1 2011-04-19 00:01:16 <luke-jr> midnightmagic: not quite the latter bit
2 2011-04-19 00:01:36 <luke-jr> mining by nature expands until difficulty is just barely high enough that the profit is minimized
3 2011-04-19 00:01:46 <luke-jr> but this also means everyone will be in pools
4 2011-04-19 00:01:54 <luke-jr> due to the 2 week cutoff
5 2011-04-19 00:02:04 <luke-jr> so there will be a few select pools with all the control
6 2011-04-19 00:02:59 <midnightmagic> no, people will be loathe to join pools because of the trust thing.
7 2011-04-19 00:03:21 <midnightmagic> i'm not talking long-tail people, i'm talking miners with >5Gh right now.
8 2011-04-19 00:03:41 <luke-jr> so there will be a few lone big-miners and pools
9 2011-04-19 00:04:17 <midnightmagic> mining is still incredibly important for exactly the reason i stated.
10 2011-04-19 00:04:33 Lachesis has quit (Read error: Operation timed out)
11 2011-04-19 00:05:13 <lianj> no mining, no transactions :)
12 2011-04-19 00:07:09 theorb has joined
13 2011-04-19 00:07:18 theorbtwo has quit (Ping timeout: 252 seconds)
14 2011-04-19 00:07:24 theorb is now known as theorbtwo
15 2011-04-19 00:08:18 <eamon> not really
16 2011-04-19 00:08:46 <eamon> all the mining could be done by one computer
17 2011-04-19 00:09:01 <eamon> so its not important at all
18 2011-04-19 00:09:09 <midnightmagic> and then the whole point of distributed trust disappears, and if it ever becomes public that there's only one computer mining, the currency crashes.
19 2011-04-19 00:09:38 <lianj> eamon: my line should still be valid
20 2011-04-19 00:09:52 <midnightmagic> to keep the network honest requires competing mining interests.
21 2011-04-19 00:10:03 <eamon> how?
22 2011-04-19 00:10:08 <lfm> "the two week cutoff" is an illusion. Sine the result is random PER HASH you have a proportionate chance of mining a block before the cutoff
23 2011-04-19 00:10:29 <eamon> midnightmagic you're not making sense
24 2011-04-19 00:10:31 <lianj> mining is required by design to verify transactions and make them happen. isnt it?
25 2011-04-19 00:10:39 <luke-jr> lfm: no.
26 2011-04-19 00:10:49 <lfm> yes
27 2011-04-19 00:11:13 <sipa> gavinandresen: got it
28 2011-04-19 00:11:16 <eamon> lianj: yes and it could all be done by one or a few low powered computers
29 2011-04-19 00:11:19 <lfm> every hash is independant
30 2011-04-19 00:11:21 <luke-jr> lfm: if it's unlikely you will get a block during a given difficulty, you are more profitable working in a pool
31 2011-04-19 00:11:26 <gavinandresen> sipa: I thought you were going to sleep....
32 2011-04-19 00:11:29 <luke-jr> lfm: doesn't change statistics
33 2011-04-19 00:11:32 <lfm> luke false
34 2011-04-19 00:11:48 <sipa> gavinandresen: i thought so too
35 2011-04-19 00:11:51 <lfm> luke you are wrong, sorry
36 2011-04-19 00:11:56 <sipa> gavinandresen: line 1030-1031 in main.h
37 2011-04-19 00:11:57 <lianj> eamon: thats why im state the mining race is BS
38 2011-04-19 00:12:15 <sipa> gavinandresen: why does that dependency check return true, in the loop?
39 2011-04-19 00:12:35 <luke-jr> lfm: no u
40 2011-04-19 00:12:50 <lfm> every hash is separate. each hash has a current difficulty. you are better off hashing now than later
41 2011-04-19 00:13:03 <sipa> it shouldn't return true unless all dependencies are ok, not just one
42 2011-04-19 00:13:20 <lfm> and pools with their overhead cannot compete with mining your own
43 2011-04-19 00:13:49 <lianj> eamon: waste and even counter productive in the sense that time spent there could be used to build other BC stuff
44 2011-04-19 00:14:19 <eamon> this should probably be in #bitcoin or #bitcoin-discussion, this is the development channel
45 2011-04-19 00:14:19 <luke-jr> lfm: every hash is separate, but the statistical probability you will find a block's hash within 2 weeks is still unlikely
46 2011-04-19 00:14:43 <lfm> luke unliklely and proportional to the time spent.
47 2011-04-19 00:15:36 <gavinandresen> sipa: thinking... thinking....
48 2011-04-19 00:15:38 <lfm> if you mulitply your chance of finding a block by the reward it works out to the same average reward as the pool except for pool overhead
49 2011-04-19 00:15:53 <lianj> eamon: true :)
50 2011-04-19 00:17:36 <sipa> gavinandresen: it's not the only problem, though
51 2011-04-19 00:17:47 mologie has joined
52 2011-04-19 00:17:58 <sipa> gavinandresen: it fixes the two clients giving a different result, but not the difference between getbalance and getbalance '*'
53 2011-04-19 00:18:10 <lianj> eamon: so where is the tests dir in the codebase? ;)
54 2011-04-19 00:18:15 <gavinandresen> sipa: yeah, it shouldn't do that check at all...
55 2011-04-19 00:20:19 <jgarzik> lianj: waiting for your contribution :)
56 2011-04-19 00:20:25 <jgarzik> unit tests for bitcoin would be fun
57 2011-04-19 00:20:25 <gavinandresen> sipa: the getbalance versus getbalance '*' is a different bug, I think getbalance with no args doesn't count 0-confirmation receives, but getbalance with any arg does (or vice-versa, I can't remember)
58 2011-04-19 00:20:35 <lianj> jgarzik: hihi
59 2011-04-19 00:21:06 <gavinandresen> RE: unit tests: patches very welcome. Somebody did a huge refactor and did some unit tests, just pulling the unit tests without the "refactor the world" would be great
60 2011-04-19 00:21:07 <lfm> gavinandresen: getbalance '*' doesnt find unlabeled addresse?
61 2011-04-19 00:21:36 Lachesis has joined
62 2011-04-19 00:21:45 <sipa> gavinandresen: neither should...
63 2011-04-19 00:21:53 <sipa> oh wait, nvm
64 2011-04-19 00:22:32 <sipa> yes, getBalance() has a check for IsConfirmed(), while GetAmounts() doesn't
65 2011-04-19 00:22:35 <eternal1> can the standard client, make a one to many tnx ?
66 2011-04-19 00:23:31 <gavinandresen> eternal1: there's a new sendmany RCP command in git head
67 2011-04-19 00:23:44 <gavinandresen> ^RCP^RPC
68 2011-04-19 00:23:59 <eternal1> mm, ok
69 2011-04-19 00:24:38 edgarallanpoe has joined
70 2011-04-19 00:24:42 edgarallanpoe is now known as ezl
71 2011-04-19 00:26:37 <jgarzik> gavinandresen: how does one get the balance of the "" account?
72 2011-04-19 00:26:51 <gavinandresen> getbalance "" 0
73 2011-04-19 00:27:30 <jgarzik> gavinandresen: doesn't seem to work here. does that need special shell quoting?
74 2011-04-19 00:28:11 <gavinandresen> jgarzik: shouldn't-- works for me in bash
75 2011-04-19 00:28:39 <gavinandresen> sipa: I'll fix the getbalance '*' bug-- it is completely ignoring minconf param, and it shouldn't.
76 2011-04-19 00:28:48 <jgarzik> gavinandresen: what I'm seeing here: listaccounts' "" balance is correct. 'getbalance ""' on shell command line returns full node balance (same as getinfo return).
77 2011-04-19 00:30:24 <gavinandresen> jgarzik: really? Works for me: http://pastebin.com/Yht9GVjq
78 2011-04-19 00:30:55 <sipa> gavinandresen: is that the reason?
79 2011-04-19 00:31:26 <sipa> gavinandresen: i still get a difference of 0.012 BTC in the output of '*' and plain
80 2011-04-19 00:31:42 <gavinandresen> sipa: yeah, that's part of the reason-- don't know about the 0.012 BTC, that's a different bug
81 2011-04-19 00:33:09 <sipa> commenting out the ischange check in getamounts makes a difference :S
82 2011-04-19 00:34:10 * jgarzik sends info to gavinandresen via PM
83 2011-04-19 00:34:27 <jgarzik> gavinandresen: the "" balance from listaccounts is correct. I've triple-checked that.
84 2011-04-19 00:34:37 <jgarzik> gavinandresen: and you can see the total wallet balance in getinfo.
85 2011-04-19 00:35:13 bk128 has joined
86 2011-04-19 00:35:52 <lulzplzkthx> So with Bitcoin, people can always know how much money you have, have made, or sent using things like BlockExplorer?
87 2011-04-19 00:37:02 <sipa> no
88 2011-04-19 00:37:09 <sipa> they can now these things about an address
89 2011-04-19 00:37:12 <sipa> but not about a person
90 2011-04-19 00:37:36 * jgarzik 's execute-bitcoind script had a dumb bug. not bitcoind's fault.
91 2011-04-19 00:37:53 <lulzplzkthx> Right, but if you give out your Bitcoin address as a donation thing, everybody can see how much it has received, and the balance of that address, yes?
92 2011-04-19 00:37:56 <lulzplzkthx> And of course, where that money is going.
93 2011-04-19 00:38:23 <lulzplzkthx> http://www.reddit.com/r/ideasfortheadmins/comments/grz9l/allow_bitcoin_as_an_alternative_payment_for/ <-- If you haven't already (anyone), would you mind upvoting this if you have a reddit account?
94 2011-04-19 00:38:25 <lfm> lulzplzkthx: give a different address to everyone
95 2011-04-19 00:41:05 <bk128> lulzplzkthx: upvoting
96 2011-04-19 00:41:38 <lulzplzkthx> Thanks bk128
97 2011-04-19 00:41:50 <lulzplzkthx> lfm: You can't for donations, and that is then monitored and can be followed where it's sent.
98 2011-04-19 00:42:02 <lulzplzkthx> Now, I don't plan on doing nefarious things, but I still feel a little uneasy about that.
99 2011-04-19 00:42:25 <lfm> lulzplzkthx: sure you could, or just change it frequently
100 2011-04-19 00:42:44 <lfm> depends how secret you wanna be. addresses are cheap.
101 2011-04-19 00:44:24 <lfm> if you give a different address to every ip address the monitors wont see the donations
102 2011-04-19 00:44:47 <midnightmagic> eamon: How am I not making sense, again?
103 2011-04-19 00:51:04 <lulzplzkthx> lfm: a different address to every ip address?
104 2011-04-19 00:51:08 <lulzplzkthx> Oh.
105 2011-04-19 00:51:12 <lulzplzkthx> Well, the'll see their own money anyway.
106 2011-04-19 00:51:43 <lulzplzkthx> How does the Bitcoin client handle sending amounts of money from multiple accounts to one? DOes it filter it into one and then move it?
107 2011-04-19 00:51:45 <midnightmagic> lulzplzkthx: or a different address for every visit.
108 2011-04-19 00:51:56 <lulzplzkthx> If so, that could potentially allow it again to be non-transparent.
109 2011-04-19 00:52:02 <lulzplzkthx> err transparent*
110 2011-04-19 00:52:21 <luke-jr> lulzplzkthx: you don't send money to accounts
111 2011-04-19 00:52:34 <luke-jr> accounts are purely client accounting
112 2011-04-19 00:52:55 <luke-jr> so it's just one integer - MOVED AMOUNT, and another integer + MOVED AMOUNT
113 2011-04-19 00:53:01 <luke-jr> you can even make your accounts negative
114 2011-04-19 00:53:09 <lulzplzkthx> I mean addresses.
115 2011-04-19 00:53:33 <lulzplzkthx> Say address 1 contains 3 bitcoins, address 2 contains 5 bitcoins. I send 8 bitcoins to address 3.
116 2011-04-19 00:53:44 <lulzplzkthx> Or are you saying that does not exist?
117 2011-04-19 00:53:46 <luke-jr> lulzplzkthx: addresses don't contain coins
118 2011-04-19 00:53:49 <lulzplzkthx> Okay.
119 2011-04-19 00:53:54 <lulzplzkthx> Bitcoins are files on your computer, okay.
120 2011-04-19 00:53:58 <luke-jr> nope
121 2011-04-19 00:54:01 <lulzplzkthx> Then what?
122 2011-04-19 00:54:07 <luke-jr> they're transaction outputs
123 2011-04-19 00:54:33 <luke-jr> in your example, what would happen is you make a single transaction
124 2011-04-19 00:54:42 <lulzplzkthx> And that single transaction would do what?
125 2011-04-19 00:54:44 <luke-jr> with 2 inputs (assuming each 'address' is an output)
126 2011-04-19 00:54:47 <lulzplzkthx> Where is the bitcoin coming from and going to?
127 2011-04-19 00:54:56 <luke-jr> and 1 output
128 2011-04-19 00:55:10 <lulzplzkthx> So the 2 inputs each contain bitcoins. Those inputs are the addresses, yaeh
129 2011-04-19 00:55:11 <lulzplzkthx> ?
130 2011-04-19 00:55:21 <lulzplzkthx> So basically, the address contains the bitcoins.
131 2011-04-19 00:55:22 <midnightmagic> i think luke enjoys being cryptic.
132 2011-04-19 00:55:32 <luke-jr> lulzplzkthx: no
133 2011-04-19 00:55:37 ezl has quit (Ping timeout: 240 seconds)
134 2011-04-19 00:55:47 <midnightmagic> he's trying to get you to think in his way, without addressing what it would take to get from where you are first.
135 2011-04-19 00:55:47 <luke-jr> coins are always outputs
136 2011-04-19 00:56:10 <luke-jr> so when you mint a new coin, you have the output of a Generation transaction
137 2011-04-19 00:56:13 <luke-jr> with 0 inputs
138 2011-04-19 00:56:40 <luke-jr> to spend that coin, you specify it plus a key+signature as an input
139 2011-04-19 00:57:42 <lulzplzkthx> Here's my point luke-jr... go to http://blockexplorer.com/block/000000000000a5c63aed1e66907f48d8f432943a29d4d86853c28f0d919c019e
140 2011-04-19 00:57:50 <lulzplzkthx> Look at the second transaction from the top.
141 2011-04-19 00:58:10 <lulzplzkthx> There are 2 from addresses, 1 to address. I can now find out how you're spending what I sent you.
142 2011-04-19 00:58:17 <lulzplzkthx> (Assuming I know who owns the To address)
143 2011-04-19 00:58:25 <luke-jr> lulzplzkthx: ok, that's true
144 2011-04-19 00:59:01 <lulzplzkthx> Here's my other confusion: When you find a block, does that work you did, in itself, create 50 bitcoins?
145 2011-04-19 00:59:19 <lulzplzkthx> In other words: Do you know you found a block BECAUSE you now have 50 bitcoins?
146 2011-04-19 00:59:29 <luke-jr> â¦no
147 2011-04-19 00:59:45 <thermal> i have an idea
148 2011-04-19 01:00:01 <thermal> that i think should *really* be implemented and included in the spec
149 2011-04-19 01:00:03 <luke-jr> your client is constantly creating block data. the miner is only trying to find a hash of that data that fits the target difficulty
150 2011-04-19 01:00:16 <thermal> rather than having one key pair, have two
151 2011-04-19 01:00:24 <luke-jr> when the miner finds a valid hash, that's when you "find a block"
152 2011-04-19 01:00:36 <lulzplzkthx> And then how are the coins created?
153 2011-04-19 01:00:38 <luke-jr> thermal: no spec needed. just do it.
154 2011-04-19 01:00:53 <luke-jr> lulzplzkthx: the block data your client is creating simply gives itself 50 BTC
155 2011-04-19 01:01:02 <thermal> this way, by having two levels, banks etc. can use the 2nd level while you can still keep the primary one safe
156 2011-04-19 01:01:06 <midnightmagic> lulzplzkthx: you as a miner package up a transaction which gives you control over the generation block's "fresh" bitcoins. there is no "from". and the rest of the network accepts and agrees that your fresh address now has control over spending those coins in future transactions.
157 2011-04-19 01:01:27 <CIA-89> bitcoin: Gavin Andresen master * r72e962c / rpc.cpp : getbalance '*' was ignoring minconf param. - http://bit.ly/eGhliN
158 2011-04-19 01:01:34 <luke-jr> thermal: ok, THAT line made no sense
159 2011-04-19 01:01:57 <luke-jr> thermal: if EITHER key is valid, then there is no purpose to keeping only one safe
160 2011-04-19 01:02:09 <lulzplzkthx> midnightmagic: What happens if two people find a block, and they conflict at some node?
161 2011-04-19 01:02:11 <thermal> ok so, i see banks and other services that will handle your bitcoins and transactions becoming commonly used
162 2011-04-19 01:02:31 <midnightmagic> lulzplzkthx: then whoever finds the next block gets to decide which one of the previous blocks is valid.
163 2011-04-19 01:02:37 <luke-jr> lulzplzkthx: then it's all about who makes the next block
164 2011-04-19 01:02:37 <thermal> the primary key would be able to revoke any child keys or something to that effect
165 2011-04-19 01:02:48 <midnightmagic> lulzplzkthx: think about it this way: every node accepts whatever is the longest chain of blocks as "canonical".
166 2011-04-19 01:02:48 <lulzplzkthx> Okay.
167 2011-04-19 01:02:48 <luke-jr> lulzplzkthx: the person who makes the next block decides which of them wins
168 2011-04-19 01:02:49 <thermal> or set a TTL
169 2011-04-19 01:03:00 <lulzplzkthx> Yeah, I knew the longest thing, but wasn't sure if they were fresh blocks.
170 2011-04-19 01:03:32 <midnightmagic> lulzplzkthx: so if there are two chains that are equally long, the network is forked, and whoever builds a longer chain resolves the fork (ideally)
171 2011-04-19 01:03:53 <lulzplzkthx> Why can't one user just decide to create a "block" without the block data? Would other clients not except it?
172 2011-04-19 01:04:15 <midnightmagic> lulzplzkthx: the corollary is that if someone keeps a secret fork that outpaces the original network, they could re-introduce the blocks later on and suddenly all the clients get a history that is completely re-written.
173 2011-04-19 01:04:31 <midnightmagic> lulzplzkthx: it has to be linked in with the previous blocks or the other clients will just reject it.
174 2011-04-19 01:04:33 <lulzplzkthx> midnightmagic: And that could really screw stuff up, yeah?
175 2011-04-19 01:04:42 <thermal> so.. you could use the primary key to issue 'proxy keys'
176 2011-04-19 01:04:46 <thermal> and also set parameters
177 2011-04-19 01:04:53 <lulzplzkthx> In fact, couldn't that keep you from keeping bitcoins you received previously?
178 2011-04-19 01:05:10 <thermal> (max transaction amount maybe etc)
179 2011-04-19 01:05:12 <lulzplzkthx> So if you got caught at a fork, paid for stuff, then found a block, you could rewrite all those payments...?
180 2011-04-19 01:05:21 <midnightmagic> lulzplzkthx: yes, definitely. it's the one "known" vulnerability of ฿.. and the reason why mining with enough competing, non-cooperating interests is really crucial. history, otherwise, is controlled by whoever has control of the majority share of computing power.
181 2011-04-19 01:05:33 <lulzplzkthx> No... it would keep the firt person's payments, not yours?
182 2011-04-19 01:05:54 <midnightmagic> lulzplzkthx: yes, a competing fork that outpaced the real network could completely re-write all history and destroy all the coins you mined for right back to the fork point.
183 2011-04-19 01:06:20 <lulzplzkthx> Interesting. What would keep a user from rewriting the history without a fork (assuming they found a block)?
184 2011-04-19 01:06:25 <lulzplzkthx> Other clients simply wouldn't accept it?
185 2011-04-19 01:06:31 <lulzplzkthx> Or encryption?
186 2011-04-19 01:06:49 <jgarzik> midnightmagic: and/or last lock-in point
187 2011-04-19 01:07:02 <midnightmagic> lulzplzkthx: pretty much. how would you suggest this history be rewritten?
188 2011-04-19 01:07:14 <midnightmagic> jgarzik: ah, right. that too.
189 2011-04-19 01:07:48 <lulzplzkthx> midnightmagic: You could just rewrite everything to go to you, if it werne't encrypted. But it's safe because it's encrypted you're saying?
190 2011-04-19 01:08:54 <midnightmagic> lulzplzkthx: the crypto in-use prevents that. you don't have the keys to sign away someone else's money to go to yourself. all you're doing as a miner is certifying the transaction, not participating in the actual transaction itself. you're just hashing it as part of your block.
191 2011-04-19 01:09:35 <lulzplzkthx> Also: If a bitcoin is constantly adding keys to it (you know, so you know who belongs it) wouldn't a) the bitcoins get huge, to the point where they could take up a large portion of your hard drive? b) how can you split up something into 10000000th's?
192 2011-04-19 01:10:58 <luke-jr> lulzplzkthx: the person who makes the next block decides which of them wins
193 2011-04-19 01:11:00 <luke-jr> err
194 2011-04-19 01:11:17 <midnightmagic> a) you're not constantly adding keys to the network as a whole. you're just keeping a little pool of them yourself, and then using them in public where everyone can see. b) it's just a numerical arithmetical split, each coin doesn't have its own separate entity representing it.
195 2011-04-19 01:12:26 <luke-jr> lulzplzkthx: hard drives grow fast
196 2011-04-19 01:12:33 <midnightmagic> it's like gpg. you give everyone your public key, but only your private key can actually do stuff like signing.
197 2011-04-19 01:12:39 <lulzplzkthx> luke-jr: and shrink. look at ssd.
198 2011-04-19 01:12:43 <luke-jr> lulzplzkthx: and the fork problem is why you should wait until 6 confirmations
199 2011-04-19 01:12:50 <luke-jr> lulzplzkthx: hence why I don't use SSD
200 2011-04-19 01:13:21 <midnightmagic> lulzplzkthx: also, since in your block you're only including the root of a merkle hash, you could technically delete old useless transactions to save HDD space over time..
201 2011-04-19 01:13:32 <lulzplzkthx> midnightmagic: Yeah, I know it's like public/private... but... okay.
202 2011-04-19 01:14:17 <lulzplzkthx> Wait, wouldn't you need all the keys since they'd be part of the blocks (which you need all of)?
203 2011-04-19 01:14:49 <luke-jr> lulzplzkthx: you don't need all the blocks, and you can have only parts of blocks
204 2011-04-19 01:15:06 <lulzplzkthx> luke-jr: It seems for the original Bitcoin client to work, you do.
205 2011-04-19 01:15:11 <lulzplzkthx> Perhaps I'm wrong.
206 2011-04-19 01:15:16 <luke-jr> lulzplzkthx: that's a flaw in that client
207 2011-04-19 01:15:19 <lulzplzkthx> Or maybe you just need the last, and the issue is it starts with first.
208 2011-04-19 01:19:18 <midnightmagic> lulzplzkthx: perhaps talking about a merkle root wasn't the best idea on my part. but say A spends 50 to B, and it's certified and now X blocks old. Now say B spends it to C. Who cares about A? you can just stem it off and delete that transaction, trusting that the network as a whole did in fact verify the old spend, and perhaps that someone, somewhere still holds the old data for forensics purposes.
209 2011-04-19 01:20:48 theymos has joined
210 2011-04-19 01:21:12 Kiba has joined
211 2011-04-19 01:21:32 eao has quit (Quit: Leaving)
212 2011-04-19 01:22:07 <Kiba> hello
213 2011-04-19 01:23:46 <lulzplzkthx> ah, okay midnightmagic
214 2011-04-19 01:23:51 <lulzplzkthx> well, does the current bitcoin client do that?
215 2011-04-19 01:23:58 <Kiba> hello
216 2011-04-19 01:24:03 <midnightmagic> lulzplzkthx: no. but it's an option in the future.
217 2011-04-19 01:24:05 noagendamarket has joined
218 2011-04-19 01:24:08 <lulzplzkthx> Okay.
219 2011-04-19 01:24:21 <midnightmagic> lulzplzkthx: transactions aren't happening often enough that normal participants can't keep the entire history on their machines.
220 2011-04-19 01:25:27 <dbitcoin> ;;bc,stats
221 2011-04-19 01:25:29 <gribble> Current Blocks: 119053 | Current Difficulty: 92347.59095209 | Next Difficulty At Block: 120959 | Next Difficulty In: 1906 blocks | Next Difficulty In About: 1 week, 5 days, 15 hours, 54 minutes, and 4 seconds | Next Difficulty Estimate: 97363.56909684
222 2011-04-19 01:26:34 <Kiba> so...
223 2011-04-19 01:26:35 <Kiba> hmm
224 2011-04-19 01:34:13 <lulzplzkthx> midnightmagic: Well, that's true BUT... my DATA folder (I just setup Portable Bitcoin) is 220 MB. That could exceed a flash drive pretty quickly.
225 2011-04-19 01:34:21 <lulzplzkthx> I have a 4 GB drive, so I need not worry for a while, but...
226 2011-04-19 01:34:44 <theymos> Client mode will probably be available soon.
227 2011-04-19 01:34:52 <midnightmagic> better watch it over time and project it out then.
228 2011-04-19 01:38:18 devon_hillard has quit (Read error: Connection reset by peer)
229 2011-04-19 01:42:55 toffoo has quit ()
230 2011-04-19 01:44:26 toffoo has joined
231 2011-04-19 01:46:19 wolfspraul has quit (Quit: leaving)
232 2011-04-19 01:46:40 <luke-jr> ;;bc,blocks
233 2011-04-19 01:46:40 <gribble> 119056
234 2011-04-19 01:47:29 <Kiba> so...
235 2011-04-19 01:48:28 theymos has quit (Remote host closed the connection)
236 2011-04-19 01:49:18 <retinal> so...
237 2011-04-19 01:49:54 * Kiba have nothing to say
238 2011-04-19 01:51:51 <noagendamarket> that would be a first Kiba :)
239 2011-04-19 01:52:56 amiller has joined
240 2011-04-19 01:54:26 <Kiba> of course, I have something to say
241 2011-04-19 01:54:44 <Kiba> http://kibabase.com/articles/sale-cycle
242 2011-04-19 01:54:51 <Kiba> this explicate a business theory
243 2011-04-19 01:55:01 <noagendamarket> http://marketcut.com/bitcoin-btc/k-factor-the-viral-coefficient-and-bitcoin/ nice article
244 2011-04-19 01:55:12 <Kiba> noagendamarket: that's bitcoinmoney's article
245 2011-04-19 01:55:15 <Kiba> sneaky link, btw
246 2011-04-19 01:55:33 <noagendamarket> orly did they steal it ?
247 2011-04-19 01:56:23 <Kiba> dunno
248 2011-04-19 01:56:30 <lulzplzkthx> Hey, does anyone know who created BitcoinMonitor?
249 2011-04-19 01:56:32 <Kiba> they did link to bitcoin money
250 2011-04-19 01:56:39 <Kiba> but it isn't obvious
251 2011-04-19 01:57:09 <Kiba> they have contact information on the site
252 2011-04-19 01:58:16 <lulzplzkthx> Well, whoever did: To help my OCD, it'd be nice if you added a favicon XD
253 2011-04-19 01:58:57 mologie has quit (Quit: Nettalk6 - www.ntalk.de)
254 2011-04-19 02:02:49 <noagendamarket> seems like there is over 9000 places with the bitcoin favicon
255 2011-04-19 02:02:57 <noagendamarket> it gets very confusing
256 2011-04-19 02:04:36 jwalck has quit (Ping timeout: 240 seconds)
257 2011-04-19 02:05:39 <Blitzboom> itâs fun when people talk about a bitcoin bubble
258 2011-04-19 02:05:49 * Kiba is happy that bitcoinweekly's revenue got upgraded to 0.04 bitcent
259 2011-04-19 02:05:54 <Blitzboom> and disregard facebook/twitter
260 2011-04-19 02:06:12 <luke-jr> ?
261 2011-04-19 02:06:17 <Kiba> the bitcoin community thought more about bitcoin's weakness more than any bitcoin critics on the planet
262 2011-04-19 02:06:20 <Blitzboom> facebook = 70 billion
263 2011-04-19 02:06:23 <Blitzboom> USD
264 2011-04-19 02:06:35 <Kiba> what about facebook and twitter?
265 2011-04-19 02:06:52 <Blitzboom> everyone is investing in them
266 2011-04-19 02:07:03 Zarutian has quit (Read error: Operation timed out)
267 2011-04-19 02:07:03 <luke-jr> not I
268 2011-04-19 02:07:05 <Kiba> and?
269 2011-04-19 02:07:21 <Blitzboom> and itâs obviously a bubble
270 2011-04-19 02:07:39 <Kiba> overuse of the word bubble tend to make kiba a dull boy
271 2011-04-19 02:07:47 <Kiba> twitter is useful
272 2011-04-19 02:07:49 <Kiba> facebook is not
273 2011-04-19 02:08:00 <CIA-89> bitcoin: Luke Dashjr * r62eb33a0bc89 spesmilo/main.py: Bugfix: Delete old TrayIcon when we're about to create a new one http://tinyurl.com/4yu7aty
274 2011-04-19 02:08:02 <CIA-89> bitcoin: Luke Dashjr * rf78d8014f28d spesmilo/main.py: Bugfix: Make settings the current window when not yet connected http://tinyurl.com/3m8t7lo
275 2011-04-19 02:08:06 <Blitzboom> you donât come to decide whatâs useful or not
276 2011-04-19 02:08:06 <Kiba> at least, for me, it is useful
277 2011-04-19 02:08:14 <Kiba> twitter is useful
278 2011-04-19 02:08:20 <Kiba> because twitter give me good content to read
279 2011-04-19 02:08:26 <Kiba> and a pulse on the bitcoin world
280 2011-04-19 02:08:31 topi`_ has quit (Ping timeout: 240 seconds)
281 2011-04-19 02:08:50 <Kiba> very good way to catch breaking news too
282 2011-04-19 02:09:03 <Kiba> twitter is not a social networking site
283 2011-04-19 02:09:07 <Kiba> it is an information network
284 2011-04-19 02:09:15 <luke-jr> Kiba: Twitter is not useful
285 2011-04-19 02:09:19 <Kiba> and kiba is not very conductive to being banal social
286 2011-04-19 02:09:21 <luke-jr> Facebook is useful only because of its monopoly
287 2011-04-19 02:09:37 topi` has joined
288 2011-04-19 02:09:44 bob__ has joined
289 2011-04-19 02:09:46 <Kiba> luke-jr: I used it to get info for bitcoinwatch's newsfeed
290 2011-04-19 02:09:49 Zarutian has joined
291 2011-04-19 02:12:48 dbitcoin has quit (Quit: Ex-Chat)
292 2011-04-19 02:13:34 Usertwelve has quit (Ping timeout: 252 seconds)
293 2011-04-19 02:17:34 Warlord has quit (Quit: GeekBouncer - http://geekbouncer.co.uk)
294 2011-04-19 02:18:13 Warlord has joined
295 2011-04-19 02:18:49 taaz has left ()
296 2011-04-19 02:19:18 blablaa has quit ()
297 2011-04-19 02:19:27 <sacarlson> well at least I got github setup but I'm still at a loss as to how to create a new proto coin but here's what I have so far https://github.com/sacarlson/bitcoin
298 2011-04-19 02:19:32 eamon has quit ()
299 2011-04-19 02:20:30 Zarutian has quit (Quit: Zarutian)
300 2011-04-19 02:20:43 jb55 has quit (Read error: Operation timed out)
301 2011-04-19 02:21:34 DavidSJ has joined
302 2011-04-19 02:22:10 <gavinandresen> sacarlson: MATURITY=1 is a really bad idea. Think about what will happen if a 1-maturity coinbase is spent, then spent again a couple times in the next 2 blocks, then there is a 3-block block re-porg
303 2011-04-19 02:22:14 <gavinandresen> re-org
304 2011-04-19 02:22:34 eternal1 has quit (Quit: Leaving.)
305 2011-04-19 02:23:02 <sacarlson> gavinandresen: it's just a temp move to quicken development
306 2011-04-19 02:23:21 jb55 has joined
307 2011-04-19 02:23:31 <sacarlson> gavinandresen: it will later be added to the config file so that it can be easily changed
308 2011-04-19 02:23:34 <gavinandresen> sacarlson: you should be using testnet-in-a-box....
309 2011-04-19 02:23:50 <gavinandresen> (it comes with already-mature coins to play with)
310 2011-04-19 02:24:23 <sacarlson> gavinandresen: that sounds cool but I want to learn to create new coins
311 2011-04-19 02:25:19 <sacarlson> gavinandresen: where can this testnet-in-a-box be found? is that in git?
312 2011-04-19 02:25:46 gasteve has joined
313 2011-04-19 02:26:12 <gavinandresen> sacarlson: https://sourceforge.net/projects/bitcoin/files/Bitcoin/testnet-in-a-box/
314 2011-04-19 02:26:28 <sacarlson> gavinandresen: very good I'll take a look
315 2011-04-19 02:30:37 gavinandresen has quit (Quit: gavinandresen)
316 2011-04-19 02:35:45 Xunie` has joined
317 2011-04-19 02:36:15 Xunie has quit (Read error: Connection reset by peer)
318 2011-04-19 02:36:52 <grbgout> hmm, has anyone looked at hadoop with bitcoins in mind?
319 2011-04-19 02:37:34 <Kiba> what's a hadoop?
320 2011-04-19 02:37:53 <grbgout> Kiba: http://hadoop.apache.org/
321 2011-04-19 02:37:57 <grbgout> Kiba: just learned about it myself.
322 2011-04-19 02:39:37 * Kiba wonders how soon will the first namecoin to btc market open
323 2011-04-19 02:39:52 <grbgout> namecoin? Oh, that DNS thing?
324 2011-04-19 02:40:35 <Kiba> yes
325 2011-04-19 02:40:56 <Kiba> people have not agreed on how many domain names should be created...
326 2011-04-19 02:41:47 <bd_> grbgout: What, you have a cluster of GPU-equipped servers in your basement to run it on? :)
327 2011-04-19 02:42:38 DavidSJ has quit (Quit: DavidSJ)
328 2011-04-19 02:43:18 <grbgout> bd_: not yet :)
329 2011-04-19 02:46:32 redengin has joined
330 2011-04-19 02:48:43 hacim has quit (Changing host)
331 2011-04-19 02:48:43 hacim has joined
332 2011-04-19 02:49:43 tabsa has quit (Ping timeout: 240 seconds)
333 2011-04-19 02:49:55 toffoo has quit ()
334 2011-04-19 02:50:13 sacarlson has quit (Ping timeout: 260 seconds)
335 2011-04-19 02:53:30 toffoo has joined
336 2011-04-19 02:57:22 <Diablo-D3> lol
337 2011-04-19 02:57:29 <Diablo-D3> we're past the 25 hour mark
338 2011-04-19 02:57:36 marlowe has quit (Ping timeout: 246 seconds)
339 2011-04-19 02:57:42 <Diablo-D3> still doesnt have all 1000 done
340 2011-04-19 03:00:15 <netxshare> new nvidia drivers
341 2011-04-19 03:00:32 <netxshare> 270.61 came out of beta
342 2011-04-19 03:05:56 sacarlson has joined
343 2011-04-19 03:07:53 bob__ has quit (Quit: Page closed)
344 2011-04-19 03:14:27 MingusDew has joined
345 2011-04-19 03:19:22 m86 has joined
346 2011-04-19 03:19:42 Jkessler has quit (Ping timeout: 264 seconds)
347 2011-04-19 03:28:15 ezl has joined
348 2011-04-19 03:29:03 redengin has quit (Ping timeout: 240 seconds)
349 2011-04-19 03:30:04 luke-jr has quit (Read error: Connection reset by peer)
350 2011-04-19 03:33:35 redengin has joined
351 2011-04-19 03:34:26 fimp has joined
352 2011-04-19 03:35:51 luke-jr has quit (otg!~luke-jr@ishibashi.dashjr.org|Read error: Connection reset by peer)
353 2011-04-19 03:36:23 Jkessler has joined
354 2011-04-19 03:36:29 Jkessler has quit (Changing host)
355 2011-04-19 03:36:29 Jkessler has joined
356 2011-04-19 03:36:36 OneFixt_ has joined
357 2011-04-19 03:37:39 fluf^arr has joined
358 2011-04-19 03:38:17 OneFixt has quit (Ping timeout: 252 seconds)
359 2011-04-19 03:39:29 fluffle has quit (Ping timeout: 255 seconds)
360 2011-04-19 03:39:36 stamit has quit (Ping timeout: 246 seconds)
361 2011-04-19 03:40:52 RenaKunisaki has joined
362 2011-04-19 03:43:48 stamit has joined
363 2011-04-19 03:44:41 taco_the_paco has joined
364 2011-04-19 03:47:14 DavidSJ has joined
365 2011-04-19 03:55:24 redengin has quit (Quit: AndroIRC)
366 2011-04-19 03:55:49 redengin has joined
367 2011-04-19 03:59:03 user23542 has joined
368 2011-04-19 04:00:29 user23542 has quit (Client Quit)
369 2011-04-19 04:03:12 NWTSPV has joined
370 2011-04-19 04:07:34 fimp has quit (Quit: This computer has gone to sleep)
371 2011-04-19 04:07:36 x6763 has quit (Ping timeout: 246 seconds)
372 2011-04-19 04:14:28 Karzy has left ()
373 2011-04-19 04:15:17 eao has joined
374 2011-04-19 04:16:44 x6763 has joined
375 2011-04-19 04:19:37 luke-jr has joined
376 2011-04-19 04:28:29 <Lachesis> wow no discussion here anymore
377 2011-04-19 04:29:20 <hacim> hush
378 2011-04-19 04:29:39 <luke-jr> 2.6.38 is bunk
379 2011-04-19 04:30:40 <luke-jr> https://bugzilla.kernel.org/show_bug.cgi?id=33662
380 2011-04-19 04:30:56 endian7000 has joined
381 2011-04-19 04:31:01 <Lachesis> does #bitcoin-market still work?
382 2011-04-19 04:31:07 <DavidSJ> doesn't seem to
383 2011-04-19 04:31:13 <Lachesis> i saw no streaming data when i was idling there (admittedly not for long)
384 2011-04-19 04:31:37 <endian7000> difficulty updates happen every K blocks, right?
385 2011-04-19 04:31:45 <endian7000> what if a well-funded adversary mines at 10 PHash/sec until the difficulty goes way up, then stops immediately after a difficulty update?
386 2011-04-19 04:32:08 <Lachesis> yeah, 2016
387 2011-04-19 04:32:18 <Lachesis> diff will only go up / down 4x at a time
388 2011-04-19 04:32:27 <Lachesis> meaning it'd cost him/her a while to raise it too high
389 2011-04-19 04:32:36 <sacarlson> endian7000: I guess that might take a week to correct?
390 2011-04-19 04:32:55 <Lachesis> sacarlson, no, it'd take 2016 blocks
391 2011-04-19 04:33:15 <Lachesis> which could take months if, say, 99% of the computing power in the client went away
392 2011-04-19 04:33:43 luke-jr has quit (Read error: Connection reset by peer)
393 2011-04-19 04:34:36 luke-jr has joined
394 2011-04-19 04:34:48 <endian7000> because governments probably have way more than 1THash sitting around
395 2011-04-19 04:35:41 <[Noodles]> 1T != 10P, besides that, why would any government bother at bitcoins current size?
396 2011-04-19 04:35:44 <Lachesis> yeah, but also better ways to break bitcoin
397 2011-04-19 04:35:57 <Lachesis> besides, with > 51% of the network power, you can double spend
398 2011-04-19 04:36:02 <Lachesis> which would destroy confidence in bitcoin
399 2011-04-19 04:37:25 zyb has joined
400 2011-04-19 04:37:30 luke-jr has quit (Remote host closed the connection)
401 2011-04-19 04:38:36 luke-jr has joined
402 2011-04-19 04:40:00 AAA_awright_ is now known as AAA_awright
403 2011-04-19 04:42:44 Cusipzzz has quit (Quit: KVIrc 4.0.2 Insomnia http://www.kvirc.net/)
404 2011-04-19 04:44:51 toffoo has quit ()
405 2011-04-19 04:46:20 toffoo has joined
406 2011-04-19 04:47:51 tenach has joined
407 2011-04-19 04:47:51 tenach has quit (Changing host)
408 2011-04-19 04:47:51 tenach has joined
409 2011-04-19 04:53:33 OneFixt_ is now known as OneFixt
410 2011-04-19 04:53:43 OneFixt has quit (Changing host)
411 2011-04-19 04:53:43 OneFixt has joined
412 2011-04-19 04:55:59 Lachesis has quit (Ping timeout: 260 seconds)
413 2011-04-19 04:56:25 Lachesis has joined
414 2011-04-19 05:04:37 alystair has joined
415 2011-04-19 05:13:30 Slix` has joined
416 2011-04-19 05:15:28 JackRabiit has joined
417 2011-04-19 05:15:50 <EPiSKiNG> betco.in has 4 hold em players just seated and ready for action!
418 2011-04-19 05:15:56 <JackRabiit> guys come play poker at betco.in , We gotta table going!
419 2011-04-19 05:16:05 JackRabiit has quit (Client Quit)
420 2011-04-19 05:16:06 <noagendamarket> cool
421 2011-04-19 05:16:53 Slix` has quit (Client Quit)
422 2011-04-19 05:25:01 DavidSJ has quit (Quit: DavidSJ)
423 2011-04-19 05:30:07 dissipate has joined
424 2011-04-19 05:34:03 skeledrew has quit (Ping timeout: 250 seconds)
425 2011-04-19 05:39:35 wolfspraul has joined
426 2011-04-19 05:40:09 wolfspra1l has joined
427 2011-04-19 05:40:20 wolfspraul has quit (Client Quit)
428 2011-04-19 05:40:22 zyb has quit (Remote host closed the connection)
429 2011-04-19 05:40:22 wolfspra1l has quit (Client Quit)
430 2011-04-19 05:40:33 wolfspraul has joined
431 2011-04-19 05:42:59 Lachesis has quit (Quit: Leaving)
432 2011-04-19 05:50:11 Warlord has quit (Changing host)
433 2011-04-19 05:50:11 Warlord has joined
434 2011-04-19 05:52:04 sosborn has joined
435 2011-04-19 06:00:57 endian7000 has quit (Quit: endian7000)
436 2011-04-19 06:03:55 ForceDestroyer has joined
437 2011-04-19 06:04:35 WakiMiko has joined
438 2011-04-19 06:04:53 slueth has joined
439 2011-04-19 06:06:11 <slueth> In ubuntu how do you get the miner to reconize the second gpu?
440 2011-04-19 06:07:38 WakiMiko_ has quit (Ping timeout: 240 seconds)
441 2011-04-19 06:07:58 Kicchiri has quit (Ping timeout: 276 seconds)
442 2011-04-19 06:14:01 danlucraft has quit (Quit: Leaving.)
