1 2011-03-03 00:00:15 <luke-jr> phantomcircuit: that hurts share accuracy
   2 2011-03-03 00:00:24 <phantomcircuit> luke-jr, yeah it's a trade off
   3 2011-03-03 00:00:40 <dazoe> i have a question about rpcminer... when connected to my local bitcoin it says target=0000000000012dcd0000000000000000000000000000000000000000000000 but when connected to deepbit.net says target=00000000ffffffffffffffffffffffffffffffffffffffffffffffffffffff why is it different?
   4 2011-03-03 00:00:43 <phantomcircuit> ;;bc,calc 4000 1
   5 2011-03-03 00:00:44 <gribble> Error: invalid syntax (<string>, line 1)
   6 2011-03-03 00:01:03 <phantomcircuit> eh how do you do that?
   7 2011-03-03 00:01:03 <luke-jr> dazoe: so deepbit can tell how much you're mining
   8 2011-03-03 00:01:13 <luke-jr> phantomcircuit: calcd to specify difficulty
   9 2011-03-03 00:01:19 <phantomcircuit> ;;bc,calcd 4000 1
  10 2011-03-03 00:01:19 <gribble> The average time to generate a block at 4000 Khps, given the supplied difficulty of 1, is 17 minutes and 53 seconds
  11 2011-03-03 00:01:39 <[Tycho]> luke-jr, may me he is using some VERY old and slow computer ? :)
  12 2011-03-03 00:01:54 <xelister> error: [score][bad] 1-Cypress-core1:  COULD NOT REPORT THE BLOCK to server! Block found, but server was unreachable!!!
  13 2011-03-03 00:01:59 <xelister> still some blocks on slush are droped
  14 2011-03-03 00:02:01 <luke-jr> [Tycho]: unlikely
  15 2011-03-03 00:02:02 <xelister> *some shares
  16 2011-03-03 00:02:33 <knotwork> ;;bc.calc 500 1
  17 2011-03-03 00:02:34 <gribble> Error: "bc.calc" is not a valid command.
  18 2011-03-03 00:02:39 <knotwork> ;;bc,calc 500 1
  19 2011-03-03 00:02:39 <gribble> Error: invalid syntax (<string>, line 1)
  20 2011-03-03 00:02:44 <knotwork> ;;bc,calcd 500 1
  21 2011-03-03 00:02:44 <gribble> The average time to generate a block at 500 Khps, given the supplied difficulty of 1, is 2 hours, 23 minutes, and 9 seconds
  22 2011-03-03 00:02:46 <[Tycho]> He said that the problem hit him at 30 GH/s
  23 2011-03-03 00:02:50 jrabbit_ has joined
  24 2011-03-03 00:03:27 <knotwork> calc is way wrong, I my client claims to run at 400 to 600 or so khash
  25 2011-03-03 00:03:40 <knotwork> and gets 3 or 4 or more blocks per hour at difficulty 1
  26 2011-03-03 00:03:49 <jgarzik> knotwork: s/blocks/shares/
  27 2011-03-03 00:03:51 <phantomcircuit> ;;bc,calc 80000000
  28 2011-03-03 00:03:52 jrabbit has quit (Disconnected by services)
  29 2011-03-03 00:03:52 <gribble> The average time to generate a block at 80000000 Khps, given current difficulty of 55590.23763914 , is 49 minutes and 44 seconds
  30 2011-03-03 00:03:57 jrabbit_ is now known as jrabbit
  31 2011-03-03 00:03:57 <jgarzik> knotwork: blocks are not generated at difficulty 1.
  32 2011-03-03 00:04:19 <knotwork> they are if you start a whole new block chain of your own from scratch
  33 2011-03-03 00:04:31 <knotwork> and only use one cpu mining it at any one time
  34 2011-03-03 00:04:35 <phantomcircuit> luke-jr, even with diff=1, my core i3 would only have a 50% change of generating a single share
  35 2011-03-03 00:04:49 <sipa> per how much time?
  36 2011-03-03 00:05:17 <jrabbit> [Tycho]: your deepbit stuff seems to be working nicely
  37 2011-03-03 00:05:51 rgm3 has joined
  38 2011-03-03 00:06:29 <xelister> ;; bc,stats
  39 2011-03-03 00:06:30 <gribble> Current Blocks: 111433 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1462 blocks | Next Difficulty In About: 1 week, 2 days, 12 hours, 14 minutes, and 4 seconds | Next Difficulty Estimate: 59700.33444072
  40 2011-03-03 00:08:20 <[Tycho]> jrabbit, thanks :)
  41 2011-03-03 00:11:00 AmpEater has joined
  42 2011-03-03 00:11:12 rgm3 has quit (Quit: rgm3)
  43 2011-03-03 00:12:09 Necr0s has quit (Ping timeout: 250 seconds)
  44 2011-03-03 00:15:23 Jeroenz0r has quit ()
  45 2011-03-03 00:18:09 Jeroenz0r has joined
  46 2011-03-03 00:18:09 Jeroenz0r has quit (Changing host)
  47 2011-03-03 00:18:09 Jeroenz0r has joined
  48 2011-03-03 00:22:31 AmpEater has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
  49 2011-03-03 00:24:24 Cusipzzz has joined
  50 2011-03-03 00:25:29 sgornick has quit (Ping timeout: 250 seconds)
  51 2011-03-03 00:25:40 <Blitzboom> bitcoin.org down?
  52 2011-03-03 00:25:55 discHead has quit (Quit: discHead)
  53 2011-03-03 00:26:43 gasteve has joined
  54 2011-03-03 00:26:50 discHead has joined
  55 2011-03-03 00:26:50 discHead has quit (Changing host)
  56 2011-03-03 00:26:50 discHead has joined
  57 2011-03-03 00:30:51 discHead has quit (Client Quit)
  58 2011-03-03 00:31:33 RichardG has quit (Quit: EPIC QUIT MESSAGE)
  59 2011-03-03 00:34:31 <[Tycho]> !seen m0mchil
  60 2011-03-03 00:39:38 Raulo_ has quit (Ping timeout: 272 seconds)
  61 2011-03-03 00:39:59 sgornick has joined
  62 2011-03-03 00:40:26 sgornick has quit (Client Quit)
  63 2011-03-03 00:40:53 sgornick has joined
  64 2011-03-03 00:47:14 <RBecker> ;;seen m0mchill
  65 2011-03-03 00:47:14 <gribble> I have not seen m0mchill.
  66 2011-03-03 00:47:20 <RBecker> ;;seen m0mchil
  67 2011-03-03 00:47:20 <gribble> m0mchil was last seen in #bitcoin-dev 1 week, 5 days, 9 hours, 54 minutes, and 32 seconds ago: <m0mchil> both boost builds > 7-8 hours... it was slow on MY machine but not THIS slow :)
  68 2011-03-03 00:47:25 <RBecker> [Tycho], ^
  69 2011-03-03 00:51:27 midnightmagic has joined
  70 2011-03-03 00:52:17 Blitzboom_ has joined
  71 2011-03-03 00:52:43 <[Tycho]> Thanks.
  72 2011-03-03 00:52:54 <RBecker> yw
  73 2011-03-03 00:54:37 Blitzboom has quit (Ping timeout: 250 seconds)
  74 2011-03-03 00:56:44 Blitzboom has joined
  75 2011-03-03 00:56:44 Blitzboom has quit (Changing host)
  76 2011-03-03 00:56:44 Blitzboom has joined
  77 2011-03-03 00:57:16 Blitzboom has quit (Remote host closed the connection)
  78 2011-03-03 00:57:34 Blitzboom has joined
  79 2011-03-03 00:57:34 Blitzboom has quit (Changing host)
  80 2011-03-03 00:57:34 Blitzboom has joined
  81 2011-03-03 00:57:38 redMBA has joined
  82 2011-03-03 00:58:57 Blitzboom_ has quit (Ping timeout: 250 seconds)
  83 2011-03-03 00:59:19 <dazoe> i've got another question... When you send some bitcoins it's status is unconfirmed till it makes it into a block. with the block difficulty incressing doesn't that mean the delay till a tranaction makes it into a block goes up?
  84 2011-03-03 01:00:01 <[Tycho]> No.
  85 2011-03-03 01:00:25 <[Tycho]> Difficulty is increasing to make time interval between blocks more consistent.
  86 2011-03-03 01:00:34 <[Tycho]> It will allways be ~10 mins.
  87 2011-03-03 01:01:13 <dazoe> so even if the number of miners drops does that mean the difficulty will go down?
  88 2011-03-03 01:01:30 Necr0s has joined
  89 2011-03-03 01:01:31 <[Tycho]> Yes.
  90 2011-03-03 01:01:38 <[Tycho]> It did once.
  91 2011-03-03 01:03:32 <dazoe> how does it change? automatic?
  92 2011-03-03 01:03:54 <lfm> yup , every node in the net makes the same calculation
  93 2011-03-03 01:04:03 <JFK911> ;;bc,stats
  94 2011-03-03 01:04:05 <gribble> Current Blocks: 111438 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1457 blocks | Next Difficulty In About: 1 week, 2 days, 10 hours, 14 minutes, and 23 seconds | Next Difficulty Estimate: 59752.66089763
  95 2011-03-03 01:05:17 <dazoe> so i take it that it uses time stamps in prev blocks to calc that?
  96 2011-03-03 01:06:01 <dazoe> and when a node reports a block other nodes check it using the same method
  97 2011-03-03 01:06:04 noagendamarket has joined
  98 2011-03-03 01:06:30 <lfm> yup
  99 2011-03-03 01:09:12 <dazoe> so what happend at the last block. will bitcoin just die off and cease to exist?
 100 2011-03-03 01:09:16 Blitzboom has quit (Remote host closed the connection)
 101 2011-03-03 01:09:28 <dazoe> in year ~2033?
 102 2011-03-03 01:10:07 <lfm> naw, the fixed rewards go to zero but the fees are still there to reward miners
 103 2011-03-03 01:10:21 <lfm> year 2133 more like
 104 2011-03-03 01:12:11 AmpEater has joined
 105 2011-03-03 01:12:31 legion0501 has quit (Read error: Connection reset by peer)
 106 2011-03-03 01:12:32 gasteve has quit (Quit: gasteve)
 107 2011-03-03 01:14:24 Blitzboom has joined
 108 2011-03-03 01:14:24 Blitzboom has quit (Changing host)
 109 2011-03-03 01:14:24 Blitzboom has joined
 110 2011-03-03 01:14:33 legion050 has joined
 111 2011-03-03 01:15:25 <dazoe> from what i under stand it's searching for a block with a hash with the first x bits 0 out of (496?) there was a question here but i just figured it out on my own...
 112 2011-03-03 01:15:25 <Syke> I might actually stop mining a few minutes to play Boom http://www.youtube.com/watch?v=Xdk0auH5dVI&feature=related
 113 2011-03-03 01:16:01 <Syke> bulletstorm
 114 2011-03-03 01:17:26 <dazoe> looks like an AWESOME game
 115 2011-03-03 01:20:15 gasteve has joined
 116 2011-03-03 01:20:15 <[Tycho]> Four-barreled shotgun ? Oh...
 117 2011-03-03 01:20:56 <[Tycho]> Is it for PC ?
 118 2011-03-03 01:23:16 <TheKid> yes
 119 2011-03-03 01:25:05 j_ has joined
 120 2011-03-03 01:25:14 Syke has quit (Ping timeout: 276 seconds)
 121 2011-03-03 01:26:04 j_ has quit (Client Quit)
 122 2011-03-03 01:27:18 <xelister> TheKid: linux?
 123 2011-03-03 01:31:25 [Tycho] has quit (Ping timeout: 240 seconds)
 124 2011-03-03 01:34:52 <TheKid> xelister: don't think so
 125 2011-03-03 01:36:18 Syke has joined
 126 2011-03-03 01:36:40 Necr0s has quit (Ping timeout: 252 seconds)
 127 2011-03-03 01:39:05 larsig has joined
 128 2011-03-03 01:44:32 [Tycho] has joined
 129 2011-03-03 01:45:57 Mango-chan has quit ()
 130 2011-03-03 01:45:58 mtve has quit (Ping timeout: 240 seconds)
 131 2011-03-03 01:45:58 Mango-chan has joined
 132 2011-03-03 01:45:58 Mango-chan has quit (Changing host)
 133 2011-03-03 01:45:58 Mango-chan has joined
 134 2011-03-03 01:45:58 <Mango-chan> is slush's pool down
 135 2011-03-03 01:45:58 <TheKid> working for me
 136 2011-03-03 01:45:58 Guest28555 has quit (Ping timeout: 240 seconds)
 137 2011-03-03 01:45:58 nameless has quit (!~root@weowntheinter.net|Ping timeout: 240 seconds)
 138 2011-03-03 01:45:59 <Mango-chan> TheKid really?
 139 2011-03-03 01:45:59 <TheKid> no I'm lying to you
 140 2011-03-03 01:45:59 rcorreia has quit (Ping timeout: 250 seconds)
 141 2011-03-03 01:45:59 <TheKid> for fun and profit
 142 2011-03-03 01:45:59 <Mango-chan> hm
 143 2011-03-03 01:45:59 <Mango-chan> i can't ping the site
 144 2011-03-03 01:45:59 redMBA has quit (Ping timeout: 250 seconds)
 145 2011-03-03 01:46:00 jnd has quit (Ping timeout: 240 seconds)
 146 2011-03-03 01:46:00 <TheKid> mining.bitcoin.cz?
 147 2011-03-03 01:46:00 <Mango-chan> yeah
 148 2011-03-03 01:46:00 <Mango-chan> Pinging mining.bitcoin.cz [178.79.147.99] with 32 bytes of data:
 149 2011-03-03 01:46:00 <Mango-chan> Request timed out.
 150 2011-03-03 01:46:00 <Mango-chan> Request timed out.
 151 2011-03-03 01:46:00 <Mango-chan> Request timed out.
 152 2011-03-03 01:46:00 <Mango-chan> Request timed out.
 153 2011-03-03 01:46:00 <Mango-chan> work on my server
 154 2011-03-03 01:46:00 <TheKid> weird
 155 2011-03-03 01:46:00 <OneFixt> Mango-chan: feel free to try the bitpenny upgrade, it's open
 156 2011-03-03 01:46:00 <Mango-chan> weird
 157 2011-03-03 01:46:00 <TheKid> pings fine for me
 158 2011-03-03 01:46:00 <Mango-chan> did they fix it?
 159 2011-03-03 01:46:00 <Mango-chan> bitpenny
 160 2011-03-03 01:46:00 <Mango-chan> i thought the owner lost money or something
 161 2011-03-03 01:46:00 <OneFixt> nope, i'm the owner, it was in private mode while being upgraded
 162 2011-03-03 01:46:01 <TheKid> OneFixt: have you been making 10
 163 2011-03-03 01:46:01 <TheKid> 10%
 164 2011-03-03 01:46:01 <TheKid> like I suspect you will?
 165 2011-03-03 01:46:01 mtve has joined
 166 2011-03-03 01:46:01 <OneFixt> TheKid: theoretically
 167 2011-03-03 01:46:02 <TheKid> I'm asking actual numbers :P
 168 2011-03-03 01:46:02 <OneFixt> sorry, the only actual numbers that I can give you are the ones in your balance =)
 169 2011-03-03 01:46:15 <tcatm> the new bitcoincharts.com is live :)
 170 2011-03-03 01:46:16 <Mango-chan> s> 270 bitcoins
 171 2011-03-03 01:46:31 <TheKid> OneFixt: if that changes, let me know ;)
 172 2011-03-03 01:46:48 <TheKid> I'm curious as to whether you are profitable, and if so, if you'll lower fees\
 173 2011-03-03 01:47:45 phantomcircuit has quit (Remote host closed the connection)
 174 2011-03-03 01:48:04 redMBA has joined
 175 2011-03-03 01:48:05 [Tycho] has quit (Changing host)
 176 2011-03-03 01:48:05 [Tycho] has joined
 177 2011-03-03 01:49:03 <larsig> tcatm, nice man
 178 2011-03-03 01:51:35 malfy has joined
 179 2011-03-03 01:51:35 rcorreia has joined
 180 2011-03-03 01:51:35 nameless1 has joined
 181 2011-03-03 01:51:35 18VAAA2SJ has joined
 182 2011-03-03 01:52:51 malfy is now known as Guest11497
 183 2011-03-03 01:52:52 <OneFixt> tcatm: very nice
 184 2011-03-03 01:54:06 jnd has joined
 185 2011-03-03 01:56:25 Necr0s has joined
 186 2011-03-03 01:58:02 malfy_ has joined
 187 2011-03-03 02:00:39 noagendamarket has quit (Read error: Connection reset by peer)
 188 2011-03-03 02:03:53 Mango-chan has quit (Ping timeout: 272 seconds)
 189 2011-03-03 02:05:23 phantomcircuit has joined
 190 2011-03-03 02:06:56 Necr0s has quit (Ping timeout: 246 seconds)
 191 2011-03-03 02:08:46 <[Tycho]> Who is the owner of bitcoincharts.com ?
 192 2011-03-03 02:08:49 <jgarzik> where's mmarker?
 193 2011-03-03 02:08:51 <jgarzik> [Tycho]: tcatm
 194 2011-03-03 02:08:54 <jgarzik> http://www.bitcoin.org/smf/index.php?topic=3486.msg58644#msg58644
 195 2011-03-03 02:08:57 <bitbot> Yet one CPU SSE2 miner for Windows Connection refused.
 196 2011-03-03 02:10:45 twobitcoins has quit (Quit: Leaving)
 197 2011-03-03 02:10:46 mmarker has joined
 198 2011-03-03 02:10:46 mmarker has quit (Changing host)
 199 2011-03-03 02:10:46 mmarker has joined
 200 2011-03-03 02:12:22 <phantomcircuit> jgarzik, why would you even do that
 201 2011-03-03 02:12:32 <phantomcircuit> what's the point of a slightly faster cpu implementation?
 202 2011-03-03 02:12:41 <phantomcircuit> it's already just barely worth it
 203 2011-03-03 02:13:45 twobitcoins has joined
 204 2011-03-03 02:14:36 <[Tycho]> ;;seen tcatm
 205 2011-03-03 02:14:36 <gribble> tcatm was last seen in #bitcoin-dev 28 minutes and 20 seconds ago: <tcatm> the new bitcoincharts.com is live :)
 206 2011-03-03 02:14:51 <tcatm> I'm here ;)
 207 2011-03-03 02:15:02 <[Tycho]> Cool.
 208 2011-03-03 02:15:33 malfy_ has quit (Read error: Connection reset by peer)
 209 2011-03-03 02:16:22 <Syke> looks good
 210 2011-03-03 02:16:26 Bister has quit (Ping timeout: 264 seconds)
 211 2011-03-03 02:17:15 <LobsterMan> is there a changelog somewhere for the new version if bitcoin?
 212 2011-03-03 02:17:37 <LobsterMan> of*
 213 2011-03-03 02:18:30 <phantomcircuit> https://github.com/bitcoin/bitcoin/commits/master
 214 2011-03-03 02:18:36 Mango-chan has joined
 215 2011-03-03 02:18:36 Mango-chan has quit (Changing host)
 216 2011-03-03 02:18:36 Mango-chan has joined
 217 2011-03-03 02:18:55 ApertureScience has quit (Quit: Linux: because a PC is a terrible thing to waste)
 218 2011-03-03 02:19:10 <phantomcircuit> i like how EURO is trading at half the value of USD for BTC
 219 2011-03-03 02:19:31 <mmarker> grrr
 220 2011-03-03 02:19:35 <mmarker> So damn close :(
 221 2011-03-03 02:20:09 <phantomcircuit> indeed if you had enough cash you could make some money arrbitraging EUR/USD via BTC
 222 2011-03-03 02:20:53 * mmarker waits to check to see if he can get a makework done...
 223 2011-03-03 02:21:33 da2ce7 has quit (Remote host closed the connection)
 224 2011-03-03 02:22:36 <mmarker> ;;bc, calcd 12000
 225 2011-03-03 02:22:37 <gribble> Error: "bc," is not a valid command.
 226 2011-03-03 02:22:46 <mmarker> ;;bc,calcd 12000 1
 227 2011-03-03 02:22:46 <gribble> The average time to generate a block at 12000 Khps, given the supplied difficulty of 1, is 5 minutes and 57 seconds
 228 2011-03-03 02:23:13 <mmarker> wish I had the probabilty curve for this...
 229 2011-03-03 02:25:11 <mmarker> Hmm, I;m assuming the pools ask for difficulty 1 blocks?
 230 2011-03-03 02:25:48 <[Tycho]> Yes.
 231 2011-03-03 02:25:58 <mmarker> damn, I'm not getting lucky then :(
 232 2011-03-03 02:26:13 <lfm> ;;bc,gend 12000 1
 233 2011-03-03 02:26:13 <gribble> The expected generation output, at 12000 Khps, given the supplied difficulty of 1, is 12069.940567 BTC per day and 502.914190292 BTC per hour.
 234 2011-03-03 02:26:20 twobitcoins has quit (Ping timeout: 276 seconds)
 235 2011-03-03 02:26:45 <lfm> ;;bc,gend 12000
 236 2011-03-03 02:26:45 <gribble> (bc,gend <an alias, 2 arguments>) -- Alias for "echo The expected generation output, at $1 Khps, given the supplied difficulty of $2, is [math calc 50*24*60*60 / (1/((2**224-1)/$2*$1*1000/2**256))] BTC per day and [math calc 50*60*60 / (1/((2**224-1)/$2*$1*1000/2**256))] BTC per hour.".
 237 2011-03-03 02:26:50 <lfm> ;;bc,gen 12000
 238 2011-03-03 02:26:51 <gribble> The expected generation output, at 12000 Khps, given current difficulty of 55590.23763914 , is 0.217123384962 BTC per day and 0.00904680770675 BTC per hour.
 239 2011-03-03 02:27:47 <mmarker> How many blocks are on the testnet?
 240 2011-03-03 02:27:56 twobitcoins has joined
 241 2011-03-03 02:28:04 <AmpEater> ;;bc,gen 1200000
 242 2011-03-03 02:28:05 <gribble> The expected generation output, at 1200000 Khps, given current difficulty of 55590.23763914 , is 21.7123384962 BTC per day and 0.904680770675 BTC per hour.
 243 2011-03-03 02:28:22 twobitcoins has quit (Client Quit)
 244 2011-03-03 02:28:26 <mmarker> Hmm
 245 2011-03-03 02:28:33 <mmarker> get H==0
 246 2011-03-03 02:28:39 <mmarker> but not enough 0's :\
 247 2011-03-03 02:28:43 ApertureScience has joined
 248 2011-03-03 02:28:51 twobitcoins has joined
 249 2011-03-03 02:29:19 <tcatm> mmarker: 10161 blocks
 250 2011-03-03 02:29:24 <mmarker> hmm
 251 2011-03-03 02:29:28 <mmarker> must still be d/l then
 252 2011-03-03 02:31:11 <mmarker> Ok, on the testnet
 253 2011-03-03 02:31:59 <mmarker> Come on little miner...work!
 254 2011-03-03 02:32:02 da2ce7 has joined
 255 2011-03-03 02:32:04 da2ce7 has quit (Changing host)
 256 2011-03-03 02:32:04 da2ce7 has joined
 257 2011-03-03 02:32:35 <mmarker> ;;bc,calcd 12400 32
 258 2011-03-03 02:32:35 <gribble> The average time to generate a block at 12400 Khps, given the supplied difficulty of 32, is 3 hours, 4 minutes, and 43 seconds
 259 2011-03-03 02:32:39 <mmarker> Wait
 260 2011-03-03 02:32:42 <mmarker> grr
 261 2011-03-03 02:33:17 <mmarker> Back to the pool!
 262 2011-03-03 02:36:33 sigkill_ has joined
 263 2011-03-03 02:37:22 <mmarker> Ok, I *know* my sha hashing is good
 264 2011-03-03 02:37:27 <lfm> why diff 32?
 265 2011-03-03 02:38:31 <mmarker> that was what was on the testnet
 266 2011-03-03 02:38:55 <lfm> oh ok
 267 2011-03-03 02:43:14 <mmarker> hmm
 268 2011-03-03 02:43:37 <mmarker> Spam incoming:
 269 2011-03-03 02:43:41 <mmarker>  Proof: 000000000000000000000000000000000000000000000000509a809aee896fb2
 270 2011-03-03 02:43:41 <mmarker> Target: 00000000ffffffffffffffffffffffffffffffffffffffffffffffffffffffff
 271 2011-03-03 02:43:42 <mmarker> TrgVal? YES (hash < target)
 272 2011-03-03 02:43:42 <mmarker> j: 1
 273 2011-03-03 02:43:42 <mmarker> nonce: 0x827759, nonce_base: 0x827758
 274 2011-03-03 02:43:42 <mmarker> HashMeter(2): 8550232 hashes, 3176.90 khash/sec
 275 2011-03-03 02:43:43 <mmarker> DBG: sending RPC call:
 276 2011-03-03 02:43:43 <mmarker> {"method": "getwork", "params": [ "000000013eca59d5ce0478d209d85a1f58c68514f62fda551e2d9a4e00002e43000000003c3e2bc3589f64417463695b3e755810bd1b1b83fbfbfaf953360f2c21a9fcef4d6efddd1b012dcd59778200000000800000000000000000000000000000000000000000000000000000000000000000000000000000000080020000" ], "id":1}
 277 2011-03-03 02:44:06 <mmarker> I'm thinking that many 0's shouldn't be in the getwork back to the server
 278 2011-03-03 02:44:43 <mmarker> Course, I don't even see my nonce in there...
 279 2011-03-03 02:46:16 AmpEater has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
 280 2011-03-03 02:46:28 <phantomcircuit> mmarker, why would your nonce be in the getwork?
 281 2011-03-03 02:46:34 <phantomcircuit> you're just going to overwrite it anyways
 282 2011-03-03 02:46:48 <xelister> COULD NOT REPORT THE BLOCK to server! Block found, but server was unreachable!!!
 283 2011-03-03 02:46:50 <xelister> x10
 284 2011-03-03 02:46:54 <mmarker> Hmm, so what gets sent to the serber.
 285 2011-03-03 02:46:56 <xelister> on slush.  all ok with slush pool?
 286 2011-03-03 02:46:56 <mmarker> err, server
 287 2011-03-03 02:47:42 <jgarzik> mmarker: 128 hex chars of 'data'
 288 2011-03-03 02:47:59 <jgarzik> mmarker: er, 256 hex chars representing 128 bytes
 289 2011-03-03 02:48:07 <mmarker> well, what's in the data...since I dont think that big strings of 0's looks right
 290 2011-03-03 02:48:35 <jgarzik> mmarker: it's right:
 291 2011-03-03 02:48:40 <mmarker> hmm
 292 2011-03-03 02:48:41 <jgarzik> "00000001cf4a5f9fa73de207c6b0975a1e8f41e156250c1ed71ddf0b00011bc9000000007a80a897fc906f858e3c2498acf916ca9086695f0b333beed324a13f1a00fd0f4d6eff871b012dcd00000000000000800000000000000000000000000000000000000000000000000000000000000000000000000000000080020000"
 293 2011-03-03 02:48:59 <jgarzik> mmarker: you're just patching 4 bytes into offset 76, and leaving the rest alone
 294 2011-03-03 02:49:06 <mmarker> The nonce?
 295 2011-03-03 02:49:12 <jgarzik> mmarker: correct
 296 2011-03-03 02:49:44 <mmarker> Of which, I don't see it in the data :(
 297 2011-03-03 02:51:06 <jgarzik> mmarker: in the hex string, it's at offset 76*2
 298 2011-03-03 02:51:59 xelister has quit (Remote host closed the connection)
 299 2011-03-03 02:52:03 <mmarker> Yes. But see my debug. The nonce should be 0x00827759
 300 2011-03-03 02:52:06 <lfm> you have to find the nonce of course
 301 2011-03-03 02:52:10 <mmarker> Which is where it's wonky!
 302 2011-03-03 02:52:55 <mmarker> wait, I see it
 303 2011-03-03 02:52:57 <lfm> generally you start at 1 and increment till you find a good nonce
 304 2011-03-03 02:53:03 <jgarzik> mmarker: yeah I see that value
 305 2011-03-03 02:53:08 <mmarker> lfm: I know! :D
 306 2011-03-03 02:53:18 <mmarker> I'm trying to figure out why my code is still acting bogus :\
 307 2011-03-03 02:54:59 <luke-jr> jgarzik: he's talking about the solution, not the work
 308 2011-03-03 02:55:41 <mmarker> hmm
 309 2011-03-03 02:56:01 <mmarker> this is wierd
 310 2011-03-03 02:56:20 <jgarzik> mmarker: you're not byte-swapping the nonce too many times, are you?
 311 2011-03-03 02:56:30 <mmarker> jgarzik: Dont think so
 312 2011-03-03 02:57:10 <jgarzik> mmarker: the byte ordering is a bit non-standard, so it requires some attention (note differences between sha256_via and sha256_generic)
 313 2011-03-03 02:57:32 <mmarker> jgarzik: Yea, I'm copying more of the 4way code, really
 314 2011-03-03 02:57:50 <mmarker> but now I have something odder going on.
 315 2011-03-03 02:58:05 <jgarzik> mmarker: have you seen http://www.bitcoin.org/smf/index.php?topic=3486.msg58644#msg58644 ?
 316 2011-03-03 02:58:08 <bitbot> Yet one CPU SSE2 miner for Windows Connection refused.
 317 2011-03-03 02:58:28 <mmarker> Yes. He's distributing the binary blob of JWASM
 318 2011-03-03 02:58:52 <mmarker> which (a), I dont want to touch with a 10 foot pole, and B, it's just a chopped up hack of the original src dump
 319 2011-03-03 02:59:14 * jgarzik has never heard of JWASM
 320 2011-03-03 02:59:41 <mmarker> Bad license, MASM duplicate
 321 2011-03-03 03:00:13 <mmarker> well, Open Wacom license
 322 2011-03-03 03:00:26 bk128 has joined
 323 2011-03-03 03:04:12 <bk128> tcatm: new bitcoincharts design looks great
 324 2011-03-03 03:04:49 <tcatm> :)
 325 2011-03-03 03:04:52 <mmarker> Ok
 326 2011-03-03 03:08:54 <mmarker> SOMEHOW
 327 2011-03-03 03:08:58 <mmarker> I have no clue HOW
 328 2011-03-03 03:09:10 <mmarker> my code is doing the sha256(sha256()) properly
 329 2011-03-03 03:09:32 <mmarker> oh, wait
 330 2011-03-03 03:09:32 <mmarker> no
 331 2011-03-03 03:09:35 <mmarker> Damnit
 332 2011-03-03 03:09:48 noagendamarket has joined
 333 2011-03-03 03:10:51 Cusipzzz has quit (Ping timeout: 264 seconds)
 334 2011-03-03 03:11:34 <mmarker> hmm, 2nd pass
 335 2011-03-03 03:11:36 <mmarker> bad
 336 2011-03-03 03:14:42 EvanR_ has joined
 337 2011-03-03 03:14:54 <EvanR_> when did diff go from 36k to 55k
 338 2011-03-03 03:16:11 <bk128> last week?
 339 2011-03-03 03:16:42 <[Tycho]> mmarker, what's wrong ?
 340 2011-03-03 03:16:49 <EvanR_> the date! whats the date!
 341 2011-03-03 03:16:54 <EvanR_> nov 5 1955
 342 2011-03-03 03:17:03 <mmarker> tycho: My first pass through my sha256 code, the hash is good
 343 2011-03-03 03:17:04 citiz3n has joined
 344 2011-03-03 03:17:05 citiz3n has quit (Changing host)
 345 2011-03-03 03:17:05 citiz3n has joined
 346 2011-03-03 03:17:10 <mmarker> the 2nd pass, it's all messed up :(
 347 2011-03-03 03:17:49 <jgarzik> mmarker: you == chromicant, right?
 348 2011-03-03 03:17:53 <mmarker> yup
 349 2011-03-03 03:18:30 <EvanR_> the nullvoid diff list seems to be down?
 350 2011-03-03 03:18:31 <jgarzik> mmarker: is your cpuminer pull request really final?  new commits keep appearing.
 351 2011-03-03 03:18:33 <EvanR_> is there another
 352 2011-03-03 03:18:54 <mmarker> jgarzik: For ARM, it should be
 353 2011-03-03 03:18:55 <jgarzik> mmarker: I'm happy to merge arm and validation, if they are working
 354 2011-03-03 03:19:04 <mmarker> arm and validation are working
 355 2011-03-03 03:19:17 <mmarker> But I should be commiting to the SSE2 branch
 356 2011-03-03 03:20:36 <jgarzik> mmarker: can you create separate pull requests for ARM and validation, please?  I like to clean up the history a bit.  Like the linux kernel, for brand new code, we don't import history like:  commit AB "skeleton code"  commit BC "it builds!"  commit CD "doh fix a bug"  commit DE "dooh another bug"
 357 2011-03-03 03:20:55 <mmarker> Ok, I can do a rebase on my end
 358 2011-03-03 03:21:00 <jgarzik> mmarker: thanks
 359 2011-03-03 03:21:16 <jgarzik> mmarker: I'll close out the current PR
 360 2011-03-03 03:23:48 <mmarker> Thank.
 361 2011-03-03 03:24:13 <mmarker> So, this is just bizzare. Round 1 of sha256, hash is good. Second round, it blows up
 362 2011-03-03 03:24:20 <mmarker> time for the debugger
 363 2011-03-03 03:29:48 <mmarker> yea, something is bizzare
 364 2011-03-03 03:31:40 EvanR has quit (Disconnected by services)
 365 2011-03-03 03:31:43 EvanR_ is now known as EvanR
 366 2011-03-03 03:31:48 EvanR has quit (Changing host)
 367 2011-03-03 03:31:49 EvanR has joined
 368 2011-03-03 03:33:55 discHead has joined
 369 2011-03-03 03:34:09 <mmarker> Ok, never good when random data comes back
 370 2011-03-03 03:35:00 <mmarker> Yay! Mystery memory corruption!
 371 2011-03-03 03:38:16 <mmarker> Hmm...Oh snap. Globals bad.
 372 2011-03-03 03:39:10 brunner has joined
 373 2011-03-03 03:40:46 <JFK911> heh
 374 2011-03-03 03:42:16 chaos_ has joined
 375 2011-03-03 03:42:30 chaos_ is now known as xg0d
 376 2011-03-03 03:43:22 <xg0d> can someone explain the mining concept to me real quick? sorry for the n00b request
 377 2011-03-03 03:44:05 <jrabbit> my sisters macbook isn't even gettign hot running sha256 operations on the gpu
 378 2011-03-03 03:44:26 <jrabbit> xg0d: do you get how they work in context of bitcoin?
 379 2011-03-03 03:44:51 <jrabbit> xg0d: basically they verify and facilitate the trades, and for doing that they're rewarded
 380 2011-03-03 03:44:56 BurtyB has quit (Ping timeout: 246 seconds)
 381 2011-03-03 03:45:15 <jgarzik> mmarker: FYI new version release tonight.  You'll want to base your PR on that.
 382 2011-03-03 03:45:30 <mmarker> I'll pull before the work
 383 2011-03-03 03:45:38 bonsaikitten has quit (Ping timeout: 246 seconds)
 384 2011-03-03 03:45:50 <xg0d> ty jrabbit
 385 2011-03-03 03:46:00 <luke-jr> jgarzik: new version of what?
 386 2011-03-03 03:46:07 <jgarzik> luke-jr: cpuminer
 387 2011-03-03 03:46:10 <luke-jr> ah
 388 2011-03-03 03:46:21 <luke-jr> was gonna say… a bit early for a new bitcoind/wx
 389 2011-03-03 03:46:45 <luke-jr> jgarzik: any AVX/etc support yet? ;)
 390 2011-03-03 03:46:55 <mmarker> hmm.
 391 2011-03-03 03:46:59 <jgarzik> luke-jr: waiting for your PR on that
 392 2011-03-03 03:47:12 <mmarker> SO, my pointers for some reason are bogusified.
 393 2011-03-03 03:47:25 bonsaikitten has joined
 394 2011-03-03 03:47:27 <luke-jr> my PR? O.o
 395 2011-03-03 03:47:37 <mmarker> Pull Request
 396 2011-03-03 03:47:40 <luke-jr> o
 397 2011-03-03 03:47:46 <mmarker> i.e. Git codin!
 398 2011-03-03 03:48:00 <luke-jr> I haven't the slightest idea where I'd even begin with something like that
 399 2011-03-03 03:48:09 <mmarker> it's pretty simple
 400 2011-03-03 03:48:12 <luke-jr> there are no examples for any AVX stuff, and I don't know x86 assembly
 401 2011-03-03 03:48:13 BurtyB has joined
 402 2011-03-03 03:48:13 <mmarker> the code is pretty clean
 403 2011-03-03 03:48:22 <mmarker> luke-jr: That's my problem
 404 2011-03-03 03:48:26 <luke-jr> I also have no idea how SHA-2 works :P
 405 2011-03-03 03:48:39 <jrabbit> luke-jr: does anyone?
 406 2011-03-03 03:48:40 <mmarker> Read wikipedia
 407 2011-03-03 03:48:43 <mmarker> It's really simple
 408 2011-03-03 03:48:48 <jrabbit> :P
 409 2011-03-03 03:48:59 <luke-jr> mmarker: if it's really simple, why didn't we have it in 1980?
 410 2011-03-03 03:49:13 <mmarker> Well, we had DES :D
 411 2011-03-03 03:49:30 <mmarker> and someone probably had something....but didn't tell anyone
 412 2011-03-03 03:50:10 <jgarzik> mmarker: for validation, I would think you'd want to pick an actual block header where a nonce was found.  that would permit timing and algorithm fitness checks.
 413 2011-03-03 03:50:30 <xg0d> mmarker lol
 414 2011-03-03 03:50:46 <mmarker> jgarzik: Probably a good idea
 415 2011-03-03 03:51:36 <mmarker> Ok, something bizzare...I must be smashing a stack somewhere
 416 2011-03-03 03:52:45 <luke-jr> so can I use AVX in C without assembly? :P
 417 2011-03-03 03:52:54 <mmarker> you can, but god, it'll suck
 418 2011-03-03 03:53:00 <luke-jr> >_<
 419 2011-03-03 03:55:58 mmarker has quit (Quit: bbiab)
 420 2011-03-03 03:56:33 xelister has joined
 421 2011-03-03 03:57:18 larsig has quit (Read error: Connection reset by peer)
 422 2011-03-03 03:57:24 <luke-jr> yeah, this is futile
 423 2011-03-03 03:57:51 citiz3n has quit (Ping timeout: 264 seconds)
 424 2011-03-03 03:57:55 <luke-jr> sure, SHA-2 looks simple enough on wiki, but how to make use of AVX or SHLD/ADC is very non-obvious
 425 2011-03-03 03:57:56 larsig has joined
 426 2011-03-03 03:58:13 <luke-jr> in fact, I don't even see a left-shift in SHA-2
 427 2011-03-03 03:58:38 gasteve has quit (Quit: gasteve)
 428 2011-03-03 04:00:05 mmarker has joined
 429 2011-03-03 04:00:27 <mmarker> back
 430 2011-03-03 04:00:35 <jrabbit> [Tycho]: do I need a seperate worker account for each instance of miner stuff?
 431 2011-03-03 04:00:36 jwalck has quit (Remote host closed the connection)
 432 2011-03-03 04:00:48 jwalck has joined
 433 2011-03-03 04:01:04 <mmarker> I wish I could wave a magic wand and get this code to magically be fixed...grrr
 434 2011-03-03 04:01:46 <xelister> mmarker: you can
 435 2011-03-03 04:01:50 <xelister> its called btc
 436 2011-03-03 04:01:57 <mmarker> haha
 437 2011-03-03 04:02:05 <xelister> srsly
 438 2011-03-03 04:02:20 <xelister> mmarker: what needs to be fixed where
 439 2011-03-03 04:02:21 <[Tycho]> jrabbit, no.
 440 2011-03-03 04:02:48 <mmarker> well, trying to figure out why a first call to a function is OK
 441 2011-03-03 04:02:51 <[Tycho]> jrabbit, you can use same account for all miners or create different workers, that makes no difference for me.
 442 2011-03-03 04:02:56 <mmarker> the 2nd call, and all hell breaks loose
 443 2011-03-03 04:03:16 <mmarker> hmm. "Use of uninitialized value"
 444 2011-03-03 04:03:19 <mmarker> Well now
 445 2011-03-03 04:03:22 <jrabbit> [Tycho]: oh ok
 446 2011-03-03 04:03:42 <mmarker> all in libc
 447 2011-03-03 04:03:46 <mmarker> thank you
 448 2011-03-03 04:05:59 FreeMoney has joined
 449 2011-03-03 04:06:52 <luke-jr> mmarker: try valgrind
 450 2011-03-03 04:07:05 <mmarker> That was valgrind
 451 2011-03-03 04:07:12 <luke-jr> lol
 452 2011-03-03 04:07:43 <FreeMoney> mtgox down?
 453 2011-03-03 04:08:35 <mmarker> hmm
 454 2011-03-03 04:09:35 <mmarker> jgarzik: Still here. What exactly is in hash1 in the call to the scanhash functions?
 455 2011-03-03 04:10:19 <doublec> FreeMoney: down for me
 456 2011-03-03 04:11:10 <FreeMoney> it told me "invalid bitcoin address" when I tried to withdraw about 2 hours ago, I wrote to Jed and got no reply yet
 457 2011-03-03 04:11:25 <mmarker> Oh maybe valgrind was right
 458 2011-03-03 04:13:59 <noagendamarket> mt gox and bitcoin central are both down
 459 2011-03-03 04:14:15 <xg0d> sadness =(
 460 2011-03-03 04:14:24 <xg0d> i wanted to buy some btcs lol
 461 2011-03-03 04:14:33 <tcatm> bitcoinmarket.com and btcex.com are still up
 462 2011-03-03 04:14:38 <xg0d> thx bro!
 463 2011-03-03 04:14:41 <mmarker> Ok, kids.
 464 2011-03-03 04:14:44 <mmarker> Today's lesson
 465 2011-03-03 04:14:52 <mmarker> Never use uninitialized variables
 466 2011-03-03 04:14:58 <xg0d> ah yes
 467 2011-03-03 04:15:01 <xg0d> this isnt php =P
 468 2011-03-03 04:15:28 <jrabbit> ./configure: line 4305: syntax error near unexpected token `LIBCURL_CHECK_CONFIG'
 469 2011-03-03 04:15:31 <jrabbit> Fuck
 470 2011-03-03 04:16:20 <jrabbit> configure.ac:48: error: possibly undefined macro: AC_MSG_ERROR
 471 2011-03-03 04:16:21 <jrabbit> Hm
 472 2011-03-03 04:16:41 <mmarker> Time to mine
 473 2011-03-03 04:16:46 <mmarker> jrabbit: install libcurl-dev
 474 2011-03-03 04:17:18 <mmarker> ;;bc,calcd 14800 1
 475 2011-03-03 04:17:19 <gribble> The average time to generate a block at 14800 Khps, given the supplied difficulty of 1, is 4 minutes and 50 seconds
 476 2011-03-03 04:17:20 <jrabbit> I jsut installed libcurl myself!
 477 2011-03-03 04:17:38 <jrabbit> let alone that osx coems with curl
 478 2011-03-03 04:17:48 <mmarker> Ok, so the odds say that I will not get it right this time either :D
 479 2011-03-03 04:17:55 <jrabbit> heh
 480 2011-03-03 04:18:11 <mmarker> jrabbit: may not have the aclocal macro in a place aclocal will find it
 481 2011-03-03 04:18:41 <mmarker> Ok, downside. If this works, my hash rate per core is only 3.7Mhash/s
 482 2011-03-03 04:18:44 <mmarker> which is sad
 483 2011-03-03 04:19:15 <jrabbit> :\
 484 2011-03-03 04:19:23 <luke-jr> mmarker: on what?
 485 2011-03-03 04:19:34 <mmarker> core i5
 486 2011-03-03 04:19:35 <mmarker> 760
 487 2011-03-03 04:19:41 <luke-jr> hmm
 488 2011-03-03 04:19:44 <[Tycho]> ;;bc,calc 290000
 489 2011-03-03 04:19:45 <gribble> The average time to generate a block at 290000 Khps, given current difficulty of 55590.23763914 , is 1 week, 2 days, 12 hours, 41 minutes, and 44 seconds
 490 2011-03-03 04:19:50 <[Tycho]> ;;bc,gen 290000
 491 2011-03-03 04:19:51 <gribble> The expected generation output, at 290000 Khps, given current difficulty of 55590.23763914 , is 5.24714846991 BTC per day and 0.218631186246 BTC per hour.
 492 2011-03-03 04:19:59 <luke-jr> I only get like 1 MH/s/core on my i5
 493 2011-03-03 04:20:07 <[Tycho]> ;;bc,gen 330000
 494 2011-03-03 04:20:08 <gribble> The expected generation output, at 330000 Khps, given current difficulty of 55590.23763914 , is 5.97089308645 BTC per day and 0.248787211936 BTC per hour.
 495 2011-03-03 04:20:18 <mmarker> Yea, this is the code from the SSE2 miner for windows in jgarzik's code
 496 2011-03-03 04:20:28 <mmarker> But I need to make sure it can validate some work!
 497 2011-03-03 04:20:33 <mmarker> which is a problem
 498 2011-03-03 04:20:41 <mmarker> and jgarzik missed another optimization :D
 499 2011-03-03 04:21:18 <FreeMoney> when I emailed bitcoin market about rolling my account over to the new site he said to have my password reset, but I never got the email. Now I get no response from him.
 500 2011-03-03 04:21:33 <luke-jr> FreeMoney: it's ok, the new site is totally broken
 501 2011-03-03 04:21:34 <FreeMoney> Did anyone else have get on? It looks pretty dead there
 502 2011-03-03 04:21:46 <FreeMoney> okay
 503 2011-03-03 04:21:51 <luke-jr> FreeMoney: if you put in an order to buy at .01, it purchases at 1.08 for you
 504 2011-03-03 04:22:00 <FreeMoney> I wonder why he doesn't revert back to the old one till he fixes it
 505 2011-03-03 04:22:13 <FreeMoney> that's pretty broken
 506 2011-03-03 04:22:20 <luke-jr> I wonder why he didn't test it XD
 507 2011-03-03 04:23:30 <mmarker> Wait
 508 2011-03-03 04:23:36 <mmarker> WHY AM I RUNNING ON ONE CORE
 509 2011-03-03 04:23:47 <mmarker> 4 cores for speed, people
 510 2011-03-03 04:24:22 <luke-jr> duh
 511 2011-03-03 04:24:32 <xg0d> one day i will have a bitcoin =] that will be a great day
 512 2011-03-03 04:24:39 <mmarker> Heh
 513 2011-03-03 04:24:45 <luke-jr> ;;bc,calc 14800
 514 2011-03-03 04:24:46 <gribble> The average time to generate a block at 14800 Khps, given current difficulty of 55590.23763914 , is 26 weeks, 4 days, 17 hours, 11 minutes, and 54 seconds
 515 2011-03-03 04:25:00 <mmarker> I really should hack my bitcoind to return the genesis block
 516 2011-03-03 04:25:06 <mmarker> to see if it can find it
 517 2011-03-03 04:25:15 noagendamarket has quit (Changing host)
 518 2011-03-03 04:25:15 noagendamarket has joined
 519 2011-03-03 04:25:15 <mmarker> be a good test for miners
 520 2011-03-03 04:25:16 <luke-jr> xg0d: I'll loan you one if you want.
 521 2011-03-03 04:25:24 <xg0d> wow really!
 522 2011-03-03 04:25:41 <luke-jr> xg0d: if you really don't have any…
 523 2011-03-03 04:25:53 <xg0d> i swear! i just learned about this today man
 524 2011-03-03 04:26:09 <phantomcircuit> so im trying to decide on an overall design for this client
 525 2011-03-03 04:26:14 <phantomcircuit> luke-jr, possibly you have some ideas here
 526 2011-03-03 04:26:20 <xg0d> hell, id be happy with a .01
 527 2011-03-03 04:26:25 <luke-jr> xg0d: be careful not to lose it
 528 2011-03-03 04:26:27 <mmarker> ;;bc,block
 529 2011-03-03 04:26:28 <gribble> Error: "bc,block" is not a valid command.
 530 2011-03-03 04:26:32 <luke-jr> xg0d: well, the Faucet will give you 0.05
 531 2011-03-03 04:26:36 <mmarker> ;;bc,current
 532 2011-03-03 04:26:36 <gribble> Error: "bc,current" is not a valid command.
 533 2011-03-03 04:26:40 <phantomcircuit> so far i have each peer separated out into it's own class running as it's own thread
 534 2011-03-03 04:26:40 <luke-jr> phantomcircuit: like what?
 535 2011-03-03 04:26:43 <mmarker> ;;bc,dieinafire
 536 2011-03-03 04:26:43 <gribble> Error: "bc,dieinafire" is not a valid command.
 537 2011-03-03 04:26:44 <xg0d> i tried that and i guess i have to find the block?
 538 2011-03-03 04:26:52 <phantomcircuit> (which on linux isn't expensive at all)
 539 2011-03-03 04:26:56 <luke-jr> xg0d: huh?
 540 2011-03-03 04:27:05 <xg0d> ok yea, idk what im talking about
 541 2011-03-03 04:27:12 <xg0d> i just know, i havnt got the .05 lol
 542 2011-03-03 04:27:16 <luke-jr> you put an address into the faucet?
 543 2011-03-03 04:27:20 <mmarker> phantomcircuit: depends on how many clients you're going to scale to
 544 2011-03-03 04:27:33 <xg0d> yes
 545 2011-03-03 04:27:34 <luke-jr> phantomcircuit: like, each connection to another peer?
 546 2011-03-03 04:27:41 <xg0d> just havnt got it yet
 547 2011-03-03 04:27:41 <luke-jr> xg0d: is your client done downloading all the blocks?
 548 2011-03-03 04:27:42 <phantomcircuit> luke-jr, yeah
 549 2011-03-03 04:27:48 <luke-jr> phantomcircuit: ugly
 550 2011-03-03 04:28:05 <mmarker> event based!
 551 2011-03-03 04:28:10 <phantomcircuit> luke-jr, it's actually the suggested method for writing new servers
 552 2011-03-03 04:28:10 <luke-jr> mmarker++
 553 2011-03-03 04:28:16 <xg0d> luke-jr: i am on block 105k
 554 2011-03-03 04:28:17 <luke-jr> phantomcircuit: suggested by idiots -.-
 555 2011-03-03 04:28:19 <phantomcircuit> nptl is faster than event based systems
 556 2011-03-03 04:28:21 <luke-jr> ;;bc,blocks
 557 2011-03-03 04:28:22 <mmarker> perthread is like smoking 7 gram rocks
 558 2011-03-03 04:28:22 <gribble> 111458
 559 2011-03-03 04:28:23 <phantomcircuit> significantly so
 560 2011-03-03 04:28:27 brunner has quit (Ping timeout: 264 seconds)
 561 2011-03-03 04:28:34 <luke-jr> xg0d: almost there
 562 2011-03-03 04:28:41 <luke-jr> phantomcircuit: no.
