1 2011-02-04 00:00:02 Cusipzzz has joined
   2 2011-02-04 00:00:21 <ArtForz> and nonces/run is again a poisson distribution
   3 2011-02-04 00:00:47 <OneFixt> well, let's start by saying that the chance to get a valid hash is 1 / 12 / 22012
   4 2011-02-04 00:01:03 <ArtForz> ?
   5 2011-02-04 00:01:26 <OneFixt> diff times harder than a diff 1 hash
   6 2011-02-04 00:01:47 <ArtForz> I am talking about basic H==0 = diff 1 here
   7 2011-02-04 00:02:12 <jgarzik> if one is running a private pool, it makes more sense for them all to work on full-target (quit when hash < target) rather then reduced-target (quit when high 32 bits == 0), right?
   8 2011-02-04 00:02:22 <ArtForz> probability for *one* H==0 in one kernel run is about 7.51%
   9 2011-02-04 00:02:50 <OneFixt> so i'm saying in the kernel run when you solve a block, you still have a 7.51% chance of getting an h==0 in that same run
  10 2011-02-04 00:03:00 <ArtForz> probabilty for *two or more* in one run is... 0.315%
  11 2011-02-04 00:03:00 <OneFixt> and you toss a coin to see which one overwrites the other
  12 2011-02-04 00:03:15 <ArtForz> 0.315% != 7.5%
  13 2011-02-04 00:03:19 <ArtForz> by quite a margin
  14 2011-02-04 00:03:54 <OneFixt> think about it
  15 2011-02-04 00:04:07 <OneFixt> we're comparing two different things
  16 2011-02-04 00:04:14 <ArtForz> I thought about this months ago
  17 2011-02-04 00:04:36 <OneFixt> your mining is different from pooled mining so this doesn't matter to your setup
  18 2011-02-04 00:05:18 <OneFixt> just walk through this with me, alright?
  19 2011-02-04 00:07:08 <ArtForz> I only check H==0 in kernel, too
  20 2011-02-04 00:08:02 <OneFixt> this is more like flipping a coin
  21 2011-02-04 00:08:05 <ArtForz> because the speed loss from doing more checking outweighs the lost solutions from 2 or more H=0 nonces in one run
  22 2011-02-04 00:08:27 <ArtForz> yes, a 2**32 sided coin, 350M times
  23 2011-02-04 00:08:52 <ArtForz> now count how often you get "0" twice in 350M flips
  24 2011-02-04 00:09:04 <ArtForz> *twice or more
  25 2011-02-04 00:09:08 <OneFixt> instead, count how often you get a valid hash
  26 2011-02-04 00:09:32 <OneFixt> winning*
  27 2011-02-04 00:10:15 <ArtForz> again, what does that have to do with % lost from H==0 collisions?
  28 2011-02-04 00:11:18 <OneFixt> i'm trying to say that on the kernel run in which your h==0 happens to be a very rare winning block, that hash can be found on any of 350,000 nonces
  29 2011-02-04 00:11:29 <ArtForz> doesnt matter
  30 2011-02-04 00:11:43 <OneFixt> finding that one does not actually reduce the probability of finding another one
  31 2011-02-04 00:11:51 <OneFixt> yes, on average, it does
  32 2011-02-04 00:11:56 <OneFixt> but not in this discrete case
  33 2011-02-04 00:12:06 <ArtForz> any h==0 has the same chance of being a winning block
  34 2011-02-04 00:12:27 <Xunie> Hmm, I'm making a p2p network myself, and I was wonder, what would be a great interval to reconnect? I know bitcoin has a great interval system based on the last connection and current time.
  35 2011-02-04 00:12:34 <Xunie> Does anyone know that algorithm?
  36 2011-02-04 00:12:54 <OneFixt> yes but!