443 2011-04-19 06:16:40 BERRI has quit (Excess Flood)
444 2011-04-19 06:19:21 BERRI has joined
445 2011-04-19 06:20:44 <lfm> slueth: genarally all you need is a d-1 parm
446 2011-04-19 06:20:50 <lfm> slueth: genarally all you need is a d=1 parm
447 2011-04-19 06:22:47 Xunie` is now known as Xunie
448 2011-04-19 06:22:52 Xunie has quit (Changing host)
449 2011-04-19 06:22:52 Xunie has joined
450 2011-04-19 06:26:54 <slueth> It actually uses my cpu when i do d=1
451 2011-04-19 06:27:03 <slueth> so I assume it is not reconizing my gpu or seeing it
452 2011-04-19 06:29:54 skeledrew has joined
453 2011-04-19 06:35:53 mvn071 has joined
454 2011-04-19 06:36:31 <mvn071> FYI https://github.com/bitcoin/bitcoin.git does 404
455 2011-04-19 06:39:40 skyewm has quit (Ping timeout: 252 seconds)
456 2011-04-19 06:41:08 <jaromil> are you in china or somewhere like that?
457 2011-04-19 06:41:11 <jaromil> it works
458 2011-04-19 06:41:50 <rly> mvn071: you are doing it wrong.
459 2011-04-19 06:41:58 <rly> jaromil: no, it doesn't.
460 2011-04-19 06:42:05 <mvn071> hin
461 2011-04-19 06:42:09 <rly> jaromil: but it is designed to give that response.
462 2011-04-19 06:42:28 <mvn071> hi sorry, it can pull it by git
463 2011-04-19 06:42:45 <mvn071> but https give 404, so it looks like an error
464 2011-04-19 06:43:48 <rly> mvn071: it is not an error if it is by design.
465 2011-04-19 06:44:06 <rly> mvn071: remove the bitcoin.git part for proof of that
466 2011-04-19 06:44:23 <mvn071> rly maybe a responce like "use git for this" makes more sense
467 2011-04-19 06:45:22 <jaromil> oh sryi thought you knew
468 2011-04-19 06:45:34 <jaromil> usually .git is to be used with git
469 2011-04-19 06:45:55 <mvn071> I did not use the words wright,wrong,error,
470 2011-04-19 06:46:00 <jaromil> good'moin btw
471 2011-04-19 06:46:17 <mvn071> aye good morging
472 2011-04-19 06:47:10 <mvn071> I was trying to get coind working on freebsd
473 2011-04-19 06:48:49 <mvn071> and there in nothing in ports... Is known to work on any BSD (not osX)?
474 2011-04-19 06:59:44 slueth has quit (Quit: Lost terminal)
475 2011-04-19 07:02:30 <topi`> jaromil: I just wrote a small article about nationalist politics and bitcoin.
476 2011-04-19 07:02:47 <topi`> jaromil: feedback would be appreciated :)
477 2011-04-19 07:02:55 <topi`> http://lorelei.kaverit.org/cgi/sana.py?the-true-finns-and-bitcoin
478 2011-04-19 07:06:32 <Diablo-D3> arent the true finns some sort of post-nazi group?
479 2011-04-19 07:06:38 noagendamarket has quit (Ping timeout: 240 seconds)
480 2011-04-19 07:07:10 <mtrlt> Diablo-D3: are you serious? :P
481 2011-04-19 07:07:42 <rly> Diablo-D3: sounds more like common sense.
482 2011-04-19 07:07:54 <Diablo-D3> so they... are.
483 2011-04-19 07:08:11 tenach has quit (Ping timeout: 258 seconds)
484 2011-04-19 07:08:25 <rly> Diablo-D3: they want to be sovereign.
485 2011-04-19 07:08:27 <Diablo-D3> I mean, the only other parties that are "common sense" are the nazis and the tea party
486 2011-04-19 07:08:30 <rly> Diablo-D3: what is wrong with that?
487 2011-04-19 07:08:31 eao has quit (Quit: Leaving)
488 2011-04-19 07:08:40 <Diablo-D3> rly: because where are they going to run off to?
489 2011-04-19 07:08:41 <Diablo-D3> canada?
490 2011-04-19 07:09:13 <rly> Diablo-D3: why can't they just sit in the snow with their weapons and fight of whoever they want?
491 2011-04-19 07:09:14 <mtrlt> what?
492 2011-04-19 07:09:26 <Diablo-D3> rly: because WWII ended many years ago.
493 2011-04-19 07:09:26 <rly> Diablo-D3: that worked in WW2 too pretty well.
494 2011-04-19 07:09:46 <rly> Diablo-D3: so, they just demonstrated that they could fight fairly well.
495 2011-04-19 07:10:10 <rly> Diablo-D3: that other countries are such pussies is a different story.
496 2011-04-19 07:10:15 <Diablo-D3> any political party that advocates war should be wiped from the face of the earth.
497 2011-04-19 07:10:25 <rly> Diablo-D3: they do not advocate that.
498 2011-04-19 07:10:36 <mtrlt> but true finns don't advocate war. why did you think they did :P
499 2011-04-19 07:10:42 <rly> Diablo-D3: also, you are implying that the US should be wiped from the face of the earth now.
500 2011-04-19 07:10:48 <Diablo-D3> rly: just Bush.
501 2011-04-19 07:11:04 <rly> The US has continuously fought wars.
502 2011-04-19 07:11:11 <Diablo-D3> Yes, under Bush.
503 2011-04-19 07:11:12 <rly> There are lists available showing that.
504 2011-04-19 07:11:20 <rly> No, since WW2.
505 2011-04-19 07:11:20 <Diablo-D3> Bush just keeps changing names and faces.
506 2011-04-19 07:11:26 <Diablo-D3> Right now, we know him as Bush.
507 2011-04-19 07:11:37 <Diablo-D3> Tommorow, he could be some other guy
508 2011-04-19 07:11:42 <mtrlt> like obama? :P
509 2011-04-19 07:11:51 <Diablo-D3> And I suspect he is also multiple people at once
510 2011-04-19 07:11:58 <Diablo-D3> Even in the same room
511 2011-04-19 07:12:01 <rly> This was about the Finnish people. I do not see anything wrong with deciding what is best for them.
512 2011-04-19 07:12:15 <Diablo-D3> So, the true finns are NOT a nazi party?
513 2011-04-19 07:12:21 <rly> It sounds like they have a working democracy, unlike most other countries.
514 2011-04-19 07:12:27 <mtrlt> well, of course not?
515 2011-04-19 07:12:35 <Diablo-D3> So why does the Internet think they are
516 2011-04-19 07:12:35 <rly> Diablo-D3: er no.
517 2011-04-19 07:12:50 <rly> Diablo-D3: I do not think they have ever had a nazi problem or something like that.
518 2011-04-19 07:12:58 <mtrlt> except in the war of lapland ;-)
519 2011-04-19 07:13:33 <mtrlt> but that was a different kind of problem
520 2011-04-19 07:13:47 <rly> I meant like the youth being attracted to such idiotic ideas.
521 2011-04-19 07:13:50 <Diablo-D3> so is there any hot women finland doesnt want anymore?
522 2011-04-19 07:13:59 * Diablo-D3 would gladly take them
523 2011-04-19 07:14:09 <rly> Diablo-D3: I don't think it quite works like that.
524 2011-04-19 07:14:37 <rly> Diablo-D3: also, the Internet is stupid.
525 2011-04-19 07:14:44 <Diablo-D3> ^ THIS
526 2011-04-19 07:14:48 <mtrlt> if you really want to know about finnish politics, you have to read finnish news from multiple sources in finnish :-)
527 2011-04-19 07:14:55 <Diablo-D3> mtrlt: fuck that
528 2011-04-19 07:15:02 <Diablo-D3> even the finnish hate your ... whatever the hell that is
529 2011-04-19 07:15:04 <Diablo-D3> its not a language
530 2011-04-19 07:15:09 <Diablo-D3> Im not sure what it is
531 2011-04-19 07:15:13 <mtrlt> thx
532 2011-04-19 07:15:19 <mtrlt> i hate your miner :-)
533 2011-04-19 07:15:27 <Diablo-D3> all nvidia users do.
534 2011-04-19 07:15:33 <mtrlt> i use ati
535 2011-04-19 07:15:44 <mtrlt> but seriously
536 2011-04-19 07:15:44 <Diablo-D3> then quit betraying the cause!
537 2011-04-19 07:15:49 <rly> Diablo-D3: did you improve yours?
538 2011-04-19 07:16:10 <rly> Diablo-D3: the python one was faster on my system a few months ago or so.
539 2011-04-19 07:16:28 <Diablo-D3> its faster
540 2011-04-19 07:16:50 <Diablo-D3> and it depends when you tried "a few months ago"
541 2011-04-19 07:16:51 TD_ has joined
542 2011-04-19 07:17:03 <Diablo-D3> I was experimenting with stuff that ended up not working "a few months ago"
543 2011-04-19 07:17:11 <Diablo-D3> as it stands, newest poclbm vs newest mine, mines faster
544 2011-04-19 07:17:30 <mtrlt> on my system, i get wildly varying performance on diablominer, like one second it's 80Mhash, another it's 160, but with poclbm i get steady 275 :P
545 2011-04-19 07:17:33 <ArtForz> ... on a 4850
546 2011-04-19 07:17:46 <Diablo-D3> ArtForz: no, I havent tested m0's for while
547 2011-04-19 07:17:54 <Diablo-D3> this is all other people's testing
548 2011-04-19 07:18:03 <Diablo-D3> mtrlt: sure you were reading the right number?
549 2011-04-19 07:18:09 <mtrlt> yea
550 2011-04-19 07:18:18 <Diablo-D3> second number on mine == since start of program
551 2011-04-19 07:18:59 <mtrlt> neither number never got as far as 200Mhps
552 2011-04-19 07:19:10 <mtrlt> tested for a minute or so though
553 2011-04-19 07:19:23 <Diablo-D3> try it again now
554 2011-04-19 07:19:40 <Diablo-D3> mtrlt: you're not on sdk 2.2 and 2.3, right?
555 2011-04-19 07:19:45 <mtrlt> nope, 2.4 afaik
556 2011-04-19 07:19:51 <Diablo-D3> good
557 2011-04-19 07:19:56 <Diablo-D3> 2.1 and 2.4 are the only safe sdks
558 2011-04-19 07:20:07 <Diablo-D3> and 2.1 is still faster
559 2011-04-19 07:20:30 <mtrlt> suggestion: add DiabloMiner-windows.bat or something :)
560 2011-04-19 07:20:43 <Diablo-D3> well, I keep asking someone to write one
561 2011-04-19 07:20:44 <Diablo-D3> but they never do
562 2011-04-19 07:20:51 <mtrlt> i'll do it then
563 2011-04-19 07:21:01 <Diablo-D3> I dont know how to write batch files that accept args
564 2011-04-19 07:21:08 <Diablo-D3> and I dont have a windows machine to test it on anyhow
565 2011-04-19 07:22:05 aninoni has joined
566 2011-04-19 07:22:42 <aninoni> hey, I've got a question, anybody up?
567 2011-04-19 07:22:51 <Diablo-D3> dont ask to ask, its rude
568 2011-04-19 07:23:21 <rly> Diablo-D3: what does one get on a 5770 with your miner?
569 2011-04-19 07:23:25 purpleposeidon has quit (Quit: ZNC - http://znc.sourceforge.net)
570 2011-04-19 07:23:43 <Diablo-D3> rly: not sure, should be above 150
571 2011-04-19 07:24:17 <sipa> aninoni: just ask your question
572 2011-04-19 07:24:36 <Diablo-D3> rly: the wiki is out of date imo
573 2011-04-19 07:24:37 danlucraft has joined
574 2011-04-19 07:24:38 <aninoni> :\ my bad. I just downloaded bitcoin on my 2nd computer, and when I hit generate coins, I don't see a "khash/sec" number anywhere. Google wasn't getting me anywhere
575 2011-04-19 07:24:48 <Diablo-D3> aninoni: blocks arent done downloading yet.
576 2011-04-19 07:25:13 <aninoni> gotcha. That's what I was forgetting
577 2011-04-19 07:25:16 <aninoni> thanks!
578 2011-04-19 07:25:40 <rly> Diablo-D3: the Python one gets 159 highish.
579 2011-04-19 07:25:54 <Diablo-D3> rly: mine should get at least that
580 2011-04-19 07:26:45 <ArtForz> wow, CL miners are slow
581 2011-04-19 07:27:34 <ArtForz> 177.2Mh at 850 core, 221 at 1060 core
582 2011-04-19 07:27:53 dissipate has quit (Ping timeout: 260 seconds)
583 2011-04-19 07:30:03 <CIA-89> DiabloMiner: Patrick McFarland master * rb9d62cd / DiabloMiner-Windows.bat : Added mtrlt's batch script for Windows - http://bit.ly/fzgcUr
584 2011-04-19 07:30:55 <Diablo-D3> ArtForz: heh
585 2011-04-19 07:30:59 <Diablo-D3> well I dont know what mine gets
586 2011-04-19 07:31:18 <rly> nvidia has CUDA, does ATI only have OpenCL?
587 2011-04-19 07:31:43 <Diablo-D3> CUDA is a closed source unstandardized language developed by nvidia to aid vendor lock in
588 2011-04-19 07:31:58 <Diablo-D3> it is inferior and obviously a giant hack job
589 2011-04-19 07:32:11 <ArtForz> ati has CAL, but thats kinda EOL
590 2011-04-19 07:32:19 <Diablo-D3> CAL isnt really the same thing
591 2011-04-19 07:32:30 <Diablo-D3> all new programs will be written in either opencl or directx compute
592 2011-04-19 07:32:37 <Diablo-D3> theres no reason to use something like cuda
593 2011-04-19 07:32:38 <ArtForz> well, still better than writing card-specific shader asm
594 2011-04-19 07:32:50 <Diablo-D3> ArtForz: cuda is higher level
595 2011-04-19 07:32:51 Tarlusk has joined
596 2011-04-19 07:33:00 <ArtForz> the high level cuda interface is
597 2011-04-19 07:33:11 aninoni has quit (Quit: Page closed)
598 2011-04-19 07:33:12 <ArtForz> which is what pretty much everyone uses
599 2011-04-19 07:33:12 <Diablo-D3> its like hlsl but for computing
600 2011-04-19 07:33:20 <Diablo-D3> its also retarded
601 2011-04-19 07:33:59 <ArtForz> iirc you can also assemble/load low level shaders using cuda, but no one does that
602 2011-04-19 07:34:38 <rly> Diablo-D3: based on what do you say that?
603 2011-04-19 07:35:04 <Diablo-D3> rly: nvidia has never hired any language designers
604 2011-04-19 07:35:16 <Diablo-D3> or, really, anyone who knows anything
605 2011-04-19 07:36:12 <ArtForz> imo, cuda gets abstractions and language design pretty wrong
606 2011-04-19 07:37:09 <rly> Diablo-D3: heh, I always got the impression the best work at nvidia (at least for graphics). Must have been good marketing then.
607 2011-04-19 07:37:13 <ArtForz> sure, hpc people write high performance compute kernels in cuda, but they also write similar stuff in fortran...
608 2011-04-19 07:37:26 <Diablo-D3> rly: no, not even close
609 2011-04-19 07:37:32 <Diablo-D3> rly: like, take chip design people
610 2011-04-19 07:37:43 <Diablo-D3> take the top 10 schools in the world for that (9 of which are in the US)
611 2011-04-19 07:38:04 <Diablo-D3> the absolute best of the class.... work for AMD or Intel or ARM.
612 2011-04-19 07:38:20 <rly> Diablo-D3: how old are you? 30ish?
613 2011-04-19 07:38:20 <Diablo-D3> whats left works for the old pre-AMD ATI or Nvidia or some other shitty outfit
614 2011-04-19 07:38:23 <Diablo-D3> rly: 27
615 2011-04-19 07:38:33 <rly> Diablo-D3: but you went to one of those 9?
616 2011-04-19 07:38:39 <Diablo-D3> nope
617 2011-04-19 07:39:06 <rly> Did your friends go and work at AMD et al?
618 2011-04-19 07:39:23 <Diablo-D3> now, the reason Radeons are now so fucking fast is those best of the best chip designers from AMD, that used to work on CPUs, are now working on GPUs as well
619 2011-04-19 07:39:29 <Diablo-D3> rly: I know people who do work for AMD, yes.
620 2011-04-19 07:39:37 <Diablo-D3> but this is a rather well known thing
621 2011-04-19 07:39:51 <ArtForz> yep
622 2011-04-19 07:40:01 <Diablo-D3> the best work for whoever has the highest salaries
623 2011-04-19 07:40:08 <ArtForz> pretty obvious really
624 2011-04-19 07:40:13 <Diablo-D3> major CPU manufacturers have the highest
625 2011-04-19 07:40:17 <Diablo-D3> oh, and IBM as well
626 2011-04-19 07:40:21 <Diablo-D3> I forgot about them
627 2011-04-19 07:40:31 <rly> Does IBM research pay a lot too?
628 2011-04-19 07:40:37 EPiSKiNG- has joined
629 2011-04-19 07:40:40 <Diablo-D3> probably
630 2011-04-19 07:40:49 <ArtForz> just look at nvidias perf/mm² and perf/W numbers even for recent parts, doesn't exactly scream "competent designers"
631 2011-04-19 07:40:53 <rly> I thought research was underpaid.
632 2011-04-19 07:40:53 <Diablo-D3> but IBM is off doing their own thing
633 2011-04-19 07:41:07 <Diablo-D3> IBM pulled out of PowerPC development mostly and is only doing POWER mostly
634 2011-04-19 07:41:11 <rly> Sure, IBM is just inventing the next 40 years of computing.
635 2011-04-19 07:41:13 <ArtForz> oh, and of course the whole fun with the deadend 3xx series
636 2011-04-19 07:41:24 <Diablo-D3> rly: sure. they invented the past 40, right? :D
637 2011-04-19 07:41:33 twobitcoins_ has joined
638 2011-04-19 07:41:36 <rly> Diablo-D3: they did, didn't they?
639 2011-04-19 07:41:43 <Diablo-D3> surprisingly, pretty much
640 2011-04-19 07:41:43 <ArtForz> aka "we created a new GPU arch that's slower than our last"
641 2011-04-19 07:41:53 <Diablo-D3> ArtForz: sub-arch
642 2011-04-19 07:41:58 <Diablo-D3> geforces havent changed since 8xxx
643 2011-04-19 07:42:09 <Diablo-D3> they just get smaller and get more pipes
644 2011-04-19 07:42:13 <ArtForz> well, the shaders themselves... not much
645 2011-04-19 07:42:20 <Diablo-D3> there hasnt been many major changes
646 2011-04-19 07:42:35 <rly> Diablo-D3: did you open up one these chips and look at them with some equipment?
647 2011-04-19 07:42:36 <Diablo-D3> Radeons otoh are now looping around to their second major redesign since AMD bought ATI
648 2011-04-19 07:42:37 <ArtForz> FFUs, frontend, mem controller, addition of DP support
649 2011-04-19 07:42:44 madveru has joined
650 2011-04-19 07:42:46 <Diablo-D3> rly: thats exceedingly difficult.
651 2011-04-19 07:42:46 TD_ has quit (Quit: TD_)
652 2011-04-19 07:43:04 <rly> Diablo-D3: I know, that's why I want to know where you get all this information.
653 2011-04-19 07:43:13 <Diablo-D3> rly: its common knowledge.
654 2011-04-19 07:43:17 <ArtForz> yep
655 2011-04-19 07:43:22 <rly> Not here.
656 2011-04-19 07:43:26 * Diablo-D3 shrugs.
657 2011-04-19 07:43:36 <rly> Then again, graphics cards are not a main interest of mine.
658 2011-04-19 07:43:45 curiosit1squared has joined
659 2011-04-19 07:43:47 dirtyfil1hy has joined
660 2011-04-19 07:43:53 <rly> I suppose you just care more about them than I do.
661 2011-04-19 07:44:06 <Diablo-D3> graphics cards didnt become interesting until people realized they've turned into ultra wide DSPs
662 2011-04-19 07:44:26 brizna has joined
663 2011-04-19 07:44:28 <ArtForz> yep, return of the vector processor ;)
664 2011-04-19 07:44:41 <Diablo-D3> the return of the return... of... the return. I think.
665 2011-04-19 07:44:52 <Diablo-D3> we've looped around that point a few times
666 2011-04-19 07:44:59 <ArtForz> yeah, 3 times or so
667 2011-04-19 07:45:00 <Diablo-D3> every so often someone goes HEY I HAVE A NEW IDEA
668 2011-04-19 07:45:08 <ArtForz> ... and reinvents LISP
669 2011-04-19 07:45:12 <ArtForz> oh wait... hardware ;)
670 2011-04-19 07:45:15 <Diablo-D3> hah
671 2011-04-19 07:45:20 <Diablo-D3> oh wait, LISP MACHINES
672 2011-04-19 07:45:22 <Diablo-D3> WAAAGH
673 2011-04-19 07:45:44 Llamab4t has joined
674 2011-04-19 07:45:54 <Diablo-D3> but yeah, everytime someone says they have a new idea, IBM invented it and perfected it and threw it out 40 years ago
675 2011-04-19 07:45:57 <Diablo-D3> I shit you not
676 2011-04-19 07:46:38 <joepie91> lol
677 2011-04-19 07:46:41 <mtrlt> :D
678 2011-04-19 07:46:41 <Diablo-D3> the entire fucking process is "new dedicated totally 100% fixed function chip" -> "cheaper next gen more flexiable chip" -> "totally generic reprogrammable chip" -> "its now a standard part of the cpy"
679 2011-04-19 07:46:44 <Diablo-D3> and then we repeat.
680 2011-04-19 07:46:54 <Diablo-D3> *cpu
681 2011-04-19 07:47:07 <ArtForz> well... dunno about that
682 2011-04-19 07:47:26 <Diablo-D3> FPU coprocessors.. bigger than the CPU.
683 2011-04-19 07:47:31 <ArtForz> yeah
684 2011-04-19 07:47:39 <Diablo-D3> now we have like 12 fucking FPU ALUs on the goddamned CPU now
685 2011-04-19 07:47:59 <rly> Simpsons did it <=> IBM did it.
686 2011-04-19 07:48:06 <Diablo-D3> L2 cache controllers in the northbridge with expensive extra fast ram in sockets
687 2011-04-19 07:48:09 <Diablo-D3> now its all in the CPU
688 2011-04-19 07:48:40 <Diablo-D3> gigantic fucking data routing chips going between CPUs in giant supercomputers
689 2011-04-19 07:48:40 Llamab4t is now known as BCBot
690 2011-04-19 07:48:40 madveru is now known as manveru
691 2011-04-19 07:48:53 <ArtForz> except we still dont have high bandwidth mem interfaces for CPUs, which you kinda want to make good use of a massive vector unit
692 2011-04-19 07:48:56 <Diablo-D3> AMD calls this "hypertransport", and its almost entirely on the CPU.
693 2011-04-19 07:49:05 <Diablo-D3> ArtForz: we're getting there.
694 2011-04-19 07:49:35 <Diablo-D3> Im waiting for AMD to start shipping SuperFusions that support like 4 level cache
695 2011-04-19 07:49:39 <ArtForz> well, not with conventional sockets
696 2011-04-19 07:49:40 <Diablo-D3> L1, L2, L3, and then "video ram"
697 2011-04-19 07:50:01 topi` has joined
698 2011-04-19 07:50:20 <ArtForz> and your exeternal mem bus is still only 128 bits wide and clocked at a snails pace
699 2011-04-19 07:50:21 <Diablo-D3> and you have this gigantic blob of like 2gb of "video ram" (which is neither exclusively for the video, nor is it ram addressable by things)
700 2011-04-19 07:50:31 <Diablo-D3> which leads to an interesting thing
701 2011-04-19 07:50:49 <Diablo-D3> "standard part of the cpu" -> "GETS FUCKING REINVENTED AND WE DO THE SAME EXACT THING AGAIN"
702 2011-04-19 07:51:01 <ArtForz> that'd be pretty much the only option, a G or so of ram on package
703 2011-04-19 07:51:13 <Diablo-D3> actually
704 2011-04-19 07:51:23 <Diablo-D3> if we could shove like a huge shitload of L3 on a chip
705 2011-04-19 07:51:27 <Diablo-D3> that'd solve a lot of problems
706 2011-04-19 07:51:35 <Diablo-D3> then agian
707 2011-04-19 07:51:36 <ArtForz> and without stacked silicon that will be one *really* expensive package
708 2011-04-19 07:51:36 <Diablo-D3> you have power
709 2011-04-19 07:51:42 <Diablo-D3> where 256MB of L3 isnt uncommon....
710 2011-04-19 07:52:03 <Diablo-D3> Im surprised stacked silicon isnt more common
711 2011-04-19 07:52:16 <Diablo-D3> it solves a lot of design issues
712 2011-04-19 07:52:27 <Pander> Sometimes "aticonfig --odgt --adapter=all" results in "ERROR - X needs to be running to perform ATI Overdrive(TM) commands" How to fix this?
713 2011-04-19 07:52:37 <ArtForz> they're still pretty much working on making it mass-manufacturable
714 2011-04-19 07:52:54 <Diablo-D3> Pander: X needs to be running.
715 2011-04-19 07:52:57 <Pander> it is
716 2011-04-19 07:53:01 <ArtForz> main problem iirc is still failures in the inter-die interface
717 2011-04-19 07:53:03 <Diablo-D3> your DISPLAY isnt set right.
718 2011-04-19 07:53:21 <MagicalTux> Pander, make sure you're doing this under the user logged in on X
719 2011-04-19 07:53:21 <Pander> I had the same yesterday, one ssh session did return correct output, another didn't (both root)
720 2011-04-19 07:53:25 <Pander> ah, that coudl be it
721 2011-04-19 07:53:28 <ArtForz> which are kinda hard to fix with current process tech when you have 2 dies with massively different thermal profiles
722 2011-04-19 07:53:42 <netxshare> should I go with the 6990 or the 6970?
723 2011-04-19 07:54:10 <Pander> thanks MagicalPenguin
724 2011-04-19 07:54:17 <ArtForz> it should work fine for flash, dram and other low-power stuff
725 2011-04-19 07:54:29 <Diablo-D3> netxshare: 6990 and then clock the GPUs back to 6970 speeds
726 2011-04-19 07:54:46 <Diablo-D3> ArtForz: well, unless they're NOT different thermal profiles ;)
727 2011-04-19 07:55:02 <ArtForz> ... how would that work?
728 2011-04-19 07:55:52 <ArtForz> you have a structured hot design on one die, a sea of ram cells on the other
729 2011-04-19 07:56:16 <netxshare> damn mrb wants 400 btc for his miner
730 2011-04-19 07:56:22 <Diablo-D3> netxshare: mrb can shove it.
731 2011-04-19 07:56:31 <ArtForz> we already kinda do the same shit with wirebonded multi-die packages
732 2011-04-19 07:56:32 <Diablo-D3> 400btc can buy you another video card
733 2011-04-19 07:56:35 <netxshare> at least he is giving the code out this time
734 2011-04-19 07:56:47 <Diablo-D3> what is his miner written in? CAL?
735 2011-04-19 07:56:52 <ArtForz> yep
736 2011-04-19 07:56:54 <netxshare> I am not sure
737 2011-04-19 07:56:56 <ArtForz> CAL IL
738 2011-04-19 07:56:59 <netxshare> oh
739 2011-04-19 07:57:06 <netxshare> that's not what I want anyways
740 2011-04-19 07:57:06 <Pander> True that this is the fastest way to get GPU temp? export DISPLAY=:0; aticonfig --odgt --adapter=1|grep Temperature|sed -e 's/.* - \(.*\) C$/\1/'
741 2011-04-19 07:57:32 <netxshare> I just want to use stream and a sha256 shader
742 2011-04-19 07:57:33 <ArtForz> quite a bit slower than my cal miner on 5xxx/68xx, faster on 69xx
743 2011-04-19 07:58:13 <ArtForz> seriously, writing a sha256 shader is not rocket science
744 2011-04-19 07:58:22 <netxshare> I still can't get nvidia's compiler to dump their ptx files
745 2011-04-19 07:58:28 <netxshare> I would not think so
746 2011-04-19 07:58:29 <Diablo-D3> ArtForz: heh
747 2011-04-19 07:58:31 <Diablo-D3> well
748 2011-04-19 07:58:32 <ArtForz> and btw, a generic sha256 shader is way too slow
749 2011-04-19 07:58:38 <Diablo-D3> except you did shit in your CL kernel that probably isnt even sane
750 2011-04-19 07:58:42 <Diablo-D3> also
751 2011-04-19 07:58:49 <ArtForz> what?
752 2011-04-19 07:58:52 <Diablo-D3> if its a lot slower than your miner... its slower than MY miner
753 2011-04-19 07:59:04 <ArtForz> err... no
754 2011-04-19 07:59:15 <ArtForz> iirc his gets 670Mh/s or so on stock 5970
755 2011-04-19 07:59:19 <ArtForz> err
756 2011-04-19 07:59:21 <ArtForz> 570
757 2011-04-19 07:59:42 <Diablo-D3> 570?
758 2011-04-19 07:59:43 <Diablo-D3> thats it?
759 2011-04-19 07:59:45 <ArtForz> yeah
760 2011-04-19 07:59:48 <ArtForz> kinda ... "meh"
761 2011-04-19 07:59:54 <Diablo-D3> someone was getting 615 with mine on a stock 5970
762 2011-04-19 08:00:03 <ArtForz> sorry, no way
763 2011-04-19 08:00:06 TheAncientGoat has joined
764 2011-04-19 08:00:13 <netxshare> what is the best option for stick 4 6970's in a computer
765 2011-04-19 08:00:15 <netxshare> mother board wise
766 2011-04-19 08:00:24 <netxshare> I am guessing going amd cpu would be the best
767 2011-04-19 08:00:26 <netxshare> price wise
768 2011-04-19 08:00:27 <ArtForz> "none", build 2 boxes
769 2011-04-19 08:00:29 <Diablo-D3> netxshare: easy... buy two 6990s.
770 2011-04-19 08:00:44 <netxshare> doing a 4x setup is a bad idea?
771 2011-04-19 08:00:50 <ArtForz> yep
772 2011-04-19 08:01:01 <Diablo-D3> its difficult unless you're Art
773 2011-04-19 08:01:05 <ArtForz> way too much fucking with shit
774 2011-04-19 08:01:07 <netxshare> http://blog.zorinaq.com/?e=42
775 2011-04-19 08:01:12 <netxshare> I think that's mrb again
776 2011-04-19 08:01:19 <netxshare> but he used 5970s
777 2011-04-19 08:01:28 <ArtForz> yeah
778 2011-04-19 08:02:01 <ArtForz> $170 for PCIe extenders
779 2011-04-19 08:02:09 zyb has joined
780 2011-04-19 08:02:15 <netxshare> =/
781 2011-04-19 08:02:15 <ArtForz> you can get 2nd mobo+cpu+ram quite a bit cheaper than that
782 2011-04-19 08:02:17 <netxshare> yeah
783 2011-04-19 08:02:20 <ArtForz> err... $140
784 2011-04-19 08:02:37 <netxshare> the 4x pcie board was 170
785 2011-04-19 08:02:37 <Diablo-D3> yeah
786 2011-04-19 08:02:42 <Diablo-D3> but art built like 9000 boxes
787 2011-04-19 08:02:57 <Diablo-D3> his house must be audable down the street
788 2011-04-19 08:03:02 <netxshare> he always does crazy things
789 2011-04-19 08:03:05 <ArtForz> and then he had to build a custom wiring harness, modify the extenders to feed the cards power directly from PSU 12V lines, ...
790 2011-04-19 08:03:23 <ArtForz> and it doesnt fit in a normal case
791 2011-04-19 08:03:27 <netxshare> yeah
792 2011-04-19 08:03:28 <Diablo-D3> ArtForz: so wait, why do you think 615 is impossible?
793 2011-04-19 08:03:32 <netxshare> I was not going to use a case
794 2011-04-19 08:03:49 <netxshare> I have plenty of room in my nice cold basement
795 2011-04-19 08:04:00 <ArtForz> because 618 is theoretical max at 100% VLIW util and using BFI_INT
796 2011-04-19 08:04:02 sacarlson has quit (Ping timeout: 260 seconds)
797 2011-04-19 08:04:15 <Diablo-D3> hrm, maybe it was 605
798 2011-04-19 08:04:23 TheKid has quit (Read error: Connection reset by peer)
799 2011-04-19 08:04:28 <ArtForz> OpenCL can't use BFI_INT, instant 6% slowdown
800 2011-04-19 08:04:42 <Diablo-D3> yes, and when it can
801 2011-04-19 08:04:45 <Diablo-D3> 6% speedup
802 2011-04-19 08:04:46 <Diablo-D3> :D
803 2011-04-19 08:05:00 <netxshare> could opencl just add support in for BFI_INT?
804 2011-04-19 08:05:04 <ArtForz> not easily
805 2011-04-19 08:05:11 <netxshare> ah
806 2011-04-19 08:05:14 <Diablo-D3> yes, actually
807 2011-04-19 08:05:18 <ArtForz> well, unless ATI fixes their drivers first
808 2011-04-19 08:05:26 <Diablo-D3> since you'd end up changing the code back to calling .. uh...
809 2011-04-19 08:05:29 <netxshare> like that will happen
810 2011-04-19 08:05:32 <Diablo-D3> whatever the fuck the name of that function is
811 2011-04-19 08:05:37 <ArtForz> IL bytecode doesn't support it
812 2011-04-19 08:05:41 <Diablo-D3> ArtForz: not YET
813 2011-04-19 08:05:45 <Diablo-D3> but look at 2.4
814 2011-04-19 08:05:48 <Diablo-D3> now requires at least 11.1
815 2011-04-19 08:05:55 <ArtForz> yeah... so... maybe in 6 months?
816 2011-04-19 08:06:01 <Diablo-D3> so theres nothing against 2.5 requiring something newer
817 2011-04-19 08:06:03 <Insti> genjix: let me know if you have server questions.
818 2011-04-19 08:06:14 <Diablo-D3> damnuit
819 2011-04-19 08:06:15 <Diablo-D3> damnit
820 2011-04-19 08:06:19 <ArtForz> by then I probably wont have a a gpu cluster anymore anyways
821 2011-04-19 08:06:21 <netxshare> so does http://bitcoindarts.movoda.net/ payout?
822 2011-04-19 08:06:21 <Diablo-D3> who here has a 5970 running 2.1?
823 2011-04-19 08:06:32 <Diablo-D3> ArtForz: give me one of your 5970s dude :<
824 2011-04-19 08:06:46 <netxshare> why do you want a 5970?
825 2011-04-19 08:06:52 <lfm> just one?
826 2011-04-19 08:06:53 <Diablo-D3> netxshare: miner development
827 2011-04-19 08:06:55 NWTSPV has left ()
828 2011-04-19 08:06:56 <Diablo-D3> lfm: just one.
829 2011-04-19 08:07:14 <netxshare> oh
830 2011-04-19 08:07:32 <Diablo-D3> I mean, Ive probably made people that much money
831 2011-04-19 08:07:39 <Diablo-D3> its the least the community could do for me :<
832 2011-04-19 08:08:00 <netxshare> ill make a custom ati miner
833 2011-04-19 08:08:08 <netxshare> but require it be a part of a pool
834 2011-04-19 08:08:11 <netxshare> and ill take a %
835 2011-04-19 08:08:17 <netxshare> that's better then 400 btc right
836 2011-04-19 08:08:50 <lfm> netxshare: ask puddinpop how well that worked for him
837 2011-04-19 08:08:51 <ArtForz> puddinpop tried that, didnt exactly work well
838 2011-04-19 08:09:10 <netxshare> oh someone did try it?
839 2011-04-19 08:09:16 <Diablo-D3> puddingpop got ass raped.
840 2011-04-19 08:09:34 <netxshare> is that why I have not seen any updates to rpc-miner
841 2011-04-19 08:09:49 <ArtForz> the only tricky part is writing a good kernel, and that's pretty easy to rip from another miner
842 2011-04-19 08:10:19 <netxshare> well his cuda kernel sucks
843 2011-04-19 08:10:39 <netxshare> if opencl can do 10-15mh/s more
844 2011-04-19 08:10:49 <ArtForz> yep
845 2011-04-19 08:10:54 <Diablo-D3> actually, can it?
846 2011-04-19 08:10:58 <netxshare> yes
847 2011-04-19 08:11:02 <Diablo-D3> I know art's kernel is now really fucking fast on nvidia
848 2011-04-19 08:11:08 <Diablo-D3> since it basically forces the compiler to work
849 2011-04-19 08:11:12 <netxshare> I get 121/122 in opencl on m0ms
850 2011-04-19 08:11:20 <Diablo-D3> it was like a 2% increase on ati
851 2011-04-19 08:11:24 <Diablo-D3> but like a 15% increase on nvidia
852 2011-04-19 08:11:54 <ArtForz> yeah, you'd expect compilers to be good at shit like invariant extraction... they're not
853 2011-04-19 08:12:15 <netxshare> I would expect nvcc to output when using the correct flags
854 2011-04-19 08:12:26 <Diablo-D3> netxshare: what is mine getting on your nvidia shit?
855 2011-04-19 08:12:38 <netxshare> have not tried it
856 2011-04-19 08:12:42 <netxshare> I dislike java
857 2011-04-19 08:12:42 brunner has quit (Ping timeout: 260 seconds)
858 2011-04-19 08:12:49 <netxshare> but ill give it a shot
859 2011-04-19 08:12:55 <Diablo-D3> yet you like python? wtf dude.
860 2011-04-19 08:13:06 <netxshare> yes
861 2011-04-19 08:13:09 <netxshare> I <3 python
862 2011-04-19 08:13:36 mtrlt has quit (Read error: Connection reset by peer)
863 2011-04-19 08:13:45 <Diablo-D3> =|
864 2011-04-19 08:13:57 <netxshare> I just hated writing code in it
865 2011-04-19 08:16:09 <Diablo-D3> heh
866 2011-04-19 08:17:18 mologie has joined
867 2011-04-19 08:18:54 sacarlson has joined
868 2011-04-19 08:19:08 <netxshare> it's mad at me
869 2011-04-19 08:19:10 zyb_ has joined
870 2011-04-19 08:19:13 zyb has quit (Remote host closed the connection)
871 2011-04-19 08:19:16 <netxshare> says it can't connect to Bitcoin
872 2011-04-19 08:19:34 <netxshare> seems to fail to connect to deepbit.net
873 2011-04-19 08:20:02 zyb_ is now known as zyb
874 2011-04-19 08:21:54 <netxshare> there we go
875 2011-04-19 08:22:29 <netxshare> 114822/114830 khash/sec
876 2011-04-19 08:23:26 <MagicalTux> no block since 2011-04-19 07:55:05
877 2011-04-19 08:23:41 <Diablo-D3> netxshare: what was m0's getting?
878 2011-04-19 08:23:47 <sipa> MagicalTux: what timezone?
879 2011-04-19 08:23:50 <netxshare> 121/122
880 2011-04-19 08:23:54 <netxshare> tho I am using some commands
881 2011-04-19 08:24:03 <netxshare> -f 64 -w 128 or something like that
882 2011-04-19 08:24:04 <Diablo-D3> which ones?
883 2011-04-19 08:24:09 <netxshare> oh and -v
884 2011-04-19 08:24:15 slush has joined
885 2011-04-19 08:24:17 <Diablo-D3> erm, -f should be a multiple or divsor of 60
886 2011-04-19 08:24:20 <MagicalTux> sipa, blockexplorer's timezone
887 2011-04-19 08:24:24 <Diablo-D3> and my miner doesnt use -v
888 2011-04-19 08:24:31 <netxshare> maybe that's not what I am using then
889 2011-04-19 08:24:33 <netxshare> let me check
890 2011-04-19 08:24:54 <Diablo-D3> and -w 64 is the default on my miner, its generally superior
891 2011-04-19 08:28:02 <netxshare> -v -a 5 -f 1 -w 256
892 2011-04-19 08:28:59 <Diablo-D3> mine doesnt use -v or -a
893 2011-04-19 08:29:00 <netxshare> these drivers do not like that setting anymore
894 2011-04-19 08:29:02 <Diablo-D3> Im not sure what -a is
895 2011-04-19 08:29:32 <netxshare> aggression
896 2011-04-19 08:29:42 <Diablo-D3> yeah but whats it do?
897 2011-04-19 08:29:59 <netxshare> help.txt in rpc-miner says
898 2011-04-19 08:30:00 <netxshare> ion=X
899 2011-04-19 08:30:00 <netxshare> Specifies how many hashes (2^(X-1)) per kernel thread will be calculated.
900 2011-04-19 08:30:00 <netxshare> The default is 6. It starts at 1 and goes to 32, with each successive
901 2011-04-19 08:30:00 <netxshare> number meaning double the number of hashes. Sane values are 1 to 12 or
902 2011-04-19 08:30:00 <netxshare> maybe 14 if you have some super card.
903 2011-04-19 08:30:17 <Diablo-D3> uh, wtf?
904 2011-04-19 08:30:23 <Diablo-D3> that meanst
905 2011-04-19 08:30:28 <Diablo-D3> -a overrides -f
906 2011-04-19 08:30:48 <Diablo-D3> since -f tries to set kernel length by time
907 2011-04-19 08:31:31 <sipa> maybe one influences worksize, and the other an inner loop?
908 2011-04-19 08:31:41 <Diablo-D3> sipa: doubt it.
909 2011-04-19 08:31:54 <netxshare> I just started to play with settings
910 2011-04-19 08:32:02 <netxshare> put those in and got much more then any other setting
911 2011-04-19 08:32:03 <Diablo-D3> because 1024 is the most optimum for 5xxx/6xxx no matter what
912 2011-04-19 08:32:21 <netxshare> -v is what gives me 122
913 2011-04-19 08:32:29 <netxshare> I just changed it to do -v only
914 2011-04-19 08:32:33 <netxshare> and I am getting -122
915 2011-04-19 08:32:40 <Diablo-D3> yeah, and -v shouldnt work
916 2011-04-19 08:32:41 <netxshare> er 121/122
917 2011-04-19 08:32:43 <Diablo-D3> its a gross hack
918 2011-04-19 08:32:46 <netxshare> lol
919 2011-04-19 08:32:54 <netxshare> maybe I should not use it then
920 2011-04-19 08:33:08 <Diablo-D3> well, it abuses vectors to implictly force vector math where it doesnt belong
921 2011-04-19 08:33:26 <Diablo-D3> the compiler is far better at doing this correctly
922 2011-04-19 08:33:39 <netxshare> what is the difference between poclbm and poclbm-mod?
923 2011-04-19 08:33:52 <Diablo-D3> dunno
924 2011-04-19 08:35:14 <netxshare> has anyone used your kernel with poclbm?
925 2011-04-19 08:35:29 <Diablo-D3> its art's kernel
926 2011-04-19 08:35:36 <netxshare> oh
927 2011-04-19 08:35:38 <Diablo-D3> both my miner and poclbm uses it
928 2011-04-19 08:35:41 <Pander> netxshare, yes bu accident
929 2011-04-19 08:35:46 <Pander> bu=by
930 2011-04-19 08:35:54 <netxshare> so they are both the same
931 2011-04-19 08:35:57 <Diablo-D3> mostly
932 2011-04-19 08:36:00 <netxshare> I see
933 2011-04-19 08:36:04 <Diablo-D3> he obviously has modifications to make -v work
934 2011-04-19 08:36:09 <netxshare> yeah
935 2011-04-19 08:36:24 <netxshare> oh I see it now // ArtForz's kernel hehe
936 2011-04-19 08:37:12 <sipa> ;;bc,blocks
937 2011-04-19 08:37:13 <gribble> 119086
938 2011-04-19 08:37:20 <netxshare> ill give cuda another shot at compiling
939 2011-04-19 08:38:10 <Diablo-D3> whats poclbm do without -v?
