1 2011-02-28 00:37:40 <andrewh> hi
  2 2011-02-28 01:42:21 <Lachesis> i'm getting the clGetPlatformIDs error on Ubuntu 10.10
  3 2011-02-28 01:42:25 <Lachesis> anyone know how to fix this?
  4 2011-02-28 01:42:59 <Lachesis> i've fixed this before but i forgot how
  5 2011-02-28 01:52:58 <temp1029> anyone here that could field a, probably simple, question?
  6 2011-02-28 01:53:15 <endian7000> ?
  7 2011-02-28 01:53:23 <Lachesis> shoot temp1029
  8 2011-02-28 01:54:26 <temp1029> how does bitcoin decide which transactions to include in a block?  and do the transactions my instance of bitcoin has included need to match with what all the other nodes have decided to include?
  9 2011-02-28 01:54:37 <Lachesis> for the second part, no
 10 2011-02-28 01:54:48 <Blitzboom> all transactions since the last block are included in the next block
 11 2011-02-28 01:56:33 <temp1029> so if i am working on making a new block and a transaction comes through do i then have to include it and start working on a "new" block or does a transaction like that go into some kind of queue, to be included in the block after the one i am working on?
 12 2011-02-28 01:58:31 <temp1029> if the way I wrote that makes any sense....
 13 2011-02-28 01:59:49 <luke-jr> temp1029: hybrid between #1 and #2
 14 2011-02-28 01:59:51 <luke-jr> ideally #1
 15 2011-02-28 02:00:09 <luke-jr> but you don't *have* to do anything I think
 16 2011-02-28 02:00:26 <ArtForz> you dont *have* to include any transactions
 17 2011-02-28 02:00:27 <TheKid> it's highly unlikely at the moment that a transaction comes in at the exact moment you find + broadcast your found block
 18 2011-02-28 02:01:00 <ArtForz> blocks not containing any transactions except for generation are perfectly ok
 19 2011-02-28 02:01:02 <TheKid> if it happens, it goes into the next block someone founds
 20 2011-02-28 02:01:06 <ArtForz> yep
 21 2011-02-28 02:01:25 <ArtForz> normal client only adds transactions to block once every 60 seconds
 22 2011-02-28 02:01:52 <luke-jr> ArtForz: what if it gets a block with the old tx list?
 23 2011-02-28 02:01:56 <luke-jr> does it still use it?
 24 2011-02-28 02:01:59 <ArtForz> ?
 25 2011-02-28 02:02:20 <luke-jr> eg, it updates the tx list, then gets results from miner
 26 2011-02-28 02:02:27 <luke-jr> results withouit that tx
 27 2011-02-28 02:02:40 <ArtForz> I think it does it the other way around
 28 2011-02-28 02:03:09 <luke-jr> so it only updates when it has a getwork, not every 60 s?
 29 2011-02-28 02:03:25 <ArtForz> well, getworks happen usually a lot more often than once a minute ...
 30 2011-02-28 02:03:30 <ArtForz> I think getwork caches old transaction list for a while
 31 2011-02-28 02:03:57 <luke-jr> think it'd piss people off or confuse them, if people started demanding fees?
 32 2011-02-28 02:03:59 <ArtForz> never really checked the internals too deeply, because the whole concept is backwards
 33 2011-02-28 02:04:16 <ArtForz> unless they happened to have a large % of the network... nope
 34 2011-02-28 02:05:13 <luke-jr> I mean multiple people :P
 35 2011-02-28 02:05:17 <luke-jr> maybe say 10%
 36 2011-02-28 02:05:24 <ArtForz> nah, not really
 37 2011-02-28 02:05:54 <ArtForz> you'd need more like 80% to make it really bothersome imo
 38 2011-02-28 02:06:51 <ArtForz> even then fee-less txes would get into one of 5 blocks on average = < 1h
 39 2011-02-28 02:07:45 <temp1029> thx ArtForz, your statement about generation only blocks tells me what i needed to know, i was having trouble wrapping my head around how all the nodes would be kept synchronous, but since the created clock is included with the solution it, they don't need to be
 40 2011-02-28 02:07:58 <temp1029> once again amazed by the elegance of Bitcoin!
 41 2011-02-28 02:08:10 <ArtForz> yep
 42 2011-02-28 02:09:36 <temp1029> one other question, after we reach approx 21mil Bitcoins, wouldn't it make sense to lock the target at it's easiest possible value, so as to generate blocks as fast as possible?  or am i missing something on that?
 43 2011-02-28 02:09:51 <ArtForz> erm, no
 44 2011-02-28 02:09:59 <[Tycho]> Then block will be generated way too fast.
 45 2011-02-28 02:10:11 <ArtForz> each block causes some overhead, so that be rather... bad
 46 2011-02-28 02:10:49 <nevezen> No more transactions after 21mil bitcoins?
 47 2011-02-28 02:10:52 <temp1029> ahhhh, got it, would overload many of the nodes with blocks being generated ever couple of minutes of seconds, right?
 48 2011-02-28 02:10:53 <ArtForz> ?
 49 2011-02-28 02:10:57 <ArtForz> yeah
 50 2011-02-28 02:10:59 <luke-jr> nevezen: &
 51 2011-02-28 02:11:10 <[Tycho]> Actually i think that waiting 10 min for each transaction is not nice :) But nothing can be done.
 52 2011-02-28 02:11:13 <luke-jr> temp1029: it would take the light speed problem home
 53 2011-02-28 02:11:22 <ArtForz> also, you'd get way more blocks independently found by 2 nodes near-simultaneously
 54 2011-02-28 02:11:22 <nevezen> when generation becomes irrelevant because of difficulty..
 55 2011-02-28 02:11:26 <ArtForz> = way more chain froks
 56 2011-02-28 02:11:34 <ArtForz> *forks
 57 2011-02-28 02:12:13 <ArtForz> reducing avg time/block increases forks, which lowers overall network strength
 58 2011-02-28 02:12:13 <temp1029> i was just thinking in terms of verifying the validity of a transaction for fast acceptence, as in retail outlets and such
 59 2011-02-28 02:12:30 <ArtForz> well, bit6coin really isnt designed for that
 60 2011-02-28 02:12:34 <ArtForz> *bitcoin
 61 2011-02-28 02:12:38 <luke-jr> temp1029: if you have blocks every minute, you would need 60 confirmations to be sure
 62 2011-02-28 02:12:44 <ArtForz> yep
 63 2011-02-28 02:12:48 <[Tycho]> There will be no retail outlets in such a distant future.
 64 2011-02-28 02:12:59 <luke-jr> lol
 65 2011-02-28 02:13:09 <temp1029> i like your thinking Tycho!
 66 2011-02-28 02:13:20 <temp1029> thanks for letting me pick your brains
 67 2011-02-28 02:13:24 <[Tycho]> May be you just missed the fact that the end of the world comes in 2012.
 68 2011-02-28 02:13:25 <luke-jr> already, you can order food and have it delivered
 69 2011-02-28 02:13:31 <[Tycho]> Not to mention World War III
 70 2011-02-28 02:13:49 <amiller> what's the significance of the 21 million
 71 2011-02-28 02:13:57 <amiller> is that number an optimal balance between some factors
 72 2011-02-28 02:13:58 <luke-jr> amiller: no significance
 73 2011-02-28 02:14:07 <temp1029> well the big bus driving down my street yesterday would disagree with you Tycho, its coming march 21st 2011....
 74 2011-02-28 02:14:10 <amiller> would 10 have been adequate
 75 2011-02-28 02:14:20 <luke-jr> amiller: 21million isn't adequate.
 76 2011-02-28 02:14:36 <luke-jr> it's not really 21 million either
 77 2011-02-28 02:14:46 <temp1029> although i could swear that same bus said the same thing about march 21st 2010.......
 78 2011-02-28 02:14:59 <amiller> i understand that since each one can be split into 8 decimal place parts, that the granularity is still very good
 79 2011-02-28 02:15:06 <luke-jr> it's 2,099,999,997,690,000
 80 2011-02-28 02:15:21 <temp1029> anyways, nite all, thx again
 81 2011-02-28 02:15:22 <endian7000> looking forward to buying pizza with mBTC...
 82 2011-02-28 02:15:47 <luke-jr> endian7000: I'm looking forward to buying pizza with TBC :P
 83 2011-02-28 02:15:48 <amiller> luke-jr, you are saying that there are that many 10^15  indivisible units, is that right
 84 2011-02-28 02:16:00 <ArtForz> thats the total number of smallest units
 85 2011-02-28 02:16:13 <amiller> great, that makes sense to me
 86 2011-02-28 02:18:10 <amiller> what is TBC
 87 2011-02-28 02:18:18 <amiller> i thought it was just a typo for btc
 88 2011-02-28 02:18:18 <luke-jr> https://en.bitcoin.it/wiki/Units
 89 2011-02-28 02:18:35 <ArtForz> luke-jr: hectobtc?
 90 2011-02-28 02:18:52 <luke-jr> ArtForz: yeah
 91 2011-02-28 02:19:13 <ArtForz> cool ;)
 92 2011-02-28 02:19:19 <luke-jr> I'd prefer to be buying/selling GBTC ;)
 93 2011-02-28 02:19:20 <luke-jr> err
 94 2011-02-28 02:19:22 <luke-jr> I'd prefer to be buying/selling GTBC ;)
 95 2011-02-28 02:21:54 <[Tycho]> mBTC... Mean Time Between Crashes ? :)
 96 2011-02-28 02:22:05 <luke-jr> [Tycho]: milli-bitcoin
 97 2011-02-28 02:22:14 <ArtForz> thatd be mtbc ;)
 98 2011-02-28 02:22:39 <luke-jr> ra-tam, ra-mill su-san fy-ton -bong, su-mill ni-san po-ton go and ni millths <-- total number of TBC ever :P
 99 2011-02-28 02:23:04 <amiller> i want to buy bong bitcoins
100 2011-02-28 02:23:35 <luke-jr> amiller: I'll sell you 1 bong-bitcoin for 40 USD
101 2011-02-28 02:24:46 <[Tycho]> What's the difference between "Tonal" and "Tonal as Hex" in that table ?
102 2011-02-28 02:24:59 <luke-jr> [Tycho]: the characters used
103 2011-02-28 02:25:09 <amiller> ie 42.94967 @ 0.931
104 2011-02-28 02:27:57 <amiller> so a new block is made every 10 minutes
105 2011-02-28 02:28:17 <TheKid> that's the target
106 2011-02-28 02:28:24 <TheKid> lately it has been far more frequent
107 2011-02-28 02:28:26 <amiller> everyone who is trying to generate blocks has a chance of winning each block
108 2011-02-28 02:28:39 <amiller> people with more computer power are more likely to generate blocks
109 2011-02-28 02:28:50 <amiller> do we know how much variety is in the people generating blocks?
110 2011-02-28 02:29:10 <amiller> if there were on person in control of 50% of the network because they had a super computer generating blocks, would that be noticeable
111 2011-02-28 02:31:17 <amiller> wait, does someone win 50BTC every ten minutes, that can't be..
112 2011-02-28 02:32:30 <[Tycho]> Why not ? They are working hard.
