1 2013-06-28 00:03:05 btcquant has quit (Quit: Textual IRC Client: www.textualapp.com)
2 2013-06-28 00:05:39 <michagogo> Anyone available with a confirmed testnet transaction to quickly help me test something?
3 2013-06-28 00:06:16 <michagogo> If so, could you please make and sign a raw transaction that spends a well-confirmed transaction, and pastebin it for me?
4 2013-06-28 00:09:27 viperhr has quit (Ping timeout: 256 seconds)
5 2013-06-28 00:12:18 KillYourTV has quit (Remote host closed the connection)
6 2013-06-28 00:13:18 execut3 has left ()
7 2013-06-28 00:13:53 LiCeUser6 has left ()
8 2013-06-28 00:14:05 KillYourTV has joined
9 2013-06-28 00:14:10 EvilPete has joined
10 2013-06-28 00:15:05 <realazthat> sipa: hi
11 2013-06-28 00:15:09 <realazthat> sipa: whats up
12 2013-06-28 00:15:39 anarchy5 has joined
13 2013-06-28 00:16:36 jaekwon has joined
14 2013-06-28 00:20:31 <nsh> realazthat, it's a trap!
15 2013-06-28 00:21:31 <realazthat> mmm?
16 2013-06-28 00:21:54 <nsh> sipa, sipa is trying to trap you into a productive and interesting discussion
17 2013-06-28 00:21:59 <realazthat> haha
18 2013-06-28 00:22:06 <realazthat> sipa does that
19 2013-06-28 00:22:07 <nsh> run for the border; it's not too late!
20 2013-06-28 00:22:11 * nsh nods
21 2013-06-28 00:23:13 <sipa> realazthat: seen https://github.com/bitcoin/bitcoin/pull/2802 ?
22 2013-06-28 00:24:08 <nsh> oh, that's a good thing!
23 2013-06-28 00:24:28 <realazthat> haha
24 2013-06-28 00:24:30 <realazthat> awesome
25 2013-06-28 00:24:50 * sipa thinks realazthat gets the subtle hint
26 2013-06-28 00:25:16 <nsh> sipa, "A full -txindex=1 -addrindex=1 index is about 2.7 GiB now." in terms of?
27 2013-06-28 00:25:24 <sipa> bytes?
28 2013-06-28 00:25:24 <nsh> diskspace?
29 2013-06-28 00:25:27 <realazthat> yes
30 2013-06-28 00:25:29 <sipa> yes
31 2013-06-28 00:25:31 <nsh> k
32 2013-06-28 00:25:55 <realazthat> sipa: the alternative is to do this outside the chain I guess
33 2013-06-28 00:25:59 <realazthat> er
34 2013-06-28 00:26:01 <realazthat> outside the program
35 2013-06-28 00:26:07 <sipa> yes
36 2013-06-28 00:26:14 <realazthat> ie. go through all transactions once and store it in a db
37 2013-06-28 00:26:27 <realazthat> but thats crazy to do over a network if its not on the same machine
38 2013-06-28 00:28:19 <realazthat> sipa: how long does it take for these type of things to make it in
39 2013-06-28 00:28:24 <realazthat> just curious
40 2013-06-28 00:29:29 <jaekwon> i don't like c++. what's the best alternative implementation of a bitcoin server for developing altcoins?
41 2013-06-28 00:30:19 <sipa> haha
42 2013-06-28 00:30:28 <nsh> jaekwon, if you don't like (read/understand) C++ well, i wouldn't recommend trying to make an altcoin
43 2013-06-28 00:30:51 <sipa> realazthat: depends how much the others like it...
44 2013-06-28 00:31:14 <jaekwon> haha. sorry, you can't stop me.
45 2013-06-28 00:32:02 <michagogo> I'm kinda confused as to whether I'm currently mining against a difficulty of 1 or a difficulty of 240
46 2013-06-28 00:32:35 <michagogo> bfgminer says 240, `./bitcoind getinfo` says 1
47 2013-06-28 00:32:43 <sipa> then trust bfgminer
48 2013-06-28 00:33:00 <michagogo> And the gbt target is 0000000001108300000000000000000000000000000000000000000000000000
49 2013-06-28 00:33:03 Skav has joined
50 2013-06-28 00:33:30 <realazthat> sipa: where do I vote
51 2013-06-28 00:33:32 <realazthat> :P
52 2013-06-28 00:33:43 <jaekwon> nsh: do you have a valid argument? I just happen to prefer anything to C++.
53 2013-06-28 00:34:38 <nsh> jaekwon, i have nothing valid. it's just that to fully understand how to do *coin it's probably good to be able to read the definitive reference, which is written in a language called C++
54 2013-06-28 00:34:52 patcon has quit (Remote host closed the connection)
55 2013-06-28 00:35:03 MobPhone has quit (Ping timeout: 252 seconds)
56 2013-06-28 00:35:57 <nsh> there are afaik no full implementations in other languages. though bits-of-proof (java) may be getting there
57 2013-06-28 00:36:57 <jaekwon> yeah. i have no need for most of the server code, my coin is pretty different so i should be ok. yeah it's either bits of proof or btcd, btcd not quite there yet.
58 2013-06-28 00:37:49 <jaekwon> basically i need the code i'm forking to be *very* modular. i hear that bitcoind is still spagetti. that won't work for me.
59 2013-06-28 00:38:10 <realazthat> you can try reimplementing it from scratch ...
60 2013-06-28 00:38:21 <realazthat> and annoying everyone in the channel on the itty bitty details\
61 2013-06-28 00:38:23 <realazthat> :P
62 2013-06-28 00:38:40 <realazthat> I've reimplemented block chain parsing in python
63 2013-06-28 00:38:43 <sipa> if he's creating an altcoin, there is no need to *re* implement anything
64 2013-06-28 00:38:47 <realazthat> that was fun
65 2013-06-28 00:38:52 <jaekwon> hehe i've made that mistake on other projects before⦠i've implemented a language starting with a new PEG parser
66 2013-06-28 00:38:59 <Luke-Jr> jaekwon: want to bet we can't stop you?
67 2013-06-28 00:39:01 <realazthat> sipa: well he insists on not looking at C++
68 2013-06-28 00:39:16 <jaekwon> realazthat: esp not messy C++
69 2013-06-28 00:39:25 <gmaxwell> sipa: did you just click [edit] instead of reply on that github issue?
70 2013-06-28 00:39:28 <realazthat> tbh I haven't looked at the bitcoin source myself
71 2013-06-28 00:39:36 <jaekwon> Luke-Jr: hey help me find the best solution, and watch the world change.
72 2013-06-28 00:39:41 <realazthat> so I don't know how messy it is
73 2013-06-28 00:39:44 <nsh> /topic #bitcoin-dev "for all your discussion about how terrible the bitcoin code is" requirements
74 2013-06-28 00:39:51 <sipa> gmaxwell: oh, wow!
75 2013-06-28 00:39:55 <sipa> i must be tired
76 2013-06-28 00:39:56 <realazthat> lol
77 2013-06-28 00:39:56 <nsh> *quote-in-right-place()
78 2013-06-28 00:40:02 <Luke-Jr> jaekwon: if you're having trouble with code, I agree that you shouldn't make an altcoin
79 2013-06-28 00:40:03 <gmaxwell> sipa: can you restore my text? github doesn't email it to me. :)
80 2013-06-28 00:40:09 <Luke-Jr> jaekwon: deciding what to do is harder than the code!
81 2013-06-28 00:40:26 <sipa> gmaxwell: the irony
82 2013-06-28 00:40:28 <sipa> sure
83 2013-06-28 00:40:30 <nsh> lol
84 2013-06-28 00:40:36 <jaekwon> Luke-Jr: I've already decided exactly what needs to be done. :)
85 2013-06-28 00:40:36 <gmaxwell> sipa: I hit edit accidentally every couple of times I use itâ though AFAIK I've not yet actually saved one that way! :P (thats how I knew what you did!)
86 2013-06-28 00:40:42 <Luke-Jr> jaekwon: you think
87 2013-06-28 00:40:53 <jaekwon> i'm not making some shitty clone. this is different.
88 2013-06-28 00:41:00 <gmaxwell> (well, I was _guessing_ it was you because you were active in here!)
89 2013-06-28 00:41:01 <Luke-Jr> jaekwon: and you think it's something Bitcoin can't just adapt out from under your altcoin?
90 2013-06-28 00:41:23 <jaekwon> luke-Jr: of course. bitcoin relies on PoW.
91 2013-06-28 00:41:31 <jaekwon> it will always be.
92 2013-06-28 00:41:37 <sipa> indeed it will
93 2013-06-28 00:41:42 <Luke-Jr> jaekwon: yes, PoW was Satoshi's major innovation that made crypto-currency impossible
94 2013-06-28 00:41:48 <realazthat> jaekwon: just tell me your idea already so I can just do it myself while you reimplement it in another language :P
95 2013-06-28 00:41:50 <sipa> Luke-Jr: ha, impossible :(
96 2013-06-28 00:41:53 <jaekwon> and a major stepping stone at that.
97 2013-06-28 00:41:58 <Luke-Jr> jaekwon: nobody for the decades before him, nor anyone after him, has ever come up with a viable alternative solution
98 2013-06-28 00:42:20 <Luke-Jr> sorry, I meant POSSIBLE above
99 2013-06-28 00:42:25 <nsh> !quotes add <Luke-Jr> jaekwon: yes, PoW was Satoshi's major innovation that made crypto-currency impossible
100 2013-06-28 00:42:26 <gribble> Error: "quotes" is not a valid command.
101 2013-06-28 00:42:27 <jaekwon> oh well, you'll just have to wait until i've implemented it.
102 2013-06-28 00:42:27 <nsh> hehe :)
103 2013-06-28 00:42:33 <realazthat> lol nsh
104 2013-06-28 00:42:37 <sipa> jaekwon: what alternate scheme will you use?
105 2013-06-28 00:42:40 <Luke-Jr> jaekwon: why not discuss it first?
106 2013-06-28 00:42:44 <jaekwon> lol nsh
107 2013-06-28 00:42:49 melvster has quit (Ping timeout: 256 seconds)
108 2013-06-28 00:42:57 <nsh> Luke-Jr and Kaminsky, sitting in a tree, working on a new PoW algorithm-y
109 2013-06-28 00:43:06 <Luke-Jr> nsh: you have me confused with jaekwon
110 2013-06-28 00:43:07 <nsh> (i am available for kids parties)
111 2013-06-28 00:43:11 * nsh smiles
112 2013-06-28 00:43:15 <Luke-Jr> nsh: you're not hired.
113 2013-06-28 00:43:19 * nsh frowns
114 2013-06-28 00:43:49 * Luke-Jr wonders how he typo'
115 2013-06-28 00:43:53 * Luke-Jr wonders how he typo'd possible as impossible
116 2013-06-28 00:44:13 <sipa> the keys are like right next to eachother
117 2013-06-28 00:44:20 <michagogo> sipa: Good news.
118 2013-06-28 00:44:20 <Luke-Jr> >.>
119 2013-06-28 00:44:31 <michagogo> It appears that the patch did in fact work perfectly
120 2013-06-28 00:44:39 MobiusL is now known as LinuxKernel
121 2013-06-28 00:44:53 <jaekwon> cuz then you guys are just going to fork it from the aweful that is bitcoind
122 2013-06-28 00:45:01 LinuxKernel is now known as MobiusL
123 2013-06-28 00:45:10 <jaekwon> :)
124 2013-06-28 00:45:18 MobiusL is now known as ReadMe
125 2013-06-28 00:45:22 <realazthat> whats an aweful
126 2013-06-28 00:45:29 ReadMe is now known as MobiusL
127 2013-06-28 00:45:43 <michagogo> And as an added bonus, the fact that I ended up just building for linux and running in the VM lets me use a host-only network to connect to my bitcoin-qt testnet node
128 2013-06-28 00:45:53 <jaekwon> an awful awful.
129 2013-06-28 00:46:01 <michagogo> Which also means I can easily wireshark the connection
130 2013-06-28 00:46:30 macboz has joined
131 2013-06-28 00:47:01 * Luke-Jr has no trouble wiresharking localhost
132 2013-06-28 00:48:27 and has joined
133 2013-06-28 00:48:36 Skav has quit (Read error: Connection reset by peer)
134 2013-06-28 00:48:42 <michagogo> Luke-Jr: It's easier when you can just capture the virtualbox interface
135 2013-06-28 00:49:33 <Luke-Jr> slightly.
136 2013-06-28 00:49:36 MobPhone has joined
137 2013-06-28 00:49:43 <Diablo-D3> http://storestteampowered.com/app/4411/
138 2013-06-28 00:49:47 <michagogo> Luke-Jr: BTW, a question about bfgminer... If bitcoind on testnet says difficulty is 1 and bfgminer says it's 240, which one is correct?
139 2013-06-28 00:50:02 <jaekwon> i'd discuss it with people in person so i can get a sense of their character. if you're in the bay area, contact me.
140 2013-06-28 00:50:04 <Luke-Jr> michagogo: both
141 2013-06-28 00:50:05 <jaekwon> cya l8rz.
142 2013-06-28 00:50:09 <michagogo> Luke-Jr: Hmm?
143 2013-06-28 00:50:21 <Luke-Jr> michagogo: bitcoind probably dropped it to 1 after BFG started on the work
144 2013-06-28 00:50:39 <SomeoneWeird> because testnet
145 2013-06-28 00:50:43 <gmaxwell> sipa: I'd told him above that I thought it was that it changed when the signature was pushed, and he was telling me that wasn't what happened. I hate this stupid discussion, it's dumb dumb dumb. But it's also a bit frustrating to have people arguing against perfectly reasonable behavior with apparent misinformation. (and if it really is buggy, I'd like to get github to fix it)
146 2013-06-28 00:50:55 <michagogo> Luke-Jr: Also, `./bitcoind getblocktemplate` shows a target of 0000000001108300000000000000000000000000000000000000000000000000
147 2013-06-28 00:51:20 <michagogo> Luke-Jr: It's been like that for quite a while -- doesn't bfgminer get new work every 2 mins?
148 2013-06-28 00:51:27 <gmaxwell> michagogo: if 'bitcoind' says 1 means getinfo says 1, IIRC getinfo always shows the last block or something like that.
149 2013-06-28 00:51:44 <Luke-Jr> michagogo: I'm not sure how it works without longpolling <.<
150 2013-06-28 00:51:54 <Luke-Jr> michagogo: I think it'll just hash it forever
151 2013-06-28 00:51:56 <michagogo> gmaxwell: If it's any different, getdifficulty also says 1
152 2013-06-28 00:52:05 <michagogo> Luke-Jr: Ah, so kill and restart?
153 2013-06-28 00:52:07 <Luke-Jr> unless you set scantime
154 2013-06-28 00:52:14 <Luke-Jr> or something like that; see README âº
155 2013-06-28 00:52:32 <Luke-Jr> oh, what gmaxwell said too
156 2013-06-28 00:52:34 MobPhone has quit (Client Quit)
157 2013-06-28 00:52:46 <sipa> getinfo indeed returns the last block
158 2013-06-28 00:52:50 <michagogo> "--scan-time|-s <arg> Upper bound on time spent scanning current work, in seconds (default: 60)"
159 2013-06-28 00:52:53 <sipa> not the 'current' new difficulty
160 2013-06-28 00:53:40 <michagogo> After a quick Google search, I'm guessing 0000000001108300000000000000000000000000000000000000000000000000 corresponds to a difficulty of 240
161 2013-06-28 00:53:54 <sipa> correct
162 2013-06-28 00:54:02 <michagogo> What I'm wondering is, how come it's been saying difficulty is 1 for several blocks
163 2013-06-28 00:54:13 <michagogo> while gbt is giving a difficulty of 240 consistently
164 2013-06-28 00:54:14 <sipa> difficulty 1 is 00000000FFFF00000000000000000000...
165 2013-06-28 00:54:49 <gmaxwell> michagogo: because diff 1 blocks are being solved once permitted, so the last block is 1 thus getinfo says 1.
166 2013-06-28 00:55:10 <michagogo> Ohhhhhh
167 2013-06-28 00:55:11 <gmaxwell> presumably ninja reflexes would allow you to see the lower target at some point.
168 2013-06-28 00:55:13 <michagogo> I see
169 2013-06-28 00:55:23 <michagogo> So every 20 minutes exactly someone solves a difficulty 1 block
170 2013-06-28 00:55:30 <michagogo> And then it goes back up to what it was?
171 2013-06-28 00:55:55 <gmaxwell> I'm not doing it right now but in the past I used to turn on hashpower on testnet only when the difficulty went to 1... just long enough to get one block. Someone else might be doing that now.
172 2013-06-28 00:55:56 <michagogo> I guess I'd assumed that it would then build back up from 1
173 2013-06-28 00:56:02 <sipa> in 99.95039% of the cases it goes back up :p
174 2013-06-28 00:56:25 <michagogo> Yeah, but I thought that would happen in 25% increments
175 2013-06-28 00:56:53 <gmaxwell> 2015/2016 would have involved fewer characters. :P
176 2013-06-28 00:57:18 <sipa> No, exactly the same.
177 2013-06-28 00:57:42 <gmaxwell> ha. Right you are.
178 2013-06-28 00:57:55 * sipa -> DreamLand
179 2013-06-28 00:59:00 <michagogo> Luke-Jr: I'm running now with -s 1
180 2013-06-28 00:59:29 <michagogo> Luke-Jr: And during the time the target went up to 00000000FFFF00000000000000000000... the reported difficulty in bfgminer stayed the same at 240
181 2013-06-28 00:59:35 <michagogo> Is that supposed to happen?
182 2013-06-28 00:59:36 ForceMajeure has joined
183 2013-06-28 01:00:03 <gmaxwell> without longpoll it won't notice quickly at least
184 2013-06-28 01:00:18 <gmaxwell> Luke-Jr: though you really should have a maximum scantime of 5 minutes or something like that.
185 2013-06-28 01:00:46 <michagogo> gmaxwell: That's 1 second
186 2013-06-28 01:01:07 <gmaxwell> odd
187 2013-06-28 01:01:21 <michagogo> Oh, wait...
188 2013-06-28 01:01:36 <michagogo> Maybe scanning is different from
189 2013-06-28 01:01:36 <michagogo> --expiry|-E <arg> Upper bound on how many seconds after getting work we consider a share from it stale (w/o longpoll active) (default: 120)
190 2013-06-28 01:02:55 wamatt has quit (Ping timeout: 268 seconds)
191 2013-06-28 01:04:49 GordonG3kko has joined
192 2013-06-28 01:05:22 GordonG3kko has quit (Remote host closed the connection)
193 2013-06-28 01:05:39 GordonG3kko has joined
194 2013-06-28 01:06:31 rdymac has joined
195 2013-06-28 01:06:31 rdymac has quit (Changing host)
196 2013-06-28 01:06:31 rdymac has joined
197 2013-06-28 01:07:09 michagogo_ has joined
198 2013-06-28 01:07:41 michagogo is now known as Guest46622
199 2013-06-28 01:07:41 Guest46622 has quit (Killed (moorcock.freenode.net (Nickname regained by services)))
200 2013-06-28 01:07:41 michagogo_ is now known as michagogo
201 2013-06-28 01:08:48 <TheLordOfTime> is it possible to run bitcoin-qt on the main network, and still run a bitcoin-qt client on the testnet
202 2013-06-28 01:16:19 FredEE_ has joined
203 2013-06-28 01:18:31 <michagogo> yes
204 2013-06-28 01:19:36 FredEE has quit (Ping timeout: 248 seconds)
205 2013-06-28 01:19:37 FredEE_ is now known as FredEE
206 2013-06-28 01:20:48 justusranvier has quit (Ping timeout: 268 seconds)
207 2013-06-28 01:26:08 wei_ has quit (Quit: wei_)
208 2013-06-28 01:27:49 <TheLordOfTime> awesome... now to hope i have the spare ram XD
209 2013-06-28 01:28:22 justusranvier has joined
210 2013-06-28 01:29:36 msvb-lab has quit (Read error: Operation timed out)
211 2013-06-28 01:33:07 bbbrian has quit (Ping timeout: 276 seconds)
212 2013-06-28 01:35:43 milone_ has joined
213 2013-06-28 01:38:17 <dan_> "listunspent" isn't working for multi-sig addresses for me (returns none), is there a good alternate way to find unspent multisig tx outputs?
214 2013-06-28 01:39:09 milone has quit (Ping timeout: 246 seconds)
215 2013-06-28 01:40:22 rdymac has quit (Quit: Saliendo)
216 2013-06-28 01:41:46 justusranvier has quit (Ping timeout: 268 seconds)
217 2013-06-28 01:42:02 <nsh> hmm
218 2013-06-28 01:47:10 msvb-lab has joined
219 2013-06-28 01:48:04 msvb-lab has quit (Client Quit)
220 2013-06-28 01:48:54 justusranvier has joined
221 2013-06-28 01:52:58 dan_ has quit (Remote host closed the connection)
222 2013-06-28 01:53:55 dan_ has joined
223 2013-06-28 01:56:04 GordonG3kko has quit (Remote host closed the connection)
224 2013-06-28 01:57:08 XCortex has joined
225 2013-06-28 01:57:50 mE\Ta has joined
226 2013-06-28 02:00:22 GordonG3kko has joined
227 2013-06-28 02:00:26 macboz_ has joined
228 2013-06-28 02:01:37 macboz has quit (Read error: Connection reset by peer)
229 2013-06-28 02:01:59 <runeks> What could be a reason that the secp256k1 library takes 7.8 seconds for a "test count" of 100? I assume that's 100 verifications in 7.8 seconds.
230 2013-06-28 02:02:07 justusranvier has quit (Ping timeout: 268 seconds)
231 2013-06-28 02:02:45 <phantomcircuit> sipa, runeks has a question
232 2013-06-28 02:03:06 GordonG3kko has quit (Remote host closed the connection)
233 2013-06-28 02:03:17 Maged has quit (Ping timeout: 256 seconds)
234 2013-06-28 02:03:48 GordonG3kko has joined
235 2013-06-28 02:04:50 sensorii has quit (Remote host closed the connection)
236 2013-06-28 02:05:23 sensorii has joined
237 2013-06-28 02:05:41 _milone has joined
238 2013-06-28 02:07:55 <runeks> Core 2 Quad 2.83 GHz CPU, by the way.
239 2013-06-28 02:08:37 roconnor has joined
240 2013-06-28 02:09:08 miscreanity has joined
241 2013-06-28 02:09:39 milone_ has quit (Ping timeout: 252 seconds)
242 2013-06-28 02:11:23 michagogo has quit (Ping timeout: 268 seconds)
243 2013-06-28 02:11:35 miscrean1ty has quit (Ping timeout: 264 seconds)
244 2013-06-28 02:11:35 devrandom has quit (Quit: leaving)
245 2013-06-28 02:12:12 <TheLordOfTime> how long does mining the testnet take CPU only? usually.
246 2013-06-28 02:12:30 devrandom has joined
247 2013-06-28 02:14:31 <runeks> TheLordOfTime: What is the difficulty?
248 2013-06-28 02:15:15 <TheLordOfTime> runeks: 240.48507088
249 2013-06-28 02:15:29 <runeks> Time to find a block is (difficulty * 2^32) / hashrate
250 2013-06-28 02:15:37 * TheLordOfTime grabs his calculator
251 2013-06-28 02:15:40 <runeks> *Average time
252 2013-06-28 02:15:48 <runeks> TheLordOfTime: What is your hashrate?
253 2013-06-28 02:16:08 <runeks> *Average time in seconds
254 2013-06-28 02:16:18 <TheLordOfTime> 2 threads, each averaging about 2300Khash/s
255 2013-06-28 02:16:21 <TheLordOfTime> (because cpu :/)
256 2013-06-28 02:16:42 <TheLordOfTime> ehh play it safe, each average 2000Khash/s
257 2013-06-28 02:17:03 <runeks> TheLordOfTime: My calculations say 62 hours, on average.
258 2013-06-28 02:17:09 <TheLordOfTime> figures
259 2013-06-28 02:17:16 <runeks> TheLordOfTime: I think I might have some testnet coins. You want some?
260 2013-06-28 02:17:28 <TheLordOfTime> ... oopsies, there goes the miner
261 2013-06-28 02:17:30 <runeks> Unless the testnet has been reset in the last year... which it probably has
262 2013-06-28 02:17:31 * TheLordOfTime kicks the ram
263 2013-06-28 02:17:59 <TheLordOfTime> runeks: if you have testnet coins that'd be great, i'd love to have some testnet coins to mess around with
264 2013-06-28 02:18:17 <runeks> TheLordOfTime: Checking now... but I think it has been reset since I got some.
265 2013-06-28 02:18:26 <TheLordOfTime> probably true
266 2013-06-28 02:18:41 <runeks> Was working on an in-browser miner that never got finished (needed testing).
267 2013-06-28 02:18:49 <runeks> In Javascript.
268 2013-06-28 02:18:50 <runeks> Very useful.
269 2013-06-28 02:19:38 <runeks> TheLordOfTime: Looks like I've got 99 testnet BTC. Give me an address. How many do you need? Is 50 good?
270 2013-06-28 02:19:39 <TheLordOfTime> i saw a site with such a java applet
271 2013-06-28 02:20:01 <TheLordOfTime> runeks: 50 is fine. mvaHx8M2aAr8uziXE32PmzHGDhNfaANTYd is my testnet address on the system i'm fiddling with
272 2013-06-28 02:20:01 <runeks> TheLordOfTime: I did it in Javascript though. Just for fun. To demonstrate that anyone can mine. It's just hard.
273 2013-06-28 02:20:11 <runeks> TheLordOfTime: Comin right up.
274 2013-06-28 02:20:13 <TheLordOfTime> at least, until my other testnet wallet syncs up
275 2013-06-28 02:20:16 <TheLordOfTime> thanks :)
276 2013-06-28 02:20:28 mE\Ta has quit (Ping timeout: 246 seconds)
277 2013-06-28 02:20:31 <nsh> runeks, code available for that js miner?
278 2013-06-28 02:20:40 milone_ has joined
279 2013-06-28 02:20:46 <runeks> nsh: I can put it on Github if you're interested.
280 2013-06-28 02:21:05 <nsh> runeks, if it's no bother i'd certainly take a look. (don't expect any contributions though)
281 2013-06-28 02:21:06 <nsh> :)
282 2013-06-28 02:21:07 <runeks> nsh: It's with a Python backend that builds the block. Prepare for ugly code though.
283 2013-06-28 02:21:17 <nsh> runeks, ugly code is my forte
284 2013-06-28 02:21:40 <runeks> nsh: You'll *love* it then!
285 2013-06-28 02:21:44 <TheLordOfTime> runeks: thanks!
286 2013-06-28 02:21:44 <runeks> TheLordOfTime: Coins sent.
287 2013-06-28 02:21:50 <runeks> c41c517357d94efa1826e8fb2144ef80026725bae0f79b1223122b21e1881a7d
288 2013-06-28 02:21:50 <TheLordOfTime> got em thanks!
289 2013-06-28 02:21:54 <runeks> Cool!
290 2013-06-28 02:22:02 <TheLordOfTime> ehehehehehehehehe
291 2013-06-28 02:22:10 <TheLordOfTime> i'm seriously taxing my RAM, and probably my swap
292 2013-06-28 02:22:21 <TheLordOfTime> i've got bitcoin-qt on the primary network and bitcoin-qt on the testnet on this one system
293 2013-06-28 02:24:50 <runeks> nsh: Looks like it's already up: https://github.com/runeksvendsen/findablock
294 2013-06-28 02:24:57 _milone has quit (Ping timeout: 268 seconds)
295 2013-06-28 02:24:59 <runeks> nsh: Do let me know if you can get it to work. I might have to create a README
296 2013-06-28 02:25:14 <nsh> thanks!
297 2013-06-28 02:27:15 <runeks> nsh: I was planning on finishing it by actually making sure that if a block is found (very unlikely) that the block is announced to the network, and the user gets paid. But I never got around to that. It does work for mining though, and will notify you if you find a block on the website.
298 2013-06-28 02:27:58 <nsh> runeks, right
299 2013-06-28 02:28:22 <nsh> runeks, using native js sha256?
300 2013-06-28 02:28:43 <nsh> apparently the latest firefox js runs are "close to not terrible" speed
301 2013-06-28 02:28:46 <nsh> *at
302 2013-06-28 02:29:17 <runeks> nsh: Using CryptoJS IIRC.
303 2013-06-28 02:29:21 <TheLordOfTime> runeks: any idea what the average testnet block time is?
304 2013-06-28 02:29:26 shesek has joined
305 2013-06-28 02:29:36 <runeks> nsh: It's a really shitty implementation though, just proof-of-concept.
306 2013-06-28 02:29:56 <nsh> runeks, right
307 2013-06-28 02:29:57 <shesek> is there someone here with access to bitcointalk's statistics?
308 2013-06-28 02:30:14 <runeks> nsh: So it's controlled by a timer, to prevent it from hanging the browser window (Javascript isn't multi-threaded). So I get around 220 hashes per seconds.
309 2013-06-28 02:30:36 <nsh> hmm
310 2013-06-28 02:30:42 <shesek> I'm building some web-based project for Bitcoin, and wondering which browsers I should support
311 2013-06-28 02:31:00 <nsh> i wonder if there's a more variable way to do it without sucking all the cycles
312 2013-06-28 02:31:10 <shesek> my guess is that most bitcoin users are somewhat tech-savvy and use modern browsers, but I wonder how many people would care if I drop support for IE <9
313 2013-06-28 02:31:12 <nsh> maybe in one of these async javascript platforms
314 2013-06-28 02:31:31 metabyte has joined
315 2013-06-28 02:31:34 <shesek> why not use a webworker?
316 2013-06-28 02:31:46 <nsh> shesek, just wrap the latest IE installer in candy and kittens and serve it to anyone hitting the site with an old browser
317 2013-06-28 02:31:52 <nsh> shesek, problem should solve itself
318 2013-06-28 02:32:06 <shesek> I'm display a message to older browsers, with a link to browserhappy
319 2013-06-28 02:32:21 <shesek> but still... I'm wondering how many users it'll effect
320 2013-06-28 02:32:24 <nsh> shesek, oh nice. (wasn't aware of web-workers)
321 2013-06-28 02:32:45 * nsh checks http://caniuse.com/#search=webworker
322 2013-06-28 02:32:50 <shesek> what is that shitty proof-of-concept implementation? I joined after you started talking about
323 2013-06-28 02:33:16 <shesek> s/$/ it/
324 2013-06-28 02:33:17 <nsh> https://github.com/runeksvendsen/findablock
325 2013-06-28 02:33:19 Subo1978_ has joined
326 2013-06-28 02:33:30 <nsh> javascript bitcoin miner
327 2013-06-28 02:33:52 super3 has joined
328 2013-06-28 02:33:56 <shesek> and you can kinda simulate webworker with an iframe
329 2013-06-28 02:34:26 <nsh> shesek, how would that go?
330 2013-06-28 02:34:56 <super3> shesek, web mining?
