1 2011-06-04 00:05:28 <Doc_M> Anyway a mac spesmilo seems like a worthy project, any thing I should know before I start playing with it?
  2 2011-06-04 00:06:01 <Doc_M> gaa wrong window again
  3 2011-06-04 00:06:11 <Doc_M> sorry folks, closing this window
  4 2011-06-04 00:11:43 <kermit> when its not cost effictive to make more bitcoins, will the transactions fees rise?
  5 2011-06-04 00:11:57 <Doc_M> Likely
  6 2011-06-04 00:12:11 <Doc_M> although its a competitive market stricture so they will likely stay small
  7 2011-06-04 00:19:04 <Doc_M> Oh spesmilo is mostly python? hells yes
  8 2011-06-04 00:21:32 <lfm> ERROR: AcceptToMemoryPool() : nonstandard transaction
  9 2011-06-04 00:21:52 <lfm> luke-jr is that you again?
 10 2011-06-04 00:23:21 <gjs278> ;;bc,mtgox
 11 2011-06-04 00:23:24 <gribble> {"ticker":{"high":14.499,"low":10.6,"vol":73811,"buy":14.38,"sell":14.39,"last":14.38}}
 12 2011-06-04 00:23:27 <gjs278> goddamn
 13 2011-06-04 00:23:36 <lfm> whew!
 14 2011-06-04 00:23:37 <luke-jr> lfm: no, but it might be someone using my branch
 15 2011-06-04 00:24:36 <lfm> at that price my phenom cpu might be economical again
 16 2011-06-04 00:25:39 <luke-jr> lfm: I sold at $20 earlier
 17 2011-06-04 00:26:09 <Doc_M> $20? I thought this was the all time high?
 18 2011-06-04 00:26:12 <lfm> seems crazy
 19 2011-06-04 00:26:23 <luke-jr> Doc_M: only 10 BTC tho
 20 2011-06-04 00:27:02 <lfm> stil $200 is not nothing
 21 2011-06-04 00:29:33 <lfm> no, my phenom isnt quite economical, darn
 22 2011-06-04 00:29:46 <luke-jr> ;;bc,calc 14000
 23 2011-06-04 00:29:47 <gribble> The average time to generate a block at 14000 Khps, given current difficulty of 434882.7217497 , is 4 years, 12 weeks, 0 days, 3 hours, 39 minutes, and 50 seconds
 24 2011-06-04 00:29:52 <luke-jr> ;;bc,gen 14000
 25 2011-06-04 00:29:53 <gribble> The expected generation output, at 14000 Khps, given current difficulty of 434882.7217497 , is 0.032380217985 BTC per day and 0.00134917574938 BTC per hour.
 26 2011-06-04 00:30:10 <luke-jr> $0.48 USD/day?
 27 2011-06-04 00:30:24 <luke-jr> I think my i5 is
 28 2011-06-04 00:30:44 <Doc_M> heh yah
 29 2011-06-04 00:31:17 <luke-jr> ;;bc,stats
 30 2011-06-04 00:31:19 <gribble> Current Blocks: 128466 | Current Difficulty: 434882.7217497 | Next Difficulty At Block: 129023 | Next Difficulty In: 557 blocks | Next Difficulty In About: 3 days, 3 hours, 11 minutes, and 42 seconds | Next Difficulty Estimate: 529521.30017713
 31 2011-06-04 00:31:25 <luke-jr> srsly? difficulty is going up?
 32 2011-06-04 00:31:28 <Doc_M> yes
 33 2011-06-04 00:31:37 <Doc_M> when price goes up more people mine
 34 2011-06-04 00:31:42 <Doc_M> so difficulty rises
 35 2011-06-04 00:31:47 <lfm> ya like another 20%
 36 2011-06-04 00:31:56 <luke-jr> I thought people were giving up
 37 2011-06-04 00:32:10 <lfm> some give up, some come back
 38 2011-06-04 00:32:14 <luke-jr> I know I probably will be shutting down the Radeon soon
 39 2011-06-04 00:32:19 <luke-jr> too much heat
 40 2011-06-04 00:32:47 <lfm> maybe depends on there electricity rates (if they're logical) or what news they see if theyre not
 41 2011-06-04 00:32:51 <Doc_M> its a rather competitive market, so marginal cost should roughly equal price
 42 2011-06-04 00:33:06 <luke-jr> actually, how do you change the fan speed?
 43 2011-06-04 00:43:57 <lfm> aticonfig --pplib-cmd "set fanspeed 0 100"
 44 2011-06-04 00:44:08 <Diablo-D3> 100% sucks
 45 2011-06-04 00:44:16 <lfm> change the 100 to percent speed
 46 2011-06-04 00:44:28 <Doc_M> I was told that 84% was good
 47 2011-06-04 00:44:41 <Doc_M> I have no personal experience to back that up
 48 2011-06-04 00:44:48 <lfm> depends on a lot of things
 49 2011-06-04 00:47:59 <luke-jr> what temp is too hot?
 50 2011-06-04 00:48:24 <jgarzik> Doc_M: http://forum.bitcoin.org/index.php?topic=10062.0  it's a pretty dumb script, but it works
 51 2011-06-04 00:49:23 <gjs278> I don't go above 80
 52 2011-06-04 00:49:28 <gjs278> even if I'm at 80 I turn my fan up
 53 2011-06-04 00:50:12 <luke-jr> erm
 54 2011-06-04 00:50:20 <luke-jr> how do you stay under 80?
 55 2011-06-04 00:50:32 <luke-jr> even with my fans at 100%, I can't get that low
 56 2011-06-04 00:50:41 <gjs278> my 5970 is at 70C right now
 57 2011-06-04 00:50:47 <gjs278> with 100% fan
 58 2011-06-04 00:50:47 <lfm> 80 celcius
 59 2011-06-04 00:51:01 <luke-jr> lfm: yeah, I'm talking C
 60 2011-06-04 00:51:12 <Doc_M> tried spacing your cards?
 61 2011-06-04 00:51:16 <gjs278> 5870 is at 50C but that's due to aftermarket cooler
 62 2011-06-04 00:51:19 <luke-jr> I only have one 5850
 63 2011-06-04 00:51:28 <gjs278> the gigabyte 5850 stays really cool
 64 2011-06-04 00:51:33 <Doc_M> have a crappy case?
 65 2011-06-04 00:51:38 <luke-jr> a case is a case
 66 2011-06-04 00:52:01 <Doc_M> not really
 67 2011-06-04 00:52:06 <gjs278> if my room gets hot enough I eventually hit 80C but at that point my cpu is idling around 40C so I open a window or something
 68 2011-06-04 00:52:09 <Doc_M> cases affect airflow
 69 2011-06-04 00:52:35 <luke-jr> room temp is 27 C
 70 2011-06-04 00:52:50 <luke-jr> outside temp is 25 C
 71 2011-06-04 00:53:23 <luke-jr> fans at 100% seem to settle the 5850 around  84 C
 72 2011-06-04 00:53:26 <gjs278> when my room temp hits 27C I definitely start to see 80C
 73 2011-06-04 01:06:59 <da2ce7> I'm back
 74 2011-06-04 01:09:11 <io_error> da2ce7: Welcome back.
 75 2011-06-04 01:12:53 <Xenland> whats up chat
 76 2011-06-04 01:13:20 <Xenland> When is the time set, on the "listtransactions" when querying bitcoind?
 77 2011-06-04 01:14:33 <lfm> txn time is (i think) set by the miner when it finds the block that txn is in
 78 2011-06-04 01:15:51 <SerajewelKS> lfm: no, the time is part of the transaction and cannot be changed once created, since that would make the signatures invalid
 79 2011-06-04 01:16:21 <lfm> but there is no time field in the txn, only in the block header
 80 2011-06-04 01:18:13 <lfm> so its not part of the txn afaik
 81 2011-06-04 01:20:03 <GarrettB> smart people
 82 2011-06-04 01:20:11 <GarrettB> can someone tell me if this is correct?
 83 2011-06-04 01:20:15 <GarrettB> the ECDSA part
 84 2011-06-04 01:20:20 <Diablo-D3> btw
 85 2011-06-04 01:20:23 <Diablo-D3> even if they ARE broken
 86 2011-06-04 01:20:27 <Diablo-D3> it doesnt effect us directly
 87 2011-06-04 01:20:51 <Diablo-D3> at this point in time, the only realistic breakage just shaves a few years off something that'd take a hundred
 88 2011-06-04 01:20:52 <GarrettB> Diablo-D3: I know, writing an article on that for The Bitcoin Sun :)
 89 2011-06-04 01:20:58 <lfm> well it depends what level of "break" it is I spoze
 90 2011-06-04 01:21:11 <Diablo-D3> and hell
 91 2011-06-04 01:21:21 <Diablo-D3> if sha256 gets broken, but doesnt produce collisions
 92 2011-06-04 01:21:23 <Diablo-D3> that does nothing
 93 2011-06-04 01:21:24 <Doc_M> eventually there may come a point where breaking ECDSA may be profitable as bitcoins deflate
 94 2011-06-04 01:21:27 <Diablo-D3> it just means the diff goes up
 95 2011-06-04 01:21:37 <Diablo-D3> and guess what happens if it doesnt get broken?
 96 2011-06-04 01:21:40 <Diablo-D3> the diff goes up
 97 2011-06-04 01:21:40 <Doc_M> and we would need to increase the strength
 98 2011-06-04 01:21:44 <Diablo-D3> no matter what
 99 2011-06-04 01:21:46 <Diablo-D3> the diff goes up
100 2011-06-04 01:21:49 <Diablo-D3> its a fact of life
101 2011-06-04 01:21:59 <Doc_M> for sha256 yes, but the ECDSA is a different matter?
