1 2010-11-21 00:12:05 <nanotube> ;;bc,stats
   2 2010-11-21 00:12:08 <gribble> Current Blocks: 93044 | Current Difficulty: 6866.89864897 | Next Difficulty At Block: 94752 | Next Difficulty In: 1708 blocks | Next Difficulty In About: 1 week, 5 days, 7 hours, 57 minutes, and 56 seconds | Next Difficulty Estimate: 6604.73964649
   3 2010-11-21 00:12:19 <nanotube> hm, so what's the deal with the decreased difficulty? is it due to the spam issues?
   4 2010-11-21 00:15:33 <tcatm> hard to tell
   5 2010-11-21 00:16:10 <nanotube> well, is the spam thing still going on? i see a little above something about "address spam"... ? is that different from the earlier transaction spam? what exactly is happening, basically. :)
   6 2010-11-21 00:17:44 <tcatm> there are some spam transactions in last blocks, but not as many as before
   7 2010-11-21 00:17:57 <nanotube> hm, bbe shows that tx volume has droppeed...
   8 2010-11-21 00:26:11 redMBA has quit (Ping timeout: 255 seconds)
   9 2010-11-21 00:26:33 <xelister> ok Im running diablo's miner
  10 2010-11-21 00:26:35 <xelister> it sucks
  11 2010-11-21 00:26:54 <xelister> desktop is slow as shit.  with -f 120  it works like m0m's -f 20
  12 2010-11-21 00:26:56 <xelister> and uses 200% CPI
  13 2010-11-21 00:26:58 <xelister> CPU
  14 2010-11-21 00:28:12 <nanotube> mm, some time ago brocktice reported it working pretty well... so either diablo changed something in the code, or you did something different.
  15 2010-11-21 00:30:33 redMBA has joined
  16 2010-11-21 00:31:47 <tcatm> anyone using internet explorer here?
  17 2010-11-21 00:34:10 <xelister> well I use the shitty 2.2 sdk. I will try 2.1
  18 2010-11-21 00:39:23 <nanotube> xelister: that could be it.
  19 2010-11-21 00:39:28 <nanotube> tcatm: not me...
  20 2010-11-21 00:39:44 <theymos> tcatm: I don't use it regularly, but I am on Windows.
  21 2010-11-21 00:42:19 <tcatm> what does bitcoincharts look like in IE?
  22 2010-11-21 00:42:32 <tcatm> i guess it's all messed up :)
  23 2010-11-21 00:43:30 <xelister> yes, that was it.  with sdk 2.1 it works fine, nanotube
  24 2010-11-21 00:43:37 <xelister> but just 100M instead 160M
  25 2010-11-21 00:44:15 <nanotube> xelister: ok cool
  26 2010-11-21 00:44:29 <nanotube> were you getting 160M using some other miner?
  27 2010-11-21 00:44:53 <xelister> yes
  28 2010-11-21 00:45:03 <xelister> on moms I was getting the expected 160M
  29 2010-11-21 00:45:11 redMBA has quit (Ping timeout: 265 seconds)
  30 2010-11-21 00:45:37 <xelister> on Diablos, also 160M with SDK 2.2 but 170%cpu java + 30%cpu xorg so and desktop is useless
  31 2010-11-21 00:45:55 <xelister> on Diablos - 100M with SDK 2.1 and almost 0% cpu. But then the generation is slow as shit
  32 2010-11-21 00:46:26 <nanotube> mm
  33 2010-11-21 00:46:27 <xelister> on m0m's I was getting 160M, untill it start locking up (perhaps after ubuntu's lib/kernel upgrade)
  34 2010-11-21 00:46:35 <xelister> making m0ms generators useless
  35 2010-11-21 00:46:48 <nanotube> well, talk to diablo about it, then. last i heard he was claiming his miner is faster than m0m's
  36 2010-11-21 00:47:08 <xelister> so we have slow as shit generator, making desktop useless ganerator, and useless hanging up generator. awesome
  37 2010-11-21 00:47:38 <xelister> perhaps we can make one that will block desktop AND generate slowly?  and be hard to install too
  38 2010-11-21 00:48:03 <nanotube> haha
  39 2010-11-21 00:48:07 <xelister> is m0m frequently here?
  40 2010-11-21 00:48:16 <nanotube> yes
  41 2010-11-21 00:48:22 <xelister> didnt seen him in days...
  42 2010-11-21 00:48:37 <nanotube> well, yes == i see him every few days
  43 2010-11-21 00:50:26 <theymos> tcatm: It looks like the layout might be slightly off, but it's usable (on IE8). http://img832.imageshack.us/img832/2048/clipboard01kp.png
  44 2010-11-21 00:50:50 <tcatm> cool, that's not too bad actually :)
  45 2010-11-21 00:58:55 <tcatm> isn't bcm supposed to output realtime data on port 27007?
  46 2010-11-21 01:00:01 <theymos> Yes.
  47 2010-11-21 01:01:23 <xelister> any idea what m0m miner start hanging?
  48 2010-11-21 01:01:59 <tcatm> had a telnet session open the whole day and i got no message after the 200
  49 2010-11-21 01:03:40 <theymos> I just tried it. It works. Maybe there were no trades.
  50 2010-11-21 01:04:02 <tcatm> there were some in #bitcoin-market
  51 2010-11-21 01:04:16 <tcatm> now i got a cancelled bid
  52 2010-11-21 01:04:43 <theymos> I just placed and canceled a bid, and it showed up in telnet like it should.
  53 2010-11-21 01:05:32 <tcatm> mhm then i didn't get the new-bid
  54 2010-11-21 01:05:36 <tcatm> strange
  55 2010-11-21 01:06:28 <theymos> I got both.
  56 2010-11-21 01:09:22 <xelister> theymos: you use miner?
  57 2010-11-21 01:09:29 <xelister> anyone using miner, can all please write
  58 2010-11-21 01:09:50 <theymos> I'm mining with the stock Bitcoin CPU miner.
  59 2010-11-21 01:10:06 <nanotube> theymos: what's your hps ? :)
  60 2010-11-21 01:10:18 <theymos> nanotube: About 2200 on two computers.
  61 2010-11-21 01:10:27 <nanotube> ;;bc,calc 2200
  62 2010-11-21 01:10:28 <gribble> The average time to generate a block at 2200 Khps, given current difficulty of 6866.89864897 , is 22 weeks, 1 day, 3 hours, 52 minutes, and 36 seconds
  63 2010-11-21 01:10:34 <nanotube> mmm :)
  64 2010-11-21 01:11:12 <xelister> system/version/latest update+restart (i.e. using newest kernel for distro?),  bitcoind svn revision,  any modifications to miner?  AMD SDK version.  Catalyst version --> what options --> Mhash, desktop lag/other problems
  65 2010-11-21 01:12:06 <nanotube> xelister: well, i know that ArtForz and Keefe both use custom miners... besides that... there must be a bunch of people using m0m's, since network power has been increasing steadily since he's released it. :)
  66 2010-11-21 01:12:27 <xelister> linux/ubuntu/2010.11.21 , 183+getwork, m0m, sdk 2.2 --> RADEON 5770; any options like -f30 -f200 --> 160Mhash as expected, desktop is nice BUT it hangs each few minutes, quicker if GPU is under load(movies etc)(
  67 2010-11-21 01:12:58 <xelister> linux/ubuntu/2010.11.10(!!!) [... all same options ...] ---> 160 Mhash, and NOT hanging. Best config
  68 2010-11-21 01:13:51 <xelister> linux/ubuntu/2010.11.21 , 183+getwork, Diablo's, edited .cl to unblock amd_bit_align as Diablo told. sdk 2.2 --> RADEON 5770; --> 160Mhash as expected, desktop is SLOW AS SHIT even -f 200, and cpu use is 200%
  69 2010-11-21 01:14:27 <xelister> linux/ubuntu/2010.11.21 [disablos, same ]  sdk 2.1 !!! --> RADEON 5770; --> 100Mhash :((( and desktop is responsive, almost 0% cpu use
  70 2010-11-21 01:14:48 <xelister> so.. since the update m0m's miner started hanging;  And diablo miner is either slow or kills desktop
  71 2010-11-21 01:14:55 <xelister> anyone else can write?
  72 2010-11-21 01:16:17 <nanotube> xelister: maybe post a forum thread with this info, too
  73 2010-11-21 01:17:00 <tcatm> i have a custom miner, but it needs a special bitcoind
  74 2010-11-21 01:39:28 <Kiba> so we briefly reached .29
  75 2010-11-21 01:40:46 <tcatm> not really
  76 2010-11-21 01:41:37 <tcatm> 0.285 .. 0.29 were the only trades with significant volume (~13k)
  77 2010-11-21 01:42:45 * Kiba delegates his bitcoin promotion effort to other people through bounties
  78 2010-11-21 01:43:21 <jgarzik> boy, this TX spam flood is highly irregular.  sometime it maxes out ~219 TX/block, sometimes it's only 14/block.  and the spammer takes breaks every now and then.
  79 2010-11-21 01:44:39 <Kiba> is he losing money, yet?
  80 2010-11-21 01:45:14 joose has joined
  81 2010-11-21 01:45:30 <joose> thankyou bitcoin devs!!!!!!
  82 2010-11-21 01:45:38 <joose> LONG LIVE BITCOIN!
  83 2010-11-21 01:45:48 <Kiba> another market anarchist?
  84 2010-11-21 01:45:59 <joose> i dont know, but i fucking LOVE bitcoin
  85 2010-11-21 01:46:05 <joose> it solves so many important problems
  86 2010-11-21 01:46:44 <joose> its the best thing since sliced bread, but i dont even buy sliced bread
  87 2010-11-21 01:46:51 <joose> cause i make my own bread
  88 2010-11-21 01:46:56 <joose> i cant wait til i can work for bitcoins
  89 2010-11-21 01:47:16 <Kiba> you can work for bitcoins
  90 2010-11-21 01:48:11 <tcatm> anyone have a well-connected bitcoin node on a public IP?
  91 2010-11-21 01:48:44 <theymos> tcatm: I do.
  92 2010-11-21 01:48:52 <theymos> theymos: theymos.ath.cx
  93 2010-11-21 01:49:22 <theymos> Also see http://www.bitcoin.org/wiki/doku.php?id=fallback_nodes
  94 2010-11-21 01:50:02 <tcatm> ah that's cool
  95 2010-11-21 01:50:48 <joose> I hope my landlord takes bitcoins :P
  96 2010-11-21 01:51:18 <joose> i think the exchange rate may go wild in the near future and eventually become nice and stable, thoughts?
  97 2010-11-21 01:51:31 <Kiba> there's no such thing as stability.
  98 2010-11-21 01:51:44 <Kiba> there is only...change.
  99 2010-11-21 01:51:47 <jgarzik> the exchange rate will continue to be volatile long into the future, whether bitcoin is wildly successful or remains a niche currency
 100 2010-11-21 01:52:12 <theymos> I don't think it will become stable in relation to fiat currencies because fiat currencies are not stable.
 101 2010-11-21 01:52:46 <Kiba> The Chinese already felt that they are the master
 102 2010-11-21 01:53:07 <Kiba> but the geeks of the world will have the last laugh
 103 2010-11-21 01:53:13 <Kiba> after all, we'll inherit the earth
 104 2010-11-21 01:53:15 genjix has quit (Ping timeout: 240 seconds)
 105 2010-11-21 01:54:05 <xelister> tcatm: carfe to post it? what language?
 106 2010-11-21 01:54:45 uberjar has joined
 107 2010-11-21 01:54:51 <tcatm> plain c
 108 2010-11-21 01:54:59 <uberjar> ftw
 109 2010-11-21 01:55:00 <xelister> Kiba: perhaps its good if chinese are strong
 110 2010-11-21 01:55:13 <xelister> at least it will teach something the United States of Arrogancce
 111 2010-11-21 01:55:21 <xelister> *arrogance
 112 2010-11-21 01:55:26 <Kiba> Arrogance?
 113 2010-11-21 01:55:31 <xelister> yes
 114 2010-11-21 01:55:43 <Kiba> the government's arrogance, perhaps
 115 2010-11-21 01:55:58 <xelister> sorry, you voted it. Now and always. And people are the same
 116 2010-11-21 01:56:05 <xelister> note, how people give addresses on internet
 117 2010-11-21 01:56:09 <lfm> government of the people by the people etc etc
 118 2010-11-21 01:56:25 <xelister> usually American will write like 32/512 whatever st., Dallas
 119 2010-11-21 01:56:25 <joose> i didnt vote
 120 2010-11-21 01:56:39 <xelister> perhaps Dallas, Colorado or whereve it lies
 121 2010-11-21 01:57:12 <joose> jgarzik: very true. I feel we have margin trading to thank for that
 122 2010-11-21 01:57:16 <xelister> no douchbags not everyone lives in USA damn it, the address is .... Iowa, United States of America
 123 2010-11-21 01:57:29 <xelister> here people write like ... Warsaw, Poland
 124 2010-11-21 01:57:40 <jgarzik> theymos: googling for bitcoin addresses pops up google cache of BBE pages.  I was curious... do you output a Last-Modified HTTP header?
 125 2010-11-21 01:58:08 <theymos> jgarzik: No.
 126 2010-11-21 01:58:11 <xelister> but that is just small detail that I found interesting to mention :)  The examples of really we-rule-the-world-and-there-be-dragons everyone knows so no reason to repeat it here
 127 2010-11-21 01:58:42 <jgarzik> theymos: that's surprising.  If it matters to you, that will save you some bot bandwidth.
 128 2010-11-21 01:58:44 <Kiba> whatever the case, I don't vote.
 129 2010-11-21 01:58:48 <Kiba> and I don't intend to vote.
 130 2010-11-21 01:58:53 <Kiba> democracy are for fools.
 131 2010-11-21 01:59:04 <Kiba> a way for the majority to plunder the rest.
 132 2010-11-21 01:59:06 <uberjar> when you vote you're giving your consent to be ruled under this particular system
 133 2010-11-21 01:59:08 <lfm> kiba people get the goverment they want
 134 2010-11-21 01:59:20 <xelister> no matter voting or not voting, this is how usa gov. operates, and this is how MOST people behave more or less unfortante
 135 2010-11-21 01:59:24 <jgarzik> people get the government that momentum wants
 136 2010-11-21 01:59:25 <jgarzik> :)
 137 2010-11-21 01:59:26 <xelister> of course not eveyrone is the same
 138 2010-11-21 01:59:45 <jgarzik> how else to explain election of a n00b
 139 2010-11-21 01:59:48 <joose> democracy sucks because it is so easy to condition and control people in a way that will make them vote themselves into getting raped
 140 2010-11-21 02:00:11 <Kiba> no, they votes to rape others
 141 2010-11-21 02:00:14 <Kiba> particularly the eldery.
 142 2010-11-21 02:00:21 <lfm> joose just the alternatives are worse
 143 2010-11-21 02:00:27 <Kiba> damn, medicaid, and medicare.
 144 2010-11-21 02:00:41 <Kiba> err
 145 2010-11-21 02:00:43 <joose> yeah, a bunch of brain dead old people watching TV, barely alive. thats your voters
 146 2010-11-21 02:00:59 <joose> lfm: would any of us know?
 147 2010-11-21 02:01:00 <Kiba> they want their social security
 148 2010-11-21 02:01:07 <Kiba> at the expense of their children.
 149 2010-11-21 02:01:16 <Kiba> and their children's children
 150 2010-11-21 02:01:22 <jgarzik> gavinandresen: http://theymos.ath.cx:64150/bbe/tx/2d82b522277d9d70a33e81400fae9b17e4fb4d5a23157b76eee6799d2f19f488 looks like somebody is trying to "steal" from the Bitcoin Faucet
 151 2010-11-21 02:01:33 <lfm> whereas the non-voters are all bright young beautiful people huh
 152 2010-11-21 02:01:41 <jgarzik> tx consolidates a bunch of faucet receipts, it looks like
 153 2010-11-21 02:01:50 <xelister> but averaged American thinks highly of himself
 154 2010-11-21 02:01:57 <xelister> would you say any of you support raping children?
 155 2010-11-21 02:02:16 <xelister> of course not, you would murder such a rapist on spot
 156 2010-11-21 02:02:18 <joose> i bet someone in here would
 157 2010-11-21 02:02:22 <jgarzik> and their children's children's children.  and their children's children's children's children.  Curse you, People's Front of Judea.
 158 2010-11-21 02:02:22 <uberjar> so you're asking if there are any catholics present ?
 159 2010-11-21 02:02:32 <xelister> then why is your army raping children and the killing them and their families
 160 2010-11-21 02:02:34 <jgarzik> and the Popular Front.  Splitters!
 161 2010-11-21 02:02:37 <xelister> and your taxes pay for it
 162 2010-11-21 02:02:50 <uberjar> not to mention your religious leaders
 163 2010-11-21 02:02:52 <nameless> !~root@weowntheinter.net|<so far off topic>
 164 2010-11-21 02:02:53 <joose> xelister: the same reason we dont have healthcare
 165 2010-11-21 02:02:56 <theymos> jgarzik: All of the page data is fetched on-the-fly, currently without caching. I don't think Last-Modified data would be valuable in this case.
 166 2010-11-21 02:02:58 <nameless> !~root@weowntheinter.net|Any one here play mirian
 167 2010-11-21 02:03:01 <xelister> and why then said rapists comes home, to get funny "sentence" of 4 or 6 years in prison for RAPING AND KILLING  CHILDREN AND ENTIRE FAIMILY?  how?
 168 2010-11-21 02:03:08 <nameless> !~root@weowntheinter.net|*miriani
 169 2010-11-21 02:03:22 <Kiba> taxes?
 170 2010-11-21 02:03:26 <Kiba> it's theft, pure and simple.
 171 2010-11-21 02:03:51 <xelister> you THINK you are against child abuse, corruption, murder, against racism and so on - but IN FACT, when childrens are killed and raped in iraq you couldn't care less
 172 2010-11-21 02:04:24 <Kiba> xelister: if you tone down your balant blanket blame of Americans
 173 2010-11-21 02:04:28 <jgarzik> theymos: it reduces the bot query frequency, if the bot gets back 301-not-modified (which Apache will generate, based on conditional http headers sent by the bot).
 174 2010-11-21 02:04:30 <Kiba> and also, please provide citations.
 175 2010-11-21 02:04:32 <xelister> well this is probably case for most other nations too,  just USA is most proud of itself and most smug ;)  This is, imho, why almost everyone in EU thinks USA == arrogand savages
 176 2010-11-21 02:04:49 <Kiba> I don't know what other people thinks.
 177 2010-11-21 02:04:50 <jgarzik> theymos: so it changes the google cache refresh frequency, even if you do no caching at all on your side
 178 2010-11-21 02:04:58 <xelister> Kiba: google it (g2g)
 179 2010-11-21 02:05:10 <Kiba> but xelister you make assertion, you provide the evidence.
 180 2010-11-21 02:05:10 <uberjar> xelister, everything the US learned about raping and killing it learned from Great Britian
 181 2010-11-21 02:05:16 <theymos> jgarzik: Does Apache do that automatically? Maybe I am doing it.
 182 2010-11-21 02:05:19 <lfm> ah so bitcoins is being use to launder terrosit funding! now it will be jumpped on!
 183 2010-11-21 02:05:27 <uberjar> the British empire has way more blood on it's hands
 184 2010-11-21 02:05:37 <xelister> Kiba: ok I back for a moment being, ok
 185 2010-11-21 02:05:56 <jgarzik> theymos: bot passed If-Modified-Since header, which will return only headers if-and-only-if your CGI script returns a Last-Modified header for comparison.  With no basis for comparison, Apache cannot perform that optimization.
 186 2010-11-21 02:06:19 <xelister> CNN:  A former U.S. soldier could face the death penalty after being convicted of raping a 14-year-old Iraqi girl and killing her and her family.
 187 2010-11-21 02:06:22 <jgarzik> ...returns a Last-Modified header, and the LM header is older than If-Modified-Since input.
 188 2010-11-21 02:06:36 <xelister> oh wait, death penalty. That is against my point :) Let me google the other example.
 189 2010-11-21 02:06:50 <lfm> hehe
 190 2010-11-21 02:07:10 <uberjar> bitcoin related noob question:  Is the bitcoin network supposed to be resistant to a "hacker attack" ? Couldn't any old script kiddy with a modest botnet of a few hundred thousand nodes easily create his own client and take over the network ?
 191 2010-11-21 02:07:29 <jgarzik> uberjar: only if the botnet has good graphics cards
 192 2010-11-21 02:07:31 <Kiba> uberjar: he have to fight against GPU farms
 193 2010-11-21 02:07:32 <lfm> uberjar, why would they?