331 2013-06-28 02:35:01 <shesek> just create an iframe and execute that code in there; it should get its own thread and not block the primary window UI
332 2013-06-28 02:35:12 <nsh> oh, simple
333 2013-06-28 02:35:18 <shesek> (it might depend on the browser, perhaps some won't create a separate thread, I'm not sure)
334 2013-06-28 02:35:47 <nsh> i wonder how much editing would be required to support mining pools
335 2013-06-28 02:35:53 <runeks> nsh: Added instructions: https://github.com/runeksvendsen/findablock
336 2013-06-28 02:36:12 <nsh> it might be somewhat nonsilly to allow a bunch of people who like a project on the web to dedicate a tab to slowly crowdfunding it with their cycles
337 2013-06-28 02:36:20 <shesek> and wouldn't a javascript implementation that uses CPU (or anything CPU-based really) takes non-realistic times to generate?
338 2013-06-28 02:36:29 Wasp2744373175 has joined
339 2013-06-28 02:36:41 Subo1978 has quit (Ping timeout: 240 seconds)
340 2013-06-28 02:36:51 michagogo has joined
341 2013-06-28 02:36:52 <nsh> shesek, not after january when dan kaminsky replaces the PoW algorithm with something that can only be solved with HTML5
342 2013-06-28 02:37:06 <runeks> nsh: The reason I made it was to demonstrate that anyone can mine. It really isn't useful for actually helping anything. At least not at 230 hashes per second.
343 2013-06-28 02:37:10 Guest97364 has quit (Changing host)
344 2013-06-28 02:37:10 Guest97364 has joined
345 2013-06-28 02:37:13 Guest97364 is now known as jgarzik
346 2013-06-28 02:37:17 <shesek> "can _only_ be solved with HTML5"?
347 2013-06-28 02:37:26 Zoop_ has quit (Disconnected by services)
348 2013-06-28 02:37:29 <nsh> shesek, jokes :)
349 2013-06-28 02:37:54 <shesek> :)
350 2013-06-28 02:38:06 <nsh> runeks, there's a lot of browsers out there...
351 2013-06-28 02:38:09 <shesek> I did hear about them planning to switch the PoW function in some conference
352 2013-06-28 02:38:20 <runeks> TheLordOfTime: Nope. I don't know the average block time of testnet. I'd guess it's set to 10 minutes like Bitcoin proper, but I think it varies a lot because the difficulty jumps up and miners retract the hashing power.
353 2013-06-28 02:38:21 <shesek> but wasn't sure they're actually doing that... are they?
354 2013-06-28 02:38:31 <nsh> shesek, the consesus is "not likely"
355 2013-06-28 02:38:34 <shesek> I heard something like "it won't last until the end of the year" or something
356 2013-06-28 02:38:38 XCortex has quit (Quit: 0out)
357 2013-06-28 02:38:50 <nsh> but dan kaminsky was super-confident about that assertion on the panel
358 2013-06-28 02:39:01 OneFixt has joined
359 2013-06-28 02:39:04 <nsh> to mostly entertainment and minor chagrin
360 2013-06-28 02:39:57 Wasp2744373175 has left ()
361 2013-06-28 02:40:10 <TheLordOfTime> runeks: ehh figures
362 2013-06-28 02:40:15 Zoop_ has joined
363 2013-06-28 02:40:17 <TheLordOfTime> blocks do seem slower
364 2013-06-28 02:40:31 <runeks> nsh: True. If a billion people mined at 230 hashes per second, someone would find a block, on average, every 4½ days.
365 2013-06-28 02:40:57 <super3> basically even with millions of people running it you would make at most like $5
366 2013-06-28 02:41:06 <super3> its not worth it
367 2013-06-28 02:41:17 <super3> unless you implemented some sort of webgl miner
368 2013-06-28 02:41:29 <super3> then you might get a reasonable hash rate
369 2013-06-28 02:41:36 <nsh> mmm
370 2013-06-28 02:41:42 <nsh> would be interesting to see
371 2013-06-28 02:41:55 <nsh> ;;title http://bitcoin.biniok.net/gl.html
372 2013-06-28 02:41:55 <gribble> Error: This url is not on the whitelist.
373 2013-06-28 02:42:00 <nsh> shup gribble
374 2013-06-28 02:42:08 <nsh> "This is a demo page that shows how to mine bitcoins via Javascript and WebGL. "
375 2013-06-28 02:42:11 <shesek> anyone running a website targeted at Bitcoin users and mind sharing browser statistics?
376 2013-06-28 02:42:58 <super3> shesek, i am
377 2013-06-28 02:43:13 <shesek> awesome :)
378 2013-06-28 02:43:20 <shesek> all I really need to know is the % of users with IE < 9
379 2013-06-28 02:43:20 <nsh> shesek, bitcointalk stats for nov 2011: http://bitcoinstats.org/post/browserstats.png
380 2013-06-28 02:43:22 <super3> shesek, trust me the numbers don't add up ive already looked into this
381 2013-06-28 02:43:26 <nsh> *bitcoinstats
382 2013-06-28 02:43:31 <runeks> super3: I know. I made as a counter-argument to the people saying you need expensive hardware to mine bitcoins, and that it's impossible for normal people. I mean, they were using that argument as though they thought you needed special access to something. I made this to demonstrate that *anyone* can mine. It's just a matter of how fast you do it.
383 2013-06-28 02:43:45 <shesek> nsh, oh cool, I didn't know they published that
384 2013-06-28 02:44:02 <nsh> shesek, was linked from a forum post probably parenthetically to make a point
385 2013-06-28 02:44:03 <nsh> https://bitcointalk.org/index.php?topic=52728.20
386 2013-06-28 02:44:10 <super3> runeks, good point
387 2013-06-28 02:44:10 <shesek> no specific IE version in there, but a total market share of 6.8% is pretty good
388 2013-06-28 02:44:33 <super3> i guess you could actually make something useful if you implmented bitcoinj's new microtransactions and a web miner
389 2013-06-28 02:44:49 and has quit (Quit: Page closed)
390 2013-06-28 02:44:50 <nsh> runeks, wouldn't it have been easier to just say "everyone can win a lottery, but people who can buy a million lottery tickets a week will win more often"?
391 2013-06-28 02:45:02 <nsh> then again, coding is better than explaining things
392 2013-06-28 02:45:17 <runeks> shesek: So you're saying that if I create an IFRAME with the actual mining stuff, it won't block the tab? The thing is though, that I would like a responsive UI. It an incrementing hash showing, and a hash rate showing as well, so it can't block the IFRAME either.
393 2013-06-28 02:45:19 <super3> nsh, unless its not commented
394 2013-06-28 02:45:29 <nsh> hehe
395 2013-06-28 02:45:36 <runeks> nsh: Sure. But not nearly as fun.
396 2013-06-28 02:45:43 * nsh nods
397 2013-06-28 02:45:49 <super3> the only proper way to do this any actually make something useful is to make a browser based coin
398 2013-06-28 02:45:52 <runeks> nsh: Also, one thing is an explanation. Another thing is experiencing it yourself.
399 2013-06-28 02:46:00 <nsh> indeed
400 2013-06-28 02:46:01 <super3> and*
401 2013-06-28 02:46:05 <shesek> runeks, just send an status update to the parent window every `(++i % 1000) == 0` or something?
402 2013-06-28 02:46:07 <nsh> super3, lolcatcoin?
403 2013-06-28 02:46:33 <super3> nsh, probably would work out
404 2013-06-28 02:46:35 <runeks> super3: Browser-based coin?
405 2013-06-28 02:46:48 <runeks> shesek: How would I do that exactly?
406 2013-06-28 02:46:49 <super3> lol blocks embeded in lolcat memes
407 2013-06-28 02:47:28 <runeks> shesek: Oh, so you're saying only run the miner in the iframe, and display the UI in the other frame?
408 2013-06-28 02:47:35 agnostic98 has quit (Remote host closed the connection)
409 2013-06-28 02:47:39 <nsh> runeks, sounds right
410 2013-06-28 02:47:49 <runeks> nsh: Sounds like a job for you :)
411 2013-06-28 02:48:12 agnostic98 has joined
412 2013-06-28 02:48:27 <nsh> i have taken a solemn vow of nonproductivity. but if i suffer a relapse it'll be first on my list of projects
413 2013-06-28 02:48:47 <runeks> nsh: Cool. Non-productivity can be good.
414 2013-06-28 02:48:54 <runeks> nsh: I'm not very productive these days either.
415 2013-06-28 02:49:17 <runeks> Which is demonstrated by me sitting up at 4:20 AM on IRC.
416 2013-06-28 02:49:23 * nsh smiles
417 2013-06-28 02:49:49 <shesek> runeks. display the UI in the main window, do all the computations inside an iframe
418 2013-06-28 02:50:00 <nsh> does your name literally mean like glyph or something in danish, runeks?
419 2013-06-28 02:50:01 <super3> https://www.khronos.org/webcl/
420 2013-06-28 02:50:08 <super3> http://bennolan.com/2011/04/28/proof-of-work-in-js.html
421 2013-06-28 02:50:09 <shesek> and send status updates to the main window from the iframe
422 2013-06-28 02:50:19 <super3> those too links provide some guidance
423 2013-06-28 02:50:27 <runeks> shesek: Right. Makes sense. How do you communicate with an iframe though?
424 2013-06-28 02:50:46 <super3> if you want to actually implement something like this you need to use webcl or something like it to access the gpu
425 2013-06-28 02:50:55 <shesek> runeks, the iframed window can access the parent window using `parent`
426 2013-06-28 02:50:59 <super3> runeks, ajax?
427 2013-06-28 02:51:00 <runeks> nsh: It means "Rune", my first name "K" for my middle name "Kjær" and "S" for my last name "Svendsen". That's all :).
428 2013-06-28 02:51:14 <nsh> right
429 2013-06-28 02:51:19 <shesek> runeks, so `window.update_statuc = function() { ... }` on the main window can be accessed as `parent.update_status()` in the iframe
430 2013-06-28 02:51:31 <nsh> i meant does Rune mean rune in Danish, but that's probably a silly question
431 2013-06-28 02:51:33 <runeks> shesek: Ah, cool.
432 2013-06-28 02:51:46 <shesek> alternatively, you can also use postMessage in newer browsers - but in that case you can probably just use webworkers instead
433 2013-06-28 02:51:51 <runeks> nsh: It means some type of old writing, IIRC.
434 2013-06-28 02:51:58 * nsh nods
435 2013-06-28 02:52:10 <shesek> runeks, tho generally speaking, its not the kind of project where support for older browsers is really needed
436 2013-06-28 02:52:12 <runeks> nsh: Like these: http://www.denstoredanske.dk/@api/deki/files/24817/=245414.801.png?size=webview
437 2013-06-28 02:52:14 <shesek> I'd just go with webworkers
438 2013-06-28 02:52:22 <nsh> runeks, or https://www.khronos.org/webcl/
439 2013-06-28 02:52:26 <nsh> or http://en.wikipedia.org/wiki/Kensington_Runestone even
440 2013-06-28 02:52:37 <nsh> (copy-paste from chromium works slower than i can alt-tab :(( )
441 2013-06-28 02:52:47 <runeks> nsh: That too
442 2013-06-28 02:53:04 agnostic98 has quit (Ping timeout: 276 seconds)
443 2013-06-28 02:53:17 <runeks> shesek: Yeah, I don't care about old browsers, generally. But I'm a hobbyist, so I can afford to not care.
444 2013-06-28 02:53:45 <runeks> nsh: How many browsers implement webcl though? I don't think it works for me (Ubuntu, open source radeon driver).
445 2013-06-28 02:53:55 <runeks> It sounds very new.
446 2013-06-28 02:54:17 <nsh> http://caniuse.com/#search=webgl
447 2013-06-28 02:54:28 <runeks> nsh: webgl==webcl?
448 2013-06-28 02:54:31 <nsh> apparently not
449 2013-06-28 02:54:37 <nsh> but caniuse doesn't return anything for webcl
450 2013-06-28 02:54:44 <nsh> it might be a reasonable proxy
451 2013-06-28 02:54:46 <runeks> nsh: Nor does it work for webgl for me.
452 2013-06-28 02:54:48 <runeks> 404
453 2013-06-28 02:54:58 <shesek> nsh, are you sure those are bitcointalk stats, and not bitcoinstats's stats?
454 2013-06-28 02:55:03 <runeks> Looks like the whole site is down for me.
455 2013-06-28 02:55:10 <TheLordOfTime> anyone here able to mine a few testnet blocks to get some of these transactions confirmed? :/
456 2013-06-28 02:55:15 <nsh> "Currently, no browsers natively support WebCL, seeing as it is new. However, non-native add-ons are used to implement WebCL. For example, Nokia developed a WebCL extension." -WP
457 2013-06-28 02:55:37 <runeks> TheLordOfTime: I can afford some spare CPU cycles.
458 2013-06-28 02:56:00 <TheLordOfTime> thank you kindly
459 2013-06-28 02:56:24 <runeks> TheLordOfTime: But it's only CPU, so it probably won't matter much.
460 2013-06-28 02:56:33 <TheLordOfTime> indeed
461 2013-06-28 02:56:39 <runeks> Everyone: Run this:
462 2013-06-28 02:56:40 <runeks> bitcoind -testnet -gen=1
463 2013-06-28 02:56:58 <TheLordOfTime> if only for a copule of blocks plz :P
464 2013-06-28 02:57:35 <runeks> TheLordOfTime: What's your block count? I have 78959 blocks.
465 2013-06-28 02:58:07 <TheLordOfTime> 93708 on my synced up testnet client
466 2013-06-28 02:58:22 <runeks> TheLordOfTime: Ok. Need to catch up first.
467 2013-06-28 02:58:31 <TheLordOfTime> shouldn't take long if you're on 64bit
468 2013-06-28 02:58:40 brson has quit (Quit: leaving)
469 2013-06-28 02:58:44 <runeks> Seems to be going pretty fast.
470 2013-06-28 02:58:46 <TheLordOfTime> this 32bit system elsewhere on my net is lagged at 80513
471 2013-06-28 02:59:15 <runeks> bitcoind should really have a "block per second" statistic in "getinfo" when doing a sync.
472 2013-06-28 02:59:32 justusranvier has joined
473 2013-06-28 02:59:32 <runeks> *blocks per second
474 2013-06-28 03:00:48 wei_ has joined
475 2013-06-28 03:01:33 <shesek> super3, mind sharing your stats?
476 2013-06-28 03:01:58 <super3> what ones do you want?
477 2013-06-28 03:02:43 <shesek> % of IE users with versions older than 9
478 2013-06-28 03:02:44 <runeks> TheLordOfTime: "hashespersec" : 4517466
479 2013-06-28 03:03:40 <TheLordOfTime> urgh
480 2013-06-28 03:03:49 <runeks> So I'll probably find a block in 63½ hours :\
481 2013-06-28 03:03:51 <TheLordOfTime> 1826786 is mine
482 2013-06-28 03:04:01 <TheLordOfTime> :/
483 2013-06-28 03:04:03 * TheLordOfTime groans
484 2013-06-28 03:04:07 <runeks> Looks like it's using all my CPU cores.
485 2013-06-28 03:04:15 <runeks> TheLordOfTime: Do you have a multi-core CPU?
486 2013-06-28 03:04:22 <TheLordOfTime> i betcha it has something to do with the fact i put one of my CPU cores to building this debian package
487 2013-06-28 03:04:26 <TheLordOfTime> runeks: dual core
488 2013-06-28 03:04:36 <runeks> TheLordOfTime: That probably explains it :)
489 2013-06-28 03:04:41 <super3> looks like 9% IE
490 2013-06-28 03:04:57 <TheLordOfTime> runeks: yeah, the fact 1 of my cores is completely dedicated to pbuilder-dist and the other is on bitcoin-qt... :P
491 2013-06-28 03:05:03 <runeks> TheLordOfTime: What GPU do you have?
492 2013-06-28 03:05:05 <TheLordOfTime> that explains LowHashRate
493 2013-06-28 03:05:09 <TheLordOfTime> runeks: too old
494 2013-06-28 03:05:29 <TheLordOfTime> > 5 years old on a laptop that should have died last year
495 2013-06-28 03:05:33 <TheLordOfTime> but hasn't
496 2013-06-28 03:05:37 <runeks> TheLordOfTime: I have a Radeon 5700, but I'd have to install the Catalyst proprietary driver, which really sucks, for it to work. *And* set everything up.
497 2013-06-28 03:05:38 <TheLordOfTime> so i can't get a free upgrade from the parents :/
498 2013-06-28 03:05:57 <super3> totally maybe 3% with version >8
499 2013-06-28 03:05:58 <TheLordOfTime> runeks: come several paychecks i'm overhauling my desktop
500 2013-06-28 03:06:04 <TheLordOfTime> like *full* overhaul
501 2013-06-28 03:06:12 <runeks> TheLordOfTime: Cool. Whatcha' getting?
502 2013-06-28 03:06:19 <runeks> My desktop is from 2008.
503 2013-06-28 03:06:35 <TheLordOfTime> runeks: parts.
504 2013-06-28 03:06:44 <TheLordOfTime> my desktop's 3 years old custombuild
505 2013-06-28 03:06:48 <TheLordOfTime> i'm redoing basically everything
506 2013-06-28 03:06:55 <TheLordOfTime> 'cept the case because it's not a bad case
507 2013-06-28 03:07:08 <TheLordOfTime> but i haven't got money to start specing it out
508 2013-06-28 03:07:16 * TheLordOfTime also hasn't decided liquid v. air cooling
509 2013-06-28 03:07:22 <runeks> I really haven't followed the CPU scene the last five years.
510 2013-06-28 03:07:33 <TheLordOfTime> quad seems to be the standard
511 2013-06-28 03:07:43 * TheLordOfTime shrugs
512 2013-06-28 03:07:44 <runeks> I get by pretty well with my Core 2 Quad. But more cores would be kinda awesome.
513 2013-06-28 03:08:44 <runeks> Only thing that sucks about my PC is that it has an Abit motherboard. They went bankrupt after 2008, and there's a bug in the BIOS that doesn't allow AHCI and a USB keyboard at the same time.
514 2013-06-28 03:09:01 robocoin_ has joined
515 2013-06-28 03:09:17 <runeks> TheLordOfTime: Do you want performance or silence, when it comes to cooling?
516 2013-06-28 03:09:44 <TheLordOfTime> runeks: the thing has to double as a server, so silence is kinda not needed, performance is the plan
517 2013-06-28 03:10:08 <TheLordOfTime> ...uhoh...
518 2013-06-28 03:10:17 <TheLordOfTime> my netbook feels hot to the touch...
519 2013-06-28 03:10:25 <runeks> TheLordOfTime: Cool. Then, personally, I'd go with air cooling. In fact, I'd do that regardless.
520 2013-06-28 03:10:26 <TheLordOfTime> almost like it'll go *POOF* *FIRE* *blue smoke*
521 2013-06-28 03:10:50 <runeks> I don't think "hot" is a problem. Unless it's 120C hot.
522 2013-06-28 03:11:44 <shesek> <super3> looks like 9% IE <super3> totally maybe 3% with version >8
523 2013-06-28 03:12:02 <shesek> so that's 6% of IE <=8?
524 2013-06-28 03:12:11 robocoin has quit (Ping timeout: 264 seconds)
525 2013-06-28 03:12:18 <shesek> or did you mean 3% with version 8 (and not >8)?
526 2013-06-28 03:12:30 <super3> also i must add the obligatory ie can go suck a dick
527 2013-06-28 03:13:14 <runeks> Indeed.
528 2013-06-28 03:13:19 <TheLordOfTime> runeks: the keyboard's hot to the touch
529 2013-06-28 03:13:28 <TheLordOfTime> not just the exhaust port and bottom of the system
530 2013-06-28 03:13:31 <TheLordOfTime> the entire thing is hot
531 2013-06-28 03:13:34 <runeks> TheLordOfTime: Don't you have thermo sensors you can query?
532 2013-06-28 03:13:34 <shesek> any chance you could look up the % of IE8 users, specifically?
533 2013-06-28 03:13:38 * TheLordOfTime should probably turn it off
534 2013-06-28 03:13:42 <shesek> pretty please? :)
535 2013-06-28 03:13:52 <TheLordOfTime> runeks: um... if i could touch the computer without getting minor 1st degree burns, yes.
536 2013-06-28 03:14:05 <TheLordOfTime> that's how hot it is
537 2013-06-28 03:14:08 <TheLordOfTime> which is bad
538 2013-06-28 03:14:31 <runeks> "*without* getting minor 1st degree burns"?
539 2013-06-28 03:14:47 owowo has quit (Quit: sayonara)
540 2013-06-28 03:15:03 <super3> oh i guess my stats were wrong
541 2013-06-28 03:15:09 <runeks> TheLordOfTime: If you can find out how to GPU solo mine on testnet, I might set it up. I can't find a guide though.
542 2013-06-28 03:15:16 <super3> that should be 5% or so >8
543 2013-06-28 03:15:28 <super3> it says 4.09% for ie9
544 2013-06-28 03:15:31 <super3> er ie8
545 2013-06-28 03:16:09 <shesek> thanks super3
546 2013-06-28 03:17:28 malaimo has quit (Ping timeout: 248 seconds)
547 2013-06-28 03:19:20 malaimo has joined
548 2013-06-28 03:20:03 <TheLordOfTime> runeks: i don't think it's different than GPU mining on realnet
549 2013-06-28 03:20:22 <runeks> TheLordOfTime: I found a guide. Do you still need it? Has the transaction not been verified?
550 2013-06-28 03:20:34 <TheLordOfTime> GUIminer, solo, host is the address, port is 18332, username, pw...
551 2013-06-28 03:20:35 <TheLordOfTime> *looks*
552 2013-06-28 03:20:45 <warren> hmm, anyone figured out how to make two separate forks of the same repo on github? the UI won't let you do it.
553 2013-06-28 03:20:55 <TheLordOfTime> runeks: finally, yes.
554 2013-06-28 03:20:57 <shesek> I wonder what would happen if someone connects an ASIC to testnet, mine a few blocks and leave
555 2013-06-28 03:21:06 <runeks> TheLordOfTime: Cool. So no need?
556 2013-06-28 03:21:14 <shesek> it'll probably be stuck for quite some time
557 2013-06-28 03:21:21 <TheLordOfTime> runeks: i'm testing some RPC calls so having one up for the next hour or so wouldn't hurt
558 2013-06-28 03:21:45 <warren> shesek: testnet has a rule that drops difficulty to 1 if there's no blocks for 20 minutes
559 2013-06-28 03:21:47 <runeks> TheLordOfTime: What do you mean exactly? I'm talking about GPU mining on testnet. Are we on the same page?
560 2013-06-28 03:21:54 <TheLordOfTime> yes on testnet.
561 2013-06-28 03:22:02 <TheLordOfTime> runeks: i'm going to be sending additional tx on testnet
562 2013-06-28 03:22:11 <runeks> TheLordOfTime: Cool. I'll see if I can get it up and running.
563 2013-06-28 03:22:13 dparrish_ is now known as dparrish
564 2013-06-28 03:22:24 <TheLordOfTime> cool, i won't need it for more than like 30 minutes
565 2013-06-28 03:22:30 <TheLordOfTime> :)
566 2013-06-28 03:22:38 <shesek> warren, oh, cool.
567 2013-06-28 03:23:01 <warren> shesek: it's especially fun when alt coins add that to their mainnet as a "fix"
568 2013-06-28 03:25:01 <shesek> seriously? which one did that?
569 2013-06-28 03:26:44 <warren> shesek: Terracoin... and hilarity followed.
570 2013-06-28 03:27:54 <shesek> oh, now that I think about it... someone actually did connect an ASIC to terracoin and left after raising the difficulty, no?
571 2013-06-28 03:28:05 <shesek> I vaguely remember something like that
572 2013-06-28 03:28:10 <warren> shesek: yeah, and adding the testnet rule was the "fix"
573 2013-06-28 03:28:49 <shesek> tho they're probably not dropping it to 1, do they?
574 2013-06-28 03:28:57 <warren> by half, IIRC
575 2013-06-28 03:29:09 <warren> in any case, you can't have a rule like that on mainnet
576 2013-06-28 03:29:31 <runeks> TheLordOfTime: What does your bitcoind say for the difficulty on testnet now? Mine says 1.
577 2013-06-28 03:29:37 <shesek> I guess an REALLY high variance from the expected average time could mean that something is really wrong and that the difficulty should be corrected
578 2013-06-28 03:29:44 <TheLordOfTime> same
579 2013-06-28 03:29:58 <runeks> Has it been reset?
580 2013-06-28 03:29:59 <TheLordOfTime> runeks: that should be relatively simpler to mine right?
581 2013-06-28 03:30:22 <TheLordOfTime> dunno
582 2013-06-28 03:30:31 <runeks> TheLordOfTime: Yes. It requires 7 MH/s for a block every 10 minutes, on average.
583 2013-06-28 03:30:49 agnostic98 has joined
584 2013-06-28 03:30:53 <runeks> So I should probably not spoil it with GPU mining now. Got the driver installed though.
585 2013-06-28 03:32:13 <TheLordOfTime> i could spin up minerd if i wanted to
586 2013-06-28 03:32:41 <runeks> What's minerd?
587 2013-06-28 03:32:53 <TheLordOfTime> sorry, cpuminer
588 2013-06-28 03:32:59 * TheLordOfTime didn't launch with gen=1
589 2013-06-28 03:33:11 <TheLordOfTime> but i might just relaunch
590 2013-06-28 03:33:24 <runeks> I'm CPU mining at 4.6 MH/s now. Should be sufficient for a block every 15-20 minutes (if it's just me).
591 2013-06-28 03:33:52 <TheLordOfTime> 1948234 is what mine's running at. which is bad.
592 2013-06-28 03:34:18 <runeks> It's a block every 10 minutes or so, since I'm mining at 4.5 MH/s.
593 2013-06-28 03:34:18 TheSeven has quit (Disconnected by services)
594 2013-06-28 03:34:25 <TheLordOfTime> mhm
595 2013-06-28 03:34:28 [7] has joined
596 2013-06-28 03:34:30 <TheLordOfTime> until the difficulty spikes again
597 2013-06-28 03:34:46 <runeks> If it does it'll be because of a lot of found blocks.
598 2013-06-28 03:34:53 <runeks> In 2016 blocks.
599 2013-06-28 03:35:43 _milone has joined
600 2013-06-28 03:35:51 <runeks> I'm taking a computer break now. I will let it run for now.
601 2013-06-28 03:36:03 eoss has joined
602 2013-06-28 03:36:23 <TheLordOfTime> ok thanks again runeks
603 2013-06-28 03:36:29 <runeks> TheLordOfTime: No prob :)
604 2013-06-28 03:39:34 milone_ has quit (Ping timeout: 268 seconds)
605 2013-06-28 03:40:15 <petertodd> Luke-Jr: re: http://gitorious.org/bitcoin/libblkmaker/commit/e7d36757ef0532b94350c6691edf41d9d5877eb6/diffs/1e44b127ba61db2e83387cc7741385d4393f2e5b looks reasonable enough, although I don't claim to follow the C all that well
606 2013-06-28 03:41:07 <petertodd> Luke-Jr: more generally I have to wonder if C is the right language for the block creation stuff in a mining app, but that's another issue
607 2013-06-28 03:41:17 <petertodd> Luke-Jr: the logic itself seems ok
608 2013-06-28 03:41:49 <jgarzik> raw API needs a good solution for providing change addresses
609 2013-06-28 03:42:46 <petertodd> jgarzik: or a good solution for accessing the coin selection code
610 2013-06-28 03:44:47 * nsh remembers what he wanted to ask
611 2013-06-28 03:45:34 <nsh> petertodd, a while back you mentioned that some people (colleagues of yours?) have a working hypothesis that quantum computation replaces NP-hard search problems with NP-hard engineering problems, or something to that effect
612 2013-06-28 03:45:35 xire has joined
613 2013-06-28 03:45:35 xire has quit (Changing host)
614 2013-06-28 03:45:35 xire has joined
615 2013-06-28 03:45:42 <TheLordOfTime> runeks: i assume you found the last block, all the TXes i have are confirmed now.l
616 2013-06-28 03:45:45 <nsh> i was wondering if you know of any literature supporting that view?
617 2013-06-28 03:45:52 <nsh> or if you could possibly ask
618 2013-06-28 03:45:55 <TheLordOfTime> well... 'cept for 5 of them
619 2013-06-28 03:47:44 <petertodd> nsh: http://www.scottaaronson.com/democritus/lec14.html is one, arguing that essentially working quantum computation is just an example of analog computing
620 2013-06-28 03:47:52 <petertodd> nsh: the rest is just engineering intuition
621 2013-06-28 03:48:04 <nsh> petertodd, thanks
622 2013-06-28 03:48:49 <nsh> oh, there's shades of Chaitin there
623 2013-06-28 03:48:59 <nsh> he's long advocated against infinite precision in natural process
624 2013-06-28 03:49:13 <nsh> or at least, skepticism against the continuum being physically real
625 2013-06-28 03:49:16 <petertodd> nsh: keep in mind that AFAIK physics isn't really at the level where we could argue that quantum computing isn't possible - we don't understand what's "under" quantum physics basically
626 2013-06-28 03:49:38 * nsh nods
627 2013-06-28 03:50:24 <petertodd> nsh: the engineering point is view then is simple: cutting noise in half requires even more than double the effort
628 2013-06-28 03:50:37 <nsh> well, it struck me as i've wondered for a while whether the difficult of writing quantum algorithms resulted primarily from our intuitive experience/familiarity with classical information and probability theory, or whether there was some actual inherent difficulty
629 2013-06-28 03:50:56 <nsh> hmm
630 2013-06-28 03:51:19 <nsh> however, reality can be safely assumed to be relatively efficient/elegant
631 2013-06-28 03:51:40 <nsh> modulo philosophical arguments about simulated reality, etc.
632 2013-06-28 03:51:40 <petertodd> well, remember that quantum algorithms are totally different from actual quantum computers...
633 2013-06-28 03:51:55 <petertodd> meh, I'm an engineer, reality seems like a bitch to me :P
634 2013-06-28 03:52:02 * nsh smiles
635 2013-06-28 03:52:20 Mylon has quit ()
636 2013-06-28 03:52:42 <nsh> perhaps it's analogous to the statement about the amount of energy in a cup of water (massive, assuming you could somehow convert matter to energy)
637 2013-06-28 03:52:54 <nsh> so perhaps any small region of space has awesome quantum computational capacity
638 2013-06-28 03:53:09 <nsh> but it's mostly irrevocably dedicated to the task of being a small region of space
639 2013-06-28 03:54:37 reneg has joined
640 2013-06-28 03:54:56 <petertodd> maybe - even if physics figures that out, I'm sure not going to understand the explanation :)
641 2013-06-28 03:55:01 reneg is now known as Guest79337
642 2013-06-28 03:56:07 <nsh> :)
643 2013-06-28 03:56:37 Guest28769 has quit (Ping timeout: 240 seconds)
644 2013-06-28 04:01:19 mrkent has quit (Ping timeout: 245 seconds)
645 2013-06-28 04:04:58 Maged has joined
646 2013-06-28 04:05:52 <CodeShark> the threshold for computational universality is so low that practically any physical system is a universal computer (only subject to space, time, energy, and entropy limitations) - if we only knew how to prepare the inputs and read the outputs.