102 2011-06-04 01:22:02 <GarrettB> Doc_M: you'd have to be very sneaky, if people caught wind that coins were being stolen they would become worthless overnight
103 2011-06-04 01:22:03 <Diablo-D3> Doc_M: erm, ecdsa is what keeps our wallets secure
104 2011-06-04 01:22:08 <lfm> and it wouldnt be all that hard to change over to other algorithms really
105 2011-06-04 01:22:18 <Diablo-D3> and ecdsa is NOT a conventional algo
106 2011-06-04 01:22:20 <Diablo-D3> by any means
107 2011-06-04 01:22:27 <Diablo-D3> its a real mindfuck on how clever it is
108 2011-06-04 01:22:46 <GarrettB> hold up
109 2011-06-04 01:22:49 <Diablo-D3> if you can break ecdsa legitimately, then you might as well go build a fucking warp engine
110 2011-06-04 01:22:56 <GarrettB> There has been a successful timing attack on ECDSA, but it is not a general attack, and requires certain very specific circumstances, which don?t occur on the bitcoin network.
111 2011-06-04 01:22:59 <GarrettB> is that correct
112 2011-06-04 01:23:01 <GarrettB> y/n
113 2011-06-04 01:23:18 <Diablo-D3> GarrettB: yes BUT
114 2011-06-04 01:23:30 <Diablo-D3> GarrettB: bitcoin doesnt use ecdsa to talk to itself
115 2011-06-04 01:23:36 <Diablo-D3> its completely offline in usage
116 2011-06-04 01:23:37 <lfm> GarrettB: ya timeing attacks would need to be onb the premisses of the attacked machine I think
117 2011-06-04 01:24:01 <GarrettB> Diablo-D3: right, but if ECDSA is broken in a certain way, someone could generate private keys from a public key, correct?
118 2011-06-04 01:24:08 <Doc_M> I mean rather, at some point it becomes profitable to crack someone's account using a supercomputer to bruitforce it
119 2011-06-04 01:24:21 <cuddlefish> GarrettB: basically
120 2011-06-04 01:24:32 <Doc_M> so you shouldn't let your accounts get too big
121 2011-06-04 01:24:41 <Doc_M> by account I meant wallet, sorry
122 2011-06-04 01:24:46 <Diablo-D3> GarrettB: yes, but it may not be the CORRECT private key
123 2011-06-04 01:24:47 <lfm> GarrettB: well completely hypothetical ya, I guess it could happen,
124 2011-06-04 01:25:05 <Diablo-D3> GarrettB: since if ecdsa can be broken, you might be looking at a huge tub of shit.
125 2011-06-04 01:25:07 <cuddlefish> GarrettB: look
126 2011-06-04 01:25:10 <Diablo-D3> PLUS
127 2011-06-04 01:25:13 <Diablo-D3> if ecdsa is broken
128 2011-06-04 01:25:16 <Diablo-D3> world goes into chaos
129 2011-06-04 01:25:18 <Diablo-D3> same way with sha256
130 2011-06-04 01:25:19 <cuddlefish> GarrettB: 1. It's only in the implementation, not the algorith
131 2011-06-04 01:25:20 <cuddlefish> m
132 2011-06-04 01:25:32 <Diablo-D3> no one will care about bitcoins when nukes are firing out of their silos
133 2011-06-04 01:25:32 <lfm> Doc_M: well that depends how you measure the risk. bitcoin may be safer than a bank!
134 2011-06-04 01:25:36 <Doc_M> of course ecdsa can be broken, the question is how much computing power it takes
135 2011-06-04 01:25:39 <Doc_M> agreed LFM
136 2011-06-04 01:25:43 <Diablo-D3> Doc_M: no
137 2011-06-04 01:25:48 <Diablo-D3> Doc_M: broken has a very specific meaning
138 2011-06-04 01:25:54 <Doc_M> I meant bruitforced
139 2011-06-04 01:25:55 <Diablo-D3> it means you found a shortcut for brute forcing it
140 2011-06-04 01:26:09 <Doc_M> yes, that just lowers the difficulty
141 2011-06-04 01:26:13 <Diablo-D3> for sha256 yes
142 2011-06-04 01:26:15 <Diablo-D3> BUT
143 2011-06-04 01:26:23 <Diablo-D3> if sha256 suddenly has collisions
144 2011-06-04 01:26:29 <Diablo-D3> even the banking industry will shit pants
145 2011-06-04 01:26:38 <cuddlefish> GarrettB: So yeah. It's only the OpenSSL ECDSA that's at issue
146 2011-06-04 01:26:40 <Diablo-D3> I dont mean merely shit in their pants
147 2011-06-04 01:26:43 <Doc_M> can sha256 could be broken by quantum computing?
148 2011-06-04 01:26:44 <cuddlefish> GarrettB: so it will be patched
149 2011-06-04 01:26:46 <Diablo-D3> I mean they will shit pressed dress pants
150 2011-06-04 01:26:49 <Diablo-D3> piles of them
151 2011-06-04 01:27:01 <lfm> thats why I say there are different levels of "breaking" like the old DES was never really broke, it just got real cheap to brute force it
152 2011-06-04 01:27:18 <Doc_M> yah
153 2011-06-04 01:27:27 <Doc_M> well, it had a few shortcuts
154 2011-06-04 01:27:28 <Diablo-D3> also, length extension attacks obviously wont and cannot work on sha256
155 2011-06-04 01:27:34 <Diablo-D3> theres been rumors of useful ones of those floating around
156 2011-06-04 01:27:47 <Diablo-D3> double sha256, and its always a 80 byte header anyhow
157 2011-06-04 01:28:18 <GarrettB> Diablo-D3: can I quote you on the warp engine thing :P
158 2011-06-04 01:28:23 <Diablo-D3> GarrettB: sure.
159 2011-06-04 01:28:31 <GarrettB> neat, what name should I use?
160 2011-06-04 01:28:33 <Diablo-D3> see also: a fucking tardis.
161 2011-06-04 01:28:37 <Diablo-D3> GarrettB: /whois me
162 2011-06-04 01:28:39 <GarrettB> haha
163 2011-06-04 01:28:47 <GarrettB> great, thanks a lot guys
164 2011-06-04 01:29:03 <davex__> wtf
165 2011-06-04 01:29:13 <davex__> so mtgox will sometimes sell your coins for less than what you entered as ask price?
166 2011-06-04 01:29:14 <lfm> and if they suddenly make quantum computers work we could be in trouble too
167 2011-06-04 01:29:31 <Diablo-D3> davex__: not that Im aware of
168 2011-06-04 01:29:51 <davex__> hmm...  quite sure i entered 14.45, sold for 14.35
169 2011-06-04 01:29:59 <Diablo-D3> thats an easy typo to make
170 2011-06-04 01:30:00 <lfm> davex__: its not spozed to
171 2011-06-04 01:30:02 <davex__> saw the order sitting there
172 2011-06-04 01:30:04 <davex__> at 14.45
173 2011-06-04 01:30:08 <cuddlefish> davep: fee?
174 2011-06-04 01:30:10 <davex__> actually
175 2011-06-04 01:30:10 <Diablo-D3> weird, pester tux
176 2011-06-04 01:30:13 <davex__> oh duh.
177 2011-06-04 01:30:14 <davex__> haha
178 2011-06-04 01:30:17 <davex__> forgot about the fee
179 2011-06-04 01:30:27 <Diablo-D3> er, the fee shouldnt be factored into that yet
180 2011-06-04 01:30:37 <Diablo-D3> then again, it doesnt have a third fee column
181 2011-06-04 01:30:39 <io_error> da2ce7: What is it you want done?
182 2011-06-04 01:30:41 <davex__> i think it is...  i had this happen one other time
183 2011-06-04 01:30:48 <davex__> then forgot about it
184 2011-06-04 01:31:00 <da2ce7> io_error: two projects...
185 2011-06-04 01:31:17 <io_error> da2ce7: And they're secret?
186 2011-06-04 01:31:25 <da2ce7> https://forum.bitcoin.org/index.php?topic=8379.0
187 2011-06-04 01:31:28 <da2ce7> https://forum.bitcoin.org/index.php?topic=11665.0
188 2011-06-04 01:31:40 <Diablo-D3> mtrlt: I s/Ma(/Ma2(/ except the #define, and the bug is gone
189 2011-06-04 01:31:46 <Diablo-D3> mtrlt: so I suspect another one needs replaced
190 2011-06-04 01:31:53 <Diablo-D3> mtrlt: interestingly, Im at 355 instead of 366.
191 2011-06-04 01:49:47 <Doc_M> If we were to switch to SHA-512 would that change the number of possible coins?  Someone on a forum said it would, but that doesn't make sense to me.
192 2011-06-04 01:53:02 <chuck> Is there anyone who can give me a quick overview of the techincal details of the bitcoin protocol/network?
193 2011-06-04 01:53:22 <chuck> I'm looking into a ruby implementation of sorts and it doesn't seem like the wiki is giving me the full picture on what goes on behind the scenes
194 2011-06-04 01:53:35 <chuck> (and the official bitcoin client reference is kind of daunting :P)
195 2011-06-04 01:58:35 <Doc_M> ok
196 2011-06-04 01:58:56 <Doc_M> it would be easier if you ask questions?
197 2011-06-04 01:59:10 <Doc_M> what part are you curious about, the client? the mining?
198 2011-06-04 02:01:08 <chuck> Doc_M, not really mining so much, more of the client. What does a client need to do to "join the network" when it starts up, does it need to listen on a port, does it need to connect to other servers, etc?
199 2011-06-04 02:02:37 <Doc_M> it can find them using peer to peer stuff or can join an IRC channel
200 2011-06-04 02:04:06 <chuck> Doc_M, so it joins an IRC channel, finds a couple nodes to connect to, and sends them a "version" message?
201 2011-06-04 02:05:24 <Doc_M> I am unsure about the details of the client, sorry
202 2011-06-04 02:05:48 <Doc_M> although i'm starting to try to port a third party client to mac, so in a couple weeks I should be more convesant
203 2011-06-04 02:07:36 <luke-jr> Doc_M: Spesmilo is just a client, not a node/wallet ;)
204 2011-06-04 02:09:02 <grondilu> I'm trying to compile bitcoind and the linker complains about missing gthread lib.  Yet I don't see any libgthread packet in my debian repo  :(
205 2011-06-04 02:09:56 <Doc_M> ah, it doesn't actually handle any wallet transactions?
206 2011-06-04 02:10:32 <grondilu> yet there IS a /usr/lib/libgthread-2.0.so.0 file.  I don't get it.