 194 2010-11-21 02:07:46 <Kiba> plus, he would compromise the integrity of bitcoins
 195 2010-11-21 02:07:49 <uberjar> lfm, look at the attack the RIAA made against torrent sites
 196 2010-11-21 02:07:52 <Kiba> AND make bitcoins worthless
 197 2010-11-21 02:08:13 <uberjar> I see.. so it's not about number of nodes it's about GPU power
 198 2010-11-21 02:08:25 <Kiba> it's about hashing rate
 199 2010-11-21 02:08:35 <uberjar> so a nice beowulf cluster made out of PS3s could do the trick
 200 2010-11-21 02:08:35 <jgarzik> uberjar: total CPU power (and currently, GPU's have more power than most CPUs)
 201 2010-11-21 02:08:48 <jgarzik> probably not
 202 2010-11-21 02:08:49 <lfm> uberjar, the riaa is trying to stop millions of dollars lost to pricay. what would motivate a botnet such as you propose? There isn't that much money in bitcoin
 203 2010-11-21 02:08:51 <Kiba> assuming, he got enough of PS3s
 204 2010-11-21 02:08:59 <jgarzik> PS3 graphics engine is nowhere near ATI
 205 2010-11-21 02:09:15 <jgarzik> that's like getting a bunch of 486's
 206 2010-11-21 02:09:24 <joose> no it couldnt, beowulf is old, antiquated bullshit
 207 2010-11-21 02:09:28 <uberjar> lfm, I'm just saying.. if bitcoin ever does "take off" and get a fair amount of money in it..  it would be an easy target
 208 2010-11-21 02:09:41 <jgarzik> uberjar: not really
 209 2010-11-21 02:09:45 <joose> the PS3 has a nice processor
 210 2010-11-21 02:09:47 <Kiba> bigger GPU farms, bigger difficulty
 211 2010-11-21 02:09:57 <joose> and we dont use beowulf man, we use MPI and openMP
 212 2010-11-21 02:09:59 <uberjar> amazon just rolled out it's GPU instances
 213 2010-11-21 02:09:59 <lfm> uberjar, if bitcoin ever does take off it would be a lot harder then to subvert
 214 2010-11-21 02:10:07 <Kiba> double spending = compromsing securities
 215 2010-11-21 02:10:08 <jgarzik> uberjar: amazon uses NV... slow
 216 2010-11-21 02:10:11 <Kiba> amazon cost money.
 217 2010-11-21 02:10:13 <joose> there are a finite number of bitcoins, as a mathematical fact
 218 2010-11-21 02:10:29 <Kiba> compromising securities = no money for script kiddie.
 219 2010-11-21 02:10:33 <joose> it will limit itself
 220 2010-11-21 02:10:37 <uberjar> so you're saying by the time bitcoin has a lot of money in it.. it would not be cost effective to use cloud computing to take over
 221 2010-11-21 02:10:39 <joose> the mineing
 222 2010-11-21 02:10:47 <uberjar> interesting
 223 2010-11-21 02:10:48 <joose> absolutely
 224 2010-11-21 02:10:52 <Kiba> cloud computing take money.
 225 2010-11-21 02:10:58 <joose> yeah that too
 226 2010-11-21 02:11:03 <lfm> uberjar, thats the plan, we will see we hope
 227 2010-11-21 02:11:09 <Kiba> and you don't gain anything by destroying bitcoins, uberjar
 228 2010-11-21 02:11:17 <Kiba> only the government would gain anything.
 229 2010-11-21 02:11:18 <jgarzik> uberjar: not only not cost effective... probably quite difficult, given the likely CPU power that will be in bitcoin in 12 months time
 230 2010-11-21 02:11:21 <jgarzik> or 24 months, ...
 231 2010-11-21 02:11:22 <joose> i wana try top set up a mine, now, im not sure how easy it is to find them these days. I do have access to very very serious GPU power, teslas, a few
 232 2010-11-21 02:11:39 <jgarzik> joose: you want ATI.  NV is crap.
 233 2010-11-21 02:11:47 <xelister> ok gound it
 234 2010-11-21 02:11:47 <joose> i didnt buy them man
 235 2010-11-21 02:11:54 <xelister> USA Marine accused of raping a 14-year-old Japanese girl was convicted to.... 4 years in prison
 236 2010-11-21 02:11:58 <joose> i use NV by habbit from the early days when it always worked way better with *nix
 237 2010-11-21 02:11:59 <uberjar> ok thanks those were good explainations.  I'll read further
 238 2010-11-21 02:12:02 <xelister> ^--- http://www.militarytimes.com/forum/showthread.php?1564663-Staff-Sgt.-Tyrone-L.-Hadnott-Lesser-charge-in-rape-case-gets-Marine-4-years
 239 2010-11-21 02:12:06 <xelister> ^--- Kiba
 240 2010-11-21 02:12:21 <lfm> xelister, that was BY JAPANESE court!
 241 2010-11-21 02:12:22 <jgarzik> xelister: are these examples statistically relevant in any way?  No.  :)
 242 2010-11-21 02:12:29 <Kiba> JAPANESE court?
 243 2010-11-21 02:12:36 <joose> that marine will have his punishment. the rest of his life will be shit, for sure. not that it already isnt when your a marine, but a known child raping marine....
 244 2010-11-21 02:12:43 <xelister> lfm: yeah, they always give 4 years to prison. I BET THE COURT WAS NOT LIKE, INFLUENCED BY USA, OR ANYTHING
 245 2010-11-21 02:12:50 <joose> 1/3 of military women get raped
 246 2010-11-21 02:12:50 <xelister> for rape of underaged
 247 2010-11-21 02:12:59 <xelister> joose: what?
 248 2010-11-21 02:13:04 <lfm> xelister, so who isnt influenced by someone
 249 2010-11-21 02:13:10 <Kiba> xelister: dude, a man got the death plenty for killing an entire family
 250 2010-11-21 02:13:22 <joose> 33% of women in the military have been raped in the military
 251 2010-11-21 02:13:22 <Kiba> but this guy got convicted and spend 4 years in jail?
 252 2010-11-21 02:13:25 <Kiba> DISPARENCY
 253 2010-11-21 02:13:31 <xelister> USA solider negr^H afroamerican, had forced a civilian Japanes girl - raped her. And got 4 years
 254 2010-11-21 02:13:33 <jgarzik> xelister: US military kills people and breaks things, and citizens pay tax money for this.  Rape almost never occurs, so it is fallacious to assume people pay tax money for rape.
 255 2010-11-21 02:13:35 <Kiba> joose: evidence?
 256 2010-11-21 02:13:46 <xelister> if it would be my decission, that solider would be killed
 257 2010-11-21 02:14:02 <xelister> and so would it be if usa president/gov would have balls or any sense of morality
 258 2010-11-21 02:14:12 <xelister> and for this, the voters are responsible
 259 2010-11-21 02:14:14 <Kiba> raping is traumatic, but you still have a life.
 260 2010-11-21 02:14:26 <lfm> killing is not moral, not even death penalty is moral
 261 2010-11-21 02:14:26 <uberjar> the US army is not raping as a policy like Japan did
 262 2010-11-21 02:14:40 <uberjar> the Japanese systematically raped eveyr woman in Nan Ching
 263 2010-11-21 02:14:42 <joose> http://articles.cnn.com/2008-07-31/us/military.sexabuse_1_sexual-assault-sexual-abuse-military-service?_s=PM:US
 264 2010-11-21 02:14:42 <uberjar> or however you spell it
 265 2010-11-21 02:14:46 <xelister> Kiba: how much would you then reocommand for raping 14 year old girl
 266 2010-11-21 02:15:03 <lfm> xelister, doesnt matter what age they are
 267 2010-11-21 02:15:04 <Kiba> xelister: I don't know. A very expensive fine. And some years in prison. Like 10 years.
 268 2010-11-21 02:15:05 <joose> the hurt, that is put upon a person by child rape, it lasts for GENERATIONS man
 269 2010-11-21 02:15:14 <joose> my ex girl, her mom was raped by her dad
 270 2010-11-21 02:15:15 <xelister> Americans are OH NOOOOE  PEDOPHILES EVERYWHERE KILL THEM...  but when their own "solider" actually RAPES someone (rapes as in forces, not as in - find 17 yo girl that loves him) - then 4 year sentence
 271 2010-11-21 02:15:22 * jgarzik waits patiently for the next block
 272 2010-11-21 02:15:23 <jgarzik> tap tap tap
 273 2010-11-21 02:15:27 <xelister> WHY was that solider not also convicted by usa court to 25 years or life
 274 2010-11-21 02:15:30 <joose> and i have been dealing with the effects of this still, a generation and social connection later
 275 2010-11-21 02:15:33 <jgarzik> xelister: soldier
 276 2010-11-21 02:15:39 * Kiba patiently waits for xelister to calm down.
 277 2010-11-21 02:15:40 <lfm> xelister, I guess there are no pedophiles in your country
 278 2010-11-21 02:15:45 <Kiba> Look, you're spectulating.
 279 2010-11-21 02:15:57 <Kiba> you say the Japanese court is influenced, but you have no evidence
 280 2010-11-21 02:16:05 <Kiba> how many years does a typical Japanese person get for raping?
 281 2010-11-21 02:16:20 <xelister> lfm: people averywhere, always, liked sex with around girls that are sexually matured, and that happens around 14-16 depending on individual afair
 282 2010-11-21 02:16:23 <joose> i wouldnt be surprised if its less than here
 283 2010-11-21 02:16:23 <jgarzik> 40 minutes since last block.  at higher difficulties, I wonder if the time between blocks varies more?
 284 2010-11-21 02:16:37 <joose> how many hash per sec you got going jgarzik?
 285 2010-11-21 02:16:44 <uberjar> Kiba, the japanese army systematically raped > 50k people within 6 days
 286 2010-11-21 02:16:57 <joose> and hardware? GPU? where can you get the GPU miner code?
 287 2010-11-21 02:16:57 <xelister> lfm: and some people learned to controll it,  while other people totally do not controll any urges and even become actuall rapist (as in have forced sex, not as in have 16 years and find same-aged girl)
 288 2010-11-21 02:16:59 <uberjar> http://en.wikipedia.org/wiki/Nanking_Massacre
 289 2010-11-21 02:17:01 <Kiba> uberjar: well, I don't mean that.. The American rpaist.
 290 2010-11-21 02:17:05 <lfm> jgarzik, shouldnt
 291 2010-11-21 02:17:05 <xelister> joose: forum. there are 2 miners
 292 2010-11-21 02:17:26 <joose> ahh, the code public?
 293 2010-11-21 02:17:33 <xelister> uberjar: where/when
 294 2010-11-21 02:18:02 <xelister> uberjar: so they are stupid nation as well.  Btw, my nations is probably also stupid in this way, dunno fortunatelly latelly there where no such incidents that I know of
 295 2010-11-21 02:18:08 <uberjar> xelister, Nanking china December 13, 1937
 296 2010-11-21 02:18:19 <Kiba> you're calling an entire people....stupid.
 297 2010-11-21 02:18:20 <xelister> joose: yes
 298 2010-11-21 02:18:29 <uberjar> I'm just saying as far as modern warfare goes the US military hardly does any raping
 299 2010-11-21 02:18:31 <lfm> xelister, your nation is better at covering them up. doesnt make you better
 300 2010-11-21 02:18:51 <xelister> Kiba: no, this do not apply to each person,  but to nation and government and peraitlly culture/tradition as a whole.
 301 2010-11-21 02:19:30 <xelister> lfm: if that is the fact then you are right. But Im not arguing that we are better. But we had far less military involvment. And certainly we are not fucking smug as Americans.
 302 2010-11-21 02:19:36 <joose> ahh FUCK, thats right, its all for windows right? the miners
 303 2010-11-21 02:19:46 <xelister> joose: no. they run perfect on linux
 304 2010-11-21 02:19:47 <jgarzik> joose: plenty of Linux GPU miners
 305 2010-11-21 02:19:57 <xelister> well "perfect" is a strong word, but they run after some configuration
 306 2010-11-21 02:20:00 <jgarzik> the wiki needs a GPU mining page
 307 2010-11-21 02:20:04 <jgarzik> for times like this
 308 2010-11-21 02:20:07 <xelister> yea
 309 2010-11-21 02:20:19 <xelister> with examples of setup, exact config, and the results + workarounds
 310 2010-11-21 02:20:20 <Kiba> xelister: doesn't make sense.
 311 2010-11-21 02:20:23 <lfm> xelister,  so all smug people must die? interesting point of view. I don't think I will support it.
 312 2010-11-21 02:20:45 <xelister> lfm: did I said someone should die? Other then rapists, that should (well, ok, or 25 to lifetime)
 313 2010-11-21 02:20:47 * Kiba subscribes to methodical individualism
 314 2010-11-21 02:21:07 <Kiba> incarnation is a cost to society.
 315 2010-11-21 02:21:12 <Kiba> an expensive fine is better.
 316 2010-11-21 02:21:29 <Kiba> but that could have hidden cost too..
 317 2010-11-21 02:21:50 <lfm> kiba so rich people can afford to be criminals?
 318 2010-11-21 02:21:51 <uberjar> it wouldn't be so expensive if the prisons were not packed full of non-violent drug offenders
 319 2010-11-21 02:22:20 <uberjar> the prison industrial complex is big business though
 320 2010-11-21 02:22:21 <Kiba> lfm: prisons have hidden cost too. Make people more violent.
 321 2010-11-21 02:22:29 <xelister> my point was that, there is always war on something. That grows out of proportion and becomes ridiculous. Like war on "pedophiles" now means "two 17 years olds in love, are SEXUAL PREDATORS if they send each other OWN naked pictures". What the fuck.  And on antoher hand, when someone ACTUALLY /RAPES/ a girl, he doesnt have too much trouble, 4 years in prison, meh, and zero (no trial) in USA. How this makes sense
 322 2010-11-21 02:22:37 <uberjar> without the prison industry america would have no economy left at all :D
 323 2010-11-21 02:23:05 <Kiba> xelister: how many years ago was that?
 324 2010-11-21 02:23:07 <xelister> not to mention now "war on air terrorists" means "lets strip search everyone including 3 yeard old girls" - also this a bit conflicts with previous war on pedo ;)
 325 2010-11-21 02:23:16 <lfm> xelister, well why have a trial if they plead guilty?
 326 2010-11-21 02:23:52 <xelister> Kiba: it was in 2008 yaer
 327 2010-11-21 02:23:55 <xelister> *year
 328 2010-11-21 02:24:07 <xelister> Kiba: http://en.wikipedia.org/wiki/2008_in_Japan -> find "Hadnott"
 329 2010-11-21 02:25:36 <xelister> lfm: in the end he got 4 years. For ACTUALL FORCING GIRL TO SEX / rape.   While, you can get MUCH HIGHER sentences, for like: smoking a bit of weed in own home; touching boobie of 17 year old girl (even if you are yourslef in that age and she is your girlfriend and wanted to do it). And so on.
 330 2010-11-21 02:25:48 <Kiba> xelister: fuck, this is so little information
 331 2010-11-21 02:25:49 <jgarzik> this is weird:
 332 2010-11-21 02:25:50 <jgarzik> http://theymos.ath.cx:64150/bbe/tx/4b3f52c3843cc7b01c24ec6dd2e04735a1069f360017bb4d68cd4b7a6b842e73
 333 2010-11-21 02:25:58 <jgarzik> two inputs, 1 BTC each.  two outputs, 1 BTC each.
 334 2010-11-21 02:26:05 <jgarzik> seems like a non-standard transaction?
 335 2010-11-21 02:26:22 <uberjar> American girls don't have to deal with squid monsters comming out of the sea and raping 5 or 6 girls at a time like they do in japan.
 336 2010-11-21 02:26:32 <xelister> Kiba: yeah this sucks that wiki doesnt say more. If you are for /reall/ interested in fate of children (not just shouting "omg war on *") then why dont you google more (and please /msg me if anything - I may be afk)
 337 2010-11-21 02:26:52 <xelister> uberjar: uh, what? ;)
 338 2010-11-21 02:26:54 <jgarzik> 50% of that is a TX to self
 339 2010-11-21 02:26:55 <Kiba> xelister: I don't have time to investigate one of the many injustice.
 340 2010-11-21 02:27:00 <jgarzik> new kind of TX spam?
 341 2010-11-21 02:27:06 <jgarzik> at least it's low volume.
 342 2010-11-21 02:27:09 <Kiba> this is. One example.
 343 2010-11-21 02:27:32 <Kiba> and no description of how proceeding goes and what the heck they found out
 344 2010-11-21 02:27:45 <xelister> Kiba: yes, americans never have time to do that, what they do have time for, is shouting next whichhunt on soemthing, while continuing to think they are the most just, free, democratic and intelligent nation in the world. That was my point :)
 345 2010-11-21 02:28:06 <Kiba> xelister: and you come to this conclusion, how?
 346 2010-11-21 02:28:15 <xelister> of course I here refere to most of population,  while some people can be in fact totally different
 347 2010-11-21 02:28:21 <Kiba> by hearing my fellow idiotic Americans who are noationalist
 348 2010-11-21 02:28:21 <uberjar> xelister, http://en.wikipedia.org/wiki/Tentacle_erotica
 349 2010-11-21 02:28:27 <tcatm> UukGoblin: are you around?
 350 2010-11-21 02:28:37 <Kiba> msot of the population? How?
 351 2010-11-21 02:28:44 <xelister> uberjar: sorry to bring the news to you, this hentai movies are not actually portraiting real-life events ;)
 352 2010-11-21 02:29:16 <xelister> while that guy did actually hurt a girl - and look, noone gives two shits, even hard to google details on that
 353 2010-11-21 02:29:23 <uberjar> it predates anime by at least a hundred years
 354 2010-11-21 02:29:35 <uberjar> there are scrolls from japan that date back to 1814 that show giant squid having their way with women
 355 2010-11-21 02:29:50 <Kiba> xelister: shit? People have jobs.
 356 2010-11-21 02:29:59 <Kiba> their own problems to worry about.
 357 2010-11-21 02:30:16 <xelister> well I have like 3 jobs and yet I took time to google it as requested
 358 2010-11-21 02:30:27 <xelister> one of which is fixing either m0ms or diablo's miner =)
 359 2010-11-21 02:30:34 <xelister> (or my box)
 360 2010-11-21 02:30:48 <Kiba> xelister: sure. You have 3 jobs. You are interested in this topic. You are aware of it. How many millions are aware of tiny incidents amongst the news?
 361 2010-11-21 02:31:16 <uberjar> you mean like tiger woods cheating on his wife ?
 362 2010-11-21 02:31:20 <uberjar> I'd say most of the poplulation
 363 2010-11-21 02:31:30 <Kiba> uberjar: well, I don't care.
 364 2010-11-21 02:31:42 <theymos> jgarzik: Bitcoin does those useless change transactions pretty frequently. I don't know why.
 365 2010-11-21 02:31:44 <xelister> Kiba: yeah - averaged American seems to not realize how much hurt his nation is doing -> which is the reason of beliving his nation is best and blessed by god, and nothing needs to be changed in the way it influences other people in and outside of usa
 366 2010-11-21 02:31:57 <Kiba> xelister: and how you determine that exactly?
 367 2010-11-21 02:32:02 <Kiba> did you survey random Americans?
 368 2010-11-21 02:32:45 <Kiba> so you hate all Americans, despite saying that you don't hate some Americans.
 369 2010-11-21 02:32:52 <xelister> well, america is alway starting war on OMG something
 370 2010-11-21 02:32:54 <uberjar> there are a handful of liberal cities in the US where people are aware of US atrocities (new york, san francisco, portland, etc)
 371 2010-11-21 02:32:57 <xelister> where something changes each 5-10 years
 372 2010-11-21 02:33:03 <Kiba> and you say majority, but you don't give me how you come to this conclusion
 373 2010-11-21 02:33:06 <xelister> as if there would be no more real and more pressing ACTUALL problems
 374 2010-11-21 02:33:27 <theymos> jgarzik: If you look at Bitcoin Faucet transactions, you will see that maybe 1 in 10 transactions have a 0.05 input, a 7 input or whatever, and then sends 0.05 and change. Could be an intentional anonymity feature.
 375 2010-11-21 02:33:57 <xelister> Kiba: it seems americans support gov in this. unless it directly affects them like in TSA scans case
 376 2010-11-21 02:34:06 <Kiba> Like I say, HOW?
 377 2010-11-21 02:34:15 <uberjar> by paying taxes
 378 2010-11-21 02:34:18 <xelister> this is my general opinion from many sources from news to talks
 379 2010-11-21 02:34:18 <Kiba> what is the basis of your conclusion? What survey? statistics?