940 2011-04-19 08:39:15 <Pander> does poclmb use all my 4 cores if I say to use -d0 ?
941 2011-04-19 08:39:24 <Pander> i.e. CPU cores
942 2011-04-19 08:40:08 <netxshare> 115/117
943 2011-04-19 08:40:27 <netxshare> nope staying at 115 now
944 2011-04-19 08:42:03 <Diablo-D3> so its basically identical to mine
945 2011-04-19 08:42:14 <Diablo-D3> Pander: it doesnt use any cpu cores, really.
946 2011-04-19 08:42:22 <Diablo-D3> its single threaded and idle 99% of the time
947 2011-04-19 08:42:54 <MagicalTux> did someone stop mining?
948 2011-04-19 08:43:04 <MagicalTux> (or is the whole network unlucky?)
949 2011-04-19 08:43:30 <Pander> Diablo-D3: if I have 2x dual radeon, should I use -d1, ..., -d4? what is -d0 then?
950 2011-04-19 08:43:47 <Pander> ah, you mean single thread on CPU or GPU?
951 2011-04-19 08:44:24 <rly> Pander: CPU
952 2011-04-19 08:44:26 <Diablo-D3> single CPU thread.
953 2011-04-19 08:44:30 <sipa> ;;bc,prob 700000000 45m
954 2011-04-19 08:44:30 <Diablo-D3> Pander: on my miner, you dont need that.
955 2011-04-19 08:44:31 <gribble> 0.991478371038
956 2011-04-19 08:44:40 <Diablo-D3> it selects all GPUs and uses them
957 2011-04-19 08:45:57 <Pander> thnx
958 2011-04-19 08:47:30 <netxshare> deepbit.net is still going
959 2011-04-19 08:47:30 <Diablo-D3> I dont know why poclbm ever did stuff that way
960 2011-04-19 08:47:40 <Diablo-D3> running multiple miners == fail
961 2011-04-19 08:47:54 <netxshare> yours selects them all and makes new threads for each one?
962 2011-04-19 08:48:25 <Diablo-D3> well, 3 threads per gpu
963 2011-04-19 08:48:33 <Diablo-D3> it does multiple queue stuffing
964 2011-04-19 08:48:47 <netxshare> well at least it does that
965 2011-04-19 08:49:00 <Diablo-D3> it also helps with pool mining
966 2011-04-19 08:49:10 <Diablo-D3> http accesses dont block mining
967 2011-04-19 08:49:21 <netxshare> does it support long poll
968 2011-04-19 08:49:32 <Diablo-D3> no, I dont see value in long poll
969 2011-04-19 08:50:01 <Diablo-D3> I'm also writing pool software
970 2011-04-19 08:50:02 <netxshare> pool ops just like it when you use ones that do
971 2011-04-19 08:50:04 <netxshare> yeah
972 2011-04-19 08:50:08 <netxshare> that's what I have been working on
973 2011-04-19 08:50:15 <Diablo-D3> so Im probably just going to implement my own mining api
974 2011-04-19 08:50:31 larsivi has joined
975 2011-04-19 08:50:50 <netxshare> I just need a setup like deepbit
976 2011-04-19 08:51:02 <Diablo-D3> well, mine'd be more like slush's
977 2011-04-19 08:51:08 <netxshare> or one that will just allow a unlimited number of computers to mine
978 2011-04-19 08:51:14 <netxshare> whats his url, ill have to check it out
979 2011-04-19 08:51:18 <netxshare> only used deepbit
980 2011-04-19 08:51:57 <Diablo-D3> http://mining.bitcoin.cz/
981 2011-04-19 08:52:31 <netxshare> I figured they all pay about the same
982 2011-04-19 08:53:06 <rly> Diablo-D3: with all the default options, yours gets only 73M/s on a 5770.
983 2011-04-19 08:53:24 <Diablo-D3> rly: sounds like your system is configured wrong
984 2011-04-19 08:53:32 <Diablo-D3> you should be on catalyst 10.9 through 11, plus sdk 2.1
985 2011-04-19 08:54:06 <rly> Diablo-D3: I did make a mistake.
986 2011-04-19 08:54:18 <Diablo-D3> oh?
987 2011-04-19 08:54:29 <rly> Diablo-D3: for some reason the other miner also kept running.
988 2011-04-19 08:54:34 <Diablo-D3> hah
989 2011-04-19 08:54:50 <rly> It now doesn't come further than 145M/s.
990 2011-04-19 08:55:06 <rly> I think the worksize is not optimal for this card.
991 2011-04-19 08:55:47 <Diablo-D3> what, a 5770?
992 2011-04-19 08:56:13 <Diablo-D3> 64 is the most optimum
993 2011-04-19 08:56:17 <rly> Diablo-D3: yes
994 2011-04-19 08:56:24 <Diablo-D3> what driver and sdk are you on?
995 2011-04-19 08:56:27 <rly> I see I made yet another mistake.
996 2011-04-19 08:56:50 <Diablo-D3> oh?
997 2011-04-19 08:57:03 <netxshare> what are you writing your pool code in?
998 2011-04-19 08:57:09 <Diablo-D3> netxshare: java.
999 2011-04-19 08:58:09 <netxshare> ah
1000 2011-04-19 08:58:58 <netxshare> do all of these pools use their own code or are they using a open source
1001 2011-04-19 08:59:43 <Diablo-D3> all closed source
1002 2011-04-19 08:59:43 <rly> Diablo-D3: about every 5 seconds I get that bitcoin is not connected, but it is hashing.
1003 2011-04-19 08:59:53 <Diablo-D3> rly: then its not connected
1004 2011-04-19 09:00:02 <Diablo-D3> what args are you feeding it?
1005 2011-04-19 09:00:18 <ersi> netxshare: Feel free to code an open one, it'll probably be used quickly
1006 2011-04-19 09:00:26 <rly> ps aux | grep bitcoind says it has been started.
1007 2011-04-19 09:00:29 <ersi> A bazillion pools spawning and dying at the same time
1008 2011-04-19 09:00:53 <rly> I also do not get why it is hashing if it is not connected.
1009 2011-04-19 09:02:01 <rly> Diablo-D3: cd $HOME/src/DiabloMiner && ./DiabloMiner-Linux.sh -w 64 -u u -p p
1010 2011-04-19 09:02:13 <Diablo-D3> rly: -w 64 is the default anyways
1011 2011-04-19 09:02:35 <Diablo-D3> rly: are you sure your -u and -p are right? do other miners connect fine?
1012 2011-04-19 09:02:46 <rly> Diablo-D3: yes and yes
1013 2011-04-19 09:03:01 <Diablo-D3> huh. is your u and p sane?
1014 2011-04-19 09:03:18 <rly> Diablo-D3: now it has been a few minutes since the last cannot connect.
1015 2011-04-19 09:03:21 <netxshare> I do like Estimated reward data on that site
1016 2011-04-19 09:03:29 <netxshare> instead of waiting 1 hour for it to update
1017 2011-04-19 09:03:30 <rly> Diablo-D3: still it is 10M slower than the Python one.
1018 2011-04-19 09:03:38 <rly> Er 14M/s
1019 2011-04-19 09:04:08 <Diablo-D3> rly: shouldnt be. try -f 1 and wait like 5 minutes
1020 2011-04-19 09:04:22 <Diablo-D3> and you still havent answered, which catalyst, which version of the sdk
1021 2011-04-19 09:04:55 <rly> Diablo-D3: shouldn't yours be faster than the Python one regardless of those variables?
1022 2011-04-19 09:05:05 <Diablo-D3> not quite
1023 2011-04-19 09:05:17 <Diablo-D3> it makes the drivers do stupid shit sometimes
1024 2011-04-19 09:08:46 <ArtForz> well, just looking at em too hard makes the drivers do stupid shit
1025 2011-04-19 09:09:14 <netxshare> mine are getting mad
1026 2011-04-19 09:09:37 <netxshare> everytime I stop the miner they stop responding
1027 2011-04-19 09:10:05 <Diablo-D3> happens if you use low -f values, its harmless
1028 2011-04-19 09:10:42 <Diablo-D3> you know what
1029 2011-04-19 09:10:48 <Diablo-D3> m0's doesnt implement loops at all
1030 2011-04-19 09:10:51 <Diablo-D3> I wonder why
1031 2011-04-19 09:11:15 <ArtForz> too high kernel exec time makes the drivers freak
1032 2011-04-19 09:13:12 <Diablo-D3> rly: you're using -w 64 on poclbm too?
1033 2011-04-19 09:14:25 brizna has quit (Ping timeout: 260 seconds)
1034 2011-04-19 09:21:32 * Diablo-D3 adds an experimental option
1035 2011-04-19 09:22:19 <netxshare> ooo
1036 2011-04-19 09:22:21 <netxshare> crashed
1037 2011-04-19 09:22:30 <netxshare> system restarted
1038 2011-04-19 09:22:42 <Diablo-D3> nice :D
1039 2011-04-19 09:23:00 <netxshare> ever since I installed these drivers
1040 2011-04-19 09:23:11 <Diablo-D3> which ones?
1041 2011-04-19 09:25:45 <rly> Diablo-D3: no -w 128. I now installed the 2.4 SDK, but now it doesn't work at all anymore. Error: Compilation from LLVMIR binary to IL text failed!
1042 2011-04-19 09:26:06 <Diablo-D3> 2.4 requires 11.1 and up
1043 2011-04-19 09:26:34 <rly> Diablo-D3: ok (the software (compiler infrastructure) should say that)
1044 2011-04-19 09:26:39 <netxshare> the dev drivers
1045 2011-04-19 09:26:43 <Diablo-D3> the website you download it from say this.
1046 2011-04-19 09:26:46 <netxshare> and then they released new ones today
1047 2011-04-19 09:26:47 <Diablo-D3> *says
1048 2011-04-19 09:27:09 <netxshare> 270.61
1049 2011-04-19 09:27:43 <Diablo-D3> thats nivida =P
1050 2011-04-19 09:36:26 edcba_ is now known as edcba
1051 2011-04-19 09:37:38 brizna has joined
1052 2011-04-19 09:37:47 brizna has quit (Client Quit)
1053 2011-04-19 09:38:08 <ArtForz> octal chicken
1054 2011-04-19 09:38:09 devon_hillard has joined
1055 2011-04-19 09:40:39 <Diablo-D3> you can now adjust loop size on my miner
1056 2011-04-19 09:42:59 <slush> I just found 5870 in bazaar, owner wrote 'picture suddenly failed during gaming'. But there is real chance that mining will still work, right?
1057 2011-04-19 09:43:19 <ArtForz> maybe, possibly
1058 2011-04-19 09:43:20 <slush> it is 7$ including shipping...
1059 2011-04-19 09:43:25 <Diablo-D3> slush: $7? try it.
1060 2011-04-19 09:43:27 <netxshare> id buy it
1061 2011-04-19 09:43:32 <ArtForz> if it has ref cooler, buy it
1062 2011-04-19 09:43:37 <slush> yes
1063 2011-04-19 09:43:44 <slush> ok, thanks :)
1064 2011-04-19 09:43:45 <ArtForz> the cooler alone can be sold for $20+
1065 2011-04-19 09:43:53 <ArtForz> so even if it's completely dead, it's a bargain
1066 2011-04-19 09:44:18 <slush> :)
1067 2011-04-19 09:45:25 <netxshare> hrm
1068 2011-04-19 09:46:23 <Diablo-D3> hah
1069 2011-04-19 09:48:02 <CIA-89> DiabloMiner: Patrick McFarland master * r9306c65 / (2 files in 2 dirs): Loop length can now be configured with -z - http://bit.ly/idgJx9
1070 2011-04-19 09:48:56 <netxshare> need to set CUDA_FORCE_PTX_JIT=1
1071 2011-04-19 09:49:36 <Diablo-D3> hey rly
1072 2011-04-19 09:49:49 <Diablo-D3> rly: download newest binary, try -z 0
1073 2011-04-19 09:50:20 <mrb_> slush: then you can sell the cooler on ebay in an ad titled "HD 5870 (cooler) - $20!!"
1074 2011-04-19 09:50:43 <rly> Diablo-D3: what does it do?
1075 2011-04-19 09:50:51 <rly> Diablo-D3: did you make changes in the last few minutes?
1076 2011-04-19 09:50:56 <Diablo-D3> yeah I just made changes
1077 2011-04-19 09:51:02 <Diablo-D3> -z 0 turns looping off
1078 2011-04-19 09:51:10 <mrb_> :)
1079 2011-04-19 09:51:24 <rly> Diablo-D3: why would you want that?
1080 2011-04-19 09:51:53 <slush> mrb_: :)
1081 2011-04-19 09:51:54 <Diablo-D3> because I want to know why m0's is faster in some cases
1082 2011-04-19 09:51:58 <Diablo-D3> thats the ONLY difference.
1083 2011-04-19 09:52:02 <rly> Diablo-D3: I am still figuring out how to get the damn 11.2 drivers to work.
1084 2011-04-19 09:52:14 <Diablo-D3> rly: why 11.2? isnt 11.4 out now?
1085 2011-04-19 09:52:24 <rly> Diablo-D3: I am trying to install it on Fedora 14.
1086 2011-04-19 09:52:30 <Diablo-D3> heh.
1087 2011-04-19 09:52:40 <rly> Diablo-D3: those people lag like hell.,
1088 2011-04-19 09:52:57 <rly> Diablo-D3: what do you use?
1089 2011-04-19 09:53:11 <Diablo-D3> debian
1090 2011-04-19 09:53:20 <rly> Diablo-D3: do they package the latest and greatest?
1091 2011-04-19 09:53:43 <Diablo-D3> in experimental, its about half a month behind
1092 2011-04-19 09:53:50 <Diablo-D3> it has 11.3
1093 2011-04-19 09:54:06 <Diablo-D3> rly: btw, you do realize 2.4 is slower than 2.1, right?
1094 2011-04-19 09:54:10 <rly> I don't see why they cannot automate that stuff.
1095 2011-04-19 09:54:14 <rly> Diablo-D3: I had been using 2.3
1096 2011-04-19 09:54:19 <Diablo-D3> use 2.1
1097 2011-04-19 09:54:31 <rly> Diablo-D3: I got 160Mh/s with 2.3, which was more than most people.
1098 2011-04-19 09:54:42 <Diablo-D3> which sounds impossible.
1099 2011-04-19 09:54:52 <rly> Except... it is not.
1100 2011-04-19 09:55:06 <Diablo-D3> 2.1 should always be faster
1101 2011-04-19 09:55:19 <Diablo-D3> it sounds like you wernt on catalyst 10.9 through 10.11
1102 2011-04-19 09:55:29 <rly> I was on 10.12
1103 2011-04-19 09:55:35 <Diablo-D3> 10.12 is... kind of shitty
1104 2011-04-19 09:55:55 <rly> So, with the most shitty configuration I got 160Mh/s ;)
1105 2011-04-19 09:56:16 <rly> Rebooting, let's hope they didn't mess things up this time.
1106 2011-04-19 09:56:31 rly has quit (Quit: WeeChat 0.3.3)
1107 2011-04-19 10:04:51 robblesz has joined
1108 2011-04-19 10:05:34 <Diablo-D3> and we never heard from rly again
1109 2011-04-19 10:06:53 <retinal> â(-｡ｰ;)â
1110 2011-04-19 10:07:22 <retinal> PG&E stinx
1111 2011-04-19 10:07:46 <retinal> $0.40352/kWh for 200%+ baseline usage
1112 2011-04-19 10:07:58 <Diablo-D3> in maine its still higher.
1113 2011-04-19 10:08:43 <retinal> well, the thing is, there's a much better utilities provider in the neighboring county
1114 2011-04-19 10:09:12 <retinal> I would be in said county if commuting didn't suck so much
1115 2011-04-19 10:10:59 rly has joined
1116 2011-04-19 10:11:02 mtrlt has joined
1117 2011-04-19 10:11:17 <Diablo-D3> you know how commuting works?
1118 2011-04-19 10:11:23 * Diablo-D3 starts his commute
1119 2011-04-19 10:11:28 * Diablo-D3 opens laptop
1120 2011-04-19 10:11:31 * Diablo-D3 has arrived at work
1121 2011-04-19 10:11:35 <Diablo-D3> wow that was easy!
1122 2011-04-19 10:12:11 <Diablo-D3> rly: it didnt work?
1123 2011-04-19 10:13:35 <retinal> Diablo-D3: no car
1124 2011-04-19 10:13:40 <retinal> â(ï¿£ã¼ï¿£)â
1125 2011-04-19 10:14:03 <retinal> mainly because insurance prices are comical for someone in my age group
1126 2011-04-19 10:14:36 <Diablo-D3> I dont have a car either.
1127 2011-04-19 10:14:38 <Diablo-D3> in maine.
1128 2011-04-19 10:14:44 zyb has quit (Remote host closed the connection)
1129 2011-04-19 10:15:11 <retinal> public transportation, then?
1130 2011-04-19 10:15:38 <Diablo-D3> in maine? it doesnt exist
1131 2011-04-19 10:15:50 <rly> Diablo-D3: 10.12 is still 'active' according to the control center.
1132 2011-04-19 10:15:56 <rly> Diablo-D3: it is crapware.
1133 2011-04-19 10:15:57 <Diablo-D3> pwned.
1134 2011-04-19 10:16:00 <Diablo-D3> try -z 0 now.
1135 2011-04-19 10:16:15 <rly> Diablo-D3: no, I first want to use 11.2.
1136 2011-04-19 10:16:21 <Diablo-D3> dude
1137 2011-04-19 10:16:27 <Diablo-D3> quit moving the goalposts
1138 2011-04-19 10:16:47 <Diablo-D3> see if -z 0 makes it as fast poclbm
1139 2011-04-19 10:16:49 <rly> Diablo-D3: I have 2.4 installed now.
1140 2011-04-19 10:16:57 <rly> Diablo-D3: I cannot test it with a broken configuration.
1141 2011-04-19 10:16:57 <Diablo-D3> dude, its a zip file
1142 2011-04-19 10:17:01 <Diablo-D3> its not like you install it
1143 2011-04-19 10:18:58 <mizerydearia> When renting an apartment, is it common practice for applicants to have their credit checked before being accepted? If so, without credit cards, us dollar and instead making use of bitcoin to handle payment for renting apartment, etc, what type of alternative to credit checks seems likely?
1144 2011-04-19 10:18:58 rly has quit (Quit: WeeChat 0.3.3)
1145 2011-04-19 10:19:09 <netxshare> $*@*#
1146 2011-04-19 10:19:26 <Diablo-D3> mizerydearia: yes.
1147 2011-04-19 10:19:41 <Diablo-D3> and you got confused somewhere along the wway
1148 2011-04-19 10:19:46 <Diablo-D3> they check credit and look for negatives
1149 2011-04-19 10:20:03 <Diablo-D3> if you have a blank report, they accept you
1150 2011-04-19 10:20:30 <Diablo-D3> they just wanna make sure you havent fucked a previous landlord
1151 2011-04-19 10:21:19 <iera> we just have to establish an alternative rating system, with irc like in #bitcoin-otc lol
1152 2011-04-19 10:21:33 <mizerydearia> hmm, well, for example, my Mom has difficulty finding place to live due to bad credit.
1153 2011-04-19 10:21:39 rly has joined
1154 2011-04-19 10:21:45 <mizerydearia> but she has money
1155 2011-04-19 10:22:03 <rly> That still didn't kill 10.12.
1156 2011-04-19 10:22:21 <rly> A package management system is supposed to make things easier.
1157 2011-04-19 10:23:25 <joepie91> guise
1158 2011-04-19 10:23:26 <joepie91> http://encyclopediadramatica.ch/File:ED_First_Day_Crash_Stats.png
1159 2011-04-19 10:23:28 <joepie91> seriously
1160 2011-04-19 10:23:37 <joepie91> that is pretty much the most insane alexa drop that I've ever seen
1161 2011-04-19 10:24:18 <AAA_awright> joepie91: Why?
1162 2011-04-19 10:24:34 <joepie91> why what? why the drop? :P
1163 2011-04-19 10:25:02 <AAA_awright> Why's it say "And nothing of value was lost"
1164 2011-04-19 10:25:22 <joepie91> oh
1165 2011-04-19 10:25:22 <joepie91> well
1166 2011-04-19 10:25:30 <joepie91> owner of encyclopedia dramatica
1167 2011-04-19 10:25:33 <joepie91> decided to pull the plug
1168 2011-04-19 10:25:34 <joepie91> out of nowhere
1169 2011-04-19 10:25:46 <AAA_awright> But also, what's this have to do with #bitcoin-dev
1170 2011-04-19 10:25:48 <joepie91> and redirect the domain to oh internet, which is basically a crappy SWF knowyourmeme-cloe
1171 2011-04-19 10:25:50 <joepie91> clone*
1172 2011-04-19 10:25:52 <joepie91> not much
1173 2011-04-19 10:25:55 <joepie91> but it was just insane to see
1174 2011-04-19 10:27:03 <da2ce7> http://www.youtube.com/watch?v=FavUpD_IjVY&feature=player_embedded#at=58
1175 2011-04-19 10:27:05 <da2ce7> zomg
1176 2011-04-19 10:27:23 <da2ce7> http://www.youtube.com/watch?v=FavUpD_IjVY
1177 2011-04-19 10:27:42 <AAA_awright> Also, everyone short Bank of America
1178 2011-04-19 10:28:37 <AAA_awright> Hmm maybe it's not going anywhere
1179 2011-04-19 10:33:32 <ezl> ;;bc,mtgox
1180 2011-04-19 10:33:33 <gribble> {"ticker":{"high":1.18,"low":1.12,"vol":17253,"buy":1.151,"sell":1.1775,"last":1.1793}}
1181 2011-04-19 10:43:44 <Diablo-D3> [06:18:05] <mizerydearia> hmm, well, for example, my Mom has difficulty finding place to live due to bad credit.
1182 2011-04-19 10:43:50 <Diablo-D3> mizerydearia: then maybe she shouldnt screw people
1183 2011-04-19 10:44:53 <mizerydearia> I agree
1184 2011-04-19 10:46:25 <ne0futur> or move to another country :p
1185 2011-04-19 10:46:43 <Diablo-D3> ne0futur: doesnt always work
1186 2011-04-19 10:47:11 * mizerydearia considers moving to Japan...I've always wanted to live there.
1187 2011-04-19 10:47:21 <Diablo-D3> unless you speak japanese, whats the point
1188 2011-04-19 10:47:28 <Diablo-D3> and they dont look kindly on gaijin anyhow
1189 2011-04-19 10:47:28 <mizerydearia> Maybe Fukushima or near Sendai prefecture might be nice place to live
1190 2011-04-19 10:47:37 <Diablo-D3> Heh.
1191 2011-04-19 10:47:41 * Diablo-D3 smacks mizerydearia
1192 2011-04-19 10:48:25 <rly> Good housing prices there.
1193 2011-04-19 10:48:34 <mizerydearia> Diablo-D3, I studied Japanese language four years in high school.
1194 2011-04-19 10:48:58 <da2ce7> ;;bc.mtgox
1195 2011-04-19 10:48:58 <gribble> Error: "bc.mtgox" is not a valid command.
1196 2011-04-19 10:48:58 <rly> No pesky neighbors, good wireless connectivity over long ranges in the middle of the town.
1197 2011-04-19 10:49:05 <da2ce7> ;;bc,mtgox
1198 2011-04-19 10:49:05 <mizerydearia> ;;bc,mtgox
1199 2011-04-19 10:49:05 <gribble> {"ticker":{"high":1.18,"low":1.12,"vol":17487,"buy":1.1521,"sell":1.1785,"last":1.178}}
1200 2011-04-19 10:49:06 <gribble> {"ticker":{"high":1.18,"low":1.12,"vol":17487,"buy":1.1521,"sell":1.1785,"last":1.178}}
1201 2011-04-19 10:49:08 <mizerydearia> o_X
1202 2011-04-19 10:49:27 <ArtForz> thats actually not *that* crazy
1203 2011-04-19 10:52:36 <mrb_> says the guy who now makes $13k/mo mining :)
1204 2011-04-19 10:53:22 <ArtForz> friend of mine basically lives in the eu equivalent of a superfund site
1205 2011-04-19 10:53:31 Cusipzzz has joined
1206 2011-04-19 10:54:29 Lartza has joined
1207 2011-04-19 10:54:47 <Diablo-D3> ArtForz: heh, surprised hes not dead
1208 2011-04-19 10:55:00 <ArtForz> mainly heavy metal and arsenic contamination
1209 2011-04-19 10:56:23 <ArtForz> mainly ground and surface water, though some of the buildings also have dangerous levels in dust
1210 2011-04-19 10:56:33 <Diablo-D3> so why hasnt the site been closed/
1211 2011-04-19 10:57:23 <ArtForz> well, it mostly is
1212 2011-04-19 10:58:03 <rly> What is a superfund site?
1213 2011-04-19 10:58:20 <Diablo-D3> rly: areas that made people rich.
1214 2011-04-19 10:58:31 <Diablo-D3> lies, corruption, and greed in physical form.
1215 2011-04-19 10:58:46 <Diablo-D3> extremely toxic to human life.
1216 2011-04-19 10:59:27 <rly> Diablo-D3: like Sillicon Valley too? Or more like the Niger delta?
1217 2011-04-19 10:59:36 <ArtForz> well, don't lick the basement walls.
1218 2011-04-19 11:00:33 <Diablo-D3> https://secure.wikimedia.org/wikipedia/en/wiki/Superfund
1219 2011-04-19 11:05:52 <rly> ArtForz: why does he live there?
1220 2011-04-19 11:06:06 <rly> Is that his secret hide out? ;)
1221 2011-04-19 11:06:12 Zarutian has joined
1222 2011-04-19 11:07:49 <rly> That seems like a lot of places o.O: https://secure.wikimedia.org/wikipedia/en/wiki/File:Superfund_sites.svg
1223 2011-04-19 11:09:47 <Diablo-D3> http://www.independent.co.uk/life-style/food-and-drink/utter-PR-fiction-but-people-love-this-shit-so-fuck-it-lets-just-print-it-2269573.html
1224 2011-04-19 11:11:42 rly has quit (Quit: WeeChat 0.3.3)
1225 2011-04-19 11:12:35 <topi`> is there any political channel for bitcoin?
1226 2011-04-19 11:13:00 <Blitzboom> #bitcoin-politics
1227 2011-04-19 11:13:36 <topi`> right
1228 2011-04-19 11:14:08 Cusipzzz has quit (Quit: KVIrc 4.0.2 Insomnia http://www.kvirc.net/)
1229 2011-04-19 11:14:57 <Diablo-D3> ahaha
1230 2011-04-19 11:15:15 <Diablo-D3> all of xel's transactions
1231 2011-04-19 11:15:16 <Diablo-D3> JUST
1232 2011-04-19 11:15:19 <Diablo-D3> got into the chain
1233 2011-04-19 11:15:41 <Diablo-D3> 35 hours give or take
1234 2011-04-19 11:15:57 <Diablo-D3> or 196 blocks
1235 2011-04-19 11:19:51 mmoya has quit (Ping timeout: 260 seconds)
1236 2011-04-19 11:25:29 amiller has quit (Ping timeout: 246 seconds)
1237 2011-04-19 11:32:17 <topi`> previously, we were talking about the feasibility or infeasibility of doing point-of-sale (POS) operations with bitcoin
1238 2011-04-19 11:32:25 <topi`> here's what I wrote about it yesterday:
1239 2011-04-19 11:32:26 <topi`> http://lorelei.kaverit.org/point-of-sale-bitcoin.txt
1240 2011-04-19 11:33:03 <topi`> I guess there are some omissions and screwups I made, but then it needs discussion and tuning.
1241 2011-04-19 11:34:06 FabianB_ has joined
1242 2011-04-19 11:35:08 FabianB has quit (Ping timeout: 260 seconds)
1243 2011-04-19 11:35:16 <sipa> topi`: all that is needed is a mobile device (smartcard, cell phone, ...) which holds a (small) wallet, and is able to create transactions to a given output
1244 2011-04-19 11:35:41 <ArtForz> yep
1245 2011-04-19 11:35:44 dbitcoin has joined
1246 2011-04-19 11:35:55 <sipa> POS sends address and value to device, device creates transactions, sends it back to POS
1247 2011-04-19 11:35:56 <ArtForz> the wallet can actually be *really* small
1248 2011-04-19 11:36:05 <ArtForz> a single keypair works
1249 2011-04-19 11:36:09 <sipa> POS verifies that the transaction is valid and no double spend, and done
1250 2011-04-19 11:36:11 <sipa> ArtForz: indeed
1251 2011-04-19 11:37:05 Speeder has joined
1252 2011-04-19 11:37:27 <ArtForz> yep
1253 2011-04-19 11:38:17 <ArtForz> well, in theory you can still pull a double-spend on that if you have "decent" gen capacity, but it's pretty hard
1254 2011-04-19 11:38:38 amiller has joined
1255 2011-04-19 11:38:48 <sipa> client device specifically doesn't need an internet connection or full p2p node, and it is decidedly safer for the seller if it doesn't
1256 2011-04-19 11:38:54 <ArtForz> yep
1257 2011-04-19 11:39:15 <sipa> as long as it is "recharged" using a trusted system
1258 2011-04-19 11:39:27 <ArtForz> a smartcard would work, but then you have to trust the terminal to accurately show inputs/outputs
1259 2011-04-19 11:39:51 <ArtForz> so ideally you'd want at least something with a display
1260 2011-04-19 11:40:45 BCBot has joined
1261 2011-04-19 11:41:13 RazielZ has joined
1262 2011-04-19 11:47:06 eternal1 has joined
1263 2011-04-19 11:50:15 bonsaikitten has quit (Ping timeout: 252 seconds)
1264 2011-04-19 11:51:18 taco_the_paco has quit (Quit: Leaving)
1265 2011-04-19 11:51:59 taco_the_paco has joined
1266 2011-04-19 11:51:59 taco_the_paco has quit (Changing host)
1267 2011-04-19 11:51:59 taco_the_paco has joined
1268 2011-04-19 11:53:11 agricocb has quit (Quit: Leaving.)
1269 2011-04-19 11:53:26 BlueMatt has joined
1270 2011-04-19 11:53:26 BlueMatt has quit (Changing host)
1271 2011-04-19 11:53:26 BlueMatt has joined
1272 2011-04-19 11:58:15 <topi`> sipa: but how can the seller know if the client has full p2p node or not?
1273 2011-04-19 11:58:23 <sipa> not
1274 2011-04-19 11:58:32 <sipa> unfortunate, so he shouldn't assume he hasn't
1275 2011-04-19 11:58:56 <topi`> ArtForz: I agree, you want to have your own display if you want to be certain that there is no trickery
1276 2011-04-19 11:59:41 <topi`> sipa: also I have a full p2p bitcoin node, and still trade with ppl on #bitcoin-otc
1277 2011-04-19 11:59:50 <topi`> and they trust I'm not double-spending!
1278 2011-04-19 12:03:42 taco_the_paco has quit (Quit: Leaving)
1279 2011-04-19 12:04:04 mologie has joined
1280 2011-04-19 12:04:04 lfm has joined
1281 2011-04-19 12:04:04 ivan has joined
1282 2011-04-19 12:04:04 jercos has joined
1283 2011-04-19 12:04:04 quellhorst has joined
1284 2011-04-19 12:05:59 gasteve has quit (Quit: gasteve)
1285 2011-04-19 12:06:04 taco_the_paco has joined
1286 2011-04-19 12:07:48 gasteve has joined
1287 2011-04-19 12:07:52 gasteve has quit (Changing host)
1288 2011-04-19 12:07:52 gasteve has joined
1289 2011-04-19 12:09:13 tabsa has joined
1290 2011-04-19 12:14:28 agricocb has joined
1291 2011-04-19 12:14:33 rly has joined
1292 2011-04-19 12:15:11 <rly> Diablo-D3: I still get the same error when I run yours, but the Python one works when I select the AMD parallel platform.
1293 2011-04-19 12:15:21 <Kiba> dun dun dunnnnnn
1294 2011-04-19 12:16:53 bonsaikitten has joined
1295 2011-04-19 12:20:36 DrEeevil has joined
1296 2011-04-19 12:21:50 ezl has quit (Ping timeout: 246 seconds)
1297 2011-04-19 12:22:12 bonsaikitten has quit (Ping timeout: 276 seconds)
1298 2011-04-19 12:23:15 eao has joined
1299 2011-04-19 12:23:31 brunner has joined
1300 2011-04-19 12:23:52 DrEeevil is now known as bonsaikitten
1301 2011-04-19 12:25:59 trifon_ has joined
1302 2011-04-19 12:31:44 Incitatus has joined
1303 2011-04-19 12:32:48 <Incitatus> hey what is the IRC to see the block feeds?
1304 2011-04-19 12:34:02 <sipa> see #bitcoin-watch
1305 2011-04-19 12:34:18 <Incitatus> thank you
1306 2011-04-19 12:37:17 zyb has joined
1307 2011-04-19 12:42:35 sgornick has quit (Quit: Leaving.)
1308 2011-04-19 12:43:13 <netxshare> how much bandwidth does bitcoin use?
1309 2011-04-19 12:43:43 sabalaba has joined
1310 2011-04-19 12:44:42 <trifon_> netxsahre, once you load all block not much.
1311 2011-04-19 12:44:44 BlueMatt has quit (Read error: Connection reset by peer)
1312 2011-04-19 12:44:56 <trifon_> it uses lot of GPU :)
1313 2011-04-19 12:45:07 <netxshare> that's what I figured
1314 2011-04-19 12:46:08 sgornick has joined
1315 2011-04-19 12:46:24 <topi`> netxshare: I tried to measure it, and it averaged out at 5 kilobytes per second
1316 2011-04-19 12:46:51 <topi`> although, I think that also included my IRC traffic etc ,)
1317 2011-04-19 12:47:46 <netxshare> thanks
1318 2011-04-19 12:49:09 <topi`> but it definitely is less than that :)
1319 2011-04-19 12:49:47 <BurtyB> I see about 30-40MB/day
1320 2011-04-19 12:50:51 <BurtyB> (that's just for the bitcoin process on windows with ~20 connections)
1321 2011-04-19 12:51:37 <tcoppi> ;;bc,stats
1322 2011-04-19 12:51:39 <gribble> Current Blocks: 119116 | Current Difficulty: 92347.59095209 | Next Difficulty At Block: 120959 | Next Difficulty In: 1843 blocks | Next Difficulty In About: 1 week, 5 days, 21 hours, 12 minutes, and 52 seconds | Next Difficulty Estimate: 92222.66560735
1323 2011-04-19 12:53:17 BlueMatt has joined
1324 2011-04-19 12:53:17 BlueMatt has quit (Changing host)
1325 2011-04-19 12:53:17 BlueMatt has joined
1326 2011-04-19 12:55:00 zyb has quit (Remote host closed the connection)
1327 2011-04-19 13:00:28 trifon__ has joined
1328 2011-04-19 13:02:14 subpar has quit (Quit: Leaving)
1329 2011-04-19 13:02:58 trifon__ is now known as trifon
1330 2011-04-19 13:03:11 <genjix> http://pastebin.com/GEHfQ0wc
1331 2011-04-19 13:03:22 trifon_ has quit (Ping timeout: 252 seconds)
1332 2011-04-19 13:03:30 <BlueMatt> genjix: wrong version of miniupnp
1333 2011-04-19 13:03:35 <genjix> wtf, how come it's including that header when UPNP=0?
1334 2011-04-19 13:03:49 * sipa agrees the UPNP tristate is confusing
1335 2011-04-19 13:03:54 <BlueMatt> USE_UPNP=0 means build upnp with it off by default, USE_UPNP= means dont build
1336 2011-04-19 13:04:08 <genjix> look at the compile line though
1337 2011-04-19 13:04:12 * BlueMatt points his finger at luke_jr who had the idea
1338 2011-04-19 13:04:18 <genjix> it has USE_UPNP=0
1339 2011-04-19 13:04:26 <genjix> but still includes that header?
1340 2011-04-19 13:04:32 <BlueMatt> genjix: ^
1341 2011-04-19 13:04:52 <genjix> ahh so UPNP is defined but it equals 0... ok thanks :)
1342 2011-04-19 13:05:07 <BlueMatt> but seriously, if anyone has a better idea please submit patches
1343 2011-04-19 13:05:31 <genjix> no no it's ok. i have no beef with your patch, but im just trying to get my patch to work first
1344 2011-04-19 13:05:33 <BlueMatt> although it might be better to use USE_UPNP=-1/0/1 or something
1345 2011-04-19 13:05:46 <genjix> i just set UPNP:= in the makefile
1346 2011-04-19 13:06:09 <BlueMatt> genjix: I know, its just quite a few people have been complaining about similar issues so...
1347 2011-04-19 13:06:36 <genjix> what does upnp? it's a discovery library for peer2peer, right?
1348 2011-04-19 13:06:46 <BlueMatt> it automatically maps the ports on your router
1349 2011-04-19 13:06:52 <BlueMatt> provided upnp is enabled on your router
1350 2011-04-19 13:06:55 ezl has joined
1351 2011-04-19 13:06:59 <genjix> nice, to route around ports being blocked
1352 2011-04-19 13:07:05 <BlueMatt> yea
1353 2011-04-19 13:07:24 <genjix> so with encrypted traffic + upnp, bitcoin can never be blocked.
1354 2011-04-19 13:07:32 <genjix> unless you shutdown the internet
1355 2011-04-19 13:07:32 <BlueMatt> not the idea
1356 2011-04-19 13:07:46 <BlueMatt> the idea of upnp is that people dont want to bother forwarding the ports on their router
1357 2011-04-19 13:07:49 <BlueMatt> so upnp does it for them
1358 2011-04-19 13:07:52 <genjix> oh ok
1359 2011-04-19 13:08:11 <genjix> but doesnt it also help with isps that block ports?
1360 2011-04-19 13:08:16 <BlueMatt> no
1361 2011-04-19 13:08:20 <genjix> k
1362 2011-04-19 13:08:32 <BlueMatt> its needed now that we have a bunch of non-techies joining and as we move into more mainstrea
1363 2011-04-19 13:08:34 <BlueMatt> m
1364 2011-04-19 13:08:49 <genjix> right
1365 2011-04-19 13:08:49 <BlueMatt> because those people never touch their router config
1366 2011-04-19 13:08:54 <Blitzboom> we need a mainstream client â¦
1367 2011-04-19 13:09:01 <Blitzboom> something like the utorrent of bittorrent
1368 2011-04-19 13:09:24 <BlueMatt> or we just need to make the current client a bit better then it can be the mainstream client ;)
1369 2011-04-19 13:09:40 <Blitzboom> BlueMatt: centralization sucks
1370 2011-04-19 13:09:53 slush has quit (Quit: Leaving.)
1371 2011-04-19 13:10:05 <Blitzboom> iâd rather have multiple widely used clients
1372 2011-04-19 13:10:06 <BlueMatt> never said we dont want more clients, but that doesnt mean all of them shouldnt be good
1373 2011-04-19 13:10:13 <Blitzboom> hmm, true
1374 2011-04-19 13:10:17 * sipa wants a libbitcoin
1375 2011-04-19 13:10:23 subpar has joined
1376 2011-04-19 13:10:34 <BlueMatt> BitcoinJ?
1377 2011-04-19 13:10:47 <sipa> bleh
1378 2011-04-19 13:10:59 <sipa> java is nice, but it's not the only language i want to use
1379 2011-04-19 13:11:17 <sipa> and if you want a de-facto library used by all "full clients", java is a bad choice i think
1380 2011-04-19 13:11:19 <BlueMatt> true, so write libbitcoin
1381 2011-04-19 13:11:34 <BlueMatt> and dont let Diablo hear you
1382 2011-04-19 13:12:55 * sipa conjures up a shitload of time
1383 2011-04-19 13:13:30 <BlueMatt> yea I dont think anyone has enough time to write it. Plus its better to use your time to generate more hype around bitcoin than writing code for bitcoin
1384 2011-04-19 13:14:50 <genjix> sipa: what about a bitcoin in python?
1385 2011-04-19 13:15:30 <sipa> genjix: i think the only viable option for a libbitcoin to become a de-facto standard library for integrating bitcoin in some other project, is C
1386 2011-04-19 13:17:30 <genjix> sipa: i see
1387 2011-04-19 13:17:43 <genjix> sipa: not c++, right?
1388 2011-04-19 13:17:48 <genjix> and how about using MySQL?
1389 2011-04-19 13:17:59 <sipa> that's a backend question
1390 2011-04-19 13:18:27 <sipa> and C++ code with a C wrapper around it for easy linking, isn't hard
1391 2011-04-19 13:18:33 <genjix> right.
1392 2011-04-19 13:18:49 <sipa> so that's possible as well, and maybe the best option if someone were to do it, to be able to reuse the bitcoin codebase
1393 2011-04-19 13:19:08 <genjix> how about to write it in C++ but provide a C api?
1394 2011-04-19 13:19:24 <sipa> yes, that's what i suggested
1395 2011-04-19 13:19:26 <genjix> because memory management in C is pretty hard/sucky
1396 2011-04-19 13:19:28 <genjix> ok
1397 2011-04-19 13:21:06 <sipa> but ultimately you want something like: btcnode_t node; btcnode_init(&node); btcwallet_t wallet; btcwallet_register(&node,&wallet); ...
1398 2011-04-19 13:22:19 <genjix> i see
1399 2011-04-19 13:31:27 <genjix> the current github release branch doesn't work.
1400 2011-04-19 13:31:34 <genjix> it's broken.
1401 2011-04-19 13:31:55 <ByteCoin> Is someone making a bitcoin lib?
1402 2011-04-19 13:32:25 <sipa> not afaik
1403 2011-04-19 13:32:48 <Stellar> fatal error: miniupnpc/miniwget.h: No such file or directory
1404 2011-04-19 13:32:53 <jaromil> ByteCoin: i'm maintaining my branch using autotools and libtool ultimately to achieve
1405 2011-04-19 13:32:54 <Stellar> is that a lib?
1406 2011-04-19 13:33:01 <iera> wasnt there some python bitcoin lib already?
1407 2011-04-19 13:33:14 <genjix> the current release repo just hangs
1408 2011-04-19 13:33:16 <sipa> Stellar: set USE_PNP= in your makefile
1409 2011-04-19 13:33:24 <genjix> im using that and it compiled
1410 2011-04-19 13:33:31 <genjix> but the daemon fails to fork
1411 2011-04-19 13:33:51 <Stellar> ok
1412 2011-04-19 13:33:57 <ByteCoin> I'm frustrated that the wallet handling rubbish is so mixed with the p2p blockchain and transaction handling stuff
1413 2011-04-19 13:34:06 <genjix> right.
1414 2011-04-19 13:34:19 <genjix> that's why a rewrite might be needed
1415 2011-04-19 13:34:37 <genjix> i might be getting some resources to setup a bitcoin foundation
1416 2011-04-19 13:34:39 <ByteCoin> One should be able to write secure wallet code
1417 2011-04-19 13:34:40 <jaromil> ByteCoin: i'm also frustrated
1418 2011-04-19 13:34:49 <genjix> that's my primary goal maybe
1419 2011-04-19 13:35:22 <ByteCoin> Has anyone looked at the bitcoinj stuff?
1420 2011-04-19 13:35:23 * sipa wished he had more time
1421 2011-04-19 13:35:59 <ByteCoin> Does bitcoinj separate the wallet from the network process?