207 2011-06-04 02:10:56 <Doc_M> goes back to rtfm
208 2011-06-04 02:11:51 <luke-jr> Doc_M: you connect it to a bitcoind (or it can manage its own bitcoind)
209 2011-06-04 02:14:02 <Doc_M> ah
210 2011-06-04 02:18:31 <grondilu> ok found it
211 2011-06-04 02:19:59 <Doc_M> yay!
212 2011-06-04 02:29:57 <Doc_M> btw is there any chance in the world we might move to an every minute a block with a smaller payout instead of 50 btc every 10 minutes
213 2011-06-04 02:30:25 <EvanR> can the client send from a specific address
214 2011-06-04 02:30:36 <EvanR> it seems to randomly choose source addresses
215 2011-06-04 02:31:31 <EvanR> Doc_M: point of that?
216 2011-06-04 02:33:56 <Doc_M> faster integration into the chain?
217 2011-06-04 02:34:00 <Doc_M> I'm just curious
218 2011-06-04 02:34:29 <EvanR> its already pretty fast
219 2011-06-04 02:34:49 <io_error> Doc_M: Any less than 10 minutes and you'd have more frequent reorgs
220 2011-06-04 02:35:00 <Doc_M> fair enough
221 2011-06-04 02:35:05 <EvanR> if youre thinking of point of sale, better to just not wait for confirmations
222 2011-06-04 02:35:23 <Doc_M> yah, I was thinking of POS apps
223 2011-06-04 02:35:26 <io_error> Right, retail doesn't need to wait for confirmations
224 2011-06-04 02:35:32 <io_error> It just needs to see the tx on the network
225 2011-06-04 02:36:05 <Doc_M> how is that more secure than cheques then?
226 2011-06-04 02:36:22 <Doc_M> I guess you could collect credit card info so the person doesn't defraud you?
227 2011-06-04 02:36:31 <io_error> Doc_M: The TX would be rejected from the network if the addresses didn't have enough money
228 2011-06-04 02:36:44 <Doc_M> I mean brick and mortar POS
229 2011-06-04 02:37:09 <Doc_M> got yah
230 2011-06-04 02:37:20 <Doc_M> so the only purpose of confirmations is to prevent cheating by miners?
231 2011-06-04 02:37:25 <io_error> It would never even get to 0/unconfirmed, it would just go straight into /dev/null
232 2011-06-04 02:37:53 <io_error> Doc_M: That's one way to look at it :)
233 2011-06-04 02:38:27 <Doc_M> I am using cheating very broadly, ok
234 2011-06-04 02:38:44 <scott`> ;;bc,stats
235 2011-06-04 02:38:46 <gribble> Current Blocks: 128489 | Current Difficulty: 434882.7217497 | Next Difficulty At Block: 129023 | Next Difficulty In: 534 blocks | Next Difficulty In About: 3 days, 0 hours, 14 minutes, and 18 seconds | Next Difficulty Estimate: 532025.65085018
236 2011-06-04 02:39:31 <EvanR> There is no way to ask Bitcoin to "create a payment transaction using the coins received from these previously received transactions.
237 2011-06-04 02:39:56 <EvanR> ""
238 2011-06-04 02:40:19 <EvanR> mmmm
239 2011-06-04 02:40:25 <EvanR> dont understand that
240 2011-06-04 02:40:43 <EvanR> seems obvious to make a raw transaction from any addresses you own
241 2011-06-04 02:41:10 <EvanR> of your choice
242 2011-06-04 02:42:54 <io_error> EvanR: Sure you can do that, but you'd have to code it up yourself.
243 2011-06-04 02:43:21 <kermit> EvanR: use different wallets
244 2011-06-04 02:43:22 <EvanR> ok
245 2011-06-04 02:43:33 <EvanR> so its not a technical fundamental limitation
246 2011-06-04 02:44:05 <EvanR> different wallets on the same computer?
247 2011-06-04 02:44:21 <kermit> EvanR: im not sure how easy that is to do
248 2011-06-04 02:49:55 <EvanR> how the hell does 'send to ip address' work ?
249 2011-06-04 02:53:38 <DavidSJ> I think it just connects to the bitcoin peer at that IP address and negotiates the rest.
250 2011-06-04 02:53:38 <Xenland> It dosen't
251 2011-06-04 02:53:41 <DavidSJ> But I could be wrong.
252 2011-06-04 02:53:45 <DavidSJ> I don't really trust it. :)
253 2011-06-04 02:54:23 <Xenland> whats a good formula for calculating the reward for share based
254 2011-06-04 02:55:30 <Xenland> EvanR: i think you can use bitcoin-php to send bitcoins from a specific addess/account in your wallet
255 2011-06-04 02:55:43 <doublec> usershare * (50 - fee)/totalshares
256 2011-06-04 03:03:42 <ZOP> Not an advert but it's going to sound like one...I knew the Zalman VF3000A would make a big difference on my GPU temps, but i didn't think I'd get down from 88C to barely 60-61C
257 2011-06-04 03:04:19 <ZOP> the thermal tape they include definitely leaves MUCH to be desired though, had to replace it on one of the ramsinks.
258 2011-06-04 03:04:32 <ZOP> (didn't get a good bond first time, then it was SOL)
259 2011-06-04 03:05:49 <ZOP> and even at full speed, the fans are lots quieter than the stock cooler.  whole world of difference.
260 2011-06-04 03:18:55 <Xenland> ;;bc,gen 800000
261 2011-06-04 03:18:57 <gribble> The expected generation output, at 800000 Khps, given current difficulty of 434882.7217497 , is 1.85029817057 BTC per day and 0.0770957571072 BTC per hour.
262 2011-06-04 03:22:18 <Xenland> ;;bc, gen 85000
263 2011-06-04 03:22:19 <gribble> Error: "bc," is not a valid command.
264 2011-06-04 03:22:27 <Xenland> ;;bc,gen 85000
265 2011-06-04 03:22:28 <gribble> The expected generation output, at 85000 Khps, given current difficulty of 434882.7217497 , is 0.196594180623 BTC per day and 0.00819142419264 BTC per hour.
266 2011-06-04 03:23:06 <scott`> ;;bc,gen 1200000
267 2011-06-04 03:23:07 <gribble> The expected generation output, at 1200000 Khps, given current difficulty of 434882.7217497 , is 2.77544725586 BTC per day and 0.115643635661 BTC per hour.
268 2011-06-04 04:12:26 <RenaKunisaki> hmm. bitcoin CLI reports a balance of zero if it hasn't downloaded any blocks yet
269 2011-06-04 04:12:39 <RenaKunisaki> had me worried for a sec... maybe should report 'unknown' or something?
270 2011-06-04 04:14:47 <DavidSJ> RenaKunisaki: at what block should it consider the balance known?
271 2011-06-04 04:14:50 <DavidSJ> Not 1, right?
272 2011-06-04 04:14:58 <DavidSJ> Not 2, right?
273 2011-06-04 04:14:58 <RenaKunisaki> once it has them all?
274 2011-06-04 04:15:04 <DavidSJ> And how does it know when it has them all?
275 2011-06-04 04:15:12 <RenaKunisaki> doesn't it know?
276 2011-06-04 04:15:36 <RenaKunisaki> at least, can't it remember how many it had last time it shut down?
277 2011-06-04 04:15:39 <DavidSJ> No, for all it knows it's in a network partition with a bunch of peers which don't have the latest blocks.
278 2011-06-04 04:15:52 <RenaKunisaki> I see
279 2011-06-04 04:16:11 <DavidSJ> Yes, and it can also remember the contents of those blocks, too.  But it sounds like you're talking about a cold start situation where it never had that info.
280 2011-06-04 04:16:27 <DavidSJ> What you could do is not go by block number but by timestamp: if it's within the last 60 minutes, for example.
281 2011-06-04 04:19:31 <DavidSJ> night
282 2011-06-04 04:19:37 <RenaKunisaki> nightr
283 2011-06-04 04:19:44 <RenaKunisaki> damn, missed.
284 2011-06-04 05:24:30 <jine> Anyone got the formula for calculation estimated earnings based on difficulty & hashrate?
285 2011-06-04 05:25:36 <jine> Got it, thanks anyway
286 2011-06-04 05:47:54 <lfm> jine: you still there?
287 2011-06-04 05:48:47 <lfm> ;;bc,calc 300000
288 2011-06-04 05:48:48 <gribble> The average time to generate a block at 300000 Khps, given current difficulty of 434882.7217497 , is 10 weeks, 2 days, 1 hour, 27 minutes, and 3 seconds
289 2011-06-04 05:49:00 <Doc_M> wow
290 2011-06-04 05:49:06 <lfm> ;;bc,gen 300000
291 2011-06-04 05:49:07 <gribble> The expected generation output, at 300000 Khps, given current difficulty of 434882.7217497 , is 0.693861813965 BTC per day and 0.0289109089152 BTC per hour.
292 2011-06-04 05:49:24 <jine> lfm: Yeah, im here.
293 2011-06-04 05:49:35 <jine> I know how to calculate it HERE, but not the math behind it.
294 2011-06-04 05:49:38 <jine> ;;bc,gen
295 2011-06-04 05:49:39 <gribble> (bc,gen <an alias, 1 argument>) -- Alias for "echo The expected generation output, at $1 Khps, given current difficulty of [bc,diff], is [math calc 50*24*60*60 / (1/((2**224-1)/[bc,diff]*$1*1000/2**256))] BTC per day and [math calc 50*60*60 / (1/((2**224-1)/[bc,diff]*$1*1000/2**256))] BTC per hour.".
296 2011-06-04 05:49:39 <lfm> k see those?
297 2011-06-04 05:49:43 <jine> But that teached me it ;)
298 2011-06-04 05:49:52 <jine> Got it from #bitcoin-mining
299 2011-06-04 05:49:57 <jine> Thanks anyway
300 2011-06-04 05:50:07 <lfm> k, np
301 2011-06-04 06:24:51 <redss> a few of us over at #bitcoin were debating the best practice for anonymization of the source of one's BTC
302 2011-06-04 06:25:58 <redss> say you start out with an account full of BTC at mtgox, in order to hide the source of the coin, what would be best practice?  What benefit does an anonymization service have, for example, over just switching addresses a few times?