 380 2010-11-21 02:34:20 <xelister> you think Im wrong
 381 2010-11-21 02:34:22 <xelister> also, waht uberjar said
 382 2010-11-21 02:34:29 <uberjar> if you pay your federal income tax you're supporting the military industrial complex
 383 2010-11-21 02:34:33 <Kiba> xelister: Don't assume you know what others think.
 384 2010-11-21 02:34:37 <xelister> Kiba: you say I am wrong to think that most Americans do support things like
 385 2010-11-21 02:34:40 <jgarzik> and the alternative to paying taxes is....?
 386 2010-11-21 02:34:47 <Kiba> NO, I didn't say you're wrong.
 387 2010-11-21 02:34:57 <Kiba> I am asking you how you come to this conclusion, specifically?
 388 2010-11-21 02:35:09 <uberjar> the alternative to paying taxes is the creation of black markets
 389 2010-11-21 02:35:14 <xelister> invading country to Destroy WoMDs ^H no wait, to remove the tyran.. ^H^H^H to bring the democracy... oh, btw that country has most oil (and afghan have tons of gold)
 390 2010-11-21 02:35:24 <Kiba> uberjar: or JAIL. Face it, we're all pussies.
 391 2010-11-21 02:35:33 <uberjar> hahhaa
 392 2010-11-21 02:35:38 <xelister> allowing a rapist military guy to get laughtable 4 years in prison, instead of executing him before military court
 393 2010-11-21 02:35:47 <Kiba> xelister: one EXAMPLE.
 394 2010-11-21 02:35:59 <Kiba> how many people support the Iraq wars?
 395 2010-11-21 02:36:10 <xelister> apprently many, dont see any /effective/ protests
 396 2010-11-21 02:36:10 <Kiba> 90%? 55% 25%?
 397 2010-11-21 02:36:14 <Kiba> I want, number.
 398 2010-11-21 02:36:20 <xelister> > 50%
 399 2010-11-21 02:36:26 <Kiba> xelister: then cite it.
 400 2010-11-21 02:36:35 <xelister> > 50% voters,  if the democracy actually works as it should.  By definition
 401 2010-11-21 02:36:44 <jgarzik> xelister: Back on topic for #bitcoin-dev...    http://www.xtranormal.com/watch/7754515/
 402 2010-11-21 02:36:45 <Kiba> support the Iraq war, SPECIFICALLY
 403 2010-11-21 02:36:53 <xelister> if you voted for a guy that does otherwise then you said, then you voted wrongly though
 404 2010-11-21 02:37:01 <Kiba> xelister: 2003.
 405 2010-11-21 02:37:07 <Kiba> 4 years of office.
 406 2010-11-21 02:37:13 <Kiba> but 9/11 occurs on 2001.
 407 2010-11-21 02:37:18 <xelister> Kiba: Im in ##politics , join me :)
 408 2010-11-21 02:37:34 <Kiba> so you're saying we should vote Bushie out.
 409 2010-11-21 02:37:44 <tcatm> theymos: to how many clients is your bitcoind connected?
 410 2010-11-21 02:37:51 * Kiba doesn't believe in voting.
 411 2010-11-21 02:38:06 <theymos> tcatm: 33
 412 2010-11-21 02:38:42 <xelister> Kiba: to end this offtopic for now (sorry, I dont want to cause too much flame here) - I think americans could at least stop looking so 'proud' about it. Ask some people in other countries do they think that americans think they are best
 413 2010-11-21 02:38:48 <tcatm> can you send me a list of those ips?
 414 2010-11-21 02:38:55 <Kiba> how 'proud' are teh Americans?
 415 2010-11-21 02:39:00 <Kiba> percents
 416 2010-11-21 02:39:02 <Kiba> surveys
 417 2010-11-21 02:39:02 <LobsterMan> so bitcoin.exe just crashed on me.....lol
 418 2010-11-21 02:39:14 <xelister> LobsterMan: woot... which revision?
 419 2010-11-21 02:39:16 <LobsterMan> Faulting application name: bitcoin.exe, version: 0.0.0.0, time stamp: 0x4ce24d1a
 420 2010-11-21 02:39:16 <LobsterMan> Faulting module name: bitcoin.exe, version: 0.0.0.0, time stamp: 0x4ce24d1a
 421 2010-11-21 02:39:16 <LobsterMan> Exception code: 0xc0000005
 422 2010-11-21 02:39:16 <LobsterMan> Fault offset: 0x00554693
 423 2010-11-21 02:39:16 <LobsterMan> Faulting process id: 0x1710
 424 2010-11-21 02:39:17 <Kiba> that what I want, not fussy saying.
 425 2010-11-21 02:39:24 <LobsterMan> -.3.15
 426 2010-11-21 02:39:32 <LobsterMan> svn r182or 812 it hink
 427 2010-11-21 02:39:36 <uberjar> xelister, the British are just as proud to be occupying all the countries which they are occupying.  French, Spanish, etc all proud
 428 2010-11-21 02:39:40 <LobsterMan> svn r182 or 812 i think
 429 2010-11-21 02:40:05 <theymos> tcatm: Why?
 430 2010-11-21 02:40:14 <xelister> uberjar: I think they are less. Although they do suck as well ;)  Kiba:  I will continue in ##politics if you would like to join me. brb ~1h.
 431 2010-11-21 02:40:32 <tcatm> theymos: i'm trying to track down the source of those spam TXs
 432 2010-11-21 02:40:37 <Kiba> I am not taking a side, xelister
 433 2010-11-21 02:40:40 <LobsterMan> changing to 186
 434 2010-11-21 02:40:50 <xelister> LobsterMan: getwork does not work with 186 afaik
 435 2010-11-21 02:41:04 <Kiba> I just wanted you to be less sloppy in explaining your dislike of Americans.
 436 2010-11-21 02:41:09 <LobsterMan> http://www.bitcoin.org/smf/index.php?topic=1334.0 according ot this it does
 437 2010-11-21 02:41:12 <xelister> LobsterMan: which miner you use?  for me, m0m's hang since recently, and Diablo's either eats all cpu, or is 40% slower them m0m :((
 438 2010-11-21 02:41:30 <LobsterMan> still using m0's
 439 2010-11-21 02:41:41 <LobsterMan> m0's has worked for me mostly
 440 2010-11-21 02:42:09 <xelister> Kiba: this is a general compiled felling from news, tv shows, talking to people, watching websites, reading about wars, and so on. Perahps I will have time to google some statistics for you too, as I googled that incidents. I will be afk now if anything then pleave me /msg :) Goodnight
 441 2010-11-21 02:42:35 <xelister> LobsterMan: for me too, untill I upgraded ubuntu :[[[
 442 2010-11-21 02:42:51 <LobsterMan> lol
 443 2010-11-21 02:43:05 <xelister> if someone sees Diablo can you tell him Im looking for him to debug his slow as ***** miner ;)
 444 2010-11-21 02:43:18 <xelister> 114m instead 160m
 445 2010-11-21 02:43:45 <xelister> 114m instead 160m on SDK 2.1.   On SDK 2.2 I get full 160m, but desktop is as slow as m0m's with -f 5
 446 2010-11-21 02:46:02 <xelister> wait
 447 2010-11-21 02:46:05 <xelister> ok...
 448 2010-11-21 02:49:32 <xelister> Diablo's on SDK 2.2 with -f 200 have latencies a bit worse then m0m's and uses 200% cpu (m0m's used 100%).     On lower -f it makes desktop unusable.     AND  On SDK 2.1, it always eats almost 0 cpu, even on -f 2, BUT it geneates just 113 M instead 160 M.
 449 2010-11-21 02:50:52 <xelister> LobsterMan: how easy was the pyopencl installation under windows?  Overall, did you had to build anything?
 450 2010-11-21 02:53:36 <theymos> tcatm: I don't see how a list of IPs would help. You can already get a list of all online nodes by looking in your addr.dat file.
 451 2010-11-21 02:54:04 <tcatm> ah just asking the nodes for more addrs works
 452 2010-11-21 02:54:12 <tcatm> i just run into "too many open files" :)
 453 2010-11-21 02:56:50 <jgarzik> tcatm: due to bitcoin?  do you set any ulimit's?
 454 2010-11-21 02:57:11 <jgarzik> tcatm: does netstat listen a bunch of bitcoin-related sockets?
 455 2010-11-21 02:57:16 <jgarzik> s/listen/list/
 456 2010-11-21 02:57:58 <xelister> tcatm:  ulimit -n   or -a and read
 457 2010-11-21 02:58:05 <tcatm> i hacked a client to connect to as many nodes as possible and print a every transactions it receives with a timestamp
 458 2010-11-21 02:58:13 <xelister> you uber node you
 459 2010-11-21 02:58:22 <tcatm> i can then see which node sent which transactions first
 460 2010-11-21 02:58:33 <xelister> same style of attack exist in freenet btw. well probably in all opennet like p2p
 461 2010-11-21 02:58:59 <xelister> sad thing is even FN had not yet protect from fully from attack as tcatm described
 462 2010-11-21 02:59:08 <jgarzik> tcatm: you're probably only seeing 10-20% of the network
 463 2010-11-21 02:59:52 bertodsera has quit (Quit: No Ping reply in 180 seconds.)
 464 2010-11-21 03:00:09 <jgarzik> still, you can indeed deduce popular relays
 465 2010-11-21 03:00:11 bertodsera has joined
 466 2010-11-21 03:01:22 <xelister> perhaps spammer is using tor
 467 2010-11-21 03:01:44 <xelister> well anyway, just the client should be protected. If some kid can do it, then what can bigger company/corp do.
 468 2010-11-21 03:05:13 <xelister> I suspect even developers may sometime do such attack, as a healthy training
 469 2010-11-21 03:07:30 <theymos> The spammer is on this channel, so he's probably using Tor now...
 470 2010-11-21 03:07:32 xelister has quit (Read error: Connection reset by peer)
 471 2010-11-21 03:11:10 <tcatm> i should use threads :/
 472 2010-11-21 03:11:39 <tcatm> 160 connections, 100% cpu usage
 473 2010-11-21 03:11:46 Diablo-D3 has joined
 474 2010-11-21 03:11:47 <jgarzik> another block with 800+ TX's
 475 2010-11-21 03:11:53 <Diablo-D3> ;;bc,estimate
 476 2010-11-21 03:11:54 <gribble> 6562.70475419
 477 2010-11-21 03:13:37 dwdollar has joined
 478 2010-11-21 03:13:55 <jgarzik> theymos: <shrug>  it was only a matter of time until somebody ran "while (true) { send $foo message to network }"
 479 2010-11-21 03:14:18 <jgarzik> so I tend to think these basic things (like Tor) are obvious
 480 2010-11-21 03:14:55 <jgarzik> 5.08 blocks/hour, over past 24h
 481 2010-11-21 03:17:22 <tcatm> there's one IP where 5x as many transactions are from than from others
 482 2010-11-21 03:17:38 <theymos> Which one?
 483 2010-11-21 03:18:02 <tcatm> 69.164.218.197
 484 2010-11-21 03:18:13 <theymos> I am connected to him.
 485 2010-11-21 03:18:59 T_X has quit (Read error: Operation timed out)
 486 2010-11-21 03:19:04 T_X has joined
 487 2010-11-21 03:20:31 * jgarzik wonders:  Alice sends to Bob 1 BTC, 10000 times.  Then Bob sends 10000 BTC to someone.  That would create an over-large transaction, would it not?
 488 2010-11-21 03:20:44 <theymos> jgarzik: Yes.
 489 2010-11-21 03:24:22 jercos has quit (Quit: Lost terminal)
 490 2010-11-21 03:26:46 xelister has joined
 491 2010-11-21 03:26:59 <xelister> what was my quit message?
 492 2010-11-21 03:27:15 <xelister> my radeon crashe'd
 493 2010-11-21 03:27:29 <theymos> Read error: Connection reset by peer
 494 2010-11-21 03:28:48 <jgarzik> tcatm: in your 'Asks' column, it should sort in ascending, not descending, order
 495 2010-11-21 03:29:04 <jgarzik> tcatm: so the highest bid, and lowest ask, are both shown in same screen location
 496 2010-11-21 03:29:43 <jgarzik> theymos: do you have a BBE also watching testnet?
 497 2010-11-21 03:30:03 <theymos> jgarzik: http://theymos.ath.cx:64150/testnet/bbe/
 498 2010-11-21 03:32:08 <tcatm> jgarzik: fixed that. thanks. i had the ask list ontop the bids first and forgot to change it
 499 2010-11-21 03:32:15 FreeMoney has joined
 500 2010-11-21 03:32:35 <jgarzik> tcatm: your stuff looks really nice.  well done.
 501 2010-11-21 03:33:05 * jgarzik goes to dig out the test network faucet info
 502 2010-11-21 03:33:22 <jgarzik> ah hah: https://freebitcoins.appspot.com/TEST/
 503 2010-11-21 03:34:01 <brocktice> Is it the spam that's dropping block production so much?
 504 2010-11-21 03:34:07 <brocktice> Because people haven't updated getwork?
 505 2010-11-21 03:34:09 <brocktice> still?
 506 2010-11-21 03:34:40 <xelister> brocktice: you mean?
 507 2010-11-21 03:34:46 <xelister> brocktice: version 183 should work fine?
 508 2010-11-21 03:34:49 <xelister> Diablo-D3: oh, hai
 509 2010-11-21 03:34:58 <jgarzik> brocktice: I don't think the spam is messing with block production much
 510 2010-11-21 03:35:15 <brocktice> Well, older versions of getwork are vulnerable
 511 2010-11-21 03:35:18 <xelister> Diablo-D3: frankly, your miner appears to suck at my config
 512 2010-11-21 03:35:20 <jgarzik> comparatively, this "attack" is quite light, and the network is handling it well, IMO
 513 2010-11-21 03:35:20 <brocktice> they rebuild the tx list every time
 514 2010-11-21 03:36:56 <xelister> Diablo-D3: on sdk 2.2 it is as good as m0m but uses over 200% cpu!  and on sdk 2.1 it uses alost 0 cpu but is 115 instead 160 Mhash
 515 2010-11-21 03:37:32 * jgarzik wishes he still had that crappy old Hostway (nee ValueWeb) dedicated server...  I'd stick it on testnet
 516 2010-11-21 03:37:49 <brocktice> xelister: 5xxx card?
 517 2010-11-21 03:38:06 <xelister> brocktice: yea, 5770. I did edited the 3 lines in diablo's .cl
 518 2010-11-21 03:38:16 <brocktice> hm
 519 2010-11-21 03:38:25 <xelister> brocktice: btw how you check catalyst driver version
 520 2010-11-21 03:38:28 <brocktice> I got slightly lower performance with his miner than m0's
 521 2010-11-21 03:38:36 <brocktice> xelister: apt-cache show fglrx
 522 2010-11-21 03:38:41 <brocktice> should give you an 8.something version
 523 2010-11-21 03:38:53 <brocktice> that is not equal to catalyst version
 524 2010-11-21 03:38:55 <brocktice> but you can infer it
 525 2010-11-21 03:39:01 <xelister> 2:8.780-0ubuntu2
 526 2010-11-21 03:39:11 <xelister> yeah, but how to check that not in package manager
 527 2010-11-21 03:39:13 <brocktice> huh
 528 2010-11-21 03:39:23 <brocktice> 8.780 is probably 10.9 or something?
 529 2010-11-21 03:39:34 <brocktice> 10.10 is 8.783
 530 2010-11-21 03:40:05 <brocktice> 8.753 is 10.7
 531 2010-11-21 03:40:17 <brocktice> those are the two I know (10.10 and 10.7)
 532 2010-11-21 03:40:46 <brocktice> xelister: but my performance difference was not that large
 533 2010-11-21 03:40:53 <brocktice> xelister: weren't you writing your own too?
 534 2010-11-21 03:40:57 Xanie has joined
 535 2010-11-21 03:41:14 <xelister> brocktice: I was going to use one as a base
 536 2010-11-21 03:41:18 <xelister> diablo's have issues
 537 2010-11-21 03:41:29 Toadyonps3 has quit (Ping timeout: 240 seconds)
 538 2010-11-21 03:41:36 <xelister> m0m was hanging for me. but perhaps m0m is easier to install after all? on windows
 539 2010-11-21 03:41:41 <brocktice> I have a daemon version of m0's that I'm still developing.
 540 2010-11-21 03:41:43 <brocktice> https://github.com/brocktice/poclbm
 541 2010-11-21 03:41:48 <brocktice> but it's geared to linux
 542 2010-11-21 03:41:50 <xelister> LobsterMan: how you installed m0ms, needed to build anythingg?
 543 2010-11-21 03:42:01 <brocktice> reminds me I need to push
 544 2010-11-21 03:43:34 <brocktice> pushed
 545 2010-11-21 03:44:08 <brocktice> jgarzik: if spam isn't messing with block production, what is?
 546 2010-11-21 03:44:27 <brocktice> bitcoin watch is showing 5.17/hour for the last 24 hours
 547 2010-11-21 03:44:44 <brocktice> unless it's attrition from the diff increase
 548 2010-11-21 03:44:46 <xelister> I wonder which miner to base on. ideas?  brocktice?
 549 2010-11-21 03:45:10 <brocktice> xelister: I recommend m0m's/mine, if you're comfortable with python
 550 2010-11-21 03:45:23 <brocktice> it's kind of the bare minimum
 551 2010-11-21 03:45:29 <xelister> Im more comfortable wiht java I thought... but python is fine
 552 2010-11-21 03:45:32 <xelister> code is triviall as shit
 553 2010-11-21 03:45:37 <xelister> except the opencl part
 554 2010-11-21 03:45:53 <brocktice> Diablo's doesn't change the opencl part anyway.
 555 2010-11-21 03:45:58 <xelister> well m0m's was hanging for me. I wonder if it does it now as well
 556 2010-11-21 03:46:13 <OneFixt> ;;bc,estimate
 557 2010-11-21 03:46:14 <gribble> 6609.04572850
 558 2010-11-21 03:46:28 <brocktice> OneFixt: bravo on the price slide call, btw
 559 2010-11-21 03:46:31 <xelister> is the spam aimed at lowering diff?
 560 2010-11-21 03:46:32 <brocktice> nice work
 561 2010-11-21 03:46:37 <brocktice> xelister: I think it may be
 562 2010-11-21 03:46:43 <OneFixt> brocktice: Thanks =)
 563 2010-11-21 03:46:50 <xelister> but main miners are not affected, are they?
 564 2010-11-21 03:46:55 <brocktice> that or punishing gpu miners that haven't updated their getwork
 565 2010-11-21 03:47:02 <brocktice> main meaning cpu? no
 566 2010-11-21 03:47:08 <xelister> so what is 183 status of affection by this? its latest rev that works with getwork patch in the current form by m0m
 567 2010-11-21 03:47:29 <brocktice> xelister: the latest getwork patch fixes it
 568 2010-11-21 03:47:31 <OneFixt> brocktice: I'm currently betting on price going up, even though it's not yet obvious.
 569 2010-11-21 03:47:34 <brocktice> I have no slowdowns.
 570 2010-11-21 03:47:44 <brocktice> OneFixt: I think with diff up supply is going to finally be reduced, so I agree.
 571 2010-11-21 03:47:48 <xelister> brocktice: which getwork patch? used on which bitcoind revision?
 572 2010-11-21 03:47:55 <brocktice> xelister: latest of m0's
 573 2010-11-21 03:47:58 <brocktice> let me check revision
 574 2010-11-21 03:47:59 <OneFixt> brocktice: My thoughts exactly.
 575 2010-11-21 03:48:24 <brocktice> that's strange, "svnversion ." reports "exported"
 576 2010-11-21 03:48:27 <brocktice> never seen that before
 577 2010-11-21 03:48:42 <brocktice> oh, duh
 578 2010-11-21 03:48:45 <xelister> you did svn export and cd into that?
 579 2010-11-21 03:48:46 <brocktice> that's a git repository
 580 2010-11-21 03:48:48 <xelister> lol
 581 2010-11-21 03:49:02 <brocktice> I have 181
 582 2010-11-21 03:49:07 <xelister> 181 is old
 583 2010-11-21 03:49:14 <xelister> 183 works still with getwork
 584 2010-11-21 03:49:15 * brocktice shrugs 
 585 2010-11-21 03:49:40 <brocktice> getwork was updated with the necessary workaround first to match 181
 586 2010-11-21 03:49:42 <xelister> how do I see if my node is affected? I generate coins like x2 slower?  and any more obvious instantly hints?