 563 2011-03-03 04:28:48 <phantomcircuit> lets see if i can find the paper
 564 2011-03-03 04:28:57 <xg0d> im so excited about this, this is the best damn idea i have ever seen.
 565 2011-03-03 04:28:58 <mmarker> Lemme find the nginx chart
 566 2011-03-03 04:29:06 <luke-jr> phantomcircuit: a standard async design doesn't have context switching overhead
 567 2011-03-03 04:29:18 <phantomcircuit> luke-jr, threading reduces the number of system calls necessary by about 50%
 568 2011-03-03 04:29:25 <luke-jr> phantomcircuit: nonsense
 569 2011-03-03 04:29:38 <luke-jr> threading requires a syscall for every thread
 570 2011-03-03 04:29:45 <luke-jr> async requires a single syscall for everything
 571 2011-03-03 04:31:52 <phantomcircuit> luke-jr, http://www.mailinator.com/tymaPaulMultithreaded.pdf
 572 2011-03-03 04:32:24 <mmarker> 2008?
 573 2011-03-03 04:32:47 <phantomcircuit> it's still accurate
 574 2011-03-03 04:32:59 MartianW has joined
 575 2011-03-03 04:33:06 <xg0d> oooohhhhh shit a lightbulb just came on. so basicly, this is similar to torrent? but the concept is applied to currency, and mining takes places because they earn a little for processing the transaction?
 576 2011-03-03 04:33:24 <phantomcircuit> mmarker, nptl was introduced into the kernel in early '08 iirc
 577 2011-03-03 04:33:40 Kiba has joined
 578 2011-03-03 04:33:45 <mmarker> Yes, but the real sexy event based systems came later
 579 2011-03-03 04:34:07 <noagendamarket> xgod  thats why we are here too
 580 2011-03-03 04:34:24 <phantomcircuit> mmarker, epoll was introduced into the kernel before nptl
 581 2011-03-03 04:34:55 <xg0d> purely brilliant
 582 2011-03-03 04:35:03 * Kiba 's posting rate is steady on the increase O_o
 583 2011-03-03 04:35:21 * Kiba is an insane posting machine o_O
 584 2011-03-03 04:36:03 <phantomcircuit> mmarker, also that paper specifically addresses epoll
 585 2011-03-03 04:36:08 <phantomcircuit> and found it to be 25% slower
 586 2011-03-03 04:37:07 <phantomcircuit> mmarker, the secret is that the event based systems require at minimum 1 extra system call
 587 2011-03-03 04:37:18 <luke-jr> phantomcircuit: it's by Google; they don't know anything
 588 2011-03-03 04:37:46 <mmarker> luke-jr: actually, there's the classic C10K paper, also by a Google employee
 589 2011-03-03 04:38:01 <mmarker> well, "paper" in the loosest sense of the word
 590 2011-03-03 04:39:06 <jgarzik> mmarker: I find http://yyz.us/bitcoin/patch.bitcoin-pow-fail useful when debugging.  It's a dumb, simple patch, but it's nice info to have.
 591 2011-03-03 04:39:52 <mmarker> Ahh. Hmm
 592 2011-03-03 04:40:17 <mmarker> Hmm... I'd probably want to start my own testnet with low difficulty for that, then.
 593 2011-03-03 04:40:18 <luke-jr> phantomcircuit: it's talking a lot about Java crap
 594 2011-03-03 04:40:26 <luke-jr> nothing is efficient in Java
 595 2011-03-03 04:40:55 <mmarker> But this is weird, yet again....no pow after 10 minutes.
 596 2011-03-03 04:41:00 <mmarker> which is just not right
 597 2011-03-03 04:41:18 <luke-jr> mmarker: I went 10 days without a POW :P
 598 2011-03-03 04:41:26 <luke-jr> back when it was supposed to take 5
 599 2011-03-03 04:41:28 <mmarker> This is at difficulty 1 on a pool
 600 2011-03-03 04:41:35 <mmarker> so, yes, it is odd
 601 2011-03-03 04:42:06 <phantomcircuit> luke-jr, the papers about java, but the actual technical problems all have nothing to do with java (except the notation that threads on linux used to suck, and that threads in the jvm on window sused to suck)
 602 2011-03-03 04:42:43 <mmagic> what the hell is up with mtgox?
 603 2011-03-03 04:42:59 <jgarzik> mmagic: moving servers
 604 2011-03-03 04:43:20 <mmagic> ah.
 605 2011-03-03 04:43:24 <jgarzik> mmarker: testing against a pool is fine, they have difficulty 1, and they pay you for testing :)
 606 2011-03-03 04:43:24 <phantomcircuit> running away with money to the bahamas
 607 2011-03-03 04:43:40 <mmarker> It's the diff 1 that I want
 608 2011-03-03 04:43:52 <mmarker> but again, statistics say I should have made a hit before now...
 609 2011-03-03 04:44:03 <mmagic> uh. jgarzik, how do you know this? (so I can go wherever it is you found that and then i won't ask lame questions in the future)
 610 2011-03-03 04:44:21 <luke-jr> phantomcircuit: except that async doesn't require more than 1 syscall per IO across all connections
 611 2011-03-03 04:44:45 <doublec> mmarker: mtgox mentioned it in #bitcoin-otc
 612 2011-03-03 04:45:17 <mmarker> Ok, the hashes are changing
 613 2011-03-03 04:45:18 <mmagic> doh. thanks doublec.
 614 2011-03-03 04:45:21 <mmarker> wait
 615 2011-03-03 04:45:23 <mmarker> God damnit
 616 2011-03-03 04:45:42 dirtyfilthy has quit (Ping timeout: 240 seconds)
 617 2011-03-03 04:45:45 mmagic has quit (Quit: Leaving)
 618 2011-03-03 04:45:55 <da2ce7> ;;bc,stats
 619 2011-03-03 04:45:56 <mmarker> Do not hash test data
 620 2011-03-03 04:45:57 <gribble> Current Blocks: 111463 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1432 blocks | Next Difficulty In About: 1 week, 2 days, 6 hours, 21 minutes, and 28 seconds | Next Difficulty Estimate: 59741.85586512
 621 2011-03-03 04:45:57 mmagic has joined
 622 2011-03-03 04:46:03 <da2ce7> ;;bc;mtgox
 623 2011-03-03 04:46:03 <gribble> Error: "bc;mtgox" is not a valid command.
 624 2011-03-03 04:46:14 <mmagic> whoah, difficulty estimate is down!
 625 2011-03-03 04:46:29 <phantomcircuit> luke-jr, huh? epoll does indeed, you make one call and are notified of what other calls you have to make
 626 2011-03-03 04:46:30 <mmagic> ;;bc,mtgox
 627 2011-03-03 04:46:30 <gribble> HTTP Error 403: Forbidden
 628 2011-03-03 04:47:25 <luke-jr> phantomcircuit: you have to make all those other calls anyway
 629 2011-03-03 04:47:39 <xg0d> luke-jr: i just got my .05 :D
 630 2011-03-03 04:47:46 <phantomcircuit> luke-jr, that's kind of my point, with threading you dont have to make the epoll call
 631 2011-03-03 04:48:10 <luke-jr> phantomcircuit: yes you do
 632 2011-03-03 04:48:15 <luke-jr> well, I use select, not epoll
 633 2011-03-03 04:48:17 <phantomcircuit> luke-jr, uh no you dont
 634 2011-03-03 04:48:25 <luke-jr> phantomcircuit: how do you handle timeouts?
 635 2011-03-03 04:48:26 * mmarker waits to get lucky
 636 2011-03-03 04:48:47 <mmarker> god, that sounds so bad.
 637 2011-03-03 04:50:07 <mmarker> hmm
 638 2011-03-03 04:50:45 <mmarker> Oh bugger
 639 2011-03-03 04:50:54 <mmarker> wrong test for H==0
 640 2011-03-03 04:51:40 <jrabbit> is it sane that a GPU would perform slower than a cpu?
 641 2011-03-03 04:52:12 <xg0d> ok it says i got the .05 as a credit, but i have noticed my address has changed. is that supposed to happen? >_<
 642 2011-03-03 04:52:14 <mmagic> jrabbit: which GPU. :)
 643 2011-03-03 04:52:30 <jrabbit>   Chipset Model:	NVIDIA GeForce 9400M
 644 2011-03-03 04:52:47 bitcoiner has joined
 645 2011-03-03 04:53:02 <mmarker> yes
 646 2011-03-03 04:53:10 <mmarker> It helps keep you anonymous
 647 2011-03-03 04:53:16 <xg0d> awesome!
 648 2011-03-03 04:53:20 <mmarker> you can still get coins at your old address.
 649 2011-03-03 04:53:31 <doublec> jrabbit: what are you getting on the 9400M?
 650 2011-03-03 04:53:32 <mmarker> oh, and BACKUP YOUR wallet.dat file!
 651 2011-03-03 04:53:39 <mmarker> lose it, you lose all your cash
 652 2011-03-03 04:53:46 <xg0d> aw man that would be a bummer
 653 2011-03-03 04:54:02 <jrabbit> 1300 khash ish
 654 2011-03-03 04:54:14 <doublec> sounds low
 655 2011-03-03 04:54:16 <mmarker> wow. That's...low
 656 2011-03-03 04:54:28 <jrabbit> yeah XD
 657 2011-03-03 04:54:30 <doublec> I get 9000 on a FX 880M
 658 2011-03-03 04:54:39 <doublec> (another mobile nvidia chip)
 659 2011-03-03 04:55:01 <jrabbit> :O
 660 2011-03-03 04:55:07 <jrabbit> holy shit
 661 2011-03-03 04:55:28 <jrabbit> I'm using poclbm
 662 2011-03-03 04:55:39 <jrabbit> on osx
 663 2011-03-03 04:55:39 <doublec> I also use poclbm
 664 2011-03-03 04:55:44 <doublec> except on linux
 665 2011-03-03 04:55:55 <jrabbit> hm
 666 2011-03-03 04:56:01 <phantomcircuit> luke-jr, rely on the OS to close the connection
 667 2011-03-03 04:56:26 <doublec> jrabbit: are you sure it's using the gpu
 668 2011-03-03 04:56:38 <doublec> jrabbit: and not the cpu via opencl?
 669 2011-03-03 04:56:40 <mmarker> Now, wait a second
 670 2011-03-03 04:56:59 <mmarker> for work to be valid at difficulty 1, does H==0, or is that not sufficient enough?
 671 2011-03-03 04:56:59 <jrabbit> omly using 30% cpu python
 672 2011-03-03 04:57:15 <jrabbit> oddly thats kind of high
 673 2011-03-03 04:57:19 <doublec> my python is using 4%
 674 2011-03-03 04:57:27 <jgarzik> mmarker: yes, H==0 is sufficient
 675 2011-03-03 04:57:28 <jrabbit> doublec: yeah its using the gpu
 676 2011-03-03 04:57:38 <xg0d> so i think it would be best to set this up on my server with a cron job to autoback up the wallet. sounds like a good time for tomorrow :D
 677 2011-03-03 04:57:39 <mmarker> Ok, then I'm still botching something:
 678 2011-03-03 04:57:40 <jrabbit> doublec: the cpu  proc ould use like 170% :P
 679 2011-03-03 04:57:46 <jgarzik> mmarker: if H==0, you will get a pool share
 680 2011-03-03 04:57:55 <mmarker> Hash: 13c4aecf c57b235b 528af9e2 42632d92
 681 2011-03-03 04:57:55 <mmarker> Hash: ce042284 e840dba6 70c5dcd2 df5e8fd7
 682 2011-03-03 04:57:55 <mmarker> Hash: 139b8104 10d721af 1eb7e3a2 2180fe6d
 683 2011-03-03 04:57:55 <mmarker> Hash: 36ae32ea fc799435 ae2d1fa3 11e6a02c
 684 2011-03-03 04:57:55 <mmarker> Hash: d35c0ab3 1da95ec8 a4409466 e6b7bcaf
 685 2011-03-03 04:57:55 <mmarker> Hash: dca89cd2 ae9a862b 4489d5f9 14fde06a
 686 2011-03-03 04:57:57 <mmarker> Hash: f1b3f64c 6466eb8f 1ed7fa96 6bf06a4a
 687 2011-03-03 04:57:59 <mmarker> Hash: c1e24eb3 106f6f5e 124f749d 0
 688 2011-03-03 04:58:01 <mmarker> j: 3
 689 2011-03-03 04:58:03 <mmarker> nonce: 0xd4a97, nonce_base: 0xd4a94
 690 2011-03-03 04:58:23 <mmarker> my 4 way hash, nonce should have been 0xd4a97...I may have the words reversed. Back to test!
 691 2011-03-03 04:59:53 prax_ has joined
 692 2011-03-03 05:00:53 prax has quit (Read error: Connection reset by peer)
 693 2011-03-03 05:01:10 <luke-jr> jrabbit: my CPU can beat that crappy GPU easy
 694 2011-03-03 05:01:22 <jrabbit> luke-jr: even the macbooks cpu beat its gpu
 695 2011-03-03 05:01:26 * nameless1 can't beat anything
 696 2011-03-03 05:01:32 <jrabbit> luke-jr: I donno why its so bad!
 697 2011-03-03 05:01:37 <nameless1> I don't even generate anymore
 698 2011-03-03 05:01:38 phantomcircuit has quit (Quit: Leaving)
 699 2011-03-03 05:01:41 <luke-jr> jrabbit: nvidia
 700 2011-03-03 05:01:42 hexidigital has quit (Read error: Connection reset by peer)
 701 2011-03-03 05:01:49 docl has quit (Read error: Connection reset by peer)
 702 2011-03-03 05:01:58 hexidigital has joined
 703 2011-03-03 05:02:03 <jrabbit> luke-jr: hahaaha
 704 2011-03-03 05:02:03 hexidigital has quit (Changing host)
 705 2011-03-03 05:02:03 hexidigital has joined
 706 2011-03-03 05:02:08 <jrabbit> luke-jr: hm
 707 2011-03-03 05:02:09 discHead has quit (Quit:)
 708 2011-03-03 05:02:09 phantomcircuit has joined
 709 2011-03-03 05:02:15 <jrabbit> well I have a horrible ATI card
 710 2011-03-03 05:02:24 <jrabbit> like ages old
 711 2011-03-03 05:02:25 <luke-jr> can't be as bad as that nvidia
 712 2011-03-03 05:02:44 docl has joined
 713 2011-03-03 05:02:51 <jrabbit> hah
 714 2011-03-03 05:03:23 <jrabbit> luke-jr: what is actually needed to run opencl stuff on a linux box?
 715 2011-03-03 05:03:29 * luke-jr ponders whether to grab this 5970 for $305
 716 2011-03-03 05:03:39 <luke-jr> jrabbit: rtfm?
 717 2011-03-03 05:03:50 <jrabbit> i mean the abstract.
 718 2011-03-03 05:07:30 <mmarker> hmm, again. the 2nd round is killing me
 719 2011-03-03 05:07:32 <mmarker> wtf
 720 2011-03-03 05:09:40 <mmarker> Grr
 721 2011-03-03 05:09:43 <mmarker> Bad cast.
 722 2011-03-03 05:14:42 redMBA has quit (Ping timeout: 276 seconds)
 723 2011-03-03 05:14:48 amiller has joined
 724 2011-03-03 05:15:43 mmarker has quit (Quit: Leaving)
 725 2011-03-03 05:16:56 FreeMoney has quit (Quit: Page closed)
 726 2011-03-03 05:25:30 bulletbill has joined
 727 2011-03-03 05:26:15 xg0d has quit (Quit: Page closed)
 728 2011-03-03 05:26:59 bulletbill has left ()
 729 2011-03-03 05:33:19 prax_ is now known as prax
 730 2011-03-03 05:57:07 doublec has quit (Quit: Leaving)
 731 2011-03-03 05:59:18 MartianW has quit (Quit: Bye all.)
 732 2011-03-03 06:08:50 bitcoiner has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.14/20110218125750])
 733 2011-03-03 06:25:59 Diablo-D3 has joined
 734 2011-03-03 06:31:01 <jgarzik> anyone with a GPU 'getwork' miner, will to give me a few minutes of GPU time to test some pool software?  All you need is a new host/port/user/pass for testing.
 735 2011-03-03 06:31:49 * Diablo-D3 's pool software is superior
 736 2011-03-03 06:32:21 <tcatm> jgarzik: sure :)
 737 2011-03-03 06:33:19 * jgarzik PMs login info
 738 2011-03-03 06:34:43 <tcatm> miner started
 739 2011-03-03 06:34:54 <Diablo-D3> you know
 740 2011-03-03 06:34:56 <tcatm> (3x 5870)
 741 2011-03-03 06:34:58 <Diablo-D3> whoever invented jsonrpc
 742 2011-03-03 06:35:01 <Diablo-D3> is an idiot
 743 2011-03-03 06:35:26 <tcatm> you don't have to use it
 744 2011-03-03 06:36:02 <tcatm> jgarzik: should I run it in pool-mode (submit every H==0)?
 745 2011-03-03 06:36:13 <Diablo-D3> tcatm: yes you kind of do
 746 2011-03-03 06:36:37 <jgarzik> tcatm: yes
 747 2011-03-03 06:36:50 hwolf has quit (Ping timeout: 240 seconds)
 748 2011-03-03 06:36:52 * tcatm restarts miner
 749 2011-03-03 06:37:04 <tcatm> PROOF OF WORK RESULT: false (booooo)
 750 2011-03-03 06:37:11 <jgarzik> tcatm: note that it is running full target, so it will reject a lot.  but pool mode should stress the server well.
 751 2011-03-03 06:38:19 <tcatm> k
 752 2011-03-03 06:39:04 <Kiba> jgarzik: how's pastecoin?
 753 2011-03-03 06:39:44 * tcatm starts to collect ideas for a bitcoin video
 754 2011-03-03 06:40:12 * Kiba hopes the bitcoin economy would pick up so he can start doing work exclusively insider the bitcoin economy
 755 2011-03-03 06:40:34 <Kiba> the problem with libertarians is that they don't adopt bitcoin
 756 2011-03-03 06:40:35 <jgarzik> tcatm: OK, you can turn it off now
 757 2011-03-03 06:40:55 <jgarzik> Kiba: got my base site going, which has a payment system.  now to hook up pastecoin to it...
 758 2011-03-03 06:41:16 <jgarzik>  1404 bitcoin   20   0  521m  65m  10m S  0.7  0.8   2:56.96 bitcoind
 759 2011-03-03 06:41:16 <jgarzik> 13745 bitcoin   20   0  148m 3764 2352 S  0.3  0.0   0:00.72 pushpoold
 760 2011-03-03 06:41:21 <jgarzik> not bad
 761 2011-03-03 06:41:42 <jgarzik> of course, a public test is many, many times the size.  but the async engine should handle that without a problem.
 762 2011-03-03 06:41:58 <tcatm> 0.72s is efficient
 763 2011-03-03 06:42:05 * jgarzik was also watching bitcoind, which had fallen over a couple times in the past
 764 2011-03-03 06:44:21 <Kiba> I see that bitcoin have 46 forks and 130 watchers
 765 2011-03-03 06:44:39 skeledrew has quit (Ping timeout: 240 seconds)
 766 2011-03-03 06:47:53 <MT`AwAy> shit
 767 2011-03-03 06:47:56 <MT`AwAy> stuff is going fasterr
 768 2011-03-03 06:48:27 <Kiba> hmm?
 769 2011-03-03 06:48:31 <[Tycho]> tcatm, are you going to create bitcoin video ?
 770 2011-03-03 06:48:46 omglolbbq has quit (Quit: Leaving.)
 771 2011-03-03 06:48:53 <Kiba> will someone ever claim the bounty's prize?
 772 2011-03-03 06:49:44 <tcatm> [Tycho]: maybe. I'm only collecting ideas for now
 773 2011-03-03 06:50:28 <MT`AwAy> Kiba, what bounty?
 774 2011-03-03 06:51:08 <[Tycho]> Sadly, they want animated video, not live action...
 775 2011-03-03 06:51:13 <Kiba> the movie bounty
 776 2011-03-03 06:51:44 <tcatm> It'll be animated if I make one.
 777 2011-03-03 06:52:09 <[Tycho]> How can you create animated one ?
 778 2011-03-03 06:52:21 <jgarzik> someone needs to film a powerpoint presentation + audio, to present a simple, basic picture of bitcoin to newbies.
 779 2011-03-03 06:52:35 bitmonster has joined
 780 2011-03-03 06:52:57 <tcatm> [Tycho]: No idea, but I'm currently installing adobe's creative suite.
 781 2011-03-03 06:52:58 TheAncientGoat has joined
 782 2011-03-03 06:53:00 <bitmonster> hi everyone
 783 2011-03-03 06:54:03 <bitmonster> does anyone know if there is a way to send sub bitcent transactions? meaning less than 0.01 bitcoins?
 784 2011-03-03 06:54:06 <Kiba> hmm, faciliating a global information economy using bitcoin
 785 2011-03-03 06:54:34 <tcatm> bitmonster: not yet (you'd have to pay fees so sending 0.01 BTC would be more efficient)
 786 2011-03-03 06:54:53 * [Tycho] just lost 0.0004 BTC due to official client bug :(
 787 2011-03-03 06:56:05 <tcatm> what's voy.com?
 788 2011-03-03 06:56:16 <Kiba> ?
 789 2011-03-03 06:56:35 <bitmonster> is there a patch for sending full precision bitcoin amounts?
 790 2011-03-03 06:56:35 <Aciid> I'm a mere figure in the pool
 791 2011-03-03 06:56:37 <Kiba> a messageboard?
 792 2011-03-03 06:56:39 <Aciid> apathy
 793 2011-03-03 06:56:51 <tcatm> 5000 hits on bitcoincharts.com with voy.com as referrer
 794 2011-03-03 06:57:19 <Aciid> tcatm: can't you see the full referer url?
 795 2011-03-03 06:57:24 <tcatm> http://voy.com/64855/
 796 2011-03-03 06:58:05 <Aciid> thats what I got from google aswell
 797 2011-03-03 06:58:16 <Kiba> I see no mention of bitcoin
 798 2011-03-03 06:58:18 <Aciid> VoyForums: The Unofficial Kitco Refugees Gold and Metals Voyagers ... - [ Käännä tämä sivu ]
 799 2011-03-03 06:58:21 <Aciid> http://bitcoincharts.com/charts/mtgoxUSD#rg180ztgSzm1g10zm2g25. [ Edit | View ]; @RIP @you might think they could only play for Tampon Bay -- mozel () ...
 800 2011-03-03 06:58:24 <Aciid> www.voy.com/64855/5/
 801 2011-03-03 06:58:38 skeledrew has joined
 802 2011-03-03 06:58:54 <Aciid> http://www.google.com/search?hl=fi&q=voy.com+%22bitcoincharts.com%22&
 803 2011-03-03 06:58:55 <Aciid> search
 804 2011-03-03 06:59:09 <[Tycho]> http://blockexplorer.com/tx/04d93bf4fad686cd60d503f798494b8d1cfbe35c5b8b414d85090fe3417606a2
 805 2011-03-03 06:59:24 <bitmonster> so is there a patch for sending full precision bitcoin amounts and if so does the network accept these transactions ?
 806 2011-03-03 06:59:38 <Kiba> Aciid: no idea what they're fricking saying
 807 2011-03-03 06:59:45 <tcatm> bitmonster: not without fees
 808 2011-03-03 07:00:16 <Aciid> Kiba: same
 809 2011-03-03 07:00:33 <Aciid> seems to be a free kusabahost
 810 2011-03-03 07:00:58 <Aciid> or some other imageboard
 811 2011-03-03 07:01:15 <nanotube> tcatm: .1001 is ok without fees.
 812 2011-03-03 07:01:24 <Aciid> yay doing 0.10% of deepbit again :3
 813 2011-03-03 07:01:28 <Aciid> I'm a somebody!
 814 2011-03-03 07:01:29 <nanotube> bitmonster: as long as they're greater than .01, network won't complain about full precision.
 815 2011-03-03 07:02:12 <bitmonster> how do i enter full precision transactions do i need to patch the client?
 816 2011-03-03 07:04:24 puddinpop has quit (Ping timeout: 260 seconds)
 817 2011-03-03 07:04:43 <Syke> This is one sick card! http://www.newegg.com/Product/Product.aspx?Item=N82E16814121388
 818 2011-03-03 07:05:23 <Aciid> I wonder how many MH/s
 819 2011-03-03 07:05:24 <Aciid> =D
 820 2011-03-03 07:06:00 <[Tycho]> Nice ont.
 821 2011-03-03 07:06:05 <[Tycho]> Better than 5970
 822 2011-03-03 07:06:06 <tcatm> 653 Mhash/s
 823 2011-03-03 07:06:47 <[Tycho]> What's the price ?
 824 2011-03-03 07:06:53 <Aciid> 999$
 825 2011-03-03 07:06:58 <MT`AwAy> anyone knows why I get "DbEnv::open: DB_RUNRECOVERY: Fatal error, run database recovery" after importing a wallet ?
 826 2011-03-03 07:07:23 <[Tycho]> 999$ is too much...
 827 2011-03-03 07:07:33 <Aciid> it's a collectors edition
 828 2011-03-03 07:07:35 <Aciid> serial numbered
 829 2011-03-03 07:09:10 <[Tycho]> Isn't it way too new to be collected ?
 830 2011-03-03 07:09:34 <[Tycho]> VooDoo 5500, 6600 indeed are collector's items :)
 831 2011-03-03 07:09:48 <nanotube> bitmonster: yes currently you'd need a patch.
 832 2011-03-03 07:09:55 <tcatm> that's marketing ;)
 833 2011-03-03 07:10:50 <bitmonster> where can i get this patch?
 834 2011-03-03 07:11:48 <[Tycho]> bitmonster, can you compile bitcoin ?
 835 2011-03-03 07:12:12 <bitmonster> no :( i cant
 836 2011-03-03 07:12:33 <[Tycho]> Then you don't need the patch.
 837 2011-03-03 07:12:54 <nanotube> heh good point [Tycho]
 838 2011-03-03 07:13:59 <bitmonster> how do i enter a full precision transation then? can i do it manually without gui?
 839 2011-03-03 07:14:31 <nanotube> bitmonster: nope
 840 2011-03-03 07:14:39 <nanotube> just wait until .21 is released
 841 2011-03-03 07:14:55 <Diablo-D3> http://nerfnow.com/
 842 2011-03-03 07:15:00 <Diablo-D3> the visual pun, it burns
 843 2011-03-03 07:15:13 <bitmonster> .21 supports full precision then?
 844 2011-03-03 07:15:36 <nanotube> bitmonster: my understanding is that that is planned for .21 release.
 845 2011-03-03 07:15:40 <[Tycho]> Diablo-D3, are you from Russia ?
 846 2011-03-03 07:17:02 Kiba has quit (Ping timeout: 276 seconds)
 847 2011-03-03 07:17:21 <bitmonster> me wait for .21 then... me sad...
 848 2011-03-03 07:19:00 bitmonster has quit (Quit: Page closed)
 849 2011-03-03 07:19:23 <JFK911> preved
 850 2011-03-03 07:19:56 <Diablo-D3> [Tycho]: no?
 851 2011-03-03 07:19:59 <cosurgi> http://www.taters.net/cgi-bin/btc/matrix.pl?axisinc=0.02
 852 2011-03-03 07:20:05 <cosurgi> internal server error :(
 853 2011-03-03 07:31:11 Lachesis has quit (Ping timeout: 240 seconds)
 854 2011-03-03 07:32:02 gasteve has joined
 855 2011-03-03 07:32:04 gasteve has quit (Changing host)
 856 2011-03-03 07:32:04 gasteve has joined
 857 2011-03-03 07:34:23 <[Tycho]> Wow, 35 Gh/s peak...
 858 2011-03-03 07:50:48 dissipate has joined
 859 2011-03-03 07:50:48 dissipate has quit (Changing host)
 860 2011-03-03 07:50:48 dissipate has joined
 861 2011-03-03 08:08:56 MeanderingCode has joined
 862 2011-03-03 08:10:25 MeanderingCode has left ()
 863 2011-03-03 08:16:33 gasteve has quit (Quit: gasteve)
 864 2011-03-03 08:16:59 larsig has quit (Ping timeout: 250 seconds)
 865 2011-03-03 08:18:31 <Diablo-D3> lol
 866 2011-03-03 08:18:35 <Diablo-D3> I made cdecker's day
 867 2011-03-03 08:18:56 <dissipate> o rly?
 868 2011-03-03 08:19:03 <Diablo-D3> yeah, I merged his patch
 869 2011-03-03 08:19:15 <dissipate> cool
 870 2011-03-03 08:21:33 devon_hillard has joined
 871 2011-03-03 08:25:27 Zarutian has joined
 872 2011-03-03 08:29:54 larsig has joined
 873 2011-03-03 08:31:34 <nathan7> la la la
 874 2011-03-03 08:32:32 larsig has quit (Read error: Connection reset by peer)
 875 2011-03-03 08:33:04 <devon_hillard> ;;bc,calc 250000
 876 2011-03-03 08:33:05 <gribble> The average time to generate a block at 250000 Khps, given current difficulty of 55590.23763914 , is 1 week, 4 days, 1 hour, 17 minutes, and 13 seconds
 877 2011-03-03 08:33:58 <devon_hillard> ;;bc,calc 70000
 878 2011-03-03 08:33:59 <gribble> The average time to generate a block at 70000 Khps, given current difficulty of 55590.23763914 , is 5 weeks, 4 days, 11 hours, 27 minutes, and 12 seconds
 879 2011-03-03 08:36:12 <devon_hillard> ;;bc,rate 65000
 880 2011-03-03 08:36:12 <gribble> Error: "bc,rate" is not a valid command.
 881 2011-03-03 08:36:45 bitterJ has joined
 882 2011-03-03 08:43:52 mmagic has quit (Quit: Leaving)
 883 2011-03-03 08:44:17 mmagic has joined
 884 2011-03-03 08:46:37 <legion050> ;;bc,calc 1400000
 885 2011-03-03 08:46:38 <gribble> The average time to generate a block at 1400000 Khps, given current difficulty of 55590.23763914 , is 1 day, 23 hours, 22 minutes, and 21 seconds
 886 2011-03-03 08:46:46 <bitterJ> hello
 887 2011-03-03 08:48:32 larsig has joined
 888 2011-03-03 08:49:39 <bitterJ> I have a tech question. I got 0.05 BTC from bitcoinfaucet 10 hours ago and it still has 0 confirmations. is this normal? (btw, my bitcoin client has only 8 connections as my ISP won't give me a public IP addrees, and apparently blocks port 8333)
 889 2011-03-03 08:51:01 larsig has quit (Read error: Connection reset by peer)
 890 2011-03-03 08:52:23 altamic has joined
 891 2011-03-03 08:52:23 altamic has quit (Changing host)
 892 2011-03-03 08:52:23 altamic has joined
 893 2011-03-03 08:52:37 <Aciid> bitterJ: start generating.
 894 2011-03-03 08:52:54 <Aciid> ++download all blocks
 895 2011-03-03 08:54:46 <[Noodles]> what exactly has generating todo with receiving?
 896 2011-03-03 08:57:18 hwolf has joined
 897 2011-03-03 09:02:10 <dissipate> [Noodles], you can't do anything until you have downloaded the entire block chain.
 898 2011-03-03 09:02:41 <[Noodles]> that's true, but generating or not doesnt matter
 899 2011-03-03 09:03:15 <[Noodles]> besides that, bitterJ said, his tx has 0conf, means he can already see it, so i guess he has the blocks
 900 2011-03-03 09:04:14 <Aciid> I thought "generating" connects to the IRC
 901 2011-03-03 09:04:27 <Aciid> or does it connect on startup?
 902 2011-03-03 09:04:32 <dissipate> [Noodles], his block chain could be corrupted
 903 2011-03-03 09:04:34 <[Noodles]> nope, "generating" just hashes ^_^
 904 2011-03-03 09:04:56 <[Noodles]> and you can shutoff irc completely
 905 2011-03-03 09:05:22 <[Noodles]> dissipate: could be, yet what has "generating" todo with it?
 906 2011-03-03 09:06:16 <dissipate> [Noodles], nothing
 907 2011-03-03 09:06:54 sshc has quit (Ping timeout: 276 seconds)
 908 2011-03-03 09:08:06 larsig has joined
 909 2011-03-03 09:08:13 dissipate has quit (Quit: Leaving)
 910 2011-03-03 09:09:29 TheKid has quit (Read error: Connection reset by peer)
 911 2011-03-03 09:11:13 Syke has quit (Read error: Operation timed out)
 912 2011-03-03 09:19:14 jackmcbarn has quit (Read error: Operation timed out)
 913 2011-03-03 09:19:30 jackmcbarn has joined
 914 2011-03-03 09:27:50 * MT`AwAy hiring part time bitcoin public relations representative, contact for details
 915 2011-03-03 09:31:40 Spenvo has joined
 916 2011-03-03 09:35:10 <Aciid> 5970 not available in finland, gotta go with 5870
 917 2011-03-03 09:37:09 <Spenvo> Hmm.  Mtgox.com is down for me.  Anyone else?
 918 2011-03-03 09:37:23 <MT`AwAy> Spenvo, http://twitter.com/MtGox/status/43238072393023488
 919 2011-03-03 09:37:40 <MT`AwAy> or other option: http://www.bitcoin.org/smf/index.php?topic=4073.0
 920 2011-03-03 09:37:42 <bitbot> '403 forbidden' at Mt.Gox Connection refused.
 921 2011-03-03 09:38:09 <Spenvo> ty
 922 2011-03-03 09:41:53 <Spenvo> Does anyone know of a good independent Bitcoin blog/specialty site?
 923 2011-03-03 09:45:59 <nevezen> what does that even mean?
 924 2011-03-03 09:46:31 <noagendamarket> Spenvo what do you mean ?
 925 2011-03-03 09:48:30 <Spenvo> Well, I'm thinking of starting a Bitcoin blog, and I was wondering what, outside of the wikis and official sites, were your favorite current blogs, etc.
 926 2011-03-03 09:49:27 <Spenvo> I've found some interesting stuff, including a Bitcoin weekly blog that's just started
 927 2011-03-03 09:49:34 <Spenvo> just wondering what you guys liked
 928 2011-03-03 09:49:38 <noagendamarket> http://www.bitcoinnews.com/    is sgornicks
 929 2011-03-03 09:50:24 <Spenvo> Sweet, this looks really cool. Great site, thanks
 930 2011-03-03 09:50:40 <noagendamarket> http://bitcoinmedia.posterous.com/   you could post there as a contributor
 931 2011-03-03 09:50:49 <noagendamarket> its a group blog
 932 2011-03-03 09:51:43 <OneFixt> sgornick: BitPenny mining pool
 933 2011-03-03 09:51:44 <OneFixt> Status: Reopened, accepting new registrations.
 934 2011-03-03 09:51:44 <OneFixt> Technically, there's no registration required =)
 935 2011-03-03 09:52:00 <noagendamarket> It regularly gets over 200 views per article
 936 2011-03-03 09:52:39 <Spenvo> :) , thanks again.  I will definitely join up
 937 2011-03-03 10:05:36 BlueMatt has joined
 938 2011-03-03 10:08:28 altamic has quit (Ping timeout: 240 seconds)
 939 2011-03-03 10:13:13 slush has joined
 940 2011-03-03 10:14:32 <devon_hillard> does anyone have an AC unit hooked up to their computer cooling kit?
 941 2011-03-03 10:14:57 altamic has joined
 942 2011-03-03 10:15:36 MartianW has joined
 943 2011-03-03 10:16:01 MartianW has quit (Client Quit)
 944 2011-03-03 10:16:08 <devon_hillard> the best way I can think of it is a watercooling system dumping heat in a chilled water tank
 945 2011-03-03 10:18:55 Phoebus has joined
 946 2011-03-03 10:20:16 rli has joined
 947 2011-03-03 10:21:05 aMR-[u] has quit (Ping timeout: 272 seconds)
 948 2011-03-03 10:25:37 Phoebus has quit (Read error: Connection reset by peer)
 949 2011-03-03 10:31:47 matt__ has joined
 950 2011-03-03 10:39:40 <devon_hillard> or just build a phase-change cooling system with a lot of insulation
 951 2011-03-03 10:41:25 matt__ has quit (Quit: Page closed)
 952 2011-03-03 10:43:53 MartianW has joined
 953 2011-03-03 10:50:18 <slush> ;;bc,stats
 954 2011-03-03 10:50:19 <gribble> Current Blocks: 111519 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1376 blocks | Next Difficulty In About: 1 week, 1 day, 15 hours, 55 minutes, and 44 seconds | Next Difficulty Estimate: 61368.88990497
 955 2011-03-03 10:50:59 <UukGoblin> ;;b,calc 1685000
 956 2011-03-03 10:51:00 <gribble> Error: "b,calc" is not a valid command.
 957 2011-03-03 10:51:03 <UukGoblin> ;;bc,calc 1685000
 958 2011-03-03 10:51:04 <gribble> The average time to generate a block at 1685000 Khps, given current difficulty of 55590.23763914 , is 1 day, 15 hours, 21 minutes, and 36 seconds
 959 2011-03-03 10:53:22 BlueMatt has quit (Ping timeout: 272 seconds)
 960 2011-03-03 10:59:42 <slush> sipa: can you add all graphs  to the homepage? Or links to the images?
 961 2011-03-03 11:00:01 <slush> I always forgot the correct urls :)
 962 2011-03-03 11:01:36 <slush> wow, my pool has half a milion page views in last month
 963 2011-03-03 11:03:15 MartianW has quit (Quit: Bye all.)
 964 2011-03-03 11:07:15 <hwolf> ;;bc,calc 7000000
 965 2011-03-03 11:07:16 <gribble> The average time to generate a block at 7000000 Khps, given current difficulty of 55590.23763914 , is 9 hours, 28 minutes, and 28 seconds
 966 2011-03-03 11:07:43 cosurgi has quit (Ping timeout: 250 seconds)
 967 2011-03-03 11:09:40 cosurgi has joined
 968 2011-03-03 11:12:59 <mizerydearia> http://sandbox.witcoin.com/ costs only 0.00001 to post, edit, reply, upvote
 969 2011-03-03 11:13:28 <bitterJ> fyi, my bitcoins got confirmed after a few hours. 24 confirmations now.
 970 2011-03-03 11:13:32 <bitterJ> thanks fo rthe advice
 971 2011-03-03 11:16:17 <Aciid> patience
 972 2011-03-03 11:16:29 <Aciid> Y U NO HAS
 973 2011-03-03 11:22:22 devon_hillard has quit (Quit: Leaving)
 974 2011-03-03 11:22:35 <andrewh> sipa: all the links on http://bitcoin.sipa.be/ (for the pics) link to the same image
 975 2011-03-03 11:22:52 BCBot has quit (Ping timeout: 264 seconds)
 976 2011-03-03 11:23:03 <sipa> andrewh: not anymore
 977 2011-03-03 11:23:12 <andrewh> anyways, http://bitcoin.sipa.be/growth-small.png -- wow, bitcoin's managed to keep a 1%/day rate
 978 2011-03-03 11:23:30 <andrewh> sipa: yay
 979 2011-03-03 11:23:47 <andrewh> looks you added another graph too
 980 2011-03-03 11:23:56 <andrewh> or two
 981 2011-03-03 11:24:02 <sipa> 3
 982 2011-03-03 11:24:07 <andrewh> or three
 983 2011-03-03 11:24:08 <andrewh> :p
 984 2011-03-03 11:24:20 <sipa> i'll try to add a zoomable one in javascript too one of these days
 985 2011-03-03 11:24:56 <andrewh> want me to help? :P
 986 2011-03-03 11:26:00 MattJD has joined
 987 2011-03-03 11:26:04 MJD has quit (Ping timeout: 264 seconds)
 988 2011-03-03 11:26:28 <sipa> well, if you can create a html page and tell me in what format it needs the data, i'll add it :)
 989 2011-03-03 11:29:14 altamic has quit (Ping timeout: 260 seconds)
 990 2011-03-03 11:30:17 <mizerydearia> It was suggested that even offering 0.00000001 bitcoins as a fee would be sufficient considering many clients still do not include fee in their transactions, however it was also suggested that the mainline bitcoin client will round 0.00000001 to 0.01.  Is this true for gavin's mainline client?  Does it round fee of 0.00000001 to 0.01?
 991 2011-03-03 11:32:31 grondilu has joined
 992 2011-03-03 11:33:02 altamic has joined
 993 2011-03-03 11:33:22 <grondilu> I have a bitcoind sendfrom command that seems to be hanging.  Should I worry?
 994 2011-03-03 11:33:49 <grondilu> I'm running bitcoin 0.3.19 and it's the first time this happens
 995 2011-03-03 11:35:48 xelister has quit (Remote host closed the connection)
 996 2011-03-03 11:35:50 <grondilu> oh ok I guess it is because I had a blender process that was eating all CPU
 997 2011-03-03 11:36:22 <grondilu> I killed the blender process but now the bitcoind command still hangs :(
 998 2011-03-03 11:36:38 <grondilu> can I safely Ctrl-C it?
 999 2011-03-03 11:37:24 RazielZ has joined
1000 2011-03-03 11:38:00 altamic has quit (Ping timeout: 260 seconds)
1001 2011-03-03 11:39:58 <RBecker> ;;bc,blocks
1002 2011-03-03 11:39:58 <gribble> 111530
1003 2011-03-03 11:41:45 Zarutian has quit (Quit: Zarutian)
1004 2011-03-03 11:51:00 altamic has joined
1005 2011-03-03 11:51:01 altamic has quit (Changing host)
1006 2011-03-03 11:51:01 altamic has joined
1007 2011-03-03 11:51:50 Spenvo has quit (Quit: Page closed)
1008 2011-03-03 11:55:19 darsk1ez has quit (Read error: Operation timed out)
1009 2011-03-03 11:57:24 xelister has joined
1010 2011-03-03 11:58:23 darsk1ez has joined
1011 2011-03-03 12:00:59 <molecular> what's with mtgox? I get 403?
1012 2011-03-03 12:01:17 <xelister> Which is better for normal GPU fan on 5970 (normally it is limited to max ~40% of max speed). I assume the fan may die on GPU eventually:  fan-speed,gpu-temperature:  70%,75C or 75%,66C or 85%,60C ?
1013 2011-03-03 12:02:08 <xelister> mtgox died?
1014 2011-03-03 12:02:12 <xelister> molecular: me too
1015 2011-03-03 12:02:52 <molecular> last trade was 40 minutes ago according to #bitcoin-market
1016 2011-03-03 12:03:05 <molecular> heh, it might even still do the trades ;)
1017 2011-03-03 12:03:14 <xelister>   mtgox announced on #bitcoin-otc: "We are moving the mtgox server to a new host. Should be back up soon"
1018 2011-03-03 12:03:23 <molecular> ah, allright, thanks for the info
1019 2011-03-03 12:03:31 <grondilu> ;;sell 100 EUR @ 0.7 EUR/BTC cash remis un samedi sur Paris dans un lieu public avec accès internet ouvert.  Il n'y aura pas de rencontre physique.  Le cash sera déposé à un endroit précis qui sera dévoilé par IRC après réception paiement.
1020 2011-03-03 12:03:32 <gribble> Order id 724 created.
1021 2011-03-03 12:03:33 BlueMatt has joined
1022 2011-03-03 12:04:02 <molecular> really? you can place orders using gribble?
1023 2011-03-03 12:04:13 <grondilu> sur you can
1024 2011-03-03 12:04:59 <grondilu> but I should have done it on #bitcoin-otc.  My bad
1025 2011-03-03 12:05:20 <molecular> c'est fou
1026 2011-03-03 12:05:46 <grondilu> ah t'es français?
1027 2011-03-03 12:07:01 <molecular> bitcoin-central is down, too
1028 2011-03-03 12:07:13 <molecular> grondilu, no, but I learned it for 3 years in school, after latin ;|
1029 2011-03-03 12:07:22 <molecular> should've done it the other way around ;)
1030 2011-03-03 12:07:51 <grondilu> well, I guess for an english native speaker, latin is about as useful as french.
1031 2011-03-03 12:07:59 <molecular> I'm german
1032 2011-03-03 12:08:14 <molecular> latin is still quite usefull, also for learning french ;)
1033 2011-03-03 12:08:20 <grondilu> indeed
1034 2011-03-03 12:08:47 purpleposeidon has quit (Ping timeout: 276 seconds)
1035 2011-03-03 12:11:14 <mizerydearia> Which page provides optimal documentation on how to use txfee?
1036 2011-03-03 12:16:09 <bk128> mizerydearia: this page has a bit of info https://en.bitcoin.it/wiki/Transaction_fee
1037 2011-03-03 12:16:20 grondilu has quit (Quit: leaving)
1038 2011-03-03 12:16:33 <mizerydearia> bk128, I see no configuration or switch info there
1039 2011-03-03 12:16:51 <mizerydearia> I understand it, but it doesn't explain so well how to use it
1040 2011-03-03 12:17:21 <bk128> yeah, I'm not too sure either.  I use the gui bitcoin client.  sorry
1041 2011-03-03 12:20:44 <bk128> mizerydearia: there are some people in #bitcoin-otc right now.  someone there might know
1042 2011-03-03 12:27:23 <molecular> does anyone here buy a vps in bitcoin?
1043 2011-03-03 12:28:43 <molecular> there's quite some offers linked from bitcoin.org trade page
1044 2011-03-03 12:36:18 genjix has joined
1045 2011-03-03 12:36:59 BlueMatt_ has joined
1046 2011-03-03 12:37:05 BlueMatt_ has quit (Changing host)
1047 2011-03-03 12:37:05 BlueMatt_ has joined
1048 2011-03-03 12:37:05 BlueMatt_ has quit (Changing host)
1049 2011-03-03 12:37:05 BlueMatt_ has joined
1050 2011-03-03 12:37:56 <BlueMatt_> what is an average number of ips in a client's cache
1051 2011-03-03 12:39:10 altamic has quit (Ping timeout: 272 seconds)
1052 2011-03-03 12:40:24 <mizerydearia> I found one floating point issue.  And it comes directly from bitcoin client.  I sent 0.02 bitcoins and in PHP it is detected as 0.020000000000000000416
1053 2011-03-03 12:40:45 <mizerydearia> That is related to 20 decimal places configured in my php.ini
1054 2011-03-03 12:41:07 <mizerydearia> precision = 20
1055 2011-03-03 12:42:14 BlueMatt has quit (Quit: Leaving)
1056 2011-03-03 12:42:35 noagendamarket has quit (Ping timeout: 276 seconds)
1057 2011-03-03 12:42:51 BlueMatt_ is now known as BlueMatt
1058 2011-03-03 12:42:59 <mizerydearia> One workaround for now is changing $btcconn->getreceivedbyaddress($bitcoin_address_witcoin, 0); to number_format($btcconn->getreceivedbyaddress($bitcoin_address_witcoin, 0), 16, ".", "");
1059 2011-03-03 12:44:50 <BlueMatt> then that is a php bug, not a bug in bitcoin?
1060 2011-03-03 12:45:57 <mizerydearia> iirc, the data returned by JSON is float data type.  PHP evaluates float data up to precision in php.ini
1061 2011-03-03 12:46:17 <mizerydearia> Previously it was suggested that for monetary or currency data that the data should be returned as a string.
1062 2011-03-03 12:46:22 <mizerydearia> There were arguments on both sides, iirc
1063 2011-03-03 12:46:26 <sipa> yes
1064 2011-03-03 12:47:31 <Diablo-D3> mizerydearia: nope
1065 2011-03-03 12:47:40 <Diablo-D3> the data returned by json is fixed point
1066 2011-03-03 12:48:17 <mizerydearia> Diablo-D3, Do you have a suggestion on proper way to evaluate the JSON data from PHP environment?
1067 2011-03-03 12:48:26 <Diablo-D3> hrm.
1068 2011-03-03 12:48:36 <Diablo-D3> depends how flexible php's json parser is
1069 2011-03-03 12:48:51 <sipa> multiply by 10^8, round to nearest integer, divide again
1070 2011-03-03 12:49:12 <Diablo-D3> you should be able to just do like $tree->node_for_the_number->toString() or whatever the fuck php does
1071 2011-03-03 12:49:14 <Diablo-D3> sipa: no
1072 2011-03-03 12:49:20 <Diablo-D3> you _cannot_ convert it to float first
1073 2011-03-03 12:49:29 <sipa> i agree you shouldn't
1074 2011-03-03 12:49:38 <Diablo-D3> you really want to process it as the original string, then regex the . out and parse as an int.
1075 2011-03-03 12:49:54 <Diablo-D3> sipa: if this was real money, it'd be illegal. as in, against the law.
1076 2011-03-03 12:49:58 <Diablo-D3> the kind you get fined for.
1077 2011-03-03 12:49:59 <sipa> the rpc call should just return an integer
1078 2011-03-03 12:50:09 bitterJ has quit (Quit: Yummy, like ircing on a cake! [ http://www.bersirc.org/ - Open Source IRC ])
1079 2011-03-03 12:50:10 altamic has joined
1080 2011-03-03 12:50:12 <Diablo-D3> no, the rpc call is correct. both ends should be sane.
1081 2011-03-03 12:50:31 <Diablo-D3> now, it COULD return both human readable and ubtc values
1082 2011-03-03 12:50:39 <Diablo-D3> which would be fine for people stuck on broken languages
1083 2011-03-03 12:52:15 <mizerydearia> For example, sending 0.02 bitcoins JSON returns the following data as observed from PHP -> string(42) "{"result":0.02000000,"error":null,"id":1} "
1084 2011-03-03 12:52:17 gasteve has joined
1085 2011-03-03 12:52:20 <molecular> dns still stale, mtgox is back up, though: 174.121.74.59
1086 2011-03-03 12:52:59 <Diablo-D3> mizerydearia: ....
1087 2011-03-03 12:53:03 <Diablo-D3> goddamnit satoshi
1088 2011-03-03 12:53:03 <mizerydearia> There is no indication from JSON what data type it is and ultimately PHP evaluates it as a float
1089 2011-03-03 12:53:14 * Diablo-D3 punches out all his blood
1090 2011-03-03 12:53:24 <Diablo-D3> it should be a fucking string.
1091 2011-03-03 12:59:50 <BlueMatt> does setConnected.count(addr.ip & 0x0000ffff) not protect against the attack which the the (addr.port != GetDefaultPort()) 2h delay is designed to prevent?
1092 2011-03-03 13:09:51 puddinpop has joined
1093 2011-03-03 13:14:14 altamic has quit (Ping timeout: 240 seconds)
1094 2011-03-03 13:14:15 grondilu has joined
1095 2011-03-03 13:14:16 d-snp has joined
1096 2011-03-03 13:14:20 <d-snp> hi :)
1097 2011-03-03 13:14:44 grondilu has quit (Client Quit)
1098 2011-03-03 13:15:43 <d-snp> is there an article somewhere that explains how bitcoin works as a transaction network?
1099 2011-03-03 13:15:56 <sipa> the bitcoin whitepaper, by satoshi?
1100 2011-03-03 13:16:19 <d-snp> alright
1101 2011-03-03 13:18:37 keystroke has joined
1102 2011-03-03 13:19:00 <Diablo-D3> BlueMatt: whats that do?