113 2011-02-28 02:32:39 <jgarzik> amiller: on average, every 10 minutes they receive 50 BTC + transaction fees
114 2011-02-28 02:34:59 <luke-jr> until 2013
115 2011-02-28 02:35:47 <amiller> i guess that's still not very much inflation given the current volume
116 2011-02-28 02:36:17 <[Tycho]> It's actually deflation, not inflation :)
117 2011-02-28 02:37:54 <amiller> is there anyone aggregating all the statistics, that would show which addresses hold the most coins, things like that
118 2011-02-28 02:38:14 <gasteve> adding more btc is inflation...however you have to consider that the more miners there are, the higher the cost of mining, which will put upward pressure on the value of btc...and the more miners, the broader those BTC are being distributed, which is a good thing overall
119 2011-02-28 02:40:22 <endian7000> ...and GPUs get cheaper :)
120 2011-02-28 02:40:44 <gasteve> http://bitcoinreport.blogspot.com/
121 2011-02-28 02:42:38 <amiller> thanks gasteve that's exactly what i had in mind
122 2011-02-28 02:43:49 <amiller> how will the block size be expected to grow?
123 2011-02-28 02:43:54 <amiller> like if it takes me an hour to download it on a new machine now
124 2011-02-28 02:44:10 <amiller> will it take twice as much space when there are twice as many bitcoin transactions
125 2011-02-28 02:46:12 <greyface> hm, not sure if i like that my address is on that list
126 2011-02-28 02:46:33 <greyface> maybe i should break my holdings up into smaller chunks
127 2011-02-28 02:47:35 <amiller> is there anyone maintaining a watch list
128 2011-02-28 02:47:47 <amiller> trying to match accounts to public personas, such as irc nicks
129 2011-02-28 02:48:16 <amiller> something like that might actually be useful to help you audit your anonynimity
130 2011-02-28 02:49:28 <gasteve> a block actually only includes the root hash of a merkle tree of transaction hashes...in the bitcoin whitepaper, it states that at a rate of 10 minutes per block, the entire block chain would increate by 4.3mb / year
131 2011-02-28 02:49:54 <hexidigital> hi, I have a small patch that fixes a missing newline in the latest available linux version: https://gist.github.com/846910
132 2011-02-28 02:50:49 <endian7000> amiller: send me a penny to 1Pguqs4RVxM6YzTt8haZR1e5z57tqMZATE and I'll add you ;)
133 2011-02-28 02:51:07 <hexidigital> also, "./bitcoind -?" breaks with the csh shell.  maybe '-help' would be a better argument?
134 2011-02-28 02:51:22 <nevezen> only 4.3MB/year?
135 2011-02-28 02:51:37 <ArtForz> for headers, yeah
136 2011-02-28 02:51:37 <gasteve> today, clients download and store all transactions I believe (not 100% sure)...it's conceivable that a client could download and verify all transactions in a block, but only retain transactions that it cares about (i.e. ones in a particular wallet)
137 2011-02-28 02:52:07 <gasteve> actually, 4.2MB/year (mistyped)
138 2011-02-28 02:52:33 <amiller> are there any evil github projects
139 2011-02-28 02:52:40 <amiller> containing code that is easy to run that would make the network worse for everyone
140 2011-02-28 02:52:44 <gasteve> but, again, that's just block headers...you would want a number of node in the network retaining all transactions
141 2011-02-28 02:52:45 <nevezen> what's the current block size now? (in MB)?
142 2011-02-28 02:52:48 <tcatm> hexidigital: thanks!
143 2011-02-28 02:53:07 <amiller> such as reference code for making people do too much work, making lots of spurious transactions things like that
144 2011-02-28 02:53:08 <gasteve> about 80 bytes
145 2011-02-28 02:53:28 <gasteve> (oh, you mean all blocks)
146 2011-02-28 02:53:40 <nevezen> 80MB
147 2011-02-28 02:53:46 <CIA-57> bitcoin: tcatm master * r6665aca / init.cpp : fix missing newline in help - http://bit.ly/g7Iq1w
148 2011-02-28 02:54:18 <hexidigital> tcatm:  np
149 2011-02-28 02:54:23 <nevezen> only transactions get added to a new block after each generation right?
150 2011-02-28 02:54:31 <nevezen> only new transactions I mean..
151 2011-02-28 02:54:37 <gasteve> 80MB is including the transactions...the headers only would be just under 9MB
152 2011-02-28 02:55:18 <dirtyfilthy> 9mb really? 2000 headers is about 160k
153 2011-02-28 02:55:25 <luke-jr> hexidigital: --help works
154 2011-02-28 02:55:40 <hexidigital> luke-jr:  indeed, but the --help says to use -? :)
155 2011-02-28 02:55:50 <luke-jr> :p
156 2011-02-28 02:55:53 <luke-jr> -h is the standard
157 2011-02-28 02:55:55 <dirtyfilthy> oh whoops, i'm thinking an order of magnitude out
158 2011-02-28 02:56:00 <gasteve> I'm going by what the white paper says the header size of a block is
159 2011-02-28 02:56:06 <gasteve> (80 bytes)
160 2011-02-28 02:56:44 <dirtyfilthy> gasteve: yeah the block chain is ~ 100k blocks not 10k, i wasn't calculating properly
161 2011-02-28 02:57:29 <gasteve> actually, if 2000 headers is 160k, then that would be in line with a total of a little under 9mb for all of the headers
162 2011-02-28 02:58:39 <dirtyfilthy> yeh
163 2011-02-28 02:59:21 <endian7000> what if I spam the system with an epic amount of transactions?
164 2011-02-28 02:59:33 <amiller> how many would you have to spam to be a problem
165 2011-02-28 02:59:42 <amiller> and from how many iP addresses?
166 2011-02-28 03:00:06 <endian7000> and is there some page summarizing aspects like this?
167 2011-02-28 03:00:27 <amiller> (i'm sure all the information is there, i've spent all day bouncing between wikis and forums and whitepapers)
168 2011-02-28 03:01:25 <amiller> how much would it cost for a government to 'shut down bitcoin' if desired
169 2011-02-28 03:01:35 <dirtyfilthy> two fiddy
170 2011-02-28 03:01:37 <Cusipzzz> bout tree fiddy
171 2011-02-28 03:01:40 <hexidigital> 5 million bitcoins
172 2011-02-28 03:01:42 <Cusipzzz> lol
173 2011-02-28 03:02:15 <jgarzik> $5 mil to p0wn bitcoin, I'd say.
174 2011-02-28 03:02:16 <endian7000> (network strength in 5970s) * (price of 5970) ?
175 2011-02-28 03:02:24 <amiller> it seems clear to me that it's not profitable to 'cheat' at bitcoin, but it seems like it would be possible to make it slightly worse for linear cost
176 2011-02-28 03:02:46 <dirtyfilthy> shutdowning down the main bitcoin site and the irc bootstrap would go a long way to fucking it up
177 2011-02-28 03:03:09 <jgarzik> not really
178 2011-02-28 03:03:37 <amiller> well in a lot of ways having more people use bitcoins and participating by hosting sites, the stronger it all gets
179 2011-02-28 03:03:49 <jgarzik> yes
180 2011-02-28 03:04:08 <gasteve> there is a maximum size to a block and a limit to the number of free transactions...a spammer would have to start paying bitcoins to get their transactions into blocks
181 2011-02-28 03:04:25 <gasteve> (paying bitcoins in the form of a transaction fee)
182 2011-02-28 03:04:30 <jgarzik> yes
183 2011-02-28 03:04:31 <gasteve> (that a miner collects)
184 2011-02-28 03:04:47 <endian7000> so a spammer could force everybody to pay transaction fees?
185 2011-02-28 03:04:48 <amiller> well a spammer could ruin free transactions for everyone
186 2011-02-28 03:04:50 <jgarzik> search the forums for MrBurns to see transaction spam
187 2011-02-28 03:04:57 <jgarzik> amiller: yes
188 2011-02-28 03:05:22 <amiller> well that would certainly slow down adoption...
189 2011-02-28 03:05:27 <amiller> are there any other effects like that?
190 2011-02-28 03:05:31 <endian7000> note to self: if I get into mining, it will be profitable for me to spam the network
191 2011-02-28 03:05:58 <jgarzik> amiller: it's inevitable that fees will be required.  nobody should be assuming bitcoin transactions are / will always be free.
192 2011-02-28 03:06:26 <amiller> in general, spammers can make the costs higher
193 2011-02-28 03:06:43 <jgarzik> when I started, many blocks only had 1 transaction (the 50 BTC generation transaction).  Looking at http://blockexplorer.com/ the average is now ~10 transactions per block, quite often.
194 2011-02-28 03:06:54 <jgarzik> even without spammers, the free slots will be used up.
195 2011-02-28 03:07:06 <hexidigital> tcatm:  FYI, I'm running this on a FreeBSD system with no changes.  Nice work for cross-platform compatibility. :)
196 2011-02-28 03:07:48 <endian7000> and we'll need sites/services to predict how much you'll need to pay
197 2011-02-28 03:08:14 <tcatm> hexidigital: That's great to hear. CLI or wxGUI?
198 2011-02-28 03:08:23 <endian7000> /aspect-for-your-node
199 2011-02-28 03:08:23 <hexidigital> tcatm:  It's a headless server, so CLI
200 2011-02-28 03:08:45 <amiller> what is the largest number of transactions in a block
201 2011-02-28 03:09:22 <amiller> what would happen if i made a thousand little addresses and gave them each .01BTC and had them all trade every minute
202 2011-02-28 03:09:41 <amiller> would that A. use up all the free transactions and B. cause a significant amount more work or C. not have any effect or not be viable
203 2011-02-28 03:09:43 <dirtyfilthy> you'd start paying transaction fees
204 2011-02-28 03:10:03 <luke-jr> no
205 2011-02-28 03:10:05 <luke-jr> everyone else would
206 2011-02-28 03:10:14 <ArtForz> nope, your transactions wouldnt make it into a block
207 2011-02-28 03:10:19 <luke-jr> and I would REALLY modify my miner to only include tx with fees
208 2011-02-28 03:10:29 <ArtForz> tx with fees always get priority
209 2011-02-28 03:10:41 <gasteve> btw, in anticipation of transaction fees, a good feature to have in the client is an ability to move keys between wallets (instead of using transactions to do that)
210 2011-02-28 03:10:47 <luke-jr> I might even release a branch that prioritizes by tx amount
211 2011-02-28 03:10:55 <amiller> so if i did that
212 2011-02-28 03:10:55 <ArtForz> for the rest ,they're scored bage on age of inputs, amount of coins, size of transaction in bytes, ...
213 2011-02-28 03:11:03 <amiller> but only as a free transaction
214 2011-02-28 03:11:12 <amiller> it would use up all of the free transactions for everyone indefinitely at no expense to me
215 2011-02-28 03:11:17 <ArtForz> you'd get about 10 transactions into every block
216 2011-02-28 03:11:39 <luke-jr> ah, ArtForz has a good point
217 2011-02-28 03:11:45 <luke-jr> your .01 coins would be newer than the rest
218 2011-02-28 03:11:46 <jgarzik> Current bitcoin transaction flow rate:  2104 transactions over the past 24 hours.  If you assume one block every 10 minutes, that is ~15 transactions per block.  That's around 3k, and the free transaction space is 27k.
219 2011-02-28 03:12:14 <amiller> it seems like i could make an app engine script that would use up all the free things once and for all
220 2011-02-28 03:12:15 <jgarzik> it is inevitable that the free tx area will be full 100% of the time, in a few years.
221 2011-02-28 03:12:38 <jgarzik> amiller: yep
222 2011-02-28 03:12:49 <ArtForz> and you wouldnt be the first to try
223 2011-02-28 03:13:09 <amiller> what happened to the others :p
224 2011-02-28 03:13:33 <amiller> i suppose it would be rude to put up a bounty to do that
225 2011-02-28 03:13:45 <ArtForz> they succeeded for a short wile, then we added rate limiting and free tx scoring
226 2011-02-28 03:14:14 <amiller> is there going to be competition for transaction fees
227 2011-02-28 03:14:24 <amiller> that seems like the natural sort of thing that would occur
228 2011-02-28 03:14:27 <ArtForz> ?