 587 2010-11-21 03:49:46 <brocktice> and it's working fine for me
 588 2010-11-21 03:49:50 <brocktice> xelister: bitcoind getwork
 589 2010-11-21 03:49:55 <brocktice> should be mostly instantaneous
 590 2010-11-21 03:50:05 <brocktice> if tx flooding is going on and you're affected, it can take several seconds
 591 2010-11-21 03:50:17 <brocktice> you'll see your GPU use percentages dropping periodically, whenever getwork is called
 592 2010-11-21 03:55:32 <xelister> yeah diablo's miner is slower and takes more cpu
 593 2010-11-21 03:55:45 * xelister beats up Diablo-D3 with a wooden club for false advertising ;)
 594 2010-11-21 03:57:16 <brocktice> xelister: Apparently it's faster with 2.2
 595 2010-11-21 03:57:20 xelister has quit (Read error: Connection reset by peer)
 596 2010-11-21 04:00:07 madclown has joined
 597 2010-11-21 04:04:35 xelister has joined
 598 2010-11-21 04:04:40 <tcatm> should i quantize the order book to 2 or 3 decimal places?
 599 2010-11-21 04:04:43 <xelister> this. sucks. donkey. cocks.
 600 2010-11-21 04:05:25 <xelister> apparently running any opengl application when EITHER of miners is running, results in X/fglrx hanging
 601 2010-11-21 04:05:35 <xelister> with high 'success' rate
 602 2010-11-21 04:05:54 <xelister> if you run serveral  windows of  ice amoeba -nosound -windowed -800x600   then do you see same problems?  apt-get install amoeba
 603 2010-11-21 04:06:12 <xelister> try also stopping the amoeba  by pressing ctrl-Z  in its console.  this 'worked' 2 out of 10 tries
 604 2010-11-21 04:06:22 <brocktice> hm, I can try it tomorrow
 605 2010-11-21 04:06:25 <brocktice> I'm not by the miner right now
 606 2010-11-21 04:06:28 <xelister> takes 20 min
 607 2010-11-21 04:06:30 <xelister> ah ok
 608 2010-11-21 04:06:47 <xelister> this all stuff is unstable as shit :<
 609 2010-11-21 04:06:50 <brocktice> also I'm about 5 min from bedtime :)
 610 2010-11-21 04:06:57 <xelister> ArtForz: what linux distro you run again?
 611 2010-11-21 04:07:00 <brocktice> xelister: you're not running it as a dedicated miner huh?
 612 2010-11-21 04:07:03 <brocktice> I think debian
 613 2010-11-21 04:07:07 <xelister> I think also
 614 2010-11-21 04:07:10 <xelister> no, its also desktop
 615 2010-11-21 04:07:23 <brocktice> I'm going to be switching to 5770s in my workstation here on Tuesday
 616 2010-11-21 04:07:29 <brocktice> for eyefinity
 617 2010-11-21 04:07:34 <xelister> well anyhow this means radeons are not good on linux is apps can hang X server
 618 2010-11-21 04:07:47 <brocktice> I'll probably start trying to mine on that when not in use, so I may hit these issues then.
 619 2010-11-21 04:07:50 <xelister> brocktice: what cards you have?
 620 2010-11-21 04:08:00 <brocktice> xelister: right now two 5970s in the miner, water cooled
 621 2010-11-21 04:08:14 <xelister> WHAT CARD YOU HAVE? / SHOW ME YOUR  CARDS    <--- a totally new meaning, thanks to bitcoin  :D
 622 2010-11-21 04:08:17 <brocktice> I also have an air-cooled 5770 I just dropped in but it was overheating and stressing the PSU
 623 2010-11-21 04:08:28 <brocktice> it's going in the workstation when the other 5770 gets here
 624 2010-11-21 04:08:34 <brocktice> I apparently needed an eyefinity 5 card
 625 2010-11-21 04:08:42 <brocktice> regular 5770 can't drive four displays, only 3
 626 2010-11-21 04:08:45 <brocktice> despite having 4 ports
 627 2010-11-21 04:08:58 <brocktice> because the hdmi port is actually wired to one of the DVIs
 628 2010-11-21 04:09:02 <brocktice> or something
 629 2010-11-21 04:10:10 <xelister> what it is about paytxfee? should I set it on my node?
 630 2010-11-21 04:10:16 <brocktice> before water cooling and without 5770: http://www.flickr.com/photos/brocktice/5143440063/
 631 2010-11-21 04:10:28 <brocktice> xelister: I don't think so, I think that's for sending
 632 2010-11-21 04:10:39 <brocktice> for mining I don't believe it makes any difference
 633 2010-11-21 04:10:49 <brocktice> once I clean up the water cooling setup I'll post a pic of that
 634 2010-11-21 04:10:54 <brocktice> it's fugly right now
 635 2010-11-21 04:10:54 <xelister> cool
 636 2010-11-21 04:11:22 <xelister> 1000 W ;] how much it costs?
 637 2010-11-21 04:11:26 <brocktice> $200
 638 2010-11-21 04:11:39 <brocktice> to go to 1.3kW it's around 260-300
 639 2010-11-21 04:11:43 <brocktice> should have got that though to start with
 640 2010-11-21 04:11:47 <brocktice> better expandability
 641 2010-11-21 04:12:23 <xelister> 1000 W
 642 2010-11-21 04:12:27 <xelister> thats more the water boiler
 643 2010-11-21 04:12:36 <xelister> :)
 644 2010-11-21 04:12:46 <brocktice> brb
 645 2010-11-21 04:19:14 <Diablo-D3> [10:48:05] * xelister beats up Diablo-D3 with a wooden club for false advertising ;)
 646 2010-11-21 04:19:14 <Diablo-D3> [10:49:37] <brocktice> xelister: Apparently it's faster with 2.2
 647 2010-11-21 04:19:21 <Diablo-D3> no, its faster with 2.1
 648 2010-11-21 04:24:25 <xelister> Diablo-D3: no, it is slower
 649 2010-11-21 04:24:33 <xelister> on this box
 650 2010-11-21 04:24:54 <xelister> 2.2  170% cpu by java + 30% cpu by xorg, around optimal speed 160,
 651 2010-11-21 04:25:04 <xelister> 2.1  ~10% cpu by java + ~10% cpu by xorg, but speed is 115
 652 2010-11-21 04:25:47 <xelister> btw, m0m's on 2.1sdk has both low-cpu, and full speed. I wonder if it stoped hanging itself
 653 2010-11-21 04:26:26 Xanie has quit (Quit: Leaving)
 654 2010-11-21 04:26:35 <xelister> Diablo-D3: would you like to debug this in next days? best around 12:00 - 3:00 in +02:00 time zone
 655 2010-11-21 04:27:27 <xelister> this box == ubuntu 10.10 64 bit with current fglrx, quad core cpu
 656 2010-11-21 04:32:31 <brocktice> ok, I'm back but now I'm going
 657 2010-11-21 04:32:36 <brocktice> be back tomorrow
 658 2010-11-21 04:33:05 <xelister> Diablo-D3: if you add any debug code, tell me how to get it and I wil try it
 659 2010-11-21 04:36:40 AAA_awright_ has joined
 660 2010-11-21 04:39:02 AAA_awright has quit (Disconnected by services)
 661 2010-11-21 04:39:06 AAA_awright_ is now known as AAA_awright
 662 2010-11-21 04:42:48 doublec has quit (Ping timeout: 245 seconds)
 663 2010-11-21 04:43:03 eureka^ has quit (Ping timeout: 250 seconds)
 664 2010-11-21 04:43:08 HarryS has quit (Read error: Operation timed out)
 665 2010-11-21 04:43:20 soultcer has quit (Quit: ZNC - http://znc.sourceforge.net)
 666 2010-11-21 04:43:53 soultcer has joined
 667 2010-11-21 04:44:18 bd_ has quit (Ping timeout: 264 seconds)
 668 2010-11-21 04:51:34 tcatm has quit (Quit: quit)
 669 2010-11-21 04:52:02 <jgarzik> theymos: BBE is refusing connections
 670 2010-11-21 04:56:40 <jgarzik> brocktice: how loud are the 5970's, when running full steam?
 671 2010-11-21 05:02:09 <theymos> jgarzik: I know. I'm recompiling Apache. It'll be back online in a little while.
 672 2010-11-21 05:03:36 <theymos> Thanks for telling me, though.
 673 2010-11-21 05:04:54 <Diablo-D3> xelister: theres nothing to really debug here
 674 2010-11-21 05:04:59 <Diablo-D3> theres no reason why it should be slower.
 675 2010-11-21 05:05:05 <Diablo-D3> I suspect m0 just doesn't account correctly
 676 2010-11-21 05:10:25 <theymos> BBE is back.
 677 2010-11-21 05:11:20 <OneFixt> I have a question about the hashing in bitcoin.
 678 2010-11-21 05:11:28 <Diablo-D3> [11:17:14] <xelister> 2.2  170% cpu by java + 30% cpu by xorg, around optimal speed 160,
 679 2010-11-21 05:11:28 <Diablo-D3> [11:17:25] <xelister> 2.1  ~10% cpu by java + ~10% cpu by xorg, but speed is 115
 680 2010-11-21 05:11:31 <Diablo-D3> hrm.
 681 2010-11-21 05:11:47 <OneFixt> If I have a system with a throughput of 1000Mbps, what does that get me in terms of bitcoins hashes per second?
 682 2010-11-21 05:13:03 <Diablo-D3> OneFixt: hrm, good question
 683 2010-11-21 05:13:09 <Diablo-D3> OneFixt: because mbps makes no sense in this context
 684 2010-11-21 05:13:18 <OneFixt> well it's megabits of hashed data
 685 2010-11-21 05:13:24 <Diablo-D3> OneFixt: if the system has a high new hash setup cost, then there is a problem
 686 2010-11-21 05:13:42 <Diablo-D3> and not only that, the number will not reflect more optimized usage when available
 687 2010-11-21 05:14:12 <Diablo-D3> OneFixt: that said, divide by 256
 688 2010-11-21 05:14:35 <Diablo-D3> er wait
 689 2010-11-21 05:14:50 <Diablo-D3> that'd be for megabyte per second
 690 2010-11-21 05:15:00 <OneFixt> I guess the question would be this: What's the size of the "blah" that gets hashed into one sha256 hash?
 691 2010-11-21 05:15:08 <Diablo-D3> divide by 2048
 692 2010-11-21 05:15:25 <Diablo-D3> OneFixt: each step of input is 128 bytes
 693 2010-11-21 05:15:31 <Diablo-D3> OneFixt: even if the whole input is less than 128
 694 2010-11-21 05:15:45 <Diablo-D3> and Im fudging that slightly... its less than 128
 695 2010-11-21 05:16:32 <Diablo-D3> OneFixt: the entire bitcoin header takes two passes for example
 696 2010-11-21 05:16:35 <jgarzik> block header is 80 bytes, I'm pretty sure
 697 2010-11-21 05:16:53 <Diablo-D3> wait, maybe I have my math wrong
 698 2010-11-21 05:16:53 <jgarzik> OneFixt: and don't forget it's sha256(sha256(data))
 699 2010-11-21 05:16:56 <Diablo-D3> jgarzik: yes
 700 2010-11-21 05:16:58 <Diablo-D3> ergo
 701 2010-11-21 05:17:00 <Diablo-D3> 256 bytes
 702 2010-11-21 05:18:02 <Diablo-D3> whoops
 703 2010-11-21 05:18:07 <Diablo-D3> its 64 bytes input per pass
 704 2010-11-21 05:18:34 <OneFixt> So if I have a magic elf which hashes at 1000Mbps, how many useful bitcoin hashes per sec would I get out of it?
 705 2010-11-21 05:18:35 <Diablo-D3> so the sha256(sha256()) requires 128 butes input
 706 2010-11-21 05:18:42 <Diablo-D3> OneFixt: divide by 8, divide by 128
 707 2010-11-21 05:18:55 <Diablo-D3> so about 1 mhash/sec
 708 2010-11-21 05:19:16 <Diablo-D3> thats assuming zero cost engine reset, and no available optimizations
 709 2010-11-21 05:19:31 <OneFixt> hmm, pretty slow
 710 2010-11-21 05:30:10 theymos has quit (Remote host closed the connection)
 711 2010-11-21 05:30:53 bd_ has joined
 712 2010-11-21 05:31:02 <OneFixt> what's the fastest hashing that anyone's done so far with an fpga?
 713 2010-11-21 05:32:09 <jrabbit> too fast
 714 2010-11-21 05:33:12 <jgarzik> sha256 isn't very parallel, is it?
 715 2010-11-21 05:33:26 <OneFixt> I seem to remember that it is actually.
 716 2010-11-21 05:33:40 <jgarzik> seems like you'd have more luck putting a ton of sha256 cores on an fpga, than making sha256 massively parallel
 717 2010-11-21 05:34:02 <jgarzik> each round is dependent upon the previous round, is it not?
 718 2010-11-21 05:34:50 <AAA_awright> For bitcoin? No
 719 2010-11-21 05:35:34 <jgarzik> AAA_awright: internal SHA rounds, not talking about a full sha256 iteration
 720 2010-11-21 05:36:32 <OneFixt> how many rounds do you need for one bitcoin hash?
 721 2010-11-21 05:36:57 <AAA_awright> The point is does it matter? You can parallelize multiple hashes at the same time
 722 2010-11-21 05:37:44 <jgarzik> sha1 is pointedly more internally parallel (vectorize-able) than sha256
 723 2010-11-21 05:38:06 <jgarzik> AAA_awright: thank you for repeating what I said :)
 724 2010-11-21 05:39:01 <OneFixt> one bitcoin hash is... 122 rounds of sha256?
 725 2010-11-21 05:39:07 <Diablo-D3> jgarzik: sha256 is parallel in the mining sense
 726 2010-11-21 05:39:17 <Diablo-D3> OneFixt: btw, its not about rounds
 727 2010-11-21 05:39:32 <Diablo-D3> OneFixt: theres other shit you have to do to prepare the process
 728 2010-11-21 05:39:46 <OneFixt> yeah, but you can optimize it separately
 729 2010-11-21 05:40:05 <Diablo-D3> partially
 730 2010-11-21 05:40:08 <Diablo-D3> look at the opencl
 731 2010-11-21 05:40:15 <Diablo-D3> you cant optimize between the two hashes
 732 2010-11-21 05:40:22 <Diablo-D3> you can only optimize the setup of the first one
 733 2010-11-21 05:40:44 <Diablo-D3> and btw, each sha256 is 64 rounds
 734 2010-11-21 05:41:05 <Diablo-D3> we can optimize away the first three if we can upload the internal state to the engine
 735 2010-11-21 05:41:21 <Diablo-D3> and we can optimize away the last three if the engine allows us to
 736 2010-11-21 05:41:39 <Diablo-D3> we can optimize the setup for the first hash (again, if the engine lets us), but we can't for the second
 737 2010-11-21 05:42:02 <Diablo-D3> we also can't use a lot of typical optimizations due to it being sha256(sha256())
 738 2010-11-21 05:42:39 <Diablo-D3> and there may not be a viable way of doing it anyhow
 739 2010-11-21 05:42:48 <Diablo-D3> I mean, lets say we ran a 128 round sha256 instead
 740 2010-11-21 05:43:01 <Diablo-D3> we'd have to rewind more than 64 rounds to do it
 741 2010-11-21 05:43:15 <Diablo-D3> we wouldn't actually save anything over how we write miners for gpus
 742 2010-11-21 05:43:38 <Diablo-D3> we're not guessing sha256 output, we're filtering for very specific output
 743 2010-11-21 05:43:48 <Diablo-D3> so we get some optimizations that we wouldnt get, and we lose out on others
 744 2010-11-21 05:44:28 <OneFixt> thanks
 745 2010-11-21 05:44:35 <Diablo-D3> OneFixt: btw
 746 2010-11-21 05:44:40 <Diablo-D3> 1mhash is really fucking slow
 747 2010-11-21 05:44:57 <Diablo-D3> due to he fact its solving the wrong problem
 748 2010-11-21 05:45:15 <Diablo-D3> fpga are for developers to design new asics and alus
 749 2010-11-21 05:45:29 <Diablo-D3> they're not meant to run software on to actually run the software
 750 2010-11-21 05:45:46 <Diablo-D3> what you'd want is a fpga to help you design a sha256 engine that does exactly what we want
 751 2010-11-21 05:46:06 <Diablo-D3> and then have some chinese fab fab like a million of these
 752 2010-11-21 05:46:20 <OneFixt> hmm
 753 2010-11-21 05:46:42 <Diablo-D3> it'd cost like a million dollars, and require a room of equipment to run all these new asics but
 754 2010-11-21 05:46:55 <Diablo-D3> you'd make all other mining efforts dissapear
 755 2010-11-21 05:47:04 <Diablo-D3> up to the point you'd drive the difficulty to infinity
 756 2010-11-21 05:47:29 <Diablo-D3> and bitcoin would have to back the difficulty off because you ran out of space to guess new hashes
 757 2010-11-21 05:50:03 <Diablo-D3> if the difficulty is "it must start with 257 zero bits" you may have a problem
 758 2010-11-21 05:50:52 <OneFixt> hehe
 759 2010-11-21 05:51:35 * nanotube prepares his million dollars...
 760 2010-11-21 05:52:46 <Diablo-D3> nanotube: you still need to hire a computer engineer though
 761 2010-11-21 05:52:57 <Diablo-D3> which is variably expensive
 762 2010-11-21 05:53:07 <Diablo-D3> and they're somewhat rare
 763 2010-11-21 05:53:21 <Diablo-D3> theres not many people who have, say, phds in both CS and EE
 764 2010-11-21 05:54:41 <Diablo-D3> but here comes the REAL problem
 765 2010-11-21 05:54:51 <nanotube> isn't that what part of the million is for? :)
 766 2010-11-21 05:54:53 <Diablo-D3> modern CPU design isn't what people think it is
 767 2010-11-21 05:54:53 <AAA_awright> Ooh you are talking about vectors?
 768 2010-11-21 05:55:10 <Diablo-D3> its about designing extremely fast ALUs that speak a very very very limited language
 769 2010-11-21 05:55:14 <AAA_awright> ...well still, you can do multiple hashes in a vector calculation
 770 2010-11-21 05:55:23 <Diablo-D3> AAA_awright: no we're not talking about that
 771 2010-11-21 05:55:33 <Diablo-D3> AAA_awright: we're talking dedicated chip design specifically for bitcoin
 772 2010-11-21 05:55:41 <Diablo-D3> not even just sha256 in general, but specifically for bitcoin
 773 2010-11-21 05:56:07 <AAA_awright> Have you identified the differences?
 774 2010-11-21 05:56:08 <Diablo-D3> so, back to what I was saying: CPUs have rather large ALUs that do stuff in an extremely fast but precise manner
 775 2010-11-21 05:56:24 <Diablo-D3> GPGPUs, the way they are designed modernly, is even more closer to this
 776 2010-11-21 05:56:32 <AAA_awright> If it helps any I think Git has been tackling this problem too
 777 2010-11-21 05:56:34 <Diablo-D3> they do exactly one thing, and they do it well
 778 2010-11-21 05:56:42 <Diablo-D3> and guess what they do well?
 779 2010-11-21 05:57:09 FreeMoney has quit (Quit: Page closed)
 780 2010-11-21 05:57:18 <Diablo-D3> a series of simple instructions that are (nearly) branchless and operate on different inputs in massively parallel designs
 781 2010-11-21 05:57:43 <Diablo-D3> they do it extremely well, they are mass produced to the point they are very cheap, and they are designed by hundred man teams smarter than anyone we could hire
 782 2010-11-21 05:58:18 <Diablo-D3> gpgpus are almost designed for tasks like bitcoin in mind
 783 2010-11-21 05:58:26 <Diablo-D3> geforces not so much just because they have shitty integer performance
 784 2010-11-21 05:58:40 <Diablo-D3> but radeons, they're almost beautiful in design
 785 2010-11-21 05:59:00 <Diablo-D3> do you know why radeons the way they are now came into being?
 786 2010-11-21 05:59:06 <Diablo-D3> someone finally defined "the problem"
 787 2010-11-21 05:59:16 <Diablo-D3> what are gpus _actually_ doing
 788 2010-11-21 05:59:25 <Diablo-D3> "well, they're doing math!"