1103 2011-03-03 13:19:15 <Diablo-D3> BlueMatt: because you dont want to accidently merge nodes on different ports
1104 2011-03-03 13:19:24 keystrike has quit (Read error: Connection reset by peer)
1105 2011-03-03 13:19:26 Netsniper has quit (Read error: Connection reset by peer)
1106 2011-03-03 13:19:27  has joined
1107 2011-03-03 13:19:27 <Diablo-D3> it could be a security NAT (and yes, they do infact exist on ipv6)
1108 2011-03-03 13:19:56 gasteve has quit (Quit: gasteve)
1109 2011-03-03 13:20:02 <bk128> d-snp: http://www.bitcoin.org/sites/default/files/bitcoin.pdf
1110 2011-03-03 13:20:14 <BlueMatt> Diablo-D3: it is designed to prevent a flood attack where I run a bunch of nodes and become the only person you connect to
1111 2011-03-03 13:20:17 <d-snp> hmm that paper is easy to read
1112 2011-03-03 13:20:19 <d-snp> very nice
1113 2011-03-03 13:21:13 <d-snp> so, what property makes sure that two nodes aren't generating the same bitcoin?
1114 2011-03-03 13:21:14 AmpEater has joined
1115 2011-03-03 13:21:25 <BlueMatt> But AFAICT, the 2h penalty for nodes which use an alternate port isn't necessary, as it already only connects to one ip per unique a.b.X.X
1116 2011-03-03 13:21:35 gasteve has joined
1117 2011-03-03 13:21:42 <BlueMatt> Thus the penalty doesn't help anything?
1118 2011-03-03 13:21:57 <sipa> d-snp: nothing, but it will cause a split in the block chain
1119 2011-03-03 13:22:09 <sipa> and eventually one of the sides will win
1120 2011-03-03 13:22:27 <bk128> d-snp: this is a decent article too http://en.wikipedia.org/wiki/Bitcoin
1121 2011-03-03 13:22:46 <d-snp> oh you sort of compete over the validity of both bitcoins? and people would invest in either one? :P
1122 2011-03-03 13:23:00 <d-snp> rendering the other branch useless?
1123 2011-03-03 13:23:13 <d-snp> or would it never be useless?
1124 2011-03-03 13:23:22 <d-snp> ehh worthless I mean
1125 2011-03-03 13:23:25 <sipa> the other branch is discarded in that case
1126 2011-03-03 13:23:34 <sipa> so that's a significant loss for the one who found the block
1127 2011-03-03 13:23:42 <d-snp> right
1128 2011-03-03 13:23:49 <d-snp> did this happen in the past?
1129 2011-03-03 13:23:53 <sipa> all the time
1130 2011-03-03 13:23:58 <sipa> very small splits
1131 2011-03-03 13:24:20 <BlueMatt> one, maybe two blocks is about all that ever happens
1132 2011-03-03 13:24:28 <sipa> but the whole network must agree on one single sequence of transactions to be able to verify things
1133 2011-03-03 13:24:34 altamic has joined
1134 2011-03-03 13:24:42 <BlueMatt> hence the chain with the largest total difficulty wins
1135 2011-03-03 13:25:09 <d-snp> if I understand correctly anyone who generates a bitcoin adds it to the block chain right? so if you're unlucky you may append your perfectly unique bitcoin to the wrong branch of a split?
1136 2011-03-03 13:25:41 <sipa> but by doing so, you increase the difficulty of that branch
1137 2011-03-03 13:25:56 <sipa> which will make it with 95% chance or so the winning one
1138 2011-03-03 13:26:47 <d-snp> alright that seems fair enough, I'll continue reading :P
1139 2011-03-03 13:27:01 <sipa> and you don't generate a bitcoin - you create a block, but doing so allows you to create a generation transaction that gives you 50 BTC
1140 2011-03-03 13:27:08 <sipa> but that's more a side effect than the purpose
1141 2011-03-03 13:28:07 <d-snp> because creating the block is the hard work right?
1142 2011-03-03 13:28:12 <sipa> yes
1143 2011-03-03 13:28:20 <sipa> ;;bc,diff
1144 2011-03-03 13:28:21 <gribble> 55590.23763914
1145 2011-03-03 13:28:49 <sipa> it currently requires on average 238761895854380 attempts to create one
1146 2011-03-03 13:29:17 <d-snp> they can only be bruteforced or are there heuristics?
1147 2011-03-03 13:29:46 <sipa> it's a bruteforce looking for a nonce that hashes to a low enough number
1148 2011-03-03 13:29:48 BlueMatt has quit (Ping timeout: 272 seconds)
1149 2011-03-03 13:30:23 <sipa> there are some optimizations, so that it only requires +- 120 sha256 rounds instead of the 192
1150 2011-03-03 13:30:49 eLX-[u] has joined
1151 2011-03-03 13:31:00 legion050 has left ()
1152 2011-03-03 13:31:05 <d-snp> right, so I'm a bit confused abit chapter 5, it says all nodes get all new transactions, and they all make blocks out of them, do the blocks all need to be about the same transactions?
1153 2011-03-03 13:31:18 <d-snp> or is it just some set of transactions that happened recently?
1154 2011-03-03 13:31:35 <sipa> when you create a tx, you broadcast i
1155 2011-03-03 13:31:37 <sipa> it
1156 2011-03-03 13:32:12 <sipa> miners cache that tx, and if they like it (purely their own discretion, but the default implementation is pretty lax) they will put it in their next block
1157 2011-03-03 13:32:23 <sipa> unless of course someone else puts it in a block before they do
1158 2011-03-03 13:32:34 <x6763> new blocks only include transactions that are not in previous blocks
1159 2011-03-03 13:32:51 <sipa> yes
1160 2011-03-03 13:34:49 <mizerydearia> Is anyone around that is part of an organization or charity that accepts bitcoin donations?
1161 2011-03-03 13:34:56 AmpEater has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
1162 2011-03-03 13:37:10 <bk128> eff does
1163 2011-03-03 13:37:28 <mizerydearia> Is anyone representative of eff here?
1164 2011-03-03 13:37:51 <bk128> not that I know of
1165 2011-03-03 13:38:07 <mizerydearia> I'm looking for initial volunteers for something
1166 2011-03-03 13:38:12 TD has joined
1167 2011-03-03 13:39:45 <d-snp> so you have to claim a transaction as yours?
1168 2011-03-03 13:39:58 <sipa> how do you mean?
1169 2011-03-03 13:40:03 <xelister> I CLAIM THIS TRANSATION
1170 2011-03-03 13:40:06 * xelister sticks a flag
1171 2011-03-03 13:40:16 <d-snp> well you say that if someone else has already put it into their block, you can't also put it into yours
1172 2011-03-03 13:40:29 <sipa> well yes
1173 2011-03-03 13:40:53 <sipa> but typically each block just has all recent transactions that were not yet put in another block
1174 2011-03-03 13:40:59 * xelister AND DIABLO's SISTER
1175 2011-03-03 13:41:00 * xelister sticks a flag
1176 2011-03-03 13:41:07 <d-snp> and since the transaction is broadcasted over the whole network the chance is rather large that someone elese is working with the same transactions?
1177 2011-03-03 13:41:11 * mizerydearia feels sorry for the flag
1178 2011-03-03 13:41:13 <sipa> d-snp: no
1179 2011-03-03 13:41:29 <sipa> since you also have to mine for your block to be built upon a particular earlier block
1180 2011-03-03 13:41:36 <xelister> mizerydearia: what volunteers?
1181 2011-03-03 13:41:53 <sipa> so either you don't know there is another block made, and you keep mining on the previous, causing a chain split if you find one
1182 2011-03-03 13:41:57 <mizerydearia> xelister, volunteers that would like to participate in a site to establish additional donations to them
1183 2011-03-03 13:42:19 <sipa> or you do know that a new block was found, build further on that one, and don't include the transactions that were put in that one either
1184 2011-03-03 13:42:55 <sipa> d-snp: or rather, yes, everyone is working with the same transactions, mostly, but the first one that finds a block with them in, gets them
1185 2011-03-03 13:43:01 <sipa> (or rather, their fees)
1186 2011-03-03 13:44:30 altamic has quit (Ping timeout: 246 seconds)
1187 2011-03-03 13:45:34 <xelister> oh, update. http://bitcoincharts.com/markets/
1188 2011-03-03 13:46:15 <mizerydearia> xelister, nice chartage!
1189 2011-03-03 13:46:38 <mizerydearia> I suggest the bitcoin charts logo link to the home page
1190 2011-03-03 13:46:51 <xelister> mizerydearia: it's not mine page :)
1191 2011-03-03 13:46:55 <d-snp> so sipa you could say you're always in a race and because it's a pseudorandom bruteforce the distribution of blocks you actually win corresponds with your computing power
1192 2011-03-03 13:47:00 <mizerydearia> xelister, I didn't say it was ^_^
1193 2011-03-03 13:47:38 <mizerydearia> but maybe if I suggest it here I can inspire others to relay the suggestion infinitely until eventually someone makes the logo a link.  Otherwise I'll let the idea fade into forever forgottenness.
1194 2011-03-03 13:47:40 <d-snp> I used too many words I probably don't fully understand in that sentence, please interpret it disregarding any bullshit :P
1195 2011-03-03 13:48:42 <d-snp> .. but it does mean there's a small chance  you never ever win a block
1196 2011-03-03 13:51:52 <mizerydearia> Does anyone know who's site witcoin.com is?
1197 2011-03-03 13:54:10 <sipa> d-snp: yes, it's a poisson proces
1198 2011-03-03 13:54:39 <sipa> but the average numbers of blocks you find furing a given time interval, is proportional to your processing power
1199 2011-03-03 14:00:02 <d-snp> alright
1200 2011-03-03 14:00:03 <sipa> ;;bc,calc 650000
1201 2011-03-03 14:00:03 <gribble> The average time to generate a block at 650000 Khps, given current difficulty of 55590.23763914 , is 4 days, 6 hours, 2 minutes, and 0 seconds
1202 2011-03-03 14:00:03 <sipa> ;;bc,gen 650000
1203 2011-03-03 14:00:03 <gribble> The expected generation output, at 650000 Khps, given current difficulty of 55590.23763914 , is 11.7608500188 BTC per day and 0.490035417449 BTC per hour.
1204 2011-03-03 14:00:04 Bister has joined
1205 2011-03-03 14:00:04 Bister is now known as Bistre
1206 2011-03-03 14:03:11 <quellhorst> hmm, my rig is using $36 in electric per month. if i did the calculation right... $.12 kwh... 420 watts used 24/7.
1207 2011-03-03 14:03:32 <d-snp> thanks a lot for the explanations sipa :)
1208 2011-03-03 14:04:04 Bistre has quit (Client Quit)
1209 2011-03-03 14:05:27 altamic has joined
1210 2011-03-03 14:09:32 altamic has quit (Ping timeout: 240 seconds)
1211 2011-03-03 14:10:24 <d-snp> quellhorst: are you selling bitcoin on a regular basis?
1212 2011-03-03 14:10:34 <d-snp> or are you just hoarding? :)
1213 2011-03-03 14:10:44 Zarutian has joined
1214 2011-03-03 14:11:13 Animeking has joined
1215 2011-03-03 14:11:14 <Animeking> so
1216 2011-03-03 14:11:21 <[Noodles]> hoarding generation-output of a 420W-rig?
1217 2011-03-03 14:11:29 <[Noodles]> ^_^
1218 2011-03-03 14:11:38 <Animeking> the bitcoin client, can act as a 'server' for the miner assuming I understand how the miner works to a lesser extent
1219 2011-03-03 14:11:39 <Animeking> ?
1220 2011-03-03 14:11:52 <sipa> yes
1221 2011-03-03 14:11:54 <[Noodles]> yes, it can
1222 2011-03-03 14:11:57 <Animeking> Does this mean, I can use multiple machines within the lan as a miner to my PC?
1223 2011-03-03 14:12:05 <[Noodles]> yes
1224 2011-03-03 14:12:05 <sipa> yes
1225 2011-03-03 14:12:08 <sipa> hehe
1226 2011-03-03 14:12:11 <Animeking> o-o wow, never knew that.
1227 2011-03-03 14:13:21 <d-snp> but they generate transactions so your machines could be distributed around the world as long as the creator transactions are addressed to you
1228 2011-03-03 14:13:53 <d-snp> they don't necessarily have to be aware of eachother right?
1229 2011-03-03 14:13:56 <sipa> actually, the client creates the transactions
1230 2011-03-03 14:14:00 <sipa> not the miners
1231 2011-03-03 14:14:12 <d-snp> oh
1232 2011-03-03 14:14:23 <sipa> they just search for nonces that make the block (including the generation tx) hash to a low number
1233 2011-03-03 14:14:27 altamic has joined
1234 2011-03-03 14:14:53 <d-snp> ah, what I didn't really learn yet, how do you know you have received a transaction?
1235 2011-03-03 14:15:16 <sipa> you just receive it :)
1236 2011-03-03 14:15:18 <[Noodles]> the blockchain knows
1237 2011-03-03 14:15:23 <[Noodles]> and you know the blockchain
1238 2011-03-03 14:15:33 <d-snp> so you have to be listening, or read the blockchain?
1239 2011-03-03 14:15:45 <d-snp> the blockchain doesn't keep a full history right?
1240 2011-03-03 14:15:50 <d-snp> at least, it isn't obliged to?
1241 2011-03-03 14:15:50 <sipa> currently it does
1242 2011-03-03 14:15:55 <sipa> but it's not necessary
1243 2011-03-03 14:16:00 <[Noodles]> well, if you want to know that you received coins, you have to check the blockchain
1244 2011-03-03 14:16:05 <sipa> only outputs of unspent transactions need to be kept
1245 2011-03-03 14:16:35 <sipa> *only unspent outputs of transactions need to be kept
1246 2011-03-03 14:16:37 <d-snp> and a transaction you don't know about is always unspent ofcourse
1247 2011-03-03 14:16:53 <sipa> no, those you just don't care about
1248 2011-03-03 14:17:06 <sipa> the block chains continuously builds an ordered sequence of transactions
1249 2011-03-03 14:17:22 <sipa> each trasnactions consumes some outputs of previous transactions, and creates some new outputs
1250 2011-03-03 14:18:46 <d-snp> I was thinking from the perspective of an account holder that had his machine turned off for the night and might have had incoming transactions during the night
1251 2011-03-03 14:19:20 <d-snp> he might turn his machine on, and open his wallet, and see new transactions there deduced from the block chain?
1252 2011-03-03 14:19:46 <sipa> his client will receive the new blocks in the chain as soon as he reconnects
1253 2011-03-03 14:19:57 <sipa> and he will notice that it contains transactions which influence his balance
1254 2011-03-03 14:20:25 <d-snp> does the network track which transactions haven't been affirmed or something?
1255 2011-03-03 14:20:41 <magnetron> is the account balance implied or explicit in a block?
1256 2011-03-03 14:20:48 Jeroenz0r_ has joined
1257 2011-03-03 14:20:58 <sipa> magnetron: implicit
1258 2011-03-03 14:21:42 <magnetron> sipa: what prevents an account from transferring more bitcoins than it has in its balance?
1259 2011-03-03 14:21:49 ApertureScience has quit (Quit: Linux: because a PC is a terrible thing to waste)
1260 2011-03-03 14:21:55 <sipa> there are no account balances
1261 2011-03-03 14:22:11 <d-snp> magnetron: you don't know what a bitcoin you don't have looks like
1262 2011-03-03 14:22:15 Jeroenz0r has quit (Read error: Connection reset by peer)
1263 2011-03-03 14:22:19 Jeroenz0r_ is now known as Jeroenz0r
1264 2011-03-03 14:22:22 <quellhorst> d-snp: never sold any, still waiting for my first block
1265 2011-03-03 14:22:50 <sipa> d-snp: no, you don't *have* bitcoins
1266 2011-03-03 14:22:55 <magnetron> d-snp: well i don't have any balance at all, it's implicit
1267 2011-03-03 14:23:07 <sipa> you have the keys to sign transaction outputs, in order to use them
1268 2011-03-03 14:23:23 <sipa> and the client tracks which transactions have such outputs
1269 2011-03-03 14:23:29 <sipa> and the sum of those is your account balance
1270 2011-03-03 14:24:16 <magnetron> sipa: i have a bitcoin account with 0.05 BTC. what prevents me from signing a transaction that says " i give EFF  ONE MILLION BITCOINS"
1271 2011-03-03 14:24:41 <sipa> magnetron: i'll call a transaction output a coin for now
1272 2011-03-03 14:24:55 <sipa> the blockchain describes transactions done with coins
1273 2011-03-03 14:25:26 <sipa> a transaction: 1) says which coins it "consumes", 2) which coins it "creates", and 3) which addresses are allowed to consume the coins created in 2
1274 2011-03-03 14:26:03 BlueMatt has joined
1275 2011-03-03 14:26:14 <sipa> the sum of the values of the coins a transaction created must equal the sum of the values of the coins it consumes
1276 2011-03-03 14:26:34 <sipa> ok?
1277 2011-03-03 14:26:36 <magnetron> yes
1278 2011-03-03 14:26:45 <sipa> to be able to spend a coin
1279 2011-03-03 14:26:51 <sipa> you must reference it
1280 2011-03-03 14:27:00 <magnetron> you mean reference the account
1281 2011-03-03 14:27:03 <sipa> no
1282 2011-03-03 14:27:11 <sipa> accounts don't even exist at this level
1283 2011-03-03 14:27:17 <BlueMatt> although tx fees are not directly included and thus the fee is the difference between inputs and outputs
1284 2011-03-03 14:27:26 <sipa> BlueMatt: i'm simplifying, but yes indeed
1285 2011-03-03 14:27:28 <magnetron> BlueMatt: yes, let's disregard that
1286 2011-03-03 14:28:22 <bk128> I'm building my next mining rig in this http://www.itworld.com/hardware/138740/asus-motherboard-box-doubles-a-pc-case :)
1287 2011-03-03 14:28:29 <sipa> magnetron: if i want to spend a coin, i explicitly need to tell the world "i am now now consuming coin number N created in transaction T, and here is a signature by me which proves i am the owner of the account stated to be allowed to use it"
1288 2011-03-03 14:28:58 <sipa> and afterwards no one may spend coin number N of transaction T ever again
1289 2011-03-03 14:28:58 pierre` has joined
1290 2011-03-03 14:29:01 <pierre`> hi
1291 2011-03-03 14:29:06 <sipa> and afterwards is defined as "later in the block chain"
1292 2011-03-03 14:29:26 <sipa> i'm using "spend" and "consume" as synonyms here
1293 2011-03-03 14:29:30 <magnetron> sipa: who makes sure that i never spend coin number N ever again?
1294 2011-03-03 14:29:34 <sipa> everyone
1295 2011-03-03 14:29:41 <magnetron> sipa: you mean the miner
1296 2011-03-03 14:29:44 <sipa> no, everyone
1297 2011-03-03 14:29:56 <sipa> no client will even relay your transaction if it's using a spent coin
1298 2011-03-03 14:30:12 <BlueMatt> and will ignore blocks with such errors
1299 2011-03-03 14:30:19 <sipa> (for now, in the future thinner clients which do not do this verification will be made)
1300 2011-03-03 14:30:28 <BlueMatt> hence the long block chain verification
1301 2011-03-03 14:31:14 <magnetron> sipa: so every client has a long table of all account balances, and never accepts blocks that would reduce the implied balance of an account below zero?
1302 2011-03-03 14:31:22 <sipa> no, no one has an account balance
1303 2011-03-03 14:31:26 <sipa> an account does not exist
1304 2011-03-03 14:31:33 <sipa> it is something only your client is aware of
1305 2011-03-03 14:31:34 <magnetron> an address
1306 2011-03-03 14:31:51 <sipa> even an address is fuzzy, and only used in the signature verification
1307 2011-03-03 14:31:59 <magnetron> private key
1308 2011-03-03 14:32:18 <sipa> at this level in the block chain, there are only transaction outputs
1309 2011-03-03 14:32:27 <sipa> and i use some of them, if i spend coins
1310 2011-03-03 14:32:36 <magnetron> what is the data type of a transaction output, then?
1311 2011-03-03 14:32:37 <sipa> there is no need to keep balances of anything
1312 2011-03-03 14:32:55 <sipa> it's a reference to a transaction, plus a number
1313 2011-03-03 14:33:15 <sipa> T:N, and refers to the Nth output of transaction with hash T
1314 2011-03-03 14:33:44 <magnetron> ok
1315 2011-03-03 14:33:49 <sipa> and yes, all "fat" clients keep a database of all unspent transaction outputs
1316 2011-03-03 14:34:00 <sipa> in orde to be able to verify that a single output is never spent twice
1317 2011-03-03 14:34:02 <magnetron> wait
1318 2011-03-03 14:34:05 <sipa> in the same chain
1319 2011-03-03 14:34:12 <magnetron> stop
1320 2011-03-03 14:34:31 * sipa stops
1321 2011-03-03 14:34:36 bitterJ has joined
1322 2011-03-03 14:34:37 <magnetron> the INPUT of a transaction must be the previous transaction output T:N
1323 2011-03-03 14:34:52 <sipa> yes
1324 2011-03-03 14:35:02 <BlueMatt> what is the protocol that thin clients will use? will they connect to a full server? or use some kind of modified protocol on a p2p network?
1325 2011-03-03 14:35:16 <sipa> just the p2p network
1326 2011-03-03 14:35:30 <sipa> but they will relay more (possibly invalid) transactions and blocks
1327 2011-03-03 14:35:42 <sipa> since they cannot do full verification
1328 2011-03-03 14:35:52 bitterJ has quit (Client Quit)
1329 2011-03-03 14:35:52 <magnetron> but how can the OUTPUT of a transaction be a T:N pair? This would mean that transactions are recursions. or that they point to transactions that haven't yet happened
1330 2011-03-03 14:36:05 <BlueMatt> sipa: is there existing support in the protocol?
1331 2011-03-03 14:36:23 bitterJ has joined
1332 2011-03-03 14:36:27 <BlueMatt> magnetron they point all the way back to their generation
1333 2011-03-03 14:36:34 <sipa> magnetron: T is the hash of an earlier transaction
1334 2011-03-03 14:36:38 <BlueMatt> ie a generation tx has no input
1335 2011-03-03 14:36:56 <sipa> magnetron: go to blockexplorer.com and click around a bit
1336 2011-03-03 14:37:14 <x6763> sipa: haha, i was just going to suggest maying using blockexplorer to help with your explanation
1337 2011-03-03 14:37:20 <x6763> *maybe
1338 2011-03-03 14:38:02 <sipa> BlueMatt: the protocol supports this, since nodes only advertise their knowledge of objects, but the other side needs to request it if they are interested
1339 2011-03-03 14:38:02 <BlueMatt> who runs the channel log? it appears to be downs
1340 2011-03-03 14:38:52 <BlueMatt> ah, so thin clients might request a block to verify a tx, and watch new blocks for their addresses but not store a db
1341 2011-03-03 14:39:32 <sipa> magnetron: oh, i misunderstood
1342 2011-03-03 14:39:47 Jeroenz0r has quit ()
1343 2011-03-03 14:39:49 <magnetron> sipa: take http://blockexplorer.com/tx/94cefb779add220ee10bdd254e4e66d19900637178f6859f7c2f54782a00c903 as an example. this seems to imply that a transaction output is an amount transferred to an address
1344 2011-03-03 14:40:13 Jeroenz0r has joined
1345 2011-03-03 14:40:13 Jeroenz0r has quit (Changing host)
1346 2011-03-03 14:40:13 Jeroenz0r has joined
1347 2011-03-03 14:40:25 <sipa> magnetron: a transaction just lists a number of outputs, each one being an amount and a script to verify who is allowed to spend it (typically an address or a pubkey indeed)
1348 2011-03-03 14:40:30 altamic has quit (Ping timeout: 250 seconds)
1349 2011-03-03 14:40:35 <BlueMatt> magnetron it pretty much is
1350 2011-03-03 14:40:54 <sipa> and an input refers back to a previous output, by stating its hash, and the number of the output
1351 2011-03-03 14:41:38 bk128 has quit (Quit: bk128)
1352 2011-03-03 14:42:00 <BlueMatt> ;;seen gavinandresen
1353 2011-03-03 14:42:00 <gribble> gavinandresen was last seen in #bitcoin-dev 16 hours, 45 minutes, and 59 seconds ago: <gavinandresen> BlueMatt:  yup, IPv6 will be an issue...   Perhaps detecting a Sybil attack will be easier than defending against it.
1354 2011-03-03 14:44:22 EvanR-work has joined
1355 2011-03-03 14:44:24 <TD> BlueMatt: SPV clients need what is basically a slightly upgraded version of todays p2p protocol
1356 2011-03-03 14:44:34 <TD> with a getmerklebranch message
1357 2011-03-03 14:44:46 <TD> a few other bits on the addr messages, perhaps
1358 2011-03-03 14:45:22 bitterJ has quit (Quit: Leaving)
1359 2011-03-03 14:45:22 <BlueMatt> oh, ok so they will be able to request a specific tx then?
1360 2011-03-03 14:45:53 <TD> you can already request transactions by their hash
1361 2011-03-03 14:46:07 <TD> today, implementing SPV means you have to wait a block to have any confidence in a transaction at all
1362 2011-03-03 14:46:13 <TD> or, you trust the node you are connected to
1363 2011-03-03 14:46:37 <TD> with a getmerklebranch message, when receiving a transaction you could get the same level of confidence a 0/unconfirmed in the official client gives you
1364 2011-03-03 14:46:44 <TD> even though you don't have a full copy of the block chain
1365 2011-03-03 14:47:14 darsk1ez has quit (Read error: Operation timed out)
1366 2011-03-03 14:47:25 <BlueMatt> how so? can one not simply request the tx input by hash?
1367 2011-03-03 14:47:50 darsk1ez has joined
1368 2011-03-03 14:48:09 <TD> yes but it doesn't tell you anything
1369 2011-03-03 14:48:22 <TD> if you don't trust the node you're connected to, they can just invent a transaction for the inputs
1370 2011-03-03 14:48:29 <TD> linking it to a place in the chain tells you the wider network accepted it
1371 2011-03-03 14:48:32 <TD> the node can't forge that
1372 2011-03-03 14:48:44 <BlueMatt> what about requesting that block
1373 2011-03-03 14:48:45 <TD> (unless they own gobs of cpu power, which is the standard case for attacks on bitcoin)
1374 2011-03-03 14:48:50 bitterJ has joined
1375 2011-03-03 14:48:57 <TD> there's no way to go from a tx hash to a block hash in the protocol today.
1376 2011-03-03 14:49:05 <TD> it's assumed you have an index of all transactions stored locally
1377 2011-03-03 14:49:10 <BlueMatt> ah, so hence the need
1378 2011-03-03 14:49:14 <TD> and in future blocks will be enormous
1379 2011-03-03 14:49:21 <TD> fortunately all you need is the merkle branch
1380 2011-03-03 14:50:00 bitterJ has quit (Read error: Connection reset by peer)
1381 2011-03-03 14:50:01 <BlueMatt> never done any research on how merkle trees work, care to explain?
1382 2011-03-03 14:50:45 <TD> it's a tree of hashes
1383 2011-03-03 14:50:58 <TD> if you're given a branch of the tree you can get back to the tree root
1384 2011-03-03 14:51:00 OneFixt has quit (Ping timeout: 250 seconds)
1385 2011-03-03 14:51:00 <TD> which is in the block headers
1386 2011-03-03 14:51:03 <sipa> you have a list of hashes
1387 2011-03-03 14:51:18 <TD> so if you have all the block headers, a transaction and a branch, you can check it was indeed a part of that block
1388 2011-03-03 14:51:23 <TD> without having the entire contents of the block
1389 2011-03-03 14:51:24 <sipa> in each "layer" you take two subsequent ones together, hash them again, giving you a higher-level hash
1390 2011-03-03 14:51:34 <sipa> you repeat this proces, untill you have only one hash left
1391 2011-03-03 14:51:39 <sipa> that's the merkle root
1392 2011-03-03 14:52:42 <BlueMatt> so "thin" clients will have a full block headers db?
1393 2011-03-03 14:54:25 <sipa> yes
1394 2011-03-03 14:54:29 <sipa> but that's only 80 bytes per block
1395 2011-03-03 14:54:43 <TD> it's not strictly necessary
1396 2011-03-03 14:55:02 <luke-jr> ;;bc,stats
1397 2011-03-03 14:55:02 <sipa> they could request those as well when needed
1398 2011-03-03 14:55:03 <gribble> Current Blocks: 111564 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1331 blocks | Next Difficulty In About: 1 week, 1 day, 3 hours, 57 minutes, and 10 seconds | Next Difficulty Estimate: 63004.66142471
1399 2011-03-03 14:55:06 <TD> if the root of the branch isn't in your collection of block headers, you can ask the other node to send you those headers to the pointw here you do have them
1400 2011-03-03 14:55:16 <TD> so you can choose your own storage vs bandwidth/latency tradeoffs
1401 2011-03-03 14:55:30 <BlueMatt> fair enough, thanks
1402 2011-03-03 14:55:32 <TD> headers are only ~5mb a year though, i think
1403 2011-03-03 14:56:20 <TD> the official software doesn't support this mode fully today
1404 2011-03-03 14:56:24 <TD> it's not a big project to add though
1405 2011-03-03 14:56:34 <BlueMatt> is there an existing command to get just a blocks headers?
1406 2011-03-03 14:56:37 <TD> yes
1407 2011-03-03 14:57:02 <TD> the client i've implemented works sort of like SPV mode, in that it requires a block to receive a transaction.
1408 2011-03-03 14:57:12 <TD> it doesn't use getheaders today but that's a good optimization i'll implement at some point
1409 2011-03-03 14:57:25 rgm3 has joined
1410 2011-03-03 14:57:52 <BlueMatt> ok so, yea not too hard to do then...I'm always amazed at how well satoshi planned this even if all
1411 2011-03-03 14:57:57 <TD> with the current state of things though, there's no point because you can't get the merkle branch. if a tx is in a block of the right difficulty, and it snaps onto a checkpoint you know is valid, it's ok
1412 2011-03-03 14:58:01 <BlueMatt> the stuff want even implemented
1413 2011-03-03 14:58:12 <TD> yes he worked on the design and coding for a couple of years before release
1414 2011-03-03 14:58:35 <TD> that's why bitcoin is taking off. it is remarkably well thought out. very few problems that need to be solved
1415 2011-03-03 14:58:41 <TD> really only DoS attacks
1416 2011-03-03 14:58:53 <TD> and some stuff to do with very fast transaction acceptance
1417 2011-03-03 14:59:10 <BlueMatt> tx acceptance problems?
1418 2011-03-03 14:59:24 <eps> the backlog i presume
1419 2011-03-03 14:59:29 <TD> no that was just a bug
1420 2011-03-03 14:59:36 <TD> i mean if you want to accept coins without waiting for a block
1421 2011-03-03 14:59:47 <TD> how do you communicate the risk of that in a meaningful way, how do you minimize that risk etc
1422 2011-03-03 14:59:48 <eps> why would you want to?
1423 2011-03-03 14:59:48 <magnetron> TD: well, currently more than 15 free small transactions per block is considered as DDoS
1424 2011-03-03 15:00:35 <BlueMatt> is there a solution that people have put out to prevent double-pay?
1425 2011-03-03 15:00:36 <TD> eps: wanting to accept coins within seconds rather than minutes is a very common need
1426 2011-03-03 15:00:43 bk128 has joined
1427 2011-03-03 15:01:46 hwolf has quit (Ping timeout: 240 seconds)
1428 2011-03-03 15:02:31 <eps> unless the everybody on the whole internet gets upgraded to superfast speeds it is always going to take a while to get confirmations
1429 2011-03-03 15:02:46 <TD> speed of hardware is irrelevant
1430 2011-03-03 15:02:47 <eps> or am i misunderstanding
1431 2011-03-03 15:02:48 <TD> the network aims for 10 minutes
1432 2011-03-03 15:02:54 <TD> taking total hash power into account
1433 2011-03-03 15:03:01 <TD> so confirmation time will always be roughly 10 mins
1434 2011-03-03 15:03:07 <BlueMatt> eps: the point is to accept before confirmations
1435 2011-03-03 15:03:21 <BlueMatt> well to minimize risk of ...
1436 2011-03-03 15:03:22 <eps> when do confirmations start?, after a block is generated?
1437 2011-03-03 15:03:33 <TD> the canconical example of where this is a problem is vending machines. fortunately the difficulty of mounting a double spend even on 0/unconfirmed transactions is pretty high
1438 2011-03-03 15:03:40 <TD> probably not worth it for just buying snacks
1439 2011-03-03 15:04:09 <TD> eps: confirmations is the number of blocks since the transaction was incorporated into the chain. 1 confirmation means 1 block. 2 means 2 blocks etc
1440 2011-03-03 15:04:18 <TD> more blocks == more certainty the transaction won't reverse
1441 2011-03-03 15:04:37 <TD> after an hour you're practically guaranteed that even a very high powered adversary cannot reverse the tx
1442 2011-03-03 15:06:38 slush has quit (Quit: Leaving.)
1443 2011-03-03 15:07:48 bk128 has quit (Quit: bk128)
1444 2011-03-03 15:14:20 newsham has quit (Ping timeout: 272 seconds)
1445 2011-03-03 15:20:11 BlueMatt has quit (Remote host closed the connection)
1446 2011-03-03 15:24:00 BlueMatt has joined
1447 2011-03-03 15:25:31 gavinandresen has joined
1448 2011-03-03 15:25:33 Zarutian has quit (Quit: Zarutian)
1449 2011-03-03 15:25:55 <Diablo-D3> hrm
1450 2011-03-03 15:26:07 <Diablo-D3> you know
1451 2011-03-03 15:27:07 <gavinandresen> luke-jr:  you know that not all integers passed to the JSON-RPC interface are money values, right?  (RE: money handling wiki page)
1452 2011-03-03 15:27:37 <Diablo-D3> gavinandresen: yeah, but what the fuck, why is the money value being passed in number representation
1453 2011-03-03 15:27:59 <Diablo-D3> it should either be a string, or be an integer in ubtc form
1454 2011-03-03 15:28:29 AmpEater has joined
1455 2011-03-03 15:29:01 <gavinandresen> yeah yeah yeah, sure fine whatever.  What does it matter, you're going to have to do proper conversion to/from the "1.00" format to display BTC values to users anyway
1456 2011-03-03 15:29:27 <gavinandresen> ... or do you think people will be OK with typing 100000000 when they want to send 1 BTC ???
1457 2011-03-03 15:30:23 Kiba has joined
1458 2011-03-03 15:30:49 Raulo has joined
1459 2011-03-03 15:32:09 BlueMatt has quit (Quit: Leaving)
1460 2011-03-03 15:32:23 BlueMatt has joined
1461 2011-03-03 15:32:23 BlueMatt has quit (Changing host)
1462 2011-03-03 15:32:23 BlueMatt has joined
1463 2011-03-03 15:32:45 <BlueMatt> when did the -dev logger break?
1464 2011-03-03 15:36:38 <BlueMatt> gavinandresen: does the connect only to one address per a.b.?.? not protect against the attack which the 2h delay on odd-port clients is designed to prevent?
1465 2011-03-03 15:36:51 <gwillen> gavinandresen: here is an example of why you never use floating point for money: http://www.javaranch.com/journal/2003/07/MoneyInJava.html
1466 2011-03-03 15:38:49 AmpEater has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
1467 2011-03-03 15:38:57 <gavinandresen> gwillen:  yeah, that's a red-herring-- note they're talking about single-precision floating point.  JSON is double-precision, which is 53 bits of precision, which is plenty to hold old 2.1 quadrillion bitcoin values.
1468 2011-03-03 15:39:31 <gavinandresen> The only way to get into trouble is to forget to ROUND instead of TRUNCATE to 8 decimal places.
1469 2011-03-03 15:39:54 <gwillen> or you could do it sensibly, with integers, and then there's no way to get into trouble at all :-P
1470 2011-03-03 15:41:20 devon_hillard has joined
1471 2011-03-03 15:41:20 devon_hillard has quit (Changing host)
1472 2011-03-03 15:41:20 devon_hillard has joined
1473 2011-03-03 15:41:53 <gavinandresen> Coulda shoulda woulda....   there is no integer numeric type in JSON.  And if we passed numbers as strings there would be a constant "why do I have to type numbers with quotes around them".
1474 2011-03-03 15:42:25 <gwillen> nobody should ever have to type raw JSON into an interface anyway
1475 2011-03-03 15:42:41 <gwillen> "why do I have to type quotes" or "why do I have to type all these digits" is clearly also a red herring
1476 2011-03-03 15:42:50 <gwillen> this is what user interfaces are for
1477 2011-03-03 15:42:55 <BlueMatt> either way not optimal, but either way not worth changing now, just have to be careful when programming
1478 2011-03-03 15:43:16 <gwillen> the lack of an integer type in JSON is indeed a problem.
1479 2011-03-03 15:44:08 <TD> protobufs ;)
1480 2011-03-03 15:44:20 <TD> there is  even a json/proto bridge
1481 2011-03-03 15:44:26 <gwillen> haha
1482 2011-03-03 15:44:27 <lfm> isnt the actual protocol in char mode anyway. its just the api that uses floats?
1483 2011-03-03 15:45:34 <luke-jr> gavinandresen: ah, that might be a problem with my change
1484 2011-03-03 15:46:11 <luke-jr> gavinandresen: human interaction is for the GUI only :p
1485 2011-03-03 15:46:46 <luke-jr> Diablo-D3: JSON-RPC is broken in more ways than just this. Hence why it's slated for replacement
1486 2011-03-03 15:46:58 <Diablo-D3> hurr "replacement"
1487 2011-03-03 15:47:05 <gavinandresen> luke-jr: says who?
1488 2011-03-03 15:47:14 <Diablo-D3> gavinandresen: trolls, mainly
1489 2011-03-03 15:47:15 <luke-jr> gavinandresen: https://www.bitcoin.org/smf/index.php?topic=3757.0
1490 2011-03-03 15:47:18 <bitbot> New, standardized wallet protocol Connection refused.
1491 2011-03-03 15:47:24 MartianW has joined
1492 2011-03-03 15:47:24 hwolf has joined
1493 2011-03-03 15:47:26 <Aciid> does anyone have any idea how many MH/s does 6970 do?
1494 2011-03-03 15:47:32 <luke-jr> … how did that bot do that? XD
1495 2011-03-03 15:47:36 <Diablo-D3> Aciid: maybe 230
1496 2011-03-03 15:47:48 <Diablo-D3> Aciid: not enough people own them
1497 2011-03-03 15:47:54 <Aciid> Diablo-D3: asked coz I seriously couldn't find the information
1498 2011-03-03 15:47:59 <Aciid> ok
1499 2011-03-03 15:48:18 <Diablo-D3> luke-jr: whoever wrote this idea up is an idiot
1500 2011-03-03 15:48:28 <Diablo-D3> Im rejecting this.
1501 2011-03-03 15:48:39 <lfm> Aciid: chech out https://en.bitcoin.it/wiki/Mining_hardware_comparison
1502 2011-03-03 15:48:40 <luke-jr> Diablo-D3: the idiot is the one who rejects it without an alternative proposal
1503 2011-03-03 15:48:53 <Diablo-D3> lfm: 69xx isnt on that yet
1504 2011-03-03 15:48:56 <luke-jr> The existing JSON-RPC is unworkable.
1505 2011-03-03 15:49:03 <lfm> doh
1506 2011-03-03 15:49:06 <Diablo-D3> luke-jr: your mom is unworkable.
1507 2011-03-03 15:49:38 <Aciid> lfm: 6970 aint there, also. would someone want to add my CPU there?
1508 2011-03-03 15:49:39 larsivi has joined
1509 2011-03-03 15:49:43 <luke-jr> JSON-RPC works only for shopping cart automata type things, not UIs.
1510 2011-03-03 15:49:55 <Diablo-D3> luke-jr: you cant code.
1511 2011-03-03 15:50:00 <luke-jr> nor miners, in some ways.
1512 2011-03-03 15:50:01 <Diablo-D3> and its quite obvious.
1513 2011-03-03 15:50:10 <luke-jr> Diablo-D3: maybe you can't. Java luser.
1514 2011-03-03 15:50:24 <Diablo-D3> luke-jr: you dont even know what you're arguing about
1515 2011-03-03 15:50:28 <BlueMatt> gavinandresen: does setConnected.count(addr.ip & 0x0000ffff) not remove the need for the addr.port != GetDefaultPort() 2h delay?
1516 2011-03-03 15:50:40 <luke-jr> Diablo-D3: unlike you, I've actually worked on a UI for this.
1517 2011-03-03 15:50:42 <lfm> Diablo-D3: hes trying to do his tonal crap and cant make it work. oh dear!
1518 2011-03-03 15:50:45 <Diablo-D3> luke-jr: you're basically trying to attempt the soap vs rest argument
1519 2011-03-03 15:50:52 MartianW has quit (Client Quit)
1520 2011-03-03 15:50:57 <Aciid> Phenom II X6 1100T, nproc 6, MH/s 22, clock 3.8ghz
1521 2011-03-03 15:50:59 <Diablo-D3> luke-jr: and you dont understand why, largely, the argument is moot.
1522 2011-03-03 15:51:00 <gavinandresen> BlueMatt: I dunno, the networking code isn't my area of expertise.
1523 2011-03-03 15:51:00 <Aciid> could have been more
1524 2011-03-03 15:51:01 <luke-jr> lfm: JSON-RPC problems are unrelated to tonal
1525 2011-03-03 15:51:16 <BlueMatt> gavin: who handles networking code?
1526 2011-03-03 15:51:26 <gavinandresen> (I DO know a lot about floating point precision issues, however....)
1527 2011-03-03 15:51:35 <BlueMatt> lol
1528 2011-03-03 15:51:56 frewsxcv has joined
1529 2011-03-03 15:51:59 AmpEater has joined
1530 2011-03-03 15:52:00 <frewsxcv> i'm creating a bitcoin extension for firefox https://addons.mozilla.org/en-US/firefox/addon/none/
1531 2011-03-03 15:52:09 <frewsxcv> here's the source code if you want to contribute https://github.com/frewsxcv/Bitcoin-Prices
1532 2011-03-03 15:52:12 <Diablo-D3> json-rpc is basically soap for json
1533 2011-03-03 15:52:21 <Diablo-D3> and guess what, I actually hate soap for a lot of things
1534 2011-03-03 15:52:30 <frewsxcv> basically it just displays mtgox for now
1535 2011-03-03 15:52:31 <Diablo-D3> and for what we're doing, rest probably is a better idea.
1536 2011-03-03 15:52:43 <AmpEater> Ohh, firefox plugin, good idea
1537 2011-03-03 15:52:50 <luke-jr> Diablo-D3: so suggest REST for the new protocol
1538 2011-03-03 15:52:51 <BlueMatt> frewsxcv: I would love to help
1539 2011-03-03 15:53:04 <Diablo-D3> luke-jr: I dont think you understand what REST is, so diaf.
1540 2011-03-03 15:53:10 <BlueMatt> frewsxcv: not that I know anything about plugin development, but Ill do what I can...
1541 2011-03-03 15:53:13 <frewsxcv> BlueMatt: it's mostly javascript. feel free to fork my code
1542 2011-03-03 15:53:14 <luke-jr> nope. probably some other retarded thing, tbh
1543 2011-03-03 15:53:52 <frewsxcv> BlueMatt: https://jetpack.mozillalabs.com/sdk/1.0b3/docs/
1544 2011-03-03 15:54:58 <luke-jr> yeah, REST looks like just a new term for AJAX more or less
1545 2011-03-03 15:55:11 <frewsxcv> rest is beautiful
1546 2011-03-03 15:55:14 <luke-jr> trying to force application-like interfaces through request interfaces
1547 2011-03-03 15:56:17 <Diablo-D3> luke-jr: they arent related.
1548 2011-03-03 15:56:27 <Diablo-D3> REST is just making your methods part of the URL.
1549 2011-03-03 15:56:34 <Diablo-D3> instead of part of the protocol
1550 2011-03-03 15:56:39 <luke-jr> Diablo-D3: it's the same crap
1551 2011-03-03 15:56:58 barburator has joined
1552 2011-03-03 15:57:00 <luke-jr> URLs are designed for requests, not bidirectional communication between applications
1553 2011-03-03 15:57:06 AmpEater has quit (Remote host closed the connection)
1554 2011-03-03 15:57:32 <Diablo-D3> so instead of doing http://localhost:8333/ -> GET -> { method: "getwork" } you do http://localhost:8333/work/ -> GET
1555 2011-03-03 15:57:44 <Diablo-D3> luke-jr: you fail at http forever.
1556 2011-03-03 15:58:05 <luke-jr> Diablo-D3: you're trying to fix a problem that doesn't even exist with that
1557 2011-03-03 15:58:27 <x6763> rest would make things a lot simpler for people writing software to access bitcoind
1558 2011-03-03 15:58:35 <Aciid> so 6870x2 would be the best ratio to buy at the moment?
1559 2011-03-03 15:59:07 <luke-jr> x6763: how so? all it does is move the method from content to URI
1560 2011-03-03 15:59:12 <gavinandresen> ... so if I call http://localhost:$RESTPORT/info  ... what's returned?
1561 2011-03-03 15:59:51 <Diablo-D3> gavinandresen: you know whats returned from json-rpc getinfo? the map node named result? that.
1562 2011-03-03 16:00:06 <frewsxcv> BlueMatt: let me know if you ever are interested
1563 2011-03-03 16:00:09 <gavinandresen> Encoded in JSON?
1564 2011-03-03 16:00:22 <BlueMatt> frewsxcv: I might be, gonna do some research and see what I can do
1565 2011-03-03 16:00:23 <gavinandresen> Or key=values on separate lines?
1566 2011-03-03 16:00:32 <gavinandresen> Or some other ad-hoc encoding?
1567 2011-03-03 16:00:48 AmpEater has joined
1568 2011-03-03 16:01:01 <Diablo-D3> gavinandresen: encoded in json
1569 2011-03-03 16:01:08 <Diablo-D3> gavinandresen: it'd look exactly the same.
1570 2011-03-03 16:01:13 <gavinandresen> Seems like you're just reinventing the JSON-RPC over HTTP standard with, as luke-jr says, the method in the URL instead of params
1571 2011-03-03 16:01:14 <Diablo-D3> gavinandresen: could be encoded in xml too
1572 2011-03-03 16:01:22 <Diablo-D3> gavinandresen: no, REST came first.
1573 2011-03-03 16:01:41 <Diablo-D3> gavinandresen: all the native json protocols on the web are restful
1574 2011-03-03 16:01:53 <luke-jr> Diablo-D3: but it doesn't solve any real problems
1575 2011-03-03 16:01:55 <Diablo-D3> gavinandresen: and all the xml protocols that arent shit are restful too
1576 2011-03-03 16:02:08 <Diablo-D3> luke-jr: no, but it proves you're an idiot
1577 2011-03-03 16:02:19 <luke-jr> Diablo-D3: no u
1578 2011-03-03 16:02:22 <Diablo-D3> you have no clue what Ive been talking about for the past 15 minutes.
1579 2011-03-03 16:02:48 <luke-jr> I think you're just trolling, so I'm going to ignore you for a bit.
1580 2011-03-03 16:03:24 <rgm3> oh hey Diablo -- I think there's some issues with the parameter parsing in yesterday's miner.zip download.   It no longer accepts -u and -p, but --url works okay.
1581 2011-03-03 16:04:24 <Diablo-D3> rgm3: derp
1582 2011-03-03 16:04:25 <luke-jr> gavinandresen: anyhow, in addition to the all-too-often-mentioned problem of using human representation in internals/protocols, JSON-RPC also requires polling listtransactions (for UIs) and getwork (for miners), and lot of smarts for UIs to figure out which transactions it already has vs needs to update with
1583 2011-03-03 16:04:50 <Diablo-D3> and btw
1584 2011-03-03 16:04:54 <Diablo-D3> REST does fix one problem
1585 2011-03-03 16:05:04 <Diablo-D3> you can websocket with rest easily
1586 2011-03-03 16:05:32 <luke-jr> gavinandresen: a bidirectional protocol is needed to properly separate UI and wallet; as noted, slush and jgarzik have already been looking into a binary protocol fixing this in the limited scope of miners (which could be folded into a generic wallet protocol)
1587 2011-03-03 16:05:37 <gavinandresen> luke-jr:  RE: the editing war on the proper money handling page:  is there a JSON-RPC library that represents double-precision floating point numbers that HAVE more than 8 digits of precision with fewer than 8 ?
1588 2011-03-03 16:06:03 <luke-jr> gavinandresen: my understanding from the commit log, is that bitcoind's library originally used 4 digits
1589 2011-03-03 16:06:09 gasteve has quit (Quit: gasteve)
1590 2011-03-03 16:06:13 <luke-jr> and someone extended it to 8
1591 2011-03-03 16:06:19 <Diablo-D3> gavinandresen: you fail
1592 2011-03-03 16:06:25 <Diablo-D3> gavinandresen: do NOT use floats for money.
1593 2011-03-03 16:06:34 Syke has joined
1594 2011-03-03 16:06:43 <[Tycho]> :))
1595 2011-03-03 16:06:45 <gavinandresen> Diablo-D3: I don't, I use Decimal
1596 2011-03-03 16:06:53 <Diablo-D3> gavinandresen: if you want to fix this, either quit representing them as fixed point in the api, or output it as a string
1597 2011-03-03 16:07:25 <luke-jr> Diablo-D3: new protocol can use base units properly. JSON-RPC is legacy.
1598 2011-03-03 16:07:44 <Diablo-D3> Im not interested in a binary protocol
1599 2011-03-03 16:07:46 <Diablo-D3> and neither is anyone else
1600 2011-03-03 16:07:53 <Diablo-D3> binary protocols, largely, are legacy.
1601 2011-03-03 16:07:57 gasteve has joined
1602 2011-03-03 16:08:10 <luke-jr> Diablo-D3: good for you, where did anyone say this would be a binary protocol?
1603 2011-03-03 16:08:19 <Diablo-D3> thats what its been implied.
1604 2011-03-03 16:08:29 <Diablo-D3> if you're in charge of it, we're all doomed.
1605 2011-03-03 16:08:39 <luke-jr> Diablo-D3: it's still open to debate on the wiki. if you want to provide input, do it.
1606 2011-03-03 16:08:56 <Diablo-D3> theres no debate to be done
1607 2011-03-03 16:08:57 <luke-jr> Diablo-D3: nobody's in charge of it
1608 2011-03-03 16:09:04 <Diablo-D3> only satoshi and gavin can make the changes
1609 2011-03-03 16:09:10 <Diablo-D3> and satoshi isnt exactly the world's best programmer here
1610 2011-03-03 16:09:30 <gavinandresen> luke-jr:  on the json-spirit home page: "Version 2.0, 14th December 2007 - Increased precision of floating point number output"
1611 2011-03-03 16:10:17 <luke-jr> looking at the git commit, it appears I had it backward: it was originally 16 digits
1612 2011-03-03 16:10:59 <luke-jr> so it's possible there is no practical case where this is a problem, but just theoretical
1613 2011-03-03 16:10:59 AmpEater has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
1614 2011-03-03 16:12:07 <Diablo-D3> anyhow, if the new protocol doesnt use existing standards, its already failed.