  37 2011-02-04 00:13:09 <OneFixt> any overwrite has a 22012 times higher chance of destroying a winning block than being one
  38 2011-02-04 00:13:14 <OneFixt> monty hall problem
  39 2011-02-04 00:13:16 <ArtForz> errr.. no
  40 2011-02-04 00:13:56 <ArtForz> the overwritten H==0 has a chance of 1 in difficulty of being a winning block, the overwriting H==0... also has a chance of 1 in difficulty of being a winning block
  41 2011-02-04 00:14:13 <ArtForz> all you did is lose one "try"
  42 2011-02-04 00:15:49 <OneFixt> =) let me write a small simulation and i'll get back to you
  43 2011-02-04 00:16:11 <OneFixt> gotta go eat first so my brain works better
  44 2011-02-04 00:17:07 <luke-jr> Xunie: use the source!
  45 2011-02-04 00:20:28 <Xunie> "Use the (fo/sou)rce, luke!)
  46 2011-02-04 00:20:38 <Xunie> s/luke!)/luke!/;
  47 2011-02-04 00:24:23 riush has joined
  48 2011-02-04 00:32:25 davex__ has quit (Read error: Connection reset by peer)
  49 2011-02-04 00:36:34 <echelon> someone highlighted me? :/
  50 2011-02-04 00:36:52 <echelon> oh nvm :)
  51 2011-02-04 00:37:00 <echelon> thought it went passed the buffer
  52 2011-02-04 00:45:34 jav has joined
  53 2011-02-04 00:46:25 altamic has quit (Quit: altamic)
  54 2011-02-04 00:46:34 <draginx> Whats the best way to pipeline transactions? So as they come through I can just store those values into a local DB of mine
  55 2011-02-04 00:46:59 <sipa> monitorreceived patch?
  56 2011-02-04 00:47:51 <echelon> that's still a patch?
  57 2011-02-04 00:47:52 <jgarzik> draginx: grab artforz's python client
  58 2011-02-04 00:48:02 <jgarzik> draginx: and watch for tx msgs
  59 2011-02-04 00:48:19 <sipa> echelon: a branch in gavin's git actually
  60 2011-02-04 00:49:14 noagendamarket has joined
  61 2011-02-04 00:50:15 kisom_dev has quit (Ping timeout: 250 seconds)
  62 2011-02-04 00:52:24 <draginx> jgarzik: u mean this? http://json-rpc.org/wiki/python-json-rpc
  63 2011-02-04 00:53:04 <draginx> hmm i wonder if this ruby client has it https://github.com/requnix/bitcoin >.> I'm indiana jones damnit! I like rubies and I hate snakes >< :P
  64 2011-02-04 00:53:25 <jgarzik> draginx: no, http://pastebin.com/ZSM7iHZw
  65 2011-02-04 00:53:58 <draginx> o kso its asynch which is good
  66 2011-02-04 00:54:26 <draginx> just trying to see how to listen to all incoming tx calls O_o but i do like this concept
  67 2011-02-04 00:54:32 <draginx> its just looking at the network stuff right?
  68 2011-02-04 00:56:01 <doublec> I run bitcoind with -noirc and an explicit -connect. But the connection count shows >1. I thought -connect meant only connect to that node?
  69 2011-02-04 00:56:41 genjix has quit (Ping timeout: 240 seconds)
  70 2011-02-04 00:56:57 <ArtForz> yes
  71 2011-02-04 00:57:11 <ArtForz> you still send addr()s though and other nodes will try to connect to you
  72 2011-02-04 00:57:16 <doublec> ah, ok
  73 2011-02-04 00:57:23 kisom_dev has joined
  74 2011-02-04 00:57:40 <doublec> this is on my phone. I've switched providers. I guess the previous one didn't allow inbound connections but this one does.
  75 2011-02-04 00:57:48 kisom_dev is now known as Guest67589
  76 2011-02-04 00:58:02 <draginx> switched to verizon with ur iphone?@
  77 2011-02-04 00:58:06 <prax> any of you guys sharp with JSON?