647 2013-06-28 04:05:52 Mylon has joined
648 2013-06-28 04:06:01 ericmuyser has quit (Ping timeout: 248 seconds)
649 2013-06-28 04:07:01 <CodeShark> but I suppose the task of figuring out how to prepare the inputs and how to read the outputs might be far more complex than the problem we're trying to compute :)
650 2013-06-28 04:08:09 xire has quit (Ping timeout: 248 seconds)
651 2013-06-28 04:09:21 sacredchao has quit (Ping timeout: 240 seconds)
652 2013-06-28 04:09:52 <nsh> CodeShark, universality only tells you whether a system can emulate a particular computation, and neglects whether it can do so efficiently
653 2013-06-28 04:10:41 <CodeShark> nsh: right - it might require exponential time when compared to the familiar computing models
654 2013-06-28 04:10:47 sacredchao has joined
655 2013-06-28 04:11:07 <CodeShark> well...I suppose you could achieve poly-time for poly-time
656 2013-06-28 04:11:12 * nsh nods
657 2013-06-28 04:11:29 <CodeShark> but poly-time might be n^100000000
658 2013-06-28 04:11:30 <CodeShark> :p
659 2013-06-28 04:11:35 <nsh> :)
660 2013-06-28 04:11:45 FlyingLeap_ has left ()
661 2013-06-28 04:11:53 <nsh> it's all just interesting properties of incredibly large natural numbers :)
662 2013-06-28 04:12:01 FlyingLeap has joined
663 2013-06-28 04:14:43 paybitcoin has quit (Read error: Connection reset by peer)
664 2013-06-28 04:16:13 paybitcoin has joined
665 2013-06-28 04:17:22 one_zero has quit (Ping timeout: 246 seconds)
666 2013-06-28 04:25:25 bbbrian has joined
667 2013-06-28 04:26:19 roconnor has quit (Quit: Konversation terminated!)
668 2013-06-28 04:34:18 nowan_ has joined
669 2013-06-28 04:36:23 FredEE_ has joined
670 2013-06-28 04:36:57 nowan has quit (Ping timeout: 246 seconds)
671 2013-06-28 04:38:47 fanquake has joined
672 2013-06-28 04:39:41 FredEE has quit (Ping timeout: 256 seconds)
673 2013-06-28 04:39:42 FredEE_ is now known as FredEE
674 2013-06-28 04:48:01 justusranvier has quit (Ping timeout: 268 seconds)
675 2013-06-28 04:48:11 melvster has joined
676 2013-06-28 04:49:23 freewil has quit (Ping timeout: 264 seconds)
677 2013-06-28 04:51:22 Guest89887 has left ()
678 2013-06-28 04:52:02 Guest89887 has joined
679 2013-06-28 04:52:42 Guest89887 has left ()
680 2013-06-28 04:52:46 has joined
681 2013-06-28 04:52:46 has quit (Clown|!Clown@static-87-79-93-140.netcologne.de|Changing host)
682 2013-06-28 04:52:46 has joined
683 2013-06-28 04:53:34 setkeh has quit (Ping timeout: 256 seconds)
684 2013-06-28 04:54:42 justusranvier has joined
685 2013-06-28 04:56:44 RazielZ has joined
686 2013-06-28 04:56:45 agnostic98 has quit (Read error: Connection reset by peer)
687 2013-06-28 04:57:05 agnostic98 has joined
688 2013-06-28 04:57:31 ericmuys_ has joined
689 2013-06-28 04:58:32 da2ce7 has joined
690 2013-06-28 04:58:33 A2501 has quit (Remote host closed the connection)
691 2013-06-28 04:58:48 eoss has quit (Read error: Connection reset by peer)
692 2013-06-28 04:59:44 justusranvier has quit (Ping timeout: 268 seconds)
693 2013-06-28 05:01:35 daktak has quit (Ping timeout: 268 seconds)
694 2013-06-28 05:03:07 daktak has joined
695 2013-06-28 05:03:41 adam3us has joined
696 2013-06-28 05:05:40 mappum has quit (Read error: Connection reset by peer)
697 2013-06-28 05:05:46 setkeh has joined
698 2013-06-28 05:06:20 one_zero has joined
699 2013-06-28 05:16:39 patcon has joined
700 2013-06-28 05:18:35 Brizzo has joined
701 2013-06-28 05:18:58 Brizzo has quit (Client Quit)
702 2013-06-28 05:19:05 Brizzo has joined
703 2013-06-28 05:19:58 yubrew has joined
704 2013-06-28 05:22:42 RoboTeddy has quit (Ping timeout: 252 seconds)
705 2013-06-28 05:23:26 freewil has joined
706 2013-06-28 05:23:42 mokawa has joined
707 2013-06-28 05:25:14 justusranvier has joined
708 2013-06-28 05:26:31 andyh2 has joined
709 2013-06-28 05:31:20 agnostic98 has quit (Remote host closed the connection)
710 2013-06-28 05:34:58 wei_ has quit (Quit: wei_)
711 2013-06-28 05:35:02 Nesetalis has quit (Read error: Connection reset by peer)
712 2013-06-28 05:37:49 mokawa has left ()
713 2013-06-28 05:37:57 mokawa has joined
714 2013-06-28 05:38:20 Nesetalis has joined
715 2013-06-28 05:38:27 Cory has quit (Read error: Connection reset by peer)
716 2013-06-28 05:39:15 Cory has joined
717 2013-06-28 05:40:14 halalmeat has joined
718 2013-06-28 05:43:15 halalmeat has quit (Client Quit)
719 2013-06-28 05:43:47 <mokawa> hi
720 2013-06-28 05:43:55 <mokawa> i need help if someone is available
721 2013-06-28 05:44:47 <nsh> <mokawa> i backed up my wallet 8 transactions ago, put the wallet in my new hard drive, reindexed and rescanned the network. but my "change" from my last transaction didnt show up in my wallet? Why not?
722 2013-06-28 05:45:00 <mokawa> yep
723 2013-06-28 05:45:01 <nsh> (please wait a while for someone to respond)
724 2013-06-28 05:45:02 <mokawa> that
725 2013-06-28 05:45:48 <flound1129> please hold.
726 2013-06-28 05:45:49 <CodeShark> does bitcoind fully support 2-of-3 signature transactions?
727 2013-06-28 05:47:13 AtashiCon has quit (Quit: AtashiCon)
728 2013-06-28 05:47:18 <CodeShark> oh, I know where I screwed up...
729 2013-06-28 05:48:09 Jackneill has joined
730 2013-06-28 05:48:09 Jackneill has quit (Changing host)
731 2013-06-28 05:48:09 Jackneill has joined
732 2013-06-28 05:48:18 AtashiCon has joined
733 2013-06-28 05:48:31 <phantomcircuit> mokawa, did you resync completely
734 2013-06-28 05:48:49 <mokawa> im doing it again just in case but my wallet says 0
735 2013-06-28 05:48:58 <mokawa> it sent my change to a wallet i dont recognize
736 2013-06-28 05:49:10 sydna has joined
737 2013-06-28 05:49:11 <mokawa> which i read online is normal because its part of the same pool
738 2013-06-28 05:50:11 <mokawa> apparently bitcoin-qt doesnt show you your change wallets but since it doesnt show up in my balance at all, it scares me since i reformatted and backed up my wallet correctly
739 2013-06-28 05:51:57 <gmaxwell> mokawa: "change wallets"? Bitcoin-qt implements a single wallet. The balance reflects all confirmed funds spendable by you plus unconfirmed change.
740 2013-06-28 05:53:15 <gmaxwell> Without more details its hard to say what you're seeing. You won't see unconfirmed transactions that you lost by the recovery until they get confirmed.
741 2013-06-28 05:53:21 RazielZ has quit (Ping timeout: 256 seconds)
742 2013-06-28 05:53:33 sydna has quit (Quit: sydna)
743 2013-06-28 05:54:11 FredEE has quit (Quit: FredEE)
744 2013-06-28 05:54:13 <mokawa> apparently here is more on the issue that i am currently seeing: https://bitcointalk.org/index.php?topic=184768.0
745 2013-06-28 05:54:19 enikanorov_ has joined
746 2013-06-28 05:54:43 <Luke-Jr> mokawa: you sound a bit confused
747 2013-06-28 05:55:01 <gmaxwell> That post certantly is (as the first response points out)
748 2013-06-28 05:55:20 <Luke-Jr> makomk: on the Overview tab, what does it say your Balance is?
749 2013-06-28 05:55:23 <gmaxwell> I wonder how many people get bad advice from the forum because they google and get some old post that has a reasonable reponse that they don't read.
750 2013-06-28 05:55:35 <mokawa> 0
751 2013-06-28 05:56:11 wrabbit has quit (Ping timeout: 256 seconds)
752 2013-06-28 05:56:15 <mokawa> i'll try to slow down and explain a little more in debt
753 2013-06-28 05:56:18 <Luke-Jr> makomk: how about Unconfirmed?
754 2013-06-28 05:56:27 <Luke-Jr> you mean in depth? :/
755 2013-06-28 05:56:29 enikanorov has quit (Ping timeout: 248 seconds)
756 2013-06-28 05:56:32 <mokawa> depth*
757 2013-06-28 05:56:35 <mokawa> lol yes
758 2013-06-28 05:58:09 <mokawa> if you look at transaction: c11d729debf725802d6a6dd37fdd52286ebb338a68836aeadd71adc1fbe6e59c
759 2013-06-28 05:58:30 <mokawa> my change is 3.36BTC but it wasnt sent back to my wallet
760 2013-06-28 05:58:40 <mokawa> https://blockchain.info/tx/c11d729debf725802d6a6dd37fdd52286ebb338a68836aeadd71adc1fbe6e59c
761 2013-06-28 05:59:52 <gmaxwell> mokawaL why do you say it wasn't sent back to your wallet?
762 2013-06-28 06:00:06 <mokawa> because my balance says zero
763 2013-06-28 06:00:33 <gmaxwell> How did you backup your wallet?
764 2013-06-28 06:00:39 <gmaxwell> Do you have bitcoin-qt running right now?
765 2013-06-28 06:00:44 <mokawa> i have it running now
766 2013-06-28 06:00:51 <mokawa> im synced
767 2013-06-28 06:00:56 <phantomcircuit> mokawa, does it list other transactions
768 2013-06-28 06:01:04 <mokawa> it lists all my recent transactions
769 2013-06-28 06:01:08 <gmaxwell> open the debugging console and type "getinfo" how many blocks does it say?
770 2013-06-28 06:01:35 <gmaxwell> then type "validateaddress 131MQgNoj82fKM4ecn8PdMtGUBfWVg8CFS"
771 2013-06-28 06:01:38 <mokawa> 243698
772 2013-06-28 06:01:39 <gmaxwell> and does it say IsMine?
773 2013-06-28 06:01:51 <phantomcircuit> ;;bc,blocks
774 2013-06-28 06:01:51 <gribble> 243698
775 2013-06-28 06:01:54 <mokawa> ismine" : false
776 2013-06-28 06:02:09 <gmaxwell> Indeed. That wallet doesn't have that key.
777 2013-06-28 06:02:14 wrabbit has joined
778 2013-06-28 06:02:15 <phantomcircuit> try again for 13V5KB68Lw5eL29bABweaL8YiPMiydyp8P
779 2013-06-28 06:02:18 <gmaxwell> Did you recently encrypt your wallet?
780 2013-06-28 06:02:30 <mokawa> mightve, do i have to decrypt it?
781 2013-06-28 06:02:45 <phantomcircuit> gmaxwell, shouldn't matter right?
782 2013-06-28 06:02:59 <gmaxwell> mokawa: No you don't have to decrypt it. But if you backed up before encrypting then that backup would have no change addresses you used after encrypting.
783 2013-06-28 06:03:07 <mokawa> address" : "13V5KB68Lw5eL29bABweaL8YiPMiydyp8P", "ismine" : false
784 2013-06-28 06:03:22 <mokawa> so i lost my btc :(
785 2013-06-28 06:03:23 <mokawa> ?
786 2013-06-28 06:03:36 <gmaxwell> mokawa: What happened to the original wallet?
787 2013-06-28 06:03:43 <phantomcircuit> he formatted
788 2013-06-28 06:03:47 <phantomcircuit> gmaxwell, wait seriously?
789 2013-06-28 06:03:50 <phantomcircuit> why is that
790 2013-06-28 06:04:05 <nsh> hmm
791 2013-06-28 06:04:10 <phantomcircuit> mokawa, try again 1FBTX17FYMnAxV9zrthvpSqhDFY46J1XEQ
792 2013-06-28 06:04:18 <gmaxwell> phantomcircuit: because the keypool was disclosed in the clear on the disk, when you encrypt for the first time it retires the keypool and fills it again.
793 2013-06-28 06:04:30 <mokawa> iscompressed" : true,
794 2013-06-28 06:04:39 <mokawa>  { "isvalid" : true, "address" : "1FBTX17FYMnAxV9zrthvpSqhDFY46J1XEQ", "ismine" : true, "isscript" : false, "pubkey" : "03bf304f828f8facfb294b69eb3c4f3d17abf420066cfec387130940a6359a52e6", "iscompressed" : true, "account" : "" }
795 2013-06-28 06:05:01 <gmaxwell> mokawa: so the sequence of events is something like: You backed up. You encrypted. You performed some transactions. You formatted. You restored. ?
796 2013-06-28 06:05:10 <mokawa> yep
797 2013-06-28 06:06:05 <gmaxwell> If so I recommend you stop writing to the disk right away and make an image of it to try to recover the deleted wallet. If you did the above your backup won't contain any change addresses used after you encrypted.
798 2013-06-28 06:06:47 <mokawa> recommend any software for recovery?
799 2013-06-28 06:07:34 <gmaxwell> normally I'd recommend this: https://bitcointalk.org/index.php?topic=25091.0 but it doesn't work on encrypted wallets. If you shut down the system and image the disk it'll probably compress pretty well and you can worry about trying to recover it later.
800 2013-06-28 06:08:03 <gmaxwell> e.g. maybe in 10 years those 3.3 btc will be worth paying someone to write a recovery tool for encrypted wallets if someone doesn't before then. :P
801 2013-06-28 06:08:34 mrkent has joined
802 2013-06-28 06:08:35 mrkent has quit (Changing host)
803 2013-06-28 06:08:35 mrkent has joined
804 2013-06-28 06:08:37 MobPhone has joined
805 2013-06-28 06:08:49 <mokawa> well shit, 300 down the drain
806 2013-06-28 06:09:23 * nsh promises to get mokawa a beer should the opportunity arise
807 2013-06-28 06:09:42 <mokawa> thanks nsh
808 2013-06-28 06:10:01 <gmaxwell> mokawa: there is a reasonable chance the data is still there if you only recently formatted... but recovering it would probably a big pain right now. :(
809 2013-06-28 06:10:39 <nsh> gmaxwell, can he not figure out the approximate filesize from the backup and narrow down the location on the disk that way?
810 2013-06-28 06:11:08 <gmaxwell> nsh: it wouldn't have been contigious on disk. a smart recovery tool wouldn't care but I'm not aware of any that work on encrypted wallets.
811 2013-06-28 06:11:48 <nsh> hmm, yeah you'd have to do some work i suppose
812 2013-06-28 06:12:16 <mokawa> so even if i got back my most recent wallet, i couldnt do anything?
813 2013-06-28 06:12:49 <gmaxwell> So the encryption process says in bold text: "IMPORTANT: Any previous backups you have made of your wallet file should be replaced with the newly generated... previous backups of the unencrypted wallet file will become useless as soon as you start using the new encrypted wallet"
814 2013-06-28 06:13:02 FredEE has joined
815 2013-06-28 06:13:05 <gmaxwell> I don't know how we can make it more obvious. :( if someone has ideas, please suggest them.
816 2013-06-28 06:13:30 <mokawa> lol GG for my wallet v_v
817 2013-06-28 06:13:38 <nsh> gmaxwell, silently backup the wallet file to numerous places on disk constantly :)
818 2013-06-28 06:13:57 <gmaxwell> mokawa: If you got the wallet from right before you formatted you'd be fine.
819 2013-06-28 06:14:07 <gmaxwell> Any backup from after you encrypted would have left you fine.
820 2013-06-28 06:14:26 <warren> sipa: https://github.com/litecoin-project/litecoin-0.8/commits/awesomecoin-cc7sec1 Here's where I encourage folks to test secp256k1...
821 2013-06-28 06:14:59 <gmaxwell> I think this is the first person I've talked to whos lost funds this way? not sure. OTOH, I've talked to two people who had stolen wallets and were foiled by this. :( Security and usibility are hard.
822 2013-06-28 06:16:19 <mokawa> it was an unexpected format unfortunately, blue screens after updating to windows 8.1 developer preview
823 2013-06-28 06:16:45 <mokawa> figured my old wallet was safe, oh well
824 2013-06-28 06:16:51 <gmaxwell> mokawa: crap. Of course it would have been trivial to recover at that point. :( just boot a linux rescue cd and copy off the wallet.
825 2013-06-28 06:17:00 <gmaxwell> but yea, you didn't realize.
826 2013-06-28 06:17:50 <gmaxwell> nsh: there is a tradeoff with more copies. Though I hope that our future wallet format will have some kind of forward error protection (if nothing else than just saving the data multiple times internally)
827 2013-06-28 06:17:58 <nsh> gmaxwell, looks like wallet encryption only mangles the privkeys in wallet.dat, so the file itself should still contain reasonable amounts of magic on disk for recognition. had a quick look to see if any existing solutions can find bdb files
828 2013-06-28 06:18:00 <nsh> but no luck
829 2013-06-28 06:18:09 <nsh> gmaxwell, right
830 2013-06-28 06:18:32 patcon has quit (Remote host closed the connection)
831 2013-06-28 06:18:37 <gmaxwell> nsh: that wallet recovery tool I linked to works by reconizing the private key records. e.g. ignoring the bdb format entirely. Presumably the same thing could be done for encrypted private keys.
832 2013-06-28 06:19:02 <nsh> gmaxwell, makes sense
833 2013-06-28 06:20:29 <gmaxwell> Esp if we've switched to determinstic wallets the most important private data will be really small, like 512 bits.. so repeating it 128 times (e.g. to span two full 4k sectors) wouldn't be a big deal.
834 2013-06-28 06:20:36 <nsh> mokawa, can you make an image file the disk and keep hold of it in case wallet-recover or another tool comes out?
835 2013-06-28 06:21:01 <mokawa> yes i can thankfully
836 2013-06-28 06:21:05 <gmaxwell> Personally I'd make an image of the disk. Then eventually a tool will come out that will be able to recover it.
837 2013-06-28 06:21:10 <mokawa> started the imaging process now
838 2013-06-28 06:21:33 <gmaxwell> you can look in the image file for evidence of BDB to confirm that it will probably be useful.
839 2013-06-28 06:22:40 toffoo has quit (Ping timeout: 252 seconds)
840 2013-06-28 06:23:16 <mokawa> def gonna try a bunch of things before giving up on this
841 2013-06-28 06:23:49 <mokawa> hopefully a wallet.dat file is intact and sequential
842 2013-06-28 06:25:14 <gmaxwell> it's might actually be mostly sequential since the encryption process fully rewrites it. At least enough that bdb's normal recovery could recover it.
843 2013-06-28 06:25:25 <gmaxwell> But I wouldn't bet on that.
844 2013-06-28 06:26:04 JZavala has joined
845 2013-06-28 06:26:10 <mokawa> if only BFL delivered, i wouldnt need to worry about a measly 3 btc
846 2013-06-28 06:26:23 freewil has quit (Quit: Leaving)
847 2013-06-28 06:29:59 <nsh> mokawa, what did you order?
848 2013-06-28 06:30:13 <mokawa> 7gh jalepeno
849 2013-06-28 06:30:34 <nsh> someone in #bitcoin received a bitforce (5gh) from BFL today
850 2013-06-28 06:30:43 <nsh> upgraded from a japaleno
851 2013-06-28 06:30:51 <nsh> or something like that
852 2013-06-28 06:31:11 <mokawa> so there really are BFL asics out there huh?
853 2013-06-28 06:31:32 <mokawa> keep hearing a lot of conspiracy theories
854 2013-06-28 06:32:36 <jgarzik> plenty of BFLs in the field now
855 2013-06-28 06:33:17 <mokawa> i hear jalapenos have a 100 day delivery period now, "supposedly"
856 2013-06-28 06:34:34 <mokawa> but they only started shipping last septembers so you can throw that out the window
857 2013-06-28 06:39:16 davout has joined
858 2013-06-28 06:39:16 davout has quit (Changing host)
859 2013-06-28 06:39:16 davout has joined
860 2013-06-28 06:42:41 <mokawa> the difficulty is going up 4 million every 10 days. anyone think this is crazy?
861 2013-06-28 06:43:50 <Eliel> no, it's just a concrete sign that a lot of asics are coming online
862 2013-06-28 06:45:21 grau has joined
863 2013-06-28 06:47:08 AtashiCon has quit (Quit: AtashiCon)
864 2013-06-28 06:47:22 AtashiCon has joined
865 2013-06-28 06:49:25 RazielZ has joined
866 2013-06-28 06:49:29 chorao has quit ()
867 2013-06-28 06:49:55 xire has joined
868 2013-06-28 06:49:55 xire has quit (Changing host)
869 2013-06-28 06:49:55 xire has joined
870 2013-06-28 06:49:59 super3 has quit (Ping timeout: 264 seconds)
871 2013-06-28 06:52:27 AtashiCon has quit (Quit: AtashiCon)
872 2013-06-28 06:52:42 AtashiCon has joined
873 2013-06-28 06:55:28 davout has quit (Remote host closed the connection)
874 2013-06-28 06:56:21 grau has quit (Remote host closed the connection)
875 2013-06-28 06:57:17 stretchwarren has quit (Ping timeout: 248 seconds)
876 2013-06-28 06:57:55 mokawa has quit (Ping timeout: 250 seconds)
877 2013-06-28 06:59:09 viperhr has joined
878 2013-06-28 06:59:53 Blitzboom has quit (Read error: Connection reset by peer)
879 2013-06-28 07:00:01 stretchwarren has joined
880 2013-06-28 07:00:04 Blitzboom has joined
881 2013-06-28 07:00:04 Blitzboom has quit (Changing host)
882 2013-06-28 07:00:04 Blitzboom has joined
883 2013-06-28 07:00:37 RoboTeddy has joined
884 2013-06-28 07:02:17 agnostic98 has joined
885 2013-06-28 07:04:03 AtashiCon has quit (Quit: AtashiCon)
886 2013-06-28 07:06:40 saulimus has joined
887 2013-06-28 07:06:47 Hunger- has quit (Read error: Operation timed out)
888 2013-06-28 07:07:24 grau has joined
889 2013-06-28 07:08:31 Eiii has quit ()
890 2013-06-28 07:14:54 Hunger- has joined
891 2013-06-28 07:17:52 justusranvier has quit (Ping timeout: 268 seconds)
892 2013-06-28 07:19:01 CodeName has joined
893 2013-06-28 07:23:13 Brizzo has quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
894 2013-06-28 07:23:44 idstam has joined
895 2013-06-28 07:25:25 justusranvier has joined
896 2013-06-28 07:29:00 patcon has joined
897 2013-06-28 07:30:15 _milone has quit (Quit: Leaving)
898 2013-06-28 07:30:49 milone has joined
899 2013-06-28 07:31:00 <petertodd> jgarzik: https://github.com/richardkiss/pycoin <- :(
900 2013-06-28 07:35:09 <sipa> runeks: the test binary does much more than just verifications, and it does everything in debug mode (with assertions enabled)
901 2013-06-28 07:35:21 gfinn has quit (Ping timeout: 240 seconds)
902 2013-06-28 07:35:22 patcon has quit (Ping timeout: 256 seconds)
903 2013-06-28 07:38:35 grau has quit (Remote host closed the connection)
904 2013-06-28 07:38:58 rainworm has joined
905 2013-06-28 07:39:24 CodeName has quit (Ping timeout: 256 seconds)
906 2013-06-28 07:42:52 <nsh> petertodd, why :( ?
907 2013-06-28 07:43:21 <petertodd> nsh: jgarzik has a python library already, and it can be made py3 compat
908 2013-06-28 07:43:31 grau has joined
909 2013-06-28 07:43:52 xire has quit (Ping timeout: 246 seconds)
910 2013-06-28 07:43:54 GordonG3kko has quit (Remote host closed the connection)
911 2013-06-28 07:43:57 <nsh> coding is not zero sum. richardkiss may solve a different set of problems, or create a different covering of solutions
912 2013-06-28 07:44:14 <petertodd> He might be. He probably isn't though.
913 2013-06-28 07:44:25 <nsh> true, but it worst it's an educational experience for him
914 2013-06-28 07:44:39 <nsh> may be worth pointing him at pynode though
915 2013-06-28 07:47:27 GordonG3kko has joined
916 2013-06-28 07:53:25 grau has quit (Remote host closed the connection)
917 2013-06-28 07:58:59 zer0def has quit (Ping timeout: 264 seconds)
918 2013-06-28 07:59:15 mappum has joined
919 2013-06-28 08:06:00 abrkn has quit ()
920 2013-06-28 08:06:07 AtashiCon has joined
921 2013-06-28 08:07:36 Thepok has joined
922 2013-06-28 08:14:05 milone_ has joined
923 2013-06-28 08:14:24 Neozonz has joined
924 2013-06-28 08:14:24 Neozonz has quit (Changing host)
925 2013-06-28 08:14:24 Neozonz has joined
926 2013-06-28 08:16:59 stretchwarren has quit (Ping timeout: 264 seconds)
927 2013-06-28 08:18:10 milone has quit (Ping timeout: 252 seconds)
928 2013-06-28 08:18:31 da2ce7 has quit (Ping timeout: 246 seconds)
929 2013-06-28 08:19:21 stretchwarren has joined
930 2013-06-28 08:21:01 MiningBuddy has quit (Remote host closed the connection)
931 2013-06-28 08:22:55 MiningBuddy has joined
932 2013-06-28 08:22:55 MiningBuddy has quit (Changing host)
933 2013-06-28 08:22:55 MiningBuddy has joined
934 2013-06-28 08:27:19 handle_ has quit (Remote host closed the connection)
935 2013-06-28 08:27:34 handle has joined
936 2013-06-28 08:27:47 t7 has joined
937 2013-06-28 08:27:56 swulf-- has joined
938 2013-06-28 08:33:01 Neozonz has quit (Ping timeout: 252 seconds)
939 2013-06-28 08:33:26 agnostic98 has quit (Remote host closed the connection)
940 2013-06-28 08:33:26 mappum has quit (Ping timeout: 240 seconds)
941 2013-06-28 08:33:35 Neozonz has joined
942 2013-06-28 08:38:06 one_zero has quit (Ping timeout: 240 seconds)
943 2013-06-28 08:38:21 Neozonz has quit (Remote host closed the connection)
944 2013-06-28 08:39:54 FredEE has quit (Quit: FredEE)
945 2013-06-28 08:40:04 one_zero has joined
946 2013-06-28 08:41:52 MoALTz_ is now known as MoALTz
947 2013-06-28 08:46:46 stretchwarren has quit (Ping timeout: 256 seconds)
948 2013-06-28 08:47:26 TD[gone] has left ("Leaving")
949 2013-06-28 08:47:30 TD[gone] has joined
950 2013-06-28 08:47:31 TD[gone] is now known as TD
951 2013-06-28 08:58:53 nx201 has quit (Ping timeout: 248 seconds)
952 2013-06-28 09:01:02 nx201 has joined
953 2013-06-28 09:02:05 justusranvier has quit (Ping timeout: 268 seconds)
954 2013-06-28 09:03:19 <michagogo> What will it look like if I manage to find a block? (running bfgminer with --coinbase-addr mvXQFs7xmJ5PJGrqYYLMnZ31DcaqXVASFA)
955 2013-06-28 09:03:43 <michagogo> (mvXQFs7xmJ5PJGrqYYLMnZ31DcaqXVASFA being an address in my main client wallet [not the no-relay one running in a VM])
956 2013-06-28 09:03:58 agnostic98 has joined
957 2013-06-28 09:06:59 <nsh> michagogo, the last scene in 2001: A Space Odyssey i my guess
958 2013-06-28 09:07:04 <nsh> *is
959 2013-06-28 09:07:25 <nsh> "my god! it's full of dust..."
960 2013-06-28 09:07:42 <michagogo> ...what?
961 2013-06-28 09:08:08 <nsh> http://www.youtube.com/watch?v=AXS8P0HksQo
962 2013-06-28 09:08:13 <sipa> michagogo: you will create a block, with a coinbase payout to mvXQFs7xmJ5PJGrqYYLMnZ31DcaqXVASFA
963 2013-06-28 09:08:13 <michagogo> I mean, how and when will the coinbase show up in my wallet?
964 2013-06-28 09:08:26 <sipa> it will show up in the wallet it credits
965 2013-06-28 09:08:37 justusranvier has joined
966 2013-06-28 09:08:52 <michagogo> Immediately?
967 2013-06-28 09:09:11 <sipa> the GUI only shows it after one confirmation i think
968 2013-06-28 09:09:21 <sipa> (so after another block is created on top of it)
969 2013-06-28 09:09:27 <michagogo> Erm, isn't a coinbase transaction inherently confirmed?
970 2013-06-28 09:09:32 <michagogo> So that's 2 confirmations
971 2013-06-28 09:09:51 reizuki__ has joined
972 2013-06-28 09:09:51 reizuki__ has quit (Changing host)
973 2013-06-28 09:09:51 reizuki__ has joined
974 2013-06-28 09:09:56 Prattler has joined
975 2013-06-28 09:10:43 agnostic98 has quit (Read error: Connection reset by peer)
976 2013-06-28 09:11:00 <sipa> michagogo: right!
977 2013-06-28 09:11:12 Nothing4You has quit (K-Lined)
978 2013-06-28 09:11:12 <sipa> after 1 extra confirm, i mean - but you're right
979 2013-06-28 09:12:02 Guest83631 has joined
980 2013-06-28 09:13:19 <imd23> sips: I can't stop reading your answers on bitcoin stack exchange. you rock!
981 2013-06-28 09:15:02 <imd23> sipa: that message was to you :p typo
982 2013-06-28 09:15:26 <sipa> yw :)
983 2013-06-28 09:15:28 Nothing4You has joined
984 2013-06-28 09:15:49 <imd23> ha
985 2013-06-28 09:16:16 valparaiso has quit (Quit: We are not gonna die. You know why? Because we are so... very... pretty.)
986 2013-06-28 09:16:45 <imd23> sipa: can I turn the key pool off from bitcoind? I will use it as a server for the core of a new online service I am developing
987 2013-06-28 09:17:09 patcon has joined
988 2013-06-28 09:17:16 <imd23> So I need bitcoind to be as stataless/user-specific/wallet-specific as possible.
989 2013-06-28 09:17:40 <imd23> I will use it as a gateway to the bit coin network.
990 2013-06-28 09:17:44 <sipa> you're not using the wallet at all?
991 2013-06-28 09:17:49 <imd23> nop.
992 2013-06-28 09:17:57 <imd23> I don't know so.
993 2013-06-28 09:18:08 <imd23> I don't think so (not know)
994 2013-06-28 09:18:21 <sipa> if you don't, the keypool doesn't matter
995 2013-06-28 09:18:26 <nsh> ".tf is the Internet country code top-level domain (ccTLD) for the French Southern and Antarctic Lands."
996 2013-06-28 09:18:56 <nsh> why would someone register a domain from such an obscure region for such an awful vhost
997 2013-06-28 09:19:01 <imd23> sipa: I mean, I need to generate address...
998 2013-06-28 09:19:14 <sipa> then you are using the wallet
999 2013-06-28 09:19:21 <imd23> sipa: maybe I can handle the generation on my service.
