 2011-05-05 05:12:48 <tippenein> thanks
563 2011-05-05 05:13:04 <tippenein> msg me your bc addr?
564 2011-05-05 05:13:07 phantomcircuit has quit (Remote host closed the connection)
565 2011-05-05 05:13:21 <sacarlson> tippenein: cool ok just a moment
566 2011-05-05 05:13:33 <tippenein> i'm gonna climb mtgox
567 2011-05-05 05:13:42 <gjs278> ;;bc,mtgox
568 2011-05-05 05:13:43 <gribble> {"ticker":{"high":3.5699,"low":3.3,"vol":13081,"buy":3.42,"sell":3.445,"last":3.4449}}
569 2011-05-05 05:14:27 <tippenein> <---should've bought when they were $.3 / bc : (
570 2011-05-05 05:15:51 amiller has joined
571 2011-05-05 05:15:51 zyb has quit (Ping timeout: 260 seconds)
572 2011-05-05 05:15:52 snowdog has joined
573 2011-05-05 05:17:47 Beremaat has joined
574 2011-05-05 05:19:20 <sacarlson> tippenein: my bc address 12XFAN22VQFNcYwa9iNB3wZBTAFvvizK3j
575 2011-05-05 05:19:35 Beremat has quit (Ping timeout: 246 seconds)
576 2011-05-05 05:22:35 sacarlson has quit (Read error: Connection reset by peer)
577 2011-05-05 05:24:03 coconnor has quit (Quit: Page closed)
578 2011-05-05 05:25:02 <tippenein> is dwolla a good way to transfer funds?
579 2011-05-05 05:25:08 lulzplzkthx has quit (Ping timeout: 252 seconds)
580 2011-05-05 05:30:38 <nanotube> tippenein: people seem to like it
581 2011-05-05 05:30:43 lulzplzkthx has joined
582 2011-05-05 05:31:02 <tippenein> some ppl seem to like Sarah Palin too.. but that doesn't make it good
583 2011-05-05 05:32:35 <midnightmagic> "it"
584 2011-05-05 05:34:16 <midnightmagic> anyone want some testnet coins?
585 2011-05-05 05:34:19 <midnightmagic> free
586 2011-05-05 05:35:20 skyewm has quit (Ping timeout: 246 seconds)
587 2011-05-05 05:39:03 lzsaver has quit (Ping timeout: 260 seconds)
588 2011-05-05 05:41:17 sacarlson has joined
589 2011-05-05 05:43:41 RenaKunisaki has joined
590 2011-05-05 05:45:11 AStove has joined
591 2011-05-05 05:46:47 bk128pc has quit ()
592 2011-05-05 05:49:49 snowdog has quit (Ping timeout: 252 seconds)
593 2011-05-05 05:50:26 zyb_ has quit (Remote host closed the connection)
594 2011-05-05 05:52:45 EPiSKiNG has quit ()
595 2011-05-05 05:54:58 <CIA-30> bitcoin: various watch * r5c3a55..878bef supybot-bitcoin-marketmonitor-personal/ (config.py bitcoin.py plugin.py): (21 commits) http://tinyurl.com/5sefxl7
596 2011-05-05 05:56:16 EPiSKiNG has joined
597 2011-05-05 05:56:39 jargon has joined
598 2011-05-05 05:57:25 <jargon> Ok, I can't do more of the source audit tonight because I have court in the morning and may potentially end up in prison again for dumbass crap I didn't even do.
599 2011-05-05 05:58:06 <jargon> I will however just trace that null root Teller memory model pointer.
600 2011-05-05 05:58:35 <jargon> It sounds to me like the developers of bitcoin don't even understand their own code.
601 2011-05-05 05:58:38 danbri has joined
602 2011-05-05 05:59:02 <doublec> the current developers didn't write it
603 2011-05-05 05:59:05 <luke-jr> jargon: Satoshi wrote the original code
604 2011-05-05 05:59:11 <luke-jr> jargon: and he's gone forever
605 2011-05-05 06:00:00 <midnightmagic> prison, jesus. what did you do?
606 2011-05-05 06:00:03 <midnightmagic> er.. not do?
607 2011-05-05 06:01:01 <citiz3n> what do you mean satoshi is gone?
608 2011-05-05 06:01:24 tippenein has quit (Remote host closed the connection)
609 2011-05-05 06:02:05 <doublec> satoshi is no longer involved in bitcoin
610 2011-05-05 06:02:13 <citiz3n> link?
611 2011-05-05 06:02:26 <luke-jr> â¦
612 2011-05-05 06:02:53 <citiz3n> i thought he was still involved somewhat
613 2011-05-05 06:02:59 <citiz3n> i never heard that he walked away from it
614 2011-05-05 06:03:05 <doublec> citiz3n: see https://en.bitcoin.it/wiki/Satoshi_Nakamoto
615 2011-05-05 06:03:17 <doublec> "His involvement in the Bitcoin project had tapered and by late 2010 it has ended. The most recent messages reportedly indicate that Satoshi is "gone for good""
616 2011-05-05 06:03:26 <doublec> there's a link to a cite there
617 2011-05-05 06:04:14 <doublec> see here http://veritas.maximilianeum.ch/bitcoin/irc/logs/2011/04/26#l1530
618 2011-05-05 06:05:07 <jargon> line 217 in db.h: DbTxn* ptxn = NULL;
619 2011-05-05 06:05:32 <jargon> I don't know if it is 216 or 217 due to line numbering in the github viewer being mis-aligned
620 2011-05-05 06:05:59 <jrabbit> lol satoshi conspiracy.
621 2011-05-05 06:06:01 ForceDestroyer has joined
622 2011-05-05 06:06:04 <doublec> jargon: what about that line?
623 2011-05-05 06:06:16 <jargon> That is the only series issue I have seen due to only doing a partial audit on that single source-file thus far.
624 2011-05-05 06:06:22 <jargon> serious issue*
625 2011-05-05 06:06:23 <doublec> it's 217 btw
626 2011-05-05 06:06:30 <doublec> in what way?
627 2011-05-05 06:07:25 <jargon> It is a Teller memory modeled database
628 2011-05-05 06:07:50 <jargon> You don't index the root pointer as NULL, ever.
629 2011-05-05 06:08:10 midnightmagic has quit (Ping timeout: 258 seconds)
630 2011-05-05 06:08:59 Kicchiri has quit (Ping timeout: 240 seconds)
631 2011-05-05 06:09:14 <doublec> That code creates a new transaction and gets a pointer to it
632 2011-05-05 06:09:17 <doublec> how is that wrong again?
633 2011-05-05 06:09:35 <doublec> and how does it relate to "index the root pointer as NULL"
634 2011-05-05 06:11:53 <jargon> just give me an hour to manually trace
635 2011-05-05 06:13:14 Beremat has joined
636 2011-05-05 06:13:15 Beremaat has quit (Read error: Connection reset by peer)
637 2011-05-05 06:13:23 <jargon> first-off, the 'vector' directive is obsolete.
638 2011-05-05 06:13:24 * [Tycho] is feeling lucky
639 2011-05-05 06:13:38 <doublec> jargon: what 'vector' directive?
640 2011-05-05 06:13:52 <jargon> line 17
641 2011-05-05 06:13:57 <jargon> db.h
642 2011-05-05 06:14:07 <doublec> line 17 is empty in db.h
643 2011-05-05 06:14:38 <jargon> extern vector<unsigned char> vchDefaultKey;
644 2011-05-05 06:14:42 <doublec> the only use of vector in db.h is the standard C++ vector class
645 2011-05-05 06:14:47 <doublec> how is that obsolete?
646 2011-05-05 06:15:07 <jargon> nobody has improved the backend for vector in 20 some years.
647 2011-05-05 06:15:25 dissipate has joined
648 2011-05-05 06:15:32 <doublec> ah, I see
649 2011-05-05 06:15:36 <doublec> you have no idea what you're talking about
650 2011-05-05 06:15:39 <doublec> now I understand
651 2011-05-05 06:16:11 PLATO has joined
652 2011-05-05 06:16:13 <PLATO> PLATO> there's a thread about distributed exchange discussion
653 2011-05-05 06:16:13 <PLATO> PLATO> i want to decide on a standard web of trust API so many sites that use webs of trust can communicate to each other
654 2011-05-05 06:16:16 <PLATO> PLATO> if anyone is interested in working on this, I have set up a mailing list for project discussion
655 2011-05-05 06:16:17 <doublec> either that, or I'm completely misunderstanding you
656 2011-05-05 06:16:19 <PLATO> PLATO> details here: http://www.bitcoin.org/smf/index.php?topic=7096.msg107062#msg107062
657 2011-05-05 06:16:30 <jargon> Vapor was a cycle-sharing distributed database that was pretty similar to what bitcoin is hush-hush striving to be.
658 2011-05-05 06:16:44 <doublec> Vapor is what you've been inhaling
659 2011-05-05 06:16:52 <doublec> you should sell it for btc
660 2011-05-05 06:17:07 <jargon> Vapor's Teller was ridonculously faster than vector
661 2011-05-05 06:17:27 <doublec> a vector is an array of data
662 2011-05-05 06:17:28 <jargon> You are trying to combine vector with Teller. it don't work that way without issues down the road.
663 2011-05-05 06:17:37 <doublec> ok, you definitely are tolling
664 2011-05-05 06:17:40 <doublec> trolling
665 2011-05-05 06:17:43 <doublec> sorry for the noise
666 2011-05-05 06:17:47 <jargon> No, I am sorry to say I am not.
667 2011-05-05 06:18:16 <jargon> Also, vector uses ass-tons more memory than the Teller method.
668 2011-05-05 06:18:39 <jargon> ..in the long run
669 2011-05-05 06:19:41 <jargon> you are trying to make a Teller DB using a vector back-end.
670 2011-05-05 06:19:50 <jargon> That is a major no-no.
671 2011-05-05 06:20:40 <jargon> It is as if someone started work on that, then halfassed a compile.
672 2011-05-05 06:23:33 mologie has joined
673 2011-05-05 06:24:11 AStove has quit ()
674 2011-05-05 06:24:16 [Tycho] has quit (Ping timeout: 258 seconds)
675 2011-05-05 06:28:36 Andrevan has quit ()
676 2011-05-05 06:32:47 <jargon> For now I am starting to trace whether a NULL Key/Value passed to the entire p2p will hijack the entire db.
677 2011-05-05 06:33:32 <jargon> Doing so may induce everyone's wallets into having zero value.
678 2011-05-05 06:33:52 <jargon> (meaning all coins are deleted in the entire p2p)
679 2011-05-05 06:34:14 <PLATO> o.O
680 2011-05-05 06:38:00 DD- has joined
681 2011-05-05 06:38:00 midnightmagic has joined
682 2011-05-05 06:41:12 <jargon> yeah the data integrity protection in bitcoin is total crap
683 2011-05-05 06:43:57 <wumpus> it's not even possible to delete coins
684 2011-05-05 06:46:55 Kiba has quit (Remote host closed the connection)
685 2011-05-05 06:48:15 <wumpus> data integrity is actually good, as everything is hashed with the previous block; it is similar to how version control systems such as git/mercurial work... unless you can rewind the block chain, there is no way to make intermediate blocks/coins disappear
686 2011-05-05 06:53:46 x5x is now known as x5x`brb
687 2011-05-05 06:56:50 <jargon> collisions and procedure jmp's
688 2011-05-05 06:57:09 <jargon> two different ways to do it
689 2011-05-05 07:00:06 <jargon> plus you don't even have so much as an event chain integrity checksum
690 2011-05-05 07:00:42 <jargon> The hashes don't mean dick.
691 2011-05-05 07:02:05 <jargon> you don't even have a register of flags for what processes did what under which forks.
692 2011-05-05 07:02:48 <jargon> You have zero integrity protection against middle-end procedure jumps
693 2011-05-05 07:03:46 <nanotube> jargon: please file bug reports on the github issue tracker?
694 2011-05-05 07:03:48 x5x`brb is now known as x5x
695 2011-05-05 07:03:58 <jargon> in-otherwords, your stack is wide open to corruption
696 2011-05-05 07:04:09 bgeron has joined
697 2011-05-05 07:04:11 <jargon> nanotube i am doing an audit, not a bug report
698 2011-05-05 07:06:08 mologie has quit (Quit: Nettalk6 - www.ntalk.de)
699 2011-05-05 07:06:37 <nanotube> an audit, if it finds anything, would normally result in bug reports.
700 2011-05-05 07:06:44 <jargon> I freelance for the United States DARPA TTO
701 2011-05-05 07:07:16 <jargon> If anyone knows how to properly do an audit, it is me.
702 2011-05-05 07:08:10 <ersi> jargon: DERP DERP DERP
703 2011-05-05 07:08:22 <jargon> nanotube, the entire db.h is jacked to Hell --and that is the only source-file I have looked at in this audit so far.
704 2011-05-05 07:08:51 <DD-> i derp for the US Derpie derp
705 2011-05-05 07:08:55 <ersi> wumpus: Be aware, jargon is a trolololo.
706 2011-05-05 07:09:28 <jargon> it lacks self-detection of procedure chain compromise.
707 2011-05-05 07:09:39 <DD-> yes
708 2011-05-05 07:09:49 <DD-> and backward encryptions methods
709 2011-05-05 07:10:06 <wumpus> ah, right,I see now
710 2011-05-05 07:10:20 <wumpus> nice technobabble, why don't you write a scifi novel instead :)
711 2011-05-05 07:10:45 <JFK911> ;;bc,mtgox
712 2011-05-05 07:10:46 <gribble> {"ticker":{"high":3.5699,"low":3.3,"vol":13026,"buy":3.42,"sell":3.445,"last":3.42}}
713 2011-05-05 07:11:26 <jargon> And program that doesn't detect a direct jump into an given procedure of-which skirts the external API, is dead in the water these days.
714 2011-05-05 07:11:27 pirrr has joined
715 2011-05-05 07:11:43 <jargon> Any program*
716 2011-05-05 07:12:29 <diki> luke, why has my balance been reset?
717 2011-05-05 07:12:43 <jargon> Fawkes@Home
718 2011-05-05 07:14:10 LightRIder is now known as LightRider|afk
719 2011-05-05 07:14:36 <jargon> using a hash as a security measure is obsolete
720 2011-05-05 07:15:34 <diki> and what isnt?
721 2011-05-05 07:15:53 noagendamarket has quit (Ping timeout: 252 seconds)
722 2011-05-05 07:15:56 <jargon> You even realize you have a turing-complete interpreter up in this bitch?
723 2011-05-05 07:16:18 <DD-> yes
724 2011-05-05 07:16:19 <diki> i use a triple md5 hashed salted password
725 2011-05-05 07:16:34 <wumpus> not only that, it can even fly a helicopter if connected right
726 2011-05-05 07:17:07 <DD-> we did build in a defense system against portal guns tho , u can find it in portal.h
727 2011-05-05 07:17:12 <DD-> very well start of the art
728 2011-05-05 07:17:48 noagendamarket has joined
729 2011-05-05 07:18:09 <wumpus> yes the state of the art defense systems inside bitcoin even surprise the secret cults of cthulhu
730 2011-05-05 07:19:07 larsivi has joined
731 2011-05-05 07:19:35 <wumpus> but the layman might not be able to distinghuish them as they work very subtle and use colored quantum hashing with unicorn dust
732 2011-05-05 07:20:35 <DD-> ah damn , didnt know that you pushed the unicorn dust version live
733 2011-05-05 07:20:39 <DD-> thought it was still in testing
734 2011-05-05 07:25:52 <wumpus> ah, testing, I knew we forgot something
735 2011-05-05 07:26:24 <wumpus> nah, who tests weapons of mass distraction anyway
736 2011-05-05 07:26:34 <jargon> You aren't even using the Keal Demux method for base58, and you really think this crap can't have it's hashing out-run?
737 2011-05-05 07:27:28 <wumpus> yes the Keal Demux would give it that extra 'oomph'
738 2011-05-05 07:27:30 <pirrr> I use 0.3.19beta and maybe you already know this, but when I suspend my PC and wake it up again (a few times), over time I lose all my connections and have to restart bitcoin. Soneone might want to look into that?
739 2011-05-05 07:27:44 <pirrr> someone*
740 2011-05-05 07:27:58 <jargon> Keal Demux instantly solves each digit, instead of cycling through counting with a remainder.
741 2011-05-05 07:28:35 <wumpus> it instantly solves digits? that must be heavy acid, man
742 2011-05-05 07:28:35 ersi has quit (Ping timeout: 260 seconds)
743 2011-05-05 07:28:49 <jargon> Just Google it you fat retard fuck,
744 2011-05-05 07:28:53 <jargon> :P
745 2011-05-05 07:29:22 <DD-> how about we feed Keal Demux a base quantum phased twelve flip-flop duplex , and send it back in time?
746 2011-05-05 07:29:28 <nanotube> pirrr: first, try latest version... and if that persists, file an issue on github (i think it may be a known problem though... so check)
747 2011-05-05 07:29:47 <jargon> Keal Demux works with irrational bases too, assuming you have a nice little laser lab.
748 2011-05-05 07:29:51 <wumpus> pirrr: sleep mode causes all network connections to be broken, which makes sense, but I'm less sure why they are not reestablished
749 2011-05-05 07:30:18 <pirrr> Allright, I'll check
750 2011-05-05 07:30:20 ersi has joined
751 2011-05-05 07:31:06 Props has joined
752 2011-05-05 07:31:07 <wumpus> I knew nature was pulling a base58 behind our backs, we should have known all along
753 2011-05-05 07:31:22 wolfspraul has joined
754 2011-05-05 07:31:27 <wumpus> maybe you should patch the source and introduce this wonderful technology
755 2011-05-05 07:31:49 <jargon> If you have a laserlab, you can solve any base-foo digit of pi in a single step with Keal Demux
756 2011-05-05 07:32:17 <DD-> i got a druglab , will that work?
757 2011-05-05 07:32:47 <jargon> You can even detect fluctuations in temporal space itself due to any given digit being deformed.
758 2011-05-05 07:33:27 <DD-> that wont work
759 2011-05-05 07:33:52 <DD-> what if space anomalies turn up with broccoli ?
760 2011-05-05 07:33:52 <wumpus> depends on the kind of drugs, DD-, I think you need hallucinogenics to remove all cognitive barriers and experience the universe as it really is, and then patch it into the timespace fabric
761 2011-05-05 07:34:11 <jargon> Watch the youtube video under mediaplague regarding your "that wont work" jackass. :)
762 2011-05-05 07:34:19 <DD-> again , the broccoli will catch you in the base58
763 2011-05-05 07:34:31 <jargon> my show on Youtube is whack as fuck
764 2011-05-05 07:34:47 <jargon> But there is a lot of ridiculously advanced mathematics in there.
765 2011-05-05 07:34:57 amiller has quit (Remote host closed the connection)
766 2011-05-05 07:35:15 <wumpus> yeah of course it will work, but fr some reason the cams are always of flakey quality and the person holding the cam can't hold it still, so effectively you see nothing
767 2011-05-05 07:35:26 <jargon> It even shows how to grow a specific biological weapon.
768 2011-05-05 07:36:28 <DD-> does this biological weapon happends to have anything todo with broccoli?
769 2011-05-05 07:36:31 <jargon> ..out of common fridge items.
770 2011-05-05 07:36:41 <jargon> no
771 2011-05-05 07:36:45 <wumpus> that's good, everyone should have biological weapons
772 2011-05-05 07:36:51 <wumpus> to protect against other biological weapons
773 2011-05-05 07:37:36 <wumpus> I usually just grow them out of my nose when I am lazy
774 2011-05-05 07:37:48 <_ape> i've been the victim of biological weapons in several elevators/movie theaters
775 2011-05-05 07:38:44 <Props> where's a quantum computer when you need one
776 2011-05-05 07:38:50 <jargon> People are starting to catch on that pretty much everything in the Supervillain Sabbath show is real despite how surreal.
777 2011-05-05 07:40:29 <wumpus> which isn't surprising, as we all know, everything is true
778 2011-05-05 07:40:58 <jargon> I even had one dude throwing death threats my way over the show.
779 2011-05-05 07:41:04 <DD-> ah a famous Keal Demux quote
780 2011-05-05 07:41:08 <DD-> 1 of my favorite
781 2011-05-05 07:41:34 <jargon> And yes, I did Fukushima, not the jews.
782 2011-05-05 07:42:38 danlucraft has joined
783 2011-05-05 07:44:26 bk128 has quit (Quit: bk128)
784 2011-05-05 07:44:26 danlucraft has quit (Read error: Connection reset by peer)
785 2011-05-05 07:45:38 [Tycho] has joined
786 2011-05-05 07:47:46 [Tycho] has quit (Changing host)
787 2011-05-05 07:47:46 [Tycho] has joined
788 2011-05-05 07:48:38 wolfspraul has quit (Quit: leaving)
789 2011-05-05 07:48:47 wolfspraul has joined
790 2011-05-05 07:50:01 <jargon> You can't even fathom the complex of such a device as-to wormhole a mile tall tsunami around Hawaii and into Japan a year later.
791 2011-05-05 07:51:27 <DD-> i can phantom is tho , dunno if thats any good
792 2011-05-05 07:51:46 BitterTea has quit (Quit: leaving)
793 2011-05-05 07:52:24 m00p has joined
794 2011-05-05 07:53:03 <jargon> crunch the numbers, see that the origin and destination don't align in the Solar calendar.
795 2011-05-05 07:55:05 <wumpus> but that's because your computatins don't take into account earth's second, dark moon Lilith
796 2011-05-05 07:55:54 <DD-> or the great moon from saturn , Titan
797 2011-05-05 08:00:43 <jargon> There is a second Earth on the far side of the sun.
798 2011-05-05 08:00:48 dissipate has quit (Read error: Connection reset by peer)
799 2011-05-05 08:00:48 <jargon> ..same orbit path.
800 2011-05-05 08:01:17 <DD-> yes
801 2011-05-05 08:01:19 <wumpus> how conveniently out of sight
802 2011-05-05 08:01:27 <DD-> between the space mirror , into the vortex!
803 2011-05-05 08:02:36 <jargon> NASA has images of it, using a probe launched perpendicular to the Solar disc.
804 2011-05-05 08:03:11 <jargon> Never thought beyond two dimensions did you?
805 2011-05-05 08:03:11 <wumpus> ssh, that's where the unicorns live, now everyone will go there to harvest unicorn meat and sell it on ebay
806 2011-05-05 08:03:18 <DD-> THE DUST!
807 2011-05-05 08:03:23 <DD-> we must save the dust
808 2011-05-05 08:05:12 <jargon> ..and yes I have made a star system model which did not follow a ring of orbits, but an ordered set of complex orbits.
809 2011-05-05 08:05:22 <jargon> ..in stability of each other.
810 2011-05-05 08:17:35 <DD-> all that for the TTO uh?
811 2011-05-05 08:17:40 dishwara has joined
812 2011-05-05 08:20:09 zebrah_ has joined
813 2011-05-05 08:20:11 <jargon> Galaxies are expendable.
814 2011-05-05 08:20:21 <zebrah_> jargon
815 2011-05-05 08:20:33 <jargon> We already annihilated the Crab Nebula into quantum foam last year.
816 2011-05-05 08:20:52 <zebrah_> Joe I know you're in here
817 2011-05-05 08:21:07 <jargon> Well do you want to talk bitcoin or run the log or what?
818 2011-05-05 08:21:58 <zebrah_> I want a lot of things. But mostly I was looking for joe
819 2011-05-05 08:23:17 <diki> i only know of joe from vivthomas
820 2011-05-05 08:23:47 <zebrah_> Hmm I don't think he's here. I don't even know his name on here. So cya
821 2011-05-05 08:23:49 zebrah_ has left ()
822 2011-05-05 08:29:56 amiller has joined
823 2011-05-05 08:30:23 <gjs278> ;;tell Diablo-D3 http://www.newegg.com/Product/Product.aspx?Item=N82E16814125318 it's $154 with mail in rebate, I just got mine today it looks pretty solid
824 2011-05-05 08:30:33 <gribble> Error: I haven't seen Diablo-D3, I'll let you do the telling.
825 2011-05-05 08:30:35 <gjs278> gribble you faggot robot
826 2011-05-05 08:32:51 bk128 has joined
827 2011-05-05 08:33:58 mikehawkburn has joined
828 2011-05-05 08:34:08 mikehawkburn has quit (Client Quit)
829 2011-05-05 08:35:39 gribble has quit (Remote host closed the connection)
830 2011-05-05 08:36:01 <DD-> now u done it gjs278 , you made gribble upset
831 2011-05-05 08:36:09 <gjs278> fuck gribble
832 2011-05-05 08:36:13 <gjs278> the one time I want to use ;;tell
833 2011-05-05 08:36:15 <gjs278> he can't do it
834 2011-05-05 08:37:20 sc8nt4u has quit (Ping timeout: 246 seconds)
835 2011-05-05 08:37:26 <jargon> http://www.youtube.com/watch?v=-Ft6gNYh0Xk eight-inch tall robot makes f- coffee
836 2011-05-05 08:44:29 <molecular> davex__, "<davex__> molecular", yes?
837 2011-05-05 08:45:36 x5x is now known as x5x`brb
838 2011-05-05 08:47:37 gribble has joined
839 2011-05-05 08:48:30 <diki> ;;bc,gen 340000
840 2011-05-05 08:48:43 <diki> ;;bc
841 2011-05-05 08:49:53 <gribble> The expected generation output, at 340000 Khps, given current difficulty of 109670.13329248 , is 3.11827513228 BTC per day and 0.129928130512 BTC per hour.
842 2011-05-05 08:49:54 <gribble> Error: "bc" is not a valid command.
843 2011-05-05 08:52:13 bk128 has quit (Quit: bk128)
844 2011-05-05 08:52:17 PLATO has quit (Remote host closed the connection)
845 2011-05-05 08:53:57 anarchyx has joined
846 2011-05-05 08:56:01 x5x`brb is now known as x5x
847 2011-05-05 08:56:56 <dirtyfilthy> oh when did the karma system come online in the forums
848 2011-05-05 08:59:49 <gjs278> ;;bc,mtgox
849 2011-05-05 08:59:56 <gribble> {"ticker":{"high":3.5699,"low":3.3,"vol":13052,"buy":3.4201,"sell":3.445,"last":3.4201}}
850 2011-05-05 09:00:32 danlucraft has joined
851 2011-05-05 09:01:21 <diki> ;;bc,calc
852 2011-05-05 09:01:22 <gribble> (bc,calc <an alias, 1 argument>) -- Alias for "echo The average time to generate a block at $1 Khps, given current difficulty of [bc,diff], is [time elapsed [math calc 1/((2**224-1)/[bc,diff]*$1*1000/2**256)]]".
853 2011-05-05 09:01:29 <diki> ;;bc,calc 340000
854 2011-05-05 09:01:30 <gribble> The average time to generate a block at 340000 Khps, given current difficulty of 109670.13329248 , is 2 weeks, 2 days, 0 hours, 49 minutes, and 41 seconds
855 2011-05-05 09:01:42 <diki> ;;bc,calc 342000
856 2011-05-05 09:01:43 <gribble> The average time to generate a block at 342000 Khps, given current difficulty of 109670.13329248 , is 2 weeks, 1 day, 22 hours, 34 minutes, and 39 seconds
857 2011-05-05 09:02:13 midnightmagic has quit (Ping timeout: 258 seconds)
858 2011-05-05 09:02:52 <diki> ;;bc,calc 345000
859 2011-05-05 09:02:53 <gribble> The average time to generate a block at 345000 Khps, given current difficulty of 109670.13329248 , is 2 weeks, 1 day, 19 hours, 15 minutes, and 3 seconds
860 2011-05-05 09:03:02 <diki> ;;bc,calc 350000
861 2011-05-05 09:03:03 <gribble> The average time to generate a block at 350000 Khps, given current difficulty of 109670.13329248 , is 2 weeks, 1 day, 13 hours, 49 minutes, and 58 seconds
862 2011-05-05 09:03:24 TheAncientGoat has joined
863 2011-05-05 09:05:16 thedrs has joined
864 2011-05-05 09:05:49 <thedrs> ;;bc,help
865 2011-05-05 09:05:50 <gribble> Alias bc,avgprc, Alias bc,bcm, Alias bc,blocks, Alias bc,btcex, Alias bc,calc, Alias bc,calcd, Alias bc,channels, Alias bc,diff, Alias bc,eligius, Alias bc,estimate, Alias bc,gen, Alias bc,gend, Alias bc,help, Alias bc,hextarget, Alias bc,labs, Alias bc,lbs, Alias bc,lukepool, Alias bc,mtgox, Alias bc,nexttarget, Alias bc,poolstats, Alias bc,prob, Alias bc,stats, Alias bc,timetonext, Alias (1 more message)
866 2011-05-05 09:06:00 <thedrs> ;;bc,gen 180000
867 2011-05-05 09:06:01 <gribble> The expected generation output, at 180000 Khps, given current difficulty of 109670.13329248 , is 1.65085154062 BTC per day and 0.0687854808592 BTC per hour.
868 2011-05-05 09:08:02 xelister has joined
869 2011-05-05 09:09:08 ahbritto_ has quit (Quit: Ex-Chat)
870 2011-05-05 09:09:43 <diki> ;;bc,eligius
871 2011-05-05 09:09:45 <gribble> 14688788.1523
872 2011-05-05 09:11:28 <thedrs> 14Ghash not bad
873 2011-05-05 09:11:38 echelon has quit (Ping timeout: 246 seconds)
874 2011-05-05 09:12:40 <thedrs> ;;bc,gen 8000
875 2011-05-05 09:12:40 <gribble> The expected generation output, at 8000 Khps, given current difficulty of 109670.13329248 , is 0.0733711795832 BTC per day and 0.00305713248263 BTC per hour.
876 2011-05-05 09:13:30 devon_hillard has joined
877 2011-05-05 09:15:17 <eps> ;;bc,stats
878 2011-05-05 09:15:19 <gribble> Current Blocks: 121980 | Current Difficulty: 109670.13329248 | Next Difficulty At Block: 122975 | Next Difficulty In: 995 blocks | Next Difficulty In About: 5 days, 3 hours, 16 minutes, and 10 seconds | Next Difficulty Estimate: 147838.93300532
879 2011-05-05 09:15:34 <eps> ;;bc,poolstats
880 2011-05-05 09:15:35 <gribble> {"ghashes_ps": "283.072", "shares": 223374, "active_workers": 2333, "round_duration": "0:56:20", "score": "138793.4033", "round_started": "2011-05-05 08:14:03", "shares_cdf": "86.96", "getwork_ps": 771}
881 2011-05-05 09:19:15 <sgornick> Anyone know anything about this? http://www.bitcoinplus.com/generate
882 2011-05-05 09:19:15 <cdecker> ;;bc,gen 343000
883 2011-05-05 09:19:16 <gribble> The expected generation output, at 343000 Khps, given current difficulty of 109670.13329248 , is 3.14578932463 BTC per day and 0.131074555193 BTC per hour.
884 2011-05-05 09:20:06 <diki> ;;bc,calc 6000
885 2011-05-05 09:20:07 <gribble> The average time to generate a block at 6000 Khps, given current difficulty of 109670.13329248 , is 2 years, 25 weeks, 3 days, 14 hours, 55 minutes, and 39 seconds
886 2011-05-05 09:25:27 <jargon> Oh, that's right, javascript.
887 2011-05-05 09:25:50 <jargon> never thought of that, did you?
888 2011-05-05 09:25:52 <wumpus> sgornick: seems to start a java applet when you click generate
889 2011-05-05 09:26:05 kseistrup has joined
890 2011-05-05 09:26:17 echelon has joined
891 2011-05-05 09:26:25 <jargon> Dude, I wrote an exact clone of DoomII in javascript, trust me, java is shit
892 2011-05-05 09:26:58 <wumpus> wonder how well CPU mining in a JRE will do, probably even slower than normal CPU mining, but heh, it's a fun idea
893 2011-05-05 09:27:08 <wumpus> yes webGL is nice
894 2011-05-05 09:27:56 <jargon> I wonder how well it will do in BlitzBasic
895 2011-05-05 09:28:06 <jargon> (considering BB is 9000x faster than C+)
896 2011-05-05 09:28:25 <wumpus> hehe
897 2011-05-05 09:29:28 <jargon> most of your mining time is wasted in Base58
898 2011-05-05 09:30:30 <jargon> As I said before, using Keal Demux instead of your shitty Base58 crap that looks like a 12yo wrote it, would annihilate damn near 100% of the lag.
899 2011-05-05 09:31:04 FabianB has quit (Ping timeout: 240 seconds)
900 2011-05-05 09:31:16 stitekili has joined
901 2011-05-05 09:32:26 <cosurgi> ;;isitdown bitcoinwatch.com
902 2011-05-05 09:32:27 <gribble> http://bitcoinwatch.com Is Up -> Check if your website is up or down?
903 2011-05-05 09:32:31 dishwara has quit (Quit: See u soon)
904 2011-05-05 09:32:33 <cosurgi> ;;isitdown mtgox.com
905 2011-05-05 09:32:33 <gjs278> it's working for me
906 2011-05-05 09:32:34 <gribble> http://mtgox.com Is Up -> Check if your website is up or down?
907 2011-05-05 09:33:09 kseistrup has quit (Quit: leaving)
908 2011-05-05 09:33:42 FabianB has joined
909 2011-05-05 09:33:44 <diki> ;
910 2011-05-05 09:33:46 <diki> ;;
911 2011-05-05 09:33:50 <diki> ;; help
912 2011-05-05 09:33:50 <gribble> The bot responds when you start a line with the ! character. A good starting point for exploring the bot is the !facts command. You can also visit the bot's website for a list of help topics and documentation: http://gribble.sourceforge.net/
913 2011-05-05 09:34:27 kseistrup has joined
914 2011-05-05 09:34:30 <jargon> Well, I'm gonna soup this bitch up and potentially generate severalt thousand bitcoins a minute.
915 2011-05-05 09:35:12 <jargon> That's what happens when you piss off the TTO with your bullshit lame coders with a Fawkes@Home client, they rob you blind.
916 2011-05-05 09:35:40 <DD-> Good luck! send a postcard when ur done
917 2011-05-05 09:36:38 <jargon> Vapor's gonna kick your ass flat to the ground
918 2011-05-05 09:36:52 <jargon> (Vapor is the "original" bitcoin)
919 2011-05-05 09:37:11 * DD- yawns
920 2011-05-05 09:37:19 stitekili has left ()
921 2011-05-05 09:37:24 <wumpus> more coding, less talk please :)
922 2011-05-05 09:41:48 <molecular> I'm trying to use amdovdrvctrl on a remote miner. setting DISPLAY=mymachine:0.0 results in no graphics card being found through adl, setting DISPLAY=:0 however results in me not seing pretty wx gui of the app. any ideas?
923 2011-05-05 09:42:16 ahbritto_ has joined
924 2011-05-05 09:42:18 <molecular> can I pass wx a different DISPLAY at init somehow?
925 2011-05-05 09:43:00 <sipa> molecular: afaik amdovdrctrl cannot be used on a remote display
926 2011-05-05 09:43:12 <sipa> what you can do howver is create a config file, and use it in batch mode
927 2011-05-05 09:43:17 <sipa> (without gui)
928 2011-05-05 09:43:47 <molecular> is there documentation or sample config file?
929 2011-05-05 09:44:03 <molecular> damn, who connects displays to their graphics card anyways?
930 2011-05-05 09:44:33 <gjs278> molecular what are you teying to do with it?
931 2011-05-05 09:44:35 <gjs278> lower your memclock?
932 2011-05-05 09:45:05 <gjs278> anyways
933 2011-05-05 09:45:10 <gjs278> you need to specify your adapter index
934 2011-05-05 09:45:12 <gjs278> like
935 2011-05-05 09:45:14 <gjs278> DISPLAY=:0.1 ./AMDOverdriveCtrl --adapter-index=3
936 2011-05-05 09:45:30 <molecular> no, I can lower memclock with my own tool ;). Just wanted to check out amdovdrvctrl out of curiosity
937 2011-05-05 09:45:35 <gjs278> but more importantly, you should set it up once, save your defaults, run it, ctrl+c it, and the memclock will stick
938 2011-05-05 09:45:37 <gjs278> oh
939 2011-05-05 09:45:41 <gjs278> what tool do you use
940 2011-05-05 09:45:42 <gjs278> for memclocks
941 2011-05-05 09:46:04 <gjs278> also amdoverctrl fucks with your fan
942 2011-05-05 09:46:15 <gjs278> like the second you actually get it running, it will lower your fan to like 30% on you
943 2011-05-05 09:46:29 <gjs278> and then you have to jack it back up really quick
944 2011-05-05 09:46:35 <molecular> gjs278, this wont work, I need DISPLAY=mydesktopmachine:0
945 2011-05-05 09:46:43 <molecular> gjs278, it wont even initialize adl
946 2011-05-05 09:46:48 <gjs278> you're lucky
947 2011-05-05 09:46:50 <sipa> molecular: i can give you mine
948 2011-05-05 09:47:00 <molecular> your what?
949 2011-05-05 09:47:18 <gjs278> if you can get it connected without the gui
950 2011-05-05 09:47:21 <gjs278> use the batch file method
951 2011-05-05 09:47:30 <sipa> molecular: http://pastebin.com/1bQKj4x8
952 2011-05-05 09:47:33 <molecular> thing is: I _want_ to see the gui, that's the whole point
953 2011-05-05 09:47:39 <gjs278> no you dont
954 2011-05-05 09:47:39 <cosurgi> molecular: what tool do you use to lower memclock? AMDOverdriveCtrl isn't working for me :(
955 2011-05-05 09:47:44 <gjs278> I'll screenshot it for you
956 2011-05-05 09:47:57 <molecular> cosurgi, I hacked one myself using adl calls
957 2011-05-05 09:48:07 <cosurgi> molecular: can you share it?
958 2011-05-05 09:48:09 <gjs278> it offers nothing except memclock lowering and crappy fan control
959 2011-05-05 09:48:18 <molecular> cosurgi, it was initially a PID fan regulator by ArtForz
960 2011-05-05 09:48:20 <gjs278> you can't overvolt
961 2011-05-05 09:48:37 <molecular> cosurgi, sure, but you have to set the values hardcoded in the code ;|
962 2011-05-05 09:48:46 <cosurgi> molecular: I remember he has done soimething like that.
963 2011-05-05 09:48:52 <cosurgi> molecular: no problem with that ;)
964 2011-05-05 09:49:07 <gjs278> if the app is remote, you have to use something like ssh X forwarding to see the gui for it
965 2011-05-05 09:49:46 <sipa> gjs278: but then it will also change the values of the X server you're connecting to
966 2011-05-05 09:49:48 <gjs278> otherwise you're just going to get the text feedback of adl
967 2011-05-05 09:49:57 <sipa> instead of the one on the machine it is running on
968 2011-05-05 09:50:02 <gjs278> yeah
969 2011-05-05 09:50:03 <molecular> gjs278, I am. other apps run fine. problem is: set DISPLAY=:0 to make adl work or set DISPLAY=desktop:0 to make wx gui forward?
970 2011-05-05 09:50:35 <molecular> gjs278, oh wait sorry, you might have a point with the x forwarding...
971 2011-05-05 09:50:46 <gjs278> yeah if you try DISPLAY=desktop:0
972 2011-05-05 09:50:50 <gjs278> it would have to go through X forwarding
973 2011-05-05 09:50:50 <molecular> but doesnt that map to localhost:10.0 or somehting?
974 2011-05-05 09:51:24 <molecular> if I just ssh to a machine and xclock shows on local display, is that "using x forwarding"?
975 2011-05-05 09:51:25 <gjs278> you're trying to read the settings of a remote machine with adl right
976 2011-05-05 09:51:26 <gjs278> not locally
977 2011-05-05 09:51:55 <molecular> gjs278, no, I want to run amdoverdrive on the machine that has the 5970
978 2011-05-05 09:52:03 <gjs278> and that's the remote one right
979 2011-05-05 09:52:07 <molecular> but show the wx gui stuff on other machine
980 2011-05-05 09:52:16 <molecular> remote to me, but local to adl
981 2011-05-05 09:52:18 <gjs278> yeah
982 2011-05-05 09:52:33 <gjs278> you have to ssh -X it or whatever
983 2011-05-05 09:52:35 <molecular> adl works fine when I set DISPLAY=:0
984 2011-05-05 09:52:39 <gjs278> the xclock method should be okay
985 2011-05-05 09:52:41 <molecular> gjs278, I'll try
986 2011-05-05 09:53:03 <molecular> I think it does X-forwarding by default (due to my ssh config)
987 2011-05-05 09:53:18 <gjs278> yeah for me I get
988 2011-05-05 09:53:19 <gjs278> Warning: untrusted X11 forwarding setup failed: xauth key data not generated
989 2011-05-05 09:53:35 <sipa> molecular: it won't work
990 2011-05-05 09:53:48 <sipa> either the GUI appearts and you're changing values for the screen you're seeing the GUI on
991 2011-05-05 09:53:52 sgornick has quit (Ping timeout: 240 seconds)
992 2011-05-05 09:54:05 <sipa> or it will appear on the X of the machine you're trying to edit, and you won't see it
993 2011-05-05 09:54:15 <gjs278> in all seriousness, you may want to just install vnc and do it that way
994 2011-05-05 09:54:29 <sipa> that's the way to go if you really want the gui
995 2011-05-05 09:54:30 <gjs278> because then you can vnc to the remote machine, and pull up the GUI with display variables
996 2011-05-05 09:54:49 <molecular> gjs278, thought about that, yes
997 2011-05-05 09:54:54 <gjs278> you can even use X11vnc
998 2011-05-05 09:54:59 <gjs278> which won't spawn a new session
999 2011-05-05 09:55:02 <molecular> but still: no way to init wx subsystem to a given display using code?
1000 2011-05-05 09:55:20 <gjs278> only forwarding I guess
1001 2011-05-05 09:55:31 <gjs278> I've gotten it to work like one time for just a notepad app
1002 2011-05-05 09:55:34 <gjs278> then I never tried it again
1003 2011-05-05 09:55:41 <molecular> heh
1004 2011-05-05 09:56:26 <molecular> so I can use x11vnc on miner machine and just vncviewer on my desktop?
1005 2011-05-05 09:56:31 <gjs278> yes
1006 2011-05-05 09:56:40 <gjs278> and that will give you the active X session of your miner
1007 2011-05-05 09:56:40 <molecular> ok, will do that, thx
1008 2011-05-05 09:56:45 <molecular> nice
1009 2011-05-05 09:56:51 <gjs278> you may have to specify DISPLAY when initially running x11vnc
1010 2011-05-05 09:57:00 midnightmagic has joined
1011 2011-05-05 09:57:01 <gjs278> you want to do x11vnc -usepw -forever
1012 2011-05-05 09:57:05 <gjs278> -usepw for obvious reasons
1013 2011-05-05 09:57:21 <gjs278> and -forever so that the vnc sessions keeps going instead of dying the first time you kill a client connection
1014 2011-05-05 09:57:43 <molecular> thx, I slightly remember having used it that way couple years back
1015 2011-05-05 09:57:50 <diki> so if i want to grab a whole html page with curl, how do i store that information if i dont know the size of the page?
1016 2011-05-05 09:58:03 <gjs278> I used it today to watch one my 5870's crash during mining on 11.4 drivers
1017 2011-05-05 09:58:10 <gjs278> diki
1018 2011-05-05 09:58:12 <gjs278> what language
1019 2011-05-05 09:58:16 <diki> c++
1020 2011-05-05 09:58:26 <gjs278> I don't think you have to know the size
1021 2011-05-05 09:58:35 <gjs278> you just have to give a variable with enough space
1022 2011-05-05 09:58:43 <diki> how much?
1023 2011-05-05 09:58:43 <gjs278> if that even
1024 2011-05-05 09:58:46 <gjs278> who knows
1025 2011-05-05 09:58:52 <gjs278> pages arent that big
1026 2011-05-05 09:58:54 <molecular> gjs278, I switched from 10.x to 11.2 -> should I expect miner crash?
1027 2011-05-05 09:58:58 <gjs278> nah
1028 2011-05-05 09:59:00 <diki> yes i know
1029 2011-05-05 09:59:02 <gjs278> I only got crash for
1030 2011-05-05 09:59:04 <gjs278> 11.4
1031 2011-05-05 09:59:07 <gjs278> back to 11.3 for me
1032 2011-05-05 09:59:14 <gjs278> because 11.3 is the only version that gave me no cpu usage
1033 2011-05-05 09:59:23 <molecular> does it make any speed difference which driver one uses?
1034 2011-05-05 09:59:27 <gjs278> nah
1035 2011-05-05 09:59:35 <gjs278> the only major player was using opencl 2.1
1036 2011-05-05 09:59:37 <molecular> ok, so if it works: dont fiddle with it ;)
1037 2011-05-05 09:59:41 <gjs278> basically
1038 2011-05-05 10:00:29 <cosurgi> molecular: did you get my reply on /msg ?
1039 2011-05-05 10:00:51 <diki> So what other method is there besides Sleep to make my thread sleep?
1040 2011-05-05 10:01:38 <diki> so it doesnt take cpu time for looping
1041 2011-05-05 10:01:45 <diki> or not much anyway
1042 2011-05-05 10:07:44 <UukGoblin> lulz jimbobway put out the heavy artillery
1043 2011-05-05 10:08:01 <UukGoblin> "he's right cause he works for google"
1044 2011-05-05 10:08:11 <UukGoblin> "and that's what satoshi would've wanted anyway"
1045 2011-05-05 10:08:26 <UukGoblin> gee I really don't want to start a fight with bitcoin community...
1046 2011-05-05 10:09:10 bitcoinbulletin has quit (Ping timeout: 260 seconds)
1047 2011-05-05 10:09:10 _ape has quit (Ping timeout: 260 seconds)
1048 2011-05-05 10:09:10 RBecker has quit (Ping timeout: 260 seconds)
1049 2011-05-05 10:09:39 _ape has joined
1050 2011-05-05 10:10:20 fahadsadah has quit (Ping timeout: 260 seconds)
1051 2011-05-05 10:10:20 npouillard has quit (Ping timeout: 260 seconds)
1052 2011-05-05 10:10:20 bernie has quit (Ping timeout: 260 seconds)
1053 2011-05-05 10:10:20 ntosme2 has quit (Ping timeout: 260 seconds)
1054 2011-05-05 10:11:20 ntosme2 has joined
1055 2011-05-05 10:11:22 _flow_ has quit (Excess Flood)
1056 2011-05-05 10:11:29 RBecker has joined
1057 2011-05-05 10:11:29 BlueMatt has quit (Ping timeout: 260 seconds)
1058 2011-05-05 10:11:30 darkskiez has quit (Ping timeout: 260 seconds)
1059 2011-05-05 10:11:47 npouillard has joined
1060 2011-05-05 10:11:59 darkskiez has joined
1061 2011-05-05 10:12:04 <sipa> UukGoblin: though i generally agree with mike, that argument from jimbobway isn't a very good one
1062 2011-05-05 10:12:17 bernie has joined
1063 2011-05-05 10:12:19 _flow_ has joined
1064 2011-05-05 10:12:45 BlueMatt has joined
1065 2011-05-05 10:12:45 <UukGoblin> sipa, what solution for timestamping would you propose?
1066 2011-05-05 10:13:01 <UukGoblin> sipa, cause a separate blockchain doesn't make much sense for me for timestamping
1067 2011-05-05 10:13:30 <UukGoblin> if you don't want to waste more time on this discussion that's fine, I also have some work to do, actually
1068 2011-05-05 10:14:14 thedrs has quit (Quit: Page closed)
1069 2011-05-05 10:14:56 fahadsadah has joined
1070 2011-05-05 10:15:11 Beffjaker has joined
1071 2011-05-05 10:15:21 bitcoinbulletin has joined
1072 2011-05-05 10:15:46 <Beffjaker> Hello Yall
1073 2011-05-05 10:16:47 <Beffjaker> i have a question about moving my wallet...
1074 2011-05-05 10:18:00 <UukGoblin> 42!
1075 2011-05-05 10:18:16 larsivi has quit (Ping timeout: 240 seconds)
1076 2011-05-05 10:18:52 <diki> ;;bc,calc 12000000
1077 2011-05-05 10:18:53 <gribble> The average time to generate a block at 12000000 Khps, given current difficulty of 109670.13329248 , is 10 hours, 54 minutes, and 12 seconds
1078 2011-05-05 10:20:12 <sipa> Beffjaker: if you don't ask the question, we won't ever know it, and - by extension - you'll probably not get an answer either
1079 2011-05-05 10:21:28 <UukGoblin> 'a reasonable answer'
1080 2011-05-05 10:21:35 <UukGoblin> 42 is an answer ;-]
1081 2011-05-05 10:21:41 <UukGoblin> just a bit of a useless one
1082 2011-05-05 10:21:49 <sipa> an answer? it's *the* answer
1083 2011-05-05 10:21:57 <UukGoblin> too :-)
1084 2011-05-05 10:21:58 <Beffjaker> ok, i'm trying to move my wallet and all adresses to the new user i have created in ubuntu, seems permisions wont let me change ownership of file
1085 2011-05-05 10:22:13 <sipa> Beffjaker: you need root to change ownership
1086 2011-05-05 10:22:24 <sipa> but copying + deleting should work fine, actually
1087 2011-05-05 10:22:30 <diki> sudo
1088 2011-05-05 10:22:36 <Beffjaker> ok simple enough
1089 2011-05-05 10:22:52 <Beffjaker> thanks
1090 2011-05-05 10:24:04 <Beffjaker> sudo....make sndwich
1091 2011-05-05 10:26:01 <UukGoblin> anyways, if we can send out 0 outputs, that means we could easily implements adding a short message along with a bitcoin transaction
1092 2011-05-05 10:26:08 <UukGoblin> that's a feature that bitcoin users want
1093 2011-05-05 10:27:16 <jargon> I found an additional solution to the "42" marble puzzle, but was too lazy to post it, and forgot what it was.
1094 2011-05-05 10:27:46 <UukGoblin> jargon, hahahahah
1095 2011-05-05 10:27:54 <UukGoblin> jargon, I like that :-)
1096 2011-05-05 10:27:58 eao has quit (Remote host closed the connection)
1097 2011-05-05 10:28:31 <jargon> Think it had to do with anti-stealth technology or something.
1098 2011-05-05 10:29:01 <UukGoblin> jargon, could you call up an airstrike for me please?
1099 2011-05-05 10:29:05 <jargon> Some Pythagorian GOD math or something.
1100 2011-05-05 10:29:18 <jargon> Where do you want it?
1101 2011-05-05 10:30:56 draaglom has joined
1102 2011-05-05 10:31:05 larsivi has joined
1103 2011-05-05 10:31:09 <jargon> Once a cruise missile barely missed my house by 800ft of altitude, and struck higher up on the mountain.
1104 2011-05-05 10:31:13 <UukGoblin> jargon, head office of National Rail East Anglia?
1105 2011-05-05 10:31:22 <jargon> Fuck nope.
1106 2011-05-05 10:31:35 <UukGoblin> damn.
1107 2011-05-05 10:32:33 <jargon> All I know, is Osama was a puss-ass-bitch if it only took two 9mm rounds to take him out,
1108 2011-05-05 10:33:12 <UukGoblin> I'm /sure/ you know more than that.
1109 2011-05-05 10:33:15 <jargon> I've face fucked concrete and did just dandy.
1110 2011-05-05 10:33:20 <DD-> Ohw ffs
1111 2011-05-05 10:33:32 <DD-> stay of ur drugs jargon and get ur ass back to TTO
1112 2011-05-05 10:33:46 <DD-> on a side note , lunch time!
1113 2011-05-05 10:33:55 <noagendamarket> http://www.bitcoin.org/smf/index.php?topic=7317.msg107297#msg107297 anyone want to help with this ?
1114 2011-05-05 10:34:43 <jargon> I already let the cat out of the bag that bitcoin is just Fawkes@Home
1115 2011-05-05 10:36:19 <jargon> ..plus it crashes at 99999 coins anyhoo.
1116 2011-05-05 10:36:34 <UukGoblin> noagendamarket, so in other words, you advise to BUY BUY BUY?
1117 2011-05-05 10:37:15 <jargon> The bottleneck in the program is the shitass interpreter and Base58 shit
1118 2011-05-05 10:37:49 <jargon> Also they are using vector instead of full-out Teller memory model.
1119 2011-05-05 10:38:00 <jargon> The client is a piece of crap.
1120 2011-05-05 10:38:24 <jargon> They don't even do integrity checks of db.h procedures
1121 2011-05-05 10:38:30 <UukGoblin> jargon, careful there ;-P
1122 2011-05-05 10:38:38 <jargon> Careful how?
1123 2011-05-05 10:38:53 <UukGoblin> I would advise against calling the client piece of crap
1124 2011-05-05 10:39:06 antivigilante_ is now known as antivigilante
1125 2011-05-05 10:39:20 <jargon> What are they going to do? Take my kid away? Ruin my 950 credit score? erase my computer? post vids of me jerkin' it?