  78 2011-02-04 00:58:27 <doublec> draginx: no I'm in New Zealand, no verizon here. I use an Nokia N900 to run bitcoind.
  79 2011-02-04 00:58:41 <draginx> aye
  80 2011-02-04 00:58:43 <draginx> prax: I am :)
  81 2011-02-04 00:59:03 <draginx> its not that hard O_o think of it as an array… (cause thats what it is lol :P)
  82 2011-02-04 00:59:28 <prax> oh okay
  83 2011-02-04 00:59:29 <draginx> http://www.jsonlint.com/ if u need help validating json data first
  84 2011-02-04 00:59:38 <prax> um I am pretty bad at coding things
  85 2011-02-04 00:59:51 <prax> gonna need to hire someone, but not for a couple months or so
  86 2011-02-04 01:00:08 <draginx> what language?
  87 2011-02-04 01:00:09 <prax> JSONmight be one thing I use, don't know much abt it
  88 2011-02-04 01:00:14 <prax> prob PHP too
  89 2011-02-04 01:00:20 <draginx> lol very easy to do mate :S
  90 2011-02-04 01:00:35 <prax> I just suck
  91 2011-02-04 01:00:41 <draginx> if ur using php 5.1+ i believe its 5.1… could be later/earlier but not php4 u can just do json_decode($variable,true);
  92 2011-02-04 01:00:43 <draginx> and then its an array :)
  93 2011-02-04 01:01:08 <prax> think im 5.1+ but idk
  94 2011-02-04 01:01:22 <prax> okay though I made a note, might bug u then
  95 2011-02-04 01:01:26 <prax> cheers
  96 2011-02-04 01:01:41 <luke-jr> doublec: awesome, so when someone steals the N900, they get all your moneys too
  97 2011-02-04 01:01:58 <luke-jr> doublec: I'll leave my bitcoind at home and access it remotely from my N900, thx :P
  98 2011-02-04 01:02:09 <doublec> luke-jr: only if they can get past the encryption on the drive
  99 2011-02-04 01:02:20 <Cusipzzz> luke-jr: nobody wants your tonal coins, you are safe :)
 100 2011-02-04 01:02:45 <doublec> luke-jr: I also don't have a home machine - only a laptop. So it's either steal the phone or the laptop.
 101 2011-02-04 01:03:35 <doublec> using -maxconnections seems to solve the problem of random people connecting
 102 2011-02-04 01:04:01 <luke-jr> doublec: N900 isn't a phone
 103 2011-02-04 01:04:31 <luke-jr> Cusipzzz: TBC and BTC are just human representations of the same underlying commodity
 104 2011-02-04 01:04:51 <Cusipzzz> i know, just like to fsck with ya
 105 2011-02-04 01:04:54 <luke-jr> doublec: filesystem encryption is worthless if you leave it mounted
 106 2011-02-04 01:05:53 <jav> doublec, I believe the latest bitcoin code also adds the option '-nolisten' .. should be another way to avoid having people connect
 107 2011-02-04 01:06:12 <doublec> luke-jr: N900 is definitely a phone
 108 2011-02-04 01:06:16 <luke-jr> doublec: nope.
 109 2011-02-04 01:06:32 <doublec> luke-jr: I must be imagining that phone call I made on it before
 110 2011-02-04 01:06:35 <sipa> can you call with it?
 111 2011-02-04 01:06:42 <doublec> sipa: yes, it's a phone
 112 2011-02-04 01:06:46 <luke-jr> doublec: ability to make calls does not make it a phone
 113 2011-02-04 01:06:53 <luke-jr> I can make calls with my desktop
 114 2011-02-04 01:07:12 <sipa> fine
 115 2011-02-04 01:07:22 <luke-jr> sipa: earlier models didn't even have cellular connectivity
 116 2011-02-04 01:07:33 * jav has an N900 too - and regards it as a phone as well =)
 117 2011-02-04 01:07:48 <luke-jr> that was a new feature in N900, and now all of a sudden it's a "phone" to a bunch of n00bs
 118 2011-02-04 01:08:01 <luke-jr> and N900 even was only spec'd to have it for data only originally
 119 2011-02-04 01:08:14 <sipa> question: does this matter?