229 2011-02-28 03:14:41 <amiller> first everyone will start requiring transaction fees, then sorting by the fee amount
230 2011-02-28 03:14:52 <hexidigital> tcatm:  I'm running the shipped binary with linux emulation.  I plan to see what's needed for native freebsd support in my "spare time" hopefully soon.  Interested in patches, assuming I can provide them?
231 2011-02-28 03:15:07 <jgarzik> hexidigital: we are always interested in patches
232 2011-02-28 03:15:13 <jgarzik> hexidigital: or clean pull requests
233 2011-02-28 03:15:29 <hexidigital> cool
234 2011-02-28 03:15:29 <luke-jr> jgarzik: yeah right, I still don't see my bugfix merged :p
235 2011-02-28 05:38:34 <grondilu> I tired to compile bitcoin-0.3.20 manually and it didn't go well.  I had to create the directory obj/nogui manually and finally I had no "bitcoind" executable created !?
236 2011-02-28 05:38:48 <grondilu> s/tired/tried/
237 2011-02-28 05:41:20 <grondilu> this is messed up
238 2011-02-28 05:41:22 <gasteve> grondilu: did you do make the "bitcoind" target?
239 2011-02-28 05:41:52 <grondilu> yes, I ran "make -f makefile.unix bitcoind"
240 2011-02-28 05:43:06 <gasteve> strange that you had to make that nogui directory...maybe it's permission related?
241 2011-02-28 05:44:07 <gasteve> when I build (I've only built on osx) from the master branch (which is a bit newer than 0.3.20), nogui and the .o files instead were automatically created
242 2011-02-28 05:44:25 <gasteve> s/instead/inside/
243 2011-02-28 05:48:52 <grondilu> well I have problems linking boost anyway
244 2011-02-28 05:49:00 <grondilu> even with 0.3.19
245 2011-02-28 05:49:58 <grondilu> maybe it's because my libboost-all-dev version is 1.42.0.1 when then bitcoin install file tlaks about version 1.40
246 2011-02-28 05:57:02 <grondilu> nevermind
247 2011-02-28 06:09:46 <alkor> Has anybody tried to compile bitcoind with clang++ on Mac OS X?
248 2011-02-28 06:10:45 <mmagic> ;;bc,stats
249 2011-02-28 06:10:47 <gribble> Current Blocks: 111029 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1866 blocks | Next Difficulty In About: 1 week, 4 days, 1 hour, 23 minutes, and 12 seconds | Next Difficulty Estimate: 65619.10652080
250 2011-02-28 06:17:12 <mmagic> yay! I see concavity!
251 2011-02-28 07:17:05 <necrodearia> Are there any charity organizations that actively participate in Bitcoin community?  e.g. are there any individuals representative of a charity that frequent this IRC channel?
252 2011-02-28 07:17:14 <alkor> Where can I access the source for the latest release of bitcoin?
253 2011-02-28 07:17:20 <necrodearia> alkor, http://bitcoin.org
254 2011-02-28 07:17:43 <alkor> Can you point me to a .tar.gz or .zip file with the source?
255 2011-02-28 07:17:49 <necrodearia> sure
256 2011-02-28 07:18:00 <alkor> I can only see binaries on the front page.
257 2011-02-28 07:18:24 <necrodearia> http://downloads.sourceforge.net/project/bitcoin/Bitcoin/bitcoin-0.3.20/bitcoin-0.3.20.01-linux.tar.gz?r=http%3A%2F%2Fwww.bitcoin.org%2F&ts=1298881084&use_mirror=surfnet
258 2011-02-28 07:19:44 <edcba> aren't source included with install anymore ?
259 2011-02-28 07:19:44 <mmagic> it's easier to get the source with git or svn
260 2011-02-28 07:19:46 <necrodearia> https://en.bitcoin.it/wiki/Trade#Donation-accepting_organizations_and_projects
261 2011-02-28 07:20:48 <edcba> mmagic: yes of course :)
262 2011-02-28 07:23:41 <mmagic> does mtgox LR withdrawal always go offline this time of day?
263 2011-02-28 07:26:52 <uni4dfx> where can i buy ASICs?
264 2011-02-28 07:27:34 <mmagic> pretty sure you can't.
265 2011-02-28 07:28:09 <dissipate> uni4dfx, ArtForz is creating some supposedly
266 2011-02-28 07:28:31 <mmagic> for sale?
267 2011-02-28 07:28:35 <uni4dfx> well he better hurry up because if the diff keeps going up like it is now they'll be worthless no matter how powerful
268 2011-02-28 07:29:10 <dissipate> mmagic, not that i know of, but if you contact him you can find out.
269 2011-02-28 07:29:21 <dissipate> uni4dfx, why is that?
270 2011-02-28 07:29:32 <mmagic> i'm pretty sure all art keeps saying is, "not now"
271 2011-02-28 07:29:47 <edcba> uni4dfx: do you know vhdl ?
272 2011-02-28 07:30:03 <uni4dfx> edcba no, who/what is that?
273 2011-02-28 07:30:04 <mmagic> they're not powerful. they're about as fast as a single 5870. they just do it in 1/6 the power (apparently)
274 2011-02-28 07:30:23 <uni4dfx> oh then they're already worthless
275 2011-02-28 07:30:26 <mmagic> uni4dfx: he's roundabout saying that you will probably need to design your own chips.
276 2011-02-28 07:30:27 <uni4dfx> power is the least expensive thing
277 2011-02-28 07:30:36 <edcba> lol
278 2011-02-28 07:30:45 <mmagic> what a strange way of looking at it.
279 2011-02-28 07:31:29 <edcba> but you should be able to get them cheaper than gpu cards if you buy enough theoritically
280 2011-02-28 07:32:51 <mmagic> the cheapness comes in when it's no longer power-economical to run gpu cards. else, you can run gpu for the life the card and be pretty much ok
281 2011-02-28 07:34:05 <dissipate> my question is how many ASICs can you run in parallel, hooked up to 1 machine?
282 2011-02-28 07:34:24 <edcba> a lot ?
283 2011-02-28 07:34:54 <dissipate> then that is way cheaper than GPUs. you can only cram so many GPUs into one motherboard.
284 2011-02-28 07:34:57 <mmagic> probably 100s
285 2011-02-28 07:35:13 <dissipate> whereas with the ASICs you could stack them up pretty good
286 2011-02-28 07:36:04 <mmagic> the infrastructure to run structured asic (because real asic is probably not feasible for a long, long time yet) is significant.
287 2011-02-28 07:36:10 <dissipate> it seems the cost savings is also in being able to hook them all up to 1 box instead of needing multiple machines for the equivalent number of GPUs
288 2011-02-28 07:37:11 <edcba> if you can daisy chain them i don't see why you'd need more than a box
289 2011-02-28 07:37:43 <ArtForz> well, you're limited by airflow and PSUs
290 2011-02-28 07:38:44 <ArtForz> my current design uses full speed usb to connect the ASIC modules to a PC
291 2011-02-28 07:39:23 <mmagic> and controller chips and USB bus chains.. :)
292 2011-02-28 07:39:40 <ArtForz> controller is a spartan6 lx16 fpga
293 2011-02-28 07:39:52 <ArtForz> I dont even use a USB transceiver :P
294 2011-02-28 07:41:53 <ArtForz> usb hubs are cheap as fuck
295 2011-02-28 07:43:26 <ArtForz> each chip is ~7W, 8-chip module is ~64W, complete 2U unit is ~290W
296 2011-02-28 07:43:39 <ArtForz> 2U unit gets 6.4Ghps
297 2011-02-28 07:44:55 <dissipate> wow
298 2011-02-28 07:45:02 <mmagic> oh, i thought you said power was 1/6 for equivalent hashing, and each chip got 300Mhash?
299 2011-02-28 07:45:10 <mmagic> that's way less power..
300 2011-02-28 07:45:11 <ArtForz> 200Mhash/chip
301 2011-02-28 07:45:15 <dissipate> going to get some major coin with with that. :D
302 2011-02-28 07:45:23 <mmagic> ahhh..  interesting
303 2011-02-28 07:45:49 <ArtForz> I could up Vcore and clock, but I'm thermally limited
304 2011-02-28 07:46:20 <mmagic> that's still more like 1/15 isn't it?
305 2011-02-28 07:46:46 <ArtForz> I calculated efficiency at 11x my 5970s
306 2011-02-28 07:47:07 <edcba> 200mhash with 16 fpga ?
307 2011-02-28 07:47:08 <mmagic> are you disregarding the m/b + cpu of the PC?
308 2011-02-28 07:47:16 <edcba> ouch
309 2011-02-28 07:47:31 <ArtForz> pretty much
310 2011-02-28 07:47:37 <ArtForz> mainboard is ~20W total
311 2011-02-28 07:48:15 <ArtForz> and I only need one PC
312 2011-02-28 07:48:51 <dissipate> i'm surprised others aren't jumping on the ASIC bandwagon
313 2011-02-28 07:49:01 <ArtForz> master 2U contains mainboard, slave 2Us just have ASIC modules, PSU and external usb
314 2011-02-28 07:49:32 <mmagic> the investment is significant, both in engineering resources and capital; the returns are volatile and risky.
315 2011-02-28 07:49:35 <ArtForz> yep
316 2011-02-28 07:49:39 <mmagic> i'm not surprised at all.
317 2011-02-28 07:49:57 <mmagic> not really much of a bandwagon. :)
318 2011-02-28 07:49:59 <ArtForz> and even at current difficulty GPUs still provide way better ROI
319 2011-02-28 07:50:07 <dissipate> anyone operate a semiconductor company and can design and fabricate custom SHA256 chips? that would be upping the ante. :O
320 2011-02-28 07:50:45 <mmagic> i'm thinking that they're better if they improve in speed+ability..
321 2011-02-28 07:50:57 <dissipate> how much mhashes could one get with a real custom chip?
322 2011-02-28 07:51:08 <ArtForz> about 1Ghps/chip shouldnt be too hard
323 2011-02-28 07:51:27 <dissipate> wow
324 2011-02-28 07:51:57 <dissipate> just an array of 500 chips and you could generate as much ghash as the entire bitcoin network.
325 2011-02-28 07:52:01 <ArtForz> yep
326 2011-02-28 07:52:18 <dissipate> if anyone has a few million dollars sitting around... :D
327 2011-02-28 07:52:24 <ArtForz> and it'll only cost you ... about 2 million
328 2011-02-28 07:52:34 <mmagic> if you're willing to spend that much there are much more lucrative things you could be doing with your money. :)
329 2011-02-28 07:52:37 <ArtForz> yep
330 2011-02-28 07:52:50 <dissipate> i have a friend who works at a semiconductor company. he is an RFIC designer. but he can't design and fabricate whatever he wants. haha.
331 2011-02-28 07:53:15 <ArtForz> not to mention you could just buy a shitload of GPUs, pull the same stunt and sell the GPUs afterwards
332 2011-02-28 07:53:46 <dissipate> well for 2 million the U.S. government could screw the bitcoin network.
333 2011-02-28 07:54:03 <dissipate> that's not cool
334 2011-02-28 07:54:08 <nevezen> and I bet 2 million is nothing for the government
335 2011-02-28 07:54:23 <ArtForz> oh, you'd need way less than that with GPUs
336 2011-02-28 07:54:32 <larsig> bitcoin price would go up:)
337 2011-02-28 07:54:48 <nevezen> would it?
338 2011-02-28 07:54:55 <dissipate> why?