1615 2011-03-03 16:12:37 <gavinandresen> Ok, I'm done arguing.  There are MUCH MUCH more important things to work on than "it might be a problem in theory"
1616 2011-03-03 16:12:40 <Diablo-D3> json-rpc, as sucky as it is, can be reasonably implemented, as a client, using any http and json parser
1617 2011-03-03 16:13:06 <Diablo-D3> gavinandresen: what, the number problem? yes, this is a serious issue
1618 2011-03-03 16:13:31 <Diablo-D3> gavinandresen: if this was real money, it'd be against the law to represent it as floats. as in, the kind of laws that fine you to death
1619 2011-03-03 16:14:40 <gavinandresen> Diablo-D3:  oh really?  show me a real-world example that doesn't involve either:  (a) a brain-dead, buggy JSON library   or (b) brain-dead, buggy code on the other end of the JSON  ... and I'll reconsider.
1620 2011-03-03 16:14:47 <luke-jr> Diablo-D3: but JSON-RPC does not support bidirectional communication AFAIK
1621 2011-03-03 16:14:59 <Diablo-D3> gavinandresen: apparently php's json lib wont let you grab the original string.
1622 2011-03-03 16:15:20 <luke-jr> Diablo-D3: no JSON lib does AFAIK
1623 2011-03-03 16:15:29 <Diablo-D3> luke-jr: erm, I can do it fine on linux
1624 2011-03-03 16:15:30 <Diablo-D3> err
1625 2011-03-03 16:15:33 <Diablo-D3> on java
1626 2011-03-03 16:15:47 <gavinandresen> Diablo-D3:  as long as it reads and write floating point with more than 8 digits of precision there is no issue.
1627 2011-03-03 16:16:03 <Diablo-D3> gavinandresen: except its icky as fuck and gives people the impression its okay to be dumb
1628 2011-03-03 16:16:07 <lfm> isnt the actual protocol in char mode anyway. its just the api that uses floats?
1629 2011-03-03 16:16:15 <luke-jr> gavinandresen: do you concur that the polling problem *cannot* be solved with JSON-RPC, or am I overlooking something?
1630 2011-03-03 16:16:17 <gavinandresen> lmf: yup
1631 2011-03-03 16:16:18 <Diablo-D3> lfm: yes, its all human readable
1632 2011-03-03 16:16:30 <Diablo-D3> and the polling problem isnt an issue of json-rpc
1633 2011-03-03 16:16:37 <TD> i think this bikeshed was painted enough already
1634 2011-03-03 16:16:41 <Diablo-D3> you could, if you were an idiot enough, do json-rpc with websockets
1635 2011-03-03 16:16:44 <Diablo-D3> _I dont suggest it_
1636 2011-03-03 16:16:46 <gavinandresen> luke-jr:  JSON-RPC allows persistent connections.  That's actually the normal case, JSON-RPC over HTTP is an extension.
1637 2011-03-03 16:16:48 <Diablo-D3> but it can be done
1638 2011-03-03 16:17:00 <TD> if somebody cares deeply enough about this, they can write a patch and debate it on a review thread, or even a proxy
1639 2011-03-03 16:17:39 <Diablo-D3> gavinandresen: and no, json-rpc over http isnt really an extension
1640 2011-03-03 16:18:10 <gavinandresen> Diablo-D3:  The JSON-RPC spec is independent of HTTP. THere is a separate spec for JSON-RPC over HTTP.
1641 2011-03-03 16:18:27 <gavinandresen> (I can dig out the links if you can't google it for yourself)
1642 2011-03-03 16:18:28 <Diablo-D3> gavinandresen: no, json-rpc 2.0 added naked over tcp
1643 2011-03-03 16:18:32 ApertureScience has joined
1644 2011-03-03 16:18:37 <Diablo-D3> 1.0/1.1 was always over http
1645 2011-03-03 16:18:46 <Diablo-D3> we're using 1.0, obviously
1646 2011-03-03 16:18:55 BCBot has joined
1647 2011-03-03 16:19:02 <Diablo-D3> 2.0 has convulted it to be true soap for json
1648 2011-03-03 16:19:23 ducki2p has quit (Remote host closed the connection)
1649 2011-03-03 16:19:33 <Diablo-D3> either way, what Im saying is, json over REST _can_ be websocketed with long push
1650 2011-03-03 16:19:44 <luke-jr> gavinandresen: persistent doesn't solve polling, unless it's bidirectional
1651 2011-03-03 16:19:47 ducki2p has joined
1652 2011-03-03 16:19:49 <Diablo-D3> browsers even support it fine
1653 2011-03-03 16:19:55 <Diablo-D3> well, browsers that do websockets anyway
1654 2011-03-03 16:20:10 <gavinandresen> Ok, I sit corrected.  I actually made bitcoin speak JSON 2.0... until I noticed that few of the JSON-RPC libraries supported it yet.
1655 2011-03-03 16:20:18 <rgm3> Is there a URL format for sharing a bitcoin address?  Something that would be appropriate for "donation" links, if the proper browser plugin was installed?
1656 2011-03-03 16:20:34 <gavinandresen> luke-jr: JSON-RPC 2.0 was designed to work well for bidirectional, persistent connections
1657 2011-03-03 16:20:38 <luke-jr> rgm3: yes
1658 2011-03-03 16:20:39 <rgm3> I suppose in the future ipv6 world the host and port portion of the url could point directly at a user's wallet proxy system
1659 2011-03-03 16:20:52 <rgm3> luke-jr: whassit
1660 2011-03-03 16:20:54 <luke-jr> rgm3: https://en.bitcoin.it/wiki/URI_Scheme
1661 2011-03-03 16:20:58 <rgm3> perfect!
1662 2011-03-03 16:21:13 <gavinandresen> BlueMatt: I never answered "who does the networking part of bitcoin" question.  Answer is:  Satoshi did.
1663 2011-03-03 16:21:14 <Diablo-D3> gavinandresen: json-rpc 2.0 is fffffffff
1664 2011-03-03 16:21:23 <Diablo-D3> I'd rather we dont move in that direction
1665 2011-03-03 16:21:32 tower is now known as towerX
1666 2011-03-03 16:21:33 <echelon> ewallet services need to allow users to set their own txfee
1667 2011-03-03 16:21:38 <Diablo-D3> standard xml/json techniques over websockets seems to be the future
1668 2011-03-03 16:21:44 <gavinandresen> I hate xml.
1669 2011-03-03 16:21:49 <BlueMatt> rgm3: can I ask what you are doing?
1670 2011-03-03 16:21:52 <luke-jr> gavinandresen: is that your preference for a new protocol, or just a "it's possible"?
1671 2011-03-03 16:22:03 <Diablo-D3> gavinandresen: well, you do realize its essentially the same shit, right?
1672 2011-03-03 16:22:08 <gavinandresen> New protocol for RPC?
1673 2011-03-03 16:22:19 <BlueMatt> gavinandresen: oh...
1674 2011-03-03 16:22:22 <gavinandresen> Diablo-D3:  yup.  Just wordier and with more dependencies
1675 2011-03-03 16:22:24 <luke-jr> gavinandresen: for communication between UIs and Wallet, or between Miner and Wallet
1676 2011-03-03 16:22:24 <Diablo-D3> gavinandresen: in any well written restful code impl I can have the same exact code produce both
1677 2011-03-03 16:22:36 <Diablo-D3> gavinandresen: especially with jersey/resteasy in java
1678 2011-03-03 16:22:53 <rgm3> luke-jr: You know how I know you wrote this?  it uses "TBC" as an optional part of the spec ;)
1679 2011-03-03 16:22:56 bk128 has joined
1680 2011-03-03 16:23:08 <gavinandresen> luke-jr:  I dunno, I figure the folks who've written miners would be the right people to decide that.
1681 2011-03-03 16:23:36 <luke-jr> rgm3: it gives examples in BTC and TBC, but the spec actually doesn't support Tonal
1682 2011-03-03 16:23:47 <gavinandresen> I don't really care what is used, as long as it doesn't add a whole bunch of dependencies to the build (bitcoin is already too hard to build on mac/linux/windows) and it is secure.
1683 2011-03-03 16:23:49 <luke-jr> rgm3: also, while I contributed, I can't take credit for it overall
1684 2011-03-03 16:24:03 <luke-jr> gavinandresen: it's not a problem just for mining ;)
1685 2011-03-03 16:24:22 <luke-jr> gavinandresen: it's been painful working on Spesmilo due to these things too
1686 2011-03-03 16:24:23 <mizerydearia> "the client lets you set it on the fly" -- How can I set the txfee in a running bitcoin daemon?
1687 2011-03-03 16:24:33 <luke-jr> mizerydearia: you can't afaik
1688 2011-03-03 16:24:36 <mizerydearia> must I restart the daemon each time I want to change the txfee amount?
1689 2011-03-03 16:24:45 <Diablo-D3> gavinandresen: but yeah, if all the sudden we switched to websocketed jsoned rest, shit would be golden
1690 2011-03-03 16:24:46 <gavinandresen> mizerydearia:  yes, that needs to be fixed.
1691 2011-03-03 16:24:58 <mizerydearia> "should make a request to add that into the api"
1692 2011-03-03 16:25:05 <gavinandresen> And actually, that is one of the MUCH MUCH more important things that needs to be worked out
1693 2011-03-03 16:25:09 <mizerydearia> hiya gavinandresen.  pm incoming
1694 2011-03-03 16:25:54 * luke-jr doesn't see how REST *can* be bidirectional
1695 2011-03-03 16:26:28 bk128 has quit (Client Quit)
1696 2011-03-03 16:26:32 <luke-jr> gavinandresen: IMO, I'd like to specify txfee with the actual send command. but that might be easier reserved for the new protocol
1697 2011-03-03 16:26:50 mmarker has joined
1698 2011-03-03 16:27:02 <luke-jr> since right now parameters are just an array
1699 2011-03-03 16:28:00 <mmarker> >.>
1700 2011-03-03 16:29:32 <gavinandresen> luke-jr:  I've been thinking about allowing RPC params to be passed as either array or object.  Object syntax would make it much easier to add extra params, and it should be easy to continue to support the old array syntax.
1701 2011-03-03 16:29:35 <mizerydearia> Yes, perhaps the sendfrom and sendtoaddress methods can provide an additional parameter that allows specifying the txfee on a per transaction basis
1702 2011-03-03 16:30:10 <luke-jr> gavinandresen: that'd be neat… but still doesn't address the other problems, so not sure if it's worth the time
1703 2011-03-03 16:30:26 <gavinandresen> what other problems?  only problem I'm aware of is polling issue.
1704 2011-03-03 16:30:49 <Diablo-D3> and the polling issue is easily fixed
1705 2011-03-03 16:30:57 <Diablo-D3> but everyones like WAAAAAAAAH I HAVE TO CODDDEEEEE
1706 2011-03-03 16:31:05 <luke-jr> that, and representing amounts as human BTC values instead of raw bitcoin amounts
1707 2011-03-03 16:31:10 <gavinandresen> ... and that could be solved either by a persistent connection to a port (downside is that won't work with Javascript in the browser) or HTTP callbacks.
1708 2011-03-03 16:31:20 <luke-jr> and possibly overhead of parsing JSON in general, dunno if that's a real problem
1709 2011-03-03 16:31:23 <Diablo-D3> gavinandresen: uh, hello?
1710 2011-03-03 16:31:28 <Diablo-D3> gavinandresen: websockets.
1711 2011-03-03 16:31:35 <Diablo-D3> browsers _support this shit_
1712 2011-03-03 16:31:36 <gavinandresen> luke-jr: I thought we just establised that is not an actual problem, but is purely theoretical
1713 2011-03-03 16:31:41 <luke-jr> Diablo-D3: WebSockets aren't supported by any browsers except Chrome
1714 2011-03-03 16:31:50 <Diablo-D3> luke-jr: except I use it everyday in firefox.
1715 2011-03-03 16:31:57 <Kiba> what is this purely theoretical problem you speak of, gavinandresen?
1716 2011-03-03 16:31:58 <Diablo-D3> and theres versions of safari shipping that have it
1717 2011-03-03 16:32:13 <Diablo-D3> and msie... 9? will have it
1718 2011-03-03 16:32:29 <gavinandresen> Kiba: money as floating-point values in the JSON-RPC api
1719 2011-03-03 16:32:39 <luke-jr> gavinandresen: human representation in protocols/internals is a design flaw no matter how well it works; JSON parsing, I don't know how problematic it is or isn't.
1720 2011-03-03 16:32:42 <Diablo-D3> Kiba: he says its theoretical, its not
1721 2011-03-03 16:32:58 <Diablo-D3> Kiba: money CANNOT be represented as floating point. even pocket change.
1722 2011-03-03 16:33:32 <Kiba> gavinandresen: I prefer percise over impercise
1723 2011-03-03 16:33:37 <lfm> luke-jr can you just force the money values to char mode and interpret them yourself?
1724 2011-03-03 16:33:48 <luke-jr> lfm: not with most libraries
1725 2011-03-03 16:33:48 <BlueMatt> If you guys hate the idea of floating point so much, make a patch and submit it?
1726 2011-03-03 16:33:50 <TD> it was already clearly explained that bitcoin money _can_ be represented as double precision floats
1727 2011-03-03 16:33:55 <gavinandresen> Sure it can.  I have $1.24999999999998  in my pocket, which any sane implementation knows is $1.25
1728 2011-03-03 16:33:56 <TD> now can we PLEASE move on !
1729 2011-03-03 16:34:01 <luke-jr> lfm: and it doesn't solve the fact that it's a *design flaw*, not an implementation problem
1730 2011-03-03 16:34:08 <lfm> luke-jr most? use one that does it?
1731 2011-03-03 16:34:08 <Diablo-D3> gavinandresen: see!
1732 2011-03-03 16:34:09 <luke-jr> BlueMatt: I did :p
1733 2011-03-03 16:34:11 <Diablo-D3> bad!
1734 2011-03-03 16:34:15 <Diablo-D3> bad gavin! bad!
1735 2011-03-03 16:34:19 <Kiba> gavinandresen: I think it's better to truncate it rather than round it up
1736 2011-03-03 16:34:28 <Diablo-D3> bad kiba!
1737 2011-03-03 16:34:28 <gavinandresen> (ugh, I said I was DONE wasting time arguing about that....)
1738 2011-03-03 16:34:28 <BlueMatt> luke-jr: where?
1739 2011-03-03 16:34:31 <luke-jr> lfm: I'm not interested in one-off protocols, I want something that multiple implementations can use sanely.
1740 2011-03-03 16:34:38 <luke-jr> BlueMatt: Gitorious branch
1741 2011-03-03 16:34:38 <Diablo-D3> either output it as a string, or use ubtc.
1742 2011-03-03 16:34:42 <Diablo-D3> there is your two solutions.
1743 2011-03-03 16:34:51 <gavinandresen> Kiba:  you can think that, and you can also think that the world is flat...
1744 2011-03-03 16:35:02 * Kiba shrugs
1745 2011-03-03 16:35:02 <BlueMatt> In any case, gavin said its not happening so I guess we are done now...?
1746 2011-03-03 16:35:25 <lfm> luke-jr but its not the protocol, its the api that is broken. just dont use the broken api, get another (or make it) for the same protocol
1747 2011-03-03 16:35:34 <luke-jr> it's easily solved by doing it right the first time in the new protocol that we need to fix other problems
1748 2011-03-03 16:35:49 <luke-jr> lfm: API = protocol
1749 2011-03-03 16:36:07 <gavinandresen> RE: websockets:   http://stackoverflow.com/questions/1253683/websocket-for-html5
1750 2011-03-03 16:36:18 <lfm> luke-jr ok now I agree with Diablo-D3 , you are a poor programmer if you think that
1751 2011-03-03 16:36:39 <luke-jr> re WebSockets, Firefox and Opera at least *removed* support
1752 2011-03-03 16:36:55 <Diablo-D3> chrome, firefox, opera, safari
1753 2011-03-03 16:36:57 <Diablo-D3> in other words
1754 2011-03-03 16:37:00 <Diablo-D3> about 75% of the internet
1755 2011-03-03 16:37:10 <TD> hmm, odd
1756 2011-03-03 16:37:10 <gavinandresen> I thought IE still had a big share
1757 2011-03-03 16:37:15 <TD> when i started my testnet client it seems to have missed a block
1758 2011-03-03 16:37:15 <luke-jr> lfm: in this case, bitcoin has no non-protocol API
1759 2011-03-03 16:37:22 <jrabbit> gavinandresen: depends on who you're talking about
1760 2011-03-03 16:37:24 <TD> now it's downloading lots of orphan blocks
1761 2011-03-03 16:37:25 <Diablo-D3> gavinandresen: I dont trust most of the numbers published
1762 2011-03-03 16:37:25 <TD> ProcessBlock: ORPHAN BLOCK, prev=0000000028b3af4b812b
1763 2011-03-03 16:37:30 sabalaba has quit (Ping timeout: 240 seconds)
1764 2011-03-03 16:37:31 <luke-jr> lfm: and only the protocol "API" has this problem: the internals don't.
1765 2011-03-03 16:37:42 <jrabbit> gavinandresen: if you're targeting the "internet" its safe to assume a good browser.
1766 2011-03-03 16:37:44 <TD> unless i somehow got onto the wrong testnet
1767 2011-03-03 16:37:52 <gavinandresen> I sit corrected, IE is down to 26% according to http://www.w3schools.com/browsers/browsers_stats.asp
1768 2011-03-03 16:37:53 <Diablo-D3> a lot of them are not MSIE either behind a proxy or with UA hiding
1769 2011-03-03 16:37:56 <lfm> luke-jr so fix the api, it just software. dont change the protocol and f--- up everyone
1770 2011-03-03 16:38:02 <jrabbit> gavinandresen: if you're targeting "people magazine readers" you can't assume such
1771 2011-03-03 16:38:04 * mmarker goes to shoot himself
1772 2011-03-03 16:38:10 <Diablo-D3> what, jesus? 26% now?
1773 2011-03-03 16:38:12 <Diablo-D3> holy crap
1774 2011-03-03 16:38:17 <luke-jr> lfm: the API isn't broken, just the protocol
1775 2011-03-03 16:38:21 <Diablo-D3> I thought it was still in the 40s
1776 2011-03-03 16:38:28 <jrabbit> 26% that has to be only one versino of IE
1777 2011-03-03 16:38:39 <Diablo-D3> but yeah, whatever the next major msie release is, it has websockets
1778 2011-03-03 16:38:42 <jrabbit> oh shit its not
1779 2011-03-03 16:38:48 <lfm> luke-jrI think you are mistaken, it is the api that uses floats, the protocol uses char strings
1780 2011-03-03 16:38:49 <luke-jr> http://hacks.mozilla.org/2010/12/websockets-disabled-in-firefox-4/
1781 2011-03-03 16:38:56 <jrabbit> no way thats not right.
1782 2011-03-03 16:39:01 <Diablo-D3> luke-jr: thats up to debate, atm. check the date.
1783 2011-03-03 16:39:07 <jrabbit> gavinandresen: that data looks skewed
1784 2011-03-03 16:39:10 <luke-jr> lfm: no, the bitcoind internals all use int64
1785 2011-03-03 16:39:15 <Diablo-D3> the data is partly skewed
1786 2011-03-03 16:39:17 <jrabbit> gavinandresen: look back in 2006/7
1787 2011-03-03 16:39:25 <Diablo-D3> most of the w3school visitors are webdevs
1788 2011-03-03 16:39:32 <jrabbit> firefox has a MASSIVE usage that was not right
1789 2011-03-03 16:39:33 <luke-jr> lfm: in any case, this is NOT the biggest problem with the current protocol
1790 2011-03-03 16:39:45 <gavinandresen> jrabbit:  ok.  Even if IE has 40% that's less than when I was working as a web developer
1791 2011-03-03 16:39:45 <luke-jr> lfm: fixing this problem still leaves much bigger problems
1792 2011-03-03 16:39:49 <jrabbit> i.e. this is baiased towards "thje internet"
1793 2011-03-03 16:40:04 <Diablo-D3> I just checked the stats on atpa and miotd
1794 2011-03-03 16:40:07 bk128 has joined
1795 2011-03-03 16:40:10 <jrabbit> gavinandresen: this is representational of people who know maybe what they're doing I think.
1796 2011-03-03 16:40:12 <Diablo-D3> msie is still between 35 and 40
1797 2011-03-03 16:40:13 <lfm> luke-jr seems like you dont understand what the term "Protocol" means. It is the definitions of what goes accross the wires between nodes
1798 2011-03-03 16:40:26 <Diablo-D3> which is basically "fuck them, they're not even close mattering"
1799 2011-03-03 16:40:26 <luke-jr> lfm: that's a *different* protocol
1800 2011-03-03 16:40:35 <jrabbit> enough that they use a technbical website (a bad one but still)
1801 2011-03-03 16:40:41 <lfm> luke-jr and it is in characters
1802 2011-03-03 16:40:54 <gavinandresen> RE: websockets:  using something as new as websockets to do something as critical as sending bitcoins gives me the heebie-jeebies
1803 2011-03-03 16:41:04 <luke-jr> lfm: I'm not talking about the p2p protocol, I'm talking about the Wallet<->Miner/UI protocol
1804 2011-03-03 16:41:26 <Diablo-D3> new?
1805 2011-03-03 16:41:29 <Diablo-D3> how is websockets new?
1806 2011-03-03 16:41:42 <lfm> luke-jryes and it is a tcp/ip protocl that at least can go over the wires
1807 2011-03-03 16:41:42 <Diablo-D3> go read the spec, its just http with additional shit
1808 2011-03-03 16:41:48 <Diablo-D3> I almost wanna just call it http 1.2
1809 2011-03-03 16:42:00 <jrabbit> native client bitcoin minign woudl be cool btw.
1810 2011-03-03 16:42:00 <Diablo-D3> not sure if it even earns the right to be called 2.0
1811 2011-03-03 16:42:14 <jrabbit> NaCl is pretty easy todo iirch
1812 2011-03-03 16:42:17 <gavinandresen> It is new because it hasn't been supported by all the major browsers for more than 5 years.
1813 2011-03-03 16:42:27 <gavinandresen> (I have a very conservative definition of "new")
1814 2011-03-03 16:42:39 <Diablo-D3> I mean, if we were all in browsers, we could just use an ajax framework here
1815 2011-03-03 16:42:48 <gavinandresen> Old == https and REST
1816 2011-03-03 16:42:59 <luke-jr> so does anyone have any constructive input for the new protocol?
1817 2011-03-03 16:43:01 <Diablo-D3> try websockets, try cometd, try comet, use a hidden iframe
1818 2011-03-03 16:43:12 <Diablo-D3> atm we're at the hidden iframe stage
1819 2011-03-03 16:43:21 <TD> hrm
1820 2011-03-03 16:43:26 <TD> somebody must be running a <.20 testnet client
1821 2011-03-03 16:43:28 <luke-jr> is JSON-RPC 2.0 the only suggestion for protocol basis?
1822 2011-03-03 16:43:33 <TD> that i keep connecting to
1823 2011-03-03 16:43:34 <lfm> luke-jr the tcp/ip packets themselves are CHARACTERS. just fix the api
1824 2011-03-03 16:43:40 <Diablo-D3> luke-jr: no, I just said json-rpc 2.0 is bad
1825 2011-03-03 16:43:46 <Diablo-D3> I do NOT want to implement that in ... anything
1826 2011-03-03 16:43:52 <Diablo-D3> json-rpc 1.0 is barely sane as it is
1827 2011-03-03 16:43:53 <TD> gavinandresen: do you have the IP address of a .20+ testnet node i can use directly? when i start my node currently it spends ages downloading orphan blocks
1828 2011-03-03 16:43:56 <TD> presumably the old chain
1829 2011-03-03 16:44:00 <luke-jr> Diablo-D3: then suggest an alternate
1830 2011-03-03 16:44:06 <Diablo-D3> moving to json rest is much easier
1831 2011-03-03 16:44:17 <luke-jr> Diablo-D3: REST is unidirectional.
1832 2011-03-03 16:44:26 <Diablo-D3> luke-jr: you fail at http forever.
1833 2011-03-03 16:44:27 <gavinandresen> TD:  the testnet faucet is running on    69.164.218.197
1834 2011-03-03 16:44:28 <luke-jr> and requires HTTP support
1835 2011-03-03 16:44:29 <BlueMatt> gavinandresen: What is the chance of getting the portoption merged into mainline for the next release?
1836 2011-03-03 16:44:31 <TD> thanks
1837 2011-03-03 16:44:38 <Diablo-D3> luke-jr: we're already using http you retard
1838 2011-03-03 16:44:43 <Diablo-D3> how the fuck do you think any of this works
1839 2011-03-03 16:45:13 <Diablo-D3> what the fuck is this, piss off diablo day?
1840 2011-03-03 16:45:22 <luke-jr> Diablo-D3: the existing JSON-RPC 1.0 implementation is not HTTP compliant. and irrelevant to a replacement protocol, mostly.
1841 2011-03-03 16:45:38 <Diablo-D3> luke-jr: no, bitcoind's http impl is not http compliant
1842 2011-03-03 16:45:54 <Diablo-D3> the json-rpc impl, afiact, actually is complaint
1843 2011-03-03 16:46:11 <[Tycho]> ;;bc,prob
1844 2011-03-03 16:46:12 <gribble> (bc,prob <an alias, at least 1 argument>) -- Alias for "math calc 1-exp(-$1*1000 * [seconds $*] / (2**32* [bc,diff]))".
1845 2011-03-03 16:46:24 <[Tycho]> ;;bc,prob 700
1846 2011-03-03 16:46:25 <gribble> Error: There's really no reason why you should have underscores or brackets in your mathematical expression.  Please remove them.
1847 2011-03-03 16:46:47 <[Tycho]> How i can use this command ?
1848 2011-03-03 16:46:49 <luke-jr> IIRC, slush and jgarzik's current conceptual mining-specific protocol is binary
1849 2011-03-03 16:46:51 <mmarker> Grr, something broke. Now my yasm code gives me garbage in gdb
1850 2011-03-03 16:47:04 <TD> there we go
1851 2011-03-03 16:47:06 <Diablo-D3> luke-jr: yes, and thats an extremely bad idea.
1852 2011-03-03 16:47:07 <TD> gavinandresen: thanks, that worked
1853 2011-03-03 16:47:11 <gavinandresen> BlueMatt:  I think it aught to be included.
1854 2011-03-03 16:47:12 <Diablo-D3> I refuse to support a binary protocol.
1855 2011-03-03 16:47:19 <[Tycho]> luke-jr, fear not, i have my own protocol proposal :)
1856 2011-03-03 16:47:28 <Diablo-D3> we're supposed to be increasing bitcoin adoption, not fucking it over.
1857 2011-03-03 16:47:34 <BlueMatt> gavinandresen: anything blocking it that I could work on...really want to see this
1858 2011-03-03 16:47:55 <[Tycho]> And, by the way, i'll support binary protocol, of course.
1859 2011-03-03 16:48:36 <[Tycho]> Binary is cool.
1860 2011-03-03 16:48:43 Lachesis has joined
1861 2011-03-03 16:49:00 <gavinandresen> BlueMatt:  worries about Sybil attacks are what is stopping it.
1862 2011-03-03 16:49:01 <[Tycho]> Don't like how those linuxoids stuff their human-readable ones everywhere.
1863 2011-03-03 16:49:16 <TD> gavinandresen: btw, i had a question about the new send buffer limits
1864 2011-03-03 16:49:31 <jgarzik> TD: ?
1865 2011-03-03 16:49:32 <TD> gavinandresen: my java implementation keeps hitting this. if i raise the buffer limit with a flag, it works ok again
1866 2011-03-03 16:49:41 <gavinandresen> TD: it'll have to wait, I have a lunch meeting I need to get to
1867 2011-03-03 16:49:45 <TD> no problem
1868 2011-03-03 16:49:50 <gavinandresen> See y'all later
1869 2011-03-03 16:49:51 <jgarzik> TD: which send-buffer limits?
1870 2011-03-03 16:50:02 gavinandresen has quit (Quit: gavinandresen)
1871 2011-03-03 16:50:04 <Lachesis> jgarzik, did ArtForz explain why he thought your poold was broken?
1872 2011-03-03 16:50:08 <TD> jgarzik: .20 now includes some anti-DoS mitigation where it disconnects nodes that it tries to send too much data too
1873 2011-03-03 16:50:12 <luke-jr> [Tycho]: why not provide input for a single spec?
1874 2011-03-03 16:50:17 <TD> i'm not totally sure what this is for. receive buffer limits, sure ...
1875 2011-03-03 16:50:19 <jgarzik> Lachesis: no
1876 2011-03-03 16:50:39 <Lachesis> jgarzik, alright - we were working through bitcoin's getwork() code and saw that it wasn't stateless
1877 2011-03-03 16:50:55 <Lachesis> so i asked him why pools weren't causing a memory leak
1878 2011-03-03 16:51:01 <Lachesis> and he thought pools didn't just call getwork
1879 2011-03-03 16:51:04 <Lachesis> so i showed him your code
1880 2011-03-03 16:51:20 <Lachesis> after some more looking, we figured out that it's storing refs
1881 2011-03-03 16:51:21 <luke-jr> [Tycho]: add your ideas to https://en.bitcoin.it/wiki/Wallet_protocol
1882 2011-03-03 16:51:28 <Lachesis> and they only change if a new txn is added to the block
1883 2011-03-03 16:51:35 <Diablo-D3> rgm3: hey, found the bug
1884 2011-03-03 16:51:42 <Lachesis> long story short - we were wrong; you did it right
1885 2011-03-03 16:51:44 <BlueMatt> ;;later tell gavinandresen in reference to the Sybil attacks and portoption: dont see why afaict, the 2h delay on non-standard ports is actually unnecessary as the outgoing connections per ip is already limited to 1 per /16 block
1886 2011-03-03 16:51:45 <gribble> The operation succeeded.
1887 2011-03-03 16:51:50 <rgm3> Diablo-D3: cool.
1888 2011-03-03 16:51:56 <Diablo-D3> rgm3: --user wouldnt have worked either.
1889 2011-03-03 16:51:59 <rgm3> i know :)
1890 2011-03-03 16:52:09 <Diablo-D3> no, I mean --user instead of -u
1891 2011-03-03 16:52:15 <rgm3> yep, i realize that
1892 2011-03-03 16:52:16 <Diablo-D3> I had an extra comma floating around
1893 2011-03-03 16:52:36 <Diablo-D3> now, otoh, "--user," would have worked.
1894 2011-03-03 16:52:51 <xelister> BlueMatt: oh, interesting
1895 2011-03-03 16:52:56 <xelister> BlueMatt: good catch
1896 2011-03-03 16:53:03 <rgm3> might be helpful to say "http://user:password@hostname:port/" in the help output too
1897 2011-03-03 16:53:26 <rgm3> also -- in the URL, is the port required?  it prolly oughta default to 8332 if not provided
1898 2011-03-03 16:53:36 <Diablo-D3> rgm3: its a standard url.
1899 2011-03-03 16:53:36 <[Tycho]> luke-jr, no, i'm talking about miner->pool protocol, not wallet's one.
1900 2011-03-03 16:53:57 <luke-jr> [Tycho]: pool is just a specialized wallet
1901 2011-03-03 16:54:16 <[Tycho]> "<luke-jr> [Tycho]: why not provide input for a single spec?" - what do you mean ?
1902 2011-03-03 16:54:41 <luke-jr> [Tycho]: https://en.bitcoin.it/wiki/Wallet_protocol is the foundation to design a specification for Wallet<->Miner/UI protocol
1903 2011-03-03 16:54:41 <Diablo-D3> rgm3: user, password, and port arent required. port defaults to 80 as required by http spec.
1904 2011-03-03 16:54:58 <Diablo-D3> rgm3: you could do http://whatever/url/shit/here, and it'd still be valid too
1905 2011-03-03 16:54:59 <rgm3> okay
1906 2011-03-03 16:55:41 skeledrew1 has joined
1907 2011-03-03 16:55:49 <[Tycho]> luke-jr, currently i just propose enhancing actual protocol with long polls. This will make miners ~1.8% more effective and requires adding just one more connection with same protocol.
1908 2011-03-03 16:56:44 bulletbill has joined
1909 2011-03-03 16:57:30 <jgarzik> Lachesis: thanks :)
1910 2011-03-03 16:57:36 greyface has quit (Read error: Connection reset by peer)
1911 2011-03-03 16:57:37 greyface has joined
1912 2011-03-03 16:57:37 da2ce7 has quit (Read error: Connection reset by peer)
1913 2011-03-03 16:57:37 da2ce7 has joined
1914 2011-03-03 16:57:38 ducki2p has quit (Ping timeout: 252 seconds)
1915 2011-03-03 16:57:38 da2ce7 has quit (Changing host)
1916 2011-03-03 16:57:39 da2ce7 has joined
1917 2011-03-03 16:57:42 <CIA-55> DiabloMiner: Patrick McFarland master * rbba5c36 / src/main/java/com/diablominer/DiabloMiner/DiabloMiner.java : Remove exactly one comma - http://bit.ly/goVZEa
1918 2011-03-03 16:57:49 skeledrew has quit (Ping timeout: 252 seconds)
1919 2011-03-03 16:58:02 ducki2p has joined
1920 2011-03-03 16:59:16 <jgarzik> TD: yes, indeed.  we don't want to queue too much data in RAM for each node.  that is an easy DoS.
1921 2011-03-03 16:59:44 <TD> the send DoS limits is trickier.
1922 2011-03-03 16:59:54 <TD> now i have to change my implementation to try and not ask the node too much, too fast
1923 2011-03-03 16:59:59 <TD> it could just throttle itself. hrmm.
1924 2011-03-03 17:00:32 Spenvo has joined
1925 2011-03-03 17:05:16 jrabbit has quit (Read error: Connection reset by peer)
1926 2011-03-03 17:05:16 <hwolf> in windows if you have several gpu miners connected to bitcoin.exe -server, how can you see your total khash/s?
1927 2011-03-03 17:05:16 keystrike has joined
1928 2011-03-03 17:05:16 skeledrew has joined
1929 2011-03-03 17:05:16 devon_hillard_ has joined
1930 2011-03-03 17:05:16 Netsniper has joined
1931 2011-03-03 17:05:16 <hwolf> running headless can you see your total khash/s?
1932 2011-03-03 17:05:16 <bulletbill> hwolf: no...not yet possible
1933 2011-03-03 17:05:16 <hwolf> k
1934 2011-03-03 17:05:16 <TD> what's with all the invalid transactions floating around
1935 2011-03-03 17:05:16 sabalaba has joined
1936 2011-03-03 17:05:16 <TD> this is more fallout from the fee bug?
1937 2011-03-03 17:05:16 <x6763> woohoo! my ram just came in...now maybe my machine won't start swapping all the time when i run eclipse and the android emulator
1938 2011-03-03 17:05:50 <bulletbill> TD: how are you monitoring this?
1939 2011-03-03 17:05:57 <TD> i see it in the logs
1940 2011-03-03 17:06:01 <bulletbill> debug.log
1941 2011-03-03 17:06:17 skeledrew1 has quit (Ping timeout: 264 seconds)
1942 2011-03-03 17:06:18  has quit (Netsniper|!~kvirc@76.251.230.92|Ping timeout: 264 seconds)
1943 2011-03-03 17:06:18 keystroke has quit (Ping timeout: 264 seconds)
1944 2011-03-03 17:06:18 RazielZ has quit (Ping timeout: 264 seconds)
1945 2011-03-03 17:06:18 brocktice has quit (Ping timeout: 264 seconds)
1946 2011-03-03 17:06:26 brocktice has joined
1947 2011-03-03 17:06:28 RazielZ has joined
1948 2011-03-03 17:06:58 <TD> yes
1949 2011-03-03 17:07:29 devon_hillard has quit (Ping timeout: 264 seconds)
1950 2011-03-03 17:08:21 `Jaka has joined
1951 2011-03-03 17:09:47 int0x27h has quit (Read error: Operation timed out)
1952 2011-03-03 17:09:49 Mango-chan has quit (Ping timeout: 240 seconds)
1953 2011-03-03 17:09:52 int0x27h_ has joined
1954 2011-03-03 17:10:02 jrabbit has joined
1955 2011-03-03 17:10:18 <bulletbill> TD: first time looking at log here. what does a typical invalid transaction entry look like?
1956 2011-03-03 17:10:37 RazielZ has quit (Ping timeout: 240 seconds)
1957 2011-03-03 17:10:44 RazielZ has joined
1958 2011-03-03 17:11:01 Animeking has quit (Ping timeout: 240 seconds)
1959 2011-03-03 17:11:42 slush has joined
1960 2011-03-03 17:13:11 <TD> ah never mind
1961 2011-03-03 17:13:18 <TD> they are probably not invalid. just out of order.
1962 2011-03-03 17:13:22 prax_ has joined
1963 2011-03-03 17:13:22 prax_ has quit (Changing host)
1964 2011-03-03 17:13:22 prax_ has joined
1965 2011-03-03 17:14:59 DoomDumasAlt has joined
1966 2011-03-03 17:14:59 DoomDumasAlt has quit (Changing host)
1967 2011-03-03 17:14:59 DoomDumasAlt has joined
1968 2011-03-03 17:14:59 DoomDumas has quit (Read error: Connection reset by peer)
1969 2011-03-03 17:15:17 <bulletbill> TD: ok. :) no problem.
1970 2011-03-03 17:15:22 <lfm> txn backlog seems to be building up again a little. 137 txn in memory, 29 in block
1971 2011-03-03 17:15:51 luke-jr has quit (Read error: Connection reset by peer)
1972 2011-03-03 17:16:09 <midnightmagic> what's the txn limit in a block again?
1973 2011-03-03 17:16:09 luke-jr has joined
1974 2011-03-03 17:16:28 <lfm> midnightmagic: something like 1MB
1975 2011-03-03 17:16:44 <TD> lfm: the clearance was one off
1976 2011-03-03 17:16:52 <TD> lfm: people will need to upgrade to a new client, to stop generating them
1977 2011-03-03 17:16:56 prax has quit (Ping timeout: 260 seconds)
1978 2011-03-03 17:17:03 <midnightmagic> so why is there a txn backlog then?
1979 2011-03-03 17:17:04 <TD> and by people i suspect i just mean  slush ....
1980 2011-03-03 17:17:41 <lfm> td so a bug was found in 0.3.20.1?
1981 2011-03-03 17:18:27 <slush> TD: some news?
1982 2011-03-03 17:20:30 <lfm> td clearance?
1983 2011-03-03 17:21:04 <luke-jr> https://en.bitcoin.it/wiki/Infrastructure
1984 2011-03-03 17:22:08 rli has left ()
1985 2011-03-03 17:22:42 <luke-jr> lfm: more like a missing feature: no way to include optional tx fees
1986 2011-03-03 17:24:42 <lfm> another feature missing (especially for slush) would be ui support for multi-output txn. He could pay all his miners with ONE txn/per hour
1987 2011-03-03 17:25:11 <luke-jr> lfm: slush isn't using a UI :p
1988 2011-03-03 17:25:24 <lfm> command line ui is a ui
1989 2011-03-03 17:25:35 <luke-jr> is he actually using CLI?
1990 2011-03-03 17:25:45 <luke-jr> I would think just direct JSON-RPC at least
1991 2011-03-03 17:26:03 <lfm> well support is missing there too
1992 2011-03-03 17:26:11 <luke-jr> yes
1993 2011-03-03 17:26:25 <luke-jr> that would be a benefit to having the new Wallet protocol be binary
1994 2011-03-03 17:26:35 <luke-jr> to allow the UI to craft its own custom tx, and just send it as-is
1995 2011-03-03 17:26:54 <lfm> binary? yuk
1996 2011-03-03 17:26:58 <TD> slush: i was referring to gavins fix
1997 2011-03-03 17:27:10 <slush> TD: oh, I have to look at it
1998 2011-03-03 17:27:57 <midnightmagic> bitcoind at git head doesn't want to daemonize
1999 2011-03-03 17:28:01 <TD> slush: http://www.bitcoin.org/smf/index.php?topic=4064.0
2000 2011-03-03 17:28:03 <bitbot> [PULL] Transaction backlog fix (part 1?) Connection refused.
2001 2011-03-03 17:28:15 <TD> posting forum links is magic?
2002 2011-03-03 17:28:26 <TD> http://www.bitcoin.org/smf/index.php?topic=4035.0
2003 2011-03-03 17:28:28 <bitbot> Coding guidelines Connection refused.
2004 2011-03-03 17:28:34 <TD> seems a bit broken
2005 2011-03-03 17:29:05 Travex has joined
2006 2011-03-03 17:29:14 <Travex> Hi guys
2007 2011-03-03 17:29:35 <Travex> Anybody can tell me how can I get an invitation for bitcoin please?
2008 2011-03-03 17:29:46 <mmarker> hmm
2009 2011-03-03 17:30:00 <luke-jr> Travex: wtf?
2010 2011-03-03 17:30:02 <TD> fascinating
2011 2011-03-03 17:30:16 Syke has quit (Ping timeout: 246 seconds)
2012 2011-03-03 17:30:49 <TD> Travex: where did you get the idea you need an invite?
2013 2011-03-03 17:30:54 <bulletbill> Travex: send me an email. there is a subscription fee
2014 2011-03-03 17:30:58 <bulletbill> :)
2015 2011-03-03 17:31:07 <lfm> Travex: you shouldnt need any invitation
2016 2011-03-03 17:31:15 <Travex> Well, since I tried to make an account on bitcoin.org
2017 2011-03-03 17:31:18 jostmey has joined
2018 2011-03-03 17:31:22 <Travex> but seem to be its closed
2019 2011-03-03 17:31:25 <Travex> the registration
2020 2011-03-03 17:31:30 <lfm> Travex: just check your email
2021 2011-03-03 17:31:35 <mmarker> *sigh*
2022 2011-03-03 17:31:50 <luke-jr> Travex: that's just for the forum
2023 2011-03-03 17:31:57 <mmarker> anyone have experiende debugging handwritten asm with gdb, it's getting lost in the source code again.
2024 2011-03-03 17:32:03 <midnightmagic> does anyone use the git merge repo?
2025 2011-03-03 17:32:17 Lobster_Man has joined
2026 2011-03-03 17:32:19 <Travex> @lfm : ive just checked  my email , nothing happen bud
2027 2011-03-03 17:32:21 <luke-jr> midnightmagic: git merge repo?
2028 2011-03-03 17:32:22 <midnightmagic> mmarker: I do but it's very time-consuming and I'm at work. :-(
2029 2011-03-03 17:32:32 <midnightmagic> luke-jr: yeah the staging one that gavin uses before committing to svn.
2030 2011-03-03 17:32:35 <luke-jr> Travex: you don't need to sign up, just download bitcoin client and run it
2031 2011-03-03 17:32:49 <luke-jr> midnightmagic: git has basically replaced svn
2032 2011-03-03 17:32:53 magnetro1 has joined
2033 2011-03-03 17:32:56 <Travex> oh i get it, but sorry for my noobish
2034 2011-03-03 17:32:59 DoomDumas has joined
2035 2011-03-03 17:32:59 DoomDumas has quit (Changing host)
2036 2011-03-03 17:32:59 DoomDumas has joined
2037 2011-03-03 17:33:10 <Travex> I want to use my GPU for bitcoin, but in every guide
2038 2011-03-03 17:33:12 magnetron has quit (Read error: Connection reset by peer)
2039 2011-03-03 17:33:13 <luke-jr> midnightmagic: you're welcome to use the Gitorious repo rather than GitHub tho
2040 2011-03-03 17:33:25 <Travex> it usually require an user name and password
2041 2011-03-03 17:33:25 <luke-jr> Travex: you need to setup your client for that
2042 2011-03-03 17:33:31 <Travex> for connecting to bitcoin ?
2043 2011-03-03 17:33:34 <luke-jr> yeah, you set that in your client
2044 2011-03-03 17:33:38 <lfm> Travex: ya forum is separate from bitcoin client and net itself
2045 2011-03-03 17:33:41 <luke-jr> run bitcoin -?
2046 2011-03-03 17:33:42 Syke has joined
2047 2011-03-03 17:33:44 <midnightmagic> luke-jr: this is partly why i'm using the staging repo to build my bitcoind, but the bitcoind as of a git pull of a few minutes ago doesn't want to daemonize; therefore, I'm asking to see if anyone else is successfully using recent git bitcoind so I can verify that for them -daemonize seems to work.
2048 2011-03-03 17:33:51 DoomDumasAlt has quit (Ping timeout: 260 seconds)
2049 2011-03-03 17:33:51 LobsterMan has quit (Ping timeout: 260 seconds)
2050 2011-03-03 17:33:57 jrabbit has quit (Ping timeout: 276 seconds)
2051 2011-03-03 17:33:57 <nanotube> Travex: try bitpenny.com mining pool.
2052 2011-03-03 17:33:58 <TD> Travex: you can just download and run the program
2053 2011-03-03 17:34:01 <midnightmagic> -daemon rather.
2054 2011-03-03 17:34:01 <mmarker> midnightmagic: Yea. What sucks is I think it was working last night...now notsomuch...and can't figure out what I broke.
2055 2011-03-03 17:34:03 <nanotube> Travex: that one is open to new registration.
2056 2011-03-03 17:34:04 <Travex> I created the forum acc yes
2057 2011-03-03 17:34:06 <TD> Travex: no need for a username/password to send/receive bitcoins
2058 2011-03-03 17:34:31 <lfm> Travex: download client from bitcoin.org homepage
2059 2011-03-03 17:34:35 t4ls0 has joined
2060 2011-03-03 17:34:35 ac_ has joined
2061 2011-03-03 17:34:36 <Travex> but think is, I want to use my GPU instead of CPU for bitcoin
2062 2011-03-03 17:34:41 <midnightmagic> mmarker: ever since i started using scm (in my case, perforce) i don't have moments like that anymore. :-( but I do wish you good luck in finding the solution.
2063 2011-03-03 17:34:53 <mmarker> hmm, I see what may be wrong
2064 2011-03-03 17:34:54 <lfm> Travex: ok gpu mining is more complex
2065 2011-03-03 17:34:59 <nanotube> Travex: the gpu miners have to connect to an instance of bitcoin, to receive work units.
2066 2011-03-03 17:35:07 <luke-jr> midnightmagic: just use &
2067 2011-03-03 17:35:09 <mmarker> the debugging info is messed up, so step in gdb doesn't know where the source lines are
2068 2011-03-03 17:35:11 <nanotube> Travex: so you have a choice: either run your own instance, or join a pool.
2069 2011-03-03 17:35:15 <mmarker> stepi to the rescue, I hope
2070 2011-03-03 17:35:20 <luke-jr> Travex: what GPU btw?
2071 2011-03-03 17:35:28 <nanotube> Travex: you of course would want your own instance so you can have your own wallet as well :) but that's a separate issue.
2072 2011-03-03 17:35:29 <Travex> 2xHD6970
2073 2011-03-03 17:35:31 <lfm> Travex: you still run bitcoin.exe just turn generation off
2074 2011-03-03 17:35:34 bk128 has quit (Read error: Connection reset by peer)
2075 2011-03-03 17:35:45 bk128 has joined
2076 2011-03-03 17:35:47 <midnightmagic> luke-jr: -daemon seems to be broken
2077 2011-03-03 17:36:03 <lfm> Travex: up to you if you want to join a pool or not
2078 2011-03-03 17:36:04 dukeleto- has joined
2079 2011-03-03 17:36:04 dukeleto- has quit (Excess Flood)
2080 2011-03-03 17:36:10 <luke-jr> midnightmagic: bisect it
2081 2011-03-03 17:36:20 <Travex> what is the benefit for joining a pool mate ?
2082 2011-03-03 17:36:22 frewsxcv_ has joined
2083 2011-03-03 17:36:22 frewsxcv_ has quit (Changing host)
2084 2011-03-03 17:36:22 frewsxcv_ has joined
2085 2011-03-03 17:36:22 <lfm> midnightmagic: -server?
2086 2011-03-03 17:36:23 saline has quit (Ping timeout: 260 seconds)
2087 2011-03-03 17:36:23 talso has quit (Ping timeout: 260 seconds)
2088 2011-03-03 17:36:23 HarryS has quit (Ping timeout: 260 seconds)
2089 2011-03-03 17:36:23 frewsxcv has quit (Ping timeout: 260 seconds)
2090 2011-03-03 17:36:24 dukeleto has quit (Ping timeout: 260 seconds)
2091 2011-03-03 17:36:30 <luke-jr> Travex: no benefit from the pool, with those GPUs
2092 2011-03-03 17:36:31 <Travex> so far as i know
2093 2011-03-03 17:36:37 <Travex> oh ,and why is that
2094 2011-03-03 17:36:40 HarryS has joined
2095 2011-03-03 17:36:41 <luke-jr> Travex: note, you will need to turn Crossfire OFF
2096 2011-03-03 17:36:47 <midnightmagic> lfm: -server is for answering json calls i think.
2097 2011-03-03 17:36:48 <luke-jr> Travex: pool is for slow miners
2098 2011-03-03 17:36:59 <lfm> Travex: your gpu are fast enuf you dont really need a pool
2099 2011-03-03 17:37:13 <Travex> oh I see
2100 2011-03-03 17:37:20 <lfm> midnightmagic: oh ok
2101 2011-03-03 17:37:23 frewsxcv_ is now known as frewsxcv
2102 2011-03-03 17:37:36 <Travex> but ok, let say I make my own instance, then how do I use my GPU for bitcoin then ?
2103 2011-03-03 17:37:40 <luke-jr> midnightmagic: when was the last version that you know it worked?
2104 2011-03-03 17:37:45 <Travex> is there any guide uotthere mates?
2105 2011-03-03 17:38:03 <luke-jr> Travex: run the bitcoin client with -rpcpassword=foo -server, and give the same password to your miner
2106 2011-03-03 17:38:08 <nanotube> luke-jr: 2x6970 is only about 600mhps no?
2107 2011-03-03 17:38:11 <nanotube> ;;bc,calc 600000
2108 2011-03-03 17:38:12 <gribble> The average time to generate a block at 600000 Khps, given current difficulty of 55590.23763914 , is 4 days, 14 hours, 32 minutes, and 10 seconds
2109 2011-03-03 17:38:19 <lfm> Travex: check forums for a gpu "miner" liek m0mchill or diablominer and install one
2110 2011-03-03 17:38:24 <luke-jr> nanotube: something like that
2111 2011-03-03 17:38:37 <nanotube> i could see there being benefit to a pool in this case. if you get unlucky, you could get nothing until diff goes up.
2112 2011-03-03 17:38:46 dukeleto has joined
2113 2011-03-03 17:38:46 dukeleto has quit (Excess Flood)
2114 2011-03-03 17:38:53 <Spenvo> Travex: If you want an in depth explanation of the bitcoin technology and its history and have some time, check out the Security Now podcast and skip ahead to about the 40 min mark: http://twit.tv/sn287
2115 2011-03-03 17:38:57 <luke-jr> nanotube: depends on luck, sure
2116 2011-03-03 17:39:03 <midnightmagic> luke-jr: I was working off one that was git pulled a few weeks ago, but i never actually stamped it or branched it so I don't know for sure.