 120 2011-02-04 01:08:16 <luke-jr> Nokia just said "heck, why not?" and added the minimal code for audio
 121 2011-02-04 01:08:21 <luke-jr> probably not
 122 2011-02-04 01:08:24 <sipa> ok
 123 2011-02-04 01:08:38 <luke-jr> unless you want to buy a phone
 124 2011-02-04 01:08:51 <luke-jr> then it matters, because N900 doesn't have the features people expect in a phone
 125 2011-02-04 01:08:58 <doublec> like?
 126 2011-02-04 01:09:12 <luke-jr> like MMS, or a phone-centric user interface
 127 2011-02-04 01:09:21 <sipa> mms?
 128 2011-02-04 01:09:25 <doublec> I can send and receive MMS fine
 129 2011-02-04 01:09:26 <luke-jr> sipa: picture SMS
 130 2011-02-04 01:09:34 <sipa> ill send an email if i want
 131 2011-02-04 01:09:38 <luke-jr> doublec: with third party software that only works by random chance
 132 2011-02-04 01:09:46 <doublec> same way as the original iPhone didnt have MMS - but it was still a phone
 133 2011-02-04 01:09:47 <luke-jr> sipa: sure, but that's not a photo interface
 134 2011-02-04 01:09:51 <luke-jr> it didn't?
 135 2011-02-04 01:09:54 <doublec> nope
 136 2011-02-04 01:09:57 <sipa> ive never sent or received an mms in my life
 137 2011-02-04 01:10:03 <luke-jr> still, the software for iPhone is phone software
 138 2011-02-04 01:10:06 <doublec> only jailbroken original iPhones had MMS originally
 139 2011-02-04 01:10:15 <doublec> through third party software
 140 2011-02-04 01:10:28 <luke-jr> for N900, you have to go launch the Phone app to get phone functionality
 141 2011-02-04 01:10:48 <doublec> just like you have to on the iPhone
 142 2011-02-04 01:10:58 <luke-jr> I'm not complaining, I think handheld computing (that is, non-phones) is the future.
 143 2011-02-04 01:11:19 <sipa> or on my android
 144 2011-02-04 01:11:22 <luke-jr> doublec: iPhone can't run arbitrary GNU/Linux/X apps, so there :p
 145 2011-02-04 01:11:29 <luke-jr> neither can Android
 146 2011-02-04 01:11:46 jav has quit (Quit: Verlassend)
 147 2011-02-04 01:16:47 <sipa> luke-jr: but an android phone is a phone, and an n900 isn't?
 148 2011-02-04 01:17:07 <luke-jr> yes, Android is a phone platform.
 149 2011-02-04 01:17:22 <luke-jr> Maemo was a general-purpose computing platform.
 150 2011-02-04 01:17:27 <sipa> but if i put debian on my phone, it stops being a phone?
 151 2011-02-04 01:17:38 <luke-jr> yes, then you've repurposed it :p
 152 2011-02-04 01:17:52 * sipa thinks this is a very fuzzy definition
 153 2011-02-04 01:17:57 <luke-jr> like if you root an XBox360 and run it as a server, it's a server, not a game console
 154 2011-02-04 01:18:06 * doublec thinks luke-jr is a very fuzzy person
 155 2011-02-04 01:18:13 <luke-jr> :p
 156 2011-02-04 01:18:27 <Syke> It all depends on your definition of "is".