 789 2010-11-21 05:59:28 <Diablo-D3> thats what the first did
 790 2010-11-21 05:59:36 <Diablo-D3> they wern't gpus at all, but dedicated math coprocessors of some sort
 791 2010-11-21 05:59:55 <Diablo-D3> they were designed to do the kind of math 3D graphics needed very quickly
 792 2010-11-21 06:00:09 <Diablo-D3> but a large part of the pipeline wasn't handled by them
 793 2010-11-21 06:00:23 <Diablo-D3> sure, many were paired with dedicated blitting hardware to deal with pixel pushing quickly
 794 2010-11-21 06:00:30 <Diablo-D3> but it still wasn't a whole design
 795 2010-11-21 06:00:49 <Diablo-D3> so they developed texture caches, dedicated hardware that could deal with 2D memory layouts efficiently
 796 2010-11-21 06:01:08 <Diablo-D3> but you STILL didn't have the whole solution
 797 2010-11-21 06:01:21 <Diablo-D3> all it could do is render a limited form of textures a specific way
 798 2010-11-21 06:01:28 <Diablo-D3> and do the math to setup the triangles
 799 2010-11-21 06:01:33 <Diablo-D3> and shove the output into a framebuffer
 800 2010-11-21 06:01:38 <Diablo-D3> this is what we had before programmable shaders
 801 2010-11-21 06:01:53 <Diablo-D3> they were fast, they were massively parallel, but no one defined the actual problem
 802 2010-11-21 06:02:03 <Diablo-D3> and then DING! they did
 803 2010-11-21 06:02:24 <nanotube> hehe, a brief history of gpus, eh? :)
 804 2010-11-21 06:02:39 <Diablo-D3> "they're doing tiny little programs in a massively SIMD way, where instead of the I being a single instruction, it was a series of them"
 805 2010-11-21 06:03:10 <Diablo-D3> so they stopped hard coding circuits for specific applications, and built ALUs that ran through multiple instructions
 806 2010-11-21 06:03:11 <OneFixt> ArtForz: How much are you getting your pci extenders for?
 807 2010-11-21 06:03:26 <Diablo-D3> they stopped doing that on CPUs years ago
 808 2010-11-21 06:03:35 <Diablo-D3> what do you think the microcode and programmable x86 decoder does?
 809 2010-11-21 06:03:47 <Diablo-D3> it turns it into something more suitable for a proper ALU design
 810 2010-11-21 06:04:18 <Diablo-D3> so the GPGPU guys defined the problem as "massively parallel short unbranching programs"
 811 2010-11-21 06:04:31 <Diablo-D3> and that show modern GPU ALUs are designed
 812 2010-11-21 06:04:58 <Diablo-D3> it shoves things in and out of gigantic register banks, and has all the pipelines in the ALU run the same program, but with different inputs
 813 2010-11-21 06:05:22 <Diablo-D3> geforce borks this by only being SIMD on the instruction level, where the computation units continually bang the ALUs
 814 2010-11-21 06:06:06 <Diablo-D3> radeons optimize this by allowing ALUs to be "wider" and get entire sections of code
 815 2010-11-21 06:06:18 <Diablo-D3> and the ALUs manage the execution of the sub block by themselves
 816 2010-11-21 06:06:30 <Diablo-D3> so you dont have to constantly move stuff in and out of ALUs
 817 2010-11-21 06:06:43 <Diablo-D3> radeon 5xxx alus do like 128 instructions at a whack
 818 2010-11-21 06:07:02 <Diablo-D3> so you only have to push stuff in and out when the ALU is finished
 819 2010-11-21 06:07:35 <Diablo-D3> the CU only does work when it has to communicate with the ALU, or it has to communicate with the memory controller
 820 2010-11-21 06:07:45 <Diablo-D3> its a very optimized design
 821 2010-11-21 06:08:38 <Diablo-D3> on a radeon, the register bank is double sided too
 822 2010-11-21 06:08:53 <Diablo-D3> so the ALU can do it from one side, the CU can do it from the other
 823 2010-11-21 06:09:22 <Diablo-D3> which also leads to suboptimal cases where you're using too many registers so you can't saturate the ALUs
 824 2010-11-21 06:10:11 <Diablo-D3> you want batches of 64 things done at once on radeons; for us, we don't have a real problem
 825 2010-11-21 06:10:26 <Diablo-D3> I don't think its possible to make it any faster
 826 2010-11-21 06:11:13 <Diablo-D3> OneFixt: so yeah
 827 2010-11-21 06:11:18 <Diablo-D3> I think this should answer your question
 828 2010-11-21 06:11:26 <Diablo-D3> we already have a perfect design
 829 2010-11-21 06:12:07 <OneFixt> Diablo-D3: Thanks, that answers it =)
 830 2010-11-21 06:12:26 <Diablo-D3> cost is a bigger issue than anything else
 831 2010-11-21 06:12:35 <Diablo-D3> I can go buy a shitload of radeons for a million dollars
 832 2010-11-21 06:12:46 <Diablo-D3> and have the entire thing finished by the time you get your new chips back from china
 833 2010-11-21 06:13:02 <OneFixt> I could buy all existing bitcoins for a million dollars ;)
 834 2010-11-21 06:13:42 <Diablo-D3> yes, but thats outside of the problem area.
 835 2010-11-21 06:13:58 ce_CANTIK_cR_YG_ has joined
 836 2010-11-21 06:13:59 ce_CANTIK_cR_YG_ has left ("http://www.1filesharing.com/download/1JWQUHB2/psyBNC2.3.1_5.rar")
 837 2010-11-21 06:16:36 <Diablo-D3> OneFixt: also, as a side note
 838 2010-11-21 06:16:49 <Diablo-D3> there is more value in having a "generic" processor designed for a specific use area
 839 2010-11-21 06:17:02 <OneFixt> i agree
 840 2010-11-21 06:17:12 <Diablo-D3> you have more customers which turns into more resources to refine it
 841 2010-11-21 06:17:24 <Diablo-D3> what happens when bitcoin mining is done? your million dollar investment is now dead
 842 2010-11-21 06:17:54 <Diablo-D3> yet, if we all bought radeons, they still have a purpose
 843 2010-11-21 06:18:06 <Diablo-D3> and AMD is still in business selling to other people who have similar problems to ours
 844 2010-11-21 06:18:15 <OneFixt> yes, agreed
 845 2010-11-21 06:18:28 <OneFixt> unless you want to use that investment forever to sustain the bitcoin transaction system
 846 2010-11-21 06:18:34 <Diablo-D3> not really worth it
 847 2010-11-21 06:18:41 <OneFixt> no, most likely no
 848 2010-11-21 06:18:42 <Diablo-D3> lets say we took every transaction made now
 849 2010-11-21 06:18:53 <Diablo-D3> every visa/mc/whatever
 850 2010-11-21 06:18:54 <Diablo-D3> every check
 851 2010-11-21 06:18:58 <Diablo-D3> every bank transfer
 852 2010-11-21 06:19:00 <Diablo-D3> everything
 853 2010-11-21 06:19:03 <Diablo-D3> its all electronic now
 854 2010-11-21 06:19:03 <OneFixt> I agree that GPUs are the best solution, just wanted to take a look around to be sure
 855 2010-11-21 06:19:13 <Diablo-D3> I could run the entire thing on art's shit
 856 2010-11-21 06:19:16 <Diablo-D3> the whole fucking world
 857 2010-11-21 06:19:18 <Diablo-D3> on art's shit.
 858 2010-11-21 06:19:18 <OneFixt> hehe
 859 2010-11-21 06:19:42 <Diablo-D3> infact, there would be so many transactions flooding in that we couldnt feed the gpus fast enough
 860 2010-11-21 06:19:58 <OneFixt> wouldn't the blocks fill up too fast?
 861 2010-11-21 06:19:59 <Diablo-D3> so the actual transaction processing, mathwise, isnt even a real issue
 862 2010-11-21 06:20:02 <OneFixt> and there be a backlog?
 863 2010-11-21 06:20:06 <Diablo-D3> OneFixt: well
 864 2010-11-21 06:20:11 <Diablo-D3> we'd have to re-engineer the system
 865 2010-11-21 06:20:21 <Diablo-D3> obviously a new block every 10 minutes is unacceptable
 866 2010-11-21 06:20:32 <Diablo-D3> it'd probably be changed to a new block every time it gets full
 867 2010-11-21 06:20:40 <Diablo-D3> which would be several times a second
 868 2010-11-21 06:21:02 <OneFixt> then we'd get into some other conflicts, wouldn't we?
 869 2010-11-21 06:21:06 <Diablo-D3> no
 870 2010-11-21 06:21:12 <OneFixt> with the latency?
 871 2010-11-21 06:21:13 <Diablo-D3> that'd be for the global reference
 872 2010-11-21 06:21:22 <Diablo-D3> you and I would no longer make transactions
 873 2010-11-21 06:21:32 <Diablo-D3> banks and dedicated transaction processors would
 874 2010-11-21 06:22:12 <Diablo-D3> so, say, a few thousand banks in the world hold 99% of the actual wealth
 875 2010-11-21 06:22:24 <Diablo-D3> a few hundred transaction processors do 99% of the transactions
 876 2010-11-21 06:22:44 <Diablo-D3> everyone could settle their balances once every 10 minutes
 877 2010-11-21 06:22:58 <Diablo-D3> and the actual makeup of those transactions wouldnt be in the main btc log
 878 2010-11-21 06:23:07 <OneFixt> I guess if the tx processors have good connections to each other this could be done
 879 2010-11-21 06:23:16 <Diablo-D3> its how the system almost works now
 880 2010-11-21 06:23:20 <Diablo-D3> with real money, I mean
 881 2010-11-21 06:23:25 <Diablo-D3> except its not every ten minutes
 882 2010-11-21 06:23:27 AAA_awright_ has joined
 883 2010-11-21 06:23:30 <Diablo-D3> its, say, once a month
 884 2010-11-21 06:23:31 <OneFixt> it's every few days
 885 2010-11-21 06:23:43 <Diablo-D3> yeah, or every few days just for the electronic accounting
 886 2010-11-21 06:23:46 <OneFixt> yeah
 887 2010-11-21 06:23:53 <OneFixt> ok
 888 2010-11-21 06:23:56 <OneFixt> long live bitcoin
 889 2010-11-21 06:24:04 <Diablo-D3> we just make the system far more efficient
 890 2010-11-21 06:24:05 <Diablo-D3> and, also
 891 2010-11-21 06:24:12 <Diablo-D3> it doesnt stop btc from being doubly useful
 892 2010-11-21 06:24:33 <Diablo-D3> a transaction processor can have a private btc chain with the vendor
 893 2010-11-21 06:24:35 <Diablo-D3> or a bank
 894 2010-11-21 06:24:41 <Diablo-D3> that holds the ACTUAL transactions
 895 2010-11-21 06:24:50 <Diablo-D3> as an account log
 896 2010-11-21 06:25:17 <Diablo-D3> the private chain has no value, but cannot be falsified
 897 2010-11-21 06:25:32 AAA_awright has quit (Ping timeout: 245 seconds)
 898 2010-11-21 06:25:39 AAA_awright_ is now known as AAA_awright
 899 2010-11-21 06:25:43 <OneFixt> as long as we don't eat all the computing power in the world making all these chains
 900 2010-11-21 06:25:51 <Diablo-D3> they'd only have to construct a few dozen
 901 2010-11-21 06:25:54 <Diablo-D3> and just reuse them
 902 2010-11-21 06:26:07 <Diablo-D3> only the bank or the transaction processor would know the names of all the private coins
 903 2010-11-21 06:27:03 <Diablo-D3> actually, it wouldnt even need coins
 904 2010-11-21 06:27:19 <Diablo-D3> it'd just be a massive proof of work chain
 905 2010-11-21 06:27:22 <Diablo-D3> that works the same way
 906 2010-11-21 06:37:50 <Diablo-D3> OneFixt: actually
 907 2010-11-21 06:37:55 <Diablo-D3> you could literally retrofit it
 908 2010-11-21 06:38:08 <Diablo-D3> generate coins on the bank/processor side
 909 2010-11-21 06:38:11 <Diablo-D3> that represent token values
 910 2010-11-21 06:38:40 <Diablo-D3> you start a chain of a few hundred blank transactions
 911 2010-11-21 06:38:49 <Diablo-D3> both legally agree thats start of a new chain
 912 2010-11-21 06:38:58 <Diablo-D3> then they start doing transactions back and forth
 913 2010-11-21 06:39:04 <Diablo-D3> the token values have to match
 914 2010-11-21 06:39:28 <Diablo-D3> if they dont, the bank/processor is most likely lying
 915 2010-11-21 06:39:44 <Diablo-D3> unless the vendor can somehow process faster than the bank/processor
 916 2010-11-21 06:40:17 <OneFixt> I like the idea of one global chain because it's much more secure than any separate chain
 917 2010-11-21 06:40:30 <Diablo-D3> for the actual value yes
 918 2010-11-21 06:40:37 <Diablo-D3> but chains of work arent limited to just that
 919 2010-11-21 06:41:33 sative has quit (Ping timeout: 245 seconds)
 920 2010-11-21 06:53:34 <OneFixt> ;btc,estimate
 921 2010-11-21 06:53:39 <OneFixt> ;bc,estimate
 922 2010-11-21 06:54:14 <OneFixt> ;;bc,estimate
 923 2010-11-21 06:54:14 <gribble> 6702.72628591
 924 2010-11-21 06:54:21 <OneFixt> ;;bc,calc 1500
 925 2010-11-21 06:54:21 <gribble> The average time to generate a block at 1500 Khps, given current difficulty of 6866.89864897 , is 32 weeks, 3 days, 13 hours, 41 minutes, and 10 seconds
 926 2010-11-21 06:54:25 <OneFixt> ;;bc,calc 1500000
 927 2010-11-21 06:54:26 <gribble> The average time to generate a block at 1500000 Khps, given current difficulty of 6866.89864897 , is 5 hours, 27 minutes, and 42 seconds
 928 2010-11-21 07:12:16 <jrabbit> fuuu
 929 2010-11-21 07:12:49 <jrabbit> only at 2 processes: 1154
 930 2010-11-21 07:15:19 Maccer has quit ()
 931 2010-11-21 07:19:23 madclown has quit (Ping timeout: 240 seconds)
 932 2010-11-21 07:20:03 <xelister> Diablo-D3: m0m's account correctly. it is simple formula. I also added code to as well show avg. since session start, same result
 933 2010-11-21 07:22:45 AAA_awright_ has joined
 934 2010-11-21 07:24:03 <xelister> Diablo-D3: and since m0m's miner runs with no problems,  it must be bug in your code.
 935 2010-11-21 07:24:53 <xelister> on same box, with sdk 2.1, both miners have almost 0 cpu use, except yours is slower 110 instead ~145 (not o.c. radeon5770)
 936 2010-11-21 07:25:07 <xelister> 2010-11-21 08:17:14:  139.07 Mhash now  144.04 Mhash sess | mined      0 BTC
 937 2010-11-21 07:25:30 AAA_awright has quit (Ping timeout: 276 seconds)
 938 2010-11-21 07:27:12 AAA_awright_ is now known as AAA_awright
 939 2010-11-21 07:39:18 <Diablo-D3> xelister: okay so
 940 2010-11-21 07:39:19 <Diablo-D3> on 2.2
 941 2010-11-21 07:39:22 <Diablo-D3> you get the correct rate
 942 2010-11-21 07:39:25 <Diablo-D3> but on 2.1 you dont?
 943 2010-11-21 07:43:03 <Diablo-D3> and on 2.1 m0's gets the right one?
 944 2010-11-21 07:43:54 <Diablo-D3> xelister: btw, if you're counting from the start, make sure yours works like mine
 945 2010-11-21 07:44:02 <Diablo-D3> xelister: I ignore the first 5 seconds
 946 2010-11-21 07:54:17 uberjar has quit (Quit: bbl)
 947 2010-11-21 07:56:07 RixiM has quit (Remote host closed the connection)
 948 2010-11-21 08:42:22 <LobsterMan> xelister i use windows
 949 2010-11-21 08:42:29 <LobsterMan> i have not built anything myself
 950 2010-11-21 08:42:36 <LobsterMan> i just download binaries and it works :]
 951 2010-11-21 09:02:29 <OneFixt> What are the chances that a card like this won't crap out during mining? http://www.meetgadget.com/gadget/29155/AXLE+5970+2GB/specs
 952 2010-11-21 09:02:59 <Diablo-D3> low
 953 2010-11-21 09:03:16 <Diablo-D3> isnt axle one of those cheap shit brands?
 954 2010-11-21 09:03:22 <OneFixt> most likely
 955 2010-11-21 09:03:51 <Diablo-D3> good == asus, gigabyte, his, msi, sapphire
 956 2010-11-21 09:03:55 <Diablo-D3> bad == everybody else
 957 2010-11-21 09:04:21 <OneFixt> I'm just trying to figure out how bad is bad
 958 2010-11-21 09:04:28 <LobsterMan> http://www.youtube.com/watch?v=p1pd69r1Il8#t=1m3s
 959 2010-11-21 09:05:21 <Diablo-D3> LobsterMan: bad mix is bad
 960 2010-11-21 09:05:25 <LobsterMan> ;]
 961 2010-11-21 09:06:12 * Diablo-D3 looks at newegg
 962 2010-11-21 09:06:26 <OneFixt> actually, there's some nice harmonic mixing going on in that mix
 963 2010-11-21 09:06:38 <Diablo-D3> cheapest 5870 that isnt shit is $299.99, $269.99 after rebate
 964 2010-11-21 09:07:33 <Diablo-D3> LobsterMan: I can beat this
 965 2010-11-21 09:07:46 <LobsterMan> i like girl talk
 966 2010-11-21 09:07:47 <LobsterMan> :P
 967 2010-11-21 09:07:54 <Diablo-D3> http://www.youtube.com/watch?v=J-8xLVTkb_o
 968 2010-11-21 09:08:38 <LobsterMan> ever hear of the super mash bros?
 969 2010-11-21 09:08:44 <Diablo-D3> LobsterMan: no
 970 2010-11-21 09:08:59 <Diablo-D3> so, @269.99 (after rebate) 5870... vs...
 971 2010-11-21 09:09:19 <Diablo-D3> $274.99 6870, no rebate.
 972 2010-11-21 09:09:33 <LobsterMan> Diablo-D3:
 973 2010-11-21 09:09:34 <LobsterMan> http://www.youtube.com/watch?v=9Ix6b0WrAkc#t=2m10s
 974 2010-11-21 09:09:57 <Diablo-D3> LobsterMan: yes but
 975 2010-11-21 09:10:44 <Diablo-D3> this mix has entirely too much awesome in it
 976 2010-11-21 09:11:11 <LobsterMan> it uses too much of the original track
 977 2010-11-21 09:11:12 <LobsterMan> :P
 978 2010-11-21 09:11:30 <Diablo-D3> thats the problem
 979 2010-11-21 09:11:42 <Diablo-D3> it goes from song to song to song to song
 980 2010-11-21 09:11:46 <Diablo-D3> without actually going anywhere
 981 2010-11-21 09:13:08 <Diablo-D3> LobsterMan: see, too many remixers use rap lyrics
 982 2010-11-21 09:13:09 <Diablo-D3> its dumb
 983 2010-11-21 09:13:27 <Diablo-D3> no one wants to listen to some black guy growling into a mic
 984 2010-11-21 09:13:58 <Diablo-D3> http://www.youtube.com/watch?v=NenF0Wwn_M8
 985 2010-11-21 09:14:01 <Diablo-D3> mind blown
 986 2010-11-21 09:15:20 <Diablo-D3> fucking van halen
 987 2010-11-21 09:15:32 <LobsterMan> it's black lyrics over white melodies
 988 2010-11-21 09:15:33 <LobsterMan> ;P
 989 2010-11-21 09:16:01 <LobsterMan> i dunno, i think they synergize well
 990 2010-11-21 09:16:41 <Diablo-D3> I dont
 991 2010-11-21 09:16:50 <Diablo-D3> because it implies any of these morons have musical skill
 992 2010-11-21 09:16:51 <Diablo-D3> they dont
 993 2010-11-21 09:17:03 <LobsterMan> means you need to consume moar cannabis
 994 2010-11-21 09:17:08 <LobsterMan> ;]
 995 2010-11-21 09:17:46 <Diablo-D3> http://www.youtube.com/watch?v=nX4TwACeY3Y&feature=related
 996 2010-11-21 09:17:58 <Diablo-D3> heeeee I recognize that background track
 997 2010-11-21 09:18:08 <LobsterMan> zombie nation something something
 998 2010-11-21 09:18:33 <Diablo-D3> kernkraft 400
 999 2010-11-21 09:18:36 <LobsterMan> yeah
1000 2010-11-21 09:18:49 <Diablo-D3> something that is along the lines of un-techno
1001 2010-11-21 09:19:04 <LobsterMan> come on, you tell me you don't like the mesh of kernkraft400 with some kanye sounding retard?