339 2011-02-28 07:54:55 <larsig> i think so
340 2011-02-28 07:55:05 <nevezen> I'd think it'd go down
341 2011-02-28 07:55:07 <ArtForz> you'd only need 2000 5870s or so
342 2011-02-28 07:55:10 <dissipate> no, it would crash once people realized a hostile takeover was underway
343 2011-02-28 07:55:14 <larsig> if they were to hold 2 million, 3 million woud be left in circulation
344 2011-02-28 07:55:29 <larsig> uh, 19 million left
345 2011-02-28 07:55:31 <dissipate> larsig, i'm not talking about them buying BTC
346 2011-02-28 07:55:50 <larsig> no, generating coins and limiting the supply
347 2011-02-28 07:55:59 <dissipate> larsig, i'm talking about them taking over the network by generating over 50% of total hash output.
348 2011-02-28 07:56:13 <larsig> oh shit
349 2011-02-28 07:56:16 <mmagic> ;;bc,stats
350 2011-02-28 07:56:18 <gribble> Current Blocks: 111040 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1855 blocks | Next Difficulty In About: 1 week, 4 days, 0 hours, 20 minutes, and 15 seconds | Next Difficulty Estimate: 65386.34026033
351 2011-02-28 07:56:27 <ArtForz> and once you have a majority of hashrate, you can prevent any transaction from going through and stop anyone else from finding blocks
352 2011-02-28 07:56:31 <mmagic> still coming down, but barely..
353 2011-02-28 07:56:44 <larsig> yikes, why 50%?
354 2011-02-28 07:56:55 <dissipate> yep, network would be hosed
355 2011-02-28 07:57:36 <dissipate> larsig, https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_computing_power
356 2011-02-28 07:57:49 <nevezen> aside from a mining point of view, would bitcoins be worth anything if a 3rd (or 2/3) of coins mined out were thrown away?
357 2011-02-28 07:58:14 <dissipate> larsig, you could generate the longest block chain, invalidating the original block chain
358 2011-02-28 07:58:25 <larsig> taken that people were somewhat depending on them, i think so yes
359 2011-02-28 07:58:34 <mmagic> larsig: they could, over time, create their own chain, work only on their own blocks, and their chain would grow faster, on average, than the main bitcoin network's chain. all they have to do is simply cease working on incoming blocks, and transmit their own blocks as they find them.
360 2011-02-28 07:58:36 <larsig> i see..
361 2011-02-28 07:59:09 <dissipate> nevezen, what do you mean by 'thrown away'?
362 2011-02-28 07:59:47 <larsig> but can they refuse the other half of doing transactions?
363 2011-02-28 07:59:49 <nevezen> they mining out of coins and then discarding their wallet.dat files :)
364 2011-02-28 08:00:46 <dissipate> nevezen, that would be reducing the supply, causing the price to go up for a given demand.
365 2011-02-28 08:00:49 <larsig> http://www.youtube.com/watch?v=2_GjZ7i4A6M
366 2011-02-28 08:01:29 <dissipate> i can see how throwing away blocks would not have been so bad back in the day, but now at nearly $1 per BTC, seems rather foolish. :P
367 2011-02-28 08:02:49 <nevezen> It would be foolish
368 2011-02-28 08:02:54 <larsig> hehe, maybe. but listen, if someone controls 50% of the network, can they stop transactions from coming into the block chain?
369 2011-02-28 08:03:01 <nevezen> but not by an entity wish real money to burn
370 2011-02-28 08:03:12 <mmagic> larsig: yes, they just mine for blocks and never put any transactions in them.
371 2011-02-28 08:03:54 <nevezen> how is that different than what's happening at the moment, mmagic?
372 2011-02-28 08:04:14 <mmagic> nevezen: i don't know what you're talking about?
373 2011-02-28 08:04:40 <nevezen> "...them mining for blocks but never putting any transactions in them."
374 2011-02-28 08:04:52 <mmagic> no, what's "happening at the moment"?
375 2011-02-28 08:05:01 <dissipate> nevezen, at the moment no single person has control of the block chain, therefore they can only introduce blocks based on old blocks.
376 2011-02-28 08:05:28 <nevezen> what about pooled mining?
377 2011-02-28 08:05:32 <mmagic> nevezen: ah, i see what you're asking. what's happening now is people who mine for blocks are putting other peoples' transactions in the mined blocks willingly.
378 2011-02-28 08:05:37 <nevezen> it would appear they have "more" control
379 2011-02-28 08:05:57 <dissipate> nevezen, no pool has over 50% of hash power. although slush's pool has come dangerously close as far as i know.
380 2011-02-28 08:05:57 <nevezen> as they're the ones outputting more hash/s than a single user
381 2011-02-28 08:07:12 <dissipate> nevezen, you have to output more than 50% of total hash rate, not just a single user.
382 2011-02-28 08:08:44 <nevezen> oh? what mechanism let's that happen mmagic?
383 2011-02-28 08:09:02 <larsig> lets say I own 49% of the cpu power, then i will probably generate coins too?
384 2011-02-28 08:09:17 <larsig> even if someone else owns 51%
385 2011-02-28 08:09:28 <dissipate> larsig, nope
386 2011-02-28 08:09:44 <larsig> yikes
387 2011-02-28 08:09:50 <dissipate> larsig, as it was explained before, the longest block chain wins
388 2011-02-28 08:09:50 <mmagic> nevezen: mutual consent.
389 2011-02-28 08:10:05 <nevezen> I think it increases your chances of generating a block, but only "increases", not guarantees..
390 2011-02-28 08:10:25 <dissipate> larsig, if i have over 50% of hash power i can generate the longest block chain and dominate the whole network without worrying about anyone else's blocks.
391 2011-02-28 08:10:47 <dissipate> larsig, others might generate blocks and form their own chains, but i would dominate all of them.
392 2011-02-28 08:10:48 <mmagic> larsig: yes, you may, but only for a certain time until the 51% person statistically kicks your ass and the long split that happened suddenly collides overtop and erases all the work of the 49% person.
393 2011-02-28 08:11:01 <nevezen> I wonder what the distinction is between a bitcoin user whom doesn't generate blocks and those whom do..
394 2011-02-28 08:11:27 <dissipate> larsig, the 49% guy might be able to keep up for awhile, but eventually he would be dominated, depending on how things played out.
395 2011-02-28 08:11:36 <mmagic> nevezen: the distinction is one of them just passes stuff along, and one of them encodes the stuff in a block.
396 2011-02-28 08:11:38 <nevezen> I recall someone mentioning a while back that the miners are the ones "supporting" the network via mining
397 2011-02-28 08:11:55 <mmagic> nevezen: correct. mining is the only thing supporting the network.
398 2011-02-28 08:12:09 <larsig> will not the 49% guy be able to both generate and spend his transactions?
399 2011-02-28 08:12:16 <dissipate> nevezen, they are. they generate blocks which secures the transactions on the network.
400 2011-02-28 08:12:23 <nevezen> so in essence, just having the software without it generating does not contribute to the network at all?
401 2011-02-28 08:12:33 <dissipate> larsig, no, he would be completely shut out.
402 2011-02-28 08:12:34 <larsig> or will his transactions become invalid once taken over by a stronger chain?
403 2011-02-28 08:12:34 <mmagic> larsig: yes, and then one day the 51% guy does 1 block more, and everyone just throws away all the 49% work.
404 2011-02-28 08:12:46 <mmagic> larsig: that's it.
405 2011-02-28 08:13:05 <mmagic> nevezen: not really. just creates more nodes to pass data around.
406 2011-02-28 08:13:12 <dissipate> nevezen, it does contribute by allowing others to connect and download the block chain and receive TXs being relayed across the network.
407 2011-02-28 08:13:14 <xelister> interesting fact!!!
408 2011-02-28 08:13:19 <xelister> Konrganizer sucks cocks in hell
409 2011-02-28 08:13:23 <mmagic> nevezen: it does allow people to participate in the network directly.
410 2011-02-28 08:13:30 <nevezen> so when do people start charging transaction fees realistically?
411 2011-02-28 08:13:32 <xelister> running Korganizer while mining == instant crash of X server
412 2011-02-28 08:14:03 <dissipate> nevezen, i believe they can charge tx fees now. but it seems the miners are content getting BTC.
413 2011-02-28 08:14:07 <edcba> why it would be korganizer's fault ?
414 2011-02-28 08:14:08 <dissipate> from mining alone
415 2011-02-28 08:14:20 <nevezen> Why would you want to be running X when it's gpu mining?
416 2011-02-28 08:14:28 <dissipate> xelister, what happened? scam?
417 2011-02-28 08:14:29 <xelister> edcba: dunno
418 2011-02-28 08:14:41 <xelister> nevezen: to use computer
419 2011-02-28 08:15:15 <nevezen> I often attribute gpu mining to not using the X server at all..
420 2011-02-28 08:15:20 <edcba> you mean you can multitask with your computer ?!!!
421 2011-02-28 08:15:28 <xelister> edcba: hm?
422 2011-02-28 08:15:29 <nevezen> just go headless
423 2011-02-28 08:15:45 <xelister> you need at least one X screen to run miner afair
424 2011-02-28 08:15:53 <nevezen> oh really?
425 2011-02-28 08:16:01 <nevezen> can't you just do it via CLI?
426 2011-02-28 08:16:06 <ArtForz> nope
427 2011-02-28 08:16:15 <xelister> nevezen: can do it from CLI, but the CLI must be attached to X session
428 2011-02-28 08:16:31 <nevezen> well that sucks. :]
429 2011-02-28 08:16:36 <xelister> X.so +0FA3C suck(&cocks) + 0x38
430 2011-02-28 08:16:43 <xelister> X.so +0CA3C hell() + 0x591
431 2011-02-28 08:16:46 <xelister> ah! as I thought.
432 2011-02-28 08:17:03 <nevezen> I'm assuming you're running KDE?
433 2011-02-28 08:17:07 <xelister> nevezen: no
434 2011-02-28 08:17:17 <nevezen> ok
435 2011-02-28 08:17:45 <xelister> lets try it again.
436 2011-02-28 08:17:53 <nevezen> I was thinking of controlling it via serial console
437 2011-02-28 08:18:16 <xelister> ok it doesnt crash when not oc it seems
438 2011-02-28 08:18:26 <xelister> so in other words...
439 2011-02-28 08:18:54 <xelister> KDE generates as much GPU use something (I guess gpu-mem transfers?) as playing a movie (this is 2nd known thing to instcrash mining X).
440 2011-02-28 08:18:58 <xelister> good job KDE developers
441 2011-02-28 08:19:11 <nevezen> btw, where do you usually download the latest blocks from?
442 2011-02-28 08:19:15 <nevezen> from other users?
443 2011-02-28 08:19:37 <xelister> I think flat rates in Honk Kong will go way up, with all this developers lining up to suck a dong.
444 2011-02-28 08:19:38 <nevezen> don't use kde
445 2011-02-28 08:19:51 <xelister> nevezen: I dont use KDE de actually
446 2011-02-28 08:20:05 <xelister> and korganizer is the less shitty TODO software linux has to offer
447 2011-02-28 08:20:15 <mmagic> you can run X headless.
448 2011-02-28 08:20:24 <nevezen> I don't really bother with those bloated desktop managers at all
449 2011-02-28 08:20:28 <mmagic> you don't need actual monitors connected to your machine. but you do have to run X
450 2011-02-28 08:20:36 <mmagic> nevezen: me neither. dwm all the way!!
451 2011-02-28 08:20:36 <xelister> nevezen: I. Do. Not. Use. KDE. Desktop. Manager.
452 2011-02-28 08:20:45 <mmagic> xelister: gnome?