 157 2011-02-04 01:18:33 <sipa> haha
 158 2011-02-04 01:18:40 <ArtForz> hhehehe
 159 2011-02-04 01:32:03 Tritonio has quit (Quit: Leaving)
 160 2011-02-04 01:36:58 Zarutian has quit (Quit: Zarutian)
 161 2011-02-04 01:54:55 skeledrew1 has joined
 162 2011-02-04 01:56:51 neocode has quit (Remote host closed the connection)
 163 2011-02-04 01:57:25 skeledrew has quit (Ping timeout: 272 seconds)
 164 2011-02-04 01:58:34 dukeleto has quit (Excess Flood)
 165 2011-02-04 01:59:08 dukeleto has joined
 166 2011-02-04 02:01:51 fauxalliance has quit (Quit: Leaving)
 167 2011-02-04 02:08:10 nevezen has quit (Quit: leaving)
 168 2011-02-04 02:10:22 prax has quit (Read error: Connection reset by peer)
 169 2011-02-04 02:12:46 <afed> ;;bc,diff
 170 2011-02-04 02:12:47 <gribble> 22012.4941572
 171 2011-02-04 02:13:29 Kiba has quit (Ping timeout: 240 seconds)
 172 2011-02-04 02:14:52 Guest67589 is now known as kisom_dev
 173 2011-02-04 02:16:55 prax has joined
 174 2011-02-04 02:17:02 Xunie has quit (Remote host closed the connection)
 175 2011-02-04 02:18:48 Xunie has joined
 176 2011-02-04 02:18:48 Xunie has quit (Changing host)
 177 2011-02-04 02:18:48 Xunie has joined
 178 2011-02-04 02:24:31 * EvanR switches to the pool
 179 2011-02-04 02:25:23 <hacim> EvanR: yeah, its a a little scary running solo when you are trying to recoup your costs, eh?
 180 2011-02-04 02:26:40 <EvanR> im used to that though ;)
 181 2011-02-04 02:26:47 <EvanR> i havent made a profit... ever
 182 2011-02-04 02:26:55 <EvanR> some shit always happens to deplete my reserves
 183 2011-02-04 02:27:09 <EvanR> bitcoin seems like independent profit at this point ;)
 184 2011-02-04 02:30:03 <EvanR> yee bitch
 185 2011-02-04 02:30:08 <EvanR> look at those blocks fly
 186 2011-02-04 02:30:12 <EvanR> lol
 187 2011-02-04 02:32:05 <hacim> hehe
 188 2011-02-04 02:33:27 <EvanR> they havent found a block in 2:22:22
 189 2011-02-04 02:33:41 <EvanR> that means the next block outta not take very long at all!
 190 2011-02-04 02:34:05 <EvanR> </gamblersfallacy>
 191 2011-02-04 02:34:30 <lfm> bad time to join pool
 192 2011-02-04 02:34:50 <EvanR> why?
 193 2011-02-04 02:35:21 <lfm> better pay within first hour or two after a block has been found
 194 2011-02-04 02:35:47 <EvanR> well better join now than later
 195 2011-02-04 02:35:56 <lfm> maybe
 196 2011-02-04 02:39:00 <EvanR> this wallet i have has balance zero, one +50 generate transaction and -50 spend transaction
 197 2011-02-04 02:39:05 <EvanR> can i just throw it away?
 198 2011-02-04 02:39:55 <hacim> if you want
 199 2011-02-04 02:40:04 <EvanR> i cant combine it with my main wallet right
 200 2011-02-04 02:40:05 <doublec> if you've published any addresses from the wallet then any coins sent to that address will be gone forever if you throw it away
 201 2011-02-04 02:40:19 <hacim> it has zero balance
 202 2011-02-04 02:40:38 <EvanR> but theres history, and as doublec said, someone may send to it
 203 2011-02-04 02:40:52 <doublec> a tool to merge wallets would be useful
 204 2011-02-04 02:40:53 <EvanR> well in principle. i didnt public any addresses from it
 205 2011-02-04 02:41:06 <doublec> I had the same situation when I shut a vps down that had a wallet that had addresses I'd used for donations
 206 2011-02-04 02:41:55 <lfm> just merge the balances with a send
 207 2011-02-04 02:42:11 <hacim> if you think there might be coin coming in you'd want to save the private key material and reuse it elsewhere
 208 2011-02-04 02:42:11 <lfm> oh nm i c
 209 2011-02-04 02:42:14 <EvanR> but future payments would be lost
 210 2011-02-04 02:42:23 <doublec> yeah it's not the balances, it's the private keys for the addresses
 211 2011-02-04 02:42:46 <hacim> then you gotta run the private keys on the p2p net until yuou dont think anymore coin will come in
 212 2011-02-04 02:43:22 <EvanR> make sure you change all your public addresses, and dont send to random addresses you wrote down without checking the primary source ;)
 213 2011-02-04 02:43:36 <lfm> well you dont have to run them, just save em
 214 2011-02-04 02:43:55 <presence> there are a couple of people in the virtual hosting game for BTC are there not?