1422 2011-04-19 13:36:11 <ByteCoin> I know it's not a full client
1423 2011-04-19 13:36:48 <sipa> i think it does
1424 2011-04-19 13:36:53 <sipa> its design is a lot cleaner
1425 2011-04-19 13:37:17 <ByteCoin> Well it's nice to know that there's support for a modular design
1426 2011-04-19 13:37:43 <ByteCoin> I have a bug in ljrbot to report... whom to contact?
1427 2011-04-19 13:38:44 <sipa> luke-jr, i suppose
1428 2011-04-19 13:38:51 <genjix> sipa: you reckon it's possible to refactor bitcoin to perfection? it *is* a tiny codebase
1429 2011-04-19 13:38:52 <ByteCoin> Cheers!
1430 2011-04-19 13:39:58 <sipa> i don't believe in perfection
1431 2011-04-19 13:40:02 <sipa> :)
1432 2011-04-19 13:40:11 <genjix> sipa: so your daemon-mode branch fixed the fork problem months ago
1433 2011-04-19 13:40:24 <genjix> was it ever merged because now i see the release branch isn't forking.
1434 2011-04-19 13:40:27 <sipa> i never implemented a daemon mode?
1435 2011-04-19 13:40:36 <genjix> oh sorry i mean tcatm
1436 2011-04-19 13:41:29 <Kiba> there is sure a lot of podcast with Gavin Andersen in it
1437 2011-04-19 13:43:07 <jaromil> genjix: do a git checkout HEAD~1 or 2 or 3
1438 2011-04-19 13:43:12 <jaromil> to go in the past
1439 2011-04-19 13:43:15 <jaromil> and see how it works
1440 2011-04-19 13:44:23 <jaromil> mmm, nvm. i can replicate the problem
1441 2011-04-19 13:45:19 <Kiba> bitcoin critics are annoying
1442 2011-04-19 13:45:25 <Kiba> they are uninformed and don't know anything
1443 2011-04-19 13:48:46 <UukGoblin> hrm
1444 2011-04-19 13:48:58 <Kiba> the major problem with the bitcoin critics is that they think bitcoin sucks so they don't join the community and learn
1445 2011-04-19 13:49:24 <UukGoblin> is there a secure and anonymous electronic voting system that'd allow me to vote from my PC, whilst ensuring every citizen gets to vote once and once only?
1446 2011-04-19 13:49:27 <Kiba> the community generated technical, economic, and business knowledge
1447 2011-04-19 13:49:35 <Kiba> but these knowledge are not assimilated to the wider public
1448 2011-04-19 13:49:47 <sipa> UukGoblin: not without having a unique identifier for each citizen
1449 2011-04-19 13:50:14 <UukGoblin> sipa, yeah, it's hard
1450 2011-04-19 13:50:23 <UukGoblin> you'd have to get the identifier physically or something
1451 2011-04-19 13:50:26 <sipa> and not without risking "friendly people" going from door to door to "help" eg. elder people with their voting
1452 2011-04-19 13:50:59 <UukGoblin> sipa, well that wouldn't be mandatory
1453 2011-04-19 13:51:19 <UukGoblin> I'm thinking from a crypto-anarchic perspective
1454 2011-04-19 13:51:31 <UukGoblin> or just cryptographic generally
1455 2011-04-19 13:51:36 <Kiba> why would I want democracy 2.0?
1456 2011-04-19 13:51:40 ezl has quit (Ping timeout: 240 seconds)
1457 2011-04-19 13:51:41 <sipa> cryptographically it's possible
1458 2011-04-19 13:51:47 <UukGoblin> it seems it's one of the problems that cryptography can't solve ;-]
1459 2011-04-19 13:51:55 <sipa> it is solved, afaik
1460 2011-04-19 13:52:01 <UukGoblin> oh?
1461 2011-04-19 13:52:11 <UukGoblin> do you happen to know a google keyword? :-]
1462 2011-04-19 13:52:24 <sipa> you can let people vote, vote once only, lets everyone verify the result without revealing who voted what
1463 2011-04-19 13:52:54 <sipa> sorry, no, a friend of mine (with a phd in cryptography) told me about it
1464 2011-04-19 13:52:55 <UukGoblin> I mean, in a computer network
1465 2011-04-19 13:53:07 <UukGoblin> where you can't verify physical presence
1466 2011-04-19 13:53:20 <UukGoblin> well
1467 2011-04-19 13:53:30 <UukGoblin> I guess you verify it once
1468 2011-04-19 13:53:37 <edcba> choose a random number x let every ppl add a or b then substract x
1469 2011-04-19 13:53:39 <UukGoblin> and then just use it later
1470 2011-04-19 13:54:15 <sipa> edcba: that allows you to verify it, but not the rest :)
1471 2011-04-19 13:54:36 <edcba> which rest ?
1472 2011-04-19 13:54:59 <UukGoblin> anonymity of the vote?
1473 2011-04-19 13:55:00 <sipa> those who voted
1474 2011-04-19 13:55:11 <edcba> i thought the goal was everyone had to vote
1475 2011-04-19 13:55:20 <sipa> they obviously should be able to verify that the vote was fair
1476 2011-04-19 13:56:32 <edcba> if you know how much ppl did vote what is the problem ?
1477 2011-04-19 13:57:05 <sipa> they also need to verify their own vote was used
1478 2011-04-19 13:58:44 <lulzplzkthx> One thing I realy like about Bitcoin is it's giving many developers an oppurtunity to create a startup that might otherwise not be possible for them.
1479 2011-04-19 13:59:08 <lulzplzkthx> If not because it *requires* Bitcoin, at least because it provides an audience.
1480 2011-04-19 13:59:28 <edcba> anyway there are plenty of cyptographic voting system
1481 2011-04-19 13:59:38 <edcba> the problem usually with paid votes iirc
1482 2011-04-19 14:00:50 <edcba> ie you must not be able to show someone who you voted for :)
1483 2011-04-19 14:02:44 <sipa> ;;bc,blocks
1484 2011-04-19 14:02:45 <gribble> 119126
1485 2011-04-19 14:02:45 <edcba> so UukGoblin your requirements are in total contradiction with what you would like for a perfect voting system
1486 2011-04-19 14:03:28 glassresistor has joined
1487 2011-04-19 14:03:31 <edcba> ie someone forcing you to vote for him *from your pc* :)
1488 2011-04-19 14:04:13 <sipa> you should be able to prove "yes i've voted", but not be able to prove "yes, i voted for <X>"
1489 2011-04-19 14:04:52 <edcba> indeed
1490 2011-04-19 14:08:17 <edcba> anyway electronic voting system just weakens it
1491 2011-04-19 14:10:37 mvn071 has quit (Ping timeout: 250 seconds)
1492 2011-04-19 14:10:45 <UukGoblin> hmm
1493 2011-04-19 14:10:58 <UukGoblin> there is that, yes
1494 2011-04-19 14:11:43 <UukGoblin> in which case, there'd need to be a way to /change/ your vote later without the person knowing ;-]
1495 2011-04-19 14:12:10 <UukGoblin> or earlier
1496 2011-04-19 14:12:49 <UukGoblin> basically make it impossible for anyone but yourself to verify your vote...
1497 2011-04-19 14:13:56 mvn071 has joined
1498 2011-04-19 14:15:26 mologie2 has joined
1499 2011-04-19 14:17:54 lfm_ has joined
1500 2011-04-19 14:18:01 quellhor1t has joined
1501 2011-04-19 14:18:01 jercos_ has joined
1502 2011-04-19 14:19:30 mologie has quit (Read error: Connection reset by peer)
1503 2011-04-19 14:19:31 lfm has quit (Ping timeout: 240 seconds)
1504 2011-04-19 14:19:32 jercos has quit (Ping timeout: 240 seconds)
1505 2011-04-19 14:19:32 quellhorst has quit (Ping timeout: 240 seconds)
1506 2011-04-19 14:19:32 ivan has quit (Ping timeout: 240 seconds)
1507 2011-04-19 14:20:21 mvn071 has quit (Quit: Leaving)
1508 2011-04-19 14:20:29 BlueMatt has quit (Read error: Operation timed out)
1509 2011-04-19 14:26:20 ivan has joined
1510 2011-04-19 14:27:21 slush has joined
1511 2011-04-19 14:28:23 stamit has quit (Quit: stamit)
1512 2011-04-19 14:30:34 <BurtyB> genjix is there any reason btc deposits don't show up in "your trades" until they are Finished?
1513 2011-04-19 14:31:16 <netxshare> http://www.bitcoin.org/smf/index.php?topic=6081.0
1514 2011-04-19 14:31:51 <netxshare> he is getting the same hash rate with two 5970's as two 6990s
1515 2011-04-19 14:37:46 Stellar has quit ()
1516 2011-04-19 14:40:43 <genjix> BurtyB: you mean when unconfirmed? i will change that
1517 2011-04-19 14:41:13 <genjix> not now though. later.
1518 2011-04-19 14:41:31 BlueMatt has joined
1519 2011-04-19 14:41:31 BlueMatt has quit (Changing host)
1520 2011-04-19 14:41:31 BlueMatt has joined
1521 2011-04-19 14:42:24 Diablo-D3 has quit (Ping timeout: 260 seconds)
1522 2011-04-19 14:43:44 mologie2 has quit (Quit: Nettalk6 - www.ntalk.de)
1523 2011-04-19 14:44:20 KBme is now known as hilter
1524 2011-04-19 14:44:52 <BurtyB> genjix yeah, would be handy to see its on it's way
1525 2011-04-19 14:46:21 hilter is now known as KBme
1526 2011-04-19 14:49:16 <luke-jr> genjix: tcatm's patch only fixed -daemon on *nix, not Windows
1527 2011-04-19 14:55:32 <topi`> which card would be a better investment for mining, a Radeon 5850 or a 6950 ? basically, the 6950 is only slightly more expensive
1528 2011-04-19 14:55:45 <topi`> and newer (barts) than the 5850 (cypress)
1529 2011-04-19 14:58:57 <mtrlt> calculate Mhps per unit of money or something
1530 2011-04-19 14:59:04 <netxshare> I am wondering it's not just a better idea to get a used 5970 over a 6990
1531 2011-04-19 14:59:46 <mtrlt> a used 5970 would be a lot cheaper, i'd buy that.
1532 2011-04-19 15:01:45 <topi`> it's hard to find on the 2nd hand market
1533 2011-04-19 15:01:55 <mtrlt> yeah
1534 2011-04-19 15:01:59 <netxshare> ebay have a bunch right now
1535 2011-04-19 15:02:02 <netxshare> 5 or 6
1536 2011-04-19 15:02:58 <mtrlt> i found one on a finnish web-auction site but decided to see how much people are going to pay and bid during the last hour or something. but as usual, i forgot about that and it was sold for like 280 EUR :P
1537 2011-04-19 15:03:22 <mtrlt> i would've bid at least 300
1538 2011-04-19 15:04:56 <netxshare> I think I could get two 5970s for close the to price of 1 6990 about 150 more
1539 2011-04-19 15:05:02 f3n has joined
1540 2011-04-19 15:07:23 <Kiba> http://bitcoinweekly.com/articles/bitcoin-knowledge-generation-and-assimilation
1541 2011-04-19 15:07:25 <Kiba> yo
1542 2011-04-19 15:07:38 <Kiba> anybody want to review my article before I promote it widely?
1543 2011-04-19 15:08:49 zhalox has joined
1544 2011-04-19 15:10:02 <Blitzboom> wow, itâs fun to look back: http://www.bitcoin.org/smf/index.php?topic=57.0
1545 2011-04-19 15:11:22 <topi`> mtrlt: same happened to me when I found a 5870 on a web auction site, waited for it to close and then it closed WITHOUT my bid... grrr
1546 2011-04-19 15:11:37 TD_ has joined
1547 2011-04-19 15:12:06 ByteCoin has quit (Remote host closed the connection)
1548 2011-04-19 15:12:23 ByteCoin has joined
1549 2011-04-19 15:12:38 molecular has quit (Ping timeout: 246 seconds)
1550 2011-04-19 15:16:06 wolfspraul has quit (Quit: leaving)
1551 2011-04-19 15:18:49 <dbitcoin> ;;bc,stats
1552 2011-04-19 15:18:50 <gribble> Current Blocks: 119133 | Current Difficulty: 92347.59095209 | Next Difficulty At Block: 120959 | Next Difficulty In: 1826 blocks | Next Difficulty In About: 1 week, 5 days, 15 hours, 19 minutes, and 8 seconds | Next Difficulty Estimate: 93136.61274233
1553 2011-04-19 15:18:54 <rly> There is a big red square with a big white one in the left top corner of my screen with an ATI driver. Why is it there and how can I get rid of it?
1554 2011-04-19 15:21:17 isilion has quit (Read error: Connection reset by peer)
1555 2011-04-19 15:21:33 trifon has quit (Ping timeout: 264 seconds)
1556 2011-04-19 15:23:59 marlowe has joined
1557 2011-04-19 15:24:08 stamit has joined
1558 2011-04-19 15:24:45 rly has quit (Quit: WeeChat 0.3.3)
1559 2011-04-19 15:26:05 <genjix> luke-jr: yeah i dont use windows. i use linux but again it's not forking for me.
1560 2011-04-19 15:26:16 <luke-jr> genjix: git master?
1561 2011-04-19 15:26:21 <genjix> jgarzik: i re-submitted my pull request. check it out and let me know.
1562 2011-04-19 15:26:25 <jaromil> i confirm git master is not forking in bg
1563 2011-04-19 15:26:25 <genjix> luke-jr: right on github.
1564 2011-04-19 15:26:42 <tcatm> genjix: bitcoind -daemon
1565 2011-04-19 15:26:47 <genjix> ok off now for a while. if anybody PMs me, i'll check it when back.
1566 2011-04-19 15:27:54 <grbgout> yeah, why the F did they change bitcoind to require the -daemon flag to detach? What's the fing point of having bitcoinD, then, in the first place!?! >_<
1567 2011-04-19 15:28:07 <grbgout> s/fing/effing/
1568 2011-04-19 15:28:23 <grbgout> I mean, g'morning everyone.
1569 2011-04-19 15:28:50 <gasteve> in deciding how to organize autotools make files, I came across the paper "recursive make considered harmful" ...I haven't read it yet, but wonder if anyone has an opinion on it
1570 2011-04-19 15:28:59 isilion has joined
1571 2011-04-19 15:29:10 <tcatm> bitcoin-nogui would be a better name
1572 2011-04-19 15:29:11 <grbgout> gasteve: could you link us to the article so we may read it too?
1573 2011-04-19 15:29:24 <gasteve> sure: http://aegis.sourceforge.net/auug97.pdf
1574 2011-04-19 15:29:27 <grbgout> tcatm: no, switching it back to the proper behavior would be better >_<
1575 2011-04-19 15:30:06 <tcatm> use the daemon switch
1576 2011-04-19 15:30:30 <grbgout> I do use it, but it shouldn't be necessary. :\
1577 2011-04-19 15:30:54 <grbgout> Where's gjs278 when you need him? ;P
1578 2011-04-19 15:30:57 <tcatm> start a forum thread to change it so -nodaemon with daemonize by default
1579 2011-04-19 15:31:00 molecular has joined
1580 2011-04-19 15:31:26 <grbgout> tcatm: I don't see why that's necessary when there's an equivalent bitcoin executable already.
1581 2011-04-19 15:31:51 <grbgout> tcatm: doesn't bitcoin -server achieve the same thing?
1582 2011-04-19 15:31:55 <tcatm> bitcoin (without d) is the GUI version
1583 2011-04-19 15:32:07 Incitatus has quit (Ping timeout: 240 seconds)
1584 2011-04-19 15:32:08 <gasteve> (it was actually referenced in some of the autotools documentation I was reading...and autotools seems to accomodate either style just fine)
1585 2011-04-19 15:32:31 <sipa> bitcoin-cli and wxbitcoin are better names, maybe
1586 2011-04-19 15:32:47 <grbgout> tcatm: fair enough, I guess. Do you see any benefit to running bitcoind in "nodaemon" mode?
1587 2011-04-19 15:32:57 <jaromil> gasteve: i agree if by recursive make it intends nested autotool repos
1588 2011-04-19 15:33:08 skeledrew has quit (Quit: Instantbird 0.3a3pre)
1589 2011-04-19 15:33:09 <tcatm> sipa: ack. maybe even add a readline rpc interface when it doesn't fork
1590 2011-04-19 15:33:18 <sipa> indeed
1591 2011-04-19 15:33:22 <sipa> nice idea
1592 2011-04-19 15:33:25 <gasteve> why would you even want the process to detach at all? you can run any process in the background easy enough
1593 2011-04-19 15:35:04 <gasteve> jaromil: while I'm thinking of it...I noticed you put GPL on some of your stuff...the base bitcoin license is MIT (or MIT like)...GPL would pretty much rule out pulling any of your changes into the master brancch
1594 2011-04-19 15:36:09 <gasteve> jaromil: do you have a link where I could read about "nested autotools repos"?
1595 2011-04-19 15:36:16 octarine has joined
1596 2011-04-19 15:37:07 skeledrew has joined
1597 2011-04-19 15:39:01 <gasteve> tcatm: I would perfer the process not fork itself and just leave that up to how you invoke it...you would pass something on the command line like --rpc (to start the RPC server) or --repl (to start a stdio REPL like interface)
1598 2011-04-19 15:39:54 <tcatm> gasteve: http://www.bitcoin.org/smf/index.php?topic=6110.msg89680#msg89680
1599 2011-04-19 15:40:18 <tcatm> only -daemon would fork
1600 2011-04-19 15:40:35 <Kiba> anybody wanna help me with my crappy article's grammar and spelling error?
1601 2011-04-19 15:40:39 <Kiba> I'll pay 3 BTC for the time
1602 2011-04-19 15:41:00 <gasteve> yeah, that's fine (but at the same time, I don't see the utility of it...and less code is more maintainable code)
1603 2011-04-19 15:41:15 <jaromil> gasteve: re: GPL it was automake -ac and i removed it
1604 2011-04-19 15:41:20 <tcatm> utility of readline interface?
1605 2011-04-19 15:41:29 <jaromil> gasteve: re: nested autotools i only have some (bad) experiences
1606 2011-04-19 15:41:33 <grbgout> gasteve: you don't see the utility of daemon processes?
1607 2011-04-19 15:41:33 <gasteve> utility of -daemon
1608 2011-04-19 15:41:37 <grbgout> ...
1609 2011-04-19 15:41:43 <grbgout> Are you trollin'?
1610 2011-04-19 15:41:46 <gasteve> no ;)
1611 2011-04-19 15:42:05 <grbgout> Are you new to unix/linux?
1612 2011-04-19 15:42:20 <gasteve> I see the utility of running processes detached from any UI (CLI or otherwise)...obviously...I just don't see the utility of building that into the executable
1613 2011-04-19 15:42:40 fimp has joined
1614 2011-04-19 15:42:49 <gasteve> (yes, I'm relatively new to unix/linux ...only be using it for about 25 years)
1615 2011-04-19 15:43:00 <jaromil> you can set things in background without forking code. the utility of forking code in unix/bsd is usually that of have uid/git separation via setuidgid.
1616 2011-04-19 15:43:11 <jaromil> uid/gid
1617 2011-04-19 15:43:29 mologie has joined
1618 2011-04-19 15:43:49 <Kiba> job taken
1619 2011-04-19 15:44:41 <grbgout> gasteve: well, I'm not familiar with the technical details of a daemon processes over using the ampersand to background it, so I can't offer a sound argument --- but that doesn't mean there isn't one to be made. I'm sure gjs278 could illuminate the benefits for us, but he doesn't seem to be around. I've only been using unix/linux for 11 years.
1620 2011-04-19 15:44:45 <tcatm> -daemon would be for backwards compatibility. Everyone running bitcoin on a server will have a monitor script/daemon that handles forking.
1621 2011-04-19 15:45:01 sabalaba has quit (Ping timeout: 276 seconds)
1622 2011-04-19 15:46:14 <gasteve> jaromil suggested the setuidgid benefit ...but I'm wondering whether something like daemontools wouldn't handle that use case just fine (not sure)
1623 2011-04-19 15:46:35 <jaromil> gasteve: sure, djb coded setuidgid for the purpose, but is not standard in gnu/linux
1624 2011-04-19 15:46:48 mmoya has joined
1625 2011-04-19 15:46:48 <tcatm> it would. nohup bitcoin-cli & would work, too.
1626 2011-04-19 15:47:10 <devrandom_> forking helps detach from the shell cleanly
1627 2011-04-19 15:47:10 <gasteve> and, isn't it bad practice anyway to use setuidgid on such a largish program like bitcoin?
1628 2011-04-19 15:47:32 <tcatm> devrandom_: man nohup
1629 2011-04-19 15:47:34 <devrandom_> otherwise you have to use ampersand and disown, or a subshell
1630 2011-04-19 15:47:52 <jaromil> gasteve: there isn't much point in setting different uid/gid of bitcoind ATM
1631 2011-04-19 15:47:53 <devrandom_> nohup is another way to do a fork... ;)
1632 2011-04-19 15:48:19 <jaromil> daemons opening ports <1000 need that much more...
1633 2011-04-19 15:48:19 <Pander> with DISPLAY=:0.0; aticonfig --pplib-cmd "set fanspeed 0 70" the fan speed of gpu is set to 70%, how to set it back to automatic?
1634 2011-04-19 15:48:47 <tcatm> I would also vote to remove -daemon and fork() from bitcoin-cli completely.
1635 2011-04-19 15:48:54 <gasteve> so anyway, my only argument was that there are plenty of options for this stuff without having to build it into bitcoin itself...and given that, wouldn't it be better to have less code in bitcoin than more code
1636 2011-04-19 15:49:21 <grbgout> The biggest case I can make for the default behavior of bitcoind being to automatically fork is expectation. That's the behavior I would expect from an executable with a d suffix: I would expect it to be a daemon, and to behave as such.
1637 2011-04-19 15:50:25 <jaromil> historically (and AFAIK) daemon means "more" that is listening for commands, rather than is running in background
1638 2011-04-19 15:50:31 <gasteve> (if were talking names.. I would have perferred bitcoin be the non-GUI executable and bitcoin-gui or bitcoin-wx be the GUI version)
1639 2011-04-19 15:50:46 <grbgout> It did exactly what I thought it would the first time I executed it, and the need for a -daemon flag (single hyphen? argh!) was counter intuitive when I built from the github repo.
1640 2011-04-19 15:50:51 <devrandom_> grbgout - agreed... it's also a pretty small amount of code
1641 2011-04-19 15:50:55 <grbgout> jaromil: which bitcoind does: it listens for the RPC commands.
1642 2011-04-19 15:51:23 <jaromil> grbgout: yes. i've seen. i also used xmlrpcpp in the past and i find it a decent solution
1643 2011-04-19 15:51:39 <jaromil> actually that library got better
1644 2011-04-19 15:51:45 <grbgout> jaromil: I'm not familiar with xmlrpcpp, link?
1645 2011-04-19 15:52:05 <jaromil> the one on sforge
1646 2011-04-19 15:52:25 <jaromil> plz lookup yourself, i hate using the browser :>
1647 2011-04-19 15:52:42 * grbgout shrugs
1648 2011-04-19 15:52:49 <grbgout> I'm content with bitcoind's rpc commands.
1649 2011-04-19 15:53:16 <grbgout> It hasn't seemed broken for my usage yet, so I don't really have an incentive to go looking for alternatives (at the moment).
1650 2011-04-19 15:54:59 <tcatm> IIRC versions before 0.3 used to break things a few times. I see no reaseon why we shouldn't change things if we discover better ways to do them.
1651 2011-04-19 15:55:15 <jaromil> grbgout: i agree with you
1652 2011-04-19 15:55:41 <grbgout> tcatm: indeed. I was just speaking specifically about looking for jaromil's xmlrpcpp suggestion. What bitcoind offers me right now works for me, so I'm happy.
1653 2011-04-19 15:55:51 <jaromil> i wasn't suggesting xmlrpcpp
1654 2011-04-19 15:56:08 <jaromil> beg pardon for confusion
1655 2011-04-19 15:56:21 <jaromil> i mixed up different proj. and for a moment thought bitcoin was using it
1656 2011-04-19 15:56:25 <grbgout> jaromil: well, "mentioning" if you prefer....
1657 2011-04-19 15:56:57 <jaromil> FWIW my opinion, i think the most urgent fix for bitcoin is the database storage
1658 2011-04-19 15:57:14 <thermal> i was going to mention the same thing
1659 2011-04-19 15:57:18 <jaromil> after having seen how libdb handles format upgrades
1660 2011-04-19 15:57:21 <thermal> there needs to be a caching method
1661 2011-04-19 15:57:24 <tcatm> I see. Well, what about a TCP connection so we could finally do: build a transaction, tell UI the details + fee required and ask for actual permission to send
1662 2011-04-19 15:57:28 <grbgout> Uhm, any tips on identifying a GPU once it's in hand? I'm buying a used card, and I know the guy has an old 48xx model --- I want to make sure he isn't going to pull a fast one, and swap the cover things.
1663 2011-04-19 15:57:36 <thermal> hdd is accessed way too much
1664 2011-04-19 15:57:52 <jaromil> yep
1665 2011-04-19 15:58:09 <thermal> ugh i can hear it now
1666 2011-04-19 15:58:18 <jaromil> a workaround can be in shell launcher, copying .bitcoin to /dev/shm at start
1667 2011-04-19 15:58:19 <eps2> what are we putting the recent rise in value of BTC down to?
1668 2011-04-19 15:58:26 stamit has quit (Quit: stamit)
1669 2011-04-19 15:58:28 <thermal> that's gonna totally destroy the lifespan of hdds
1670 2011-04-19 15:58:29 <jaromil> since debian has already a wrapper to launch it....
1671 2011-04-19 15:58:41 <tcatm> eps2: -> #bitcoin-discussion
1672 2011-04-19 15:59:07 stamit has joined
1673 2011-04-19 15:59:20 <jaromil> not sure on win. osx has /dev/shm too AFAIK
1674 2011-04-19 15:59:46 <jaromil> but that's a workaround - still it touches less code as possible which i understand is a (wise IMHO) choice ATM
1675 2011-04-19 16:00:43 <tcatm> bitcoin intentionally writes every block/tx to disk as soon as it is processed
1676 2011-04-19 16:01:08 <grbgout> tcatm: is there a rationale provided anywhere?
1677 2011-04-19 16:01:14 <grbgout> comments, maybe?
1678 2011-04-19 16:01:14 <thermal> should atleast be optional
1679 2011-04-19 16:01:26 <tcatm> grbgout: avoid dataloss
1680 2011-04-19 16:01:28 <ArtForz> yesd, who needs transactional integrity
1681 2011-04-19 16:02:04 <thermal> it's not needed every time a block is processed. every minute would be enough
1682 2011-04-19 16:02:44 <thermal> it's not going to matter if a minute worth of processing is lost
1683 2011-04-19 16:03:04 <ArtForz> so, how do you sync shit before commiting to real disk?
1684 2011-04-19 16:03:05 <tcatm> Is it a problem during normal block broadcast? We could actually disable it during initial chain download.
1685 2011-04-19 16:03:25 <thermal> just use memory
1686 2011-04-19 16:03:30 <thermal> it's practically the same thing
1687 2011-04-19 16:03:35 <ArtForz> err... what?
1688 2011-04-19 16:03:44 <thermal> physical memory, RAM
1689 2011-04-19 16:03:52 <tcatm> Yep, as long as you have UPS or a laptop
1690 2011-04-19 16:03:55 <ArtForz> *headdesk*
1691 2011-04-19 16:04:40 <thermal> did i say something wrong?
1692 2011-04-19 16:04:40 <sacarlson> well I put my two cents in here https://github.com/sacarlson/bitcoin
1693 2011-04-19 16:05:03 <sacarlson> that's about all it's worth
1694 2011-04-19 16:05:11 <ArtForz> so... what is sync()ing while in initial block dl mode?
1695 2011-04-19 16:05:17 <thermal> i love it that everyone is using github for this
1696 2011-04-19 16:05:31 <tcatm> I'm not sure bitcoin could recover from out-of-sync blkindex/blk0001.dat, though...
1697 2011-04-19 16:05:38 <sacarlson> thermal: that's what they told me to use
1698 2011-04-19 16:05:58 <ArtForz> tcatm: it really can't properly
1699 2011-04-19 16:06:11 <thermal> sacarlson: then they told you right :)
1700 2011-04-19 16:06:18 <tcatm> Can we build the blkindex after download? Or every 500 blocks?
1701 2011-04-19 16:06:28 <ArtForz> err... I think we already do
1702 2011-04-19 16:06:40 <ArtForz> well, at least we only sync every 500 blocks
1703 2011-04-19 16:06:51 <sacarlson> thermal: yes but I had some other tell me later to use another but the account I created was already here
1704 2011-04-19 16:06:56 <ArtForz> thats why I'm asking if somethign is still syncing ever block somewhere, ebcause thats more like a bug then
1705 2011-04-19 16:07:23 <tcatm> doesn't WriteToDisk()'s fflush() cause a sync?
1706 2011-04-19 16:08:02 <tcatm> hm looks like it should not
1707 2011-04-19 16:08:40 <ArtForz> I'm pretty sure in initial block dl mode it should only sync/commit every 500 blocks
1708 2011-04-19 16:08:53 <tcatm> yep. if (!IsInitialBlockDownload() || (nBestHeight+1) % 500 == 0) fsync()
1709 2011-04-19 16:09:04 <ArtForz> though it's possible we still have somethign somewhere causing excessive syncs
1710 2011-04-19 16:09:32 <tcatm> btw, any progress on the CreateBlock() fixes?
1711 2011-04-19 16:10:26 <ArtForz> not really, kinda got sidetracked with designing the v2 pcbs for my miner
1712 2011-04-19 16:10:57 <jaromil> ArtForz: i was profiling using gprof yesterday, yet i've found nothing really anomalous. but then gprof is not the best for i/o profiling
1713 2011-04-19 16:11:17 <tcatm> run with strace during initialdownload?
1714 2011-04-19 16:11:18 <jaromil> AFAIK there are better profiling tools on OSX
1715 2011-04-19 16:11:38 <sacarlson> I have yet to find the needed information to create a new genesis block as I have seen gavinadneresen do but not sure how he get's the numbers to create it
1716 2011-04-19 16:11:52 <ArtForz> it's also not a really small change, as guaranteeing the (all tx in cache are checked and valid to go into next block) invariant needs changes to reorg, reaccept and something else I currently forgot
1717 2011-04-19 16:12:47 <thermal> are you guys using Cocoa?
1718 2011-04-19 16:13:15 <ArtForz> and I dont quite know what to do with for example now-invalid tx after chain reorg in reorg or reacceptwallettx
1719 2011-04-19 16:13:31 <ArtForz> the current code just throws em in cache and lets ProcessBlock sort it out
1720 2011-04-19 16:14:02 <ArtForz> which imo is a slightly weird approach
1721 2011-04-19 16:14:30 <tcatm> true... is a recent diff available somewhere, though?
1722 2011-04-19 16:14:34 gavinandresen has joined
1723 2011-04-19 16:14:54 <thermal> did the restructruing into classes code end up getting used? someone mentioned it recently...
1724 2011-04-19 16:15:30 <ArtForz> not quite, still working on making a utility "add all tx from this vector to cache if possible" function
1725 2011-04-19 16:16:33 <ArtForz> reorg needs it anyways and it means I dont have copypasta for acceptwallettx
1726 2011-04-19 16:16:37 <thermal> once i read the spec again (properly this time rather than skim) and fully take it all in, i can join in the coding fun :)
1727 2011-04-19 16:16:52 <ArtForz> actually the spec really only tells half the story
1728 2011-04-19 16:17:07 <thermal> hmm well that's not good
1729 2011-04-19 16:17:30 <thermal> i'll prolly read through the forums a bit too, then
1730 2011-04-19 16:18:22 <grbgout> ArtForz: are you releasing your miner pcbs under open-source hardware?
1731 2011-04-19 16:18:33 <ArtForz> nope
1732 2011-04-19 16:18:38 <grbgout> dang
1733 2011-04-19 16:18:44 <ArtForz> would be kinda useless without the custom ASICs...
1734 2011-04-19 16:19:02 <ArtForz> well, okay, the spartan6 controller could be useful for general prototyping
1735 2011-04-19 16:19:49 <ArtForz> S6 LX16, config flash, full speed usb, 2*24 pin 0.1" header
1736 2011-04-19 16:20:19 * grbgout totally knows what all that means.
1737 2011-04-19 16:20:27 <ArtForz> on a 25x72mm pcb
1738 2011-04-19 16:20:49 <sacarlson> ArtForz: I looked at altera how bit the new programable chips are not sure what they can do
1739 2011-04-19 16:20:54 <grbgout> hmm, actually, I think I know most of what that means, except what the S6 LX16 is
1740 2011-04-19 16:20:55 <netxshare> ^_^
1741 2011-04-19 16:20:59 <sacarlson> bit = big
1742 2011-04-19 16:21:04 <ArtForz> Spartan6 LX16
1743 2011-04-19 16:21:26 <ArtForz> $15 fpga that does usb<->custom sync serial bus
1744 2011-04-19 16:22:02 <ArtForz> also does temp monitoring via a cheap i2c sensor and pwm fan control
1745 2011-04-19 16:22:04 TheAncientGoat has quit (Ping timeout: 276 seconds)
1746 2011-04-19 16:22:24 <grbgout> nice
1747 2011-04-19 16:22:48 <ArtForz> and it's still only ~20% used
1748 2011-04-19 16:23:25 <ArtForz> a usb1.1 engine is really pretty small
1749 2011-04-19 16:23:29 <tcatm> any recommendation for a good FPGA eval board?
1750 2011-04-19 16:23:48 danbri has quit (Ping timeout: 252 seconds)
1751 2011-04-19 16:24:16 <luke-jr> [12:14:53] <grbgout> ArtForz: are you releasing your miner pcbs under open-source hardware?
1752 2011-04-19 16:24:19 <netxshare> I was looking at fpga's the other day dev-starter kits are rather highly priced
1753 2011-04-19 16:24:24 <luke-jr> ArtForz: let me rephrase his question:
1754 2011-04-19 16:24:25 <ArtForz> I kinda like the stuff from digilent for xilinx
1755 2011-04-19 16:24:41 <luke-jr> ArtForz: what would be the price to get your miner pcbs under open-source hardware?
1756 2011-04-19 16:24:46 <netxshare> more then I expected I guess I should say, tho they come with a ton of things on them.
1757 2011-04-19 16:24:53 <ArtForz> well ... yeah
1758 2011-04-19 16:25:43 <JFK911> ArtForz: what all does the spartan do? Handles entire USB protocol in one endpoint?
1759 2011-04-19 16:25:58 <ArtForz> yeah
1760 2011-04-19 16:26:27 <JFK911> dont like the ft232 eh
1761 2011-04-19 16:26:38 danbri_ has joined
1762 2011-04-19 16:26:39 <luke-jr> â¦
1763 2011-04-19 16:26:51 <luke-jr> ArtForz: hundreds? thousands? more? O.o
1764 2011-04-19 16:26:57 <ArtForz> well, then I'd need to implement a interface to that in hdl
1765 2011-04-19 16:27:04 <netxshare> that reminds me of something I found yesterday
1766 2011-04-19 16:27:10 <JFK911> oh right uarts arent very easy
1767 2011-04-19 16:27:30 topace has quit (Remote host closed the connection)
1768 2011-04-19 16:27:32 <ArtForz> btw, if you don't care about usb if compliance, you can directly interface a 3.3V I/O fpga with usb
1769 2011-04-19 16:27:38 <ArtForz> no need for even a transceiver
1770 2011-04-19 16:27:39 <JFK911> i have seen other people do usb in fpga but microcontrollers are real easy -> spi output
1771 2011-04-19 16:27:48 <JFK911> yeah, but bitbanging usb sounds like no-fun
1772 2011-04-19 16:28:02 <ArtForz> actually it's rather simple in hardware
1773 2011-04-19 16:28:03 <tcatm> bitbanging usb is simple
1774 2011-04-19 16:28:06 <luke-jr> sirius: I just installed wxGTK just for your bounty :P
1775 2011-04-19 16:28:14 <netxshare> http://www2.electronicproducts.com/Crypto_authentication_IC_uses_SHA_256-article-ICDJH04_May2011-html.aspx
1776 2011-04-19 16:28:25 <netxshare> $0.41 ea/1,000
1777 2011-04-19 16:28:33 <ArtForz> the usb engine with in/out endpoint fifos is like 10% of a LX16
1778 2011-04-19 16:28:37 <thermal> so, how's this for an idea
1779 2011-04-19 16:28:37 <JFK911> ok heres my perspective: its always been done for me, so i always took the PHY for granted and I dont want to learn it :)
1780 2011-04-19 16:29:09 <thermal> eftpos-like cards, except open
1781 2011-04-19 16:29:21 <JFK911> i have seen pic bitbang usb before
1782 2011-04-19 16:29:34 <JFK911> eftpos is meant to do one or two operations in a reasonable time
1783 2011-04-19 16:29:36 topace has joined
1784 2011-04-19 16:29:40 <ArtForz> well, the actual phy part is rather simpkle, theres one on opencores
1785 2011-04-19 16:29:51 <thermal> with a configuration on the card, specifying a URL for communication
1786 2011-04-19 16:29:52 <JFK911> opencores :D
1787 2011-04-19 16:30:06 <ArtForz> the tricky part is the protocol engine
1788 2011-04-19 16:30:06 <thermal> the reader must have Internet
1789 2011-04-19 16:30:20 <thermal> the idea is that it's open to anyone that wants to setup any system they want
1790 2011-04-19 16:30:22 <JFK911> I dont want to implement basic protocol either
1791 2011-04-19 16:30:28 <ArtForz> though I currently kinda cheat
1792 2011-04-19 16:30:42 PsyForce has joined
1793 2011-04-19 16:30:49 <JFK911> To define configurations and endpoints is not so bad.
1794 2011-04-19 16:30:51 <ArtForz> basically it just implements barely enough of a control EP to work
1795 2011-04-19 16:31:03 <thermal> it could simply use SSL/HTTPS
1796 2011-04-19 16:31:15 <ArtForz> well, my interface to the other logic is ... 2 FIFOs
1797 2011-04-19 16:31:28 <ArtForz> 8-bit wide, 128 entries deep
1798 2011-04-19 16:31:47 <ArtForz> with simple full/empty flags for in/out
1799 2011-04-19 16:31:55 <thermal> then, it's just a matter of satisfying the merchant's request (e.g. pay $x to this location)
1800 2011-04-19 16:32:12 <ArtForz> so for the rest of the logic, the whole usb part looks like a bytewise stream pipe
1801 2011-04-19 16:32:49 <midnightmagic> time to build a slime-mold sha2 solver
1802 2011-04-19 16:32:58 stamit has left ()
1803 2011-04-19 16:33:55 <thermal> just how much faster is hardware processing?
1804 2011-04-19 16:34:13 <sacarlson> ArtForz: sounds like the simple altera chip design we used for a radar circuit to capture high speed analog to digital data in pulses
1805 2011-04-19 16:34:34 <ArtForz> yeah
1806 2011-04-19 16:36:24 <ArtForz> I *really* like to throw FGPAs or at least a CPLD or two in custom digital designs
1807 2011-04-19 16:36:46 <ArtForz> changing your HDL and reprogramming sure as hell beats green wire
1808 2011-04-19 16:37:15 <sacarlson> ArtForz: I was looking at a more resent design that used fpga for the software radio project
1809 2011-04-19 16:37:44 <ArtForz> I kidna like gnuradios design, but their fpga is rather... dated
1810 2011-04-19 16:38:07 <ArtForz> *gnuradios USRP
1811 2011-04-19 16:38:13 <sacarlson> ArtForz: yes
1812 2011-04-19 16:38:36 <lulzplzkthx> Sorry to keep posting this, but I want to try and keep this from goetting voted down, can anyone who hasn't already please upvote this: http://www.reddit.com/r/ideasfortheadmins/comments/grz9l/allow_bitcoin_as_an_alternative_payment_for/
1813 2011-04-19 16:38:39 <netxshare> so does http://www.bitcoingadgets.com ship to the us?
1814 2011-04-19 16:38:40 <lulzplzkthx> It would be greatly appreciated.
1815 2011-04-19 16:39:29 <tcatm> netxshare: they ship worldwide (iirc)
1816 2011-04-19 16:39:54 <netxshare> I wonder if I could get a cell phone jammer here safely
1817 2011-04-19 16:39:54 <ArtForz> hmmm... wonder why no one changed USRP2 to use a S6LX50
1818 2011-04-19 16:40:36 <netxshare> what bitcoin clients does everyone use, just the default?
1819 2011-04-19 16:40:45 <ArtForz> should be about the same capacity as a S3-2000, can clock a lot higher and *is* support by free(beer) ISE WebPack
1820 2011-04-19 16:41:01 <tcatm> netxshare: default, mybitcoin and mtgox are common
1821 2011-04-19 16:41:18 <sacarlson> ArtForz: they probly have old farts like me programing them
1822 2011-04-19 16:41:38 <BlueMatt> netxshare: probably a question to ask on #bitcoin-otc
1823 2011-04-19 16:41:42 <netxshare> thanks tactm
1824 2011-04-19 16:41:47 <lulzplzkthx> netxshare: Check out Spesmilo.
1825 2011-04-19 16:42:36 <sacarlson> lulzplzkthx: I'll check it out where is it?
1826 2011-04-19 16:42:43 <eps2> ;;bc,stats
1827 2011-04-19 16:42:47 <gribble> Current Blocks: 119139 | Current Difficulty: 92347.59095209 | Next Difficulty At Block: 120959 | Next Difficulty In: 1820 blocks | Next Difficulty In About: 1 week, 5 days, 15 hours, 50 minutes, and 20 seconds | Next Difficulty Estimate: 92610.43988269
1828 2011-04-19 16:42:49 <netxshare> does it support multi wallets?
1829 2011-04-19 16:43:07 <sacarlson> lulzplzkthx: I think it's what I needed to control bitcoind from a gui correct
1830 2011-04-19 16:43:14 MingusDew has quit (Quit: http://folding.stanford.edu/)
1831 2011-04-19 16:43:34 <lulzplzkthx> This is neat: http://goo.gl/8ELJk
1832 2011-04-19 16:43:54 <lulzplzkthx> sacarlson: https://en.bitcoin.it/wiki/Spesmilo
1833 2011-04-19 16:44:02 <lulzplzkthx> I'm not sure netxshare . I haven't tried it.
1834 2011-04-19 16:44:17 <lulzplzkthx> I don't think so though.
1835 2011-04-19 16:44:23 <lulzplzkthx> I might set about creating my own client sometime soon...
1836 2011-04-19 16:44:37 <netxshare> that's what I am about to start working on
1837 2011-04-19 16:44:46 <netxshare> client pool server and php backend
1838 2011-04-19 16:44:46 <sacarlson> lulzplzkthx: I googled it and will try it
1839 2011-04-19 16:45:42 <netxshare> I would like be able to have php support wallets for the users on the pool
1840 2011-04-19 16:47:48 <lulzplzkthx> netxshare: Please check out my cURL implementation of JSOn-RPC if you plan on using JSON-RPC, as it seems to be much more reliable.