303 2011-06-04 06:30:18 <redss> If there's a link to the issues, that would also be greatly appreciated! ;)
304 2011-06-04 06:31:18 <RoboTeddy> is there anything written about how bitcoin could (theoretically) switch to a new hash, if sha256 proved to be weak a decade from now, or something?
305 2011-06-04 06:36:49 <edcba> RoboTeddy: shouldn't be very hard
306 2011-06-04 06:37:08 <edcba> at least for clients and cpu miners
307 2011-06-04 06:42:27 <eianpsego> edcba, wouldn't such a switch cause an immediate collapse in value?
308 2011-06-04 06:42:39 <vrs> RoboTeddy: client changes
309 2011-06-04 06:43:08 <vrs> eianpsego: you can build onto the existing blockchain, you just need to consent on where to
310 2011-06-04 06:43:31 <eianpsego> vrs, consent how? there is no centralization
311 2011-06-04 06:43:37 <vrs> redss: mixnets
312 2011-06-04 06:43:55 <vrs> eianpsego: there is, it's called client developers
313 2011-06-04 06:44:24 <eianpsego> vrs, that sounds like the "flag day" argument used for ipv6
314 2011-06-04 06:44:43 <eianpsego> problem* not argument
315 2011-06-04 06:45:09 <vrs> how so?
316 2011-06-04 06:45:48 <eianpsego> vrs, how can you form consensus between developers of different clients to switch on a specific day
317 2011-06-04 06:46:34 <vrs> there are like, two full featured clients out there afaik
318 2011-06-04 06:46:50 <vrs> the official one and bitcoinj
319 2011-06-04 06:47:08 <vrs> miners are another matter though
320 2011-06-04 06:47:14 <eianpsego> well, my assumption was that the underlying crypto wouldn't be broken for a few years
321 2011-06-04 06:47:30 <eianpsego> and I imagine that more clients will spring up
322 2011-06-04 06:47:32 <gmaxwell> 'few'
323 2011-06-04 06:47:57 <gmaxwell> RoboTeddy: there have been forum threads on making a migration.
324 2011-06-04 06:47:59 <eianpsego> maxwell, I was trying to be ultra conservative
325 2011-06-04 06:48:06 <vrs> i think bitcoin clients need a mechanism to switch to a new hash algorithm without updating the client
326 2011-06-04 06:48:39 <gmaxwell> vrs: then the 'new' hash whatever it is would itself be a security liability
327 2011-06-04 06:49:08 <samfisher> hi
328 2011-06-04 06:49:32 <samfisher> is it still ok to mine with graphic cards? what card should i buy?
329 2011-06-04 06:49:35 <gmaxwell> 01:45 < eianpsego> vrs, how can you form consensus between developers of different clients to switch on a specific day
330 2011-06-04 06:49:40 <gmaxwell> by block number.
331 2011-06-04 06:49:48 <gmaxwell> samfisher: #bitcoin-mining
332 2011-06-04 06:50:03 <gmaxwell> (not by day)
333 2011-06-04 06:50:19 <eianpsego> gmaxwell, is the idea that the mining pools would agree? Don't they have incentive to actively work against the devs?
334 2011-06-04 06:50:43 <gmaxwell> The past 30 years of cryptographic discoverys support the idea that no well tested alogorithim is likely to become insecure overnight.
335 2011-06-04 06:51:18 <gmaxwell> eianpsego: no. The mining pools income isn't improved by the hash being weak.
336 2011-06-04 06:51:35 <eianpsego> no no, I don't mean weak. I mean this idea of a flag day
337 2011-06-04 06:51:47 <eianpsego> I wouldn't switch to new software if I had a malicious intent
338 2011-06-04 06:52:14 <gmaxwell> eianpsego: if I made all the mining pools 4x faster starting tomorrow, they'd just burn though 2016 blocks quickly get a little extra money and then be exactly where they are now.
339 2011-06-04 06:52:31 <gmaxwell> eianpsego: not just miners, all clients.
340 2011-06-04 06:52:46 <gmaxwell> but anyone who doesn't change will be left behind most likely.
341 2011-06-04 06:52:59 <gmaxwell> eianpsego: so long as most of the hashpower does change.
342 2011-06-04 06:53:10 <vrs> even if sha256 gets weakened, it would have to be utterly owned and broken to break the double sha256 bitcoin is using
343 2011-06-04 06:53:20 <gmaxwell> indeed.
344 2011-06-04 06:53:34 <eianpsego> wait
345 2011-06-04 06:53:36 <RoboTeddy> how does double sha256 work? do you literally take hash(hash(value)) instead of hash(value) ?
346 2011-06-04 06:53:41 <vrs> yes
347 2011-06-04 06:53:44 <gmaxwell> Well, we already take advantage of a 'weakness' for the type of work we do.
348 2011-06-04 06:54:00 <RoboTeddy> for a weak hash like md5& is double-md5 still safe?
349 2011-06-04 06:54:14 <gmaxwell> we're able to precompute the inital expansion, and terminate 4 rounds early.
350 2011-06-04 06:54:20 <eianpsego> maxwell, the idea that they would be able to get a "little" extra money - wouldn't they be able to double spend their brains out for a blocks?
351 2011-06-04 06:54:56 <eianpsego> for a *few blocks
352 2011-06-04 06:55:01 <gmaxwell> eianpsego: that would be a lot of incentive for everyone else to install the software that enforces the new hash, now wouldn't it? :)
353 2011-06-04 06:55:19 <eianpsego> you mean the other mining pools?
354 2011-06-04 06:55:24 <lfm> RoboTeddy: yes it is sha256(sha256(blkhdr))
355 2011-06-04 06:55:40 <eianpsego> (my assumption is that the big pools form the majority of the network)
356 2011-06-04 06:55:45 <gmaxwell> Everyone who is honest, and then they'd ignore the dishonest nodes blocks.
357 2011-06-04 06:55:51 <RoboTeddy> lfm, vrs: thanks
358 2011-06-04 06:57:29 <lfm> eianpsego: the pools dont control the market price. if bitcoin is undermined by anyone the market would probably collapse and no one wins
359 2011-06-04 06:58:18 <eianpsego> lfm, if btc are exchanged for real currency, of course someone wins - the first guy (or pool) to exploit the system
360 2011-06-04 06:58:40 <lfm> or the last
361 2011-06-04 06:58:46 <gmaxwell> eianpsego: this is, of course speculation on top of speculation.
362 2011-06-04 06:58:59 <RoboTeddy> is it known how much computing power is currently in the network?
363 2011-06-04 06:59:02 <eianpsego> maxwell, I don't imagine this flag day will go without a hitch
364 2011-06-04 06:59:03 <vrs> yes
365 2011-06-04 06:59:27 <gmaxwell> Weaknesses in sha256 eventually are likely sure. But e.g. "omg infinite blocks for free" are not.
366 2011-06-04 06:59:29 <vrs> http://bitcoinwatch.com/
367 2011-06-04 06:59:54 <lfm> RoboTeddy: you cant know exactly for sure but you can estimate it. see www3.telus.net/millerlf/hashes.png
368 2011-06-04 07:00:15 <vrs> you can estimate it via the block rate
369 2011-06-04 07:00:38 <gmaxwell> eianpsego: it wouldn't be more economically damaging than someone taking over a passenger plane and flying it into a big building was, in any case.
370 2011-06-04 07:01:08 <lfm> gmaxwell: isnt that big enuf for ya?
371 2011-06-04 07:01:23 <gmaxwell> Survivable in any case.
372 2011-06-04 07:01:31 <RoboTeddy> hmm, anyone know how to translate btchashes/s into a more traditional metric for computational power? I'm wondering how the bitcoin network compares to the list top supercomputers
373 2011-06-04 07:01:35 <eianpsego> gmaxwell, my only reason for bringing this hypothetical crap up is the idea that this cutover would somehow be seemless and painless - like rainbows and kittens
374 2011-06-04 07:02:09 <lfm> RoboTeddy: well if you name a specific cpu or gpu we can tell you the hashes/s for that chip
375 2011-06-04 07:02:10 <gmaxwell> eianpsego: it would be if the attack is like every other attack on old and trusted crypto has been: useless at first and slowly becoming more useful.
376 2011-06-04 07:02:10 <vrs> you can't really map it 1:1, but some people have done calculations in the forums
377 2011-06-04 07:02:53 <gmaxwell> eianpsego: if you can make the transistion happen two years out then it's not worse e.g. than having to cope with things like daylight savings date changes or the like.
378 2011-06-04 07:03:16 <eianpsego> gmaxwell, that was an excellent point - hah
379 2011-06-04 07:03:27 <eianpsego> gmaxwell, well played :P
380 2011-06-04 07:03:28 <lfm> RoboTeddy: one problem is most supercomputing is measured in floating point operations (flops) whereas bitcoin is strictly fixed point so the hardware doesnt always performa predicably
381 2011-06-04 07:03:50 <RoboTeddy> lfm: ah, I see. so, apples to oranges
382 2011-06-04 07:04:21 <eianpsego> gmaxwell, I guess if the majority of people get DST right, they can figure out how to download the new f'ing software
383 2011-06-04 07:04:27 <lfm> RoboTeddy: yup, some supercomputers might be really terrible for bitcoin hashing
384 2011-06-04 07:04:42 <eianpsego> gmaxwell, congrats. you've shut me up lol
385 2011-06-04 07:04:52 <gmaxwell> eianpsego: there have been some flag-blockish things in bitcoin in the past, and I'm sure there will be more in the future, unrelated to crypto changes.. e.g. p2p behavior and the like. DOS mitigation.
386 2011-06-04 07:05:35 <lfm> and the overflow bug event
387 2011-06-04 07:06:26 <gmaxwell> oh yea.. ten billion coins for me please!
388 2011-06-04 07:06:29 <eianpsego> gmaxwell, in all honesty, I'd love to have this thing succeed
389 2011-06-04 07:07:00 <eianpsego> gmaxwell, I'm just worried about systemwide attacks of some sort
390 2011-06-04 07:07:09 <lfm> actually I think most people dont really understand dst and time zones! grin
391 2011-06-04 07:07:43 <gmaxwell> eianpsego: the core design has seen a lot of review. It looks solid. People will beat on rough edges and there will be some emergencies no doubt.. but that doesn't really mean any fundimental danger.