2117 2011-03-03 17:39:06 <luke-jr> nanotube: but if you're unlucky, you might only get 35% from pool like I do
2118 2011-03-03 17:39:08 <midnightmagic> luke-jr: I'm looking into the code now..
2119 2011-03-03 17:39:21 <Travex> @Spenvo oh ok thx
2120 2011-03-03 17:39:26 <lfm> nanotube: on average works out the same
2121 2011-03-03 17:39:31 <luke-jr> midnightmagic: don't bother until you bisect
2122 2011-03-03 17:39:42 <Travex> @lfm so, what is the user name, or I just need the passworld oly ?
2123 2011-03-03 17:39:42 <Raulo> I also confirm broken -daemon in the git version
2124 2011-03-03 17:39:57 <midnightmagic> luke-jr: i'm not gonna bisect a running bitcoind and let my miners sit idle.
2125 2011-03-03 17:40:05 <luke-jr> :p
2126 2011-03-03 17:40:07 <lfm> Travex: user name? for forum? make your own
2127 2011-03-03 17:40:11 <luke-jr> midnightmagic: test with another one then
2128 2011-03-03 17:40:15 <amiller> midnightmagic, virtualbox it
2129 2011-03-03 17:40:18 <Travex> here i mean
2130 2011-03-03 17:40:23 <Travex> <luke-jr> Travex: run the bitcoin client with -rpcpassword=foo -server, and give the same password to your miner
2131 2011-03-03 17:40:24 <Travex> This
2132 2011-03-03 17:40:33 <luke-jr> Travex: use -rpcuser or smth
2133 2011-03-03 17:40:35 <midnightmagic> luke-jr: in half a moment if i can't find the answer by eyeballing it, I'll do that other stuff.
2134 2011-03-03 17:40:37 <luke-jr> try -? to get a list of otpions
2135 2011-03-03 17:40:43 <nanotube> lfm: right, on average :)
2136 2011-03-03 17:40:47 <lfm> Travex: that is make your own put it in the bitcoin.conf file
2137 2011-03-03 17:40:54 dukeleto has joined
2138 2011-03-03 17:40:54 dukeleto has quit (Excess Flood)
2139 2011-03-03 17:41:10 <slush> luke-jr stop spreading this fud, please. You made conclusion after two unlucky days and you are spreading it around again and again
2140 2011-03-03 17:41:24 <phantomcircuit> so i just realized something
2141 2011-03-03 17:41:26 <lfm> nanotube: ya i know, some people can't handle "on average"
2142 2011-03-03 17:41:29 <Travex> oh, I see guys, thanks for the geat helps
2143 2011-03-03 17:41:35 <Kiba> anybody would like to advertise on my signature space?
2144 2011-03-03 17:41:37 <Travex> Imma trying, still totally newb to this =D
2145 2011-03-03 17:41:45 <phantomcircuit> the average peer connection takes practically zero bandwidth
2146 2011-03-03 17:42:02 <luke-jr> slush: I said explicitly if he's unlucky
2147 2011-03-03 17:42:10 <phantomcircuit> it would be trivial to consume 100% of the available connection slots for connectible peers
2148 2011-03-03 17:42:16 <phantomcircuit> which would cause the network to fail
2149 2011-03-03 17:42:27 <lfm> phantomcircuit: well zero out of broadband. it might be significant if your still on a modem or something
2150 2011-03-03 17:42:30 dukeleto has joined
2151 2011-03-03 17:42:31 dukeleto has quit (Excess Flood)
2152 2011-03-03 17:42:46 <luke-jr> slush: besides, don't you WANT fewer people on your pool?
2153 2011-03-03 17:42:56 <phantomcircuit> lfm, i could probably do it from my home dsl :|
2154 2011-03-03 17:43:40 <xelister> when working for a pool,  my miner still can know if he calcuating the winning share (with correct diff)?
2155 2011-03-03 17:43:48 <xelister> of course?
2156 2011-03-03 17:43:50 <lfm> phantomcircuit: oh your talking about DOSing? ya I spoze. if you want to just target certain nodes or something
2157 2011-03-03 17:44:16 <luke-jr> xelister: English?
2158 2011-03-03 17:44:16 <phantomcircuit> lfm, no im saying i think i could take down the entire network by making it impossible to connect to another node
2159 2011-03-03 17:44:18 slush1 has joined
2160 2011-03-03 17:44:26 <luke-jr> [12:33:44] <luke-jr> slush: I said explicitly if he's unlucky
2161 2011-03-03 17:44:27 <luke-jr> [12:34:28] <luke-jr> slush: besides, don't you WANT fewer people on your pool?
2162 2011-03-03 17:44:44 <phantomcircuit> lfm, it would take a while to be effective because of already existing connections, but eventually i would have consumed all of the available connection slots
2163 2011-03-03 17:44:46 <slush1> luke-jr and if I remember well, you did 12 btc instead of 16, it is not 35% of expected reward
2164 2011-03-03 17:44:57 <lfm> phantomcircuit: maybe. why bother
2165 2011-03-03 17:45:01 <luke-jr> slush1: 35% of what I am making self-mining.
2166 2011-03-03 17:45:07 <slush1> luke-jr yes, I wanted, because I had problems with many connections
2167 2011-03-03 17:45:17 <luke-jr> slush1: which is 141% of expected
2168 2011-03-03 17:45:36 slush has quit (Ping timeout: 252 seconds)
2169 2011-03-03 17:45:36 ducki2p has quit (Ping timeout: 252 seconds)
2170 2011-03-03 17:45:37 <midnightmagic> also, tonal sucks.
2171 2011-03-03 17:45:38 <phantomcircuit> lfm, if i can force each node to be an island, i can setup a single peer that is connectable, and control the shape of the network from a single high bandwidth node
2172 2011-03-03 17:45:38 * midnightmagic ducks.
2173 2011-03-03 17:45:40 <slush1> luke-jr you cannot calculate it this style and you know that. If you have more than expected calculation, you are just temporary lucky
2174 2011-03-03 17:45:47 <phantomcircuit> lfm, specifically i could segment the network
2175 2011-03-03 17:45:59 ducki2p has joined
2176 2011-03-03 17:46:13 <luke-jr> slush1: I don't believe in luck.
2177 2011-03-03 17:46:16 <phantomcircuit> lfm, neat huh
2178 2011-03-03 17:46:25 <slush1> luke-jr ok, stop this discussion, I just don't like your statements, which looks differently if others don't know whole situation
2179 2011-03-03 17:46:30 <lfm> phantomcircuit: I spoze you could for a while. people might notice and take corrective measures
2180 2011-03-03 17:47:05 <phantomcircuit> lfm, sure but for that short period of time i would be able to block anybody else from producing a valid block chain
2181 2011-03-03 17:47:28 <[Tycho]> How can you not believe in luck when working on a pseudorandom task ? :)
2182 2011-03-03 17:47:42 <nanotube> phantomcircuit: that's a valid point - connection slots should enforce ip uniqueness, that'd help.
2183 2011-03-03 17:48:31 cosurg1 has joined
2184 2011-03-03 17:48:31 <luke-jr> slush1: do you disagree, that there is little point to someone with 500 MH/s using a pool?
2185 2011-03-03 17:48:45 DrEeevil has joined
2186 2011-03-03 17:48:57 <Syke> luke-jr, a pool is great if you only want to mine for a day or two
2187 2011-03-03 17:48:58 <slush1> luke-jr I'm using pool even with 2.5 ghash
2188 2011-03-03 17:48:58 <mmarker> Jesus. I need help. Again, uninitialized registers usually contain garbage. I need to remember this
2189 2011-03-03 17:48:59 frewsxcv_ has joined
2190 2011-03-03 17:48:59 frewsxcv_ has quit (Changing host)
2191 2011-03-03 17:48:59 frewsxcv_ has joined
2192 2011-03-03 17:49:00 <phantomcircuit> nanotube, yeah
2193 2011-03-03 17:49:03 cosurgi has quit (Ping timeout: 240 seconds)
2194 2011-03-03 17:49:04 JStoker has quit (Ping timeout: 240 seconds)
2195 2011-03-03 17:49:04 bonsaikitten has quit (Read error: Connection reset by peer)
2196 2011-03-03 17:49:04 frewsxcv has quit (Ping timeout: 240 seconds)
2197 2011-03-03 17:49:06 dukeleto has joined
2198 2011-03-03 17:49:07 dukeleto has quit (Excess Flood)
2199 2011-03-03 17:49:13 KuT-Sickness has quit (Ping timeout: 240 seconds)
2200 2011-03-03 17:49:24 <lfm> luke-jr true in fact those sorts of nodes tend to clog up a pool with too frequent getwork calls
2201 2011-03-03 17:49:30 <Travex> Ok guys
2202 2011-03-03 17:49:31 JStoker has joined
2203 2011-03-03 17:49:31 <Travex> So now
2204 2011-03-03 17:49:42 <Travex> my Hd6970 is making
2205 2011-03-03 17:49:46 <Syke> any long-term mining should not be using a pool
2206 2011-03-03 17:49:51 <Travex> 310000khash/s
2207 2011-03-03 17:49:59 <Travex> so, how do i know
2208 2011-03-03 17:49:59 <luke-jr> slush1: if you don't mind my asking: why? :p
2209 2011-03-03 17:50:08 <luke-jr> ;;bc,calc 310000
2210 2011-03-03 17:50:09 <gribble> The average time to generate a block at 310000 Khps, given current difficulty of 55590.23763914 , is 1 week, 1 day, 21 hours, 56 minutes, and 27 seconds
2211 2011-03-03 17:50:10 <Syke> Travex, grats!
2212 2011-03-03 17:50:10 <lfm> Travex: wtg, now you wait. could be weeks still
2213 2011-03-03 17:50:10 <Travex> when my first coin will be generated ?
2214 2011-03-03 17:50:19 <Travex> ehhh
2215 2011-03-03 17:50:22 <mmarker> Travex: between now and never
2216 2011-03-03 17:50:23 <nanotube> Travex: coins are generated in 50-block chunks
2217 2011-03-03 17:50:23 <luke-jr> Travex: probably within 8 days
2218 2011-03-03 17:50:25 <Travex> so can i resume this process ?
2219 2011-03-03 17:50:29 <mmarker> No
2220 2011-03-03 17:50:29 <nanotube> ;;bc,calc 300000
2221 2011-03-03 17:50:30 <gribble> The average time to generate a block at 300000 Khps, given current difficulty of 55590.23763914 , is 1 week, 2 days, 5 hours, 4 minutes, and 20 seconds
2222 2011-03-03 17:50:32 <nanotube> Travex: --^
2223 2011-03-03 17:50:33 <mmarker> Wekk, yea
2224 2011-03-03 17:50:34 <luke-jr> Travex: it isn't a process
2225 2011-03-03 17:50:37 <luke-jr> Travex: it's a chance
2226 2011-03-03 17:50:39 <Travex> wow
2227 2011-03-03 17:50:42 dukeleto has joined
2228 2011-03-03 17:50:42 dukeleto has quit (Excess Flood)
2229 2011-03-03 17:50:44 <mmarker> you're playing the lottery
2230 2011-03-03 17:50:44 <Travex> I see
2231 2011-03-03 17:50:52 <mmarker> The faster you go, the better the chance
2232 2011-03-03 17:50:54 <Travex> So i have to keep my computer up for a week
2233 2011-03-03 17:50:55 <nanotube> Travex: hence the suggestion to join a pool, if you want smoother income distribution :)
2234 2011-03-03 17:50:58 <Travex> without any interuption ?
2235 2011-03-03 17:51:01 <nanotube> Travex: no, you can keep it up and down
2236 2011-03-03 17:51:03 <lfm> Travex: run it whenever you can, stop it whenever you wish but if you can leave it running is better for your chances
2237 2011-03-03 17:51:10 <[Tycho]> Pool gives maximum average income between difficulty steps.
2238 2011-03-03 17:51:12 <nanotube> Travex: there's no such thing as "resuming", every hash is standalone.
2239 2011-03-03 17:51:14 dukeleto has joined
2240 2011-03-03 17:51:14 dukeleto has quit (Excess Flood)
2241 2011-03-03 17:51:19 <Travex> Oh I see
2242 2011-03-03 17:51:37 <Travex> Because I dont see any indication in bitcoin client
2243 2011-03-03 17:51:40 <Travex> just the khash number running
2244 2011-03-03 17:51:41 <mmarker> Ok, message expansion looks good
2245 2011-03-03 17:51:44 <Travex> so its a bit, werid !
2246 2011-03-03 17:51:52 <mmarker> and SHA-256 runs SLOW at -O0
2247 2011-03-03 17:51:56 <slush1> luke-jr because I'm not a gambler. I like the feeling that my 100 BTC come to my wallet more or little at the same time as teoretic calculations said.
2248 2011-03-03 17:51:57 <lfm> Travex: each "hash" is a separate run. you are doing millions of hashes per sec
2249 2011-03-03 17:52:02 <luke-jr> Travex: personally, I leave mine mining 24/7 minus when I play PS2 games
2250 2011-03-03 17:52:04 <Travex> unlike when I use compute4cash
2251 2011-03-03 17:52:15 <slush1> luke-jr but it is completely about personality and I understand that somebody with 500 mhash/s don't want to use pool
2252 2011-03-03 17:52:23 <slush1> and like the feeling of expectation
2253 2011-03-03 17:52:39 <[Tycho]> 500 MH/s is way too slow for solo mining.
2254 2011-03-03 17:53:03 <lfm> iiiieeee ya compuet4cash is kinda a rip-off if you can just set it up for yourself like you just did
2255 2011-03-03 17:53:07 <[Tycho]> I would mine solo only if faster than 10 GH/s :)
2256 2011-03-03 17:53:09 <Syke> in the long run, anyone with a decent miner will get more coins mining solo.
2257 2011-03-03 17:53:35 <[Tycho]> Syke, it heavily depends on luck.
2258 2011-03-03 17:53:43 <slush1> Syke: this does not be necessary true, if you calculate mised coins because of dificulty increase
2259 2011-03-03 17:54:37 <[Tycho]> Syke, with current difficulty you have chance to find a small number of blocks until next step.
2260 2011-03-03 17:54:39 <nanotube> Travex: compute4cash is exactly bitcoin mining... just with a different UI.
2261 2011-03-03 17:54:53 Netsniper has quit (Read error: Connection reset by peer)
2262 2011-03-03 17:55:18 Netsniper has joined
2263 2011-03-03 17:55:39 DrEeevil is now known as AmazingPudding
2264 2011-03-03 17:55:43 <Travex> well, ok so everything I can do now is , just wai t?
2265 2011-03-03 17:55:43 <Travex> wait ?
2266 2011-03-03 17:55:45 magnetro1 has quit (Read error: Operation timed out)
2267 2011-03-03 17:55:45 <Travex> Guys  :D ?
2268 2011-03-03 17:55:45 KuT-Sickness has joined
2269 2011-03-03 17:55:46 magnetron has joined
2270 2011-03-03 17:55:46 <Travex> yeah, but at least it provides an indicator
2271 2011-03-03 17:55:46 <Travex> =D
2272 2011-03-03 17:55:46 dukeleto has joined
2273 2011-03-03 17:55:46 <Travex> like how many WU youve dont
2274 2011-03-03 17:55:46 <Travex> something like that
2275 2011-03-03 17:55:46 <Travex> done*
2276 2011-03-03 17:55:46 jrabbit has joined
2277 2011-03-03 17:55:47 <luke-jr> slush1: where is the protocol stuff you and jgarzik were working on btw?
2278 2011-03-03 17:55:47 jrabbit has quit (Changing host)
2279 2011-03-03 17:55:47 jrabbit has joined
2280 2011-03-03 17:55:50 <lfm> with pool you have better chance to find smaller number of coins
2281 2011-03-03 17:56:02 <nanotube> Travex: if you want indicators, join a pool.
2282 2011-03-03 17:56:03 <Syke> slush1, there are no "missed coins" mining solo
2283 2011-03-03 17:56:09 <mmarker> Yea. I'd NEVER find a block with what I'm using
2284 2011-03-03 17:56:21 <mmarker> but running it 24/7 in a pool...lit's a teeny trickle, but it's there
2285 2011-03-03 17:56:27 <Syke> you could just as easily solo mine double the expected rate. on the long term, solo mining will generate more
2286 2011-03-03 17:56:35 <Travex> Oh nano, which pool then ?
2287 2011-03-03 17:56:37 <slush1> luke-jr this week I'm working on my regular job project, I had to finish it before next work on the pool
2288 2011-03-03 17:56:44 <slush1> so no progress this week
2289 2011-03-03 17:56:45 <lfm> mmarker: solo you always have a chance, maybe small chance
2290 2011-03-03 17:56:55 <BlueMatt> would someone mind building from https://github.com/TheBlueMatt/bitcoin with the default port set to 9635 and opening the client with -addnode=87.155.89.184
2291 2011-03-03 17:56:58 <BlueMatt> to test?
2292 2011-03-03 17:57:19 <Syke> [Tycho], no, it doesn't depend on luck. It depends on randomness.
2293 2011-03-03 17:57:24 <mmarker> lfm: With my army of ARM boxen, try next to none at the current level. But yea, there's a chance
2294 2011-03-03 17:57:30 <Syke> and over time, the randomness evens out
2295 2011-03-03 17:57:53 <amiller> i can't wait till someone starts selling pluggable sha256 asics
2296 2011-03-03 17:57:54 <bk128> luck is randomness in your favor.....
2297 2011-03-03 17:58:09 <lfm> Syke: thats just a point of view
2298 2011-03-03 17:58:36 slush1 is now known as slush
2299 2011-03-03 18:01:01 [Tycho] has quit (Read error: Connection reset by peer)
2300 2011-03-03 18:01:43 [Tycho] has joined
2301 2011-03-03 18:02:10 dukeleto has joined
2302 2011-03-03 18:02:42 Diablo-D3 has quit (Ping timeout: 264 seconds)
2303 2011-03-03 18:02:44 <mmarker> hmm, can gribble calculate how fast you're hashing based on how many coins youre making a day?
2304 2011-03-03 18:02:44 <bk128> luck is randomness
2305 2011-03-03 18:02:44 <Syke> it's math, not a point of view
2306 2011-03-03 18:02:44 Kiba has quit (Ping timeout: 250 seconds)
2307 2011-03-03 18:02:44 <lfm> mmarker: try bc,gen?
2308 2011-03-03 18:02:44 <bk128> ;;ping
2309 2011-03-03 18:02:44 <gribble> pong
2310 2011-03-03 18:02:44 <lfm> ;;bc,gen 320000
2311 2011-03-03 18:02:44 <gribble> The expected generation output, at 320000 Khps, given current difficulty of 55590.23763914 , is 5.78995693232 BTC per day and 0.241248205513 BTC per hour.
2312 2011-03-03 18:02:46 <mmarker> ;;bc,gen 3200
2313 2011-03-03 18:02:47 <gribble> The expected generation output, at 3200 Khps, given current difficulty of 55590.23763914 , is 0.0578995693232 BTC per day and 0.00241248205513 BTC per hour.
2314 2011-03-03 18:02:47 <Syke> solo mining doesn't waste time due to latency, doesn't waste time due to operator fees, etc.
2315 2011-03-03 18:02:53 <mmarker> ;;bc,gen 320
2316 2011-03-03 18:02:53 <gribble> The expected generation output, at 320 Khps, given current difficulty of 55590.23763914 , is 0.00578995693232 BTC per day and 0.000241248205513 BTC per hour.
2317 2011-03-03 18:02:53 <mmarker> ;;bc,gen 600
2318 2011-03-03 18:02:53 <amiller> ;;bc,gen 13000
2319 2011-03-03 18:02:53 <gribble> The expected generation output, at 600 Khps, given current difficulty of 55590.23763914 , is 0.0108561692481 BTC per day and 0.000452340385337 BTC per hour.
2320 2011-03-03 18:02:53 <gribble> The expected generation output, at 13000 Khps, given current difficulty of 55590.23763914 , is 0.235217000375 BTC per day and 0.00980070834898 BTC per hour.
2321 2011-03-03 18:02:53 <mmarker> there. My total hash rate is ~600khash/sec
2322 2011-03-03 18:02:53 <amiller> this is a lot different than it was before....
2323 2011-03-03 18:02:53 <amiller> has something changed?
2324 2011-03-03 18:02:54 <midnightmagic> it's math which depends on the presumed random distribution of sha hashes
2325 2011-03-03 18:02:54 <lfm> Travex: ya, with "luck" you should see a block every few days, each block is 50 bitcoins (worth almost $50US)
2326 2011-03-03 18:02:54 <bk128> pools reduce risk at a cost.  it's pretty simple
2327 2011-03-03 18:02:54 <Syke> hehe, feel free to show sha one not being pseudo-random
2328 2011-03-03 18:02:54 <Syke> and how a pool improves it
2329 2011-03-03 18:02:54 <Travex> I see
2330 2011-03-03 18:02:54 <midnightmagic> yeah, i know. i'm just saying.
2331 2011-03-03 18:02:54 <Travex> How about the difficulty factor ?
2332 2011-03-03 18:02:55 <lfm> Travex: assuming you leav it running 24/7
2333 2011-03-03 18:02:55 <midnightmagic> it's not proven.
2334 2011-03-03 18:02:55 <amiller> midnightmagic, the randomness of sha256 is directly related to the security of SSL and a number of other important technologies that are relied on heavily by the rest of the world
2335 2011-03-03 18:02:55 <bk128> Syke: are you more likely to win a scratch off card or the mega millions lottery?
2336 2011-03-03 18:02:55 <amiller> so while it hasn't been proven, if turns out not to be true then we have many bigger problems!
2337 2011-03-03 18:02:56 <lfm> Travex: the difficulty tends to go up over time making it harder for each miner to find blocks but it keeps the overall net mining at a constant rate of 10 blocks per hour or so
2338 2011-03-03 18:02:56 <midnightmagic> amiller: yes, but it's not proven. that's why crypto people are sitll trying to break it.
2339 2011-03-03 18:02:56 <Syke> bk128, can't answer that without knowing the payouts/odds
2340 2011-03-03 18:02:56 <Travex> alright, last question
2341 2011-03-03 18:02:56 [Tycho] has quit (Changing host)
2342 2011-03-03 18:02:56 [Tycho] has joined
2343 2011-03-03 18:02:56 <Travex> Can I combine my GPU with CPU ?
2344 2011-03-03 18:02:56 <Travex> for running bitcoin
2345 2011-03-03 18:02:59 <amiller> Travex, yes!
2346 2011-03-03 18:03:09 <Syke> Travex, yes, but not worth it
2347 2011-03-03 18:03:14 <bk128> Syke: do you agree that pools increase your odds of getting coins?
2348 2011-03-03 18:03:15 Diablo-D3 has joined
2349 2011-03-03 18:03:20 <Travex> Because the GPU is much faster ?
2350 2011-03-03 18:03:20 <Syke> bk128, no
2351 2011-03-03 18:03:27 <Syke> Travex, yes
2352 2011-03-03 18:03:31 <Syke> 100 times or so
2353 2011-03-03 18:03:36 <lfm> Travex: ya, you can run both but most people do not find the cpu mining to be profitable compared to the extra power cost
2354 2011-03-03 18:03:36 <amiller> Travex, it's also much more power efficient
2355 2011-03-03 18:03:37 Zarutian has joined
2356 2011-03-03 18:03:37 <Travex> well, yeah sadly
2357 2011-03-03 18:03:55 <Travex> well cause im having electricity-free
2358 2011-03-03 18:03:55 <amiller> Travex, if you pay your own power bills, GPUs may be worthwhile but CPU is not
2359 2011-03-03 18:03:56 <bk128> Syke: can you explain why you think being in a pool doesn't increase your odds of getting coins?
2360 2011-03-03 18:04:11 <Travex> yeah, but still wanna try with my CPU =D
2361 2011-03-03 18:04:17 <Travex> how am i gonna do that
2362 2011-03-03 18:04:40 <Syke> bk128, because a pool does exactly the same thing, but with latency/fees/downtime, the payout is always less than 100%.
2363 2011-03-03 18:04:42 <lfm> Travex: also the cpu is about 100x slower so you might not want the extra heat involved
2364 2011-03-03 18:04:54 <Travex> yeah, youre right
2365 2011-03-03 18:05:10 <nanotube> Travex: try jgarzik's cpu miner if you want to play with your cpu.
2366 2011-03-03 18:05:12 <Travex> at full load my 2600k heats up at 75 degree
2367 2011-03-03 18:05:13 <Travex> lame
2368 2011-03-03 18:05:17 <bk128> Syke: i asked if it increases your odds of getting any coins
2369 2011-03-03 18:05:32 <Travex> thanks nano , im using momchild gpu miner
2370 2011-03-03 18:05:36 <Syke> "any" being non-zero?
2371 2011-03-03 18:05:42 <Travex> lol this is like playing lottery eh ?
2372 2011-03-03 18:05:51 <bk128> maybe it's a bad analogy
2373 2011-03-03 18:05:57 <Syke> bk128, at what hash-rate and in what time-frame?
2374 2011-03-03 18:05:59 <bk128> but there are some parallels
2375 2011-03-03 18:06:00 <amiller> Travex, not at all - the expected value of the lottery is a loss of money
2376 2011-03-03 18:06:04 <lfm> Travex: the other effect is if you use your system for other things you might notice some slowdown so thats up to you
2377 2011-03-03 18:06:11 <amiller> Travex, the expected value of bitcoins is a positive gain
2378 2011-03-03 18:06:12 <Syke> I'm not talking about cpu-mining for 1 hour.
2379 2011-03-03 18:06:24 <bk128> Syke: any hashrate any timeframe.
2380 2011-03-03 18:06:34 <Syke> then no.
2381 2011-03-03 18:06:40 <Travex> and why is that
2382 2011-03-03 18:06:42 <Syke> not long-term with a decent miner
2383 2011-03-03 18:06:52 <Travex> lets say im unlucky
2384 2011-03-03 18:06:59 <Travex> and I have to wait for
2385 2011-03-03 18:07:00 <Travex> 28 days
2386 2011-03-03 18:07:03 <[Noodles]> decent miner != any hashrate
2387 2011-03-03 18:07:04 <Travex> to get my first 50 bucks
2388 2011-03-03 18:07:10 bulletbill has joined
2389 2011-03-03 18:07:15 <Travex> but since, I have to pay 100 bucks for my electrictity bill
2390 2011-03-03 18:07:26 <Travex> therefore have minus 50 bucks
2391 2011-03-03 18:07:29 <amiller> Travex, if you're unlucky then you might not get what you expect - but on average you'll win, that's the opposite of a lottery
2392 2011-03-03 18:07:40 <Travex> yeah, youre right
2393 2011-03-03 18:07:54 <amiller> Travex, if you want to reduce your risk and guarantee yourself that you'll get closer to the 'average value', then you can join a pool
2394 2011-03-03 18:07:54 <lfm> Travex: should be quite unlikely, you might alos get 400 btc so it averages out theoreticlly
2395 2011-03-03 18:07:54 <amiller> suppose there was a pool for lotteries
2396 2011-03-03 18:07:55 <Travex> lottery is negative skewness
2397 2011-03-03 18:08:26 <bk128> Travex: so is cpu mining :)
2398 2011-03-03 18:08:38 <Travex> ha, your in your opinions, joining a pool is a "better" option ?
2399 2011-03-03 18:08:44 <Travex> or I just solo
2400 2011-03-03 18:08:44 <lfm> no
2401 2011-03-03 18:08:54 <lfm> Travex: with your gpu just solo
2402 2011-03-03 18:08:54 <bk128> Travex: but actually, bitcoin mining is like buying a billion lotto tickets a second
2403 2011-03-03 18:09:05 <amiller> Travex, that depends on your 'tolerance for risk', it's like stocks vs mutual funds
2404 2011-03-03 18:09:09 <midnightmagic> long-term or if you have a few gpu, solo mining is a superior option from any angle you want to look at it.
2405 2011-03-03 18:09:21 <Syke> Travex, how long do you expect to mine? 1 week? 1 year?
2406 2011-03-03 18:09:22 EvanR has joined
2407 2011-03-03 18:09:22 genjix has joined
2408 2011-03-03 18:09:22 <Travex> Well, some how Im a risk lover
2409 2011-03-03 18:09:24 <amiller> bk128, terrible analogy - the more lottery tickets you buy, the poorer you become!
2410 2011-03-03 18:09:26 <lfm> bk1 ya and "drawrs 10 times per hour
2411 2011-03-03 18:09:31 <Travex> lets say one week ?
2412 2011-03-03 18:09:34 <Travex> or 2
2413 2011-03-03 18:09:34 <lfm> bk1 ya and "draws" 10 times per hour
2414 2011-03-03 18:09:38 <amiller> Travex, if you want the risk, mine solo
2415 2011-03-03 18:09:39 <bk128> amiller: the more hashes you do, the more you spend on power
2416 2011-03-03 18:09:49 <Syke> solo = more risk and more payout
2417 2011-03-03 18:09:50 <Travex> alrighty, I get it
2418 2011-03-03 18:09:52 <bk128> each hash and each ticket is a chance to win
2419 2011-03-03 18:10:06 <Travex> but in the worse case scenario
2420 2011-03-03 18:10:14 <JFK911> ;;bc,stat
2421 2011-03-03 18:10:14 <gribble> Error: "bc,stat" is not a valid command.
2422 2011-03-03 18:10:18 <JFK911> ;;bc,statw
2423 2011-03-03 18:10:18 <gribble> Error: "bc,statw" is not a valid command.
2424 2011-03-03 18:10:19 <JFK911> ;;bc,stats
2425 2011-03-03 18:10:21 <gribble> Current Blocks: 111586 | Current Difficulty: 55590.23763914 | Next Difficulty At Block: 112895 | Next Difficulty In: 1309 blocks | Next Difficulty In About: 1 week, 1 day, 0 hours, 42 minutes, and 50 seconds | Next Difficulty Estimate: 63068.49968428
2426 2011-03-03 18:10:25 <bk128> Syke: yes, that's what I've been saying.  more risk solo mining
2427 2011-03-03 18:10:32 <Travex> I ccan still make my first 50 coins in about 28 days average right ?
2428 2011-03-03 18:10:33 <JFK911> 63000 xD
2429 2011-03-03 18:10:48 <lfm> Travex: whats you hash/sec again?
2430 2011-03-03 18:10:55 <Syke> yes, and the longer you wait, and the better your miner is, the risk is very low.
2431 2011-03-03 18:11:05 <amiller> Travex, yes, and keep checking with gribble to see how your rate changes
2432 2011-03-03 18:11:06 <amiller> Travex, have you seen the difficulty-over-time graph?
2433 2011-03-03 18:11:11 <Travex> with overclocking, its about 400000/s now
2434 2011-03-03 18:11:24 <Travex> Nope, i have no idea ;(
2435 2011-03-03 18:11:30 <Travex> havent joined any pool
2436 2011-03-03 18:11:32 <Travex> indeed
2437 2011-03-03 18:11:48 <lfm> Travex: ok use "/msg gribble bc,calc 400000"
2438 2011-03-03 18:11:58 sgornick has quit (Quit: Leaving.)
2439 2011-03-03 18:12:16 <bk128> Syke: my computers have only found 3 blocks but have been paid 160 btc by slush's pool.  I'm unlucky
2440 2011-03-03 18:12:20 <Travex> you mean using bitcoin client ?
2441 2011-03-03 18:12:29 <Travex> ah
2442 2011-03-03 18:12:30 <lfm> Travex: no here in irc
2443 2011-03-03 18:12:31 <Travex> sry my bad
2444 2011-03-03 18:12:33 <bk128> Syke: only been in it a week though
2445 2011-03-03 18:12:34 <Travex> sry ;D
2446 2011-03-03 18:12:35 <Travex> got it
2447 2011-03-03 18:12:40 <Travex> done
2448 2011-03-03 18:12:55 <Syke> bk128, too short a time frame
2449 2011-03-03 18:13:03 <bk128> I know
2450 2011-03-03 18:13:40 <Travex> 6 days, 21 hours, 48 minutes, and 15 seconds
2451 2011-03-03 18:13:45 <Travex> lewl =D
2452 2011-03-03 18:13:48 <Travex> with 24/7
2453 2011-03-03 18:13:50 <Travex> I suppose ?
2454 2011-03-03 18:13:59 <Syke> bk128, that's not unlucky, that's just being impatient
2455 2011-03-03 18:14:15 <lfm> ya at 24/7
2456 2011-03-03 18:14:24 larsivi has quit (Remote host closed the connection)
2457 2011-03-03 18:14:35 larsivi has joined
2458 2011-03-03 18:14:40 citiz3n has joined
2459 2011-03-03 18:14:40 citiz3n has quit (Changing host)
2460 2011-03-03 18:14:40 citiz3n has joined
2461 2011-03-03 18:15:02 <bk128> Syke: at what point is it unlucky?
2462 2011-03-03 18:15:27 greyface has quit (Ping timeout: 252 seconds)
2463 2011-03-03 18:15:27 gasteve has quit (Ping timeout: 252 seconds)
2464 2011-03-03 18:15:28 ducki2p has quit (Ping timeout: 252 seconds)
2465 2011-03-03 18:15:28 MattJD has quit (Ping timeout: 252 seconds)
2466 2011-03-03 18:15:30 MJD has joined
2467 2011-03-03 18:15:33 amiller has quit (Ping timeout: 252 seconds)
2468 2011-03-03 18:15:33 <Syke> go at least a month. 3 or 4 blocks is nothing
2469 2011-03-03 18:15:33 greyface has joined
2470 2011-03-03 18:16:30 <slush> Syke: that's the point. With the pool, there is no reason to wait a month
2471 2011-03-03 18:16:47 <Syke> sure there is, you get more coins solo
2472 2011-03-03 18:16:53 ducki2p has joined
2473 2011-03-03 18:16:55 <slush> Syke: After three months of pool mining, I found 80 blocks and got 4017 BTC
2474 2011-03-03 18:16:57 amiller has joined
2475 2011-03-03 18:17:02 <Blitzboom> Syke: dude
2476 2011-03-03 18:17:08 <Blitzboom> the difficulty readjusts
2477 2011-03-03 18:17:14 <Blitzboom> and therefore the probability of finding a block
2478 2011-03-03 18:17:35 <Syke> so?
2479 2011-03-03 18:17:41 <Syke> every hash is independant
2480 2011-03-03 18:17:48 <Blitzboom> if you do not, you better join a pool
2481 2011-03-03 18:22:06 <bulletbill> anyone get boost exceptions on 3.20?
2482 2011-03-03 18:23:33 m87 has joined
2483 2011-03-03 18:23:48 rgm3 has quit (Ping timeout: 240 seconds)
2484 2011-03-03 18:23:48 Netsniper has quit (Ping timeout: 240 seconds)
2485 2011-03-03 18:23:49 xelister has quit (Remote host closed the connection)
2486 2011-03-03 18:23:49 Lachesis has quit (Ping timeout: 250 seconds)
2487 2011-03-03 18:23:49 Syke has quit (Ping timeout: 240 seconds)
2488 2011-03-03 18:23:49 DoomDumas has quit (Ping timeout: 250 seconds)
2489 2011-03-03 18:23:49 m86 has quit (Ping timeout: 250 seconds)
2490 2011-03-03 18:23:49 <mmarker> I really think I should write the simulation for expected block payout as a function of time
2491 2011-03-03 18:23:49 <mmarker> as a little webapp
2492 2011-03-03 18:23:49 <mmarker> with sliders, so people can see how it actually works.
2493 2011-03-03 18:24:00 <amiller> ;;bc,gen 1200
2494 2011-03-03 18:24:00 <amiller> ;;bc,gen 130
2495 2011-03-03 18:24:00 Syke_ has joined
2496 2011-03-03 18:24:13 dazoe has quit (Ping timeout: 250 seconds)
2497 2011-03-03 18:24:33  has joined
2498 2011-03-03 18:24:33 <lfm> pool is kinda like ;;bc,gen while solo is like ;;bc,calc
2499 2011-03-03 18:24:33 <nanotube> lfm: hehe right
2500 2011-03-03 18:24:34 xelister_ has joined
2501 2011-03-03 18:24:35 <gribble> The expected generation output, at 1200 Khps, given current difficulty of 55590.23763914 , is 0.0217123384962 BTC per day and 0.000904680770675 BTC per hour.
2502 2011-03-03 18:24:36 <gribble> The expected generation output, at 130 Khps, given current difficulty of 55590.23763914 , is 0.00235217000375 BTC per day and 9.80070834898e-05 BTC per hour.
2503 2011-03-03 18:25:04 <amiller> ;;bc,gen 1200000
2504 2011-03-03 18:25:05 <gribble> The expected generation output, at 1200000 Khps, given current difficulty of 55590.23763914 , is 21.7123384962 BTC per day and 0.904680770675 BTC per hour.
2505 2011-03-03 18:25:12 <Blitzboom> you understand probability, do you?
2506 2011-03-03 18:25:12 <Blitzboom> sure it is possible
2507 2011-03-03 18:25:12 <Blitzboom> winning in the lottery is possbile too
2508 2011-03-03 18:25:12 bk128 has quit (Quit: bk128)
2509 2011-03-03 18:25:13 <Blitzboom> my chances of getting something at the pool are better than solo
2510 2011-03-03 18:25:13 <Blitzboom> and something is better than nothing
2511 2011-03-03 18:25:13 <Blitzboom> much better
2512 2011-03-03 18:25:13 sgornick has joined
2513 2011-03-03 18:25:13 <Blitzboom> i’d only consider soloing if i were pretty sure i’d find a block in 2016 blocks
2514 2011-03-03 18:25:13 <Blitzboom> i did, actually
2515 2011-03-03 18:25:15 <Travex> which pool for example ?
2516 2011-03-03 18:25:15 <Blitzboom> doesn’t matter. i use slush’s though
2517 2011-03-03 18:25:15 BlueSTARS has joined
2518 2011-03-03 18:25:15 <Blitzboom> but deepbit seems nice too
2519 2011-03-03 18:25:15 DoomDumasAlt has joined
2520 2011-03-03 18:25:16 bk128 has joined
2521 2011-03-03 18:25:16 <bk128> what's with all the netsplits?
2522 2011-03-03 18:25:17 <bk128> freenode ddos again?
2523 2011-03-03 18:25:17 dazoe has joined
2524 2011-03-03 18:25:17 luke-jr has joined
2525 2011-03-03 18:25:17 BlueSTARS has quit (Changing host)
2526 2011-03-03 18:25:17 BlueSTARS has joined
2527 2011-03-03 18:25:17 DoomDumasAlt has quit (Changing host)
2528 2011-03-03 18:25:17 DoomDumasAlt has joined
2529 2011-03-03 18:25:17 bk128 has quit (Changing host)
2530 2011-03-03 18:25:17 bk128 has joined
2531 2011-03-03 18:25:30 larsig has quit (Ping timeout: 264 seconds)
2532 2011-03-03 18:25:37 Spenvo has quit (Quit: Page closed)
2533 2011-03-03 18:25:42 T_X has joined
2534 2011-03-03 18:26:10 <luke-jr> ok so y'all need to upgrade your internets… :p
2535 2011-03-03 18:26:19 <BlueSTARS> lol
2536 2011-03-03 18:26:27 <Syke_> Blitzboom, what is your hashrate?
2537 2011-03-03 18:26:35 <BlueSTARS> ;;bc,gen 666
2538 2011-03-03 18:26:36 <gribble> The expected generation output, at 666 Khps, given current difficulty of 55590.23763914 , is 0.0120503478654 BTC per day and 0.000502097827724 BTC per hour.
2539 2011-03-03 18:26:39 <BlueSTARS> ;;bc,gen 666000
2540 2011-03-03 18:26:40 <gribble> The expected generation output, at 666000 Khps, given current difficulty of 55590.23763914 , is 12.0503478654 BTC per day and 0.502097827724 BTC per hour.
2541 2011-03-03 18:26:46 <Blitzboom> 330 mhash/s
2542 2011-03-03 18:27:00 <amiller> ;;bc,gen 230000
2543 2011-03-03 18:27:01 <gribble> The expected generation output, at 230000 Khps, given current difficulty of 55590.23763914 , is 4.1615315451 BTC per day and 0.173397147713 BTC per hour.
2544 2011-03-03 18:27:32 <lfm> geez I set off a flurry
2545 2011-03-03 18:27:46 <Syke_> mine for a couple months solo and you'll have more coins than if you were in a pool
2546 2011-03-03 18:28:00 <mmarker> FINALLY
2547 2011-03-03 18:28:04 bk128 has quit (Quit: bk128)
2548 2011-03-03 18:28:05 <Blitzboom> uhm … why, Syke_?
2549 2011-03-03 18:28:20 <mmarker> SHA-2 hash1 and hash2 are equal to hash1 and hash2 from the standard
2550 2011-03-03 18:28:31 <mmarker> Let's see if I can successfully getwork
2551 2011-03-03 18:28:32 <Syke_> solo doesn't pay fees, solo doesn't loose effectiveness due to latency
2552 2011-03-03 18:28:38 <lfm> ya that bc,gen assumes it all runs perfectly. if there is any downtime or communications problems or bugs it will prolly be less
2553 2011-03-03 18:28:42 <Blitzboom> yes
2554 2011-03-03 18:28:48 <Blitzboom> but it’s possible i get nothing
2555 2011-03-03 18:28:57 <Blitzboom> until the difficulty rises yet again
2556 2011-03-03 18:29:01 dishwara has joined
2557 2011-03-03 18:29:02 <Syke_> extremely unlikely, and also just as likely to get extra!
2558 2011-03-03 18:29:06 <Blitzboom> and it will be even more probable i get nothing
2559 2011-03-03 18:29:10 <mmarker> where's slush
2560 2011-03-03 18:29:28 <mmarker> There's a mathmatical way to prove this, but I need to know how slush divvys out the cash
2561 2011-03-03 18:29:56 <jgarzik> mmarker: /me scrolls back...  so your asm is validating internally?  whee...
2562 2011-03-03 18:29:58 <Syke_> Blitzboom, you have maybe a 10% chance to get nothing in 2 weeks
2563 2011-03-03 18:30:21 <Blitzboom> doesn’t have to be 14 days
2564 2011-03-03 18:30:33 DoomDumasAlt has quit (Ping timeout: 260 seconds)
2565 2011-03-03 18:30:39 <Blitzboom> rather calculate with ten
2566 2011-03-03 18:30:44 <slush> mmarker: what?
2567 2011-03-03 18:30:46 kellopyy has joined
2568 2011-03-03 18:31:01 Travex has left ()
2569 2011-03-03 18:31:03 tg has quit (Ping timeout: 260 seconds)
2570 2011-03-03 18:31:12 luke-jr has quit (Excess Flood)
2571 2011-03-03 18:31:14 T_X_ has joined
2572 2011-03-03 18:31:19 niekie_ has joined
2573 2011-03-03 18:31:30 luke-jr has joined
2574 2011-03-03 18:31:31 int0x27h_ has quit (Read error: Connection reset by peer)
2575 2011-03-03 18:31:35 niekie has quit (Read error: Connection reset by peer)
2576 2011-03-03 18:31:36 flok has quit (Ping timeout: 260 seconds)
2577 2011-03-03 18:31:36 BlueMatt has quit (Ping timeout: 260 seconds)
2578 2011-03-03 18:31:37 soultcer has quit (Ping timeout: 260 seconds)
2579 2011-03-03 18:31:37 darkskiez has quit (Ping timeout: 260 seconds)
2580 2011-03-03 18:31:38 jostmey has left ()
2581 2011-03-03 18:31:41 RBecker has quit (Read error: Connection reset by peer)
2582 2011-03-03 18:31:42 Edogaa has quit (Disconnected by services)
2583 2011-03-03 18:31:53 Edogaa has joined
2584 2011-03-03 18:31:58 T_X has quit (Read error: Connection reset by peer)
2585 2011-03-03 18:32:02 <Edogaa> anyone having difficulty with mtgox?
2586 2011-03-03 18:32:03 flok has joined
2587 2011-03-03 18:32:10 tg` has joined
2588 2011-03-03 18:32:11 darsk1ez has quit (Ping timeout: 260 seconds)
2589 2011-03-03 18:32:25 <lfm> Edogaa: what sort of difficulty?
2590 2011-03-03 18:32:32 darkskiez has joined
2591 2011-03-03 18:32:33 tg` is now known as tg
2592 2011-03-03 18:32:33 kellopyy is now known as Vilja
2593 2011-03-03 18:32:36 BlueMatt has joined
2594 2011-03-03 18:32:38 darsk1ez has joined
2595 2011-03-03 18:32:46 Daviey has joined
2596 2011-03-03 18:32:46 magnetron has quit (Ping timeout: 260 seconds)
2597 2011-03-03 18:32:48 KBme has joined
2598 2011-03-03 18:32:53 int0x27h has joined
2599 2011-03-03 18:33:06 <Edogaa> well, i type in www.mtgox.com (well the www. isn't neccessary but eh) and then here is what i see
2600 2011-03-03 18:33:09 magnetron has joined
2601 2011-03-03 18:33:10 <Edogaa> You don't have permission to access / on this server."
2602 2011-03-03 18:33:22 <Edogaa> "that kind...
2603 2011-03-03 18:33:32 <BlueMatt> wait for the dns to propogate
2604 2011-03-03 18:33:42 <BlueMatt> or clear your dns cache (if you run your own servers)
2605 2011-03-03 18:33:50 BlueSTARS is now known as Lachesis
2606 2011-03-03 18:34:17 soultcer has joined
2607 2011-03-03 18:34:30 <lfm> BlueMatt: mtgox moved server?
2608 2011-03-03 18:34:56 cosurg1 is now known as cosurgi
2609 2011-03-03 18:35:19 <jrabbit> But yeah someone ought to make an NaCl thingy maybe I'll work with cpuminer
2610 2011-03-03 18:35:24 <lfm> Edogaa: I dont see that. maybe as BlueMatt  sez
2611 2011-03-03 18:35:37 <Edogaa> -.-
2612 2011-03-03 18:35:43 <Edogaa> I see.
2613 2011-03-03 18:36:07 <Edogaa> Anyways, so does difficulty on bitcoins rise overtime according to how powerful the computers mining currently are?
2614 2011-03-03 18:36:10 dishwara has quit (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
2615 2011-03-03 18:36:27 DoomDumas has joined
2616 2011-03-03 18:36:27 DoomDumas has quit (Changing host)
2617 2011-03-03 18:36:27 DoomDumas has joined
2618 2011-03-03 18:36:34 <lfm> Edogaa: i Get Address: 74.52.143.58
2619 2011-03-03 18:36:34 <Blitzboom> Syke_: but seriously, isn’t the cash flow worth the possible cut?
2620 2011-03-03 18:36:41 <lfm> Edogaa: i Get Address: 74.52.143.58 for mtgix.com
2621 2011-03-03 18:36:52 dishwara has joined
2622 2011-03-03 18:36:57 <Blitzboom> i don’t really care if i don’t get 5 or 10% less
2623 2011-03-03 18:37:00 <Edogaa> GALACTIC INDUSTRIES :D
2624 2011-03-03 18:37:07 <Blitzboom> -don’t
2625 2011-03-03 18:37:10 da2ce7 has quit (Read error: Connection reset by peer)
2626 2011-03-03 18:37:10 echelon has quit (Ping timeout: 252 seconds)
2627 2011-03-03 18:37:10 ducki2p has quit (Ping timeout: 252 seconds)
2628 2011-03-03 18:37:10 amiller has quit (Ping timeout: 252 seconds)
2629 2011-03-03 18:37:12 Syke_ has quit (Ping timeout: 252 seconds)
2630 2011-03-03 18:37:12 greyface has quit (Ping timeout: 252 seconds)
2631 2011-03-03 18:37:17 da2ce7 has joined
2632 2011-03-03 18:37:57 <lfm> Blitzboom: I dont really care if daya to day is smooth
2633 2011-03-03 18:38:04 da2ce7 has quit (Changing host)
2634 2011-03-03 18:38:04 da2ce7 has joined
2635 2011-03-03 18:38:22 Syke_ has joined
2636 2011-03-03 18:38:34 greyface has joined
2637 2011-03-03 18:38:38 <Blitzboom> i don’t like lottery
2638 2011-03-03 18:38:49 magnetron has quit (Read error: Connection reset by peer)
2639 2011-03-03 18:39:39 <jrabbit> Yep NaCl time.
2640 2011-03-03 18:39:52 magnetron has joined
2641 2011-03-03 18:40:39 <Syke_> Blitzboom, then why are you mining?
2642 2011-03-03 18:40:56 T_X has joined
2643 2011-03-03 18:40:56 T_X has quit (Changing host)
2644 2011-03-03 18:40:56 T_X has joined
2645 2011-03-03 18:40:57 fahadsadah_ has joined
2646 2011-03-03 18:40:58 niekie has joined
2647 2011-03-03 18:40:59 <Blitzboom> because it’s a nice way to get BTC
2648 2011-03-03 18:41:04 <Blitzboom> for now, that is
2649 2011-03-03 18:41:05 KBme has quit (Ping timeout: 260 seconds)
2650 2011-03-03 18:41:05 fahadsadah has quit (Ping timeout: 260 seconds)
2651 2011-03-03 18:41:05 dsg has quit (Ping timeout: 260 seconds)
2652 2011-03-03 18:41:05 BlueMatt has quit (Ping timeout: 260 seconds)
2653 2011-03-03 18:41:10 <lfm> Syke_: he thinks pool isn't gambling
2654 2011-03-03 18:41:15 <Blitzboom> it is
2655 2011-03-03 18:41:26 <lfm> isnt lottery
2656 2011-03-03 18:41:28 dsg has joined
2657 2011-03-03 18:41:28 dsg has quit (Changing host)
2658 2011-03-03 18:41:28 dsg has joined
2659 2011-03-03 18:41:31 luke-jr has quit (Read error: Connection reset by peer)
2660 2011-03-03 18:41:31 <Blitzboom> but it reduces the probability of getting nothing
2661 2011-03-03 18:41:31 tg has quit (Ping timeout: 260 seconds)
2662 2011-03-03 18:41:31 niekie_ has quit (Ping timeout: 260 seconds)
2663 2011-03-03 18:41:31 T_X_ has quit (Ping timeout: 260 seconds)
2664 2011-03-03 18:41:32 gwillen has quit (Ping timeout: 260 seconds)
2665 2011-03-03 18:41:36 luke-jr_ has joined
2666 2011-03-03 18:41:36 BlueMatt has joined
2667 2011-03-03 18:41:37 luke-jr_ is now known as luke-jr
2668 2011-03-03 18:41:50 <Blitzboom> you’ll get your share of the next block the pool gets
2669 2011-03-03 18:41:55 <Syke_> Blitzboom, you're paying for that reduction
2670 2011-03-03 18:41:59 <Blitzboom> i know i do
2671 2011-03-03 18:42:14 <Syke_> and that's fine
2672 2011-03-03 18:42:41 <lfm> Blitzboom: and reduces the probability of getting windfall, it all averages out
2673 2011-03-03 18:42:42 RBecker has joined
2674 2011-03-03 18:42:43 <Syke_> you could also wait a few weeks, and not pay, and still get that reduction
2675 2011-03-03 18:42:51 gwillen has joined
2676 2011-03-03 18:42:51 gwillen has quit (Changing host)
2677 2011-03-03 18:42:51 gwillen has joined
2678 2011-03-03 18:43:05 <Blitzboom> the problem is
2679 2011-03-03 18:43:18 KBme has joined
2680 2011-03-03 18:43:26 <Blitzboom> if i miss a block in 2016 block, the chance to miss again will probably be higher
2681 2011-03-03 18:43:37 <Blitzboom> so the present is more important
2682 2011-03-03 18:43:54 xelister_ has quit (Read error: Connection reset by peer)
2683 2011-03-03 18:43:55 <lfm> ya I understand, for some situations a windfall is almost as much trouble as extended period(s) of nothing
2684 2011-03-03 18:44:07 <Blitzboom> it’s also psychologically straining
2685 2011-03-03 18:45:16 <slush> btw the worst pool round needed 420000 shares to find a block. It was 12x more shares than current difficulty. I cannot imagine my feelings if I hit the same probability with single GPU...