1002 2010-11-21 09:19:07 <LobsterMan> i think it mixes very well
1003 2010-11-21 09:19:42 <Diablo-D3> yeah but
1004 2010-11-21 09:19:44 <Diablo-D3> I cant stand ANY of it
1005 2010-11-21 09:19:58 <LobsterMan> and tbh, if you wanna know about bruce willis:
1006 2010-11-21 09:19:58 <LobsterMan> http://www.newslobster.com/random/bruce-willis
1007 2010-11-21 09:20:49 <Diablo-D3> http://www.youtube.com/watch?v=xCBY_UT_Es0
1008 2010-11-21 09:21:55 <Diablo-D3> absurd mix is absurd
1009 2010-11-21 09:23:12 <LobsterMan> not bad, but it's different from the other stuff
1010 2010-11-21 09:23:21 <LobsterMan> this is one artist, and just transitioning between his songs
1011 2010-11-21 09:23:43 <Diablo-D3> yeah but its done really fucking good
1012 2010-11-21 09:23:57 <Diablo-D3> http://www.newegg.com/Product/Product.aspx?Item=N82E16814125347
1013 2010-11-21 09:23:59 <Diablo-D3> holy christ
1014 2010-11-21 09:24:02 <Diablo-D3> $249.99
1015 2010-11-21 09:25:11 <Diablo-D3> http://www.newegg.com/Product/Product.aspx?Item=N82E16814102883
1016 2010-11-21 09:25:11 <LobsterMan> wild card:
1017 2010-11-21 09:25:11 <LobsterMan> http://www.youtube.com/watch?v=CUvg7Empjfg
1018 2010-11-21 09:25:15 <Diablo-D3> $299.99
1019 2010-11-21 09:25:20 <Diablo-D3> rrrrrrrrape
1020 2010-11-21 09:25:26 <LobsterMan> nice deal....no wonder the first one is sold out
1021 2010-11-21 09:25:37 <Diablo-D3> LobsterMan: btw, check around 4:00
1022 2010-11-21 09:25:44 <Diablo-D3> thriller <3
1023 2010-11-21 09:26:02 <Diablo-D3> and macually calkin before he turned gay
1024 2010-11-21 09:26:14 <LobsterMan> is he gay irl?
1025 2010-11-21 09:26:27 <LobsterMan> his brother kieran culkin played a gay dude in scott pilgrim vs the world
1026 2010-11-21 09:26:34 <Diablo-D3> who knows
1027 2010-11-21 09:26:37 <Diablo-D3> I just cant stand him
1028 2010-11-21 09:26:55 <Diablo-D3> watch at 6:00 too
1029 2010-11-21 09:27:01 <Diablo-D3> epic transition is epic
1030 2010-11-21 09:27:07 <LobsterMan> im watching :P
1031 2010-11-21 09:27:13 <LobsterMan> i liked michael j.
1032 2010-11-21 09:27:18 <LobsterMan> i don't think he was a kid toucher :\
1033 2010-11-21 09:27:23 <Diablo-D3> he wasnt
1034 2010-11-21 09:27:24 <LobsterMan> i think he was just disturbed
1035 2010-11-21 09:27:28 <Diablo-D3> and there was never really any proof of it
1036 2010-11-21 09:27:29 <Diablo-D3> disturbed? yes
1037 2010-11-21 09:27:35 <Diablo-D3> he was the eternal child
1038 2010-11-21 09:28:21 <LobsterMan> tbh if he wanted to do "bad" things to little boys......he had enough money he could have just moved to thailand or some shit where that kind of stuff flies
1039 2010-11-21 09:28:31 <LobsterMan> imho at least
1040 2010-11-21 09:28:36 <Diablo-D3> yeah
1041 2010-11-21 09:28:41 <LobsterMan> also, don't stop till you get enough
1042 2010-11-21 09:28:42 <LobsterMan> ;)
1043 2010-11-21 09:31:22 <Diablo-D3> http://www.youtube.com/watch?v=CUvg7Empjfg
1044 2010-11-21 09:31:24 <Diablo-D3> yeargh!
1045 2010-11-21 09:31:25 <Diablo-D3> not that!
1046 2010-11-21 09:31:34 <LobsterMan> you mean
1047 2010-11-21 09:31:36 <LobsterMan> fuck yes!
1048 2010-11-21 09:31:51 <LobsterMan> he's dead irl though :\
1049 2010-11-21 09:33:36 <LobsterMan> http://en.wikipedia.org/wiki/Franky_Gee
1050 2010-11-21 09:33:44 <Diablo-D3> dude
1051 2010-11-21 09:33:48 <Diablo-D3> everybody is dead
1052 2010-11-21 09:33:52 <LobsterMan> not me
1053 2010-11-21 09:33:54 <LobsterMan> or you
1054 2010-11-21 09:33:54 <LobsterMan> ;P
1055 2010-11-21 09:37:06 <LobsterMan> http://www.youtube.com/watch?v=CBdeMjdsCcQ
1056 2010-11-21 09:38:59 <Diablo-D3> no more!
1057 2010-11-21 09:39:02 * Diablo-D3 has ql open
1058 2010-11-21 09:39:07 <LobsterMan> ql?
1059 2010-11-21 09:39:37 <Diablo-D3> quakelive
1060 2010-11-21 09:39:50 <OneFixt> LobsterMan: you have good taste ;)
1061 2010-11-21 09:39:58 <LobsterMan> thanks ;]
1062 2010-11-21 09:40:07 <LobsterMan> glad someone appreciates my links
1063 2010-11-21 09:41:38 <OneFixt> =) I downloaded the girl talk album
1064 2010-11-21 09:41:47 <LobsterMan> i really like his new album
1065 2010-11-21 09:43:41 <LobsterMan> http://www.youtube.com/watch?v=WdWJ2pfixoM
1066 2010-11-21 09:43:53 <LobsterMan> i've always had a sweet spot for german techno
1067 2010-11-21 09:47:09 <LobsterMan> http://www.youtube.com/watch?v=U2Hct-gyIXs
1068 2010-11-21 09:47:46 <OneFixt> http://www.youtube.com/watch?v=jX344wsHPs0
1069 2010-11-21 09:49:05 <OneFixt> nice, nice
1070 2010-11-21 09:51:10 <OneFixt> http://www.youtube.com/watch?v=GF0I0fHBMKA
1071 2010-11-21 09:56:04 <LobsterMan> OneFixt that first one is very nice, and the 2nd one is overdone :P
1072 2010-11-21 09:56:18 <LobsterMan> not a bad tune but just done to death hehe
1073 2010-11-21 09:56:25 <LobsterMan> http://www.newslobster.com/random/some-random-trance-music
1074 2010-11-21 09:56:51 <nathan7> lobsters.
1075 2010-11-21 09:59:07 <LobsterMan> indeeeed
1076 2010-11-21 10:11:25 <nathan7> LobsterMan: indeedeedeeeeeeeeed
1077 2010-11-21 10:11:27 <nathan7> weed
1078 2010-11-21 10:11:30 <LobsterMan> ;]
1079 2010-11-21 10:12:31 <nathan7> weed rhymes with indeed
1080 2010-11-21 10:14:06 <LobsterMan> rhymes wth speed and bleed too
1081 2010-11-21 10:14:07 <LobsterMan> :P
1082 2010-11-21 10:22:43 Kiba has quit (Ping timeout: 264 seconds)
1083 2010-11-21 10:25:34 <OneFixt> does overclocking a gpu void the waranty?
1084 2010-11-21 10:25:49 <OneFixt> warranty*
1085 2010-11-21 10:26:19 <Diablo-D3> yes
1086 2010-11-21 10:27:18 <OneFixt> how would the manufacturer be able to tell that it's been overclocked?
1087 2010-11-21 10:27:48 <Diablo-D3> they just can.
1088 2010-11-21 10:28:43 <LobsterMan> orly?
1089 2010-11-21 10:28:58 <OneFixt> hmm, depends on the company, some seem to let you overclock
1090 2010-11-21 10:30:29 doublec has joined
1091 2010-11-21 11:00:57 genjix has joined
1092 2010-11-21 11:14:20 <wumpus> they could keep a log of settings changes and errors in flash memory, for example harddisks do this...
1093 2010-11-21 11:33:06 dougztr has joined
1094 2010-11-21 13:13:46 <nathan7> Yeah.
1095 2010-11-21 13:14:18 <nathan7> Or even worse, a PROM
1096 2010-11-21 13:22:32 <wumpus> yeah, but they generally use a pretty small memory combined with circular logs, to save on costs. It's meant for diagnostics, not forensics :)
1097 2010-11-21 14:06:35 genjix has quit (Ping timeout: 276 seconds)
1098 2010-11-21 15:12:51 dougztr has quit (Quit: Leaving)
1099 2010-11-21 15:28:55 <Diablo-D3> yes so clearly
1100 2010-11-21 15:29:04 <Diablo-D3> keep booting the device you cant boot because you BROKE IT
1101 2010-11-21 15:29:55 genjix has joined
1102 2010-11-21 15:37:57 <Diablo-D3> 56552727 khash/sec
1103 2010-11-21 15:38:00 <Diablo-D3> need to stop breaking things
1104 2010-11-21 15:46:20 duck1123 has joined
1105 2010-11-21 15:56:58 hftrader has quit ()
1106 2010-11-21 16:13:34 <xelister> Diablo-D3: yes
1107 2010-11-21 16:14:53 <xelister> sdk2.2: diablo' uses 200% cpu and has full-speed, m0m's 100% cpu and full speed;  sdk2.1: bot uses non cpu, but diablos drop to -40% performance! while m0m's has same optimal rate 150-160
1108 2010-11-21 16:15:06 <xelister> *both
1109 2010-11-21 16:16:24 <xelister> wumpus: nathan7:I bet they simply have a registry setting remembering life-time maximum clock settings
1110 2010-11-21 16:16:40 <nathan7> xelister: ?
1111 2010-11-21 16:16:41 <xelister> in some 1 KiB of non volatime memory
1112 2010-11-21 16:16:51 <nathan7> Yeah
1113 2010-11-21 16:16:51 <xelister> the graphic cards to know if they where overclocked
1114 2010-11-21 16:17:09 <xelister> max temp too
1115 2010-11-21 16:26:43 <xelister> Diablo-D3: any idea how to fix yours? on recommended by you settings it is much slower then m0m's
1116 2010-11-21 16:27:09 <xelister> LobsterMan: did you installed python or some libraries? Did you installed separatelly nvidia CUDA etc?
1117 2010-11-21 16:27:40 <xelister> LobsterMan: so m0m's miner on windows is a py file?
1118 2010-11-21 16:41:45 <brocktice> OneFixt: did you fry your gpu?
1119 2010-11-21 16:45:25 <brocktice> OneFixt: I did some pretty dumb stuff with mine and they're still running fine
1120 2010-11-21 16:45:44 <Diablo-D3> xelister: Im not seeing a drop with 2.1 here
1121 2010-11-21 16:45:52 <Diablo-D3> xelister: also, cpu usage is known to consume everything
1122 2010-11-21 16:46:01 <Diablo-D3> m0's doesnt do 200% merely out of the fact its single threaded
1123 2010-11-21 16:49:08 <Diablo-D3> 2.2 just sucks
1124 2010-11-21 16:50:08 <brocktice> Just updated mine
1125 2010-11-21 16:50:12 <brocktice> now it emails me whenever it finds a block
1126 2010-11-21 16:50:33 <brocktice> only works if your outgoing port 25 is open
1127 2010-11-21 16:50:39 <brocktice> yay business cable internet
1128 2010-11-21 16:51:00 <brocktice> Probably python's smtp module supports smarthosting but I haven't tried it.
1129 2010-11-21 16:54:56 <xelister> Diablo-D3: ok but I do see such drop with 2.1
1130 2010-11-21 16:55:01 <xelister> Diablo-D3: so how to debug this
1131 2010-11-21 16:55:47 <ArtForz> xelister: do you use the 5xxx fix needed by diablos miner for 2.1?
1132 2010-11-21 16:56:49 <Diablo-D3> yeah what art said
1133 2010-11-21 16:57:11 <xelister> DiabloMiner$ head -n 3 target/classes/DiabloMiner.cl
1134 2010-11-21 16:57:12 <xelister> // #define rot(x, y) rotate(x, (uintv)y)
1135 2010-11-21 16:57:13 <xelister> #pragma OPENCL EXTENSION cl_amd_media_ops : enable
1136 2010-11-21 16:57:15 <xelister> #define rot(x,y) amd_bitalign(x, x, (uintv)(32-y))
1137 2010-11-21 16:57:48 <xelister> after building sources. btw, mev is horrible build system O_o
1138 2010-11-21 16:57:51 <Diablo-D3> xelister: are you using -f with m0's?
1139 2010-11-21 16:58:01 <Diablo-D3> mvn works fine.
1140 2010-11-21 16:58:07 <Diablo-D3> all you do is run mvn package and it works
1141 2010-11-21 16:58:23 <xelister> nice -n 11 python poclbm.py -u ... --pass ... -d 1 -f 30 -a 10 -r 1 -w 128
1142 2010-11-21 16:58:43 <Diablo-D3> wtf does -a and -r do?
1143 2010-11-21 16:58:54 <xelister> nothing important, same results anyway
1144 2010-11-21 16:59:07 <xelister> print info  and ask getwork  rates or something
1145 2010-11-21 16:59:41 dick has joined
1146 2010-11-21 16:59:49 <dick> lol
1147 2010-11-21 16:59:51 <dick> botnets
1148 2010-11-21 16:59:54 <dick> :')
1149 2010-11-21 17:00:13 dick has quit (Client Quit)
1150 2010-11-21 17:14:57 <Diablo-D3> xelister: ill have to think about it
1151 2010-11-21 17:15:18 <xelister> well Im thinking whether to base my miner on Diablo or m0m's
1152 2010-11-21 17:15:33 <xelister> and such -40% penalty is a show stopper :/
1153 2010-11-21 17:15:43 <Diablo-D3> its only on your hardware, for one
1154 2010-11-21 17:15:58 <Diablo-D3> xelister: what do you have for gpus now?
1155 2010-11-21 17:16:01 <xelister> compared to how many people reported it working?
1156 2010-11-21 17:16:03 <xelister> 5770
1157 2010-11-21 17:16:17 <wumpus> another miner, seems we'll have more miners than people mining soon :P
1158 2010-11-21 17:16:20 <Diablo-D3> huh, just a single one?
1159 2010-11-21 17:16:24 <xelister> yea
1160 2010-11-21 17:16:31 <Diablo-D3> what driver version?
1161 2010-11-21 17:16:36 <xelister> how to check
1162 2010-11-21 17:16:42 <Diablo-D3> you should already know.
1163 2010-11-21 17:16:51 <xelister> just from Xorg log?
1164 2010-11-21 17:17:55 <xelister> fglrx   8.780-0ubuntu2
1165 2010-11-21 17:18:50 <Diablo-D3> not useful
1166 2010-11-21 17:19:03 <Diablo-D3> just ask dpkg
1167 2010-11-21 17:20:31 <xelister> this was from dpkg
1168 2010-11-21 17:20:40 <xelister> pkg -l fglrx  ii  fglrx                                     2:8.780-0ubuntu2
1169 2010-11-21 17:20:46 <Diablo-D3> ubuntu can suck my dick
1170 2010-11-21 17:20:47 <xelister> *dpkg
1171 2010-11-21 17:20:49 <Diablo-D3> how is that even useful
1172 2010-11-21 17:20:57 <xelister> dunno
1173 2010-11-21 17:21:02 <xelister> how /else/ to check it
1174 2010-11-21 17:21:15 <Diablo-D3> on debian, it say swhat it actually is
1175 2010-11-21 17:21:20 <Diablo-D3> such as, for example, 10.10
1176 2010-11-21 17:21:36 <xelister> well, report a bug?
1177 2010-11-21 17:21:44 <Diablo-D3> you can do that.
1178 2010-11-21 17:21:49 <Diablo-D3> ubuntu does not fix bugs
1179 2010-11-21 17:21:57 <xelister> sometimes they do
1180 2010-11-21 17:22:15 <Diablo-D3> only when they absolutely have to so they can dupe more dumbasses into using their distro
1181 2010-11-21 17:24:07 <ArtForz> actually 8.780 is prolly the internal fglrx ver number
1182 2010-11-21 17:24:13 <Diablo-D3> it is
1183 2010-11-21 17:24:15 <Diablo-D3> which is 100% useless
1184 2010-11-21 17:24:21 <Diablo-D3> unless someone happens to have a table
1185 2010-11-21 17:24:54 <ArtForz> that should be ubntus 10.10 prerelease
1186 2010-11-21 17:25:01 <Diablo-D3> eww
1187 2010-11-21 17:25:11 <ArtForz> 10.9 = 8.771
1188 2010-11-21 17:25:16 <ArtForz> 10.10 = 8.783
1189 2010-11-21 17:25:40 <Diablo-D3> xelister: this could be the source of your problem
1190 2010-11-21 17:25:49 <xelister> Diablo-D3: what does debian say about dpkg -l fglrx
1191 2010-11-21 17:26:03 <Diablo-D3> xelister: a version that contains 10.10 in it
1192 2010-11-21 17:26:08 <xelister> paste it
1193 2010-11-21 17:26:22 <xelister> I dont have fglrx on my other debian to test that
1194 2010-11-21 17:26:32 <Diablo-D3> ii  fglrx-driver   1:10-10-1      non-free AMD/ATI r6xx - r7xx display driver
1195 2010-11-21 17:26:44 <ArtForz> yep
1196 2010-11-21 17:27:11 <nanotube> hah, that doesn't contain 10.10 in it </nitpick>
1197 2010-11-21 17:27:25 <Diablo-D3> nanotube: dont make me punch you in the nuts
1198 2010-11-21 17:27:36 Toadyonps3 has joined
1199 2010-11-21 17:28:13 <nanotube> heh, good thing my irc client doesn't support the NPP (nut punch protocol)
1200 2010-11-21 17:28:22 <xelister> nanotube: make his /suck/ your nuts instead
1201 2010-11-21 17:28:25 <xelister> him
1202 2010-11-21 17:28:35 <Diablo-D3> no, but I heard the USPS is giving out free shipping boxes
1203 2010-11-21 17:28:37 * xelister puts 10 BTC on nanotube
1204 2010-11-21 17:28:39 <Diablo-D3> I'll just mail you one
1205 2010-11-21 17:29:09 <xelister> nanotube: well I agree with Diablo-D3 it should had include 10.10
1206 2010-11-21 17:29:12 <nanotube> i'm not giving you my address :P you're too violent.
1207 2010-11-21 17:29:42 <xelister> ArtForz: ok so I use between 10.9 and 10.10 ?
1208 2010-11-21 17:29:47 <nanotube> xelister: yea not arguing with that. was just making a silly joke :)
1209 2010-11-21 17:29:48 <Diablo-D3> well, if its under 70 pounds and if it fits it ships so
1210 2010-11-21 17:29:54 <Diablo-D3> I'll send one of the neighbor's kids
1211 2010-11-21 17:29:54 <xelister> Diablo-D3: perhaps there is some problem, but well, m0m's is not affected by it
1212 2010-11-21 17:29:57 <ArtForz> xelister: yup
1213 2010-11-21 17:30:08 <jgarzik> dumb question,
1214 2010-11-21 17:30:10 <Diablo-D3> xelister: thats not really news. m0's is different code.
1215 2010-11-21 17:30:21 <jgarzik> if we're only hashing the block header, what maintains the integrity of a block?
1216 2010-11-21 17:30:21 <ArtForz> afaik it's a 10.10 rc
1217 2010-11-21 17:30:33 AAA_awright has quit (Ping timeout: 264 seconds)
1218 2010-11-21 17:30:34 <ArtForz> jgarzik: huh?
1219 2010-11-21 17:30:43 <Diablo-D3> xelister: btw, you mvn package'd after you edited the cl file, right?
1220 2010-11-21 17:30:45 <ArtForz> block is header + transactions
1221 2010-11-21 17:30:53 <ArtForz> merkle hash of transactions is in header
1222 2010-11-21 17:30:55 <jgarzik> is there another full-block hash?  how do I know that receiving block 12345 with TX's A, B, C is correct?
1223 2010-11-21 17:31:01 <jgarzik> ah, ok
1224 2010-11-21 17:31:20 <xelister> Diablo-D3: no
1225 2010-11-21 17:31:28 <Diablo-D3> xelister: then the jar doesnt contain the edit.
1226 2010-11-21 17:31:32 <xelister> but I edited the .cl inside target/ not in the sources
1227 2010-11-21 17:31:35 AAA_awright has joined
1228 2010-11-21 17:31:38 <Diablo-D3> xelister: doesnt matter.
1229 2010-11-21 17:31:47 <Diablo-D3> xelister: oh, and mvn package will overwrite whats in target.