1841 2011-04-19 16:48:20 danbri_ is now known as danbri
1842 2011-04-19 16:48:39 Zarutian has quit (Quit: Zarutian)
1843 2011-04-19 16:48:46 <netxshare> I think I saw it the forums, but I am going to move away from JSON-RPC
1844 2011-04-19 16:48:50 <finnomenon> http://timothyblee.com/2011/04/19/bitcoins-collusion-problem/
1845 2011-04-19 16:49:08 <finnomenon> pretty interesting problem he points out
1846 2011-04-19 16:49:32 <lulzplzkthx> netxshare: What do you plan to use?
1847 2011-04-19 16:50:15 <lulzplzkthx> Do any domain egistrars accept Bitcoin yet?
1848 2011-04-19 16:50:28 <tcatm> lulzplzkthx: kalyhost
1849 2011-04-19 16:50:53 <jgarzik> finnomenon: old and well-discussed, not very interesting :)
1850 2011-04-19 16:50:59 <jgarzik> finnomenon: read the comments
1851 2011-04-19 16:51:17 <lulzplzkthx> Thanks tcatm
1852 2011-04-19 16:51:32 <lulzplzkthx> I'm thinking about writing an essay on Bitcoin for class (have to read something to my English class.)
1853 2011-04-19 16:51:53 <lulzplzkthx> Also, I wonder if submitting info about Bitcoin to newspapers like New York Times would be helpful to Bitcoin at all?
1854 2011-04-19 16:52:05 <Kiba> dunno
1855 2011-04-19 16:52:34 <lulzplzkthx> Might get some notice if they decide to publish something, yeah?
1856 2011-04-19 16:52:45 <lulzplzkthx> Only issue I see is a wave of noobies who don't know how to attach a file to an email.
1857 2011-04-19 16:52:54 <finnomenon> I guess they have journalists that already know about bitcoin
1858 2011-04-19 16:52:56 <Kiba> jgarzik: I seem to have misread the problem stated in that blog post
1859 2011-04-19 16:53:23 <Kiba> jgarzik: but if it is old and well-discussed, then critics are usually less informed about bitcoin
1860 2011-04-19 16:53:27 <lulzplzkthx> I think Bitcoin needs a "Press" section on bitcoin.org
1861 2011-04-19 16:53:53 <Kiba> eventually, the mediastorm will makes journalists notice
1862 2011-04-19 16:54:00 <Kiba> lulzplzkthx: already have that
1863 2011-04-19 16:54:13 <netxshare> I could see someone doing that
1864 2011-04-19 16:54:33 <netxshare> I agree with you luzplzkthx
1865 2011-04-19 16:54:37 <netxshare> there are many news sites
1866 2011-04-19 16:54:43 <netxshare> just needs links to them
1867 2011-04-19 16:54:45 <Kiba> there's a mailing list for the rpess
1868 2011-04-19 16:55:02 <PsyForce> hi all, is this an appropriate place to ask technical-ish questions? the discussion channel didn't seem to be too fruitful..
1869 2011-04-19 16:55:37 <netxshare> never hurts to ask?
1870 2011-04-19 16:55:47 <lulzplzkthx> PsyForce: Yeah. The discussion channel hardly has anyone ever.
1871 2011-04-19 16:55:56 <lulzplzkthx> I don't even idle there anymore.
1872 2011-04-19 16:56:10 <PsyForce> right then, I've just found out about bitcoin and am using ubuntu
1873 2011-04-19 16:56:34 <PsyForce> I unpacked the tar.gz and seem to be able to just run it without 'installing' it further
1874 2011-04-19 16:56:37 <PsyForce> is that normal?
1875 2011-04-19 16:56:47 <Kiba> don't know
1876 2011-04-19 16:56:50 <Kiba> I don't run window
1877 2011-04-19 16:56:58 <PsyForce> also, I am curious about what the headless version is and how to use it
1878 2011-04-19 16:57:12 LtBrenton_ is now known as LtBrenton
1879 2011-04-19 16:57:13 <midnightmagic> pretty much, yeah. you can check and see if it's begun downloading the block chain (~220MB, so be prepared to wait a while) into ~/.bitcoin/
1880 2011-04-19 16:57:14 LtBrenton has quit (Changing host)
1881 2011-04-19 16:57:14 LtBrenton has joined
1882 2011-04-19 16:57:32 <netxshare> has anyone messed with blackmarket.py ?
1883 2011-04-19 16:57:35 <midnightmagic> the headless version is called bitcoind, and it can be run in the backgroun. no gui.
1884 2011-04-19 16:57:38 RazielZ has quit (Read error: Connection reset by peer)
1885 2011-04-19 16:58:32 <PsyForce> if I'm running the headless version, can I open the gui to check on it without messing things up?
1886 2011-04-19 16:58:50 <PsyForce> currently I have the gui open, it's generating blocks
1887 2011-04-19 16:59:16 <lulzplzkthx> PsyForce: you can always just run "bitcoind getinfo" while another bitcoind is running in the background
1888 2011-04-19 16:59:26 <PsyForce> ok
1889 2011-04-19 16:59:29 <lulzplzkthx> "bitcoind x" will run x from the API call list https://en.bitcoin.it/wiki/Original_Bitcoin_client/API_calls_list
1890 2011-04-19 16:59:48 <lulzplzkthx> Another example would be "bitcoind getbalance"
1891 2011-04-19 17:00:06 <lulzplzkthx> Class, later.
1892 2011-04-19 17:00:12 <PsyForce> thanks
1893 2011-04-19 17:00:15 <PsyForce> :)
1894 2011-04-19 17:00:16 <sacarlson> finnomenon: well my prediction is there will be many bitcoins (different genesis) running at some point with different rules each making up a different share of the market that the banks will have to compete with
1895 2011-04-19 17:00:33 <PsyForce> so how do I know if it's downloading the block chain?
1896 2011-04-19 17:01:30 <midnightmagic> i'm not sure what it looks like in the GUI.. sorry man. all I know is what it looks like in the debug log. :-) it only happens once, so the memory of it is distant for me.
1897 2011-04-19 17:02:53 <PsyForce> hehe, ok maybe I'll close the gui and start it up headless...
1898 2011-04-19 17:03:58 <midnightmagic> yeah, timothy is correct this time.
1899 2011-04-19 17:04:17 Teslah has joined
1900 2011-04-19 17:04:49 skyewm has joined
1901 2011-04-19 17:04:58 <sacarlson> lulzplzkthx: that's what I've been having to use (headless) since I presently don't have the libs available to compile the bitcoin bin
1902 2011-04-19 17:07:16 <sacarlson> lulzplzkthx: I also found that I could control my custom bitcoind that was already running with the bitcoind I installed from the ubuntu repository so I see the python gui to be perfect for me
1903 2011-04-19 17:07:57 Hyper-Core has joined
1904 2011-04-19 17:09:52 Hyper-Core has quit (Excess Flood)
1905 2011-04-19 17:11:28 danlucraft has quit (Quit: Leaving.)
1906 2011-04-19 17:12:00 <netxshare> so far all the python clients on the wiki seem to all require you run bitcoin
1907 2011-04-19 17:12:48 <phantomcircuit> netxshare, http://github.com/phantomcircuit/bitcoin-alt
1908 2011-04-19 17:13:11 <netxshare> nice thanks
1909 2011-04-19 17:14:24 <LobsterMan> ;;bc,stats
1910 2011-04-19 17:14:25 <gribble> Current Blocks: 119143 | Current Difficulty: 92347.59095209 | Next Difficulty At Block: 120959 | Next Difficulty In: 1816 blocks | Next Difficulty In About: 1 week, 5 days, 14 hours, 40 minutes, and 0 seconds | Next Difficulty Estimate: 92752.44055717
1911 2011-04-19 17:14:35 <LobsterMan> ;;bc,calc 115000
1912 2011-04-19 17:14:36 <gribble> The average time to generate a block at 115000 Khps, given current difficulty of 92347.59095209 , is 5 weeks, 4 days, 22 hours, 2 minutes, and 35 seconds
1913 2011-04-19 17:14:38 <LobsterMan> >_>
1914 2011-04-19 17:16:08 trifon has joined
1915 2011-04-19 17:16:20 <netxshare> weak I have to use git to download
1916 2011-04-19 17:17:17 <phantomcircuit> netxshare, it's not a complete client ...
1917 2011-04-19 17:17:30 <phantomcircuit> also there is a giant downloads button
1918 2011-04-19 17:17:31 <phantomcircuit> lol
1919 2011-04-19 17:19:05 TD_ has quit (Quit: TD_)
1920 2011-04-19 17:19:22 <thermal> idea: in the right click context menu of the system tray icon, there should be a 'Copy Bitcoin Address to Clipboard' option
1921 2011-04-19 17:19:23 <netxshare> nothing is there when I click on it
1922 2011-04-19 17:19:26 PsyForce has left ()
1923 2011-04-19 17:19:52 <netxshare> Sorry, there aren't any downloads for this repository.
1924 2011-04-19 17:19:54 ezl has joined
1925 2011-04-19 17:20:09 <thermal> or simply left mouse clicking the icon once could have the same effect
1926 2011-04-19 17:20:41 <netxshare> should add in like bitcoin://
1927 2011-04-19 17:20:49 <netxshare> so you can add links on websites to address
1928 2011-04-19 17:21:09 <netxshare> so the client can handle input and open a payment
1929 2011-04-19 17:21:23 <tcatm> js-remote can do that
1930 2011-04-19 17:21:51 twobitcoins_ is now known as twobitcoins
1931 2011-04-19 17:21:54 <netxshare> oh I see
1932 2011-04-19 17:22:02 <netxshare> did not know that
1933 2011-04-19 17:22:29 <thermal> and knowing is half the battle!
1934 2011-04-19 17:22:44 <thermal> http://www.youtube.com/watch?v=WkJQ3seomLo
1935 2011-04-19 17:23:12 amiller has quit (Ping timeout: 246 seconds)
1936 2011-04-19 17:23:24 <thermal> probably should have watched it myself first
1937 2011-04-19 17:24:36 <sacarlson> how about a bitcoin in ruby? my favorite language
1938 2011-04-19 17:29:54 <Kiba> Marginal Revolution mentioned Bitcoin
1939 2011-04-19 17:30:08 Zarutian has joined
1940 2011-04-19 17:30:09 <phantomcircuit> sacarlson, how about no
1941 2011-04-19 17:30:39 <sacarlson> phantomcircuit: that's fine if your python works I'm happy with that
1942 2011-04-19 17:30:52 <luke-jr> hrm
1943 2011-04-19 17:31:00 <luke-jr> got the IPC part figured out, but I really don't know wxâ¦
1944 2011-04-19 17:31:11 <luke-jr> is it legal for ThreadB to interact with wx stuff?
1945 2011-04-19 17:31:12 <phantomcircuit> luke-jr, whatcha doin
1946 2011-04-19 17:31:31 <luke-jr> phantomcircuit: bitcoin URI support for orig client
1947 2011-04-19 17:31:39 <ezl> ;;bc,mtgox
1948 2011-04-19 17:31:40 <gribble> {"ticker":{"high":1.19,"low":1.1202,"vol":14166,"buy":1.1812,"sell":1.1895,"last":1.1894}}
1949 2011-04-19 17:34:37 <sacarlson> phantomcircuit but at some point with my new multi coin client if I can't find a gui to run it I will have to write my own and maybe in ruby
1950 2011-04-19 17:35:03 <phantomcircuit> multicoin client?
1951 2011-04-19 17:35:45 <sacarlson> phantomcircuit yes that can transact in unliminted number of coin types
1952 2011-04-19 17:35:57 KBme has quit (Read error: Operation timed out)
1953 2011-04-19 17:37:11 <phantomcircuit> sacarlson, the fuck would you do that?
1954 2011-04-19 17:37:17 <sacarlson> https://github.com/sacarlson/bitcoin
1955 2011-04-19 17:37:45 <sacarlson> phantomcircuit just simply have different config files to accept different types of coin
1956 2011-04-19 17:38:03 <phantomcircuit> when you say different types of coins
1957 2011-04-19 17:38:09 <phantomcircuit> you mean a different genesis block?
1958 2011-04-19 17:38:15 <sacarlson> phantomcircuit yes
1959 2011-04-19 17:38:24 <phantomcircuit> lol that's not gonna happen
1960 2011-04-19 17:38:31 <sacarlson> phantomcircuit and different networks and different rules
1961 2011-04-19 17:38:38 <midnightmagic> it could if the mining effort could be piggy-backed.
1962 2011-04-19 17:38:49 <sacarlson> phantomcircuit well as you can see it already has
1963 2011-04-19 17:38:52 <midnightmagic> similar to the overlap of bitcoins with bitdns that satoshi described.
1964 2011-04-19 17:39:50 <phantomcircuit> midnightmagic, uh they cant really piggy back
1965 2011-04-19 17:40:01 KBme has joined
1966 2011-04-19 17:40:04 <phantomcircuit> mining effort would have to be diverted
1967 2011-04-19 17:40:08 <midnightmagic> yes they can.
1968 2011-04-19 17:40:09 <sacarlson> it will allow for quicker transactions with many networks running at the same time
1969 2011-04-19 17:40:40 <midnightmagic> so read satoshi's posts about bitdns. it's quite possible the same way people mine in pools; you just report for more than one difficulty level.
1970 2011-04-19 17:40:44 <midnightmagic> so => go
1971 2011-04-19 17:41:02 <sacarlson> but different coins will have different value depending on market value and time to create
1972 2011-04-19 17:41:12 <jgarzik> sacarlson: yes
1973 2011-04-19 17:41:27 <midnightmagic> i think it's a splendid idea.
1974 2011-04-19 17:41:39 <phantomcircuit> midnightmagic, uhm what?
1975 2011-04-19 17:41:55 <midnightmagic> i said, I think it's a splendid idea. :)
1976 2011-04-19 17:42:06 <sacarlson> it won't take 10 mintes to fully transact it will only take secounds
1977 2011-04-19 17:42:42 <midnightmagic> it could enable the conversion of one coin type to another, and allow for one coin's backing by the more established ฿..
1978 2011-04-19 17:42:50 <sacarlson> since each network will have different that will randomly create coin
1979 2011-04-19 17:43:29 <phantomcircuit> sacarlson, uh that wont work
1980 2011-04-19 17:43:35 zylche has quit (Remote host closed the connection)
1981 2011-04-19 17:43:42 <sacarlson> midnightmagic: yes at some point dependent on the genesis time of creation determine it's true value
1982 2011-04-19 17:44:06 <sacarlson> phantomcircuit and why not?
1983 2011-04-19 17:44:16 <phantomcircuit> sacarlson, you'd have coins on only a single network
1984 2011-04-19 17:44:22 <phantomcircuit> so it would still be 10 minutes
1985 2011-04-19 17:44:57 <sacarlson> phantomcircuit no the code will later provide haveing more than one network and that's the way I have it now on a private net
1986 2011-04-19 17:45:05 <midnightmagic> anyway, it could be a wonderful idea. i have no idea about the implementation so i can't comment much past that.
1987 2011-04-19 17:45:28 <phantomcircuit> sacarlson, uh that's literally impossible
1988 2011-04-19 17:45:50 <sacarlson> phantomcircuit my private net that runs already is impossible?
1989 2011-04-19 17:46:14 <phantomcircuit> sacarlson, it's impossible to increase the time between block generation
1990 2011-04-19 17:46:17 <phantomcircuit> consider
1991 2011-04-19 17:46:28 <phantomcircuit> you want to send 10USD to me
1992 2011-04-19 17:46:33 <phantomcircuit> you have coins on 2 networks
1993 2011-04-19 17:46:47 <phantomcircuit> each network generates a block every 10 minutes
1994 2011-04-19 17:46:49 <midnightmagic> sacarlson: don't worry about him, he's just mouthy and doesn't think non-phantomcircuit ways of doing things can be viable.
1995 2011-04-19 17:46:53 <sacarlson> phantomcircuit remind you each net generates it's own blocks not included in the others chain
1996 2011-04-19 17:46:53 <phantomcircuit> lets assume perfect overlap
1997 2011-04-19 17:46:55 <midnightmagic> :-)
1998 2011-04-19 17:46:57 <tabsa> jgarzik: about the miner collusion problem, has it been discussed that a solution for that might be all new bitcoin clients coming with "Generate coins" enabled by default, only limited to 10-20% max CPU?
1999 2011-04-19 17:47:03 <phantomcircuit> sacarlson, just listen
2000 2011-04-19 17:47:09 <phantomcircuit> midnightmagic, lol
2001 2011-04-19 17:47:26 <midnightmagic> ;-)
2002 2011-04-19 17:48:11 <midnightmagic> no offence intended of course, you crazy mofo
2003 2011-04-19 17:48:13 <phantomcircuit> sacarlson, so you send me 10 USD in coins on the main network
2004 2011-04-19 17:48:21 <phantomcircuit> you have one small problem
2005 2011-04-19 17:48:49 <sacarlson> I would think in this method it would require less electric power that the minners will have to evaluate and change to the chain that best pays
2006 2011-04-19 17:49:08 <phantomcircuit> just because a block hasn't been generated in the last 9 minutes does not increase the odds of a block being generated
2007 2011-04-19 17:49:10 <phantomcircuit> thus
2008 2011-04-19 17:49:18 <phantomcircuit> if they both generate every 10 minutes on average
2009 2011-04-19 17:49:33 <phantomcircuit> they both have the same odds of generating a block at any given moment
2010 2011-04-19 17:49:35 <midnightmagic> sacarlson: consider merging the two networks into the same work unit so mining effort doesn't have to be diverted.
2011 2011-04-19 17:50:02 <sacarlson> phantomcircuit two nets will double the speed of 10 mintes to an average of 5 mins and 1000 nets will make that time again smaller
2012 2011-04-19 17:50:18 <phantomcircuit> sacarlson, wrong, because you have to pick a net to send the coins on
2013 2011-04-19 17:50:30 <midnightmagic> whoa.
2014 2011-04-19 17:50:31 <phantomcircuit> you cant use all of them simultaneously for a single (logical) transaction
2015 2011-04-19 17:50:33 * midnightmagic goes back to sleep.
2016 2011-04-19 17:51:14 <sacarlson> phantomcircuit my mutli coin client can be controlled or automated to pick the best path to transact in the shotest time
2017 2011-04-19 17:51:29 <phantomcircuit> sacarlson, you're not paying attention
2018 2011-04-19 17:51:38 <phantomcircuit> there *isnt* a shortest time
2019 2011-04-19 17:51:49 <phantomcircuit> unless of course the networks agree to different times
2020 2011-04-19 17:51:51 <sacarlson> phantomcircuit your not thinking in paralel
2021 2011-04-19 17:52:32 <phantomcircuit> have you ever taken a statistics course?
2022 2011-04-19 17:52:37 <sacarlson> phantomcircuit the code of each network is different the rules are differnent the risks are different
2023 2011-04-19 17:53:19 <sacarlson> it will be up to the market to decide
2024 2011-04-19 17:54:03 amiller has joined
2025 2011-04-19 17:54:12 eternal1 has quit (Quit: Leaving.)
2026 2011-04-19 17:54:12 <phantomcircuit> sacarlson, lol you dont get it at all
2027 2011-04-19 17:54:56 <sacarlson> phantomcircuit so you want to keep only one coin as the basis of the universe?
2028 2011-04-19 17:55:12 <phantomcircuit> yes?
2029 2011-04-19 17:55:45 <phantomcircuit> there's absolutely no reason to change to a new genesis
2030 2011-04-19 17:55:48 devrandom_ has quit (Remote host closed the connection)
2031 2011-04-19 17:56:27 <sacarlson> phantomcircuit well look at the world today do we only have USD ? I assure you there will be others
2032 2011-04-19 17:56:37 <phantomcircuit> rofl
2033 2011-04-19 17:56:48 devrandom_ has joined
2034 2011-04-19 17:56:51 <phantomcircuit> we have multiple currencies for a reason
2035 2011-04-19 17:57:06 <phantomcircuit> there is no reason to have multiple genesis blocks
2036 2011-04-19 17:57:27 <sacarlson> they already exist your too late to try to stop them
2037 2011-04-19 17:57:38 <phantomcircuit> i dont have to stop them
2038 2011-04-19 17:57:42 <phantomcircuit> they'll fail all on their own
2039 2011-04-19 17:57:43 <phantomcircuit> lol
2040 2011-04-19 17:57:51 <sacarlson> no the market will or won't
2041 2011-04-19 17:58:37 <phantomcircuit> im pretty sure they'll fail all on their own
2042 2011-04-19 17:58:38 <LtBrenton> gah, I wish mtgox would show unconfirmed transactions as 'pending'
2043 2011-04-19 17:58:58 <sacarlson> but I think there are other things to trade other that currency that has no value in itself why not trade companies with it
2044 2011-04-19 18:00:36 <sacarlson> I am looking into the design of new rules to trade equities and bonds and comodites using the bitcoin protocol with different genesis
2045 2011-04-19 18:01:22 <phantomcircuit> hmm
2046 2011-04-19 18:01:26 <phantomcircuit> that could work actually
2047 2011-04-19 18:01:40 <phantomcircuit> except one small problem
2048 2011-04-19 18:01:46 <phantomcircuit> they would be small networks
2049 2011-04-19 18:02:00 <phantomcircuit> it would be trivial to dominate them
2050 2011-04-19 18:02:05 <sacarlson> phantomcircuit it should open up a micro trading envirment that only the big boy once could play
2051 2011-04-19 18:02:33 <phantomcircuit> what?
2052 2011-04-19 18:03:44 <sacarlson> you could sell parts of a $1000 company into over 1000 shares so the risk is so small between the parties that any one could have a part of the action
2053 2011-04-19 18:04:10 <phantomcircuit> uh
2054 2011-04-19 18:04:26 <phantomcircuit> that's not exactly a big boys club
2055 2011-04-19 18:04:31 LtBrenton has quit (Read error: Connection reset by peer)
2056 2011-04-19 18:04:36 <sacarlson> with no cost to transact to sell at any time to another buyer
2057 2011-04-19 18:04:55 LtBrenton has joined
2058 2011-04-19 18:05:49 <sacarlson> In the markets we now play in we pay $9 to make a transaction so this brings it down to the smaller players
2059 2011-04-19 18:06:09 <phantomcircuit> sacarlson, you realize that the cost would not be trivial right?
2060 2011-04-19 18:06:30 <phantomcircuit> you'd have to have enough miner power to avoid a network take over
2061 2011-04-19 18:06:31 <sacarlson> phantomcircuit not I don't
2062 2011-04-19 18:06:55 TD_ has joined
2063 2011-04-19 18:07:41 <sacarlson> phantomcircuit no the rules on these will only allow a maximum number of coins mined until you fall back to a dedicated server for this coin
2064 2011-04-19 18:08:00 tenach has joined
2065 2011-04-19 18:08:32 <sacarlson> or only allow a cirtain group of miners that are controled by the company to participate in mining
2066 2011-04-19 18:09:27 <sacarlson> the rules must change a bit for this to work
2067 2011-04-19 18:11:14 CyanDynamo1 has quit (Quit: Leaving.)
2068 2011-04-19 18:12:30 <sacarlson> but for other genesis of coins and networks the rules are wide open
2069 2011-04-19 18:13:29 <lulzplzkthx> Are you talking about a Bitcoin stock exchange?
2070 2011-04-19 18:14:13 <sacarlson> lulzplzkthx: yes and bond and comodities and open to anything
2071 2011-04-19 18:14:21 <sacarlson> lulzplzkthx: contracts
2072 2011-04-19 18:14:32 <lulzplzkthx> I see.
2073 2011-04-19 18:14:34 <sacarlson> lulzplzkthx: futures options
2074 2011-04-19 18:15:13 <sacarlson> I see it as the future of trade
2075 2011-04-19 18:15:14 <phantomcircuit> sacarlson, implementing futures is actually impossible afaict
2076 2011-04-19 18:15:36 Zarutian has quit (Quit: Zarutian)
2077 2011-04-19 18:16:11 <sacarlson> phantomcircuit it's just another market just it's value changes over time it's up to the buyers and sellers what the value is at any point in time
2078 2011-04-19 18:16:12 <lulzplzkthx> I'm so excited about Bitcoin, everyone needs to use it.
2079 2011-04-19 18:16:32 <BlueMatt> lulzplzkthx: then spread the word :)
2080 2011-04-19 18:16:38 <lulzplzkthx> I wish I could host a lecture, or informational thing on it in my area, but a) I don't think I really know quite enough about it. Maybe enough to get people started. b) I don't think people would like a lecture led by a 15-year-old.
2081 2011-04-19 18:16:41 <phantomcircuit> sacarlson, how do you force someone to deliver on a futures contract if you dont know who they are?
2082 2011-04-19 18:16:42 <lulzplzkthx> But don't worry, I am spreading the word.
2083 2011-04-19 18:16:55 <lulzplzkthx> Posting on Facebook frequently, asking sites to use Bitcoins, telling friends (both online and offline).
2084 2011-04-19 18:17:21 <BlueMatt> phantomcircuit: you cant, but its trust
2085 2011-04-19 18:17:32 <lulzplzkthx> Use the WoT?
2086 2011-04-19 18:17:36 <lulzplzkthx> From OTC?
2087 2011-04-19 18:17:40 <sacarlson> phantomcircuit on reputation as we do in the #bitcoin-otc if you fail to come threw on privios contracts you unlikely to continue to trade
2088 2011-04-19 18:17:49 <BlueMatt> phantomcircuit: its not like if I buy a futures contract today I have any idea who is selling it
2089 2011-04-19 18:18:17 tenach has quit (Ping timeout: 258 seconds)
2090 2011-04-19 18:18:43 <phantomcircuit> BlueMatt, a normal futures contract has the force of law behind it, and iirc there is a chain of liability
2091 2011-04-19 18:19:13 <sacarlson> phantomcircuit I've heard that before but last time you got robed what did the law do for you?
2092 2011-04-19 18:19:23 <BlueMatt> if you actually sign a contract for bitcoin it would have the force of law...though i suppose thats not the point
2093 2011-04-19 18:19:32 <phantomcircuit> sacarlson, a robbery is anonymous ;)
2094 2011-04-19 18:19:52 <phantomcircuit> BlueMatt, the point is that the futures contract is not anonymous or even pseudo anonymous
2095 2011-04-19 18:19:58 <BlueMatt> Anyway, it has been done on otc and I think its fine, though I would never do it with someone with low trust
2096 2011-04-19 18:20:04 <sacarlson> phantomcircuit so look at enron what good did the law do for them?
2097 2011-04-19 18:20:44 <sacarlson> phantomcircuit you have to gain trust over time that's the real law
2098 2011-04-19 18:20:44 <phantomcircuit> BlueMatt, well implementing a WoT is trivial
2099 2011-04-19 18:20:55 <phantomcircuit> i think ill build a btc futures market that operates over tor
2100 2011-04-19 18:20:59 <phantomcircuit> just for shits and giggles
2101 2011-04-19 18:21:18 <phantomcircuit> and when i say shits and giggles i mean 1%
2102 2011-04-19 18:21:24 <BlueMatt> phantomcircuit: if there is a central authority that prevents people from backing out (ie you have to have the money in your account at all times)
2103 2011-04-19 18:21:46 <phantomcircuit> a futures market where you have to keep money on deposit isn't of much use
2104 2011-04-19 18:22:19 <BlueMatt> if it was a mybitcoin alt which people use anyway why not?
2105 2011-04-19 18:22:50 <Kiba> I think bitcoin critics are going to be surprised by namecoin
2106 2011-04-19 18:23:12 <sacarlson> phantomcircuit we are working on an escrow over bitcoin but there are third parties
2107 2011-04-19 18:23:28 <phantomcircuit> escrow is trivial
2108 2011-04-19 18:23:38 <sacarlson> phantomcircuit or you just work on trust up to you
2109 2011-04-19 18:24:33 seafoodchef has joined
2110 2011-04-19 18:25:03 <seafoodchef> hello?
2111 2011-04-19 18:25:07 <Kiba> 4 more unique visitors than I'll have my 200 unique visitors today
2112 2011-04-19 18:25:09 <Kiba> MUHAHAHHAHA
2113 2011-04-19 18:25:20 <phantomcircuit> Kiba, ?
2114 2011-04-19 18:25:29 <seafoodchef> anyone wanna play penny poker?
2115 2011-04-19 18:26:07 tenach has joined
2116 2011-04-19 18:26:07 tenach has quit (Changing host)
2117 2011-04-19 18:26:07 tenach has joined
2118 2011-04-19 18:26:55 danlucraft has joined
2119 2011-04-19 18:26:55 <sacarlson> seafoodchef: I'll play you poker in pokerth for .02btc for 3000 chips
2120 2011-04-19 18:27:11 <Kiba> phantomcircuit: http://bitcoinweekly.com is getting traffics
2121 2011-04-19 18:27:47 Teslah has quit (Ping timeout: 240 seconds)
2122 2011-04-19 18:28:54 <seafoodchef> what is pokerth?
2123 2011-04-19 18:29:11 <sacarlson> seafoodchef: see my pm
2124 2011-04-19 18:30:34 tenach is now known as tenach|AFK
2125 2011-04-19 18:33:25 berkes has joined
2126 2011-04-19 18:35:04 kermit has quit (Ping timeout: 252 seconds)
2127 2011-04-19 18:36:00 amiller has quit (Ping timeout: 246 seconds)
2128 2011-04-19 18:37:43 kermit has joined
2129 2011-04-19 18:40:04 mmoya has quit (Ping timeout: 248 seconds)
2130 2011-04-19 18:43:46 <lulzplzkthx> Glad to see the Faucet is back up.
2131 2011-04-19 18:43:53 <lulzplzkthx> Has any kind of new anti-cheating mechanisms been implemented?
2132 2011-04-19 18:46:40 ByteCoin has quit (Ping timeout: 246 seconds)
2133 2011-04-19 18:46:52 citiz3n has joined
2134 2011-04-19 18:48:02 CyanDynamo has joined
2135 2011-04-19 18:51:47 CodePHP has joined
2136 2011-04-19 18:52:21 CodePHP is now known as AFK_CodePHP
2137 2011-04-19 18:57:44 seafoodchef has quit (Ping timeout: 252 seconds)
2138 2011-04-19 19:02:17 DukeOfURL has joined
2139 2011-04-19 19:02:23 AFK_CodePHP is now known as AFK_PHPAdam
2140 2011-04-19 19:02:33 AFK_PHPAdam is now known as AFK_CodePHP
2141 2011-04-19 19:02:36 AFK_CodePHP is now known as AFK_PHPAdam
2142 2011-04-19 19:07:37 JuanDaugherty has joined
2143 2011-04-19 19:09:06 <JuanDaugherty> more on "what if somebody bought all the bitcoins?": at the current record price, and 150K metric tonnes, the price of all the gold is 7.2 trillion USD.
2144 2011-04-19 19:12:13 <JuanDaugherty> since no one person has even 100 billion USD, no non-sovereign entity has liquid assets in the trillions, etc. ... OTOH many individuals could corner the bitcoin market if they so chose.
2145 2011-04-19 19:12:57 <phantomcircuit> JuanDaugherty, sure and at current prices that would only be about 5 million USD
2146 2011-04-19 19:13:11 <phantomcircuit> except that if you started to buy up large quantities of BTC the price would rocket up
2147 2011-04-19 19:14:38 <JuanDaugherty> https://en.bitcoin.it/wiki/FAQ#What_if_someone_bought_up_all_the_existing_Bitcoins?
2148 2011-04-19 19:15:09 <JuanDaugherty> phantomcircuit, you think so?
2149 2011-04-19 19:15:44 <phantomcircuit> JuanDaugherty, i have no doubt
2150 2011-04-19 19:15:56 <JuanDaugherty> no doubt it would increase
2151 2011-04-19 19:16:28 <phantomcircuit> JuanDaugherty, if you placed a market order for even 10k BTC on mtgox the price would instantly reach 1.5USD
2152 2011-04-19 19:16:43 <JuanDaugherty> but "rocket up", i.e. many times it current rate you mention?
2153 2011-04-19 19:17:27 <JuanDaugherty> people can change their asking price, no?
2154 2011-04-19 19:19:27 <jgarzik> JuanDaugherty: price goes up as supply gets more scarce
2155 2011-04-19 19:20:09 <jgarzik> JuanDaugherty: it's probably impossible to buy all bitcoins
2156 2011-04-19 19:20:13 <phantomcircuit> JuanDaugherty, there's currently 85k USD in asks on mtgox
2157 2011-04-19 19:20:36 <JuanDaugherty> when are all the coins supposed to be generated by?
2158 2011-04-19 19:20:40 <tcatm> 235510.45802222 BTC total asks on all exchanges
2159 2011-04-19 19:20:44 tenach is now known as AFK!~tenach@unaffiliated/tenach|tenach
2160 2011-04-19 19:20:52 <phantomcircuit> so with about 100k USD in purchases you would effectively double the price
2161 2011-04-19 19:21:53 <phantomcircuit> so with 250k or 4.5% of currently generated bitcoins you could effectively double the price
2162 2011-04-19 19:22:00 <JuanDaugherty> i don't follow that phantomcircuit ...
2163 2011-04-19 19:22:35 <JuanDaugherty> what law/mechanism dictates that, what's the formula?
2164 2011-04-19 19:22:44 <phantomcircuit> JuanDaugherty, if you purchased all of the available bitcoins on markets right now you would only have 4.5% of the total BTC market and would have also doubled the asking price
2165 2011-04-19 19:22:52 Lartza has quit (Quit: Lähdössä)
2166 2011-04-19 19:23:04 <JuanDaugherty> you mean the eventual 21 million BTC>
2167 2011-04-19 19:23:08 <JuanDaugherty> s/>/?/
2168 2011-04-19 19:23:08 <phantomcircuit> no
2169 2011-04-19 19:23:16 <phantomcircuit> i mean of what's currently generated
2170 2011-04-19 19:23:19 <phantomcircuit> ;;bc,stats
2171 2011-04-19 19:23:21 <gribble> Current Blocks: 119163 | Current Difficulty: 92347.59095209 | Next Difficulty At Block: 120959 | Next Difficulty In: 1796 blocks | Next Difficulty In About: 1 week, 5 days, 2 hours, 21 minutes, and 12 seconds | Next Difficulty Estimate: 95627.14047157
2172 2011-04-19 19:23:55 <phantomcircuit> JuanDaugherty, right now on exchanges there is 235510 BTC for sale, which is only about 4% of the btc currently generated
2173 2011-04-19 19:24:18 KBme has quit (Remote host closed the connection)
2174 2011-04-19 19:24:34 <phantomcircuit> are you still not following?
2175 2011-04-19 19:24:43 <JuanDaugherty> and the generator own the other 96% ?
2176 2011-04-19 19:24:49 <JuanDaugherty> *generators
2177 2011-04-19 19:24:50 <phantomcircuit> yes
2178 2011-04-19 19:25:18 <phantomcircuit> now of course once you buy ~300k USD in btc the asks on the market will reflect that
2179 2011-04-19 19:25:42 AFK_PHPAdam has quit (Ping timeout: 246 seconds)
2180 2011-04-19 19:27:03 <JuanDaugherty> the bc market has been operating less than a year hasn't it?
2181 2011-04-19 19:27:30 jercos_ is now known as jercos
2182 2011-04-19 19:27:50 <phantomcircuit> JuanDaugherty, there isn't a single market, and btc has been active since 2009
2183 2011-04-19 19:28:18 <JuanDaugherty> well, network, market, pp community, whatever you want to call it
2184 2011-04-19 19:28:27 <JuanDaugherty> *p2p
2185 2011-04-19 19:28:46 <phantomcircuit> those are literally all different things
2186 2011-04-19 19:28:51 KBme has joined
2187 2011-04-19 19:30:35 <JuanDaugherty> you know there should be a trottle control in the GUI of the clients when generating is enabled
2188 2011-04-19 19:30:44 <gjs278> lol
2189 2011-04-19 19:30:54 <gjs278> yeah throttle down 4k mhashes to 3k
2190 2011-04-19 19:30:58 Stellar has joined
2191 2011-04-19 19:31:10 <jgarzik> JuanDaugherty: I have a patch to remove coin generation completely
2192 2011-04-19 19:31:15 <phantomcircuit> gjs278, 4k mhashes would be quite a lot ;)
2193 2011-04-19 19:31:25 <luke-jr> ok so
2194 2011-04-19 19:31:30 <gjs278> I get 4.5 a thread, that shit needs to be throttled
2195 2011-04-19 19:31:42 <JuanDaugherty> I was surprised to see what the cpu utilization was on OSX
2196 2011-04-19 19:31:43 <luke-jr> got the basics working
2197 2011-04-19 19:31:46 <luke-jr> but need help with wx
2198 2011-04-19 19:31:49 <luke-jr> who knows wx?
2199 2011-04-19 19:32:02 <gjs278> I accidentally increased the difficultly to 400k just now, sorry phantomcircuit
2200 2011-04-19 19:32:32 <jgarzik> I wonder how many mining machines would go offline, if that were true
2201 2011-04-19 19:32:35 <JuanDaugherty> ;;bc,stats
2202 2011-04-19 19:32:36 <gribble> Current Blocks: 119165 | Current Difficulty: 92347.59095209 | Next Difficulty At Block: 120959 | Next Difficulty In: 1794 blocks | Next Difficulty In About: 1 week, 5 days, 1 hour, 2 minutes, and 0 seconds | Next Difficulty Estimate: 95870.48282327
2203 2011-04-19 19:33:48 <JuanDaugherty> jgarzik, why do you need that, you have to turn it on anyway?
2204 2011-04-19 19:34:11 <phantomcircuit> JuanDaugherty, there's basically nobody who should be runnign the cpu miner
2205 2011-04-19 19:34:32 <JuanDaugherty> by which you mean turning on generation?
2206 2011-04-19 19:34:54 <lulzplzkthx> Skynet is now self-aware.
2207 2011-04-19 19:35:33 <phantomcircuit> JuanDaugherty, the generation option provided by the default client
2208 2011-04-19 19:35:48 <phantomcircuit> one of the gpu clients should really be included
2209 2011-04-19 19:36:29 eternal1 has joined
2210 2011-04-19 19:36:58 <JuanDaugherty> yes, that's what I thought. Remind me to ask why you say that later.
2211 2011-04-19 19:37:05 <gjs278> yeah but then they'd be stealing my blocks
2212 2011-04-19 19:37:40 <JuanDaugherty> who are "they" gjs278 ?
2213 2011-04-19 19:37:42 <lulzplzkthx> Does anyone know who created tradebitcoin.com?
2214 2011-04-19 19:37:51 <gjs278> the new people who couldn't figure out poclbm
2215 2011-04-19 19:37:58 <JuanDaugherty> ah
2216 2011-04-19 19:38:59 <JuanDaugherty> thx, for info
2217 2011-04-19 19:39:25 JuanDaugherty has left ("Exeunt Channel")
2218 2011-04-19 19:39:50 <sacarlson> lulzplzkthx: well the whois shows Jonathan Jeffus
2219 2011-04-19 19:41:05 <lulzplzkthx> ;;bc,calc 75
2220 2011-04-19 19:41:06 <gribble> The average time to generate a block at 75 Khps, given current difficulty of 92347.59095209 , is 167 years, 36 weeks, 1 day, 7 hours, 34 minutes, and 0 seconds
2221 2011-04-19 19:41:28 <lulzplzkthx> ;;bc,calc 75300
2222 2011-04-19 19:41:28 <gribble> The average time to generate a block at 75300 Khps, given current difficulty of 92347.59095209 , is 8 weeks, 4 days, 23 hours, 8 minutes, and 49 seconds
2223 2011-04-19 19:41:46 AFK_CodePHP has joined
2224 2011-04-19 19:42:04 AFK_CodePHP is now known as CodePHP
2225 2011-04-19 19:45:11 Zarutian has joined
2226 2011-04-19 19:46:55 lyspooner has joined
2227 2011-04-19 19:49:05 <dinox> anyone heard of any working on iphone app for bitcoin yet?
2228 2011-04-19 19:49:23 <lulzplzkthx> Ew, iPhone.
2229 2011-04-19 19:49:25 <lulzplzkthx> Get Android. ;D
2230 2011-04-19 19:49:39 <lulzplzkthx> dinox: Check the forums. You might want to start a bounty for it though.
2231 2011-04-19 19:50:01 Speeder has quit (Quit: Speeder)
2232 2011-04-19 19:50:40 <sipa> there are bounties for android clients
2233 2011-04-19 19:51:18 <topi`> anyone read the "bitcoins collusion problem" post? I wonder if we could estimate the number of discrete nodes participating in mining
2234 2011-04-19 19:51:34 <lulzplzkthx> sipa: I know there are. That's why I suggested one for iPhone.
2235 2011-04-19 19:51:35 <topi`> just to see if all the mining power is under hands of less than 100 people, or not
2236 2011-04-19 19:52:00 <sipa> i think one for iphone will be difficult to get accepted
2237 2011-04-19 19:52:47 <topi`> sipa: definitely :D that's the problem with central power structures, like the App Store
2238 2011-04-19 19:53:19 <sipa> don't get me started about it :)
2239 2011-04-19 19:54:28 <lulzplzkthx> Well, at the very least it could exist for jailbroken devices.
2240 2011-04-19 19:55:42 <dinox> if you dont break any appstore rules it wont be that hard to get it accepted
2241 2011-04-19 19:56:10 <sipa> for one, it needs to execute 'scripts' :)
2242 2011-04-19 19:57:57 <dinox> why? Wouldnt the bitcoin clieant work perfectly well without scripts?
2243 2011-04-19 19:58:13 <gasteve> yes, it would
2244 2011-04-19 19:58:21 <sipa> actually, yes
2245 2011-04-19 19:58:24 agricocb has quit (Quit: Leaving.)
2246 2011-04-19 19:58:36 <gasteve> you don't need the phone actually doing transaction verification
2247 2011-04-19 19:58:54 <sipa> depends what you want to application to do
2248 2011-04-19 19:59:11 <sipa> just being able to spend money would be very useful already
2249 2011-04-19 19:59:23 <dinox> yep
2250 2011-04-19 19:59:40 <dinox> spend/recieve money would be enough
2251 2011-04-19 19:59:57 <dinox> no need for a iphone miner :)
2252 2011-04-19 19:59:59 sethsethseth has joined
2253 2011-04-19 20:00:41 <topi`> the iphone could mine, if the future Apple A4 chips would implement sha2 in hardware :)
2254 2011-04-19 20:01:00 <sipa> maybe it does, we don't know :)
2255 2011-04-19 20:01:01 <dinox> hehe
2256 2011-04-19 20:01:05 <BlueMatt> still be slow compared to gpus
2257 2011-04-19 20:01:12 <gasteve> no thanks, my battery life sucks as it is
2258 2011-04-19 20:01:29 <dinox> sould be like 900mhz
2259 2011-04-19 20:01:37 <dinox> 1 sha2-hash every clock
2260 2011-04-19 20:01:47 <dinox> = 450 MHash/s :)
2261 2011-04-19 20:01:47 <topi`> sipa: I'm thinking of having a "small" bitcoin wallet on the phone, and the larger wallet on your home computer
2262 2011-04-19 20:02:03 <sipa> topi`: my thought exactly
2263 2011-04-19 20:02:03 <BlueMatt> 1 sha256 hash/clock? wtf hardware are you thinking about
2264 2011-04-19 20:02:19 <BlueMatt> or just a remote?