 215 2011-02-04 02:43:59 <EvanR> might suck to have to periodically cycle through all your old wallets to see if anything showed up in them ;)
 216 2011-02-04 02:44:07 <lfm> ya
 217 2011-02-04 02:44:32 <lfm> its also bug prone
 218 2011-02-04 02:45:18 <presence> How many btc/month for hosting
 219 2011-02-04 02:45:38 <doublec> presence: https://www.autovps.net/?Currency=BTC
 220 2011-02-04 02:45:43 <EvanR> i ran my miners for like 2 days with no blocks. i just switched to the pool so now i feel like i wasted work ;)
 221 2011-02-04 02:45:53 <EvanR> if i had just continued mining a little longer id get 50... right?! ;)
 222 2011-02-04 02:46:12 <lfm> or maybe nothing for years
 223 2011-02-04 02:46:23 <EvanR> 2 days was my average
 224 2011-02-04 02:46:29 <doublec> "I've put so much money into this slot machine, I can't stop now, it has to pay out soon, right?"
 225 2011-02-04 02:46:34 <EvanR> lol
 226 2011-02-04 02:46:49 <EvanR> better to stop earlier than later, i guess
 227 2011-02-04 02:47:52 <lfm> if you're losing and your odd predict you will continue to lose on average then yes it is better to stop sooner than later
 228 2011-02-04 02:48:27 <lfm> odds
 229 2011-02-04 02:48:30 <EvanR> so whats the optimal configuration of fan speed and clock speed to heat my room. we have a freezing rain severe weather warning tonight
 230 2011-02-04 02:48:34 <EvanR> xD
 231 2011-02-04 02:49:16 <EvanR> is slow fan and high temperature better? or high speed fan and cooler gpus
 232 2011-02-04 02:50:12 <lfm> depends how hi is hi
 233 2011-02-04 02:51:06 <lfm> up to about 70-80c you should be fine higher than that you may be asking for trouble
 234 2011-02-04 02:51:52 <EvanR> and blowing faster is better?
 235 2011-02-04 02:52:00 <EvanR> at that temperature
 236 2011-02-04 02:52:16 <lfm> if you're too hot then ya speed up your fan if oyu can
 237 2011-02-04 02:53:08 <EvanR> no, im too cold ;)
 238 2011-02-04 02:53:13 <lfm> remember fans use electricity too so dont use em if you dont need em. below 70c you can slow down the fan to save power
 239 2011-02-04 02:53:28 <EvanR> no im trying to use more power
 240 2011-02-04 02:53:48 <lfm> you have free power? or you need more heat?