1126 2011-05-05 10:39:22 genjix has joined
1127 2011-05-05 10:39:27 genjix has quit (Changing host)
1128 2011-05-05 10:39:27 genjix has joined
1129 2011-05-05 10:39:50 <UukGoblin> kick you out of the #?
1130 2011-05-05 10:39:56 <jargon> create bogus warrants for my arrest?
1131 2011-05-05 10:40:33 <jargon> synthesize my voice over an emergency number with ramblings about 24/7 multi-blocks?
1132 2011-05-05 10:40:54 <BlueMatt> jargon: there is a difference between joining a chan and saying random stuff and joining a chan and accusing bitcoin of having shit security/a shitty client without any real evidence
1133 2011-05-05 10:40:59 <BlueMatt> now please stfu
1134 2011-05-05 10:41:14 <jargon> I've already mouthed off on half the country's police band, east coast inward
1135 2011-05-05 10:41:56 <jargon> Took 50 dispatchers before one realized "NSA"
1136 2011-05-05 10:42:59 <jargon> BlueMatt, I was doing an audit of their code.
1137 2011-05-05 10:43:24 <jargon> and it *is* a wasteful code in how fucking slow the program does pretty much any low-level task.
1138 2011-05-05 10:43:49 <BlueMatt> patches welcome ;)
1139 2011-05-05 10:43:49 <noagendamarket> its not itunes
1140 2011-05-05 10:43:55 <noagendamarket> :)-
1141 2011-05-05 10:44:14 <BlueMatt> (or atleast hard evidence/direct references to places where the client could be made better)
1142 2011-05-05 10:44:22 <UukGoblin> heh, a moral dillema: he IS sometimes fun, but then again, he IS talking back
1143 2011-05-05 10:44:30 <sipa> BlueMatt: i think you should reply here: http://www.bitcoin.org/smf/index.php?topic=7313.0
1144 2011-05-05 10:44:41 <jargon> replace the Base58 in bignum.h with fucking Keal Demux
1145 2011-05-05 10:44:52 <jargon> You'll get shitload faster performance
1146 2011-05-05 10:44:55 <BlueMatt> hey when did we get reputation vote buttons on smf?
1147 2011-05-05 10:45:27 danbri has quit (Remote host closed the connection)
1148 2011-05-05 10:45:27 <BlueMatt> UukGoblin: yea the only reason I have ops is because last time jargon was in here, he should have been kicked but nanotube wasn't around...
1149 2011-05-05 10:45:41 <UukGoblin> BlueMatt, mhm, I used to have @ here too
1150 2011-05-05 10:46:02 <BlueMatt> ask nanotube to make gribble give you ops when you join
1151 2011-05-05 10:46:17 <UukGoblin> I'd give him one last chance and then kick him if he keeps spouting shit at the client or the devs
1152 2011-05-05 10:46:28 <jargon> has anyone seen n0ah around?
1153 2011-05-05 10:46:32 <UukGoblin> nanotube, could I have gribble give me ops? :-)
1154 2011-05-05 10:46:49 <jargon> oh shit, this *is* #bitcoin-dev
1155 2011-05-05 10:47:03 <BlueMatt> UukGoblin: yep, I agree
1156 2011-05-05 10:47:03 <jargon> thought I was in a different Anon chan
1157 2011-05-05 10:47:50 <jargon> I got 3 different Anonymous mingling IRC channels open on 3 different networks.
1158 2011-05-05 10:48:42 <BlueMatt> sipa: thanks, replying
1159 2011-05-05 10:48:51 danbri has joined
1160 2011-05-05 10:50:26 <UukGoblin> hmm has the bitcoin client officially moved to git, or is the svn repo still being used?
1161 2011-05-05 10:50:46 <BlueMatt> svn is either mirrored or just mirrored on releases (I dont remember)
1162 2011-05-05 10:50:49 <BlueMatt> but officially its git
1163 2011-05-05 10:50:57 <UukGoblin> kcool
1164 2011-05-05 10:51:58 genjix has quit (Quit: leaving)
1165 2011-05-05 10:52:04 genjix has joined
1166 2011-05-05 10:52:05 genjix has quit (Changing host)
1167 2011-05-05 10:52:05 genjix has joined
1168 2011-05-05 10:52:45 <UukGoblin> hey, this Script looks powerful
1169 2011-05-05 10:53:52 <BlueMatt> what script?
1170 2011-05-05 10:54:06 <UukGoblin> script.cpp:56
1171 2011-05-05 10:54:36 <UukGoblin> actually probably just script.cpp in general
1172 2011-05-05 10:55:16 <BlueMatt> yea scripting is really powerful, shame we have IsStandard, but we do kinda need it
1173 2011-05-05 10:55:37 <wumpus> you can play with scripting but only on the testnet
1174 2011-05-05 10:55:39 <UukGoblin> I wonder if this Script could be used for announcing of miner's transaction fee policies...
1175 2011-05-05 10:55:58 <UukGoblin> wumpus, thank you for permission ;-)
1176 2011-05-05 10:56:09 <BlueMatt> you could do something then just OP_DROP
1177 2011-05-05 10:56:13 <wumpus> that's not permission, it's simply hw the client is wired
1178 2011-05-05 10:56:14 <BlueMatt> or you could but it in coinbase
1179 2011-05-05 10:56:43 <BlueMatt> standard clients (miners) will reject any non-standard tx (except on testnet)
1180 2011-05-05 10:56:55 <UukGoblin> I realize that, yeah
1181 2011-05-05 10:57:07 <BlueMatt> ok, just making sure ;)
1182 2011-05-05 10:57:30 <UukGoblin> I'm trying not to comment on whether that's good or bad ;-P
1183 2011-05-05 10:57:42 <UukGoblin> cause I spend half of yesterday arguing with TD about it ;-]
1184 2011-05-05 10:57:58 <BlueMatt> meh, its terrible to let scripting go to waste but necessary
1185 2011-05-05 10:58:13 <wumpus> yes it's too unpredictable
1186 2011-05-05 10:58:16 <BlueMatt> (but I dont really want to discuss this atm either)
1187 2011-05-05 11:00:08 <UukGoblin> I think I'm gonna make a post later... this could become dangerous... if TD and others are sooo much against "polluting" of the blockchain, and if they actually make some code to prevent the 0.00 output hack (as described by theymos), they'll leave others no choice but to burn bitcoins... and burning bitcoins is the worst possible solution for everyone
1188 2011-05-05 11:00:28 x5x is now known as x5x`brb
1189 2011-05-05 11:00:30 <ersi> let's use jargon as a scripting engine
1190 2011-05-05 11:02:04 <sipa> UukGoblin: burning coins? that isn't necessary at all
1191 2011-05-05 11:02:10 <BlueMatt> I have a feeling that would be even more random than other scripting
1192 2011-05-05 11:02:28 <wumpus> it would be more like bitcasino than bitcoin
1193 2011-05-05 11:02:31 <sipa> [mike] used it in his initial explanation as an example
1194 2011-05-05 11:02:36 <UukGoblin> is there a description of this whole scripting thing? I'm trying to understand how all these OP_DROP and OP_CHECKSIG and stuff work
1195 2011-05-05 11:02:54 <BlueMatt> wiki scripting page is decent, otherwise its script.cpp
1196 2011-05-05 11:03:13 <sipa> by the way, i am a bit surprised 0 outputs are allowed
1197 2011-05-05 11:03:29 <sipa> which probably only means they haven't beed exploited yet
1198 2011-05-05 11:03:31 <UukGoblin> sipa, what I'm saying is this: IF you block the 0-output hack, THEN the only way to achieve timestamping by the main blockchain might be by burning coins
1199 2011-05-05 11:04:02 <sipa> and using a separate chain which cross-mining as described by TD?
1200 2011-05-05 11:04:14 <UukGoblin> BlueMatt, thanks
1201 2011-05-05 11:04:32 <jgarzik> UukGoblin: I don't think burning bitcoins is "worst possible solution"
1202 2011-05-05 11:04:44 <jgarzik> it clearly provides an incentive not to spam the main chain
1203 2011-05-05 11:04:48 <wumpus> can't you just send the coins somewhere instead of burning them?
1204 2011-05-05 11:04:55 <sipa> wumpus: yes you can
1205 2011-05-05 11:05:08 <UukGoblin> sipa, awesome a separate chain might be, but some people will just not like it. I'm trying not to take a side here now, but some people will want to use the main blockchain instead. And there'll be nothing to stop them.
1206 2011-05-05 11:05:31 <UukGoblin> wumpus, where would you send them though?
1207 2011-05-05 11:05:38 <wumpus> well even if the miner got them it'd be better than making them disappear
1208 2011-05-05 11:05:57 <UukGoblin> wumpus, yes, and all I'm advocating is for it to be possible for the miner to get them
1209 2011-05-05 11:06:14 <sipa> and jgarzik: i agree with you on that, but i don't feel confortable with the idea that there is a useful reason for burning coins
1210 2011-05-05 11:06:15 <wumpus> otherwise you could send them to one of the developers, I don't care :)
1211 2011-05-05 11:06:24 <jargon> should I upload the replacement Base54 procedure?
1212 2011-05-05 11:06:29 <jargon> 58*
1213 2011-05-05 11:06:33 <sipa> UukGoblin: and as suggested by mike, i think you can just send them to an address of your own
1214 2011-05-05 11:06:37 <UukGoblin> but if it's not possible to send out a 0 output together with a fee, then the only way to put a hash in a transaction will be a 0.01-type of output to a non-existing address that happens to be your hash
1215 2011-05-05 11:06:39 <wumpus> UukGoblin: ok agreed in that case
1216 2011-05-05 11:06:52 <BlueMatt> jargon: if you have a patch, hard advice, absolutely
1217 2011-05-05 11:07:07 <jargon> only problem is if I post the patch, everyone is going to fucking make coins like a mad fuck until it hits 99999 coins and crashes
1218 2011-05-05 11:07:16 <BlueMatt> ...?
1219 2011-05-05 11:07:29 <ersi> BlueMatt: He's a supermega troll...
1220 2011-05-05 11:07:37 <ersi> Only spouts bullshit
1221 2011-05-05 11:07:45 * jargon posts the patch then jackass
1222 2011-05-05 11:07:45 <ersi> been at it for days/weeks
1223 2011-05-05 11:07:47 <wumpus> just fork the client under the name 'bitjackpot'
1224 2011-05-05 11:07:48 <UukGoblin> ersi, I think BlueMatt's dillema is whether to kick him out or not ;-]
1225 2011-05-05 11:07:49 <BlueMatt> ersi: yea, thought so...just hoped he had actually changed his mind
1226 2011-05-05 11:08:05 <ersi> I'll bet 50BTC he'll never
1227 2011-05-05 11:08:21 <BlueMatt> Ill take the bet (and it totally wont influence my decision)
1228 2011-05-05 11:08:38 <jargon> oh right. base58.h, not bignum.h
1229 2011-05-05 11:08:44 * jargon patches and posts
1230 2011-05-05 11:09:08 <jgarzik> sipa: well, the idea is that nobody wants to burn coins. Same incentive as failing a clearcoin escrow transaction.
1231 2011-05-05 11:09:49 <sipa> jgarzik: another reason to dislike it: burning a coin is a transaction that will remain in the block chain forever, and will never ever be able to be pruned
1232 2011-05-05 11:09:50 <ersi> jargon: take it easy when you write that patch, we don't want a second skynet or anything
1233 2011-05-05 11:09:53 <BlueMatt> jgarzik: sipa UukGoblin what, exactly, is the solution to what problem?
1234 2011-05-05 11:10:38 <sipa> BlueMatt: UukGoblin wants to use the bitcoin block chain as a general timestamping service, and therefore needs arbitrary date to be placed in it
1235 2011-05-05 11:11:01 <BlueMatt> wait...what?
1236 2011-05-05 11:11:13 <UukGoblin> arbitrary data, yes
1237 2011-05-05 11:11:21 <jgarzik> BlueMatt: same old spamming-the-block-chain discussion
1238 2011-05-05 11:11:26 <UukGoblin> (not that I want /want/, just talking about a possibility)
1239 2011-05-05 11:11:27 <wumpus> hm that could be bad, arbitrary data
1240 2011-05-05 11:11:35 <BlueMatt> an just the old put arbitrary data in chain/txes
1241 2011-05-05 11:12:15 <UukGoblin> what I'm arguing (again) is simply this: it will always be possible to put arbitrary data into the main blockchain.
1242 2011-05-05 11:12:20 <wumpus> at least if too many people start using it
1243 2011-05-05 11:12:23 <sipa> yes it will
1244 2011-05-05 11:12:24 <BlueMatt> I dont like the idea, but I would love to see a _much_ more generous IsStandard
1245 2011-05-05 11:12:29 <sipa> and it is not ment for that
1246 2011-05-05 11:12:45 <jargon> fucking Hell, my eyes can't even focus on the fucking github thin as shit font.
1247 2011-05-05 11:12:46 <UukGoblin> sipa, regardless of what it is meant for and what it's not - it will always be possible.
1248 2011-05-05 11:13:08 <sipa> so nobody can stop you doing so, but measures may be be agreed upon by those who do not want it in the block chain
1249 2011-05-05 11:13:11 <BlueMatt> but currently it is odd/an odd hack/costs money
1250 2011-05-05 11:13:12 <jargon> I got court in 6 hours, If I don't return I am most likely in prison.
1251 2011-05-05 11:13:15 <wumpus> if you can put arbitrary data in the blockchain, it'd be a file sharing service in no time :P
1252 2011-05-05 11:13:26 <UukGoblin> sipa, but it's impossible to set up measures preventing that.
1253 2011-05-05 11:13:49 <ersi> jargon: Alright. We'll send the rescue team if you don't show up in 6½h
1254 2011-05-05 11:13:51 <UukGoblin> BlueMatt, my idea of timestamping is that it should cost money
1255 2011-05-05 11:14:10 <BlueMatt> where/why does timestamping come into this?
1256 2011-05-05 11:14:21 <UukGoblin> now, because it's impossible to prevent it, you might as well do it the right way, rather than rely on hacks, is my thinking.
1257 2011-05-05 11:14:37 <jargon> ersi, I already had my gf send a nuclear bomber from the airbase to buzz and 90 the prison yard on a dime with bomb doors open revealing two ten megaton warheads
1258 2011-05-05 11:14:38 <UukGoblin> BlueMatt, timestamping of arbitrary data by the main blockchain
1259 2011-05-05 11:14:47 <ersi> jargon: Yes, I know.
1260 2011-05-05 11:14:50 <DD-> hahaha
1261 2011-05-05 11:14:55 <DD-> oh thats the best one so far
1262 2011-05-05 11:15:09 <jargon> That is how I got out last time.
1263 2011-05-05 11:15:09 <UukGoblin> BlueMatt, you make some data, you make a hash out of it, and then make a bitcoin transaction with this hash. This proves the hash existed before the next block was generated, ergo, timestamps your data.
1264 2011-05-05 11:15:40 <BlueMatt> I disagree, make them work/hack around/pay to put their data in the chain
1265 2011-05-05 11:15:51 <ersi> jargon: Yeah, I heard about it the first time. But we'll send a seperate team to rescue you. So you'll have a even larger probability of being free'd
1266 2011-05-05 11:15:54 <wumpus> if you can put arbitrary data in the block chain, you could even put your wallet in it (encrypted of course)
1267 2011-05-05 11:16:05 <UukGoblin> BlueMatt, yes, your bitcoin transaction should involve a payment of some sort.
1268 2011-05-05 11:16:13 <sipa> and the block chain is not a large shared filesystem
1269 2011-05-05 11:16:30 <sipa> it's indeed impossible to prevent people from using it as such
1270 2011-05-05 11:16:33 <UukGoblin> sipa, it is, however a limited one.
1271 2011-05-05 11:17:04 <wumpus> hopefully this won't be bitcoin's doom
1272 2011-05-05 11:17:04 <UukGoblin> we should just make it unfeasibly expensive
1273 2011-05-05 11:17:24 <UukGoblin> rather than block it off entirely
1274 2011-05-05 11:17:33 <sipa> it is one that people agree to maintain (store, transfer, verify... which will eventually cost a massive amount of money, not only for miners), because it enables them to do something that is valuable to them: send money around the world without hassle[, other advantages of bitcoin]
1275 2011-05-05 11:17:52 <wumpus> but if you make it too expensive, won't they start burning coins again?
1276 2011-05-05 11:18:07 <UukGoblin> wumpus, if it's cheaper to burn a coin than to pay a storage fee then yes
1277 2011-05-05 11:18:15 <sipa> but as soon as it gets used on a large scale for another purpose, those people may disagree about keeping that network running, lose interest, and undermine bitcoin's value as a whole
1278 2011-05-05 11:18:16 <jargon> bitcoins aren't currency, they are cpu cycle claims tickets
1279 2011-05-05 11:18:30 <wumpus> almost a classic tragedy of the commons
1280 2011-05-05 11:18:45 <BlueMatt> wow first mostly right comment jargon made
1281 2011-05-05 11:19:17 <ersi> Boggles the mind
1282 2011-05-05 11:19:17 <jargon> when you pay someone, you are granting them low-level access of an interpreter on your machine.
1283 2011-05-05 11:19:31 <UukGoblin> sipa, nothing to stop it from that. You could embed a child porn transaction and it could get accepted in the main blockchain and no-one could do anything about it. That would indeed be very bad for bitcoin's reputation.
1284 2011-05-05 11:19:37 <BlueMatt> and now were back
1285 2011-05-05 11:19:40 <ersi> jargon: sparcdalb?
1286 2011-05-05 11:19:44 <UukGoblin> BlueMatt, yeah, surprising.
1287 2011-05-05 11:19:51 <jargon> the more you pay coins, the slower you can mine them.
1288 2011-05-05 11:19:51 <sipa> UukGoblin: my point exactly
1289 2011-05-05 11:19:57 <jgarzik> And all this is old news :)
1290 2011-05-05 11:19:59 <ersi> jargon: rackdalb?
1291 2011-05-05 11:20:05 <jgarzik> Nobody is claiming data spam can be prevented
1292 2011-05-05 11:20:23 <jgarzik> but that does not mean we want to encourage data spam
1293 2011-05-05 11:20:35 * BlueMatt agrees with jgarzik, nothing we can really do about it, its already expensive and unless you have a suggestion to make it more so...
1294 2011-05-05 11:20:35 <jargon> I wrote the original bitcoin called Vapor, remember?
1295 2011-05-05 11:20:48 <UukGoblin> jgarzik, well I'm claiming I'd rather encourage it and charge for it, than pointlessly try to ask people not to do it
1296 2011-05-05 11:20:52 <jargon> Vapor was like Steam combined with bitcoin in the same process tree.
1297 2011-05-05 11:21:00 <sipa> you can charge for it in a separate chain
1298 2011-05-05 11:21:07 <ersi> jargon: But like a Teller majingy?
1299 2011-05-05 11:21:14 <BlueMatt> yea Vapor, it was announced the just became vaporware (it is pun thursday, after all)
1300 2011-05-05 11:21:33 <jargon> Vapor really was vaporware
1301 2011-05-05 11:21:35 <UukGoblin> sipa, how?
1302 2011-05-05 11:21:36 <jargon> literally
1303 2011-05-05 11:21:54 <jgarzik> UukGoblin: [mike] explained that in his post
1304 2011-05-05 11:21:54 <ersi> jargon: wasn't it jargonware?
1305 2011-05-05 11:22:08 <sipa> UukGoblin: please, read TD's post carefully
1306 2011-05-05 11:22:26 <jargon> none of the games really existed, a computer just hacks the crap out of your computer vi Vapor's engine in real time, allowing you a light show and funky sounds that resemble games like quake iii and doom 3
1307 2011-05-05 11:22:31 <UukGoblin> he did not explain how to charge for something alternative with bitcoins
1308 2011-05-05 11:22:38 <jgarzik> yes, he did
1309 2011-05-05 11:22:55 <sipa> Paying for alternative resources with BitCoins
1310 2011-05-05 11:22:59 <sipa> is a bold title in his post
1311 2011-05-05 11:23:08 <UukGoblin> and if he did, and it's what I think you mean it is, then it's equivalent in terms of storage and bandwidth to my solution
1312 2011-05-05 11:23:16 <BlueMatt> TD is ridiculously smart, when he says something, be sure to read it carefully ;)
1313 2011-05-05 11:23:25 RazielZ has joined
1314 2011-05-05 11:23:50 peck has quit (Ping timeout: 260 seconds)
1315 2011-05-05 11:24:13 <jargon> bitcoin has turned into Fawkes@Home
1316 2011-05-05 11:24:31 <UukGoblin> sipa, jgarzik: ok yeah he changed that, I noticed. The new version talks about transferring coins from your account to your account. It is more damaging to the blockchain than just sending out a non-standard timestamping transaction.
1317 2011-05-05 11:24:32 <ersi> Damn, you got us jargon
1318 2011-05-05 11:25:13 <UukGoblin> so this solution is a pointless argument, really.
1319 2011-05-05 11:25:40 <jargon> I have no idea what I did with my cigarette, for all I know the condo is going to burn the fuck down.
1320 2011-05-05 11:26:12 <UukGoblin> (not to mention it's not actually paying anyone, but that's irrelevant)
1321 2011-05-05 11:27:04 <UukGoblin> ok it's lunchtime for me
1322 2011-05-05 11:27:33 <sipa> compared with what kind of non-standard timestamping transaction?
1323 2011-05-05 11:27:59 <sipa> the send-0-coins-to-inexisting-hash-of-data-claimed-to-be-an-address?
1324 2011-05-05 11:28:06 <jgarzik> overall, I'm unconcerned -- the status quo seems to be working just fine
1325 2011-05-05 11:28:31 <jgarzik> and plenty in the community seem interested in keeping bitcoin a currency, rather than a data posting service
1326 2011-05-05 11:28:40 <jargon> send NULL coins, really, see what happens
1327 2011-05-05 11:29:29 <sipa> UukGoblin: for that to work, the 0-output can never be pruned, ultimately weighing more heavy on the network's resources than normal transactions
1328 2011-05-05 11:29:29 <jargon> didn't I already post that finding in my audit 3 days ago?
1329 2011-05-05 11:30:33 <jargon> better yet, use a zero length hash
1330 2011-05-05 11:31:12 * noagendamarket doesnt want to turn bitcoin into freenet
1331 2011-05-05 11:31:13 jroot has joined
1332 2011-05-05 11:31:36 jargon has joined
1333 2011-05-05 11:31:38 <jargon> more like freeforallnet
1334 2011-05-05 11:31:47 * sipa wonders why there were still people not ignoring that guy
1335 2011-05-05 11:31:54 * wumpus neither
1336 2011-05-05 11:32:21 <ersi> Don't worry, he'll find a nuclear plant and free it of it's energy to mine derps into the block chain
1337 2011-05-05 11:32:29 peck has joined
1338 2011-05-05 11:32:37 <ersi> so that he can get 99999 BTC and crash the world
1339 2011-05-05 11:32:41 <wumpus> hehe
1340 2011-05-05 11:32:45 <wumpus> good luck with that
1341 2011-05-05 11:34:00 <BlueMatt> jgarzik: can you kickban jargon from #bitcoin as well?
1342 2011-05-05 11:34:45 <genjix> g++ $1 -lboost_system -lboost_thread -Wall -Wextra -std=c++0x -pedantic
1343 2011-05-05 11:34:49 <genjix> /usr/bin/ld: cannot find -lgcc_s
1344 2011-05-05 11:34:49 <jgarzik> BlueMatt: I'm not much for kicking and banning
1345 2011-05-05 11:34:49 <noagendamarket> lol @ derpchain
1346 2011-05-05 11:34:53 <genjix> collect2: ld returned 1 exit status
1347 2011-05-05 11:34:58 <genjix> natty narwhal - any ideas?
1348 2011-05-05 11:35:06 <jgarzik> BlueMatt: I put jargon on /ignore
1349 2011-05-05 11:35:14 <sipa> same here
1350 2011-05-05 11:35:43 <sipa> but i think a kickban was justified here
1351 2011-05-05 11:35:46 <ersi> well, he's quite destructive in that he derails people who do not know he's just full of bullshit
1352 2011-05-05 11:36:05 <BlueMatt> especially when #bitcoin is designed for noobs and the like
1353 2011-05-05 11:36:14 <ersi> ye
1354 2011-05-05 11:36:34 <ersi> I'll admit to buying some of what he said in the beginning
1355 2011-05-05 11:37:07 <wumpus> yeah before he started with really crackpot things I also took him a bit seriusly.. he's a distraction at least, it's good he was kicked out of the dev channel
1356 2011-05-05 11:39:11 <wumpus> he should start #bitcoin-clowns
1357 2011-05-05 11:44:05 cdeckerAnd has joined
1358 2011-05-05 11:45:02 <DD-> omg u banned jargon :(
1359 2011-05-05 11:45:23 <DD-> back to work i guess
1360 2011-05-05 11:45:34 <BlueMatt> sorry ;)
1361 2011-05-05 11:45:42 BERRI has quit (Read error: Operation timed out)
1362 2011-05-05 11:46:06 <UukGoblin> sipa, yes, this kind of non-standard tx. How can a transaction from your account to another account of yours ever be pruned from the network?
1363 2011-05-05 11:46:18 <jgarzik> you can always try and convince nanotube. I really don't like banning people, and tend to dislike IRC pissing matches in general :) But I do make it a point to give nanotube/gribble ops
1364 2011-05-05 11:46:31 <sipa> UukGoblin: by spending it...
1365 2011-05-05 11:46:42 <sipa> as soon as you don't need the resource anymore
1366 2011-05-05 11:46:50 <UukGoblin> sipa, but if you want to keep its meaning, you won't spend it
1367 2011-05-05 11:47:02 <sipa> then i don't mind it staying in the block chain either
1368 2011-05-05 11:47:03 <BlueMatt> jgarzik: I agree, I prefer to never kick/ban anyone, but there is a point where its just kinda ridiculous
1369 2011-05-05 11:47:26 <BlueMatt> (for that matter, I never really put anyone on /ignore either)
1370 2011-05-05 11:47:34 Zarutian has joined
1371 2011-05-05 11:47:43 <sipa> UukGoblin: i just don't like data that will without any way of recovery need to be stored forever
1372 2011-05-05 11:47:46 <UukGoblin> sipa, then what's the point of the discussion? timestamping is just a bunch of such transactions
1373 2011-05-05 11:47:52 <ersi> Quite pointless to have @ if one ignores whatever needs moderation
1374 2011-05-05 11:47:53 <ersi> :p
1375 2011-05-05 11:48:31 <BlueMatt> jgarzik: speaking of which, why does gribble/nanotube not have @ on #bitcoin from ChanServ?
1376 2011-05-05 11:49:09 <UukGoblin> are 0-output transactions being kept forever? also, i was kinda under the impression that ALL transactions are being kept forever
1377 2011-05-05 11:49:22 <sipa> currently, all transactions are kept forever, yes
1378 2011-05-05 11:49:32 <sipa> but that's just an implementation issue
1379 2011-05-05 11:49:48 <UukGoblin> so what's the problem?
1380 2011-05-05 11:49:50 <sipa> transaction outputs that are 1) spent and 2) buried in the block chain can be pruned
1381 2011-05-05 11:50:06 <UukGoblin> i thought the txs are held in the blocks
1382 2011-05-05 11:50:16 <sipa> read the whitepaper
1383 2011-05-05 11:50:22 <UukGoblin> client may choose to store a subset of blocks/txs if he so desires
1384 2011-05-05 11:50:33 <sipa> no, that's something different
1385 2011-05-05 11:50:48 <sipa> A) not all clients need to store the full blockchain, but miners always will have to
1386 2011-05-05 11:51:09 <sipa> B) spent and buried transaction outputs *CAN* be pruned by everyone, by design
1387 2011-05-05 11:51:38 <UukGoblin> B) but not blocks in which these transactions were stored in, right?
1388 2011-05-05 11:51:49 <sipa> block headers will always need to be kept
1389 2011-05-05 11:52:00 <sipa> but those are 80 bytes per 10 minutes
1390 2011-05-05 11:52:04 <sipa> that's all
1391 2011-05-05 11:52:22 <cdeckerAnd> I guess it would be desirable for lightweight clients to track only the inputs he owns right?
1392 2011-05-05 11:52:27 <sipa> cdeckerAnd: indeed
1393 2011-05-05 11:52:35 <UukGoblin> doesn't the block's hash include all the txs in it?
1394 2011-05-05 11:52:41 <sipa> UukGoblin: yes
1395 2011-05-05 11:52:47 marlowe has quit (Read error: Operation timed out)
1396 2011-05-05 11:52:48 <sipa> that's why we have the merkle tree scheme
1397 2011-05-05 11:53:02 <sipa> so transactions can be pruned from blocks
1398 2011-05-05 11:53:18 <sipa> cdeckerAnd: that's the so called lightweight or client mode
1399 2011-05-05 11:53:25 <UukGoblin> sipa, so... if you discard all block's data except for headers, you won't be able to verify its validity anymore
1400 2011-05-05 11:53:34 <sipa> UukGoblin: please, read satoshi's paper
1401 2011-05-05 11:53:41 <cdeckerAnd> Full/pruned block history is only needed for transaction verification, right?
1402 2011-05-05 11:54:18 <UukGoblin> sipa, gah. i did a while ago
1403 2011-05-05 11:54:30 <sipa> UukGoblin: this is the entire reason why bitcoin works per transaction output instead of balance per address
1404 2011-05-05 11:54:36 <sipa> exactly so that transaction *CAN* be pruned
1405 2011-05-05 11:54:48 <sipa> and no permanent storage is required except for block headers
1406 2011-05-05 11:54:56 <CIA-30> bitcoin: genjix exper * rcd84cefa1ad2 intersango/DATABASE: added account column to bank_statement for when you have multiple bank accounts. http://tinyurl.com/6gatogl
1407 2011-05-05 11:54:58 <CIA-30> bitcoin: genjix exper * r4358b880ea6d intersango/util.php: Merge branch 'exper' of gitorious.org:intersango/master into exper http://tinyurl.com/67c4l8x
1408 2011-05-05 11:56:10 <sipa> and you will always be able to verify the validity of block headers, as they only depend on the previous block header hash (which is known) and the merkle root (which is known inside the block header)
1409 2011-05-05 11:56:14 <UukGoblin> sipa, how can you verify a block's validity if you don't have its transactions?
1410 2011-05-05 11:56:27 <cdeckerAnd> Do we actually need the headers? Couldn't we just keep the latest developments block headers and prune blocks without unspent outputs?
1411 2011-05-05 11:56:35 <UukGoblin> ok...
1412 2011-05-05 11:56:39 <UukGoblin> lag
1413 2011-05-05 11:57:08 <sipa> and once you know you won't need to verify the validity of transactions themselves anymore (since they are verifiably spent), you can drop those
1414 2011-05-05 11:57:11 <cdeckerAnd> Ah I hate autocorrection on my phone
1415 2011-05-05 11:57:26 <UukGoblin> ok, that rings a bell now
1416 2011-05-05 11:57:56 <sipa> and for those 0-outputs, either they will be pruned immediately, in which case they become useless for storing data
1417 2011-05-05 11:58:10 <sipa> or they will never be pruned, in which case it is a permanent storage cost to the network
1418 2011-05-05 11:58:41 <UukGoblin> i see what you're saying
1419 2011-05-05 11:58:56 <jgarzik> sipa: that's an issue with bitcoin anyway. some amount of bitcoins will always be burned regardless of any practices or standard usages
1420 2011-05-05 11:59:03 <UukGoblin> i need to dig deeper into the merkle tree then
1421 2011-05-05 11:59:03 <sipa> UukGoblin: thanks for listening, in that case :)
1422 2011-05-05 11:59:24 <UukGoblin> my argument still holds though ;)
1423 2011-05-05 11:59:36 <jgarzik> sipa: the network will slowly degrade over time
1424 2011-05-05 11:59:49 <UukGoblin> i'll explain the details in my post
1425 2011-05-05 11:59:59 <sipa> jgarzik: unfortunately, yes, and coins will always be split to smaller and smaller amounts too
1426 2011-05-05 12:00:15 <sipa> but storage capacities will keep growing too
1427 2011-05-05 12:00:30 <cdeckerAnd> God this lecture is boring
1428 2011-05-05 12:01:06 agricocb has quit (Quit: Leaving.)
1429 2011-05-05 12:01:24 <cdeckerAnd> We could force users to send their oldest inputs to themselves
1430 2011-05-05 12:02:08 <cdeckerAnd> So after a grace period of 2 - 3 years we can prune also unspent inputs
1431 2011-05-05 12:02:47 <sipa> i don't like that idea
1432 2011-05-05 12:02:51 <tcatm> wouldn't that create new unspent outputs and lots of traffic?
1433 2011-05-05 12:03:11 <cdeckerAnd> It's what I will be doing on my Android implementation to keep the input fragmentation low
1434 2011-05-05 12:03:20 <wumpus> i don't like that idea either, so if you forget to send the coins to yourself, you lose them?
1435 2011-05-05 12:03:46 <cdeckerAnd> Yeah but we consolidate multiple inputs into one output
1436 2011-05-05 12:04:06 <sipa> that's true
1437 2011-05-05 12:04:18 <cdeckerAnd> It can be part of the client to select oldest inputs first
1438 2011-05-05 12:04:43 <cdeckerAnd> Most people would have high enough turnover
1439 2011-05-05 12:04:55 <tcatm> you could "defragment" inputs during normal transactions just by including more inputs and a bigger change output
1440 2011-05-05 12:05:03 <sipa> i do want to be able to send my savings to a single address
1441 2011-05-05 12:05:09 <sipa> export that address's private key
1442 2011-05-05 12:05:14 <sipa> engrave that in stone
1443 2011-05-05 12:05:20 <sipa> store it in a safe
1444 2011-05-05 12:05:25 <cdeckerAnd> Sounds good
1445 2011-05-05 12:05:30 <sipa> and be sure that after 10 years i can still access its funds
1446 2011-05-05 12:05:44 <wumpus> exactly sipa -- it shouldn't be that when you find an old wallet, its contents are worthless just because the coins have been reused
1447 2011-05-05 12:05:50 <jgarzik> yeah, that's the only solution to unspent TX's -- require coins to be resent every X amount of time, otherwise they are pruned regardless
1448 2011-05-05 12:06:03 <jgarzik> but it breaks the 'savings' case just described
1449 2011-05-05 12:06:07 <cdeckerAnd> You could just use the same input and output adtesses
1450 2011-05-05 12:06:11 <BlueMatt> god P=NP is 40 years old...I feel so young
1451 2011-05-05 12:06:24 <wumpus> bitcoin is supposed to be like coins, not like commodities that rot after a while
1452 2011-05-05 12:06:39 genjix has quit (Quit: leaving)
1453 2011-05-05 12:06:46 <wumpus> though that's interesting in its own right
1454 2011-05-05 12:06:48 genjix has joined
1455 2011-05-05 12:07:36 <cdeckerAnd> Have to get back to my lecture we'll discuss it later again
1456 2011-05-05 12:07:51 <cdeckerAnd> See you ^^
1457 2011-05-05 12:08:13 cdeckerAnd has quit (Quit: Bye)
1458 2011-05-05 12:08:15 <wumpus> later
1459 2011-05-05 12:09:55 larsivi has quit (Remote host closed the connection)
1460 2011-05-05 12:12:48 Speeder has joined
1461 2011-05-05 12:13:28 larsivi has joined
1462 2011-05-05 12:17:20 m00p has quit (Quit: Leaving)
1463 2011-05-05 12:21:18 <jgarzik> in a future where only Big Nodes and Big Companies can afford to fully exist on the P2P network, I wonder if 'coin expiration' will be implemented
1464 2011-05-05 12:22:02 agricocb has joined
1465 2011-05-05 12:22:37 <wumpus> it'd be a way of levying tax, as you probably pay tx costs every time you pay to yourself
1466 2011-05-05 12:26:58 <noagendamarket> wont that just make all the remaining coins worth more ?
1467 2011-05-05 12:33:50 <jgarzik> yes
1468 2011-05-05 12:34:12 <jgarzik> but will that balance the cost of carrying millions of unspent, tiny amounts of bitcoins?
1469 2011-05-05 12:34:40 <jgarzik> those TX's remain in TX indexes forever
1470 2011-05-05 12:36:33 <tcatm> do we have any statistics of how much data that actually is?
1471 2011-05-05 12:37:11 <jgarzik> not right now, we currently keep everything even spent tx's :)
1472 2011-05-05 12:37:19 <jgarzik> and throw away micro-change
1473 2011-05-05 12:38:11 <sipa> well, if every txout is 1 ucBTC, and in a separate tx, we'll need about 2 exabytes of data to store the block chain
1474 2011-05-05 12:38:11 <tcatm> iirc it was 1/3 unspent last time I checked
1475 2011-05-05 12:38:16 <wumpus> but with a deflationary currency, what is now microchange will at a certain moment be a large amount
1476 2011-05-05 12:38:29 <wumpus> so no, you can't just throw away micro-change
1477 2011-05-05 12:40:54 <tcatm> hrm, the sqldump of my blockchain is 1.8 GB
1478 2011-05-05 12:40:55 original_speeder has joined
1479 2011-05-05 12:42:31 <wumpus> if you make bitcoin inflationary you could be more careless with change, it'd be more natural to 'throw away old coins' in that case
1480 2011-05-05 12:42:58 Speeder has quit (Ping timeout: 264 seconds)
1481 2011-05-05 12:42:58 original_speeder is now known as Speeder
1482 2011-05-05 12:43:16 <wumpus> then again, that has a lot of other drawbacks
1483 2011-05-05 12:44:07 <tcatm> I think there was a patch to prevent throwing away change whenever possible...
1484 2011-05-05 12:44:14 larsivi has quit (Ping timeout: 240 seconds)
1485 2011-05-05 12:45:00 <CIA-30> bitcoin: genjix exper * r0f9c354ce4e6 intersango/cron/bankd/match_withdrawals.php: Merge branch 'master' into exper http://tinyurl.com/3f3blw5
1486 2011-05-05 12:47:27 <UukGoblin> sipa, ok I read the paper again; looks like some nodes will always have to store all blocks' contents. But how does a lightweight node know which blocks to request in order to verify a transaction?
1487 2011-05-05 12:49:29 marlowe has joined
1488 2011-05-05 12:49:31 eao has joined
1489 2011-05-05 12:50:07 draaglom has quit (Ping timeout: 240 seconds)
1490 2011-05-05 12:50:27 gavinandresen has joined
1491 2011-05-05 12:50:47 <UukGoblin> oh hang on they can't verify a transaction
1492 2011-05-05 12:51:22 <sacarlson> wow 8 hours for my bitcoind on main with clean slate hasn't fully downloaded all the blocks yet. what block should I see here or is this number corect 103911
1493 2011-05-05 12:51:37 <UukGoblin> ;;blocks
1494 2011-05-05 12:51:38 <gribble> Error: "blocks" is not a valid command.
1495 2011-05-05 12:51:40 <sipa> ;;bc,blocks
1496 2011-05-05 12:51:40 <UukGoblin> ;;bc,blocks
1497 2011-05-05 12:51:41 <gribble> 122024
1498 2011-05-05 12:51:42 <gribble> 122024
1499 2011-05-05 12:51:51 <sacarlson> wow that's slow
1500 2011-05-05 12:51:57 draaglom has joined
1501 2011-05-05 12:52:01 <sacarlson> I show I have 8 connections
1502 2011-05-05 12:52:44 <jgarzik> so...
1503 2011-05-05 12:52:53 <jgarzik> has anyone tested GUI in latest bitcoin.git?
1504 2011-05-05 12:53:01 <UukGoblin> hrm... so how does it know the money it receives aren't double-spent?
1505 2011-05-05 12:53:03 <jgarzik> even an "it works" report would be nice
1506 2011-05-05 12:53:13 <jgarzik> current upstream bitcoin/bitcoin.git
1507 2011-05-05 12:53:29 <sipa> jgarzik: sec
1508 2011-05-05 12:54:02 <sacarlson> jgarzik: I only run the bitcoind from git on my ubuntu 10.04 so no gui results from me
1509 2011-05-05 12:54:59 <sacarlson> other than spesmilo that works fine
1510 2011-05-05 12:55:25 <tcatm> jgarzik: compiles and seems to work
1511 2011-05-05 12:55:32 <UukGoblin> the paper is kinda vague on that... it says that disk space can be reclaimed by removing the blocks' contents, and that lightweight nodes might exist, but it doesn't seem to describe how to do it
1512 2011-05-05 12:55:36 <BlueMatt> why does bitcoin-git still show "Limit bitcoin generation to ..." in the options?
1513 2011-05-05 12:55:42 <UukGoblin> or I'm missing something (more likely)
1514 2011-05-05 12:55:43 <wumpus> I guess you're only interested in results with ubuntu 10.10?
1515 2011-05-05 12:55:48 <jgarzik> tcatm: 'generate coins' menu item no longer appears, as expected?
1516 2011-05-05 12:55:52 <BlueMatt> (though it doesnt mine)
1517 2011-05-05 12:55:55 <BlueMatt> jgarzik: yes
1518 2011-05-05 12:56:22 <tcatm> jgarzik: yes
1519 2011-05-05 12:56:28 <jgarzik> BlueMatt: I missed 'limit bitcoin generation' it sounds like
1520 2011-05-05 12:56:43 WakiMiko has quit (Read error: Operation timed out)
1521 2011-05-05 12:56:49 <BlueMatt> jgarzik: yep
1522 2011-05-05 12:56:52 eternal1 has joined
1523 2011-05-05 12:56:55 <jgarzik> BlueMatt: keep in mind, I never run GUI, so that patch was written blind
1524 2011-05-05 12:57:08 <BlueMatt> you gonna fix it or you want me to?
1525 2011-05-05 12:57:10 <jgarzik> BlueMatt: you're welcome to submit a pull request removing 'limit...'
1526 2011-05-05 12:57:19 <sipa> GUI works indeed
1527 2011-05-05 12:57:27 <jgarzik> BlueMatt: I nominate you :)
1528 2011-05-05 12:57:31 <wumpus> well it works on 10.04
1529 2011-05-05 12:57:37 <sipa> (ubuntu 10.10 here)
1530 2011-05-05 12:57:39 <wumpus> generate bitcoins is indeed gone, good :D
1531 2011-05-05 12:57:45 <BlueMatt> 11.04 works here (with wx-trunk)
1532 2011-05-05 12:57:46 <ersi> Thumbs up
1533 2011-05-05 12:58:49 <sacarlson> wumpus: were did you get the gui libs for ubuntu 10.04? I didn't see them in ppa
1534 2011-05-05 12:59:02 <wumpus> built from source
1535 2011-05-05 12:59:18 <wumpus> there is indeed not the right wx version in a ppa
1536 2011-05-05 12:59:32 larsivi has joined
1537 2011-05-05 12:59:58 <UukGoblin> ah, I missed the term 'spent' :-]
1538 2011-05-05 13:00:00 <BlueMatt> 2.9 is "beta" hence it is not included in any package managers
1539 2011-05-05 13:00:25 WakiMiko has joined
1540 2011-05-05 13:00:37 <sacarlson> wumpus: then I wouldn't say it works in 10.04 yet, I'll stick with spesmilo until the libs are around
1541 2011-05-05 13:00:57 Diablo-D3 has joined
1542 2011-05-05 13:01:16 <diki> Nvidia seems to be going down in sales
1543 2011-05-05 13:01:23 <Diablo-D3> thank god
1544 2011-05-05 13:01:27 <diki> And AMD up by 15%
1545 2011-05-05 13:01:51 <Diablo-D3> bitcoin == that 15%
1546 2011-05-05 13:01:53 <Diablo-D3> I swear
1547 2011-05-05 13:01:54 <UukGoblin> "by linking it [the transaction] to a place in the chain, he [a lightweight node] can see that a network node has accepted it" <- /how/ do you link the transaction to a place in the chain, I wonder...
1548 2011-05-05 13:02:02 <xelister> Diablo-D3: =)
1549 2011-05-05 13:02:19 <Diablo-D3> UukGoblin: add it to a block's tx list?
1550 2011-05-05 13:03:16 <sipa> UukGoblin: some protocol message that may not exist yet, for figuring out where a particulat markle node is in the chain
1551 2011-05-05 13:03:26 <diki> im thinking of grabbing a 5830 to help mine
1552 2011-05-05 13:03:39 <UukGoblin> right
1553 2011-05-05 13:03:40 <Diablo-D3> diki: no
1554 2011-05-05 13:03:42 <Diablo-D3> 5830s suck
1555 2011-05-05 13:03:46 <diki> why?
1556 2011-05-05 13:03:52 <Diablo-D3> they use as much power as a 5850, but have the mining power of around a 5770
1557 2011-05-05 13:03:58 <Diablo-D3> they're horrid
1558 2011-05-05 13:04:01 <sipa> UukGoblin: after which it can fetch the block to verify that tx is in there indeed
1559 2011-05-05 13:04:04 <Diablo-D3> either grab a 5770 or a 5850
1560 2011-05-05 13:04:13 <diki> the 5770 can CF
1561 2011-05-05 13:04:21 <diki> can't CF with a 5850
1562 2011-05-05 13:04:24 <UukGoblin> sipa, well presumably won't be able to fetch the full block
1563 2011-05-05 13:04:25 <Diablo-D3> cf?
1564 2011-05-05 13:04:28 <diki> crossfire
1565 2011-05-05 13:04:32 <UukGoblin> but just the transactions that are still kept...
1566 2011-05-05 13:04:42 <Diablo-D3> diki: erm, they all cf... and cf is bad for mining anyhow
1567 2011-05-05 13:05:04 <diki> CF for gaming, and not for mining
1568 2011-05-05 13:05:09 <UukGoblin> but it will be verify that the needed transaction is there because it will be able to verify the merkle root
1569 2011-05-05 13:05:10 <sipa> UukGoblin: of course, pruned transactions are not known anymore anyway
1570 2011-05-05 13:05:20 <UukGoblin> s/will be v/will be able to v/
1571 2011-05-05 13:05:29 fimp has joined
1572 2011-05-05 13:05:30 <diki> Meaning: 1.) Cf for games 2.) When i mine, i will disable it
1573 2011-05-05 13:05:41 <diki> and 5830 is within a certain price range
1574 2011-05-05 13:05:47 <Diablo-D3> diki: yeah but youi can cf together, say, a 5770 and a 5850
1575 2011-05-05 13:05:54 <Diablo-D3> and like I said, do NOT buy a 5830
1576 2011-05-05 13:05:56 <diki> no you cant
1577 2011-05-05 13:06:02 <diki> 5770 with a 57XX series
1578 2011-05-05 13:06:08 <Diablo-D3> diki: you can.
1579 2011-05-05 13:06:15 <diki> i've asked...you cant
1580 2011-05-05 13:06:21 <Diablo-D3> you can.
1581 2011-05-05 13:06:24 <diki> proof?
1582 2011-05-05 13:06:35 <Diablo-D3> cf is a load balancing feature of the driver.
1583 2011-05-05 13:06:43 <Diablo-D3> its not part of the hardware.
1584 2011-05-05 13:07:11 <Diablo-D3> as a side note, the cf bridge also does nothing
1585 2011-05-05 13:07:12 <UukGoblin> hrm, if all nodes on a network would become these lightweight type of nodes, they could lose important transaction information...
1586 2011-05-05 13:07:17 <BlueMatt> jgarzik: done. https://github.com/bitcoin/bitcoin/pull/193
1587 2011-05-05 13:07:25 <wumpus> forum db server broke down
1588 2011-05-05 13:07:32 <Diablo-D3> unplugging it does not turn off cf, and it doesnt slow down cf usage either
1589 2011-05-05 13:07:48 <diki> http://forums.anandtech.com/showthread.php?t=2027686
1590 2011-05-05 13:07:54 <BlueMatt> anyone have a script which acts as a half-node or other method and dumps the list of peers?
1591 2011-05-05 13:07:59 sabalaba has joined
1592 2011-05-05 13:08:06 <Diablo-D3> diki: that said, do NOT buy a 5830
1593 2011-05-05 13:08:12 <Diablo-D3> it just burns watts for no performance
1594 2011-05-05 13:08:15 sabalaba has quit (Read error: Connection reset by peer)
1595 2011-05-05 13:09:01 <Diablo-D3> uses 175 watts max, a 5850 uses 151, a 5870 uses 188, a 5770 uses 108
1596 2011-05-05 13:09:33 sabalaba has joined
1597 2011-05-05 13:10:06 <diki> 5850 uses 175
1598 2011-05-05 13:10:07 sabalaba has quit (Read error: Connection reset by peer)
1599 2011-05-05 13:10:10 <diki> while 5830 uses 150
1600 2011-05-05 13:10:14 <Diablo-D3> nope.
1601 2011-05-05 13:10:22 <davex__> so it looks like amdoverdrivectrl basically has to be run from GUI to actually change anything useful...
1602 2011-05-05 13:10:23 <davex__> right?
1603 2011-05-05 13:10:37 <Diablo-D3> davex__: I wasnt aware it was a console app too
1604 2011-05-05 13:10:47 <davex__> yeah it seems to take a couple parameters
1605 2011-05-05 13:10:51 <Diablo-D3> diki: all those cards at stock speed uses those wattages
1606 2011-05-05 13:10:56 <diki> http://www.tomshardware.com/reviews/radeon-hd-5850,2433-13.html.
1607 2011-05-05 13:11:08 Incitatus has joined
1608 2011-05-05 13:11:10 <Diablo-D3> diki: dude, amd publishes thes enumbers
1609 2011-05-05 13:11:17 <Diablo-D3> 5830 is shit.
1610 2011-05-05 13:11:38 <diki> everywhere i read people say you can't crossfire 5770 with a 5850
1611 2011-05-05 13:11:49 <Diablo-D3> diki: people do it.
1612 2011-05-05 13:11:51 <sipa> davex__: no, it can run in batch mode, if you have the settings in a file
1613 2011-05-05 13:12:16 <Diablo-D3> diki: what card do you have now?
1614 2011-05-05 13:12:28 <davex__> sipa: but you have to save the settings from the UI?
1615 2011-05-05 13:12:39 <sipa> i have no idea, i haven't ever seen the UI
1616 2011-05-05 13:12:42 <davex__> ok
1617 2011-05-05 13:12:42 <diki> a 5850
1618 2011-05-05 13:12:49 <Diablo-D3> diki: buy another 5850 then.
1619 2011-05-05 13:12:54 <sipa> i got the settings file from someone, modified some values, and ran it
1620 2011-05-05 13:12:55 <diki> way off the price range
1621 2011-05-05 13:13:01 <davex__> ah
1622 2011-05-05 13:13:01 <Diablo-D3> diki: cf DOES work best when your cards are matched
1623 2011-05-05 13:13:07 <noagendamarket> http://freedomwatch.uservoice.com/forums/16625-freedom-watch-show-ideas/suggestions/1778101-bitcoin
1624 2011-05-05 13:13:07 <noagendamarket> <------vote
1625 2011-05-05 13:13:19 <diki> i know, otherwise the card downclock to the slower card's clocks
1626 2011-05-05 13:13:21 <Diablo-D3> diki: I mean, not just arch revision, but the model itself
1627 2011-05-05 13:13:23 <sipa> BlueMatt: seems to work
1628 2011-05-05 13:13:25 <Diablo-D3> diki: no, this isnt nvidia.
1629 2011-05-05 13:13:27 <davex__> maybe i'll just write my own adl console app, might be easier
1630 2011-05-05 13:13:44 <Diablo-D3> diki: AMD drivers are much smarter and will just balance the load right
1631 2011-05-05 13:13:57 <diki> Ok, but i do need proof about a 5770
1632 2011-05-05 13:14:07 <diki> the arch is different
1633 2011-05-05 13:14:32 <Diablo-D3> some games are just stupid and do shit like every other frame
1634 2011-05-05 13:14:43 <Diablo-D3> so you get like 2x slowest card performance
1635 2011-05-05 13:15:06 <Diablo-D3> diki: the arch is actually the same
1636 2011-05-05 13:15:33 <diki> ok read here http://www.overclock.net/amd-ati/743238-5850-5770-crossfire.html
1637 2011-05-05 13:15:40 <diki> i need a specific chipset
1638 2011-05-05 13:15:41 <Diablo-D3> dude, I dont care about gamer sites.
1639 2011-05-05 13:15:49 <diki> that's not a gamer site
1640 2011-05-05 13:15:51 <diki> it's an OC site
1641 2011-05-05 13:15:58 <Diablo-D3> thats the same thing!
1642 2011-05-05 13:16:42 <diki> the official chart also says you cant
1643 2011-05-05 13:16:42 <diki> http://game.amd.com/us-en/content/images/crossfirex/CF_combo_chart.jpg
1644 2011-05-05 13:16:58 <Diablo-D3> welp, people are doing it
1645 2011-05-05 13:17:19 <diki> only lucid hydra chipsets can
1646 2011-05-05 13:17:43 <diki> mine is just a plain old 785G
1647 2011-05-05 13:17:57 <diki> and it has only a x4 pci-e
1648 2011-05-05 13:18:02 <Diablo-D3> lucid hydra is also retarded
1649 2011-05-05 13:18:08 <diki> but there won't be much of a difference
1650 2011-05-05 13:18:12 <Diablo-D3> its trying to implement block rendering inside of the chip
1651 2011-05-05 13:18:26 <Diablo-D3> diki: uh, if you dont have two x16 slots, you cant put two cards in
1652 2011-05-05 13:18:34 <diki> yes you can
1653 2011-05-05 13:18:53 <Diablo-D3> well, yes, if you have a riser that is x16 in, whatever you need out
1654 2011-05-05 13:18:54 <diki> at least i think it can
1655 2011-05-05 13:19:03 <diki> gotta test with my card
1656 2011-05-05 13:19:04 <Diablo-D3> but I doubt you have one
1657 2011-05-05 13:19:12 <Diablo-D3> the card, physically, wont fit if the slot isnt x16 long.