2686 2011-03-03 18:45:22 Guest52339 has joined
2687 2011-03-03 18:45:31 <slush> But in the pool, it took only 10 hours
2688 2011-03-03 18:46:50 <[Tycho]> Scary 10 hours.
2689 2011-03-03 18:46:51 <lfm> trust the math luke
2690 2011-03-03 18:46:53 <Syke_> yesterday my 680 mhps miner found 2 blocks, I can't imagine what it would feel like to have only gotten 10 coins.
2691 2011-03-03 18:46:56 TD has quit (Quit: TD)
2692 2011-03-03 18:47:09 <Blitzboom> i was lucky too, Syke_
2693 2011-03-03 18:47:14 <Blitzboom> got five blocks in four days
2694 2011-03-03 18:47:22 <Syke_> yes, for every bad luck there is good luck
2695 2011-03-03 18:47:27 <Blitzboom> exactly
2696 2011-03-03 18:47:34 <Blitzboom> and i don’t like gambling
2697 2011-03-03 18:47:34 <Syke_> but pool operators only parrot the bad luck
2698 2011-03-03 18:47:44 <Blitzboom> right
2699 2011-03-03 18:47:53 <Blitzboom> plus fee and invalid blocks
2700 2011-03-03 18:47:56 <slush> [Tycho]: yes, it was, it still lowered daily reward for everybody a bit as it was the worse day in whole history, but even with this, every miner had at least some money for this day
2701 2011-03-03 18:48:18 <[Tycho]> You can use insurance against luck at all - by using pay-per-share pools :)
2702 2011-03-03 18:48:39 <slush> [Tycho]: but as the risk takes the pool operator, he have to rise fees ;)
2703 2011-03-03 18:48:59 <lfm> [Tycho]: still takes luck to find shares too
2704 2011-03-03 18:48:59 <Blitzboom> or lower the reward
2705 2011-03-03 18:49:42 <[Tycho]> lfm, with GPU you will certainly find some.
2706 2011-03-03 18:50:06 zygf_ has quit (Read error: Connection reset by peer)
2707 2011-03-03 18:50:06 Guest52339 has quit (Quit: Reconnecting)
2708 2011-03-03 18:50:15 <slush> I think this discussion is pointless. Everybody know the advantages and disadvantages of solo mining, share based pool and pay per share pool. So everything is only about preferences
2709 2011-03-03 18:50:21 <lfm> [Tycho]: with anything you will eventually find shares OR solo blocks, just takes time
2710 2011-03-03 18:50:24 zygf has joined
2711 2011-03-03 18:50:34 <JFK911> ;;bc,mtgox
2712 2011-03-03 18:50:34 <gribble> {"ticker":{"high":0.94,"low":0.91,"vol":2847,"buy":0.911,"sell":0.9392,"last":0.9392}}
2713 2011-03-03 18:50:41 <Blitzboom> slush: is there maybe a mathematical way of knowing?
2714 2011-03-03 18:50:41 <[Tycho]> I still can't understand some CPU-users who mine 100-700 khps.
2715 2011-03-03 18:50:53 luke-jr has quit (Ping timeout: 260 seconds)
2716 2011-03-03 18:50:54 darkskiez has quit (Ping timeout: 260 seconds)
2717 2011-03-03 18:50:54 Takyoji has quit (Ping timeout: 260 seconds)
2718 2011-03-03 18:50:55 jackmcbarn has quit (Ping timeout: 260 seconds)
2719 2011-03-03 18:50:57 <Blitzboom> but i guess there isn’t. difficulty is unpredictable
2720 2011-03-03 18:50:59 <UukGoblin> [Tycho], what's so hard to understand?
2721 2011-03-03 18:51:07 <Syke_> [Tycho], I can't understand gigahash miners pooling.
2722 2011-03-03 18:51:09 <slush> [Tycho]: me too
2723 2011-03-03 18:51:17 <lfm> [Tycho]: and I cant understand people who buy lottery tickets, so?
2724 2011-03-03 18:51:26 BlueMatt has quit (Ping timeout: 260 seconds)
2725 2011-03-03 18:51:26 darsk1ez has quit (Ping timeout: 260 seconds)
2726 2011-03-03 18:51:27 x6763 has quit (Ping timeout: 260 seconds)
2727 2011-03-03 18:51:32 <UukGoblin> I have a server, it has no GPU in it, I don't pay for power, so why not run a miner?
2728 2011-03-03 18:51:36 <[Tycho]> lfm, with GPU you will certainly find some shares in EVERY round. But not with CPU.
2729 2011-03-03 18:51:40 <slush> But with last difficulty rise, I see many CPU users disconnected from the pool
2730 2011-03-03 18:51:48 <slush> So I think they close mining at all
2731 2011-03-03 18:52:08 <[Tycho]> One of my users with 700 Khps wrote in forum that he didn't find any in 3 rounds in a row.
2732 2011-03-03 18:52:23 <lfm> [Tycho]: difference in degree, not in principle
2733 2011-03-03 18:52:49 <slush> lfm: but cpu mining eat much more resources on every side
2734 2011-03-03 18:53:11 <slush> so it is about degree
2735 2011-03-03 18:53:44 <[Tycho]> "<Syke_> [Tycho], I can't understand gigahash miners pooling" - before running my pool i was using slush's one with my near-1GHs rig and his stats showed that i didn't found a single block in 2 weeks. So if mining solo that would be a disaster.
2736 2011-03-03 18:54:11 x6763 has joined
2737 2011-03-03 18:55:42 Akiraaaa has joined
2738 2011-03-03 18:58:48 AmazingPudding has quit (Ping timeout: 240 seconds)
2739 2011-03-03 18:58:48 noot has quit (Ping timeout: 250 seconds)
2740 2011-03-03 18:58:48 <mmarker> Ok. Fingers crossed...as we wait for a work unit to be delivered...
2741 2011-03-03 18:58:49 noot has joined
2742 2011-03-03 18:58:49 <mmarker> I need to work on the performance a tad :\
2743 2011-03-03 18:58:49 JStoker has quit (Ping timeout: 250 seconds)
2744 2011-03-03 18:58:49 jercos has quit (Ping timeout: 250 seconds)
2745 2011-03-03 18:58:49 jercos has joined
2746 2011-03-03 18:58:49 <mmarker> ;;bc,calcd 4200 1
2747 2011-03-03 18:58:50 kupo has quit (Ping timeout: 240 seconds)
2748 2011-03-03 18:58:51 tg` has joined
2749 2011-03-03 18:58:51 RBecker has quit (Ping timeout: 260 seconds)
2750 2011-03-03 18:58:51 dsg has quit (Ping timeout: 260 seconds)
2751 2011-03-03 18:58:52 dsg has joined
2752 2011-03-03 18:58:52 luke-jr has joined
2753 2011-03-03 18:58:52 dsg has quit (Changing host)
2754 2011-03-03 18:58:52 dsg has joined
2755 2011-03-03 18:58:52 KBme has quit (Ping timeout: 260 seconds)
2756 2011-03-03 18:58:52 BlueMatt has joined
2757 2011-03-03 18:58:52 RBecker has joined
2758 2011-03-03 18:58:52 BlueMatt has quit (Changing host)
2759 2011-03-03 18:58:52 BlueMatt has joined
2760 2011-03-03 18:58:53 tg` has quit (Client Quit)
2761 2011-03-03 18:58:53 darkskiez has joined
2762 2011-03-03 18:58:53 tg` has joined
2763 2011-03-03 18:58:54 darsk1ez has joined
2764 2011-03-03 18:58:54 KBme has joined
2765 2011-03-03 18:58:54 Takyoji has joined
2766 2011-03-03 18:58:55 theboos has joined
2767 2011-03-03 18:58:55 theboos has quit (Changing host)
2768 2011-03-03 18:58:55 theboos has joined
2769 2011-03-03 18:58:55 jackmcbarn has joined
2770 2011-03-03 18:58:56 bonsaikitten has joined
2771 2011-03-03 18:58:56 dishwara has quit (Changing host)
2772 2011-03-03 18:58:56 dishwara has joined
2773 2011-03-03 18:58:57 <jrabbit> jgarzik: ping
2774 2011-03-03 18:58:58 bonsaikitten has quit (Client Quit)
2775 2011-03-03 18:58:58 tg` is now known as tg
2776 2011-03-03 18:58:58 RBecker has quit (Changing host)
2777 2011-03-03 18:58:58 RBecker has joined
2778 2011-03-03 18:58:58 x6763 has quit (Ping timeout: 250 seconds)
2779 2011-03-03 18:58:58 devon_hillard_ has quit (Ping timeout: 250 seconds)
2780 2011-03-03 18:58:58 <mmarker> Damn
2781 2011-03-03 18:58:58 <mmarker> ;;bc,calcd 4200 1
2782 2011-03-03 18:58:58 <lfm> [Tycho]: and there IS a chance you wont either
2783 2011-03-03 18:58:59 <gribble> The average time to generate a block at 4200 Khps, given the supplied difficulty of 1, is 17 minutes and 2 seconds
2784 2011-03-03 18:58:59 kupo has joined
2785 2011-03-03 18:58:59 <gribble> The average time to generate a block at 4200 Khps, given the supplied difficulty of 1, is 17 minutes and 2 seconds
2786 2011-03-03 18:59:00 <jgarzik> jrabbit: hello
2787 2011-03-03 18:59:00 bonsaikitten has joined
2788 2011-03-03 18:59:00 <mmarker> Ugh.
2789 2011-03-03 18:59:01 <jrabbit> jgarzik: I'm going to try and get cpuminer to run in nacl...
2790 2011-03-03 18:59:20 <jrabbit> how much do you use libcurl?
2791 2011-03-03 18:59:32 x6763 has joined
2792 2011-03-03 18:59:45 * nameless1 hates libcurl
2793 2011-03-03 18:59:48 <mmarker> jgarzik: stupid git question. is there an easy way to copy a branch and then rebase it to merge the commits? Dont quite want to lose my history in submitting the pull request.
2794 2011-03-03 18:59:57 <jgarzik> jrabbit: JSON-RPC is an HTTP protocol.  cpuminer uses curl for HTTP.
2795 2011-03-03 19:00:12 <Syke_> [Tycho], a week or two is too short a time frame.
2796 2011-03-03 19:00:35 <jrabbit> jgarzik: ok well I mean google provides some fugly APIs for stuff like that, maybe I can cut libcurl out
2797 2011-03-03 19:00:47 <lfm> disaster in his head, just as many find more blocks than average
2798 2011-03-03 19:00:59 <Syke_> if you just want to mine a week or two and get out, then by all means, get in a pool
2799 2011-03-03 19:01:24 <lfm> if you just want to mine a week or two and get out, then by all means, mine solo
2800 2011-03-03 19:01:28 skeledrew1 has joined
2801 2011-03-03 19:01:31 kupo has quit (Changing host)
2802 2011-03-03 19:01:32 kupo has joined
2803 2011-03-03 19:01:41 <jgarzik> mmarker: you can 'git cherry-pick $COMMIT' for a single commit in another branch, but merging commits together means 'git show $COMMIT | patch -sp1' + 'git show $COMMIT | patch -sp1' + 'git commit -a'
2804 2011-03-03 19:02:23 <Syke_> [Tycho], you just as likely could have earned several blocks
2805 2011-03-03 19:02:53 <Syke_> over time, you will earn more solo
2806 2011-03-03 19:02:55 <dishwara> : [Tycho] thanks for ur ppol
2807 2011-03-03 19:03:02 <mmarker> jgarzik: That's what I was afraid of.
2808 2011-03-03 19:03:13 Tril has quit (Ping timeout: 250 seconds)
2809 2011-03-03 19:03:15 <Diablo-D3> jgarzik: thats squashing commits
2810 2011-03-03 19:03:22 <Diablo-D3> and theres a command for that
2811 2011-03-03 19:03:28 <mmarker> I probably need to improve my git behavior. Damn used to committing every minor change so I don't lose things
2812 2011-03-03 19:03:41 <Diablo-D3> [01:51:39] <jgarzik> jrabbit: JSON-RPC is an HTTP protocol.  cpuminer uses curl for HTTP.
2813 2011-03-03 19:03:48 <jrabbit> I got that.
2814 2011-03-03 19:03:48 <Diablo-D3> are we still fucking talking about this?
2815 2011-03-03 19:03:54 * jrabbit is now working
2816 2011-03-03 19:04:00 <jrabbit> well mostly fuckign with git
2817 2011-03-03 19:04:01 <Diablo-D3> mmarker: nothing wrong with committing that way
2818 2011-03-03 19:04:05 noot has quit (Ping timeout: 250 seconds)
2819 2011-03-03 19:04:12 <Diablo-D3> mmarker: just make sure you use the git way of doing things
2820 2011-03-03 19:04:19 zygf_ has joined
2821 2011-03-03 19:04:20 <mmarker> Diablo-D3: YNot at all. Means I have to rebase if I want a clean pull request
2822 2011-03-03 19:04:23 <Diablo-D3> mmarker: create a topic specific branch, and then commit only to that
2823 2011-03-03 19:04:25 <mmarker> which is annoying as hell.
2824 2011-03-03 19:04:31 skeledrew has quit (Ping timeout: 250 seconds)
2825 2011-03-03 19:04:31 EvanR-work has quit (Ping timeout: 250 seconds)
2826 2011-03-03 19:04:32 hexidigital has quit (Ping timeout: 250 seconds)
2827 2011-03-03 19:04:39 <Diablo-D3> then merge it into master when you want to commit to origin
2828 2011-03-03 19:04:49 <Diablo-D3> and no, you dont have to rebase
2829 2011-03-03 19:04:57 zygf has quit (Ping timeout: 250 seconds)
2830 2011-03-03 19:04:58 eps has quit (Ping timeout: 250 seconds)
2831 2011-03-03 19:05:00 <Diablo-D3> you can stash uncomitted work
2832 2011-03-03 19:06:00 noot has joined
2833 2011-03-03 19:06:18 EvanR-work has joined
2834 2011-03-03 19:06:24 tg has quit (Quit: Reconnecting)
2835 2011-03-03 19:06:29 <mmarker> diablo: Yea, How I've set up my tree is the server had a devel branch, and the main branch from jgarzik's fork. I branch off of devel for new features, then merge back into devel with --no-ff to keep a history
2836 2011-03-03 19:06:40 tg has joined
2837 2011-03-03 19:06:46 <mmarker> It's easier to see in github's network graph
2838 2011-03-03 19:06:48 <mmarker> or gitk
2839 2011-03-03 19:06:51 theboos has quit (Quit: Page closed)
2840 2011-03-03 19:07:07 hexidigital has joined
2841 2011-03-03 19:07:07 hexidigital has quit (Changing host)
2842 2011-03-03 19:07:07 hexidigital has joined
2843 2011-03-03 19:07:15 <Diablo-D3> mmarker: you should use branches instead of committing into master.
2844 2011-03-03 19:07:47 <mmarker> I don't commit to master.
2845 2011-03-03 19:08:08 bonsaikitten is now known as DrEeevil
2846 2011-03-03 19:08:14 <mmarker> commit to branch, merge in when feature is done
2847 2011-03-03 19:08:31 <mmarker> but not merge to master, of course
2848 2011-03-03 19:10:18 larsivi has quit (Read error: Operation timed out)
2849 2011-03-03 19:11:05 <mmarker> https://github.com/chromicant/cpuminer/network
2850 2011-03-03 19:11:14 <mmarker> the twisty line that's devel is the mess
2851 2011-03-03 19:11:29 <mmarker> each little loop was a local branch that was a feature
2852 2011-03-03 19:12:30 <jrabbit> jgarzik: do you just dump the json object to jansson? I'm seeing json_rpc_call(CURL *curl,...) which reads to my limited c knowleges as if you're passign curl stuff?
2853 2011-03-03 19:13:08 f3n has joined
2854 2011-03-03 19:13:17 <jgarzik> jrabbit: CURL* is the handle for the current HTTP connection, and all HTTP settings, etc.
2855 2011-03-03 19:13:27 f3n has quit (Changing host)
2856 2011-03-03 19:13:27 f3n has joined
2857 2011-03-03 19:14:14 * mmarker fires up the core i5 for testing
2858 2011-03-03 19:14:28 <jrabbit> hm i mgiht try and compile in libcurl for simplicity's sake for now
2859 2011-03-03 19:16:07 larsivi has joined
2860 2011-03-03 19:16:45 <lfm> why is that simpler?
2861 2011-03-03 19:17:01 JStoker has joined
2862 2011-03-03 19:17:11 <mmarker> jgarzik: Oh, there's another sha optimization you can do, but it makes the C code much more complicated.
2863 2011-03-03 19:19:06 eps has joined
2864 2011-03-03 19:19:23 <mmarker> ;;bc,calcd 12800 1
2865 2011-03-03 19:19:23 <gribble> The average time to generate a block at 12800 Khps, given the supplied difficulty of 1, is 5 minutes and 35 seconds
2866 2011-03-03 19:19:29 <jgarzik> mmarker: you could describe it in a forum post, or Create an Issue on github for it
2867 2011-03-03 19:19:43 <mmarker> jgarzik: Or, I can actually code it
2868 2011-03-03 19:19:45 TheAncientGoat has quit (Ping timeout: 276 seconds)
2869 2011-03-03 19:19:46 <mmarker> :D
2870 2011-03-03 19:19:50 <jgarzik> mmarker: or that :)
2871 2011-03-03 19:20:03 <mmarker> Basically, you can skip the first 3 rounds of SHA on the first hash
2872 2011-03-03 19:20:14 <mmarker> but that requires you to special case the code
2873 2011-03-03 19:20:21 rgm3 has joined
2874 2011-03-03 19:20:24 <mmarker> well, not "skip"
2875 2011-03-03 19:20:36 <mmarker> but they're common across nonces for a given block
2876 2011-03-03 19:22:13 <jgarzik> mmarker: poclbm has that optimization, IIRC
2877 2011-03-03 19:22:15 <rgm3> Why might someone want to override worksize in DiabloMiner?  I tried -w 128 for kicks, but I can't really tell what changed.
2878 2011-03-03 19:22:45 <Diablo-D3> nothing should change
2879 2011-03-03 19:23:06 <Diablo-D3> optimum worksize for radeons is a multiple of 64
2880 2011-03-03 19:23:07 <phantomcircuit> nMaxOutboundConnections = min(nMaxOutboundConnections, (int)GetArg("-maxconnections", 125));
2881 2011-03-03 19:23:12 <phantomcircuit> shouldnt that be inbound connections?
2882 2011-03-03 19:23:22 <mmarker> Yea, again, it's only a few cycles. On a cpu, though, those damn majority bit functions are just icky.
2883 2011-03-03 19:23:22 <Diablo-D3> the only time you up it from 64 is if you need to offset memory latency
2884 2011-03-03 19:23:29 <Diablo-D3> we dont use memory, so it doesnt apply to us
2885 2011-03-03 19:23:33 <Diablo-D3> thus, 64 is most optimum
2886 2011-03-03 19:23:44 <[Tycho]> dishwara, you are welcome :)
2887 2011-03-03 19:23:45 Vladimir__ has joined
2888 2011-03-03 19:23:47 <rgm3> okay, thanks.
2889 2011-03-03 19:23:47 * mmarker waits to win the getwork lotto
2890 2011-03-03 19:23:50 echelon has joined
2891 2011-03-03 19:24:06 <Diablo-D3> _now_ some people strangely are doing well on 64
2892 2011-03-03 19:24:10 <Diablo-D3> err
2893 2011-03-03 19:24:12 <Diablo-D3> on 128
2894 2011-03-03 19:24:24 amiller has joined
2895 2011-03-03 19:24:24 <Diablo-D3> but that just seems to be a nasty driver bug
2896 2011-03-03 19:24:27 <rgm3> I'm doing exactly the same
2897 2011-03-03 19:24:33 <jgarzik> phantomcircuit: inbound check is elsewhere in net.cpp
2898 2011-03-03 19:24:38 <[Tycho]> Syke_, yes, i could. But there was real proved possibility of getting nothing. With pool mining i can always be sure to get my average value.
2899 2011-03-03 19:24:55 <phantomcircuit> else if (nInbound >= GetArg("-maxconnections", 125) - MAX_OUTBOUND_CONNECTIONS)
2900 2011-03-03 19:25:09 <Diablo-D3> rgm3: yeah, so 64 is best
2901 2011-03-03 19:25:12 bulletbill has left ()
2902 2011-03-03 19:25:12 <phantomcircuit> yeah
2903 2011-03-03 19:25:14 <phantomcircuit> so it is
2904 2011-03-03 19:25:14 <Diablo-D3> rgm3: 64 is also now the default in my miner
2905 2011-03-03 19:25:32 <jrabbit> jgarzik: aww crap this seems very over my head (Putting libcurl into nacl is a bad/ill advised plan)
2906 2011-03-03 19:25:45 <jgarzik> phantomcircuit: MAX_OUTBOUND_CONNECTIONS is hardcoded to 8, so you don't get a lot by default anyway
2907 2011-03-03 19:25:47 <rgm3> Diablo-D3: Yes, your miner is the only miner that I could get functioning on my Mac Pro.
2908 2011-03-03 19:25:56 * mmarker still waits...
2909 2011-03-03 19:26:06 djoot has joined
2910 2011-03-03 19:26:07 <jrabbit> though what I'm seeing is that all that needs to be done is a hack from curl stuff to PPB_URLLoader
2911 2011-03-03 19:26:24 <jrabbit> http://src.chromium.org/viewvc/chrome/trunk/src/ppapi/c/ppb_url_loader.h?view=markup
2912 2011-03-03 19:26:28 <mmarker> Oh
2913 2011-03-03 19:26:33 <mmarker> someone shoot me
2914 2011-03-03 19:26:56 <jrabbit> mmarker: lol why
2915 2011-03-03 19:27:04 citiz3n_ has joined
2916 2011-03-03 19:27:23 <mmarker> Urm
2917 2011-03-03 19:27:23 <rgm3> Diablo-D3: I get 10Mh/s from your miner, while the official bitcoin client does about 5Mh/s using the dual quad-core intel xeon 2.26GHz processor (8 hyperthreaded cores)
2918 2011-03-03 19:27:29 <rgm3> (16 virtual cpus)
2919 2011-03-03 19:27:38 <mmarker> So. Uhh. For a block to be valid in a pool, H==0
2920 2011-03-03 19:27:40 <rgm3> so in total I manage 15Mh/s
2921 2011-03-03 19:27:44 <mmarker> I was looking at D
2922 2011-03-03 19:27:54 citiz3n has quit (Ping timeout: 264 seconds)
2923 2011-03-03 19:27:59 <mmarker> So either I'm looking for that high difficulty block, or I'm a dumbass
2924 2011-03-03 19:28:00 citiz3n_ is now known as citiz3n
2925 2011-03-03 19:28:06 citiz3n has quit (Changing host)
2926 2011-03-03 19:28:06 citiz3n has joined
2927 2011-03-03 19:29:57 f3n has quit ()
2928 2011-03-03 19:32:22 <mmarker> GRR
2929 2011-03-03 19:32:46 <mmarker> Found a valid hash, didn't validate
2930 2011-03-03 19:32:53 <mmarker> wonder if my byteorder is messed up
2931 2011-03-03 19:32:58 <rgm3> sad trombone
2932 2011-03-03 19:37:11 <Diablo-D3> rgm3: dont use cpu mining.
2933 2011-03-03 19:43:27 RazielZ has quit ()
2934 2011-03-03 19:45:18 * mmarker scratches his head
2935 2011-03-03 19:45:41 <Aciid> ;;bc,gen 500000
2936 2011-03-03 19:45:42 <gribble> The expected generation output, at 500000 Khps, given current difficulty of 55590.23763914 , is 9.04680770675 BTC per day and 0.376950321114 BTC per hour.
2937 2011-03-03 19:47:34 <citiz3n> ;;bc,gen 3480000
2938 2011-03-03 19:47:35 <gribble> The expected generation output, at 3480000 Khps, given current difficulty of 55590.23763914 , is 62.965781639 BTC per day and 2.62357423496 BTC per hour.
2939 2011-03-03 19:48:08 sethsethseth has quit (Quit: ~ Trillian Astra - www.trillian.im ~)
2940 2011-03-03 19:50:23 <mmarker> Well now
2941 2011-03-03 19:50:28 <mmarker> this is an interesting bug
2942 2011-03-03 19:52:28 <[Tycho]> Randomness is funny. First two days after difficulty increase i didn't even notice any deviation from average income @ slush's pool :)
2943 2011-03-03 19:54:14 <mmarker> minor nonce increment bug fixed...
2944 2011-03-03 19:56:31 <Aciid> [Tycho]: SSL coming soon?
2945 2011-03-03 19:57:18 <mmarker> Tycho: I hate randomness right now.
2946 2011-03-03 19:57:28 <mmarker> Trying to do a live test of the protocol...
2947 2011-03-03 19:57:45 <[Tycho]> mmarker, use GPU.
2948 2011-03-03 19:58:06 <mmarker> tycho: I'm testing my CPU routine!
2949 2011-03-03 19:58:22 <[Tycho]> Aciid, yes. I think that SSL is far from providing complete protection, but i'll do it for people :)
2950 2011-03-03 19:58:54 <[Tycho]> Actually i can do something else too. For example, e-mail verification for changing bitcoin address.
2951 2011-03-03 20:00:59 <Aciid> [Tycho]: yeah
2952 2011-03-03 20:01:26 <Aciid> I appreciate that you take the security as a matter of concern =)
2953 2011-03-03 20:03:33 <mmarker> SUCK IT TREBEK!
2954 2011-03-03 20:03:48 <mmarker> PROOF OF WORK RESULT: true (yay!!)
2955 2011-03-03 20:03:57 * mmarker goes to commit
2956 2011-03-03 20:04:30 <mmarker> Bam, another one
2957 2011-03-03 20:04:39 <mmarker> 3200khash/sec per core
2958 2011-03-03 20:04:40 <mmarker> slow
2959 2011-03-03 20:04:43 <mmarker> :(
2960 2011-03-03 20:04:53 <mmarker> But now I know the core code works...
2961 2011-03-03 20:05:01 <Aciid> please do share...
2962 2011-03-03 20:05:13 <mmarker> I'm commiting, pushing to github, and posting
2963 2011-03-03 20:05:22 <Aciid> I can do 20MH/s on all six
2964 2011-03-03 20:05:25 <Aciid> together
2965 2011-03-03 20:05:34 <Aciid> single code would be fun to try
2966 2011-03-03 20:05:43 <lfm> mmarker: what cpu model is that?
2967 2011-03-03 20:06:07 <mmarker> Core i5 760
2968 2011-03-03 20:06:24 <mmarker> You need to be running an x86_64 bit version of Linux, though :D
2969 2011-03-03 20:06:34 <Aciid> no problem.
2970 2011-03-03 20:06:37 <mmarker> Sorry kids, no 32 bit, no windows. Get a real machine
2971 2011-03-03 20:06:48 <lfm> ya leeme atem
2972 2011-03-03 20:06:55 <Aciid> push the code, im anxious
2973 2011-03-03 20:08:12 <mmarker> I've pushed. Going to do one last validation to make sure this all wasn't a fluke
2974 2011-03-03 20:08:28 gavinandresen has joined
2975 2011-03-03 20:08:47 <mmarker> We;re good!
2976 2011-03-03 20:08:55 <mmarker> lemme comment out one extra printk I missed...
2977 2011-03-03 20:12:33 <EvanR-work> i use 32 bit software exclusively
2978 2011-03-03 20:12:37 <EvanR-work> im that bad ass
2979 2011-03-03 20:16:25 <phantomcircuit> yeah im pretty sure this slow dos would be extremely effective at segmenting the network
2980 2011-03-03 20:16:32 <phantomcircuit> im going to try it out on the test net
2981 2011-03-03 20:17:44 Akiraaaa has quit (Read error: Connection reset by peer)
2982 2011-03-03 20:18:00 <phantomcircuit> although the attack could take days
2983 2011-03-03 20:18:05 <phantomcircuit> even weeks
2984 2011-03-03 20:18:50 <Aciid> mmarker: wheres your code at?
2985 2011-03-03 20:19:38 <mmarker> See forum
2986 2011-03-03 20:19:55 <mmarker> http://github.com/chromicant/cpuminer
2987 2011-03-03 20:20:02 <mmarker> SSE2 branch
2988 2011-03-03 20:20:22 <mmarker> It's still a tad slow. There's some further work I need to do on it.
2989 2011-03-03 20:20:26 <Diablo-D3> rgm3: you're on ubuntu, right?
2990 2011-03-03 20:20:53 barburator has quit (Quit: Page closed)
2991 2011-03-03 20:21:49 <phantomcircuit> so anyways i take it nobody cares about my ridiculously slow network segmentation attack?
2992 2011-03-03 20:21:51 <BlueMatt> jgarzik, phantomcircuit: shouldnt it be the other way around, ie limit incoming and use a lot of outgoing connections?
2993 2011-03-03 20:22:24 <phantomcircuit> BlueMatt, it doesn't really matter
2994 2011-03-03 20:22:25 <rgm3> Diablo-D3: why wouldn't I want to use my CPU mining in addition to your GPU miner?
2995 2011-03-03 20:22:28 <BlueMatt> or maybe atleast say, have atleast x outgoing
2996 2011-03-03 20:22:43 <phantomcircuit> BlueMatt, either way if i can use up all of the networks connectable slots i can control the network
2997 2011-03-03 20:22:46 <rgm3> Diablo-D3: I have two systems, an Ubuntu 10.10 laptop that cannot mine
2998 2011-03-03 20:22:53 <rgm3> and a Mac Pro that can
2999 2011-03-03 20:23:00 <BlueMatt> phantomcircuit: No, outoing is limited to one ip per /24
3000 2011-03-03 20:23:08 <BlueMatt> outgoing*
3001 2011-03-03 20:23:12 <phantomcircuit> BlueMatt, for example i could force the production of 2 very long segmented block chains
3002 2011-03-03 20:23:21 <phantomcircuit> oh
3003 2011-03-03 20:23:22 <rgm3> the laptop has a radeon X1400, I would have to downgrade my xorg and kernel versions to get the proprietary drivers working, and i'm unwilling.
3004 2011-03-03 20:23:33 <phantomcircuit> BlueMatt, currently there are no limits for the # of conns/ip
3005 2011-03-03 20:23:43 <phantomcircuit> just limits on the number of total connections
3006 2011-03-03 20:23:51 <BlueMatt> phantomcircuit: true, but for outgoing there is a limit
3007 2011-03-03 20:23:58 dishwara has quit (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
3008 2011-03-03 20:24:13 <phantomcircuit> BlueMatt, there is a limit for both incoming and outgoing on the total # of connections
3009 2011-03-03 20:24:21 <BlueMatt> phantomcircuit: It would work on the current version and it is a problem, but it is fairly easy to fix
3010 2011-03-03 20:24:29 <phantomcircuit> BlueMatt, incoming + outgoing < 125 (default), and outgoing <= 8
3011 2011-03-03 20:24:30 <mmarker> Hmm. Plus side with this new code. More hashes from my budget vps's
3012 2011-03-03 20:24:33 <mmarker> muhahahaha
3013 2011-03-03 20:25:13 <Raulo> mmaker: I'm testing tour code
3014 2011-03-03 20:25:19 <phantomcircuit> BlueMatt, sure but all id have to do in order to get around any limit is obtain more ips
3015 2011-03-03 20:25:26 <BlueMatt> phantomcircuit: Actually I take that back, I believe it only limits to one per /16 on startup
3016 2011-03-03 20:25:31 <phantomcircuit> BlueMatt, if i could get more ips than there are nodes in the network... game over
3017 2011-03-03 20:25:32 prax_ has quit (Quit: Leaving)
3018 2011-03-03 20:25:41 <Raulo> No validation yet but on Intel core i5 it is twice faster than 4way
3019 2011-03-03 20:25:57 <mmarker> Raulo: Yea, that's slow
3020 2011-03-03 20:26:02 prax has joined
3021 2011-03-03 20:26:02 prax has quit (Changing host)
3022 2011-03-03 20:26:02 prax has joined
3023 2011-03-03 20:26:04 <lfm> mmarker: is there spozed to be as .asm file?
3024 2011-03-03 20:26:07 <BlueMatt> phantomcircuit: very true, but that is expensive and not _that_ easy, but yes it is a potential problem
3025 2011-03-03 20:26:08 <Raulo> On AMD K10 it is 20% slower than 4way
3026 2011-03-03 20:26:12 <mmarker> lfm: you.
3027 2011-03-03 20:26:14 <mmarker> yup
3028 2011-03-03 20:26:25 <mmarker> lfm: need the SSE2 branch (git checkout -b sse2 sse2
3029 2011-03-03 20:26:31 <phantomcircuit> BlueMatt, well the attack itself is trivial, seriously it's like 15 loc
3030 2011-03-03 20:26:37 <Diablo-D3> [03:14:07] <rgm3> Diablo-D3: why wouldn't I want to use my CPU mining in addition to your GPU miner?
3031 2011-03-03 20:26:44 <Diablo-D3> rgm3: because it usually fucks over gpu mining.
3032 2011-03-03 20:26:51 <mmarker> Raulo: Hmm. Not surprised. The AMD probably has a different SSE unit
3033 2011-03-03 20:26:55 <phantomcircuit> BlueMatt, and getting > 3k ips would be pretty easy also
3034 2011-03-03 20:27:06 <rgm3> well now that you mention it, the GPU mining *does* crash about once / day
3035 2011-03-03 20:27:07 <BlueMatt> phantomcircuit: plus if we prefer opening new connections instead of incoming connections, you would have to have a HUGE percent of the network's ips
3036 2011-03-03 20:27:30 <BlueMatt> which would be hard
3037 2011-03-03 20:27:30 <Diablo-D3> rgm3: yeah which shouldnt happen. which version of cat and the sdk?
3038 2011-03-03 20:27:36 <rgm3> Diablo-D3: by "fucks over" do you mean invalidates its progress?
3039 2011-03-03 20:27:40 <Raulo> It's disappointing because the 4way version was so mach faster on K10 than on Intel's
3040 2011-03-03 20:27:42 <rgm3> uh...
3041 2011-03-03 20:27:54 <phantomcircuit> BlueMatt, what? no id have to have the same number of ips as are in the network, which currently would be pretty easy
3042 2011-03-03 20:27:57 <mmarker> raulo: Need to figure out why
3043 2011-03-03 20:27:58 <rgm3> I'm sorry I don't know what cat means or what SDK you're talking about
3044 2011-03-03 20:28:05 <phantomcircuit> BlueMatt, im sure i can find 2.3k proxies...
3045 2011-03-03 20:28:07 <mmarker> I dont have a K10, so someone would need to profile it
3046 2011-03-03 20:28:07 <Diablo-D3> rgm3: it slows it down greadly
3047 2011-03-03 20:28:13 <Diablo-D3> rgm3: catalyst, stream sdk
3048 2011-03-03 20:28:25 <rgm3> NVIDIA GeForce GT 120 512MB
3049 2011-03-03 20:28:27 <Raulo> I can do it but I don't know how
3050 2011-03-03 20:28:30 <rgm3> that's the Mac Pro.
3051 2011-03-03 20:28:34 <rgm3> it's where I use DiabloMiner
3052 2011-03-03 20:28:47 <Raulo> OK. Validated on K10. Produces correct hash
3053 2011-03-03 20:28:47 <rgm3> no SDK, original mac drivers.
3054 2011-03-03 20:28:53 <Diablo-D3> rgm3: oh, you're on nvidia nad osx
3055 2011-03-03 20:28:55 <Diablo-D3> rgm3: feh
3056 2011-03-03 20:29:07 <Diablo-D3> rgm3: also... holy shit thats a slow card
3057 2011-03-03 20:29:18 <rgm3> :)
3058 2011-03-03 20:29:22 <mmarker> Raulo; yay. But boux on the slowness
3059 2011-03-03 20:29:27 <rgm3> like i said, 10-12 Mhash/s
3060 2011-03-03 20:29:28 <BlueMatt> phantomcircuit: No, if you assume I want to have all connections to you, and we open a bunch of connections instead of accepting a ton of incoming ones, you would need more than the rest of the network to keep the probability low
3061 2011-03-03 20:29:36 <rgm3> with CPU it can bump to about 16
3062 2011-03-03 20:29:52 <Diablo-D3> rgm3: geforce gt 120 = ~134 gflops....
3063 2011-03-03 20:30:03 bk128-Droid has joined
3064 2011-03-03 20:30:52 <gavinandresen> phantomcircuit: catch me up, what's the attack?
3065 2011-03-03 20:31:06 <Diablo-D3> gt220 ~= 192 and does about 10 mhash/sec
3066 2011-03-03 20:31:11 <Diablo-D3> rgm3: and you're getting what?
3067 2011-03-03 20:31:19 <rgm3> i would like to upgrade it, but I don't know much about mac compatability.  I'd want something officially supporetd
3068 2011-03-03 20:31:21 <BlueMatt> gavinandresen: get a bunch of ips diverse around the net and try to control all connections to clients
3069 2011-03-03 20:31:32 <phantomcircuit> BlueMatt, no i would simply have to take up 100% of the connection slots, 117/node, ~3k nodes, 300k connections
3070 2011-03-03 20:32:07 <Diablo-D3> rgm3: you said 10, right?
3071 2011-03-03 20:32:17 <BlueMatt> gavinandresen: has some problems but it does indicate that we should prefer to open more connections on startup than to just accept all incoming ones and get 70 incoming and 4 outgoing
3072 2011-03-03 20:32:18 <Diablo-D3> rgm3: because you're getting higher than what I estimate for your card.
3073 2011-03-03 20:32:35 <rgm3> Diablo-D3: It says "local cu 4"  and it says 5400/5100 khas/sec
3074 2011-03-03 20:32:44 <rgm3> then, I factor in the 5600 khash/sec from the CPU miner
3075 2011-03-03 20:32:44 <BlueMatt> phantomcircuit: Currently, yea no problem after a while you would get those slots, assuming client stays on 24/7
3076 2011-03-03 20:32:53 <Diablo-D3> rgm3: ahh, just 5?
3077 2011-03-03 20:32:56 <Diablo-D3> that sounds about right
3078 2011-03-03 20:33:02 <rgm3> now it says 5500/5100
3079 2011-03-03 20:33:09 <Diablo-D3> you only need one number =P
3080 2011-03-03 20:33:11 <phantomcircuit> BlueMatt, id actually do better with clients going offline, it would open slots up that i could then fill
3081 2011-03-03 20:33:16 <rgm3> okay so i mistakenly assumed those two numbers were per-core or something
3082 2011-03-03 20:33:23 <Diablo-D3> first number is 15 sec avg, second number is forever avg
3083 2011-03-03 20:33:31 <rgm3> sheesh, label them, eh?
3084 2011-03-03 20:33:32 <rgm3> ;)
3085 2011-03-03 20:33:35 <BlueMatt> phantomcircuit: However if you assume the client rejects incoming connections and only makes outgoing ones, you would have to control a large percent of the ips so that a new connection would almost always be opened to you
3086 2011-03-03 20:33:59 <phantomcircuit> BlueMatt, if every peer rejected all incoming connections...
3087 2011-03-03 20:33:59 <BlueMatt> phantomcircuit: But when the client starts the 8 new connections wont be to you unless you have a lot of the network
3088 2011-03-03 20:34:04 <gavinandresen> phantomcircuit: but with nodes randomly connecting you need a pretty darn large percentage to make it an effective attack...
3089 2011-03-03 20:34:19 genjix has quit (Remote host closed the connection)
3090 2011-03-03 20:34:24 <rgm3> | 5731/5437 khash/sec
3091 2011-03-03 20:34:38 <rgm3> so yeah, 5.5ish
3092 2011-03-03 20:34:41 defaced has joined
3093 2011-03-03 20:34:43 <phantomcircuit> gavinandresen, neither of you are getting the attack :P
3094 2011-03-03 20:34:51 <rgm3> plus the CPU = 11.
3095 2011-03-03 20:34:54 <Diablo-D3> rgm3: thats about right... 9.83/192*134 minus 15%
3096 2011-03-03 20:35:00 <gavinandresen> So explain it to us in little words we'd understand.
3097 2011-03-03 20:35:03 <BlueMatt> phantomcircuit: Then exactly what is it?
3098 2011-03-03 20:35:05 <Diablo-D3> er 25%
3099 2011-03-03 20:35:10 <phantomcircuit> i was about to
3100 2011-03-03 20:35:12 <phantomcircuit> jeez
3101 2011-03-03 20:35:15 <BlueMatt> gavinandresen: Did you get gribble's message from me?
3102 2011-03-03 20:35:25 <Diablo-D3> rgm3: and the 25% is the OSX penalty
3103 2011-03-03 20:35:28 <rgm3> looks like i can get a radeon 9800 pro
3104 2011-03-03 20:35:34 <gavinandresen> BlueMatt: yup.  haven't had a chance to glance at your pull request.
3105 2011-03-03 20:35:36 <defaced> is there anywhere that i could purchase bitcoins with say.. my paypal or a cc?
3106 2011-03-03 20:35:51 <Lachesis> defaced, mtgox?
3107 2011-03-03 20:35:57 <defaced> i thought it was down
3108 2011-03-03 20:35:59 <Lachesis> defaced, or #bitcoin-otc
3109 2011-03-03 20:36:01 <Lachesis> ah
3110 2011-03-03 20:36:03 <BlueMatt> gavinandresen: its just your portoption + the 2h delay removed
3111 2011-03-03 20:36:06 <defaced> lachesis ty bro
3112 2011-03-03 20:36:09 <Lachesis> np
3113 2011-03-03 20:36:15 <gavinandresen> phantomcircuit:  (sorry I'm grumpy today, I'm pissed at myself for the -maxsendbuffer screw up)
3114 2011-03-03 20:36:25 anon92 has joined
3115 2011-03-03 20:36:31 <Diablo-D3> rgm3: you can get any card that currently ships in a mac
3116 2011-03-03 20:36:36 <phantomcircuit> gavinandresen, each node will only accept 117 inbound connections, once those slots are full the node rejects inbound connections, there are 3k nodes on the network currently, so ~300k inbound connection slots, if i can fill all of those connection slots then 100% of new nodes will eventually be directed towards the nodes which i control
3117 2011-03-03 20:36:41 <phantomcircuit> get it now?
3118 2011-03-03 20:36:58 <Diablo-D3> rgm3: if its agp, you have to buy a mac specific one if you dont wanna fuck with shoving it in a pc and reflashing the firmware
3119 2011-03-03 20:36:58 <tcatm> gavinandresen: is there any good reaseon why -daemon is ignored in init.cpp?
3120 2011-03-03 20:37:05 <defaced> looks like i gotta dl irc for the first time in 10 years =] feelsgood
3121 2011-03-03 20:37:10 <Diablo-D3> rgm3: but generally, its not fucking worth it
3122 2011-03-03 20:37:11 <rgm3> yeah, i totally don't want to fuck with anything
3123 2011-03-03 20:37:12 <Lachesis> phantomcircuit, yeah, but don't you need at least 117 IPs for that?
3124 2011-03-03 20:37:18 <BlueMatt> phantomcircuit: Yea, but you are still forgetting about the initial outgoing connections
3125 2011-03-03 20:37:19 <Diablo-D3> rgm3: osx sucks extremely bad on opencl
3126 2011-03-03 20:37:26 <Diablo-D3> rgm3: you lose 25% of your speed just by using osx
3127 2011-03-03 20:37:26 <rgm3> i wanna pay somebody money and slap a better card in this box
3128 2011-03-03 20:37:31 <Lachesis> BlueMatt, he's hoping to get new connections
3129 2011-03-03 20:37:33 <rgm3> i understand
3130 2011-03-03 20:37:34 <Lachesis> people who just came online
3131 2011-03-03 20:37:36 <phantomcircuit> Lachesis, yes, but obtaining 117 ips is trivial
3132 2011-03-03 20:37:39 <gavinandresen> tcatm:  I never understood that code, give it a good scrub if you can convince everybody how it should work.
3133 2011-03-03 20:37:40 <Lachesis> phantomcircuit, fair enough
3134 2011-03-03 20:37:41 <Diablo-D3> rgm3: and osx doesnt support a lot of cards.
3135 2011-03-03 20:37:46 <rgm3> i realize that.
3136 2011-03-03 20:37:53 <rgm3> it's kinda sad
3137 2011-03-03 20:38:10 <phantomcircuit> BlueMatt, it's a very slow attack, i would have to wait for the currently connections to fail (or otherwise knock them over)
3138 2011-03-03 20:38:13 <BlueMatt> phantomcircuit: Those 8 outgoing connections need to be most likely to be exclusively your nodes, hence you need to have a large percent of the network
3139 2011-03-03 20:38:19 <tcatm> gavinandresen: k, I'll try to clean it up
3140 2011-03-03 20:38:27 <rgm3> i was hoping there'd be some OSX experts in here that could recommend a fast card
3141 2011-03-03 20:38:29 <phantomcircuit> BlueMatt, you're still not getting the attack
3142 2011-03-03 20:38:38 <Raulo> mmarker: In practice it is 33% faster on Core i5 when all 4 virtual threads are used (as one would use it anyway)
3143 2011-03-03 20:38:46 gasteve has joined
3144 2011-03-03 20:38:48 <phantomcircuit> BlueMatt, by filling up everybody elses connection slots i am guaranteeing that i am 100% of the connectable network
3145 2011-03-03 20:39:08 <phantomcircuit> BlueMatt, it would take 117 ips and a trivial amount of bandwidth to launch the attack
3146 2011-03-03 20:39:09 <BlueMatt> phantomcircuit: Oh, ok
3147 2011-03-03 20:39:16 <phantomcircuit> hell currently i could do this from my lame dsl
3148 2011-03-03 20:39:40 <BlueMatt> phantomcircuit: Ok good point then, yea incoming connections do need to be more carefully protected
3149 2011-03-03 20:39:46 <phantomcircuit> and segment the entire network however i wanted, eventually depending on how well the existing connections last
3150 2011-03-03 20:39:55 <Raulo> mmarker: Apparently the 4way version utlizes the cpus via virtual threads somehow and the difference is smaller than for single CPU. It's good but I had some hopes for 100% speedup as peaple claimed for the Windows version
3151 2011-03-03 20:40:01 <Lachesis> phantomcircuit, why is there a 117 connection limit?
3152 2011-03-03 20:40:06 towerX is now known as tower
3153 2011-03-03 20:40:12 <gasteve> well...finally gave up on solo mining and hooked switched my 5970 to slush's pool
3154 2011-03-03 20:40:17 <phantomcircuit> Lachesis, to prevent my previous memory exhaustion attack
3155 2011-03-03 20:40:21 <phantomcircuit> Lachesis, hehe
3156 2011-03-03 20:40:22 <Lachesis> phantomcircuit, lol
3157 2011-03-03 20:40:30 <Lachesis> so how about this
3158 2011-03-03 20:40:42 <Lachesis> instead of rejecting new connections, boot old ones occasionally
3159 2011-03-03 20:40:45 <Raulo> mmarker: On K10 it is slow, though and it will probably hard to make the code more optimal on K10.
3160 2011-03-03 20:41:00 <phantomcircuit> Lachesis, that would actually make this attack easier
3161 2011-03-03 20:41:07 purpleposeidon has joined
3162 2011-03-03 20:41:11 <BlueMatt> phantomcircuit: So, solution would be something like limiting incoming connections from ip ranges? Assuming no one can get access to a large number of ips not in the same blocks
3163 2011-03-03 20:41:18 <mmarker> Raulo: Need to see what CPUs people have. Intel may be better in this regard.
3164 2011-03-03 20:41:22 <phantomcircuit> since im going to be trying to reconnection far more frequently the newly opened slots would just go to me
3165 2011-03-03 20:41:28 <Lachesis> BlueMatt, or just preferring them
3166 2011-03-03 20:41:33 <BlueMatt> phantomcircuit: Which I think getting 117 ips on completely separate blocks might be harder than you think
3167 2011-03-03 20:41:37 <phantomcircuit> BlueMatt, actually that's not gonna help either
3168 2011-03-03 20:41:39 <Lachesis> i.e. prefer to connect to a ton of different net blocks
3169 2011-03-03 20:41:40 <Lachesis> why?
3170 2011-03-03 20:41:48 <Lachesis> also, raise the number to 1000
3171 2011-03-03 20:41:51 <Lachesis> or >
3172 2011-03-03 20:41:52 <gavinandresen> phantomcircuit: have you done a back-of-the-envelope on how much traffic you'll have to sustain?  If the entire bitcoin network is connected through you, just initial block downloads is going to be a lot of GB of data....
3173 2011-03-03 20:41:53 <BlueMatt> Lachesis: how do you "prefer" an incomming connection? reject ones?
3174 2011-03-03 20:41:54 <phantomcircuit> because i can get 117 ips in random blocks
3175 2011-03-03 20:42:06 <phantomcircuit> gavinandresen, it's huge
3176 2011-03-03 20:42:12 <Lachesis> BlueMatt, boot old ones from a redundant netblock
3177 2011-03-03 20:42:13 <mmarker> I need to read a little about the K10 to see what may be going wrong. The actual assembly is pretty small
3178 2011-03-03 20:42:24 <BlueMatt> Lachesis: oh, ok yea
3179 2011-03-03 20:42:25 <gavinandresen> So you spend a lot of money on bandwidth to... do what?
3180 2011-03-03 20:42:35 <Lachesis> gavinandresen, control the network
3181 2011-03-03 20:42:46 <Lachesis> segment it, more accurately
3182 2011-03-03 20:43:14 <gavinandresen> Ok.  So it is segmented, and half the people only see half the transactions (or whatever)...
3183 2011-03-03 20:43:18 <phantomcircuit> gavinandresen, it would make it so that instead of having > 50% of the CPu time (or GPU time whatever)
3184 2011-03-03 20:43:32 <phantomcircuit> i would just have to have more CPU time than the fastest node
3185 2011-03-03 20:43:44 <phantomcircuit> and i could do all kinds of shit like double spending
3186 2011-03-03 20:43:44 <Raulo> mmarker: My i5 is 1.19 MH/s per GHz per physical core in the sse2_64 version (was 0.89 for 4way)
3187 2011-03-03 20:43:53 <Lachesis> phantomcircuit, would raising the number of permitted inbound connections help?