453 2011-02-28 08:20:47 <nevezen> dwm or ratpoison
454 2011-02-28 08:20:52 <slush> bk128: ping
455 2011-02-28 08:20:55 <ArtForz> WM? my miners dont have no window manager :P
456 2011-02-28 08:20:59 <nevezen> oh but ratpoison, I think, still requires X
457 2011-02-28 08:21:14 <xelister> nevezen: I bet korganizer will crash as well when executed inside ratpoison wm.
458 2011-02-28 08:21:19 <mmagic> i use dwm for this mining machine on which I am irc'ing.
459 2011-02-28 08:21:42 <nevezen> well I basically use screen, and console 100%
460 2011-02-28 08:21:47 <ArtForz> my miners only have a minimal X server and xterm
461 2011-02-28 08:21:53 <xelister> Sho_ sets a ban on *!*@unaffiliated/xelister
462 2011-02-28 08:21:54 <xelister> what a fag
463 2011-02-28 08:22:02 <mmagic> what channel?
464 2011-02-28 08:22:05 <xelister> why developers take so persoanlly when we notice they suck dongs
465 2011-02-28 08:22:08 <ArtForz> as VT switching with fglrx = bad idea
466 2011-02-28 08:22:16 <xelister> mmagic: kde obviously
467 2011-02-28 08:22:28 <mmagic> you just went in there to harp on them?
468 2011-02-28 08:22:35 <xelister> mmagic: no, to call the fags
469 2011-02-28 08:22:36 <xelister> =)
470 2011-02-28 08:22:46 <xelister> them
471 2011-02-28 08:22:53 <nevezen> does it make much of a difference in performance, between running and using X than just running it?
472 2011-02-28 08:22:59 <ArtForz> yes
473 2011-02-28 08:23:01 <xelister> ArtForz: btw I checked, I do NOT run KMS and still vt switches gives instacrash
474 2011-02-28 08:23:12 <mmagic> okay, that's pretty awesome.
475 2011-02-28 08:23:17 <nevezen> by how much?
476 2011-02-28 08:23:29 <xelister> nevezen: minimal desktop use takes like 2-3% of speed. so not big problem
477 2011-02-28 08:23:32 <ArtForz> yep
478 2011-02-28 08:24:07 <xelister> mmagic: I know, right? but by mistake I said the line reserved for ati developers normally
479 2011-02-28 08:24:17 <ArtForz> worst for hashrate murdering is dragging windows around
480 2011-02-28 08:24:31 <ArtForz> well, except for accelerated video playback and the pile of shit that is flash
481 2011-02-28 08:24:32 <nevezen> dragging windows around with special effects :)
482 2011-02-28 08:24:42 <xelister> ArtForz: or dragging a MOUSE CURSORT above kde application hehe. in older version of diablo/sdk especially
483 2011-02-28 08:24:43 <nevezen> composite
484 2011-02-28 08:24:47 <ArtForz> nope, just plain "dragging windows around"
485 2011-02-28 08:24:51 <nevezen> or compviz, ehehe
486 2011-02-28 08:24:54 <mmagic> xelister: fuckin-A man.
487 2011-02-28 08:25:09 <xelister> although some may point out that actually macintosh users are the fags
488 2011-02-28 08:25:24 <xelister> we need to invent new cursewords for software world.
489 2011-02-28 08:26:37 <xelister> so what's new.  It seems no longer it makes sense to buy mining boxes and expeect them to return own value?
490 2011-02-28 08:26:39 <mmagic> 27/02/2011 20:14, 0000da67, invalid or stale  <---- AAARG!
491 2011-02-28 08:27:35 <xelister> mmagic: on what are you mining? when got last block?
492 2011-02-28 08:28:04 <mmagic> xelister: like what hardware?
493 2011-02-28 08:28:18 <nevezen> I'm beginning to see alot of invalid/stale states when using the slush pool
494 2011-02-28 08:28:19 <xelister> yea
495 2011-02-28 08:28:28 <mmagic> xelister: just a 5970..
496 2011-02-28 08:28:29 <dissipate> xelister, once people start dominating with ASICs, time to dump GPUs.
497 2011-02-28 08:28:31 <nevezen> I guess my hardware has finally become irrelevant?
498 2011-02-28 08:28:48 <slush> nevezen: how long time ago?
499 2011-02-28 08:28:52 <slush> nevezen: it started?
500 2011-02-28 08:28:58 <mmagic> dissipate: gonna be a long time from now..
501 2011-02-28 08:29:21 <dissipate> mmagic, how so? ArtForz is gearing up.
502 2011-02-28 08:29:41 <dissipate> mmagic, are you thinking of custom semiconductors?
503 2011-02-28 08:30:03 <nevezen> Whenever I run m0mchil's opencl miner for about 15 hours, about 1/3 to 2/3's are stale
504 2011-02-28 08:30:22 <nevezen> last I noticed was a few days ago
505 2011-02-28 08:30:54 <mmagic> dissipate: one person, experimenting. everyone else is almost certainly using gpu
506 2011-02-28 08:31:09 <dissipate> mmagic, true
507 2011-02-28 08:31:25 <dissipate> mmagic, he will be the first or one of the first
508 2011-02-28 08:31:31 <dissipate> mmagic, then others will jump on the bandwagon
509 2011-02-28 08:31:46 <xelister> mmagic: you got the stale while mining solo?
510 2011-02-28 08:32:03 <dissipate> nevezen, i would get that too. not sure what caused it.
511 2011-02-28 08:32:08 <mmagic> xelister: yes.
512 2011-02-28 08:32:11 <mmagic> xelister: second time.
513 2011-02-28 08:32:14 <xelister> mmagic: auch
514 2011-02-28 08:32:22 <mmagic> xelister: fucking tell me about it.. :-(
515 2011-02-28 08:32:39 <xelister> mmagic: when did you mines last time? I can't mine any block since like a week+ on 5770. it was once per week then
516 2011-02-28 08:32:43 <xelister> *suppoosed to be ;)
517 2011-02-28 08:32:47 <xelister> a bit of bad luck I guess
518 2011-02-28 08:33:08 <nevezen> I just assumed someone beat me to a proof faster than I did..
519 2011-02-28 08:33:12 <mmagic> xelister: yesterday..  i have a few 5970 so i'll still be doing daily blocks for a while yet
520 2011-02-28 08:33:26 <nevezen> therefore, old hardware
521 2011-02-28 08:36:45 <mmagic> ... i hope. :-(
522 2011-02-28 08:38:00 <nevezen> so why is it bad to be buying mining hardware now?
523 2011-02-28 08:38:15 <nevezen> just that the return won't be as expected from all the hype? :)
524 2011-02-28 08:40:30 <slush> nevezen: it's weird. There were some problems days ago, but now it should be fixed. Do you say you're still receiving invalid/stale?
525 2011-02-28 08:40:47 <slush> nevezen: which nickname on the pool do you have?
526 2011-02-28 08:41:31 <xelister> nevezen: yes
527 2011-02-28 08:42:28 <mmagic> nevezen: there is a rapid drop in profit margins as miners get desperate to sell and are willing to take a hit in their profits to do so.
528 2011-02-28 08:43:44 <mmagic> nevezen: if i hadn't started back in december i wouldn't have started now; expected returns are stretching out pretty long-term..
529 2011-02-28 08:46:13 <nevezen> so miners are making/mining less coins
530 2011-02-28 08:46:48 <nevezen> killing miners' enthusiasm perhaps? :)
531 2011-02-28 08:46:52 <validus> whats a 6970 end up hitting near on khash/s?
532 2011-02-28 08:47:03 <nevezen> slush: as is 'nevezen'
533 2011-02-28 08:47:31 <nevezen> so it isn't slow hardware slush?
534 2011-02-28 08:47:52 <slush> no, you should see only 1% of invalid in normal case
535 2011-02-28 08:50:38 <slush> nevezen: you are not mining now, right?
536 2011-02-28 08:50:54 <slush> nevezen: so your troubles are related to situation before many days...?
537 2011-02-28 08:55:49 <Expletive> So I was wondering if anyone knew a motherboard that could run 4 or more PCI-E 2.0 cards all at x16 speed?
538 2011-02-28 08:56:53 <Expletive> You know why.
539 2011-02-28 08:57:57 <Expletive> Also, has anyone benchmarked hash rates for the 6970?
540 2011-02-28 08:58:41 <ArtForz> yes, and yes.
541 2011-02-28 09:00:58 <validus> whats it average?
542 2011-02-28 09:01:44 <ArtForz> well, if you could avoid the sdk 2.3 opencl brokenness, it'd be exactly as fast as a 5870
543 2011-02-28 09:02:53 <Expletive> Really?
544 2011-02-28 09:03:36 <Expletive> Also, what about that motherboard?
545 2011-02-28 09:04:31 <ArtForzZz> asus p6t7 supercomputer
546 2011-02-28 09:04:43 <Expletive> Thanks
547 2011-02-28 09:05:02 <ArtForzZz> 7 x16 slots, up to 4 * 2.0 x16 or 1 * 2.0 x16 + 6* 2.0 x8
548 2011-02-28 09:05:14 <ArtForzZz> and expensive as hell
549 2011-02-28 09:07:39 <Expletive> Well if you fill it with 7 5870s how long until it pays for itself?
550 2011-02-28 09:07:46 <ArtForz> a loong while
551 2011-02-28 09:07:53 <sipa> very hard to predict
552 2011-02-28 09:08:00 <edcba> depends on bitcoin miners
553 2011-02-28 09:08:00 <sipa> but definitely not immediately
554 2011-02-28 09:08:10 <ArtForz> cheaper to just get a whole bunch of 2*x16 @ x8 boards
555 2011-02-28 09:08:38 <Expletive> ArtForz: Then you have to get a bunch of CPUs/ PSUs/ETC
556 2011-02-28 09:08:47 <ArtForz> you need multiple PSUs anyways
557 2011-02-28 09:08:57 <ArtForz> and cpu+ram are pretty damn cheap
558 2011-02-28 09:09:04 <Expletive> Also, how would you keep something like that cool, anyway?
559 2011-02-28 09:09:05 <ArtForz> just get a sempron 140 and 1G of ddr3
560 2011-02-28 09:09:31 <ArtForz> you need to use risers anyway, so spread the cards around
561 2011-02-28 09:10:43 <Expletive> I'm just dreaming at the moment.
562 2011-02-28 09:11:05 <Expletive> I only have a 6870 and I just found my first block today. So I'm excited.
563 2011-02-28 09:11:36 <ArtForz> I have 6*5770, 8*5870, 28*5970, 2*6870, 2*6970
564 2011-02-28 09:12:23 <xelister> Expletive: what MHash are you getting?
565 2011-02-28 09:12:38 <Expletive> xelister: Around 230
566 2011-02-28 09:13:41 <ArtForz> thats actually not too bad, considering you have to use a crap SDK version
567 2011-02-28 09:15:19 <Expletive> I have catalyst 11.2
568 2011-02-28 09:16:08 <Expletive> It's around 232 or so normally but I'm using -f 60 to not slow down desktop performance and I'm getting between 228 and 230 now
569 2011-02-28 09:16:38 <Expletive> Took me a week to find my first block
570 2011-02-28 09:17:49 <Expletive> It says it matures in 78 more blocks
571 2011-02-28 09:18:01 <Expletive> What does that mean?
572 2011-02-28 09:18:36 <ArtForz> it takes 120 blocks before you can spend a generation
573 2011-02-28 09:18:43 <mmagic> Expletive: be careful you know in advance the growth curves for difficulty if you wanna get a big machine like that..
574 2011-02-28 09:19:08 <Expletive> mmagic: I can't afford a big machine like that.
575 2011-02-28 09:19:31 <ArtForz> well, imo its a bad idea to build something like that for bitcoin only with the way difficulty has been growing lately
576 2011-02-28 09:19:33 <Expletive> But I told somebody about bitcoin and they want to get in on it.