1658 2011-05-05 13:19:48 <diki> if you ask me, the size is absolutely the same
1659 2011-05-05 13:19:53 <sipa> strange
1660 2011-05-05 13:20:00 <sipa> if i recompile bitcoin while it is running
1661 2011-05-05 13:20:07 <sipa> i get an exception when exiting
1662 2011-05-05 13:20:16 <sipa> UNKNOWN EXCEPTION
1663 2011-05-05 13:20:16 <sipa> bitcoin in CMyApp::OnUnhandledException()
1664 2011-05-05 13:20:19 <Diablo-D3> diki: if its the same, then its a x16 slot. your mobo is probably referring to how its wired, which is unrelated to this
1665 2011-05-05 13:20:36 <Diablo-D3> sipa: probably got a terminate request from the OS
1666 2011-05-05 13:20:38 <CIA-30> bitcoin: Luke Dashjr * r6bdfec682228 pushpool/personal/msg.c: Allow miners to change timestamp http://tinyurl.com/424bd8g
1667 2011-05-05 13:20:39 <diki> 1 x PCI Express x16 slot, running at x4 is what i have
1668 2011-05-05 13:20:50 <Diablo-D3> diki: then its a x16 slot
1669 2011-05-05 13:20:59 <diki> the x4 speed won't make a big difference
1670 2011-05-05 13:21:00 <Diablo-D3> diki: x16 slots are not required to be wired at x16
1671 2011-05-05 13:21:15 <diki> as i've seen videos of a 6990 running without a difference
1672 2011-05-05 13:21:22 <wumpus> yes pretty interesting things can happen on some platforms if you modify an executable while running
1673 2011-05-05 13:21:24 <sipa> Diablo-D3: no, even when the exiting is hours later
1674 2011-05-05 13:21:26 <diki> only 200 3dmarks less
1675 2011-05-05 13:21:31 <Diablo-D3> sipa: weird.
1676 2011-05-05 13:21:36 <sipa> Diablo-D3: it happens when i want to close it
1677 2011-05-05 13:21:51 <Diablo-D3> diki: people mine off of x1 1.0 slots, thats 1/32th the speed of a x16 2.0 slot
1678 2011-05-05 13:21:59 <Diablo-D3> diki: gaming only suffers slightly
1679 2011-05-05 13:22:10 <CIA-30> bitcoin: Luke Dashjr * r6bdfec682228 pushpool-personal/msg.c: Allow miners to change timestamp http://tinyurl.com/424bd8g
1680 2011-05-05 13:22:15 <diki> the slot can't get saturated completely
1681 2011-05-05 13:22:20 <diki> with information
1682 2011-05-05 13:22:26 <luke-jr> oh, I was in the wrong channel :P
1683 2011-05-05 13:22:27 thedrs has joined
1684 2011-05-05 13:22:27 <Diablo-D3> it'll only saturate when uploading textures
1685 2011-05-05 13:22:29 <diki> i don't even see the need of pci-e 3
1686 2011-05-05 13:22:44 <Diablo-D3> diki: its about command latency
1687 2011-05-05 13:22:45 <wumpus> many OSses simply mmap the executable into memory, so if you change the file, the actual executed code changes
1688 2011-05-05 13:22:59 <sipa> wumpus: that could explain things
1689 2011-05-05 13:23:05 <Diablo-D3> wumpus: yeah, but if the pages are locked, thats not an issue
1690 2011-05-05 13:23:07 <wumpus> at least with solaris and shared libraries this is the case, dunno about others
1691 2011-05-05 13:23:29 <wumpus> haven't had the issue on linux yet
1692 2011-05-05 13:23:35 <diki> i do hope the 600w supply will be enough
1693 2011-05-05 13:23:56 <diki> no matter how i calculate even the PSU operating at 80% efficiency should still allow CF
1694 2011-05-05 13:23:59 <Diablo-D3> 600w peak or 600w constant?
1695 2011-05-05 13:24:06 <diki> i am not sure
1696 2011-05-05 13:24:10 <Diablo-D3> find out
1697 2011-05-05 13:24:13 <diki> but even 550 should be enough
1698 2011-05-05 13:24:14 <Diablo-D3> most PSUs love to lie
1699 2011-05-05 13:24:22 <netxshare> I Thought they all do
1700 2011-05-05 13:24:24 <Diablo-D3> 600 peak could mean 450 or less constant
1701 2011-05-05 13:24:33 <netxshare> what brand is it?
1702 2011-05-05 13:24:57 <Diablo-D3> you need about 300 for two 5850s and then, say, another 150-200 for the computer
1703 2011-05-05 13:25:00 <diki> says 576 MAX
1704 2011-05-05 13:25:10 <netxshare> ep
1705 2011-05-05 13:25:11 <Diablo-D3> diki: thats not useful
1706 2011-05-05 13:25:16 <thedrs> hi all
1707 2011-05-05 13:25:17 <diki> nothing other than that
1708 2011-05-05 13:25:23 <Diablo-D3> diki: probably a chinese firecracker brand
1709 2011-05-05 13:25:27 <diki> but all the 12v 4 rails give a total of 56 amps
1710 2011-05-05 13:25:28 <Diablo-D3> I'd buy a new psu
1711 2011-05-05 13:25:39 <netxshare> they like to go pop pop
1712 2011-05-05 13:25:43 <Diablo-D3> yuck a multirail psu
1713 2011-05-05 13:25:47 <netxshare> buy a good brand too
1714 2011-05-05 13:25:49 <netxshare> aye
1715 2011-05-05 13:25:54 <diki> It's a Chieftec 600w
1716 2011-05-05 13:25:59 <Diablo-D3> LOL CHIEFTEC
1717 2011-05-05 13:25:59 <Diablo-D3> fail
1718 2011-05-05 13:26:01 <netxshare> single rail please
1719 2011-05-05 13:26:05 <Diablo-D3> thats a chinese firecracker brand
1720 2011-05-05 13:26:09 <diki> single rail is not always better than multi
1721 2011-05-05 13:26:15 <Diablo-D3> yes it is.
1722 2011-05-05 13:26:16 <Diablo-D3> yes. it is.
1723 2011-05-05 13:26:17 <netxshare> uh
1724 2011-05-05 13:26:20 <diki> no proof
1725 2011-05-05 13:26:26 <netxshare> go read some reviews
1726 2011-05-05 13:26:28 <netxshare> on psus
1727 2011-05-05 13:26:28 * [Tycho] uses chieftec PSUs, they are very good
1728 2011-05-05 13:26:29 <netxshare> REAL ones
1729 2011-05-05 13:26:43 <Diablo-D3> [Tycho]: yeah, but dont you have a giant mining farm?
1730 2011-05-05 13:26:47 <xelister> [Tycho]: says windows user
1731 2011-05-05 13:26:51 <[Tycho]> Yes, i do.
1732 2011-05-05 13:26:54 <netxshare> http://www.hardwaresecrets.com/article/410
1733 2011-05-05 13:26:56 <netxshare> start there
1734 2011-05-05 13:26:58 molecular has quit (Read error: Operation timed out)
1735 2011-05-05 13:27:01 <Diablo-D3> [Tycho]: a) you should know better, b) if one pops you havent lost anything
1736 2011-05-05 13:27:04 <xelister> >_>
1737 2011-05-05 13:27:06 <netxshare> and read the whole thing
1738 2011-05-05 13:27:11 <netxshare> it will help
1739 2011-05-05 13:27:42 <Diablo-D3> worst case is it fries all the hardware in your box
1740 2011-05-05 13:27:53 <xelister> and your apartment
1741 2011-05-05 13:28:05 <netxshare> it would most likely just blow up first
1742 2011-05-05 13:28:07 <thedrs> small q: i saw on the bitcoin hardware page high hash rate statistics for miners specifically compiled for Core2Duo (E7300 getting 7.76) - i currently get 2.5 and that is on a E8400 ... i am using rpcminer-cpu ... can someone point me to binaries (windowz) of a miner compiled for core2duo ?
1743 2011-05-05 13:28:22 <Diablo-D3> thedrs: dont bother cpu mining, its a waste
1744 2011-05-05 13:28:33 <netxshare> maybe he has like 1 million cpus
1745 2011-05-05 13:28:41 <netxshare> and does not pay power
1746 2011-05-05 13:28:48 <thedrs> i know i know .... still when you have full classes just waiting to be used at night ...
1747 2011-05-05 13:29:01 <eps> heh
1748 2011-05-05 13:29:04 <UukGoblin> ok my new argument is: if you allow a new non-standard transaction for timestamping, it'll be easier for nodes who want to discard it, to discard it :-]
1749 2011-05-05 13:29:07 <diki> well if it blows up, i'll pray to the warranty
1750 2011-05-05 13:29:08 <netxshare> you wont even earn a bitcoin
1751 2011-05-05 13:29:09 <Diablo-D3> thedrs: power costs more than the coins are worth
1752 2011-05-05 13:29:11 <UukGoblin> meh
1753 2011-05-05 13:29:12 <netxshare> dude diki
1754 2011-05-05 13:29:14 <netxshare> don't do it
1755 2011-05-05 13:29:16 <UukGoblin> I need a way of writing this up
1756 2011-05-05 13:29:17 <netxshare> it will
1757 2011-05-05 13:29:27 <netxshare> 24/7 load on that psu
1758 2011-05-05 13:29:27 <Diablo-D3> UukGoblin: yes, but I want that option in bitcoin: hard reject mode.
1759 2011-05-05 13:29:40 <thedrs> and back to the question ... ? where can i find better binaries ?
1760 2011-05-05 13:29:51 <diki> look, it's not like i have enough money to buy even a single 5750
1761 2011-05-05 13:29:58 <diki> so a new psu is also way off for now
1762 2011-05-05 13:30:01 <Diablo-D3> thedrs: that sounds like bullshit, I have a e8500, nothing is going to make it do more than 3 mhash.
1763 2011-05-05 13:30:11 <diki> the money will come strictly from btc FOR btc
1764 2011-05-05 13:30:14 <Diablo-D3> thedrs: this is why everyone gpu mines.
1765 2011-05-05 13:30:23 <thedrs> diablo-d3: https://en.bitcoin.it/wiki/Mining_hardware_comparison
1766 2011-05-05 13:30:30 <netxshare> okay
1767 2011-05-05 13:30:32 <netxshare> I wish you luck
1768 2011-05-05 13:30:45 <netxshare> but when your psu dies and you have no money to buy a new one what will you do?
1769 2011-05-05 13:30:55 <Diablo-D3> thedrs: that clearly says its using ufasoft
1770 2011-05-05 13:31:13 <Diablo-D3> that sounds like a lot of bs though
1771 2011-05-05 13:31:17 <netxshare> what
1772 2011-05-05 13:31:21 <netxshare> 18.9MH/s
1773 2011-05-05 13:31:24 <netxshare> on an i7
1774 2011-05-05 13:31:31 <netxshare> I guess I never tested my cpu
1775 2011-05-05 13:31:39 <thedrs> what an idiot i am - i missed that column
1776 2011-05-05 13:32:25 <thedrs> 14MH/s on i5-2400
1777 2011-05-05 13:32:32 <diki> i've predicted i need about 150btc to buy new stuff
1778 2011-05-05 13:32:36 <diki> and i have a long list
1779 2011-05-05 13:32:40 <netxshare> what card do you have?
1780 2011-05-05 13:32:48 <Diablo-D3> that sounds very much bullshit though
1781 2011-05-05 13:32:48 <diki> New cooler,more ram, better case, another gpu to mine
1782 2011-05-05 13:32:51 <thedrs> me ? X1050 :((
1783 2011-05-05 13:33:05 <netxshare> no, diki
1784 2011-05-05 13:33:09 <thedrs> and mother board cards
1785 2011-05-05 13:33:10 <netxshare> why more ram?
1786 2011-05-05 13:33:15 <netxshare> just get another gpu
1787 2011-05-05 13:33:17 <Diablo-D3> thedrs: a radeon x1050? wont work
1788 2011-05-05 13:33:18 <netxshare> and mine faster
1789 2011-05-05 13:33:18 <UukGoblin> this is awesome, generally.
1790 2011-05-05 13:33:21 <diki> i have 2GB of ram
1791 2011-05-05 13:33:22 <netxshare> then get other crap
1792 2011-05-05 13:33:35 <diki> you don't know what it's like when it's all used up
1793 2011-05-05 13:33:37 <Diablo-D3> diki: you dont need more to min
1794 2011-05-05 13:33:37 <thedrs> Diablo-D3: no ... it starts from 4XXX
1795 2011-05-05 13:33:39 <Diablo-D3> *mine
1796 2011-05-05 13:33:46 marlowe has quit (Ping timeout: 276 seconds)
1797 2011-05-05 13:33:47 <UukGoblin> you could actually use the main bitcoin chain, and make it lightweight at the same time. Some nodes would store the extra information, some would discard it
1798 2011-05-05 13:33:49 <diki> No it starts after 4200
1799 2011-05-05 13:33:57 <Diablo-D3> thedrs: I wrote a miner, I know what cards work ;)
1800 2011-05-05 13:34:03 <netxshare> thedrs: https://en.bitcoin.it/wiki/Category:Miners
1801 2011-05-05 13:34:10 <Diablo-D3> and for the record, an igp 42xx is not a 4xxx
1802 2011-05-05 13:34:22 <Diablo-D3> you need igp 6xxx or later to mine, and its still pointless
1803 2011-05-05 13:34:35 <thedrs> Diable-D3: so are you saying it could work on an X1050 ?
1804 2011-05-05 13:34:44 * UukGoblin needs some time to reorganize his synapses
1805 2011-05-05 13:34:47 <Diablo-D3> thedrs: no, Im saying you have no chance in hell
1806 2011-05-05 13:34:52 <netxshare> haha
1807 2011-05-05 13:35:12 <UukGoblin> nanotube, does this bitdns thing, these transactions with fees, does it care if they are pruned?
1808 2011-05-05 13:35:30 marlowe has joined
1809 2011-05-05 13:35:30 <Diablo-D3> UukGoblin: the big problem is
1810 2011-05-05 13:35:41 <Diablo-D3> UukGoblin: bitcoin needs to be fixed to reject trash tx
1811 2011-05-05 13:35:46 <Diablo-D3> they are not valid bitcoin tx, they need to be removed
1812 2011-05-05 13:36:22 <UukGoblin> no, not reject
1813 2011-05-05 13:36:22 <thedrs> :))
1814 2011-05-05 13:36:23 <Diablo-D3> if they're valid on some other network, then let some other network deal with it
1815 2011-05-05 13:36:27 <Diablo-D3> yes, reject
1816 2011-05-05 13:36:28 <UukGoblin> yeah
1817 2011-05-05 13:36:32 <netxshare> my god
1818 2011-05-05 13:36:32 <netxshare> http://www.hardocp.com/images/news/1304550897OFGPuJch3o_1_1_l.jpg
1819 2011-05-05 13:36:39 <UukGoblin> the block shouldn't get dropped because of that though
1820 2011-05-05 13:36:45 <Diablo-D3> UukGoblin: yes it should
1821 2011-05-05 13:36:50 <Diablo-D3> because the block cannot be verified.
1822 2011-05-05 13:36:50 <diki> is that a cooler master cooler?
1823 2011-05-05 13:37:02 <Diablo-D3> netxshare: THE FUCK
1824 2011-05-05 13:37:05 <Diablo-D3> WHAT THE FUCK HAPPENED
1825 2011-05-05 13:37:28 <diki> mine has been in a similar state
1826 2011-05-05 13:37:32 <UukGoblin> Diablo-D3, why not? it has the right previous block's hash, it has a verifiable merkle hash root, it has a verifiable proof-of-work - all can be verified
1827 2011-05-05 13:37:35 <diki> not as much but still
1828 2011-05-05 13:37:36 <diki> close
1829 2011-05-05 13:37:48 <Diablo-D3> UukGoblin: but the tx itself cannot be verified.
1830 2011-05-05 13:37:57 <diki> tx?
1831 2011-05-05 13:38:00 <UukGoblin> Diablo-D3, it could
1832 2011-05-05 13:38:08 <Diablo-D3> it cant on a stock bitcoin client.
1833 2011-05-05 13:38:28 <UukGoblin> it can, stock client accepts 0 outputs apparently
1834 2011-05-05 13:38:29 <Diablo-D3> which means its an attack on the network and needs to be rejected.
1835 2011-05-05 13:38:54 <UukGoblin> and if it's cannibalized not to, then there's still the option of burning a coin
1836 2011-05-05 13:38:55 <Diablo-D3> Im going to have to code a client now
1837 2011-05-05 13:39:00 <diki> Eyecancer!Eyecancer!
1838 2011-05-05 13:39:02 mologie has joined
1839 2011-05-05 13:39:10 <Diablo-D3> one that rejects broken blocks
1840 2011-05-05 13:39:42 <UukGoblin> Diablo-D3, that'd be silly, it'd advocate coin burning.
1841 2011-05-05 13:39:50 <Diablo-D3> UukGoblin: not at all
1842 2011-05-05 13:40:04 <UukGoblin> Diablo-D3, you'd have no other way of storing arbitrary data in your blockchain than by burning.
1843 2011-05-05 13:40:06 <Diablo-D3> bitdns and all that other shit would cease to work
1844 2011-05-05 13:40:16 <UukGoblin> Diablo-D3, (if what you're proposing is to reject 0 outputs)
1845 2011-05-05 13:40:17 <Diablo-D3> my client would reject them.
1846 2011-05-05 13:40:30 <UukGoblin> Diablo-D3, they would not cease to work if they used coin burning
1847 2011-05-05 13:40:49 <Diablo-D3> and it wouldnt be fucking coin burning if the bitdns clients would give a valid bitcoin address
1848 2011-05-05 13:41:10 <Diablo-D3> which is another reason to ban that entire project
1849 2011-05-05 13:41:21 <Diablo-D3> if they want to do their shit, they can do their shit on their own network
1850 2011-05-05 13:41:24 <diki> if you ask me, bitcoin should have a way of verifying if an address is valid
1851 2011-05-05 13:41:29 <jgarzik> I don't see anything wrong with zero outputs. That's just dumping money on a miner.
1852 2011-05-05 13:41:30 <Diablo-D3> diki: cant.
1853 2011-05-05 13:41:39 <sipa> diki: define valid?
1854 2011-05-05 13:41:39 <diki> but it should
1855 2011-05-05 13:41:46 <sipa> that someone owns it?
1856 2011-05-05 13:41:49 <Diablo-D3> diki: how do you validate it.
1857 2011-05-05 13:41:49 <diki> well, an address that doesnt exist
1858 2011-05-05 13:41:52 molecular has joined
1859 2011-05-05 13:42:03 <Diablo-D3> but how do you know it doesnt exist?
1860 2011-05-05 13:42:14 <sipa> diki: and would an once existing address whose key is lost, count as valid?
1861 2011-05-05 13:42:21 <diki> using a similar method where the blocks are stored
1862 2011-05-05 13:42:31 <Diablo-D3> diki: but the addresses are only public keys
1863 2011-05-05 13:42:41 <diki> or the way they are verified, when you generate a new addres you should broadcast it to the other clients
1864 2011-05-05 13:42:44 <Diablo-D3> how do you know the public key is valid without having a list of all private keys?
1865 2011-05-05 13:42:59 <diki> it doesnt matter who it belongs
1866 2011-05-05 13:42:59 <Diablo-D3> if you're broadcasting the ownership of a public key, how is bitcoin anonymous?
1867 2011-05-05 13:43:03 <sipa> diki: why would you?
1868 2011-05-05 13:43:04 <UukGoblin> Diablo-D3, bitdns servers giving out their addresses might be interesting, but hard to verify and make fair
1869 2011-05-05 13:43:07 <diki> what matters is it belongs to someone
1870 2011-05-05 13:43:29 <sipa> if i have an address, it surely belonged to someone sometime
1871 2011-05-05 13:43:34 <Diablo-D3> diki: yes, which means whoever said it exists first owns it, and then we know where to send the FBI
1872 2011-05-05 13:43:35 <sipa> or i just conjured it up
1873 2011-05-05 13:43:45 <Diablo-D3> plus, addresses are produced constantly
1874 2011-05-05 13:43:50 <Diablo-D3> its a 2^160 space.
1875 2011-05-05 13:43:52 <diki> the point is, it exists
1876 2011-05-05 13:44:04 <Diablo-D3> yes, and bitcoin is anonymous
1877 2011-05-05 13:44:08 <UukGoblin> jgarzik, by 0 output I mean one output to an arbitrary address with 0 value, not empty output, actually.
1878 2011-05-05 13:44:10 <sipa> diki: what problem would it solve, exactly?
1879 2011-05-05 13:44:11 <diki> i never said it wasn't
1880 2011-05-05 13:44:24 <diki> it would solve the problem of sending coins to a non-existing address
1881 2011-05-05 13:44:28 <diki> at least most of the time
1882 2011-05-05 13:44:33 <Diablo-D3> diki: so if a node broadcasts it knows of a new address, then that node obviously owns the address if no one else has broadcast that yet
1883 2011-05-05 13:44:53 <Diablo-D3> which means I can directly correlate an address to a person
1884 2011-05-05 13:44:58 <diki> now that you mention it, it would solve duplicate addresses if they exist
1885 2011-05-05 13:44:59 luke-jr has quit (Read error: Operation timed out)
1886 2011-05-05 13:45:10 <diki> you can't corelate an address to anyone
1887 2011-05-05 13:45:17 <UukGoblin> Diablo-D3, and banning other projects is like trying to ban bitcoin. Good luck with that.
1888 2011-05-05 13:45:28 <BlueMatt> does noone really have a script which somehow gets a list of nodes which it dumps?
1889 2011-05-05 13:45:30 <Diablo-D3> UukGoblin: if clients reject bad tx, whats the problem
1890 2011-05-05 13:45:44 <Diablo-D3> diki: you cant have duplicate addreses.
1891 2011-05-05 13:45:45 <UukGoblin> Diablo-D3, the problem is you have no way of telling which tx is bad
1892 2011-05-05 13:45:54 <UukGoblin> all look the same
1893 2011-05-05 13:46:06 <Diablo-D3> diki: and also, yes, if a node says theres a new address, the first node that said this IS THE OWNER OF THE ADDRESS
1894 2011-05-05 13:46:06 <diki> who said you can't?
1895 2011-05-05 13:46:15 <Diablo-D3> diki: its a 2^160 space
1896 2011-05-05 13:46:19 <Diablo-D3> its easier to win powerball.
1897 2011-05-05 13:46:22 <diki> that really doesnt tell me anything
1898 2011-05-05 13:46:24 <diki> im bad at math
1899 2011-05-05 13:46:44 <netxshare> 2^160 = 1.46150164 Ã 1048
1900 2011-05-05 13:46:51 <UukGoblin> diki, it's easier to win a lottery. that should speak for itself
1901 2011-05-05 13:47:01 <netxshare> er it forgot the 48
1902 2011-05-05 13:47:06 <diki> they're all staged anyway
1903 2011-05-05 13:47:18 <Diablo-D3> more than one quindecillion
1904 2011-05-05 13:47:55 luke-jr has quit (otg!~luke-jr@ishibashi.dashjr.org|Ping timeout: 260 seconds)
1905 2011-05-05 13:47:59 <netxshare> the lottery is staged?
1906 2011-05-05 13:48:08 <UukGoblin> Diablo-D3, what exactly do you want to change with the stock client? Is that rejecting of 0-valued outputs?
1907 2011-05-05 13:48:17 <Diablo-D3> UukGoblin: rejecting tx that make no sense.
1908 2011-05-05 13:48:27 luke-jr has joined
1909 2011-05-05 13:48:37 <UukGoblin> Diablo-D3, example?
1910 2011-05-05 13:48:40 <Diablo-D3> UukGoblin: like, an address is going to be between x and x characters, start with 1, etc
1911 2011-05-05 13:48:55 <Diablo-D3> and the coins sent cant be too big or too small
1912 2011-05-05 13:48:55 marlowe has quit (Ping timeout: 240 seconds)
1913 2011-05-05 13:48:56 <sipa> transactions do not contain addresses
1914 2011-05-05 13:49:02 luke-jr has joined
1915 2011-05-05 13:49:04 <Diablo-D3> sipa: target address.
1916 2011-05-05 13:49:05 <sipa> they just contain the 160-bit hash
1917 2011-05-05 13:49:12 <sipa> and any 160-bit number there is valid
1918 2011-05-05 13:49:18 <diki> yes it is
1919 2011-05-05 13:49:21 <diki> at least where i live
1920 2011-05-05 13:49:30 <Diablo-D3> diki: go take your paranoia elsewhere
1921 2011-05-05 13:49:40 <netxshare> where do you live
1922 2011-05-05 13:49:42 <diki> i dont live in the us
1923 2011-05-05 13:49:46 <netxshare> so?
1924 2011-05-05 13:49:47 <diki> i live europe
1925 2011-05-05 13:49:49 <netxshare> rofl
1926 2011-05-05 13:49:51 <netxshare> they are legit
1927 2011-05-05 13:49:57 <diki> In a country that has a lot of gypsys
1928 2011-05-05 13:49:58 <CIA-30> bitcoin: various personal * r36534c..85fcc0 pushpool-personal/ (30 files in 2 dirs): (116 commits) http://tinyurl.com/3egzu9w
1929 2011-05-05 13:50:08 <Diablo-D3> yes, only the niggers, jews, italians, etc can win the fucking lottery.
1930 2011-05-05 13:50:13 <sipa> don't trust them, don't play with them - at least i don't
1931 2011-05-05 13:50:17 <Diablo-D3> its always against us white people isnt it
1932 2011-05-05 13:50:22 <diki> my points is...most people that win..are never heard of again
1933 2011-05-05 13:50:23 <Diablo-D3> goddamned fucking losers.
1934 2011-05-05 13:50:24 <netxshare> damn us
1935 2011-05-05 13:50:24 <sipa> if people want to spend money on that, fine
1936 2011-05-05 13:50:34 <Diablo-D3> diki: of course you dont! they're too busy being rich!
1937 2011-05-05 13:50:39 <diki> or being dead
1938 2011-05-05 13:50:40 <UukGoblin> a non-standard transaction with a fee to the miner is better than burning of coins.
1939 2011-05-05 13:50:48 <UukGoblin> that's my claim and I can support it. :-]
1940 2011-05-05 13:50:51 <Diablo-D3> if I had 50 million euros in my pocket, why the fuck would I bother dealing with you poor people?!
1941 2011-05-05 13:51:00 <netxshare> hells yea
1942 2011-05-05 13:51:08 <Diablo-D3> UukGoblin: then the fee should automatically become 50btc.
1943 2011-05-05 13:51:25 <sipa> UukGoblin: yes, but both are worse than just not using the bitcoin financial block chain for your own purposes
1944 2011-05-05 13:51:37 <Diablo-D3> UukGoblin: the outpoint coins of a block must be "vaiid"
1945 2011-05-05 13:51:49 <UukGoblin> sipa, yes, but that's unstoppable, really.
1946 2011-05-05 13:51:49 <Diablo-D3> UukGoblin: ie, gen -> 50, plus all the input coins of the tx.
1947 2011-05-05 13:51:56 Xunie has joined
1948 2011-05-05 13:51:56 <Diablo-D3> UukGoblin: the "fee" is whatever is left.
1949 2011-05-05 13:51:56 <UukGoblin> back to where we started anyway
1950 2011-05-05 13:52:05 <sipa> UukGoblin: and i never claimed it was unstoppable
1951 2011-05-05 13:52:11 <Diablo-D3> UukGoblin: anything that doesnt have a matching output becomes part of the fee
1952 2011-05-05 13:52:28 <Diablo-D3> so if there is no gen pay tx, then someones getting part of that fee.
1953 2011-05-05 13:52:32 <sipa> you can't stop me from throwing a brick through your wall either
1954 2011-05-05 13:52:40 <sipa> but as it doesn't benefit either of us, i won't
1955 2011-05-05 13:52:56 <Diablo-D3> UukGoblin: bitdns nodes cant generate new blocks, right?
1956 2011-05-05 13:53:02 <UukGoblin> Diablo-D3, yes I know how fees are stored
1957 2011-05-05 13:53:10 Incitatus has quit (Ping timeout: 264 seconds)
1958 2011-05-05 13:53:27 <UukGoblin> Diablo-D3, I don't know what bitdns you speak of
1959 2011-05-05 13:53:34 <Diablo-D3> namecoin or whatever the fuck it is
1960 2011-05-05 13:53:38 <Diablo-D3> the project thats attacking bitcoin
1961 2011-05-05 13:53:48 <netxshare> how is it attacking bitcoin?
1962 2011-05-05 13:53:55 <UukGoblin> no-one's attacking anyone
1963 2011-05-05 13:54:09 <UukGoblin> I have no idea what you're talking about now
1964 2011-05-05 13:54:17 <sipa> namecoin is completely separate from bitcoin
1965 2011-05-05 13:54:21 <Diablo-D3> its filling the chain with broken tx and burning coins.
1966 2011-05-05 13:54:27 <netxshare> that's what I thought
1967 2011-05-05 13:54:28 <sipa> that's not a thread at all, but imho just pointless
1968 2011-05-05 13:54:45 diki has quit ()
1969 2011-05-05 13:54:56 <[Noodles]> namecoin has its own chain, it's not connected to bitcoin at all
1970 2011-05-05 13:55:06 <sipa> s/Thread/Threat/
1971 2011-05-05 13:55:08 pnicholson has joined
1972 2011-05-05 13:55:22 <netxshare> are a lot of people using namecoin?
1973 2011-05-05 13:55:43 <Diablo-D3> okay then
1974 2011-05-05 13:55:46 <luke-jr> ;;bc,calc [bc,eligius]
1975 2011-05-05 13:55:46 <Diablo-D3> what the fuck are we talking about?
1976 2011-05-05 13:55:47 <gribble> The average time to generate a block at 16721739.3391 Khps, given current difficulty of 109670.13329248 , is 7 hours, 49 minutes, and 28 seconds
1977 2011-05-05 13:55:56 <netxshare> I dunno
1978 2011-05-05 13:56:02 <netxshare> about getting me some food
1979 2011-05-05 13:56:05 <netxshare> I pay BTC
1980 2011-05-05 13:56:08 <Diablo-D3> if no one is putting trash tx in the block list, whats the problem?
1981 2011-05-05 13:56:21 <UukGoblin> netxshare, no, I think it's bad by design, but I might run it if cross-mining is implemented cause it might be profitable.
1982 2011-05-05 13:56:35 <netxshare> yeah
1983 2011-05-05 13:56:40 <netxshare> I was thinking about buying some name coins
1984 2011-05-05 13:56:43 <netxshare> namecoins*
1985 2011-05-05 13:56:49 <Diablo-D3> UukGoblin: seriously, whats the whole issue
1986 2011-05-05 13:56:52 <netxshare> someone was selling 1000nc for 7BTC
1987 2011-05-05 13:56:59 <UukGoblin> Diablo-D3, I was talking about a p2p timestamping service. You broadcast a hash of your data, you send a fee to the miner, you get your data included in the bitcoin blockchain and thus timestamped.
1988 2011-05-05 13:57:10 <UukGoblin> s/data included/hash of data included/
1989 2011-05-05 13:57:12 <Diablo-D3> UukGoblin: except thats mostly invalid
1990 2011-05-05 13:57:41 <UukGoblin> Diablo-D3, what's invalid
1991 2011-05-05 13:57:47 <Diablo-D3> that whole concept
1992 2011-05-05 13:57:53 <jgarzik> an arbitrary-data chain could be designed -so- much more elegantly, and efficiently
1993 2011-05-05 13:57:56 <Diablo-D3> because the bitcoin clients would have to "ignore" no output blocks.
1994 2011-05-05 13:57:58 <jgarzik> if not shoehorned into bitcoin
1995 2011-05-05 13:58:13 <jgarzik> main chain
1996 2011-05-05 13:58:18 <Diablo-D3> and not count that towards the 21m coin limit
1997 2011-05-05 13:58:26 <UukGoblin> Diablo-D3, some could, some wouldn't have to.
1998 2011-05-05 13:58:30 glassresistor has quit (Read error: Operation timed out)
1999 2011-05-05 13:58:40 <Diablo-D3> otherwise it burns coins no matter if you have an output or not
2000 2011-05-05 13:58:49 <Diablo-D3> and if its burning coins, I want that completely rejected
2001 2011-05-05 13:59:03 <UukGoblin> Diablo-D3, you cannot reject burning coins! every bitcoin user can do that
2002 2011-05-05 13:59:15 <sipa> ... what is wrong with TD's proposal?
2003 2011-05-05 13:59:25 <Diablo-D3> UukGoblin: I can stop using bitcoin.
2004 2011-05-05 13:59:25 <sipa> it defines a elegant arbitrary chain
2005 2011-05-05 13:59:32 <Diablo-D3> sipa: because people are fucking retarded
2006 2011-05-05 13:59:34 <sipa> it links it to the block chain with almost no overhead
2007 2011-05-05 13:59:36 <UukGoblin> sipa, TD's proposal is pretty OK, I just don't see a use for a separate chain.
2008 2011-05-05 13:59:40 <Diablo-D3> UukGoblin: I think you misunderstood what I saidm btw
2009 2011-05-05 13:59:53 <Diablo-D3> UukGoblin: if clients count the number of coins in existence via the number of blocks
2010 2011-05-05 13:59:55 <CIA-30> bitcoin: Luke Dashjr personal * rfddf99997dfa pushpool-personal/msg.c: Allow miners to change timestamp within reason http://tinyurl.com/6dlwa7y
2011 2011-05-05 13:59:58 <luke-jr> jgarzik: ^
2012 2011-05-05 14:00:06 <Diablo-D3> UukGoblin: then a block with no outputs just burned 50 coins
2013 2011-05-05 14:00:26 <noagendamarket> oh god
2014 2011-05-05 14:00:30 <noagendamarket> not this again
2015 2011-05-05 14:00:31 <UukGoblin> Diablo-D3, blocks don't have outputs, transactions do
2016 2011-05-05 14:00:32 <sipa> Diablo-D3: we're not talking about generation tx's
2017 2011-05-05 14:00:41 <Diablo-D3> UukGoblin: you missed the entire point of what I just said
2018 2011-05-05 14:00:52 <Diablo-D3> sipa: yes, but if you have some fucking homo timestamping client, its going to be genning blocks
2019 2011-05-05 14:00:53 <sipa> UukGoblin: well, i don't want the bitcoin chain used for other things than financial transactions - doing so would decrease bitcoin's *value* for *me*
2020 2011-05-05 14:01:10 <Diablo-D3> sipa: and its going to be genning nonsensical blocks that fuck shit up
2021 2011-05-05 14:01:11 <jgarzik> luke-jr: testing and verified that endianness is correct?
2022 2011-05-05 14:01:17 <sipa> Diablo-D3: have you read the proposal even?
2023 2011-05-05 14:01:17 <luke-jr> jgarzik: no
2024 2011-05-05 14:01:26 <jgarzik> luke-jr: I think you're missing a byteswap
2025 2011-05-05 14:01:33 <Diablo-D3> sipa: the proposal sounds like bs
2026 2011-05-05 14:01:35 <UukGoblin> sipa, well, I don't want other miners mining bitcoin blocks, that decreases the bitcoin's value to me too.
2027 2011-05-05 14:01:45 <luke-jr> maybe, but I thought getwork data was always x86-endian
2028 2011-05-05 14:01:47 <Diablo-D3> UukGoblin: blocks HAVE outputs
2029 2011-05-05 14:01:54 <Diablo-D3> UukGoblin: its a list of all the outputs of the tx.
2030 2011-05-05 14:02:13 <Diablo-D3> UukGoblin: so if I have a block that has nothing but 1 gen tx, the block as an output of 50btc.
2031 2011-05-05 14:02:34 <sipa> UukGoblin: please elaborate?
2032 2011-05-05 14:02:42 <sipa> ah, yes
2033 2011-05-05 14:02:51 <sipa> well you don't have to use bitcoin if you don't like the scheme
2034 2011-05-05 14:02:53 <UukGoblin> Diablo-D3, OK, yeah, we're not talking about the gen tx
2035 2011-05-05 14:02:59 <UukGoblin> sipa, likewise :-]
2036 2011-05-05 14:03:20 <UukGoblin> I think there's still too much misunderstanding... I'll try to write up another post tonight or tomorrow
2037 2011-05-05 14:03:40 <sipa> UukGoblin: except that you need other miners (in a socio-economical way) for bitcoin to have value
2038 2011-05-05 14:04:02 <Diablo-D3> UukGoblin: yes, but how do you put bs tx into the chain without mining broken blocks?
2039 2011-05-05 14:04:16 <UukGoblin> Diablo-D3, easy
2040 2011-05-05 14:04:29 <UukGoblin> Diablo-D3, bitcoind sendtoaddress myfuckingimportantdata 0.01
2041 2011-05-05 14:04:29 <jgarzik> luke-jr: it's not that simple :)
2042 2011-05-05 14:04:58 x5x`brb is now known as x5x
2043 2011-05-05 14:05:11 <UukGoblin> Diablo-D3, that's substandard, what I'd rather have and what would benefit everyone much more would be bitcoind sendnonstandard myfuckingimportantdata fee=0.01
2044 2011-05-05 14:05:40 <jgarzik> there are better ways to pay someone to store data
2045 2011-05-05 14:05:52 <UukGoblin> I believe currently the nonstandard is positive input, 0-valued output to an address of myfuckingimportantdata
2046 2011-05-05 14:06:14 <UukGoblin> but if you take action against it, you'll make things worse for everyone, because I'll have to revert to the original method
2047 2011-05-05 14:06:26 <Diablo-D3> UukGoblin: feh.
2048 2011-05-05 14:06:41 <Diablo-D3> whatever, someone is just going to store child porn in the chain anyhow
2049 2011-05-05 14:06:48 <UukGoblin> yes.
2050 2011-05-05 14:06:52 <jgarzik> 99.9% of current users have an interest in bitcoin-the-currency
2051 2011-05-05 14:06:56 <jgarzik> not data storage
2052 2011-05-05 14:07:12 <UukGoblin> jgarzik, bitcoin-the-currency is data storage
2053 2011-05-05 14:07:17 <xelister> ( )( ) <-- omg cp
2054 2011-05-05 14:07:20 <UukGoblin> every node stores data
2055 2011-05-05 14:07:21 <sipa> it is transaction storage
2056 2011-05-05 14:07:31 <UukGoblin> and transactions are data
2057 2011-05-05 14:07:41 <sipa> and nodes agree to store transactions becomes they find it valuable to have a transaction network
2058 2011-05-05 14:07:44 mologie has quit (Quit: Nettalk6 - www.ntalk.de)
2059 2011-05-05 14:07:54 <sipa> nodes may stop doing that if it's mainly used for other purposes
2060 2011-05-05 14:08:10 <xelister> I can make jgarzik store lots of childporn by one mouseclick right now
2061 2011-05-05 14:08:20 <UukGoblin> it'll never be used for other purposes because each nonstandard transaction will actually be a financial transaction
2062 2011-05-05 14:08:38 <UukGoblin> a fee from the othersystem-user to the block miner.
2063 2011-05-05 14:08:42 <UukGoblin> that's a transaction
2064 2011-05-05 14:08:42 <sipa> but that's not the purpose of those transactions, right?
2065 2011-05-05 14:08:44 <UukGoblin> face it.
2066 2011-05-05 14:08:48 <sipa> i'm not talking about fees
2067 2011-05-05 14:08:57 <sipa> it will require fees, sure, like everything else
2068 2011-05-05 14:09:07 <sipa> but the purpose of the transaction will not be money transfer
2069 2011-05-05 14:09:13 <UukGoblin> it will
2070 2011-05-05 14:09:16 <sipa> *sigh*
2071 2011-05-05 14:09:18 <UukGoblin> not its sole purpose
2072 2011-05-05 14:09:20 <UukGoblin> but it will
2073 2011-05-05 14:09:38 <sipa> if you want timestamping, then timestamping is the purpose
2074 2011-05-05 14:09:46 <jgarzik> regardless of the technical definition (distributed, decentralized, notarized data storage), current users prefer financial uses, and have incentives to want to avoid non-financial transactions in the block chain. It is not as simple as miners having an incentive to receive money for data storage. Miners also have an incentive in seeing bitcoin become a popular, global currency. Data storage is just a parasitic use
2075 2011-05-05 14:09:46 <jgarzik> on top of the financial uses.
2076 2011-05-05 14:09:50 larsivi has quit (Ping timeout: 240 seconds)
2077 2011-05-05 14:09:58 <UukGoblin> well if you want to buy a loaf of bread, then that loaf of bread is the purpose.
2078 2011-05-05 14:10:00 <sipa> the fact that you need to pay for it is a by-product because you want it in the financial block chain
2079 2011-05-05 14:10:17 <sipa> the bitcoin finanacial chain however serves in its own right to process money transfers
2080 2011-05-05 14:10:29 <luke-jr> jgarzik: short of deploying this on Eligius, I don't have a sensible way to test it
2081 2011-05-05 14:10:42 <jgarzik> luke-jr: you'll know immediately if it works or not...
2082 2011-05-05 14:10:53 <jgarzik> luke-jr: because just about every share would get rejected
2083 2011-05-05 14:11:28 <luke-jr> no, every share accepted now would be caught by ==
2084 2011-05-05 14:11:49 <UukGoblin> so sipa and Diablo-D3 - do you both think that allowing for 0-valued outputs is a bug and should be removed?
2085 2011-05-05 14:11:50 <thedrs> hey luke-jr
2086 2011-05-05 14:11:54 <luke-jr> thedrs: ?
2087 2011-05-05 14:12:24 <sipa> UukGoblin: for now, i consider it a bug yes (except in generation tx's)
2088 2011-05-05 14:12:31 <jgarzik> sipa: agreed
2089 2011-05-05 14:12:43 <sipa> unless there is some rationale behind it i am not aware of
2090 2011-05-05 14:12:48 <sipa> but i don't see it
2091 2011-05-05 14:12:53 <UukGoblin> damn.
2092 2011-05-05 14:12:56 <UukGoblin> damn damn damn.
2093 2011-05-05 14:13:26 <UukGoblin> you do realize this will encourage coin burning, right?
2094 2011-05-05 14:13:49 <sipa> i hope it will disencourage people from using the block chain as a data store
2095 2011-05-05 14:14:20 glassresistor has joined
2096 2011-05-05 14:14:22 <UukGoblin> it won't
2097 2011-05-05 14:14:24 glassresistor has quit (Changing host)
2098 2011-05-05 14:14:24 glassresistor has joined
2099 2011-05-05 14:14:32 <UukGoblin> maybe a bit, maybe few, but not all of them
2100 2011-05-05 14:14:32 <gavinandresen> sipa: the 0.01-per-kilobyte fee should discourage people
2101 2011-05-05 14:14:59 <UukGoblin> gavinandresen, ohai, gee, do you come in the right time! thanks! :-D
2102 2011-05-05 14:15:04 <gavinandresen> I agree with UukGoblin: ban 0-value-txouts and people will just find another sneaky way of shoving in data.
2103 2011-05-05 14:15:33 <jgarzik> gavinandresen: true -- but won't we waste time tracking whether or not those are spent?
2104 2011-05-05 14:15:38 <gavinandresen> If not in the TxOuts, they'll just generate a bunch of sends-to-selfs and hide bits in the 'k' param of the ECDSA sigs....
2105 2011-05-05 14:16:02 <jgarzik> gavinandresen: just thinking about it from a database/engineering sense, not in terms of sneaking in data
2106 2011-05-05 14:16:05 <Speeder> there ARE people using blockchains for other things than storing BTC?
2107 2011-05-05 14:16:15 <gavinandresen> jgarzik: optimizing away 0-value txouts would be easy. As would dropping data before an OP_DROP
2108 2011-05-05 14:16:28 <UukGoblin> jgarzik, if you made a special non-standard transaction of arbitrary data, you could very easily stop wasting time trying to validate, because you'll know there's nothing to validate.
2109 2011-05-05 14:16:32 <sipa> gavinandresen: define optimizing away?
2110 2011-05-05 14:16:46 <gavinandresen> ... if storage is a problem. Which it isn't.
2111 2011-05-05 14:16:51 <jgarzik> gavinandresen: IMO OP_DROP is valid, 0-value txout is not
2112 2011-05-05 14:17:27 <gavinandresen> sipa: you tell me-- what's the issue with 0-value txouts-- sucking up disk space?
2113 2011-05-05 14:17:47 <sipa> gavinandresen: sucking up disk space that will never be able to be pruned from the block chain
2114 2011-05-05 14:17:54 <UukGoblin> the 0-value txouts could just be pruned in future.
2115 2011-05-05 14:17:58 <sipa> as opposed to TD's proposal which does allow that
2116 2011-05-05 14:18:10 <jgarzik> gavinandresen: I'm still mulling whether or not a 32-byte opaque field would be useful. People do indeed want to store a hash of an external transaction, with a bitcoin TX.
2117 2011-05-05 14:18:23 <gavinandresen> Yeah, what UukGoblin said. You can certainly prune them, they're all spent.
2118 2011-05-05 14:18:46 <sipa> gavinandresen: if 0-txouts are pruned, they are not useful to his purpose anymore either
2119 2011-05-05 14:18:47 <gavinandresen> jgarzik: satoshi suggested 512 bytes, so SHA-3 hashes will fit.
2120 2011-05-05 14:18:49 <jgarzik> gavinandresen: I don't think UukGoblin meant they could be pruned immediately
2121 2011-05-05 14:19:01 <jgarzik> sipa: exactly
2122 2011-05-05 14:19:04 <UukGoblin> sipa, they are, because they wouldn't be pruned from nodes that want to store them
2123 2011-05-05 14:19:10 <wumpus> sha3 hashes are 4096 bits?
2124 2011-05-05 14:19:21 <sipa> UukGoblin: pruning means that every node can delete that data
2125 2011-05-05 14:19:32 <gavinandresen> Huh? THere's a reason the RPC gettransaction only gets transactions that are in your wallet... (remember that debate?)
2126 2011-05-05 14:19:35 <UukGoblin> sipa, some nodes could choose not to
2127 2011-05-05 14:19:40 <sipa> UukGoblin: of course
2128 2011-05-05 14:19:50 <UukGoblin> sipa, then there's no problem, is there?
2129 2011-05-05 14:19:51 <sipa> UukGoblin: but miners won't be able to do that
2130 2011-05-05 14:20:01 <sipa> they always need the full block chain, except for prunable parts
2131 2011-05-05 14:20:06 <UukGoblin> no, miners won't prune them
2132 2011-05-05 14:20:11 <sipa> *sigh*
2133 2011-05-05 14:20:16 <gavinandresen> Miners don't need 0-value txouts
2134 2011-05-05 14:20:17 <UukGoblin> miners will take the fee and live happily ever after
2135 2011-05-05 14:20:28 <sipa> miners need to be able to verify every transaction
2136 2011-05-05 14:20:33 <sipa> if they do not have the full block chain
2137 2011-05-05 14:20:37 <sipa> they cannot do that
2138 2011-05-05 14:21:26 FabianB_ has joined
2139 2011-05-05 14:21:44 <UukGoblin> sipa, they don't need a full block chain, they can use a pruned block chain if I understand correctly.
2140 2011-05-05 14:21:51 x5x is now known as x5x`brb
2141 2011-05-05 14:21:53 <sipa> yes
2142 2011-05-05 14:22:00 <sipa> and either we decide that 0-txouts are prunable
2143 2011-05-05 14:22:07 <UukGoblin> oh they are
2144 2011-05-05 14:22:13 <sipa> in which case your system that builds upon them break
2145 2011-05-05 14:22:20 <UukGoblin> it doesn't
2146 2011-05-05 14:22:55 <UukGoblin> well it may a bit, actually. But the general concept is still fine.
2147 2011-05-05 14:23:21 <sipa> if the network does not guarantee that your timestamped data that is stored in these 0-txouts will be kept forever
2148 2011-05-05 14:23:24 <sipa> is it still useful for you?
2149 2011-05-05 14:23:25 <UukGoblin> the transaction will be kept until the miner spends the fee, which will generally be pretty logn
2150 2011-05-05 14:23:31 <UukGoblin> maybe not actually
2151 2011-05-05 14:23:45 g007 has joined
2152 2011-05-05 14:23:46 <gavinandresen> My gut tells me that we'd regret disallowing 0-value Txouts. NO_OP / NULL / 0 usually turns out to be very handy
2153 2011-05-05 14:23:57 <UukGoblin> sipa, yes, because I can have an alternative datastore which will keep these transactions
2154 2011-05-05 14:24:19 <sipa> UukGoblin: but you can't guarantee that in the future people will be able to verify your timestamped data
2155 2011-05-05 14:24:23 <luke-jr> jgarzik: ntohl seems to help
2156 2011-05-05 14:24:32 <sipa> since the link with the block chain may be lost
2157 2011-05-05 14:24:36 <UukGoblin> gavinandresen, exactly what I was saying. This would also encourage coin burning because that'd be the only resort to store data in the blockchain. And coin burning is bad.
2158 2011-05-05 14:25:00 <sipa> gavinandresen: i may agree with that
2159 2011-05-05 14:25:06 g007 has quit (Client Quit)
2160 2011-05-05 14:25:23 <sipa> gavinandresen: but is a 0-value txout with a non-empty script useful, as far as you can see?
2161 2011-05-05 14:25:41 <jgarzik> sipa: generic data storage
2162 2011-05-05 14:25:47 <sipa> apart from that :D
2163 2011-05-05 14:26:06 <jgarzik> sipa: I suppose it proves you own something
2164 2011-05-05 14:26:13 <jgarzik> sipa: i.e. control a key
2165 2011-05-05 14:26:19 <sipa> good point
2166 2011-05-05 14:26:20 <gavinandresen> sipa: could be. I dunno, just off the top of my head what if we wanted a transaction-proof-of-work scheme? Use sendmany and generate 0-value txouts until you get a transaction id hash that is less than a target.....
2167 2011-05-05 14:26:25 FabianB has quit (Ping timeout: 276 seconds)
2168 2011-05-05 14:26:35 <gavinandresen> (I don't know why we'd want to do that, but...)
2169 2011-05-05 14:26:58 <sipa> jgarzik: wait, you can do that with an empty scriptPubKey too
2170 2011-05-05 14:27:15 <sipa> the signature on the input of the transaction proves you own an address
2171 2011-05-05 14:27:48 eternal11 has joined
2172 2011-05-05 14:27:50 * luke-jr ponders giving higher priority to transactions with more leading zeros
2173 2011-05-05 14:27:51 <jgarzik> I think 0-value txouts are a bug and potentially waste db space, but that's just a gut feeling
2174 2011-05-05 14:29:10 <UukGoblin> sipa, well, if there's at least one node that stores the timestamped data, then they will be able to verify. And that node can be the sender's node.
2175 2011-05-05 14:29:19 <jgarzik> gavinandresen: script.cpp already limits an individual stack item to 520 bytes, so SHA-3 hash w/ OP_DROP would work with existing script engine
2176 2011-05-05 14:29:33 <jgarzik> and older clients
2177 2011-05-05 14:29:38 eternal1 has quit (Ping timeout: 240 seconds)
2178 2011-05-05 14:29:58 <UukGoblin> sipa, the sender's node will have to store the hash (obviously) + one branch of the merkle tree
2179 2011-05-05 14:30:14 <gavinandresen> jgarzik: perfect. Satoshi thinking ahead again probably...