 241 2011-02-04 02:53:50 <EvanR> i disabled the central heater to save money
 242 2011-02-04 02:53:56 <EvanR> but its freezing outside
 243 2011-02-04 02:53:56 <EvanR> lol
 244 2011-02-04 02:54:31 <EvanR> freezing outside this room too, but i blocked the cracks in the door and the window
 245 2011-02-04 02:54:53 <lfm> what does your central heat run on? electricity? or gas or oil? sometimes eletricity is not a good choice for heating
 246 2011-02-04 02:55:01 <EvanR> yes its electricity
 247 2011-02-04 02:55:06 <EvanR> costs a lot
 248 2011-02-04 02:55:20 <lfm> ok just turn up your fans to 100% then
 249 2011-02-04 02:55:34 <EvanR> k
 250 2011-02-04 02:55:39 <lfm> it still might not be enuf heat of course
 251 2011-02-04 02:55:46 <EvanR> now this will cause the gpus to run at like 57C
 252 2011-02-04 02:56:08 <EvanR> wouldnt it be nice if i could overclock to like 80C ;)
 253 2011-02-04 02:56:19 <lfm> ya thats fine, they are still producing the same heat, its just being blown into the room
 254 2011-02-04 02:56:20 <EvanR> but im already around the limit before it crashes :(
 255 2011-02-04 02:56:22 <ArtForz> so?
 256 2011-02-04 02:56:36 <ArtForz> higher exhaust velocity is better, gets more air movement
 257 2011-02-04 02:56:49 <ArtForz> otherwise you get *very* nice layering
 258 2011-02-04 02:56:59 <EvanR> yes thats fine, but id like it to be hotter
 259 2011-02-04 02:56:59 <lfm> and higher fan power produces heat in the fan
 260 2011-02-04 02:57:17 <EvanR> overclocking does that, and makes more coins... but crashes X
 261 2011-02-04 02:57:22 <EvanR> :(
 262 2011-02-04 02:57:51 <lfm> EvanR did you try increasing the voltage? (its dangerous tho)
 263 2011-02-04 02:57:54 <EvanR> heh
 264 2011-02-04 02:57:56 <EvanR> no
 265 2011-02-04 02:58:51 <EvanR> how about cpu mining at the same time ;)
 266 2011-02-04 02:59:05 <lfm> yup, turn up the cpus too
 267 2011-02-04 02:59:13 <EvanR> m0's work on the cpu?
 268 2011-02-04 02:59:22 <EvanR> i saw it listed, but i thought it was a gpu miner
 269 2011-02-04 02:59:40 <lfm> just use main bitcoin or cpuminer from jgarzik
 270 2011-02-04 03:00:24 <lfm> I think puddinpoop has some cpu miners too
 271 2011-02-04 03:00:58 <EvanR> ive used jgarzik before so ill use that
 272 2011-02-04 03:00:58 <presence> so dont run x
 273 2011-02-04 03:01:04 <presence> why do you need x anyway
 274 2011-02-04 03:01:12 <EvanR> because
 275 2011-02-04 03:01:24 <EvanR> ;)
 276 2011-02-04 03:01:24 <lfm> the opencl support is iin the X drivers
 277 2011-02-04 03:01:35 <presence> you can load thr drivers without running x
 278 2011-02-04 03:01:58 <EvanR> hehe, disable X, and you can overclock to whatever!
 279 2011-02-04 03:02:15 <lfm> presence: ok, go ahead, and if you get the chance detail for the rest of us how you do it
 280 2011-02-04 03:02:50 <presence> hell, I have the video card passed through to the VM on my esx4i server and it works fine :D
 281 2011-02-04 03:03:15 <lfm> ok esx eh, not for me
 282 2011-02-04 03:03:30 <presence> I use it for other reasons.  Why not put a card in it and mine?
 283 2011-02-04 03:03:42 <presence> better than standing up a host solely to use the gpu
 284 2011-02-04 03:07:43 <EvanR> god dammit
 285 2011-02-04 03:07:46 alowm has quit (Quit: leaving)
 286 2011-02-04 03:08:22 <EvanR> again with cpuminer automake crap
 287 2011-02-04 03:09:04 sabalaba has quit (Ping timeout: 240 seconds)
 288 2011-02-04 03:09:10 <EvanR> does anyone have binaries
 289 2011-02-04 03:09:48 <lfm> automake is good, whats wrong with your system?