1000 2013-06-28 09:19:47 <imd23> sipa: but I need to generate and get the priv key, not store it in the wallet.
1001 2013-06-28 09:20:00 <CodeShark> feel free to use my refactor: https://github.com/CodeShark/CoinClasses/blob/master/src/CoinKey.h
1002 2013-06-28 09:20:04 <CodeShark> and https://github.com/CodeShark/CoinClasses/blob/master/src/CoinKey.cpp
1003 2013-06-28 09:20:28 <imd23> hehe, awesome CodeShark
1004 2013-06-28 09:21:43 <sipa> imd23: you need the wallet to learn about transactions crediting your addresses, and to spend the received coins
1005 2013-06-28 09:21:51 <sipa> imd23: you can do it all outside of bitcoind of course
1006 2013-06-28 09:22:03 <sipa> but something will need to watch/listen for transactions
1007 2013-06-28 09:22:15 valparaiso_afk has joined
1008 2013-06-28 09:22:23 <imd23> bitcoind can listen for transactions.
1009 2013-06-28 09:22:24 <CodeShark> I don't recommend using bitcoind for watching addresses at all :)
1010 2013-06-28 09:22:30 valparaiso_afk is now known as valparaiso
1011 2013-06-28 09:22:31 <imd23> ?
1012 2013-06-28 09:22:37 <sipa> imd23: yes, for transactions to its wallet
1013 2013-06-28 09:22:59 <CodeShark> it's generally a BAD idea to listen for transactions on the same machine that stores the private keys
1014 2013-06-28 09:23:07 <sipa> agree
1015 2013-06-28 09:23:14 Nothing4You is now known as NOTHING4YOU
1016 2013-06-28 09:23:28 <imd23> It won't be the same machine
1017 2013-06-28 09:23:36 <imd23> I am using txindex=1
1018 2013-06-28 09:23:41 stretchwarren has joined
1019 2013-06-28 09:23:42 <imd23> isn't this for listening to all transactions?
1020 2013-06-28 09:23:48 <CodeShark> what I'm saying is that the machine that listens for transactions shouldn't be storing any private keys at all
1021 2013-06-28 09:23:54 <imd23> it won't
1022 2013-06-28 09:24:07 <sipa> imd23: no, that allows you to look up any transaction
1023 2013-06-28 09:24:08 <imd23> one server will have bitcoind and other our App/service
1024 2013-06-28 09:24:23 <CodeShark> you'll need addrindex, too
1025 2013-06-28 09:24:29 <sipa> but it doesn't allow you to answer the question "given a set of addresses, tell me all relevant transactions"
1026 2013-06-28 09:24:30 <imd23> ok. so I can then use getrawtransaction is that ok?
1027 2013-06-28 09:24:36 <sipa> and that's really what addrindex is NOT for
1028 2013-06-28 09:24:41 <CodeShark> hehe
1029 2013-06-28 09:24:43 patcon has quit (Ping timeout: 252 seconds)
1030 2013-06-28 09:24:43 michagogo is now known as MICHAGOGO
1031 2013-06-28 09:24:44 <CodeShark> right, sipa
1032 2013-06-28 09:24:58 <imd23> ok.. let met think what i am missing here...
1033 2013-06-28 09:25:02 <CodeShark> to listen for transactions, use this: https://github.com/CodeShark/CoinClasses/blob/master/examples/listener2/listener2.cpp :)
1034 2013-06-28 09:25:47 <CodeShark> I need to add queueing to that
1035 2013-06-28 09:26:24 <imd23> oh fuck, you are right. I need the anweser of that question sipa :(
1036 2013-06-28 09:26:42 <sipa> imd23: please do NOT do that
1037 2013-06-28 09:26:45 mrkent has quit (Read error: Operation timed out)
1038 2013-06-28 09:27:02 <imd23> sipa: dont do what?
1039 2013-06-28 09:27:06 <sipa> as it restrains you to forever rely on a fully indexed full blockchain
1040 2013-06-28 09:27:28 <sipa> if you need to watch payments to a set of addresses, just watch transactions as they happen
1041 2013-06-28 09:27:42 <CodeShark> perhaps watch-only wallets really should be added soon, sipa
1042 2013-06-28 09:27:47 <sipa> CodeShark: *absolutely*
1043 2013-06-28 09:27:57 <sipa> imd23: bitcoind can do that, but only if you have the private keys
1044 2013-06-28 09:28:01 <CodeShark> I don't particularly like that solution fully - but it's easy to use and better than most of what's out there
1045 2013-06-28 09:28:02 <imd23> sorry guys I don't get it...
1046 2013-06-28 09:28:06 <imd23> don't do what?
1047 2013-06-28 09:28:22 <sipa> imd23: do not try to query "give me all transactions for a given set of addresses"
1048 2013-06-28 09:28:34 <imd23> oh , but⦠is it possible?
1049 2013-06-28 09:28:34 <sipa> imd23: instead "look for transactions to a set of addresses as they happen"
1050 2013-06-28 09:28:41 <sipa> yes it's possible, but don't
1051 2013-06-28 09:28:57 <imd23> ok, I won't I promise
1052 2013-06-28 09:29:07 <imd23> may I know how? just to get the internals.
1053 2013-06-28 09:29:12 <imd23> and what's the correct way
1054 2013-06-28 09:29:34 <CodeShark> I also think a streaming API (whether the 0mq pull request or something else) is necessary
1055 2013-06-28 09:29:44 <CodeShark> the p2p protocol is too low level for most app development
1056 2013-06-28 09:29:51 <sipa> well, if you use bitcoind (which is very limited for this task), you'll need to have a wallet online with the keys in it you want to watch
1057 2013-06-28 09:30:09 <sipa> and then you can use -walletnotify to get a script executed when a payment is received
1058 2013-06-28 09:30:16 <sipa> or use the listreceivedbyaddress RPC call
1059 2013-06-28 09:30:27 <CodeShark> REST is wholely inadequate for the task of listening for incoming transactions and confirmations
1060 2013-06-28 09:30:45 <imd23> awesome. It won't work for me that ...
1061 2013-06-28 09:30:50 <imd23> so what's my alternative?
1062 2013-06-28 09:30:52 <CodeShark> so we need 0mq or websockets or something
1063 2013-06-28 09:30:56 <sipa> imd23: what won't work?
1064 2013-06-28 09:31:33 <Scrat> CodeShark: for the love of thor no websockets
1065 2013-06-28 09:31:40 <imd23> I can't think why I will need to look up for past addreses and get the transactions. But I definitely need to be notified when a transaction is made to an addrs
1066 2013-06-28 09:31:44 <CodeShark> Scrat: why not?
1067 2013-06-28 09:32:16 <CodeShark> websockets makes a streaming API accessible to a much larger set of application developers than pretty much anything else ATM
1068 2013-06-28 09:32:24 <CodeShark> for better or worse
1069 2013-06-28 09:32:43 <Scrat> pretty sure 0mq is more accessible
1070 2013-06-28 09:33:36 <CodeShark> 0mq can easily go through firewalls and run in browsers?
1071 2013-06-28 09:34:20 <Scrat> websockets were only recently standardized. though I'm guessing long term you're right
1072 2013-06-28 09:34:56 shesek has quit (Ping timeout: 256 seconds)
1073 2013-06-28 09:35:54 <imd23> so⦠you guys⦠what's my best alternative to get notified or be able to poll an RPC or any other service/server, to get a list of TX given an ADRRs
1074 2013-06-28 09:36:02 <imd23> address sorry
1075 2013-06-28 09:36:14 <CodeShark> in any case, the messaging protocol isn't really so much the issue as the fact that bitcoind lacks a high level streaming API
1076 2013-06-28 09:36:24 <CodeShark> which is a HUGE shortcoming
1077 2013-06-28 09:36:52 <BlueMatt> MICHAGOGO: eventually...but not this week probably
1078 2013-06-28 09:37:08 <CodeShark> if it had that, we wouldn't be having these discussions about getrawtransaction and listreceivedbyaddress
1079 2013-06-28 09:37:46 <MICHAGOGO> BlueMatt: Yeah, I found a way around it
1080 2013-06-28 09:39:02 agnostic98 has joined
1081 2013-06-28 09:39:21 <imd23> CodeShark: I got it.. I mean, any other alternative. Maybe an online API from a company you know or a script , whetever outside bitcoind as it's currently not possible.
1082 2013-06-28 09:39:45 <TD> what's not currently possible?
1083 2013-06-28 09:39:47 <CodeShark> I already showed you what I use, imd23 :)
1084 2013-06-28 09:40:28 <CodeShark> I added a filter and a queue to https://github.com/CodeShark/CoinClasses/blob/master/examples/listener2/listener2.cpp
1085 2013-06-28 09:40:35 <CodeShark> and a subscription mechanism
1086 2013-06-28 09:40:43 <CodeShark> but I haven't published all of that
1087 2013-06-28 09:42:13 agnostic98 has quit (Read error: Connection reset by peer)
1088 2013-06-28 09:42:13 <jaekwon> best way to dive into the bitcoind source and understand it all? just read it?
1089 2013-06-28 09:42:42 <imd23> CodeShark: awesome, I was looking into that
1090 2013-06-28 09:43:02 <imd23> CodeShark: My App is in ruby on rails I know something about cpp
1091 2013-06-28 09:43:24 <imd23> CodeShark: but i need to connect it to ruby to make it useful to me in this case
1092 2013-06-28 09:43:42 <CodeShark> right - to really consider this tool complete I'd have to expose an API that can be called from other languages
1093 2013-06-28 09:44:11 Guest79337 has quit (Read error: Connection reset by peer)
1094 2013-06-28 09:44:20 reneg has joined
1095 2013-06-28 09:44:31 <CodeShark> pub/sub with filtering options, I suppose
1096 2013-06-28 09:44:37 reneg is now known as Guest9044
1097 2013-06-28 09:44:46 <CodeShark> and a since timestamp
1098 2013-06-28 09:45:11 anarchy5 has quit (Ping timeout: 276 seconds)
1099 2013-06-28 09:45:15 <CodeShark> enter block height, a set of addresses, etc...
1100 2013-06-28 09:46:17 <CodeShark> I've been debating whether to continue developing this tool as a separate process or whether to try to build in these capabilities to bitcoind itself
1101 2013-06-28 09:46:20 <imd23> i am trying to compile the listener2 sample
1102 2013-06-28 09:46:41 <imd23> it would be awesome to be IN bitcoind
1103 2013-06-28 09:47:12 <imd23> i get
1104 2013-06-28 09:47:13 <imd23> "../../src/BigInt.h:150:9: error: use of undeclared identifier 'OPENSSL_free'"
1105 2013-06-28 09:47:24 <CodeShark> oh, right - it's an OPENSSL version issue
1106 2013-06-28 09:47:33 <CodeShark> simple fix...
1107 2013-06-28 09:47:57 <CodeShark> #define __NO_OPENSSL_FREE as your first include
1108 2013-06-28 09:48:12 <CodeShark> I mean before your first include
1109 2013-06-28 09:48:19 <imd23> yeah
1110 2013-06-28 09:48:21 <imd23> lets try
1111 2013-06-28 09:48:29 <imd23> mmm...
1112 2013-06-28 09:48:31 <imd23> no?
1113 2013-06-28 09:48:41 <imd23> like this
1114 2013-06-28 09:48:42 <imd23> #define __NO_OPENSSL_FREE
1115 2013-06-28 09:48:43 <imd23> #include <CoinNodeAbstractListener.h>
1116 2013-06-28 09:48:44 <imd23> ?
1117 2013-06-28 09:48:45 <CodeShark> yes
1118 2013-06-28 09:48:50 <imd23> didn't work
1119 2013-06-28 09:48:57 <CodeShark> what error did you get?
1120 2013-06-28 09:49:01 <imd23> oh wait
1121 2013-06-28 09:49:03 <imd23> dif error
1122 2013-06-28 09:49:08 <imd23> ../../src/BigInt.h:153:43: warning: 'BN_hex2bn' is deprecated [-Wdeprecated-declarations]
1123 2013-06-28 09:49:28 <imd23> and the same again
1124 2013-06-28 09:49:29 <imd23> ../../src/BigInt.h:160:9: error: use of undeclared identifier 'OPENSSL_free'
1125 2013-06-28 09:50:03 <CodeShark> oh, you need to define it for the entire make
1126 2013-06-28 09:50:18 <imd23> oh
1127 2013-06-28 09:50:19 <imd23> ok
1128 2013-06-28 09:50:39 <imd23> how?
1129 2013-06-28 09:50:40 <imd23> :p
1130 2013-06-28 09:51:06 <matjeh> BN_hex2bn is deprecated since when?
1131 2013-06-28 09:51:16 <imd23> I have no idea what it is :p
1132 2013-06-28 09:51:22 <imd23> I am just compiling haha
1133 2013-06-28 09:51:23 <imd23> sorry
1134 2013-06-28 09:51:37 <imd23> if I can help, just tell me
1135 2013-06-28 09:51:50 <CodeShark> I believe it's -D__NO_OPENSSL_FREE=1 in gcc
1136 2013-06-28 09:52:48 <matjeh> imd23: what version of openssl are you on? (run "openssl version")
1137 2013-06-28 09:52:57 <imd23> PRECATED_IN_MAC_OS_X_VERSION_10_7_AND_LATER;
1138 2013-06-28 09:52:58 <imd23> ^
1139 2013-06-28 09:52:59 <imd23> ../../src/CoinNodeSocket.cpp:303:19: error: use of undeclared identifier 'CLOCK_REALTIME'
1140 2013-06-28 09:53:00 <imd23> clock_gettime(CLOCK_REALTIME, &ts);
1141 2013-06-28 09:53:01 <imd23> ^
1142 2013-06-28 09:53:08 <imd23> OpenSSL 0.9.8x 10 May 2012
1143 2013-06-28 09:53:16 stretchwarren has quit (Ping timeout: 268 seconds)
1144 2013-06-28 09:57:05 <CodeShark> I should really replace that code with something more portable
1145 2013-06-28 09:58:00 <imd23> :p
1146 2013-06-28 09:58:05 <imd23> so....
1147 2013-06-28 09:58:20 <imd23> what's going wrong here?
1148 2013-06-28 09:58:26 <imd23> I want to compile it :)
1149 2013-06-28 09:58:35 <CodeShark> researching it
1150 2013-06-28 09:58:38 <CodeShark> gimme a moment :)
1151 2013-06-28 09:58:45 <imd23> oh yeah no problem
1152 2013-06-28 09:58:45 <imd23> haha
1153 2013-06-28 09:58:47 <imd23> sorry
1154 2013-06-28 09:59:09 <CodeShark> I'm not crazy about boost::posix_time but it might make it more portable
1155 2013-06-28 10:01:58 msvb-lab has joined
1156 2013-06-28 10:03:07 <imd23> so I change that fund?
1157 2013-06-28 10:03:08 <imd23> func
1158 2013-06-28 10:04:10 <CodeShark> I should probably just replace all that code in CoinNodeSocket with boost
1159 2013-06-28 10:04:20 <CodeShark> including the thread stuff
1160 2013-06-28 10:05:03 <imd23> ok
1161 2013-06-28 10:07:52 datagutt has quit (Read error: Operation timed out)
1162 2013-06-28 10:08:14 <CodeShark> in any case, sounds like you've got an incompatible version of time.h
1163 2013-06-28 10:09:07 <imd23> ok...
1164 2013-06-28 10:09:16 <imd23> so...
1165 2013-06-28 10:09:20 <imd23> time.h is part of?
1166 2013-06-28 10:09:59 <CodeShark> GNU C library
1167 2013-06-28 10:11:16 daybyter has joined
1168 2013-06-28 10:11:35 <imd23> weird.
1169 2013-06-28 10:11:45 <imd23> how do they do this? http://blockchain.info/api/api_websocket
1170 2013-06-28 10:11:51 <imd23> what service are they running?
1171 2013-06-28 10:12:06 <CodeShark> he's running a custom build of bitcoind
1172 2013-06-28 10:12:34 <CodeShark> or at least that's what he was running last time I spoke to him
1173 2013-06-28 10:13:00 <MICHAGOGO> I'm not sure I understand the reasoning behind the testnet reset-to-1 logic
1174 2013-06-28 10:13:14 <CodeShark> so that nobody can run up the difficulty and then leave, MICHAGOGO
1175 2013-06-28 10:13:22 <MICHAGOGO> Not what I meant
1176 2013-06-28 10:13:34 <imd23> CodeShark: is it open source or did he keep it private?
1177 2013-06-28 10:13:37 <MICHAGOGO> How come, if there's no block for 20 mins and the difficulty is reset to 1, that reset is only in effect for one single block?
1178 2013-06-28 10:13:51 <CodeShark> imd23: I think it's private as far as I know
1179 2013-06-28 10:13:52 <MICHAGOGO> Why not simply have it build back up to the 10-minute point?
1180 2013-06-28 10:14:20 <CodeShark> MICHAGOGO: I think that's a bug
1181 2013-06-28 10:14:43 <MICHAGOGO> Also: the current testnet height is a palindrome
1182 2013-06-28 10:14:49 <CodeShark> imd23: the problem with custom builds is merges
1183 2013-06-28 10:14:53 <MICHAGOGO> ;;seen sipa
1184 2013-06-28 10:14:53 <gribble> sipa was last seen in #bitcoin-dev 43 minutes and 55 seconds ago: <sipa> imd23: what won't work?
1185 2013-06-28 10:15:32 <CodeShark> imd23: IMHO, the only workable solutions are either to have a streaming API built into bitcoind itself or to run the streaming server as a separate process that connects to bitcoind via p2p
1186 2013-06-28 10:16:04 abrkn has joined
1187 2013-06-28 10:16:11 <CodeShark> by "built into bitcoind" I mean as part of the official build
1188 2013-06-28 10:16:51 <CodeShark> otherwise it's too much of a pain in the ass to merge each time there's a new release
1189 2013-06-28 10:17:02 Xeno-Genesis has joined
1190 2013-06-28 10:18:05 Xeno-Genesis has quit (Client Quit)
1191 2013-06-28 10:18:19 Xeno-Genesis has joined
1192 2013-06-28 10:19:01 <CodeShark> imd23: I'll rework CoinNodeSocket to be more portable - but won't be able to have that done until probably sometime next week
1193 2013-06-28 10:20:07 <CodeShark> I actually think having the filtering/queueing handled by a separate process is in some ways superior architecturally - although of course, if these things were built into bitcoind itself it would make deployment simpler
1194 2013-06-28 10:20:30 catcowllama has joined
1195 2013-06-28 10:20:30 catcowllama has quit (Changing host)
1196 2013-06-28 10:20:30 catcowllama has joined
1197 2013-06-28 10:21:11 catcow has quit (Ping timeout: 264 seconds)
1198 2013-06-28 10:21:16 <CodeShark> bitcoind already has plenty of bloat that is significantly heavier than a filtering/queueing/streaming service
1199 2013-06-28 10:21:46 nomailing has joined
1200 2013-06-28 10:21:54 <CodeShark> I'd like to see it become a streamlined, minimal verification/relay agent
1201 2013-06-28 10:22:11 <CodeShark> and have these other services run as either separate processes or dynamically loadable modules
1202 2013-06-28 10:22:37 <CodeShark> especially the wallet/RPC stuff and the GUI
1203 2013-06-28 10:22:56 <CodeShark> but that's not going to happen anytime soon :p
1204 2013-06-28 10:23:14 <Diablo-D3> ;;ticker
1205 2013-06-28 10:23:14 <gribble> MtGox BTCUSD ticker | Best bid: 96.86984, Best ask: 97.00000, Bid-ask spread: 0.13016, Last trade: 97.07347, 24 hour volume: 34601.36005370, 24 hour low: 95.70000, 24 hour high: 103.12300, 24 hour vwap: 99.41184
1206 2013-06-28 10:23:53 <CodeShark> having said all that, a lightweight filtering/queueing/streaming agent is certainly doable
1207 2013-06-28 10:24:26 <CodeShark> with perhaps even smaller footprint than some of the unused features of bitcoind like the cpuminer :)
1208 2013-06-28 10:24:26 stretchwarren has joined
1209 2013-06-28 10:26:10 imd23 has quit (Read error: Connection reset by peer)
1210 2013-06-28 10:26:21 tcatm has quit (Ping timeout: 252 seconds)
1211 2013-06-28 10:26:54 imd23 has joined
1212 2013-06-28 10:27:17 <MICHAGOGO> Is it just me, or is everyone's testnet mempool empty?
1213 2013-06-28 10:27:43 paracyst has quit ()
1214 2013-06-28 10:27:48 justusranvier has quit (Ping timeout: 268 seconds)
1215 2013-06-28 10:32:33 catcow has joined
1216 2013-06-28 10:32:33 catcow has quit (Changing host)
1217 2013-06-28 10:32:33 catcow has joined
1218 2013-06-28 10:33:47 catcowllama has quit (Ping timeout: 264 seconds)
1219 2013-06-28 10:34:48 justusranvier has joined
1220 2013-06-28 10:34:48 justusranvier has quit (Changing host)
1221 2013-06-28 10:34:48 justusranvier has joined
1222 2013-06-28 10:37:31 tcatm has joined
1223 2013-06-28 10:47:30 RoboTeddy has quit (Remote host closed the connection)
1224 2013-06-28 10:49:06 fanquake has left ()
1225 2013-06-28 10:50:15 nowan_ has quit (Read error: Operation timed out)
1226 2013-06-28 10:50:37 nowan has joined
1227 2013-06-28 10:53:33 andyh2 has quit (Quit: Leaving...)
1228 2013-06-28 11:00:29 ericmuys_ has quit (Ping timeout: 268 seconds)
1229 2013-06-28 11:02:11 macboz_ has quit (Quit: This computer has gone to sleep)
1230 2013-06-28 11:05:47 _milone has joined
1231 2013-06-28 11:05:57 yubrew has quit (Remote host closed the connection)
1232 2013-06-28 11:07:07 yubrew has joined
1233 2013-06-28 11:07:12 yubrew has quit (Remote host closed the connection)
1234 2013-06-28 11:08:58 milone_ has quit (Ping timeout: 246 seconds)
1235 2013-06-28 11:09:01 ericmuyser has joined
1236 2013-06-28 11:09:21 fishfish has joined
1237 2013-06-28 11:10:05 agnostic98 has joined
1238 2013-06-28 11:12:30 agnostic98 has quit (Read error: Connection reset by peer)
1239 2013-06-28 11:12:31 melvster has quit (Ping timeout: 240 seconds)
1240 2013-06-28 11:13:52 ericmuyser has quit (Ping timeout: 246 seconds)
1241 2013-06-28 11:23:05 mrkent has joined
1242 2013-06-28 11:23:05 mrkent has quit (Changing host)
1243 2013-06-28 11:23:05 mrkent has joined
1244 2013-06-28 11:24:32 viperhr1 has joined
1245 2013-06-28 11:25:15 melvster has joined
1246 2013-06-28 11:26:27 rainworm has quit ()
1247 2013-06-28 11:28:09 viperhr has quit (Ping timeout: 248 seconds)
1248 2013-06-28 11:29:13 one_zero has quit ()
1249 2013-06-28 11:30:49 zer0def has joined
1250 2013-06-28 11:31:57 eian has joined
1251 2013-06-28 11:33:48 BW^- has joined
1252 2013-06-28 11:34:10 <BW^-> td: where in the code is send of the version handshake?
1253 2013-06-28 11:35:22 abrkn\ has joined
1254 2013-06-28 11:35:51 datagutt has joined
1255 2013-06-28 11:37:24 abrkn has quit (Read error: Operation timed out)
1256 2013-06-28 11:38:01 <BW^-> bluematt: ^ ?
1257 2013-06-28 11:39:18 tcatm has quit (Ping timeout: 252 seconds)
1258 2013-06-28 11:43:25 tcatm has joined
1259 2013-06-28 11:44:08 <nsh> BW^-, grep for PROTOCOL_VERSION
1260 2013-06-28 11:44:21 daybyter has quit (Quit: Konversation terminated!)
1261 2013-06-28 11:44:33 <nsh> net.cpp: PushMessage("version", PROTOCOL_VERSION, nLocalServices, nTime, addrYou, addrMe,
1262 2013-06-28 11:44:54 Thepok has quit (Read error: Connection reset by peer)
1263 2013-06-28 11:46:53 <BW^-> nsh: I'm talking BitcoinJ . you too?
1264 2013-06-28 11:47:06 <nsh> oh, no, sorry
1265 2013-06-28 11:47:17 <BW^-> nsh: anyhow thanks for clarifying :))
1266 2013-06-28 11:47:22 <nsh> np
1267 2013-06-28 11:47:25 <BW^-> how many KLOCS are BitcoinD now?
1268 2013-06-28 11:47:30 <BW^-> is
1269 2013-06-28 11:48:47 <nsh> ~40 iirc
1270 2013-06-28 11:48:50 <nsh> (may be wrong)
1271 2013-06-28 11:49:09 Thepok has joined
1272 2013-06-28 11:49:24 <BW^-> 40? hm
1273 2013-06-28 11:51:05 mrkent has quit (Ping timeout: 248 seconds)
1274 2013-06-28 11:52:02 <nsh> (not including libs i presume)
1275 2013-06-28 11:52:36 Jackneill has quit (Remote host closed the connection)
1276 2013-06-28 11:52:42 <nsh> "21:51 Ferroh are you saying that leveldb will be imported into the source tree? / 21:51 gmaxwell Yes. / 21:53 sipa it's around 18k lines of code / 21:53 sipa bitcoin itself is twice that"
1277 2013-06-28 11:52:48 <nsh> - http://bitcoinstats.com/irc/bitcoin-dev/logs/2012/10/04
1278 2013-06-28 11:53:46 Thepok has quit (Ping timeout: 256 seconds)
1279 2013-06-28 11:53:53 fishfish has quit (Read error: Connection reset by peer)
1280 2013-06-28 11:54:25 <nsh> the variable is called VersionMessage in bitcoinj, BW^-
1281 2013-06-28 11:55:11 <nsh> ./core/src/main/java/com/google/bitcoin/core/TCPNetworkConnection.java
1282 2013-06-28 11:55:14 <BW^-> nsh: yeah i know, i'm just interested where it's actually sent
1283 2013-06-28 11:55:31 <BW^-> nsh: interesting, so BitcoinD is 37klocks (!) - BitcoinJ is that much or a bit more.
1284 2013-06-28 11:55:53 <nsh> that was last november, so it's probably about a few k more now
1285 2013-06-28 11:56:11 <BW^-> nsh: i just checked.
1286 2013-06-28 11:56:17 <BW^-> nsh: is all BitcoinD's logics in the src/* files ?
1287 2013-06-28 11:56:51 <BW^-> so let's see, there's a share directory but that seems to be UI-related things
1288 2013-06-28 11:57:09 <nsh> afaik everything's within src
1289 2013-06-28 11:59:50 <BW^-> ok
1290 2013-06-28 12:00:05 <BW^-> nsh: and not in its subdirs, with the exception of the json/ directory , that contains some things too right?
1291 2013-06-28 12:00:21 <BW^-> nsh: how does BTCD use OS threads, does it use any at all?
1292 2013-06-28 12:00:52 <nsh> eek, i am not a dev :)
1293 2013-06-28 12:00:54 <BW^-> aha
1294 2013-06-28 12:00:56 <BW^-> any dev around? :))
1295 2013-06-28 12:01:43 <nsh> q@Qbox:/media/truecrypt2/bitcoin/bitcoin$ git diff --stat 4b825dc642cb6eb9a060e54bf8d69288fbee4904 | tail -1
1296 2013-06-28 12:01:43 <nsh> 581 files changed, 232719 insertions(+)
1297 2013-06-28 12:02:02 <nsh> but that's will leveldb and boost? included in the source tree
1298 2013-06-28 12:02:07 <nsh> *with
1299 2013-06-28 12:02:29 <nsh> ("This shows the differences from the empty tree to your current working tree. Which happens to count all lines in your current working tree.") - http://stackoverflow.com/questions/4822471/count-number-of-lines-in-a-git-repository
1300 2013-06-28 12:02:44 <nsh> also that includes all config, etc. etc.
1301 2013-06-28 12:04:23 xnyhps has quit (Ping timeout: 264 seconds)
1302 2013-06-28 12:09:02 datagutt has quit (Quit: kthxbai)
1303 2013-06-28 12:09:53 datagutt has joined
1304 2013-06-28 12:11:30 grau has joined
1305 2013-06-28 12:12:06 sturles has quit (Quit: I'll be back)
1306 2013-06-28 12:15:43 justusranvier has quit (Ping timeout: 268 seconds)
1307 2013-06-28 12:16:27 sturles has joined
1308 2013-06-28 12:16:47 sturles has quit (Changing host)
1309 2013-06-28 12:16:47 sturles has joined
1310 2013-06-28 12:17:36 CodeShark has quit (Remote host closed the connection)
1311 2013-06-28 12:19:20 yubrew has joined
1312 2013-06-28 12:20:24 eian has quit (Ping timeout: 246 seconds)
1313 2013-06-28 12:20:44 milone_ has joined
1314 2013-06-28 12:22:56 justusranvier has joined
1315 2013-06-28 12:24:47 _milone has quit (Ping timeout: 264 seconds)
1316 2013-06-28 12:30:55 <BlueMatt> BW^-: peer version handshake or payment channel version handshake?
1317 2013-06-28 12:31:25 Xeno-Genesis has quit (Remote host closed the connection)
1318 2013-06-28 12:31:51 <TD> BW^-: look in the Peer class
1319 2013-06-28 12:32:53 JZavala has quit (Ping timeout: 276 seconds)
1320 2013-06-28 12:33:39 dansmithbtc2 has joined
1321 2013-06-28 12:34:39 dansmithbtc2 has quit (Changing host)
1322 2013-06-28 12:34:39 dansmithbtc2 has joined
1323 2013-06-28 12:35:01 Transisto has quit (Ping timeout: 252 seconds)
1324 2013-06-28 12:35:16 Transisto has joined
1325 2013-06-28 12:37:14 tcatm has quit (Ping timeout: 252 seconds)
1326 2013-06-28 12:40:19 <BW^-> td: where/what row?
1327 2013-06-28 12:40:44 <BW^-> bluematt: wait, so when a new outbound TCP connection is established, there's a "handshake" in the form of BitcoinJ sending a version message and that's all, right?
1328 2013-06-28 12:41:01 <BlueMatt> control-f: "version"
1329 2013-06-28 12:41:01 <BW^-> bluematt: it's the location where the send of that one is committed i'm looking for :)
1330 2013-06-28 12:41:15 <BW^-> bluematt: yeah i saw it all over, but nowhere like "connection.send(version);"
1331 2013-06-28 12:41:36 <BW^-> would be happy to understand where exactly that row is :)
1332 2013-06-28 12:41:52 tcatm has joined
1333 2013-06-28 12:45:38 <BW^-> any BitcoinD dev around?
1334 2013-06-28 12:46:18 <TD> BW^-: TCPNetworkConnection and search for write(channel, myVersionMessage);
1335 2013-06-28 12:46:25 <TD> although that part of the code is kind of a mess
1336 2013-06-28 12:50:53 <sipa> BW^-: yes
1337 2013-06-28 12:51:11 <BW^-> sipa: cool. what's BitcoinD's threading model?
1338 2013-06-28 12:51:14 <BW^-> sipa: one OS thread only?