577 2011-02-28 09:19:38 <mmagic> Expletive: k. :)
578 2011-02-28 09:19:49 <ArtForz> but if you want a badass box for fucking with GPGPU anyways... why not ;)
579 2011-02-28 09:20:28 <Expletive> ArtForz: So how much money have you made so far?
580 2011-02-28 09:21:15 <Expletive> I mean, with a setup like that you have to be making some bank, right?
581 2011-02-28 09:21:17 <ArtForz> about -$20k
582 2011-02-28 09:21:36 <Expletive> Dang.
583 2011-02-28 09:21:52 <ArtForz> and yes, thats negative 20 grand
584 2011-02-28 09:21:55 <xelister> yey, you can get like minus 20 all night vip hookers for that
585 2011-02-28 09:22:09 <Expletive> Shit. What happened there?
586 2011-02-28 09:22:19 <xelister> can you please donate thie -20k usd to Ati?
587 2011-02-28 09:22:26 <ArtForz> mainly, $50k for ASICs happened
588 2011-02-28 09:22:57 <Expletive> You spent 50 grand on shoes?
589 2011-02-28 09:23:04 <TD> :-)
590 2011-02-28 09:23:08 <Expletive> What are you, a woman?
591 2011-02-28 09:23:11 <ArtForz> before that I was about $20k ahead
592 2011-02-28 09:23:16 <TD> ArtForz takes mining seriously
593 2011-02-28 09:23:33 <xelister> ArtForz: if you are going to jump, donate the boxes to opensource first
594 2011-02-28 09:24:17 <Expletive> But what do mean by ASICs since you probably don't mean shoes
595 2011-02-28 09:24:18 <xelister> with low W/hash you may still win right?
596 2011-02-28 09:24:34 <ArtForz> application specific integrated circuit
597 2011-02-28 09:24:39 <xelister> Expletive: AwesomeSexyIntegreatedCircuit
598 2011-02-28 09:24:50 <ArtForz> basically got myself 20Ghps of custom chips
599 2011-02-28 09:25:04 <satamusic> $50k in ASICs D:
600 2011-02-28 09:25:09 <xelister> always you can later sell them to mafia working to crack SSLs
601 2011-02-28 09:25:11 <Expletive> Oh, shit
602 2011-02-28 09:25:11 <xelister> =)
603 2011-02-28 09:25:38 <xelister> - Hey Sang-Chang, that crazy dude from EU is calling again
604 2011-02-28 09:25:42 <ArtForz> contemplating if I should sell off my GPUs and get another batch
605 2011-02-28 09:25:43 <UukGoblin> ;;bc,stats
606 2011-02-28 09:25:45 <gribble> Current Blocks: 111049 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1846 blocks | Next Difficulty In About: 1 week, 3 days, 22 hours, 32 minutes, and 32 seconds | Next Difficulty Estimate: 65540.00030519
607 2011-02-28 09:26:28 <Expletive> I'll give you 50 BTC for a 5970
608 2011-02-28 09:26:35 <UukGoblin> fuckin' harsh
609 2011-02-28 09:26:45 <UukGoblin> people who make up these difficulties should be shot
610 2011-02-28 09:26:48 <mmagic> difficulty estimate is up again. :)
611 2011-02-28 09:26:52 <xelister> woot
612 2011-02-28 09:27:09 <xelister> lets bet will it pass 100,000 in Marc
613 2011-02-28 09:27:24 <xelister> and when it will be 1 milion =)
614 2011-02-28 09:27:27 <ArtForz> guess I'll just pay the next batch out of pocket and sell the GPUs when I have those up
615 2011-02-28 09:27:37 <Expletive> Why is difficulty going up so much?
616 2011-02-28 09:27:49 <ArtForz> mainly... because price went up so much
617 2011-02-28 09:28:06 <mmagic> it was slashdotted and a bunch of people didn't count on the fact that other people are as smart as they are.
618 2011-02-28 09:28:48 <mmagic> and very few of them seem to be capable of rational projections. :)
619 2011-02-28 09:29:05 <Keefe> good way to put it
620 2011-02-28 09:29:06 <ArtForz> well, we should be back to <20%/period growth pretty soon
621 2011-02-28 09:29:51 <ArtForz> at least if price doesnt rise much above $1/btc
622 2011-02-28 09:29:52 <sipa> aha, my 3-day-window estimated speed decreased a bit, finally :)
623 2011-02-28 09:29:54 <Keefe> many of the new miners probably just looked at current price and difficulty, and extended that indefinitely and said hey this will pay off in just 3 weeks
624 2011-02-28 09:30:07 <mmagic> what i'm projecting is a lot of 5970 on ebay really soon by those people who bought outside their means.
625 2011-02-28 09:30:19 <Expletive> So how much does a 5970 hash for?
626 2011-02-28 09:30:24 <Keefe> 650 mhps
627 2011-02-28 09:30:28 <Keefe> if OC'd
628 2011-02-28 09:30:28 <mmagic> 650?
629 2011-02-28 09:30:33 <ArtForz> about 560Mhps stock, 650 OCed
630 2011-02-28 09:30:50 <mmagic> ... and well-cooled.
631 2011-02-28 09:30:56 <ArtForz> yep
632 2011-02-28 09:30:57 <Expletive> Why don't most GPUs support water cooling?
633 2011-02-28 09:31:10 <Keefe> most do, if you replace the cooler :P
634 2011-02-28 09:31:21 <ArtForz> ^...
635 2011-02-28 09:31:21 <TD> ArtForz: how much of the 50k was one-off prep work vs the per unit chip price?
636 2011-02-28 09:31:22 <mmagic> because water cooling is fiddly and nobody wants to be a plumber.
637 2011-02-28 09:31:26 <ArtForz> 30k one-off
638 2011-02-28 09:31:28 <TD> i mean will the next 20Gh cost you 50k too?
639 2011-02-28 09:31:30 <lfm> water cooling just adds expense
640 2011-02-28 09:31:31 <TD> ok
641 2011-02-28 09:31:35 <ArtForz> 20k for 100 chips
642 2011-02-28 09:31:45 <TD> so it's $1000 per gigahash?
643 2011-02-28 09:32:08 <ArtForz> well, if support components, PCBs, cases, ... grow on trees, yes
644 2011-02-28 09:32:13 <Keefe> you'd be better off putting a big box fan on your computer if you had to, instead of setting up water cooling
645 2011-02-28 09:32:14 <TD> ok
646 2011-02-28 09:32:28 <Expletive> So you did you mortgage your house to pay for that or what?
647 2011-02-28 09:32:33 <UukGoblin> sipa, I've got a new graph idea: difficulty / price ;-]
648 2011-02-28 09:32:53 <sipa> yeah, people have suggested that a few times
649 2011-02-28 09:33:13 <ArtForz> people have already done that ;)
650 2011-02-28 09:33:19 <sipa> indeed
651 2011-02-28 09:33:26 <ArtForz> http://bitcoin.atspace.com/income.html
652 2011-02-28 09:33:26 <Keefe> i have a chart of that, a week old iirc
653 2011-02-28 09:33:57 <Keefe> ah there we go. just be aware it's log
654 2011-02-28 09:34:00 <UukGoblin> aha! goes down!
655 2011-02-28 09:34:18 <Expletive> Income seems to be going steadily down.
656 2011-02-28 09:34:22 <mmagic> that is an irritatingly log graph
657 2011-02-28 09:34:33 <ArtForz> yep
658 2011-02-28 09:35:11 <ArtForz> still adjusting down for the Mhps/$ jump from GPU mining
659 2011-02-28 09:35:32 <Expletive> So how long until difficulty rises so high it'd be impossible to see return without a significant hardware investment?
660 2011-02-28 09:35:52 <ArtForz> that... depends on a lot of assumptions
661 2011-02-28 09:35:53 <mmagic> Expletive: that's kinda the million-dollar question.
662 2011-02-28 09:36:01 <UukGoblin> Expletive, 39 days, 4 hours, 32 minutes, 16 seconds
663 2011-02-28 09:36:04 <ArtForz> if you got free power, "never" is a good guess
664 2011-02-28 09:36:15 <UukGoblin> oh sorry no
665 2011-02-28 09:36:17 <UukGoblin> 42!
666 2011-02-28 09:36:22 <sipa> i
667 2011-02-28 09:36:56 <Blitzboom> mmagic: why a million dollar
668 2011-02-28 09:37:06 <Blitzboom> a million bitcoin is more impressive
669 2011-02-28 09:37:06 <UukGoblin> million bitcoin
670 2011-02-28 09:37:14 <mmagic> Blitzboom: just an expression, sheesh.
671 2011-02-28 09:37:16 <mmagic> =]
672 2011-02-28 09:37:23 <Blitzboom> change them ;)
673 2011-02-28 09:37:26 <ArtForz> a millibitcoin... not so much ;)
674 2011-02-28 09:37:50 <Keefe> it's the million dollar question cause that's about what it'd cost to get a totally custom chip fabbed :)
675 2011-02-28 09:37:54 <sipa> now, if you'd say 42 megabitcoin...
676 2011-02-28 09:38:00 <mmagic> i vote we overload megabit with one more meaning just to mock the SI people.
677 2011-02-28 09:38:27 <Blitzboom> haha, we should really label all decimals
678 2011-02-28 09:38:28 <Expletive> Also, if difficulty is there to prevent hardware from making bitcoin generation trivial, what is there to prevent sufficiently powerful hardware from counterfeiting bitcoins themselves?
679 2011-02-28 09:38:40 <slush> is 850W psu enough for 3x5870 and 1x 5970 ? Without voltage tuning
680 2011-02-28 09:38:48 <UukGoblin> mmagic, +1
681 2011-02-28 09:39:00 <Keefe> Expletive: same thing
682 2011-02-28 09:39:02 <mmagic> Expletive: lots of participation.
683 2011-02-28 09:39:07 <ArtForz> slush: you like living dangerous, eh?
684 2011-02-28 09:39:15 <UukGoblin> Expletive, more powerful hardware
685 2011-02-28 09:39:16 <slush> hehe :)
686 2011-02-28 09:39:32 <ArtForz> well, it *should* be enough at stock clocks
687 2011-02-28 09:39:39 <slush> 2x 5870 are like single 5970 and I'm running two 5970 on single PSU
688 2011-02-28 09:39:47 <ArtForz> OCed... probably not
689 2011-02-28 09:39:59 <sipa> slush: we use a 850W psu for 2x 5970, and it's drawing 650W in total
690 2011-02-28 09:40:00 <slush> so the question is if 5970 and 5870 on the same psu will work
691 2011-02-28 09:40:27 <sipa> so, i suppose it's possible to add another 150W to it, but no guarantees :)
692 2011-02-28 09:40:36 <Expletive> ArtForz: So since you're getting rid of them, I'll give you 50 BTC for a 5970
693 2011-02-28 09:40:46 <UukGoblin> 51!
694 2011-02-28 09:40:53 <Keefe> 500
695 2011-02-28 09:40:57 <ArtForz> well, AX850 is a pretty good PSU, can handle up to 950W or so before getting iffy
696 2011-02-28 09:41:34 <ArtForz> ripple gets worse quick beyond 100% load on +12 though
697 2011-02-28 09:41:42 <slush> I have CORSAIR CMPSU-850TX
698 2011-02-28 09:41:44 <ArtForz> and efficiency takes a massive nosedive
699 2011-02-28 09:41:59 <Expletive> UukGoblin, Keefe he has 28 of them. It's not like he can't sell them to us for 50 BTC each.