392 2011-06-04 07:08:12 <gmaxwell> eianpsego: things will be better off once there are more implementations, right now there is a lot of monoculture.
393 2011-06-04 07:08:13 <eianpsego> gmaxwell, I have discussed this system with a few people that work with U.S. defense agencies
394 2011-06-04 07:08:27 <eianpsego> gmaxwell, I'm told that if the U.S. wants to shut this down, they'd just buy out all the bitcoins
395 2011-06-04 07:08:37 <lfm> but you are welcome to try, just dont expect people to panic unless you have worked out the problem very thuroughly
396 2011-06-04 07:08:48 <gmaxwell> eianpsego: sounds good to me.
397 2011-06-04 07:09:09 <gmaxwell> eianpsego: more likely they'd get the media to attack it. BITCOIN: TERROR DOLLAR. ;)
398 2011-06-04 07:09:25 <eianpsego> gmaxwell, haha
399 2011-06-04 07:09:28 <lfm> or kidd7y porn dealers or something
400 2011-06-04 07:09:33 <RoboTeddy> all they need is a few headlines about bitcoin being used to transfer money around by terrorists, yeah
401 2011-06-04 07:09:41 <RoboTeddy> i guess we should probably move into #bitcoin-politics with this
402 2011-06-04 07:10:01 <eianpsego> my god...I didn't realize there was such a channel wow
403 2011-06-04 07:10:04 <gmaxwell> yea. Find some random actors to talk behind voice disguisers "Oh yes, I've bought 6 year old girls for bitcoin. Got me a good deal"
404 2011-06-04 07:10:24 <RoboTeddy> eianpsego: smarter people in here though ;)
405 2011-06-04 07:11:14 <RoboTeddy> when the US wanted to shut down online poker, they just made it really hard to exchange the poker currency for USD&. seems like they could easily do the same here (correct me if I'm wrong)
406 2011-06-04 07:11:35 <lfm> RoboTeddy: your right actually
407 2011-06-04 07:11:42 <eianpsego> robo, you might be right. The guy I spoke to was a crypto guy
408 2011-06-04 07:12:19 <gmaxwell> My thinking is still more along the lines of "the transition point between 'too small to worry about' and 'too big to fail' is fairly narrow"
409 2011-06-04 07:12:40 <RoboTeddy> the transition point in terms of time?
410 2011-06-04 07:12:41 <gmaxwell> whereas online poker wouldn't have been economically important enough to enough people to avoid blowing it up.
411 2011-06-04 07:12:51 <gmaxwell> RoboTeddy: time, level of usage, etc.
412 2011-06-04 07:13:36 <lfm> and gambling has an inherent aura of sinnfullness around it right off the start
413 2011-06-04 07:13:55 <gmaxwell> well, those opposeed to BTC will try to argue the same thing.
414 2011-06-04 07:13:55 <RoboTeddy> well so do terror kiddy molestation drug dollars
415 2011-06-04 07:14:00 <gmaxwell> yep
416 2011-06-04 07:14:24 <lfm> RoboTeddy: ya but thats an extra step from sin to bitcoin
417 2011-06-04 07:14:27 <RoboTeddy> the volumes involved in poker were likely /much/ higher than in bitcoin now
418 2011-06-04 07:14:39 <RoboTeddy> well, it's already an extra step for poker& poker is a game of skill
419 2011-06-04 07:14:46 <gmaxwell> I think the key point is that cash is actually better for all of these things, except that it's not online (by itself)... and bitcoin is generally somewhat less anonymous than cash.
420 2011-06-04 07:15:09 <lfm> RoboTeddy: yaya, betting on ponies is skill according to some people too
421 2011-06-04 07:15:29 <gmaxwell> RoboTeddy: narrower segment of the population though, or er. maybe not. But presumably bitcoin will be broader before anyone really carres.
422 2011-06-04 07:15:33 <RoboTeddy> lfm: yeah but poker actually IS skill& same people at the world championships every year, they don't all get there each time by luck :P
423 2011-06-04 07:16:19 <lfm> RoboTeddy: yet those people at the top readily admit there IS a portion of luck involved
424 2011-06-04 07:16:39 <RoboTeddy> lfm: well of course, there's a good deal of variance, but that's true in a lot of skill games
425 2011-06-04 07:16:59 <RoboTeddy> anyway we should probably stop talkign about this in #bitcoin-dev
426 2011-06-04 07:17:00 <eianpsego> lfm, I wonder if that is because it benefits those at the top to perpetuate that idea
427 2011-06-04 07:17:18 <lfm> RoboTeddy: anyway, betting on skill or luck is still gambling and still the same aura of sin
428 2011-06-04 07:18:13 <lfm> RoboTeddy: pure luck on lotteries is pure skill and yet somehow it has convinced people it is less sinfull
429 2011-06-04 07:18:28 <lfm> is pure non skill
430 2011-06-04 07:18:53 <RoboTeddy> mm I would agree that the actual amount of skill involved in an activity and the perceived morality of the activity are entirely separate
431 2011-06-04 07:21:11 <lfm> ya if anyone has any more on - topic questions please break in
432 2011-06-04 07:24:39 <RoboTeddy> there's only high variance when the players are somewhat matched in skill. if you put a top-level poker player against a total newbie, the newbie will be /crushed/, almost entirely regardless of the cards
433 2011-06-04 07:25:14 <Doc_M> crap its 4 am
434 2011-06-04 07:25:28 <lfm> depends where ya are
435 2011-06-04 08:04:42 <samfisher> tcatm: the only thing that's out of place on your website are the ads
436 2011-06-04 08:04:56 <samfisher> otherwise it's very neat
437 2011-06-04 08:10:40 <Phoebus> Bitcoin ethics: http://forum.bitcoin.org/index.php?topic=11921.0
438 2011-06-04 08:15:27 <roconnor> Interesting, only the outTransaction script is used as the subscript when verifying signatures.
439 2011-06-04 08:15:42 <roconnor> I was concatinating the two halves of the script
440 2011-06-04 08:16:44 <lfm> roconnor: well the sigs are in the in-txn script that are verified in the out-txn
441 2011-06-04 08:17:23 <roconnor> lfm: yes, but the transaction is mangled before hashing.
442 2011-06-04 08:18:18 <lfm> oh? I havnt looked into it at that level of detail yet
443 2011-06-04 08:18:37 <roconnor> I'm having fun implementing my own bitcoin library
444 2011-06-04 08:18:53 <lfm> you arnt the google guy are ya?
445 2011-06-04 08:19:02 <roconnor> no, that is TD
446 2011-06-04 08:19:21 <sipa> roconnor: how complete is it?
447 2011-06-04 08:19:26 <lfm> what lang are you using then?
448 2011-06-04 08:20:00 <roconnor> sipa: I seem to have very basic signature verification working.
449 2011-06-04 08:20:05 <roconnor> lfm: Haskell
450 2011-06-04 08:20:37 <roconnor> sipa: as in I can verify the two types of regular signatures
451 2011-06-04 08:21:28 <sipa> but no full script support?
452 2011-06-04 08:21:38 <roconnor> sipa: but I still need to build a proper blockchain.  Right now I'm just running allong all the transactions in the mainline blockchain and checking them.
453 2011-06-04 08:21:48 <sipa> i see
454 2011-06-04 08:22:02 <sipa> would be very nice to have a Haskell bitcoin library
455 2011-06-04 08:22:33 <roconnor> sipa: At the moment it is just a matter of impelmenting op_codes
456 2011-06-04 08:23:03 <roconnor> sipa: the harder part is supporting the various Hashtype Values
457 2011-06-04 08:23:04 <lfm> doesnt seem much point in worrying about the unused op-codes
458 2011-06-04 08:23:44 <roconnor> half the opcodes seem disabled anyways :D
459 2011-06-04 08:25:39 <roconnor> sipa: I'm upto 900 lines of code + a couple of extra modules for SHA2 (which I had already written) RIPEMD and EcDsaSecp256k1.
460 2011-06-04 08:26:07 <sipa> nice
461 2011-06-04 08:26:12 <sipa> anywhere online?
462 2011-06-04 08:26:16 <roconnor> right now I'm just dumping everything into one big module.  I figure I can refactor it later once I actually understand how it all works. :D
463 2011-06-04 08:26:21 <roconnor> not online yet
464 2011-06-04 08:28:03 <lfm> I been using libgcrypt for sha256 and ripem. dunno for sure if it would cover the ecdsa
465 2011-06-04 08:28:27 <roconnor> lfm: my crypto routines are slow! :D
466 2011-06-04 08:28:49 <lfm> ya libgcrypt isnt the fastest but I like em anyway
467 2011-06-04 08:28:51 <roconnor> although my elliptic curve functions are about 100x faster than yesterday.
468 2011-06-04 08:29:08 <sipa> roconnor: how long does a signature verification take?
469 2011-06-04 08:29:13 <gmaxwell> man getting rid of the openssl dep would be nice.
470 2011-06-04 08:29:36 <roconnor> sipa: I don't have good data on that at the moment
471 2011-06-04 08:29:37 <gjs278> ;;bc,mtgox
472 2011-06-04 08:29:38 <gribble> {"ticker":{"high":17.41,"low":12.5311,"vol":67125,"buy":17.1983,"sell":17.3,"last":17.2}}
473 2011-06-04 08:29:42 <gjs278> what the fuck
474 2011-06-04 08:29:47 <sipa> gmaxwell: it'll still be some crypto library
475 2011-06-04 08:30:14 <gjs278> ;;bc,stats
476 2011-06-04 08:30:16 <lfm> gmaxwell: why dont you like openssl?
477 2011-06-04 08:30:18 <gribble> Current Blocks: 128545 | Current Difficulty: 434882.7217497 | Next Difficulty At Block: 129023 | Next Difficulty In: 478 blocks | Next Difficulty In About: 2 days, 15 hours, 59 minutes, and 56 seconds | Next Difficulty Estimate: 536459.06598599
478 2011-06-04 08:30:42 <gmaxwell> sipa: really bitcoin ought to have an internal fast code for the signature validation.. the generic stuff is slllow.