2180 2011-05-05 14:30:21 <jgarzik> #define OP_MEMO OP_DROP
2181 2011-05-05 14:31:05 <luke-jr> except that the blocks themselves are still SHA-2 :p
2182 2011-05-05 14:31:31 redengin has joined
2183 2011-05-05 14:31:42 <BlueMatt> so satoshi planned OP_DROP as a way to add a memo to a tx
2184 2011-05-05 14:32:00 <jgarzik> well I wouldn't go that far
2185 2011-05-05 14:32:09 <UukGoblin> btw, short notes with transactions would be pretty useful to have, there were users asking about this
2186 2011-05-05 14:32:21 <jgarzik> satoshi didn't really want non-financial data in block chain, and I agree w/ that
2187 2011-05-05 14:32:28 <jgarzik> but I do see the need for a memo field
2188 2011-05-05 14:32:30 <BlueMatt> problem is when people start adding crap to the memos
2189 2011-05-05 14:32:38 <noagendamarket> yea
2190 2011-05-05 14:32:40 <UukGoblin> all the memos can be pruned
2191 2011-05-05 14:32:45 <BlueMatt> like kiddieporn and such
2192 2011-05-05 14:32:51 <noagendamarket> you get the twitter dm problem
2193 2011-05-05 14:32:57 <jgarzik> ship a sha256(order data) for merchant, into the block chain
2194 2011-05-05 14:33:17 NOTAL has joined
2195 2011-05-05 14:33:31 NOTAL has quit (Remote host closed the connection)
2196 2011-05-05 14:33:35 <UukGoblin> memos can be kept by the network's blockchain for a short time and then pruned
2197 2011-05-05 14:33:48 <UukGoblin> sender and receiver will keep their memos indefinitely, along with a merkle branch
2198 2011-05-05 14:33:57 <UukGoblin> (unless of course they don't want to)
2199 2011-05-05 14:33:59 NOTAL has joined
2200 2011-05-05 14:34:06 <BlueMatt> if you are concerned about storage, much easier to implement the tx pruning already planned
2201 2011-05-05 14:34:12 <BlueMatt> (well maybe)
2202 2011-05-05 14:34:20 <BlueMatt> not really planned, but easily possible
2203 2011-05-05 14:34:29 <jgarzik> well you have the "eat at joe's" problem
2204 2011-05-05 14:34:38 <jgarzik> spend 0.01 to store your plaintext message for all time
2205 2011-05-05 14:34:39 <BlueMatt> ?
2206 2011-05-05 14:34:43 <BlueMatt> oh
2207 2011-05-05 14:34:57 <CIA-30> bitcoin: Luke Dashjr personal * rda1471417bf2 pushpool-personal/msg.c: Allow miners to change timestamp within reason http://tinyurl.com/62zw2q4
2208 2011-05-05 14:34:58 <UukGoblin> and if they are legally in the block chain, using a kind of OP_MEMO instead of stuff smuggled in as an address, they'd be even easier to prune.
2209 2011-05-05 14:34:59 <jgarzik> "Visit jgarzikrocks.com"
2210 2011-05-05 14:35:44 <UukGoblin> jgarzik, well you can't really stop that, especially if someone's willing to burn coins
2211 2011-05-05 14:35:56 <UukGoblin> meh, full circle now I think
2212 2011-05-05 14:36:08 <UukGoblin> (if not a third full circle)
2213 2011-05-05 14:36:16 <jgarzik> luke-jr: verified it works, if you #if-0 the test 'timestampRcvd == timestampSent'?
2214 2011-05-05 14:36:37 <jgarzik> UukGoblin: right now you won't really have more than a couple plaintext characters per TX
2215 2011-05-05 14:36:47 <jgarzik> UukGoblin: so it's not feasible or cheap or visible right now
2216 2011-05-05 14:37:05 <UukGoblin> jgarzik, hey, that's what I've been saying all along... make it expensive in tx fees, but not impossible
2217 2011-05-05 14:37:20 <gavinandresen> jgarzik: I think the benefits will outweigh the costs. refund and initiate-a-key-exchange-for-private-customer-support and other nifty future features become easier to implement.
2218 2011-05-05 14:37:24 <UukGoblin> it's MUCH BETTER to pay a tx fee to have your data stored than it is to burn the coins forever
2219 2011-05-05 14:37:28 Beffjaker has quit (Ping timeout: 252 seconds)
2220 2011-05-05 14:37:36 <BlueMatt> how visible is it, really, if you add a memo to your tx? its shown only to users of blockexplorer and the tx recipient?
2221 2011-05-05 14:37:58 <jgarzik> the kiddie-porn-in-chain problem is really tough.
2222 2011-05-05 14:38:07 <jgarzik> what happens when a judge issues an order :/
2223 2011-05-05 14:38:09 <UukGoblin> BlueMatt, depends what memo you send. It could be encrypted with the recipient's pubkey
2224 2011-05-05 14:38:13 <lulzplzkthx> what jgarzik ?
2225 2011-05-05 14:38:25 <lulzplzkthx> data in a transaction?
2226 2011-05-05 14:38:28 <UukGoblin> jgarzik, then we can prune the offending txs manually, perhaps?
2227 2011-05-05 14:38:28 <gavinandresen> You delete those transactions.
2228 2011-05-05 14:38:32 <lulzplzkthx> encrypt the data.
2229 2011-05-05 14:38:40 <jgarzik> I definitely see plenty of good uses
2230 2011-05-05 14:38:42 <BlueMatt> lulzplzkthx: you put kiddie porn in the chain and a judge says delete all copies of bitcoin and the chain
2231 2011-05-05 14:38:43 <jgarzik> 100% agreed there
2232 2011-05-05 14:38:47 <jgarzik> I've long wanted a memo field
2233 2011-05-05 14:38:53 <jgarzik> for a external-tx hash
2234 2011-05-05 14:38:56 <lulzplzkthx> BlueMatt: But if it's encrypted with the public/private, then what?
2235 2011-05-05 14:39:01 <lulzplzkthx> Encrypt everything, like Tor does.
2236 2011-05-05 14:39:08 <BlueMatt> lulzplzkthx: what if some idiot doesnt bother to encrypt it?
2237 2011-05-05 14:39:18 <lulzplzkthx> BlueMatt: Then don't see it as a valid transaction.
2238 2011-05-05 14:39:28 <UukGoblin> in UK you have to give up an encryption key
2239 2011-05-05 14:39:30 <lulzplzkthx> The client should encrypt the message w/ the same key it encrypts the Bitcoin.
2240 2011-05-05 14:39:31 <BlueMatt> if this person hates bitcoin and wants a judge to force everyone to delete the chain...
2241 2011-05-05 14:39:39 <sipa> just a question: are 0-txouts spendable?
2242 2011-05-05 14:39:40 <jgarzik> hmmmm
2243 2011-05-05 14:39:49 <UukGoblin> BlueMatt, that'd be like saying to destroy all computers and the internet because one has kiddie porn on it
2244 2011-05-05 14:39:54 <lulzplzkthx> BlueMatt: But if it's not seen as valid, it won't get into the chain.
2245 2011-05-05 14:40:01 <BlueMatt> UukGoblin: but a judge wouldnt...
2246 2011-05-05 14:40:14 <BlueMatt> lulzplzkthx: it is valid if you put it in a "memo" field
2247 2011-05-05 14:40:24 <lulzplzkthx> BlueMatt: So encrypt what's put in the memo field.
2248 2011-05-05 14:40:33 <lulzplzkthx> That should be part of the Bitcoin client, to encrypt that.
2249 2011-05-05 14:40:59 <jgarzik> lulzplzkthx: you cannot control what a malicious attacker puts into the memo field
2250 2011-05-05 14:41:10 <lulzplzkthx> jgarzik: But you can control what a normal client accepts into the block chain.
2251 2011-05-05 14:41:24 <sipa> lulzplzkthx: you cannot distinguish encrypted and unencrypted data
2252 2011-05-05 14:41:25 eternal11 has quit (Ping timeout: 252 seconds)
2253 2011-05-05 14:41:28 <UukGoblin> lulzplzkthx, you can't check if a message is encrypted
2254 2011-05-05 14:41:43 <BlueMatt> lulzplzkthx: I hate bitcoin and love kiddie porn, so I put it in the memo field unencrypted and tell a judge that bitcoin is storing kiddie porn...
2255 2011-05-05 14:42:01 <UukGoblin> lulzplzkthx, then we kill you and prune the kiddie porn
2256 2011-05-05 14:42:07 <lulzplzkthx> BlueMatt: I understand the issue, you're ignoring the solution which sipa and UukGoblin have responded to.
2257 2011-05-05 14:42:16 <lulzplzkthx> Hmm.
2258 2011-05-05 14:42:22 <luke-jr> jgarzik: eyeball-verified.
2259 2011-05-05 14:42:27 <UukGoblin> s/lulzplzkthx/BlueMatt/ ;-]
2260 2011-05-05 14:42:27 <lulzplzkthx> Max length of memo that can't fit kiddie porn?
2261 2011-05-05 14:42:34 <jgarzik> luke-jr: cool
2262 2011-05-05 14:42:38 <jgarzik> luke-jr: wanna email a patch?
2263 2011-05-05 14:42:46 <UukGoblin> lulzplzkthx, multiple memos in multiple transactions
2264 2011-05-05 14:42:51 <BlueMatt> lulzplzkthx: If I have 1 BTC, I can split it and put it in 100 txes
2265 2011-05-05 14:42:57 <lulzplzkthx> Hmm.
2266 2011-05-05 14:43:01 <luke-jr> jgarzik: why not just pull? then you get the mysql-without-password too :P
2267 2011-05-05 14:43:08 <gavinandresen> lulzplzkthx: encrypt the memo with what key?
2268 2011-05-05 14:43:14 <jgarzik> luke-jr: pull is fine, too. git url and branch?
2269 2011-05-05 14:43:16 <lulzplzkthx> gavinandresen: The same one the bitcoin's encrypted with.
2270 2011-05-05 14:43:21 eternal1 has joined
2271 2011-05-05 14:43:31 <gavinandresen> lulzplzkthx: bitcoins aren't encrypted. They're signed.
2272 2011-05-05 14:43:32 <sipa> seems 0-txouts are spendable, as far as i can see
2273 2011-05-05 14:43:37 <BlueMatt> lulzplzkthx: its ecdsa not rsa, you cant encrypt with dsa
2274 2011-05-05 14:43:45 <lulzplzkthx> Okay. I don't know then. :P
2275 2011-05-05 14:43:54 <sipa> you can encrypt with an algorithm that uses the same sort of keys though
2276 2011-05-05 14:44:10 <lulzplzkthx> sipa: Is that the whole multiplying thing?
2277 2011-05-05 14:44:11 <UukGoblin> you can ask the recipient of bitcoins for a public key along with his address
2278 2011-05-05 14:44:13 <luke-jr> jgarzik: git://gitorious.org/~Luke-Jr/bitcoin/luke-jrs-pushpool.git personal
2279 2011-05-05 14:44:14 <gavinandresen> Sure, but the sender and receiver need to agree on an encryption key somehow
2280 2011-05-05 14:44:25 <UukGoblin> and encrypt to his pubkey with GPG
2281 2011-05-05 14:44:30 * lulzplzkthx is talking shit and will stop now.
2282 2011-05-05 14:44:58 <UukGoblin> gavinandresen, again, I can't thank you enough for dropping into the discussion at a perfect time :-]
2283 2011-05-05 14:45:07 <luke-jr> jgarzik: any idea why ufasoft hates pushpool?
2284 2011-05-05 14:45:09 <UukGoblin> gavinandresen, damn, I start feeling I should donate ;-]
2285 2011-05-05 14:45:10 <BlueMatt> sipa: there is an algo for encryption with ecdsa keys?
2286 2011-05-05 14:45:38 <gavinandresen> UukGoblin: the faucet is getting low again if you can't resist the urge to donate
2287 2011-05-05 14:45:45 <jgarzik> luke-jr: yeah -- pushpool needs to handle the lack of authorization by sending back some special http reply code, that triggers a standard http retry w/ auth
2288 2011-05-05 14:45:58 <luke-jr> jgarzik: special how? it already sends 401 O.o
2289 2011-05-05 14:46:09 DD- has left ()
2290 2011-05-05 14:46:19 <sipa> BlueMatt: ecdsa uses ec keys, and ec keys can be used in this: http://en.wikipedia.org/wiki/Integrated_Encryption_Scheme
2291 2011-05-05 14:46:28 <luke-jr> ah
2292 2011-05-05 14:46:35 <luke-jr> The response MUST include a WWW-Authenticate header field (section 14.47) containing a challenge applicable to the requested resource.
2293 2011-05-05 14:46:39 <UukGoblin> gavinandresen, added 1 more BTC
2294 2011-05-05 14:46:46 <BlueMatt> sipa: hm..., well maybe lulzplzkthx's idea is pretty good...
2295 2011-05-05 14:46:52 <gavinandresen> sipa: but ordinary bitcoin transactions don't use the full ECDSA public key
2296 2011-05-05 14:46:56 <BlueMatt> sipa: as a sidenote, how the f do you reopen a pull request?
2297 2011-05-05 14:46:56 <gavinandresen> UukGoblin: thanks!
2298 2011-05-05 14:47:27 <sipa> gavinandresen: indeed, you'd need the full pubkey instead of just the address...
2299 2011-05-05 14:48:05 <lulzplzkthx> ;;bc,mtgox
2300 2011-05-05 14:48:05 <gribble> {"ticker":{"high":3.5699,"low":3.4001,"vol":11955,"buy":3.4211,"sell":3.445,"last":3.4211}}
2301 2011-05-05 14:48:11 <BlueMatt> ok, so I take that back...
2302 2011-05-05 14:48:19 <sipa> BlueMatt: i clicked 'reopen' :p
2303 2011-05-05 14:48:46 <BlueMatt> where the hell is that button?
2304 2011-05-05 14:49:15 <lulzplzkthx> Any devs who can deal with it: There's a bug in my code for balance in the tooltip. The balance won't update until the UI is opened again.
2305 2011-05-05 14:49:22 <lulzplzkthx> I tried a couple ways to fix it, but couldn't.
2306 2011-05-05 14:49:29 <sipa> BlueMatt: maybe eh... not everyone can do that
2307 2011-05-05 14:50:03 <BlueMatt> must be a comitter feature
2308 2011-05-05 14:50:12 <sipa> BlueMatt: but if you closed it yourself, i expect you to be able to reopen yourself
2309 2011-05-05 14:50:44 <gavinandresen> lulzplzkthx: Can you please submit a patch to remove that functionality until somebody figures out how to make it work?
2310 2011-05-05 14:50:49 <BlueMatt> When I accidentally push removing anything useful, automatically closing the commit I cant reopen
2311 2011-05-05 14:51:34 <lulzplzkthx> Sure gavinandresen.
2312 2011-05-05 14:51:37 <jgarzik> luke-jr: pulled
2313 2011-05-05 14:51:43 <gavinandresen> lulzplzkthx: thanks
2314 2011-05-05 14:52:10 <BlueMatt> gavinandresen: is that bug such a big deal that the entire feature should be removed?
2315 2011-05-05 14:52:21 <jgarzik> luke-jr: patches/pulls to fix ufasoft accepted :)
2316 2011-05-05 14:53:23 <gavinandresen> BlueMatt: I'd rather not add features that are "broken but kinda nice when they work"
2317 2011-05-05 14:54:10 <jgarzik> heh.
2318 2011-05-05 14:54:22 <jgarzik> consider a tx invalid if OP_DROP contains too much readable ASCII
2319 2011-05-05 14:54:25 <BlueMatt> lulzplzkthx: oh, you mean the balance is stuck at x until the entire client is closed/opened or ?
2320 2011-05-05 14:54:27 <jgarzik> valid, otherwise.
2321 2011-05-05 14:54:31 <sipa> jgarzik: lol!
2322 2011-05-05 14:54:40 <BlueMatt> jgarzik: and when I submit a jpeg?
2323 2011-05-05 14:54:44 <jgarzik> BlueMatt: that's fine
2324 2011-05-05 14:54:45 diki has joined
2325 2011-05-05 14:54:54 <diki> how long ago was the current difficulty set?
2326 2011-05-05 14:55:00 <UukGoblin> jgarzik, I think that's what Diablo-D3 was proposing. Mental.
2327 2011-05-05 14:55:01 bgeron has quit (Ping timeout: 276 seconds)
2328 2011-05-05 14:55:08 <gavinandresen> speaking of kinda broken.... jgarzik: can you finish removing the generate options from the GUI? (or is somebody else working on that)?
2329 2011-05-05 14:55:12 <UukGoblin> diki, ;;bc,stats
2330 2011-05-05 14:55:16 <BlueMatt> jgarzik: heh, and I think you have it around the other way
2331 2011-05-05 14:55:18 <diki> Diablo..could you be the one that made the cpu miner?
2332 2011-05-05 14:55:21 <BlueMatt> gavinandresen: I have a pull...
2333 2011-05-05 14:55:23 <diki> diablo miner
2334 2011-05-05 14:55:31 <gavinandresen> BlueMatt: great
2335 2011-05-05 14:55:39 <jgarzik> gavinandresen: BlueMatt took care of it, I ACK'd the pull
2336 2011-05-05 14:55:49 <lulzplzkthx> Submited a pull request.
2337 2011-05-05 14:56:00 <UukGoblin> that's a nice controversial change ;-]
2338 2011-05-05 14:56:02 <CIA-30> bitcoin: Matt Corallo master * r3451c5a / (ui.cpp ui.h uibase.cpp uibase.h uiproject.fbp): Remove the option to generate coins from the Options Dialog. - http://bit.ly/mcjiZW
2339 2011-05-05 14:56:09 <CIA-30> bitcoin: Gavin Andresen master * r5f56b28 / (ui.cpp ui.h uibase.cpp uibase.h uiproject.fbp):
2340 2011-05-05 14:56:10 <CIA-30> bitcoin: Merge pull request #193 from TheBlueMatt/nogenoption
2341 2011-05-05 14:56:10 <CIA-30> bitcoin: Remove the option to generate coins from the Options Dialog. - http://bit.ly/k91vUr
2342 2011-05-05 14:56:10 <lulzplzkthx> BlueMatt: The balance in the tooltip stays the same until the UI is displayed.
2343 2011-05-05 14:56:19 <lulzplzkthx> So if I double click on the UI in the taskbar, it will update the balance tooltip.
2344 2011-05-05 14:56:30 <diki> Goblin, that doesnt say when the current difficulty was set
2345 2011-05-05 14:56:40 <lulzplzkthx> But if it sits in the tray, and then it receives 10 BTC, I won't know until I display the UI.
2346 2011-05-05 14:56:46 <BlueMatt> lulzplzkthx: ah, ok...I wish we had a real wxdev here
2347 2011-05-05 14:56:52 <lulzplzkthx> :P yeah.
2348 2011-05-05 14:57:05 <jgarzik> gavinandresen: in general, 'ACK' is shorthand, in Linux kernel land, for indicating that it is OK to merge a commit, and that the approver has fully reviewed the commit (perhaps even tested it, if you're really lucky :))
2349 2011-05-05 14:57:11 <lulzplzkthx> Well, I could even make it work if I could figure out a way to update on transaction receiving, but I'm not very experienced with C++.
2350 2011-05-05 14:57:17 <lulzplzkthx> So I'm not sure how I'd go cross class with it.
2351 2011-05-05 14:57:22 <jgarzik> gavinandresen: so, you'll see that in bitcoin github, FYI
2352 2011-05-05 14:57:29 <gavinandresen> jgarzik: what does ICK mean in kernel-land?
2353 2011-05-05 14:57:33 <lulzplzkthx> Nor do I know wx. Qt would be my GUI framework of choice
2354 2011-05-05 14:57:38 <UukGoblin> ;;bc,stats
2355 2011-05-05 14:57:40 <gribble> Current Blocks: 122038 | Current Difficulty: 109670.13329248 | Next Difficulty At Block: 122975 | Next Difficulty In: 937 blocks | Next Difficulty In About: 4 days, 18 hours, 15 minutes, and 43 seconds | Next Difficulty Estimate: 149529.41201686
2356 2011-05-05 14:57:43 <sipa> jgarzik: you also have NAK? :)
2357 2011-05-05 14:57:57 <UukGoblin> diki, ah, damn.
2358 2011-05-05 14:58:00 <jgarzik> gavinandresen: yes, it's NAK -- hell no
2359 2011-05-05 14:58:03 <jgarzik> sipa: ^^
2360 2011-05-05 14:58:22 <UukGoblin> diki, you could dig it up with some maths (subtraction) and blockexplorer to get you a time
2361 2011-05-05 14:58:26 <jgarzik> aggressive deleting of email responses is 'abstain'
2362 2011-05-05 14:58:35 <diki> i mentioned before i am bad with math
2363 2011-05-05 14:58:38 <UukGoblin> actually might require some multiplication and addition too
2364 2011-05-05 14:59:01 <sipa> jgarzik, gavinandresen: what do you think about transactions that conflict with the block chain?
2365 2011-05-05 14:59:21 <TD> good day, folks
2366 2011-05-05 14:59:29 <UukGoblin> hi TD :-]
2367 2011-05-05 14:59:32 <BlueMatt> TD: hi
2368 2011-05-05 14:59:47 <BlueMatt> it would appear that the gang's all here
2369 2011-05-05 14:59:47 <diki> May i ask who is updating bitcoin program(that manages the transactions etc) and what happened to satoshi?
2370 2011-05-05 14:59:52 <gavinandresen> sipa: you mean like accidental double-spends?
2371 2011-05-05 14:59:54 <CIA-30> bitcoin: various * rf686c4..945a26 pushpool/ (msg.c db-sqlite.c server.c db-mysql.c): (6 commits) http://tinyurl.com/3k2tc7h
2372 2011-05-05 14:59:55 <lulzplzkthx> hello TD
2373 2011-05-05 14:59:56 <sipa> gavinandresen: yes
2374 2011-05-05 15:00:10 <sipa> and other weird things that may leave you with a corrupted wallet
2375 2011-05-05 15:00:13 * TD is implementing handling of that in bitcoinj at the moment
2376 2011-05-05 15:00:17 <TD> it's a headache, that's for sure
2377 2011-05-05 15:00:19 <UukGoblin> diki, gavinandresen is doing work on the client now, and https://en.bitcoin.it/wiki/Satoshi
2378 2011-05-05 15:00:22 <BlueMatt> diki: gavinandresen, sipa, jgarzik for the most part. But check the github log to see
2379 2011-05-05 15:00:22 <gavinandresen> sipa: I think they are a very small problem for a very small number of people, and probably not worth a lot of engineering effort.
2380 2011-05-05 15:00:31 <lulzplzkthx> gavin, pull #194 for the removal of the balance in tooltip
2381 2011-05-05 15:00:36 <TD> i'm thinking of introducing some concept of a "dead" transaction into the API, that is, a tx that is pending but is guaranteed to never confirm unless there's a re-org
2382 2011-05-05 15:00:39 <gavinandresen> lulzplzkthx: thanks
2383 2011-05-05 15:00:41 <sipa> gavinandresen: i have a working patch for that
2384 2011-05-05 15:00:51 <TD> not so much because it's useful to have in the api but just because it gets confusing to test otherwise :)
2385 2011-05-05 15:01:03 <sipa> gavinandresen: but it simply greys out the transactions in the gui
2386 2011-05-05 15:01:14 <sipa> i'm not sure what should be done gui/interfacewise with them
2387 2011-05-05 15:01:33 <BlueMatt> sipa: mark them rejected?
2388 2011-05-05 15:02:12 <sipa> yes, but things like eg. the ability to remove them from your wallet
2389 2011-05-05 15:02:18 <sipa> or trying to recreate them
2390 2011-05-05 15:02:32 <UukGoblin> gavinandresen, I know there's a lot of bugs waiting, but did you see my "configurable transaction fees" idea at http://www.bitcoin.org/smf/index.php?topic=7240.0 ?
2391 2011-05-05 15:02:42 <TD> probably just mark them as red for now or something
2392 2011-05-05 15:02:48 <TD> it should be really rare if the rest works
2393 2011-05-05 15:03:05 skyewm has joined
2394 2011-05-05 15:03:11 <BlueMatt> I dont think there is any reason to try to remove them (too much effort for no real gain), but maybe some kind of resend button in the double-click-dialog
2395 2011-05-05 15:03:18 <cosurgi> ArtForz: this is segfaulting for my 5870, any idea? ADL_Overdrive5_ODPerformanceLevels_Set(idx, perfLevels))
2396 2011-05-05 15:03:36 <sipa> gavinandresen: i agree that it's only a problem for a small group of people, but for those, it can be very painful - it could result in a wallet that is unable to do any valid transactions anymore
2397 2011-05-05 15:03:49 <xelister> ArtForz: I can set min 500 using AMDoverdrive, any idea why not 300? need to program bios for that?
2398 2011-05-05 15:04:03 <cosurgi> ArtForz: when 5870 is only a single GPU it works. But on another PC there are three cards: 5850,5870,5870. And there it segfaults
2399 2011-05-05 15:04:10 <BlueMatt> sipa: in general, "rejected" is probably fine until its actually a common problem (or feature request)
2400 2011-05-05 15:04:18 <sipa> good point
2401 2011-05-05 15:04:38 <BlueMatt> IMHO of course
2402 2011-05-05 15:05:19 mologie has joined
2403 2011-05-05 15:05:38 <gavinandresen> sipa, TD: I'd vote for doing what's pretty easy, and if somebody has a wallet that is messed up they should use something like bitcointools to reset.
2404 2011-05-05 15:06:56 <gavinandresen> sipa: actually, a wallet backup function that just exported private keys would fix it, right? Export just private keys, then use that as your wallet (with auto-rescan) and bad transactions go away.
2405 2011-05-05 15:06:56 <BlueMatt> gavinandresen: you mentioned you would rather see bugs fixed than new features...IMHO this is a bug that needs fixed?
2406 2011-05-05 15:07:17 <sipa> gavinandresen: all non-confirmed transactions would go away
2407 2011-05-05 15:07:22 <sipa> but, yes
2408 2011-05-05 15:07:56 <luke-jr> jgarzik: why is the auth code copied in two places? -.-
2409 2011-05-05 15:07:57 <sipa> gavinandresen: but as i said, it works already, but it's a rather large change
2410 2011-05-05 15:08:02 <gavinandresen> BlueMatt: is what a bug? rejected transactions hanging around?
2411 2011-05-05 15:08:15 <BlueMatt> gavinandresen: them not displaying in a way that is noob-friendly
2412 2011-05-05 15:08:19 skyewm has quit (Ping timeout: 246 seconds)
2413 2011-05-05 15:08:30 danlucraft has quit (Read error: No route to host)
2414 2011-05-05 15:08:31 <gavinandresen> BlueMatt: noobs don't run into the problem, so no.
2415 2011-05-05 15:08:49 <gavinandresen> You have to work fairly hard to generate rejected transactions.
2416 2011-05-05 15:08:53 <BlueMatt> gavinandresen: noobs do stupid things (like copying their wallet and spending in multiple places)
2417 2011-05-05 15:08:53 danlucraft has joined
2418 2011-05-05 15:09:04 <BlueMatt> (at the same time)
2419 2011-05-05 15:09:12 <jgarzik> gavinandresen: well, except an under-fee'd TX that never confirms
2420 2011-05-05 15:09:14 <sipa> just copy a wallet and do two transactions within the same block, while one of both wallets is offline
2421 2011-05-05 15:09:33 <gavinandresen> BlueMatt: ... and sipa's latest code should do the right thing 99.99% of the time. You'd have to work hard to double-spend in two places quick enough....
2422 2011-05-05 15:09:36 <jgarzik> gavinandresen: that remains a problem that Aunt Tillie might run into, and have difficulty recovering from
2423 2011-05-05 15:09:36 <sipa> and if walletdump/import gets included in the main client, that will be even more commonplace
2424 2011-05-05 15:09:38 <diki> I have a question I have two PCs currently running, this one with a Phenom 955 and 5850. I am using the Phenom 955 with 3 threads and 3 cores to mine as well as the GPU. I am also using the other(older) pc's CPU which Pentium 4 to mine. 5850 gives me 340mhash/s, the phenom gives me 9 and the pentium 0,600mhash/s
2425 2011-05-05 15:09:54 <CIA-30> bitcoin: various op-drop * r8bfacd..66fb32 bitcoind-personal/ (31 files in 10 dirs): (38 commits) http://tinyurl.com/4yufspj
2426 2011-05-05 15:09:54 <gavinandresen> jgarzik: latest code makes it hard to under-fee
2427 2011-05-05 15:10:01 <jgarzik> yeah
2428 2011-05-05 15:10:01 <CIA-30> bitcoin: Luke Dashjr personal * r7344c120bbf5 pushpool-personal/server.c: Include mandatory WWW-Authenticate header in 401 replies (fixes Ufasoft miner) http://tinyurl.com/43nnqnk
2429 2011-05-05 15:10:04 <diki> The question is, do they help each other without sending shares?
2430 2011-05-05 15:10:32 <jgarzik> btw: git://github.com/jgarzik/bitcoin.git op-drop
2431 2011-05-05 15:10:50 sp1ky has joined
2432 2011-05-05 15:11:01 <jgarzik> does the obvious, supports relaying <= 520 byte OP_DROP
2433 2011-05-05 15:11:41 <Diablo-D3> [10:06:53] <UukGoblin> so sipa and Diablo-D3 - do you both think that allowing for 0-valued outputs is a bug and should be removed?
2434 2011-05-05 15:11:44 <Diablo-D3> well
2435 2011-05-05 15:11:46 <Diablo-D3> I dunno
2436 2011-05-05 15:11:54 <cosurgi> jgarzik: Aunt Tillie also needs wallet encryption, or first stupid virus will steal all her money.
2437 2011-05-05 15:11:57 <Diablo-D3> it just allows for too much abuse
2438 2011-05-05 15:12:04 <Diablo-D3> UukGoblin: btw, you still have to pay fees on it, right?
2439 2011-05-05 15:12:13 <jgarzik> cosurgi: that's why I wrote the #crypter branch, that adds wallet encryption
2440 2011-05-05 15:12:27 <diki> i fail to understand how in this day and age people still talk in bytes.
2441 2011-05-05 15:12:41 * jgarzik talks in sentences
2442 2011-05-05 15:12:43 <Diablo-D3> diki: because we dont
2443 2011-05-05 15:12:45 <diki> for a long time, i've been talking in terabytes
2444 2011-05-05 15:12:46 <Diablo-D3> I talk in ints often
2445 2011-05-05 15:12:50 <TD> jgarzik: what's the use case for putting half a kilobyte of data in the scriptsig?
2446 2011-05-05 15:13:00 <luke-jr> jgarzik: might want to merge that last commit too; tested against Ufasoft successfully
2447 2011-05-05 15:13:13 <BlueMatt> TD: the whole OP_DROP/OP_MEMO thing
2448 2011-05-05 15:13:46 <sipa> in the scriptSig? i'd expect it to be in the scriptPubKey
2449 2011-05-05 15:14:06 eao has quit (Read error: Connection reset by peer)
2450 2011-05-05 15:14:10 <Diablo-D3> luke-jr: did you add time changing to LP?
2451 2011-05-05 15:14:20 <jgarzik> luke-jr: same branch, not rebased?
2452 2011-05-05 15:14:20 <luke-jr> Diablo-D3: english?
2453 2011-05-05 15:14:27 <luke-jr> jgarzik: correct
2454 2011-05-05 15:14:28 <UukGoblin> Diablo-D3, yes, you'd pay fees as normal or more
2455 2011-05-05 15:14:30 <Diablo-D3> miner-side increase of the time field
2456 2011-05-05 15:14:35 <Diablo-D3> UukGoblin: meh, maybe not so bad
2457 2011-05-05 15:14:38 diki has quit ()
2458 2011-05-05 15:14:57 <luke-jr> jgarzik: shall I rebase?
2459 2011-05-05 15:15:16 <UukGoblin> ok, so p2p dns implementation step 1: allow for a memo field in bitcoin transactions
2460 2011-05-05 15:15:28 <jgarzik> TD: it's a strawman patch, designed to provoke discussion :) but to answer your question specifically, in the context of adding a merchant-friendly "memo" field to each transaction, <gavinandresen> jgarzik: satoshi suggested 512 bytes, so SHA-3 hashes will fit.
2461 2011-05-05 15:15:45 <TD> and what goes in the memo?
2462 2011-05-05 15:15:49 <jgarzik> luke-jr: that would be nice, but not necessary
2463 2011-05-05 15:15:59 <luke-jr> jgarzik: done
2464 2011-05-05 15:16:00 <jgarzik> TD: a hash of a merchant-specific order id
2465 2011-05-05 15:16:06 <BlueMatt> TD: what goes on the memo line of a check?
2466 2011-05-05 15:16:14 <jgarzik> TD: hash of something tied to the external system
2467 2011-05-05 15:16:26 <TD> why is that necessary when the merchant already has the address they generated?
2468 2011-05-05 15:16:34 <gavinandresen> TD: or really really tiny child porn
2469 2011-05-05 15:16:38 <TD> lol
2470 2011-05-05 15:16:41 <BlueMatt> I dont think its useful for merchant order ids
2471 2011-05-05 15:16:46 <TD> children are tiny!
2472 2011-05-05 15:16:52 <BlueMatt> oh god
2473 2011-05-05 15:16:59 <gavinandresen> sorry! sorry!
2474 2011-05-05 15:17:16 <TD> merchants already have a unique transaction id - they have to, it's practically a requirement of bitcoin already.
2475 2011-05-05 15:17:35 <TD> they can link that with whatever data they like (login name, delivery address, whatever)
2476 2011-05-05 15:17:35 <jgarzik> TD: the merchant may require signed data to initiate the TX. it's useful to tie signed data to a TX.
2477 2011-05-05 15:17:55 <Diablo-D3> [11:10:19] <UukGoblin> ok, so p2p dns implementation step 1: allow for a memo field in bitcoin transactions
2478 2011-05-05 15:17:58 <Diablo-D3> now that I'd want
2479 2011-05-05 15:18:26 <Diablo-D3> UukGoblin: cant you encrypt things knowing the target public key?
2480 2011-05-05 15:18:38 <BlueMatt> Whether its useful for merchant order ids or not, its still very useful for other memos on txes
2481 2011-05-05 15:18:47 <Diablo-D3> its useful for humans
2482 2011-05-05 15:18:50 <Diablo-D3> its the memo field on a check
2483 2011-05-05 15:19:02 <TD> is this something US specific? i've never seen a cheque with a memo field
2484 2011-05-05 15:19:09 <Diablo-D3> TD: not really
2485 2011-05-05 15:19:18 <jgarzik> TD: yeah, all US checks have a memo field, in my experience
2486 2011-05-05 15:19:27 <jgarzik> luke-jr: pulled, thanks again
2487 2011-05-05 15:19:29 <Diablo-D3> TD: obviously commercial checks dont have a memofield
2488 2011-05-05 15:19:33 <UukGoblin> all .eu SEPA transfers have an option of a memo field
2489 2011-05-05 15:19:34 <BlueMatt> TD: http://byrdauto.com/ESW/Images/4write_a_check_step4_writing.jpg?xcache=1308
2490 2011-05-05 15:19:39 <Diablo-D3> its just a little space in the corner that allows you to write what its for
2491 2011-05-05 15:19:40 <TD> well, i can see why it'd be necessary. you can't create account numbers at will with cheques
2492 2011-05-05 15:19:50 <Diablo-D3> TD: no
2493 2011-05-05 15:19:52 <Diablo-D3> like for example
2494 2011-05-05 15:19:54 <Diablo-D3> lets say Im a guy
2495 2011-05-05 15:19:55 <CIA-30> bitcoin: Luke Dashjr * ree98e0fd6204 pushpool/server.c: Include mandatory WWW-Authenticate header in 401 replies (fixes Ufasoft miner) http://tinyurl.com/3sbvdzz
2496 2011-05-05 15:19:56 <CIA-30> bitcoin: Jeff Garzik personal * rf686c43e2041 pushpool-personal/ (db-mysql.c db-sqlite.c msg.c server.c): Merge branch 'personal' of git://gitorious.org/~Luke-Jr/bitcoin/luke-jrs-pushpool into tmp http://tinyurl.com/3ksch82
2497 2011-05-05 15:19:58 <CIA-30> bitcoin: Luke Dashjr personal * ree98e0fd6204 pushpool-personal/server.c: Include mandatory WWW-Authenticate header in 401 replies (fixes Ufasoft miner) http://tinyurl.com/3lr6uu5
2498 2011-05-05 15:20:00 <Diablo-D3> and my former gf is a fucking bitch
2499 2011-05-05 15:20:01 <CIA-30> bitcoin: Jeff Garzik personal * rfa8e5cd98541 pushpool-personal/README: Better document dependencies. http://tinyurl.com/3nx3pkg
2500 2011-05-05 15:20:05 <Diablo-D3> and shes suing me for not repaying a check
2501 2011-05-05 15:20:07 <Diablo-D3> er
2502 2011-05-05 15:20:09 <jgarzik> TD: you cannot assume that one bitcoin address == one transaction
2503 2011-05-05 15:20:10 <Diablo-D3> not repaying a loan
2504 2011-05-05 15:20:23 <Diablo-D3> and I go to the judge "hey judge, I repaid it, heres the proof"
2505 2011-05-05 15:20:35 <Diablo-D3> without anything written in the memo, she goes "that was just a gift" and then I lose
2506 2011-05-05 15:20:37 <TD> jgarzik: why not?
2507 2011-05-05 15:20:44 <Diablo-D3> but if I wrote in the memo field "repaying loan" then BAM shes fucked
2508 2011-05-05 15:20:46 <Diablo-D3> I win
2509 2011-05-05 15:20:52 <jgarzik> TD: because people can and do reuse bitcoin addresses for multiple transactions
2510 2011-05-05 15:20:55 <BlueMatt> TD: for individual to individual txes you cant
2511 2011-05-05 15:21:02 <BlueMatt> ie public donation address
2512 2011-05-05 15:21:04 <Diablo-D3> also, a lot of companies prefer you write your account id in the memo field
2513 2011-05-05 15:21:17 <Diablo-D3> although thats kind of moot with bitcoin
2514 2011-05-05 15:21:21 <jgarzik> TD: it is only a /recommendation/ that you use a fresh addr
2515 2011-05-05 15:21:24 <UukGoblin> anyone can send money to a bitcoin address
2516 2011-05-05 15:21:37 <UukGoblin> but not anyone can put the same secret data into a memo
2517 2011-05-05 15:21:41 <Diablo-D3> so
2518 2011-05-05 15:21:42 <TD> BlueMatt: but for individual to indvidual txns you probably discussed it outside of bitcoin anyway. and if there needs to be a complex explanation, 500 bytes might not be enough
2519 2011-05-05 15:21:43 <Diablo-D3> Im thinking
2520 2011-05-05 15:21:53 <Diablo-D3> we should be able to send memos in the tx
2521 2011-05-05 15:21:55 <Diablo-D3> give it like
2522 2011-05-05 15:22:00 <TD> i mean, why is bitcoin going to beat email?
2523 2011-05-05 15:22:01 <Diablo-D3> enough room to encrypt 140 bytes
2524 2011-05-05 15:22:07 <TD> it can never do a better job of messaging than email or even twitter
2525 2011-05-05 15:22:08 <jgarzik> TD: hardly. I can see it being very useful to send signed data + money
2526 2011-05-05 15:22:10 <Diablo-D3> TD: its not about sending messages
2527 2011-05-05 15:22:13 <wumpus> TD: then put a link + hash of content in there
2528 2011-05-05 15:22:15 <Diablo-D3> TD: its about gluing a message to the tx
2529 2011-05-05 15:22:17 <CIA-30> bitcoin: Gavin Andresen master * r5c4aeb7 / ui.cpp :
2530 2011-05-05 15:22:17 <CIA-30> bitcoin: Merge pull request #194 from johnmaguire2013/tray
2531 2011-05-05 15:22:17 <CIA-30> bitcoin: Removed tooltip balance until it's fixed. - http://bit.ly/k6grz8
2532 2011-05-05 15:22:30 <BlueMatt> TD: true, but if I want to donate to jgarzik, I might want to add "for cpuminer" to the memo to tell him why
2533 2011-05-05 15:22:34 <UukGoblin> TD, email or twitter don't cost bitcoins
2534 2011-05-05 15:22:45 <TD> you could also just send him a mail and then he can reply :-)
2535 2011-05-05 15:22:54 <Diablo-D3> TD: but then I forget what the tx is for
2536 2011-05-05 15:23:00 <jgarzik> I'm ok with limiting OP_DROP to 32 bytes
2537 2011-05-05 15:23:04 <jgarzik> sha256 hash size
2538 2011-05-05 15:23:12 <Diablo-D3> also, bitcoin ALREADY does memos
2539 2011-05-05 15:23:20 <Diablo-D3> it just cant do them on address targets
2540 2011-05-05 15:23:20 <sipa> jgarzik: 32 bytes should be more than enough, i agree
2541 2011-05-05 15:23:22 <Diablo-D3> it works on ip targets
2542 2011-05-05 15:23:24 <gavinandresen> If we've got the fee schedule right then the size shouldn't matter
2543 2011-05-05 15:23:37 <Diablo-D3> gavinandresen: no, because then people will just PAY for abuse
2544 2011-05-05 15:23:52 <Diablo-D3> then again, we have block maximum
2545 2011-05-05 15:23:52 <jgarzik> "it's like crack: the first few bytes are free"
2546 2011-05-05 15:23:55 <gavinandresen> (the default fee rules... lets not get started on 'but miners can set fees blah blah blah')
2547 2011-05-05 15:23:57 <TD> yeah, IP-to-IP messaging was kind of a misfeature too
2548 2011-05-05 15:24:09 <sipa> and maybe it's a bit against the initial intent of using separate addresses if you want to distinguish transactions, but it is extremely useful in e-commerce
2549 2011-05-05 15:24:14 <Diablo-D3> gavinandresen: does main client support fee jacking yet?
2550 2011-05-05 15:24:18 <Diablo-D3> gavinandresen: because I want that feature
2551 2011-05-05 15:24:24 <sipa> fee jacking?
2552 2011-05-05 15:24:26 <gavinandresen> ... and the free-transaction-priority rules
2553 2011-05-05 15:24:36 <gavinandresen> (bigger-k transactions are already penalized)
2554 2011-05-05 15:24:36 <Diablo-D3> sipa: insanely high min fees to get into blocks
2555 2011-05-05 15:24:38 <TD> the work that's already being done on signing messages with bitcoin addresses seems more useful
2556 2011-05-05 15:24:50 <Diablo-D3> gavinandresen: ask xelister about that time he sent me 10btc in 1000 0.01 btc tx
2557 2011-05-05 15:24:59 <Diablo-D3> gavinandresen: it cost me almost 4btc to send 685 btc to mtgox
2558 2011-05-05 15:25:06 eternal11 has joined
2559 2011-05-05 15:25:08 <Diablo-D3> gavinandresen: across multiple tx because I kept hitting max tx size
2560 2011-05-05 15:25:12 <TD> putting the data in the scriptSig sort of confuses me
2561 2011-05-05 15:25:20 <TD> which input gets it?
2562 2011-05-05 15:25:28 <TD> the first by convention?
2563 2011-05-05 15:25:38 <sipa> wait... a message would go in scriptPubKey, no?
2564 2011-05-05 15:25:42 diki has joined
2565 2011-05-05 15:25:50 <xelister> Diablo-D3: Problem, Diablo-D3? /:-D
2566 2011-05-05 15:25:55 <TD> well, jgarziks patch just allows the same scriptPubKey as before but with OP_DROP as the first opcode
2567 2011-05-05 15:26:03 <TD> so you'd have to put the message as the last element in the scriptSig I guess
2568 2011-05-05 15:26:14 <Diablo-D3> gavinandresen: it took him 35 hours to get all those tx into blocks.]
2569 2011-05-05 15:26:24 <sipa> well, now we're at it, i think in the future we may want to put a (configurable) penalty for *splitting* coins, defines as a measure of coinsize in the output minus a measure of coinsize in the input
2570 2011-05-05 15:27:33 <gavinandresen> TD: I agree, I was thinking the message goes in the TxOut. Doesn't makes sense in scriptsig TxIn (what if txn is a sendmany?)
2571 2011-05-05 15:27:56 <BlueMatt> has to be in txout
2572 2011-05-05 15:28:16 <UukGoblin> this is funny
2573 2011-05-05 15:28:29 <UukGoblin> just yesterday, half a channel was telling me I'm wrong and will damage the bitcoin network
2574 2011-05-05 15:28:29 <noagendamarket> Diablo-D3 at least he didnt stick trololol in the tx
2575 2011-05-05 15:28:38 eternal1 has quit (Ping timeout: 240 seconds)
2576 2011-05-05 15:28:40 <UukGoblin> and now you're all talking about adding the feature I wanted :-]
2577 2011-05-05 15:28:41 <Diablo-D3> noagendamarket: he might as well have.
2578 2011-05-05 15:28:56 danlucraft has quit (Remote host closed the connection)
2579 2011-05-05 15:28:56 <Diablo-D3> UukGoblin: meh, honestly, I wouldnt mind having memo field put into btc.
2580 2011-05-05 15:28:58 <noagendamarket> blocktroll
2581 2011-05-05 15:29:16 danlucraft has joined
2582 2011-05-05 15:29:39 <UukGoblin> Diablo-D3, that single memo field can be used to implement the whole bitdns with all the implications of "polluting" the main blockchain
2583 2011-05-05 15:29:46 <UukGoblin> Diablo-D3, not that I'd mind to have it
2584 2011-05-05 15:29:49 <BlueMatt> UukGoblin: slightly different implementation...but its needed...just hard to get right
2585 2011-05-05 15:30:08 <sipa> UukGoblin: and i still object to using it for that
2586 2011-05-05 15:30:21 dbitcoin has quit (Ping timeout: 246 seconds)
2587 2011-05-05 15:30:32 <noagendamarket> seems like it would happen imo
2588 2011-05-05 15:30:40 <UukGoblin> sipa, ugh... you want to have a memo field whilst telling people what they can and can't use it for?
2589 2011-05-05 15:30:47 <jgarzik> huh?
2590 2011-05-05 15:30:49 thedrs has left ()
2591 2011-05-05 15:30:56 <jgarzik> yes, it's in the scriptPubkey of the txout
2592 2011-05-05 15:30:57 <BlueMatt> UukGoblin: only sort of, bitdns wouldnt really be good as you cant really reject txes based on memos which create dns problems
2593 2011-05-05 15:31:18 <gavinandresen> jgarzik: (I haven't actually looked at your patch...)
2594 2011-05-05 15:31:38 <UukGoblin> BlueMatt, all that resolution would be done way outside the bitcoin blockchain
2595 2011-05-05 15:31:39 <sipa> UukGoblin: no, but it isn't permanent storage in the block chain
2596 2011-05-05 15:31:42 <BlueMatt> jgarzik: looking at the patch...doesnt the spendcoins stuff need updated as well
2597 2011-05-05 15:32:03 <jgarzik> BlueMatt: yes. it's just the first step, enabling some of the relaying, so that people can experiment
2598 2011-05-05 15:32:11 <BlueMatt> jgarzik: fair enough...
2599 2011-05-05 15:32:13 <UukGoblin> sipa, and BitStamp doesn't need a permanent storage in the block chain either
2600 2011-05-05 15:32:26 <noagendamarket> messages will get pruned from most of thelock chain
2601 2011-05-05 15:32:33 <sipa> indeed
2602 2011-05-05 15:32:33 <BlueMatt> Id like to see purge of old txes first, just to point out that the storage isnt permanent to people looking to abuse it
2603 2011-05-05 15:32:40 <sipa> BlueMatt: agree
2604 2011-05-05 15:32:56 <noagendamarket> agreed
2605 2011-05-05 15:32:57 <BlueMatt> UukGoblin: BitStamp?
2606 2011-05-05 15:33:23 <UukGoblin> BlueMatt, a generalized p2p timestamping service, allowing for other stuff like BitDNS to be based upon it
2607 2011-05-05 15:33:24 <gavinandresen> BlueMatt, sipa: fast download for non-mining newbies aught to be first.
2608 2011-05-05 15:33:34 <BlueMatt> gavinandresen: that too
2609 2011-05-05 15:33:37 <noagendamarket> he wants to timestamp his mmko game
2610 2011-05-05 15:33:37 <noagendamarket> lol
2611 2011-05-05 15:33:40 tippenein has joined
2612 2011-05-05 15:33:54 <jgarzik> the list of templates in script.cpp's Solver() are the list of transactions that will pass isStandard()
2613 2011-05-05 15:34:02 <jgarzik> it just checks the ops
2614 2011-05-05 15:34:20 <sipa> gavinandresen: how would you do that?
2615 2011-05-05 15:34:25 <BlueMatt> UukGoblin: isnt that just "if its after, its timestamp is later" or do you actually want to add timestamps?
2616 2011-05-05 15:34:30 wolfspraul has quit (Quit: leaving)
2617 2011-05-05 15:34:32 <BlueMatt> sipa: I think he means thin clients
2618 2011-05-05 15:34:43 <jgarzik> for example, std tx #2 is: OP_DUP << OP_HASH160 << OP_PUBKEYHASH << OP_EQUALVERIFY << OP_CHECKSIG
2619 2011-05-05 15:34:48 <jgarzik> but the entire stack is OP_DUP OP_HASH160 d29e146b169d7dce3e7ed2ce544f9acc99fb93f6 OP_EQUALVERIFY OP_CHECKSIG
2620 2011-05-05 15:34:49 <gavinandresen> Yeah, finish the headersonly branch
2621 2011-05-05 15:34:58 <jgarzik> note the difference between the two
2622 2011-05-05 15:35:30 <sipa> thin client is definitely needed
2623 2011-05-05 15:35:34 <jgarzik> sipa: agreed
2624 2011-05-05 15:35:41 <UukGoblin> BlueMatt, I don't understand what you're asking. I want a service of "Here's my hash, here's my BTC fee, go on and timestamp it. Later on, I want to be able to prove that my hash existed before certain time X"
2625 2011-05-05 15:35:46 <BlueMatt> so...who actually has time to do that?
2626 2011-05-05 15:35:50 <jgarzik> sipa: next big step is simply expiring spent TX's
2627 2011-05-05 15:35:56 <UukGoblin> where X is the time you broadcast the hash + a few minutes/hours
2628 2011-05-05 15:36:05 <jgarzik> sipa: once you have that working (and thus partial block chain working), thin client is easy
2629 2011-05-05 15:36:59 <BlueMatt> UukGoblin: ah, so just adding a time to the chain...not sure where that is necessary, but that is actually alot more work as I can put a time in as tomorrow now and no one will reject it...ie you have to convince everyone to change clients
2630 2011-05-05 15:37:22 <sipa> jgarzik: i had some idea about splitting up main.{cpp,h} and creating two OO classes from it, CWallet and CFullNode, where a wallet would register to a fullnode to receive updates about transactions and blocks
2631 2011-05-05 15:37:36 <BlueMatt> sipa: PLEASE DO
2632 2011-05-05 15:37:49 <UukGoblin> BlueMatt, a bitcoin block inside a chain has a timestamp, so as long as it includes my transaction, it's timestamped.
2633 2011-05-05 15:37:51 <sipa> i'm already spending way more time on this than i should
2634 2011-05-05 15:38:00 <sipa> and in the future i'll have less
2635 2011-05-05 15:38:10 <BlueMatt> sipa: so do it now before you have less ;)
2636 2011-05-05 15:38:15 <UukGoblin> BlueMatt, miners only have a narrow window in which they can put the timestamp in
2637 2011-05-05 15:38:22 <BlueMatt> UukGoblin: so your idea is effectively already in the client?
2638 2011-05-05 15:38:26 <UukGoblin> BlueMatt, yes.
2639 2011-05-05 15:38:33 <gavinandresen> Speaking of re-organizing... when should I pull jaromil's autotools re-org?
2640 2011-05-05 15:38:34 kermit has quit (Ping timeout: 252 seconds)
2641 2011-05-05 15:38:36 <UukGoblin> BlueMatt, I was trying to make sure it doesn't get removed
2642 2011-05-05 15:38:40 <BlueMatt> ok, so why have you been fighting so hard to get it included?
2643 2011-05-05 15:38:50 <UukGoblin> BlueMatt, I was fighting to not get it removed
2644 2011-05-05 15:38:50 <BlueMatt> gavinandresen: when it works with x compiling
2645 2011-05-05 15:39:01 <BlueMatt> UukGoblin: it was going to be removed?
2646 2011-05-05 15:39:04 sgornick has joined
2647 2011-05-05 15:39:27 <xelister> á(°â¡Â°)á hi
2648 2011-05-05 15:39:39 dbitcoin has joined
2649 2011-05-05 15:39:41 <BlueMatt> everyone's here
2650 2011-05-05 15:39:42 <UukGoblin> BlueMatt, hopefully not, but I could see how some people might want it removed, judging from what they were saying about "blockchain pollution"
2651 2011-05-05 15:39:52 <UukGoblin> BlueMatt, and "we don't want non-bitcoin data in our blockchain"
2652 2011-05-05 15:40:23 <BlueMatt> UukGoblin: I think you are very safe...if no one is trying to remove it, its fine...plus everything is discussed at length before any kind of major changes like that
2653 2011-05-05 15:40:51 da2ce7 has quit (Ping timeout: 246 seconds)
2654 2011-05-05 15:41:39 da2ce7 has joined
2655 2011-05-05 15:41:40 da2ce7 has quit (Changing host)
2656 2011-05-05 15:41:40 da2ce7 has joined
2657 2011-05-05 15:42:02 <BlueMatt> anyone know of a good c++ primer for c++ noobs but decent programming-thinkers to help people get started making useful additions to bitcoin?
2658 2011-05-05 15:44:03 <UukGoblin> BlueMatt, hm... interesting question. I know a good site about C++ but it's unfortunately in polish.
2659 2011-05-05 15:44:14 <UukGoblin> BlueMatt, I assume you don't want C knowledge...
2660 2011-05-05 15:44:40 <BlueMatt> Ill take anything, really just need to spend more time with bitcoin itself but...