700 2011-02-28 09:42:04 <UukGoblin> ArtForz, ah, btw... if I want to test a cheap PSU, would be enough to plug it into a mobo and some gfx cards, and wire an oscilloscope to the +12V line in parallel and look for "noise"?
701 2011-02-28 09:42:15 <ArtForz> UukGoblin: yup
702 2011-02-28 09:42:17 <Keefe> but why should he, when he can sell them on ebay for over $500 each
703 2011-02-28 09:42:36 <ArtForz> DMM to monitor absolute voltage, any old 'scope to look at ripple
704 2011-02-28 09:43:05 <Keefe> but you can't measure whether the components are about to blow
705 2011-02-28 09:43:09 <ArtForz> I use a mix of a active load and a bunch of 12V bulbs as dummy loads for testing
706 2011-02-28 09:43:28 <Expletive> Keefe: The dollar is way down. It's hardly with the paper it's printed on. The bitcoin however, has nowhere to go but up.
707 2011-02-28 09:43:50 <Keefe> Expletive: then he sells them for $500 and immediately buys 600 BTC with that
708 2011-02-28 09:44:12 <eps> one thing i haven't seen yet, bitcoin loans
709 2011-02-28 09:44:15 <xelister> or 6000 BTC if the bubble bursts =)
710 2011-02-28 09:44:16 <UukGoblin> ah, 12V bulbs
711 2011-02-28 09:44:20 <ArtForz> a box full of H4 bulbs from junkyard cost me less than a single 50W power resistor ;)
712 2011-02-28 09:44:21 <Expletive> Keefe: Then he'd do it for altruism.
713 2011-02-28 09:44:40 <Keefe> Expletive: when he's $20K in the hole? i don't think so :P
714 2011-02-28 09:44:46 <UukGoblin> yeah these high-power resistors are... expensive
715 2011-02-28 09:44:59 <Blitzboom> lol bitcoin loans
716 2011-02-28 09:45:05 <Blitzboom> take one and in a year youre broke
717 2011-02-28 09:45:17 <mmagic> LOL
718 2011-02-28 09:45:20 <ArtForz> well, you *can* get by with lower rated power Rs if you watercool em
719 2011-02-28 09:45:28 <UukGoblin> hahah
720 2011-02-28 09:45:48 <ArtForz> a 10W resistor handles 50W submerged without any problem
721 2011-02-28 09:46:01 <UukGoblin> :-D
722 2011-02-28 09:46:17 <UukGoblin> I'll stick to bulbs and real load then
723 2011-02-28 09:46:20 <Expletive> Is there a difference between 2GB 5970s and 4GB 5970s?
724 2011-02-28 09:46:27 <ArtForz> yes, 2GB and about $500
725 2011-02-28 09:46:36 <mmagic> cheers all, time to sleep.
726 2011-02-28 09:46:36 <xelister> not 4 mining
727 2011-02-28 09:46:41 <mmagic> o
728 2011-02-28 09:46:57 <ArtForz> well, the 4GB ones usually are 5870x2, not 5970
729 2011-02-28 09:47:05 <xelister> unless the 4gb ones would have /worse/ support in drivers... everything is possible with Ati
730 2011-02-28 09:47:06 <Blitzboom> cu
731 2011-02-28 09:47:20 <ArtForz> = 1.1625V core, 850MHz core, 1200Mhz mem
732 2011-02-28 09:47:44 <Keefe> what's the 5970 stock V?
733 2011-02-28 09:47:50 <ArtForz> 1.05
734 2011-02-28 09:47:52 <Expletive> ArtForz: So you should totally open source those bitcoin ASICs
735 2011-02-28 09:48:08 <ArtForz> and the 5870x2s have beefier VRMs
736 2011-02-28 09:48:27 <UukGoblin> Expletive, I'm against... see what happened when ppl opensourced gfx miners?
737 2011-02-28 09:48:34 <ArtForz> 5970 has 3*50A phases per GPU, 5870x2 has 4*50A phases
738 2011-02-28 09:49:05 <Expletive> UukGoblin: Well if bitcoin values plummet, nobody will bother mining anymore.
739 2011-02-28 09:49:07 <UukGoblin> oh so there is a bit more difference
740 2011-02-28 09:49:26 <UukGoblin> Expletive, so the diff will go down
741 2011-02-28 09:49:36 <Expletive> Exactly
742 2011-02-28 09:49:39 <ArtForz> thats also the reason why OCed 5970s like to overheat their VRMs so much
743 2011-02-28 09:49:39 <Keefe> same cores, just better power and default settings
744 2011-02-28 09:49:44 <ArtForz> yep
745 2011-02-28 09:50:03 <Expletive> Everyone will have a lot of bitcoins and they won't be worth anything
746 2011-02-28 09:50:21 <ArtForz> the 3-phase VRM is run barely below spec at stock voltage/clocks
747 2011-02-28 09:50:23 <Keefe> iirc, the 5970 ref boards have room for extra vrms and power connections
748 2011-02-28 09:50:28 <ArtForz> yep
749 2011-02-28 09:50:34 <Expletive> Also, how do you report income from bitcoin mining on your taxes?
750 2011-02-28 09:50:36 <ArtForz> and the 5870x2s populate those ;)
751 2011-02-28 09:51:41 <UukGoblin> Expletive, funnier: how do you report loss / bankruptcy! :-]
752 2011-02-28 09:52:15 <ArtForz> same way you report income from goldfarming :P
753 2011-02-28 09:52:22 <UukGoblin> if the gov't helps banks during crisis times... and honest people pay taxes off bitcoin income... why shouldn't the gov't help miners in times of high difficulty and low price!
754 2011-02-28 09:53:22 <UukGoblin> free 5970s to lone mothers!
755 2011-02-28 09:53:41 <xelister> mtgox too big to fall
756 2011-02-28 09:59:27 <eps> anyone here seen inside job?
757 2011-02-28 10:00:05 <Expletive> I wish bitcoinsextoys.com had a better selection.
758 2011-02-28 10:09:50 <Expletive> So why is the exchange rate from Bitcoins to Yen so terrible?
759 2011-02-28 10:10:14 <MacRohard> what's terrible about it?
760 2011-02-28 10:10:38 <MacRohard> probably just not alot of people trading in yen i guess
761 2011-02-28 10:10:43 <Expletive> https://btcex.com/site/index/pair/5 I might be reading that chart but it seems way down
762 2011-02-28 10:11:28 <MacRohard> yea looks like tehre's just no trades
763 2011-02-28 10:11:51 <Expletive> Wasn't Bitcoin created by a japanese guy?
764 2011-02-28 10:11:58 <Keefe> can you get JPY in and out of that exchange?
765 2011-02-28 10:12:04 <MacRohard> maybe. maybe not.
766 2011-02-28 10:12:20 <MacRohard> Keefe, bank wire to some japanese company
767 2011-02-28 10:12:24 <Expletive> Who is Satoshi Nakamoto anyway?
768 2011-02-28 10:12:34 <Blitzboom> noone knows
769 2011-02-28 10:13:33 <Expletive> Is that because people like ArtForz might want to kill him?
770 2011-02-28 10:14:13 <Blitzboom> uh & what?
771 2011-02-28 10:14:51 <Keefe> wtf?
772 2011-02-28 10:15:00 <Expletive> Well if not for Satoshi Nakamoto and bitcoin, ArtForz  wouldn't be $20k in the hole.
773 2011-02-28 10:15:13 <Expletive> That seems like a good enough motive for murder.
774 2011-02-28 10:15:14 <Blitzboom> ArtForz is in debt?
775 2011-02-28 10:15:15 <sipa> Satoshi is everywhere. He is all around us, even now in this very room. You can see him when you at your screen or when you turn on your miner. You can feel him when you send a transaction, when you go to BBE, when you declare your income. He is the guy that you've been told created it all, to blind you from the truth.
776 2011-02-28 10:15:26 <Keefe> no, ArtForz just made an investment
777 2011-02-28 10:15:31 <Blitzboom> no, wait
778 2011-02-28 10:15:41 <Keefe> which will likely pay off big time
779 2011-02-28 10:15:51 <hazek> sup people
780 2011-02-28 10:15:53 <Blitzboom> satoshi is an entity created in the network
781 2011-02-28 10:16:12 <ArtForz> yep, it's called a high risk investment
782 2011-02-28 10:16:38 <Blitzboom> you must be pretty wealthy if you can afford to lose most of it
783 2011-02-28 10:16:40 <knotwork> do miners need the genesis block or any past transactions or could they work on any blockchain
784 2011-02-28 10:16:55 <ArtForz> well, not really wealthy, not really poor either
785 2011-02-28 10:17:00 <Blitzboom> although its not really an investment, because you can always sell your hardware
786 2011-02-28 10:17:13 <knotwork> on a whim just with latest head of chain info about next block tha chain wants solved?
787 2011-02-28 10:17:28 <ArtForz> thats the main problem with the ASICs, no resale value
788 2011-02-28 10:17:35 <Expletive> Who would buy $50K in bitcoin mining ASICs chips?
789 2011-02-28 10:17:52 <ArtForz> GPU boxes you can resell at a 30% or so loss
790 2011-02-28 10:18:26 <Blitzboom> so its not that high a risk
791 2011-02-28 10:18:31 <ArtForz> yep
792 2011-02-28 10:18:44 <Blitzboom> buying bitcoins is riskier
793 2011-02-28 10:19:09 <ArtForz> I'd say yes
794 2011-02-28 10:19:47 <ArtForz> while my ASICs arent really useful for anything besides bitcoin-sha256
795 2011-02-28 10:20:03 <Blitzboom> i dont think theres any other scenario than bitcoin becoming either worthless or incredibly valuable
796 2011-02-28 10:20:25 <Expletive> Who made that bitcoin income chart?
797 2011-02-28 10:20:43 <knotwork> it shouldn't become worthless because the hashing power behind it should have value for
798 2011-02-28 10:20:58 <knotwork> any competing / co-operating systems
799 2011-02-28 10:21:08 <Blitzboom> meh, what value is that?
800 2011-02-28 10:21:11 <slush> knotwork: hashing power itself has no value
801 2011-02-28 10:21:33 <knotwork> anyone wanting to start such distributed systems needs a plan to try to secure them against attack
802 2011-02-28 10:21:37 <slush> knotwork: the value has the final product, bitcoin, as transaction network
803 2011-02-28 10:21:48 <knotwork> that could be lead toward a market in hashing
804 2011-02-28 10:22:05 <Expletive> ArtForz: Are your ASICs running right now?
805 2011-02-28 10:22:11 <knotwork> what currently keeps each nation in Freeciv Galactic Milieu from starting its own bitcoin fork
806 2011-02-28 10:22:27 <comboy> ArtForz: they don't just compute sha256? can't you use them for breaking simlocks or something?
807 2011-02-28 10:22:30 <ArtForz> Expletive: half of em are
808 2011-02-28 10:22:33 <devon_hillard> Has anyone tested the HD 6970 yet?
809 2011-02-28 10:22:34 <knotwork> to compete with the Martian's BotCoins is how to protect against the Martian miners taking over
810 2011-02-28 10:22:42 <knotwork> the new national currency
811 2011-02-28 10:22:44 <Expletive> Why only half?
812 2011-02-28 10:23:17 <knotwork> but if many many many nations started such forks and miners shopped around looking for which
813 2011-02-28 10:23:36 <ArtForz> comboy: the sl3 stuff is sha1
814 2011-02-28 10:23:38 <knotwork> one to do a block for based on current exchange rates that might help them to stbilise
815 2011-02-28 10:24:00 <ArtForz> the chips can do pbkdf-sha256... but no one uses that for anything
816 2011-02-28 10:24:07 <devon_hillard> ArtForz: you use ASICs to crack hashes?