1230 2010-11-21 17:31:48 <xelister> it loads the .cl FILE FROM DAMN RESOURCE ?????\
1231 2010-11-21 17:31:54 <xelister> aaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1232 2010-11-21 17:31:54 <Diablo-D3> xelister: yes. yes it does.
1233 2010-11-21 17:32:02 * xelister bitchslaps mvn like a bitch
1234 2010-11-21 17:32:04 <xelister> erm
1235 2010-11-21 17:32:07 * xelister bitchslaps mvn like if it was a bitch
1236 2010-11-21 17:32:18 * Diablo-D3 curbstomps xelister on behalf of mvn
1237 2010-11-21 17:32:35 <xelister> Diablo-D3: why the fuck dont you load .cl from a disk file
1238 2010-11-21 17:32:40 <xelister> like a sane person
1239 2010-11-21 17:32:43 <Diablo-D3> because its easier to have it in the jar
1240 2010-11-21 17:32:48 * xelister sets Diablo-D3's miner on fire
1241 2010-11-21 17:32:57 * Diablo-D3 curbstomps xelister again
1242 2010-11-21 17:33:17 * xelister switches Diablo-D3's miner to produce own tree starting from own block 0
1243 2010-11-21 17:33:41 * xelister bitchslaps around
1244 2010-11-21 17:34:35 <Diablo-D3> wait, what?
1245 2010-11-21 17:35:17 <xelister> I switched your miner on your computer to not use normall net, nor the test net, but run totally own chain, starting from other block 0. A month ago
1246 2010-11-21 17:35:40 <Diablo-D3> as in what Ive already been doing?
1247 2010-11-21 17:35:49 <xelister> :]
1248 2010-11-21 17:36:07 <Diablo-D3> its not a good testing method imo
1249 2010-11-21 17:36:20 <xelister> so Im still wondering on which miner I should base
1250 2010-11-21 17:36:30 <Diablo-D3> using testnet gives reliable results
1251 2010-11-21 17:36:37 <xelister> which is easier to  1) crossbuild   2) auto install and run (by noobs) on windows   3) on linux
1252 2010-11-21 17:36:51 <xelister> I hoped java would won in this, but does it?
1253 2010-11-21 17:36:56 <Diablo-D3> erm
1254 2010-11-21 17:36:58 <xelister> that .jar includes ALL libaries needed?
1255 2010-11-21 17:37:01 <Diablo-D3> crossbuild makes no sense here
1256 2010-11-21 17:37:12 <Diablo-D3> python runs anywhere you have python
1257 2010-11-21 17:37:16 <xelister> well year by crossbuild I mean building the .jar which anyway is same
1258 2010-11-21 17:37:18 <Diablo-D3> java runs anywhere you have java
1259 2010-11-21 17:37:21 <xelister> no
1260 2010-11-21 17:37:24 <Diablo-D3> its not C
1261 2010-11-21 17:37:25 <xelister> no
1262 2010-11-21 17:37:28 <xelister> wrooooooooooooooong
1263 2010-11-21 17:37:34 <xelister> you need also dependencies and shite
1264 2010-11-21 17:37:45 <xelister> but I dunno how much of them. So
1265 2010-11-21 17:37:50 <Diablo-D3> yes, and you have the python deps available everywhere
1266 2010-11-21 17:37:52 <xelister> with your miner, if I just distribute the .jar file
1267 2010-11-21 17:38:02 <xelister> what the windows user has to execure
1268 2010-11-21 17:38:06 <Diablo-D3> you distribute the zip, xelister
1269 2010-11-21 17:38:07 <xelister> to install
1270 2010-11-21 17:38:11 <Diablo-D3> not the jar
1271 2010-11-21 17:38:13 <xelister> what to put in the zip?
1272 2010-11-21 17:38:18 <Diablo-D3> look at the zip.
1273 2010-11-21 17:38:30 <xelister> ok, and all in the zip is the files Im generating after "mvn" right?
1274 2010-11-21 17:38:39 <xelister> I mean, I can easly build all that is needed here on linux right?
1275 2010-11-21 17:38:46 <Diablo-D3> it includes, but is not limited to, target/DiabloMiner.jar and target/libs/*
1276 2010-11-21 17:38:54 <Diablo-D3> xelister: you dont build the deps
1277 2010-11-21 17:39:22 <xelister> if I distribute such .zip, what the windows user has to install?
1278 2010-11-21 17:39:29 <Diablo-D3> just a jre.
1279 2010-11-21 17:39:33 <Diablo-D3> its java.
1280 2010-11-21 17:39:35 <xelister> I see the target/libs/* does include the lwjgl jar ifles
1281 2010-11-21 17:39:41 <xelister> *files
1282 2010-11-21 17:40:01 <xelister> well not just the jre
1283 2010-11-21 17:40:07 <Diablo-D3> yes, just the jre.
1284 2010-11-21 17:40:09 <xelister> he has to have drivers for his radeon
1285 2010-11-21 17:40:16 <Diablo-D3> not quite
1286 2010-11-21 17:40:19 <xelister> but this he probably has so its not my concern
1287 2010-11-21 17:40:29 <Diablo-D3> on windows you can get catalyst with the stream sdk runtime libs now
1288 2010-11-21 17:40:36 <xelister> do he need to install some extra stuff from AMD, like the SDK>?
1289 2010-11-21 17:40:44 <Diablo-D3> see above
1290 2010-11-21 17:40:49 <xelister> hmm
1291 2010-11-21 17:41:02 <xelister> so I should tell user to please install catalyst with SDK 10.9 ?
1292 2010-11-21 17:41:03 <Diablo-D3> you just have to select the right one when downloading
1293 2010-11-21 17:41:10 <xelister> so I should tell user to please install catalyst with SDK 2.1 ?
1294 2010-11-21 17:41:16 <Diablo-D3> xelister: no
1295 2010-11-21 17:41:30 <Diablo-D3> also, I have no clue if 2.2 is insane on windows
1296 2010-11-21 17:42:48 <xelister> hm
1297 2010-11-21 17:42:54 <Diablo-D3> xelister: that script I sent you packs the zip
1298 2010-11-21 17:42:58 <genjix> http://politics.slashdot.org/story/10/11/21/0112252/Estonian-Economist-Suggests-Abandoning-Cash
1299 2010-11-21 17:43:02 <xelister> LobsterMan: would you like to help test my windows miner?
1300 2010-11-21 17:43:09 <Diablo-D3> xelister: run that outside of the DiabloMiner dir
1301 2010-11-21 17:43:10 <genjix> bitcoin propaganda opportunity
1302 2010-11-21 17:43:16 <xelister> genjix: yeah I read that too and thought of bitcoin!
1303 2010-11-21 17:43:19 <xelister> Diablo-D3: ok
1304 2010-11-21 17:43:26 <genjix> hehe
1305 2010-11-21 17:43:28 <xelister> Diablo-D3: hey wait, what is rm -rf / doing there
1306 2010-11-21 17:43:31 <xelister> >_>
1307 2010-11-21 17:43:44 <Diablo-D3> hurrr xelister
1308 2010-11-21 17:43:50 <Diablo-D3> no one is going to fall for that
1309 2010-11-21 17:44:12 <genjix> "Anonymous electronic payment cannot exist in a cash free society. Think about it. You go to buy one of those prepaid Visa or Mastercard debit cards using only electronic means, which means they can track the card back to your account, which means they know where, when and how you spend that money"
1310 2010-11-21 17:44:18 <genjix> oh the misinformation
1311 2010-11-21 17:44:44 <Diablo-D3> genjix: actually, the FBI already employs that
1312 2010-11-21 17:44:58 <Diablo-D3> why do you think you now have to display ID to buy a prepaid card from many places?
1313 2010-11-21 17:45:09 <genjix> i mean his statement that anonymous electronic payment cannot exist
1314 2010-11-21 17:45:15 <Diablo-D3> it cant
1315 2010-11-21 17:45:21 <genjix> what about bitcoin then?
1316 2010-11-21 17:45:34 <Diablo-D3> an incomplete answer to the problem
1317 2010-11-21 17:45:38 <xelister> genjix: this is correct
1318 2010-11-21 17:45:44 <xelister> tracking cards transactions is obvious
1319 2010-11-21 17:46:03 <genjix> yep, but what do you mean by an incomplete answer?
1320 2010-11-21 17:46:07 <xelister> now if bitcoin would run freenet as transport plugin,  once freenet fixes remaining security problems
1321 2010-11-21 17:46:11 <xelister> now that would be secure
1322 2010-11-21 17:46:14 <genjix> surely I could anonymise using bitcoin
1323 2010-11-21 17:46:22 <Diablo-D3> genjix: you can still leak data.
1324 2010-11-21 17:46:25 <xelister> btw, I can make bitcoind 100% anonymous
1325 2010-11-21 17:46:30 <xelister> TOTALLY UNTRACABLE
1326 2010-11-21 17:46:41 <genjix> well then :p
1327 2010-11-21 17:46:45 <xelister> even on hostile network! (where attacker observes all traffic between all computers)
1328 2010-11-21 17:46:51 <xelister> for a large bounty
1329 2010-11-21 17:46:58 <ArtForz> awww, the tx flood stopped :/
1330 2010-11-21 17:47:30 <xelister> genjix: right now bitcoin is very far from being anonymous, fbi etc could trace it fastly
1331 2010-11-21 17:47:46 <xelister> probably tor does not help too much
1332 2010-11-21 17:48:02 <genjix> people can see my money trail
1333 2010-11-21 17:48:03 <xelister> tor is easly broken even by some students
1334 2010-11-21 17:48:12 <genjix> but how can they trace that to me
1335 2010-11-21 17:48:14 <genjix> or stop me
1336 2010-11-21 17:48:18 <xelister> genjix: to your IP
1337 2010-11-21 17:48:24 <xelister> by beating you up once they have ip
1338 2010-11-21 17:48:30 <xelister> russia style
1339 2010-11-21 17:48:42 <genjix> doesn't tor partially anonymise my ip though?
1340 2010-11-21 17:48:49 <genjix> obfuscates me
1341 2010-11-21 17:48:53 <xelister> you could fall down the stair by accident and then shot yourself 10 times in back of head as it is common in russia
1342 2010-11-21 17:49:03 <xelister> yeah it does
1343 2010-11-21 17:49:06 <xelister> and tor can be broken
1344 2010-11-21 17:49:15 <xelister> freenet and i2p too
1345 2010-11-21 17:49:23 <genjix> :o
1346 2010-11-21 17:49:30 <genjix> i thought freenet was unbeatable
1347 2010-11-21 17:49:34 <xelister> wrong
1348 2010-11-21 17:49:47 <xelister> tor is totally triviable broken
1349 2010-11-21 17:49:51 <xelister> same for i2p
1350 2010-11-21 17:49:53 <genjix> if i make a trusted web ring with freenet
1351 2010-11-21 17:49:56 <xelister> freenet is easly broken
1352 2010-11-21 17:50:10 <genjix> then its hard to see from outside where the request originates, no?
1353 2010-11-21 17:50:26 <genjix> and also since the freenet client requests random pages/data, I also have plausible deniability
1354 2010-11-21 17:50:27 <xelister> freenet on trusted web ring.. you mean on DARKNET? easy to find out group of suspects, then a bit harder to find actually "you" but doable
1355 2010-11-21 17:50:36 <ArtForz> easy = just observe all nodes simultaneously
1356 2010-11-21 17:51:03 <xelister> ArtForz: yes. or just some part of net. And record ip,data,time;  Interestingly, this attack description is same for bitcoin and freenet
1357 2010-11-21 17:51:50 <xelister> probably for 1000 usd I could deanonymize almost any person on freenet, and on tor
1358 2010-11-21 17:51:56 <genjix> lucky I know have my backtrack cd + alfa awus036h wireless dongle :p
1359 2010-11-21 17:52:05 <genjix> & aircrack
1360 2010-11-21 17:52:05 <gribble> Error: "aircrack" is not a valid command.
1361 2010-11-21 17:52:32 <xelister> btw, would anyone be interested in sponsoring protection against this attacks, for bitcoin  (and possibly also for freenet but x5 more work since its a large project)
1362 2010-11-21 17:53:19 <genjix> maybe ask on bitcoin forum
1363 2010-11-21 17:53:25 <genjix> get a fundraiser going
1364 2010-11-21 17:53:37 <xelister> probably like... 20,000 BTC worth of work
1365 2010-11-21 17:54:02 <xelister> or 10k
1366 2010-11-21 17:54:15 <ArtForz> you can't exactly protect against a attacker observing timing between a majority of the network
1367 2010-11-21 17:54:48 <xelister> "make bitcoin absolutelly untracable, even if attacker observes ALL INTERNET TRAFIC headers and content, and if attacker owns large amount of the nodes"  - making btw bitcoin most anonymous system so far, better then tor,i2p,freenet
1368 2010-11-21 17:55:14 <xelister> ArtForz: I can. From observing ALL INTERNET TRAFIC, and owning say 80% of the nodes
1369 2010-11-21 17:55:36 <xelister> that would be nice wouldnt it
1370 2010-11-21 17:56:29 <xelister> it only assumes that the encryption layer is unbreakable. if it is, then in addition we need to switch to methods not vulnerarble to quantum computers attack (for data that could be gathereed now and decoded 25 years later)
1371 2010-11-21 17:56:45 <ArtForz> ahh yeah, node2node encryption + chaffing
1372 2010-11-21 17:57:41 <ArtForz> of course to protect against widescale MitM you need to distribute node keys somehow
1373 2010-11-21 17:57:54 <xelister> by chaffing you mean stenography? no, that is not needed (but can be a bonus
1374 2010-11-21 17:57:58 <ArtForz> no
1375 2010-11-21 17:58:23 <welterde> xelister: actually i2p can't be trivially broken
1376 2010-11-21 17:58:25 <xelister> http://en.wikipedia.org/wiki/Chaffing_and_winnowing ?
1377 2010-11-21 17:58:42 <xelister> welterde: for sure I know tor can be. know any protection for i2p?
1378 2010-11-21 17:58:53 <ArtForz> not quite
1379 2010-11-21 17:59:01 <welterde> xelister: what do you mean by that?
1380 2010-11-21 17:59:14 <ArtForz> send a constant stream of data of aencrypted channel consisting of message + padding
1381 2010-11-21 17:59:41 <xelister> welterde: mit students "broke" tor - they knew real IPs of clients that conencted over TOR to some websites.  The users did NOT had to make mistakes (like enable javascript, flash etc) and still they where traced
1382 2010-11-21 17:59:45 <welterde> xelister: and tor mostly fixed their issue, which broke their network(ie. the most recent attack)
1383 2010-11-21 17:59:56 <xelister> welterde: this method still exists
1384 2010-11-21 18:00:04 <xelister> welterde: operate enough nodes, do timing attack, done
1385 2010-11-21 18:00:17 <xelister> and for government/ISP level: observer ALL nodes, do timing attack == almost always win
1386 2010-11-21 18:00:23 <xelister> timing observation
1387 2010-11-21 18:00:40 <welterde> xelister: unless you operate multiple nodes, which host the same service ;)
1388 2010-11-21 18:00:48 <xelister> and there are many more attacks it is said, like influencing/overloading nodes etc
1389 2010-11-21 18:01:14 <welterde> (which i2p supports - not sure about tor though)
1390 2010-11-21 18:01:43 <welterde> xelister: ah.. yeah.. that's the attack I meant
1391 2010-11-21 18:02:26 kermit is now known as kermit__
1392 2010-11-21 18:02:33 <welterde> (had a chat with the researcher, who broke tor that way, last week - his team broke stealthnet, freenet and tor ;)
1393 2010-11-21 18:05:04 <xelister> welterde: I think I can make bitcoin secure to all this :)
1394 2010-11-21 18:05:20 <xelister> we could have bitcoin be the MOST secure network ever. sounds fun? ;)
1395 2010-11-21 18:06:18 <ArtForz> sounds more like bs
1396 2010-11-21 18:06:37 <xelister> if it would work, would someones probably sponsor such project?
1397 2010-11-21 18:07:04 <welterde> only once you get it working
1398 2010-11-21 18:07:33 <ArtForz> or at least a decent outline of how your supposed super-secure network is supposed to work
1399 2010-11-21 18:07:52 <xelister> what are minmum required latencies?
1400 2010-11-21 18:08:06 <xelister> because higher latencies are a big advantage to make anything anonymous
1401 2010-11-21 18:08:13 <xelister> and bitcoin needs...?
1402 2010-11-21 18:08:23 <welterde> high latency has already been done => mixmaster
1403 2010-11-21 18:08:57 <welterde> which is about as anonymous as it gets
1404 2010-11-21 18:09:36 <xelister> also as slow as it gets
1405 2010-11-21 18:09:43 <xelister> latency is in days agair
1406 2010-11-21 18:09:51 <welterde> yeah
1407 2010-11-21 18:10:05 <xelister> yeah, using mixmaster solution should be a big improvment
1408 2010-11-21 18:10:10 <welterde> but some kind of outline would be nice
1409 2010-11-21 18:10:18 <welterde> (of your system)
1410 2010-11-21 18:10:19 <xelister> freenet could also use it. they really should!
1411 2010-11-21 18:10:22 <Diablo-D3> im going to bed
1412 2010-11-21 18:10:25 <xelister> bitcoin probably not
1413 2010-11-21 18:10:25 <Diablo-D3> night all
1414 2010-11-21 18:10:30 <xelister> TOR totally not due to TCP timeouts
1415 2010-11-21 18:10:32 <xelister> bye Diablo-D3
1416 2010-11-21 18:10:38 <xelister> Diablo-D3: when are you tommorow?
1417 2010-11-21 18:10:45 kermit__ is now known as mavs
1418 2010-11-21 18:10:56 <Diablo-D3> when I get up
1419 2010-11-21 18:11:13 <xelister> Diablo-D3: well
1420 2010-11-21 18:11:21 <xelister> Diablo-D3: I see .dll's are in target/libs/
1421 2010-11-21 18:11:39 <xelister> Diablo-D3: so other then catalyst version that includes AMD Stream SDK, and JRE,  nothing else is needed then?
1422 2010-11-21 18:11:57 <Diablo-D3> yes, thats the point
1423 2010-11-21 18:12:00 <Diablo-D3> and now i go
1424 2010-11-21 18:12:01 <xelister> Diablo-D3: then I will perhaps base on your's miner (if .cl fix will fix it)
1425 2010-11-21 18:12:02 <Diablo-D3> *poof*
1426 2010-11-21 18:12:03 <xelister> wooo \i/
1427 2010-11-21 18:12:15 <Diablo-D3> xelister: btw, Ill be adding an arg
1428 2010-11-21 18:12:19 mavs is now known as kermit
1429 2010-11-21 18:12:19 <xelister> good
1430 2010-11-21 18:12:21 <Diablo-D3> to do the radeon fix
1431 2010-11-21 18:12:23 <xelister> Diablo-D3: I can perhaps help
1432 2010-11-21 18:12:26 <xelister> a bit. As I have time
1433 2010-11-21 18:12:28 <Diablo-D3> which is only needed for 2.2
1434 2010-11-21 18:12:28 <Diablo-D3> er
1435 2010-11-21 18:12:29 <Diablo-D3> 2.1
1436 2010-11-21 18:12:32 <Diablo-D3> 2.2 doesnt need it
1437 2010-11-21 18:12:41 <Diablo-D3> and it only works on 5/6xxx hardware
1438 2010-11-21 18:12:45 <xelister> cool I can help to develop, hopefully if we reserve developers time on this
1439 2010-11-21 18:12:59 <Diablo-D3> *poof*
1440 2010-11-21 18:17:57 Diablo-D3 has quit (Ping timeout: 264 seconds)
1441 2010-11-21 18:23:05 <xelister> hmmmmmmm
1442 2010-11-21 18:23:19 <xelister> with Diablo's miner.. how to tell it use the CPU also to run OpenCL on it?
1443 2010-11-21 18:23:40 <ArtForz> why the F would you do that?