479 2011-06-04 08:31:16 <sipa> i don't like implementing cryptographic primitives ourselves
480 2011-06-04 08:31:54 <lfm> ya its nice to use standard libraries. they are more portable
481 2011-06-04 08:31:54 <roconnor> sipa: I was thinking of using sparks to verify transactions :D
482 2011-06-04 08:32:01 <gmaxwell> lfm: personally schizophrenic memory management, but if it doesn't bother you, great.., plus redhat patches out all the ecc code, so its an extra burden building there.
483 2011-06-04 08:32:29 <lfm> gmaxwell: ok, ya, good points
484 2011-06-04 08:32:35 <sipa> gmaxwell: i don't mind using another library than openssl
485 2011-06-04 08:33:10 <gmaxwell> sipa: ::shrugs:: the Secp256k1 is a special case, especially relevant to bitcoin, and important to performance.
486 2011-06-04 08:33:36 <gmaxwell> (oh also, did you know that Secp256k1 is unusually vulnerable to induced fault attacks? sucks for anyone trying to implement bitcoin directly on a smartcard. :) )
487 2011-06-04 08:33:45 <lfm> marginally important to performance
488 2011-06-04 08:33:53 <roconnor> It might be best to carve out the relevent code from openssl and put it right into bitcoin
489 2011-06-04 08:34:02 <roconnor> so that we can optimize the modulo p operation.
490 2011-06-04 08:34:19 <sipa> not sure - has anyone ever tried implementing secp256k1 specifically with all possible optimizations?
491 2011-06-04 08:35:04 <gmaxwell> sipa: I couldn't find any evidence of that, I went looking... didn't look for that long thought.
492 2011-06-04 08:35:16 <sipa> if it's significantly faster (more than let's say 20%), it may be worth it indeed to use a specialized version of the openssl or other library code
493 2011-06-04 08:35:18 <gmaxwell> (thats where I noticed the paper on the fault attacks)
494 2011-06-04 08:35:37 <roconnor> fault attacks?
495 2011-06-04 08:35:47 <sipa> timing attack?
496 2011-06-04 08:36:08 <sipa> ah
497 2011-06-04 08:36:13 <uppe> possum attack?
498 2011-06-04 08:36:39 <roconnor> sipa: a fault attack is a timing attack?
499 2011-06-04 08:36:47 <gmaxwell> Not timing. E.g. radiate or underpower the smart card... then extrat the private key from data in the signature.
500 2011-06-04 08:36:52 <gmaxwell> er extract
501 2011-06-04 08:36:59 <roconnor> ah
502 2011-06-04 08:37:18 <sipa> interesting
503 2011-06-04 08:38:21 <lfm> ya particulary aiming at smart cards it seems, trying to make them fault then analizing the results
504 2011-06-04 08:38:59 <gmaxwell> Yes. Well.. smart cards are interesting because fault attacks actually violate the security model. If someone can fault your server you have bigger problems. :)
505 2011-06-04 08:39:12 <lfm> hehe
506 2011-06-04 08:39:35 <lfm> they might start attacking smart phones that way
507 2011-06-04 08:54:08 <roconnor> Am I right in observing that someone erased all the tranactions between May 11 and May 25 on the testnet?
508 2011-06-04 08:54:39 <roconnor> compare http://blockexplorer.com/testnet/b/19308 with the next block
509 2011-06-04 08:54:45 <sipa> yes, ArtForz :)
510 2011-06-04 08:55:02 <roconnor> ok
511 2011-06-04 08:55:48 <roconnor> at least I'm not crazy :D
512 2011-06-04 08:56:01 <roconnor> no wonder I couldn't find some transactions I was looking for :P
513 2011-06-04 09:08:51 <samfisher> if i want to send 10 bitcoins and i want to set a 0.01 fee, the total sum that will go to the payee is 99.99?
514 2011-06-04 09:08:56 <samfisher> sorry
515 2011-06-04 09:08:57 <samfisher> 9.99
516 2011-06-04 09:09:01 <sipa> no
517 2011-06-04 09:09:07 <mtrlt> no, but you will pay 10.01
518 2011-06-04 09:09:08 <sipa> you'll lose 10.01 BTC
519 2011-06-04 09:09:14 <samfisher> so i have to have in my wallet 10.01
520 2011-06-04 09:09:20 <mtrlt> yes
521 2011-06-04 09:09:37 <samfisher> and what is this fee exactly? it goes to a miner?
522 2011-06-04 09:09:40 <mtrlt> yes
523 2011-06-04 09:10:06 <samfisher> why is it encouraged to pay fees?
524 2011-06-04 09:10:10 <mtrlt> the next miner that solves a block and has included your transaction in there
525 2011-06-04 09:10:31 <mtrlt> well, as the block creation reward goes down, there has to be some incentive to mine
526 2011-06-04 09:10:43 <samfisher> oh, i understand
527 2011-06-04 09:10:55 <samfisher> was this in satoshi's original project?
528 2011-06-04 09:11:25 <edcba> it is not 'encouraged'
529 2011-06-04 09:11:44 <edcba> but it does encourage miners to process your transaction first
530 2011-06-04 09:11:44 <samfisher> edcba: mandatory?
531 2011-06-04 09:11:53 <samfisher> oh i se
532 2011-06-04 09:12:11 <edcba> (or to process it at all)
533 2011-06-04 09:12:39 <samfisher> they might not want to process it? where do my money go in this case?
534 2011-06-04 09:12:59 <edcba> your money is spent but is not received
535 2011-06-04 09:13:20 <samfisher> so i loose them
536 2011-06-04 09:13:34 <edcba> in fact you could double spend them
537 2011-06-04 09:13:54 <edcba> not with original client but...
538 2011-06-04 09:14:04 <samfisher> but how?
539 2011-06-04 09:14:12 <sipa> samfisher: unless your transaction is confirmed, you did not actually lose anything
540 2011-06-04 09:14:21 <sipa> though the client will not let you
541 2011-06-04 09:14:31 <samfisher> sipa: but they will be deducte from my balance
542 2011-06-04 09:14:31 <sipa> it will mark the coins spent as soon as the tx is sent out
543 2011-06-04 09:14:49 <samfisher> and how do i do that?
544 2011-06-04 09:14:54 <sipa> do what?
545 2011-06-04 09:14:58 <samfisher> i remember a strange thing
546 2011-06-04 09:15:18 <samfisher> i was paid some bitcoins for work then wanted to move them to another wallet
547 2011-06-04 09:15:35 <samfisher> i haven't paid any fee (didn't knew about that)
548 2011-06-04 09:15:45 <samfisher> and it seems i miss some coins
549 2011-06-04 09:16:05 <sipa> miss?
550 2011-06-04 09:16:08 <samfisher> yes
551 2011-06-04 09:16:20 <sipa> i'm not following
552 2011-06-04 09:16:24 <samfisher> i should have had like 34.3
553 2011-06-04 09:16:31 <samfisher> and i only have 32.7
554 2011-06-04 09:16:56 <sipa> ?
555 2011-06-04 09:17:45 <samfisher> ok.. i had this wallet with 15.4 coins. someone paid me to another wallet with 17.9
556 2011-06-04 09:18:14 <samfisher> when i added up what i should had it was more than what i really had
557 2011-06-04 09:18:20 <diki> this is silly
558 2011-06-04 09:18:29 <cyberdo> sorry for whining about not having testnet coins, but does anyone have like 1 test-BTC to spare?
559 2011-06-04 09:18:37 <diki> 16 dollars
560 2011-06-04 09:18:44 <diki> wtf
561 2011-06-04 09:18:59 <mtrlt> cyberdo: sure, address please
562 2011-06-04 09:19:06 <samfisher> i might not remember well but it stunned me when the math i did wasn't the same with the math made by the client
563 2011-06-04 09:19:20 <cyberdo> mtrlt: mifoTYUhNrUWqEvJsd5kLz7D5KRhs5hoo4
564 2011-06-04 09:19:29 <samfisher> you guys now said the client might mark btc as spent
565 2011-06-04 09:19:32 <cyberdo> thanks, man... there faucet's still down
566 2011-06-04 09:19:54 <sipa> samfisher: it will mark them spend as soon as you spend them
567 2011-06-04 09:19:57 <sipa> *spent
568 2011-06-04 09:20:25 <samfisher> sipa: so there's no change what i described might happen?
569 2011-06-04 09:20:45 <sipa> no idea what happened there
570 2011-06-04 09:21:46 <CIA-103> DiabloMiner: Patrick McFarland master * r8e13bc0 / src/main/resources/DiabloMiner.cl : Ahah, I think I fixed mtrlt's bug - http://bit.ly/kc3AGB
571 2011-06-04 09:23:51 <cyberdo> mtrlt: thank you
572 2011-06-04 09:23:54 <mtrlt> Diablo-D3: i'll never do any optimizations again because these things are buggy as fuck
573 2011-06-04 09:24:15 <mtrlt> cyberdo: np :P
574 2011-06-04 09:28:19 <Diablo-D3> mtrlt: well, as a side rule
575 2011-06-04 09:28:26 <Diablo-D3> Ma anywhere near a constant is bad
576 2011-06-04 09:28:36 <Diablo-D3> gotta use Ma2
577 2011-06-04 09:28:58 <mtrlt> Diablo-D3: but the row you edited in the last commit didn't even use a constant :p
578 2011-06-04 09:29:06 <Diablo-D3> its on the line
579 2011-06-04 09:34:09 <Diablo-D3> mtrlt: seriously, the compiler is magic
580 2011-06-04 09:34:20 <Diablo-D3> and bfi_int actually IS buggy through this code path
581 2011-06-04 09:34:23 <diki> too many workers on slush's pool
582 2011-06-04 09:34:35 <diki> my payout dropped
583 2011-06-04 09:34:42 <Diablo-D3> AMD never enabled it through opencl use because they couldnt fix the issue
584 2011-06-04 09:34:54 <Diablo-D3> apparently SDK 2.5 will automatically use bfi_int
585 2011-06-04 09:34:58 <roconnor> I think the large mining pools should conspire to only build on each other's blocks.