1339 2013-06-28 12:51:30 <sipa> oh no
1340 2013-06-28 12:51:33 <sipa> there are many threads
1341 2013-06-28 12:51:51 <sipa> the network thread and the message handler thread are the most important i guess
1342 2013-06-28 12:51:58 <sipa> but there are RPC threads
1343 2013-06-28 12:52:01 abrkn has joined
1344 2013-06-28 12:52:10 <sipa> and a reindex thread if you're reindexing
1345 2013-06-28 12:52:18 <MICHAGOGO> I'm not sure I understand the reasoning behind the testnet reset-to-1 logic. How come, if there's no block for 20 mins and the difficulty is reset to 1, that reset is only in effect for one single block? Why not simply have it build back up to the 10-minute point?
1346 2013-06-28 12:52:48 <sipa> and signature verification threads
1347 2013-06-28 12:53:38 <sipa> and a thread that opens new connections
1348 2013-06-28 12:53:52 <sipa> and a thread that occasionally dumps the address database to disk
1349 2013-06-28 12:54:15 <sipa> and a thread that fetches DNS seeds (only at startup)
1350 2013-06-28 12:55:08 <BW^-> sipa: aha interesting
1351 2013-06-28 12:55:19 <BW^-> sipa: how do they speak with each other, message passing with heap-allocated message structures?
1352 2013-06-28 12:55:50 rdymac has joined
1353 2013-06-28 12:55:50 rdymac has quit (Changing host)
1354 2013-06-28 12:55:50 rdymac has joined
1355 2013-06-28 12:55:52 <sipa> they synchronize on heap data structures, indeed
1356 2013-06-28 12:56:10 <sipa> the network thread will receive to per-peer buffers, and send data from per-peer buffers
1357 2013-06-28 12:56:25 abrkn\ has quit (Ping timeout: 268 seconds)
1358 2013-06-28 12:56:53 <sipa> and the message handler thread iterates over nodes, processing messages in those buffers - typically resulting in more messages being pushed to that (and other nodes') buffers
1359 2013-06-28 12:57:02 tcatm has quit (Ping timeout: 252 seconds)
1360 2013-06-28 12:58:55 xnyhps has joined
1361 2013-06-28 12:59:38 rdymac has quit (Client Quit)
1362 2013-06-28 13:00:27 egis has joined
1363 2013-06-28 13:00:34 grau has quit (Remote host closed the connection)
1364 2013-06-28 13:00:36 <BW^-> sipa: hm aha
1365 2013-06-28 13:00:49 <BW^-> sipa: wait, i want to understand this at a close-mechanistic levle -
1366 2013-06-28 13:01:23 <BW^-> sipa: so, btcd initiates a new connection with another host. i should be able to find that by looking for connect() in the code.
1367 2013-06-28 13:01:26 <BW^-> sipa: then, what happens then -
1368 2013-06-28 13:01:41 <BW^-> sipa: the connection structure including the socket, are added in a global list of active connections?
1369 2013-06-28 13:01:47 <sipa> indeed
1370 2013-06-28 13:01:51 <sipa> vNodes
1371 2013-06-28 13:01:52 <BW^-> sipa: and.. there's *one* networking thread that select():s on all those for new data?
1372 2013-06-28 13:01:57 <sipa> indeed
1373 2013-06-28 13:01:58 <BW^-> sipa: where are those defined?
1374 2013-06-28 13:02:01 <sipa> net.cpp
1375 2013-06-28 13:02:19 <BW^-> sipa: noted. and then, the networking thread will download data up to it detected it got a full message from the other host, and then decode it into a message structure?
1376 2013-06-28 13:02:24 <BW^-> sipa: is that done by the networking thread too?
1377 2013-06-28 13:02:43 hnz has quit (Ping timeout: 246 seconds)
1378 2013-06-28 13:03:03 <sipa> the network thread packetizes incoming data into messages, which are pushed on a queue to be processed
1379 2013-06-28 13:03:20 <sipa> and the message handler thread, when it processes that node, will process those
1380 2013-06-28 13:03:21 <BW^-> sipa: aha, so the byte array of received bytes for an individual message are put on a queue to be processed?
1381 2013-06-28 13:03:22 grau has joined
1382 2013-06-28 13:03:33 <BW^-> sipa: aha noted. where's the message handler thread defined?
1383 2013-06-28 13:03:34 <sipa> BW^-: go look at the code
1384 2013-06-28 13:03:44 <BW^-> sipa: yes i'm doing :)
1385 2013-06-28 13:03:53 <MICHAGOGO> Does ;;later tell hold multiple messages per recipient?
1386 2013-06-28 13:03:59 <BW^-> sipa: this kind of overview things tend to take Lots of time to figure out, this is why i'm asking
1387 2013-06-28 13:04:04 <sipa> the message handler thread is started in net.cpp, but mostly lives in main.cpp (ProcessMessages and SendMessages)
1388 2013-06-28 13:04:41 <BW^-> sipa: so ProcessMessages receives the individual message byte vectors and does all handling of them - aha ok
1389 2013-06-28 13:04:57 <BW^-> sipa: SendMessages does what, packetize from internal structure to byte array and sends on to net module?
1390 2013-06-28 13:05:04 <HM2> BW^-, this is fairly typical networking behaviour, nothing fancy
1391 2013-06-28 13:05:09 <MICHAGOGO> sipa: Did you just get message1, message2, or both?
1392 2013-06-28 13:05:29 <sipa> BW^-: no, it looks at various node state structure to see if there are things to be sent
1393 2013-06-28 13:05:39 <sipa> MICHAGOGO: i see both message1 message2
1394 2013-06-28 13:05:47 <MICHAGOGO> Okay, thanks
1395 2013-06-28 13:06:00 fishfish has joined
1396 2013-06-28 13:06:32 <BW^-> sipa: aha that's what SendMessages does - noted
1397 2013-06-28 13:06:45 <sipa> BW^-: it's more a constructmessages than sendmessages
1398 2013-06-28 13:07:42 saulimus has quit (Quit: saulimus)
1399 2013-06-28 13:07:51 <BW^-> sipa: aha, so SendMessages traverses internal structures to figure out if messages should be sent on the network, and if there are, it constructs those and passes them on in byte array form to the networking thread
1400 2013-06-28 13:08:14 <sipa> BW^-: it puts them on the queue of messages-to-be-sent for that peer
1401 2013-06-28 13:08:39 <sipa> but there are other ways through which messages can end up there
1402 2013-06-28 13:08:44 hnz has joined
1403 2013-06-28 13:09:03 <sipa> for example if a peer sends a message that needs a response, it's ProcessMessages that puts the response there, typically
1404 2013-06-28 13:10:03 <BW^-> sipa: is that pretty much the only exception?
1405 2013-06-28 13:10:22 tcatm has joined
1406 2013-06-28 13:10:23 tcatm has quit (Changing host)
1407 2013-06-28 13:10:23 tcatm has joined
1408 2013-06-28 13:10:34 <BW^-> sipa: so it's the message handler thread only that runs both ProcessMessages and SendMessages
1409 2013-06-28 13:11:02 abrkn has quit ()
1410 2013-06-28 13:11:04 <BW^-> sipa: the thread that opens new connections is another one than the networking thread, and that one does the outbound connect():s?
1411 2013-06-28 13:11:10 abrkn has joined
1412 2013-06-28 13:11:55 agnostic98 has joined
1413 2013-06-28 13:14:12 <BW^-> what's the point with the signature verification threads?
1414 2013-06-28 13:14:16 o3u has quit (Ping timeout: 246 seconds)
1415 2013-06-28 13:14:27 agnostic98 has quit (Read error: Connection reset by peer)
1416 2013-06-28 13:14:33 MiningBuddy- has joined
1417 2013-06-28 13:14:38 MiningBuddy has quit (Remote host closed the connection)
1418 2013-06-28 13:14:49 <TD> BW^-: more performance
1419 2013-06-28 13:14:50 <BW^-> including, why are threads needed for that, because it's too cpu-expensive to do in the message handler thread?
1420 2013-06-28 13:14:54 <BW^-> mhm
1421 2013-06-28 13:15:21 <TD> if you have 4 cores, you need 4 threads to use all of them obviously
1422 2013-06-28 13:15:57 <BW^-> mhm. where in the code is the verification thread?
1423 2013-06-28 13:16:38 Blitzboom has quit (Read error: Operation timed out)
1424 2013-06-28 13:16:50 Blitzboom has joined
1425 2013-06-28 13:17:03 rdymac has joined
1426 2013-06-28 13:17:03 rdymac has quit (Changing host)
1427 2013-06-28 13:17:03 rdymac has joined
1428 2013-06-28 13:17:12 <sipa> BW^-: if you don't have separate signature verification threads, you never really load more than a single core
1429 2013-06-28 13:17:29 <sipa> in particular when doing an import or reindex
1430 2013-06-28 13:17:41 <sipa> as you're not bound by networm overhead in that case
1431 2013-06-28 13:17:58 Blitzboom has quit (Disconnected by services)
1432 2013-06-28 13:17:59 Blitzboom_ has joined
1433 2013-06-28 13:18:20 <BW^-> mhm
1434 2013-06-28 13:18:28 daemon has joined
1435 2013-06-28 13:18:31 <sipa> it's pretty essential to speed up chain processing
1436 2013-06-28 13:18:48 <BW^-> sipa: basically bitcoind's work is a function of network input (originally seeded by its peer discovery) and RPC commands, right?
1437 2013-06-28 13:18:49 <BW^-> that's all?
1438 2013-06-28 13:19:03 <sipa> i guess
1439 2013-06-28 13:19:07 <BW^-> cool.
1440 2013-06-28 13:19:16 <BW^-> all the database handling, where is that taken care of?
1441 2013-06-28 13:19:24 Blitzboom_ has left ()
1442 2013-06-28 13:19:26 <BW^-> is that like an aspect of the message handler thread?
1443 2013-06-28 13:19:29 <BW^-> 's work
1444 2013-06-28 13:19:35 <sipa> or the import/reindex thread
1445 2013-06-28 13:19:41 <BW^-> aha
1446 2013-06-28 13:19:53 <BW^-> sipa: what's the reindexing about - what indexes does btcd maintain?
1447 2013-06-28 13:20:03 <MICHAGOGO> BW^-: An index of blocks on disk
1448 2013-06-28 13:20:20 <MICHAGOGO> As in, where each block is in the files, etc
1449 2013-06-28 13:20:24 <sipa> the term actually dates back to a time when the only database was a block/transaction index
1450 2013-06-28 13:20:30 <MICHAGOGO> Also utxos?
1451 2013-06-28 13:20:32 <sipa> today, there is no transaction index anymore
1452 2013-06-28 13:20:40 <sipa> but there is a separate database with unspent transaction outputs
1453 2013-06-28 13:20:45 <sipa> which is technically not an index
1454 2013-06-28 13:20:48 <BW^-> michagogo: as in, a dataset where key = block identifier and value = block data?
1455 2013-06-28 13:20:48 daemon has left ()
1456 2013-06-28 13:20:48 <sipa> (it's a copy)
1457 2013-06-28 13:21:02 <MICHAGOGO> BW^-: No, it's like a directory
1458 2013-06-28 13:21:12 <MICHAGOGO> Which .dat file the blocks are in
1459 2013-06-28 13:21:15 <sipa> BW^-: the block index is a key-value store where the keys are block hashes, and the values contain the block header + where on disk to find it
1460 2013-06-28 13:21:19 dustjn has joined
1461 2013-06-28 13:21:19 <MICHAGOGO> And where in that file
1462 2013-06-28 13:22:15 <BW^-> sipa: ah right, so this brings to: what are all the databases in bitcoind? so, the block database with key=block hash and value=block header + (effectively) block data . and then an unspent transactions index. any more?
1463 2013-06-28 13:22:45 <sipa> there's 3 datasets that are maintained: the blocks (raw files on disk), the block index (blockhash -> position on disk), and the UTXO set (txid -> list of unspent outputs for that tx)
1464 2013-06-28 13:23:01 <sipa> there is an optional transaction index (txid -> position on disk) as well
1465 2013-06-28 13:23:06 <sipa> (see the -txindex option)
1466 2013-06-28 13:23:11 <BW^-> aha
1467 2013-06-28 13:23:17 <BW^-> why not just inline the block data in the block index?
1468 2013-06-28 13:23:26 <sipa> because it's append only data
1469 2013-06-28 13:23:29 <MICHAGOGO> BW^-: That defeats the point of an index
1470 2013-06-28 13:23:48 <sipa> it's much faster to just write it once, than to incorporate it in a database that is 100 times smaller
1471 2013-06-28 13:25:51 gaantr2 has quit (Ping timeout: 256 seconds)
1472 2013-06-28 13:26:02 darkee has quit (Ping timeout: 240 seconds)
1473 2013-06-28 13:26:35 gaantr2 has joined
1474 2013-06-28 13:26:58 gaantr2 is now known as Guest21718
1475 2013-06-28 13:27:13 darkee has joined
1476 2013-06-28 13:30:11 tcatm has quit (Ping timeout: 264 seconds)
1477 2013-06-28 13:34:43 grau has quit (Read error: Connection reset by peer)
1478 2013-06-28 13:35:18 grau has joined
1479 2013-06-28 13:35:42 wiretapped has quit (Ping timeout: 240 seconds)
1480 2013-06-28 13:36:48 wiretapped has joined
1481 2013-06-28 13:38:35 valparaiso has quit (Ping timeout: 264 seconds)
1482 2013-06-28 13:38:41 Eliel has quit (Ping timeout: 256 seconds)
1483 2013-06-28 13:40:02 Subo1978_ has quit (Ping timeout: 240 seconds)
1484 2013-06-28 13:40:23 Subo1978 has joined
1485 2013-06-28 13:41:04 <jgarzik> mornin'
1486 2013-06-28 13:41:12 <nsh> o/
1487 2013-06-28 13:41:15 metabyte has quit ()
1488 2013-06-28 13:41:21 <TD> hey
1489 2013-06-28 13:42:58 Eliel has joined
1490 2013-06-28 13:43:00 agnostic98 has joined
1491 2013-06-28 13:44:31 agnostic98 has quit (Read error: Connection reset by peer)
1492 2013-06-28 13:44:34 signor777 has joined
1493 2013-06-28 13:46:05 o3u has joined
1494 2013-06-28 13:46:27 ericmuyser has joined
1495 2013-06-28 13:46:59 kyledrake has joined
1496 2013-06-28 13:47:05 valparaiso_afk has joined
1497 2013-06-28 13:47:27 valparaiso_afk is now known as valparaiso
1498 2013-06-28 13:48:38 <BW^-> sipa: so is reindexing, a reorganization of all the block data and corresponding update of the block database?
1499 2013-06-28 13:48:43 <BW^-> *block index
1500 2013-06-28 13:49:58 <BW^-> The thread that occasionally dumps the address database to disk - what address database is that?
1501 2013-06-28 13:51:50 <MICHAGOGO> BW^-: That's the addresses of your peers
1502 2013-06-28 13:51:58 <MICHAGOGO> peers.dat, IIRC
1503 2013-06-28 13:52:14 <MICHAGOGO> For bootstrapping on next start
1504 2013-06-28 13:54:12 <sipa> BW^-: reindexing basically means throw away the block index and the UTXO database, and rebuild them from scratch using the block data on disk
1505 2013-06-28 13:55:18 taha has joined
1506 2013-06-28 13:57:27 <BW^-> MICHAGOGO: aah. so that would be just set of hostname/ip + port number right?
1507 2013-06-28 13:57:55 <BW^-> sipa: aha. so today, levelDB is used for the block index and UTXO DB but not for the block data?
1508 2013-06-28 13:58:00 <sipa> BW^-: correct
1509 2013-06-28 13:58:12 signor777 has left ()
1510 2013-06-28 13:59:13 omnibrain has joined
1511 2013-06-28 13:59:44 <BW^-> sipa: cool. where's the code that does block database reading and writing , and reindexing (i.e. throw away block idx& utxo db and reconstruct from block db)
1512 2013-06-28 14:01:02 <BW^-> db.cpp I guess.
1513 2013-06-28 14:01:36 <BW^-> hm, it's starting to make sense. :)
1514 2013-06-28 14:01:47 <sipa> db.cpp is BDB
1515 2013-06-28 14:01:54 <sipa> which is only used for the wallet
1516 2013-06-28 14:02:03 <sipa> the leveldb glue code is in leveldb.cpp
1517 2013-06-28 14:02:17 <BW^-> sipa: aha. and the code for handling block data reading and writing?
1518 2013-06-28 14:02:20 <sipa> the read/write code for the UTXO and block index databases is in txdb.cpp
1519 2013-06-28 14:02:35 <BW^-> sipa: btw, you said append only - so once something ha been appended to the block data, it is never removed?
1520 2013-06-28 14:02:41 <BW^-> and never reordered?
1521 2013-06-28 14:02:43 <sipa> correct
1522 2013-06-28 14:03:01 justusranvier has quit (Ping timeout: 268 seconds)
1523 2013-06-28 14:03:10 <sipa> the UTXO set is abstracted via a class CCoinsView in main
1524 2013-06-28 14:04:05 agricocb has quit (Remote host closed the connection)
1525 2013-06-28 14:05:44 _milone has joined
1526 2013-06-28 14:06:14 rdymac has quit (Quit: This computer has gone to sleep)
1527 2013-06-28 14:07:20 valparaiso has quit (Ping timeout: 268 seconds)
1528 2013-06-28 14:08:25 handle has quit (Remote host closed the connection)
1529 2013-06-28 14:08:45 handle has joined
1530 2013-06-28 14:08:56 milone_ has quit (Ping timeout: 240 seconds)
1531 2013-06-28 14:10:06 justusranvier has joined
1532 2013-06-28 14:11:02 Elmf has quit (Ping timeout: 240 seconds)
1533 2013-06-28 14:11:54 omnibrain has quit (Quit: Page closed)
1534 2013-06-28 14:13:29 Elmf has joined
1535 2013-06-28 14:14:07 valparaiso_afk has joined
1536 2013-06-28 14:14:26 valparaiso_afk is now known as valparaiso
1537 2013-06-28 14:15:21 he1kki_ has quit (Read error: Operation timed out)
1538 2013-06-28 14:16:05 jaekwon has quit (Read error: Connection reset by peer)
1539 2013-06-28 14:16:16 jaekwon has joined
1540 2013-06-28 14:16:34 yano is now known as YANO
1541 2013-06-28 14:18:31 he1kki has joined
1542 2013-06-28 14:19:16 kyledrake has quit (Remote host closed the connection)
1543 2013-06-28 14:19:57 fanquake has joined
1544 2013-06-28 14:20:19 <MICHAGOGO> Luke-Jr: Hi
1545 2013-06-28 14:21:45 fanquake has quit (Client Quit)
1546 2013-06-28 14:22:01 fanquake has joined
1547 2013-06-28 14:27:50 rdymac has joined
1548 2013-06-28 14:27:56 taha has quit (Ping timeout: 276 seconds)
1549 2013-06-28 14:31:15 Eiii has joined
1550 2013-06-28 14:31:15 Eiii has quit (Changing host)
1551 2013-06-28 14:31:15 Eiii has joined
1552 2013-06-28 14:33:54 kwikness has joined
1553 2013-06-28 14:33:57 MICHAGOGO has quit (Quit: Shabbat Shalom)
1554 2013-06-28 14:36:43 <BW^-> sipa: is there any particular code being trigged when there is a change of longest chain?
1555 2013-06-28 14:37:01 <BW^-> sipa: ah right, and, all blocks have a "height" from block zero right?
1556 2013-06-28 14:37:08 polrpaul has joined
1557 2013-06-28 14:37:33 <sipa> "best chain" is measured in amount of hashing work
1558 2013-06-28 14:37:45 <sipa> not length (though in 99% of the cases it'd be identical)
1559 2013-06-28 14:37:49 <BW^-> right, so "height" in the sense, sum of the amount of hashing work of all underlying/previous blocks
1560 2013-06-28 14:38:03 <BW^-> sipa: where is the amount of hashing work associated with a block detected?
1561 2013-06-28 14:38:06 <BW^-> in the code
1562 2013-06-28 14:38:58 <sipa> CBlockIndex bnChainWork
1563 2013-06-28 14:39:53 RazielZ has quit (Ping timeout: 256 seconds)
1564 2013-06-28 14:40:10 <sipa> a new incoming block goes through ProcessBlock, AcceptBlock, AddBlockIndex, SetBestChain, ConnectBlock
1565 2013-06-28 14:40:23 Guest9044 has quit (Quit: -a- Connection Timed Out)
1566 2013-06-28 14:41:23 reneg has joined
1567 2013-06-28 14:41:46 reneg is now known as Guest4402
1568 2013-06-28 14:44:13 agricocb has joined
1569 2013-06-28 14:44:41 agnostic98 has joined
1570 2013-06-28 14:48:17 agnostic98 has quit (Read error: Connection reset by peer)
1571 2013-06-28 14:54:12 tcatm has joined
1572 2013-06-28 14:54:12 tcatm has quit (Changing host)
1573 2013-06-28 14:54:12 tcatm has joined
1574 2013-06-28 14:54:57 kyledrake has joined
1575 2013-06-28 14:58:17 grau has quit (Remote host closed the connection)
1576 2013-06-28 14:58:26 nomailing has quit (Quit: nomailing)
1577 2013-06-28 14:59:17 grau has joined
1578 2013-06-28 15:03:42 <jgarzik> argh! wiki is down.
1579 2013-06-28 15:03:52 <jgarzik> always happens right after I email a wiki link.
1580 2013-06-28 15:04:38 rdymac has quit (Quit: Saliendo)
1581 2013-06-28 15:11:10 taha has joined
1582 2013-06-28 15:13:44 <Luke-Jr> >_<
1583 2013-06-28 15:15:30 taha has quit (Ping timeout: 256 seconds)
1584 2013-06-28 15:15:39 agnostic98 has joined
1585 2013-06-28 15:16:19 wiretapped has quit (Remote host closed the connection)
1586 2013-06-28 15:16:35 wiretapped has joined
1587 2013-06-28 15:17:34 agnostic98 has quit (Read error: Operation timed out)
1588 2013-06-28 15:19:24 dan_ has quit (Remote host closed the connection)
1589 2013-06-28 15:21:19 kyledrake has quit (Remote host closed the connection)
1590 2013-06-28 15:22:54 <nsh> jgarzik, there should be a site that directs someone to a given url or transparently to the latest google/IA-cache on not-200
1591 2013-06-28 15:29:51 dan_ has joined
1592 2013-06-28 15:30:36 RazielZ has joined
1593 2013-06-28 15:33:55 reizuki__ has quit (Quit: Konversation terminated!)
1594 2013-06-28 15:38:29 rdymac has joined
1595 2013-06-28 15:39:46 rdymac has quit (Remote host closed the connection)
1596 2013-06-28 15:42:19 bmcgee has joined
1597 2013-06-28 15:43:24 imd23 has quit (Ping timeout: 264 seconds)
1598 2013-06-28 15:46:14 <BW^-> sipa,td: how are bitcoinj and bitcoind implementationally different, so that the further is better adapted for mobile devices?
1599 2013-06-28 15:46:22 <BW^-> perhaps to start with, bitcoind does not implement the simple verification mode
1600 2013-06-28 15:46:32 <BW^-> though apart from that, is there any particular implementational difference?
1601 2013-06-28 15:46:40 <BW^-> like ,btcd keeps something in RAM that btcj does not?
1602 2013-06-28 15:46:50 <TD> SPV is the main one. of course there are lots of other differences. bitcoinj does not hold all block headers in RAM
1603 2013-06-28 15:46:51 <TD> bitcoind does
1604 2013-06-28 15:47:05 imd23 has joined
1605 2013-06-28 15:48:32 A2501 has joined
1606 2013-06-28 15:52:58 <bmcgee> Hey I'm experimenting with bitcoind's JSON-RPC interface. I send a getblocktemplate request with the longpoll capability but I don't receive a longpollid back. Is there something else I should be doing?
1607 2013-06-28 15:53:42 <BW^-> td: how does that affect performance, holding the headers in ram or not?
1608 2013-06-28 15:53:49 <BW^-> td: how do yo umake it spin well in btcj?
1609 2013-06-28 15:54:07 <TD> obviously holding them in ram is better for performance. but some androids don't have enough memory
1610 2013-06-28 15:54:57 <sipa> bmcgee: bitcoind doesn't support longpolling
1611 2013-06-28 15:55:08 <bmcgee> ah, that makes sense then
1612 2013-06-28 15:55:15 bbbrian has quit (Ping timeout: 268 seconds)
1613 2013-06-28 15:55:38 <bmcgee> any alternatives which do?
1614 2013-06-28 15:56:02 <sipa> poolservers
1615 2013-06-28 15:57:22 <BW^-> td: how much memory are we talking for this?
1616 2013-06-28 15:57:26 rdymac has joined
1617 2013-06-28 15:57:27 rdymac has quit (Changing host)
1618 2013-06-28 15:57:27 rdymac has joined
1619 2013-06-28 15:57:33 <TD> it's 4mb per year. so about 16mb at the moment, i guess
1620 2013-06-28 15:57:44 <TD> more when you take into account the overhead of the map
1621 2013-06-28 15:57:47 bbbrian has joined
1622 2013-06-28 15:57:58 <BW^-> td: which map - and then like how much?
1623 2013-06-28 15:58:06 <TD> HashMap or std::map
1624 2013-06-28 15:58:18 <TD> i don't know how much. i never measured it. apps on android don't get much to play with
1625 2013-06-28 15:58:23 <TD> like maybe 32mb for everything if you're lucky
1626 2013-06-28 15:58:38 spaceSub has quit (Quit: WeeChat 0.3.7)
1627 2013-06-28 16:00:36 <t7> that can not be true
1628 2013-06-28 16:02:44 rdymac has quit (Quit: Saliendo)
1629 2013-06-28 16:03:30 <TD> http://static.googleusercontent.com/external_content/untrusted_dlcp/source.android.com/en/us/compatibility/4.2/android-4.2-cdd.pdf
1630 2013-06-28 16:03:33 <TD> see section 3.7
1631 2013-06-28 16:03:41 <TD> app memory available is linked to screen size
1632 2013-06-28 16:03:45 <TD> lowest is 16mb
1633 2013-06-28 16:03:50 <TD> highest for xlarge/xhdpi is 128mb
1634 2013-06-28 16:04:39 <TD> for a strong device like a galaxy nexus or nexus 4 you get 64mb
1635 2013-06-28 16:04:50 bmcgee has quit (Quit: bmcgee)
1636 2013-06-28 16:04:50 <TD> but then the artwork is all higher resolution as well to handle the higher screen densities
1637 2013-06-28 16:05:20 <TD> so, using a quarter of your RAM budget on the highest-end phones available (more than a quarter really given that a quarter is for the most efficient representation possible) ..... well you can see why we don't do that
1638 2013-06-28 16:05:46 milone_ has joined
1639 2013-06-28 16:09:18 _milone has quit (Ping timeout: 246 seconds)
1640 2013-06-28 16:13:22 A2501 has quit (Remote host closed the connection)
1641 2013-06-28 16:14:16 agnostic98 has joined
1642 2013-06-28 16:16:08 bbbrian has quit (Ping timeout: 256 seconds)
1643 2013-06-28 16:17:18 chmod755 has joined
1644 2013-06-28 16:17:35 metabyte has joined
1645 2013-06-28 16:17:43 fanquake has left ()
1646 2013-06-28 16:19:10 fishfish has quit (Quit: Bye!)
1647 2013-06-28 16:19:56 egis has quit (Read error: Connection reset by peer)
1648 2013-06-28 16:20:31 tcatm has quit (Ping timeout: 252 seconds)
1649 2013-06-28 16:23:52 tcatm has joined
1650 2013-06-28 16:23:52 tcatm has quit (Changing host)
1651 2013-06-28 16:23:52 tcatm has joined
1652 2013-06-28 16:24:53 A2501 has joined
1653 2013-06-28 16:27:42 bloke has joined
1654 2013-06-28 16:29:19 t7 has quit (Quit: ChatZilla 0.9.90 [Firefox 22.0/20130618035212])
1655 2013-06-28 16:31:08 TD is now known as TD[gone]
1656 2013-06-28 16:31:22 super3 has joined
1657 2013-06-28 16:33:00 wamatt has joined
1658 2013-06-28 16:39:23 pecket has quit (Quit: I'm not stupid. I'm just unlucky when I think.)
1659 2013-06-28 16:41:51 rdymac_ has joined
1660 2013-06-28 16:43:14 reneg has joined
1661 2013-06-28 16:43:38 reneg is now known as Guest91630
1662 2013-06-28 16:45:21 testnode9_ has quit (Ping timeout: 246 seconds)
1663 2013-06-28 16:45:48 Guest4402 has quit (Ping timeout: 264 seconds)
1664 2013-06-28 16:48:33 yubrew has quit (Remote host closed the connection)
1665 2013-06-28 16:50:42 _milone has joined
1666 2013-06-28 16:51:36 nx201 has quit (Ping timeout: 276 seconds)
1667 2013-06-28 16:52:56 rdymac_ has quit (Read error: Connection reset by peer)
1668 2013-06-28 16:54:00 Guest91630 has quit (Read error: Connection reset by peer)
1669 2013-06-28 16:54:23 reneg_ has joined
1670 2013-06-28 16:54:27 milone_ has quit (Ping timeout: 246 seconds)
1671 2013-06-28 16:54:28 btcven has joined
1672 2013-06-28 16:57:31 btcven is now known as rdymac
1673 2013-06-28 16:57:51 testnode9_ has joined
1674 2013-06-28 16:58:01 rdymac is now known as Guest80370
1675 2013-06-28 16:58:10 justusranvier has quit (Ping timeout: 268 seconds)
1676 2013-06-28 16:58:48 Guest80370 is now known as btcven
1677 2013-06-28 16:59:46 btcven is now known as rdymac_
1678 2013-06-28 16:59:57 bloke has left ()
1679 2013-06-28 17:03:50 yubrew has joined
1680 2013-06-28 17:04:30 justusranvier has joined
1681 2013-06-28 17:06:52 patcon has joined
1682 2013-06-28 17:09:24 MobPhone has quit (Ping timeout: 256 seconds)
1683 2013-06-28 17:11:01 ericmuyser has quit (Remote host closed the connection)
1684 2013-06-28 17:11:07 theorbtwo has quit (Remote host closed the connection)
1685 2013-06-28 17:11:58 awaxa has joined
1686 2013-06-28 17:14:23 xiaolai has joined
1687 2013-06-28 17:14:52 theorbtwo has joined
1688 2013-06-28 17:15:50 xiaolai has quit (Client Quit)
1689 2013-06-28 17:16:22 xiaolai has joined
1690 2013-06-28 17:16:34 xiaolai has quit (Client Quit)
1691 2013-06-28 17:16:35 <grau> Will anyone else be in London on Tuesday?
1692 2013-06-28 17:17:17 lixiaolai has joined
1693 2013-06-28 17:17:35 lixiaolai has quit (Client Quit)
1694 2013-06-28 17:18:15 lixiaolai has joined
1695 2013-06-28 17:19:23 lixiaolai has quit (Client Quit)
1696 2013-06-28 17:19:33 patcon has quit (Ping timeout: 276 seconds)
1697 2013-06-28 17:19:52 lixiaolai has joined
1698 2013-06-28 17:22:17 lixiaolai has quit (Client Quit)
1699 2013-06-28 17:22:34 peetaur2 has joined
1700 2013-06-28 17:27:05 <sipa> grau: not me
1701 2013-06-28 17:27:20 <grau> sad.