2661 2011-05-05 15:44:47 <UukGoblin> perhaps xelister or cosurgi could get someone to translate it...? it's http://www.intercon.pl/~sektor/cbx/
2662 2011-05-05 15:45:17 <UukGoblin> BlueMatt, hrm, there should be lots of generic resources on C and C++ on google, but I haven't looked in a while
2663 2011-05-05 15:45:28 <jgarzik> gavinandresen: jaromil's branch is real close. one final blocker (just posted comment)
2664 2011-05-05 15:45:37 <jgarzik> gavinandresen: easy to fix
2665 2011-05-05 15:45:38 <BlueMatt> yay chrome, open a link and it goes ahead and asks if I want to autotranslate
2666 2011-05-05 15:46:02 <UukGoblin> BlueMatt, polish->english autotranslation will really suck
2667 2011-05-05 15:46:13 <BlueMatt> UukGoblin: yea, but readable
2668 2011-05-05 15:46:15 marlowe has joined
2669 2011-05-05 15:46:16 <xelister> BlueMatt: we're interested in coding for BitCoin client
2670 2011-05-05 15:46:17 * jgarzik -> lunch
2671 2011-05-05 15:46:45 <xelister> (in 4 weeks best, once we clear currently orderd bitcoin work :)
2672 2011-05-05 15:47:03 <xelister> UukGoblin: translate that?
2673 2011-05-05 15:47:10 LightRider is now known as afk!~LightRide@unaffiliated/lightrider|LightRider
2674 2011-05-05 15:47:18 <xelister> I think its super old site?
2675 2011-05-05 15:47:21 <UukGoblin> xelister, dunno, it's a very nice page I like
2676 2011-05-05 15:47:32 <UukGoblin> it's about 6 years old I know
2677 2011-05-05 15:48:01 <xelister> yeap
2678 2011-05-05 15:48:03 <UukGoblin> if you want c++0x knowledge it's not good but apart from that it's pretty up-to-date ;-]
2679 2011-05-05 15:48:07 <xelister> yeap
2680 2011-05-05 15:48:10 <BlueMatt> jgarzik: Id really like to see a way to specify location of libraries manually (which pretty directly allows x compiling support which is what is needed for the new build process for win32)
2681 2011-05-05 15:48:15 <BlueMatt> gavinandresen: ^
2682 2011-05-05 15:48:16 <xelister> but what you suggested? yea this could be translated.
2683 2011-05-05 15:48:26 <xelister> I think auto google translate would make some sense to start with
2684 2011-05-05 15:48:27 <UukGoblin> BlueMatt, LD_LIBRARY_PATH?
2685 2011-05-05 15:48:33 <xelister> but translating it all is a long task, lots of text
2686 2011-05-05 15:48:49 <BlueMatt> UukGoblin: no for the different deps separately
2687 2011-05-05 15:48:50 <UukGoblin> xelister, I suggested perhaps figuring out how many BTC would it cost to translate ;-]
2688 2011-05-05 15:48:59 <xelister> UukGoblin: ok. but who wants to pay?
2689 2011-05-05 15:49:24 <UukGoblin> BlueMatt, perhaps make a special directory with symlinks to each dep?
2690 2011-05-05 15:49:40 <UukGoblin> xelister, dunno ;-] /maybe/ BlueMatt but also maybe not ;-]
2691 2011-05-05 15:49:43 <BlueMatt> sounds too complicated, why cant it just be in autotools, not too hard
2692 2011-05-05 15:49:55 <UukGoblin> BlueMatt, oh, depends what you want, yeah
2693 2011-05-05 15:50:20 <BlueMatt> na, I think I know enough C++, really just need to spend more time with boost to figure some of that out...but what I really need is more in-depth bitcoin study
2694 2011-05-05 15:50:35 <BlueMatt> All I know is enough to get around the edges adding new features, not enough to really solve bugs
2695 2011-05-05 15:50:35 <UukGoblin> oh, /bitcoin/ specifically
2696 2011-05-05 15:51:03 <UukGoblin> then I don't know. There's some useful stuff at the bitcoin.it wiki
2697 2011-05-05 15:51:21 <BlueMatt> not really sadly...
2698 2011-05-05 15:51:27 <gavinandresen> The trick with C++ is knowing what features to avoid. Don't even be tempted to do multiple inheritance, for example...
2699 2011-05-05 15:51:53 <BlueMatt> lol, ok. Ill keep that in mind, I suppose that all falls under KISS
2700 2011-05-05 15:52:02 <BlueMatt> maybe Ill spend a couple days straight just reading code
2701 2011-05-05 15:52:12 <gavinandresen> yup. Many of the boost libraries go off the deep end, in my opinion
2702 2011-05-05 15:52:52 <xelister> BlueMatt: theres a #boost even
2703 2011-05-05 15:52:59 citiz3n has quit ()
2704 2011-05-05 15:53:11 <BlueMatt> xelister: Ill keep that in mind, though they do seem to have pretty good docs
2705 2011-05-05 15:53:12 <xelister> gavinandresen: well depends, boost uses lots of exotic stuff
2706 2011-05-05 15:54:06 skyewm has joined
2707 2011-05-05 15:57:35 <wumpus> gavinandresen: indeed, the trick with C++ is to find a safe subset that you're comfortable with. The problem is that different people have different subsets, so there are always conflicts about what features to use and what features to not use.
2708 2011-05-05 15:58:17 <sipa> i've learned a lot of C++ by studying the source code :D
2709 2011-05-05 15:58:20 <sipa> (of bitcoin)
2710 2011-05-05 15:58:41 <UukGoblin> whatever you do, just don't use goto ;-]
2711 2011-05-05 15:58:42 <wumpus> person A hates templates with a passion, person B hates exceptions, person C hates 'new' or using objects for things.. etc.. in any project, it's always a war
2712 2011-05-05 15:58:42 <TD> hmmm ... interesting way to learn :-)
2713 2011-05-05 15:59:07 <sipa> UukGoblin: as a matter of fact, i do have a git branch in which there is code written by me, that does use goto :D
2714 2011-05-05 15:59:22 <UukGoblin> sipa, I know, it's sometimes useful ;-P
2715 2011-05-05 15:59:37 <wumpus> UukGoblin: unless writing a VM :)
2716 2011-05-05 15:59:45 <UukGoblin> quite useful in C to clean up before a function return, but that's avoidable in C++ if you use exceptions
2717 2011-05-05 16:00:14 <wumpus> goto is useful in some performance edge cases, but apart from that, yeah it's certainly not recommended
2718 2011-05-05 16:00:36 <eps> http://dpaste.de/OvGU/ is a goto needed here?
2719 2011-05-05 16:00:38 <wumpus> at least if someone has to maintain the code later on
2720 2011-05-05 16:00:48 <wumpus> for game programming, everything is fair I guess :P
2721 2011-05-05 16:01:10 <sipa> eps: needed, no; acceptable, imho yes
2722 2011-05-05 16:01:20 <xelister> wumpus: both of this persons you mentioned are idiots btw ;)
2723 2011-05-05 16:01:39 <wumpus> xelister: I agree.. but please don't start this war here ;)
2724 2011-05-05 16:01:41 <eps> some dude posted it on twitter saying goto was the only clean non hacky way of refactoring it
2725 2011-05-05 16:02:16 <wumpus> xelister: the underlying problem is that no one agrees on what part of C++ is good to use
2726 2011-05-05 16:03:15 <wumpus> or how to use the features, so different projects use wildly different styles (also for memory allocation, reference counted pointers versus RAI versus explicit new/delete etc)... i hate it
2727 2011-05-05 16:04:18 <UukGoblin> eps, there's no cleanup so I'd stick it in a function and use returns
2728 2011-05-05 16:04:41 noagendamarket has quit (Ping timeout: 258 seconds)
2729 2011-05-05 16:05:08 <wumpus> UukGoblin: good point
2730 2011-05-05 16:05:09 <UukGoblin> wumpus, I'd rather have choice of features than use java ;-]
2731 2011-05-05 16:05:10 <sipa> UukGoblin: agree, actually
2732 2011-05-05 16:05:38 <wumpus> UukGoblin: yeah, C++ is fine for solo projects, or games and other write-only code :P
2733 2011-05-05 16:06:18 <xelister> wumpus: all C++.
2734 2011-05-05 16:06:24 <xelister> wumpus: just use it correctly
2735 2011-05-05 16:06:27 <xelister> use boost and std
2736 2011-05-05 16:06:42 <xelister> question could be, should most complex parts of C++ used, like that signals thin
2737 2011-05-05 16:06:47 <wumpus> it's also telling that one of the first topics on the bitcoin forum about the code was to refactor it... I mean, wtf
2738 2011-05-05 16:07:08 <wumpus> someone gives you code for free and the first thing you do is refactor it to your preferences
2739 2011-05-05 16:07:21 <xelister> refacotring is good
2740 2011-05-05 16:07:40 <wumpus> yeah, if you coordinate it with the author :P but styles, in C++, vary so wildly
2741 2011-05-05 16:07:50 <xelister> yea
2742 2011-05-05 16:08:01 <xelister> *a wild c++ style appears*
2743 2011-05-05 16:08:27 <wumpus> you like boost and std.. right, I do too. There's a lot of people out there that hate them though, for variuos usually silly reasons :)
2744 2011-05-05 16:08:47 <xelister> well they are idiots everywhere that hate good tools
2745 2011-05-05 16:08:52 <xelister> these guys hating std
2746 2011-05-05 16:08:59 <xelister> jgarzik hating freenet :P
2747 2011-05-05 16:09:16 <xelister> chinesefags hating non-ID-card-needed internet
2748 2011-05-05 16:09:17 <wumpus> the "let's ignore std::string and use our own string implementation because it is faster" folks.. . right, if it's your app, but if it's a library that needs to interoperate... :/
2749 2011-05-05 16:09:18 <UukGoblin> xelister, hey, freenet is java. It's asking to be hated. ;-]
2750 2011-05-05 16:09:41 <UukGoblin> let's move to #bitcoin-flames perhaps? ;-]
2751 2011-05-05 16:09:52 <xelister> catholicfags hating the rubber
2752 2011-05-05 16:10:04 <wumpus> that's true xelister
2753 2011-05-05 16:12:02 <JFK911> surveillance is right for chinese tho
2754 2011-05-05 16:12:16 <UukGoblin> yeah that bogs me
2755 2011-05-05 16:12:19 <JFK911> most chinese are ok with it
2756 2011-05-05 16:12:23 <luke-jr> wumpus: obviously the correct solution is to make std::string faster
2757 2011-05-05 16:12:37 <UukGoblin> they seem like nice folk, but think surveilance is good? /me suspects severe brainwashing
2758 2011-05-05 16:12:46 <luke-jr> but srsly, Qt > C++
2759 2011-05-05 16:13:00 <wumpus> luke-jr: indeed. Or even better, benchmark your program, make sure strings are the bottleneck, and if not leave it well enough alone
2760 2011-05-05 16:13:09 <luke-jr> :P
2761 2011-05-05 16:13:28 <wumpus> and if strings are the bottleneck you might be doing something else wrong in your design
2762 2011-05-05 16:13:38 <genjix> xelister: are you polish?
2763 2011-05-05 16:13:43 <JFK911> UukGoblin: i found that many chinese would be frank with me (im white so its clear im not a police officer)
2764 2011-05-05 16:13:56 <luke-jr> tbh, I much prefer C for strings
2765 2011-05-05 16:13:58 <JFK911> and yeah, the general opinion is, "we're disorderly and need order"
2766 2011-05-05 16:14:09 <genjix> C for strings? wtf you mean char arrays?
2767 2011-05-05 16:14:13 <wumpus> C strings? omg, no
2768 2011-05-05 16:14:14 <genjix> evil.
2769 2011-05-05 16:14:16 <luke-jr> even C++ adds so much overhead just by making it OO
2770 2011-05-05 16:14:18 <JFK911> of course, i wasnt speaking with dissidents
2771 2011-05-05 16:14:18 <wumpus> not bufferoverflow hell
2772 2011-05-05 16:14:19 <UukGoblin> JFK911, yeah it's weird, "we need order so we'd rather be lied to"
2773 2011-05-05 16:14:19 <luke-jr> genjix: yes
2774 2011-05-05 16:14:28 <genjix> there is no overhead in C++
2775 2011-05-05 16:14:29 <mtrlt> c strings lol
2776 2011-05-05 16:14:31 <luke-jr> wumpus: only bufferoverflow if you don't know what you're doing
2777 2011-05-05 16:14:33 <genjix> it's your imagination
2778 2011-05-05 16:14:34 <wumpus> I like Python strings, and Java strings
2779 2011-05-05 16:14:41 <wumpus> and std::string
2780 2011-05-05 16:14:44 <mtrlt> what kind of string manipulation do you do where overhead matters?
2781 2011-05-05 16:14:45 <UukGoblin> glib_string for C strings ;-]
2782 2011-05-05 16:14:48 bgeron has joined
2783 2011-05-05 16:14:50 <wumpus> and glib_string
2784 2011-05-05 16:14:51 <luke-jr> mtrlt: none :P
2785 2011-05-05 16:14:56 <wumpus> but C strings.. man.. strcpy should DIE
2786 2011-05-05 16:14:57 * luke-jr stabs UukGoblin
2787 2011-05-05 16:14:57 <mtrlt> :D
2788 2011-05-05 16:15:05 <xelister> genjix: im unpolished >_>
2789 2011-05-05 16:15:16 <UukGoblin> genjix, he's polish, don't trust him ;-]
2790 2011-05-05 16:15:24 <xelister> UukGoblin is polish too
2791 2011-05-05 16:15:31 <wumpus> mtrlt: unless you're writing an XML parser, which would be a bad thing to do anyway, I can't think of any examples :)
2792 2011-05-05 16:15:31 <genjix> ahh cool
2793 2011-05-05 16:15:33 <mtrlt> kurwa
2794 2011-05-05 16:15:34 <xelister> but also britfag =) you now brits, more CCTVs then people
2795 2011-05-05 16:15:43 <genjix> because we have a group here in warsaw
2796 2011-05-05 16:15:44 <xelister> kurwa \ó/
2797 2011-05-05 16:15:48 <genjix> kurwa
2798 2011-05-05 16:15:54 <genjix> czesch
2799 2011-05-05 16:15:54 <xelister> genjix: "here"? your in pl?
2800 2011-05-05 16:15:57 <genjix> yes
2801 2011-05-05 16:16:05 <xelister> but you are polish?
2802 2011-05-05 16:16:08 <genjix> nope
2803 2011-05-05 16:16:16 <UukGoblin> nah, he doesn't have a polish accent ;-]
2804 2011-05-05 16:16:26 <UukGoblin> genjix, what are you doing in warsaw then?
2805 2011-05-05 16:16:26 <mtrlt> ;-] <- polish smilie
2806 2011-05-05 16:16:29 <mtrlt> :P
2807 2011-05-05 16:16:38 <genjix> is xelister or UukGoblin in poland ATM?
2808 2011-05-05 16:16:45 <xelister> im in pl atm.. ;)
2809 2011-05-05 16:16:46 <UukGoblin> I'm not
2810 2011-05-05 16:17:08 <genjix> where in poland?
2811 2011-05-05 16:17:17 <xelister> at your gf's place? :}
2812 2011-05-05 16:17:23 <UukGoblin> genjix, actually, it was the way you said 'cze¶æ' ;-]
2813 2011-05-05 16:17:26 <genjix> near szechzin, gdansk or krakow
2814 2011-05-05 16:17:53 genjix has left ()
2815 2011-05-05 16:18:15 DukeOfURL has joined
2816 2011-05-05 16:18:51 genjix has joined
2817 2011-05-05 16:18:57 <genjix> oops misclicked
2818 2011-05-05 16:19:04 Speeder has quit (Read error: No route to host)
2819 2011-05-05 16:20:00 <luke-jr> http://bazaar.launchpad.net/~armagetronad-dev/armagetronad/0.2.8-armagetronad-work/view/head:/src/tools/tDirectories.cpp#L212
2820 2011-05-05 16:20:02 <UukGoblin> eps, hang on, ZEDSHAW? the bloke who ranted at ruby community? He's mental! ;-]
2821 2011-05-05 16:20:04 <luke-jr> do that in normal C++
2822 2011-05-05 16:20:05 <luke-jr> :P
2823 2011-05-05 16:20:32 <xelister> UukGoblin: lol!
2824 2011-05-05 16:20:35 genjix has left ()
2825 2011-05-05 16:20:38 <eps> UukGoblin, yeah ruby people hate him
2826 2011-05-05 16:20:41 <xelister> that was the Z guy with super rant blog?
2827 2011-05-05 16:20:43 genjix has joined
2828 2011-05-05 16:20:52 genjix has quit (Quit: leaving)
2829 2011-05-05 16:20:59 <eps> i think he is a python dude
2830 2011-05-05 16:20:59 genjix has joined
2831 2011-05-05 16:21:00 genjix has quit (Changing host)
2832 2011-05-05 16:21:00 genjix has joined
2833 2011-05-05 16:21:00 <xelister> lashing out on rubby startups with "fucking facebook clones but with childrent faces in them" etc?
2834 2011-05-05 16:21:01 genjix has quit (Client Quit)
2835 2011-05-05 16:21:09 genjix has joined
2836 2011-05-05 16:21:09 genjix has quit (Changing host)
2837 2011-05-05 16:21:09 genjix has joined
2838 2011-05-05 16:21:10 <UukGoblin> xelister, yeah, that's the one! :-]
2839 2011-05-05 16:21:13 <xelister> lol
2840 2011-05-05 16:21:16 <xelister> world is so small.
2841 2011-05-05 16:21:18 <wumpus> lol he's crazy
2842 2011-05-05 16:21:24 <eps> dunno, i think i came late to this particular party
2843 2011-05-05 16:21:26 <wumpus> what does it have to do with the programming language
2844 2011-05-05 16:21:35 <UukGoblin> luke-jr, what do you mean? looks c++ to me
2845 2011-05-05 16:21:36 <eps> he doesn't seem so crazy these days
2846 2011-05-05 16:21:42 <xelister> >_>
2847 2011-05-05 16:21:47 <xelister> dunno, perhaps we was correct in his rant
2848 2011-05-05 16:21:49 <xelister> Im not sure
2849 2011-05-05 16:21:53 <xelister> dont develop rubby
2850 2011-05-05 16:21:56 <UukGoblin> path management is fucking hard even in perl ;-]
2851 2011-05-05 16:21:57 <wumpus> he has some good articles about python usage
2852 2011-05-05 16:22:15 <UukGoblin> xelister, well I don't like ruby much either, but the way he ranted at them... huh ;-]
2853 2011-05-05 16:22:20 <wumpus> but his stance on ruby is strange, I mean python and ruby, apart from the syntax, are very close
2854 2011-05-05 16:22:28 <xelister> UukGoblin: so what. like Diablo-D3 =)
2855 2011-05-05 16:22:30 Speeder has joined
2856 2011-05-05 16:22:41 <UukGoblin> xelister, nah, better. Diablo-D3 is good too though
2857 2011-05-05 16:22:51 <luke-jr> UukGoblin: it's not.
2858 2011-05-05 16:22:53 <luke-jr> UukGoblin: it's C
2859 2011-05-05 16:22:57 <UukGoblin> xelister, but he's a java proponent
2860 2011-05-05 16:23:06 <UukGoblin> I can't agree with java ;-]
2861 2011-05-05 16:23:06 <luke-jr> except for the debug block
2862 2011-05-05 16:23:13 <UukGoblin> luke-jr, oh, that's what you mean.
2863 2011-05-05 16:23:43 <luke-jr> to do it in C++ would take longer and be less efficient
2864 2011-05-05 16:23:59 <wumpus> I can agree with nearly any language, except for visual basic :P
2865 2011-05-05 16:24:33 <eps> brainfuck?
2866 2011-05-05 16:24:34 * luke-jr gives wumpus Whitespace
2867 2011-05-05 16:25:14 * UukGoblin chucks malbolge into the picture
2868 2011-05-05 16:25:21 <wumpus> lol, I doubt that those are ever the right match for a problem
2869 2011-05-05 16:25:26 <xelister> UukGoblin: =)
2870 2011-05-05 16:25:54 <xelister> oh bitcoiners
2871 2011-05-05 16:25:56 <xelister> u so crazy
2872 2011-05-05 16:26:11 <UukGoblin> luke-jr, I agree, C is great for such pointery string manipulation. But it's also very prone to bugs ;-]
2873 2011-05-05 16:26:42 <BlueMatt> I prefer straight C, easier to understand to me, a complete noob in both
2874 2011-05-05 16:27:20 TheAncientGoat has quit (Ping timeout: 276 seconds)
2875 2011-05-05 16:27:57 <wumpus> I guess you could use google go, it's like C but without pointers :P
2876 2011-05-05 16:28:07 <UukGoblin> luke-jr, how about some sort of for_each(path.split("/"), mkdir(_1))?
2877 2011-05-05 16:28:29 <UukGoblin> I'd have to look up the actual functions and details but that's my generic C++ solution to mkdir_recurse
2878 2011-05-05 16:30:19 <UukGoblin> wumpus, C without pointers? sacrilege
2879 2011-05-05 16:30:39 <UukGoblin> malbolge. srsly. at least it's challenging.
2880 2011-05-05 16:31:19 Marcel has joined
2881 2011-05-05 16:32:39 kika has joined
2882 2011-05-05 16:33:01 amiller has quit (Ping timeout: 246 seconds)
2883 2011-05-05 16:34:59 kika has quit (Client Quit)
2884 2011-05-05 16:36:50 <anarchyx> anyone know how i can use aticonfig over ssh? its complaining about not being in X.. but I dont have remote access to the ubuntu desktop.. just to ssh
2885 2011-05-05 16:36:56 kika has joined
2886 2011-05-05 16:37:02 <UukGoblin> anarchyx, DISPLAY=:0 aticonfig
2887 2011-05-05 16:37:12 <anarchyx> tnx :)
2888 2011-05-05 16:38:08 <UukGoblin> you'll have to run that as the same user who ran X, unless you make some nice xauth policy for other users
2889 2011-05-05 16:38:19 <anarchyx> yup works :)
2890 2011-05-05 16:38:46 <UukGoblin> happy mining ;-]
2891 2011-05-05 16:38:53 <anarchyx> tnx :)
2892 2011-05-05 16:39:15 <kika> hey guys what if my bitcoin client wans to send a txt ( because i want to send some bitcoins i own to someone else), then my bitcoin client will broadcast an inv message to every node it connects to, and then every node will check if my txt is valid right?
2893 2011-05-05 16:39:26 <kika> *tx
2894 2011-05-05 16:39:36 <sipa> yes
2895 2011-05-05 16:39:52 <kika> sipa: and what could happen if all the nodes reject the tx because those are malicious nodes?
2896 2011-05-05 16:40:07 <kika> sipa: my tx will never go to the main chain right ?
2897 2011-05-05 16:40:31 <UukGoblin> kika, you'll need to find some non-malicious ones or encourage them with a fee
2898 2011-05-05 16:40:35 caedes has quit (Ping timeout: 248 seconds)
2899 2011-05-05 16:41:03 <UukGoblin> kika, IIRC the mainstream client will re-send the transaction for you several times if it has trouble getting forwarded
2900 2011-05-05 16:41:07 <kika> UukGoblin: what if i send a txt and it gets only by malicious nodes, then after 1 hour, i close my bitcoin client
2901 2011-05-05 16:41:24 <kika> UukGoblin: my bitcoin client stores the tx on disk so later it can re-send it ?
2902 2011-05-05 16:41:37 <UukGoblin> hmm, don't know that, actually, but I think it should
2903 2011-05-05 16:41:41 <TD> it does
2904 2011-05-05 16:41:47 <TD> it resends forever until it sees the tx get into a block
2905 2011-05-05 16:41:53 <kika> TD: it keeps the txs on disk when you close the client?
2906 2011-05-05 16:41:57 <sipa> kika: yes
2907 2011-05-05 16:41:58 phantomcircuit has joined
2908 2011-05-05 16:42:21 NOTAL has quit (Ping timeout: 260 seconds)
2909 2011-05-05 16:42:31 <kika> sipa: here it says... "Only the headers are necessary for verifying received transactions"
2910 2011-05-05 16:42:42 <kika> "full blocks are only needed by generators"
2911 2011-05-05 16:42:59 <kika> how come i need only the block headers to verify if txs are valid?
2912 2011-05-05 16:44:00 <kika> if i dont know the previous txs outputs to match with the txs inputs im not able to understand how come i can only validate a tx by having only the block header
2913 2011-05-05 16:44:09 <UukGoblin> hmm I think you do need a few full blocks to be able to verify a tx... otherwise you can trust the network who has the full blocks to verify it
2914 2011-05-05 16:44:30 <kika> here on the wiki
2915 2011-05-05 16:44:32 <kika> it says
2916 2011-05-05 16:44:34 <kika> https://en.bitcoin.it/wiki/Network
2917 2011-05-05 16:44:39 <kika> "0.3.18 introduced new headers and getheaders messages. These are not currently used, but they will be used by non-generators to download block headers without the corresponding block bodies. Only the headers are necessary for verifying received transactions -- full blocks are only needed by generators."
2918 2011-05-05 16:44:56 <kika> i dont know how that can be possible
2919 2011-05-05 16:45:12 <kika> unless my node asks another node if the tx is valid
2920 2011-05-05 16:45:24 <UukGoblin> kika, check satoshi's paper, sections 7 and 8
2921 2011-05-05 16:45:36 <kika> url?
2922 2011-05-05 16:46:29 <UukGoblin> http://www.bitcoin.org/bitcoin.pdf
2923 2011-05-05 16:47:00 caedes has joined
2924 2011-05-05 16:47:02 <kika> as i understand the bitcoin network can keep live as long as almost one node has the full main chain stored right?
2925 2011-05-05 16:47:24 <BlueMatt> no
2926 2011-05-05 16:47:33 <lulzplzkthx> No?
2927 2011-05-05 16:47:40 <BlueMatt> as long as all miners have the chain of non-spent txes
2928 2011-05-05 16:47:46 <eps> depends what you mean by alive
2929 2011-05-05 16:48:00 <BlueMatt> all other clients have to have the full list of block headers (well, sort of)
2930 2011-05-05 16:48:02 <phantomcircuit> BlueMatt, uh all miners dont need to have the queue'd transactions
2931 2011-05-05 16:48:24 <sipa> nodes miners connect to
2932 2011-05-05 16:48:24 <gavinandresen> kika: if one node has the full main chain then yes, that is sufficient.
2933 2011-05-05 16:48:30 <Diablo-D3> hrm
2934 2011-05-05 16:48:35 <Diablo-D3> my reject loss here is pretty good
2935 2011-05-05 16:48:38 <BlueMatt> no I mean the txes that havn't been spent yet (ie the ones already in blocks)
2936 2011-05-05 16:48:38 <phantomcircuit> kika, yes the network survives so long as at least one node has the full chain
2937 2011-05-05 16:48:39 <Diablo-D3> 21 over the past 2000
2938 2011-05-05 16:48:45 <gavinandresen> I think BlueMatt is pointing out that is actually more than sufficient... you only need the un-spent txns
2939 2011-05-05 16:48:53 <phantomcircuit> BlueMatt, oh
2940 2011-05-05 16:48:54 <kika> i think there would be possible to create CPU miners that be faster than GPU miners
2941 2011-05-05 16:49:00 <kika> what do you guys think?
2942 2011-05-05 16:49:03 <BlueMatt> no
2943 2011-05-05 16:49:09 <sipa> kika: impossible
2944 2011-05-05 16:49:09 <phantomcircuit> kika, lol no
2945 2011-05-05 16:49:22 <Diablo-D3> kika: yes, if your cpus were 9000ghz.
2946 2011-05-05 16:49:25 <sipa> i mean, some CPU's will be faster than some GPU's
2947 2011-05-05 16:49:27 <kika> is there any faster way to mine other than by using GPU's ?
2948 2011-05-05 16:49:32 <gavinandresen> kika: I worked at silicon graphics for years, and there used to be a debate that general-purpose CPUs would put graphics cards out of business...
2949 2011-05-05 16:49:38 <sipa> kika: not yet
2950 2011-05-05 16:49:56 <wumpus> kika: no, the fastest CPUs cannot compete with the fastest (especially AMD) GPUs hashrate-wise
2951 2011-05-05 16:50:00 <Diablo-D3> kika: think of GPUs as gigantic parallel DSPs
2952 2011-05-05 16:50:05 <Diablo-D3> also, think of the wattages involved
2953 2011-05-05 16:50:14 <JFK911> now sgi is essentially dead!
2954 2011-05-05 16:50:15 Incitatus has joined
2955 2011-05-05 16:50:28 <gavinandresen> JFK911: yup, killed by nvidia and ati
2956 2011-05-05 16:50:35 <JFK911> hm
2957 2011-05-05 16:50:50 <wumpus> gavinandresen: like 3DFX :(
2958 2011-05-05 16:50:55 <JFK911> in my mind i went to their funeral when they announced they were selling x86 pc's
2959 2011-05-05 16:51:10 <JFK911> and that was long before ati was any good
2960 2011-05-05 16:51:38 <JFK911> i was shocked to find out they were still operating last year
2961 2011-05-05 16:51:42 <Diablo-D3> 3DFX didnt die
2962 2011-05-05 16:51:47 <Diablo-D3> it got in nvidia's tummy
2963 2011-05-05 16:51:52 <UukGoblin> gavinandresen, general-purpose CPUs? what came out of this debate?
2964 2011-05-05 16:51:53 <Diablo-D3> and THEN died
2965 2011-05-05 16:52:17 <gavinandresen> special-purpose GPUs won. THey're getting less special-purpose these days, though
2966 2011-05-05 16:52:22 <JFK911> for context, sgi's mips could have shown a lot of promise against contemporary intel
2967 2011-05-05 16:52:33 <JFK911> because back then x86 was horrid
2968 2011-05-05 16:52:33 <UukGoblin> yeah they're even called GPGPUs
2969 2011-05-05 16:52:50 danlucraft has quit (Quit: Leaving.)
2970 2011-05-05 16:52:55 <Diablo-D3> well, its how all hardware works
2971 2011-05-05 16:53:11 <UukGoblin> x86 was always pretty horrid ;-]
2972 2011-05-05 16:53:28 <phantomcircuit> still is :|
2973 2011-05-05 16:53:35 <Diablo-D3> "we need to make this faster" -> "expensive special purpose chip" -> "more flexible less expensive special purpose" -> "even less expensive general purpose chip" -> "lets just fold it back into the cpu"
2974 2011-05-05 16:53:47 <JFK911> i still dont like it, however i think that now, after 35 revs or whatever, it's somewhat usable
2975 2011-05-05 16:53:56 <JFK911> as crazy as it is
2976 2011-05-05 16:54:39 <kika> so basically orhpahn blocks are blocks that lack a predecesor?
2977 2011-05-05 16:54:48 <wumpus> yes, still is, even though x86/64 is much better. Compared to arm, x86 has a horrid over-complicated instruction set...
2978 2011-05-05 16:54:50 <eps> we should just write everything in java, then we wouldn't have to worry about the chip it runs on
2979 2011-05-05 16:55:01 <sipa> kika: blocks without a path that links them to the genesis block
2980 2011-05-05 16:55:11 <sipa> oh wait, not sure
2981 2011-05-05 16:55:13 <eps> 1.. 2.. 3..
2982 2011-05-05 16:55:14 <Diablo-D3> eps: well, seeing as I can run a subset of java on gpus now... ;)
2983 2011-05-05 16:55:16 <eps> only joking
2984 2011-05-05 16:55:26 <wumpus> eps: or Python, or Ruby, or Javascript...
2985 2011-05-05 16:55:47 <UukGoblin> Malbolge!
2986 2011-05-05 16:55:49 <kika> sipa: how a node knows whats the last black on the main chain?
2987 2011-05-05 16:55:51 <wumpus> only if you do heavy number crunching it really matters
2988 2011-05-05 16:55:54 <kika> *block
2989 2011-05-05 16:56:05 <kika> *which one is the last block
2990 2011-05-05 16:56:32 <kika> i understand that the main chain is the one with the higher combined difficulty
2991 2011-05-05 16:56:34 TD has quit (Quit: Ex-Chat)
2992 2011-05-05 16:56:51 <wumpus> UukGoblin: you have a strange obsession with malbolge :)
2993 2011-05-05 16:56:51 <kika> but how come a node knows whats the latest block on the main chain?
2994 2011-05-05 16:57:29 <luke-jr> [12:51:35] <kika> i understand that the main chain is the one with the higher combined difficulty
2995 2011-05-05 16:57:34 <luke-jr> where do people get this?
2996 2011-05-05 16:57:42 <kika> if a node is not well connected to other nodes, that node then may only be seeing blocks on side branches of the main branch i think
2997 2011-05-05 16:57:59 <luke-jr> kika: side branches are rare
2998 2011-05-05 16:58:01 <kika> and that node might think taht the side branch of the main branch is the main chain
2999 2011-05-05 16:58:24 <kika> suppose i have a node on Australia that can only see Australian nodes
3000 2011-05-05 16:58:26 <luke-jr> kika: the main chain is the one with the highest block height
3001 2011-05-05 16:58:31 <UukGoblin> wumpus, I know, do you think you can cure it?
3002 2011-05-05 16:58:32 <luke-jr> no need to look at combined difficulty
3003 2011-05-05 16:58:43 <Diablo-D3> kika: you fail at the internet forever
3004 2011-05-05 16:58:48 <kika> luke-jr: what do you mean by block height?
3005 2011-05-05 16:58:49 <devrandom> luke-jr - actually it's the one with the highest combined difficulty
3006 2011-05-05 16:58:55 <luke-jr> kika: the number you see
3007 2011-05-05 16:59:04 <wumpus> UukGoblin: I'm afraid not, I think it's fatal, the language isn't named after a level of hell by accident
3008 2011-05-05 16:59:05 <luke-jr> devrandom: which is always the one with the highest block count
3009 2011-05-05 16:59:12 <kika> luke-jr: its the combined difficulty of all the blocks on the chain really
3010 2011-05-05 16:59:12 <devrandom> no
3011 2011-05-05 16:59:33 <wumpus> UukGoblin: it's a language of unspeakable horrors
3012 2011-05-05 16:59:34 <devrandom> luke-jr it's not the same thing when you cross a difficulty change
3013 2011-05-05 16:59:48 <wumpus> UukGoblin: but still, i'd prefer it to visual basic any time :)
3014 2011-05-05 16:59:59 <diki> luke, is it possible to lower the maturity to less than 120 blocks ?
3015 2011-05-05 17:00:00 <UukGoblin> wumpus, we need to add OP_CRZY to Script
3016 2011-05-05 17:00:07 <luke-jr> diki: yes
3017 2011-05-05 17:00:16 <diki> No i am asking if you can do it for your pool
3018 2011-05-05 17:00:18 <devrandom> if you cross a difficulty change and the side-chain difficulty is lower, then the side-chain can be longer, yet have a lower combined difficulty
3019 2011-05-05 17:00:27 <luke-jr> diki: no, it's a client-side thing
3020 2011-05-05 17:00:51 <UukGoblin> imagine Script being able to execute Malbolge code...
3021 2011-05-05 17:00:51 <xelister> UukGoblin: oh bitcoiners. u so crazy
3022 2011-05-05 17:00:59 <wumpus> UukGoblin: it supports sub-scripts in arbitrary crazy languages?
3023 2011-05-05 17:01:01 <luke-jr> devrandom: the only way the side-chain difficulty could be lower, is if it took longer to get that last block
3024 2011-05-05 17:02:04 <luke-jr> Diablo-D3 says we should replace Bitcoin scripts with JVM
3025 2011-05-05 17:02:15 <Diablo-D3> I did no such thing
3026 2011-05-05 17:02:15 <xelister> and add OpenCL there
3027 2011-05-05 17:02:20 <luke-jr> :D
3028 2011-05-05 17:02:21 <xelister> then the Chainblock would be mining itself \o/
3029 2011-05-05 17:02:31 <Diablo-D3> zomg skynet
3030 2011-05-05 17:02:41 <xelister> then it gets AIs and overtakes the world
3031 2011-05-05 17:02:41 <UukGoblin> oh hang on... I've just invented something... A pure bitcoin kickstarter... if not enough funds are gathered until a pre-set time, all of them get released to the owners, otherwise they go to the project devs...
3032 2011-05-05 17:02:42 <devrandom> luke-jr if it took longer to get the last block before the difficulty change
3033 2011-05-05 17:02:53 <devrandom> luke-jr - but the subsequent blocks can be faster
3034 2011-05-05 17:03:16 <diki> Actually AI are a must in today's age
3035 2011-05-05 17:03:18 <luke-jr> devrandom: only to the degree that the last block waited
3036 2011-05-05 17:03:28 <diki> i am sick of probable causes for errors
3037 2011-05-05 17:03:38 <diki> i need to real problem in a program or other failure
3038 2011-05-05 17:03:51 <diki> scratch the last part
3039 2011-05-05 17:04:09 <diki> If AIs were invented we will no longer see just an error if a program couldnt execute
3040 2011-05-05 17:04:11 agricocb has quit (Ping timeout: 240 seconds)
3041 2011-05-05 17:04:14 <diki> we will see the detailed problem
3042 2011-05-05 17:04:17 <devrandom> luke-jr - anyway, I'm saying this based on looking at the code
3043 2011-05-05 17:04:26 <diki> Not just "Error, something happened" but the cause of the problem
3044 2011-05-05 17:04:29 <devrandom> luke-jr - it does look at difficulty sum, not height
3045 2011-05-05 17:04:40 <topi`> anyone seen tcatm around? I'm rewriting his sha256_4way.c in NEON assembler
3046 2011-05-05 17:04:46 <wumpus> diki: you don't need AIs for that, just better error reporting/handling :P
3047 2011-05-05 17:04:47 <topi`> or rather, intrinsics
3048 2011-05-05 17:04:55 <CIA-30> bitcoin: genjix exper * re7f893114e3a intersango/switcher.php: BUGFIX: do not display withdraw page when not logged in. http://tinyurl.com/5w5zv5q
3049 2011-05-05 17:04:58 <CIA-30> bitcoin: genjix exper * r59d048130fc5 intersango/DATABASE: new entries for international withdrawal. http://tinyurl.com/688abd2
3050 2011-05-05 17:05:01 <CIA-30> bitcoin: genjix exper * r49ce8bbb4add intersango/withdraw.php: international_withdraw now supports extra fields. http://tinyurl.com/6ak4xqo
3051 2011-05-05 17:05:03 <diki> let's say you have a problem with your internet connection
3052 2011-05-05 17:05:04 skyewm has quit (Read error: Operation timed out)
3053 2011-05-05 17:05:04 <CIA-30> bitcoin: genjix exper * r983ec0a73e07 intersango/view_request.php: view_request (international) supports extra fields. http://tinyurl.com/6fsdv7m
3054 2011-05-05 17:05:09 <diki> AIs are needed to tell you what is the problem
3055 2011-05-05 17:05:13 <diki> and i mean the exact problem
3056 2011-05-05 17:05:33 joepie92 has quit (Quit: Nettalk6 - www.ntalk.de)
3057 2011-05-05 17:05:42 <diki> Windows 7 comes close to that, but a real AI would handle it much better
3058 2011-05-05 17:05:44 <UukGoblin> diki, disagreed, people should always be able to find problems themselves.
3059 2011-05-05 17:05:51 <diki> And games would become better
3060 2011-05-05 17:06:01 <diki> cause an AI will be able to create new content on the fly
3061 2011-05-05 17:06:03 <devrandom> luke-jr - you could have a side chain where the last 2017 blocks are deleted, then the next block would cause a transition to difficulty 1
3062 2011-05-05 17:06:06 <wumpus> it's like 2001's Dave, but for your internet connection?
3063 2011-05-05 17:06:08 <diki> think of new plot
3064 2011-05-05 17:06:09 joepie92 has joined
3065 2011-05-05 17:06:37 <Diablo-D3> devrandom: yeah, but whichever chain has the highest aggregate difficulty wins.
3066 2011-05-05 17:06:59 <devrandom> Diablo-D3 - that's what I'm trying to say to luke-jr
3067 2011-05-05 17:07:52 <Diablo-D3> ahh
3068 2011-05-05 17:07:57 <Diablo-D3> well, luke tends to troll >_>
3069 2011-05-05 17:08:07 <Diablo-D3> hes like the dkingston of the bitcoin community
3070 2011-05-05 17:08:13 LightRider is now known as LightRider|afk
3071 2011-05-05 17:09:11 <devrandom> Diablo-D3 - we're all kinda eccentric here
3072 2011-05-05 17:09:56 <devrandom> luke-jr - then after the transition to 1 you could have 4032 blocks at the lower diff every 5 minutes
3073 2011-05-05 17:09:57 <diki> Trololol, eh? :D
3074 2011-05-05 17:09:59 agricocb has joined
3075 2011-05-05 17:10:09 <devrandom> luke-jr - so you'll get ahead of the main chain in terms of number of blocks
3076 2011-05-05 17:10:20 <devrandom> but still be behind in terms of difficulty sum
3077 2011-05-05 17:10:28 <devrandom> that's why it's done by difficulty sum
3078 2011-05-05 17:10:35 phantomcircuit has quit (Quit: Leaving)
3079 2011-05-05 17:10:44 phantomcircuit has joined
3080 2011-05-05 17:12:23 agricocb1 has joined
3081 2011-05-05 17:13:19 <kika> devrandom: so basically if i run a node in Australia, that only connects to other Australian nodes, and all the other australian nodes arent advertising me any blocks they get from the main chain ( outside Australia ) and only advertise me side blocks inside australia, then all my transactions will become invalid as soon as i my node can get access to the main chain and all the txs will be invalid in the main chain right?
3082 2011-05-05 17:13:57 <devrandom> yeah, your node will do a big reorganize once you see the heavier chain
3083 2011-05-05 17:14:10 agricocb has quit (Ping timeout: 240 seconds)
3084 2011-05-05 17:14:13 <devrandom> and any txs that conflict with the main chain will become forever invalid
3085 2011-05-05 17:14:14 <kika> devrandom: ill loose all my coins?
3086 2011-05-05 17:14:21 <devrandom> (i.e. double spends)
3087 2011-05-05 17:14:40 <devrandom> you'd just lose stuff that was sent to you in that was spent differently in the main chain
3088 2011-05-05 17:15:06 <devrandom> i.e. someone sent to you in aus, but maliciously sent to someone else outside aus
3089 2011-05-05 17:15:18 fetokun has joined
3090 2011-05-05 17:15:46 lumos has joined
3091 2011-05-05 17:15:48 <devrandom> you'd also lose any mined coins
3092 2011-05-05 17:16:09 <devrandom> for the time that you were mining on the side chain
3093 2011-05-05 17:16:39 <genjix> hey devrandom. arent you the original developer of i2p?
3094 2011-05-05 17:16:48 <devrandom> nope
3095 2011-05-05 17:16:52 <kika> devrandom: i see those blocks i have will become invalid for me right
3096 2011-05-05 17:17:04 <kika> devrandom: all the stuff on those blocks will become useless
3097 2011-05-05 17:17:09 <devrandom> genjix uh oh, does my nick clash with someone?
3098 2011-05-05 17:17:13 <BlueMatt> ;;seen jaromil
3099 2011-05-05 17:17:13 <gribble> jaromil was last seen in #bitcoin-dev 2 days, 7 hours, 27 minutes, and 25 seconds ago: <jaromil> :) i was sorry but had stuff to do and was very annoyed by the weather
3100 2011-05-05 17:17:38 <kika> devrandom: right ? all my aus blocks will become useless, right ?
3101 2011-05-05 17:17:39 <tippenein> ;;bc
3102 2011-05-05 17:17:40 <gribble> Error: "bc" is not a valid command.
3103 2011-05-05 17:17:44 <genjix> devrandom: maybe im mistaken then
3104 2011-05-05 17:18:40 <devrandom> kika - well, if someone sent you coins in aus, and they didn't on purpose double spent them outside aus, then this transaction will be eventually incorporated in the main chain
3105 2011-05-05 17:18:45 <devrandom> kika - so you'd not lose anything
3106 2011-05-05 17:18:47 <kika> devrandom: so how come i can know if my node is working on the real main chain?
3107 2011-05-05 17:19:37 <kika> devrandom: how do i know if my node is on the main or in a side chain?
3108 2011-05-05 17:19:51 <devrandom> kika - you can't know for sure. the only way to increase your confidence is to connect to a variety of nodes and to check the blockchain using some other way, like blockexplorer
3109 2011-05-05 17:20:18 <kika> devrandom: if someone hacks blockexplorer then we are fried
3110 2011-05-05 17:20:18 <devrandom> another way to tell you are on a side chain is a sudden drop in the rate of blocks
3111 2011-05-05 17:20:25 <devrandom> since the side chain will have less hash power
3112 2011-05-05 17:20:49 <devrandom> the rate would have to drop to less than 50% of the original rate
3113 2011-05-05 17:21:00 <kika> devrandom: so you say in side chain a block wont be generated every 10 min right ?
3114 2011-05-05 17:21:08 <devrandom> right
3115 2011-05-05 17:21:30 <jgarzik> BlueMatt: check out mingw32 fedora setup
3116 2011-05-05 17:21:33 <BlueMatt> gavinandresen: you asked a while ago about autotools, been looking through it and it looks like it doesnt even try any kind of win32 stuff
3117 2011-05-05 17:21:38 <jgarzik> BlueMatt: proper lib path automatically happens
3118 2011-05-05 17:21:56 <lulzplzkthx> so i'm rying to be all cutting edge with html5 and stuff... using input type='email', oh yeah. except then chrome decides if it's not a valid email it'll give the nastiest little popup i've ever seen regarding it >_<
3119 2011-05-05 17:21:56 <BlueMatt> jgarzik: its an issue in the autotools branch, not mingw/etc
3120 2011-05-05 17:22:04 <BlueMatt> AFAICT
3121 2011-05-05 17:22:06 <jgarzik> BlueMatt: for mingw32, you shouldn't need much special stuff. you run mingw32-configure to set all paths properly
3122 2011-05-05 17:22:21 <BlueMatt> you run what now?
3123 2011-05-05 17:22:25 <jgarzik> BlueMatt: you don't have to do anything special, just to get libs to work
3124 2011-05-05 17:22:43 <jgarzik> BlueMatt: mingw32-configure is a special script that runs configure with all the proper paths for your system
3125 2011-05-05 17:22:54 <BlueMatt> hm, dont have that over here
3126 2011-05-05 17:22:57 <jgarzik> BlueMatt: see cpuminer, which is built on win32 and linux, using this system
3127 2011-05-05 17:23:05 <jgarzik> BlueMatt: -./configure
3128 2011-05-05 17:23:05 <BlueMatt> the problem I have with fedora is its not easy to xcompile boost or wx on it
3129 2011-05-05 17:23:09 <tippenein> seems to me an online Ewallet is pointless if yo know how to backup files... is this true?
3130 2011-05-05 17:23:10 <jgarzik> BlueMatt: +mingw32-configure
3131 2011-05-05 17:23:15 <kika> right now 122065 blocks exist?
3132 2011-05-05 17:23:25 <BlueMatt> kika: ,,bc,blocks
3133 2011-05-05 17:23:28 <jgarzik> BlueMatt: boost is already cross-compiled for your :)
3134 2011-05-05 17:23:28 <gribble> 122065
3135 2011-05-05 17:23:29 <jgarzik> you
3136 2011-05-05 17:23:41 <jgarzik> BlueMatt: wx is not trivial, but it's doable using the standard instructions
3137 2011-05-05 17:23:46 <BlueMatt> jgarzik: can you lookup what all the script does if you have time?
3138 2011-05-05 17:24:01 <BlueMatt> meh, still easier on ubuntu afaict
3139 2011-05-05 17:24:18 <BlueMatt> plus fedora keeps crashing when I try to run it in a vm so...
3140 2011-05-05 17:24:19 <jgarzik> BlueMatt: <shrug> once I recompiled openssl, everything else was easy
3141 2011-05-05 17:24:31 <BlueMatt> openssl xcompiles really easily
3142 2011-05-05 17:24:32 <kika> so my bitcoin client to be in the main chain will need to indicate it has 122065 blocks, right?
3143 2011-05-05 17:24:58 <BlueMatt> (on ubuntu)
3144 2011-05-05 17:25:00 <xelister> blacklotus anti ddos hosting @ mtgox is not working?
3145 2011-05-05 17:25:07 <devrandom> kika - yes
3146 2011-05-05 17:25:28 <kika> devrandom: if it has more or less blocks then that means im not on the main chain?
3147 2011-05-05 17:25:45 <devrandom> if it has less, it could be that you didn't download all of them yet
3148 2011-05-05 17:26:00 <devrandom> if more, then something is fishy
3149 2011-05-05 17:26:18 <kika> devrandom: i think someone needs to create a desktop app to see like blockexplorer the blocks that each client downloaded so i can explore my downloaded blocks
3150 2011-05-05 17:26:50 <devrandom> you can run the following command:
3151 2011-05-05 17:27:00 <devrandom> bitcoind --printblockindex
3152 2011-05-05 17:27:06 <devrandom> bitcoin must not be running
3153 2011-05-05 17:27:21 <devrandom> the output is in ~/.bitcoin/debug.log I think
3154 2011-05-05 17:27:22 lumos is now known as dick
3155 2011-05-05 17:27:28 <kika> devrandom: that will print all my blocks??
3156 2011-05-05 17:27:35 <kika> devrandom: or only the headers? including the txs and everything?
3157 2011-05-05 17:27:38 <devrandom> you can compare the tail of that to #bitcoinwatch or blockexplorer
3158 2011-05-05 17:27:47 <devrandom> just the block hashes
3159 2011-05-05 17:27:53 BitMark has joined
3160 2011-05-05 17:28:00 <kika> k
3161 2011-05-05 17:28:00 <devrandom> which is enough if we trust hashing
3162 2011-05-05 17:28:00 <kika> thxs
3163 2011-05-05 17:28:04 <kika> good bye
3164 2011-05-05 17:28:05 <devrandom> sure
3165 2011-05-05 17:28:11 <kika> we talk soon
3166 2011-05-05 17:28:14 <kika> i need to leave
3167 2011-05-05 17:28:17 kika has quit (Quit: Page closed)
3168 2011-05-05 17:28:26 <devrandom> later
3169 2011-05-05 17:28:29 <jgarzik> BlueMatt: mingw32-configure is definitely part of standard mingw32 distribution, and not Fedora. might want to file a bug.
3170 2011-05-05 17:28:40 BitMark has quit (Client Quit)
3171 2011-05-05 17:28:41 Aahzmundus has joined
3172 2011-05-05 17:29:34 <BlueMatt> jgarzik: hm, on fedora it appears to call rpm ... ie wouldnt work on ubuntu?
3173 2011-05-05 17:29:48 BitMark has joined
3174 2011-05-05 17:30:24 bgeron has quit (Ping timeout: 252 seconds)
3175 2011-05-05 17:34:27 <BlueMatt> jgarzik: where do you see that mingw32-configure is part of standard mingw32 distribution?
3176 2011-05-05 17:34:42 <Diablo-D3> isnt that part of the msys distro?
3177 2011-05-05 17:34:49 <BlueMatt> I think so
3178 2011-05-05 17:35:00 <Diablo-D3> last time I decided to give a fuck, it was
3179 2011-05-05 17:35:03 AnonX has joined
3180 2011-05-05 17:36:29 <jaromil> BlueMatt: hola
3181 2011-05-05 17:36:29 <topi`> jgarzik: the sha256_4way.c is using 128-bit SSE2 xor (mm_xor_si128) but AFAIU you'd get the same result with NEON 32x4 eor
3182 2011-05-05 17:36:48 <topi`> jgarzik: I am assuming you know SSE2 like your pockets?
3183 2011-05-05 17:37:07 <topi`> there is no 128-bit xor in NEON
3184 2011-05-05 17:38:21 <jaromil> FWIW i'll be on-air for the hacktivist hour here at 21:30 CET (in 2 hours from now) http://signal.hackerspaces.org/ i guess we'll also talk about bitcoin ;)
3185 2011-05-05 17:38:33 <topi`> jaromil: cool ! :)
3186 2011-05-05 17:38:45 <BlueMatt> jaromil: nice
3187 2011-05-05 17:38:56 <BlueMatt> any progress on autotools win32/cross compiling?
3188 2011-05-05 17:39:12 sp1ky has quit (Ping timeout: 252 seconds)
3189 2011-05-05 17:40:08 <jaromil> BlueMatt: any progress would pop on the pull req, however no. it's at the state you first tried it. i cannot reproduce easily a build environment, too many things on my desk. if we are in a urge to merge i suggest we make sure the old build system works as usual
3190 2011-05-05 17:40:34 <jaromil> all doe must not be too difficult... wondering about the problem, should be related to the fact --with-boost variable paths aren't honored
3191 2011-05-05 17:40:41 <BlueMatt> no, I dont think theres any urgency (well, some but not too much)
3192 2011-05-05 17:41:02 <topi`> since bitcoin is such a big community one would think there would be *someone* who could hack up win32 autotools support
3193 2011-05-05 17:41:15 <BlueMatt> the only problem I see is it automatically adds -I/usr/... and doesnt let me skip that and replace with -I/usr/i586-mingw32msvc/...