817 2011-02-28 10:24:23 <comboy> ArtForz: oh, thought it's also 256
818 2011-02-28 10:24:32 <ArtForz> devon_hillard: partially, yes
819 2011-02-28 10:24:46 <devon_hillard> ArtForz: what sort?
820 2011-02-28 10:24:59 <ArtForz> structured ASIC
821 2011-02-28 10:25:12 <ArtForz> fully pipelined double-sha256 engine
822 2011-02-28 10:26:07 <devon_hillard> sounds... expensive
823 2011-02-28 10:26:11 <ArtForz> it is
824 2011-02-28 10:26:23 <devon_hillard> I don't understand this 'intermediate between ASIC and FPGA'
825 2011-02-28 10:26:40 <ArtForz> well, thats what it is
826 2011-02-28 10:26:58 <devon_hillard> have you posted public info about it?
827 2011-02-28 10:27:11 <ArtForz> nope
828 2011-02-28 10:27:12 <ArtForz> in a true ASIC everything is custom
829 2011-02-28 10:27:31 <ArtForz> in a FPGA hardware is fixed, you just load a bitstream into SRAM
830 2011-02-28 10:27:31 <comboy> ArtForz: so currently it works like this opencl kernel? have you considered chip that would take getwork as input?
831 2011-02-28 10:28:03 <UukGoblin> comboy, nothing of the sort ;-]
832 2011-02-28 10:28:24 <UukGoblin> it's hardware
833 2011-02-28 10:28:27 <comboy> UukGoblin: you mean to difficult?
834 2011-02-28 10:28:30 <UukGoblin> it takes logic signals as input
835 2011-02-28 10:28:37 <Expletive> ArtForz: How many of these could you afford right now? http://www.bitcoinsextoys.com/index.php?main_page=product_info&products_id=170
836 2011-02-28 10:28:39 <comboy> UukGoblin: wow
837 2011-02-28 10:29:15 <UukGoblin> there has to be something that gets the sha stuff from the PC, possibly via some serial interface, and feeds it on the appropriate pins to the ASICs
838 2011-02-28 10:29:22 <comboy> UukGoblin: I mean that currently there is thing to hash as input, but theoretically it could have work as input and iterate nonces itself, couldnt it?
839 2011-02-28 10:29:26 <ArtForz> yes, a Spartan6 LX16 FPGA
840 2011-02-28 10:30:22 <ArtForz> full-speed USB to host, 8 point-to-point synchronous serial links to ASICs
841 2011-02-28 10:30:40 <UukGoblin> comboy, not really, the use case is a bit different here... regular sha chips will take data on input and produce sha on output quickly-ish
842 2011-02-28 10:31:01 <UukGoblin> comboy, here we want one data and a gazillion sha sums and give a nonce on output if it produces a low enough sha result
843 2011-02-28 10:31:22 <ArtForz> turns out using a FPGA there was a really good idea, as I managed to mess up the timing of the serial interface ...
844 2011-02-28 10:31:43 <devon_hillard> ArtForz: how much computing power would a spartan 3e starter kit get me?
845 2011-02-28 10:31:45 <comboy> UukGoblin: exactly, so that was my question, does it do that, or does it only compute double sha and the rest is outside
846 2011-02-28 10:31:55 <ArtForz> devon_hillard: not much
847 2011-02-28 10:31:56 <UukGoblin> ArtForz, well done :-)
848 2011-02-28 10:32:23 <ArtForz> it does nonce-incrementing internally
849 2011-02-28 10:32:31 <comboy> thanks
850 2011-02-28 10:32:58 <ArtForz> a S6 LX150 in -3 speed grade gets about 70Mh/s
851 2011-02-28 10:33:22 <devon_hillard> ArtForz: would you dare to guesstimate the hashes/s from spartan 3e or some other starter kit?
852 2011-02-28 10:33:31 <ArtForz> pfff, thats a bit tricky
853 2011-02-28 10:33:53 <ArtForz> havent played with S3/E/A/... in a long time
854 2011-02-28 10:33:57 <comboy> I'm curious about it too
855 2011-02-28 10:34:05 <ArtForz> S6 is just way better in every regard
856 2011-02-28 10:34:20 <devon_hillard> would be fun to have a verilog or vhdl bitcoin client
857 2011-02-28 10:34:34 <eps> so i have 8600GT which i can run at work (free power) and gets be 6000 k/hash
858 2011-02-28 10:34:39 <eps> is it worth bothering?
859 2011-02-28 10:34:44 <comboy> daveparrish: miner would be enough
860 2011-02-28 10:34:50 <ArtForz> ;;bc,gen 6000
861 2011-02-28 10:34:51 <gribble> The expected generation output, at 6000 Khps, given current difficulty of 55590.23763914 , is 0.108561692481 BTC per day and 0.00452340385337 BTC per hour.
862 2011-02-28 10:34:56 <devon_hillard> eps: if you don't pay for the power, sure
863 2011-02-28 10:35:00 <devon_hillard> :)
864 2011-02-28 10:35:27 <ArtForz> well, lets try a wild-ass guess
865 2011-02-28 10:35:30 <devon_hillard> way better than hogging up the CPU with the default client
866 2011-02-28 10:35:42 <eps> to clarify, i can only it run it 6000 overnight, during the day i need to use that machine so its runs at 3000
867 2011-02-28 10:35:50 <eps> i am currently on slushs pool
868 2011-02-28 10:36:01 <eps> but the rewards seem so low...
869 2011-02-28 10:36:10 <devon_hillard> eps: just grab a 5570 or 5750
870 2011-02-28 10:36:26 <devon_hillard> eps: low-power cards
871 2011-02-28 10:36:30 <eps> it is my work desktop, can't really replace th graphics card
872 2011-02-28 10:36:43 <ArtForz> I'd say... about 4Mhps for a S3E-500
873 2011-02-28 10:36:58 <comboy> huh
874 2011-02-28 10:37:01 <comboy> not much
875 2011-02-28 10:37:38 <devon_hillard> eps: sure you can, unscrew the case, stick the other card in :)
876 2011-02-28 10:37:42 <ArtForz> well, S3 isnt exactly a very fast arch
877 2011-02-28 10:37:57 <eps> devon_hillard, open plan office :(
878 2011-02-28 10:38:16 <devon_hillard> eps: just say the card is broken and you had a spare at home
879 2011-02-28 10:38:51 <eps> i actually have a 5770 at home, was testing it last night got around 15000 khash
880 2011-02-28 10:38:58 <eps> but i use that card
881 2011-02-28 10:39:03 <ArtForz> thats way too low
882 2011-02-28 10:39:10 <devon_hillard> should do much more than that
883 2011-02-28 10:39:13 <comboy> and do you know maybe how much power this S6 LX150 consume?
884 2011-02-28 10:39:20 <ArtForz> a 5770 should be getting around 150000
885 2011-02-28 10:39:34 <eps> oh yeah, sorry i dropped a 0 ;)
886 2011-02-28 10:39:38 <devon_hillard> about 10 times more, eps
887 2011-02-28 10:39:53 <ArtForz> comboy: simulation says 6.something W
888 2011-02-28 10:39:58 <eps> if i bought a card, would it pay for itself...
889 2011-02-28 10:40:17 <comboy> ArtForz: thx
890 2011-02-28 10:41:06 <devon_hillard> eps: if you have a card already, that's a sunk cost, so it's not hurting to be mining anyway
891 2011-02-28 10:41:16 <comboy> huh, so that seems really nice
892 2011-02-28 10:41:39 <comboy> I wonder how long it will take until we're gonna have open source vhdl miner
893 2011-02-28 10:41:45 <ArtForz> well, hash/W it's pretty decent compared to GPUs, hash/$ it sucks
894 2011-02-28 10:42:19 <devon_hillard> ArtForz: what timeframe would you suppose your setup would pay for itself?
895 2011-02-28 10:42:40 <ArtForz> depends on future difficulty growth
896 2011-02-28 10:42:55 <devon_hillard> assuming linear growth?
897 2011-02-28 10:43:13 <ArtForz> 7 months
898 2011-02-28 10:43:28 <comboy> so how much this  S6 LX150 cost? I cant really find anything
899 2011-02-28 10:43:32 <Keefe> devon_hillard: you mean exponential?
900 2011-02-28 10:43:46 <ArtForz> comboy: about $150 in bulk qty
901 2011-02-28 10:44:05 <comboy> that sounds quite good for this power consumption
902 2011-02-28 10:44:05 <Keefe> iow, linear on a log graph
903 2011-02-28 10:44:42 <devon_hillard> ArtForz: of course, you can sell your services as a 'security consultant'... a hash-breaking machine must be valuable in that context
904 2011-02-28 10:44:52 <comboy> maybe it will become next generation home heater after 5970 age :) or I guess it doesnt head as much with this power consumption
905 2011-02-28 10:45:18 <comboy> *heat
906 2011-02-28 10:45:23 <ArtForz> well, for general hash breaking something FPGA based imo makes more sense
907 2011-02-28 10:45:23 <devon_hillard> in fact, the presence of these machines makes now any hash-based security scheme obsolete
908 2011-02-28 10:46:10 <nextgens> devon_hillard> that's not true
909 2011-02-28 10:46:15 <nextgens> you've iterated salted hashes
910 2011-02-28 10:46:17 <devon_hillard> for any dollar you can spend, the NSA can spend 1 million
911 2011-02-28 10:46:28 <nextgens> and you've costier hashes than the sha2 family
912 2011-02-28 10:46:30 <comboy> daveparrish: do you know the difference between finding block and breaking sha256? :)
913 2011-02-28 10:46:36 <devon_hillard> nextgens: salts only prevent rainbow tables
914 2011-02-28 10:46:42 <nextgens> the NIST is selecting sha2 members depending on how fast they perform
915 2011-02-28 10:46:43 <Keefe> devon_hillard: we're only "cracking" ~48 bits out of 256
916 2011-02-28 10:47:10 <devon_hillard> Keefe: the assumption is that the attacker already knows your salt
917 2011-02-28 10:47:28 <devon_hillard> that's the point of salts, to prevent rainbow attacks, even when they know your salt
918 2011-02-28 10:47:36 <nextgens> yeah
919 2011-02-28 10:47:38 <nextgens> salt is public
920 2011-02-28 10:47:42 <ArtForz> ~48 bits is already > 8-char alphanum...
921 2011-02-28 10:47:45 <devon_hillard> but when they know the salt, the search space is significantly less than 256 bits
922 2011-02-28 10:47:57 <nextgens> ArtForz> that's why you use a hash
923 2011-02-28 10:48:07 <devon_hillard> the password space is much smaller than what can fit in 256 bits
924 2011-02-28 10:48:13 <ArtForz> the hash wont help you if the password is bad
925 2011-02-28 10:48:14 <comboy> ArtForz: I dont think it counds this way, we break 48bits of sha, not sha of 48bits
926 2011-02-28 10:48:23 <comboy> *counts, shit
927 2011-02-28 10:48:25 <nextgens> ArtForz> I'm not disputing that :)
928 2011-02-28 10:48:54 <nextgens> ArtForz> but the assumption for password cracking is that there's only so much entropy the humain brain remembers
929 2011-02-28 10:48:58 <nextgens> and that doesn't grow over time
930 2011-02-28 10:48:59 <ArtForz> simple fact is, we're doing ~500Gh/s, thats about equivalent to 1T sha256 hashes/sec ...
931 2011-02-28 10:49:03 <nextgens> contrary to processing power
932 2011-02-28 10:49:14 <devon_hillard> there is only one or two alternative to standard hashes: bcrypt and scrypt
933 2011-02-28 10:49:27 <devon_hillard> or simply a public-key scheme