1702 2013-06-28 17:27:36 <grau> YOu are my favorite core dev as you know :)
1703 2013-06-28 17:30:06 <grau> It will be rather a VC not a tech event however.
1704 2013-06-28 17:30:25 <grau> you would be probably bored.
1705 2013-06-28 17:30:51 <tonikt> Hi guys. I was wondering whether there have been any discussion on changes in the network protocol?
1706 2013-06-28 17:31:11 <tonikt> Like being able to download a block in fragments
1707 2013-06-28 17:31:16 <sipa> grau: i figured :)
1708 2013-06-28 17:31:31 <grau> tonkit: there is already bloom filter download for blocks
1709 2013-06-28 17:31:34 <tonikt> Or: see a transaction size, before downloading it
1710 2013-06-28 17:31:37 <sipa> tonikt: BIP37 allows that, in a way
1711 2013-06-28 17:31:46 <tonikt> grau: bloom filter is not good for full node
1712 2013-06-28 17:32:03 <grau> block fragments are no good for full node either
1713 2013-06-28 17:32:12 <tonikt> grau: why not?
1714 2013-06-28 17:32:13 <sipa> you can have two complementary bloom filters set on two connections
1715 2013-06-28 17:32:20 <sipa> and download half blocks from each
1716 2013-06-28 17:32:40 <tonikt> so the answer is 'no'?
1717 2013-06-28 17:33:05 <gmaxwell> tonikt: Not a very worthwhile thing in my opinion, and as sipa points out its already possible.
1718 2013-06-28 17:33:05 <grau> tonkit: a fragment can not prove that funds are not spent in the other part
1719 2013-06-28 17:33:06 <sipa> the answer is 'partially', BIP37 supports (some way) of downloading blocks in fragments
1720 2013-06-28 17:33:15 <tonikt> I believe the network could really use being able to download one block from diffetent peers in i.e. 32KB chunks
1721 2013-06-28 17:33:28 <sipa> that'd only slow things down imho
1722 2013-06-28 17:33:45 <tonikt> gmaxwell: but I was checking these bloom filters and it wasn't possible to download a block like that
1723 2013-06-28 17:33:47 <sipa> and there is no way to prove that a transaction has a certain size without actually sending that transactions
1724 2013-06-28 17:33:57 <sipa> tonikt: it is
1725 2013-06-28 17:34:13 <gmaxwell> tonikt: it is, but as sipa and I are telling youâ it's not obviously useful.
1726 2013-06-28 17:34:14 <grau> tonkit: it is already allowed by the protocol to use several peers for download. BOP actually does that by block not by block fragment
1727 2013-06-28 17:34:23 <tonikt> sipa: yes - it isnt possible to confirm it, but if you see a mismatch you can at least ban the peer
1728 2013-06-28 17:34:36 <sipa> tonikt: set nHashFunctions=1, and use random complementary bits in the filter
1729 2013-06-28 17:34:41 <tonikt> grau: I al talking about a fully synchronized node
1730 2013-06-28 17:34:45 <gmaxwell> tonikt: what are you talking about?
1731 2013-06-28 17:34:57 <tonikt> you suddenly get an INV with a new block - you want to download it ASAP
1732 2013-06-28 17:35:03 <sipa> what is the actual problem you're trying to solve?
1733 2013-06-28 17:35:18 <tonikt> ... so why not to split the work into parts and downlaod it in paralell?
1734 2013-06-28 17:35:32 <sipa> we've just told you how to do that using BIP37
1735 2013-06-28 17:35:51 <tonikt> wait, I cannot read that BIP37 cause wiki is broken
1736 2013-06-28 17:35:58 <gmaxwell> Because doing so will _slow_ transmission except to the extent that it gets you an unfair share of the channel capacity.
1737 2013-06-28 17:36:00 <tonikt> maybe you are talking about a different thing :)
1738 2013-06-28 17:36:15 <sipa> 19:06:43 < sipa> tonikt: set nHashFunctions=1, and use random complementary bits in the filter
1739 2013-06-28 17:36:19 <devrandom> TD[gone]: what's the situation with netty?
1740 2013-06-28 17:36:33 ProfMac has joined
1741 2013-06-28 17:36:48 <sipa> tonikt: in practice it'd be very hard to coordinate that, but if block sizes would grow a lot, that may be a viable strategy
1742 2013-06-28 17:37:01 <tonikt> just to be clear: BIP37 was about downloading the header and the tx hashes, followed by the actual transactions?
1743 2013-06-28 17:37:11 <sipa> BIP37 = bloom filtering
1744 2013-06-28 17:37:13 <tonikt> yes
1745 2013-06-28 17:37:26 <tonikt> so how does it help me to split a block download amoung my 50 peers?
1746 2013-06-28 17:37:31 <sipa> so you give node A a random filter that selects 50% of the transactions
1747 2013-06-28 17:37:47 <sipa> and you give node B a complementary filter that selects the other 50% of the transactions
1748 2013-06-28 17:38:00 <tonikt> yes, but I have more than 2 peers.
1749 2013-06-28 17:38:15 <sipa> ok, then give node A a filter that selects 33% of the transactions
1750 2013-06-28 17:38:16 <tonikt> your solution seems more like a work around
1751 2013-06-28 17:38:35 <sipa> it's a very neat solution, as you don't need to keep track of what to download from whom
1752 2013-06-28 17:38:39 <tonikt> ok, sorry. let me ask a question then
1753 2013-06-28 17:38:43 <sipa> they figure it out themself
1754 2013-06-28 17:38:58 <grau> tonkit: and you seem to work around a problem not really there until blocks are the size we know.
1755 2013-06-28 17:39:19 <gmaxwell> tonikt: you're just going to end up in N connections all in slow start. Plus users setting your house on fire because you use _all_ their bandwidth in a burst once the connections come out of slowstart.
1756 2013-06-28 17:39:20 <sipa> maybe this will become an actual problem if the block size limit is increased
1757 2013-06-28 17:39:25 <sipa> if it is, we'll deal with it
1758 2013-06-28 17:39:33 <tonikt> why cant there be like a command "getsomething" that would return me the length of that something, plus a list of hashes of its data split into i.e. 4KB chunks
1759 2013-06-28 17:39:35 ericmuyser has joined
1760 2013-06-28 17:39:43 <tonikt> ... and that would be same for txs and blocks
1761 2013-06-28 17:40:01 <gmaxwell> sipa: sure if there are larger blocks then eventually at some point it makes sense. A few hundred K is really sketchy for any benefit there.
1762 2013-06-28 17:40:20 <tonikt> gmaxwell: no - now I use their bandwidth in a burst, asking each peer for entire block
1763 2013-06-28 17:40:40 <tonikt> I'm talking about a situation when we have a new block mined
1764 2013-06-28 17:40:45 <tonikt> which is every 10 minutes
1765 2013-06-28 17:40:50 <gmaxwell> You don't ask each peer for the entire block.
1766 2013-06-28 17:41:00 ProfMac has quit (Ping timeout: 250 seconds)
1767 2013-06-28 17:41:03 <tonikt> well, you should if you care to have it ASAP :)
1768 2013-06-28 17:41:10 <tonikt> I do :)
1769 2013-06-28 17:41:13 <gmaxwell> You ask a single peer for the entire block, or as sipa points out multiple peers for mutually exclusive subsets.
1770 2013-06-28 17:41:30 <tonikt> yes, I understand that is the status
1771 2013-06-28 17:41:32 <gmaxwell> tonikt: then you're abusing the network.
1772 2013-06-28 17:41:40 <gmaxwell> and also hurting your own performance.
1773 2013-06-28 17:41:41 <tonikt> but I was talking about a possible improvement
1774 2013-06-28 17:41:47 <tonikt> that is how bittorrent works
1775 2013-06-28 17:41:48 grau_ has joined
1776 2013-06-28 17:41:58 <tonikt> yes, call me an abuser :)
1777 2013-06-28 17:42:02 <sipa> do you use bittorrent for 1 MB data?
1778 2013-06-28 17:42:16 <tonikt> sometimes
1779 2013-06-28 17:42:17 <gmaxwell> sipa: normally torrents use 4-16 mb chunks.
1780 2013-06-28 17:42:21 <sipa> gmaxwell: i know
1781 2013-06-28 17:42:38 <tonikt> no
1782 2013-06-28 17:42:44 <gmaxwell> tonikt: what you're describing there sucks, because you can't tell who's screwed you and given you invalid data. With what sipa told you to do you can tell.
1783 2013-06-28 17:42:46 <sipa> it can use smaller chunks
1784 2013-06-28 17:42:49 <tonikt> I'm sure they use smaller ones - like 32KB
1785 2013-06-28 17:43:00 <sipa> gmaxwell said "normally"
1786 2013-06-28 17:43:16 <gmaxwell> tonikt: no, yesâ it can, but thats not what actually gets used normally (except for tiny files, which torrent takes much longer to transfer than http)
1787 2013-06-28 17:43:34 grau has quit (Ping timeout: 252 seconds)
1788 2013-06-28 17:43:43 <tonikt> OK, I get it - someone can send me a corrupt list of block's chunks hashes
1789 2013-06-28 17:43:47 <tonikt> but then...
1790 2013-06-28 17:43:50 <sipa> this is an engineering question, and it depends on very specific data (latency, bandwidth, variation, distribution)
1791 2013-06-28 17:43:54 <sipa> once it becomes a problem
1792 2013-06-28 17:44:02 <sipa> we can find an appropriate solution
1793 2013-06-28 17:44:10 <sipa> and there are several ways to deal with it
1794 2013-06-28 17:44:10 <tonikt> why don't you add a protocol command "give me this tx from this block?"
1795 2013-06-28 17:44:27 <sipa> because that would be absolutely horrible for the peers
1796 2013-06-28 17:44:38 <sipa> they need to look up the block for you, but only give you a small piece of it
1797 2013-06-28 17:44:51 <tonikt> sipa: with all due respect, but for you to find a solution, one needs to wait 2 years in average :)
1798 2013-06-28 17:45:17 JKL1234- has quit (Quit: I'm using a Free IRC Bouncer from BNC4FREE - http://bnc4free.com/)
1799 2013-06-28 17:45:20 <tonikt> It is actually quite easy to calculate
1800 2013-06-28 17:45:35 <tonikt> how much time you need to download lets say 1MB block
1801 2013-06-28 17:45:47 <tonikt> having 1Mbps connection - about 10 seconds
1802 2013-06-28 17:45:47 <sipa> right now the problem simply doesn't exist with 1 MB blocks, unless on very slow links (mobile?) where you don't want full blocks anyway
1803 2013-06-28 17:46:12 <tonikt> 10 senconds from peer to peer (and that's not counting checking it)
1804 2013-06-28 17:46:33 <tonikt> dont you think it is already enough to try decreasing it be a few folds?
1805 2013-06-28 17:47:49 bmcgee has joined
1806 2013-06-28 17:47:56 <gmaxwell> tonikt: go convince bittorrent to never use chunks larger than 100k and come back. :P
1807 2013-06-28 17:48:01 <tonikt> I believe all the API is there already, except that it should be able to download a tx that is already mined, though by specifying a block where it is
1808 2013-06-28 17:48:18 <sipa> if you're on a 1 Mbps connection, there's no way you'll get it faster than in 10s anyway
1809 2013-06-28 17:48:19 <tonikt> gmaxwell: but bittorrent does not care about latency
1810 2013-06-28 17:48:26 <tonikt> while bitcoins hould
1811 2013-06-28 17:49:05 <Vinnie_win> sup fools
1812 2013-06-28 17:49:07 <sipa> if it's your peer that is limited to 1Mbps, but your connection is faster, you won't be downloading from him (as he'll be slower to announce it to you anyway)
1813 2013-06-28 17:49:27 <tonikt> ok, so you dont want to change the net protocol, before it becoming a problem
1814 2013-06-28 17:49:29 <gmaxwell> tonikt: generally parallel fetching is not great for latency.
1815 2013-06-28 17:49:29 <tonikt> fine
1816 2013-06-28 17:49:44 <gmaxwell> As you end up waiting on the most latent response before you can validate any of it.
1817 2013-06-28 17:50:04 <sipa> tonikt: i'm against changing the protocol before having clear information about the benefits
1818 2013-06-28 17:50:20 <sipa> and no, i don't think right now there is much that can be improved
1819 2013-06-28 17:50:26 <tonikt> gmaxwell: think if I could ask a node for a block and all its transaction hashes - and other nodes for transactions
1820 2013-06-28 17:50:27 <gmaxwell> tonikt: no, we're also saying that for the current maximum blocksize what you're suggesting is likely to _hurt_. Without a bunch of analysis it would be hard to know.
1821 2013-06-28 17:50:28 <sipa> in the future that can certainly change
1822 2013-06-28 17:50:34 <bmcgee> Hey I'm trying to learn more about the role of a pool server when using getblocktemplate. As I understand it with block generation back in the hands of the miner, does the pool server no longer create shares of lower difficulty as before, but now simply verifies the work is being done? Is there any docs available explaining the whole process a bit more?
1823 2013-06-28 17:50:50 <tonikt> I just dont see a problem with adding a block hash to inv while asking for tx data
1824 2013-06-28 17:51:03 <tonikt> it doesn't seem like a development challenge
1825 2013-06-28 17:51:09 <sipa> it's not
1826 2013-06-28 17:51:11 <tonikt> and it could solve the problem
1827 2013-06-28 17:51:16 wamatt has quit (Quit: wamatt)
1828 2013-06-28 17:51:18 <sipa> it's a maintainance overhead
1829 2013-06-28 17:51:25 <sipa> and a compatibility burden
1830 2013-06-28 17:51:30 <gmaxwell> We could also throw some virgins into volcanos, that might "solve the problem"
1831 2013-06-28 17:51:42 <gmaxwell> (not that you've established that there is even a problem to be solved)
1832 2013-06-28 17:51:55 <tonikt> ok, I get it - you don't see a problem
1833 2013-06-28 17:52:04 <sipa> *yet*
1834 2013-06-28 17:52:04 freewil has joined
1835 2013-06-28 17:52:08 <tonikt> you are obviously not so much a perfectionists, as I am :P
1836 2013-06-28 17:52:29 <sipa> well we're not dealing with a nicely theoretical problem where the optimal soluton is obvious
1837 2013-06-28 17:52:34 <sipa> everything has downsides
1838 2013-06-28 17:52:41 robocoin_ is now known as robocoin
1839 2013-06-28 17:52:47 <gmaxwell> tonikt: I don't see evidence of perfectionism in you in this discussion. If there were you'd be doing some careful analysis to establish some tests to determine the level of improvement possible.
1840 2013-06-28 17:53:14 <gmaxwell> Instead you're just shooting from the hip with a blind guess at something that would maybe help or hurt a problem which may exist in the future.
1841 2013-06-28 17:53:40 <tonikt> gmaxwell: all I can tell you is that, if I want to download a block ASAP, I ask each of my peers for the entire one - is this perfect for you?
1842 2013-06-28 17:54:20 <gmaxwell> tonikt: that won't actually fetch you the block faster than asking a single peer in many cases.
1843 2013-06-28 17:54:21 <tonikt> if I could download it in parts/transactions - that would be perfect, unless I'd screw up my implementation
1844 2013-06-28 17:54:31 <gmaxwell> No, in fact it wouldn't be.
1845 2013-06-28 17:54:51 <tonikt> :) no, it would
1846 2013-06-28 17:54:56 <sipa> you'd still have to wait for the slowest one to respond
1847 2013-06-28 17:54:59 <gmaxwell> tonikt: or look at it another wayâ why not ask them each for one bit of it?
1848 2013-06-28 17:55:11 <gmaxwell> as sipa says, you have to wait for the slowest response.
1849 2013-06-28 17:55:26 <sipa> if the only constraint is bandwidth, and processing speed and latency don't exist, your solution is optimal
1850 2013-06-28 17:55:29 freewil has quit (Client Quit)
1851 2013-06-28 17:55:41 <gmaxwell> sipa: and overhead doesn't exist.
1852 2013-06-28 17:55:46 <tonikt> guys, do I really need to explain you how to implement it?
1853 2013-06-28 17:55:47 sanders has quit (Read error: Operation timed out)
1854 2013-06-28 17:55:47 <sipa> and attackers
1855 2013-06-28 17:56:02 sanders has joined
1856 2013-06-28 17:56:15 <tonikt> you just have a list of txs to download and your peers - whenever any of them is busy, you ask him for the next ts
1857 2013-06-28 17:56:22 <gmaxwell> Right, so in a world where the only constrain is the remote peers bandwidth, and process speed, latency, attackers, and overhead don't existâ then indeed, thats optimal.
1858 2013-06-28 17:56:25 <tonikt> it must be faster
1859 2013-06-28 17:56:25 <sipa> how do you know he's busy?
1860 2013-06-28 17:56:35 <sipa> by waiting?
1861 2013-06-28 17:56:50 <tonikt> i know he is busy because he has not responded to my previous data request yet
1862 2013-06-28 17:56:54 <gmaxwell> It's unknowable without waiting because of latency.
1863 2013-06-28 17:57:05 <tonikt> so it does not make any sense to ask him for more data
1864 2013-06-28 17:57:06 <gmaxwell> tonikt: if he's 80ms away he _cannot_ answer faster than 80ms.
1865 2013-06-28 17:57:20 <tonikt> sure - but you have 1000 txs
1866 2013-06-28 17:57:27 <sipa> and you need all of them
1867 2013-06-28 17:57:35 <tonikt> eventually he will answer and you will have 900+ left anyway
1868 2013-06-28 17:57:45 <gmaxwell> so you're going to fetch them one at a time without pipelining? lol good luck with that.
1869 2013-06-28 17:57:48 <tonikt> and then you will ask him for the 913th one
1870 2013-06-28 17:57:49 <bmcgee> ⦠I get the impression now's not a good time for asking potentially stupid questions â¦
1871 2013-06-28 17:58:07 <tonikt> gmaxwell: it's at least 200+ bytes
1872 2013-06-28 17:58:12 <sipa> tonikt: that is all very sensible, given the right bandwidth/latency tradeoffs
1873 2013-06-28 17:58:12 <gmaxwell> bmcgee: well, your question doesn't have a neat answer.
1874 2013-06-28 17:58:21 <tonikt> but youre right, it would be better to ask for several tx at the same time
1875 2013-06-28 17:58:28 <sipa> tonikt: it's something you certainly want to do at the ~megabyte level
1876 2013-06-28 17:58:30 <tonikt> except that some of thme may be 100kb big
1877 2013-06-28 17:58:31 <gmaxwell> tonikt: lol. just the TCP overhead from the request is going to instantly give you 50% overhead on 200 bytes.
1878 2013-06-28 17:58:50 <sipa> (and we don't by the way, so let's fix that first)
1879 2013-06-28 17:59:02 <tonikt> gmaxwell: now it gives you much more than 50%
1880 2013-06-28 17:59:05 <bmcgee> gmaxwell: understood, if at some point one of you guys could spend a few mins and field some general questions it might point me in a better direction. I'm happy to hunt down the info, I'm just not sure what to be looking for as yet
1881 2013-06-28 17:59:13 <tonikt> most of the txs you have already anyway
1882 2013-06-28 17:59:31 <gmaxwell> tonikt: no, it doesn't the overhead on transfering a block is about 2%.
1883 2013-06-28 17:59:38 <sipa> tonikt: again BIP37 to the rescue (it doesn't send transactions it knows you already have, without extra latency)
1884 2013-06-28 17:59:56 wamatt has joined
1885 2013-06-28 17:59:58 <gmaxwell> okay, ignoring that. But as sipa says bip37 takes care of that.
1886 2013-06-28 18:00:17 <gmaxwell> bmcgee: getblocktemplate potentially allows what you decribed but it's up to the pool to decide how they use it.
1887 2013-06-28 18:00:49 <gmaxwell> bmcgee: e.g. they could tell you what to have the coinbase pay to.. you mine but provide the whole block yourself.. and send them back low diff shares to prove that you're working.
1888 2013-06-28 18:01:08 <tonikt> gmaxwell: but using bip37 is working around a solution. if I need this tx from this block - why do I need to bother with bloom filters and statistics?
1889 2013-06-28 18:01:13 <gmaxwell> bmcgee: or it can be used just like traditional pooled mining where you're just a dumb computing engine on a stick and the pool does all the transaction selection.
1890 2013-06-28 18:01:29 <sipa> tonikt: because it allows you to do things without extra latency
1891 2013-06-28 18:01:47 <sipa> tonikt: you don't have to be told about the list of transactions first, and you don't have to reply with which transactions you want
1892 2013-06-28 18:01:54 <gmaxwell> tonikt: statistics?! you set a series of complemetary bits. and don't have to taken another 80ms of round trip time to send extra tiny requests with redundant hashes.
1893 2013-06-28 18:02:06 <tonikt> sipa: but the biggest latency comes not from the ping - it somes from the data that needs to be finished, before they can be used
1894 2013-06-28 18:02:16 <gmaxwell> @#@*$(@#
1895 2013-06-28 18:02:39 pjorrit has left ()
1896 2013-06-28 18:02:39 pjorrit has joined
1897 2013-06-28 18:02:48 <sipa> tonikt: on a 10 Mbit/s link, a not outrageous 200ms ping time (meaning 400ms extra for a roundtrip) means 400 kilobyte that could have been downloaded while they just waited for you to answer
1898 2013-06-28 18:02:48 <gmaxwell> tonikt: go look at the bandwidth numbers you gave before! on a 1mbit connection the data to transfer a fee hundred bytes is way less than typical latency.
1899 2013-06-28 18:02:56 <gmaxwell> er the time to transfer.
1900 2013-06-28 18:02:56 <sipa> that's more than an average block
1901 2013-06-28 18:03:26 ProfMac has joined
1902 2013-06-28 18:04:11 <tonikt> ok guys, whatever. I see you have your world and don't really want to notice mine. I guess I will have to wait for it to become a problem :)
1903 2013-06-28 18:04:41 <tonikt> but if I might add something, not as a question, but as a proposal
1904 2013-06-28 18:05:18 <gmaxwell> Yes, my world has latency in it. Not sure where you get one that doesn't, but I'd like one of those. :P
1905 2013-06-28 18:05:33 <tonikt> 1) allow to ask for a size of a transaction/block before downloading it (so you can ban anyone who is trying to send you more)
1906 2013-06-28 18:05:43 <sipa> tonikt: what if the peer lies?
1907 2013-06-28 18:05:57 <gmaxwell> sipa: there are no attackers in tonikt's world.
1908 2013-06-28 18:06:04 <tonikt> 2) imagine that you are connected to 30 peers and a new 1MB block have just been mined: what is the fastest way to download it from your peers?
1909 2013-06-28 18:06:14 <sipa> depends on your bandwidth and latency
1910 2013-06-28 18:06:23 sanders has quit (Read error: Operation timed out)
1911 2013-06-28 18:06:25 <gmaxwell> tonikt: if you think you can do better, just write a second transfer protocol. If it's better it should be easy to demonstrate. You'll probably learn something in the process.
1912 2013-06-28 18:06:25 <tonikt> sipa: if the peer lies, than you will find it out and ban it
1913 2013-06-28 18:06:37 <sipa> with very low bandwidth and low latency at the same time, downloading in parallel will certainly be faster
1914 2013-06-28 18:07:07 <tonikt> gmaxwell: believe me, I can write a protocol, but it would be quite silly to test it in my home network
1915 2013-06-28 18:07:20 <gmaxwell> tonikt: then use a network simulator.
1916 2013-06-28 18:07:33 <gmaxwell> It's pretty straight forward to simulate actual network behavior.
1917 2013-06-28 18:07:52 <sipa> tonikt: well to deploy it, we'd first need a way to download *anything* in parallel first
1918 2013-06-28 18:08:02 <tonikt> gmaxwell: but I dont need to simulate it to know that downloadin a block in parts from several peers at the same time will be faster
1919 2013-06-28 18:08:31 <gmaxwell> tonikt: But you're incorrect. The way you're describing that involves lots of round trips will actually be _slower_ in a case where there is considerable latency.
1920 2013-06-28 18:09:03 <gmaxwell> The exact balance depends on a number of factors.
1921 2013-06-28 18:09:19 <gmaxwell> Basically you never want to make a request that is smaller than the bandwidth delay product.
1922 2013-06-28 18:09:23 <tonikt> gmaxwell: no, becasue shorter transactions could go in bulk (like 20 * 200+ bytes)
1923 2013-06-28 18:09:44 <tonikt> .. thats why you need to have a way to find out tx size before downloadin it
1924 2013-06-28 18:09:46 sanders has joined
1925 2013-06-28 18:10:15 <sipa> and just finding out that size means an extra round-trip, which (in some cases) may be slower than just downloading the whole block
1926 2013-06-28 18:10:17 <gmaxwell> Now you're transmitting a bunch of data to make those decisions, and then you can do nothing until it shows up. During that time you could have sent a whole block.
1927 2013-06-28 18:11:32 <tonikt> I think we should be talking numbers here, otherwise it's just baseless accusations
1928 2013-06-28 18:11:43 <gmaxwell> There are numbers above.
1929 2013-06-28 18:12:01 <tonikt> Can we agree that an average node would have 1mbps upload speed?
1930 2013-06-28 18:12:05 pecket has joined
1931 2013-06-28 18:12:24 <tonikt> download is bigger - I know
1932 2013-06-28 18:12:30 <gmaxwell> Then when you get bad data it's impossible to tell who is giving you bad data until you have the whole block... which means that you have to fetch it from one peer if some peers is giving you bad data.. pretty cheap dos attack.
1933 2013-06-28 18:12:59 <tonikt> gmaxwell: of course you can say who sent you bad data, because you ask for transactions, which hashes you know
1934 2013-06-28 18:13:26 <gmaxwell> tonikt: e.g. I give you the wrong hashes for the block.
1935 2013-06-28 18:13:45 <tonikt> gmaxwell: with a proper difficulty? :)
1936 2013-06-28 18:13:50 <gmaxwell> huh?!
1937 2013-06-28 18:13:53 <tonikt> I can live with that
1938 2013-06-28 18:14:01 <gmaxwell> ...
1939 2013-06-28 18:14:12 <tonikt> I will compare the hashes against the merkle from the block
1940 2013-06-28 18:14:27 <sipa> which you don't have yet?
1941 2013-06-28 18:14:35 <gmaxwell> So now you have to fetch the whole merkle tree first. keep adding overhead.. (you'll end up with bip37 in a few more minutes)
1942 2013-06-28 18:14:36 <tonikt> The header is 80 bytes long
1943 2013-06-28 18:14:59 <sipa> the average transaction is 250 bytes or so
1944 2013-06-28 18:15:13 <sipa> a txid is 32
1945 2013-06-28 18:15:26 <tonikt> the minimal transaction is 250 or so :)
1946 2013-06-28 18:15:27 <sipa> that means you have to download 1/8 of the block's size before you can make any decision
1947 2013-06-28 18:15:57 <tonikt> I can live with that as well
1948 2013-06-28 18:16:10 <gmaxwell> tonikt: and again, you don't need to change the p2p protocol to expirementâ you can just use an alternative p2p protocol, and simulate actual internet conditions.
1949 2013-06-28 18:16:11 <tonikt> its still 8:1 compression :)
1950 2013-06-28 18:16:41 <tonikt> I know what I can experiment with, guys
1951 2013-06-28 18:16:59 sandbote has joined
1952 2013-06-28 18:17:01 <tonikt> its just that I dont need to experiment to know that it would be a good thing to do
1953 2013-06-28 18:17:15 daybyter has joined
1954 2013-06-28 18:17:16 freewil has joined
1955 2013-06-28 18:17:17 <gmaxwell> There are some bandwidth delay mixtures where some strategies are better than other ones, and different strategies are better in other conditions.
1956 2013-06-28 18:17:34 <tonikt> like this think recently, that they fixed
1957 2013-06-28 18:17:48 <tonikt> a peer sends you a longer tx than it should be
1958 2013-06-28 18:18:09 <sipa> that was actually an intentional design decision
1959 2013-06-28 18:18:12 <tonikt> you should ban it - but you cant, because it is likely a legit client, with a bug :)
1960 2013-06-28 18:18:27 <gmaxwell> tonikt: imagineâ for a momentâ that you peers are on mars with a 40 minute latency, and your bandwidth to each peer is 1gbit/sec, and your pay 1 BTC per megabyte transfered in aggregate. What is the optimal strategy?
1961 2013-06-28 18:18:47 <tonikt> gmaxwell: in this case I get your point
1962 2013-06-28 18:18:55 <tonikt> ... but I thought that we were on Earth
1963 2013-06-28 18:19:00 <phantomcircuit> gmaxwell, put your btc node on earth and send it instructions
1964 2013-06-28 18:19:49 <gmaxwell> tonikt: I used an extreme example because you keep rejecting the idea that different situations require different tradeoffs and you keep suggesting ones which have additional round trips, when it's possible to do this _without_ adding them.. so clearly you're not thinking about _something_.
1965 2013-06-28 18:19:51 <sipa> tonikt: banning would be a very bad idea - not because there are buggy clients that add random junk to transactions, but because you'd be hurting the ones forwarding an attacker's junk, not the attacker themself
1966 2013-06-28 18:20:13 <tonikt> so really, please, add an option to ask for tx/block size without a need to download it and allow to do getdata for tx giving a block hash as a reference - that's all I ask :)
1967 2013-06-28 18:20:19 <gmaxwell> banning on transitive behavior is a superfantastic way to convert mining dos attacks into network partitioning.
1968 2013-06-28 18:20:40 <gmaxwell> tonikt: We will not add that. Sorry.
1969 2013-06-28 18:20:41 milone_ has joined
1970 2013-06-28 18:20:57 <tonikt> gmaxwell: I know :)
1971 2013-06-28 18:21:18 <tonikt> but dont tell me later, that I did not suggest it :P
1972 2013-06-28 18:21:23 <sipa> tonikt: the first is unauthenticated data (someone can just lie, and you can claim you protect against it, but your optimal behaviour still depends on them being honest - i really prefer solutions that do not need such an assumption)
1973 2013-06-28 18:21:26 <gmaxwell> tonikt: Please create an alternative transportâ in the process you'll learn something about the evils of roundtrips for performance, come up with a better proposal which is potentially useful.
1974 2013-06-28 18:21:58 <gmaxwell> Even a protocol that depends on honesty would be not the end of the world as an alternative transport: just use it between friends.
1975 2013-06-28 18:22:05 FredEE has joined
1976 2013-06-28 18:22:19 <tonikt> sipa: but if someone lies, you will find it out and you will ban it - that is the whole point
1977 2013-06-28 18:22:22 <gmaxwell> but it's not something that makes a lot of sense as the standard p2p protocol.
1978 2013-06-28 18:22:35 <sipa> tonikt: and you'll still have lost time doing so
1979 2013-06-28 18:22:46 <sipa> tonikt: something the attacker may not care about, but you do
1980 2013-06-28 18:23:10 <tonikt> sipa: yes, but you will pay this time to ban the bastard. now you have the same problem, but you cannot ban the bastared
1981 2013-06-28 18:23:16 <gmaxwell> tonikt: IPs are cheap, we regularly get trolls on IRC with access to thousands of IPs. Someone doing that could force all your nodes into wasting a ton of bandwidth and fall back to single peer fetching.