3194 2011-05-05 17:41:28 <BlueMatt> and CPPFLAGS or similar just adds it to the end which doesnt work
3195 2011-05-05 17:42:25 eternal11 has quit (Quit: Leaving.)
3196 2011-05-05 17:43:07 amiller has joined
3197 2011-05-05 17:43:20 <jaromil> its about reproducing your build environment on my side and then smacking it a bit into the m4 code or something
3198 2011-05-05 17:43:44 <jaromil> one step i did yesterday i've reinstalled my quad tower and have it ready for some crunching of this kind
3199 2011-05-05 17:43:57 <jaromil> considering i'm not a miner i'll devolve cycles to that as soon as i have time
3200 2011-05-05 17:44:01 <jaromil> realistically not until the we
3201 2011-05-05 17:44:16 <jaromil> if someone else does it i'm happy
3202 2011-05-05 17:44:27 <jaromil> as topi says, good occasion to get involved
3203 2011-05-05 17:44:33 <BlueMatt> sadly, I cant make heads or tails of how to make a configure.ac script...
3204 2011-05-05 17:44:44 <BlueMatt> (there seems to be almost no documentation on the topic)
3205 2011-05-05 17:44:47 NOTAL has joined
3206 2011-05-05 17:44:48 gavinandresen has quit (Quit: gavinandresen)
3207 2011-05-05 17:45:06 <jaromil> all i can say, after 10 years of that: its a horror
3208 2011-05-05 17:45:08 marlowe has quit (Quit: leaving)
3209 2011-05-05 17:45:31 marlowe has joined
3210 2011-05-05 17:46:18 <BlueMatt> hm, signal.hackerspaces.org's IPv6 doesnt work making it lag like hell
3211 2011-05-05 17:46:26 <BlueMatt> IPv4 is fine though
3212 2011-05-05 17:48:08 berto- has joined
3213 2011-05-05 17:48:45 <berto-> what's the best ubuntu distro to go for mining with an AR-6990?
3214 2011-05-05 17:52:53 <phantomcircuit> BlueMatt, using he?
3215 2011-05-05 17:53:24 <BlueMatt> phantomcircuit: yea
3216 2011-05-05 17:53:32 <BlueMatt> working fine (Im on freenode via v6)
3217 2011-05-05 17:53:32 <phantomcircuit> they've been having problems
3218 2011-05-05 17:53:48 <BlueMatt> you mean he bgp problems?
3219 2011-05-05 17:53:55 <phantomcircuit> yeah
3220 2011-05-05 17:54:03 agricocb1 has quit (Quit: Leaving.)
3221 2011-05-05 17:54:05 <phantomcircuit> yesterday i was able to access ipv6.google.com but nothing else
3222 2011-05-05 17:54:12 <BlueMatt> hm, odd. they are pretty much the biggest v6 provider anywhere
3223 2011-05-05 17:54:14 <phantomcircuit> and nobody could access my side
3224 2011-05-05 17:55:49 <BlueMatt> hm, odd. Though thats not exactly a bgp problem
3225 2011-05-05 17:56:12 <BlueMatt> that sounds more like a personal problem on he's end (with tunnelbroker, maybe not he itself)
3226 2011-05-05 17:56:21 <BlueMatt> (well he's bgp/peers)
3227 2011-05-05 18:01:47 Incitatus has quit (Quit: Leaving)
3228 2011-05-05 18:02:30 AnonX has quit (Ping timeout: 252 seconds)
3229 2011-05-05 18:07:39 agricocb has joined
3230 2011-05-05 18:08:47 DukeOfURL has quit (Read error: Operation timed out)
3231 2011-05-05 18:09:40 Teppy has left ()
3232 2011-05-05 18:09:44 Marcel has left (HSD!~Marcel|HS@router2.hsdev.com|)
3233 2011-05-05 18:09:52 phantomcircuit has quit (Read error: Operation timed out)
3234 2011-05-05 18:10:07 phantomcircuit has joined
3235 2011-05-05 18:10:14 ArtForzZz has joined
3236 2011-05-05 18:11:05 MartianW has joined
3237 2011-05-05 18:13:38 ArtForz has quit (Ping timeout: 240 seconds)
3238 2011-05-05 18:17:19 amiller has quit (Ping timeout: 246 seconds)
3239 2011-05-05 18:18:48 <anarchyx> ;;bc,stats
3240 2011-05-05 18:18:50 <gribble> Current Blocks: 122075 | Current Difficulty: 109670.13329248 | Next Difficulty At Block: 122975 | Next Difficulty In: 900 blocks | Next Difficulty In About: 4 days, 12 hours, 15 minutes, and 0 seconds | Next Difficulty Estimate: 150780.29001788
3241 2011-05-05 18:19:44 <mtrlt> estimate going up :O
3242 2011-05-05 18:19:55 <mtrlt> wonder if some people are giving up on mining :P
3243 2011-05-05 18:20:40 <MartianW> Probably just variance.
3244 2011-05-05 18:22:00 <BlueMatt> or ever increasing miners
3245 2011-05-05 18:22:20 <mtrlt> or BFI_INT proliferation
3246 2011-05-05 18:22:28 <BlueMatt> that too
3247 2011-05-05 18:23:58 <jgarzik> pretty soon, a 5970 will generate one block a month
3248 2011-05-05 18:24:05 <mtrlt> :(
3249 2011-05-05 18:24:33 <BitMark> has block rate ever been above 12 per hour?
3250 2011-05-05 18:24:53 <Speeder> how someone is making seveal slaes with the same order?
3251 2011-05-05 18:25:03 <Speeder> there are since a long time ago a 3 BTC bid
3252 2011-05-05 18:25:11 <Speeder> yet I see multiple 3 BTC buys in that bid
3253 2011-05-05 18:25:15 <Speeder> maybe a super-fast bot?
3254 2011-05-05 18:25:26 <Speeder> (it is kinda getting annoying, since it is blocking my sales...)
3255 2011-05-05 18:25:49 <BitMark> Speeder: might be a darkpool order in front of that order
3256 2011-05-05 18:25:57 <Speeder> :/
3257 2011-05-05 18:26:02 <mtrlt> :(
3258 2011-05-05 18:26:05 <BitMark> Speeder: need to offer a lower price then
3259 2011-05-05 18:26:06 * Speeder really hates darkpools
3260 2011-05-05 18:26:20 <Speeder> darkpools have priority, they are annoying
3261 2011-05-05 18:26:22 <Speeder> they block the market
3262 2011-05-05 18:26:27 <BitMark> no they dont
3263 2011-05-05 18:26:31 <Speeder> they do
3264 2011-05-05 18:26:44 <Speeder> with a darkpool you do not get into a queue like everyone else
3265 2011-05-05 18:26:52 <Speeder> you can block the market as you wish
3266 2011-05-05 18:27:20 danlucraft has joined
3267 2011-05-05 18:29:40 MartianW has left ("Bye all.")
3268 2011-05-05 18:31:31 <gjs278> ;;bc,mtgox
3269 2011-05-05 18:31:31 <gribble> {"ticker":{"high":3.5,"low":3.365,"vol":14401,"buy":3.4009,"sell":3.405,"last":3.41}}
3270 2011-05-05 18:33:13 <gjs278> I really wish coincard wasn't shutdown because there's no way in hell I'm selling my coins to mtgox if they're still going to have loading issues
3271 2011-05-05 18:33:23 <gjs278> >Sorry, you must have Javascript enabled to use Mt. Gox
3272 2011-05-05 18:33:26 <gjs278> everytime I see that shit
3273 2011-05-05 18:33:28 <gjs278> I die
3274 2011-05-05 18:34:32 dvide has quit ()
3275 2011-05-05 18:35:20 AnonX has joined
3276 2011-05-05 18:42:04 sipa has quit (Read error: Connection reset by peer)
3277 2011-05-05 18:42:44 sipa has joined
3278 2011-05-05 18:42:55 sipa has quit (Read error: Connection reset by peer)
3279 2011-05-05 18:46:10 pirrr has quit (Ping timeout: 240 seconds)
3280 2011-05-05 18:46:56 <cosurgi> anybody with multiple cards is using AMDOverdriveCtrl ?
3281 2011-05-05 18:47:08 <Diablo-D3> should work
3282 2011-05-05 18:47:28 <topi`> jaromil: i'm streaming the ogg from hackerspace.org. when do you start your preaching :)
3283 2011-05-05 18:47:44 sipa has joined
3284 2011-05-05 18:48:08 <cosurgi> I have problems underclocking mem, so I want to try AMDOverdriveCtrl, but I have no console access, and I must try this with a config file.
3285 2011-05-05 18:48:14 <xelister> cosurgi: well works with both cores of r5970
3286 2011-05-05 18:48:25 <cosurgi> I mean - only remote access
3287 2011-05-05 18:48:44 <xelister> btw, ArtForzZz how to use AMDOver.. to set mem clock 300 when normally only 500 can be set there
3288 2011-05-05 18:49:10 <cosurgi> I found OVERDRIVE_PROFILE files, but they are for single GPU only. I need it for three cards: 5850+5870+5870
3289 2011-05-05 18:49:14 pirrr has joined
3290 2011-05-05 18:49:32 marlowe has quit (Ping timeout: 260 seconds)
3291 2011-05-05 18:49:35 <molecular> hey cosurgi... didn't work for 3 cards?
3292 2011-05-05 18:49:42 <cosurgi> molecular: no, it segfaults
3293 2011-05-05 18:50:13 <devrandom> BlueMatt - new text about the build process on https://gitian.org/
3294 2011-05-05 18:50:34 <cosurgi> molecular: I debugged it a bit, and I located the segfault: it happens exactly when ADL_Overdrive5_ODPerformanceLevels_Set is called.
3295 2011-05-05 18:51:20 <cosurgi> molecular: so I am helpless here. I tried changing various things, like only accessing the first card, and skip others, etc..
3296 2011-05-05 18:51:21 <molecular> cosurgi, strange. same driver versions and stuff on that system than on the one with the one card?
3297 2011-05-05 18:51:36 <cosurgi> molecular: yes. same versions
3298 2011-05-05 18:51:54 <BlueMatt> devrandom: nice, working on getting this crap to cross compile in the new autotools stuff so that it can be done soon, I have a branch now which should work but its completely untested so...
3299 2011-05-05 18:51:59 <diki> for me, downclocking the mem while clocking the core causes crashes
3300 2011-05-05 18:52:03 <BlueMatt> (well with the current build system it should work)
3301 2011-05-05 18:52:12 <cosurgi> molecular: those debian installs were cloned from each other, after I configured the first one with fglrx.
3302 2011-05-05 18:52:16 <diki> not immediately, but they occur
3303 2011-05-05 18:52:20 AStove has joined
3304 2011-05-05 18:53:05 <devrandom> BlueMatt - cross compiling sounds like fun...
3305 2011-05-05 18:53:06 <molecular> cosurgi, that's weird.. I can't image it having to do with the fact that there are more cards either.
3306 2011-05-05 18:53:19 <BlueMatt> devrandom: not...at...all
3307 2011-05-05 18:53:39 <cosurgi> molecular: so I want to try with AMDOverdriveCtrl from command line. But I need a config file with 3 cards. I don't know how to add more cards to this config file. Here's a config file that I found on forums, with one card:
3308 2011-05-05 18:53:43 <BlueMatt> devrandom: wrote this, untested but theres a chance it works: https://github.com/TheBlueMatt/bitcoin/blob/crosscompile/contrib/gitian-win32.yml
3309 2011-05-05 18:53:45 <cosurgi> <?xml version="1.0" encoding="utf-8"?>
3310 2011-05-05 18:53:45 <cosurgi> <OVERDRIVE_PROFILE> <PERFORMANCE_LEVEL level="2" gpu="90000" mem="30000" voltage="1088"/> <PERFORMANCE_LEVEL level="1" gpu="55000" mem="30000" voltage="1038"/> <PERFORMANCE_LEVEL level="0" gpu="15700" mem="30000" voltage="1000"/> <FAN_SETTING percentage="80"/> <FAN_CTRL enabled="no"/>
3311 2011-05-05 18:53:50 <cosurgi> </OVERDRIVE_PROFILE>
3312 2011-05-05 18:53:53 <cosurgi> now, how to add two more cards?
3313 2011-05-05 18:53:58 <cosurgi> http://www.bitcoin.org/smf/index.php?topic=6672.0
3314 2011-05-05 18:54:10 <cosurgi> http://www.bitcoin.org/smf/index.php?topic=6672.msg97668#msg97668
3315 2011-05-05 18:54:24 <devrandom> BlueMatt - just trying to be encouraging ;)
3316 2011-05-05 18:54:34 <molecular> cosurgi, i tried earlier to use overdrive remotely. ended up using x11vnc on miner and vncviewer
3317 2011-05-05 18:54:41 <devrandom> BlueMatt - if it fails you can use on-target to ssh in and manually run steps
3318 2011-05-05 18:54:51 <cosurgi> can anybody paste their <OVERDRIVE_PROFILE> with multiple cards?
3319 2011-05-05 18:54:54 <molecular> cosurgi, maybe it saves a config-file?
3320 2011-05-05 18:55:02 <cosurgi> molecular: I tried x11vnc already. It was extremely slow.
3321 2011-05-05 18:55:08 <BlueMatt> devrandom: alright, Ill get around to trying that soon, but it does theoretically work, I build win32 nightlies on a ubuntu server
3322 2011-05-05 18:55:17 <cosurgi> molecular: yes it saved. But with just one card inside!
3323 2011-05-05 18:55:21 <molecular> cosurgi, same problem here... can't even manage to enter stuff in xterm
3324 2011-05-05 18:55:22 <luke-jr> cosurgi: back on Eligius yet? :P
3325 2011-05-05 18:55:22 <cosurgi> brb
3326 2011-05-05 18:55:24 <devrandom> BlueMatt - cool
3327 2011-05-05 18:55:38 <luke-jr> cosurgi: I play PS2 games with x11vnc all the time
3328 2011-05-05 18:55:48 <gjs278> how could x11vnc be slow
3329 2011-05-05 18:55:54 <gjs278> unless you were on frames 1
3330 2011-05-05 18:56:03 <cosurgi> ok, back
3331 2011-05-05 18:56:08 <luke-jr> well, -f 1 kills performance with VNC or not
3332 2011-05-05 18:56:13 <gjs278> exactly
3333 2011-05-05 18:56:17 <gjs278> but local vnc
3334 2011-05-05 18:56:20 <gjs278> is always very quick
3335 2011-05-05 18:56:26 <luke-jr> yeah
3336 2011-05-05 18:56:40 <luke-jr> if I can play PS2 games through it, it's good enough for anything
3337 2011-05-05 18:56:41 <luke-jr> :P
3338 2011-05-05 18:56:48 <cosurgi> hm, ok. I can try x11vnc with mining stopped for a while.
3339 2011-05-05 18:57:01 <gjs278> don't even stop mining, just set frames to 60
3340 2011-05-05 18:57:24 <cosurgi> luke-jr: sorry, not yet. I wanted to wait until your pool gets to 30
3341 2011-05-05 18:58:32 <luke-jr> cosurgi: 30?
3342 2011-05-05 18:58:46 <diki> do yor gpus also mine at 98% load?
3343 2011-05-05 18:58:54 <diki> mine does
3344 2011-05-05 18:58:55 <gjs278> around that
3345 2011-05-05 18:58:55 Blitzboom has quit (Remote host closed the connection)
3346 2011-05-05 18:58:57 <gjs278> 98 to 100
3347 2011-05-05 18:59:04 <diki> never 100
3348 2011-05-05 18:59:12 Blitzboom has joined
3349 2011-05-05 18:59:12 Blitzboom has quit (Changing host)
3350 2011-05-05 18:59:13 Blitzboom has joined
3351 2011-05-05 18:59:15 <cosurgi> luke-jr: hashrate. now you have 12...
3352 2011-05-05 18:59:25 <cosurgi> ;;bc,lukepool
3353 2011-05-05 18:59:26 <luke-jr> cosurgi: oh, it was 31 a few days ago
3354 2011-05-05 18:59:26 <gribble> 12569937.6196
3355 2011-05-05 18:59:32 <luke-jr> but that's kinda⦠chicken and egg
3356 2011-05-05 18:59:46 <diki> isnt that 100+?
3357 2011-05-05 18:59:49 <diki> ghash
3358 2011-05-05 18:59:59 <luke-jr> 12.5 GH
3359 2011-05-05 19:00:02 manifold has joined
3360 2011-05-05 19:00:03 amiller has joined
3361 2011-05-05 19:01:04 eao has joined
3362 2011-05-05 19:01:05 eternal1 has joined
3363 2011-05-05 19:01:18 <cosurgi> luke-jr: ok, hmm.. I had strange things in logs from your pool. maybe because you were experimenting: I received two times exactly the same getwork
3364 2011-05-05 19:01:32 <cosurgi> luke-jr: two different GPUs received it. but it was the same getwork
3365 2011-05-05 19:01:38 <diki> how does one understand when a piece of code needs to be compiled to a dll?
3366 2011-05-05 19:01:42 <cosurgi> luke-jr: it happened 75 times.
3367 2011-05-05 19:01:53 <luke-jr> cosurgi: how do you identify that?
3368 2011-05-05 19:01:58 <Diablo-D3> diki: when it needs to be linked to multiple applications
3369 2011-05-05 19:02:01 <cosurgi> luke-jr: I am logging getworks
3370 2011-05-05 19:02:14 <cosurgi> luke-jr: for experimenting.
3371 2011-05-05 19:02:14 <diki> and when does that happen?
3372 2011-05-05 19:02:19 <Diablo-D3> diki: libraries, frameworks.
3373 2011-05-05 19:02:26 <gjs278> Diablo-D3 that gigabyte card is $154 now with mail in rebate
3374 2011-05-05 19:02:28 <luke-jr> cosurgi: like, exactly the same?
3375 2011-05-05 19:02:33 <Diablo-D3> gjs278: hrrrrrrm
3376 2011-05-05 19:02:35 <gjs278> good for any purchase from may 1st to may 30th
3377 2011-05-05 19:02:35 <cosurgi> luke-jr: yes. exactly
3378 2011-05-05 19:02:42 <luke-jr> jgarzik: ^
3379 2011-05-05 19:02:46 <cosurgi> luke-jr: a wasted hashpower.
3380 2011-05-05 19:03:25 <luke-jr> cosurgi: what's your address?
3381 2011-05-05 19:03:36 <cosurgi> hold on..
3382 2011-05-05 19:04:05 <cosurgi> luke-jr: see msg
3383 2011-05-05 19:04:47 <cosurgi> luke-jr: I am not dating my logs, but I could try to roughly identify the time when it happened.
3384 2011-05-05 19:05:17 <topi`> jgarzik: online?
3385 2011-05-05 19:05:27 <luke-jr> cosurgi: I see less than 0.005% duplicate shares from you. I would expect to see nearly 50% if two equal miners were getting the same work
3386 2011-05-05 19:05:35 <Diablo-D3> yeah what luke said
3387 2011-05-05 19:05:40 <topi`> jgarzik: what was the last version of poold.py that you touched?
3388 2011-05-05 19:05:44 <Diablo-D3> http://www.newegg.com/Product/Product.aspx?Item=N82E16814125357
3389 2011-05-05 19:05:45 <Diablo-D3> LOL JESUS
3390 2011-05-05 19:05:50 <Diablo-D3> three fucking fans
3391 2011-05-05 19:05:51 <topi`> i'm trying to setup my own pool
3392 2011-05-05 19:06:24 <cosurgi> luke-jr: you don't understand. It's not happening all the time.
3393 2011-05-05 19:06:26 <luke-jr> cosurgi: in fact, you have less invalid shares than average
3394 2011-05-05 19:06:27 <Diablo-D3> VROOOM
3395 2011-05-05 19:06:28 <luke-jr> cosurgi: ?
3396 2011-05-05 19:06:38 NOTAL has quit (Read error: Operation timed out)
3397 2011-05-05 19:06:52 skyewm has joined
3398 2011-05-05 19:07:04 RazielZ has quit (Quit: Leaving)
3399 2011-05-05 19:07:08 <cosurgi> luke-jr: it only happened, to be exact, 80 times. Out of which 31 were solved (submitted share) and 49 were unsolved (didn't find a nonce)
3400 2011-05-05 19:07:21 <luke-jr> oh
3401 2011-05-05 19:07:42 NOTAL has joined
3402 2011-05-05 19:07:54 <Diablo-D3> cosurgi: er, how are you logging this?
3403 2011-05-05 19:08:02 <cosurgi> luke-jr: I can paste you a midstate of those duplicate getworks, that;s what I have in logs.
3404 2011-05-05 19:08:08 <Diablo-D3> because like, with my miner, if you kept outputting the same getwork repeatedly
3405 2011-05-05 19:08:10 <luke-jr> sure
3406 2011-05-05 19:08:12 <Diablo-D3> and its a LP miner
3407 2011-05-05 19:08:21 <Diablo-D3> you're printing the same getwork instance repeatedly
3408 2011-05-05 19:08:34 <cosurgi> Diablo-D3: I hacked phoenix for that.
3409 2011-05-05 19:08:44 <Diablo-D3> same thing applies though
3410 2011-05-05 19:08:45 <cosurgi> Diablo-D3: I'm not logging duplicate times.
3411 2011-05-05 19:09:09 <cosurgi> Diablo-D3: because it wasn't happenieng when I wasn';t using luke's pool
3412 2011-05-05 19:09:29 <cosurgi> and I'm logging about 60 000 getworks per day
3413 2011-05-05 19:09:35 <cosurgi> and only 80 were duplicate
3414 2011-05-05 19:10:09 <Diablo-D3> could be just a stupid hiccup
3415 2011-05-05 19:10:28 <cosurgi> yeah, luke-jr was experimenting with pool that day
3416 2011-05-05 19:10:54 <luke-jr> oh
3417 2011-05-05 19:10:59 <luke-jr> could be when I restarted the pool then
3418 2011-05-05 19:11:23 manifold has quit (Remote host closed the connection)
3419 2011-05-05 19:14:38 theymos has joined
3420 2011-05-05 19:16:32 phantomcircuit has quit (Read error: Connection reset by peer)
3421 2011-05-05 19:19:11 tenach has joined
3422 2011-05-05 19:19:11 tenach has quit (Changing host)
3423 2011-05-05 19:19:11 tenach has joined
3424 2011-05-05 19:20:10 <nanotube> UukGoblin: ok, you've got autoop. also: the registration transactions won't get pruned in the bitdns design, since those outputs will never be spent. (they're 0). its the second output that maintains your control of domain that gets spent for updates.
3425 2011-05-05 19:20:31 fimp has quit (Quit: This computer has gone to sleep)
3426 2011-05-05 19:20:45 threepio has quit (Read error: Connection reset by peer)
3427 2011-05-05 19:20:54 <theymos> You could spend a 0-confirmation output if you wanted to, I think.
3428 2011-05-05 19:22:04 <Diablo-D3> 5850 = 151 watts, 6870 = 151 watts
3429 2011-05-05 19:22:05 <Diablo-D3> hrrrrrrm
3430 2011-05-05 19:22:41 <diki> 6870==5770
3431 2011-05-05 19:22:49 <Diablo-D3> diki: nope
3432 2011-05-05 19:23:18 <Diablo-D3> 68xx gives smaller dies, less power usage, less heat output, and about the same 3D speed as the matching 58xx
3433 2011-05-05 19:23:28 <diki> my point is
3434 2011-05-05 19:23:35 <Diablo-D3> your point is pointless
3435 2011-05-05 19:23:38 <diki> it replaces 5770
3436 2011-05-05 19:23:50 <Diablo-D3> diki: no, its an UPGRADE for the 5770
3437 2011-05-05 19:23:54 Incitatus has joined
3438 2011-05-05 19:24:01 <Diablo-D3> its the smallest 6xxx thats an upgrade
3439 2011-05-05 19:24:09 bk128 has joined
3440 2011-05-05 19:24:29 <topi`> damn, it seems the NEON insn set specifies only 16 128bit wide SIMD registers. If I look at tcatm's 4way-sse2 code, there's more than 16 vector variables used.
3441 2011-05-05 19:24:39 theymos has quit (Remote host closed the connection)
3442 2011-05-05 19:24:49 <topi`> w0-w15 and then a,b,c,d,e,f and friends
3443 2011-05-05 19:25:06 <Diablo-D3> topi`: yup.
3444 2011-05-05 19:25:14 <jaromil> topi`: i'm sitting there. but i don't preach :) and even if i would, with walter here i could never, we have often opposite points of view :)
3445 2011-05-05 19:25:17 <Diablo-D3> diki: dude, I explain this shit to people on a daily basis, I know what Im talking about
3446 2011-05-05 19:25:19 <topi`> that's going to suck if regs need to be shuffled.
3447 2011-05-05 19:25:30 <Diablo-D3> diki: _I wrote the second most used miner_
3448 2011-05-05 19:25:31 <diki> no you dont seem to
3449 2011-05-05 19:25:35 <topi`> does SSE2 have more than 16 128b registers then
3450 2011-05-05 19:26:11 Aahzmundus has quit (Read error: Connection reset by peer)
3451 2011-05-05 19:26:11 <Diablo-D3> topi`: hes probably reg stuffing like a faggot, or windowing them
3452 2011-05-05 19:26:25 <topi`> ah, I'll just write some stupid code that throws the "extra" regs needed into 2 integer regs
3453 2011-05-05 19:26:33 Aahzmundus has joined
3454 2011-05-05 19:27:05 <topi`> yes, it seems there's windowing, so I might change that
3455 2011-05-05 19:27:07 <Diablo-D3> diki: a 68xx has a smaller die than the 58xx family, costs less to make, uses less watts, outputs less heat, and gives the same 3D power
3456 2011-05-05 19:27:22 <Diablo-D3> topi`: x86 loves to run out of regs, so it cant be THAT different ;)
3457 2011-05-05 19:27:39 <topi`> Diablo-D3: it seems they have reduced the amount of cores somewhat from 6870 w/ regards to the 5x series
3458 2011-05-05 19:27:43 <Diablo-D3> diki: the 68xx series is not an upgrade, but its a replacement.
3459 2011-05-05 19:27:52 <Diablo-D3> topi`: yes, but that means pretty much nothing.
3460 2011-05-05 19:27:55 <tcatm> topi`: you should base your code on a recent opencl miner. it has much better performance than 4way
3461 2011-05-05 19:27:55 phantomcircuit has joined
3462 2011-05-05 19:28:04 <Diablo-D3> topi`: less cores, but more efficient use of them
3463 2011-05-05 19:28:08 <diki> different naming scheme
3464 2011-05-05 19:28:19 <topi`> tcatm: then I need to learn to read opencl intrinsics :)
3465 2011-05-05 19:28:27 <diki> the 4850 was replaced by 5850
3466 2011-05-05 19:28:31 <topi`> tcatm: I take it your code is not the most optimal one for SSE2 then?
3467 2011-05-05 19:28:45 <diki> amd changed the names then
3468 2011-05-05 19:28:49 <Diablo-D3> diki, topi`: 6870 will give you 5870s worth of 3D power, 5850s worth of mining power, and 5850s usage of watts.
3469 2011-05-05 19:28:54 <Diablo-D3> diki: yes, I explain this to people daily
3470 2011-05-05 19:28:58 <diki> so 6870 replaces 5770
3471 2011-05-05 19:29:07 <Diablo-D3> no, 6870 upgrades the 5770
3472 2011-05-05 19:29:17 bitcoiner has quit (Ping timeout: 252 seconds)
3473 2011-05-05 19:29:20 <tcatm> topi`: it was last year. opencl miners don't have any opencl specific code. you can just compile them as C code
3474 2011-05-05 19:29:44 <topi`> tcatm: I guess so, but to use NEON, I must use the NEON intrinsics
3475 2011-05-05 19:29:45 <edcba> instead of explaining daily just write that on webpage and refers to it
3476 2011-05-05 19:29:52 <topi`> the gcc -ftree-vectorize just won't cut it
3477 2011-05-05 19:30:04 <Diablo-D3> diki: I explain to people daily that AMD moved all the numbers up one so x9xx isnt used by one card
3478 2011-05-05 19:30:17 <Diablo-D3> the UPGRADE for the 5870 is a 6970.
3479 2011-05-05 19:30:49 <Diablo-D3> the REPLACEMENT, IE, YOU GET NO ADDITIONAL 3D PERFORMANCE, is the 6870.
3480 2011-05-05 19:30:54 bitcoiner has joined
3481 2011-05-05 19:30:55 <Diablo-D3> this isnt fucking rocket science.
3482 2011-05-05 19:31:20 <phantomcircuit> Diablo-D3, the 6xxx series are a different architecture right?
3483 2011-05-05 19:31:28 <Diablo-D3> phantomcircuit: they're two different ones
3484 2011-05-05 19:31:30 <cosurgi> Diablo-D3: what's the performance difference between 5870 and 6970 ?
3485 2011-05-05 19:31:32 threepio has joined
3486 2011-05-05 19:31:35 ArtForzZz is now known as ArtForz
3487 2011-05-05 19:31:38 <BlueMatt> you know your approaching the problem wrong when you have to run 20 sed 's/.../...' s after running ./configure...:(
3488 2011-05-05 19:31:46 <Diablo-D3> phantomcircuit: 68xx is a mashup of the best parts of 5xxx
3489 2011-05-05 19:31:53 <ArtForz> 6xxx are *3* different architectures
3490 2011-05-05 19:31:57 <Diablo-D3> phantomcircuit: a little 58xx here, a little 57xx there, even some 56xx
3491 2011-05-05 19:32:06 <Diablo-D3> phantomcircuit: so 68xx, basically, its a 5xxx.
3492 2011-05-05 19:32:10 <Diablo-D3> phantomcircuit: 69xx is the new arch
3493 2011-05-05 19:32:50 <Diablo-D3> ArtForz: no
3494 2011-05-05 19:32:55 <Diablo-D3> I do not recognize 67xx as a 6xxx
3495 2011-05-05 19:32:58 <phantomcircuit> so basically amd decided that sane model numbers was just for their cpus
3496 2011-05-05 19:32:58 <phantomcircuit> :|
3497 2011-05-05 19:33:00 <Diablo-D3> its a rebadged 57xx.
3498 2011-05-05 19:33:03 <ArtForz> 67xx == 57xx arch, 68xx is upgreaded 57xx shaders with new frontend, mem controller, tex unit, ..., 69xx has the same stuff as 68xx (some of it doubled up) and the new VLIW4 shaders
3499 2011-05-05 19:33:19 <topi`> tcatm: also, the poclbm *.cl code seems to use W0..W15 and a,b,c,d,e,f,g,h
3500 2011-05-05 19:33:23 <Diablo-D3> phantomcircuit: no, THIS 6xxx wasnt supposed to happen
3501 2011-05-05 19:33:26 <topi`> that's a lot of registers.
3502 2011-05-05 19:33:30 <ArtForz> 65/66xx == same arch as 68xx, just a lot less units
3503 2011-05-05 19:33:40 <Diablo-D3> phantomcircuit: 7xxx was supposed to be 6xxx
3504 2011-05-05 19:33:50 <Diablo-D3> phantomcircuit: 69xx is basically a preview of the 7xxx technology
3505 2011-05-05 19:34:08 <phantomcircuit> i stand by my statement :|
3506 2011-05-05 19:34:18 <Diablo-D3> phantomcircuit: the 7xxx fab size wasnt ready in time
3507 2011-05-05 19:34:32 <Diablo-D3> ArtForz: dont you have a 5850 or 5870 you dont want anymore?
3508 2011-05-05 19:34:36 <ArtForz> nope
3509 2011-05-05 19:34:38 <tcatm> topi`: 4way uses w0..w15 and a..h, too
3510 2011-05-05 19:34:48 <Diablo-D3> ArtForz: do you have ANYTHING you dont want anymore?
3511 2011-05-05 19:34:57 <ArtForz> nope
3512 2011-05-05 19:35:01 <Diablo-D3> ArtForz: not even for btc?
3513 2011-05-05 19:35:04 <topi`> tcatm: is there any useful way of reducing the amount? since NEON only has 16 128b registers
3514 2011-05-05 19:35:05 <ArtForz> the only part that more or less didnt get changed due to 32nm cancellation was 68xx (which should've been 67xx)
3515 2011-05-05 19:35:09 <phantomcircuit> tcatm, topi` those are the variable names used by the sha standard iirc
3516 2011-05-05 19:35:15 <cosurgi> ArtForz: do you have an idea why ADL_Overdrive5_ODPerformanceLevels_Set(..) is working OK on PC with single 5870, but is segfaulting on another PC (cloned debian configuration) with 5830+5870+5870 ?
3517 2011-05-05 19:35:23 <ArtForz> nope
3518 2011-05-05 19:35:30 <Diablo-D3> ArtForz: jesus christ dude :<
3519 2011-05-05 19:35:58 <tcatm> topi`: hrm, you need about 23 or 24 real registers
3520 2011-05-05 19:36:17 <ArtForz> tcatm: at least
3521 2011-05-05 19:36:29 <Diablo-D3> phantomcircuit: btw, you think 69xx is so impressive? wait for 79xx.
3522 2011-05-05 19:36:36 <ArtForz> 8 for a..h, 16 for W... leaves you with 0 temp regs
3523 2011-05-05 19:36:39 <Diablo-D3> phantomcircuit: thats what 69xx was supposed to look like
3524 2011-05-05 19:36:45 <topi`> tcatm: then it becomes infeasible to try to use NEON to accelerate sha256 :(
3525 2011-05-05 19:37:14 <Diablo-D3> phantomcircuit: it should fry everybody's balls off
3526 2011-05-05 19:37:21 <ArtForz> and 69xx had to be kinda crippled due to the 32nm->45nm process "grow"
3527 2011-05-05 19:37:31 <tcatm> topi`: you still have a cache, don't you?
3528 2011-05-05 19:37:42 <Diablo-D3> yeah
3529 2011-05-05 19:37:46 <Diablo-D3> ArtForz: http://www.newegg.com/Product/Product.aspx?Item=N82E16814125318
3530 2011-05-05 19:37:50 <topi`> tcatm: there's only 32k L1 data cache and I think it's write-through
3531 2011-05-05 19:37:51 <Diablo-D3> ArtForz: do you think its worth $155?
3532 2011-05-05 19:38:12 <ArtForz> at 32nm it probably would've had ~2k shaders
3533 2011-05-05 19:38:26 <topi`> tcatm: it will probably end up in L2 cache, which will then take the latencies up to 5-6 cycles
3534 2011-05-05 19:39:00 <ArtForz> Diablo-D3: kinda pricey, but nice cooler
3535 2011-05-05 19:39:14 <Diablo-D3> ArtForz: thats all they basically have left
3536 2011-05-05 19:39:30 <Diablo-D3> its either that, or this 6870
3537 2011-05-05 19:39:32 <Diablo-D3> http://www.newegg.com/Product/Product.aspx?Item=N82E16814125357
3538 2011-05-05 19:40:03 <Diablo-D3> ArtForz: oh, and thats an OC'ed 5850
3539 2011-05-05 19:40:26 <ArtForz> WTF @ that 6870 cooler design
3540 2011-05-05 19:40:38 <Diablo-D3> MORE FANS THAN GOD
3541 2011-05-05 19:40:39 <tcatm> topi`: that's okay if you can pipeline access to the cache
3542 2011-05-05 19:40:40 <ArtForz> check the fin orientation below the fans
3543 2011-05-05 19:40:48 <topi`> trying to analyze the data dependencies in the cl code. it seems that W0 is only used in the calculation of W0, W2, W7 and W15
3544 2011-05-05 19:41:06 <topi`> tcatm: I see what you mean, but anyway this is a complicated problem
3545 2011-05-05 19:41:06 <Diablo-D3> the cheapest 6950 is $243
3546 2011-05-05 19:41:13 <Diablo-D3> er
3547 2011-05-05 19:41:18 <Diablo-D3> $223 after rebate
3548 2011-05-05 19:41:24 <Diablo-D3> ArtForz: so what do you think I should get?
3549 2011-05-05 19:41:44 <ArtForz> hmm, tough question
3550 2011-05-05 19:41:53 <Diablo-D3> although the 6950 has that 200 watt budget
3551 2011-05-05 19:42:05 <ArtForz> I'd probably get something 6-series
3552 2011-05-05 19:42:14 TheKid has joined
3553 2011-05-05 19:42:16 TheKid has quit (Changing host)
3554 2011-05-05 19:42:16 TheKid has joined
3555 2011-05-05 19:42:25 <Diablo-D3> ArtForz: yeah, but with that 5850 I get to stick with sdk 2.1
3556 2011-05-05 19:42:51 <ArtForz> 5xxx is on its way out, and knowing DAAMIT, driver/SDK support is going to get worse in newer versions
3557 2011-05-05 19:42:51 <Diablo-D3> 5850 and 6870 have about the same mhash, so the extra $30 only gets me more 3D speed (and not much)
3558 2011-05-05 19:43:30 <topi`> tcatm: to me it seems that these W0..W15 vars are just working vars, since they get overwritten in the middle of the algo... so A,B,C,D and friends are the "right" state variables for sha256 calculation, right?
3559 2011-05-05 19:43:45 <ArtForz> it really depends if you want a mining-only card or something that can be used for normal stuff later
3560 2011-05-05 19:44:06 <Diablo-D3> well, a 5850 would already double my 3D speed give or take
3561 2011-05-05 19:44:11 <ArtForz> but then 6870 does have a new video decoder engine, more hardware for 3D stuff and dual DP 1.3 outputs
3562 2011-05-05 19:44:26 <Diablo-D3> video decoder engine is pointless, I do that all in CPU
3563 2011-05-05 19:44:29 <Diablo-D3> mplayer > *
3564 2011-05-05 19:44:42 <phantomcircuit> Diablo-D3, meh
3565 2011-05-05 19:44:48 <ArtForz> = 6-head with a single card with active DP -> 2*DVI splitters
3566 2011-05-05 19:44:56 <Diablo-D3> ArtForz: yeah but I have one monitor, and its dvi
3567 2011-05-05 19:45:08 <tcatm> topi`: yep
3568 2011-05-05 19:45:10 <ArtForz> eww
3569 2011-05-05 19:45:12 <Diablo-D3> 6950 really blows up my heat/watt budget
3570 2011-05-05 19:45:24 <Diablo-D3> ArtForz: dont eww me, its a 26" 1920x1200.
3571 2011-05-05 19:45:27 <Diablo-D3> its glorious
3572 2011-05-05 19:45:28 twilno is now known as stilio
3573 2011-05-05 19:45:41 <ArtForz> thats the res of *one* of my screens
3574 2011-05-05 19:45:47 <Diablo-D3> screw you >_>
3575 2011-05-05 19:45:53 <ArtForz> and I still want more resolution
3576 2011-05-05 19:45:55 <Diablo-D3> not everyone can be rich like you
3577 2011-05-05 19:46:35 <ArtForz> saving on monitors is stupid
3578 2011-05-05 19:46:43 <Diablo-D3> ArtForz: okay so
3579 2011-05-05 19:46:50 <ArtForz> they outlive computers at least by a ratio of 3:1
3580 2011-05-05 19:46:52 <Diablo-D3> how many mhash/sec should I get on a 5850 max at stock
3581 2011-05-05 19:46:55 <topi`> tcatm: it seems SSE2 does not have any rotate instructions for vectors, how is rotr() done then?
3582 2011-05-05 19:47:05 <ArtForz> topi`: shl, shr, or
3583 2011-05-05 19:47:12 <topi`> right.
3584 2011-05-05 19:47:36 <topi`> I guess this would be necessary on NEON, too...
3585 2011-05-05 19:48:10 <ArtForz> some cpus have stuff like shift-and-add, which would cut it to 2 opcodes
3586 2011-05-05 19:48:11 <Diablo-D3> because it seems to be like 290 @ 765
3587 2011-05-05 19:48:14 <topi`> I still wonder why the regular C code is faster on Core2duo than the sha256_4way.c
3588 2011-05-05 19:48:38 <topi`> you'd think that the SSE2 code can do 4 32-bit ops instaed of only 1
3589 2011-05-05 19:48:40 <bk128> Diablo-D3: remember that debian box I set up with 2 5870s and your miner? ssh locked up a month ago and I cant even get into it, but it's been running great nonstop for 2 months
3590 2011-05-05 19:48:40 echelon has quit (Ping timeout: 246 seconds)
3591 2011-05-05 19:48:42 <ArtForz> maybe C2 doesn't have enough shadow regs and hits cache
3592 2011-05-05 19:48:44 <bk128> just not going to touch it :)
3593 2011-05-05 19:48:47 <Diablo-D3> bk128: LOL
3594 2011-05-05 19:48:53 gjs278 has quit (Remote host closed the connection)
3595 2011-05-05 19:49:00 <Diablo-D3> bk128: its still shitting out blocks?
3596 2011-05-05 19:49:34 gjs278 has joined
3597 2011-05-05 19:49:41 <topi`> ArtForz: so effectively one could use the shift-and-add opcode in NEON to fuse shl/shr and or?
3598 2011-05-05 19:49:42 <Diablo-D3> okay so, 290@765, $155 == $1.90 per mhash
3599 2011-05-05 19:49:47 <bk128> I keep getting payed by slush, so I just figured i'm going to leave it until my parents tell me it's on fire or until I stop getting paid. It's not even at my place, it's an hour away
3600 2011-05-05 19:49:50 <ArtForz> yeah
3601 2011-05-05 19:50:00 <bk128> paid*
3602 2011-05-05 19:50:19 <topi`> that doesn't sound too bad, if and if gcc can order the registers in such a way that they don't stall the whole shit
3603 2011-05-05 19:50:41 <gjs278> you just have to get 50 coins
3604 2011-05-05 19:50:47 <gjs278> and it's paid for
3605 2011-05-05 19:50:51 <ArtForz> btw, same trick works on nvidia
3606 2011-05-05 19:50:51 <Diablo-D3> er wait Im doing that wrong
3607 2011-05-05 19:51:13 Speeder has quit (Quit: Speeder)
3608 2011-05-05 19:51:16 <Diablo-D3> 5850 290@765, $155 = 53 cents per mhash
3609 2011-05-05 19:51:18 <ArtForz> doesnt have a native rotate, but it does have shift-and-add
3610 2011-05-05 19:51:42 <gjs278> you get 1.90 for the mhash to watt ratio
3611 2011-05-05 19:51:46 <Diablo-D3> 6870 290@915, $180, 62 cents per mhash
3612 2011-05-05 19:51:47 <gjs278> on a 5850 I think
3613 2011-05-05 19:52:06 <gjs278> because it uses 151 max
3614 2011-05-05 19:52:26 <gjs278> and you can pull in a 290 so 1.92 for mhash/watt
3615 2011-05-05 19:53:01 <gjs278> which isn't too bad if you gen 50 coins at $3.40 each you'll get $170 so $15 for electricity to run it 23 days at 150k difficulty on any pool
3616 2011-05-05 19:53:03 <topi`> ArtForz: so basically where Radeon can do the whole hash with x instructions, Nvidia would need x*2 ?
3617 2011-05-05 19:53:10 <ArtForz> kinda
3618 2011-05-05 19:53:17 <ArtForz> sha256 isn't *only * shifts ;)
3619 2011-05-05 19:53:19 <topi`> of course, not every op is a rotate op
3620 2011-05-05 19:53:22 <topi`> yeah
3621 2011-05-05 19:53:46 <ArtForz> difference between one-opcode rotr and 3-opcode rotr is about +50% insns
3622 2011-05-05 19:54:06 <Diablo-D3> 6950 about 325@800, $223 = 68 cents an mhash
3623 2011-05-05 19:54:16 <bk128> how effective are the coolers on this? http://www.newegg.com/Product/Product.aspx?Item=N82E16814125357 I'm worried because my xfx 5870s with the fan in the center (non-blower) never goes below 78C at 80% fan speed and stock clocks
3624 2011-05-05 19:54:16 <Diablo-D3> but if it accepts the firmware
3625 2011-05-05 19:54:35 <bk128> even though I have a 2 amp delta fan pushing 150+ cfm over it
3626 2011-05-05 19:54:44 <Diablo-D3> 6950->70 about 350@800, $223 = 63 cents
3627 2011-05-05 19:54:46 <Diablo-D3> so fuck it
3628 2011-05-05 19:54:49 <Diablo-D3> I guess I buy that 5850
3629 2011-05-05 19:54:51 <Diablo-D3> it gives me the best deal
3630 2011-05-05 19:55:16 <diki> can someone explain in layman's terms what long polling means
3631 2011-05-05 19:55:32 <gjs278> it means your miner will always have work that will count
3632 2011-05-05 19:55:35 <Diablo-D3> diki: miner runs an async getwork request, it doesnt return until theres new work
3633 2011-05-05 19:55:40 BitMark_ has joined
3634 2011-05-05 19:55:40 BitMark_ has quit (Changing host)
3635 2011-05-05 19:55:40 BitMark_ has joined
3636 2011-05-05 19:55:55 <diki> is it implemented in push pool by jgarzik?
3637 2011-05-05 19:55:59 Skittles has joined
3638 2011-05-05 19:56:06 BitMark has quit (Read error: Connection reset by peer)
3639 2011-05-05 19:56:06 BitMark_ is now known as BitMark
3640 2011-05-05 19:56:16 <bk128> Diablo-D3: wait, why would you go with a 58** over the 68**?
3641 2011-05-05 19:56:23 <Diablo-D3> bk128: see above.
3642 2011-05-05 19:56:24 <gjs278> stream
3643 2011-05-05 19:56:33 <Skittles> hey guys, how many blocks does it take until I will start seeing a balance gain on my account?
3644 2011-05-05 19:56:36 <gjs278> enjoy your copy of shogun
3645 2011-05-05 19:56:45 <Diablo-D3> bk128: dollar efficiency, that 5850 wins
3646 2011-05-05 19:56:48 <ArtForz> 58xx generally is the better GPGPU card
3647 2011-05-05 19:56:58 <topi`> any estimates on the effectiveness of long polling
3648 2011-05-05 19:57:07 <diki> When Crysis 2 came out, i tried finding out the packer...i was unable to
3649 2011-05-05 19:57:10 <ArtForz> but then 5xxx probably won't have much resale value once 7xxx is out
3650 2011-05-05 19:57:15 <diki> few days later we see a crack..fully unpacked
3651 2011-05-05 19:57:22 <Diablo-D3> and the 5850 and 6870 use the same watts (151) and the 6950 uses 200
3652 2011-05-05 19:57:22 <bk128> when do you think 7**** is coming out?
3653 2011-05-05 19:57:28 <Diablo-D3> so the 5850 seems to be the best deal
3654 2011-05-05 19:57:35 <Diablo-D3> and that one comes OC ready and pre-OC'ed
3655 2011-05-05 19:57:44 echelon has joined
3656 2011-05-05 19:57:44 <bk128> so how good do you think the fans are on the http://www.newegg.com/Product/Product.aspx?Item=N82E16814125318
3657 2011-05-05 19:57:50 <bk128> so does it perform like a 5870?
3658 2011-05-05 19:57:54 <ArtForz> first cards *maybe* Q4 (though it could be a 28nm 6750/70)
3659 2011-05-05 19:58:02 <Diablo-D3> bk128: no.
3660 2011-05-05 19:58:07 <diki> btw
3661 2011-05-05 19:58:12 <Diablo-D3> bk128: I should be able to do about 300 mhash/sec on that 5850
3662 2011-05-05 19:58:27 <diki> i dont know what put on the screws on my 5850...i was unable to take them off
3663 2011-05-05 19:58:30 <bk128> ok. have you used a card with that cooler design before?
3664 2011-05-05 19:58:37 <Diablo-D3> bk128: no
3665 2011-05-05 19:58:50 <ArtForz> decent avail of 7xxx ... Q2 '12 or so
3666 2011-05-05 19:58:54 <Diablo-D3> but it looks about as efficient as the third party one I slapped on my 4850
3667 2011-05-05 19:59:01 <bk128> I'm worried about cards without blower fans now after buying those shitty xfx 5870s
3668 2011-05-05 19:59:07 <Diablo-D3> 7xxx probably wont be out in full force until q3 12
3669 2011-05-05 19:59:14 <BlueMatt> jaromil: ok, got bitcoin xcompile on autotools via mingw32, just had to run a 10-line set of sed's after ./configure
3670 2011-05-05 19:59:20 <Diablo-D3> bk128: worst case is I rip the fans out and zip tie two 120s to it
3671 2011-05-05 19:59:27 <diki> Anyone propose how to take them off, the screws so i can clean my heatsink?
3672 2011-05-05 19:59:38 <bk128> Diablo-D3: you'd rather void the warranty to keep it cooler?
3673 2011-05-05 19:59:39 <Diablo-D3> bk128: or really just one
3674 2011-05-05 19:59:50 <Diablo-D3> bk128: the fans will come off with like four screws
3675 2011-05-05 19:59:54 <Diablo-D3> Ill keep the heatsinks
3676 2011-05-05 19:59:57 <bk128> not on my xfx cards
3677 2011-05-05 19:59:58 <gjs278> taking off the cooler isnt hard
3678 2011-05-05 20:00:04 <gjs278> xfx even lets you take off the cooler
3679 2011-05-05 20:00:08 <Skittles> any GTX 480 owners out there?
3680 2011-05-05 20:00:08 <gjs278> and replace it 3rd party
3681 2011-05-05 20:00:10 <jaromil> BlueMatt: LOL at 10-line sed
3682 2011-05-05 20:00:10 <diki> i'd reconsider
3683 2011-05-05 20:00:12 <bk128> really? w/o voiding warranty?
3684 2011-05-05 20:00:14 <Diablo-D3> I completely replaced the heatsink on my 4850
3685 2011-05-05 20:00:17 <gjs278> that's what they claim
3686 2011-05-05 20:00:19 <Diablo-D3> which did void the warrenty
3687 2011-05-05 20:00:22 <diki> my 5850's cooler...couldnt
3688 2011-05-05 20:00:28 <bk128> i'll have to check with them
3689 2011-05-05 20:00:31 <gjs278> check the xfx warranty they claim they support 3rd parties
3690 2011-05-05 20:00:33 <Diablo-D3> but I dont care, I intentionally bought a 4850 with a really shitty heatsink so I could replace it
3691 2011-05-05 20:00:34 <diki> i gave all my strenght
3692 2011-05-05 20:00:34 <BlueMatt> jaromil: yea...its not exactly the way thats supposed to work...
3693 2011-05-05 20:00:43 <Diablo-D3> 4xxx gen heatsinks, third party or not, were shit
3694 2011-05-05 20:00:45 <diki> *strength
3695 2011-05-05 20:00:45 <BlueMatt> jaromil: but its a start, at least I know what exactly needs to change
3696 2011-05-05 20:00:47 <jaromil> BlueMatt: i'm all for including that and sending it to autotools mailinglist LOL
3697 2011-05-05 20:00:55 n00B_ has joined
3698 2011-05-05 20:01:08 <BlueMatt> jaromil: yea...that will sit well
3699 2011-05-05 20:01:12 <Diablo-D3> lets see what I can afford
3700 2011-05-05 20:01:13 * jaromil known complainer about autotools
3701 2011-05-05 20:01:25 Incitatus has quit (Read error: Connection reset by peer)
3702 2011-05-05 20:01:32 <jaromil> BlueMatt: i'm just sitting in revspace.nl now waiting for the radio show
3703 2011-05-05 20:01:35 <Skittles> How many blocks does it take to generate a bitcoin?
3704 2011-05-05 20:01:35 <Diablo-D3> sigh Ill have to make a large payment on my credit card =/
3705 2011-05-05 20:01:41 <jaromil> so won't be reactive on that
3706 2011-05-05 20:01:44 <jaromil> but it sounds good
3707 2011-05-05 20:01:49 <diki> And i have scene access :D
3708 2011-05-05 20:01:49 <phantomcircuit> jaromil, i fucking despise autotools
3709 2011-05-05 20:01:54 <BlueMatt> jaromil: can you ping me when that radio thing starts?
3710 2011-05-05 20:01:55 <gjs278> autotools pwns
3711 2011-05-05 20:01:56 <jaromil> good enough to sit in a README that we should write
3712 2011-05-05 20:02:02 Incitatus has joined
3713 2011-05-05 20:02:06 <jaromil> phantomcircuit: it's a PINA COLADA
3714 2011-05-05 20:02:16 <topi`> jaromil: so does the show start now? it's 10pm CET
3715 2011-05-05 20:02:19 <jaromil> means EPIC PITA
3716 2011-05-05 20:02:44 <jaromil> topi`: yea i'm here with the guys it sayz on air its gonna start
3717 2011-05-05 20:02:52 <topi`> Skittles: every block creates plenty of bitcoins for some lucky guy
3718 2011-05-05 20:03:03 <topi`> jaromil: cool :)
3719 2011-05-05 20:03:15 <jaromil> topic: bitcoin!
3720 2011-05-05 20:03:23 <jaromil> so i'll be picking up your reactions from here
3721 2011-05-05 20:03:35 <luke-jr> Skittles: what video card?