3188 2011-03-03 20:44:01 <phantomcircuit> Lachesis, not really
3189 2011-03-03 20:44:04 <BlueMatt> Now Im curious, what was the -maxsendbuffer error?
3190 2011-03-03 20:44:13 <mmarker> Raulo: that's good.
3191 2011-03-03 20:44:17 <Lachesis> surely some nodes could handle 10,000 connections or something similar
3192 2011-03-03 20:44:21 <Lachesis> what made us pick 117?
3193 2011-03-03 20:44:23 <phantomcircuit> Lachesis, and it would also make the previous memory usage bug likely to trigger on older system
3194 2011-03-03 20:44:23 <Raulo> mmarker: and my K10 Opteron 2.6 GHz is 1.12 MH/s per physical core in 4 way
3195 2011-03-03 20:44:24 <phantomcircuit> s
3196 2011-03-03 20:44:28 <Lachesis> besides Halo?
3197 2011-03-03 20:44:35 anon92 has quit (Quit: anon92)
3198 2011-03-03 20:44:35 <gavinandresen> phantomcircuit:  ... so you'd generate blocks with just creation transactions in them and transmit to both sides of the split....
3199 2011-03-03 20:44:38 <Raulo> K10 is good too
3200 2011-03-03 20:44:47 <gavinandresen> (you obviously can't generate with any real transactions becuase they'd be rejected)
3201 2011-03-03 20:44:48 <BlueMatt> Lachesis: no, managing connections can create a load on router, especially cheap ones
3202 2011-03-03 20:44:49 <gasteve> anyone know why it takes so many blocks for slush's pool to confirm?  seems like its around 12-15 hours or so before he considers it confirmed...that seems like overkill to me...if you're well connected to the network, it would seem 10-20 blocks would be a sufficient confirmation...is slush just being extra paranoid?  or is there some good reason for it?
3203 2011-03-03 20:44:50 <mmarker> Raulo: Yea, this sounds like I need to read up on the K10 architecture
3204 2011-03-03 20:45:08 <Raulo> Correction: K10 is 1.12 MH/s per physical core per 1GHz
3205 2011-03-03 20:45:21 <Lachesis> BlueMatt, yeah, but some people won't have crappy old boxes with no ram on old routers
3206 2011-03-03 20:45:35 <BlueMatt> Lachesis: so...break the clients for those who do?
3207 2011-03-03 20:45:38 <Lachesis> no
3208 2011-03-03 20:45:39 <phantomcircuit> gavinandresen, no id just block peers that generate valid blocks from transmitting to other peers, thus i would only have to be generating blocks faster than any one segment (beating ArtForz would be hard :P) to control the network
3209 2011-03-03 20:45:51 <Lachesis> pick inbound connection numbers based on local benchmarks
3210 2011-03-03 20:45:53 <Lachesis> or as a config optin
3211 2011-03-03 20:45:55 <Lachesis> option
3212 2011-03-03 20:46:12 <phantomcircuit> Lachesis, it is a config option, but it's not really going to help
3213 2011-03-03 20:46:14 <BlueMatt> Lachesis: ok, but that still leaves those with default clients as able to be screwed
3214 2011-03-03 20:46:18 <Lachesis> BlueMatt, no
3215 2011-03-03 20:46:18 <Raulo> mmarker: Funny thing. Those fast results for K10 are with Intel Compiler. Apparently Intel compiler is better in producing code for K10 than for Intel's processors
3216 2011-03-03 20:46:19 <gavinandresen> phantomcircuit:  ah, I see, you'd transmit new transactions but block new blocks.
3217 2011-03-03 20:46:23 <Lachesis> he couldn't fill all the slots on a supernode
3218 2011-03-03 20:46:32 <Lachesis> would need 10000 ips
3219 2011-03-03 20:46:51 <phantomcircuit> gavinandresen, exactly, thus only having to be faster than the fastest segment instead of faster than 50% of the network
3220 2011-03-03 20:46:53 <Lachesis> so it could act as a hub to let others flow around it
3221 2011-03-03 20:47:15 <slush> gasteve: I'm not paranoid, this is enforced by the bitcoin network
3222 2011-03-03 20:47:15 <Lachesis> phantomcircuit, ^^ supernodes?
3223 2011-03-03 20:47:17 <gavinandresen> But I still don't see how you could double-spend if you're transmitting transactions.
3224 2011-03-03 20:47:23 <phantomcircuit> Lachesis, the problem is that obtaining ips isn't hard
3225 2011-03-03 20:47:26 <mmarker> Raulo: Lemme guess, GCC is slower?
3226 2011-03-03 20:47:29 <slush> gasteve: You can spend mined coins after 100 confirmations
3227 2011-03-03 20:47:30 <mmarker> I can tell you why.
3228 2011-03-03 20:47:31 <Lachesis> phantomcircuit, how many can you obtain?
3229 2011-03-03 20:47:43 <slush> gasteve: And the client itself enforce 20 confirmations more
3230 2011-03-03 20:47:43 <Lachesis>  / how many can i obtain right now?
3231 2011-03-03 20:47:45 <BlueMatt> gavinandresen: You dont have to transmit txes, you can chose what you want to
3232 2011-03-03 20:47:59 <phantomcircuit> Lachesis, i can buy time on a botnet right now for pennies a day
3233 2011-03-03 20:48:04 <mmarker> Raulo: I've seen this with ARM. the 4way code uses SSE intrinsics, so the compiler handles assigning registers and the like
3234 2011-03-03 20:48:05 <phantomcircuit> Lachesis, so a lot
3235 2011-03-03 20:48:11 <Raulo> mmarker: Slower for K10 than i5? No. But slower than Intel Compiler
3236 2011-03-03 20:48:12 <phantomcircuit> Lachesis, especially if im making $ doing it
3237 2011-03-03 20:48:15 <Lachesis> phantomcircuit, how much is a lot? 10k? 100k?
3238 2011-03-03 20:48:21 <mmarker> gcc just blows when it comes to it. Adds lots of code that isn't necessary.
3239 2011-03-03 20:48:27 <gavinandresen> ... but if you feed different transactions to either side of the network but are feeding blocks to both sides... then you have to choose which of your double-spends to put into your blocks
3240 2011-03-03 20:48:29 <phantomcircuit> Lachesis, upwards of 100k
3241 2011-03-03 20:48:38 <Lachesis> jeez
3242 2011-03-03 20:48:41 <gasteve> slush: ah, is that the rate limiting spam transaction defense?
3243 2011-03-03 20:48:41 <Lachesis> that's a lot of the internet in bots
3244 2011-03-03 20:48:42 <bk128-Droid> gasteve: when you generate a block, you have to wait for it to mature.  Takes longer than a normal transfer confirmation
3245 2011-03-03 20:48:44 <gavinandresen> (unless you want your transactions to never be confirmed, which would be stupid)
3246 2011-03-03 20:48:45 <mmarker> Intel's compiler probably does a better job handling the intrinsics.
3247 2011-03-03 20:48:49 EvanR has quit (Ping timeout: 276 seconds)
3248 2011-03-03 20:48:51 <BlueMatt> gavinandresen: Youd have to, say pick one client, and make special blocks for that one
3249 2011-03-03 20:48:53 <mmarker> hence, faster 4way code
3250 2011-03-03 20:48:56 <phantomcircuit> Lachesis, that's actually a tiny fraction of the internet :P
3251 2011-03-03 20:49:05 <Lachesis> considering we only have 4billion ips
3252 2011-03-03 20:49:09 <phantomcircuit> Lachesis, however pennies a day * 100k is a lot
3253 2011-03-03 20:49:18 <Lachesis> oh pennies / compjuter
3254 2011-03-03 20:49:22 <Lachesis> not per 100k :)
3255 2011-03-03 20:49:22 <BlueMatt> gavinandresen: Buy your stuff, make blocks for that client, then let that one back on the real chain when you get your stuff
3256 2011-03-03 20:49:23 <phantomcircuit> so id have to be making significant profits on double spending
3257 2011-03-03 20:49:43 <Raulo> I wonder how the compiler 4way asm code compares with this hand written
3258 2011-03-03 20:49:56 <phantomcircuit> 4way asm doesn't scale so well
3259 2011-03-03 20:49:57 <slush> gasteve: no, it's not related to spam protection. But block must be matured enough to be spent. Prevention of blockchain forking, I think
3260 2011-03-03 20:49:57 <mmarker> raulo: It's probably similar
3261 2011-03-03 20:49:59 <BlueMatt> phantomcircuit: Yes, it could be a problem, but the bandwidth requirements are pretty big...In any case anyone have suggestions on how to fix it
3262 2011-03-03 20:50:14 <Raulo> mmarker: But somehow it is 20% faster for K10 :)
3263 2011-03-03 20:50:16 EvanR has joined
3264 2011-03-03 20:50:18 <mmarker> but the code really minimizes the #of memory reads and writes
3265 2011-03-03 20:50:24 <gavinandresen> BlueMatt:  yeah, that sybil attack works.  I'd actually like to see the major exchanges and merchants connect to each other directly to avoid that.
3266 2011-03-03 20:50:25 <Lachesis> BlueMatt, phantomcircuit, i think preferring connections from diverse netblocks is a good idea either way
3267 2011-03-03 20:50:28 <mmarker> which are slow
3268 2011-03-03 20:50:34 <mmarker> AND, I use more registers
3269 2011-03-03 20:50:38 <mmarker> which is also faster
3270 2011-03-03 20:50:39 <phantomcircuit> BlueMatt, the client has to be more efficient about handling peers, currently the baseline problem is that a huge amount of memory can be used with garbage
3271 2011-03-03 20:50:41 <BlueMatt> Lachesis: yea
3272 2011-03-03 20:50:44 Spenvo has joined
3273 2011-03-03 20:50:55 <Lachesis> also what phantomcircuit just said, and gavinandresen
3274 2011-03-03 20:51:03 <BlueMatt> Kicking out old connections when new connections come in is an ok idea, but that many new connections/killing old ones isnt really a good idea
3275 2011-03-03 20:51:07 <gasteve> slush: quite interesting...so much to learn
3276 2011-03-03 20:51:36 <phantomcircuit> gavinandresen, reserved slots would pretty much reduce this attack to nothing
3277 2011-03-03 20:51:41 <phantomcircuit> gavinandresen, i hadn't thought of that
3278 2011-03-03 20:51:59 <Lachesis> however, that undecentralized bitcoin :)
3279 2011-03-03 20:51:59 <mmarker> The nice thing is this code should translate to my ARM toy project, and may get "faster" per tick still
3280 2011-03-03 20:52:04 <Lachesis> not completely, obviously
3281 2011-03-03 20:52:10 <phantomcircuit> Lachesis, well not really
3282 2011-03-03 20:52:23 <gavinandresen> I haven't looked at the bitcoin -addnode code-- does it try to reconnect if the -added node disconnects?
3283 2011-03-03 20:52:25 <BlueMatt> Yea, that would beat some of the decentralization
3284 2011-03-03 20:52:26 <Raulo> mmarker: I can send you the Intel compiler asm code if you like. You might have some ideas. Unfortunately I have zero experience with assembler
3285 2011-03-03 20:52:27 <mmarker> (arm has some instructions x86 doesn't...which are applicable to calculating SHA-256)
3286 2011-03-03 20:52:31 <Lachesis> phantomcircuit, it moves it from a pure flat protocol toward a federated one
3287 2011-03-03 20:52:39 <phantomcircuit> Lachesis, if the top generators reserve slots for each other id have to take on all of them as a block
3288 2011-03-03 20:52:39 <Lachesis> obviously still not centralized
3289 2011-03-03 20:52:40 <mmarker> Raulo: post it on gist.github.com
3290 2011-03-03 20:52:54 <Lachesis> phantomcircuit, why?
3291 2011-03-03 20:52:58 <Raulo> OK. Need to figure how.
3292 2011-03-03 20:53:05 <Lachesis> why not just ignore them and try to trick some merchants
3293 2011-03-03 20:53:21 <phantomcircuit> Lachesis, segmenting the network is only useful if you have more CPU time than any of the segments
3294 2011-03-03 20:53:33 <BlueMatt> Or if you have the inbuilt seed list and connect to one of them it could prevent the attack which would keep anyone from trying which would keep the decentralization
3295 2011-03-03 20:53:46 <phantomcircuit> Lachesis, yeah i guess that would work
3296 2011-03-03 20:54:02 <mmarker> Pretty simple. it's like pastie
3297 2011-03-03 20:54:05 <Lachesis> phantomcircuit, it seems like you just need to somehow control all of one guy's links to the network
3298 2011-03-03 20:54:06 <phantomcircuit> BlueMatt, an inbuilt seed list cant stop this attack
3299 2011-03-03 20:54:15 <BlueMatt> Assuming the seeds would have unlimited incoming connections
3300 2011-03-03 20:54:15 <phantomcircuit> Lachesis, yeah you're right actually
3301 2011-03-03 20:54:24 <gavinandresen> Sybil attack detection is something I've been thinking about; if a node experiences a sudden change in average block generation rate then perhaps it should drop connections and connect to some seed nodes, or go into "safe mode" where a human can try to figure out what's going on.
3302 2011-03-03 20:54:27 <BlueMatt> which they obviously wont, but it would increase the cost
3303 2011-03-03 20:54:28 <Lachesis> ah, what bluematt said
3304 2011-03-03 20:54:34 <mmarker> need to go, like now...oops!
3305 2011-03-03 20:54:35 mmarker has quit (Quit: WeeChat 0.3.2)
3306 2011-03-03 20:54:36 <phantomcircuit> Lachesis, yeah sounds like a web of trust basically
3307 2011-03-03 20:54:46 <Lachesis> that would do the same thing, though
3308 2011-03-03 20:54:57 <Lachesis> if they had unlimited connections, they would be expensive to run
3309 2011-03-03 20:54:58 <gasteve> it figures that when I join slush's pool that now it's taking a while to find a block...seems I've brought my solo mining luck over to the pool   :\
3310 2011-03-03 20:55:00 <Lachesis> and different from other nodes
3311 2011-03-03 20:55:03 <BlueMatt> I agree with gavinandresen, though maybe a combination would be best?
3312 2011-03-03 20:55:12 <gavinandresen> Defense in depth is always good
3313 2011-03-03 20:55:14 mmarker has joined
3314 2011-03-03 20:55:38 <BlueMatt> Detection would be pretty easy assuming blocks aren't getting passed
3315 2011-03-03 20:55:56 <Lachesis> gavinandresen, acting in response to a sudden drop in generation rate might be dangerous
3316 2011-03-03 20:56:06 <BlueMatt> Lachesis: why?
3317 2011-03-03 20:56:23 <BlueMatt> Currently yea slush's pool goes down: bad but as more pools come online and more clients...
3318 2011-03-03 20:56:23 <slush> gasteve: Thanks to score based system, it's better to jump inside the round for you :)
3319 2011-03-03 20:56:27 <gavinandresen> Yup; I wish (again) I'd taken statistics, I'm sure there's a way to calculate probability that a drop in generation rate is natural variation versus sybil attack
3320 2011-03-03 20:56:43 <slush> BlueMatt: going down?
3321 2011-03-03 20:56:48 <gavinandresen> (and then figure out how long a very, very slow sybil attack might take)
3322 2011-03-03 20:56:49 <Lachesis> if we do have a legitimate rapid fall in hashrate, it could provide downward feedback, knocking more generators offline and causing more problems
3323 2011-03-03 20:56:51 <phantomcircuit> anyways
3324 2011-03-03 20:57:04 <BlueMatt> slush: ie all your miners go offline
3325 2011-03-03 20:57:04 <phantomcircuit> these are all attacks i've found in my spare time
3326 2011-03-03 20:57:13 <phantomcircuit> (spare time lol get it all my time is spare!)
3327 2011-03-03 20:57:18 <gavinandresen> "safe mode" means don't generate any transactions, not "stop mining"
3328 2011-03-03 20:57:33 <slush> BlueMatt: I don't understand... are 900 connected workers "all miners go offline"?
3329 2011-03-03 20:57:34 <Lachesis> gavinandresen, ah fair enough
3330 2011-03-03 20:57:46 <BlueMatt> gavinandresen: doesn't have to, could just not accept txes as "confirmed"
3331 2011-03-03 20:58:01 <Lachesis> BlueMatt, would be nice to warn the user too
3332 2011-03-03 20:58:09 <BlueMatt> slush: I was referring to a sudden drop in network hashing potentially being caused by your server going down
3333 2011-03-03 20:58:12 <phantomcircuit> slush, you think your pool would shrink if you required specialized software?
3334 2011-03-03 20:58:17 <Lachesis> "Warning! The Bitcoin network is behaving strangely."
3335 2011-03-03 20:58:22 <gavinandresen> BlueMatt:  right, my mistake-- you care about receiving, not sending.
3336 2011-03-03 20:58:22 <slush> BlueMatt: ?
3337 2011-03-03 20:58:24 <BlueMatt> Yea, I agree
3338 2011-03-03 20:58:28 <gasteve> slush: I need to read that paper about the attack and share vs score...but on the surface, it didn't seem obvious to me why dropped from a round would be any particular issue...I think it claimed that dropping a ~40% into the round was optimal for cheeting...however, if people drop at 40%, then you are removing compute power, thereby extending the round...so you would have actually dropped say only 30% into the round
3339 2011-03-03 20:58:46 <BlueMatt> slush: In the attack conversation, not related to what you were saying, and not directed at you...
3340 2011-03-03 20:58:50 <Lachesis> slush, how's pushwork coming?
3341 2011-03-03 20:58:51 <phantomcircuit> slush, cause i took one of luke-jr's ideas and more or less have a functional miner proxy for pools
3342 2011-03-03 20:58:56 <BlueMatt> slush: sorry to include your name
3343 2011-03-03 20:59:04 <Lachesis> slush, SO MANY PINGS!
3344 2011-03-03 20:59:15 <slush> hehe :)
3345 2011-03-03 21:00:01 <slush> BlueMatt: oh.
3346 2011-03-03 21:00:13 <slush> BlueMatt: the drop in the global hashrate was not caused by my pool in any case
3347 2011-03-03 21:00:15 <Diablo-D3> [03:49:59] <Lachesis> "Warning! The Bitcoin network is behaving strangely."
3348 2011-03-03 21:00:20 <Diablo-D3> hows that different from normal?
3349 2011-03-03 21:00:21 <Diablo-D3> dohohoho
3350 2011-03-03 21:00:23 <Lachesis> Diablo-D3, lol
3351 2011-03-03 21:00:43 <gasteve> (so, if someone else dropped from the pool and you stay in...the round would be extended due to the loss of compute power and you would pick up more shares as a result)
3352 2011-03-03 21:00:52 <slush> Lachesis: I'm finishing one project from my daily job this week, so no progress in push for now
3353 2011-03-03 21:01:01 <gasteve> (need to read that paper on pool cheating more closely)
3354 2011-03-03 21:01:34 <slush> gasteve: yes, read the paper before :)
3355 2011-03-03 21:02:17 <slush> phantomcircuit: If you're referring "getwork proxy", it is temporary solution
3356 2011-03-03 21:02:41 <slush> phantomcircuit: as there will be some agreement between developers, there is no problem to standardize this interface and include it into all miners
3357 2011-03-03 21:03:02 <phantomcircuit> slush, yeah someone else already wrote a getwork proxy?
3358 2011-03-03 21:03:05 <slush> phantomcircuit: current getwork protocol is really horrible from side of effectivity and should be deprecated soon
3359 2011-03-03 21:03:29 <slush> phantomcircuit: I don't know
3360 2011-03-03 21:04:53 <BlueMatt> What was this -maxsendbuffer error?
3361 2011-03-03 21:05:09 <phantomcircuit> slush, basically the miner needs an address/version/diff from the pool via a push protocol, to send shares to the pool, and to get merkle root from a local peer
3362 2011-03-03 21:05:12 <phantomcircuit> slush, right?
3363 2011-03-03 21:05:33 <slush> no
3364 2011-03-03 21:05:45 <slush> he will receive merkle from the server, too
3365 2011-03-03 21:06:02 <Diablo-D3> they slush
3366 2011-03-03 21:06:14 <phantomcircuit> slush, any particular reason why?
3367 2011-03-03 21:06:16 <Diablo-D3> how did you get bitcoin to insert the poison transaction?
3368 2011-03-03 21:06:28 <phantomcircuit> that would seem to substantially increase the traffic
3369 2011-03-03 21:06:32 <slush> phantomcircuit: any particular reason why not?
3370 2011-03-03 21:06:46 <phantomcircuit> address/version/diff are all static
3371 2011-03-03 21:06:56 <phantomcircuit> which would leave only sending shares for verification
3372 2011-03-03 21:07:03 <slush> phantomcircuit: in the worse case, one packet for every new bitcoin block for every miner is enough
3373 2011-03-03 21:07:18 <Lachesis> what all is needed to mine?
3374 2011-03-03 21:07:26 <phantomcircuit> oh you only send out a new merkle root when a new block is started?
3375 2011-03-03 21:07:34 <BlueMatt> would someone mind compiling with the default port set to 9635 so that I can make sure portoption works 100%?
3376 2011-03-03 21:07:41 <phantomcircuit> that might explain why it takes so long for tx's to get into a block xD
3377 2011-03-03 21:07:59 <Lachesis> what does the merkle root encode?
3378 2011-03-03 21:08:01 <slush> phantomcircuit: I also considered to calculate merkles locally on every miner, but it is much more complicated. For example, I need all source data for the block to check share validity
3379 2011-03-03 21:08:20 <slush> Diablo-D3: what do you mean?
3380 2011-03-03 21:08:36 <phantomcircuit> slush, yeah i guess that's right, you'd need a list of all the tx's in the share
3381 2011-03-03 21:08:42 <phantomcircuit> fair enough
3382 2011-03-03 21:08:57 <slush> phantomcircuit: yes, which is MUCH MORE traffic than sending just a merkle to the miner
3383 2011-03-03 21:09:02 <Diablo-D3> slush: how do you insert the poisioned transaction?
3384 2011-03-03 21:09:12 <slush> Diablo-D3: I'm doing something like this?
3385 2011-03-03 21:09:18 <phantomcircuit> slush, i thought you were updating the merkle more regularly than that
3386 2011-03-03 21:09:29 <Diablo-D3> slush: uh, yes, how else do you stop me from reusing the block
3387 2011-03-03 21:09:48 <phantomcircuit> Lachesis, it's a hash of all the tx's in the block
3388 2011-03-03 21:10:00 <slush> phantomcircuit: currently the miner needs frequent updates.
3389 2011-03-03 21:10:12 <phantomcircuit> slush, for what?
3390 2011-03-03 21:10:14 <luke-jr> slush: I have suggested to combine the effort to standardize a miner protocol to work for clients as well, FYI
3391 2011-03-03 21:10:20 <slush> phantomcircuit: once the server will proactively send new job to them, the 5s ask rate won't be necessary
3392 2011-03-03 21:10:24 <luke-jr> slush: where can I read the current progress for mining btw?
3393 2011-03-03 21:10:55 <phantomcircuit> slush, version/prev_hash/merkle_root/bits, the miner can generate the timestamp/nonce themself
3394 2011-03-03 21:11:05 <slush> phantomcircuit: but it is more complicated
3395 2011-03-03 21:11:15 <phantomcircuit> just start each miner at a different nonce offset
3396 2011-03-03 21:11:31 <slush> Diablo-D3: stopping to what? You can submit the share from the miner directly, no problem here
3397 2011-03-03 21:11:47 <slush> Diablo-D3: well, you cannot, because you don't have source data of the block
3398 2011-03-03 21:11:57 <Diablo-D3> slush: thats the _only_ thing stopping me?
3399 2011-03-03 21:12:07 <slush> Diablo-D3: are you serious? :)
3400 2011-03-03 21:12:19 <slush> Diablo-D3: I expect that you know how is the stuff working :)
3401 2011-03-03 21:12:21 <phantomcircuit> what poison transaction?
3402 2011-03-03 21:12:35 <slush> Diablo-D3: there is 50BTC reward to the pool inside the merkle
3403 2011-03-03 21:12:50 <slush> Diablo-D3: submitting it to the bitcoin network directly does not bring you anything
3404 2011-03-03 21:13:21 <Diablo-D3> slush: no, I mean, I thought the method was more complex than that
3405 2011-03-03 21:13:22 <phantomcircuit> slush, hehe it might if someone has a custom broken client
3406 2011-03-03 21:13:22 <phantomcircuit> xD
3407 2011-03-03 21:14:58 <slush> Diablo-D3: Are you talking about current implementation or about push protocol?
3408 2011-03-03 21:15:09 <Lachesis> phantomcircuit, sure you can generate the nonce yourself
3409 2011-03-03 21:15:19 <Lachesis> but you need to be kept apprised of when new txn are added by the pool or the block changes
3410 2011-03-03 21:15:56 <Diablo-D3> slush: current impl
3411 2011-03-03 21:16:01 <phantomcircuit> i thought slush just said the merkle_root is only updated when a new block is found
3412 2011-03-03 21:16:19 <Diablo-D3> phantomcircuit: no
3413 2011-03-03 21:16:21 <slush> phantomcircuit: yes, in the most optimized case (from side of network transfers)
3414 2011-03-03 21:16:29 <Diablo-D3> phantomcircuit: its also updated when a new potential tx is added
3415 2011-03-03 21:16:33 <phantomcircuit> Lachesis, either way pushing the merkle_root would be trivially expensive
3416 2011-03-03 21:16:36 <slush> phantomcircuit: in the normal fashion, miners will receive new merkle every minute
3417 2011-03-03 21:16:41 <Lachesis> phantomcircuit, yes, but that requires push support
3418 2011-03-03 21:16:50 <slush> phantomcircuit: the new merkle will include new bitcoin transactions
3419 2011-03-03 21:17:00 <Lachesis> instead of the miners polling
3420 2011-03-03 21:17:24 <Raulo> mmarker: https://gist.github.com/853566
3421 2011-03-03 21:17:27 <Lachesis> there are two nonces, right?
3422 2011-03-03 21:17:34 <Lachesis> nNonce and nExtraNonce
3423 2011-03-03 21:17:39 defaced has left ()
3424 2011-03-03 21:17:46 <Lachesis> nNonce is incremented by the miner and nExtraNonce makes sure each miner gets a different block to work on?
3425 2011-03-03 21:17:50 <phantomcircuit> slush, ok so the pool has to be able to push address/version/root/diff/nonce slice
3426 2011-03-03 21:18:01 <slush> The main concern of push is a) Miners does not need polling, they will be notified instantly when merkle changed b) Pool will drive the load for itself, so one machine will be able to handle much more clients
3427 2011-03-03 21:18:16 <slush> Lachesis: exactly
3428 2011-03-03 21:18:31 <slush> Lachesis: every miner will receive unique merkle, thanks to specific extranonce for every miner
3429 2011-03-03 21:18:34 <phantomcircuit> lol nExtraNonce?
3430 2011-03-03 21:18:36 <Lachesis> cool
3431 2011-03-03 21:18:45 <Lachesis> it's in the getwork code - i was reading it the other day
3432 2011-03-03 21:18:58 <phantomcircuit> slush, the merkle for each minor is unique?
3433 2011-03-03 21:19:00 <phantomcircuit> slush, wat
3434 2011-03-03 21:19:04 <slush> phantomcircuit: yes
3435 2011-03-03 21:19:11 <phantomcircuit> how/why
3436 2011-03-03 21:19:18 <slush> phantomcircuit: to avoid crunching duplicated work
3437 2011-03-03 21:19:18 larsig has joined
3438 2011-03-03 21:19:41 <Diablo-D3> slush: so the only thing that stops me from stealing winning shares is its non-trivial to recreate the header?
3439 2011-03-03 21:19:42 <phantomcircuit> slush, where is the extranonce ?
3440 2011-03-03 21:19:47 <phantomcircuit> is it in a transaction?
3441 2011-03-03 21:19:49 <slush> phantomcircuit: miner will receive merkle/midstate/target/whatever and will change ntime and iterate nonce
3442 2011-03-03 21:20:16 <Lachesis> does m0mchil's miner change ntime right now?
3443 2011-03-03 21:20:47 <slush> Diablo-D3: well, you need source data of the block. And in the block source there is TX for 50BTC targeted to pool private key
3444 2011-03-03 21:20:55 <mmarker> This with the Intel compiler, Raulo?
3445 2011-03-03 21:21:15 <slush> Diablo-D3: So when you find a block in your miner, "recreate header" in some way and push it to bitcoin network, you don't receive anything
3446 2011-03-03 21:21:20 <slush> Diablo-D3: but pool receive the reward
3447 2011-03-03 21:21:28 <phantomcircuit> slush, where is the miner specific nonce?
3448 2011-03-03 21:21:36 <mmarker> Oh, yes, it is.
3449 2011-03-03 21:21:50 <slush> Lachesis: ntime change was removed from poclbm, because it does problems with pool mining
3450 2011-03-03 21:21:59 <mmarker> Raulo: so this runs faster on the Athlon K10 than my code
3451 2011-03-03 21:22:05 <Diablo-D3> slush: er, so the poisoned header is in the block already from the client?
3452 2011-03-03 21:22:06 <slush> Lachesis: but with the push protocol, ntime changes will be absolutely necessary
3453 2011-03-03 21:22:07 <mmarker> By how much?
3454 2011-03-03 21:22:10 <Lachesis> slush, due to out of date clock?
3455 2011-03-03 21:22:11 <gasteve> "recreate header" basically means you have to find a new solution...so there's no benefit in that over solo mining in the first place
3456 2011-03-03 21:22:21 <Lachesis> i.e. my clock being 20 minutes faster than yours?
3457 2011-03-03 21:22:40 <Diablo-D3> Lachesis: if your clock is that far off, you shouldnt even be on the internet
3458 2011-03-03 21:22:44 <Lachesis> Diablo-D3, lol
3459 2011-03-03 21:22:48 <slush> Diablo-D3: what is "poisoned header"? If you mean "header with 50 BTC TX to the pool", then yes
3460 2011-03-03 21:22:54 <phantomcircuit> Diablo-D3, lots of clocks are off by that much...
3461 2011-03-03 21:23:05 <Diablo-D3> lots of random shit breaks when both ends are further apart than 10 minutes
3462 2011-03-03 21:23:17 <Lachesis> slush, what problems does changing ntime cause?
3463 2011-03-03 21:23:18 <slush> Lachesis: server will send you "official ntime"
3464 2011-03-03 21:23:19 <Diablo-D3> mine is ntp set.
3465 2011-03-03 21:23:26 <slush> Lachesis: then the miner will just increase it
3466 2011-03-03 21:23:29 <Lachesis> Diablo-D3, mine too
3467 2011-03-03 21:23:49 <phantomcircuit> slush, offsets are so much fun
3468 2011-03-03 21:23:51 <Diablo-D3> slush: well, how does the "mint new coins" tx work? it has the winning person's address in it?
3469 2011-03-03 21:24:01 <Lachesis> Diablo-D3, no, it has the pool's address in it
3470 2011-03-03 21:24:11 <Diablo-D3> Lachesis: the pool IS the winning person
3471 2011-03-03 21:24:22 <Lachesis> then yes
3472 2011-03-03 21:24:39 <Diablo-D3> okay, so, new block found, gen new address, new block found, etc?
3473 2011-03-03 21:24:39 <phantomcircuit> Diablo-D3, the first tx in every block is special, it mints 50 (or whatever it is at that point) BTC
3474 2011-03-03 21:24:50 <Diablo-D3> phantomcircuit: yes Im aware of that
3475 2011-03-03 21:24:51 <Lachesis> Diablo-D3, i'm kinda a freak about time actually - i have a watch that sets to atomic time, a phone that can set to either NTP, CDMA, or GPS time, 4 different computers syncing to ntp time, and a old desk clock that i reset every few days because it's a few seconds off
3476 2011-03-03 21:24:54 <x6763> phantomcircuit: the extranonce is in a transaction (typically the coin generation transaction) since there's no place for it in the block header...changing the extranonce changes a transaction hash, which changes the merkle root
3477 2011-03-03 21:25:00 <slush> Lachesis: yes, miner just need take diff from server time and local time. Unless the miner's second takes same time as server's second, it won't make the troubles :)
3478 2011-03-03 21:25:17 <Diablo-D3> Lachesis: my watch is 1 hour off two months out of the year.
3479 2011-03-03 21:25:22 <Lachesis> lol
3480 2011-03-03 21:25:28 <Lachesis> that would kill me
3481 2011-03-03 21:25:39 <Diablo-D3> Lachesis: it automatically does daylight savings.... on pre-Bush dates.
3482 2011-03-03 21:25:46 <Diablo-D3> it keeps perfect time though
3483 2011-03-03 21:25:52 <Diablo-D3> havent had to set it for like 3 years
3484 2011-03-03 21:26:09 <phantomcircuit> x6763, lol i take it at some point there was a block for which there was no solution and the extra nonce was thrown into the transaction?
3485 2011-03-03 21:26:28 <phantomcircuit> x6763, that sounds like a hack to me :P
3486 2011-03-03 21:26:39 <slush> phantomcircuit: x6763 is correct
3487 2011-03-03 21:26:49 <phantomcircuit> slush, k
3488 2011-03-03 21:26:52 <Diablo-D3> slush: so, when a new block is submitted to the pool, you also new address at the same time?
3489 2011-03-03 21:26:54 <x6763> phantomcircuit: yeah, sounds like a hack, but yeah, the nonce in the block header overflows quite often
3490 2011-03-03 21:27:03 <phantomcircuit> that's really bad from an engineering perspective
3491 2011-03-03 21:27:11 <slush> phantomcircuit: this is the reason why every job will be unique, even with the same prevhash and transactions inside
3492 2011-03-03 21:27:24 <Diablo-D3> phantomcircuit: the nonce isnt what you think it is
3493 2011-03-03 21:27:28 <slush> phantomcircuit: because server will generate unique extranonce and then unique merkle root from it
3494 2011-03-03 21:27:52 <phantomcircuit> Diablo-D3, uhm it's exactly want i think it is, a counter
3495 2011-03-03 21:28:22 <Diablo-D3> phantomcircuit: you know how a UUID works? it uses multiple pieces of data to make a guaranteed unique value without having to check with a central authority
3496 2011-03-03 21:28:30 <Diablo-D3> phantomcircuit: nonce is effectively part of a UUID
3497 2011-03-03 21:28:40 <phantomcircuit> slush, yes but why do that instead of just extending the block with a large nonce?
3498 2011-03-03 21:28:49 <slush> Diablo-D3: well, there will be probably only one private key for receiving block transactions
3499 2011-03-03 21:28:58 <phantomcircuit> Diablo-D3, lol i know exactly what the nonce is used for
3500 2011-03-03 21:29:09 <slush> Diablo-D3: so it will be hardcoded in pool config and pushed in 0th transaction for every merkle
3501 2011-03-03 21:29:14 <Diablo-D3> phantomcircuit: yeah but nonce doesnt mean guaranteed unique without a CA... it just means you cant use it more than once
3502 2011-03-03 21:29:30 <phantomcircuit> Diablo-D3, also modern UUID generation is just an rng
3503 2011-03-03 21:29:37 <Diablo-D3> phantomcircuit: and no, modern UUID isnt a rng
3504 2011-03-03 21:29:40 <slush> phantomcircuit: a) 32bit nonce is part of protocol. I cannot rise it without update of all clients
3505 2011-03-03 21:29:43 <slush> b) what it would solve?
3506 2011-03-03 21:29:48 <phantomcircuit> Diablo-D3, check the libuuid
3507 2011-03-03 21:29:53 <gasteve> I think my 5970 is really good at not finding valid blocks
3508 2011-03-03 21:29:59 <slush> gasteve: lol
3509 2011-03-03 21:30:00 <Diablo-D3> phantomcircuit: some uuid variants use a rn for PART of the value
3510 2011-03-03 21:30:21 <phantomcircuit> slush, would reduce the calculations necessary on the servers half
3511 2011-03-03 21:30:45 <Diablo-D3> phantomcircuit: Ive written uuids that do something like sha256 of cluster unique name, sha256 of unix time, sha256 of nanotime, sha256 of tid, and then sha256 all of that.
3512 2011-03-03 21:31:00 <Diablo-D3> phantomcircuit: theres lots of ways of doing UUIDs, not just limited to the 5 in the RFC
3513 2011-03-03 21:31:06 <phantomcircuit> Diablo-D3, that's just silly
3514 2011-03-03 21:31:15 <Diablo-D3> phantomcircuit: anyhow, nonces can be repeated... as long as the two other values dont match.
3515 2011-03-03 21:31:32 <phantomcircuit> Diablo-D3, ps i know everything you've just said
3516 2011-03-03 21:31:34 <slush> phantomcircuit: yes, bigger nonce can solve the pain of recalculating merkle for every miner. But it is impossible to do
3517 2011-03-03 21:31:39 <Diablo-D3> phantomcircuit: so why bitch
3518 2011-03-03 21:31:46 somebody_ has joined
3519 2011-03-03 21:31:54 <Diablo-D3> slush: find me a block the pool did on blockexplorer
3520 2011-03-03 21:32:04 <phantomcircuit> Diablo-D3, you're dumb so im going to ignore you
3521 2011-03-03 21:32:09 <slush> http://blockexplorer.com/block/000000000000fbce375dd2d8202218368ee14e2855ce1353aa1524371ea01af7
3522 2011-03-03 21:32:24 <phantomcircuit> slush, i dont know, putting a nonce in the transaction block just feels wrong :P
3523 2011-03-03 21:32:41 <slush> phantomcircuit: ask satoshi :)
3524 2011-03-03 21:33:01 <Diablo-D3> slush: now, which txes are yours? just the gen 50?
3525 2011-03-03 21:33:07 <slush> Diablo-D3: yes
3526 2011-03-03 21:33:14 <x6763> phantomcircuit: it just goes into the script space for the coin generation transaction's input (which doesn't point to any previous transaction output)
3527 2011-03-03 21:33:19 <phantomcircuit> slush, lol right i thought he disappeared
3528 2011-03-03 21:33:21 <Diablo-D3> slush: and that address is only ever used once and thrown away?
3529 2011-03-03 21:33:40 <slush> Diablo-D3: currently, yes
3530 2011-03-03 21:33:54 <phantomcircuit> x6763, ah, well that's a hack if i've ever seen one
3531 2011-03-03 21:34:05 <phantomcircuit> apparently satoshi failed to do some basic math
3532 2011-03-03 21:34:08 <Diablo-D3> slush: huh. maybe thats okay then.
3533 2011-03-03 21:34:08 <slush> Diablo-D3: in the future, I like the idea of the same address for all generated blocks
3534 2011-03-03 21:34:11 <knotwork> I have added some bitcoind commands to one of my IRC bots, but doing so I have discovered a problem
3535 2011-03-03 21:34:25 <Diablo-D3> slush: yeah, but with same address you have to do what I was asking earlier
3536 2011-03-03 21:34:35 <knotwork> it seems that sometimes a command to the daemon simply hangs and it looks like the daemon hangs too
3537 2011-03-03 21:34:53 <phantomcircuit> seems kind of obvious to me that a 32bit nonce isn't enough to guarantee there is a valid 256 bit hash
3538 2011-03-03 21:35:11 <slush> Diablo-D3: sorry, too long discussion. What did you asking earlier? :)
3539 2011-03-03 21:35:12 <phantomcircuit> ironically the higher the difficulty gets the less likely that will be a problem
3540 2011-03-03 21:35:20 <Diablo-D3> slush: you have to take the block apart and poison it with a transaction from the known pool address to your hidden address
3541 2011-03-03 21:35:36 jnd has joined
3542 2011-03-03 21:35:37 <slush> I have to go for the moment, will be back soon
3543 2011-03-03 21:35:37 bk128-Droid has quit (Quit: Bye)
3544 2011-03-03 21:35:41 <Diablo-D3> phantomcircuit: btw, yes, you can take a given header, do all nonces for it, and not get a valid new block
3545 2011-03-03 21:35:54 <slush> No, I don't need. Or I don't know why I need
3546 2011-03-03 21:36:16 <phantomcircuit> Diablo-D3, i just fucking said that, stop trying to tell me things i already know
3547 2011-03-03 21:36:45 <Diablo-D3> slush: you need to have something in the block that the attacker cannot possibly know
3548 2011-03-03 21:37:07 <Diablo-D3> also, WHAT THE FUCK
3549 2011-03-03 21:37:10 <phantomcircuit> slush, lol i get the feeling Diablo-D3 has no idea how btc works
3550 2011-03-03 21:37:16 <Diablo-D3> look at that one HUGE transaction
3551 2011-03-03 21:37:17 <phantomcircuit> slush, like, not a clue
3552 2011-03-03 21:37:40 altamic has joined
3553 2011-03-03 21:37:40 altamic has quit (Changing host)
3554 2011-03-03 21:37:40 altamic has joined
3555 2011-03-03 21:37:43 <Diablo-D3> that should trigger tx fee multiple times out of spite
3556 2011-03-03 21:37:46 <Lachesis>  Diablo-D3 why do you need something the attacker can't know?
3557 2011-03-03 21:37:52 <Diablo-D3> Lachesis: security reasons
3558 2011-03-03 21:37:58 <Lachesis> why?
3559 2011-03-03 21:38:07 <Lachesis> everyone gets the block when you've confirmed it
3560 2011-03-03 21:38:14 <Lachesis> that's the whole point
3561 2011-03-03 21:38:15 <phantomcircuit> Lachesis, lol dont bother, he doesnt understand how btc actually works
3562 2011-03-03 21:38:18 * phantomcircuit is amused
3563 2011-03-03 21:38:18 <Diablo-D3> Lachesis: just to completely cockblock any potential attack vectors on stealing winning shares
3564 2011-03-03 21:38:25 <Diablo-D3> phantomcircuit: troll elsewhere
3565 2011-03-03 21:38:27 <Lachesis> Diablo-D3, how would you steal them?
3566 2011-03-03 21:38:44 <Diablo-D3> Lachesis: using more CPU power than God; there is no such thing as too much paranoia
3567 2011-03-03 21:38:46 <phantomcircuit> Diablo-D3, rofl im not even trolling, you clearly dont understand the fundamental problem being solved here
3568 2011-03-03 21:38:53 <Lachesis> Diablo-D3, in that case, why steal winning shares?
3569 2011-03-03 21:38:58 <Lachesis> why not just double spend?
3570 2011-03-03 21:39:02 <phantomcircuit> lololololololololololololololol
3571 2011-03-03 21:39:04 <Lachesis> or put the pool out of business by pwning the market
3572 2011-03-03 21:39:08 <Diablo-D3> Lachesis: double spending should actually be harder.
3573 2011-03-03 21:39:10 * phantomcircuit goes off to lmfao
3574 2011-03-03 21:39:24 sipa has quit (Remote host closed the connection)
3575 2011-03-03 21:39:26 <Diablo-D3> Lachesis: but the problem with the winning share you actually dont have a shitload of values to check for a lot of things
3576 2011-03-03 21:39:46 <Lachesis> fewer variables, more definite statements pls
3577 2011-03-03 21:39:53 <Lachesis> what values are you checking for things?
3578 2011-03-03 21:40:12 <Diablo-D3> Lachesis: merkleroot, basically
3579 2011-03-03 21:40:49 <phantomcircuit> Lachesis, lulz he has no idea what he's talking about
3580 2011-03-03 21:40:50 <Diablo-D3> I just cant figure out how to build a merkleroot that doesnt just reconstruct the header and send it to the pool anyhow
3581 2011-03-03 21:40:53 <phantomcircuit> Lachesis, i love this
3582 2011-03-03 21:41:26 <Lachesis> what about it? if you're find a winning hash using a merkle root that includes a coinbase directed at the pool, submitting it yourself would be worthless
3583 2011-03-03 21:41:46 <Lachesis> and changing a single byte in the argument to a hash changes the entire hash
3584 2011-03-03 21:41:48 <Lachesis> cascade effect
3585 2011-03-03 21:41:58 <Diablo-D3> Lachesis: yes, Im aware of what the cascade effect is
3586 2011-03-03 21:42:05 * Diablo-D3 only schools people on that in here daily
3587 2011-03-03 21:42:20 <Lachesis> Diablo-D3, then what's your proposed attack?
3588 2011-03-03 21:42:40 <phantomcircuit> Diablo-D3, can i have some of what you're smoking? unless it's crack, i dont want that
3589 2011-03-03 21:42:44 gavinandresen has quit (Ping timeout: 252 seconds)
3590 2011-03-03 21:43:33 <Raulo> mmarker. yes, it runs faster on K10
3591 2011-03-03 21:44:40 bk128-Droid has joined
3592 2011-03-03 21:44:52 <Diablo-D3> Lachesis: well, I cant figure out one that doesnt involve inserting trash txes and hoping for magic fairy dust.
3593 2011-03-03 21:45:05 <Lachesis> Diablo-D3, what's that one
3594 2011-03-03 21:45:15 <Lachesis> b/c i don't think there is one that doesn't involve breaking SHA
3595 2011-03-03 21:45:23 <Diablo-D3> thats pretty much it
3596 2011-03-03 21:45:30 <Lachesis> Diablo-D3, then it's no concern
3597 2011-03-03 21:45:37 <Diablo-D3> still, I dont like the feel of that one.
3598 2011-03-03 21:45:44 <Lachesis> and more importantly, even if there's secret data in the header
3599 2011-03-03 21:45:47 <Lachesis> if sha is broken
3600 2011-03-03 21:45:49 <Lachesis> it won't help
3601 2011-03-03 21:48:19 gavinandresen has joined
3602 2011-03-03 21:49:45 genjix has joined
3603 2011-03-03 21:50:06 <Diablo-D3> Lachesis: thats not stopping my strong sense of paranoia.
3604 2011-03-03 21:50:55 <Lachesis> Diablo-D3, you ought to see someone about that
3605 2011-03-03 21:51:14 <Lachesis> or at least freak out about something more important
3606 2011-03-03 21:51:21 <Lachesis> stolen shares aren't the risk here
3607 2011-03-03 21:51:24 <Lachesis> breaking bitcoin is
3608 2011-03-03 21:51:31 <Diablo-D3> people start small.
3609 2011-03-03 21:51:58 <Diablo-D3> Lachesis: and btw, every hurdle added, no matter how small, is enough to stop people from even trying
3610 2011-03-03 21:52:49 <Diablo-D3> and btw, Lachesis, cultivating paranoia is a good thing for people who write software.
3611 2011-03-03 21:53:00 <Aciid> tru
3612 2011-03-03 21:54:07 <Lachesis> Diablo-D3, i agree with your last statement, at least in a certain abstract sense
3613 2011-03-03 21:54:13 <Lachesis> but it's important to be _efficiently_ paranoid
3614 2011-03-03 21:54:24 <Lachesis> worry about bottlenecks and weakest links
3615 2011-03-03 21:54:41 <Lachesis> and importantly, critical things
3616 2011-03-03 21:54:56 <Lachesis> if shares could be stolen, that'd be hard on pools, but definitely fixable
3617 2011-03-03 21:55:00 bk128-Droid has quit (Quit: Bye)
3618 2011-03-03 21:55:19 <Diablo-D3> okay, lets try it this way
3619 2011-03-03 21:55:29 <Diablo-D3> whats the cons of generating new addresses every 10 minutes
3620 2011-03-03 21:55:36 <Lachesis> there aren't any
3621 2011-03-03 21:55:44 <Diablo-D3> so, ultimately, I still win
3622 2011-03-03 21:55:51 <Lachesis> what?
3623 2011-03-03 21:55:52 <Syke_> bloats the wallet
3624 2011-03-03 21:56:03 <Lachesis> what did you win?
3625 2011-03-03 21:56:17 <Diablo-D3> Syke_: in a world of 4GB HDs, I dont think thats really an issue
3626 2011-03-03 21:56:18 <Lachesis> Syke_, yeah but 256 bits or whatever every 10 minutes is nothign
3627 2011-03-03 21:56:37 <Diablo-D3> Lachesis: gen new address vs not, gen new still wins
3628 2011-03-03 21:56:38 smaniak has joined
3629 2011-03-03 21:56:44 <Syke_> you didn't ask for only big cons
3630 2011-03-03 21:56:49 <Lachesis> Diablo-D3, i don't understand
3631 2011-03-03 21:56:52 <Lachesis> Syke_, fair enough
3632 2011-03-03 21:57:13 <Lachesis> let's look at it this way: for what are the pros of generating new addresses every ten minutes?
3633 2011-03-03 21:57:15 <Diablo-D3> Lachesis: theres no reason why I shouldnt gen a new address on every new block found by the pool
3634 2011-03-03 21:57:28 <Diablo-D3> its reasonable paranoia
3635 2011-03-03 21:57:30 <Lachesis> why are you generating addresses at all?
3636 2011-03-03 21:57:46 <Diablo-D3> to prevent any attacks.
3637 2011-03-03 21:57:47 <Lachesis> but the mere act of discussing it is taking away from useful work
3638 2011-03-03 21:57:47 <lfm> Diablo-D3: someone is looking into your memory?
3639 2011-03-03 21:57:48 <Syke_> every mined block does generate a new address
3640 2011-03-03 21:57:51 <Lachesis> slush scould be doing useful stuff
3641 2011-03-03 21:57:58 <Lachesis> (well and is)
3642 2011-03-03 21:58:13 <Diablo-D3> Lachesis: slush already is gen on new block
3643 2011-03-03 21:58:16 <Lachesis> if he were wasting his time implementing a system to generate new addresses every block (which, btw, I thought already happened)
3644 2011-03-03 21:58:21 <Lachesis> ..
3645 2011-03-03 21:58:27 <Diablo-D3> Lachesis: he wants to switch that to keep using same address
3646 2011-03-03 21:58:28 <Lachesis> what the hell are you proposing we do, and why?
3647 2011-03-03 21:58:39 <Lachesis> ok ok
3648 2011-03-03 21:58:40 <BlueMatt> gavinandresen: your patch seems great short-term (tested, working for me), but what are we going to do about this issue long term?  Is it a good idea to allow fewer blocks to be sent in one group?
3649 2011-03-03 21:58:42 <Lachesis> so why is that bad?
3650 2011-03-03 21:58:43 <Diablo-D3> Im not proposing anything, Im just saying its a bad idea
3651 2011-03-03 21:58:49 <Lachesis> and why does he want to switch it?
3652 2011-03-03 21:59:04 <Lachesis> the upshot is that everyone will know what coins came from the pool, right?
3653 2011-03-03 21:59:05 <Diablo-D3> btw, why cant me regurgitating the private key of the pool just be an attack vector?
3654 2011-03-03 21:59:30 <Lachesis> if you can break the crypto, you've broken bitcoin
3655 2011-03-03 21:59:45 <Diablo-D3> not break as much as Art has like 50 5970s.
3656 2011-03-03 21:59:51 <Diablo-D3> we only need to crack one key
3657 2011-03-03 22:00:01 <Lachesis> then what do you get?
3658 2011-03-03 22:00:07 <Diablo-D3> all future pool earnings
3659 2011-03-03 22:00:09 <Lachesis> when the pool finds a block, you wait to steal it out from under them?