1982 2013-06-28 18:23:29 <gmaxwell> (and make you take many times longer to fetch the block)
1983 2013-06-28 18:23:41 <gmaxwell> I fully endorse my mining competition adopting such a protocol. :P
1984 2013-06-28 18:23:45 <tonikt> gmaxwell: 99kb is probably cheaper than in IP
1985 2013-06-28 18:23:47 <tonikt> an*
1986 2013-06-28 18:24:09 wamatt has quit (Quit: wamatt)
1987 2013-06-28 18:24:12 <tonikt> so I can download 99kb from the IP, just to ban it for being wrong
1988 2013-06-28 18:24:26 polrpaul has quit (Quit: Page closed)
1989 2013-06-28 18:24:26 _milone has quit (Ping timeout: 252 seconds)
1990 2013-06-28 18:24:30 <tonikt> especially if I know the size up front and I do not donload anything bigger than 10kb
1991 2013-06-28 18:24:55 <gmaxwell> or, you could, you know, use a protocol which doesn't depend on unauthenticated data and which doesn't require extra round trips.. and still fetches in parallel (if the bandwidth/latency ratios make it profitable to do so)
1992 2013-06-28 18:24:55 <tonikt> .. and if I see the 10001th byte - I can it already at that moment
1993 2013-06-28 18:25:00 peter has joined
1994 2013-06-28 18:25:22 peetaur2 has quit (Ping timeout: 252 seconds)
1995 2013-06-28 18:25:23 peter is now known as Guest29918
1996 2013-06-28 18:25:51 Guest29918 is now known as peetaur2
1997 2013-06-28 18:25:57 <gmaxwell> And, in fact, BIP37 already gives us that, along with automatic (zero round trip) elimiating of known-already-sent data.
1998 2013-06-28 18:26:08 bbbrian has joined
1999 2013-06-28 18:26:33 <tonikt> bip37 was a nice invention. the only problem with it is that nobody wants to use it
2000 2013-06-28 18:26:59 <tonikt> why wont you once invent something that people would want to use it, for a change? ;)
2001 2013-06-28 18:27:07 <gmaxwell> what are you talking about??
2002 2013-06-28 18:27:08 <gmaxwell> lol
2003 2013-06-28 18:27:17 <gmaxwell> every peer connected to me at the moment supports bip37.
2004 2013-06-28 18:27:36 <tonikt> supports, but does not get adventage of it
2005 2013-06-28 18:27:38 <gmaxwell> Now I think you're just trolling.
2006 2013-06-28 18:28:02 <tonikt> I guess you can always find someone who'd kick me out :)
2007 2013-06-28 18:28:09 <sipa> between satoahi clients, he's right
2008 2013-06-28 18:28:18 <gmaxwell> I suppose I could. :P
2009 2013-06-28 18:28:29 <gmaxwell> but seriously, what the heck.
2010 2013-06-28 18:28:35 <sipa> but my cell phone just loves bip37
2011 2013-06-28 18:29:00 <gmaxwell> tonikt: all the bitcoinj clients happily use itâ but we don't think parallel fetching is currently useful in the satoshi client.
2012 2013-06-28 18:29:46 <phantomcircuit> iirc the fetch queue ends up pulling more blocks even before the queue has been processed right
2013 2013-06-28 18:29:50 <gmaxwell> After a bunch of archectural changes it might be useful.
2014 2013-06-28 18:29:53 <sipa> it's definitely useful at the block level
2015 2013-06-28 18:29:56 <phantomcircuit> so the pipeline stays full
2016 2013-06-28 18:30:00 <sipa> and we don't even do it there
2017 2013-06-28 18:30:05 <sipa> and we absolutely should
2018 2013-06-28 18:30:44 <gmaxwell> sipa: ::nods::
2019 2013-06-28 18:31:07 roconnor has joined
2020 2013-06-28 18:31:29 <sipa> phantomcircuit: yes, you can have up to 500 queued getdata requests
2021 2013-06-28 18:31:38 <sipa> whuch may take minutes to download
2022 2013-06-28 18:31:43 nova90 has joined
2023 2013-06-28 18:31:46 <gmaxwell> Doesn't involve the unauthenticated data / latency tradeoffs. And couple hundred k blocks are large enough that there isn't an overhead tax from doing thatâ beyond fetching the headers seperately.
2024 2013-06-28 18:32:42 kanzure has quit (Ping timeout: 248 seconds)
2025 2013-06-28 18:33:39 kanzure has joined
2026 2013-06-28 18:33:44 brson has joined
2027 2013-06-28 18:34:14 wamatt has joined
2028 2013-06-28 18:36:10 andyh2 has joined
2029 2013-06-28 18:37:04 peetaur2 has quit (Read error: Operation timed out)
2030 2013-06-28 18:37:23 <gmaxwell> bmcgee: So, GBT.
2031 2013-06-28 18:37:48 <bmcgee> ah, i see things have dissipated...
2032 2013-06-28 18:38:21 peetaur2 has joined
2033 2013-06-28 18:38:23 <bmcgee> so i'm just trying to understand the role of the pool server a bit more
2034 2013-06-28 18:38:27 <gmaxwell> bmcgee: I personally think that ideally what would happen is that people would use GBT to do solo mining but with pooled payments. But there would need to be mining software support for working on work using data from two sources (a local bitcoind and a remote pool)
2035 2013-06-28 18:39:27 wamatt has quit (Ping timeout: 246 seconds)
2036 2013-06-28 18:39:27 <gmaxwell> bmcgee: the fundamental role of a poolserver is to accept sub-difficulty work, test it against some criteria, and compute payment credit.
2037 2013-06-28 18:39:32 <tonikt> so anyway guys, to wrap up, I did not mean to be mean, just to indicate my needs. and I appreciate your advise, but I am not going to make a network simulation just to convince you :)
2038 2013-06-28 18:40:20 <bmcgee> it's the sub-difficulty bit i'm having a hard time understanding. here's my understanding so far, please validate:
2039 2013-06-28 18:40:42 <tonikt> and I still don't understand why you would not want to have a possibility to find out tx/block size before downloading it
2040 2013-06-28 18:40:47 JKL1234- has joined
2041 2013-06-28 18:41:10 <tonikt> or why you would not want to be able to download tx from a specific block
2042 2013-06-28 18:41:42 <gavinandresen> downloading tx from a specific block opens up a denial-of-service attack
2043 2013-06-28 18:41:43 <gmaxwell> tonikt: because it's a dos attack vector, both for the serving node and for anyone relying on it.
2044 2013-06-28 18:41:56 RoboTeddy has joined
2045 2013-06-28 18:42:33 <tonikt> how is downloading tx from a block more of an attack than downloading a block?
2046 2013-06-28 18:42:34 <bmcgee> pool server hits up a bitcoind for block template, tweaks the block template to lower the difficulty a bit, miners work away, submit valid blocks, pool server checks the blocks meet the lower difficulty, and every so often the lower difficutly blocks happen to meet the original criteria. pool server submits to it's bitcoind and any rewards is distributed to the miners. That's my rough understanding, but i'm not entirely
2047 2013-06-28 18:42:35 <bmcgee> convinced it's correct. It's surprisingly difficult to find this stuff out, or maybe i'm looking in the wrong place
2048 2013-06-28 18:42:57 <gmaxwell> tonikt: and while I wouldn't demand as much as a simulation, I'd certantly do a lot of calculating or simulating before making that kind of change for myself in order to convince myself that some tradeoff was good (even absent attackers). Thats just good engineering.
2049 2013-06-28 18:43:13 <gmaxwell> bmcgee: so, not quite.
2050 2013-06-28 18:43:33 santoscork has joined
2051 2013-06-28 18:43:45 <gavinandresen> tonikt: an attacker can, with minimal bandwidth usage, cause a server to thrash their disk by asking for random transactions in random blocks.
2052 2013-06-28 18:43:50 <gmaxwell> bmcgee: There is a difficulty field in the block header which must be accurate. The pool server instead tells the miner a seperate target in addition to the correct header one.
2053 2013-06-28 18:44:19 <tonikt> gavinandresen: but the attacker can as well ask for random block - cant he?
2054 2013-06-28 18:44:37 <gmaxwell> tonikt: and generally, ignoring all thisâ probably any major changes to the p2p protocol should be tested by using a second p2p protocol. We could use some transport diversity in any case.
2055 2013-06-28 18:45:02 <bmcgee> gmaxwell: by separate target what do you mean?
2056 2013-06-28 18:45:03 <tonikt> gmaxwell: IMO changes I proposed do not care any side effects, so why do I need to convince you?
2057 2013-06-28 18:45:13 <gavinandresen> tonikt: sure, but then he has to have the bandwidth to "eat" the block that is delivered, and since blocks are a lot bigger than transactions you get many fewer disk seeks per second (which is the attack)
2058 2013-06-28 18:45:22 <tonikt> I am talking about extensions - that would keep backward compatibility
2059 2013-06-28 18:45:50 <tonikt> gavinandresen: sorry, it sis not a convincing argument
2060 2013-06-28 18:45:56 <gmaxwell> bmcgee: the miner knows that he's mining a XXXXXXXX difficulty block but he's been told to report back shares that meet difficulty x.
2061 2013-06-28 18:45:57 <gavinandresen> Anyway, do what gmaxwell says: go implement a variation on the existing p2p protocol, and then get everybody to use it because it is way better.
2062 2013-06-28 18:46:33 <gmaxwell> In the course of actually _implementing_ something you'll learn things that even I don't know, and in the end likely end up with something that is good... and the convincing will be easy.
2063 2013-06-28 18:46:59 <gmaxwell> Or at least you'll have a concrete proposal that I can look at and say _here_ is where I can make it catch fire.
2064 2013-06-28 18:47:08 <bmcgee> gmaxwell: ah, so the miner should be firing shares towards the pool pretty regularly, which helps verify it's not just sitting there doing nothing and taking the reward?
2065 2013-06-28 18:47:12 <gavinandresen> I still dont understand what problem tonikt has that needs solving....
2066 2013-06-28 18:47:17 <gmaxwell> bmcgee: Exactly.
2067 2013-06-28 18:47:28 sivu_ has left ()
2068 2013-06-28 18:47:38 <tonikt> The thing is that I am alone and testing a protocol that is suposed to download anew block in parts by myself, it wont work
2069 2013-06-28 18:47:39 <gmaxwell> gavinandresen: He wants to fetch single blocks in parallel because ... uh. Underpants!
2070 2013-06-28 18:47:50 sivu has joined
2071 2013-06-28 18:48:00 <gavinandresen> oh, well if it is Underpants, then I understand completely
2072 2013-06-28 18:48:00 <tonikt> gmaxwell: you know why because
2073 2013-06-28 18:48:06 <tonikt> because I dont live on Mars
2074 2013-06-28 18:48:17 <gavinandresen> tonikt: what user-visible problem are you solving?
2075 2013-06-28 18:48:24 <bmcgee> gmaxwell: so as the difficutly is adjusted the pool should adjust accordingly to keep regular submissions. not too low so as to get flooded, but not too high so as to make it easier for a miner to take the piss
2076 2013-06-28 18:48:36 <gmaxwell> tonikt: but you do live on a world that has latency, if not 40 minutes of it.
2077 2013-06-28 18:48:43 <tonikt> gavinandresen: 10 seconds needed to download a freshly mined block
2078 2013-06-28 18:48:52 <tonikt> .. I want to replace it with 1 second
2079 2013-06-28 18:48:56 <gavinandresen> ⦠because....
2080 2013-06-28 18:48:57 <gmaxwell> bmcgee: right.
2081 2013-06-28 18:49:09 <gavinandresen> Are you trying to get fewer stales for your mining pool or something?
2082 2013-06-28 18:49:24 <gavinandresen> (THAT would be a problem I could understand)
2083 2013-06-28 18:49:38 <tonikt> gavinandresen: sort of. I dont have a purpose other than making it more perfect
2084 2013-06-28 18:49:56 <gavinandresen> "making it more perfect" will make me say "okey dokey"
2085 2013-06-28 18:50:00 <pjorrit> so just try to solve it see what you come up with
2086 2013-06-28 18:50:01 <tonikt> its not about problems -its about progress
2087 2013-06-28 18:50:04 <bmcgee> gmaxwell: I can see your earlier point now though. With getblocktemplate there isn't much to stop a miner from finding the right block, submitting itself and just sending the pool lower difficulty shares. Or is there a mechanism to help with this?
2088 2013-06-28 18:50:13 <gmaxwell> gavinandresen: though his proposal above required adding an extra round trip time to fetching, and can't validate until the slowest peer respondsâ not the best way to start off reducing latency.
2089 2013-06-28 18:50:24 <gavinandresen> gmaxwell: "okey dokey"
2090 2013-06-28 18:50:44 <tonikt> gmaxwell: you still dont get it that you dont need to wait for the slowest peer, do you?
2091 2013-06-28 18:50:47 <gmaxwell> bmcgee: sure there is, the same one the works everwhere: You don't credit people unless the coinbase would have paid the pool.
2092 2013-06-28 18:50:51 <gavinandresen> ⦠I don't like angels-dancing-on-the-heads-of-pins technical arguments
2093 2013-06-28 18:51:15 <tonikt> yuo guys suck :)
2094 2013-06-28 18:51:27 <gmaxwell> gavinandresen: we gave him a simple way that didn't require the round trip to get a concurrent fetch: use bip37 with mutually exclusive filters.
2095 2013-06-28 18:51:27 <gavinandresen> ⦠and I'm extra grumpy because I just spent a whole day in win32/gitian/mingw/qt/openssl/static-compiling hell
2096 2013-06-28 18:51:39 <gmaxwell> Man that would make anyone grumpy. Sorry. :(
2097 2013-06-28 18:52:13 <pjorrit> that sounds horrible, here have a beer
2098 2013-06-28 18:52:25 <gmaxwell> Or a small drum of vodka? :P
2099 2013-06-28 18:52:27 <bmcgee> gmaxwell: so i need to understand the coinbase better then. that's awared if you submit the right block isnt it? and i'm guessing when it's awarded the entire network sees it, so the pool can see a miner decided to take the piss?
2100 2013-06-28 18:52:37 <tonikt> as an engineer I feel sorry for you. myslef, I always try to improve things, just for a fun of it. you - you seem to be saying, we are not touching it because nobody has reported a problem
2101 2013-06-28 18:52:44 <pjorrit> i just eh.. i'll have to get some vodka soon :)
2102 2013-06-28 18:52:47 <tonikt> .. expect me, but who I am anyway :)
2103 2013-06-28 18:52:51 <gmaxwell> tonikt: That isn't what anyone here is saying.
2104 2013-06-28 18:53:03 <tonikt> not anyone, no
2105 2013-06-28 18:53:27 <gavinandresen> tonikt: I might say that. Lots of seemingly innocuous improvements turn out to have serious security problems.
2106 2013-06-28 18:53:36 <pjorrit> there's always infinite things to do, solving non-problems is not something that goes on the todo list
2107 2013-06-28 18:53:46 <gavinandresen> ⦠so there has to be a significant benefit to any improvement, or the risk isn't worth it
2108 2013-06-28 18:54:17 <pjorrit> try solving your problem yourself see what new problems you run into see what incredible thnigs you'll learn
2109 2013-06-28 18:54:22 <tonikt> gavinandresen: this is an actual argument - if you can only make it specific
2110 2013-06-28 18:54:32 <gmaxwell> pjorrit: yea, I think thats always great advice.
2111 2013-06-28 18:55:17 <gavinandresen> tonikt: well, that is why I was asking what user-visible problem was being solved. If the benefit is only "theoretically more pure" then that is not enough.
2112 2013-06-28 18:55:30 wamatt has joined
2113 2013-06-28 18:55:44 <tonikt> gavinandresen: so, are you also against the idea of extending the net proto to be able to fetch tx/block along with its hash?
2114 2013-06-28 18:55:56 <tonikt> tx/block length *
2115 2013-06-28 18:56:10 <tonikt> I think its a good idea that has no risks
2116 2013-06-28 18:56:11 <gavinandresen> extending the inv message to include the length seems fine to me
2117 2013-06-28 18:56:18 <gavinandresen> ⦠haven't thought a lot about it, though
2118 2013-06-28 18:56:26 <gmaxwell> tonikt: "was asking what user-visible problem was being solved" ...
2119 2013-06-28 18:57:22 <tonikt> gavinandresen: that's great. at least one normal man here :)
2120 2013-06-28 18:57:25 <gavinandresen> and as gmaxwell hints: how is knowing the length before fetching useful?
2121 2013-06-28 18:58:02 <tonikt> about the getdata for tx in mined blocks - that would be the ultimate solution to allow downloading blocks in parts
2122 2013-06-28 18:58:04 <gavinandresen> I suppose you could ignore all transactions larger than 50Kbytes just because⦠but you'll have to fetch them if somebody else puts them in a block later....
2123 2013-06-28 18:58:08 <gmaxwell> tonikt: why are you unwilling to use BIP37 to accomplish your goal? This requires no further protocol changes. (and it will result in a more efficient implementation, since you don't take an extra round trip)
2124 2013-06-28 18:58:20 <tonikt> ... which will be very handy when blocks grown bigger
2125 2013-06-28 18:58:23 <gavinandresen> I don't see that anybody would ever ignore large blocks
2126 2013-06-28 18:58:55 <tonikt> gavinandresen: it would be only to discover peers to ban
2127 2013-06-28 18:59:13 <tonikt> yes, also to dont route big txs
2128 2013-06-28 18:59:20 <tonikt> but mostly against dishonest peers
2129 2013-06-28 18:59:23 brson has quit (Ping timeout: 246 seconds)
2130 2013-06-28 18:59:31 <tonikt> like the last time
2131 2013-06-28 18:59:40 <gavinandresen> we already don't route transactions of 100Kbytes big (as of a release or two ago)
2132 2013-06-28 19:00:08 <tonikt> you don't route - but it doesn't mean that you cannot get one
2133 2013-06-28 19:00:11 <gmaxwell> ⦠(dishonest peers: a problem that doesn't exist if BIP37 is used for parallel fetch.)
2134 2013-06-28 19:00:22 <tonikt> and what do you do when you get 10MB transaction?
2135 2013-06-28 19:00:25 <tonikt> nothing
2136 2013-06-28 19:00:31 <maaku> what's the difference between Sign() and SignCompact()?
2137 2013-06-28 19:01:03 <petertodd> ok, serious question: Lets suppose I want to make a P2P protocol change for some new experimental feature - what's the best way to mark nodes supporting my new feature so they can discover each other selectively, while still remaining a part of the main network? I'd use a service bit, but I'm more thinking long-term so multiple such experiments can be conducted.
2138 2013-06-28 19:01:29 <petertodd> For sake of argument, lets say I want to make a P2P flood-fill message layer.
2139 2013-06-28 19:01:38 <gavinandresen> tonikt: "okey dokey" -- I can send your node a 10MB "gavin_is_messing_with_you" message, and it will simply be ignored
2140 2013-06-28 19:01:45 <tonikt> gmaxwell: again, I cannot use BIP37 because it does not allow me to split my block into parts in an optimial way proportionally among my peers
2141 2013-06-28 19:02:15 <tonikt> gavinandresen: exactly - and according to the rules, I should not even ban you
2142 2013-06-28 19:02:34 <gmaxwell> tonikt: if he wants to do that though he can do it even if you ban him.
2143 2013-06-28 19:02:40 <tonikt> ... but if the inv gave me the size..
2144 2013-06-28 19:02:49 wamatt has quit (Read error: Operation timed out)
2145 2013-06-28 19:03:05 <tonikt> gmaxwell: I will ban him at the first byte extra over the size given before
2146 2013-06-28 19:03:07 bmcgee has quit (Quit: bmcgee)
2147 2013-06-28 19:03:10 <tonikt> ... so I do not mind
2148 2013-06-28 19:03:15 <gmaxwell> tonikt: "split my block into parts in an optimial way proportionally among my peers" how do you define optimal?
2149 2013-06-28 19:03:35 <tonikt> gmaxwell: I will do it in a realtime
2150 2013-06-28 19:03:50 <gmaxwell> tonikt: or rather, why are you using a defintion which will result in _slower_ transfers. I thought the problem you were trying to solve was to minimize the time to fetch a block?
2151 2013-06-28 19:03:56 <tonikt> I will split my block into N jobs and execute these jobs among M peers
2152 2013-06-28 19:04:36 <gmaxwell> Right and with BIP37 you can slip your block into N jobs and execute those jobs among M peers also.
2153 2013-06-28 19:04:36 mappum has joined
2154 2013-06-28 19:04:38 <tonikt> gmaxwell: it does not result in slower transfers - you have to arguments to support that, excpet your Mars theory
2155 2013-06-28 19:05:21 <tonikt> gmaxwell: no, you cannot, because it is all about statistics and you can never be sure that your hashing pttrern has covered it all
2156 2013-06-28 19:05:43 <gmaxwell> You're waving your arms there.
2157 2013-06-28 19:05:59 <tonikt> no, I'm actually rolling a join :)
2158 2013-06-28 19:06:12 <tonikt> fingers on the paper :)
2159 2013-06-28 19:06:12 <gmaxwell> Adding a round trip making things slower is pretty simple. Take your bandwidth, multiply it by the round trip time and thats how much data you could have sent instead.
2160 2013-06-28 19:06:34 <tonikt> the round trip is like 100-200ms
2161 2013-06-28 19:06:41 <tonikt> downlaodin 1MB of data is like 10s
2162 2013-06-28 19:06:48 <gmaxwell> sure, so take 200 ms, a 10mbit link could have sent a whole block in that time.
2163 2013-06-28 19:06:50 <tonikt> start counting man
2164 2013-06-28 19:07:09 <tonikt> no, it could not
2165 2013-06-28 19:07:10 A2501 has quit (Remote host closed the connection)
2166 2013-06-28 19:07:35 <gmaxwell> Yes it could, an average size block transfers in 200ms on 10 mbit. (ignoring the half rtt for the request)
2167 2013-06-28 19:07:52 <gmaxwell> (and the latency to fetch the block and such)
2168 2013-06-28 19:07:53 <maaku> are compact signatures only used for message signing?
2169 2013-06-28 19:07:56 <tonikt> pings on 10mb link are much slower
2170 2013-06-28 19:08:00 <gmaxwell> maaku: yes.
2171 2013-06-28 19:08:07 <tonikt> when I say 200ms, I mean more 256kb
2172 2013-06-28 19:08:33 <tonikt> lower, not slower :)
2173 2013-06-28 19:09:25 <tonikt> i'm sure you can do the math
2174 2013-06-28 19:09:29 andyh2 has quit (Quit: Leaving...)
2175 2013-06-28 19:09:59 <gmaxwell> Again, what are you trying to accomplish? Yes, BIP37 will cause some small imbalancing, though for 1000 transactions in a block we'd expect the imbalance to be very very small. To try to avoid it you'd instead propose adding a lot of overhead, an extra round trip, and a protocol change.
2176 2013-06-28 19:10:48 <tonikt> gmaxwell: BIP37 is completely not feasible to split a block download into parts that would only download txs that I dont know
2177 2013-06-28 19:10:59 <tonikt> dont you understand that?
2178 2013-06-28 19:11:03 <gmaxwell> tonikt: it exactly achieve that, for free in fact.
2179 2013-06-28 19:11:08 <gmaxwell> er achieves*
2180 2013-06-28 19:11:17 <maaku> gmaxwell: thank you
2181 2013-06-28 19:11:36 <gmaxwell> tonikt: it will already exclude transactions that it previously told you about or ones you told it about, without requiring additional data transmission or latency.
2182 2013-06-28 19:11:50 <gmaxwell> "dont you understand that?"
2183 2013-06-28 19:11:53 <gmaxwell> :P
2184 2013-06-28 19:12:17 <tonikt> what if I miss one transaction in a new block - how do I ask for it using BIP37?
2185 2013-06-28 19:13:07 <BCB> https://bitcointalk.org/index.php?topic=245689.0
2186 2013-06-28 19:13:12 <gmaxwell> tonikt: you ask for the block with bip37 and it will magically not tell you ones it knows you already have. Even better, it does it in a way where you can detect if the peer is dishonest about the block content without comparing to another peer.
2187 2013-06-28 19:13:32 <gmaxwell> BCB: thats a really useless pool. What does it even mean?!?
2188 2013-06-28 19:13:43 <tonikt> gmaxwell: and how does it know which txs I have?
2189 2013-06-28 19:14:15 <BCB> gmaxwell: it is actually interesting to me. Please fell free to express your opinion there.
2190 2013-06-28 19:14:44 <gmaxwell> BCB: Are you for or against FREEDOM??? thats what I want to know!
2191 2013-06-28 19:15:03 <gmaxwell> tonikt: because it knows what it has sent you, and it knows what you've advertised to it already.
2192 2013-06-28 19:15:23 <BCB> kind of intentionally left open ended to get a snap shot of what "regulation" means to the community (who care to vote or post).
2193 2013-06-28 19:15:25 wamatt has joined
2194 2013-06-28 19:15:31 <gmaxwell> tonikt: and if you are using BIP37 to do fractional fetching, it knows which fractions you are not requesting (because you requested them from someone else)
2195 2013-06-28 19:16:22 darkee has quit (Ping timeout: 240 seconds)
2196 2013-06-28 19:16:31 <gmaxwell> BCB: someone who thinks "regulation means putting theives in jail" might say I'm all for thatâ while someone who thinks "regulation means installing spyware to observe every transaction I make" will say heck no. Even if those two people are otherwise identical in their opinions.
2197 2013-06-28 19:16:47 <tonikt> gmaxwell: so you had said me that 50+% overhead is too much, but now you tell me to ask each of my 20+ nodes to give me everything that they have not sent me before?
2198 2013-06-28 19:17:18 <BCB> I agree, however the polling has already produced an interesting trend.
2199 2013-06-28 19:17:19 <tonikt> it is abusing a protocol that was invented for something completely else
2200 2013-06-28 19:17:44 <gmaxwell> tonikt: Using it for this was a design consideration, and the protocol was revised to better accomidate this use case, in fact.
2201 2013-06-28 19:18:02 <gmaxwell> tonikt: And you're still not understanding.
2202 2013-06-28 19:18:14 <gmaxwell> Perhaps you should consider decreasing your use of memory impeading drugs.
2203 2013-06-28 19:18:17 darkee has joined
2204 2013-06-28 19:18:21 <tonikt> gmaxwell: I think I understand it very well
2205 2013-06-28 19:18:25 <gmaxwell> impeding*
2206 2013-06-28 19:18:33 <tonikt> and I have analyzed it
2207 2013-06-28 19:18:41 <gmaxwell> You clear do not and have not.
2208 2013-06-28 19:19:20 <tonikt> there is no way to split a block download among peers in an optimal way, and still avoid far over 50% overhead
2209 2013-06-28 19:19:20 <gmaxwell> You can ask each peer to give you a mutually exclusive fragment of the block, minus everything it has sent you or seen you advertise to it previously.
2210 2013-06-28 19:19:58 <tonikt> I need specific tx ids - bloom filters are not suitable for that. I dont understand why you deny it
2211 2013-06-28 19:20:45 peter has joined
2212 2013-06-28 19:20:45 peetaur2 has quit (Ping timeout: 246 seconds)
2213 2013-06-28 19:20:53 <gmaxwell> tonikt: What user visible problem are you trying to solve where you need specific txids?
2214 2013-06-28 19:21:08 peter is now known as Guest35468
2215 2013-06-28 19:22:22 <tonikt> :) oh please. maybe you are on too much drugs. i've been saying it all the time: download a new block as quickly as possible, by splitting the block's payload download among all of my peers
2216 2013-06-28 19:22:43 <gavinandresen> I vote we all ignore tonikt now.
2217 2013-06-28 19:23:08 <gmaxwell> gavinandresen: thank you for not stabbing me instead. :)
2218 2013-06-28 19:23:09 <tonikt> it can be done - I can do it, but there is no protocol to get advantage on
2219 2013-06-28 19:23:51 * Luke-Jr is tempted to make a forum poll with totally gibberish answers and see what the results are
2220 2013-06-28 19:23:53 <tonikt> yeah, you guys keep ignoring me - this way you don't need to address hard issues
2221 2013-06-28 19:24:38 <sipa> i think that's actually it
2222 2013-06-28 19:24:43 <sipa> we need to address hard issurs
2223 2013-06-28 19:24:57 <sipa> not this one that can be easily solved when it becomes a problem
2224 2013-06-28 19:25:04 <tonikt> so which hard issues are you addressing instead?
2225 2013-06-28 19:25:34 <petertodd> tonikt: payment protocol
2226 2013-06-28 19:25:48 <nsh> tonikt, https://github.com/bitcoin/bitcoin/issues (371 open, 2,429 closed)
2227 2013-06-28 19:25:50 <tonikt> i know you sipa - you made a great improvement in the ec math
2228 2013-06-28 19:25:58 <sipa> heh
2229 2013-06-28 19:26:06 <sipa> if hat's all i get credit for
2230 2013-06-28 19:26:11 <tonikt> though, dont you think that a similar improvement in block propagation time, would be useful?
2231 2013-06-28 19:26:20 catcowllama has joined
2232 2013-06-28 19:26:20 catcowllama has quit (Changing host)
2233 2013-06-28 19:26:20 catcowllama has joined
2234 2013-06-28 19:26:23 <sipa> yes, when it becomes a problem
2235 2013-06-28 19:26:33 <sipa> tackle low-hanging fruit first
2236 2013-06-28 19:26:45 <tonikt> but ec math has not been a problem yet
2237 2013-06-28 19:26:47 <sipa> that used to be database processing
2238 2013-06-28 19:26:52 <tonikt> and yet, you made it
2239 2013-06-28 19:26:53 wamatt has quit (Quit: wamatt)
2240 2013-06-28 19:27:00 <tonikt> progress - thats all that matters
2241 2013-06-28 19:27:02 <Luke-Jr> sipa: block propagation time is already a problem
2242 2013-06-28 19:27:08 catcow has quit (Ping timeout: 252 seconds)
2243 2013-06-28 19:27:21 <sipa> Luke-Jr: sure, but not because of single peer bandwidth limitations
2244 2013-06-28 19:27:27 <Luke-Jr> right
2245 2013-06-28 19:27:35 <tonikt> yo dont know why
2246 2013-06-28 19:27:43 <tonikt> I dare to say that also because of that reasons
2247 2013-06-28 19:28:01 <tonikt> if you can turn 8s into 1s
2248 2013-06-28 19:28:04 <tonikt> thats a change
2249 2013-06-28 19:28:15 eoss has joined
2250 2013-06-28 19:28:21 <sipa> tonikt: database processing in bitcoin was slow, so i rewrote the database/validation logic
2251 2013-06-28 19:28:41 <petertodd> tonikt: come back to us when you know what that 7 seconds does for your orphan rate, and by "know" give us the worked out equations
2252 2013-06-28 19:28:41 <Luke-Jr> tonikt: I suggest you might want to study the code a bit and help out, instead of just complaining
2253 2013-06-28 19:28:45 <sipa> then signature verification became a bottleneck, so i added parallel verification
2254 2013-06-28 19:29:12 <sipa> then it was still slowing things down when reindexing, so i worked on speeding up the ec math
2255 2013-06-28 19:29:22 <tonikt> Luke-Jr: I proposed some simple changes in the protocol, that were laughed at
2256 2013-06-28 19:29:28 <sipa> next will be network improvements indeed
2257 2013-06-28 19:29:32 <Luke-Jr> tonikt: where's the code?