3722 2011-05-05 20:03:41 <topi`> tcatm: you said you can compile *.cl sources with gcc, but I ran into lots of trouble, odd errors from gcc
3723 2011-05-05 20:03:41 <gjs278> I thin he has nvidia
3724 2011-05-05 20:03:45 <jaromil> actiually
3725 2011-05-05 20:03:47 <Skittles> I own GTX 480 Superclocked
3726 2011-05-05 20:03:47 <jaromil> not from here
3727 2011-05-05 20:03:54 <Skittles> eVGA's
3728 2011-05-05 20:03:56 <luke-jr> Skittles: worthless garbage
3729 2011-05-05 20:04:02 <Skittles> lol AMD fan?
3730 2011-05-05 20:04:07 <Skittles> ATI*
3731 2011-05-05 20:04:10 <luke-jr> Skittles: AMD blows nvidia away
3732 2011-05-05 20:04:11 <gjs278> well
3733 2011-05-05 20:04:15 <gjs278> for bitcoin purposes
3734 2011-05-05 20:04:20 <gjs278> you won't be able to generate much at all
3735 2011-05-05 20:04:27 <luke-jr> gjs278: for everything. it's just more obvious with bitcoin
3736 2011-05-05 20:04:29 <jaromil> ok last bothering about this in this chan: http://signal.hackerspaces.org hactivist hour radio in 5 mins on! like to interact? joinr #bitcoin-politics
3737 2011-05-05 20:04:43 <luke-jr> Skittles: GTX480 can maybe do 100 MH/s
3738 2011-05-05 20:04:56 <bk128> ArtForz: so ati is going to have 28nm chips out before intel?
3739 2011-05-05 20:04:56 <gjs278> at 250 watts too
3740 2011-05-05 20:04:58 SaneSmith has joined
3741 2011-05-05 20:05:05 <luke-jr> Skittles: so it's equivalent to a Radeon 4890
3742 2011-05-05 20:05:09 <BlueMatt> ha, not only does it compile, but it actually runs
3743 2011-05-05 20:05:27 <luke-jr> except Radeon 4890 only uses 190 W
3744 2011-05-05 20:05:29 <ArtForz> bk128: huh?
3745 2011-05-05 20:05:29 <diki> Die shrinks appear, wattage and temps increase
3746 2011-05-05 20:05:39 <diki> should be the other way around
3747 2011-05-05 20:05:45 <Skittles> luke-jr: I got it almost a year ago
3748 2011-05-05 20:06:06 <gjs278> if you want to mess around, go ahead and gen with it, but you won't get much
3749 2011-05-05 20:06:24 <bk128> I thought you said the 7xxx was coming out q4 of this year and is 28nm
3750 2011-05-05 20:06:44 <luke-jr> Skittles: 4890 was a Q1 2009 card
3751 2011-05-05 20:06:45 <ArtForz> intel is doing 28nm ?
3752 2011-05-05 20:06:50 <Diablo-D3> intels not doing 28nm yet
3753 2011-05-05 20:06:52 <Diablo-D3> not until next year
3754 2011-05-05 20:06:53 <bk128> ivy bridge/ haskell right?
3755 2011-05-05 20:06:58 <Diablo-D3> their sandy cunt one
3756 2011-05-05 20:07:01 <gjs278> need less nanometers
3757 2011-05-05 20:07:01 <Diablo-D3> er
3758 2011-05-05 20:07:01 <diki> Moore's law
3759 2011-05-05 20:07:05 <Diablo-D3> their post-sandy cunt one
3760 2011-05-05 20:07:08 <bk128> Diablo-D3: sb is awesome
3761 2011-05-05 20:07:09 <ArtForz> I thought they'll go from 32 to 22
3762 2011-05-05 20:07:09 <Diablo-D3> aka sandier cunt
3763 2011-05-05 20:07:10 <diki> I cant wait for graphene transistors
3764 2011-05-05 20:07:13 <Diablo-D3> bk128: sb?
3765 2011-05-05 20:07:18 <bk128> sandybridge
3766 2011-05-05 20:07:27 <Skittles> luke-jr: it plays crysis 2 maxed and it blows my old geforce 6600GT out of the water lol
3767 2011-05-05 20:07:31 <bk128> oh, maybe it is 22 not 28
3768 2011-05-05 20:07:38 <luke-jr> Skittles: worthless comparison
3769 2011-05-05 20:07:47 <diki> I think ivy is 22nm
3770 2011-05-05 20:07:53 <gjs278> I want 1nm
3771 2011-05-05 20:07:59 <luke-jr> Skittles: for bitcoin mining, the only thing that matters is how it compares to what everyone else is using
3772 2011-05-05 20:08:01 <diki> 1nm=graphene transistor
3773 2011-05-05 20:08:03 <bk128> you're right, intel is skipping 28
3774 2011-05-05 20:08:11 <diki> and no leakage
3775 2011-05-05 20:08:12 <luke-jr> ;;bc,gen 100000
3776 2011-05-05 20:08:12 <jaromil> BlueMatt: is there a way you can push into my branch on github?
3777 2011-05-05 20:08:13 <gribble> The expected generation output, at 100000 Khps, given current difficulty of 109670.13329248 , is 0.91713974479 BTC per day and 0.0382141560329 BTC per hour.
3778 2011-05-05 20:08:16 <Diablo-D3> bk128: feh, I hate intel
3779 2011-05-05 20:08:18 <jaromil> like i can give you access?
3780 2011-05-05 20:08:23 <gjs278> sandy bridge is good ifyou want to overclock to 5ghz and kill the chip in 3 years
3781 2011-05-05 20:08:23 <Diablo-D3> ;;bc,gen 300000
3782 2011-05-05 20:08:23 <gribble> The expected generation output, at 300000 Khps, given current difficulty of 109670.13329248 , is 2.75141923437 BTC per day and 0.114642468099 BTC per hour.
3783 2011-05-05 20:08:24 <bk128> and not haskell/ haswell, but that's even after ivy
3784 2011-05-05 20:08:26 <gjs278> but at $300 it's worth it
3785 2011-05-05 20:08:28 <jaromil> or i can merge you
3786 2011-05-05 20:08:45 <gjs278> seirously though if I had a sandy bridge 2600k I would be at 5hz right now
3787 2011-05-05 20:08:47 <ArtForz> yup, intel is skipping 28nm
3788 2011-05-05 20:08:48 <Diablo-D3> ;;bc,calc 300000
3789 2011-05-05 20:08:48 <bk128> Diablo-D3: I found that the i5 2500 was better performance/ dollar than any amd chip
3790 2011-05-05 20:08:49 <gribble> The average time to generate a block at 300000 Khps, given current difficulty of 109670.13329248 , is 2 weeks, 4 days, 4 hours, 8 minutes, and 18 seconds
3791 2011-05-05 20:08:54 <luke-jr> gjs278: why?
3792 2011-05-05 20:08:54 <BlueMatt> jaromil: dont think so, the small change I actually made is just a couple lines, then you have to run 20 lines of sed, so Im gonna do some more work then send it over
3793 2011-05-05 20:08:57 <ArtForz> while TSMC skipped 32nm
3794 2011-05-05 20:08:58 <gjs278> because
3795 2011-05-05 20:09:03 <gjs278> it's too cheap not to fry it
3796 2011-05-05 20:09:05 <topi`> BlueMatt: what compiles and runs?
3797 2011-05-05 20:09:18 <bk128> ArtForz: whats tsmc? foundry for amd?
3798 2011-05-05 20:09:19 <Skittles> luke-jr: I'm new to bitcoin mining... didn't think it used GPU power
3799 2011-05-05 20:09:22 <Diablo-D3> so it should take me.... 16 days to pay this card off
3800 2011-05-05 20:09:25 <BlueMatt> topi`: bitcoin compiled with autotools xcompiled for win32 via mingw
3801 2011-05-05 20:09:25 <luke-jr> gjs278: I don't think SB can clock that slow
3802 2011-05-05 20:09:31 <ArtForz> nope
3803 2011-05-05 20:09:34 <luke-jr> Skittles: CPU mining is even more worthless
3804 2011-05-05 20:09:36 <gjs278> sandy bridge can definitely go 5
3805 2011-05-05 20:09:43 <luke-jr> gjs278: 5 Hz is slow
3806 2011-05-05 20:09:44 <topi`> BlueMatt: right :)
3807 2011-05-05 20:09:45 <ArtForz> AMDs foundry is Global Foundries
3808 2011-05-05 20:09:49 <gjs278> gggg
3809 2011-05-05 20:09:50 <diki> yeah sb does 5 on air
3810 2011-05-05 20:09:57 <Skittles> luke-jr: haha alright wheres the GPU client?
3811 2011-05-05 20:10:02 <gjs278> poclbm-gui
3812 2011-05-05 20:10:06 <luke-jr> gjs278: I think only the best SB can reach 5 GHz if it's single-threaded
3813 2011-05-05 20:10:11 <luke-jr> Skittles: there are many
3814 2011-05-05 20:10:20 <gjs278> I think people get their i5's that high
3815 2011-05-05 20:10:24 <gjs278> if you have a certain batch
3816 2011-05-05 20:10:29 <gjs278> but you can't hyperthread on the i7 at 5
3817 2011-05-05 20:10:58 <Skittles> luke-jr: what do u use? and can I run CPU/GPU same time like I did for Fold@Home ?
3818 2011-05-05 20:11:04 <luke-jr> gjs278: no multithreading *at all*
3819 2011-05-05 20:11:14 <gjs278> I'm positive they have multithreading
3820 2011-05-05 20:11:14 <Diablo-D3> I hope that 5850 will still be in stock tommorow
3821 2011-05-05 20:11:16 <luke-jr> Skittles: CPU mining costs more in electricity than it earns
3822 2011-05-05 20:11:26 <luke-jr> gjs278: they do, but they can't reach 5 GHz using it
3823 2011-05-05 20:11:26 <diki> you buying a 5850?
3824 2011-05-05 20:11:30 <diki> why nit 5870?
3825 2011-05-05 20:11:36 <diki> does 350 at stock
3826 2011-05-05 20:11:37 <gjs278> 5870's new are all out of stock
3827 2011-05-05 20:11:41 <luke-jr> diki: 5850 is like half the price
3828 2011-05-05 20:11:46 <Skittles> luke-jr: I get free electricity :-)
3829 2011-05-05 20:11:46 <diki> no 5870 at all?
3830 2011-05-05 20:11:51 <diki> anywhere?
3831 2011-05-05 20:12:00 <luke-jr> diki: 5870 is over $200
3832 2011-05-05 20:12:04 <luke-jr> diki: 5850 is $140
3833 2011-05-05 20:12:11 <diki> so?
3834 2011-05-05 20:12:13 <gjs278> it's more than $200 even, anywhere new is out of the cheaper ones
3835 2011-05-05 20:12:14 <luke-jr> so much cheaper
3836 2011-05-05 20:12:25 <gjs278> diki it's like $280 to buy any 5870 new
3837 2011-05-05 20:12:33 <bk128> ArtForz: I bought 3 of these for my box with 2 5870s, turns out I only need one http://search.digikey.com/scripts/DkSearch/dksus.dll?keywords=603-1325-ND&x=0&y=0
3838 2011-05-05 20:12:43 forever-d has joined
3839 2011-05-05 20:12:51 <ArtForz> well, it's 30W
3840 2011-05-05 20:12:54 <bk128> ArtForz: And I have to limit the power with a bunch of 10w resistors
3841 2011-05-05 20:12:58 <bk128> :D
3842 2011-05-05 20:13:03 <ArtForz> I use 3*5W for my 4*5970 boxes ...
3843 2011-05-05 20:13:13 <ArtForz> my 2*5870s and 2*5970s only have 1*5W
3844 2011-05-05 20:13:33 <ArtForz> eww, no PWM in, that sucks
3845 2011-05-05 20:13:39 <bk128> I know :/
3846 2011-05-05 20:13:51 fetokun has quit (Quit: It's mr sandman's fault)
3847 2011-05-05 20:13:55 <ArtForz> high powered fans with no PWM input = fail
3848 2011-05-05 20:14:12 <bk128> Could I do shitty pwm with a 555 and an irf510 that I have lying around?
3849 2011-05-05 20:14:16 <ArtForz> yea
3850 2011-05-05 20:14:26 <ArtForz> thats kinda how I PWM my non-PWM fans
3851 2011-05-05 20:14:31 <bk128> do I need a diode across the fan?
3852 2011-05-05 20:14:36 <ArtForz> yep
3853 2011-05-05 20:14:40 <ArtForz> attiny + bunch of FETs
3854 2011-05-05 20:14:45 <bk128> and can I drive the fet directly with the 555 pin 3?
3855 2011-05-05 20:14:54 MartianW has joined
3856 2011-05-05 20:14:55 <bk128> dont have any attinys
3857 2011-05-05 20:14:58 <ArtForz> at 12V, easily
3858 2011-05-05 20:15:04 <ArtForz> with 5V MCUs, it's kidan problematic
3859 2011-05-05 20:15:15 <luke-jr> Skittles: note that nobody really uses nvidia anymore, so poclbm isn't optimized for it
3860 2011-05-05 20:15:31 <gjs278> luke-jr http://hardforum.com/showthread.php?t=1601937 this is the official 5+ club I doubt these guys would run their systems on one thread
3861 2011-05-05 20:15:46 <ArtForz> doing +12V fan PWM you use rather low freqs, so you don't need shitloads of drive for the FET
3862 2011-05-05 20:15:57 <luke-jr> gjs278: since only idiots care about raw GHz, I bet they do
3863 2011-05-05 20:15:59 devrandom has quit (Ping timeout: 246 seconds)
3864 2011-05-05 20:16:02 <gjs278> yeah
3865 2011-05-05 20:16:13 <ArtForz> if you feel like it, do a npn+pnp follower on pin3
3866 2011-05-05 20:16:15 <gjs278> they fold
3867 2011-05-05 20:16:17 <gjs278> and fold
3868 2011-05-05 20:16:19 <gjs278> all day
3869 2011-05-05 20:16:26 <gjs278> and they burn chips for it
3870 2011-05-05 20:17:56 <gjs278> I only run at 4ghz on the i7 9xx because anything else requires too much voltage for too little gain
3871 2011-05-05 20:18:02 BitterTea has joined
3872 2011-05-05 20:18:11 <gjs278> even 4.1ghz is a huge jump in voltage thats just not worth it
3873 2011-05-05 20:18:11 <BitterTea> Hey guys, has anyone noticed this? http://bitcoincharts.com/bitcoin/
3874 2011-05-05 20:18:20 <BitterTea> There seem to be a LOT of unconfirmed tranasctions today
3875 2011-05-05 20:18:24 <MartianW> How can I see transactions that haven't made it into a block yet?
3876 2011-05-05 20:18:32 <BitterTea> MartianW: http://bitcoincharts.com/bitcoin/
3877 2011-05-05 20:18:44 <n00B_> So is it totally stupid (beyond mere educational value) for me to run Bitcoin on a MacBook Pro Core i7 for mining?
3878 2011-05-05 20:18:44 io_error has joined
3879 2011-05-05 20:18:51 <BlueMatt> BitterTea: many, many of those appear to have unconfirmed inputs, probably spam
3880 2011-05-05 20:18:55 <BitterTea> Or add the -debug switch, then you can double click on a txn in your client for more info
3881 2011-05-05 20:19:07 <MartianW> Thanks.
3882 2011-05-05 20:19:15 <luke-jr> n00B_: pretty stupid, yeah
3883 2011-05-05 20:19:34 <BitterTea> BlueMatt: Ah, like someone sent themselves 0.01 BTC, then used that 0.01 BTC from the output as an input immediately?
3884 2011-05-05 20:19:39 <gjs278> lol n00B_ don't do that, your laptop wont appreciate the heat
3885 2011-05-05 20:19:51 <berto-> has anyone tried the ar6990 with ubuntu 11.04?
3886 2011-05-05 20:20:01 <BlueMatt> BitterTea: well, now that I look closely, its not actually that unconfirmed, but that is a pretty common issue
3887 2011-05-05 20:20:18 <berto-> is this the right channel to be asking this question?
3888 2011-05-05 20:20:33 <BlueMatt> berto-: try #bitcoin
3889 2011-05-05 20:20:40 <berto-> BlueMatt: cool, thanks.
3890 2011-05-05 20:20:46 <BlueMatt> or here if its development related, doesnt matter really
3891 2011-05-05 20:20:52 <gjs278> pretty sure #bitcoin says not to talk there
3892 2011-05-05 20:21:01 <BlueMatt> gjs278: no thats discussion
3893 2011-05-05 20:21:08 <BlueMatt> it used to be #bitcoin but we have that now
3894 2011-05-05 20:21:18 <gjs278> I used to get that
3895 2011-05-05 20:21:20 <gjs278> Tue Mar 22 2011] [04:04:27] <necrodearia> gjs278 :Nobody chats here!!!!!! Please join one of the following channels:
3896 2011-05-05 20:21:21 <gjs278> that crap
3897 2011-05-05 20:22:06 <bk128> ArtForz: would this work? is that a good r value for pin 3? and what do you mean a npn/pnp follower? http://i.imgur.com/KGNHB.png
3898 2011-05-05 20:22:11 <necrodearia> O_o
3899 2011-05-05 20:22:16 <bk128> also, what's a good freq? .5khz?
3900 2011-05-05 20:22:18 <gjs278> that's right
3901 2011-05-05 20:22:22 <gjs278> you highlight me weeks ago
3902 2011-05-05 20:22:23 <necrodearia> gjs278, That was when the channel wasn't used
3903 2011-05-05 20:22:24 <gjs278> right back at you
3904 2011-05-05 20:22:39 <ArtForz> your D is backwards
3905 2011-05-05 20:22:42 <gjs278> I eventually took bitcoin off of my autojoin
3906 2011-05-05 20:22:43 MartianW has quit (Quit: Bye all.)
3907 2011-05-05 20:22:52 <gjs278> just so I wouldnt get that highlight everytime
3908 2011-05-05 20:22:54 <necrodearia> gjs278, You can join now
3909 2011-05-05 20:22:57 <gjs278> oh I did
3910 2011-05-05 20:23:25 <necrodearia> http://signal.hackerspaces.org/ will soon talk about Bitcoin.
3911 2011-05-05 20:23:32 <ArtForz> 220R is a bit high, I'd use 47 or so
3912 2011-05-05 20:23:46 <bk128> ArtForz: ah crap it is
3913 2011-05-05 20:23:54 <topi`> what kind of flags do I need to give gcc so that I can try -ftree-vectorize on a Core2 ?
3914 2011-05-05 20:23:56 <nanotube> necrodearia: coool :)
3915 2011-05-05 20:24:15 <bk128> ArtForz: what's a good freq?
3916 2011-05-05 20:24:16 <diki> anyone familiar with symbian c++?
3917 2011-05-05 20:24:49 <ArtForz> usually a few Hz - 200 Hz or so
3918 2011-05-05 20:25:19 <necrodearia> nanotube, jaromil mentioned it and is talking now
3919 2011-05-05 20:25:23 <ArtForz> much higher than that and it gets annoying unless you go for > 20kHz (and then you need L+C on the output = you're building a step-down)
3920 2011-05-05 20:25:33 <gjs278> if I run at 825/300 what voltage do you think would be my min for a 5970
3921 2011-05-05 20:25:58 <gjs278> I don't feel like crashing and testing at random
3922 2011-05-05 20:26:18 <ArtForz> "depends"
3923 2011-05-05 20:26:29 <gjs278> yeah 1.05 or whatever I'm at now is the winner
3924 2011-05-05 20:26:40 <ArtForz> somewhere from 1.00 to 1.05 probably
3925 2011-05-05 20:26:56 <bk128> so the 555 can handle the current to drive the gate at 200hz?
3926 2011-05-05 20:27:03 <ArtForz> yep
3927 2011-05-05 20:27:13 <ArtForz> bipolar 555s usually can drive > 100mA peak
3928 2011-05-05 20:27:17 <ArtForz> at 12V VCC
3929 2011-05-05 20:27:21 <nanotube> necrodearia: don't have time to listen atm - but looking forward to your summary :)
3930 2011-05-05 20:27:32 <ArtForz> CMOS 555s... not so much
3931 2011-05-05 20:28:05 BitMark is now known as BitMark_afk
3932 2011-05-05 20:28:13 <bk128> how would I set up the npn/pnp driver you were talking about? is that like a darlington?
3933 2011-05-05 20:28:17 <ArtForz> no
3934 2011-05-05 20:28:19 lyspooner has joined
3935 2011-05-05 20:28:24 <ArtForz> simple emitter follower pair
3936 2011-05-05 20:28:24 Skittles has quit (Quit: Page closed)
3937 2011-05-05 20:28:25 <Diablo-D3> okay so
3938 2011-05-05 20:28:29 * Diablo-D3 adds the 5850 to his cart
3939 2011-05-05 20:28:32 <ArtForz> sec
3940 2011-05-05 20:28:35 redwyrm has joined
3941 2011-05-05 20:28:38 redwyrm has left ()
3942 2011-05-05 20:28:40 eternal1 has quit (Quit: Leaving.)
3943 2011-05-05 20:28:43 redwyrm has joined
3944 2011-05-05 20:29:24 <necrodearia> nanotube, Well, what particularly interested me was jaromil mentioned it in #witcoin and suggested that witcoin would be referenced. But even mentioning in #witcoin alone was enough for me to be attentive.
3945 2011-05-05 20:29:44 <topi`> so whats so great about witcoin?
3946 2011-05-05 20:29:52 <topi`> never really bothered to register :)
3947 2011-05-05 20:30:20 <gjs278> I need a helpful half cartoon video to explain it to me
3948 2011-05-05 20:30:38 <BitterTea> topi`: I hadn't either, but it's pretty cool
3949 2011-05-05 20:30:56 <BitterTea> Like reddit where posting, replying, and voting costs small (configurable per "subreddit") amounts of BTC
3950 2011-05-05 20:31:05 <jaromil> topi`: i'm going to answer your question what's so great about bitcoin
3951 2011-05-05 20:31:18 <topi`> jaromil: wit, not bit :)
3952 2011-05-05 20:31:22 <BitterTea> Of which, some is sent to the poster, the site, and optionally a charity (I think)
3953 2011-05-05 20:31:29 <jaromil> ops wit
3954 2011-05-05 20:31:32 <jaromil> i meant wit
3955 2011-05-05 20:31:38 <topi`> jaromil: did you go to have a coffee? since all I hear is just some music
3956 2011-05-05 20:31:44 <bk128> ArtForz: is this it? http://i.cmpnet.com/techonline/images/community/content/old/news/features/advanced_power/Fig9.gif
3957 2011-05-05 20:32:33 <diki> what does it meen if paypal has 5% fees?
3958 2011-05-05 20:32:40 <diki> I want to buy a cooler is which is X dollars
3959 2011-05-05 20:32:55 <diki> and on the site says with paypal i'll have 5% fees from them
3960 2011-05-05 20:33:00 <diki> how much will the sum rise to?
3961 2011-05-05 20:33:00 <jaromil> topi`: beer. plz come to #bitcoin-politics to talk :)
3962 2011-05-05 20:33:09 <topi`> right
3963 2011-05-05 20:33:11 <BitterTea> Either you'll have to send 5% more than the purchase price, or the seller eats the 5% out of the purchase price, I think
3964 2011-05-05 20:33:27 <diki> how do i calculate the exact amount of these 5 percent?
3965 2011-05-05 20:33:35 <diki> divide,multiply?
3966 2011-05-05 20:33:37 <BlueMatt> X*.05
3967 2011-05-05 20:33:40 <BlueMatt> = 5%
3968 2011-05-05 20:33:47 devrandom has joined
3969 2011-05-05 20:33:49 <BlueMatt> X*1.05 = new price
3970 2011-05-05 20:33:51 <edcba> crazy skills
3971 2011-05-05 20:33:55 <ArtForz> bk128: yup
3972 2011-05-05 20:34:17 <edcba> what if the price decreased of 5% ?
3973 2011-05-05 20:34:22 <diki> so 54*5=270?
3974 2011-05-05 20:34:24 <edcba> how do i calc that ?
3975 2011-05-05 20:34:25 n00B_ has quit (Quit: Page closed)
3976 2011-05-05 20:35:01 * edcba stops making fun
3977 2011-05-05 20:35:02 <BlueMatt> edcba: X*.95?
3978 2011-05-05 20:35:15 DukeOfURL has joined
3979 2011-05-05 20:35:16 <edcba> too complex :)
3980 2011-05-05 20:35:20 <diki> yeah
3981 2011-05-05 20:35:30 <BlueMatt> lol ok
3982 2011-05-05 20:36:58 xelister has quit (Ping timeout: 246 seconds)
3983 2011-05-05 20:37:02 <diki> so if total price of the cooler is 62 usd and paypal fees are 5% then the amount is?
3984 2011-05-05 20:37:05 <netxshare> random data
3985 2011-05-05 20:37:08 <netxshare> bruteforce
3986 2011-05-05 20:37:10 <netxshare> data
3987 2011-05-05 20:37:11 <netxshare> bits
3988 2011-05-05 20:37:13 <netxshare> can
3989 2011-05-05 20:37:13 <netxshare> be
3990 2011-05-05 20:37:39 <gjs278> diki what cooler is it
3991 2011-05-05 20:37:44 <diki> Hyper 212+
3992 2011-05-05 20:37:45 <gjs278> don't pay 5% extra for paypal
3993 2011-05-05 20:37:47 <gjs278> thats bs
3994 2011-05-05 20:38:12 <gjs278> http://www.newegg.com/Product/Product.aspx?Item=N82E16835103065
3995 2011-05-05 20:38:22 <gjs278> $47 with shipping included
3996 2011-05-05 20:38:23 <BlueMatt> use bitcoin instead :)
3997 2011-05-05 20:38:30 <netxshare> bad cooler
3998 2011-05-05 20:38:34 <diki> way more when we convert it to my currency
3999 2011-05-05 20:38:36 <diki> and it's not bad
4000 2011-05-05 20:38:42 <netxshare> there is better
4001 2011-05-05 20:38:45 <diki> but it's better than the stock amd crap
4002 2011-05-05 20:38:46 <netxshare> for that price
4003 2011-05-05 20:38:47 <gjs278> it's definitely not a "bad" cooler but of course there's batter
4004 2011-05-05 20:38:49 <netxshare> much better
4005 2011-05-05 20:38:54 <netxshare> there for bad
4006 2011-05-05 20:38:55 <netxshare> !
4007 2011-05-05 20:38:56 <anarchyx> ;;bc,stats
4008 2011-05-05 20:38:58 <gribble> Current Blocks: 122089 | Current Difficulty: 109670.13329248 | Next Difficulty At Block: 122975 | Next Difficulty In: 886 blocks | Next Difficulty In About: 4 days, 10 hours, 48 minutes, and 44 seconds | Next Difficulty Estimate: 150176.92831220
4009 2011-05-05 20:39:02 <anarchyx> wow@150k
4010 2011-05-05 20:39:16 <diki> Hyper 212+ is way better than listening to 6400 rpm
4011 2011-05-05 20:39:23 <netxshare> lol
4012 2011-05-05 20:39:28 <netxshare> I enjoy my A50 just fine
4013 2011-05-05 20:39:32 <Diablo-D3> you know
4014 2011-05-05 20:39:33 <netxshare> it was 20 bucks
4015 2011-05-05 20:39:39 <netxshare> that == win
4016 2011-05-05 20:39:44 <Diablo-D3> why dont they make fans whos drone is supersonic?
4017 2011-05-05 20:39:53 <gjs278> nh-d14 was like $80
4018 2011-05-05 20:40:02 <Diablo-D3> they could be as loud as fuck, but you couldnt hear them
4019 2011-05-05 20:40:22 <gjs278> they should make two fans that produce the exact opposite frequencies so the noise cancels each other out
4020 2011-05-05 20:40:24 <ArtForz> btw, for a slightly crazy fan controller ... http://i.imgur.com/75RyO.png *g*
4021 2011-05-05 20:40:26 <diki> the stock amd fan is loud
4022 2011-05-05 20:40:31 <diki> i already got hearing problems
4023 2011-05-05 20:40:34 <redwyrm> hahahah
4024 2011-05-05 20:40:41 <netxshare> http://www.google.com/products/catalog?hl=en&q=nh-d14&bav=on.2,or.r_gc.r_pw.&um=1&ie=UTF-8&cid=11004785383205699382&sa=X&ei=EwrDTcWGBMectwe7rMSrBQ&ved=0CEcQ8wIwAQ#
4025 2011-05-05 20:40:42 <netxshare> haha
4026 2011-05-05 20:40:44 <diki> many months listening to 6400 rpm
4027 2011-05-05 20:40:50 <redwyrm> what's the best place to recommend features for the main Bitcoin client?
4028 2011-05-05 20:40:50 <BlueMatt> gjs278: brilliant, sadly thats not exactly how it works
4029 2011-05-05 20:40:55 <gjs278> I know
4030 2011-05-05 20:40:59 <Diablo-D3> okay
4031 2011-05-05 20:40:59 <Diablo-D3> Ive
4032 2011-05-05 20:41:00 <redwyrm> is there a thread for this on the forums?
4033 2011-05-05 20:41:02 <Diablo-D3> PURCHASED THE CARD
4034 2011-05-05 20:41:06 <gjs278> thats what noise cancelling headphones try to do at least
4035 2011-05-05 20:41:07 <BlueMatt> ArtForz: noone but you can read that
4036 2011-05-05 20:41:22 <gjs278> Diablo-D3 welcome to the club
4037 2011-05-05 20:41:23 <gjs278> I have the card
4038 2011-05-05 20:41:28 <gjs278> except I have nowhere to put it
4039 2011-05-05 20:41:32 <Diablo-D3> http://www.newegg.com/Product/Product.aspx?Item=N82E16814125318
4040 2011-05-05 20:41:35 <Diablo-D3> bam. mine.
4041 2011-05-05 20:41:35 <gjs278> becuase I thought my brothers board would work but nope
4042 2011-05-05 20:41:35 <diki> give it to me :D
4043 2011-05-05 20:41:38 <Diablo-D3> gjs278: well
4044 2011-05-05 20:41:46 <Diablo-D3> this now means I can implement bfi_int
4045 2011-05-05 20:41:49 <gjs278> yeah
4046 2011-05-05 20:41:56 <gjs278> this means I had to buy a 1x riser
4047 2011-05-05 20:41:57 <phantomcircuit> redwyrm, like that?
4048 2011-05-05 20:41:58 <netxshare> YAY Diablo
4049 2011-05-05 20:42:06 <netxshare> you should have just asked me to test your code
4050 2011-05-05 20:42:08 <netxshare> I would have ran it
4051 2011-05-05 20:42:08 <gjs278> do you think I can do a 5850 a 5870 and a 5970 on an 850w
4052 2011-05-05 20:42:09 <netxshare> =(
4053 2011-05-05 20:42:16 <netxshare> or given you access to my system
4054 2011-05-05 20:42:19 <bk128> ArtForz: wow is that what you use? for pwm fans?
4055 2011-05-05 20:42:19 <Diablo-D3> netxshare: its hard to write code when you cant repeatedly test it locally
4056 2011-05-05 20:42:24 <ArtForz> nope
4057 2011-05-05 20:42:26 <redwyrm> phantomcircuit: like what? huh?
4058 2011-05-05 20:42:35 <ArtForz> thats for siimple fans with tach
4059 2011-05-05 20:42:39 <Diablo-D3> netxshare: remote debugging is fucking shit, honestly
4060 2011-05-05 20:42:45 <netxshare> I understand
4061 2011-05-05 20:42:45 Marcel has joined
4062 2011-05-05 20:42:55 <ArtForz> I used those with the scythe UK3Ks
4063 2011-05-05 20:42:56 <gjs278> plus free 5850 in 3 - 4 weeks
4064 2011-05-05 20:43:10 <ArtForz> all 3 fans ore on one FET, so they can only be controlled as a group
4065 2011-05-05 20:43:10 <gjs278> and SHOGUN 2
4066 2011-05-05 20:43:21 <ArtForz> independent rpm monitoring
4067 2011-05-05 20:43:29 <gjs278> they always package allstar games with hardware
4068 2011-05-05 20:44:06 <ArtForz> notice the BC327 5->12V level shifter followed by a BC327+BC337 driver stage ;)
4069 2011-05-05 20:44:27 <ArtForz> and yeah, that design is kidna weird, as I designed it to be easily buildable on protoboard
4070 2011-05-05 20:44:28 <gjs278> I really hope my motherboard doesn't fuck this 1x shit up on me. for right now, I can only get video output from my first card in the 16x slot, so it better not think the 1x is now the first card
4071 2011-05-05 20:45:20 <ArtForz> I like using 0603 SMDs on 0.1" hole spacing protoboard
4072 2011-05-05 20:45:50 devon_hillard has quit (Read error: Connection reset by peer)
4073 2011-05-05 20:45:50 <ArtForz> as you can "cheat" them onto pads already occupied with a through-hole
4074 2011-05-05 20:49:22 bitdragon has joined
4075 2011-05-05 20:50:50 <Diablo-D3> [04:38:12] <gjs278> and SHOGUN 2
4076 2011-05-05 20:50:56 <Diablo-D3> is that some sort of video game?
4077 2011-05-05 20:50:59 <gjs278> yeah
4078 2011-05-05 20:51:03 <gjs278> that's what the card will come with
4079 2011-05-05 20:51:08 <gjs278> they give you a steam code to download the game
4080 2011-05-05 20:51:22 <gjs278> Get Shogun game w/ combo purchase, limited offer
4081 2011-05-05 20:51:28 <gjs278> unless you didn't do the shogun combo
4082 2011-05-05 20:51:53 <gjs278> in which case nothing of value was lost seriously it's not a good game
4083 2011-05-05 20:52:32 devon_hillard has joined
4084 2011-05-05 20:52:32 devon_hillard has quit (Changing host)
4085 2011-05-05 20:52:32 devon_hillard has joined
4086 2011-05-05 20:52:47 <Diablo-D3> gjs278: heh
4087 2011-05-05 21:03:57 tabsa has joined
4088 2011-05-05 21:05:46 eao has quit (Quit: Leaving)
4089 2011-05-05 21:05:58 Kiba has joined
4090 2011-05-05 21:06:17 glassresistor has quit (Remote host closed the connection)
4091 2011-05-05 21:07:13 ersi has quit (Read error: Operation timed out)
4092 2011-05-05 21:07:55 <diki> steam sucks
4093 2011-05-05 21:08:11 <diki> i suggest selling the game for BTC
4094 2011-05-05 21:08:16 <diki> at full price converted to btc
4095 2011-05-05 21:08:23 <diki> maybe a few less for motivation
4096 2011-05-05 21:08:29 <diki> and then send the code
4097 2011-05-05 21:08:44 BitMark_afk is now known as BitMark
4098 2011-05-05 21:08:53 <luke-jr> diki: I tried tat
4099 2011-05-05 21:08:55 <luke-jr> nobody buys
4100 2011-05-05 21:08:56 <diki> result?
4101 2011-05-05 21:08:59 <diki> darn
4102 2011-05-05 21:09:00 <diki> why?
4103 2011-05-05 21:09:04 <luke-jr> you tell me
4104 2011-05-05 21:09:10 <luke-jr> wanna change it any buy?
4105 2011-05-05 21:09:13 <luke-jr> and*
4106 2011-05-05 21:09:27 <diki> if it was about crysis 2 i would
4107 2011-05-05 21:09:32 <diki> but i already bought it real cheap
4108 2011-05-05 21:09:48 <diki> ~6 coins
4109 2011-05-05 21:11:01 dick has quit (Read error: Connection reset by peer)
4110 2011-05-05 21:11:27 <phantomcircuit> luke-jr, what game?
4111 2011-05-05 21:11:31 <diki> ;;isitdown http://mtgox.com/
4112 2011-05-05 21:11:31 <gribble> http://http%3A -> Huh? Error... - Check if your website is up or down?
4113 2011-05-05 21:11:35 <diki> ;;isitdown http://mtgox.com
4114 2011-05-05 21:11:35 <gribble> http://http%3A -> Huh? Error... - Check if your website is up or down?
4115 2011-05-05 21:11:39 <diki> ;;isitdown mtgox.com
4116 2011-05-05 21:11:44 <gribble> http://mtgox.com Is Down -> Check if your website is up or down?
4117 2011-05-05 21:11:59 <diki> so how much is 1BTC then?
4118 2011-05-05 21:12:07 <diki> i hope it's above 3$
4119 2011-05-05 21:12:08 <phantomcircuit> currently about 3.8USD
4120 2011-05-05 21:12:11 <phantomcircuit> something like that
4121 2011-05-05 21:12:12 <diki> zomg
4122 2011-05-05 21:12:15 <luke-jr> phantomcircuit: Civ5
4123 2011-05-05 21:12:42 <ArtForz> err... currently about 3.4USD
4124 2011-05-05 21:12:47 <gjs278> bitcoins are 3.40
4125 2011-05-05 21:12:52 <anarchyx> going for 10
4126 2011-05-05 21:12:58 ersi has joined
4127 2011-05-05 21:13:00 <phantomcircuit> luke-jr, how much you want for it?
4128 2011-05-05 21:13:20 <luke-jr> phantomcircuit: make an offer
4129 2011-05-05 21:14:22 <diki> you think it would sell for 3,3?
4130 2011-05-05 21:14:23 <luke-jr> even 12 BTC would be good, but heck, it's not selling at all, so I'll probably take anything reasonable
4131 2011-05-05 21:16:28 <diki> if i can sell my 6,5 btc for at least 3,3USD i think i will have enough for the 212+ to order
4132 2011-05-05 21:18:51 <edcba> $3.4/BTC nice
4133 2011-05-05 21:19:03 <edcba> it keeps rising lol
4134 2011-05-05 21:19:17 <edcba> i still don't understand why but i don't really care lol
4135 2011-05-05 21:20:16 <ArtForz> actually it dropped again, we were ~$4 already
4136 2011-05-05 21:20:33 <edcba> oh buy ! buy !
4137 2011-05-05 21:20:51 <gjs278> I blame the fall on mtgox
4138 2011-05-05 21:20:57 <gjs278> and paypal
4139 2011-05-05 21:20:58 <ArtForz> huw?
4140 2011-05-05 21:21:18 <gjs278> because the coins were harder to liquidate with the ddos and paypal turning off coincard
4141 2011-05-05 21:21:20 <edcba> i blame ArtForz not selling his bitcoins !
4142 2011-05-05 21:21:27 <ArtForz> the drop was 5 days ago, before the DDoS ...
4143 2011-05-05 21:21:37 <gjs278> when did the ddos start
4144 2011-05-05 21:22:34 Marcel has left (HSD!~Marcel|HS@router2.hsdev.com|)
4145 2011-05-05 21:22:38 jaybny has joined
4146 2011-05-05 21:22:56 <ArtForz> also 5 days ago
4147 2011-05-05 21:23:08 <ArtForz> but it really didnt do much for the first day or so
4148 2011-05-05 21:23:22 <gjs278> I think coincard was april 30th
4149 2011-05-05 21:23:26 <ArtForz> yep
4150 2011-05-05 21:23:26 <gjs278> or may 1st
4151 2011-05-05 21:23:40 <ArtForz> deal with paypal, get screwed, news at 11
4152 2011-05-05 21:23:44 <gjs278> either way when I saw that I definitely sold what I had on me
4153 2011-05-05 21:23:48 forever-d has quit (Ping timeout: 264 seconds)
4154 2011-05-05 21:24:31 <ArtForz> I just don't get why people keep thinking *they* won't get screwed by paypal
4155 2011-05-05 21:24:54 <CIA-30> bitcoin: droidbird exper * r797f6a99b1c5 intersango/util.php: added a function to check rsa signatures. http://tinyurl.com/5u6en84
4156 2011-05-05 21:24:58 <CIA-30> bitcoin: Chris Howie * rd3c7bee32585 mining-proxy/htdocs/common.inc.php: Detect if site_uri ends with a slash to fix incorrect URI generation http://tinyurl.com/63rntdl
4157 2011-05-05 21:25:18 <gjs278> I knew they would get shutodwn but it was still pretty nice
4158 2011-05-05 21:25:20 <Kiba> ArtForz: what do you propose we use?
4159 2011-05-05 21:25:43 <ArtForz> shrug, getting kinda hard
4160 2011-05-05 21:26:29 <Kiba> I guess we have to relies on intercontinental network of bitcoin exchangers on the ground
4161 2011-05-05 21:26:38 <gjs278> it's the fucking fees on everything else
4162 2011-05-05 21:26:43 <ArtForz> a few years ago - get a merch acct and a CC processor, but thanks to the "war on gambling", thats now also tricky as hell
4163 2011-05-05 21:27:00 <gjs278> I hate conversion fees for these fake dollar services
4164 2011-05-05 21:27:08 <gjs278> paypal I just don't have to deal with it
4165 2011-05-05 21:27:18 <ArtForz> lol?
4166 2011-05-05 21:27:36 <ArtForz> paypal fees are worse than CC
4167 2011-05-05 21:27:48 <gjs278> I've nevre paid a paypal fee
4168 2011-05-05 21:27:48 allied has joined
4169 2011-05-05 21:28:00 <ArtForz> lemme guess, you also think CCs don't have fees, right
4170 2011-05-05 21:28:02 <gjs278> the only time I ever get a fee is from project wonderful when they do their payouts
4171 2011-05-05 21:28:06 <gjs278> I don't use cc's
4172 2011-05-05 21:28:11 <gjs278> just debit for now
4173 2011-05-05 21:28:23 <gjs278> I know if you use a cc through paypal to pay there's definitely a fee
4174 2011-05-05 21:28:31 <ArtForz> paypal fees for merchants are several % (at least they used to a few years back)
4175 2011-05-05 21:28:53 <gjs278> project wonderful made me cover the fee
4176 2011-05-05 21:28:58 <gjs278> but when I pay using paypal on like newegg
4177 2011-05-05 21:28:59 <gjs278> no fee
4178 2011-05-05 21:29:09 <ArtForz> *headdesk*
4179 2011-05-05 21:29:19 <gjs278> unless they're covering it all of this time
4180 2011-05-05 21:29:19 <ArtForz> the *merchant* pays the fees
4181 2011-05-05 21:29:25 <ArtForz> same for CCs
4182 2011-05-05 21:29:26 <gjs278> yeah
4183 2011-05-05 21:29:31 <gjs278> but I only care if I have to pay
4184 2011-05-05 21:29:32 <ArtForz> it's in the fucking merchant agreement
4185 2011-05-05 21:29:37 SerajewelKS has joined
4186 2011-05-05 21:29:44 <gjs278> if they're going to foot it that's fine by me
4187 2011-05-05 21:29:50 <ArtForz> yeah, and they totally don't add that fee to the price, no siree
4188 2011-05-05 21:29:55 <gjs278> they don't
4189 2011-05-05 21:30:04 <gjs278> otherwise they're adding it to the price of people who pay with cc as well
4190 2011-05-05 21:30:07 <ArtForz> yep
4191 2011-05-05 21:30:11 <gjs278> or direct bank deposit like on amazon
4192 2011-05-05 21:30:28 <gjs278> which is actually pretty convenient
4193 2011-05-05 21:30:29 robotarmy has joined
4194 2011-05-05 21:30:38 <ArtForz> guess how they cover the 2-4% fees CC companies charge their merchants?
4195 2011-05-05 21:30:41 <ArtForz> they get priced in
4196 2011-05-05 21:30:43 <ArtForz> only way
4197 2011-05-05 21:30:50 <redwyrm> it seems to me like a better format for the wallet.dat is a directory with each pubkey in its own file with the Bitcoin address as the filename
4198 2011-05-05 21:31:08 <redwyrm> and perhaps using JSON to store the pubkey along with metadata
4199 2011-05-05 21:31:42 <redwyrm> on a side note, is there planned support in the future for encrypted wallets?
4200 2011-05-05 21:32:55 <gjs278> put it in a truecrypt container
4201 2011-05-05 21:33:03 pnicholson has quit (Quit: pnicholson)
4202 2011-05-05 21:34:42 <ArtForz> redwyrm: are you trying to kill the FS?
4203 2011-05-05 21:34:57 <ArtForz> it's not that uncommon to have 10000s of keypairs in a wallet
4204 2011-05-05 21:35:11 <gjs278> all of those precious inodes
4205 2011-05-05 21:35:20 slush1 has quit (Quit: Leaving.)
4206 2011-05-05 21:35:30 NOTAL has quit (Remote host closed the connection)
4207 2011-05-05 21:37:55 xelister has joined
4208 2011-05-05 21:39:06 <diki> block 122100 is close
4209 2011-05-05 21:39:26 <JFK911> ;;bc,stats
4210 2011-05-05 21:39:27 <gribble> Current Blocks: 122099 | Current Difficulty: 109670.13329248 | Next Difficulty At Block: 122975 | Next Difficulty In: 876 blocks | Next Difficulty In About: 4 days, 9 hours, 21 minutes, and 48 seconds | Next Difficulty Estimate: 150367.37959783
4211 2011-05-05 21:39:49 <gjs278> deepbit is eating blocks non stop
4212 2011-05-05 21:40:00 <diki> meaning?
4213 2011-05-05 21:40:02 <gjs278> I saw like 6 blocks in a row under 10 minutes
4214 2011-05-05 21:42:11 <JFK911> 150k lol
4215 2011-05-05 21:43:02 <diki> ;;bc,calc 1000000
4216 2011-05-05 21:43:03 <gribble> The average time to generate a block at 1000000 Khps, given current difficulty of 109670.13329248 , is 5 days, 10 hours, 50 minutes, and 29 seconds
4217 2011-05-05 21:43:04 kseistrup has quit (Quit: leaving)
4218 2011-05-05 21:43:09 <diki> ;;bc,calc 10000000
4219 2011-05-05 21:43:10 <gribble> The average time to generate a block at 10000000 Khps, given current difficulty of 109670.13329248 , is 13 hours, 5 minutes, and 2 seconds
4220 2011-05-05 21:43:20 <diki> ;;bc,calc 100000000
4221 2011-05-05 21:43:20 <gribble> The average time to generate a block at 100000000 Khps, given current difficulty of 109670.13329248 , is 1 hour, 18 minutes, and 30 seconds
4222 2011-05-05 21:43:38 <redwyrm> ArtForz: it takes more than that many files to kill an FS
4223 2011-05-05 21:43:53 <ArtForz> yes
4224 2011-05-05 21:43:59 <diki> Wouldn't it be cool if there was an eggdrop about blocks being found here?
4225 2011-05-05 21:44:01 <ArtForz> otherwise mail spools wouldnt work
4226 2011-05-05 21:44:17 <redwyrm> and in any case, you could have AA/ for keys starting with 1AA..., AB/ for keys starting with 1AB... etc.
4227 2011-05-05 21:44:33 <ArtForz> but it will still cause a lot of rather unneccesary thrashing
4228 2011-05-05 21:45:00 <tcatm> diki: like #bitcoin-watch?
4229 2011-05-05 21:45:15 <redwyrm> ArtForz: okay, you convinced me :)
4230 2011-05-05 21:45:20 <redwyrm> but waht about my other question?
4231 2011-05-05 21:45:26 <redwyrm> encrypted wallet support
4232 2011-05-05 21:45:33 <redwyrm> seems like a no-brainer
4233 2011-05-05 21:45:42 <BitterTea> redwyrm: It's in the planning/implementation stages
4234 2011-05-05 21:45:43 skizza1 has joined
4235 2011-05-05 21:45:47 <ArtForz> people are already working on that
4236 2011-05-05 21:45:50 <redwyrm> my home directory is encrypted but I'm sure many people's aren't
4237 2011-05-05 21:45:53 <BitterTea> There are many ways you can go about wallet encryption though
4238 2011-05-05 21:45:53 <redwyrm> ah, cool :)
4239 2011-05-05 21:45:58 <redwyrm> BitterTea: right
4240 2011-05-05 21:46:06 <BitterTea> Personally, I prefer individually encrypted private keys
4241 2011-05-05 21:46:09 <ArtForz> seems the idea is to keep the wallet format mostly and just encrypt the privkeys
4242 2011-05-05 21:46:24 <BitterTea> Oh yeah, they are going with that model
4243 2011-05-05 21:46:26 skizza1 has left ()
4244 2011-05-05 21:46:32 <BitterTea> Basically, you'll only have to enter a password to spend money
4245 2011-05-05 21:46:55 scottlad has joined
4246 2011-05-05 21:47:44 <ArtForz> yep
4247 2011-05-05 21:47:54 <ArtForz> good enough to stop simple drive-by wallet stealing
4248 2011-05-05 21:48:43 <ArtForz> ofc it wont protect against keylogging trojans, but at that point your'e already SOL unless the wallet is on some secure external computing device
4249 2011-05-05 21:48:54 <redwyrm> an on-screen keyboard would be awesome, btw
4250 2011-05-05 21:49:10 <ArtForz> for what?
4251 2011-05-05 21:49:13 <redwyrm> if people can raise 5BTC for this feature, I'll implement it!
4252 2011-05-05 21:49:15 <redwyrm> for the password
4253 2011-05-05 21:49:17 <gjs278> so you cant get keylogged
4254 2011-05-05 21:49:21 <ArtForz> soudns completely useless
4255 2011-05-05 21:49:23 <gjs278> lol
4256 2011-05-05 21:49:27 <gjs278> how is it useless
4257 2011-05-05 21:49:28 bitdragon has quit (Quit: Leaving)
4258 2011-05-05 21:49:32 <BitterTea> Keyloggers today also mouse log
4259 2011-05-05 21:49:33 <gjs278> unless they screenshot or record every mouse clock
4260 2011-05-05 21:49:35 <ArtForz> and decent malware kit nowadays can capture that
4261 2011-05-05 21:49:36 <BitterTea> And take screenshots, etc
4262 2011-05-05 21:49:39 <gjs278> ok
4263 2011-05-05 21:49:42 <redwyrm> ArtForz: screenshot+mouse logging is harder than keylogging
4264 2011-05-05 21:49:44 <gjs278> you would need a screenshot
4265 2011-05-05 21:49:47 <gjs278> every mouseclick
4266 2011-05-05 21:49:51 <gjs278> because the numbers move
4267 2011-05-05 21:49:55 <gjs278> they don't sit still
4268 2011-05-05 21:50:08 <redwyrm> or at least a screenshot every time the bitcoin window is un-minimized or moved
4269 2011-05-05 21:50:25 <gjs278> like it has 120 buttons but the buttons rearrange themselves every click
4270 2011-05-05 21:50:34 <gjs278> it would take long as fuck to type in your password
4271 2011-05-05 21:50:34 <scottlad> Anyone interested in buying some hosting or exchange some btc for hosting
4272 2011-05-05 21:50:43 <ArtForz> yes, lots of online banking systems use that scheme
4273 2011-05-05 21:50:46 <BitterTea> Here's what I'd prefer... store private keys encrypted in the wallet. Require either a password or hardware authentication token to decrypt the keys before use
4274 2011-05-05 21:51:31 <BitterTea> Not sure how feasible the second option is, but I think that would be the next step up from passphrase protected private keys
4275 2011-05-05 21:51:37 <ArtForz> I'm working on something a bit different
4276 2011-05-05 21:51:42 <gjs278> a password seems like it would be more than enough
4277 2011-05-05 21:51:45 <ArtForz> the private keys never leave the external device ;)
4278 2011-05-05 21:51:52 <BitterTea> Ooh, I like it
4279 2011-05-05 21:53:25 <ArtForz> I should write a document explaining the process, it's rather simple
4280 2011-05-05 21:53:30 Kiba has quit (Remote host closed the connection)
4281 2011-05-05 21:53:41 draag has joined
4282 2011-05-05 21:54:15 <BitterTea> How does Bitcoin access the private keys? Does it send data to the device to get signed?
4283 2011-05-05 21:54:20 <ArtForz> yep
4284 2011-05-05 21:54:42 <ArtForz> bitcoin sends transaction to be signed and the transactions that provide the input to the to-be-signed tx
4285 2011-05-05 21:55:10 <luke-jr> ArtForz: isn't that TPM?
4286 2011-05-05 21:55:29 <ArtForz> what?
4287 2011-05-05 21:55:33 <ArtForz> okay, you can leave out the input txes if you don't want to have the device show total input amount
4288 2011-05-05 21:55:39 <ArtForz> no
4289 2011-05-05 21:55:46 <ArtForz> TPM is something entriely different
4290 2011-05-05 21:56:09 draaglom has quit (Ping timeout: 240 seconds)
4291 2011-05-05 21:56:49 <BitterTea> How long until such a device can be purchased and used with Bitcoin, do you think?
4292 2011-05-05 21:57:36 <ArtForz> stil lworking on the firmware
4293 2011-05-05 21:58:09 <ArtForz> already got a prototype of the hw breadboarded
4294 2011-05-05 21:58:38 <BitMark> can the device generate its own keys?
4295 2011-05-05 21:58:48 <ArtForz> currently, no
4296 2011-05-05 21:58:54 <ArtForz> but I'm planning for that
4297 2011-05-05 21:59:09 <BitMark> cool
4298 2011-05-05 21:59:17 lyspooner has quit (Quit: ChatZilla 0.9.86.1 [Firefox 3.6.17/20110420140830])
4299 2011-05-05 21:59:18 <gjs278> include a backdoor
4300 2011-05-05 21:59:34 <BitMark> backdoor wouldnt get him much
4301 2011-05-05 21:59:45 <ArtForz> that'd be hard, as it'll be open hardware
4302 2011-05-05 21:59:47 <BitterTea> ArtForz: What about backups?