3660 2011-03-03 22:00:14 <Diablo-D3> Lachesis: bingo
3661 2011-03-03 22:00:15 <Lachesis> no, b/c the pool does payouts
3662 2011-03-03 22:00:19 <Lachesis> so you only get one block
3663 2011-03-03 22:00:24 <Lachesis> before they see you've done that and change the key
3664 2011-03-03 22:00:30 <Diablo-D3> Lachesis: yes BUT
3665 2011-03-03 22:00:39 <Lachesis> slush, why do you want to switch to using the same key?
3666 2011-03-03 22:00:40 <Diablo-D3> never using a new address again stops the attack altogether
3667 2011-03-03 22:00:56 <Diablo-D3> if you break the key, you get nothing, as no new blocks will be assigned to it
3668 2011-03-03 22:01:08 <Lachesis> unless you break it during the current block?
3669 2011-03-03 22:01:16 <Diablo-D3> Lachesis: thats rather impossible
3670 2011-03-03 22:01:20 <Diablo-D3> keys harder to break than sha256
3671 2011-03-03 22:01:24 <Lachesis> this whole thing is pretty impossible man
3672 2011-03-03 22:01:24 <Diablo-D3> a lot harder
3673 2011-03-03 22:01:31 <Diablo-D3> its impossible _now_
3674 2011-03-03 22:01:34 <Diablo-D3> but what if Im using that key for years
3675 2011-03-03 22:02:10 <lfm> if you can do this you can do much more soon afterward
3676 2011-03-03 22:02:15 <Lachesis> what he said
3677 2011-03-03 22:02:19 <gasteve> what algorithm is used for the public/private keys used in a transaction?
3678 2011-03-03 22:02:48 <Diablo-D3> gasteve: one of the EC ones
3679 2011-03-03 22:02:48 <lfm> the public key algo is ecm
3680 2011-03-03 22:03:01 <Diablo-D3> its basically extra fuck you hard
3681 2011-03-03 22:03:04 <Lachesis> ecdsa
3682 2011-03-03 22:03:17 <Diablo-D3> yeah
3683 2011-03-03 22:03:20 asdf__ has joined
3684 2011-03-03 22:03:41 <gasteve> so, aren't the odds that the sun would burn out before you could crack the private key?
3685 2011-03-03 22:03:51 <lfm> the address hashes are sah256 and ripem I think
3686 2011-03-03 22:04:05 <Diablo-D3> gasteve: Art keeps buying new 5970s.
3687 2011-03-03 22:04:33 <gasteve> that statement was assuming you had access to the most powerful supercomputer on earth
3688 2011-03-03 22:04:40 <Diablo-D3> and no
3689 2011-03-03 22:04:44 <Diablo-D3> you just need sufficient luck
3690 2011-03-03 22:04:52 <phantomcircuit> gasteve, yes
3691 2011-03-03 22:04:59 <Lachesis> Diablo-D3, that's silly
3692 2011-03-03 22:05:08 <Lachesis> you can steal so much more with "significant luck"
3693 2011-03-03 22:05:12 <Lachesis> hell, why not just mine for yourself
3694 2011-03-03 22:05:19 <phantomcircuit> gasteve, with current attacks it would take you approximately forever to break the crypto being used
3695 2011-03-03 22:05:20 <Diablo-D3> still
3696 2011-03-03 22:05:21 <Lachesis> with significant luck, maybe you'll get a block every second
3697 2011-03-03 22:05:31 <lfm> ok so your "attack" is to guess the key?
3698 2011-03-03 22:05:34 <Diablo-D3> Lachesis: you only need to steal from the pool once to destroy it
3699 2011-03-03 22:05:37 <phantomcircuit> gasteve, attacking the p2p network and the client however are both pretty easy right now
3700 2011-03-03 22:05:45 <Diablo-D3> lfm: yeah, thats about the only legitimate attack vector I can find
3701 2011-03-03 22:05:58 <phantomcircuit> Diablo-D3, that's no even a real attack
3702 2011-03-03 22:06:06 <Lachesis> phantomcircuit, yeah it is
3703 2011-03-03 22:06:06 <phantomcircuit> "luck" isn't an attack
3704 2011-03-03 22:06:07 <phantomcircuit> >.>
3705 2011-03-03 22:06:09 <Lachesis> it's the biggest possible attack
3706 2011-03-03 22:06:13 <Lachesis> just a really really improbable one
3707 2011-03-03 22:06:16 <Lachesis> and we should ignore it
3708 2011-03-03 22:06:26 <Diablo-D3> its the same reasoning why the client now obsessively changes your address after each TX
3709 2011-03-03 22:06:31 <Lachesis> Diablo-D3, value = probability of success * reward of success
3710 2011-03-03 22:06:36 <Lachesis> no
3711 2011-03-03 22:06:38 <lfm> Diablo-D3: I know a better attack, track down slush and break into his house when he's out and steal his keys
3712 2011-03-03 22:06:42 <Lachesis> lol yeah
3713 2011-03-03 22:06:46 <Lachesis> or his datacenter
3714 2011-03-03 22:06:48 <Lachesis> hopefully :
3715 2011-03-03 22:06:49 <Diablo-D3> Lachesis: that implies criminals are logical
3716 2011-03-03 22:06:49 <Lachesis> :)
3717 2011-03-03 22:06:50 <phantomcircuit> lfm, not a bad plan
3718 2011-03-03 22:06:58 <Diablo-D3> and I doubt slush has anything local to him
3719 2011-03-03 22:07:01 <Lachesis> Diablo-D3, umm no it doesn't
3720 2011-03-03 22:07:03 <phantomcircuit> slush, hey can i buy you a pizza??
3721 2011-03-03 22:07:05 <Lachesis> he probably has ssh
3722 2011-03-03 22:07:17 <Lachesis> into his linode or whatever is hosting it
3723 2011-03-03 22:07:18 <Diablo-D3> yes, but that just means he re-images the VPS
3724 2011-03-03 22:07:19 <phantomcircuit> Lachesis, dat keylogger
3725 2011-03-03 22:07:33 <Lachesis> Diablo-D3, but you get his wallet!
3726 2011-03-03 22:07:36 <Lachesis> or you're secret
3727 2011-03-03 22:07:45 <Diablo-D3> your != you're!
3728 2011-03-03 22:07:50 <Lachesis> yep
3729 2011-03-03 22:07:57 <Lachesis> but i meant you are secret
3730 2011-03-03 22:08:04 <Lachesis> i mean, break in, rootkit his computer, steal some physical shit, and leave
3731 2011-03-03 22:08:15 <Diablo-D3> Lachesis: well
3732 2011-03-03 22:08:19 <Diablo-D3> his wallet presumably has nothing in it
3733 2011-03-03 22:08:24 <Lachesis> umm right
3734 2011-03-03 22:08:25 <Lachesis> but he has ssh
3735 2011-03-03 22:08:27 <Diablo-D3> his real wallet has more in it
3736 2011-03-03 22:08:34 <Lachesis> which would get you his linode
3737 2011-03-03 22:08:38 <Lachesis> which would get you the pool
3738 2011-03-03 22:08:42 <Diablo-D3> hrm
3739 2011-03-03 22:08:47 <Lachesis> point is, there are cheaper, easier, and more probable attacks
3740 2011-03-03 22:08:49 <Diablo-D3> but the easier way is to just enter the fucking DC
3741 2011-03-03 22:08:52 <Lachesis> and they're STILL improbable
3742 2011-03-03 22:08:54 <lfm> Diablo-D3: still easier than "guess his keys"
3743 2011-03-03 22:08:56 <Diablo-D3> since linode hosts in really ghetto DCs.
3744 2011-03-03 22:08:59 <Lachesis> lfm +1
3745 2011-03-03 22:09:30 <Diablo-D3> lfm: the argument, btw, is doing what slush does now, gen a new key after each new block.... or what slush proposes to change it to, not genning a new key.
3746 2011-03-03 22:09:40 xelister has joined
3747 2011-03-03 22:09:40 xelister has quit (Changing host)
3748 2011-03-03 22:09:40 xelister has joined
3749 2011-03-03 22:09:52 <Lachesis> btw, the current client does NOT generate a new key after every block because of lost key security concerns
3750 2011-03-03 22:10:00 <Lachesis> it does it to help ensure privacy
3751 2011-03-03 22:10:04 genjix has left ()
3752 2011-03-03 22:10:12 <Diablo-D3> lfm: Im saying slush should keep it for security concerns.
3753 2011-03-03 22:10:19 <xelister> US [develops devices to] scan people at rail and bus stations and special events [even] without their knowledge
3754 2011-03-03 22:10:25 <Diablo-D3> and theres no legitimate argument on why to change.
3755 2011-03-03 22:10:26 Vladimir__ has quit (Ping timeout: 272 seconds)
3756 2011-03-03 22:10:36 <Diablo-D3> keeping it the way it is has no penalty.
3757 2011-03-03 22:10:38 <Lachesis> Diablo-D3, i don't know why he wants to change either
3758 2011-03-03 22:10:39 genjix has joined
3759 2011-03-03 22:10:42 <lfm> xelister: face regog?
3760 2011-03-03 22:10:48 <lfm> xelister: face recog?
3761 2011-03-03 22:10:51 <Lachesis> but i would assume if he's changing stuff, there's a reason
3762 2011-03-03 22:11:01 <BlueMatt> tcatm: -daemon patch works for me :)
3763 2011-03-03 22:11:04 <xelister> lfm: body scans.   Perhaps dick recognition?
3764 2011-03-03 22:11:06 <BlueMatt> tcatm: thanks for the fix
3765 2011-03-03 22:11:15 <Lachesis> BlueMatt, what was the bug
3766 2011-03-03 22:11:27 <xelister> lfm: TSA moving to streets
3767 2011-03-03 22:11:38 <BlueMatt> Lachesis: https://github.com/bitcoin/bitcoin/issues/94
3768 2011-03-03 22:11:49 <Lachesis> cool
3769 2011-03-03 22:11:53 <lfm> xelister: cuz I heard they tried faces at a superbowl a year ago and it was useless
3770 2011-03-03 22:11:55 <xelister> how soon uneducated niggers will be able to molest your child and watch your wife naked even on streets?
3771 2011-03-03 22:12:10 <Aciid> all because of TSA
3772 2011-03-03 22:12:11 <Lachesis> xelister, what?
3773 2011-03-03 22:12:24 <Lachesis> actually nevermind
3774 2011-03-03 22:12:26 <xelister> Lachesis: well, nowdays uneducated niggers can rape children in airports
3775 2011-03-03 22:12:31 <Lachesis> i see
3776 2011-03-03 22:12:31 <xelister> by USA definition of rape == groping
3777 2011-03-03 22:12:36 <Lachesis> ahhh
3778 2011-03-03 22:12:36 <Lachesis> ok
3779 2011-03-03 22:12:37 <Aciid> hahah
3780 2011-03-03 22:12:39 <xelister> if said niggers work for TSA
3781 2011-03-03 22:12:45 <xelister> and now they want to move this body scanners
3782 2011-03-03 22:12:47 <xelister> even to streets
3783 2011-03-03 22:12:48 <Diablo-D3> so
3784 2011-03-03 22:12:49 <xelister> :)
3785 2011-03-03 22:12:49 <Diablo-D3> basically
3786 2011-03-03 22:12:51 <Lachesis> can educated white people do it?
3787 2011-03-03 22:12:52 smaniak has quit (Quit: Page closed)
3788 2011-03-03 22:12:56 <Diablo-D3> pool software really needs to do _nothing_.
3789 2011-03-03 22:12:57 <lfm> xelister: uneducated rednecks too
3790 2011-03-03 22:13:10 <Aciid> no, lowlife fat people that is
3791 2011-03-03 22:13:10 <xelister> Lachesis: sure. but educated people choose non TSA jobs. Perhaps except for pedos
3792 2011-03-03 22:13:19 <xelister> lfm: yes
3793 2011-03-03 22:13:20 <Lachesis> xelister, kik
3794 2011-03-03 22:13:21 <Lachesis> lol*
3795 2011-03-03 22:13:24 <Lachesis> damn right hand
3796 2011-03-03 22:13:26 <Diablo-D3> obama is now talking to the guys on the shuttle and the iss
3797 2011-03-03 22:13:38 <Lachesis> we should try to get bitcoin in orbit
3798 2011-03-03 22:13:50 <Aciid> ´with obamas help
3799 2011-03-03 22:13:52 <Aciid> yes we can
3800 2011-03-03 22:13:57 <xelister> Amd would export Radeons to Mars
3801 2011-03-03 22:14:02 <Lachesis> lol
3802 2011-03-03 22:14:22 <Diablo-D3> HOPE its whats for dinner
3803 2011-03-03 22:14:35 * Diablo-D3 loads the pig into the minecraft cannon
3804 2011-03-03 22:14:54 <slush> lol, pretty long discussion here
3805 2011-03-03 22:14:58 <Lachesis> apparently it costs you $300,000 to send a byte to the internet from the iss
3806 2011-03-03 22:14:59 <Lachesis> http://arstechnica.com/gaming/news/2011/02/lord-british-wants-to-take-you-to-space-and-hes-closer-than-you-think.ars/2
3807 2011-03-03 22:15:11 <slush> I'd like to have only one receiving address because it is easier to manage
3808 2011-03-03 22:15:27 <slush> I can have absolutely independent pool servers and accounting server
3809 2011-03-03 22:15:29 <Lachesis> slush, lets everyone know which coins were generated by the pool
3810 2011-03-03 22:15:34 <Lachesis> not that i'm sure that's bad
3811 2011-03-03 22:15:43 <slush> Once the pool server found the block, reward appear on the another server
3812 2011-03-03 22:15:50 <Lachesis> slush, true
3813 2011-03-03 22:15:59 <slush> Also wallet.dat is now rising for no reason
3814 2011-03-03 22:16:00 <Lachesis> so the pool server wouldn't give a damn about blocks it found
3815 2011-03-03 22:16:14 <[Tycho]> What function in bitcoind src decides about including tx in block or not ?
3816 2011-03-03 22:16:21 <slush> Lachesis: list of the blocks is already public, I have no problem with this
3817 2011-03-03 22:16:24 <Lachesis> yeah
3818 2011-03-03 22:16:29 <Lachesis> as long as you have a mechanism to change the key in place
3819 2011-03-03 22:16:34 <slush> Also sending rewards from the single wallet is better, there is less inputs to TX
3820 2011-03-03 22:16:43 <Lachesis> oh true
3821 2011-03-03 22:16:52 <Lachesis> Diablo-D3 is paranoid about the wrong things, but he is right that tons of key reuse is dangerous
3822 2011-03-03 22:17:06 <Diablo-D3> http://www.youtube.com/watch?v=JeV0XB-w_yM
3823 2011-03-03 22:17:08 <Diablo-D3> there
3824 2011-03-03 22:17:08 <Lachesis> just for fear that your wallet get compromised somehow
3825 2011-03-03 22:17:20 <Diablo-D3> the sub orbital animal canon
3826 2011-03-03 22:17:34 legion050 has joined
3827 2011-03-03 22:17:48 <slush> OK, there is quite small amount in the wallet. Currently something around 2000 BTC
3828 2011-03-03 22:18:10 <slush> If the server will be compromised, the hacker steal this amount in all case
3829 2011-03-03 22:18:24 <slush> Spreading money between more addresses does not help
3830 2011-03-03 22:18:51 <slush> And switching to another private key is just about changing config & restart pool servers
3831 2011-03-03 22:19:56 <Lachesis> slush, alright
3832 2011-03-03 22:20:04 <slush> So I don't see any real reason why fiddle with generating address for every block & distribute to central wallet
3833 2011-03-03 22:20:13 <Diablo-D3> "fiddle"
3834 2011-03-03 22:20:15 <Diablo-D3> not really
3835 2011-03-03 22:20:16 <Lachesis> the fear is that someone would steal your wallet and then wait until the balance in the pool was high
3836 2011-03-03 22:20:21 <Diablo-D3> you just submit block and then do gen
3837 2011-03-03 22:20:30 <Diablo-D3> yeah what Lachesis said
3838 2011-03-03 22:21:28 <slush> Lachesis: but it is still about same amount
3839 2011-03-03 22:21:40 xelister has quit (Remote host closed the connection)
3840 2011-03-03 22:21:48 <Diablo-D3> slush: yeah but they could repeatedly forge attack txs
3841 2011-03-03 22:21:56 <Lachesis> slush, yeah i know it's a low risk
3842 2011-03-03 22:22:07 <Lachesis> just pointing out that key reuse is bad
3843 2011-03-03 22:22:13 <slush> It is just sending buffer, before people withdraw their rewards
3844 2011-03-03 22:22:20 <slush> I also check balance every hour before sending
3845 2011-03-03 22:22:37 <slush> So I can detect troubles very quickly
3846 2011-03-03 22:23:04 <slush> Well, I still don't see any real reason.
3847 2011-03-03 22:23:09 <slush> It's maybe less elegant
3848 2011-03-03 22:23:12 <slush> But not less secure
3849 2011-03-03 22:23:42 <phantomcircuit> the odds that someone can steal the wallet but not change the way shares are divided seems small
3850 2011-03-03 22:24:18 sipa1024 has joined
3851 2011-03-03 22:24:30 <Lachesis> why is http://www.mtgox.com/ down when https://www.mtgox.com/ is up?
3852 2011-03-03 22:25:11 mmarker has quit (Quit: hometime)
3853 2011-03-03 22:25:29 xelister has joined
3854 2011-03-03 22:25:43 <xelister> http://mtgox.com/trade/megaChart - shows empty data?
3855 2011-03-03 22:25:53 <phantomcircuit> Lachesis, they're both up
3856 2011-03-03 22:26:04 <BlueMatt> Lachesis: ip change, clear your dns cache
3857 2011-03-03 22:26:29 <phantomcircuit> xelister, what BlueMatt said
3858 2011-03-03 22:26:56 sabalaba has quit (Ping timeout: 272 seconds)
3859 2011-03-03 22:27:57 Lobster_Man is now known as LobsterMan
3860 2011-03-03 22:28:11 LobsterMan has quit (Changing host)
3861 2011-03-03 22:28:11 LobsterMan has joined
3862 2011-03-03 22:28:44 <mizerydearia> I need an idea that is fun and simple for offering free monies to new users of a website...  any ideas?
3863 2011-03-03 22:29:12 <Blitzboom> reflinks. haha
3864 2011-03-03 22:29:23 <mizerydearia> reflinks?
3865 2011-03-03 22:29:49 <Blitzboom> explain what you offer free money for
3866 2011-03-03 22:29:50 <gavinandresen> mizerydearia:  Kiba might have an idea
3867 2011-03-03 22:30:07 * mizerydearia waves at kiba who happens to not be here
3868 2011-03-03 22:31:03 <mizerydearia> Blitzboom, It is for witcoin.com.  Courtesy of sgornick, I want to establish opportunity new users to the site that either do not have any bitcoins or are too poor to consider investing their own bitcoins to be able to use the site, primarily the sandbox, which is not free, but costs only 0.00001 bitcoins to post, reply, edit
3869 2011-03-03 22:31:05 <mizerydearia> and upvote
3870 2011-03-03 22:31:47 <Blitzboom> ah
3871 2011-03-03 22:32:03 <Blitzboom> do you think there’s demand for that?
3872 2011-03-03 22:32:16 <phantomcircuit> iirc transactions with amounts that low will probably never get added to a block
3873 2011-03-03 22:32:43 <mizerydearia> Blitzboom, Demand for free money?  I'm not sure
3874 2011-03-03 22:32:54 BlueMatt has quit (Quit: Ex-Chat)
3875 2011-03-03 22:33:03 <mizerydearia> phantomcircuit, The amounts are internal for the site
3876 2011-03-03 22:33:22 <phantomcircuit> mizerydearia, oh so they have to preload to your site?
3877 2011-03-03 22:33:26 somebody_ has quit (Quit: Leaving)
3878 2011-03-03 22:33:47 Spenvo has quit (Quit: Page closed)
3879 2011-03-03 22:33:58 <mizerydearia> phantomcircuit, Yep.  users can buy witcoins using bitcoins. 1 witcoin == 1 bitcoin, always.  And at any time users can withdraw their witcoins and convert them into bitcoins.
3880 2011-03-03 22:34:35 <phantomcircuit> ok them
3881 2011-03-03 22:34:48 <amiller> i want to make a challenge course for bitcoin, something i can give to friends as a gift, that will guide them through a bunch of neat things you can do
3882 2011-03-03 22:35:07 <amiller> with a focus on anonymity etc
3883 2011-03-03 22:35:27 <xelister> amiller: about anonimity, see the forum thread on BtcFn
3884 2011-03-03 22:39:14 <[Tycho]> What is the current size limit for free transactions in one block ?
3885 2011-03-03 22:42:38 <phantomcircuit> [Tycho], from the mainline client?
3886 2011-03-03 22:42:47 <phantomcircuit> because some of the miners dont follow those rules
3887 2011-03-03 22:43:13 asdf__ has quit (Quit: Page closed)
3888 2011-03-03 22:43:22 <mizerydearia> gavinandresen, What idea of Kiba's were you considering?
3889 2011-03-03 22:43:25 <[Tycho]> Yes, mainline.
3890 2011-03-03 22:43:45 <[Tycho]> if (nFees < CENT && GetBoolArg("-limitfreerelay")) - does that means that it's not enabled by default ?
3891 2011-03-03 22:44:13 <gavinandresen> mizerydearia: I don't know, Kiba asked me about tweaking the faucet so you had to play a game or somethign to get coins... sounds like what you want to do for witcoin
3892 2011-03-03 22:44:27 <mizerydearia> mm
3893 2011-03-03 22:47:04 <mizerydearia> Maybe in the theme of user-generated content I can implement a kind of user-generated q&a trivia implementation
3894 2011-03-03 22:47:16 xelister has quit (Ping timeout: 252 seconds)
3895 2011-03-03 22:47:49 xelister has joined
3896 2011-03-03 22:48:18 <mizerydearia> submit a question and answer and receive 0.00000001 witcoins.  For every correct guess from another user, receive another 0.0000001 witcoins.
3897 2011-03-03 22:49:04 <mizerydearia> However, I will also establish a pot in which users can donate to it and that's where the funds will come from.
3898 2011-03-03 22:49:19 <mizerydearia> similar to bitcoin faucet
3899 2011-03-03 22:50:21 xelister has quit (Remote host closed the connection)
3900 2011-03-03 22:51:15 doublec has joined
3901 2011-03-03 22:52:07 mmarker has joined
3902 2011-03-03 22:52:31 <mizerydearia> Or maybe statistical q&a primarily generating questions around the blockchain
3903 2011-03-03 22:52:45 <mizerydearia> answer correctly for witcoins
3904 2011-03-03 22:53:06 <slush> [Tycho]: Are you modifying rules for the free TX somehow?
3905 2011-03-03 22:53:48 <phantomcircuit> [Tycho], not necessarily there are defaults or arguments
3906 2011-03-03 22:57:19 Vlad__ has joined
3907 2011-03-03 22:57:49 <mizerydearia> Or, much easier: Every 60 seconds allow any user to click a button and receive 0.00000010 witcoins.  First person to click it gets witcoins.  0.0001440 witcoins / day and 0.0043200 witcoins / month
3908 2011-03-03 22:59:04 <[Tycho]> 19.2 years and they'll finally get one bitcoin ? :)
3909 2011-03-03 22:59:08 <mizerydearia> ^_^
3910 2011-03-03 22:59:11 <gavinandresen> mizerydearia:  do you live in a jurisdiction where online gambling is legal?  You've gotta be careful about games of chance that give money....
3911 2011-03-03 22:59:12 <mmarker> Heh.
3912 2011-03-03 22:59:13 <[Tycho]> slush, no, just exploring.
3913 2011-03-03 22:59:29 <mizerydearia> gavinandresen, I'm in Wisconsin, us
3914 2011-03-03 22:59:44 <gavinandresen> mizerdearia:  cool, I lived in Madison for two years...
3915 2011-03-03 22:59:49 <phantomcircuit> gavinandresen, arguably not a game of chance
3916 2011-03-03 22:59:50 <mizerydearia> nice
3917 2011-03-03 23:00:00 <mizerydearia> a game of timing
3918 2011-03-03 23:00:03 <validus> largest 6 pack is in wisconsin
3919 2011-03-03 23:00:24 <mizerydearia> but of course I will have to implement it in a way that bots can't game the system
3920 2011-03-03 23:00:26 <gavinandresen> phantomcircuit:  yeah.... "arguably" and Law Enforcement together make me nervous.
3921 2011-03-03 23:00:39 * mizerydearia agrees with gavinandresen 
3922 2011-03-03 23:00:39 <mmarker> Gavin: university?
3923 2011-03-03 23:00:53 <phantomcircuit> gavinandresen, pretty unlikely they're gonna care about gambling sub dollar amounts of a virtual currency
3924 2011-03-03 23:00:55 <gavinandresen> mmarker: my wife was a professor at UW for a while
3925 2011-03-03 23:01:17 <gavinandresen> phantomcircuit:  easy for you to say!
3926 2011-03-03 23:01:31 <mizerydearia> Besides, if I become targeted and am unable to be prosecuted for gambling, somehow I will all of a sudden find rape charges to criminally prosecute me and then yay guantanamo bay
3927 2011-03-03 23:01:32 <mmarker> Oh, cool.
3928 2011-03-03 23:01:40 <phantomcircuit> LADIES AND GENTLEMEN I SHALL BREAK THE LAW
3929 2011-03-03 23:01:55 <slush> Diablo-D3: Thanks for device selection patch!
3930 2011-03-03 23:01:56 <phantomcircuit> THE FIRST PERSON TO GUESS THE NUMBER IN MY HEAD WILL RECEIVE 0.01 BTC
3931 2011-03-03 23:02:00 <mmarker> 3
3932 2011-03-03 23:02:00 <mizerydearia> 29
3933 2011-03-03 23:02:05 <gavinandresen> phantomcircuit: 11 !
3934 2011-03-03 23:02:05 <validus> 9
3935 2011-03-03 23:02:06 <Blitzboom> 2
3936 2011-03-03 23:02:14 <mizerydearia> phantomcircuit, Does it have decimals?
3937 2011-03-03 23:02:16 <slush> Diablo-D3: Now I can use only one core for testing on testnet, hooray!
3938 2011-03-03 23:02:22 <phantomcircuit> mizerydearia, no
3939 2011-03-03 23:02:25 <validus> 1337
3940 2011-03-03 23:02:37 <mizerydearia> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
3941 2011-03-03 23:02:38 <gavinandresen> where's luke-jr so he can guess whatever that tonal number is
3942 2011-03-03 23:02:40 <validus> 333 cuz your only half evil
3943 2011-03-03 23:02:43 <amiller> 3.141592654......et
3944 2011-03-03 23:02:48 <mmarker> Heh.
3945 2011-03-03 23:02:57 <amiller> planck's constant
3946 2011-03-03 23:03:04 <mizerydearia> one google
3947 2011-03-03 23:03:05 <gavinandresen> If it's not eleven I don't want to win.
3948 2011-03-03 23:03:05 <mizerydearia> one google-plex
3949 2011-03-03 23:03:10 <phantomcircuit> mizerydearia, only about 100k more to go!
3950 2011-03-03 23:03:14 <mizerydearia> hehe
3951 2011-03-03 23:03:56 <mizerydearia> hint?
3952 2011-03-03 23:04:09 <mmarker> Greater than 1
3953 2011-03-03 23:04:13 <mizerydearia> obviously
3954 2011-03-03 23:04:18 <phantomcircuit> it's less than 2^32 and more than 0
3955 2011-03-03 23:04:25 <phantomcircuit> it's less than 2^32 and more than 100
3956 2011-03-03 23:04:26 <mizerydearia> it's greater than 100 also
3957 2011-03-03 23:04:48 <amiller> phantomcircuit, a billion
3958 2011-03-03 23:04:49 <phantomcircuit> and it's not divisible by 7
3959 2011-03-03 23:05:17 <phantomcircuit> amiller, no
3960 2011-03-03 23:05:27 larsivi has quit (Remote host closed the connection)
3961 2011-03-03 23:05:32 <amiller> oh well
3962 2011-03-03 23:05:34 <amiller> that was my best guess
3963 2011-03-03 23:05:36 <eps> 2 billion
3964 2011-03-03 23:05:49 larsivi has joined
3965 2011-03-03 23:05:57 <amiller> eps, that's absurd
3966 2011-03-03 23:05:58 <phantomcircuit> guesses can now be submitted via !guess
3967 2011-03-03 23:06:03 <phantomcircuit> i will answer them in batch
3968 2011-03-03 23:06:09 <phantomcircuit> after lunch
3969 2011-03-03 23:06:11 <eps> well that is me out of ideas then
3970 2011-03-03 23:06:26 <amiller> !guess trick question, phantomcircuit's head is empty
3971 2011-03-03 23:09:13 <phantomcircuit> first person to say 111112 gets it
3972 2011-03-03 23:09:19 <amiller> 111112
3973 2011-03-03 23:09:30 <phantomcircuit> winrar
3974 2011-03-03 23:09:48 <amiller> i didn't count, that was spuposed to be a wildly different number of ones in front
3975 2011-03-03 23:09:52 <phantomcircuit> address plz
3976 2011-03-03 23:10:05 <phantomcircuit> lulz
3977 2011-03-03 23:10:09 <phantomcircuit> fail at failing
3978 2011-03-03 23:10:11 <phantomcircuit> epic fail
3979 2011-03-03 23:10:27 <phantomcircuit> installing pygtk from source
3980 2011-03-03 23:10:38 <amiller> 1NcVsJ8Xj4gxmNuywwVpA3xwGpvcP4XPAb
3981 2011-03-03 23:10:46 <phantomcircuit> so far it's finished installing the actual code and has been installing the docs for about 30 minutes
3982 2011-03-03 23:10:47 <phantomcircuit> lulz
3983 2011-03-03 23:11:24 <phantomcircuit> amiller, congratulations you are now a co-conspirator
3984 2011-03-03 23:11:30 <amiller> hey, messages are included in the block chain aren't they
3985 2011-03-03 23:11:49 <phantomcircuit> when the feds show up about this illegal 0.01 BTC game of chance we shall both be in trouble
3986 2011-03-03 23:11:56 <amiller> phantomcircuit, technically i won a race, not a game of chance
3987 2011-03-03 23:12:15 <amiller> if messages are included in the block chain, we can commit bitcoin graffiti
3988 2011-03-03 23:12:17 <phantomcircuit> yes just like technically bingo is a game of skill
3989 2011-03-03 23:12:26 <phantomcircuit> amiller, what messages?
3990 2011-03-03 23:12:27 <amiller> wait
3991 2011-03-03 23:12:28 molecular has quit (Ping timeout: 276 seconds)
3992 2011-03-03 23:12:35 <amiller> i think you can leave a note in your transaction
3993 2011-03-03 23:13:09 <phantomcircuit> maybe in the client but there is nothing like that in the actual protocol
3994 2011-03-03 23:13:23 <phantomcircuit> i guess you could put a note in the script but that wouldnt work right now
3995 2011-03-03 23:13:27 molecular has joined
3996 2011-03-03 23:13:34 <x6763> what is the purpose of having a variable number of hash_start's in the "getblocks" command?
3997 2011-03-03 23:13:34 <luke-jr> gavinandresen: ?
3998 2011-03-03 23:13:38 <amiller> i thought it could, it just is disincentivised
3999 2011-03-03 23:13:42 <amiller> let me try to find a reference
4000 2011-03-03 23:13:47 <gavinandresen> luke-jr: ?
4001 2011-03-03 23:13:47 <phantomcircuit> x6763, splits
4002 2011-03-03 23:13:50 <sipa1024> it would be a non-standard scri[t
4003 2011-03-03 23:13:52 <sipa1024> script
4004 2011-03-03 23:13:58 <sipa1024> but you can put a comment in there
4005 2011-03-03 23:14:05 <luke-jr> gavinandresen: what tonal #?
4006 2011-03-03 23:14:19 <gavinandresen> luke-jr: the one phantomcircuit is thinking of
4007 2011-03-03 23:14:23 <luke-jr> which is?
4008 2011-03-03 23:14:34 <phantomcircuit> 111112
4009 2011-03-03 23:14:38 <phantomcircuit> contest has ended
4010 2011-03-03 23:14:38 <gavinandresen> luke-jr: I dunno, we're supposed to !guess
4011 2011-03-03 23:14:39 <luke-jr> sipa1024: just include a small fee, and your non-standard tx can be included!
4012 2011-03-03 23:14:43 <phantomcircuit> felonies all around
4013 2011-03-03 23:14:53 <sipa1024> luke-jr: how much % of the network do you have?
4014 2011-03-03 23:15:33 <x6763> phantomcircuit: hmmm...ok, thanks...i guess i'll have to look at it some more later when i have time...at least now i have somewhat of an idea what it's for
4015 2011-03-03 23:15:41 <phantomcircuit> luke-jr, ps found another attack, this time it would allow me to segment the network into arbitrary chunks, but would have to handle 100% of the networks traffic
4016 2011-03-03 23:15:44 <phantomcircuit> which might be a bit much
4017 2011-03-03 23:15:55 <luke-jr> sipa1024: .01 ?
4018 2011-03-03 23:15:58 <amiller> if comments are enabled
4019 2011-03-03 23:16:04 <gavinandresen> Actually, seriously, if you just give away money it probably doesn't count as gambling.
4020 2011-03-03 23:16:13 <amiller> i'm afraid that another possible attack is to put child pornography in plain text into the blocks
4021 2011-03-03 23:16:16 <amiller> and announce it
4022 2011-03-03 23:16:23 <phantomcircuit> gavinandresen, yeah i guess
4023 2011-03-03 23:16:37 <sipa1024> luke-jr: that means it would take on average a week before my tx ends up in a block mined by you? :)
4024 2011-03-03 23:16:38 <phantomcircuit> amiller, script size is too small for that
4025 2011-03-03 23:16:47 <amiller> it would be etched in history forever, easy to prove that it was there, and if it's announced publicly then anyone passing around blocks knows that they're doing it
4026 2011-03-03 23:16:47 <luke-jr> sipa1024: 10 days at current difficulty
4027 2011-03-03 23:16:51 <phantomcircuit> amiller, you might be able to do something else though
4028 2011-03-03 23:17:05 <gavinandresen> amiller: there's a thread in the forums about that.  My conclusion:  not an issue.
4029 2011-03-03 23:17:07 <luke-jr> sipa1024: been thinking about running a pool with my policies
4030 2011-03-03 23:17:20 <amiller> gavinandresen, is there a term i can use to find that thread?
4031 2011-03-03 23:17:23 <luke-jr> btw, how much MH/s would a NVIDIA Quadro FX 1800 get?
4032 2011-03-03 23:17:31 <phantomcircuit> amiller, "I WANT SOME KIDDIE PORN PLEASE"
4033 2011-03-03 23:17:35 noagendamarket has joined
4034 2011-03-03 23:17:40 <AAA_awright> ...why are clients allowing plain text to be distributed?
4035 2011-03-03 23:17:50 <phantomcircuit> AAA_awright, as a script
4036 2011-03-03 23:17:51 <amiller> don't type in caps, this is a public channel and people could be watching
4037 2011-03-03 23:17:53 <phantomcircuit> in a tx
4038 2011-03-03 23:17:57 <AAA_awright> Never mind enough information to recreate entire images
4039 2011-03-03 23:17:59 TD has joined
4040 2011-03-03 23:17:59 <phantomcircuit> amiller, what?
4041 2011-03-03 23:18:09 <phantomcircuit> AAA_awright, precisely
4042 2011-03-03 23:18:20 <luke-jr> AAA_awright: what do you think all that tx spam was?
4043 2011-03-03 23:18:23 <gavinandresen> amiller: http://www.bitcoin.org/smf/index.php?topic=2901.0
4044 2011-03-03 23:18:25 <AAA_awright> Why are we allowing ARBRITARTY EXECUTABLE CODE in the block chain?!?
4045 2011-03-03 23:18:27 <bitbot> What if I stored child porn in the block chain? Connection refused.
4046 2011-03-03 23:18:30 <amiller> thank you gavinandresen
4047 2011-03-03 23:18:53 <amiller> bitbot, how would you verify that it's not child porn
4048 2011-03-03 23:19:08 <amiller> well ishould read the forum
4049 2011-03-03 23:19:19 <amiller> it is exactly on topic
4050 2011-03-03 23:19:23 <luke-jr> AAA_awright: for the fee
4051 2011-03-03 23:19:27 <mizerydearia> amiller, That may help to establish better precedence on how absurd laws regarding such things are.
4052 2011-03-03 23:19:28 <sipa1024> luke-jr: a quadro fx 1800 has a G94 chip, the same as Geforce 9600 GT
4053 2011-03-03 23:19:28 <amiller> why is child pornography such an obvious example?
4054 2011-03-03 23:19:34 <AAA_awright> luke-jr: What about the fee?
4055 2011-03-03 23:19:39 <phantomcircuit> amiller, ps the illegal part wouldn't be bitcoin it would be the decoding program
4056 2011-03-03 23:19:47 <luke-jr> AAA_awright: I get it.
4057 2011-03-03 23:19:49 <AAA_awright> You get the money generated that block end of story right?
4058 2011-03-03 23:19:55 <phantomcircuit> amiller, im sure i could write a program to build anything from arbitrary info
4059 2011-03-03 23:20:24 <sipa1024> luke-jr: at 550 MHz instead of 650
4060 2011-03-03 23:20:28 <phantomcircuit> amiller, unlike most things simple possession even without knowledge is a serious felony
4061 2011-03-03 23:20:40 <phantomcircuit> amiller, indeed i cant think of anything else that is
4062 2011-03-03 23:20:49 <phantomcircuit>  15:12:30 up 19:42,  4 users,  load average: 21.46, 9.15, 10.29
4063 2011-03-03 23:20:50 <phantomcircuit> lulz
4064 2011-03-03 23:20:54 <AAA_awright> I'm liking my hash-based theory better http://www.bitcoin.org/smf/index.php?topic=2035.0
4065 2011-03-03 23:20:56 <bitbot> Hash-based chainless transactions theory Connection refused.
4066 2011-03-03 23:22:30 <AAA_awright> bitbot: ?
4067 2011-03-03 23:22:51 <slush> lol
4068 2011-03-03 23:23:16 <AAA_awright> Is that a bot?
4069 2011-03-03 23:23:19 <slush> yes
4070 2011-03-03 23:23:27 TD has quit (Quit: TD)
4071 2011-03-03 23:23:29 <AAA_awright> Dangit bots aren't supposed to speak unless spoken to
4072 2011-03-03 23:23:42 <phantomcircuit> AAA_awright, i have no idea what you're trying to say
4073 2011-03-03 23:23:52 <slush> I "loled" firstly with "bitbot: What if I stored child porn in the block chain? Connection refused."
4074 2011-03-03 23:24:01 <phantomcircuit> narratives tend to poorly convey technical thoughts
4075 2011-03-03 23:24:04 <AAA_awright> I thought that was intentional
4076 2011-03-03 23:24:10 <AAA_awright> as in a person
4077 2011-03-03 23:24:14 yawniek has quit (Ping timeout: 240 seconds)
4078 2011-03-03 23:24:19 <AAA_awright> phantomcircuit: What nattative?
4079 2011-03-03 23:24:19 <amiller> gavinandresen, i think your counter examples still miss a possible distinction
4080 2011-03-03 23:24:32 <AAA_awright> *narrative
4081 2011-03-03 23:24:34 <amiller> eh i'll just put it on the forum itself
4082 2011-03-03 23:26:02 <phantomcircuit> AAA_awright, the way you wrote that hash-based thing reads like a story
4083 2011-03-03 23:26:43 <AAA_awright> phantomcircuit: It starts like any other technical article, an introduction to the problem, a synopsis, and a step-by-step description of a possible solution.
4084 2011-03-03 23:27:08 frewsxcv_ is now known as frewsxcv
4085 2011-03-03 23:29:18 LobsterMan has quit (Remote host closed the connection)
4086 2011-03-03 23:30:11 <AAA_awright> phantomcircuit: The idea is you have a recorded state of the hash of the what private keys own how much money, and in such a way that examining the hash you can prove after a transaction that you have more money and someone else has less money
4087 2011-03-03 23:30:28 bk128 has joined
4088 2011-03-03 23:30:42 <phantomcircuit> AAA_awright, that's basically impossible
4089 2011-03-03 23:30:43 <AAA_awright> You prove that you have more because you know your own private key, and because the total amount of money is constant by design, someone therefore must have less.
4090 2011-03-03 23:30:48 <amiller> this is such a fun topic
4091 2011-03-03 23:31:29 <AAA_awright> phantomcircuit: I'm seeking a proof of that. It's almost certainly impossable for a constant length hash, but a variable length hash should be possible, I don't see why not
4092 2011-03-03 23:31:40 <phantomcircuit> AAA_awright, a hash which reveals information about the data isn't useful for much else
4093 2011-03-03 23:31:46 <phantomcircuit> AAA_awright, so you'd have to write your own
4094 2011-03-03 23:31:50 <AAA_awright> It's not a cryptographic hash
4095 2011-03-03 23:32:02 <AAA_awright> In that sense
4096 2011-03-03 23:32:03 <phantomcircuit> i mean
4097 2011-03-03 23:32:37 <phantomcircuit> i dont see how this helps at all
4098 2011-03-03 23:33:16 <AAA_awright> phantomcircuit: It eliminates the need for a block chain and the need to record how much money every account has
4099 2011-03-03 23:34:04 <AAA_awright> Maybe not the block chain, but it very well may save a lot of time and effort and storage
4100 2011-03-03 23:34:19 <phantomcircuit> AAA_awright, the problem you have to solve to do that is basically what BTC is designed to solve
4101 2011-03-03 23:34:53 <phantomcircuit> if you want to safe on time/storage just take a snapshot of the pruned transactions and archive everything before that
4102 2011-03-03 23:35:03 <phantomcircuit> save*
4103 2011-03-03 23:35:41 <AAA_awright> phantomcircuit: You're still explicitly recording owner/value pairs
4104 2011-03-03 23:36:01 <AAA_awright> Which you have to keep locally
4105 2011-03-03 23:37:02 <AAA_awright> With this, if it exists, and in theory (a lot of qualifiers), you wouldn't know how much is owned without the owning private key, though you could make a payment to someone's public key
4106 2011-03-03 23:37:16 slush has quit (Changing host)
4107 2011-03-03 23:37:16 slush has joined
4108 2011-03-03 23:37:52 <AAA_awright> And for the most flexibility maybe it requires an interactive channel like you use for diffie-hellman
4109 2011-03-03 23:38:35 <Diablo-D3> [05:53:57] <slush> Diablo-D3: Now I can use only one core for testing on testnet, hooray!
4110 2011-03-03 23:38:39 <Diablo-D3> slush: heh?
4111 2011-03-03 23:38:44 <Diablo-D3> slush: also, thank cdecker, thats his work
4112 2011-03-03 23:39:46 <slush> Diablo-D3: now, when I want to test something on testnet, i have to stop whole 2.4 ghash rig and point it to testnet. Annoying, because I have to do it many times per day
4113 2011-03-03 23:39:57 <slush> well, not now, but before this patch
4114 2011-03-03 23:40:20 <amiller> AAA_awright, suppose someone tried to 'mine for a collision' to your hash problem. They'd be able to try random combinations of balances, and keep trying until the hash matched. You're relying solely on the strength of your 'non cryptographic hash' function. Clients have no easy way of rejecting those blocks now.
4115 2011-03-03 23:40:41 <Diablo-D3> slush: erm, stop it?
4116 2011-03-03 23:40:46 <Diablo-D3> all you have to do is also run another client
4117 2011-03-03 23:40:51 <Diablo-D3> it'll eventually reach half and half
4118 2011-03-03 23:41:24 <AAA_awright> amiller: The information contained isn't cryptographic as such, the information itself is cryptographic... I don't know the exact mechanics of it maybe it is cryptographically secure, maybe it uses some special signing hash idk
4119 2011-03-03 23:41:28 <slush> Diablo-D3: no, it reach 2.3 and ~0.02 for some reason :)
4120 2011-03-03 23:41:59 sipa1024 is now known as sipa
4121 2011-03-03 23:42:54 <amiller> AAA_awright, but you're misusing cryptographic security to mean 'impossible to crack' rather than 'expensive/unlikely' to crack.
4122 2011-03-03 23:43:26 <Diablo-D3> slush: I said eventually.
4123 2011-03-03 23:43:27 <amiller> there ARE collisions, bitcoin relies on them to find blocks - once a false block using your system is accepted, there isn't enough information for other clients to find it
4124 2011-03-03 23:44:22 <sipa> bitcoin doesn't rely on collisions at all
4125 2011-03-03 23:44:35 <sipa> it relies on partial hashing to zero
4126 2011-03-03 23:44:36 <amiller> AAA_awright, also, if your system doesn't require transmitting the whole block record, how would you be able to merge forked blocks without losing data?
4127 2011-03-03 23:44:36 <AAA_awright> amiller: Where did I claim that?
4128 2011-03-03 23:44:46 <AAA_awright> amiller: I said "mathematically hard"
4129 2011-03-03 23:45:15 gasteve_ has joined
4130 2011-03-03 23:45:26 <AAA_awright> amiller: It's the same problem Bitcoin has right now, what happens when the block chain diverges? Someone's gonna lose their transactions
4131 2011-03-03 23:46:09 <amiller> but how would you find out who did?
4132 2011-03-03 23:46:10 <AAA_awright> Also, if someone were to break such a hash, it's not hard to program an exemption to a client saying "This branch of history invalid"
4133 2011-03-03 23:46:11 mmarker has quit (Quit: Bye)
4134 2011-03-03 23:46:13 <d-snp> if I would start my own bitcoin currency, would it be possible to use the calculations people made for the current network to generate blocks on my currency?
4135 2011-03-03 23:46:39 <mizerydearia> gavinandresen, Also, regarding gambling: I think it is technically not gambling in the sense of illegality considering for example, "guess my number" attempts are free, have no cost and therefore the "gambler" does not lose anything whatsoever.
4136 2011-03-03 23:47:05 <mizerydearia> except time
4137 2011-03-03 23:47:07 <mizerydearia> and time is money
4138 2011-03-03 23:47:09 <mizerydearia> so, nevermind
4139 2011-03-03 23:47:23 <AAA_awright> mizerydearia: Time is free to the government
4140 2011-03-03 23:47:39 <AAA_awright> 0% interest rates and such, what a life!
4141 2011-03-03 23:47:48 <amiller> mizerydearia, and money walks.... trespassing
4142 2011-03-03 23:47:58 <d-snp> hmm.. I guess not, since it would be very much like just supporting a branch that the rest of the bitcoin clients ignore..
4143 2011-03-03 23:48:00 <mizerydearia> and talks, freedom of speech
4144 2011-03-03 23:48:04 <AAA_awright> mizerydearia: I don't think you can write off time in traffic. Not yet, at least. Maybe I'm wrong who knows.
4145 2011-03-03 23:48:49  is now known as Netsniper|!~kvirc@76.251.230.92|Netsniper
4146 2011-03-03 23:48:51 <AAA_awright> amiller: Find out who did what?
4147 2011-03-03 23:49:35 <sipa> d-snp: why would you want to have your own bitcoin-like currency?
4148 2011-03-03 23:50:05 <amiller> AAA_awright, find out if a previous transaction you thought you made was in there
4149 2011-03-03 23:50:15 <d-snp> I don't know, I guess it would be kind of pointless since it would have the same processor time backing as the current one has :P
4150 2011-03-03 23:50:31 zygf_ is now known as zygf
4151 2011-03-03 23:50:57 sipa has quit (Changing host)
4152 2011-03-03 23:50:57 sipa has joined
4153 2011-03-03 23:51:07 <AAA_awright> amiller: By the fact a transaction that you applied to your store increased how much money you can send
4154 2011-03-03 23:51:36 <AAA_awright> And noting that it didn't change the total amount of money represented in the hash
4155 2011-03-03 23:51:41 <sipa> what if you make two copies of it
4156 2011-03-03 23:51:46 <sipa> on two systems
4157 2011-03-03 23:51:54 <sipa> and tries to spend the gained money twice?
4158 2011-03-03 23:52:04 <sipa> *try
4159 2011-03-03 23:52:06 <amiller> AAA_awright, how do you query the block to find out how much money you can spend?
4160 2011-03-03 23:52:24 <AAA_awright> I have no clue, that's one of the challenges presented in my post
4161 2011-03-03 23:52:42 <sipa> the only solution - i fear - is a block chain
4162 2011-03-03 23:52:50 <sipa> that serializes everything
4163 2011-03-03 23:52:50 <amiller> it seems to me that you can't query it if the information isn't in the block
4164 2011-03-03 23:53:09 <AAA_awright> At worst maybe you have to brute force it, but that can't be too hard, you double a "test" spend and see when it stops cleanly working
4165 2011-03-03 23:53:30 <AAA_awright> er, you iterate a "Test" spend doubling the amount, etc
4166 2011-03-03 23:53:33 <amiller> AAA_awright, the reason the extra informaiton is carried to avoid double spending it's to make it possible to repair by an individual client
4167 2011-03-03 23:54:30 <d-snp> how often does a block get added to the chain at the moment?
4168 2011-03-03 23:54:32 <amiller> er, it's not to avoid double spending, but to make it possible ...
4169 2011-03-03 23:55:04 akem has joined
4170 2011-03-03 23:55:25 <amiller> http://www.bitcoinmonitor.com/ you can watch them as they have come in recently, d-snp
4171 2011-03-03 23:55:34 <amiller> but it's supposed to be 10 minutes, always
4172 2011-03-03 23:55:45 <amiller> always, but on average
4173 2011-03-03 23:56:34 <amiller> and the difficulty is adjusted to maintain that
4174 2011-03-03 23:56:54 bk128 has quit (Quit: bk128)
4175 2011-03-03 23:57:02 <d-snp> oO
4176 2011-03-03 23:57:10 <d-snp> isn't that terribly slow? why not 1 per minute?
4177 2011-03-03 23:57:11 <sipa> while difficulty is continuously rising however, the difficulty will be consistently too low
4178 2011-03-03 23:57:30 <sipa> d-snp: the blocks are the one thing the whole network must agree on
4179 2011-03-03 23:57:37 OneFixt has joined
4180 2011-03-03 23:57:38 OneFixt has quit (Changing host)
4181 2011-03-03 23:57:38 OneFixt has joined
4182 2011-03-03 23:58:07 <d-snp> but an accepted block is the only thing that makes sure a transaction is actually valid right?
4183 2011-03-03 23:58:08 <sipa> the delay between them makes sure everyone has the previous one when the next one is found
4184 2011-03-03 23:58:18 <sipa> yes an no
4185 2011-03-03 23:58:37 <d-snp> why is it necessary that everyone has the previous one?
4186 2011-03-03 23:58:47 <sipa> otherwise there could be two versions
4187 2011-03-03 23:58:51 <sipa> of a block
4188 2011-03-03 23:58:58 <sipa> in one you spend your coins on bread
4189 2011-03-03 23:59:06 <sipa> in the other on wine
4190 2011-03-03 23:59:08 <d-snp> but there already could be right?
4191 2011-03-03 23:59:14 <sipa> ?
4192 2011-03-03 23:59:48 <sipa> the block chain is the one thing that orders transactions, globally
4193 2011-03-03 23:59:55 <amiller> there already can be two versions, but the network has a straightforward way of coming to an agreement on one