2265 2011-04-19 20:02:34 <edcba> http://timothyblee.com/2011/04/19/bitcoins-collusion-problem/
2266 2011-04-19 20:02:36 <edcba> hqhqhq
2267 2011-04-19 20:02:39 <edcba> hahaha
2268 2011-04-19 20:02:39 <topi`> BlueMatt: dedicated sha2 hardware could put out 1 hash every clock if it was fully pipelined
2269 2011-04-19 20:02:56 <topi`> edcba: don't laugh, those are valid questions.
2270 2011-04-19 20:03:08 <dinox> BlueMatt: Im just speculationg... doesnt ArtFortz hardware do 1 hash per clock?
2271 2011-04-19 20:03:11 <BlueMatt> topi`: and you think that is even remotely worth it for apple to do?
2272 2011-04-19 20:03:15 <sipa> dinox: yes
2273 2011-04-19 20:03:16 <topi`> what if there would be big corps who would build mining rigs, no, mining datacenters?
2274 2011-04-19 20:03:24 <topi`> and they could collude against the rest
2275 2011-04-19 20:03:31 <sipa> in 500 pipeline stages
2276 2011-04-19 20:03:32 <BlueMatt> dinox: and ArtForz's hardware doesnt run @900Mhz
2277 2011-04-19 20:03:38 nguyen has joined
2278 2011-04-19 20:03:38 <topi`> BlueMatt: sha2 has lots of other uses than just bitcoin
2279 2011-04-19 20:03:44 <edcba> big corps "may" be a problem
2280 2011-04-19 20:03:50 <sipa> topi`: but sha256 hardware isn't aimed for high throughput of many parallel hashes, it's aimed at low latency of one hash
2281 2011-04-19 20:03:53 <nguyen> how you view current coin total in wallet on bitcoind?
2282 2011-04-19 20:03:54 <BlueMatt> topi`: yes, but its still not worth it for apple to put it in the processor
2283 2011-04-19 20:04:03 <edcba> but colluding nodes is stupid
2284 2011-04-19 20:04:05 <sipa> nguyen: getbalance or getinfo
2285 2011-04-19 20:04:08 <nguyen> thx
2286 2011-04-19 20:04:24 <topi`> sipa: a valid point.
2287 2011-04-19 20:04:39 <edcba> that's like saying middle class ppl will earn each $1M
2288 2011-04-19 20:04:48 <edcba> now your money worth shit
2289 2011-04-19 20:04:55 <topi`> BlueMatt: the apple processor already has AES in hardware, so why not
2290 2011-04-19 20:05:00 <sipa> in fact, it will never be aimed at doing more than one, which most likely will make a decent cpu faster than it (not in latency but in thoughput)
2291 2011-04-19 20:05:27 nguyen has quit (Quit: Lost terminal)
2292 2011-04-19 20:05:42 <BlueMatt> it kinda defeats the purpose of risc
2293 2011-04-19 20:06:18 B0g4r7 has joined
2294 2011-04-19 20:06:18 <topi`> BlueMatt: the current system-on-a-chip designs, like the Apple A4 or TI OMAP4, are actually very complex beasts that contain lots of different hardware modules
2295 2011-04-19 20:06:53 kite11 has joined
2296 2011-04-19 20:07:02 <dinox> too bad we dont get access to them
2297 2011-04-19 20:07:05 <kite11> "{
2298 2011-04-19 20:07:06 <kite11> :P
2299 2011-04-19 20:07:27 <kite11> wat iz bitcoin
2300 2011-04-19 20:07:43 <ArtForz> I actually wonder why no one is doing pipelined sha or aes accels
2301 2011-04-19 20:07:56 <kite11> ????
2302 2011-04-19 20:07:58 <ArtForz> it would make sense for ssl accels
2303 2011-04-19 20:08:06 <BlueMatt> I know but it is still defeating the purpose. In any case, I guess that's not the point. It's always an evaluation of how much effort you put in vs how much you get out. AES gets a decent amount out because aes is used quite often. However, sha might be used occasionally by some apps but typically not used repetitively or in high volume
2304 2011-04-19 20:08:15 kite11 has quit (Client Quit)
2305 2011-04-19 20:08:23 <topi`> ArtForz: have you had a look how the AES acceleration works in e.g. OMAP3 ?
2306 2011-04-19 20:08:30 <BlueMatt> ArtForz: even though they shouldn't most ssl connections still use md5 :(
2307 2011-04-19 20:08:42 <ArtForz> well, even better
2308 2011-04-19 20:09:01 <topi`> what's wrong with md5?
2309 2011-04-19 20:09:09 <B0g4r7> It's broken.
2310 2011-04-19 20:09:21 <ArtForz> well, not really broken broken
2311 2011-04-19 20:09:28 <netxshare> it's fine if it's salted
2312 2011-04-19 20:09:30 <B0g4r7> Broken enough.
2313 2011-04-19 20:09:40 <ArtForz> depends on the scenario
2314 2011-04-19 20:09:46 <netxshare> mostly broken by rainbow tables
2315 2011-04-19 20:09:49 <BlueMatt> still best to avoid it
2316 2011-04-19 20:09:57 <ArtForz> errr.. no
2317 2011-04-19 20:10:13 <BlueMatt> it is better to use md5 over sha?
2318 2011-04-19 20:10:14 <sipa> rainbow table attacks are related to how it is used
2319 2011-04-19 20:10:24 <dinox> md5 has more problem than rainbowtables
2320 2011-04-19 20:10:32 <netxshare> yeah
2321 2011-04-19 20:10:35 <dinox> the algoritm is not very strong
2322 2011-04-19 20:10:37 <sipa> but md5 is fundamentally broken, in the sense that an attack of less than its hash strength is possible
2323 2011-04-19 20:10:45 <ArtForz> md5 has second preimage collisions
2324 2011-04-19 20:10:52 <dinox> yep
2325 2011-04-19 20:11:00 <ArtForz> and not even one with arbitrary message
2326 2011-04-19 20:11:12 <dinox> what about sha1?
2327 2011-04-19 20:11:22 <ArtForz> same thing, just harder
2328 2011-04-19 20:11:27 <BlueMatt> sha2?
2329 2011-04-19 20:11:30 <ArtForz> sha2 family is probably secure
2330 2011-04-19 20:11:41 <ArtForz> the problem with md5 and sha1 is the linear W update
2331 2011-04-19 20:11:43 <BlueMatt> then why wouldnt you want to use sha2 over md5?
2332 2011-04-19 20:11:51 <ArtForz> because sha2 is slower
2333 2011-04-19 20:12:03 <BlueMatt> by that significant of an amount?
2334 2011-04-19 20:12:15 <dinox> how hard to change bitcoin-mining hash to sha3 or something?
2335 2011-04-19 20:12:24 <BlueMatt> on a side note, can I ask what your day job is ArtForz?
2336 2011-04-19 20:12:27 <ArtForz> dunno, I'd guess about a factor of 2
2337 2011-04-19 20:12:28 <dinox> just an update?
2338 2011-04-19 20:12:41 <BlueMatt> ah, well that would make a difference then
2339 2011-04-19 20:12:54 <ArtForz> day job? EE
2340 2011-04-19 20:13:03 <netxshare> lol
2341 2011-04-19 20:13:17 <dinox> not too hard to guess :P
2342 2011-04-19 20:13:35 <sipa> sha3 doesn't exist yet :)
2343 2011-04-19 20:13:35 <ArtForz> well, more like "whenever I feel like it unless theres a major emergency" job
2344 2011-04-19 20:13:49 <sipa> well, it probably already exists, but it didn't get that name yet
2345 2011-04-19 20:13:57 <BlueMatt> meaning?
2346 2011-04-19 20:14:03 <ArtForz> I don't really have fixed work hours
2347 2011-04-19 20:14:08 <BlueMatt> ah
2348 2011-04-19 20:14:19 <ArtForz> well, as long as projects are finished on time
2349 2011-04-19 20:15:17 <dinox> ArtForz: what is approx. GHash/s for your array of asic's?
2350 2011-04-19 20:15:30 <ArtForz> 19.2
2351 2011-04-19 20:15:36 <dinox> hehe
2352 2011-04-19 20:15:41 <dinox> nice job
2353 2011-04-19 20:15:47 <sipa> ArtForz: he asked approximately, not exactly!
2354 2011-04-19 20:15:53 agricocb has joined
2355 2011-04-19 20:15:56 <dinox> lol
2356 2011-04-19 20:15:57 <ArtForz> well, it is approximately
2357 2011-04-19 20:16:16 <netxshare> does it have a display on it?
2358 2011-04-19 20:16:21 <ArtForz> nope
2359 2011-04-19 20:16:23 <dinox> will it repay
2360 2011-04-19 20:16:30 <dinox> ? :P
2361 2011-04-19 20:16:36 <ArtForz> dunno, so far it looks like yes
2362 2011-04-19 20:16:40 <sipa> and does it "piouw piouw peng peng sssssssh" when it found a block?
2363 2011-04-19 20:17:02 <dinox> leds all over it which flashes
2364 2011-04-19 20:17:14 <dinox> hehe
2365 2011-04-19 20:17:22 <dinox> do you have any pics?
2366 2011-04-19 20:17:25 <ArtForz> nope
2367 2011-04-19 20:17:45 <ArtForz> well, I can easily hear when somethign is not working
2368 2011-04-19 20:18:06 <netxshare> loud?
2369 2011-04-19 20:18:07 <dinox> smell the smoke atleast :P
2370 2011-04-19 20:18:13 <ArtForz> 5 80x38mm 12W fans per box normally running at ~40%
2371 2011-04-19 20:18:21 <netxshare> how hot does it get?
2372 2011-04-19 20:18:35 <dinox> how many boxes?
2373 2011-04-19 20:18:39 <ArtForz> 3
2374 2011-04-19 20:19:04 <dinox> ah, how many asic's then?
2375 2011-04-19 20:19:09 <ArtForz> 96
2376 2011-04-19 20:19:11 <ArtForz> 3*32
2377 2011-04-19 20:19:17 <dinox> nice
2378 2011-04-19 20:19:22 DukeOfURL has quit (Quit: ChatZilla 0.9.86.1 [Firefox 3.6.16/20110319135224])
2379 2011-04-19 20:19:30 <ArtForz> well, the heatsinks barely hit 40°C, but thanks to crappy cavity-up plastic BGA thats already within 10°C of Tjmax
2380 2011-04-19 20:19:59 <dinox> would be quiet a bit of dollars there
2381 2011-04-19 20:20:13 <netxshare> so do you use your computer to request a hash or is it done over network?
2382 2011-04-19 20:20:17 <dinox> but with 19.2GH... it will probably repay
2383 2011-04-19 20:20:42 <ArtForz> about $75k including physical synthesis and mask costs
2384 2011-04-19 20:21:34 <dinox> you must be rich to afford that
2385 2011-04-19 20:21:39 <ArtForz> not really
2386 2011-04-19 20:22:00 <netxshare> do your computers still make a profit from running?
2387 2011-04-19 20:22:02 <ArtForz> yeah
2388 2011-04-19 20:22:21 <netxshare> so it's not to late to buy some and run miners?
2389 2011-04-19 20:22:27 <ArtForz> dunno
2390 2011-04-19 20:22:38 <netxshare> well I am about to find out
2391 2011-04-19 20:22:45 <ArtForz> I just like the long-term aspect of my approach
2392 2011-04-19 20:22:46 <BlueMatt> of course ArtForz is rich
2393 2011-04-19 20:23:03 <netxshare> I just like the whole bitcoin idea
2394 2011-04-19 20:23:07 <dinox> or he will be
2395 2011-04-19 20:23:12 <netxshare> if it pays off the cost ill be happy
2396 2011-04-19 20:23:18 <dinox> when 1 btc = 1k$
2397 2011-04-19 20:23:55 <sethsethseth> wait why does it cost $75k to make 19ghash?
2398 2011-04-19 20:23:56 <netxshare> tho, I might have some people in china backing me
2399 2011-04-19 20:24:10 <ArtForz> dev costs
2400 2011-04-19 20:24:43 <sipa> netxshare: i started mining januari 19th
2401 2011-04-19 20:24:48 <netxshare> did it take you a while to get sha2 running on it well?
2402 2011-04-19 20:24:48 <sipa> today i broke even
2403 2011-04-19 20:24:49 <ArtForz> cost per Mh for further small-scale runs is < $1.50/Mh
2404 2011-04-19 20:24:57 <dinox> how much power to drive that ArtForz ?
2405 2011-04-19 20:25:07 <netxshare> sipa what are you running
2406 2011-04-19 20:25:09 <ArtForz> typically less than 300W per 2U
2407 2011-04-19 20:25:11 <sipa> 2x 5970
2408 2011-04-19 20:25:20 <ArtForz> at 100% fanspeed close to 320W
2409 2011-04-19 20:25:25 <netxshare> buy them at retail?
2410 2011-04-19 20:25:34 <dinox> 3 x 320W then?
2411 2011-04-19 20:25:37 <ArtForz> yeah
2412 2011-04-19 20:25:56 <dinox> 19.2GHash at 1kW is really good
2413 2011-04-19 20:26:02 <ArtForz> and yes, thats > 20Mh/J
2414 2011-04-19 20:26:16 <netxshare> that's not bad sipa
2415 2011-04-19 20:26:24 <netxshare> what miner are you running?
2416 2011-04-19 20:26:31 <ArtForz> well, it still sucks compared to a real ASIC on a modern process
2417 2011-04-19 20:26:37 fimp has quit (Quit: This computer has gone to sleep)
2418 2011-04-19 20:26:46 <sethsethseth> i like sipa's idea of a client alert when a block is found
2419 2011-04-19 20:27:02 <sethsethseth> i want some kind of firework to go off on my desktop
2420 2011-04-19 20:27:07 <netxshare> lol
2421 2011-04-19 20:27:13 <ArtForz> but the mining market is way too small to recoup investment on that
2422 2011-04-19 20:27:17 <netxshare> there is not a way to tell?
2423 2011-04-19 20:27:23 <luke-jr> so it seems wx goes nuts if you try to do things from more than one thread
2424 2011-04-19 20:27:26 <dinox> ArtForz: you should make a diy-set for sale :)
2425 2011-04-19 20:27:31 <luke-jr> what would be a sane way to trigger ThreadA from ThreadB?
2426 2011-04-19 20:27:40 <B0g4r7> Wasn't that the original idea?
2427 2011-04-19 20:27:42 <sipa> luke-jr: condition variable?
2428 2011-04-19 20:27:43 <luke-jr> where ThreadA is a wx event loop
2429 2011-04-19 20:27:49 <B0g4r7> "a diy-set for sale"
2430 2011-04-19 20:28:22 <dinox> would probably sell well
2431 2011-04-19 20:28:30 <netxshare> I am guessing it was much more work to put sha2 on the ASIC then the ATI shader
2432 2011-04-19 20:28:50 <ArtForz> yeah, I'm currently working on making the thing easier to build and smaller
2433 2011-04-19 20:28:51 <dinox> and we get rid of those powerhungry gpu's...
2434 2011-04-19 20:29:08 <ArtForz> next target is 32 chips in a 1U
2435 2011-04-19 20:29:09 <gasteve> so, each of those boxes is 2U, taking 320W and computing 6.4Gh/s?
2436 2011-04-19 20:29:11 <BlueMatt> netxshare: not really, but you can do a ton of optimization on bitcoin mining which helps a ton on hash/s
2437 2011-04-19 20:29:18 <ArtForz> yep
2438 2011-04-19 20:29:34 trifon_ has joined
2439 2011-04-19 20:29:47 <gasteve> that's awesome...I hope you do sell them in some capacity...I'd love to learn about FPGA programming and such
2440 2011-04-19 20:30:13 <ArtForz> a decent FPGA could do about the same perf/W, but costs about 5x as much
2441 2011-04-19 20:30:23 <sethsethseth> wouldn't it be super easy to have the client open a sound file when a block is found? going to post a bounty on this....
2442 2011-04-19 20:30:35 <B0g4r7> Yeah, I'd buy one of your units just for the neat factor and to learn from.
2443 2011-04-19 20:30:43 <ArtForz> well, or about 2.5x if you go with spartan6s
2444 2011-04-19 20:31:04 <gasteve> oh, so this isn't an FPGA?
2445 2011-04-19 20:31:07 <ArtForz> nope
2446 2011-04-19 20:31:15 <BlueMatt> sethsethseth: a simple script should be able to look through the output of a miner
2447 2011-04-19 20:31:16 <dinox> this is real deal
2448 2011-04-19 20:31:20 <ArtForz> not quite
2449 2011-04-19 20:31:23 <ArtForz> structured ASIC
2450 2011-04-19 20:31:25 <B0g4r7> sethsethse: What about a helper process that queries bitcoind, and itself plays the sound when a block is found?
2451 2011-04-19 20:31:38 <ArtForz> basically like altera hardcopy, just on a older process and cheaper
2452 2011-04-19 20:31:40 DukeOfURL has joined
2453 2011-04-19 20:31:47 <B0g4r7> To me that sounds like a cleaner solution.
2454 2011-04-19 20:31:53 <dinox> i can make that in about 3 min
2455 2011-04-19 20:32:03 <dinox> or anyone probably can
2456 2011-04-19 20:32:17 <B0g4r7> Really, the less cruft that gets added to bitcoind itself the better IMO.
2457 2011-04-19 20:32:26 <netxshare> can't you just edit the code
2458 2011-04-19 20:32:35 <netxshare> and have it do whatever you want
2459 2011-04-19 20:32:38 <B0g4r7> I actually think that the entire GUI should live outside of bitcoind.
2460 2011-04-19 20:32:50 <dinox> ArtForz: Some prototype asic?
2461 2011-04-19 20:32:53 trifon has quit (Ping timeout: 260 seconds)
2462 2011-04-19 20:32:57 <BlueMatt> B0g4r7: you realize bitcoind has no gui
2463 2011-04-19 20:32:59 <ArtForz> structured ASIC
2464 2011-04-19 20:33:03 <BlueMatt> and no wx deps?
2465 2011-04-19 20:33:26 <dinox> gotta search that on google
2466 2011-04-19 20:33:27 <phantomcircuit> sethsethseth, how much?
2467 2011-04-19 20:33:30 <B0g4r7> ...
2468 2011-04-19 20:33:34 <ArtForz> fixed silicon and lower metal layers, custom routing in upper metal layers
2469 2011-04-19 20:33:44 <luke-jr> BlueMatt: but it does have glib deps
2470 2011-04-19 20:33:55 <luke-jr> B0g4r7: https://en.bitcoin.it/wiki/Wallet_protocol
2471 2011-04-19 20:34:00 <netxshare> I rather have a cli client myself
2472 2011-04-19 20:34:09 <sethsethseth> i'll go 10btc right now since it only takes 3min:)
2473 2011-04-19 20:34:11 <BlueMatt> netxshare: the run bitcoind
2474 2011-04-19 20:34:22 <luke-jr> BlueMatt: that's not a CLI client, just a JSON-RPC server
2475 2011-04-19 20:34:33 <BlueMatt> and bitcoin is a cli client
2476 2011-04-19 20:34:33 <netxshare> and I dislike json-rpc
2477 2011-04-19 20:34:36 <luke-jr> BlueMatt: tcatm is planning to actually make a real CLI client
2478 2011-04-19 20:34:38 <B0g4r7> So there's bitcoin.exe, and there's bitcoind...
2479 2011-04-19 20:34:42 <phantomcircuit> sethsethseth, windows or linux
2480 2011-04-19 20:34:42 <BlueMatt> nice
2481 2011-04-19 20:34:45 <ArtForz> that way you just need a single metal mask instead of the > dozen masks required for a completely custom chip
2482 2011-04-19 20:34:46 <luke-jr> B0g4r7: there's also Spesmilo âº
2483 2011-04-19 20:34:51 <BlueMatt> B0g4r7: there is a bitcoind.exe in another folder
2484 2011-04-19 20:34:59 <BlueMatt> oh...sorry
2485 2011-04-19 20:35:10 <dinox> ArtForz: AH, its like asic lite :)
2486 2011-04-19 20:35:12 <B0g4r7> So bitcoind does not include any of the GUI code whatsoever?
2487 2011-04-19 20:35:12 <BlueMatt> B0g4r7: yea pretty much just bitcoin mainline for now
2488 2011-04-19 20:35:13 <sethsethseth> windows
2489 2011-04-19 20:35:14 <ArtForz> yep
2490 2011-04-19 20:35:20 Teslah has joined
2491 2011-04-19 20:35:24 <luke-jr> B0g4r7: just glib/gtk stuff
2492 2011-04-19 20:35:27 <B0g4r7> Somehow I thought it was more of a runtime switch.
2493 2011-04-19 20:35:27 <phantomcircuit> sethsethseth, that would be more than 3 minutes
2494 2011-04-19 20:35:27 <luke-jr> nothing wx though
2495 2011-04-19 20:35:40 <ArtForz> base silicon is basically a huge sea of LUT+FF, carry chains, ram blocks, I/O cells, like a FPGA, just without the routing matrix
2496 2011-04-19 20:35:41 <B0g4r7> Well that's good.
2497 2011-04-19 20:35:46 <phantomcircuit> sethsethseth, and do you mean when a new block is received from the network or when you generate a new block?
2498 2011-04-19 20:35:51 <gasteve> what does one of those boxes cost?
2499 2011-04-19 20:35:55 <ArtForz> and the routing matrix takes up about 2/3 of the chip area on a FPGA
2500 2011-04-19 20:35:58 <dinox> how much would a real asic cost, if you order 1000?
2501 2011-04-19 20:36:00 <sethsethseth> generate
2502 2011-04-19 20:36:07 <ArtForz> way too much
2503 2011-04-19 20:36:17 <B0g4r7> What I'd like to see if bitcoin.exe containing only the GUI, and none of bitcoind.
2504 2011-04-19 20:36:19 <dinox> 1M$+?
2505 2011-04-19 20:36:22 <B0g4r7> For full seperation.
2506 2011-04-19 20:36:24 <gasteve> (I meant one of your 2U boxes that does 6.4Gh/s)
2507 2011-04-19 20:36:30 <sipa> B0g4r7: many people would like that
2508 2011-04-19 20:36:31 Stellar has quit (Remote host closed the connection)
2509 2011-04-19 20:36:31 <dinox> 10M$?
2510 2011-04-19 20:36:34 <ArtForz> a single mask set at 45nm is ~4M
2511 2011-04-19 20:36:47 <dinox> hehe
2512 2011-04-19 20:36:52 <ArtForz> so assuming you need one respin... about 10M
2513 2011-04-19 20:36:53 <dinox> thats not cheap
2514 2011-04-19 20:36:59 <phantomcircuit> ArtForz, and why is it so expensive?
2515 2011-04-19 20:37:12 <BlueMatt> B0g4r7: I think just about everyone likes that, just no one has the time to program it (there are better things to do to promote bitcoin)
2516 2011-04-19 20:37:14 <phantomcircuit> <--- literally has no idea
2517 2011-04-19 20:37:33 <ArtForz> gasteve: about $10k
2518 2011-04-19 20:37:35 <B0g4r7> heh
2519 2011-04-19 20:37:40 <luke-jr> B0g4r7: Spesmilo is just a GUI, that connects to bitcoind
2520 2011-04-19 20:38:05 <B0g4r7> Hm, I'll have to check it out.
2521 2011-04-19 20:38:31 <netxshare> AcceptBlock()
2522 2011-04-19 20:38:41 <netxshare> err
2523 2011-04-19 20:38:53 <luke-jr> RejectBlock(netxshare)
2524 2011-04-19 20:38:57 <ArtForz> v2 should lower that a bit, reduced support part and PCB costs
2525 2011-04-19 20:39:00 <BlueMatt> B0g4r7: full disclosure, luke-jr is the dev of spesmilo
2526 2011-04-19 20:39:04 <luke-jr> BlueMatt: a dev*
2527 2011-04-19 20:39:13 <BlueMatt> excuse me, one of 2
2528 2011-04-19 20:39:15 <B0g4r7> Yeah, I was seeing that on the page.
2529 2011-04-19 20:39:37 <luke-jr> BlueMatt: one of 4 or 5, if you give credit where it's due
2530 2011-04-19 20:39:51 <B0g4r7> I lol'd when I saw Tonal support in the featureset. :)
2531 2011-04-19 20:39:54 <BlueMatt> well the biggest chunk of code is luke's
2532 2011-04-19 20:40:00 <luke-jr> BlueMatt: genjix's
2533 2011-04-19 20:40:01 noagendamarket has joined
2534 2011-04-19 20:40:07 <BlueMatt> B0g4r7: yea, luke-jr loves his tonal
2535 2011-04-19 20:40:13 <BlueMatt> luke-jr: ok after genjix, you
2536 2011-04-19 20:40:16 <luke-jr> :p
2537 2011-04-19 20:40:37 <BlueMatt> then a bunch of small chunks which are fairly small
2538 2011-04-19 20:40:51 <luke-jr> I made it work on Windows btw
2539 2011-04-19 20:40:54 <luke-jr> no fancy installer tho
2540 2011-04-19 20:41:19 stonetz has joined
2541 2011-04-19 20:41:19 <luke-jr> â¦yet! :P
2542 2011-04-19 20:42:44 <BlueMatt> would be awesome to have a fancy installer which installs bitcoind and then someone work to split off wxui and write the wallet protocol
2543 2011-04-19 20:42:49 <BlueMatt> though that is a lot of work
2544 2011-04-19 20:43:51 <sethsethseth> phantom: maybe make a thread and recruit more donations for it
2545 2011-04-19 20:44:51 <gasteve> ArtForz: have you considered licensing the IP and working with several others to build and res ell the boxes? if it's possible to get the price to $3-4k for a 1U system, I imagine they'd sell pretty well and it would be good to get this specialized hardware into a diverse group of miners (good for the integrity of bitcoin I mean)
2546 2011-04-19 20:45:01 zhalox has quit (Read error: Connection reset by peer)
2547 2011-04-19 20:45:31 marlowe has quit (Quit: leaving)
2548 2011-04-19 20:45:44 zhalox has joined
2549 2011-04-19 20:45:52 <ArtForz> kinda hard when the chips alone are >$5k even in decent volume
2550 2011-04-19 20:46:23 <gasteve> oh...I though you said there were like 32 in each
2551 2011-04-19 20:46:30 <ArtForz> there are
2552 2011-04-19 20:46:47 <gasteve> it wouldn't be possible to build a system with fewer of them?
2553 2011-04-19 20:46:56 <ArtForz> sure would, but whats the point?
2554 2011-04-19 20:47:22 <gasteve> to build a machine that takes up less space and has a lower entry point
2555 2011-04-19 20:47:25 <BlueMatt> and you said you weren't rich, yet you spent >300k on mining rigs
2556 2011-04-19 20:47:27 <gasteve> (entry price point)
2557 2011-04-19 20:47:29 skyewm has quit (Ping timeout: 276 seconds)
2558 2011-04-19 20:47:42 <ArtForz> BlueMatt: way less
2559 2011-04-19 20:47:59 <BlueMatt> 32/rack I thought you had several racks?
2560 2011-04-19 20:48:13 <ArtForz> ?
2561 2011-04-19 20:48:27 <BlueMatt> oh $5k per rack not per asic
2562 2011-04-19 20:48:29 <ArtForz> gasteve: v2 is already supposed to put 32 chips in 1u, maybe 40
2563 2011-04-19 20:48:52 <ArtForz> so ... why build less dense boxes?
2564 2011-04-19 20:49:20 <gasteve> ok...but the point is not maximizing the space...it's making it so that people wouldn't have to pay $10k minimum to buy one
2565 2011-04-19 20:49:22 <johnlockwood> what is the site that is like ebay, but with bitcoins
2566 2011-04-19 20:49:32 <BlueMatt> bidding pond
2567 2011-04-19 20:49:39 nguyen has joined
2568 2011-04-19 20:49:49 <ArtForz> thats gonna make shit even more expensive per Mh#
2569 2011-04-19 20:49:50 <gasteve> (and if the box could be made such that more of the s-asics could be added later, even better)
2570 2011-04-19 20:49:50 <johnlockwood> ah, thanks
2571 2011-04-19 20:50:15 <nguyen> why does poclbm only list one adapter when aticonfig lists two! how frustrating run my damn gpu!
2572 2011-04-19 20:50:27 <ArtForz> well, I'm still working on another "fun" chassis design for V2
2573 2011-04-19 20:50:55 <gasteve> yes, I know...but if you had 8 chips, that would be 1.6Gh/s right ...that's faster and more efficient than a dual 5970 that takes up 4U
2574 2011-04-19 20:51:02 <ArtForz> yep
2575 2011-04-19 20:51:11 <gasteve> well, I would buy that!
2576 2011-04-19 20:51:11 <ArtForz> I was thinking "desktop" 16-chip unit
2577 2011-04-19 20:51:59 <ArtForz> 2 modules, single 80mm fan, external +12V supply, USB to host PC
2578 2011-04-19 20:52:40 <gasteve> 2 modules meaning you could build a 1U box with one module (of 16 asics) and add a second module later if you wanted?
2579 2011-04-19 20:52:56 <ArtForz> one module = 8 asics
2580 2011-04-19 20:53:08 <TD_> i don't think anyone is going to buy ASIC based miners unless they are serious and have some real capital to spend
2581 2011-04-19 20:53:19 <gasteve> ah...and you could buy a 1U machine with one module and add a second later?
2582 2011-04-19 20:53:20 <TD_> so minimizing the cost doesn't seem like a great use of time
2583 2011-04-19 20:53:37 <BlueMatt> TD_: there are many people who are serious about mining
2584 2011-04-19 20:53:40 <ArtForz> TD_: actually it kinda is
2585 2011-04-19 20:53:57 <BlueMatt> plus if bitcoin ever takes off the more miners the better
2586 2011-04-19 20:54:02 <netxshare> this thing looks sweet
2587 2011-04-19 20:54:02 <netxshare> http://www.copacobana.org/photos/photo_1.jpg
2588 2011-04-19 20:54:12 <netxshare> http://www.copacobana.org/photos/photo_7.jpg
2589 2011-04-19 20:54:15 <netxshare> FPGA dimm cards
2590 2011-04-19 20:54:31 <ArtForz> yeah, yawn
2591 2011-04-19 20:54:43 <TD_> i won't call a miner "serious" until they build their own datacenter connected to their own solar or wind farm ;)
2592 2011-04-19 20:54:54 <ArtForz> that design is flawed on so many levels
2593 2011-04-19 20:55:13 <BlueMatt> If I were a bank/mybitcoin-style site/etc I would want to run a 1u rack just to give a tiny bit of insurance to the network
2594 2011-04-19 20:55:32 <netxshare> I just like the fact they were in dimm
2595 2011-04-19 20:56:00 <TD_> i doubt anyone will sink a rack of ASICs just out of altruism, it'd be for profit
2596 2011-04-19 20:56:05 <BlueMatt> ArtForz is like Diablo-D3 except for hardware :)
2597 2011-04-19 20:56:17 TD has quit (Disconnected by services)
2598 2011-04-19 20:56:18 TD_ is now known as TD
2599 2011-04-19 20:56:31 <phantomcircuit> TD, solar/wind is expensive
2600 2011-04-19 20:56:39 <TD> there was an interesting discussion inside google today about how the economics of mining will play out as it transitions to fees
2601 2011-04-19 20:56:43 TDX_ has joined
2602 2011-04-19 20:56:45 <phantomcircuit> even if you're using it 24/7
2603 2011-04-19 20:56:47 <ArtForz> thing is, who cares about increased pwoer efficiency if it takes *years* to make back the difference in initial cost vs. GPUs?
2604 2011-04-19 20:57:03 <netxshare> http://picocomputing.com/sc_series.html 80k
2605 2011-04-19 20:57:05 <gjs278> at a datacenter, use all of the power you want
2606 2011-04-19 20:57:07 <BlueMatt> TD: If I were a bank/mybitcoin site/etc I would want some tiny amount of insurance (or maybe try to get my (client's) txes in blocks faster)
2607 2011-04-19 20:57:15 <ArtForz> gjs278: err... no.
2608 2011-04-19 20:57:19 <TD> BlueMatt: why do that rather than just pay fees?
2609 2011-04-19 20:57:25 <gjs278> you clearly odn't have a big enough contract then
2610 2011-04-19 20:57:44 <TD> mining and running a bank are pretty different businesses
2611 2011-04-19 20:57:48 <ArtForz> you clearly have no clue
2612 2011-04-19 20:57:51 <BlueMatt> TD: I doubt mining will always be profitable, in fact, I doubt it will continue to be profitable for very long (as long as the price doesnt keep going up like this, Im betting on correction eventually)
2613 2011-04-19 20:57:54 <gjs278> lol ok
2614 2011-04-19 20:58:03 <gjs278> I don't manage the bill for the level 3 contract we have at work I guess
2615 2011-04-19 20:58:08 <TD> anyway, one guy believes the market will fail over the long run because mining is a tragedy of the commons
2616 2011-04-19 20:58:31 <gasteve> on the topic of datacenters...I wonder what the economics of colocation would be (I used to colo some servers years ago, but haven't looked into it recently) ...colo might be an unnecessary luxury for mining though...it's not like you need high availability of a mining rig...if one goes down, you fix it and put it back in service...clean the dust off every now and then, etc
2617 2011-04-19 20:58:56 <TD> in a purely fee based network, there'll always be some miner who will include low fee transactions. as fees drop and miners stop hashing because it's no longer profitable, difficulty drops and the miners that are left are once again engaged in a race to the bottom
2618 2011-04-19 20:58:58 <BlueMatt> TD: not mining for profit, but just mining to keep the network secure, Im sure an IT guy could convince their superior that it is part of the cost of running a bitcoin business ;)
2619 2011-04-19 20:58:59 <netxshare> would it not just be cheaper to rent a office space
2620 2011-04-19 20:59:03 <netxshare> and get a connection
2621 2011-04-19 20:59:18 <gasteve> yes, probably so
2622 2011-04-19 20:59:23 <netxshare> tho I guess it depends
2623 2011-04-19 20:59:25 lyspooner has quit (Quit: ChatZilla 0.9.86.1 [Firefox 3.6.16/20110319135224])
2624 2011-04-19 20:59:29 <gasteve> or better yet, warehouse space
2625 2011-04-19 20:59:34 <TD> because network difficulty is a shared resource. a few miners might do it for altruistic reasons, but you'd not be able to get significant difficulty that way
2626 2011-04-19 20:59:34 <netxshare> yeah that's what I was thinkiing
2627 2011-04-19 20:59:39 <BlueMatt> I doubt you would have almost any advantage of coloing for mining
2628 2011-04-19 21:00:02 tabsa_ has joined
2629 2011-04-19 21:00:04 <netxshare> if you had 1 server connected to the net for bitcoin and the rest running miners on a localnetwork
2630 2011-04-19 21:00:07 <BlueMatt> TD: you would have to, IMHO mining will never be profitable in the long run if bitcoin actually does take off
2631 2011-04-19 21:00:10 <netxshare> that should cut down on traffic
2632 2011-04-19 21:00:35 <TD> mining should be at least a little profitable. if it's not bitcoin will certainly fail.
2633 2011-04-19 21:00:41 <gasteve> BlueMatt: mining will always be profitable
2634 2011-04-19 21:00:52 <gasteve> just not very profitable
2635 2011-04-19 21:00:59 Teslah has quit (Quit: Leaving)
2636 2011-04-19 21:01:24 <gasteve> I am interested in hearing that "tragedy of the commons" theory though
2637 2011-04-19 21:01:27 <BlueMatt> gasteve: IMHO for that to be, the price of bitcoins will go up so high that it will present a barrier to adoption
2638 2011-04-19 21:01:33 <TD> but that was the point my colleague was making. what incentive do non-miner participants have to encourage a high difficulty, other than a vague sense of altruism and "for the greater good" type thinking?
2639 2011-04-19 21:01:52 <TD> answer: none at all. merchants and traders would be perfectly happy with a difficulty of 1, except for the insecurity of it.
2640 2011-04-19 21:01:55 <BlueMatt> also,
2641 2011-04-19 21:01:58 <BlueMatt> http://www.bitcoin.org/smf/index.php?topic=5320.0
2642 2011-04-19 21:02:03 <TD> but how do you decide what the "right" difficulty is?
2643 2011-04-19 21:02:12 <netxshare> bitcoin4gold.com!
2644 2011-04-19 21:03:11 tabsa has quit (Ping timeout: 250 seconds)
2645 2011-04-19 21:03:12 tabsa_ is now known as tabsa
2646 2011-04-19 21:03:26 <BlueMatt> TD: IMHO if bitcoin actually does take off, there will be enough altruism or people who invest in large farms powered by green energy (or some other energy that is a one-time-cost) that the difficulty will be plenty to prevent attacks
2647 2011-04-19 21:03:34 <TD> i really doubt that
2648 2011-04-19 21:04:02 <phantomcircuit> BlueMatt, green energy is not a one time cost :|
2649 2011-04-19 21:04:02 <TD> difficulty would never be very high by relying on volunteers alone
2650 2011-04-19 21:04:21 <TD> it's kind of like saying the web will always have search engines even if one has to be run by volunteers
2651 2011-04-19 21:04:23 <BlueMatt> if bitcoin does actually take off, would the fdic not decide it should insure bitcoins? or at least some other variation on it from some other country?
2652 2011-04-19 21:04:28 <TD> sure, it's true to a certain extent, but the results would be far inferior
2653 2011-04-19 21:04:31 gat3way has quit (Read error: Operation timed out)
2654 2011-04-19 21:04:48 <gasteve> all you need is sufficient fee based transaction volume to ensure a decent payout for mining
2655 2011-04-19 21:04:57 <TD> no. that was the guys point.
2656 2011-04-19 21:05:01 <TD> fees will trend to zero
2657 2011-04-19 21:05:10 <BlueMatt> TD: true, but IMHO there would be enough to keep the difficulty high enough that attacks aren't a problem
2658 2011-04-19 21:05:19 <gasteve> when the volume of transactions gets high enough, people will pay to clear transactions quickly
2659 2011-04-19 21:05:20 <BlueMatt> and I highly disagree that fees will trend to zero
2660 2011-04-19 21:05:32 <TD> why should they? the only reason you might do that today is due to artificial block size limits
2661 2011-04-19 21:05:47 <TD> but those limits need to go away in the long run for scalability
2662 2011-04-19 21:06:01 <gasteve> if you're a merchant, you will want to make sure your transaction get embedded in the block chain very quickly
2663 2011-04-19 21:06:14 <TD> in such a world, the only reason a TX would not make it into the current block, is if a miner deliberately chooses to exclude it because there are insufficient fees
2664 2011-04-19 21:06:21 <gasteve> and they will
2665 2011-04-19 21:06:35 <gasteve> (some already are I think)
2666 2011-04-19 21:06:54 <TD> let's say all pending transactions add up to a fee of only 10 BTC. but you as a miner need to make at least 20 BTC for it to be worth the electricity costs of mining
2667 2011-04-19 21:06:59 <BlueMatt> I guess I disagree on what "bitcoin has made it" means, IMHO bitcoin will never replace any currency as a country's backbone but will remain as a method of online payments ie digital goods etc
2668 2011-04-19 21:07:00 <TD> so you just don't work on this block
2669 2011-04-19 21:07:17 <TD> your electricity costs to take part would be higher than the expected chance of reward
2670 2011-04-19 21:07:34 <TD> now this keeps happening. users are annoyed because their transactions confirm slowly ... but it's a temporary annoyance
2671 2011-04-19 21:07:44 <TD> after only two weeks the network adapts to the fact that you are only mining occasionally
2672 2011-04-19 21:07:51 <gasteve> TD: you simply sit idle until you collect enough transactions that it's worthwhile to start working on a block
2673 2011-04-19 21:08:03 <TD> now users are happy because blocks are rolling in after 10 minutes again, and your mining power has been permanently lost
2674 2011-04-19 21:08:08 <TD> yes, i said that :)
2675 2011-04-19 21:08:23 CodePHP is now known as AFK_PHPAdam
2676 2011-04-19 21:08:42 <gasteve> I don't get the part about "your mining power has been permanantly lost"?
2677 2011-04-19 21:09:03 <BlueMatt> gasteve: people have stopped mining, and because no one is paying a fee, they have stopped for good
2678 2011-04-19 21:09:08 <TD> users aren't paying high enough fees for you to be profitable. instead of users increasing their fees to keep the difficulty at the same level, the difficulty simply drops
2679 2011-04-19 21:09:24 <TD> and users don't care! all they know is they didn't have to pay any fees and the slowness sorted itself out
2680 2011-04-19 21:09:29 <gasteve> if difficulty drops, then so does the cost of electricity to mine...so the equation/threshold you require to start mining is lowered
2681 2011-04-19 21:09:31 <TD> difficulty is something that isn't their specific responsibility
2682 2011-04-19 21:09:32 <phantomcircuit> TD, after what 2 weeks?
2683 2011-04-19 21:09:43 gat3way has joined
2684 2011-04-19 21:10:11 <TD> phantomcircuit: difficulty adjustment
2685 2011-04-19 21:10:22 gavinandresen has quit (Quit: gavinandresen)
2686 2011-04-19 21:10:31 <TD> gasteve: no your electricity costs are always the same. if you start mining again, the difficulty will rise. but you can't start mining again because there aren't sufficient fees.
2687 2011-04-19 21:10:44 <TD> i'm talking about a world where only fees are funding miners
2688 2011-04-19 21:10:47 <TD> after the inflation is finished
2689 2011-04-19 21:10:57 ByteCoin has joined
2690 2011-04-19 21:10:59 <TD> so the per block reward isn't relevant .... only the sum of fees attached to transactions
2691 2011-04-19 21:11:35 eternal1 has quit (Quit: Leaving.)
2692 2011-04-19 21:11:46 <gasteve> that's circular logic...if difficulty falls, mining is cheaper (less electricity required to mine a block)...so, you don't require as much in the way of transaction fees to make it profitable to mine again
2693 2011-04-19 21:12:51 <BlueMatt> I think at that point, some kind of group would decide they need to insure the bitcoin network (IMHO if bitcoin is actually widely used enough for this problem to occur, some govt org or group of banks who operate in btc would decide there is too big a risk in the system and either insure bitcoin or stop using them) If they stop using them well then yes, you have a point and the bitcoin network has failed. However, IMHO bitcoin will never be b
2694 2011-04-19 21:12:51 <BlueMatt> ig enough for this to be a problem
2695 2011-04-19 21:12:53 <TD> it's recursive. difficulty falls because you stopped mining, as it wasn't profitable. but why should users pay fees now? they learned that they can just pay less and tough it out for two weeks, then transactions confirm nice and quickly again.
2696 2011-04-19 21:13:38 <BlueMatt> it will always be small enough that altruistic miners can provide enough insurance (or miners who dont pay much/anything for power, etc)
2697 2011-04-19 21:13:48 <TD> ultimately, from the perspective of a selfish user, paying no fees at all works fine. it means nearly no PoW strength in the end but the system still appears to work in the same way. and why should I pay huge fees to make super high difficulties profitable, when my neighbour can pay nothing and get the benefit?
2698 2011-04-19 21:14:02 <gasteve> well, what if miners stopped processing free transactions altogether? they could after all
2699 2011-04-19 21:14:20 <B0g4r7> Suppose network power suddenly dropped by 95%.
2700 2011-04-19 21:14:23 <sethsethseth> i made a post in the projects forum to recruit more donations for "play sound upon block generation"
2701 2011-04-19 21:14:31 <JFK911> sethsethseth: good idea
2702 2011-04-19 21:14:32 <B0g4r7> Would it then take like half a year for difficulty to readjust?