4303 2011-05-05 21:59:48 <BitMark> would need physical access to exploit
4304 2011-05-05 21:59:55 <CIA-30> bitcoin: droidbird exper * r69a1c6e14c47 intersango/api.php: added a module for http api functions. http://tinyurl.com/6exjnry
4305 2011-05-05 21:59:56 <CIA-30> bitcoin: droidbird exper * r192cb214c9ac intersango/www/api/ (buyBTC.php sellBTC.php): added sellBTC and buyBTC api methods. http://tinyurl.com/67ynd35
4306 2011-05-05 21:59:58 <CIA-30> bitcoin: droidbird exper * r7ae9d6711d45 intersango/util.php: set uid on session after successfull signature check. http://tinyurl.com/5rrh4vb
4307 2011-05-05 22:00:00 <ArtForz> BitterTea: might be possible, but ofc thats a risky thing to do
4308 2011-05-05 22:00:16 mologie has quit (Quit: Nettalk6 - www.ntalk.de)
4309 2011-05-05 22:00:22 <BitterTea> Yeah, kind of defeats the purpose
4310 2011-05-05 22:00:38 <BitMark> make a safe for creating keys and installing them onto the device
4311 2011-05-05 22:00:38 <BitMark> store back ups in the safe
4312 2011-05-05 22:00:39 <ArtForz> hrrrrm
4313 2011-05-05 22:00:54 <BitMark> keys never leave safe not on this device
4314 2011-05-05 22:01:00 <ArtForz> yea
4315 2011-05-05 22:01:01 <tcatm> backup to microsd directly from device?
4316 2011-05-05 22:01:05 <BitMark> safe is never ever connected to internet
4317 2011-05-05 22:01:11 <gjs278> cloning machine, atom for atom
4318 2011-05-05 22:01:17 <ArtForz> yep, thats what I was thinking
4319 2011-05-05 22:01:50 <ArtForz> well, theres another way
4320 2011-05-05 22:01:57 <BitMark> how many keys would fit on the device?
4321 2011-05-05 22:02:21 <ArtForz> currently, 8
4322 2011-05-05 22:02:39 <ArtForz> though the code I got atm can only handle 1
4323 2011-05-05 22:02:44 <BitMark> ah
4324 2011-05-05 22:03:04 <ArtForz> "just use a larger eeprom" isn't exactly rocket science ;)
4325 2011-05-05 22:03:10 <BitMark> :)
4326 2011-05-05 22:03:10 leedo has joined
4327 2011-05-05 22:03:18 <phantomcircuit> oh wow
4328 2011-05-05 22:03:24 <ArtForz> and yes, that means it's not secure against local attacks on the physical device
4329 2011-05-05 22:03:32 <BitMark> nothing is
4330 2011-05-05 22:03:34 <phantomcircuit> mssql cannot handle indexes larger than 900bytes
4331 2011-05-05 22:03:38 <phantomcircuit> FFFFFFFFFFFFFUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU
4332 2011-05-05 22:03:40 <BitMark> unless encrypted
4333 2011-05-05 22:03:41 <phantomcircuit> :|
4334 2011-05-05 22:03:46 <ArtForz> exactly
4335 2011-05-05 22:03:59 <Diablo-D3> phantomcircuit: er, what?
4336 2011-05-05 22:04:12 <ArtForz> and having people enter a "secure enough" password on a PINpad is infeasible
4337 2011-05-05 22:04:15 <gjs278> isnt every index larger than 900 bytes
4338 2011-05-05 22:04:30 <Diablo-D3> dude, an index thats under 900 bytes has like 5 rows in it
4339 2011-05-05 22:04:36 <ArtForz> you'd need 20+ digits
4340 2011-05-05 22:04:37 <phantomcircuit> lol no
4341 2011-05-05 22:04:43 <gjs278> unless you mean it cant index a column over 900bytes
4342 2011-05-05 22:04:44 <phantomcircuit> the sum of the columns in the index
4343 2011-05-05 22:04:51 <BitterTea> ArtForz: What about some sort of physical security?
4344 2011-05-05 22:05:04 <phantomcircuit> gjs278, the sum of the columns must be less than 900 bytes
4345 2011-05-05 22:05:17 <BitterTea> Perhaps overkill?
4346 2011-05-05 22:05:19 <phantomcircuit> which means i cant have a uniqueness constraint for a urls column
4347 2011-05-05 22:05:21 <phantomcircuit> :|
4348 2011-05-05 22:05:22 <gjs278> what's the sum of a + a
4349 2011-05-05 22:05:30 <ArtForz> BitterTea: that stuff is hard, and tends to add a lot of cost
4350 2011-05-05 22:05:39 <redwyrm> ArtForz: right, PIN pads can only be used for cases where an attacker doesn't have physical control
4351 2011-05-05 22:05:42 <gjs278> actually
4352 2011-05-05 22:05:44 <gjs278> why use mssql
4353 2011-05-05 22:05:49 <gjs278> what does it have you can't get from the others
4354 2011-05-05 22:05:58 <BitMark> ArtForz could store decryption key on separate device, but that probably doesnt gain much
4355 2011-05-05 22:06:02 <redwyrm> like over the network, or a physically secure device
4356 2011-05-05 22:06:15 <phantomcircuit> gjs278, it's not my choice, like usual
4357 2011-05-05 22:06:19 <gjs278> yeah
4358 2011-05-05 22:06:21 <gjs278> enjoy
4359 2011-05-05 22:06:23 <ArtForz> actually you can do something like that already
4360 2011-05-05 22:06:47 sytse has joined
4361 2011-05-05 22:06:59 <ArtForz> have bitcoin running on a non-networked computer
4362 2011-05-05 22:07:15 <Diablo-D3> better yet
4363 2011-05-05 22:07:18 <Diablo-D3> an unnetworkable one
4364 2011-05-05 22:07:37 <gjs278> destroy the ethernet port
4365 2011-05-05 22:07:42 <gjs278> destroy the pci slots
4366 2011-05-05 22:07:45 <Diablo-D3> no bluetooth, no wifi, and put an ethernet lock on the port
4367 2011-05-05 22:07:45 <gjs278> and the usb ports
4368 2011-05-05 22:07:49 <redwyrm> or uh... take out the cards
4369 2011-05-05 22:07:54 <redwyrm> instead of destroying them :)
4370 2011-05-05 22:07:55 <gjs278> no it must be destroyed
4371 2011-05-05 22:08:02 <gjs278> with thermite
4372 2011-05-05 22:08:05 <redwyrm> hahaha
4373 2011-05-05 22:08:09 <ArtForz> nuke it from orbit
4374 2011-05-05 22:08:21 <redwyrm> speaking of which
4375 2011-05-05 22:08:21 <ArtForz> generate key, encrypt + back up wallet, write down address, send bitcoins to address from any other computer ;)
4376 2011-05-05 22:08:30 <redwyrm> does anyone know where I can buy enriched uranium with BTC?
4377 2011-05-05 22:08:32 <gjs278> p. sure the security on the nuclear launch codes is less than what people are willing to do for their bitcoin wallet
4378 2011-05-05 22:08:40 <Diablo-D3> [05:59:44] <gjs278> unless you mean it cant index a column over 900bytes
4379 2011-05-05 22:08:47 <redwyrm> ;)
4380 2011-05-05 22:08:47 AStove has quit ()
4381 2011-05-05 22:08:49 <redwyrm> (PM me)
4382 2011-05-05 22:08:51 <Diablo-D3> oh, indexing large blobs or text fields? yeah, fuck that dude
4383 2011-05-05 22:08:59 <Diablo-D3> dont use db indexes for that kind of shit
4384 2011-05-05 22:09:02 <gjs278> yeah
4385 2011-05-05 22:09:08 <gjs278> you make a reference column somewhere else
4386 2011-05-05 22:09:12 <gjs278> for a large text field table
4387 2011-05-05 22:09:12 <ArtForz> gjs278: hey, at least the launch code isn't 000000 anymore *g*
4388 2011-05-05 22:09:25 <Diablo-D3> no, if its a text column, use a dedicated text search package
4389 2011-05-05 22:09:46 <gjs278> I don't see any reason to index a text column either way
4390 2011-05-05 22:09:48 <gjs278> I never where that shit
4391 2011-05-05 22:09:53 <CIA-30> bitcoin: droidbird exper * rdcab3b6a85ac intersango/util.php: add a function to check if user has an uploaded rsa key. http://tinyurl.com/4yut4nv
4392 2011-05-05 22:09:58 <Diablo-D3> gjs278: well, you effectively cant
4393 2011-05-05 22:10:02 <Diablo-D3> you're going to want to do REAL searches on it
4394 2011-05-05 22:10:06 <gjs278> yeah
4395 2011-05-05 22:10:07 <Diablo-D3> which requires a real package to do it
4396 2011-05-05 22:10:17 <phantomcircuit> i would be cool with indexes of hashes
4397 2011-05-05 22:10:26 <phantomcircuit> i just want to guarantee uniqueness
4398 2011-05-05 22:10:47 <io_error> perl ms\do_
4399 2011-05-05 22:11:10 jrabbit has quit (Ping timeout: 258 seconds)
4400 2011-05-05 22:11:28 <sytse> BlueMatt: this weekend..
4401 2011-05-05 22:11:35 <phantomcircuit> fuck i guess that's what i have to do
4402 2011-05-05 22:11:36 <io_error> perl ms\do_
4403 2011-05-05 22:11:43 <BlueMatt> sytse: fair enough, see you then
4404 2011-05-05 22:11:50 <phantomcircuit> hash the lowercase url and set the hash to unique
4405 2011-05-05 22:11:51 <phantomcircuit> sigh
4406 2011-05-05 22:11:54 <phantomcircuit> ridiculousness
4407 2011-05-05 22:12:32 leedo has quit (Quit: Page closed)
4408 2011-05-05 22:13:24 BitterTea has left ()
4409 2011-05-05 22:14:12 <jaybny> ;;bc,mtgox
4410 2011-05-05 22:14:12 <gribble> {"ticker":{"high":3.479,"low":3.333,"vol":16477,"buy":3.332,"sell":3.38,"last":3.333}}
4411 2011-05-05 22:14:32 <Diablo-D3> [06:05:19] <phantomcircuit> i would be cool with indexes of hashes
4412 2011-05-05 22:14:33 <Diablo-D3> and then
4413 2011-05-05 22:14:34 <Diablo-D3> BAM
4414 2011-05-05 22:14:36 <jaybny> anyway to see our orders before comission on mtgox?
4415 2011-05-05 22:14:37 <Diablo-D3> rainbowtables
4416 2011-05-05 22:15:41 devrandom has quit (Ping timeout: 246 seconds)
4417 2011-05-05 22:15:57 danlucraft has quit (Quit: Leaving.)
4418 2011-05-05 22:16:44 amiller has quit (Ping timeout: 246 seconds)
4419 2011-05-05 22:18:56 jrabbit has joined
4420 2011-05-05 22:19:07 <redwyrm> under what circumstances will the Bitcoin client delete a public key from the wallet?
4421 2011-05-05 22:20:11 <redwyrm> is anybody working on a Bitcoin iPhone app?
4422 2011-05-05 22:20:19 <luke-jr> yes
4423 2011-05-05 22:20:24 <ArtForz> never
4424 2011-05-05 22:20:29 <redwyrm> haha
4425 2011-05-05 22:20:47 <redwyrm> would it violate Apple's app store rules?
4426 2011-05-05 22:21:00 <luke-jr> shrug
4427 2011-05-05 22:21:01 <redwyrm> (assuming no other problems like poor interface)
4428 2011-05-05 22:21:19 <luke-jr> Apple's appstore rules are draconian, so probably
4429 2011-05-05 22:21:40 <redwyrm> a lot of people won't jailbreak their phones because they don't want their warranties voided (like my girlfriend)
4430 2011-05-05 22:21:56 <redwyrm> so if it's not in the appstore, it won't see nearly as much use
4431 2011-05-05 22:22:01 <redwyrm> therefore...
4432 2011-05-05 22:22:01 tenach has quit (Quit: reloading shtuff.)
4433 2011-05-05 22:22:07 <luke-jr> redwyrm: jailbreaking won't void your warranty
4434 2011-05-05 22:22:08 <redwyrm> it's worth it to see how it can be approved
4435 2011-05-05 22:22:18 agricocb has quit (Quit: Leaving.)
4436 2011-05-05 22:22:19 <BlueMatt> redwyrm: there are a ton of question as to implementation for mobile clients
4437 2011-05-05 22:22:29 <BlueMatt> shoving a regular bitcoin client into a phone is a terrible idea
4438 2011-05-05 22:22:33 <redwyrm> BlueMatt: I agree
4439 2011-05-05 22:22:34 <BlueMatt> (for many reasons)
4440 2011-05-05 22:22:37 <luke-jr> of course, you might have to sue Apple to get them to honour itâ¦
4441 2011-05-05 22:22:44 <redwyrm> how about a standard wallet service interface
4442 2011-05-05 22:22:54 <BlueMatt> and so far there isnt much of a better option (any kind of p2p client on a phone is a terrible idea)
4443 2011-05-05 22:22:56 <luke-jr> in reality, it's your own fault for buying crap like Apple
4444 2011-05-05 22:23:00 <redwyrm> so the user could just check off which wallet sites they trust
4445 2011-05-05 22:23:04 <luke-jr> redwyrm: working on it. want to help?
4446 2011-05-05 22:23:12 <redwyrm> luke-jr: yes
4447 2011-05-05 22:23:12 <BitMark> any ideas on what would keep a pretty standard tx from getting encoded to block chain?
4448 2011-05-05 22:23:24 <luke-jr> redwyrm: https://en.bitcoin.it/wiki/Wallet_protocol
4449 2011-05-05 22:23:27 <redwyrm> I worked on Rock Prodigy for almost a year
4450 2011-05-05 22:23:43 <BlueMatt> BitMark: if it IsStandard() size vs fee is about it
4451 2011-05-05 22:23:46 <luke-jr> redwyrm: Draft 0 is probably too overcomplicated, but might be useful for ideas
4452 2011-05-05 22:23:46 <redwyrm> http://itunes.apple.com/us/app/rock-prodigy-guitar/id407303228?mt=8
4453 2011-05-05 22:24:00 <BitMark> ie http://bitcoincharts.com/bitcoin/ seems really long
4454 2011-05-05 22:24:36 <BitMark> 0b5f1cc70ed93f0a7a2cf442e440b153e425e0ff55f5c7fee84a7252ec96ba10 is the one i am looking at
4455 2011-05-05 22:24:44 <luke-jr> BitMark: did you pay a fee?
4456 2011-05-05 22:25:06 <BitMark> this is a pay out from deepbit, doesn't look like a fee was paid
4457 2011-05-05 22:25:18 <redwyrm> luke-jr: I'm AaronM on the forums, btw
4458 2011-05-05 22:25:34 <BitMark> you'd think he would encode his payouts in his blocks
4459 2011-05-05 22:26:05 <BlueMatt> redwyrm: why cant people use the same nick here+irc?
4460 2011-05-05 22:26:20 <BlueMatt> uh irc+smf sorry
4461 2011-05-05 22:26:53 scottlad has left ()
4462 2011-05-05 22:27:07 <redwyrm> BlueMatt: because gribble won't let me use my GPG key with AaronM since I already registered it with redwyrm
4463 2011-05-05 22:27:14 <Blitzboom> because bitcoin is PSEUDONYMOUS
4464 2011-05-05 22:27:33 <BlueMatt> redwyrm: you can change your gpg key iirc, or just msg nanotube and ask him
4465 2011-05-05 22:27:34 <redwyrm> BlueMatt: gribble is a #bitcoin-otc bot if you didn't know already)
4466 2011-05-05 22:27:37 <redwyrm> oh
4467 2011-05-05 22:27:39 <redwyrm> okay
4468 2011-05-05 22:27:44 <BitMark> luke-jr: how hard would it be to get your pool to prioritize your payouts without fees?
4469 2011-05-05 22:27:48 <BlueMatt> redwyrm: hes also on here ,,bc,mtgox
4470 2011-05-05 22:27:49 <gribble> {"ticker":{"high":3.479,"low":3.333,"vol":16199,"buy":3.34,"sell":3.382,"last":3.383}}
4471 2011-05-05 22:27:55 <luke-jr> BitMark: why would I do that?
4472 2011-05-05 22:28:08 <BitMark> as a perk to your mining affiliates
4473 2011-05-05 22:28:21 <redwyrm> oh, okay
4474 2011-05-05 22:28:44 <luke-jr> BitMark: it has no way to know who is a participating miner
4475 2011-05-05 22:29:02 <luke-jr> BitMark: I would like to do something like that, though
4476 2011-05-05 22:29:18 <BitMark> luke-jr: normally your payouts are in the genesis tx
4477 2011-05-05 22:29:31 <luke-jr> â¦
4478 2011-05-05 22:29:53 <AAA_awright> Anyone know of a plan to pull out the P2P, non-Bitcoin stuff as a library?
4479 2011-05-05 22:30:06 <BitMark> luke-jr: but in the off chance you had to make catch up payouts, you would be doing your affiliates a service by making sure you were including the payout tx in your blocks
4480 2011-05-05 22:30:20 <luke-jr> AAA_awright: multiple plans, just not a priority
4481 2011-05-05 22:30:35 <luke-jr> BitMark: oh, it already does that
4482 2011-05-05 22:30:39 <luke-jr> BitMark: I had to yesterday
4483 2011-05-05 22:30:45 jaybny has left ()
4484 2011-05-05 22:30:48 amiller has joined
4485 2011-05-05 22:30:52 <luke-jr> though someone else included it for some reason
4486 2011-05-05 22:30:53 <BitMark> ah cool did that tx offer a fee?
4487 2011-05-05 22:30:56 <luke-jr> no
4488 2011-05-05 22:31:03 <luke-jr> my codebase never offers a fee
4489 2011-05-05 22:31:11 ThomasV has joined
4490 2011-05-05 22:31:28 <BitMark> thats fine if someone else includes it, but just a matter of goodwill to make sure your pool includes it as well
4491 2011-05-05 22:31:28 joepie93 has joined
4492 2011-05-05 22:31:52 joepie93 is now known as joepie91
4493 2011-05-05 22:32:23 <AAA_awright> luke-jr: Do don't have a link with more information about that, do you?
4494 2011-05-05 22:32:23 <BitMark> in other words it would seem disengenious if you made a payout, then generated a block without the payout
4495 2011-05-05 22:32:25 <luke-jr> BitMark: my codebase will always include a transaction to or from its own addresses
4496 2011-05-05 22:32:33 <BitMark> very cool
4497 2011-05-05 22:32:34 <luke-jr> AAA_awright: no
4498 2011-05-05 22:32:41 <AAA_awright> What's more important?
4499 2011-05-05 22:32:59 <BitMark> or is that just something thats obvious :)
4500 2011-05-05 22:33:19 joepie92 has quit (Ping timeout: 240 seconds)
4501 2011-05-05 22:33:33 <BitMark> cool either way
4502 2011-05-05 22:33:33 <luke-jr> BitMark: maybe I will create a patch that also-sends to a pool address, and have the pool add that to your balance or something
4503 2011-05-05 22:33:45 <luke-jr> AAA_awright: bugs, features
4504 2011-05-05 22:34:31 <AAA_awright> No specific, major features?
4505 2011-05-05 22:34:35 Jaagu has joined
4506 2011-05-05 22:34:37 <BitMark> so you would also prioritize all txes to from pool addresses?
4507 2011-05-05 22:34:55 <CIA-30> bitcoin: droidbird exper * ra7efa938ed6e intersango/util.php: added a function to store the rsa key for a user. http://tinyurl.com/424n2z5
4508 2011-05-05 22:34:56 <CIA-30> bitcoin: droidbird exper * r99a5cf6ace14 intersango/ (addkey.php addkey_action.php profile.php switcher.php): added a mechanism so users can link an rsa public key for their user, to use on api operations. http://tinyurl.com/3wwvr5o
4509 2011-05-05 22:34:59 <luke-jr> BitMark: I don't know that it prioritizes them, but it does take them in lieu of a fee
4510 2011-05-05 22:35:13 <BitMark> luke-jr: thats what i meant
4511 2011-05-05 22:35:30 <BitMark> s/prioritize/include/
4512 2011-05-05 22:37:47 <cosurgi> what's power consumption of 6990 ?
4513 2011-05-05 22:38:21 <cosurgi> me is searching for PSU that will handle 6990+6990+5830
4514 2011-05-05 22:38:27 <ArtForz> ugh
4515 2011-05-05 22:38:38 <cosurgi> /me
4516 2011-05-05 22:38:49 <cosurgi> ArtForz: that 5830 ? a leftower
4517 2011-05-05 22:38:59 <cosurgi> ArtForz: else, why "ugh" ?
4518 2011-05-05 22:39:18 <cosurgi> is it crazy high?
4519 2011-05-05 22:39:24 <ArtForz> ~ 350W stock
4520 2011-05-05 22:39:41 <ArtForz> > 400W in 6970x2 mode
4521 2011-05-05 22:40:34 <ArtForz> so for 2 and a 5830 ... > 1kW PSU
4522 2011-05-05 22:40:54 <cosurgi> ouch, so in total I would need 400+400+175=975
4523 2011-05-05 22:41:09 <ArtForz> a decent 1200W should be able to handle it
4524 2011-05-05 22:41:10 <cosurgi> and 975 should be about half of PSU's power ?
4525 2011-05-05 22:41:21 <ArtForz> well, you can forget half in that config
4526 2011-05-05 22:41:35 x6763 has quit (Ping timeout: 246 seconds)
4527 2011-05-05 22:41:44 <cosurgi> hmm.. I could do some swapping then.
4528 2011-05-05 22:41:49 <ArtForz> even stock you're looking at 850W or so
4529 2011-05-05 22:41:56 <ArtForz> OCed, about 1kW
4530 2011-05-05 22:41:58 <cosurgi> if my other PSUs can handle that.
4531 2011-05-05 22:42:14 <ArtForz> so that means 1200-1500W PSU would be a good size ...
4532 2011-05-05 22:42:52 <ArtForz> might be easier to just go with dual 6990 and a good 1kW, those are quite a bit cheaper than the 1200 and 1500s
4533 2011-05-05 22:43:11 <ArtForz> or 2 decent 750-850W PSUs
4534 2011-05-05 22:43:45 <cosurgi> how about 5830+5850+6990 ? but then the existing PC will have 5850+5870+6990 and already has 750W PSU ....
4535 2011-05-05 22:44:08 <ArtForz> ugh
4536 2011-05-05 22:44:09 <cosurgi> rather bad.
4537 2011-05-05 22:44:13 <ArtForz> yea
4538 2011-05-05 22:45:02 <CIA-30> bitcoin: genjix exper * r0b0bea869de7 intersango/DATABASE: added public key field to users table. http://tinyurl.com/3lh7hm3
4539 2011-05-05 22:45:06 <ArtForz> I'd say a good 1200W PSU should do it, but if you got the space and the skills, dual 850s would probably be cheaper
4540 2011-05-05 22:45:56 <cosurgi> no space. the box must be more-or-less good looking outside. It's in server room, not at home.
4541 2011-05-05 22:46:15 <cosurgi> admins won't like too ugly looking box
4542 2011-05-05 22:46:23 forrestv has quit (Remote host closed the connection)
4543 2011-05-05 22:46:38 <ArtForz> well, say hello to a > $250 PSU ...
4544 2011-05-05 22:46:55 <cosurgi> :/
4545 2011-05-05 22:47:25 <ArtForz> too bad you can't have an adapter plate made
4546 2011-05-05 22:47:43 <ArtForz> otherwise I'd say go with something like dual zippy 850W 1U PSUs
4547 2011-05-05 22:48:51 <cosurgi> you mean 850W 1U that go into a 19" rack?
4548 2011-05-05 22:48:58 <ArtForz> yep
4549 2011-05-05 22:49:07 <cosurgi> what's an adapter plate?
4550 2011-05-05 22:49:37 <cosurgi> in server room, there are 19" racks
4551 2011-05-05 22:50:10 <ArtForz> sheet metal part that has cutouts and holes for screwing in 2 1U psus, and the W/H and hole pattern of a 3U redundant PSU
4552 2011-05-05 22:50:54 <cosurgi> hmm..
4553 2011-05-05 22:51:06 <ArtForz> basically same thing you use to mount a ATX psu in a spot for a 3U redundant, just with different cutouts and holes so it holds 2 1U PSUs
4554 2011-05-05 22:51:34 <cosurgi> and that would be cheaper than 1200W PSU ?
4555 2011-05-05 22:51:42 <ArtForz> not really, but probably lasts longer
4556 2011-05-05 22:51:54 <cosurgi> I see.
4557 2011-05-05 22:52:59 <BlueMatt> why did we get the ability to vote on reputation on the forums and now its gone?
4558 2011-05-05 22:53:32 <BlueMatt> sirius: whats going on?
4559 2011-05-05 22:53:35 <BlueMatt> ;;seen sirius
4560 2011-05-05 22:53:36 <gribble> sirius was last seen in #bitcoin-dev 21 hours, 42 minutes, and 0 seconds ago: <sirius> luke-jr: no
4561 2011-05-05 22:53:37 DougJ has joined
4562 2011-05-05 22:53:51 <cosurgi> why chieftec 1200W is only 80% price of corsair 1200W ? is it so much worse?
4563 2011-05-05 22:54:56 <Diablo-D3> yes.
4564 2011-05-05 22:54:58 <Diablo-D3> its shit.
4565 2011-05-05 22:55:04 <cosurgi> that makes sense :)
4566 2011-05-05 22:55:09 bk128 has quit (Quit: bk128)
4567 2011-05-05 22:55:35 <diki> mine hasnt died yet
4568 2011-05-05 22:55:43 <diki> i think it's fine
4569 2011-05-05 22:57:29 <BitMark> ArtForz: what kind of failure rate are you getting on PSUs?
4570 2011-05-05 22:57:51 <ArtForz> so far pretty low other than the Legion X2 failures
4571 2011-05-05 22:57:53 x6763 has joined
4572 2011-05-05 22:58:15 <ArtForz> all my SuperFlower 80+ golds are still ok
4573 2011-05-05 23:01:17 redwyrm has quit (Quit: leaving)
4574 2011-05-05 23:02:29 ThomasV has left ("Quitte")
4575 2011-05-05 23:05:48 <BitMark> nice
4576 2011-05-05 23:05:48 skyewm has quit (Ping timeout: 276 seconds)
4577 2011-05-05 23:06:33 <gjs278> 5850, 5870, 5970, what's the min power supply for that
4578 2011-05-05 23:07:05 <cosurgi> (151+188+294)*1.8
4579 2011-05-05 23:07:13 <gjs278> alright
4580 2011-05-05 23:07:16 <cosurgi> 1000W
4581 2011-05-05 23:07:25 <diki> i dont know whats going on with my 5850
4582 2011-05-05 23:07:32 <gjs278> thats' more than 1000w though
4583 2011-05-05 23:07:36 <diki> but sometimes suddenly the fan goes crazy
4584 2011-05-05 23:07:40 <gjs278> what's the 1.8 for
4585 2011-05-05 23:07:45 <diki> up and down with fan speeds and my temp goes to 91
4586 2011-05-05 23:07:47 <cosurgi> safety factor
4587 2011-05-05 23:07:50 <diki> and the card throttles itself
4588 2011-05-05 23:08:00 <gjs278> do you think I could pull it off on an 850
4589 2011-05-05 23:08:09 <cosurgi> not for long
4590 2011-05-05 23:08:13 <gjs278> good enough
4591 2011-05-05 23:08:18 <cosurgi> diki: is it gigabyte?
4592 2011-05-05 23:08:29 <diki> no
4593 2011-05-05 23:08:30 <diki> sapphire
4594 2011-05-05 23:08:36 <cosurgi> even worse...
4595 2011-05-05 23:08:36 <gjs278> set the fan speed to 100 and never look back
4596 2011-05-05 23:08:41 <diki> i do
4597 2011-05-05 23:08:53 <diki> then suddenly it starts going up and down wit the speeds
4598 2011-05-05 23:09:02 <diki> allowing the card to reach 90+c temps
4599 2011-05-05 23:09:05 <gjs278> something is hijacking it then
4600 2011-05-05 23:09:07 <diki> then starts to throttle
4601 2011-05-05 23:09:12 <cosurgi> I just RMAed gigabyte OC, one of its two fans died.
4602 2011-05-05 23:09:16 <diki> and goes up and down with the voltage too
4603 2011-05-05 23:09:23 <diki> will destroy it like that
4604 2011-05-05 23:09:27 <cosurgi> stopped rotating. I couldn't even rotate it with my fingers
4605 2011-05-05 23:09:54 <cosurgi> a week later I RMAed second 5850 gigabyte, same reason: one of two fans died.
4606 2011-05-05 23:10:00 <diki> that is why i need to make an event program
4607 2011-05-05 23:10:11 <diki> to shut down the miner if temp >= 91
4608 2011-05-05 23:10:19 <gjs278> for multiple power supplies
4609 2011-05-05 23:10:21 <gjs278> all I need to do
4610 2011-05-05 23:10:26 <gjs278> is plug the power supply into the wall
4611 2011-05-05 23:10:37 <gjs278> and plug two pci-e cords off of it into the card right
4612 2011-05-05 23:10:37 <cosurgi> it won't turn on.
4613 2011-05-05 23:10:43 <gjs278> gay
4614 2011-05-05 23:10:52 <diki> lesbian
4615 2011-05-05 23:10:54 <cosurgi> it needs signal from mobo to turn on.
4616 2011-05-05 23:11:25 <cosurgi> or you must use paperclip on holes IIRC 13 and 14 ?? sth like that..
4617 2011-05-05 23:11:35 <cosurgi> to make a shortcut
4618 2011-05-05 23:11:35 <gjs278> ugh
4619 2011-05-05 23:12:06 Jaagu has quit (Quit: Leaving.)
4620 2011-05-05 23:12:42 <AAA_awright> How does Bitcoin decide between a block collision?
4621 2011-05-05 23:13:00 <ArtForz> if chains are the same length, first one wins
4622 2011-05-05 23:13:01 <grbgout> whichever was first
4623 2011-05-05 23:13:20 <cosurgi> AAA_awright: block chains with higher haspower used to make them, win.
4624 2011-05-05 23:13:21 <grbgout> ArtForz: are you in -mining?
4625 2011-05-05 23:13:35 <ArtForz> yes
4626 2011-05-05 23:13:37 FellowTraveler has quit (Ping timeout: 252 seconds)
4627 2011-05-05 23:13:41 <xelister> AAA_awright: most sum of work wins
4628 2011-05-05 23:13:47 <ArtForz> yes
4629 2011-05-05 23:13:53 <grbgout> ArtForz: would you take a look at the last thing I mentioned about my 5850
4630 2011-05-05 23:14:01 <grbgout> please?
4631 2011-05-05 23:14:04 <ArtForz> I use length to refer to sum-of-difficulty, like the paper does
4632 2011-05-05 23:14:30 NickelBot has quit (Remote host closed the connection)
4633 2011-05-05 23:16:02 <AAA_awright> xelister: Sum of what?
4634 2011-05-05 23:16:25 <xelister> ArtForz: btw can you please tell is it possible to set r5970 to have mem clock 300 MHz always? how? RBE or just AMDOver..?
4635 2011-05-05 23:16:33 <ArtForz> yes
4636 2011-05-05 23:16:38 <ArtForz> both methods work
4637 2011-05-05 23:16:46 <BitMark> whats RBE?
4638 2011-05-05 23:16:53 NickelBot has joined
4639 2011-05-05 23:16:54 <ArtForz> bios editor
4640 2011-05-05 23:17:26 <gjs278> AMD over drive is easier
4641 2011-05-05 23:17:33 <gjs278> you just set a default and ctrl+c it
4642 2011-05-05 23:17:37 <gjs278> and then the 300 sticks
4643 2011-05-05 23:17:39 <xelister> ArtForz: with just AMDOver, the GUI shows me 500 not 300 as lowest possible setting it seems
4644 2011-05-05 23:17:46 devon_hillard has quit (Read error: Connection reset by peer)
4645 2011-05-05 23:17:56 peddie has joined
4646 2011-05-05 23:18:07 <gjs278> even on the lowest setting it lists 500?
4647 2011-05-05 23:18:50 forrestv has joined
4648 2011-05-05 23:18:50 <xelister> ctrl-c it?
4649 2011-05-05 23:18:57 tabsa has quit (Ping timeout: 240 seconds)
4650 2011-05-05 23:19:02 <xelister> in LOW the <----|----> slider is 500 to 1500
4651 2011-05-05 23:19:11 <diki> ;;rate BitMark 1
4652 2011-05-05 23:19:11 <gribble> Error: For identification purposes, you must be identified via GPG to use the rating system.
4653 2011-05-05 23:19:11 <xelister> on memory mhz
4654 2011-05-05 23:19:16 <diki> ;;auth diki
4655 2011-05-05 23:19:16 <gribble> Request successful for user diki. Your challenge string is: 50f95d1f9f7439a322a2dce009626af22d28d9966bd8e360dc49789c7b3ac094
4656 2011-05-05 23:19:27 <diki> ;;rate BitMark 1
4657 2011-05-05 23:19:27 <gribble> Error: For identification purposes, you must be identified via GPG to use the rating system.
4658 2011-05-05 23:19:31 <gjs278> haha
4659 2011-05-05 23:19:36 <diki> lol
4660 2011-05-05 23:19:40 <gjs278> just pm the bot
4661 2011-05-05 23:19:42 <diki> ;;gpg auth
4662 2011-05-05 23:19:43 <gribble> (gpg auth <nick>) -- Initiate authentication for user <nick>. You must have registered a GPG key with the bot for this to work. You will be given a random passphrase to clearsign with your key, and submit to the bot with the 'verify' command. Your passphrase will expire within 5 minutes.
4663 2011-05-05 23:19:43 <gjs278> and you can mess around there
4664 2011-05-05 23:19:47 <diki> ;;gpg auth diki
4665 2011-05-05 23:19:47 <gribble> Request successful for user diki. Your challenge string is: 375c840541f867ad7cf5594effe5bec7262d47dfd442aa92287099cdac5aa8ee
4666 2011-05-05 23:19:58 <diki> ;;rate BitMark 1
4667 2011-05-05 23:19:59 <gribble> Error: For identification purposes, you must be identified via GPG to use the rating system.
4668 2011-05-05 23:20:02 <diki> darn
4669 2011-05-05 23:20:10 <BitMark> need to sign the challenge then verify
4670 2011-05-05 23:20:16 <BitMark> a pain i know
4671 2011-05-05 23:20:19 * xelister edits the .xml od AMDover liek a boss
4672 2011-05-05 23:20:22 <diki> zomg
4673 2011-05-05 23:20:38 <diki> gotta open *that* program
4674 2011-05-05 23:20:43 <BitMark> yup
4675 2011-05-05 23:21:14 <xelister> ArtForz: AMDOver says it can not apply the settings if I edit the profile's XML to set 300000 instead 500000 on memory on 0,1,2 profile
4676 2011-05-05 23:21:47 <ArtForz> huh, weird
4677 2011-05-05 23:22:28 dleary has left ()
4678 2011-05-05 23:23:33 <gjs278> if you can't set 300 in amdover
4679 2011-05-05 23:23:38 <gjs278> I wouldn't try and set 300 in rbe
4680 2011-05-05 23:23:45 <ArtForz> why not?
4681 2011-05-05 23:23:49 threepio has quit (Read error: Connection reset by peer)
4682 2011-05-05 23:24:01 <ArtForz> 300 is stadnard idle@desktop clock, it *has* to work
4683 2011-05-05 23:24:04 <gjs278> because whenever I set lower than my min and I go to adjust the coreclock
4684 2011-05-05 23:24:12 <gjs278> it says I'm out of range
4685 2011-05-05 23:24:14 <diki> ;;gpg verify http://paste.pocoo.org/show/383742/
4686 2011-05-05 23:24:15 <gribble> Error: Challenge string not present in signed message.
4687 2011-05-05 23:24:19 <diki> what?
4688 2011-05-05 23:24:22 <gjs278> I tried it for my 5870 and I always tried to force 900 back on me
4689 2011-05-05 23:24:28 <gjs278> it*
4690 2011-05-05 23:24:41 <ArtForz> 5870 iirc hash higher min clock than 5970
4691 2011-05-05 23:24:50 <xelister> WOW guys
4692 2011-05-05 23:24:56 <diki> ;;gpg http://paste.pocoo.org/show/383744/
4693 2011-05-05 23:24:56 <gribble> Error: The "GPG" plugin is loaded, but there is no command named "http://paste.pocoo.org/show/383744/" in it. Try "list GPG" to see the commands in the "GPG" plugin.
4694 2011-05-05 23:24:56 <xelister> 1.5 USD/BTC
4695 2011-05-05 23:25:03 <ArtForz> rmeember, 5870 has diff. mem chips
4696 2011-05-05 23:25:04 <xelister> now 1.3 USD/BTC !!! what is going on
4697 2011-05-05 23:25:07 <diki> ;;gpg verify http://paste.pocoo.org/show/383744/
4698 2011-05-05 23:25:08 <gribble> You are now authenticated for user 'diki' with key 4C5FCC9FF6533A65
4699 2011-05-05 23:25:11 <gjs278> ;;bc,mtgox
4700 2011-05-05 23:25:12 <gribble> {"ticker":{"high":3.4498,"low":3.333,"vol":15696,"buy":3.3551,"sell":3.36,"last":3.355}}
4701 2011-05-05 23:25:16 <diki> ;; auth diki
4702 2011-05-05 23:25:17 <gribble> Request successful for user diki. Your challenge string is: cf86cbc1b78463d45552931467cfc27a07ab9be24702b3ef157d59040ec743b9
4703 2011-05-05 23:25:18 <gjs278> what are you talking about
4704 2011-05-05 23:25:25 <diki> ;;rate BitMark 1
4705 2011-05-05 23:25:25 <gribble> Rating entry successful. Your rating of 1 for user BitMark has been recorded.
4706 2011-05-05 23:25:30 <diki> phew
4707 2011-05-05 23:25:32 <diki> finished
4708 2011-05-05 23:25:42 <gjs278> xelister what is this 1.5usd/btc nonsense
4709 2011-05-05 23:25:52 <luke-jr> gjs278: market rate
4710 2011-05-05 23:25:52 <xelister> market crashed
4711 2011-05-05 23:25:55 <BitMark> diki: yay :)
4712 2011-05-05 23:25:58 <xelister> the bottom is falling out
4713 2011-05-05 23:26:01 * xelister jumps the ship
4714 2011-05-05 23:26:01 <luke-jr> gjs278: it's down to 0.01 now
4715 2011-05-05 23:26:02 <gjs278> I see 3.35 on mtgox
4716 2011-05-05 23:26:04 <andrew12> i'm not very good at explaining bitcoin to people
4717 2011-05-05 23:26:05 <andrew12> http://pastebin.com/4pPz4XiP
4718 2011-05-05 23:26:06 <gjs278> luke-jr is right
4719 2011-05-05 23:26:08 <luke-jr> gjs278: wrong market
4720 2011-05-05 23:26:10 <gjs278> oh
4721 2011-05-05 23:26:11 <gjs278> ok
4722 2011-05-05 23:26:12 <xelister> hachiya: mtgox was ddosed and is not updating
4723 2011-05-05 23:26:13 jackmcbarn has quit (Ping timeout: 264 seconds)
4724 2011-05-05 23:26:13 <gjs278> which market
4725 2011-05-05 23:26:16 <luke-jr> gjs278: OTC
4726 2011-05-05 23:26:46 <xelister> oh god 0.01 usd/btc price again
4727 2011-05-05 23:26:51 <Diablo-D3> wait what?
4728 2011-05-05 23:26:57 <xelister> how will I repay all the radeon cards
4729 2011-05-05 23:27:25 <ArtForz> löl
4730 2011-05-05 23:27:38 kelp has quit (Quit: Bye!)
4731 2011-05-05 23:27:42 jackmcbarn has joined
4732 2011-05-05 23:28:10 * xelister jumps the ship and waits to be rescued by HMS trololololol
4733 2011-05-05 23:28:11 <andrew12> xelister: you have to pay the radeons?
4734 2011-05-05 23:28:13 <diki> i also ordered my cooler thanks to bitmark
4735 2011-05-05 23:29:51 <gjs278> lies
4736 2011-05-05 23:30:05 tabsa has joined
4737 2011-05-05 23:30:26 <diki> bitcoincharts says otherwise
4738 2011-05-05 23:31:47 bk128 has joined
4739 2011-05-05 23:33:51 riush has left ("Leaving")
4740 2011-05-05 23:34:01 x5x`brb is now known as x5x
4741 2011-05-05 23:34:35 bk128_ has joined
4742 2011-05-05 23:34:37 threepio has joined
4743 2011-05-05 23:34:57 <CIA-30> bitcoin: droidbird exper * rb5f24e786477 intersango/util.php: fix check to see if user has an rsa key linked. http://tinyurl.com/3qg7ozs
4744 2011-05-05 23:35:19 <diki> xelister, there was this post saying that the iphone 4 white version is thickers, cause Steve job wanted to troll. And someone mentioned "He's probably sitting in his office saying "Trololol" :D
4745 2011-05-05 23:35:19 bk128 has quit (Read error: Connection reset by peer)
4746 2011-05-05 23:35:19 bk128_ is now known as bk128
4747 2011-05-05 23:35:28 <diki> *thicker
4748 2011-05-05 23:35:40 <phantomcircuit> diki, it isn't even thicker
4749 2011-05-05 23:35:43 <phantomcircuit> people are so stupid
4750 2011-05-05 23:36:15 <diki> we've seen the proof
4751 2011-05-05 23:36:16 <diki> it is
4752 2011-05-05 23:37:13 sethsethseth has joined
4753 2011-05-05 23:37:40 <sethsethseth> ;;bc,stats
4754 2011-05-05 23:37:42 <gribble> Current Blocks: 122108 | Current Difficulty: 109670.13329248 | Next Difficulty At Block: 122975 | Next Difficulty In: 867 blocks | Next Difficulty In About: 4 days, 9 hours, 14 minutes, and 39 seconds | Next Difficulty Estimate: 149580.96055506
4755 2011-05-05 23:38:00 <BitMark> this should show the current block rate
4756 2011-05-05 23:38:17 <BitMark> and possibly the percentage above 6/s
4757 2011-05-05 23:38:29 <xelister> diki: =)
4758 2011-05-05 23:38:43 <BitMark> i saw it at 11/s yesterday
4759 2011-05-05 23:38:45 <luke-jr> ;;bc,calc [bc,eligius]
4760 2011-05-05 23:38:46 <bk128> ArtForz: do you have a schematic from this circuit? http://i.imgur.com/75RyO.png still trying to figure out what's going on.
4761 2011-05-05 23:38:47 <gribble> The average time to generate a block at 14789004.0559 Khps, given current difficulty of 109670.13329248 , is 8 hours, 50 minutes, and 49 seconds
4762 2011-05-05 23:39:12 <diki> i was find the word trololol to be extremely funny
4763 2011-05-05 23:39:19 <diki> *i always
4764 2011-05-05 23:39:26 <xelister> diki: indeed =)
4765 2011-05-05 23:39:37 d4de has quit (Ping timeout: 276 seconds)
4766 2011-05-05 23:39:57 <BitMark> what circuit is that?
4767 2011-05-05 23:41:02 gjs278 has quit (Read error: Connection reset by peer)
4768 2011-05-05 23:41:12 <bk128> fan controller. I think it monitors rpm from all 3 fans and uses a buck converter to control the fan speeds
4769 2011-05-05 23:41:15 tabsa has quit (Ping timeout: 259 seconds)
4770 2011-05-05 23:41:16 peddie has left ("peace!")
4771 2011-05-05 23:41:24 <BitMark> ah
4772 2011-05-05 23:43:42 diki has quit ()
4773 2011-05-05 23:44:33 skyewm has joined
4774 2011-05-05 23:44:51 gjs278 has joined
4775 2011-05-05 23:45:01 <CIA-30> bitcoin: genjix exper * r949cbaf64300 intersango/cron/process_orders.php: BUGFIX: hack for extra field alias that OrderInfo expects. http://tinyurl.com/3qhnhco
4776 2011-05-05 23:45:03 <CIA-30> bitcoin: genjix exper * r30b5aa1ff230 intersango/help.php: you *can* cancel partially completed orders. http://tinyurl.com/453mgdr
4777 2011-05-05 23:45:06 <CIA-30> bitcoin: genjix exper * rb188f40b2683 intersango/help.php: Merge branch 'master' into exper http://tinyurl.com/3zp9s8v
4778 2011-05-05 23:46:04 <xelister> ____ ____ _ ____ _____ _
4779 2011-05-05 23:46:05 <xelister> / ___|| _ \ / \ | _ \_ _|/ \
4780 2011-05-05 23:46:07 <xelister> \___ \| |_) / _ \ | |_) || | / _ \
4781 2011-05-05 23:46:08 <xelister> ___) | __/ ___ \| _ < | |/ ___ \
4782 2011-05-05 23:46:10 <xelister> |____/|_| /_/ \_\_| \_\|_/_/ \_\
4783 2011-05-05 23:46:10 vorlov has joined
4784 2011-05-05 23:46:22 <BlueMatt> ...?
4785 2011-05-05 23:46:28 <xelister> wait. wrong channel.
4786 2011-05-05 23:46:59 <gjs278> my power just went out
4787 2011-05-05 23:47:00 <gjs278> 277 days
4788 2011-05-05 23:47:04 <gjs278> uptime on one machine
4789 2011-05-05 23:47:05 <gjs278> and it's all over
4790 2011-05-05 23:47:15 <gjs278> and my fucking .bash_history file went with it
4791 2011-05-05 23:47:30 <BlueMatt> ups ftw?
4792 2011-05-05 23:47:47 <gjs278> 13gb encrypted backup ftw
4793 2011-05-05 23:47:54 UukGoblin has left ()
4794 2011-05-05 23:47:54 UukGoblin has joined
4795 2011-05-05 23:47:55 <gjs278> except its taking forever because gpg isnt multithreaded
4796 2011-05-05 23:48:07 <gjs278> so it can do like 30mb/s
4797 2011-05-05 23:48:07 <BlueMatt> hey UukGoblin got ops
4798 2011-05-05 23:48:21 <ArtForz> bk128: sure, sec
4799 2011-05-05 23:48:27 <UukGoblin> nanotube, thanks for a@.
4800 2011-05-05 23:48:44 <bk128> ArtForz: cool, do you have the attiny code too?
4801 2011-05-05 23:49:06 <ArtForz> nope
4802 2011-05-05 23:49:23 <ArtForz> well, I have it on one of my archive HDs, but I'm not going to start looking for it
4803 2011-05-05 23:49:34 <UukGoblin> nanotube, are you sure that 0 output txs will never be pruned by the stock client?
4804 2011-05-05 23:50:30 * BlueMatt finally found the tiny problems with jgarzik's crypter (took forever to find a typo of 3 characters
4805 2011-05-05 23:50:36 <BlueMatt> I feel like I just wasted an hour
4806 2011-05-05 23:50:54 <phantomcircuit> BlueMatt, welcome to everybody elses life
4807 2011-05-05 23:50:55 <xelister> BlueMatt: hm? where
4808 2011-05-05 23:51:09 <UukGoblin> oh, did ArtForz release his asic code? :-O
4809 2011-05-05 23:51:27 <ArtForz> did hell freeze over yet?
4810 2011-05-05 23:51:35 <UukGoblin> ah, no
4811 2011-05-05 23:51:39 <gjs278> I hacked it
4812 2011-05-05 23:51:48 <BlueMatt> xelister: what?
4813 2011-05-05 23:51:56 <ArtForz> http://i.imgur.com/ik5EZ.png
4814 2011-05-05 23:52:30 <phantomcircuit> gjs278, gpg for backups?
4815 2011-05-05 23:52:35 <bk128> ArtForz: surprised you're still using max232s :)
4816 2011-05-05 23:52:59 <ArtForz> I still got a few sticks of em, might as well use em
4817 2011-05-05 23:53:11 <bk128> hah yeah
4818 2011-05-05 23:53:18 <xelister> yo ArtForz
4819 2011-05-05 23:53:23 <xelister> your openCL kernel for mining
4820 2011-05-05 23:53:29 <xelister> how many hashes it calcualtes?
4821 2011-05-05 23:53:36 <ArtForz> ?
4822 2011-05-05 23:53:38 <xelister> is it doing 1024 (the default LOOP defined in .cl) ?
4823 2011-05-05 23:53:41 <ArtForz> yep
4824 2011-05-05 23:53:47 <xelister> per one run of .cl program, 1024 hashes?
4825 2011-05-05 23:53:53 <ArtForz> yep
4826 2011-05-05 23:53:56 <ArtForz> 1024 nonces per kernel run
4827 2011-05-05 23:54:09 <xelister> ok, and there is 1 kernel run per one run of inner loop of diablos miner?
4828 2011-05-05 23:54:10 <ArtForz> not recommended for interactive usage ;)
4829 2011-05-05 23:54:20 <ArtForz> no, a few 10-100k
4830 2011-05-05 23:54:32 pirrr has quit (Read error: Operation timed out)
4831 2011-05-05 23:54:34 <xelister> err = CL10.clEnqueueNDRangeKernel(queue, kernel, 1, null, workSizeTemp, localWorkSize, null, null);
4832 2011-05-05 23:54:35 <xelister> CL10.clEnqueueReadBuffer(queue, output[bufferIndex], CL10.CL_TRUE, 0, buffer[bufferIndex], null, null);
4833 2011-05-05 23:54:45 <xelister> this will execute how much kernel runs and how to controll that?
4834 2011-05-05 23:54:53 <CIA-30> bitcoin: genjix exper * rf0bd1c614f89 intersango/www/style.css: added textarea styling. http://tinyurl.com/4yzt8xu
4835 2011-05-05 23:54:54 <ArtForz> workSizeTemp
4836 2011-05-05 23:55:02 <UukGoblin> i think newer diablo has an option to set number of kernel loops
4837 2011-05-05 23:55:08 <ArtForz> and thats indirectly controlled by -f
4838 2011-05-05 23:55:14 <xelister> UukGoblin: it has, -z
4839 2011-05-05 23:55:18 <xelister> UukGoblin: and this option sucks ass
4840 2011-05-05 23:55:31 <iera> ArtForz: what does this schematic do? i only know the 232 is used for serial conversion
4841 2011-05-05 23:55:31 <ArtForz> which is useful, as long.running kernels have the tendency to kill interactive perf
4842 2011-05-05 23:55:34 <xelister> -z 1 .. 5 makes the miner LOCK THE FUCK UP for up to 5 minutes (X freezes)
4843 2011-05-05 23:55:35 <UukGoblin> ArtForz, i'm still impressed @ you running 4x 5970
4844 2011-05-05 23:55:41 <xelister> higher z make it act as default
4845 2011-05-05 23:55:51 <ArtForz> iera: it's a rsr232 controlled/readout PWM fan controller
4846 2011-05-05 23:55:54 <ArtForz> *rs232
4847 2011-05-05 23:55:57 <iera> ah ok
4848 2011-05-05 23:56:10 <ArtForz> for fans up to 40W or so
4849 2011-05-05 23:56:24 <BitMark> why not just set fans on max rpm?
4850 2011-05-05 23:56:38 <ArtForz> fan lifetime, noise
4851 2011-05-05 23:57:00 <UukGoblin> ArtForz, tried 4x 5970 with different boxes, psus, risers - All fail compilation at #6 or #7th chip
4852 2011-05-05 23:57:01 <BitMark> are you still mining in your house?
4853 2011-05-05 23:57:07 <ArtForz> yep
4854 2011-05-05 23:57:10 <ArtForz> most fans only have a nominal lifetime of 40-60kh at 100%
4855 2011-05-05 23:57:25 <ArtForz> at 50% they're usually north of 200khr
4856 2011-05-05 23:57:25 <xelister> ArtForz: can you turn off your rig for 12 hours per day for a week?
4857 2011-05-05 23:57:31 <xelister> I will pay you x2 what you would had mined >_>
4858 2011-05-05 23:57:55 <ArtForz> that sounds kinda stupid...
4859 2011-05-05 23:58:11 <xelister> so, want?
4860 2011-05-05 23:58:14 <xelister> >_>
4861 2011-05-05 23:58:34 <BitMark> ArtForz: may I lease your rig while your not using it? :)
4862 2011-05-05 23:58:42 <lulzplzkthx> LOL ^
4863 2011-05-05 23:58:53 <ArtForz> lol
4864 2011-05-05 23:58:57 * xelister calls dea
4865 2011-05-05 23:59:11 <xelister> what is de version of deafags?
4866 2011-05-05 23:59:13 <lulzplzkthx> > turn off your rig
4867 2011-05-05 23:59:19 <UukGoblin> xelister, too much breaking bad