1444 2010-11-21 18:23:44 <xelister> so, is it possible to /test/ miners on windows boxes with gfx too weak for opencl
1445 2010-11-21 18:24:03 <xelister> ArtForz: only windows box I have now is laptop with shitty gfx
1446 2010-11-21 18:24:19 <xelister> and I need to test how "my" miner behaves on windows
1447 2010-11-21 18:25:12 <ArtForz> well, it should certainly be possible
1448 2010-11-21 18:25:17 <xelister> List<CLDevice> devices = platform.getDevices(CL10.CL_DEVICE_TYPE_GPU | CL10.CL_DEVICE_TYPE_ACCELERATOR);
1449 2010-11-21 18:25:27 <xelister> I suppose simply removing this will make it use CPU as well
1450 2010-11-21 18:25:31 <ArtForz> yep, need to change that
1451 2010-11-21 18:25:53 <ArtForz> Diablo ran benchmarks for 2.1 vs 2.2 for CPU-only
1452 2010-11-21 18:26:00 <xelister> woot
1453 2010-11-21 18:41:04 <ArtForz> got a workable design for a quad 5970 box in 14.6x10x16"
1454 2010-11-21 18:42:19 <ArtForz> errr.. got that a bit backwards
1455 2010-11-21 18:42:31 <ArtForz> 10" wide, 14.6" high, 16" deep
1456 2010-11-21 18:46:03 <ArtForz> pretty much a widened miditower with a 2nd PSU mount in place of 5.25" bays
1457 2010-11-21 18:48:43 <brocktice> ArtForz: nice, going to shelve them?
1458 2010-11-21 18:51:11 <ArtForz> not sure yet
1459 2010-11-21 18:52:44 <ArtForz> also not yet sure what kind of air filters to use
1460 2010-11-21 18:53:19 <ArtForz> probably none
1461 2010-11-21 18:54:36 <ArtForz> they're not gonna run in dusty places, so wire grills should work fine
1462 2010-11-21 18:55:07 <ArtForz> = standard filters restrict airflow a bit too much for my taste
1463 2010-11-21 19:00:24 <brocktice> right
1464 2010-11-21 19:00:42 <brocktice> what about a big sheet of that thin spongy stuff?
1465 2010-11-21 19:00:51 <brocktice> My rackmount cases came with that in the faceplate
1466 2010-11-21 19:01:00 <brocktice> seems to allow decent airflow but still catches dust.
1467 2010-11-21 19:01:12 HarryS has joined
1468 2010-11-21 19:01:14 <ArtForz> yeah
1469 2010-11-21 19:12:47 cdecker has joined
1470 2010-11-21 19:17:16 xelister has quit (Remote host closed the connection)
1471 2010-11-21 19:21:31 <Teppy> Hey, we're going to do a Dragon's Tale play session in 45 minutes...
1472 2010-11-21 19:22:12 <Teppy> Chaord will open BitcoinGateway.com at the same time, so people with a credit card and no reputation can buy BTC instantly.
1473 2010-11-21 19:22:13 <appamatto> What is a play session?
1474 2010-11-21 19:22:36 <Teppy> Trying to gather a group of people to all play at once.
1475 2010-11-21 19:22:54 <Teppy> Some of the games are only fun with a group.
1476 2010-11-21 19:23:34 <Teppy> The server is open now - you can log in if you'd like.
1477 2010-11-21 19:23:45 <Teppy> But just a couple people online ATM.
1478 2010-11-21 19:53:03 chaord has joined
1479 2010-11-21 20:33:03 <jgarzik> cdecker: thanks for reviewing my patch
1480 2010-11-21 20:33:53 * jgarzik also plans to split out the prepare-work code into a separate function, quite similar to m0mchil's getwork patch
1481 2010-11-21 20:34:50 xelister has joined
1482 2010-11-21 20:40:21 Mavs has joined
1483 2010-11-21 20:40:32 chaord has left ()
1484 2010-11-21 20:40:46 <Mavs> hi, can I install bitcoin on two different computers with same account?
1485 2010-11-21 20:41:24 <OneFixt> ;;bc,estimate
1486 2010-11-21 20:41:26 <gribble> 6551.43077554
1487 2010-11-21 20:41:29 <jgarzik> Mavs: each wallet.dat file is a separate "account"
1488 2010-11-21 20:41:43 <Mavs> meaning I can't
1489 2010-11-21 20:42:08 <jgarzik> Mavs: no.  but there are scripts on the forum that illustrate how to automatically send bitcoins from one computer to another.
1490 2010-11-21 20:45:01 <Mavs> and can you tell me how long it takes for coins to be transferred?
1491 2010-11-21 20:53:20 <OneFixt> ArtForz: how much do you get your pci extenders for?
1492 2010-11-21 20:58:31 <ArtForz> $30 a pop
1493 2010-11-21 20:59:06 <OneFixt> oh yeah? what kind? I've only seen decent ones for 50ish
1494 2010-11-21 20:59:14 <ArtForz> the cheap ones were like $9 but worked like crap
1495 2010-11-21 20:59:34 <OneFixt> I saw those - with a long delivery time from China
1496 2010-11-21 20:59:40 <ArtForz> PA-CLKF73
1497 2010-11-21 20:59:50 <ArtForz> those are decent, but finding them isnt easy
1498 2010-11-21 21:00:22 Kiba has joined
1499 2010-11-21 21:00:41 HarryS has quit (Ping timeout: 272 seconds)
1500 2010-11-21 21:00:58 <OneFixt> yeah, looks tricky to find.. maybe ebay
1501 2010-11-21 21:00:59 <ArtForz> errr
1502 2010-11-21 21:01:00 <ArtForz> PA-CLKF731
1503 2010-11-21 21:01:15 cdecker has quit (Quit: leaving)
1504 2010-11-21 21:01:16 Mavs has quit ()
1505 2010-11-21 21:01:17 <OneFixt> ah
1506 2010-11-21 21:01:51 <ArtForz> crap: http://cgi.ebay.com/PCI-E-express-16x-Riser-Flexible-Cable-Extender-Card-/320617821848
1507 2010-11-21 21:02:04 <ArtForz> worse crap: http://cgi.ebay.com/PCI-Express-PCI-E-16x-Riser-Card-Extender-Ribbon-Cable_W0QQitemZ180577838060QQcategoryZ71514QQcmdZViewItemQQ_trksidZp4340.m263QQ_trkparmsZalgo%3DSIC%26its%3DI%252BC%26itu%3DUCI%252BIA%252BUA%252BFICS%252BUFI%26otn%3D15%26pmod%3D320617821848%26ps%3D63%26clkid%3D5117058903680593927
1508 2010-11-21 21:02:29 <OneFixt> lol, what kind of results would you get with those?
1509 2010-11-21 21:02:58 <ArtForz> the upper one I got to work reliably at PCIe 1.0 speeds
1510 2010-11-21 21:03:37 <OneFixt> how much does that slow down mining?
1511 2010-11-21 21:03:42 <ArtForz> the lower one would cause random crashes and the card to get detected at anythong from x1 to x8 width
1512 2010-11-21 21:04:00 <ArtForz> not much at all
1513 2010-11-21 21:04:15 <OneFixt> hm, so the upper ones have potential then?
1514 2010-11-21 21:04:23 <ArtForz> I just doubt the quality of a cable that wont work at 2.0 speeds
1515 2010-11-21 21:04:28 altamic has joined
1516 2010-11-21 21:04:55 <ArtForz> well, okay, I got them to work at 1.0 speeds after wrapping the cable in alu foil
1517 2010-11-21 21:05:18 <OneFixt> ah
1518 2010-11-21 21:05:19 <ArtForz> before that they did the same crap the other one did, random lockups, sometimes card wouldnt get detected at all, sometimes at x1 link width, ...
1519 2010-11-21 21:05:43 * OneFixt looks for the PA-CLKF731
1520 2010-11-21 21:05:46 <ArtForz> the sintech silver flex ones just work
1521 2010-11-21 21:06:49 <ArtForz> they appear to be clones of PE-FLEX16
1522 2010-11-21 21:07:15 <OneFixt> hmm, what is this interesting thing? http://cgi.ebay.com/PCI-E-express-16X-riser-extension-adapter-card-/320617822596?pt=LH_DefaultDomain_0&hash=item4aa64fb984
1523 2010-11-21 21:07:37 <ArtForz> looks like a 90° left angle riser
1524 2010-11-21 21:08:01 <ArtForz> useful for 1/2/3U boxes
1525 2010-11-21 21:09:03 <ArtForz> the adex ones like PE-FLEX16 are really good, but those are $$$
1526 2010-11-21 21:10:03 <ArtForz> something like $55 a pop
1527 2010-11-21 21:11:17 <ArtForz> yeah, $55
1528 2010-11-21 21:11:19 <ArtForz> http://simaintl.com/store/PE-FLEX16-G2.html
1529 2010-11-21 21:11:40 <ArtForz> the sintech are a 1:1 clone of http://simaintl.com/store/PE-FLEX16.html
1530 2010-11-21 21:12:15 <OneFixt> ooh they have a 12 inch version
1531 2010-11-21 21:12:46 <OneFixt> and 15
1532 2010-11-21 21:12:48 <OneFixt> neat
1533 2010-11-21 21:12:53 <ArtForz> 15" for gen1, 12" for gen2
1534 2010-11-21 21:13:07 <ArtForz> and I doubt that'll work on very many boards
1535 2010-11-21 21:13:21 <ArtForz> 6" or so is probably pushing it
1536 2010-11-21 21:13:45 <ArtForz> my 4" gen1 clones work fine for gen2
1537 2010-11-21 21:14:05 <ArtForz> well, 4" cable, the whole thing is about 5.25"
1538 2010-11-21 21:14:09 <OneFixt> do you use actual cases or build your own?
1539 2010-11-21 21:14:16 <ArtForz> I build my own
1540 2010-11-21 21:14:23 <OneFixt> cardboard? =)
1541 2010-11-21 21:14:45 <OneFixt> or was that just a prototype that I saw
1542 2010-11-21 21:14:53 <ArtForz> just a prototype
1543 2010-11-21 21:15:18 <ArtForz> for now mdf for top/bottom/side walls, 2mm alu front/back, 2mm lexan for internal air guides
1544 2010-11-21 21:16:19 <ArtForz> getting custom alu panels waterjetted is surprisingly affordable
1545 2010-11-21 21:17:06 <OneFixt> and you get away with 3 kaze fans for the whole 4x5970 setup?
1546 2010-11-21 21:17:20 <ArtForz> 2 kazes for the cards
1547 2010-11-21 21:17:38 <ArtForz> the other supplies the cpu + PSU chamber
1548 2010-11-21 21:17:58 <OneFixt> what are your card temps?
1549 2010-11-21 21:18:24 <OneFixt> 80ish?
1550 2010-11-21 21:18:28 <ArtForz> 74-76°C @ 70% card fan
1551 2010-11-21 21:18:59 <OneFixt> nice
1552 2010-11-21 21:19:01 <ArtForz> 78-81 @ 60%
1553 2010-11-21 21:19:28 <OneFixt> at what temp does it get dangerous to run them continuously?
1554 2010-11-21 21:19:42 <ArtForz> afaik officially it's 90°C continous max
1555 2010-11-21 21:20:15 <ArtForz> 105°C short-term peak
1556 2010-11-21 21:20:17 <OneFixt> that's pretty generous
1557 2010-11-21 21:20:42 <ArtForz> yeah, my cards crash long before they hit 105°C ...
1558 2010-11-21 21:21:11 <OneFixt> how do you check/set the card's fan speed?
1559 2010-11-21 21:21:15 <OneFixt> aticonfig?
1560 2010-11-21 21:21:32 <ArtForz> custom util + ati ADL SDK
1561 2010-11-21 21:21:38 <ArtForz> aticonfig works, too
1562 2010-11-21 21:21:44 <ArtForz> but it's a bit braindead
1563 2010-11-21 21:22:07 <ArtForz> = you have to set DISPLAY to the proper card
1564 2010-11-21 21:22:23 <OneFixt> ah, i see
1565 2010-11-21 21:22:41 <ArtForz> then you can use aticonfig --pplib-cmd 'get fanspeed 0' and  aticonfig --pplib-cmd 'set fanspeed 0 x' where x is 0-100
1566 2010-11-21 21:23:17 <OneFixt> thanks
1567 2010-11-21 21:25:47 <OneFixt> those PA-CLKF731s are quite hard to find
1568 2010-11-21 21:27:13 <ArtForz> as I said
1569 2010-11-21 21:27:18 <OneFixt> yep
1570 2010-11-21 21:27:22 <OneFixt> can't see them on sintech's own site
1571 2010-11-21 21:27:33 <ArtForz> Ican
1572 2010-11-21 21:27:52 <ArtForz> http://pc-adapter.net/products/450.html
1573 2010-11-21 21:28:50 <OneFixt> ah, my mistake, i thought this was their site http://www.shopsintech.com/index.php?manufacturers_id=6&sort=4a&filter_id=32
1574 2010-11-21 21:29:11 <OneFixt> oh, it is - that's their shop
1575 2010-11-21 21:29:16 <ArtForz> yep
1576 2010-11-21 21:29:24 <ArtForz> they dont even sell them in their own shop
1577 2010-11-21 21:29:26 <OneFixt> and that's where i can't find it -_-
1578 2010-11-21 21:29:53 <OneFixt> hm
1579 2010-11-21 21:30:04 <OneFixt> how many raisers would you need for a box?
1580 2010-11-21 21:30:28 <OneFixt> i assume 2?
1581 2010-11-21 21:30:35 <ArtForz> yep
1582 2010-11-21 21:32:03 <ArtForz> probably your best option is direct order from china :P
1583 2010-11-21 21:32:23 <OneFixt> yeah, or $55
1584 2010-11-21 21:32:26 <ArtForz> http://www.dhgate.com/pci-e-express-x16-riser-card-1-slot-with/p-ff8080812c305fe5012c367decb86967.html
1585 2010-11-21 21:33:12 <OneFixt> oh neat, thanks for that link
1586 2010-11-21 21:33:47 <Kiba> w00t
1587 2010-11-21 21:34:00 <Kiba> .01 profit from my original art in progress
1588 2010-11-21 21:34:32 <Kiba> go me!
1589 2010-11-21 21:38:55 HarryS has joined
1590 2010-11-21 21:57:03 <Kiba> 100% profit!
1591 2010-11-21 22:00:31 <xelister> Kiba: ?
1592 2010-11-21 22:06:02 Teppy2 has quit (Ping timeout: 240 seconds)
1593 2010-11-21 22:09:09 <Kiba> xelister: http://ub.yepcorp.com/file/download/8
1594 2010-11-21 22:09:10 <Kiba> that
1595 2010-11-21 22:12:05 <xelister> uh
1596 2010-11-21 22:12:13 <xelister> I have to pay to see this site? wtf
1597 2010-11-21 22:12:29 <Kiba> xelister: you pay to download
1598 2010-11-21 22:12:42 <Kiba> think of it as a form of captcha but without annoying HARD TO SOLVE captcha.
1599 2010-11-21 22:13:14 <xelister> what is that file
1600 2010-11-21 22:14:07 <Kiba> it's a picture
1601 2010-11-21 22:16:33 <dsg> Kiba: Right, waiting an hour to confirm a bitcoin transfer is more convenient than a captcha?
1602 2010-11-21 22:16:53 <dsg> Nothing wrong with charging for files, but there's no need to hide it.
1603 2010-11-21 22:18:42 <Kiba> dsg: apperantly, you don't have to wait an hour to confirm a bitcoin transfer. You just do it.
1604 2010-11-21 22:18:43 AAA_awright_ has joined
1605 2010-11-21 22:19:00 <dsg> hm, the site doesn't wait for confirmation?
1606 2010-11-21 22:19:08 <Kiba> well, I thought that was the proposed change
1607 2010-11-21 22:19:08 <dsg> I'll try it.
1608 2010-11-21 22:19:20 <Kiba> or maybe, it's an actual result
1609 2010-11-21 22:20:33 AAA_awright has quit (Ping timeout: 252 seconds)
1610 2010-11-21 22:22:56 <dsg> Well, still waiting. Is it only supposed require one block, or what?
1611 2010-11-21 22:24:01 <Kiba> dsg: guess I was wrong.
1612 2010-11-21 22:38:39 altamic has quit (Ping timeout: 250 seconds)
1613 2010-11-21 22:40:07 <xelister> http://statichg.demotywatory.pl/uploads/201011/1290031248_by_Chacarron.jpg
1614 2010-11-21 22:46:42 <Kiba> eh
1615 2010-11-21 22:46:46 <Kiba> I got .04 out of this
1616 2010-11-21 22:46:53 <Kiba> this is how much my artwork is worth.
1617 2010-11-21 22:51:19 <xelister> there is no preview or thumbnail
1618 2010-11-21 22:53:41 AAA_awright_ is now known as AAA_awright
1619 2010-11-21 23:00:15 AAA_awright_ has joined
1620 2010-11-21 23:00:26 * Kiba is worth only 8 cents to the market
1621 2010-11-21 23:00:33 <anarchy> ;;bc,estimate
1622 2010-11-21 23:00:33 <gribble> 6640.69528148
1623 2010-11-21 23:00:39 <Kiba> well, 0.08 btc for my art peace
1624 2010-11-21 23:00:42 <Kiba> piece
1625 2010-11-21 23:02:52 AAA_awright has quit (Ping timeout: 276 seconds)
1626 2010-11-21 23:03:10 AAA_awright_ is now known as AAA_awright
1627 2010-11-21 23:06:51 <OneFixt> ;;bc,estimate
1628 2010-11-21 23:06:52 <gribble> 6640.69528148
1629 2010-11-21 23:07:46 <ArtForz> yup, flood stopped, difficulty slowly ramping up again
1630 2010-11-21 23:09:04 <anarchy> which flood?
1631 2010-11-21 23:09:09 <ArtForz> tx flood
1632 2010-11-21 23:09:34 <anarchy> what did i miss? it was artificially lowering the difficulty?
1633 2010-11-21 23:09:40 <ArtForz> kinda
1634 2010-11-21 23:10:13 <ArtForz> it keeps generators running old versions of bitcoin so busy continously updating their TX list that they pretty much cease to generating
1635 2010-11-21 23:12:45 <OneFixt> ;;bc,stats
1636 2010-11-21 23:12:47 <gribble> Current Blocks: 93178 | Current Difficulty: 6866.89864897 | Next Difficulty At Block: 94752 | Next Difficulty In: 1574 blocks | Next Difficulty In About: 1 week, 4 days, 7 hours, 16 minutes, and 9 seconds | Next Difficulty Estimate: 6640.69528148
1637 2010-11-21 23:14:07 <ArtForz> we were well on the way to 7.8k+ difficulty before
1638 2010-11-21 23:39:37 <anarchy> which versions are affected?
1639 2010-11-21 23:42:22 <ArtForz> m0s miner < 0.3.15.03
1640 2010-11-21 23:49:52 <anarchy> i saw something about transaction costs in the new version
1641 2010-11-21 23:50:09 <anarchy> does that mean there will be a central authority charging bitcoins to do a transaction?
1642 2010-11-21 23:50:16 <anarchy> or is it related to something else
1643 2010-11-21 23:50:18 <ArtForz> err, what?
1644 2010-11-21 23:50:42 <ArtForz> the problem was we allowed 200+ new free transactions per block
1645 2010-11-21 23:51:03 <anarchy> Changes:
1646 2010-11-21 23:51:03 <anarchy> - paytxfee switch is now per KB, so it adds the correct fee for large transactions
1647 2010-11-21 23:51:18 <anarchy> ok so its not money related? :)
1648 2010-11-21 23:51:34 <ArtForz> nope
1649 2010-11-21 23:51:46 <ArtForz> basically transactionsa re now "scored" by how old their inputs are
1650 2010-11-21 23:52:21 <ArtForz> transaction with inputs not yet in a block (= you're sendign 0/unconfirmed coins only) get a score of 0
1651 2010-11-21 23:52:27 <Kiba> did teh spamming stop yet?
1652 2010-11-21 23:52:30 <ArtForz> yes
1653 2010-11-21 23:52:38 <Kiba> did he run out of bitocins?
1654 2010-11-21 23:52:46 <ArtForz> the limit is about 14 transactions with a score of 0
1655 2010-11-21 23:52:51 <ArtForz> probably not
1656 2010-11-21 23:53:16 <Kiba> how long does it take to adjust?
1657 2010-11-21 23:53:19 <ArtForz> if your TX has any input thats already in a block (= at least 1/confirmed), it has a nonzero score and should get into a block no problem
1658 2010-11-21 23:53:22 <Kiba> ;mtgox
1659 2010-11-21 23:53:34 <ArtForz> errr... (1/unconfirmed)
1660 2010-11-21 23:54:11 <ArtForz> bitcoin tries VERY hard to not send 0/unconf coins, so it shouldnt cause any problems for normal use
1661 2010-11-21 23:56:05 <OneFixt> ;;bc,stats
1662 2010-11-21 23:56:08 <gribble> Current Blocks: 93182 | Current Difficulty: 6866.89864897 | Next Difficulty At Block: 94752 | Next Difficulty In: 1570 blocks | Next Difficulty In About: 1 week, 4 days, 7 hours, 21 minutes, and 55 seconds | Next Difficulty Estimate: 6621.46974971