2703 2011-04-19 21:14:34 <JFK911> cash register sound
2704 2011-04-19 21:14:50 <sethsethseth> yes!
2705 2011-04-19 21:15:05 <B0g4r7> The one from Pink Floyd's Money?
2706 2011-04-19 21:15:10 <tcatm> sethsethseth: IIRC Hal has such a patch
2707 2011-04-19 21:15:14 AFK_PHPAdam is now known as CodePHP
2708 2011-04-19 21:15:43 <TD> yes, they could pick a minimum fee level and exclude any transactions that don't meet that level. more likely they'd be very low fee transactions. but the cost of including a 0.01 fee transaction is zero to somebody who is already mining, so why not stick them in and claim the profits no matter how small and poxy?
2709 2011-04-19 21:16:01 <TD> B0g4r7: yes potentially.
2710 2011-04-19 21:16:17 <gasteve> TD: because of exactly your argument above
2711 2011-04-19 21:16:28 <B0g4r7> That sounds like an easy way to sabotage the network for someone with lots of resources.
2712 2011-04-19 21:16:41 <gasteve> if they just allow low or no fee transaction, then they undermine their own pricing power
2713 2011-04-19 21:16:52 <tcatm> couldn't clients enforce fees (just like they do now) by not relaying transactions without enough fees?
2714 2011-04-19 21:17:01 glassresistor has quit (Ping timeout: 246 seconds)
2715 2011-04-19 21:17:39 <TD> gasteve: it wouldn't be you allowing the low fee transactions, it'd be a competitor. if you aren't mining because there aren't enough high fee transactions, not much stops your competitor from stepping in and taking them. even if they mine less aggressively it doesn't matter, the network will eventually adapt and they will take the profits.
2716 2011-04-19 21:18:13 <gasteve> I think clients should enforce a time limit on processing of no free transactions...maybe after 24 hours, clients require that blocks include those no fee transactions or they don't get accepted (I think that's done already, but much shorter than 24h)
2717 2011-04-19 21:18:39 <gasteve> TD: yep...but that just means my competitor is beating me
2718 2011-04-19 21:18:40 <B0g4r7> I guess a transactor could look to recent blocks to see the level of fees in the included transactions to know how much of a fee to include to have their transaction be processed.
2719 2011-04-19 21:18:42 <ByteCoin> What's the name of the IRC room which shows transasctions and new blocks as they happen?
2720 2011-04-19 21:18:51 <tcatm> ByteCoin: #bitcoin-watch
2721 2011-04-19 21:18:58 <ByteCoin> thanks
2722 2011-04-19 21:19:05 <phantomcircuit> TD, if (average cost to mine a block) < ( minted value + combined tx value) then mine()
2723 2011-04-19 21:19:11 <phantomcircuit> TD, seems simple enough to me
2724 2011-04-19 21:19:17 <gasteve> (i.e. my competitor, for one reason or another is able to accept those low or no fee transactions and stay in business)
2725 2011-04-19 21:19:38 <BlueMatt> phantomcircuit: his point is that means eventually not enough mining power to ensure the network is secure
2726 2011-04-19 21:19:55 <phantomcircuit> BlueMatt, seems unlikely
2727 2011-04-19 21:20:09 <gasteve> but clients are actually the ones that ensure the network is secure, not miners
2728 2011-04-19 21:20:32 <BlueMatt> phantomcircuit: if you follow his argument, it isnt really that unlikely. In fact, assuming there are no altruistic miners, he is completely right
2729 2011-04-19 21:20:34 <phantomcircuit> gasteve, uhhhhh
2730 2011-04-19 21:20:39 <TD> my point is really, users collectively set fees, but it's impossible to value the difficulty of the network. how hard should the PoW be?
2731 2011-04-19 21:20:50 robotarmy has joined
2732 2011-04-19 21:20:58 <TD> right now difficulty is essentially a function of USD exchange rate and current inflation
2733 2011-04-19 21:20:58 <phantomcircuit> BlueMatt, id be assuming that you have competition
2734 2011-04-19 21:21:05 <gasteve> (mining horse power could be brought to bear on the network of clients, yes, but it's the clients that set the rules on what is and isn't a good block)
2735 2011-04-19 21:21:21 <B0g4r7> What if Goldman Sachs started mining?
2736 2011-04-19 21:21:28 <BlueMatt> TD: it is a good point, to put it differently, mining has many positive externalities that aren't taken into account when fees are set
2737 2011-04-19 21:21:32 <B0g4r7> Along with their buddies.
2738 2011-04-19 21:21:38 <TD> with no inflation, it's up to users. but they have strong incentives to provide either no fees or the absolute lowest fee possible. somebody will try and pick up those fees even if it's just a one guy with one CPU
2739 2011-04-19 21:22:04 <gasteve> at best, a powerful group of miner can disrupt the network
2740 2011-04-19 21:22:05 <TD> as long as the network users can stand slowness for a while they can basically force fees lower and lower
2741 2011-04-19 21:22:09 <BlueMatt> I just disagree that it will happen as I think some people will gladly take a loss to mine
2742 2011-04-19 21:22:30 <phantomcircuit> TD, i dont think most people would trade free transactions for waiting 2 weeks
2743 2011-04-19 21:22:42 <BlueMatt> phantomcircuit: not his point
2744 2011-04-19 21:22:58 <CIA-89> bitcoin: Luke Dashjr bitcoinuri * r80febb56f8b8 bitcoind-personal/makefile.unix: Merge commit 'f0b7952c' into bitcoinuri http://tinyurl.com/3qqz8mb
2745 2011-04-19 21:22:59 <CIA-89> bitcoin: Luke Dashjr bitcoinuri * rad1ee3274a7f bitcoind-personal/ (ui.cpp ui.h): Allow passing a default amount to CSendDialog constructor http://tinyurl.com/3ppgtqm
2746 2011-04-19 21:23:01 <TD> phantomcircuit: they wouldn't have to wait 2 weeks for very long. if hardly anyone was mining anymore because there were no fees, the network would adapt and then they'd have to wait 10 mins again
2747 2011-04-19 21:23:01 <phantomcircuit> BlueMatt, then i dont get it
2748 2011-04-19 21:23:02 <CIA-89> bitcoin: Luke Dashjr bitcoinuri * r6590bc8b94a5 bitcoind-personal/ (makefile.unix wxipcclient.cpp wxipcserver.cpp): Boost-based IPC for bitcoin: URI support http://tinyurl.com/3mesqnu
2749 2011-04-19 21:23:30 <BlueMatt> phantomcircuit: I doubt it would ever get near 2 weeks, maybe 20 min/block at max
2750 2011-04-19 21:23:36 <BlueMatt> the network/economy doesnt move that fast
2751 2011-04-19 21:23:39 <luke-jr> hrm, I didn't mean to do it that way
2752 2011-04-19 21:24:22 <TD> anyway this was the guys argument
2753 2011-04-19 21:24:33 <gasteve> BlueMatt: about taking a loss to mine...consider merchants...and if bitcoin has gown sufficiently large, the merchants would have a strong incentive to ensure the integrity of the system...and if that meant subsidizing mining operations, I think they would certainly do that
2754 2011-04-19 21:24:42 <TD> that you have a shared resource that everyone benefits from (difficulty) but it's in each actors interest to minimize their investment in it
2755 2011-04-19 21:24:47 <TD> i'm not sure his argument holds
2756 2011-04-19 21:25:06 <gasteve> TD: it's a good thing to think about...but I'm not convinced it's an issue
2757 2011-04-19 21:25:24 <BlueMatt> gasteve: I dont know about normal merchants maybe some mining, but IMHO someone big like govt or group of banks would come in and mine because they have more money than they know what to do with
2758 2011-04-19 21:25:33 echelon has quit (Quit: Leaving)
2759 2011-04-19 21:25:45 <BlueMatt> I mean normal merchants will probably mine a tiny bit, just not much
2760 2011-04-19 21:25:48 <gasteve> BlueMatt: subsidizing doesn't mean they have to operate it themselves
2761 2011-04-19 21:26:08 <TD> because theoretically miners could charge whatever the market can bear. if inflation goes from 50 btc to 25 btc, but you're still making a little bit of profit regardless (rather than a lot), you can just say "i only include txns that have a 1 btc fee" and continue mining. it doesn't matter that hardly anyone includes a 1btc fee because you can still make enough off inflation to pay your own costs. so eventually people would give up waiting
2762 2011-04-19 21:26:08 <TD> start paying the fees
2763 2011-04-19 21:26:08 <BlueMatt> oh you mean pay into a large miner, well they might
2764 2011-04-19 21:26:20 <gasteve> (they could contribute a monthly fee to a miner who will always prioritize their transactions ahead of others)
2765 2011-04-19 21:26:25 <TD> it should stabilize out at the maximum people are willing to pay in fees for a given acceptance time
2766 2011-04-19 21:26:27 <luke-jr> gasteve: subsidizing mining = fees
2767 2011-04-19 21:26:34 <BlueMatt> I could defiantly see that happening
2768 2011-04-19 21:26:36 <TD> but that only works if there's a smooth transition from inflation to fees
2769 2011-04-19 21:26:43 <BlueMatt> luke-jr: he means subsidation in the opposite direction
2770 2011-04-19 21:26:52 <luke-jr> BlueMatt: that makes no sense
2771 2011-04-19 21:26:56 <gasteve> luke-jr: right...the easiest way to subsidize would be to pay fees ;)
2772 2011-04-19 21:26:57 echelon has joined
2773 2011-04-19 21:27:06 <BlueMatt> luke-jr: he means merchants paying for someone like Art to mine
2774 2011-04-19 21:27:13 <luke-jr> BlueMatt: that's fees
2775 2011-04-19 21:27:24 <BlueMatt> luke-jr: they are receiving money not paying
2776 2011-04-19 21:27:32 <luke-jr> â¦
2777 2011-04-19 21:27:35 <TD> yes you can solve the tragedy of the commons by assuming actors don't act purely in rational self interest
2778 2011-04-19 21:27:35 <luke-jr> no
2779 2011-04-19 21:27:38 <luke-jr> fees = merchant pays
2780 2011-04-19 21:27:50 <TD> it does happen. but it seems better to have a system that works with self interest ...
2781 2011-04-19 21:27:54 <luke-jr> oh
2782 2011-04-19 21:28:00 <luke-jr> you mean receiver paying instead of sender?
2783 2011-04-19 21:28:07 <luke-jr> that's doable with fees too
2784 2011-04-19 21:28:23 <edcba> 23:24 < TD> yes you can solve the tragedy of the commons by assuming actors don't act purely in rational self interest
2785 2011-04-19 21:28:30 <luke-jr> just take your pending Output, and use it as an Input
2786 2011-04-19 21:28:33 <edcba> it's quite refuting it
2787 2011-04-19 21:28:34 <gasteve> but, I could also see some scenario where a merchant pays a miner a monthly fee to basically back stop their transactions buy guaranteed they'll get into a block in a certain amount of time...the transactions could even be sent directly to that miner
2788 2011-04-19 21:29:11 <TD> gasteve: why have a specific arrangement like that instead of just automatically adjusting fees until you get the acceptance time you want?
2789 2011-04-19 21:29:24 <luke-jr> exactly
2790 2011-04-19 21:29:34 <gasteve> TD: you just answered your own question ;)
2791 2011-04-19 21:29:44 <luke-jr> when you see TransactionToMe, create a new Transaction using its output as your input, and pay a fee on that one
2792 2011-04-19 21:30:04 BlueMatt has quit (Read error: Connection reset by peer)
2793 2011-04-19 21:30:21 <CIA-89> bitcoin: Luke Dashjr bitcoinuri * rd88d78aa8ad6 bitcoind-personal/ (ui.cpp ui.h): Allow passing a default amount to CSendDialog constructor http://tinyurl.com/3pmog9f
2794 2011-04-19 21:30:22 <CIA-89> bitcoin: Luke Dashjr bitcoinuri * r7eb88f0cb6af bitcoind-personal/ (makefile.unix wxipcclient.cpp wxipcserver.cpp): Boost-based IPC for bitcoin: URI support http://tinyurl.com/3z4d86u
2795 2011-04-19 21:30:46 BlueMatt has joined
2796 2011-04-19 21:31:54 <gasteve> if I'm a miner and I make a guarantee to keep enough horsepower to produce a block at least once every say 6 hours...I could charge merchants a fee...they could send transactions directly to me (in addition to broadcast)...I could provide some additional double spend protections...and I could guarantee they get into a block that I mine within 6 hours if they don't get picked up by another miner...I might even roll that into some kind of instant
2797 2011-04-19 21:31:54 <gasteve> clearing/double spend insurance product
2798 2011-04-19 21:31:56 <BlueMatt> luke-jr: no I meant a merchant pays a miner (more realistically a big group of merchants) who then get some small incentive like prioritized txes
2799 2011-04-19 21:32:04 CodePHP is now known as AFK_PHPAdam
2800 2011-04-19 21:32:34 <luke-jr> sirius: bounty code: https://www.bitcoin.org/smf/index.php?topic=5171.msg89908#msg89908
2801 2011-04-19 21:32:35 <BlueMatt> and the merchant's it dept passes it off as a cost of doing business in the btc economy
2802 2011-04-19 21:32:40 <BlueMatt> ie the cost of safety of the network
2803 2011-04-19 21:32:52 <luke-jr> BlueMatt: that's exactly what fees are
2804 2011-04-19 21:33:09 <BlueMatt> luke-jr: not if you are receiving the payments
2805 2011-04-19 21:33:14 <luke-jr> BlueMatt: yes, even then
2806 2011-04-19 21:33:19 <luke-jr> BlueMatt: you can add a fee to it
2807 2011-04-19 21:33:26 AFK_PHPAdam is now known as CodePHP
2808 2011-04-19 21:33:28 blablaa has joined
2809 2011-04-19 21:33:29 <BlueMatt> how do i pay a miner when I am receiving a payment?
2810 2011-04-19 21:33:33 <gasteve> luke-jr: the current transaction fee mechanism is one (goo d) way to handle it...bu t I don't think it's necessarily the only way
2811 2011-04-19 21:33:34 <BlueMatt> I can add a fee to the spend
2812 2011-04-19 21:33:43 <luke-jr> BlueMatt: use the Output as as Input in a change+fee-only tx
2813 2011-04-19 21:34:02 <luke-jr> your fee txn now depends on the payment
2814 2011-04-19 21:34:07 <BlueMatt> luke-jr: that doesnt prioritize the initial tx
2815 2011-04-19 21:34:09 <luke-jr> it should
2816 2011-04-19 21:34:12 <BlueMatt> oh
2817 2011-04-19 21:34:23 <luke-jr> IIRC it does, with the original client
2818 2011-04-19 21:34:28 <BlueMatt> I see, yea but that complicates the matter quite a bit
2819 2011-04-19 21:34:29 <luke-jr> not 100% sure on that, but if not, it should be fixed ;P
2820 2011-04-19 21:35:09 <BlueMatt> still simpler for a merchant to pay someone else to deal with that crap
2821 2011-04-19 21:35:10 <gasteve> it would hardly make sense to have a transaction in a block precede one of its input transactions
2822 2011-04-19 21:35:33 <gasteve> yeah, simplicity is worth something
2823 2011-04-19 21:36:18 <BlueMatt> esp if you consider that most merchants will probably simply be using a paypal-like processor which might mine to provide some protection for its customers (ie can reasonably insure its funds) and provide faster txes for its customers
2824 2011-04-19 21:36:38 <gasteve> I could easily see a merchant paying say 30btc/month to a miner that just deals with all of this and lets them just deal in plain, no fee transactions
2825 2011-04-19 21:37:12 <luke-jr> gasteve: it wouldn't precede it
2826 2011-04-19 21:37:30 <gasteve> luke-jr: that's my point
2827 2011-04-19 21:38:04 <gasteve> if you picked up a fee transaction, you would have to include any non-free (but as yet unmined) transactions that were it's inputs in order to have a valid block
2828 2011-04-19 21:38:30 <luke-jr> right, that's why it should work
2829 2011-04-19 21:38:45 <luke-jr> anyone else care to add onto sirius's bounty? https://www.bitcoin.org/smf/index.php?topic=5171.msg89908#msg89908
2830 2011-04-19 21:38:52 mmoya has joined
2831 2011-04-19 21:39:11 <luke-jr> I spent basically all day on it, which I'd usually charge no less than 200 USD for :P
2832 2011-04-19 21:39:19 <BlueMatt> luke-jr: wait, did you add bitcoin: uri handling in the main cilent???
2833 2011-04-19 21:39:23 <luke-jr> BlueMatt: yes
2834 2011-04-19 21:39:40 <BlueMatt> you want a pull req made? and post an address for a donate if you do please ;)
2835 2011-04-19 21:39:46 <luke-jr> (on the other hand, it took all day because I wasn't familiar with boost/wx APIs, so maybe I deserve less ;)
2836 2011-04-19 21:39:58 <edcba> luke-jr: does it work on all OS ?
2837 2011-04-19 21:40:05 <luke-jr> edcba: good question, care to test?
2838 2011-04-19 21:40:10 <luke-jr> it *should* afaik
2839 2011-04-19 21:40:15 <luke-jr> just boost and wx
2840 2011-04-19 21:40:40 <BlueMatt> testing now :)
2841 2011-04-19 21:40:53 <luke-jr> 19ut7h2sp9jKf5dpnK36FCPGu8L1cHnPSE to donate for it ;)
2842 2011-04-19 21:41:00 phantomcircuit has quit (Remote host closed the connection)
2843 2011-04-19 21:41:03 <luke-jr> when it hits 50 BTC (sirius's bounty), I'll MIT license it
2844 2011-04-19 21:41:06 <BlueMatt> what branch?
2845 2011-04-19 21:41:15 <BlueMatt> oh uri my bad
2846 2011-04-19 21:41:19 * BlueMatt is blind sometimes
2847 2011-04-19 21:41:28 <luke-jr> git pull git://scm.dashjr.org/var/scmroot/git/bitcoin/bitcoind/luke-jr.git bitcoinuri
2848 2011-04-19 21:41:54 <luke-jr> DO merge it, as I intentionally based it on the oldest compatible revision :p
2849 2011-04-19 21:41:57 <edcba> seems there are some security issues
2850 2011-04-19 21:42:05 <luke-jr> edcba: ?
2851 2011-04-19 21:42:25 <edcba> minor
2852 2011-04-19 21:42:49 <BlueMatt> wait why are there now 2 binaries?
2853 2011-04-19 21:43:02 noagendamarket has quit (Ping timeout: 248 seconds)
2854 2011-04-19 21:43:03 <edcba> oh maybe not
2855 2011-04-19 21:43:12 <luke-jr> BlueMatt: so you don't need to load the whole bitcoin binary every time you click a link
2856 2011-04-19 21:43:13 <edcba> how does tokenizer handle duplicates ?
2857 2011-04-19 21:43:24 <luke-jr> edcba: it's just like Perl's split func
2858 2011-04-19 21:43:33 <edcba> that i doesn't know either :p
2859 2011-04-19 21:43:56 <luke-jr> edcba: it just splits a string into pieces
2860 2011-04-19 21:43:56 <BlueMatt> luke-jr: so bitcoin is a "server" and the the "client" is made to handle bitcoin: links which then get passed to bitcoin-prope?
2861 2011-04-19 21:43:56 <edcba> but i think it's not good
2862 2011-04-19 21:44:00 <BlueMatt> r
2863 2011-04-19 21:44:14 <luke-jr> BlueMatt: right
2864 2011-04-19 21:44:23 <luke-jr> BlueMatt: but the server/client is just Boost IPC
2865 2011-04-19 21:44:24 <BlueMatt> interesting approach, but ok
2866 2011-04-19 21:44:24 <edcba> ie bitcoin:1ecz....zee?amount=12&amount=10000
2867 2011-04-19 21:44:34 <luke-jr> edcba: it would use the last one
2868 2011-04-19 21:44:41 <luke-jr> edcba: but both values there are invalid
2869 2011-04-19 21:44:43 <edcba> in your implementation
2870 2011-04-19 21:44:46 <luke-jr> you need X0 or X8
2871 2011-04-19 21:45:14 <edcba> the problem is to be consistent accross systems
2872 2011-04-19 21:45:35 <edcba> better to warn in our case
2873 2011-04-19 21:45:37 <luke-jr> edcba: it's not a problem⦠URIs that don't conform to the spec mostly have undefined behaviourâ¦
2874 2011-04-19 21:45:46 <luke-jr> in all cases, the user must click send
2875 2011-04-19 21:46:07 <BlueMatt> luke-jr: can we get address checking before opening the dialog?
2876 2011-04-19 21:46:17 <luke-jr> BlueMatt: hm, maybe
2877 2011-04-19 21:46:22 <luke-jr> but do what if it's invalid?
2878 2011-04-19 21:46:38 <edcba> i said it was minor because it *may* cause problems in future
2879 2011-04-19 21:46:48 <BlueMatt> I'd say either nothing or popup saying you clicked an invalid link
2880 2011-04-19 21:47:00 <luke-jr> BlueMatt: then the user can't even change it if they want toâ¦
2881 2011-04-19 21:47:03 <BlueMatt> also, why does uri not open bitcoin when called?
2882 2011-04-19 21:47:09 <BlueMatt> luke-jr: ok good point, forget it
2883 2011-04-19 21:47:12 <luke-jr> BlueMatt: it assumes bitcoin is always running
2884 2011-04-19 21:47:18 <edcba> anyway for now it's a really good news to have bitcoin uri support
2885 2011-04-19 21:47:25 <BlueMatt> well that needs fixed, but otherwise looks really cool
2886 2011-04-19 21:47:30 <edcba> you'll make a lot of websites happy :)
2887 2011-04-19 21:47:50 <BlueMatt> are you going to work on opening bitcoin or should I take a look at it and learn all about wx and boost tomorrow?
2888 2011-04-19 21:48:18 <edcba> BlueMatt: give him some bitcoins to motivate him :p
2889 2011-04-19 21:48:21 <netxshare> so the uri was added?
2890 2011-04-19 21:48:32 <edcba> not yet lol
2891 2011-04-19 21:48:35 <netxshare> oh
2892 2011-04-19 21:48:40 <edcba> but it's in good way now
2893 2011-04-19 21:48:47 <netxshare> cool
2894 2011-04-19 21:48:59 <edcba> it's almost developped thanx to luke-jr
2895 2011-04-19 21:49:27 <netxshare> bleh
2896 2011-04-19 21:49:33 <netxshare> I hate thinking of domain names
2897 2011-04-19 21:49:36 <luke-jr> BlueMatt: do people not leave bitcoin running?
2898 2011-04-19 21:49:40 <edcba> nope
2899 2011-04-19 21:49:45 <luke-jr> hmm
2900 2011-04-19 21:49:47 <netxshare> why not?
2901 2011-04-19 21:49:47 <BlueMatt> luke-jr: people shouldnt have to
2902 2011-04-19 21:49:59 <edcba> because it doesn't restart automatically because it uses same port than vmware...
2903 2011-04-19 21:50:11 <BlueMatt> because it shouldnt sit around and eat resources when you aren't using it
2904 2011-04-19 21:50:29 <netxshare> really? you need them that much?
2905 2011-04-19 21:50:36 <BlueMatt> one should be able to open it when they want to send/receive money, other than that it doesnt need to be running 24/7
2906 2011-04-19 21:50:40 <edcba> it should eat some resources imo
2907 2011-04-19 21:50:52 KBme has quit (Remote host closed the connection)
2908 2011-04-19 21:50:58 <netxshare> you mean you can't close it out?
2909 2011-04-19 21:51:11 <edcba> that would assure bitcoin not being taken over by small ressources
2910 2011-04-19 21:51:14 <BlueMatt> netxshare: you can, but that doesnt mean it should start up at startup
2911 2011-04-19 21:51:23 <luke-jr> BlueMatt: I'll think about it if the donations yield enough to cover my time on this part ;)
2912 2011-04-19 21:51:23 <netxshare> oh okay
2913 2011-04-19 21:51:26 <BlueMatt> edcba: in no way the same thing
2914 2011-04-19 21:51:37 <netxshare> yeah, I thought it was odd it started up on it's own
2915 2011-04-19 21:51:38 <gasteve> well, we really need to have a separate client daemon from the wallet, etc ...it should be encouraged for people to keep a client running, but keeping your wallet online all the time might not be something you'd want to do (especially since the wallet.dat isn't currently encrypted)
2916 2011-04-19 21:51:47 <tcatm> auto starting bitcoin could be tricky. what if the user uses a different datadir?
2917 2011-04-19 21:51:55 <gasteve> (it should be encouraged to keep a client running because it helps the network)
2918 2011-04-19 21:51:56 <edcba> oh yes i think a separate binary to quickly handle the uri is near
2919 2011-04-19 21:51:57 <edcba> neat
2920 2011-04-19 21:52:23 <tcatm> bitcoinuri already complains about bitcoin not running so the user can start it and click the URI again
2921 2011-04-19 21:52:25 <BlueMatt> gasteve: how does keeping a client open help the network?
2922 2011-04-19 21:52:50 <gasteve> your client validates transactions and blocks and forwards them to other peers
2923 2011-04-19 21:52:53 <netxshare> yeah
2924 2011-04-19 21:53:05 <BlueMatt> gasteve: but if you are offline, less people have to forward txes to you
2925 2011-04-19 21:53:10 <BlueMatt> ie less total network bw
2926 2011-04-19 21:53:20 <BlueMatt> which is better
2927 2011-04-19 21:53:29 <edcba> and network is easier to dupe
2928 2011-04-19 21:53:39 <gasteve> it's better to have more peers in the network
2929 2011-04-19 21:53:55 <tcatm> luke-jr: let's try it. what's your bitcoin-uri with a reasonable amount to donate?
2930 2011-04-19 21:53:56 <netxshare> I would think it would spread faster with more people on
2931 2011-04-19 21:53:58 <BlueMatt> edcba: again, not the same thing that is mining not open nodes
2932 2011-04-19 21:54:10 <BlueMatt> gasteve: no its not, how is it better?
2933 2011-04-19 21:54:27 <gasteve> (and, for most people, you only have 8 peers you connect to...so it's not that much more traffic)
2934 2011-04-19 21:54:48 <BlueMatt> netxshare: if its a smaller network, it would spread faster. More people just means you have to forward everything to more people
2935 2011-04-19 21:55:05 KBme has joined
2936 2011-04-19 21:55:05 <BlueMatt> gasteve: no its not much more bw, but that doesnt mean its better for the network to have more nodes
2937 2011-04-19 21:55:09 <luke-jr> tcatm: bitcoin:19ut7h2sp9jKf5dpnK36FCPGu8L1cHnPSE?amount=1X8
2938 2011-04-19 21:55:10 topace has quit (Ping timeout: 246 seconds)
2939 2011-04-19 21:55:13 topace has joined
2940 2011-04-19 21:55:14 <netxshare> don't you only forward it to the 8 people you are connected to?
2941 2011-04-19 21:55:17 <BlueMatt> more nodes with open ports, yes, more nodes in general no
2942 2011-04-19 21:55:24 <edcba> what is 1X8 ?
2943 2011-04-19 21:55:25 <luke-jr> tcatm: feel free to change the amount
2944 2011-04-19 21:55:28 <gasteve> BlueMatt: it's better because more peers means more nodes validating transactions and blocks...which means less vulnerability to someone taking over the network and instituting their own acceptance rules
2945 2011-04-19 21:55:32 <luke-jr> edcba: 1 BTC
2946 2011-04-19 21:55:36 <tcatm> seems to work :)
2947 2011-04-19 21:55:47 <luke-jr> anyone want to do bitcoin:19ut7h2sp9jKf5dpnK36FCPGu8L1cHnPSE?amount=1X10
2948 2011-04-19 21:55:48 <BlueMatt> gasteve: the network will never be small enough for that to be a problem
2949 2011-04-19 21:55:49 <luke-jr> :p
2950 2011-04-19 21:56:03 <gasteve> really?
2951 2011-04-19 21:56:12 <edcba> 1x0 is smallest unit ?
2952 2011-04-19 21:56:16 <BlueMatt> gasteve: that would mean sybil attack which has been discussed at length
2953 2011-04-19 21:56:16 <luke-jr> edcba: yes
2954 2011-04-19 21:56:23 <gasteve> people seem awfully worried about such attacks
2955 2011-04-19 21:56:48 <BlueMatt> gasteve: it is possible, but that is a question of how many connections you have, not how many nodes there are
2956 2011-04-19 21:56:53 <gasteve> so, what if no one ran the client on a regular basis? or what if only Gavin did?
2957 2011-04-19 21:57:05 <BlueMatt> gasteve: then yes it would be a huge issue
2958 2011-04-19 21:57:09 <gasteve> ;)
2959 2011-04-19 21:57:19 <netxshare> I think there is a reason they had it autostart
2960 2011-04-19 21:57:36 <BlueMatt> but just having windows users not have their nodes open is not a problem, there will always be enough people who have their nodes open
2961 2011-04-19 21:57:46 <BlueMatt> like me who runs a bitcoin server for js-ui from anywhere
2962 2011-04-19 21:57:51 <BlueMatt> or a merchant
2963 2011-04-19 21:57:52 <BlueMatt> or a miner
2964 2011-04-19 21:58:13 <BlueMatt> there are plenty who run nodes 24/7 and people who hardly use bitcoin dont need to
2965 2011-04-19 21:58:15 <gasteve> I keep a client running 24x7 (with nothing in its wallet!)
2966 2011-04-19 21:58:21 <BlueMatt> why?
2967 2011-04-19 21:58:37 noagendamarket has joined
2968 2011-04-19 21:58:37 <gasteve> well, I guess, why not
2969 2011-04-19 21:58:44 * tcatm has a few bitcoinds running, too. mostly from testing stuff
2970 2011-04-19 21:58:57 * luke-jr has at least 2 running 24/7
2971 2011-04-19 21:58:58 <BlueMatt> I have no problem with that, it just has no use to you
2972 2011-04-19 21:59:13 <luke-jr> one for -watch (and hopefully a pool soon), and one for myself
2973 2011-04-19 21:59:40 <gasteve> (compared to the electricity I'm burning by mining, the client is nothing)
2974 2011-04-19 21:59:41 <BlueMatt> as a side note, if you have a thin client it is better to not be on the network 24/7 as the ratio of full-to-thin clients is important for network ddos-ability
2975 2011-04-19 22:00:06 <BlueMatt> gasteve: for most people it isnt a problem, Im just saying for my grandmother its better to just not add more crap to that already-slow computer
2976 2011-04-19 22:00:24 <gasteve> BlueMatt, you could just connect through a single, other peer and avoid that issue I'd think
2977 2011-04-19 22:00:42 <gasteve> BlueMatt: sure
2978 2011-04-19 22:01:06 <netxshare> I think there should be a site that allows users to receive bitcoins with out running any software
2979 2011-04-19 22:01:07 <BlueMatt> gasteve: but thin clients will forward any valid-looking tx, full clients actually do some checking
2980 2011-04-19 22:01:11 <netxshare> they can just simply login
2981 2011-04-19 22:01:18 <netxshare> and see their balance
2982 2011-04-19 22:01:19 <BlueMatt> mybitcoin.com
2983 2011-04-19 22:01:24 <netxshare> well there you goin
2984 2011-04-19 22:01:27 <netxshare> perfect
2985 2011-04-19 22:01:34 <netxshare> then your grandma has no problem
2986 2011-04-19 22:01:34 <netxshare> :D
2987 2011-04-19 22:01:34 <BlueMatt> and I agree that that will be the future for most people
2988 2011-04-19 22:01:45 <netxshare> yeah
2989 2011-04-19 22:01:52 <netxshare> something more user friendly and paypalish
2990 2011-04-19 22:02:05 <BlueMatt> what ever I was just pointing out there is no reason why bitcoin needs to be run 24/7, it adds nothing to the network
2991 2011-04-19 22:02:21 sharperguy has joined
2992 2011-04-19 22:02:26 LaoShanLong has joined
2993 2011-04-19 22:02:32 <BlueMatt> netxshare: I agree, especially because if users use one site and the merchant they are paying does as well, no confirmation waiting :)
2994 2011-04-19 22:02:36 <gasteve> so, how many full time, full clients do you need for good robustness of the network (for availability and as a defense against any sort of attack)
2995 2011-04-19 22:02:38 <gasteve> ?
2996 2011-04-19 22:03:14 <BlueMatt> gasteve: you mean prevent sybil or ddos?
2997 2011-04-19 22:03:40 <gasteve> either I suppose...and anything else one could imagine
2998 2011-04-19 22:03:53 <netxshare> yeah
2999 2011-04-19 22:03:58 <netxshare> setting up a api
3000 2011-04-19 22:04:01 echelon has quit (Remote host closed the connection)
3001 2011-04-19 22:04:10 <netxshare> should be easy to write a simple site like that
3002 2011-04-19 22:04:18 <netxshare> then throw on a credit card processor
3003 2011-04-19 22:04:25 <netxshare> to convert monies to BTC
3004 2011-04-19 22:04:43 * tcatm is working on such a site (based on js-remote)
3005 2011-04-19 22:04:53 <netxshare> nooo
3006 2011-04-19 22:04:55 TD has quit (Quit: TD)
3007 2011-04-19 22:05:04 * luke-jr likes how he implemented bitcoin URIs without really touching the bitcoin client code at all
3008 2011-04-19 22:05:05 <netxshare> it's not even worth trying to make anything
3009 2011-04-19 22:05:33 <tcatm> sure it is
3010 2011-04-19 22:05:37 echelon has joined
3011 2011-04-19 22:05:41 <netxshare> just go buy bitpay.me
3012 2011-04-19 22:06:26 <netxshare> I have a few other ideas
3013 2011-04-19 22:06:44 <netxshare> but I am betting they will be up and running too
3014 2011-04-19 22:06:56 <netxshare> by the time I have a chance to start mine
3015 2011-04-19 22:07:45 <luke-jr> BlueMatt: actually, your change might be trivial. discuss which behaviour would be preferred on forums ;)
3016 2011-04-19 22:08:20 <netxshare> is there any way to clear the Transactions tab
3017 2011-04-19 22:08:38 <netxshare> I rather have a recent tab
3018 2011-04-19 22:08:41 <netxshare> instead of a all
3019 2011-04-19 22:08:58 <BlueMatt> for sybil: depends on the number of outgoing connections people make from their clients. Currently, an attacker would need ~783 nodes for the P of connecting exclusively to the attacker to be .1% (probably a good baseline). Because many of the current 1857 connection-accepting clients will always be up no matter what sybil isnt an issue (they are probably mostly people like me, luke, etc who will always keep their nodes open)
3020 2011-04-19 22:08:58 <netxshare> that only displays the trans from the current running then the rest shows them all
3021 2011-04-19 22:09:04 <luke-jr> BlueMatt: reality is, if bitcoin were to crash with the current code, bitcoinuri would silently "do nothing" and when the user opens bitcoin again, they would get send popups for all the ones pending
3022 2011-04-19 22:09:05 <BlueMatt> ddos is an entirely different issue
3023 2011-04-19 22:09:36 <BlueMatt> luke-jr: sidenote, currently am getting no errors when bitcoin isnt running
3024 2011-04-19 22:09:46 <BlueMatt> if nothing else bitcoinuri client should say "bitcoin isnt running"
3025 2011-04-19 22:09:55 <BlueMatt> which your code apparently does? but isnt working for me
3026 2011-04-19 22:10:02 <luke-jr> BlueMatt: did you close bitcoin cleanly?
3027 2011-04-19 22:10:05 <JFK911> ;;bc,mtgox
3028 2011-04-19 22:10:05 <gribble> {"ticker":{"high":1.1979,"low":1.1237,"vol":14158,"buy":1.1928,"sell":1.1979,"last":1.1979}}
3029 2011-04-19 22:10:06 <BlueMatt> luke-jr: yes
3030 2011-04-19 22:10:12 <luke-jr> BlueMatt: strange, what OS?
3031 2011-04-19 22:10:19 <BlueMatt> luke-jr: linux
3032 2011-04-19 22:10:24 <luke-jr> that's a kernel
3033 2011-04-19 22:10:34 Pander has quit (Remote host closed the connection)
3034 2011-04-19 22:10:44 <BlueMatt> hm, no now it is working...well I thought i closed it cleanly
3035 2011-04-19 22:10:46 <BlueMatt> must not have
3036 2011-04-19 22:10:48 <luke-jr> XD
3037 2011-04-19 22:11:09 <BlueMatt> as a side note, why does Cntl-C not close bitcoin cleanly?
3038 2011-04-19 22:11:13 <BlueMatt> it really should
3039 2011-04-19 22:11:13 <luke-jr> btw, is BTC seriously about to hit 1.20?
3040 2011-04-19 22:11:17 <JFK911> yes
3041 2011-04-19 22:11:23 <BlueMatt> luke-jr: yesterday high was 1.199
3042 2011-04-19 22:11:24 <JFK911> then its going to DUMP
3043 2011-04-19 22:11:26 <luke-jr> BlueMatt: I agree, it should.
3044 2011-04-19 22:11:32 <JFK911> because everyone i know has sell orders at 1.20
3045 2011-04-19 22:11:33 <luke-jr> JFK911: I hope so
3046 2011-04-19 22:11:53 <luke-jr> I kinda sold all my BTC back at $1
3047 2011-04-19 22:12:08 <JFK911> well the next run up will get it even higher
3048 2011-04-19 22:12:22 <sipa> BlueMatt: it doesn't?
3049 2011-04-19 22:12:42 <sipa> quite sure there was a patch for that recently
3050 2011-04-19 22:12:45 <BlueMatt> sipa: specifically, luke-jr's code doesn't close cleanly
3051 2011-04-19 22:12:54 <sethsethseth> i dont get the fee structure. sometimes i send 300btc and its .03 fee and sometimes .01
3052 2011-04-19 22:13:18 <BlueMatt> hopefully not too high, it is clearly fairly overvalued (in that if large holders were to sell the market would collapse) and high prices create some barriers to adoption IMHO
3053 2011-04-19 22:13:22 <luke-jr> BlueMatt: it does, if the rest does :P
3054 2011-04-19 22:13:41 <luke-jr> BlueMatt: did you merge it to a current branch?
3055 2011-04-19 22:13:43 <BlueMatt> sethsethseth: it depends on the tx size, ie the amount of inputs used
3056 2011-04-19 22:13:54 <BlueMatt> luke-jr: no just checkedout your branch
3057 2011-04-19 22:13:58 <luke-jr> BlueMatt: my bitcoinuri branch is intentionally ancient
3058 2011-04-19 22:14:01 <luke-jr> merge it
3059 2011-04-19 22:14:08 <LtBrenton> well I've already sold my positions out to USD, I'm ready for the dump :D
3060 2011-04-19 22:14:13 marlowe has joined
3061 2011-04-19 22:14:30 <JFK911> what will happen then:
3062 2011-04-19 22:14:33 gavinandresen has joined
3063 2011-04-19 22:14:34 <netxshare> 192.71 MH/s sllooowww
3064 2011-04-19 22:14:37 <JFK911> everyone who paid 1.18 will feel burned
3065 2011-04-19 22:14:48 <JFK911> lots of FUD will circulate about btc
3066 2011-04-19 22:14:50 <JFK911> btc will die
3067 2011-04-19 22:15:09 <netxshare> because it dropped?
3068 2011-04-19 22:15:23 <JFK911> yeah merchants and users will be scared away
3069 2011-04-19 22:15:29 <BlueMatt> JFK911: I dont think anyone will complain about paying a high price, look at the graph it fluctuates so much people should know
3070 2011-04-19 22:15:40 <BlueMatt> JFK911: plus what I'm saying is I want stability
3071 2011-04-19 22:15:42 <netxshare> that happens all the time with normal currency and stocks
3072 2011-04-19 22:15:56 <netxshare> not as large of a drop
3073 2011-04-19 22:15:56 <BlueMatt> and if this price keeps going up like this that is not stability
3074 2011-04-19 22:15:58 m86 has quit (Ping timeout: 258 seconds)
3075 2011-04-19 22:16:14 <JFK911> well this because forex and stock exchanges are global markets
3076 2011-04-19 22:16:21 <JFK911> there are more than a few thousand users playing
3077 2011-04-19 22:16:35 <BlueMatt> netxshare: its different for btc because your supply is usd based whereas for most companies they get paid in what they pay their supply in
3078 2011-04-19 22:16:39 <JFK911> here, every individual has a greater effect on the whole system
3079 2011-04-19 22:16:48 <BlueMatt> JFK911: ^
3080 2011-04-19 22:17:05 <BlueMatt> yes it is crazy unstable, but it presents a barrier to adoption
3081 2011-04-19 22:17:19 <JFK911> we saw bitcoin inflation when the price was at 0.50-0.60
3082 2011-04-19 22:17:21 <netxshare> it will even out with time
3083 2011-04-19 22:17:25 <BlueMatt> luke-jr: yea, merge makes it work
3084 2011-04-19 22:18:18 <BlueMatt> netxshare: I hope so, but with some of the large holder holding on due to the rapid increases, I don't know how soon that will come
3085 2011-04-19 22:18:21 <luke-jr> mostly, you could actually compile that wxipcserver.cpp into a .so library, and use it with LD_PRELOAD :p
3086 2011-04-19 22:18:42 <BlueMatt> luke-jr: could is one thing, should is slightly different ;)
3087 2011-04-19 22:18:50 <netxshare> once I get my mining rigs going I plan to hold on to the BTC
3088 2011-04-19 22:19:01 <JFK911> why didnt you mine last year
3089 2011-04-19 22:19:05 <JFK911> it was a lot easier
3090 2011-04-19 22:19:08 <netxshare> because I suck
3091 2011-04-19 22:19:10 <JFK911> last year i was laughing at btc
3092 2011-04-19 22:19:12 <JFK911> ya
3093 2011-04-19 22:19:13 <luke-jr> lol
3094 2011-04-19 22:19:24 <netxshare> I saw posts about it but ignored them
3095 2011-04-19 22:19:39 <MBS> i wish i would have found out about btc earlier, lol
3096 2011-04-19 22:19:52 <JFK911> would you have mined three cent bitcoins?
3097 2011-04-19 22:20:01 <BlueMatt> MBS: we all wish we started mining earlier
3098 2011-04-19 22:20:08 <netxshare> I only came here because I messaged art the other day and he told me about it
3099 2011-04-19 22:20:10 <MBS> didnt start til a few days ago, lol
3100 2011-04-19 22:20:11 <netxshare> so I checked it out
3101 2011-04-19 22:20:13 <MBS> so i wont get shit all
3102 2011-04-19 22:20:20 <netxshare> and was like damn it to hell why did I not look at this sooner
3103 2011-04-19 22:20:22 <luke-jr> netxshare: beg art for some, he has plenty
3104 2011-04-19 22:20:22 <JFK911> mbs whats your hash rate
3105 2011-04-19 22:20:28 <MBS> ~160Mh/s
3106 2011-04-19 22:20:39 <MBS> if i could afford it would get a 5870 to add to that
3107 2011-04-19 22:20:44 <JFK911> u might get a block this month, maybe you should pool up for now
3108 2011-04-19 22:20:49 <MBS> yeah im pooling
3109 2011-04-19 22:20:50 <BlueMatt> I saw bitcoin at one point like 6 months ago, but at that time I dismissed it and didn't bother to learn anything about it...now I wish I had started mining then
3110 2011-04-19 22:20:58 eternal1 has joined
3111 2011-04-19 22:21:03 <luke-jr> I never heard a word about bitcoin until Jan 1 this year
3112 2011-04-19 22:21:03 <MBS> have like 6.98 so far since like thursday, lol
3113 2011-04-19 22:21:06 <netxshare> lol
3114 2011-04-19 22:21:16 <MBS> and like .47 pending from deepbt
3115 2011-04-19 22:21:21 <luke-jr> mizerydearia was crazy enough to go into #freenode-newyears drunk offering 1 BTC to the first bitcoin address
3116 2011-04-19 22:21:31 <luke-jr> I'm like "wtf? that sounds interestingâ¦"
3117 2011-04-19 22:21:41 <johnlockwood> is the hashrate the khash/s on the bottom of the client window?