586 2011-06-04 09:35:04 <Diablo-D3> so I'll probably end up auto-detecting 2.5 and turning bfi_int off
587 2011-06-04 09:35:16 <Diablo-D3> roconnor: ... that would be fucking hilarious
588 2011-06-04 09:35:28 <roconnor> Diablo-D3: It would give them more money
589 2011-06-04 09:35:42 <Diablo-D3> it would cause them to reach ~60% hash power and they can double spend.
590 2011-06-04 09:36:29 <mtrlt> i just hope they release 2.5 soon :P
591 2011-06-04 09:36:44 <roconnor> Well they'd have to consipre to double spend which might be harder to make the users of the pool wary of them.
592 2011-06-04 09:37:08 <roconnor> but to simply consipre to only build on each other blocks; the think the pool users would actually like that.
593 2011-06-04 09:38:42 <roconnor> *I think
594 2011-06-04 09:43:30 <denisx> Diablo-D3: isn't it demotivating to know that the difficulty will destroy all your optimizations as long as you don't keep them for yourself?
595 2011-06-04 09:45:14 <gjs278> no
596 2011-06-04 09:45:33 <mtrlt> it means the network gets more efficient, therefore more difficult to break :p
597 2011-06-04 09:45:37 <gjs278> whenever he adds an optimization to his miner, that just means the network will have more hashpower
598 2011-06-04 09:45:38 <gjs278> yes
599 2011-06-04 09:45:52 <gjs278> the attackers won't be using DiabloMiner
600 2011-06-04 09:46:05 <gjs278> but the people preventing it will be
601 2011-06-04 09:48:57 <gmaxwell> 0_o
602 2011-06-04 09:49:03 <gmaxwell> (from #bitcoin )
603 2011-06-04 09:49:13 <gmaxwell> (thats the genesis block check)
604 2011-06-04 09:51:43 <davep> is bitcoin full of endian-unsafe optimisations? or have a non-zero number of them? :P
605 2011-06-04 09:52:18 <sipa> gmaxwell: disk corruption?
606 2011-06-04 09:52:36 <sipa> oh no, that's the initialization
607 2011-06-04 09:52:52 <Seta> Somebody can help me here? http://pastebin.com/LimRhnSw
608 2011-06-04 09:56:18 <gmaxwell> Mystery solved.
609 2011-06-04 09:56:20 <gmaxwell> sipa: PPC
610 2011-06-04 09:58:23 <Diablo-D3> denisx: yes, but at least I can say I have th ebest miner ever.
611 2011-06-04 10:00:06 <zamgo> best cpu & gpu?
612 2011-06-04 10:00:42 <Seta> Ahh this are the guys helped me b4 :D im happy to find the channel again (Heyho here ^^)
613 2011-06-04 10:00:47 <diki> Seta:delete any nvidia crap you might have. IF you have an ati card
614 2011-06-04 10:01:00 <Seta> ok ill try thx
615 2011-06-04 10:01:10 <diki> if you have an nvidia card dont do anything and download a cuda miner
616 2011-06-04 10:02:23 <Seta> its an ati
617 2011-06-04 10:02:54 <Diablo-D3> night all
618 2011-06-04 10:04:16 <Seta> When i remove all nvidia stuff i get:No module named pyopencl, if i reinstall pyopencl i got the nvidia stuff again
619 2011-06-04 10:05:52 <Seta> pyopencl.LogicError: clGetPlatformIDs failed: invalid/unknown error code :/
620 2011-06-04 10:07:38 <Seta> please help a sad ati noob :/
621 2011-06-04 10:10:09 <jgarzik> Seta: #bitcoin-mining
622 2011-06-04 10:12:39 <gmaxwell> So, I've been helping someone who doesn't want to git hit with the minfee in .21 and I can't figure out why he's being subjected to it at all
623 2011-06-04 10:13:13 <gmaxwell> He rxed 1 btc in a single transaction about 82 blocks ago, hes trying to send it out in a single transaction now.
624 2011-06-04 10:14:27 <gmaxwell> I'm running the priority check by hand (1e8 * 60*60*9 / 259).. and that looks fine.
625 2011-06-04 10:14:43 <gmaxwell> Why the heck else would he get the 'This transaction is over the size limit.'
626 2011-06-04 10:15:09 <ali1234> Seta: it's a bug with ubuntu and probably debian packages. it doesn't hurt to have nvidia driver installed, just ignore it
627 2011-06-04 10:15:09 <sipa> maybe some other input coin(s) were selected?
628 2011-06-04 10:15:11 <gmaxwell> (I'm assuming its 259 bytes, I obviously haven't seen the txn, but I can't see why it would be unusually big single input single output)
629 2011-06-04 10:15:17 <gmaxwell> good call
630 2011-06-04 10:15:24 <gmaxwell> lemme ask
631 2011-06-04 10:15:35 <sipa> is the transaction sent, or not?
632 2011-06-04 10:16:09 <Seta> ali1234, it doesnt work anyway...
633 2011-06-04 10:16:27 <ali1234> Seta: idk about that. make sure you have fglrx installed, not the free driver (default)
634 2011-06-04 10:17:01 <gmaxwell> sipa: no.
635 2011-06-04 10:17:19 <gmaxwell> The person has two other addresses one with 0.01 one with 0.02 .. so I don't see why they'd be related.
636 2011-06-04 10:17:35 <gmaxwell> unless it's doing something moronic like 0.02+1 = 1+0.02
637 2011-06-04 10:18:33 <sipa> that's what i fear
638 2011-06-04 10:19:49 <gmaxwell> yea... the smaller amounts are older.
639 2011-06-04 10:19:55 <gmaxwell> My thats dumb.
640 2011-06-04 10:26:14 <diki> why does the blockchain sync so slow?
641 2011-06-04 10:26:49 <gmaxwell> sipa: so I tried having him send 0.03 btc to himself (to trick out the selection algorithim, but it wouldn't do that either)
642 2011-06-04 10:29:45 <Sedra> 0.3.19 gmaxwell you can set 0.00 fees ;)
643 2011-06-04 10:30:12 <tommygunner> whats the issue with the fee? the setting doesnt work in specific releases?
644 2011-06-04 10:30:41 <gmaxwell> I know, not the issue.
645 2011-06-04 10:31:47 <mtrlt> does wallet.dat really contain every transaction you've done too?
646 2011-06-04 10:32:00 <sipa> yes
647 2011-06-04 10:32:11 <sipa> all sent and received transactions
648 2011-06-04 10:32:13 <mtrlt> hmh
649 2011-06-04 10:32:27 <sipa> that's how your balance is calculated
650 2011-06-04 10:32:32 <mtrlt> i wanna just extract the private keys from there. didn't find any documentation on the file format.
651 2011-06-04 10:32:44 <sipa> it's a bdb database file
652 2011-06-04 10:33:18 <mtrlt> okay, nice
653 2011-06-04 10:33:54 <sipa> btw, http://forum.bitcoin.org/index.php?topic=8091.0
654 2011-06-04 10:37:56 <gmaxwell> okay. I've discovered that bitcoin has some unique development challenges.
655 2011-06-04 10:38:24 <gmaxwell> Normally when a user has a reproducable bug I'll ask them to give me the data so I can reproduce it on a system with instrumented code.
656 2011-06-04 10:38:34 <gmaxwell> Amazingly this doesn't work so well for bitcoin!
657 2011-06-04 10:39:45 <sipa> i hope we'll have a modularized system somewhere in the future, where it's possible eg. replace the network node with a scriptable dummy node, to write test cases
658 2011-06-04 10:45:01 <gmaxwell> sipa: here in particular, the data I'd want for the reproduction is their wallet. which I could get in this case because its barely used... e.g. backup, downgrade to .19, spend all and wait for confirms, give wallet to me, destroy wallet and make a new one. but thats a mess.
659 2011-06-04 11:32:47 <eamon> I'm after pressing some shortcut on my key board (i had it upside downj and changing the battery) and now my screen is in a different res. It brought up (monitor / monitor + projector / projector) something like that, what is the shortcut key so I can cycle it back to monitor. My keyboard has no special function keys or anything. just the normal qwerty keys. My screen is locked to 1600x900
660 2011-06-04 11:33:58 <mads-> Hi. I'd like to get involved. :)
661 2011-06-04 11:44:50 <sipa> ;;bc,stats
662 2011-06-04 11:44:53 <gribble> Current Blocks: 128581 | Current Difficulty: 434882.7217497 | Next Difficulty At Block: 129023 | Next Difficulty In: 442 blocks | Next Difficulty In About: 2 days, 9 hours, 5 minutes, and 30 seconds | Next Difficulty Estimate: 540656.02390994
663 2011-06-04 12:23:37 <cyberdo> how come "sendtoaddress <bitcoinaddress> <amount>" sets a fee > 0 even if I set the fee on the command line to 0, and getinfo() says it's 0?
664 2011-06-04 12:24:14 <somecoiner> what version?
665 2011-06-04 12:24:49 <cyberdo> 0.3.21
666 2011-06-04 12:25:30 <cyberdo> testnet, if i makes any difference
667 2011-06-04 12:30:54 <sipa> cyberdo: 0.3.21 has an extended rule for requiring minimum fees for some transactions, to prevent spam
668 2011-06-04 12:31:10 <sipa> and those rules are applied automatically (which shouldn't happen, i think)
669 2011-06-04 12:31:49 <cyberdo> sipa: thanks
670 2011-06-04 12:32:23 <cyberdo> isn't that a bit unnessecary in an OS-software?
671 2011-06-04 12:32:52 <cyberdo> also: at the current exchange rate, 0.01 BTC is quite much for small transactions
672 2011-06-04 12:33:13 <kik-> My account balance doesn't seem to update, I transferred coins yesterday and it's still at 0.00. What could be the problem?
673 2011-06-04 12:34:29 <somecoiner> Is your block count going up?
674 2011-06-04 12:34:47 <kik-> yes, it's at 128586 atm
675 2011-06-04 12:34:58 <cyberdo> sipa: btw. Can I disable it?
676 2011-06-04 12:35:00 <kik-> and 32 connections
677 2011-06-04 12:35:14 <sipa> cyberdo: not yet
678 2011-06-04 12:42:36 <Gekz> what's the format of the wallet.dat file?