2258 2013-06-28 19:29:37 <sipa> but there is much lower handing fruit than this
2259 2013-06-28 19:29:49 <sipa> like for example actually downloading blocks in parallel
2260 2013-06-28 19:29:50 <tonikt> Luke-Jr: it was a net protocol change proposal
2261 2013-06-28 19:30:00 <Luke-Jr> sipa: IMO HD wallets are higher priority :p
2262 2013-06-28 19:30:13 <Luke-Jr> tonikt: net is nothign without code
2263 2013-06-28 19:30:37 <tonikt> Really I think if we could combine downloading blocks in paralell with sipa's new path, that wold improve blocks propagation dramatically
2264 2013-06-28 19:30:47 <sipa> right now we couldn't even use tonikt 's protocol even if it gave an advantage
2265 2013-06-28 19:30:53 <petertodd> tonikt: either come back with some nice equations, or better yet, come back with an implementation/simulation and convince us with data
2266 2013-06-28 19:30:53 <tonikt> And this is something that the devs could be proud of
2267 2013-06-28 19:30:58 <sipa> as we don't even ffing download blocks in parallel
2268 2013-06-28 19:31:01 <Luke-Jr> tonikt: what? parallel block downloads do *nothing* for propagation time
2269 2013-06-28 19:31:08 <tonikt> instead of just waiting for thing to get wrong and fic them then
2270 2013-06-28 19:31:29 <sipa> tonikt: premature optimization is the root of all evil
2271 2013-06-28 19:31:40 <sipa> this is not a math problem
2272 2013-06-28 19:31:42 <rdymac_> does anyone knows about what happens to unSYSTEM.net webpage?
2273 2013-06-28 19:31:48 <sipa> this is a live and hard to control system
2274 2013-06-28 19:31:50 <tonikt> guys, I can make a client that gives you my proposed API, but lets face it
2275 2013-06-28 19:31:58 <tonikt> none of you will dare to connect to it :)
2276 2013-06-28 19:32:02 <sipa> no
2277 2013-06-28 19:32:06 <sipa> we couldn't
2278 2013-06-28 19:32:13 <tonikt> not to mention getting adventage of it
2279 2013-06-28 19:32:18 <tonikt> exactly
2280 2013-06-28 19:32:20 <sipa> as i said: much lower hanging fruit
2281 2013-06-28 19:32:22 <petertodd> sipa: heh, it's not a math problem, and tonikt can't even give us some equations of a simplified model anyway
2282 2013-06-28 19:32:44 <gavinandresen> speaking of low-hanging fruit⦠is the pull-tester still broken?
2283 2013-06-28 19:32:57 justusranvier has quit (Ping timeout: 268 seconds)
2284 2013-06-28 19:32:57 <tonikt> petertodd: I can ell you that you need about 10s to download a block from 1mbps peer
2285 2013-06-28 19:33:07 <sipa> gavinandresen: matt was kinda busy this week...
2286 2013-06-28 19:33:16 <tonikt> petertodd: if this number is not convincing enough for you
2287 2013-06-28 19:33:25 <tonikt> .. then I have nothing better
2288 2013-06-28 19:34:56 <petertodd> tonikt: I can tell you you've already misunderstood the problem. I actually looked into this exact issue a week or two ago, looked into the comp-sci literature etc. It's not as simple a problem as you think.
2289 2013-06-28 19:35:14 <tonikt> Why?
2290 2013-06-28 19:35:41 <tonikt> I just need a net API that would allow me to download a specific tx from a specific mined block?
2291 2013-06-28 19:35:48 <tonikt> Is it really so hard to make?
2292 2013-06-28 19:36:03 <petertodd> tonikt: You think you need that API, you're wrong.
2293 2013-06-28 19:36:20 <tonikt> I dont need an API?
2294 2013-06-28 19:36:32 <tonikt> Bloom filters?
2295 2013-06-28 19:36:55 mrkent has joined
2296 2013-06-28 19:36:55 mrkent has quit (Changing host)
2297 2013-06-28 19:36:55 mrkent has joined
2298 2013-06-28 19:37:37 grau_ has quit (Remote host closed the connection)
2299 2013-06-28 19:38:05 <Luke-Jr> tonikt: first you flood the block headers, followed immediately by a list of txids in it. then you can get the txs you don't have by txid the normal way
2300 2013-06-28 19:38:09 <tonikt> Do I really need to give you numbers that downloading a block from which 50% txs I already have while the other half I can just split among the 20+ peers is at least 10x faster than downloadin this block from the first oeer that sent me the inv?
2301 2013-06-28 19:38:16 <petertodd> tonikt: Do you know what a random graph is?
2302 2013-06-28 19:38:53 <tonikt> Luke-Jr: The problem is that currently you cannot get tx data for mined blocks
2303 2013-06-28 19:39:06 <tonikt> petertodd: no
2304 2013-06-28 19:39:20 <petertodd> tonikt: Then go learn before you waste our time any further.
2305 2013-06-28 19:39:30 <Luke-Jr> tonikt: sure you can
2306 2013-06-28 19:39:33 melvster has quit (Read error: No route to host)
2307 2013-06-28 19:39:38 wamatt has joined
2308 2013-06-28 19:39:41 <tonikt> petertodd: I dont feel like going now. Its raining
2309 2013-06-28 19:39:41 justusranvier has joined
2310 2013-06-28 19:39:44 melvster has joined
2311 2013-06-28 19:40:01 <petertodd> /ignore tonikt
2312 2013-06-28 19:40:14 <tonikt> Luke-Jr: No, you cant. Thats the problem - and that is what I prosed
2313 2013-06-28 19:41:04 <tonikt> Luke-Jr: If you could ask a node for a specific tx - that would be perfect.
2314 2013-06-28 19:41:06 <Luke-Jr> yeah, i think my patience is about out too
2315 2013-06-28 19:41:30 <Luke-Jr> tonikt: YOU CAN.
2316 2013-06-28 19:41:33 * Luke-Jr is done
2317 2013-06-28 19:41:45 <tonikt> Luke-Jr: tell me how
2318 2013-06-28 19:43:39 Guest35468 has quit (Read error: Operation timed out)
2319 2013-06-28 19:45:08 Guest35468 has joined
2320 2013-06-28 19:50:53 wamatt has quit (Quit: wamatt)
2321 2013-06-28 19:57:31 RoboTeddy has quit (Remote host closed the connection)
2322 2013-06-28 20:00:56 daybyter has quit (Quit: Konversation terminated!)
2323 2013-06-28 20:05:23 gavinandresen has quit (Quit: gavinandresen)
2324 2013-06-28 20:07:13 paracyst has joined
2325 2013-06-28 20:07:23 xire has joined
2326 2013-06-28 20:07:23 xire has quit (Changing host)
2327 2013-06-28 20:07:23 xire has joined
2328 2013-06-28 20:09:36 RazielZ has quit (Ping timeout: 256 seconds)
2329 2013-06-28 20:11:56 agnostic98 has quit (Remote host closed the connection)
2330 2013-06-28 20:17:38 RoboTeddy has joined
2331 2013-06-28 20:17:42 RoboTeddy has quit (Remote host closed the connection)
2332 2013-06-28 20:17:59 RoboTeddy has joined
2333 2013-06-28 20:18:02 CodeShark has joined
2334 2013-06-28 20:18:20 CodeShark has quit (Client Quit)
2335 2013-06-28 20:18:41 CodeShark has joined
2336 2013-06-28 20:19:20 mrkent has quit (Ping timeout: 256 seconds)
2337 2013-06-28 20:19:35 yubrew has quit (Remote host closed the connection)
2338 2013-06-28 20:25:30 mappum has quit (Ping timeout: 246 seconds)
2339 2013-06-28 20:27:38 grau has joined
2340 2013-06-28 20:29:12 digitalmagus2 has joined
2341 2013-06-28 20:30:00 sandbote has quit (Remote host closed the connection)
2342 2013-06-28 20:32:51 freewil has quit (Ping timeout: 256 seconds)
2343 2013-06-28 20:32:57 grau has quit (Remote host closed the connection)
2344 2013-06-28 20:33:14 mappum has joined
2345 2013-06-28 20:34:19 <BCB> anyone have an opinion on ripple
2346 2013-06-28 20:35:37 sandbote has joined
2347 2013-06-28 20:36:10 <Luke-Jr> BCB: hard to have an opinion on it when the code is not available for review!
2348 2013-06-28 20:36:29 <sipa> Luke-Jr: ask Vinnie_win :)
2349 2013-06-28 20:37:25 BTCOxygen has quit (Ping timeout: 246 seconds)
2350 2013-06-28 20:37:26 JKL1234- has quit (Changing host)
2351 2013-06-28 20:37:26 JKL1234- has joined
2352 2013-06-28 20:40:52 <BCB> Luke-Jr: that seems to be the common opinion
2353 2013-06-28 20:42:46 patcon has joined
2354 2013-06-28 20:43:44 <CodeShark> ripple as an idea is awesome - my only two gripes: 1) XRP distribution is centralized, 2) server implementation has a very small developer base and is not open source
2355 2013-06-28 20:44:40 <gmaxwell> CodeShark: you don't share my gripe?
2356 2013-06-28 20:44:49 agnostic98 has joined
2357 2013-06-28 20:44:54 <CodeShark> what's your gripe, gmaxwell?
2358 2013-06-28 20:45:26 xire has quit (Ping timeout: 246 seconds)
2359 2013-06-28 20:46:23 <MC1984_> so have i got this straight
2360 2013-06-28 20:46:31 <gmaxwell> well I have twoâ that in order to even have XRP they imposed a need for global consensus on something which should only require local consensus (e.g. the gateway could effectively track who owns which of its tokens), and they've adopted a system for global consensus which isn't obviously viable and convergent.
2361 2013-06-28 20:46:38 jgarzik has quit (Quit: weekend)
2362 2013-06-28 20:47:12 <gmaxwell> And no one involved with ripple can explain to me the minimum necessary criteria for their consensus to be convergent.
2363 2013-06-28 20:47:24 <MC1984_> bitcoinj now has a thing where an off chain txn takes place via direct TCP socket between clients, and is essentially bonded via threat of issuance of a real bitcoin txn if someone doesnt play fair
2364 2013-06-28 20:47:24 <sipa> "everyone agrees"
2365 2013-06-28 20:47:28 <gmaxwell> ("all servers run by one party" is obviously convergent, but not at all decenteralized)
2366 2013-06-28 20:48:52 <gmaxwell> CodeShark: you can read down in this thread for my thoughts on that: https://bitcointalk.org/index.php?topic=144471.0
2367 2013-06-28 20:48:56 BTCOxygen has joined
2368 2013-06-28 20:49:09 <MC1984_> and the microtxn are eventually settled in one big lump every so often
2369 2013-06-28 20:49:56 <MC1984_> seems like a great idea but txn under 5430s still cannot be bonded right
2370 2013-06-28 20:51:53 justusranvier has quit (Ping timeout: 268 seconds)
2371 2013-06-28 20:53:20 BTCOxygen has quit (Ping timeout: 256 seconds)
2372 2013-06-28 20:54:09 brson has joined
2373 2013-06-28 20:54:32 <CodeShark> oh, a third gripe (actually more of an uncertainty than a gripe) on which I agree with you, gmaxwell - UNL distribution
2374 2013-06-28 20:54:47 agnostic_ has joined
2375 2013-06-28 20:55:12 BTCOxygen has joined
2376 2013-06-28 20:55:20 <gmaxwell> yea, basically my second point above is deeply tied to the content and distribution of the UNL.
2377 2013-06-28 20:56:50 <gmaxwell> As I show later in the thread there are some trust topoligies possible which are _sure_ to get wedged. So you have to have some way to prevent them or resolve the wedges. Some of the comments about the security model that were made before made it sound like wedges were deeply impossible to resolve (stuff like blacklisting any 'cheating' nodes).
2378 2013-06-28 20:58:05 santoscork has quit (Quit: Quiet while I make like a cat)
2379 2013-06-28 20:58:18 agnostic98 has quit (Ping timeout: 248 seconds)
2380 2013-06-28 20:58:50 justusranvier has joined
2381 2013-06-28 20:59:20 <CodeShark> I think the only thing that would keep consensus is that the biggest players on the network would have an incentive not to try to cheat
2382 2013-06-28 21:00:08 <gmaxwell> There is also something that I'd call an "altcoin attack". Say, for example, this premine that put like 80 billion XRP in the hands of a few people.
2383 2013-06-28 21:00:14 <CodeShark> yeah
2384 2013-06-28 21:00:20 <CodeShark> I know what you're about to say :)
2385 2013-06-28 21:00:25 <CodeShark> I've had that concern, too
2386 2013-06-28 21:00:34 <gmaxwell> If the software were open source a competing clique could show up and say "trust us instead and we'll reassign the 80 billion equally to all users"
2387 2013-06-28 21:00:51 <CodeShark> right, just change a few constants and redeploy as a new network
2388 2013-06-28 21:01:08 <gmaxwell> and, of course, it's in everyone's interest to do do thatâ except the founders.. but where does it end?
2389 2013-06-28 21:01:37 adam3us has left ()
2390 2013-06-28 21:01:41 <CodeShark> so the founders are looking to get at least a few gateways up and running and to push the brand before open sourcing
2391 2013-06-28 21:01:53 <gmaxwell> (The same would be true for bitcoin IFF it were the norm for users to be SPV node only)
2392 2013-06-28 21:02:15 reneg_ has quit (Ping timeout: 246 seconds)
2393 2013-06-28 21:03:22 reneg has joined
2394 2013-06-28 21:03:45 reneg is now known as Guest68541
2395 2013-06-28 21:10:05 <CodeShark> as for convergence, you must sacrifice either C, A, or P :)
2396 2013-06-28 21:10:15 <CodeShark> ripple tries to sacrifice A
2397 2013-06-28 21:10:27 <CodeShark> when necessary
2398 2013-06-28 21:11:39 <CodeShark> but that could still leave it open to DoS
2399 2013-06-28 21:13:16 Guest35468 has quit (Quit: Konversation terminated!)
2400 2013-06-28 21:13:37 <gmaxwell> It's not clear to me exactly how that plays out though. E.g. Is it full quorum sensing where knocking out a couple nodes can cause a cascaded quorum failing and the whole thing goes down (when trust is sparse)?
2401 2013-06-28 21:13:56 <gmaxwell> or can knocking out a few nodes in the sparse trust model make it inconsistent?
2402 2013-06-28 21:14:04 deadweasel has quit (Quit: leaving)
2403 2013-06-28 21:14:29 freewil has joined
2404 2013-06-28 21:14:36 <gmaxwell> If the firstâ can that even be reliable enough with publically selected particiants? (even if not quite an anonymous network)
2405 2013-06-28 21:14:44 <CodeShark> I believe it is the first
2406 2013-06-28 21:17:21 tholenst has joined
2407 2013-06-28 21:19:43 sandbote has quit (Remote host closed the connection)
2408 2013-06-28 21:21:31 <CodeShark> and ripple is certainly NOT an anonymous network
2409 2013-06-28 21:21:43 Krellan has joined
2410 2013-06-28 21:21:57 <midnightmagic> huh. i thought ripple would be OSS by now.
2411 2013-06-28 21:22:04 <gmaxwell> CodeShark: but is it decenteralized? Can a non anonymous network be decenteralized?
2412 2013-06-28 21:22:38 <gmaxwell> (at leastâ can one be in a world where state level powers like to regulate the @#$@ out of things that look like ripple?)
2413 2013-06-28 21:24:57 <CodeShark> ripple puts fundamental costs to adding new validators to the network or creating new identities
2414 2013-06-28 21:26:34 <CodeShark> gmaxwell: as long as it is difficult for new validators to jump into the network a government can target the few trusted validators
2415 2013-06-28 21:27:31 <CodeShark> take down enough of them and the whole thing grinds to a halt
2416 2013-06-28 21:29:02 Elmf has quit (Ping timeout: 240 seconds)
2417 2013-06-28 21:29:24 <CodeShark> the great thing about bitcoin is that the more that governments try to take down validators (i.e. miners) the more incentive there is for miners elsewhere to join...and miners do not need to be trusted by anyonw
2418 2013-06-28 21:30:13 wamatt has joined
2419 2013-06-28 21:32:56 <petertodd> CodeShark: Don't assume governments will take down miners. IMO much more likely is for them to simply ask them to abide by some simple regulations...
2420 2013-06-28 21:33:31 <CodeShark> petertodd: I was trying to present a worst-case scenario
2421 2013-06-28 21:33:38 <CodeShark> not the most likely scenario
2422 2013-06-28 21:34:02 <gmaxwell> petertodd: its not even really compariable, the trusted clique in ripple can freely reassign balances.
2423 2013-06-28 21:34:17 <petertodd> CodeShark: So was I - we're better off IMO if miners are shutdown than simply harnessed.
2424 2013-06-28 21:34:21 <gmaxwell> so not only is 'mining' not anonymous, full control of the state of the system is not anonymous.
2425 2013-06-28 21:35:33 <petertodd> gmaxwell: That's a giant target I agree...
2426 2013-06-28 21:37:16 <petertodd> gmaxwell: A better ripple would, as you know, ditch the consensus. Really interesting would be applying fancy zero-knowledge crypto to the process of finding paths through the social trust networks.
2427 2013-06-28 21:37:17 patcon has quit (Remote host closed the connection)
2428 2013-06-28 21:37:45 <petertodd> gmaxwell: IE, do keep participants maximally anonymous.
2429 2013-06-28 21:37:51 patcon has joined
2430 2013-06-28 21:38:14 qbasicer has quit (Remote host closed the connection)
2431 2013-06-28 21:38:54 <CodeShark> how do you do decentralized ordering?
2432 2013-06-28 21:39:14 <CodeShark> who decides the correct order in which transactions are applied?
2433 2013-06-28 21:40:12 A2501 has joined
2434 2013-06-28 21:40:13 <petertodd> CodeShark: Can you get away without that? Can the consensus on order be local to your trust group?
2435 2013-06-28 21:40:13 <gmaxwell> CodeShark: There isn't any need for that in classical ripple. The only parties involved in a transaction need to come to agreement, because they're not exchanging a global resource.
2436 2013-06-28 21:40:44 <CodeShark> gmaxwell: say you put an offer to buy X of currency A for a maximum of Y of currency B
2437 2013-06-28 21:40:59 <gmaxwell> CodeShark: e.g. If you want to accept a gmaxwell IOU from petertodd there is no reason that anymore more than gmaxwell, petertodd, and you need to be involved.
2438 2013-06-28 21:41:14 <CodeShark> you want to get the best price possible
2439 2013-06-28 21:42:03 <petertodd> CodeShark: That's an anti-spam problem, not a consensus problem. When the party buying your coins commits, they find the trust path and only the members of that path need to have consensus.
2440 2013-06-28 21:42:30 patcon has quit (Ping timeout: 246 seconds)
2441 2013-06-28 21:43:41 <CodeShark> gmaxwell: what if petertodd only has IOUs from issuer XYZ but I only accept gmaxwell IOUs?
2442 2013-06-28 21:44:07 <CodeShark> it then becomes necessary to involve someone else
2443 2013-06-28 21:44:13 Elmf has joined
2444 2013-06-28 21:44:27 <gmaxwell> CodeShark: (re above: what you want for that is unjammablity, not consensusâ it doesn't matter if you and bob see subtly different orderbooks, to get the best price you just need to see the relevant best order)
2445 2013-06-28 21:44:30 <petertodd> CodeShark: That's the whole point of ripple: you find a path. But the consensus along the path is localized to pairs (?) of participants because it's just IOU's between mutually trusting entities.
2446 2013-06-28 21:44:43 <gmaxwell> CodeShark: sure, but then just that one extra person. Not the whole world.
2447 2013-06-28 21:44:54 <gmaxwell> Or the N extra people, stillâ not the whole world.
2448 2013-06-28 21:45:14 Guest83631 has quit (Ping timeout: 248 seconds)
2449 2013-06-28 21:46:05 <gmaxwell> considering that high level of centeralization on trusted nodes and gateways that even the weird consensus-ripple has and can't dispense with... this doesn't seem like a big constraint to me, esp considering the big scaling advantage it can give (no global consensus) and availablity benefits (no global consensus failures)
2450 2013-06-28 21:47:10 <CodeShark> is the problem tractable?
2451 2013-06-28 21:47:39 Guest68541 has quit (Ping timeout: 256 seconds)
2452 2013-06-28 21:48:00 <CodeShark> you guys might be right - gonna have to ponder this some more
2453 2013-06-28 21:48:11 <petertodd> CodeShark: which problem exactly?
2454 2013-06-28 21:48:33 <CodeShark> pathfinding that relies solely on pairwise negotiation
2455 2013-06-28 21:48:48 jMyles has quit (Ping timeout: 264 seconds)
2456 2013-06-28 21:48:51 <CodeShark> and a way of achieving cryptographic proof
2457 2013-06-28 21:49:43 <petertodd> CodeShark: Of course it's tractable! Pathfinding isn't a big deal.
2458 2013-06-28 21:50:12 <petertodd> CodeShark: Insects do decentralized pathfinding all the time.
2459 2013-06-28 21:50:40 <CodeShark> the ant cryptocurrency - lol
2460 2013-06-28 21:50:49 <CodeShark> or the ant crypto payment network
2461 2013-06-28 21:51:34 agnostic98 has joined
2462 2013-06-28 21:51:35 agnostic_ has quit (Read error: Connection reset by peer)
2463 2013-06-28 21:51:51 <gmaxwell> Yea, the thing is that you don't actually need consensus for pathfinding. The system can give different paths for peter and I even if we have the same trust graphs.
2464 2013-06-28 21:51:56 <petertodd> Remember too that it's ok if a path winds up half-completed; at every step of the way you've just made a change to the balance between two peers.
2465 2013-06-28 21:52:30 <gmaxwell> There is some harryness about doing an atomic release of a half completed path so that a new path could be established.
2466 2013-06-28 21:52:37 <petertodd> There is a DoS potential if you make lots of useless adjustments, but anti-DoS is a much easier problem than consensus.
2467 2013-06-28 21:52:57 <gmaxwell> But you could just invoke a random beacon or other simple consensus-timestamp deadline for that.
2468 2013-06-28 21:53:15 <petertodd> gmaxwell: You sure about that? A half completed path just means debt has been pushed around - push it back.
2469 2013-06-28 21:53:27 <gmaxwell> and in practice I would expect that most trust paths would be short and thus thats a non-issue. E.g. pairwise can't fail half completed.
2470 2013-06-28 21:53:41 Namworld has joined
2471 2013-06-28 21:53:57 <gmaxwell> petertodd: you don't want to push it back unless you're sure it actually failed. And one of the nodes in the trust path has gone offline.
2472 2013-06-28 21:54:47 <gmaxwell> petertodd: and the network is capacated: I'm only willing to lend you X.
2473 2013-06-28 21:54:51 <petertodd> gmaxwell: Sure, but that's what timeouts and the like are for, and subsequent efforts to make the transaction go through are likely to reuse the partially completed path.
2474 2013-06-28 21:55:32 <gmaxwell> yea, sure. I do think there is still a corner case where an atomic kill is needed, but regardless: if thats true all you need is a clock.
2475 2013-06-28 21:56:04 <petertodd> Yeah, and it can probably be a regular clock, not a fancy random beacon thing.
2476 2013-06-28 21:57:28 Diapolis has quit (Read error: Connection reset by peer)
2477 2013-06-28 21:58:00 Diapolis has joined
2478 2013-06-28 22:00:13 Thepok has joined
2479 2013-06-28 22:02:02 Squid_ has joined
2480 2013-06-28 22:03:25 kwikness has quit (Quit: Leaving.)
2481 2013-06-28 22:05:09 Squidicuz has quit (Ping timeout: 252 seconds)
2482 2013-06-28 22:05:25 andyh2 has joined
2483 2013-06-28 22:05:29 Retik has joined
2484 2013-06-28 22:05:47 Subo1978 has quit (Remote host closed the connection)
2485 2013-06-28 22:06:14 Subo1978 has joined
2486 2013-06-28 22:10:22 tholenst has quit (Quit: Page closed)
2487 2013-06-28 22:12:24 Retik has quit (Quit: Leaving)
2488 2013-06-28 22:14:00 mappum has quit (Ping timeout: 276 seconds)
2489 2013-06-28 22:14:24 qbasicer has joined
2490 2013-06-28 22:14:32 <BW^-> main.cpp row 4119 and up to EOF is all exclusively about mining right?
2491 2013-06-28 22:14:36 <BW^-> src/main.cpp
2492 2013-06-28 22:15:08 dan_ has quit (Remote host closed the connection)
2493 2013-06-28 22:15:48 <BW^-> what's CDiskBlockPos and CDiskTxPos about?
2494 2013-06-28 22:15:56 <BW^-> they're two classes
2495 2013-06-28 22:17:07 <imd23> Hi all . :) I am still looking for a way to subscribe to addresses and be notified about deposits/transactions made to the them.
2496 2013-06-28 22:17:23 <imd23> any service you know?
2497 2013-06-28 22:18:05 qbasicer has quit (Remote host closed the connection)
2498 2013-06-28 22:22:38 i2pRelay has quit (Remote host closed the connection)
2499 2013-06-28 22:23:00 andyh2 has quit (Quit: Leaving...)
2500 2013-06-28 22:23:22 i2pRelay has joined
2501 2013-06-28 22:25:54 andyh2 has joined
2502 2013-06-28 22:26:41 <CodeShark> imd23: you weren't able to build CoinClasses?
2503 2013-06-28 22:28:26 imd23_ has joined
2504 2013-06-28 22:29:26 darkee has quit (Remote host closed the connection)
2505 2013-06-28 22:29:30 imd23 has quit (Ping timeout: 248 seconds)
2506 2013-06-28 22:29:30 imd23_ is now known as imd23
2507 2013-06-28 22:29:47 FabianB has joined
2508 2013-06-28 22:29:52 mappum has joined
2509 2013-06-28 22:30:05 darkee has joined
2510 2013-06-28 22:30:28 FabianB_ has quit (Ping timeout: 246 seconds)
2511 2013-06-28 22:30:51 A2501 has quit (Remote host closed the connection)
2512 2013-06-28 22:31:31 <TheLordOfTime> imd23: blockchain.info "watch address" feature
2513 2013-06-28 22:31:43 <TheLordOfTime> even if you don't import the privkey you can use their notifiers to watch addresses
2514 2013-06-28 22:31:55 <TheLordOfTime> not that it makes a nice solution, but...
2515 2013-06-28 22:31:57 <TheLordOfTime> still.
2516 2013-06-28 22:33:53 reneg has joined
2517 2013-06-28 22:33:55 agricocb has quit (Quit: Leaving.)
2518 2013-06-28 22:33:56 reneg is now known as Guest50296
2519 2013-06-28 22:33:59 Guest50296 has quit (Max SendQ exceeded)
2520 2013-06-28 22:34:25 reneg_ has joined
2521 2013-06-28 22:34:35 andyh2 has quit (Quit: Linkinus - http://linkinus.com)
2522 2013-06-28 22:37:59 Thepok has quit (Quit: Nettalk6 - www.ntalk.de)
2523 2013-06-28 22:39:57 A2501 has joined
2524 2013-06-28 22:42:45 eoss has quit (Remote host closed the connection)
2525 2013-06-28 22:47:25 <BW^-> does "nonce" carry a particular meaning here?
2526 2013-06-28 22:48:40 <chmod755> BW^-, http://en.wikipedia.org/wiki/Cryptographic_nonce
2527 2013-06-28 22:49:21 Luke-Jr has quit (Excess Flood)
2528 2013-06-28 22:49:56 parse_error has joined
2529 2013-06-28 22:50:11 Luke-Jr has joined
2530 2013-06-28 22:54:59 CodeName has joined
2531 2013-06-28 22:55:25 <BW^-> chmod755: aha
2532 2013-06-28 22:55:29 <BW^-> what's the point with the memory pool?
2533 2013-06-28 22:58:09 Luke-Jr has quit (Ping timeout: 260 seconds)
2534 2013-06-28 22:59:54 theorbtwo has quit (Ping timeout: 248 seconds)
2535 2013-06-28 23:02:43 <BW^-> in-RAM cache of tx:es, is that it?
2536 2013-06-28 23:02:59 * nsh frowns
2537 2013-06-28 23:04:33 <gmaxwell> BW^-: it's not a cache. It's storage for unconfirmed transactions.
2538 2013-06-28 23:05:12 theorbtwo has joined
2539 2013-06-28 23:05:51 <BW^-> gmaxwell: aha. all currently unconfirmed transactions there are in all the network and known to the peer, are stored there, in RAM?
2540 2013-06-28 23:06:01 CodeName has quit (Ping timeout: 252 seconds)
2541 2013-06-28 23:07:10 A2501 has quit (Remote host closed the connection)
2542 2013-06-28 23:10:18 <gmaxwell> assuming they pass the node's validation.
2543 2013-06-28 23:13:08 rlifchitz has quit (Ping timeout: 256 seconds)
2544 2013-06-28 23:18:54 <imd23> member:identifier:thelordoftime awesome. but I want to serve it myself. I need THAT implementation but in my servers :)
2545 2013-06-28 23:19:00 <imd23> thelordoftime
2546 2013-06-28 23:19:06 BTCOxygen has quit (Ping timeout: 248 seconds)
2547 2013-06-28 23:19:12 <imd23> thelordoftime do you know what do they use to implement that service?
2548 2013-06-28 23:19:38 one_zero has joined
2549 2013-06-28 23:21:59 <TheLordOfTime> imd23: unfortunately no. but i do know they watch the blockchain for stuff...
2550 2013-06-28 23:24:19 agricocb has joined
2551 2013-06-28 23:25:27 rlifchitz has joined
2552 2013-06-28 23:25:31 <imd23> thelordoftime : ok, and do you know how can I implement that? i mean, not from scratch but using any other already programmed client.. I don't care about the language...
2553 2013-06-28 23:27:28 sandbote has joined
2554 2013-06-28 23:29:17 <TheLordOfTime> also unfortunately no, but i'm after a similar implementation, if you find out how to do it give me a ping, because that'd be very useful for tracking one of my sites' donations addresses.
2555 2013-06-28 23:30:38 MobPhone has joined
2556 2013-06-28 23:32:00 NOTHING4YOU is now known as Nothing4You
2557 2013-06-28 23:32:34 sark2 has quit (!~sark@37-252-106-157.ip.skylogicnet.com|Ping timeout: 241 seconds)
2558 2013-06-28 23:35:15 rdponticelli has joined
2559 2013-06-28 23:36:01 sark2 has joined
2560 2013-06-28 23:41:25 <imd23> TheLordOfTime I am currently investigating bitcoinjs-server
2561 2013-06-28 23:41:31 <imd23> it's in js, nodejs.
2562 2013-06-28 23:44:11 sandbote has quit (Remote host closed the connection)
2563 2013-06-28 23:49:05 nimdAHK has joined
2564 2013-06-28 23:54:16 nova90 has quit (Quit: nova90)
2565 2013-06-28 23:57:48 super3 has quit (Ping timeout: 264 seconds)
2566 2013-06-28 23:59:18 mrkent has joined
2567 2013-06-28 23:59:18 mrkent has quit (Changing host)
2568 2013-06-28 23:59:18 mrkent has joined
2569 2013-06-28 23:59:22 reneg has joined
2570 2013-06-28 23:59:45 reneg is now known as Guest38532