3118 2011-04-19 22:21:46 <MBS> yeah
3119 2011-04-19 22:21:53 <BlueMatt> johnlockwood: dont mine in the client
3120 2011-04-19 22:21:54 <netxshare> It seems like people tend to give out BTC freely
3121 2011-04-19 22:21:54 <JFK911> you're wasting power
3122 2011-04-19 22:21:56 <BlueMatt> you wont ever get anything
3123 2011-04-19 22:21:57 <luke-jr> johnlockwood: turn that off
3124 2011-04-19 22:22:02 <johnlockwood> oh
3125 2011-04-19 22:22:03 <JFK911> use gpu miner
3126 2011-04-19 22:22:04 <netxshare> or maybe they just give their addresses out and hope people do
3127 2011-04-19 22:22:14 <luke-jr> netxshare: well, it's cheap to make low payments
3128 2011-04-19 22:22:16 <luke-jr> netxshare: and safe
3129 2011-04-19 22:22:20 <BlueMatt> johnlockwood: at the very least use jgarzik's cpuminer in a pool
3130 2011-04-19 22:22:29 <luke-jr> netxshare: so people don't mind paying for, eg, good info
3131 2011-04-19 22:22:30 <MBS> i wonder if i can get my friend to sell me or let my use his 5750 over the summer
3132 2011-04-19 22:22:32 <johnlockwood> BlueMatt: where do I get that?
3133 2011-04-19 22:22:37 <netxshare> ah
3134 2011-04-19 22:22:46 <MBS> he uses linux, so hes not a huge fan of atis drivers anymore
3135 2011-04-19 22:22:50 <BlueMatt> johnlockwood: http://www.bitcoin.org/smf/index.php?topic=1925.0
3136 2011-04-19 22:22:50 <johnlockwood> I only heard about bitcoins on sunday
3137 2011-04-19 22:22:52 <luke-jr> although I only have 1 BTC so far for my bitcoin: URI patch :|
3138 2011-04-19 22:23:04 <luke-jr> johnlockwood: unless you have a Radeon, forget about mining
3139 2011-04-19 22:23:10 <netxshare> it's a start luke, it's a start.
3140 2011-04-19 22:23:12 <MBS> would be too hot in my dsektop with my 5770, would saw off the end of my servers only pcie x1 slot, and put it in there
3141 2011-04-19 22:23:40 <johnlockwood> my cpu is a 4core i7
3142 2011-04-19 22:23:42 <netxshare> I earn about 1.50-1.75 btc a day right now
3143 2011-04-19 22:23:44 <netxshare> lol
3144 2011-04-19 22:23:48 <MBS> same
3145 2011-04-19 22:23:56 <netxshare> stupid nvidia
3146 2011-04-19 22:23:56 <BlueMatt> johnlockwood: if you want to make money considering power, only a high-end gpu will do that. If you just want to try out mining, the best way is either a gpu miner or jgarzk's cpuminer in a pool
3147 2011-04-19 22:24:16 <MBS> if i could sell one of the many things i have been trying to sell, could easily get a 5830 on sale
3148 2011-04-19 22:24:23 <luke-jr> johnlockwood: all CPUs are garbage for mining
3149 2011-04-19 22:24:32 <BlueMatt> johnlockwood: my i7 920 ocd to 3.8 GHz gets maybe 15 MHash/s which is verry little
3150 2011-04-19 22:24:36 <johnlockwood> my GPI is ATI Radeon HD 4850
3151 2011-04-19 22:24:39 <BlueMatt> ie a couple cents a day
3152 2011-04-19 22:24:51 <luke-jr> BlueMatt: haha, I get 14 MH/s normal-clocked i5 :P
3153 2011-04-19 22:24:56 <MBS> yeah
3154 2011-04-19 22:25:00 <luke-jr> johnlockwood: that one isnt worth running either -.-
3155 2011-04-19 22:25:02 <MBS> but cpu mining is better than nothing, lol
3156 2011-04-19 22:25:07 <luke-jr> johnlockwood: so just treat BTC like any other currency
3157 2011-04-19 22:25:12 <netxshare> what cpu miner are you using?
3158 2011-04-19 22:25:13 <MBS> get ~7k hash from the athlon in my home server
3159 2011-04-19 22:25:15 <BlueMatt> luke-jr: I haven't tried in months that was old cpuminer before all the new stuff so I dont know really
3160 2011-04-19 22:25:25 <BlueMatt> ok I guess I need to try it again...
3161 2011-04-19 22:25:25 <MBS> and 3k hash when im doing nothing on desktop from 1 core
3162 2011-04-19 22:25:39 <netxshare> I was getting 4k with bitcoin
3163 2011-04-19 22:25:41 <luke-jr> MBS: nope
3164 2011-04-19 22:25:42 <BlueMatt> johnlockwood: a 4850 will get you a decent amount
3165 2011-04-19 22:25:46 <netxshare> have not tested anything else
3166 2011-04-19 22:25:49 <luke-jr> MBS: because you're paying more for electricity to run it
3167 2011-04-19 22:25:57 <luke-jr> MBS: idle CPU uses less electric than a busy one
3168 2011-04-19 22:25:59 <MBS> lol well its running anyway so :p
3169 2011-04-19 22:26:05 <MBS> well i guess its doing different stuff
3170 2011-04-19 22:26:15 <MBS> normally its backing up 3TB of stuff up to crashplan XD
3171 2011-04-19 22:26:16 <luke-jr> if it isn't idle, it couldn't be mining ;)
3172 2011-04-19 22:26:21 <MBS> and thats what normally maxxes it out
3173 2011-04-19 22:26:31 <johnlockwood> I'd rather make BTC/money by building a website marketplace
3174 2011-04-19 22:26:31 <MBS> except now mining takes up more than crashplan
3175 2011-04-19 22:26:48 <netxshare> and wth does rpc-miner give me
3176 2011-04-19 22:26:49 <netxshare> Target = 00000000ffffffffffffffffffffffffffffffffffffffffffffffffffffff
3177 2011-04-19 22:26:51 <MBS> up to like 40% backed up though
3178 2011-04-19 22:27:08 <luke-jr> netxshare: pool
3179 2011-04-19 22:27:24 <netxshare> it does not like deepbit.net?
3180 2011-04-19 22:27:33 <netxshare> it throws errors
3181 2011-04-19 22:28:28 larsivi has quit (Ping timeout: 240 seconds)
3182 2011-04-19 22:28:28 <luke-jr> netxshare: that is the correct target for any pool
3183 2011-04-19 22:28:40 <BlueMatt> ok, I get more like 20 Mhash on my cpu
3184 2011-04-19 22:28:57 <BlueMatt> still no where near worth mining
3185 2011-04-19 22:29:24 <luke-jr> XD
3186 2011-04-19 22:29:52 <netxshare> I dunno
3187 2011-04-19 22:29:57 <netxshare> 20m more would help me
3188 2011-04-19 22:29:58 <netxshare> lol
3189 2011-04-19 22:30:14 <luke-jr> not likely
3190 2011-04-19 22:30:18 <luke-jr> ;;bc,calc 20000
3191 2011-04-19 22:30:18 <gribble> The average time to generate a block at 20000 Khps, given current difficulty of 92347.59095209 , is 32 weeks, 5 days, 12 hours, 44 minutes, and 54 seconds
3192 2011-04-19 22:30:22 <luke-jr> 32 weeks helps?
3193 2011-04-19 22:30:30 <netxshare> yes
3194 2011-04-19 22:30:44 <netxshare> I take 50mil from another computer
3195 2011-04-19 22:30:52 <netxshare> or 50-60mil
3196 2011-04-19 22:31:04 <BlueMatt> well on a pool it would make a tiny bit, but thats only if you dont pay for power
3197 2011-04-19 22:32:14 <netxshare> yeah
3198 2011-04-19 22:33:29 <netxshare> I wonder if ill be able to find enough 5970s
3199 2011-04-19 22:35:31 <netxshare> bah
3200 2011-04-19 22:35:45 <netxshare> only xfx worth buying has a waterblock on it
3201 2011-04-19 22:36:28 lulzplzkthx has quit (Ping timeout: 240 seconds)
3202 2011-04-19 22:36:55 <BlueMatt> I need a new gcard and need to finally add it to my waterloop which was designed for a gcard to begin with
3203 2011-04-19 22:37:01 <MBS> just buy twice as many 5870s lol
3204 2011-04-19 22:37:36 <MBS> if you dont mind updating bios of each one, this one is really cheap
3205 2011-04-19 22:37:37 <MBS> http://www.newegg.com/Product/Product.aspx?Item=N82E16814125316&cm_re=5870-_-14-125-316-_-Product
3206 2011-04-19 22:37:47 ezl has quit (Ping timeout: 264 seconds)
3207 2011-04-19 22:38:01 CodePHP is now known as AFK_PHPAdam
3208 2011-04-19 22:38:04 <MBS> 220 - 30 MIR and -30 for selling total war shogun 2
3209 2011-04-19 22:38:38 <netxshare> 5870
3210 2011-04-19 22:38:56 <netxshare> wiki says it does 340ish
3211 2011-04-19 22:39:39 <netxshare> I would end up buying more systems
3212 2011-04-19 22:40:07 <MBS> how many do you plan on doing?
3213 2011-04-19 22:41:00 <netxshare> 10 system with 2 cards
3214 2011-04-19 22:41:09 <netxshare> well 11
3215 2011-04-19 22:41:13 <netxshare> maybe 2
3216 2011-04-19 22:41:18 <netxshare> for starting out
3217 2011-04-19 22:41:20 <MBS> so 20 5970s
3218 2011-04-19 22:41:26 AFK_PHPAdam is now known as CodePHP
3219 2011-04-19 22:41:29 CodePHP is now known as AFK_PHPAdam
3220 2011-04-19 22:41:31 <netxshare> that's the goal
3221 2011-04-19 22:41:37 <netxshare> if I can get them used
3222 2011-04-19 22:42:00 <netxshare> I might not be able to buy them all once at retail price
3223 2011-04-19 22:42:24 <MBS> buy in batches, lol
3224 2011-04-19 22:42:34 lulzplzkthx has joined
3225 2011-04-19 22:42:54 <MBS> damn thats gunna be some power use lol
3226 2011-04-19 22:43:04 <netxshare> I have someone in china looking for hardware
3227 2011-04-19 22:43:05 <MBS> almost 10 kilowatts
3228 2011-04-19 22:43:32 <netxshare> dunno if ill be able to find the 5970's there
3229 2011-04-19 22:43:49 <netxshare> but with all the internet cafes we might be lucky
3230 2011-04-19 22:44:13 <netxshare> we will see what happens this friday
3231 2011-04-19 22:44:30 AFK_PHPAdam is now known as CodePHP
3232 2011-04-19 22:44:36 <netxshare> so should at least get either a 6990 or 2x 5970s
3233 2011-04-19 22:45:14 <MBS> ive been wondering if there are alot of black market ati cards in china, lol
3234 2011-04-19 22:47:44 <netxshare> might check Korea too
3235 2011-04-19 22:49:05 phantomcircuit has joined
3236 2011-04-19 22:49:58 octarine has left ()
3237 2011-04-19 22:55:21 EPiSKiNG- has quit (Ping timeout: 260 seconds)
3238 2011-04-19 22:55:47 sethsethseth has quit (Read error: Connection reset by peer)
3239 2011-04-19 22:56:57 sethsethseth has joined
3240 2011-04-19 22:57:24 phantomcircuit has quit (Quit: Leaving)
3241 2011-04-19 22:59:22 ByteCoin has quit (Remote host closed the connection)
3242 2011-04-19 22:59:39 ByteCoin has joined
3243 2011-04-19 22:59:45 <netxshare> what sdk does everyone use when mining ati cards on linux?
3244 2011-04-19 23:00:33 <luke-jr> 2.1
3245 2011-04-19 23:01:06 <ByteCoin> ljr: Is that your bot on bitcoin-watch?
3246 2011-04-19 23:01:20 <luke-jr> yes
3247 2011-04-19 23:01:47 <ByteCoin> ljr: Do you know about the problem with printing bitcoin values >1?
3248 2011-04-19 23:01:57 <luke-jr> ââ?
3249 2011-04-19 23:02:31 <ByteCoin> If you look, when it prints transactions with integer amounts of bitcoins it says .00
3250 2011-04-19 23:02:41 <luke-jr> so?
3251 2011-04-19 23:03:07 <ByteCoin> It should say 1.00 not .00 for 1 BTC
3252 2011-04-19 23:03:15 <luke-jr> â¦
3253 2011-04-19 23:03:19 <luke-jr> it says 1.00
3254 2011-04-19 23:03:40 <ByteCoin> Txn a5b736377cc762244ba34eebd42ae48d0435870234b3df29ebd78cb5c77041ec: 1AdGX6EGrZR1pzWoo35eQxaKMA9bEDJqfk .00 BTC
3255 2011-04-19 23:03:49 <luke-jr> [18:32:00] <ljrbot> Txn c087c5c7585b2529f5fe4c9723e3c76d2a5ce1c2f712e27f375ca6ef627f9731: 13RRwPE3B3L4hxA5SMQdbJkRcZjALxYpAt 1.00 BTC, 12ryMU3mp98T47aGNGkjoc6FQTpmadHsYk 1.00 BTC
3256 2011-04-19 23:03:56 <ByteCoin> That's a recent cut and paste
3257 2011-04-19 23:04:16 <luke-jr> [18:59:21] <ljrbot> Txn a5b736377cc762244ba34eebd42ae48d0435870234b3df29ebd78cb5c77041ec: 1AdGX6EGrZR1pzWoo35eQxaKMA9bEDJqfk 317.00 BTC
3258 2011-04-19 23:04:25 <luke-jr> ByteCoin: something wrong with your client
3259 2011-04-19 23:05:11 <ByteCoin> I'm using the Empathy client on a standard Ubuntu install
3260 2011-04-19 23:05:23 <ByteCoin> Why would it munge stuff?
3261 2011-04-19 23:05:26 <luke-jr> report a bug
3262 2011-04-19 23:05:33 <luke-jr> colour codes
3263 2011-04-19 23:06:24 <ByteCoin> Oh. Is it printed in different colours on your client?
3264 2011-04-19 23:06:35 <ByteCoin> Can you colour IRC stuff?
3265 2011-04-19 23:06:41 <luke-jr> yes
3266 2011-04-19 23:06:50 <ByteCoin> Hmmm...
3267 2011-04-19 23:07:56 <ByteCoin> Ta for explaining
3268 2011-04-19 23:08:08 <BlueMatt> can he, theoretically yes; should he, no
3269 2011-04-19 23:08:15 <BlueMatt> as you see, it causes problems in some clients
3270 2011-04-19 23:08:42 <luke-jr> ByteCoin: escaped, [Tuesday, April 19, 2011] [6:59:21 pm] <ljrbot> ^C13,1Txn ^_a5b736377cc762244ba34eebd42ae48d0435870234b3df29ebd78cb5c77041ec^_^C15,1: ^C07,1^_1AdGX6EGrZR1pzWoo35eQxaKMA9bEDJqfk^_ ^C11,01317.00 BTC^C15,01
3271 2011-04-19 23:09:12 <luke-jr> I suppose I could workaround it by moving the ^C before the space
3272 2011-04-19 23:09:24 <luke-jr> save 1 byte too
3273 2011-04-19 23:09:54 <ByteCoin> Can anyone recommend a good IRC client for Ubuntu? I *have* heard that empathy is not great
3274 2011-04-19 23:10:04 <luke-jr> ByteCoin: Konversation works nice
3275 2011-04-19 23:10:06 <BlueMatt> ByteCoin: XChat is alright
3276 2011-04-19 23:10:13 <BlueMatt> not XChat-gnome, straight XChat
3277 2011-04-19 23:10:18 <luke-jr> XChat is buggy\
3278 2011-04-19 23:10:20 <BlueMatt> xchat-gnome is shit
3279 2011-04-19 23:10:27 <BlueMatt> luke-jr: Ive never had any problems
3280 2011-04-19 23:10:29 <luke-jr> but you can at least fix XChat with settings
3281 2011-04-19 23:10:34 <BlueMatt> though honestly I dont really like it that much
3282 2011-04-19 23:10:37 <luke-jr> BlueMatt: its colour mappings are wrong
3283 2011-04-19 23:10:40 <luke-jr> the default ones
3284 2011-04-19 23:10:47 <BlueMatt> luke-jr: hence why no one should use colors
3285 2011-04-19 23:10:54 <luke-jr> BlueMatt: hence why XChat should fix their bugs
3286 2011-04-19 23:10:59 <BlueMatt> they just dont work in most clients
3287 2011-04-19 23:11:02 <luke-jr> they do
3288 2011-04-19 23:11:05 <BlueMatt> at least not the same in most clients
3289 2011-04-19 23:11:07 Netsniper has joined
3290 2011-04-19 23:11:08 <luke-jr> just not the rare buggy clients
3291 2011-04-19 23:11:20 <BlueMatt> luke-jr: IIRC some 40% of people on here used XChat?
3292 2011-04-19 23:11:23 <luke-jr> no
3293 2011-04-19 23:11:25 <luke-jr> 40% use irssi
3294 2011-04-19 23:11:28 <luke-jr> which works fine
3295 2011-04-19 23:11:39 <luke-jr> X-Chat was like 3% IIRC
3296 2011-04-19 23:11:44 <ByteCoin> Konversation is saying it needs 260MB and to download 76MB! Seems big for an IRC program!
3297 2011-04-19 23:11:45 <BlueMatt> no it was way more than that
3298 2011-04-19 23:11:53 <luke-jr> ByteCoin: that's because Ubuntu is GTK crap
3299 2011-04-19 23:12:02 <luke-jr> ByteCoin: most of that 260 MB is standard system stuff
3300 2011-04-19 23:12:19 <ByteCoin> Hmmm...
3301 2011-04-19 23:13:10 <BlueMatt> X-Chat was a huge chunk if not the majority up there in the top 3 or 4 clients
3302 2011-04-19 23:13:20 tenach has quit (Quit: moving locations.)
3303 2011-04-19 23:13:24 <BlueMatt> s/majority/plurality/
3304 2011-04-19 23:13:30 <ByteCoin> Last weekend's post of mine doesn't seem to have attracted much attention. I thought it would rock the bitcoin world... nevermind
3305 2011-04-19 23:13:41 <BlueMatt> ByteCoin: which post
3306 2011-04-19 23:13:42 <BlueMatt> ?
3307 2011-04-19 23:13:49 <ByteCoin> will get URL
3308 2011-04-19 23:14:19 <ByteCoin> http://www.bitcoin.org/smf/index.php?topic=5965.msg87757#msg87757
3309 2011-04-19 23:14:22 <BlueMatt> Apr 13 20:39:39 <luke-jr> 65 irssi, 34 X-Chat, 11 Quassel, 4 Konversation, 19 mIRC
3310 2011-04-19 23:14:38 <BlueMatt> second only to irssi
3311 2011-04-19 23:15:27 <ByteCoin> I suppose the important thing is to get it on record for the future
3312 2011-04-19 23:15:46 Netsniper has quit (Ping timeout: 258 seconds)
3313 2011-04-19 23:16:47 amiller has joined
3314 2011-04-19 23:17:58 <ByteCoin> Phew! xchat also requires a similar amount to be downloaded/stored to install!
3315 2011-04-19 23:18:12 <BlueMatt> ByteCoin: I agree with TD, what is the application?
3316 2011-04-19 23:18:38 <BlueMatt> and how are these txes less traceable than any other tx with one address in between?
3317 2011-04-19 23:18:46 DukeOfURL has quit (Remote host closed the connection)
3318 2011-04-19 23:19:30 skyewm has joined
3319 2011-04-19 23:19:32 <ByteCoin> I see... I have to explain the untraceability
3320 2011-04-19 23:19:50 <luke-jr> ByteCoin: added a workaround
3321 2011-04-19 23:20:14 <ByteCoin> ljr: Will check
3322 2011-04-19 23:20:22 lfm_ has quit (Ping timeout: 258 seconds)
3323 2011-04-19 23:20:53 <ByteCoin> Ahhh! That's better!
3324 2011-04-19 23:21:05 <gjs278> I don't see how people use irssi
3325 2011-04-19 23:21:12 <dirtyfil1hy> it rulez
3326 2011-04-19 23:21:16 toffoo has quit ()
3327 2011-04-19 23:21:32 <ByteCoin> ljr: Thx for that! Saves me installing a new client!
3328 2011-04-19 23:22:42 <ByteCoin> BlueMatt: Can I try out my untraceability explanation on you before I post to the forum?
3329 2011-04-19 23:22:53 <BlueMatt> sure
3330 2011-04-19 23:23:28 <ByteCoin> Ok. You sell a service and publish an address for receiving coins A. People send coins to A.
3331 2011-04-19 23:23:56 <luke-jr> ByteCoin: I'd still suggest Konversation so you can see it in its glory :P
3332 2011-04-19 23:24:03 <ByteCoin> If you change A, then you probably still need to publish them all so A1 A2 etc....
3333 2011-04-19 23:24:08 <luke-jr> https://en.bitcoin.it/wiki/Bitcoin-Watch
3334 2011-04-19 23:24:13 <luke-jr> (screenshot)
3335 2011-04-19 23:24:47 <ByteCoin> You want to buy from someone who publishes an address Z using your coins
3336 2011-04-19 23:25:15 <ByteCoin> So people know that the coins spent by address A are yours and they know that coins received by Z are his
3337 2011-04-19 23:25:40 <BlueMatt> luke-jr: that is exactly what it looks like in X-Chat, except with a white background and a black one on the text
3338 2011-04-19 23:25:44 <ByteCoin> No matter what you do, the block chain records that coins are going from A to Z no matter how many intermediate addressses you send them to
3339 2011-04-19 23:25:45 <BlueMatt> hence the hate of the colors in general
3340 2011-04-19 23:25:52 <luke-jr> BlueMatt: not by X-Chat default
3341 2011-04-19 23:25:55 <gjs278> I hate the theme of xchat
3342 2011-04-19 23:25:58 <gjs278> that's all
3343 2011-04-19 23:26:01 <luke-jr> BlueMatt: X-Chat default has all the colours dark
3344 2011-04-19 23:26:03 <BlueMatt> luke-jr: I never changed any color crap
3345 2011-04-19 23:26:10 <luke-jr> maybe your distro did
3346 2011-04-19 23:26:21 <BlueMatt> oh I see what you mean, yea colors are a bit darker
3347 2011-04-19 23:26:25 <ByteCoin> Can you see how the transaction between A and Z is traceable?
3348 2011-04-19 23:27:15 <BlueMatt> ByteCoin: yes (assuming Z is published as yours, which it usually isnt) but I see the point
3349 2011-04-19 23:27:23 <ByteCoin> Oooh. Quite like the colours!
3350 2011-04-19 23:28:14 <BlueMatt> luke-jr: Out of curiosity, how does one destination address in that screenshot have one less character than the rest, that shouldnt be right?
3351 2011-04-19 23:28:21 <ByteCoin> So if Z is let's say turns out to be a terrorist then you could be accused of financing them...
3352 2011-04-19 23:28:47 <BlueMatt> ByteCoin: of course, but I don't see how your thing changes that
3353 2011-04-19 23:29:55 <ByteCoin> Ok. So with my scheme your receiving address A is fixed and published and so is Z.
3354 2011-04-19 23:30:03 <luke-jr> BlueMatt: some addresses are shorter
3355 2011-04-19 23:30:06 <BlueMatt> I thought all addresses were the same length? why is 1pHwBRbG6Xes5fWMS2ysZtxBhpYrv6Fi5 valid?
3356 2011-04-19 23:30:19 <BlueMatt> luke-jr: I thought they were all the same length?
3357 2011-04-19 23:30:25 <luke-jr> BlueMatt: I don't know why, but they aren't fixed length.
3358 2011-04-19 23:30:27 <BlueMatt> ie same size pub key
3359 2011-04-19 23:30:35 <luke-jr> something to do with trimming leading zeros
3360 2011-04-19 23:30:36 <BlueMatt> luke-jr: does it have to do with the encoding?
3361 2011-04-19 23:30:39 <luke-jr> at some point
3362 2011-04-19 23:30:43 <BlueMatt> ah, ok
3363 2011-04-19 23:30:46 <BlueMatt> so its the encoding
3364 2011-04-19 23:30:52 <BlueMatt> makes sense
3365 2011-04-19 23:30:52 <ByteCoin> Bluematt: It's a feature ofthe base58 encoding
3366 2011-04-19 23:30:54 <tcatm> BlueMatt: because of bitcoin's strange base58 variant (padding)
3367 2011-04-19 23:30:57 <ByteCoin> Screwey
3368 2011-04-19 23:31:05 * luke-jr wonders if you can add the leading zeros in and still use it validly
3369 2011-04-19 23:31:13 <ByteCoin> Don't think so
3370 2011-04-19 23:31:27 <ByteCoin> Pretty sure nor
3371 2011-04-19 23:31:31 <ByteCoin> not
3372 2011-04-19 23:32:12 <ByteCoin> Ok. So with my scheme, you use the keys for A and Z to come up with a transfer key T and you send the coins to T
3373 2011-04-19 23:32:49 <ByteCoin> Then the owner of Z also "has control" of T so they transfer the coins to Q
3374 2011-04-19 23:33:40 <ByteCoin> So even though you're using publicly known receiving addresses A and Z the only transfers that can be seen are A to T and T to Q but not Q to Z
3375 2011-04-19 23:33:48 lfm_ has joined
3376 2011-04-19 23:34:12 <ByteCoin> there's nothing linking Q to Z even though Q is controlled by Z
3377 2011-04-19 23:34:18 <BlueMatt> ByteCoin: great, the problem is that it is very rare, in fact almost never does it happen that Z is published
3378 2011-04-19 23:34:53 <ByteCoin> Doesn't it happen all the time? "If you like my post send coins to this address..."
3379 2011-04-19 23:34:57 <BlueMatt> in the end Z is only known by the next recipient of the coins (which is the same in your scenario)
3380 2011-04-19 23:35:28 <BlueMatt> ByteCoin: I mean for txes which might be received by someone who is being secretive (ie terrorists)
3381 2011-04-19 23:35:29 <ByteCoin> People publish receiving addresses all the time
3382 2011-04-19 23:36:01 <BlueMatt> ByteCoin: as donations, not in the course of regular txes
3383 2011-04-19 23:36:08 <ByteCoin> Ok. So if you're being secretive how do you accept payments?
3384 2011-04-19 23:36:16 <BlueMatt> donations are, in fact, just about the only time addresses are published
3385 2011-04-19 23:36:34 <ByteCoin> Ok How do you know whom to pay for a normal tx?
3386 2011-04-19 23:36:53 <BlueMatt> ByteCoin: they tell you privately, not published
3387 2011-04-19 23:37:01 hubb has joined
3388 2011-04-19 23:37:03 <BlueMatt> I see the point, I just dont see the point
3389 2011-04-19 23:37:08 <ByteCoin> So they need a secure channel of communication....
3390 2011-04-19 23:37:13 dsg has quit (Ping timeout: 260 seconds)
3391 2011-04-19 23:37:19 dsg has joined
3392 2011-04-19 23:37:39 <ByteCoin> Wouldn't it be nice not to have to use a secure channel to tell someone how to pay you>
3393 2011-04-19 23:37:40 <BlueMatt> Also, could someone with considerable resources (ie the fbi, cia, nsa, etc) not figure out which txes belong to a Z given a list of potential Qs?
3394 2011-04-19 23:37:44 <ByteCoin> ?
3395 2011-04-19 23:37:49 <BlueMatt> ByteCoin: true
3396 2011-04-19 23:38:06 <ByteCoin> No I don't believe they can....
3397 2011-04-19 23:38:21 <ByteCoin> The security is aprox the same as for forging signatures etc...
3398 2011-04-19 23:38:33 <lfm_> well there is big steps for both public to private message to secure channel
3399 2011-04-19 23:38:50 <ByteCoin> lfm: Clarify?
3400 2011-04-19 23:39:27 <hubb> are the vendors listed on https://en.bitcoin.it/wiki/Trade vetted at all? or can the list be edited by anyone with a wiki account
3401 2011-04-19 23:39:30 <BlueMatt> ByteCoin: I see the point and it looks really cool, but in the end, I think it would be much better for any terrorists to publish receiving addresses in their existing secure channels
3402 2011-04-19 23:39:37 <lfm_> theres a difference between a private message and a secure channel just as there is a difference bewteen a public broadcast and a private messge
3403 2011-04-19 23:39:39 <BlueMatt> hubb: edited by anyone
3404 2011-04-19 23:39:42 <jgarzik> hubb: not vetted
3405 2011-04-19 23:39:53 <hubb> so trade at your own peril, then
3406 2011-04-19 23:39:57 <BlueMatt> hubb: yep
3407 2011-04-19 23:40:02 <hubb> thx
3408 2011-04-19 23:40:04 <ByteCoin> lfm: Difference between private and secure?
3409 2011-04-19 23:40:12 <BlueMatt> hubb: or go to #bitcoin-otc and look at people's WoT ratings
3410 2011-04-19 23:40:13 <gjs278> if I clicked your name
3411 2011-04-19 23:40:16 <gjs278> and sent you a private message
3412 2011-04-19 23:40:18 <gjs278> nobody could read it
3413 2011-04-19 23:40:23 <gjs278> but it wasnt securely sent
3414 2011-04-19 23:40:25 <BlueMatt> jgarzik: if luke-jr MITs his uri code, whats the chance it would get merged?
3415 2011-04-19 23:40:49 <ByteCoin> What is security in this context?
3416 2011-04-19 23:41:01 <ByteCoin> Stopping other people reading it?
3417 2011-04-19 23:41:05 <BlueMatt> ByteCoin: inability of the nsa to see what you are doing
3418 2011-04-19 23:41:05 <lfm_> theres many levels and you have to find whats right for your secureity needs, Ya ordinary private messages are not generally availabvle to the public, secure messages are not available to anyone but the intended recipient
3419 2011-04-19 23:41:17 <ByteCoin> What's the difference with private?
3420 2011-04-19 23:41:26 <gjs278> someone could intercept the private message
3421 2011-04-19 23:41:30 <ByteCoin> lfm: I see
3422 2011-04-19 23:41:33 <BlueMatt> ie freenode can
3423 2011-04-19 23:41:35 <gjs278> the scure message wouldn't be intercepted
3424 2011-04-19 23:41:37 <gjs278> yeah
3425 2011-04-19 23:41:40 <gjs278> or it could be intercepted
3426 2011-04-19 23:41:43 <gjs278> but nobody could read it
3427 2011-04-19 23:41:45 <jgarzik> BlueMatt: no idea what that means
3428 2011-04-19 23:41:46 <gjs278> except you
3429 2011-04-19 23:41:52 <jgarzik> BlueMatt: isn't URI code usually for browsers?
3430 2011-04-19 23:41:53 <BlueMatt> gjs278: probably not if you are using ssl/tls, but ok
3431 2011-04-19 23:41:54 <luke-jr> hubb: it's vetted by others on the wiki
3432 2011-04-19 23:41:57 <gjs278> yeah
3433 2011-04-19 23:41:58 <luke-jr> hubb: after the fact
3434 2011-04-19 23:42:00 <gjs278> that's the point
3435 2011-04-19 23:42:04 <gjs278> ssl would be a form of security
3436 2011-04-19 23:42:10 <BlueMatt> jgarzik: luke-jr wrote a bitcoin: url handler for bitcoin-mainline
3437 2011-04-19 23:42:17 <BlueMatt> jgarzik: mostly here:
3438 2011-04-19 23:42:20 <gjs278> but if I didn't use ssl, just sending a private message would only be considered private
3439 2011-04-19 23:42:23 <BlueMatt> http://luke.dashjr.org/programs/bitcoin/w/bitcoind/luke-jr.git/commitdiff/7eb88f0cb6af6dd106fbec818f81af9b13318380
3440 2011-04-19 23:42:40 <ByteCoin> Ok. Fine. I can't be bothered with implementing private only. For me to think it's good it has to be secure
3441 2011-04-19 23:43:11 <ByteCoin> So using your definition (which is fine) transactions are private but not "secure"
3442 2011-04-19 23:43:27 <ByteCoin> Secure against people knowing to whom you're transferring
3443 2011-04-19 23:43:32 <BlueMatt> ByteCoin: I agree, and I think your idea is great it provides security at the bitcoin-level. I just think it is more appealing for a terrorist cell to create security at the address-trading level ie communication level
3444 2011-04-19 23:44:14 <ByteCoin> Well, you know how public key crypto means you don't have to distribute secret keys securely...
3445 2011-04-19 23:44:24 <ByteCoin> It was a revolutiion
3446 2011-04-19 23:44:28 <hubb> luke-jr: after the fact, meaning after a successful trade?
3447 2011-04-19 23:44:41 <ByteCoin> Well currently you need to securely distribute receiving addresses.
3448 2011-04-19 23:45:00 <luke-jr> hubb: meaning someone else can go remove it if they think it's dishonest
3449 2011-04-19 23:45:04 <BlueMatt> ByteCoin: and my problem is that that is easy to do
3450 2011-04-19 23:45:18 <ByteCoin> How do you do it?
3451 2011-04-19 23:45:54 <BlueMatt> ByteCoin: current terrorist-methods ie public key crypto or maybe even steganography
3452 2011-04-19 23:46:12 <gjs278> use giant suitcases of money instead
3453 2011-04-19 23:46:12 blablaa has quit ()
3454 2011-04-19 23:46:30 <ByteCoin> Let's steer away from terrorists to better recipients.. Say you wanted to donate bitcoins to wikileaks...
3455 2011-04-19 23:46:46 <gjs278> create a new address, send the money
3456 2011-04-19 23:47:03 <ByteCoin> Let's say all wikileaks websites were monitored
3457 2011-04-19 23:47:11 <ByteCoin> How would wikileaks get the address out?
3458 2011-04-19 23:47:22 <ByteCoin> The address you need to transfer to?
3459 2011-04-19 23:47:23 <lfm_> ByteCoin: you could use pgp email
3460 2011-04-19 23:47:29 <BlueMatt> ByteCoin: the problem I have is that I'm not convinced the nsa cant prove Q belongs to Z given Q and Z
3461 2011-04-19 23:47:54 eao has quit (Quit: Leaving)
3462 2011-04-19 23:48:11 <gjs278> generate the address as an image
3463 2011-04-19 23:48:13 <gjs278> and serve out the image
3464 2011-04-19 23:48:21 <BlueMatt> ByteCoin: wikileaks: create infinite receiving addresses and send them around
3465 2011-04-19 23:48:22 <ByteCoin> lfm: Woudln't it be nice not to have to bother. Wikileaks can publish their bitcoin receiving address anywhere and you can transfer money to it without being linked.
3466 2011-04-19 23:48:43 <ByteCoin> BlueMatt: Wikileaks sends them around how?
3467 2011-04-19 23:48:50 <BlueMatt> ByteCoin: though wikileaks is a good use case for your idea
3468 2011-04-19 23:48:57 <BlueMatt> I'm still not convinced it is secure...
3469 2011-04-19 23:49:18 <lfm_> ByteCoin: use something like mybitcoin.com . then the address is only linked to mybitcoin, not to any specific user
3470 2011-04-19 23:49:24 <ByteCoin> Bluematt: Break it and you get an instant professorship
3471 2011-04-19 23:49:38 Cusipzzz has joined
3472 2011-04-19 23:49:49 <BlueMatt> oh wait sorry...hadn't read it right the first time
3473 2011-04-19 23:49:55 <ByteCoin> lfm: Centralized system.. vulnerable to being subborned by feds etc..
3474 2011-04-19 23:50:20 <JFK911> ;;bc,mtgox
3475 2011-04-19 23:50:22 <gribble> {"ticker":{"high":1.1979,"low":1.1436,"vol":12517,"buy":1.174,"sell":1.1918,"last":1.174}}
3476 2011-04-19 23:50:36 <BlueMatt> ByteCoin: ok, you've convinced me. Its a cool idea. For most of us though, it has no use case. For a select few (wikileaks, etc) it might be useful
3477 2011-04-19 23:50:45 <lfm_> ByteCoin: ok so pgp email is too much trouble but mybitcoin is not secure enuf, I dont see where you're going any more.
3478 2011-04-19 23:50:45 <ByteCoin> This conversation is very handy for me guys. Thanks already...
3479 2011-04-19 23:51:13 <gjs278> secure hasslefree pick one
3480 2011-04-19 23:51:23 <ByteCoin> lfm: Pgp requires an active user on the wikileaks side distributing addresses
3481 2011-04-19 23:51:36 <luke-jr> ;;rate kirix 1 passed my fraud checks, so I sold him 20 BTC
3482 2011-04-19 23:51:36 <gribble> Rating entry successful. Your rating of 1 for user kirix has been recorded.
3483 2011-04-19 23:51:48 <luke-jr> woops, wrong chan
3484 2011-04-19 23:51:50 <luke-jr> ^^
3485 2011-04-19 23:51:58 <ByteCoin> My scheme the wikileaks receiving address can be fixed forever and they just redeem the coins you send them!
3486 2011-04-19 23:52:00 <gjs278> that's all you have to do
3487 2011-04-19 23:52:12 <BlueMatt> luke-jr: you have fraud checks?
3488 2011-04-19 23:52:15 <gjs278> ;;rate luke-jr 99 a+++ trader
3489 2011-04-19 23:52:21 <luke-jr> BlueMatt: yes
3490 2011-04-19 23:52:29 <BlueMatt> like...?
3491 2011-04-19 23:52:36 <luke-jr> BlueMatt: mainly, I verify the location of IP, PayPal, and phone number match up
3492 2011-04-19 23:52:43 <luke-jr> and I talk to them on the phone to confirm all the details
3493 2011-04-19 23:52:46 <gjs278> you ask for someone's phone number for bitcoins
3494 2011-04-19 23:52:47 <luke-jr> and I make sure it's a cell phone
3495 2011-04-19 23:53:00 <luke-jr> gjs278: first-time sale, ye
3496 2011-04-19 23:53:04 <BlueMatt> luke-jr: why does it have to be a cell?
3497 2011-04-19 23:53:12 <luke-jr> BlueMatt: so it can't be some payphone
3498 2011-04-19 23:53:19 <BlueMatt> ah
3499 2011-04-19 23:53:21 <gjs278> well
3500 2011-04-19 23:53:21 <lfm_> ByteCoin: the web site could give out different addresses to every visitor
3501 2011-04-19 23:53:25 <gjs278> you can't even call most payphones
3502 2011-04-19 23:53:33 <luke-jr> gjs278: here you can
3503 2011-04-19 23:53:34 <ByteCoin> lfm: Requires a running website
3504 2011-04-19 23:53:42 <gjs278> lies, there are no payphones anymore
3505 2011-04-19 23:53:43 <lfm_> huh
3506 2011-04-19 23:53:57 <lfm_> wikileaks isnt a website now?
3507 2011-04-19 23:53:58 <gjs278> lfm_: requires internet, your idea fails
3508 2011-04-19 23:54:04 <ByteCoin> lfm: My scheme requires an entry in a newspaper or flyer or word of moutn
3509 2011-04-19 23:54:10 <luke-jr> gjs278: bitcoin requires internt
3510 2011-04-19 23:54:14 <gjs278> that's the joke
3511 2011-04-19 23:54:19 chef_ has joined
3512 2011-04-19 23:54:44 <BlueMatt> lfm_: problem is if wikileaks publishes addresses, a couple screenshots will get out or a couple people will start pushing a couple addresses which will get the most donations
3513 2011-04-19 23:54:57 <ByteCoin> ljr: Bitcoin requires internet but there's no big arrow going "Wikileaks using the internet here now"
3514 2011-04-19 23:55:05 <ByteCoin> Couild be anyone
3515 2011-04-19 23:55:20 <gjs278> we need to start printing out bitcoins
3516 2011-04-19 23:55:38 <BlueMatt> gjs278: print out some scratch-off cards
3517 2011-04-19 23:55:39 <ByteCoin> BlueMatt: Don't understand last comment to lfm
3518 2011-04-19 23:56:01 <luke-jr> BlueMatt: why would I trust it? ;)
3519 2011-04-19 23:56:02 <gjs278> I just write down the contents of my wallet.dat
3520 2011-04-19 23:56:17 <BlueMatt> ie if wikileaks starts putting out a new address for each visitor a couple addresses will end up getting tweeted and screenshotted and listed everywhere
3521 2011-04-19 23:56:35 <BlueMatt> luke-jr: it was mostly sarcastic, I know the problems with scratch-off
3522 2011-04-19 23:56:55 <chef_> im in the low stakes no limit at betco.in if anyone wants to play
3523 2011-04-19 23:56:59 <lfm_> ok so some donations might get traced, big deal. your saying it unavoidable
3524 2011-04-19 23:57:16 <gjs278> the donations getting traced is only if you're stupid enough to use someone elses donate address too
3525 2011-04-19 23:57:25 <BlueMatt> lfm_: no his idea means the donations cant be traced
3526 2011-04-19 23:57:26 <gjs278> they could just put a giant disclaimer telling people to always gen one before donating
3527 2011-04-19 23:57:45 <lfm_> BlueMatt: they should be tweeting the web page to get the addresses
3528 2011-04-19 23:57:59 <ByteCoin> gis278 surely joking
3529 2011-04-19 23:58:06 <BlueMatt> lfm_: what people should be doing usually has little bearing on what they actually do
3530 2011-04-19 23:58:13 <gjs278> I type a lot of jokes
3531 2011-04-19 23:58:27 <ByteCoin> gis: Funny now...
3532 2011-04-19 23:58:52 Teslah has joined
3533 2011-04-19 23:59:00 <gjs278> if wikileaks generates an address for each donator, and you donate through that address and never tell anyone, what's the risk
3534 2011-04-19 23:59:01 <lfm_> BlueMatt: so what is his idea to solve this? I thot he just didnt like any of the ideas
3535 2011-04-19 23:59:10 <gjs278> that someone is intercepting their traffic and seeing all of the addresses they generate?
3536 2011-04-19 23:59:15 <BlueMatt> lfm_: http://www.bitcoin.org/smf/index.php?topic=5965.msg87757#msg87757
3537 2011-04-19 23:59:25 <ByteCoin> gis: Feds intercept outgoing wikileaks messages.
3538 2011-04-19 23:59:27 <BlueMatt> lfm_: the point is that he has solved it and hence the discussion
3539 2011-04-19 23:59:40 <ByteCoin> Feds know which transactions are crediting wikileaks
3540 2011-04-19 23:59:49 <gjs278> so then do lfm_'s idea and encrypt the address
3541 2011-04-19 23:59:55 <gjs278> and make the user decrypt it