679 2011-06-04 12:43:40 <kik-> Gekz: what do you mean?
680 2011-06-04 12:45:07 <Gekz> is it a bdb, for instance.
681 2011-06-04 12:45:25 <sytse> utsl?
682 2011-06-04 12:46:00 <Gekz> I don't want to use the source luke if there's a simple answer.
683 2011-06-04 12:46:06 <sytse> and, yes.
684 2011-06-04 12:46:17 <sytse> db.h:class CWalletDB : public CDB
685 2011-06-04 12:46:33 <sytse> class CDB
686 2011-06-04 12:47:40 <sytse> protected.. seriously. Shouldn't these variables be private, the methods be public and the inheritance be private?
687 2011-06-04 12:48:13 <Gekz> so it looks like a berkeley db, okay.
688 2011-06-04 12:48:32 <Gekz> the source code for bitcoin is so fugly
689 2011-06-04 12:48:48 <sipa> sytse: yes, should be - but there's so much to fix in the code organisation
690 2011-06-04 12:49:02 <sytse> :)
691 2011-06-04 12:49:05 <sipa> first some cleaner separation between modules is needed
692 2011-06-04 12:49:21 <Gekz> so why was bdb chosen over something like sqlite or xml?
693 2011-06-04 12:49:21 <sytse> yes, that would be a higher priority
694 2011-06-04 12:49:37 <sipa> then we can start talking about moving to making everything nicely private and protected, with accessor functions
695 2011-06-04 12:50:57 <sipa> Gekz: over sqlite: no idea - over xml: it doesn't really contain human-readable data, and needs consistent and atomic updating without re-writing the entire file each time
696 2011-06-04 12:51:13 <Gekz> okay.
697 2011-06-04 12:51:13 <sytse> indeed
698 2011-06-04 12:51:23 <Gekz> I just know that bdb is a pain in the ass to use for most things
699 2011-06-04 12:51:35 <Gekz> for instance, how would one merge two wallets?
700 2011-06-04 12:51:39 <sytse> sqlite is too ;-)
701 2011-06-04 12:51:55 <Gekz> without having to send everything from one to the other and disposing of the first?
702 2011-06-04 12:51:59 <sytse> sqlite performance can be.. troubling afaik
703 2011-06-04 12:52:34 <Gekz> doesn't bdb have this habit of changing formats on different platforms and versions, too?
704 2011-06-04 12:52:47 <sytse> (btw, I have had problems with sqlite losing integrity of the database after a crash, ie, my chrome session and settings disappeared)
705 2011-06-04 12:52:55 <sytse> Gekz: yes.
706 2011-06-04 12:53:34 <Gekz> I'm just wondering if there isn't a more platform-portable, useful format we could use instead of bdb
707 2011-06-04 12:53:41 <sytse> I know the enlightenment project switched away from using a system version of bdb a long time ago because of this
708 2011-06-04 12:54:24 <Gekz> I just worry that wallet corruption may become more prominent
709 2011-06-04 12:58:13 <sytse> too bad boost doesn't have something simple but reliable and cross-platform for that
710 2011-06-04 12:59:31 <sytse> although it's almost impossible to make it reliable on any os, kernel and filesystem I suppose
711 2011-06-04 13:01:53 <Gekz> sytse: not unless you invent yet another storage format :D
712 2011-06-04 13:04:52 <sipa> Gekz: merging two wallets is a pain anyway
713 2011-06-04 13:04:57 <sipa> and splitting them is even harder
714 2011-06-04 13:05:14 <sipa> you can't just do that without special intervention from a program that understands what data is in there
715 2011-06-04 13:05:23 <Gekz> my understanding of the db is it essentially holds the key combination for each of the addresses and nothing more
716 2011-06-04 13:05:27 <Gekz> is that a correct assessment?
717 2011-06-04 13:05:46 <sipa> no
718 2011-06-04 13:05:47 <Gekz> sipa: sure, but specialised SQL is easier than specialised "BDB" parsing haha
719 2011-06-04 13:05:50 <sipa> it contains a lot of things
720 2011-06-04 13:05:57 <Gekz> what else is in the wallet?
721 2011-06-04 13:06:26 <sipa> private keys, known addresses, incoming and outgoing transactions, unconfirmed transactions, settings, information about the best known block chain
722 2011-06-04 13:06:49 <sipa> it's really all personal data
723 2011-06-04 13:08:59 <Gekz> client settings are stored in a db and not in a separate config file?
724 2011-06-04 13:09:12 <Gekz> sounds like the db itself needs to be modularised :P
725 2011-06-04 13:10:04 <sipa> definitely
726 2011-06-04 13:10:29 <lizthegrey> why can't it just be written as a plain text file?
727 2011-06-04 13:10:45 <lizthegrey> at least as far as private keys and known addresses are concerned?
728 2011-06-04 13:10:57 <Gekz> well, I was wondering that too
729 2011-06-04 13:11:03 <Gekz> like ssh does
730 2011-06-04 13:11:10 <sipa> 16:50:57 < sipa> Gekz: over sqlite: no idea - over xml: it doesn't really contain human-readable data, and needs consistent and atomic updating without re-writing  the entire file each time
731 2011-06-04 13:11:16 <Gekz> it would allow for a much more safe and platform independent manner of management
732 2011-06-04 13:11:37 <Gekz> you can append to files
733 2011-06-04 13:11:53 <sipa> you need more, like marking transactions spent
734 2011-06-04 13:12:23 <sipa> i favor exporting/importing of wallets' essential information to/from human readable form over turning everything into a human readable form
735 2011-06-04 13:13:27 <JFK911> ;;bc,mtgox
736 2011-06-04 13:13:28 <gribble> {"ticker":{"high":17.41,"low":13,"vol":53370,"buy":17,"sell":17.092,"last":17}}
737 2011-06-04 13:13:32 <JFK911> i cant believe this
738 2011-06-04 13:15:44 <rlifchitz> where can i find the source code for the gribble bot?
739 2011-06-04 13:15:51 <kW_> Hello! Can anyone tell which exact boost version is needed for compiling?
740 2011-06-04 13:16:22 <Gekz> sipa: actually, if you could simply EXPORT it into a safe format that can be imported
741 2011-06-04 13:16:37 <sipa> Gekz: https://github.com/bitcoin/bitcoin/pull/220
742 2011-06-04 13:16:38 <Gekz> that would solve this problem entirely
743 2011-06-04 13:17:08 <Gekz> well done
744 2011-06-04 13:17:11 <Gekz> I can shut up now then
745 2011-06-04 13:18:38 <sanity> my genetic algorithm has come up with this (to speculate successfully on the MtGox exchange): https://gist.github.com/1007968
746 2011-06-04 13:19:09 <sanity> but this is based on only a few days of data (per-minute ticker samples) - I need a lot more :-(
747 2011-06-04 13:22:50 <phantomcircuit> sanity, what is that?
748 2011-06-04 13:22:58 <phantomcircuit> oh
749 2011-06-04 13:22:58 <sipa> what are 'ticker samples' ?
750 2011-06-04 13:23:08 <sanity> phantomcircuit: it is a computer program i evolved to speculate on mtgox
751 2011-06-04 13:23:25 <sanity> sipa: just download https://mtgox.com/code/data/ticker.php once a minute
752 2011-06-04 13:23:29 <phantomcircuit> MagicalTux, do you have complete records of activity on mtgox ?
753 2011-06-04 13:23:38 <forrestv> sanity, that's awesome (:
754 2011-06-04 13:23:53 <Gekz> what language is that
755 2011-06-04 13:24:01 <Gekz> some kind of pseudolisp?
756 2011-06-04 13:24:05 <sanity> Gekz: it is a domain-specific language I invented for this
757 2011-06-04 13:24:20 <sanity> Gekz: it has a lot of primitive operators to-do with means and standard deviations etc
758 2011-06-04 13:24:47 <sanity> my problem is that i only started logging data 3 days ago
759 2011-06-04 13:25:04 <Gekz> why did you make a language just for this
760 2011-06-04 13:25:06 <sanity> and so all the programs I evolve are variations on "be judicious about when you get in, but when you get in - stay in"
761 2011-06-04 13:25:37 <sanity> Gekz: typically in genetic programming you want a language with primitives that relate to the problem domain.
762 2011-06-04 13:26:06 <diki> hm..MS is buying Nvidia
763 2011-06-04 13:26:08 <Gekz> I don't know what genetic programming is
764 2011-06-04 13:26:10 <diki> i wonder what will happen
765 2011-06-04 13:26:28 <Gekz> I assumed it was simply functional programming
766 2011-06-04 13:26:30 <Gekz> MS buying nvidia?
767 2011-06-04 13:26:32 <Gekz> oh ffs.
768 2011-06-04 13:26:42 <diki> well, something along those lines anyway
769 2011-06-04 13:27:09 <diki> http://sg.news.yahoo.com/microsoft-pact-nvidia-could-result-future-takeover-070058453.html
770 2011-06-04 13:27:14 <sanity> Gekz: basically you create a random population of simple algorithms, test them all on a problem, and combine and mutate the best ones to form the next generation.  then repeat
771 2011-06-04 13:28:39 <sanity> Gekz: this is what each generation looks like : https://gist.github.com/1007982
772 2011-06-04 13:29:03 <Gekz> ugliest syntax ever
773 2011-06-04 13:29:05 <sanity> Gekz: the second column indicates the profit, eg. .96 means 4% increase in revenue per day
774 2011-06-04 13:29:12 <sanity> Gekz: its not designed to be human readable
775 2011-06-04 13:29:24 <sanity> Gekz: or human writeable
776 2011-06-04 13:29:25 <sipa> sanity: i have a sample of https://mtgox.com/code/data/getDepth.php (almost) every minute from the past few months
777 2011-06-04 13:29:51 <sanity> sipa: oh?  can I infer buy/sell/last/volume from that data?
778 2011-06-04 13:30:02 <sipa> no
779 2011-06-04 13:30:11 <sanity> sipa: what exactly does that data show?
780 2011-06-04 13:30:24 <sipa> the amounts and price points of all bids and asks
781 2011-06-04 13:30:50 <sipa> you can derive which amount you could sell/buy at which price