 290 2011-02-04 03:10:11 <EvanR> nothing
 291 2011-02-04 03:10:20 <EvanR> this is the first times its failed
 292 2011-02-04 03:10:24 <EvanR> i always build my own stuff
 293 2011-02-04 03:10:27 <lfm> its missing some libs eh?
 294 2011-02-04 03:10:47 <EvanR> ill pastebin the error
 295 2011-02-04 03:12:48 shortcircuit has left ()
 296 2011-02-04 03:13:32 <EvanR> http://codepad.org/yp280REl
 297 2011-02-04 03:14:48 <lfm> you cant just run ./configure instead?
 298 2011-02-04 03:15:19 <EvanR> oh, it comes with configure now.
 299 2011-02-04 03:15:20 <EvanR> ill try
 300 2011-02-04 03:15:28 <lfm> doh
 301 2011-02-04 03:15:31 <EvanR> no
 302 2011-02-04 03:15:39 <EvanR> let me start over
 303 2011-02-04 03:15:57 <ne0futur> re install imlib-devel
 304 2011-02-04 03:15:58 <EvanR> no, it doesnt come with configure
 305 2011-02-04 03:16:00 <ne0futur> clean all
 306 2011-02-04 03:16:14 <EvanR> i saw a half built version, which failed
 307 2011-02-04 03:16:28 <EvanR> all 'packages' come with no configure, configured
 308 2011-02-04 03:16:49 <EvanR> autoconf throws that error when configure attempts to generate
 309 2011-02-04 03:17:03 <EvanR> 2.65
 310 2011-02-04 03:17:08 <EvanR> :(
 311 2011-02-04 03:17:43 <EvanR> last time i complained jgarzik gave me a package with stuff configured and i just have to configure make
 312 2011-02-04 03:22:04 <jgarzik> EvanR: download cpuminer 0.6 tarball from http://yyz.us/bitcoin/
 313 2011-02-04 03:22:32 <EvanR> ah
 314 2011-02-04 03:22:43 <EvanR> bookmarked
 315 2011-02-04 03:22:59 <EvanR> i tried cryptopp_asm32 on my 0.3 binary on my new miner... segfaulted
 316 2011-02-04 03:23:02 <EvanR> ill try .6
 317 2011-02-04 03:23:03 epicurus has joined
 318 2011-02-04 03:23:24 <lfm> EvanR I just downloaded cpuminer 0.6 and it has configure
 319 2011-02-04 03:23:45 <lfm> tar.gz
 320 2011-02-04 03:24:02 <EvanR> my browser doesnt show me a tar.gz for .6 in github
 321 2011-02-04 03:24:09 <EvanR> just master
 322 2011-02-04 03:24:21 <lfm> https://www.bitcoin.org/smf/index.php?topic=1925.0
 323 2011-02-04 03:25:47 <EvanR> cryptopp_asm32 works!
 324 2011-02-04 03:25:56 <EvanR> 150% increase in speed
 325 2011-02-04 03:26:05 <EvanR> 1.2 Mhps
 326 2011-02-04 03:26:14 <EvanR> sempron ;)
 327 2011-02-04 03:26:31 <lfm> wtg
 328 2011-02-04 03:27:15 <EvanR> 3hours block...
 329 2011-02-04 03:27:27 <EvanR> any minute now
 330 2011-02-04 03:29:23 <EvanR> ;;bc,calc 30000000
 331 2011-02-04 03:29:23 <gribble> The average time to generate a block at 30000000 Khps, given current difficulty of 22012.4941572 , is 52 minutes and 31 seconds
 332 2011-02-04 03:29:32 kiba has joined
 333 2011-02-04 03:29:58 <jgarzik> EvanR: you don't want the .tar.gz from github; that is not a distribution tarball, so you'd have to deal with autoconf/automake nastiness
 334 2011-02-04 03:30:06 <jgarzik> distribution tarball has configure, and saves you a lot of work
 335 2011-02-04 03:30:14 <EvanR> yeah i figured
 336 2011-02-04 03:30:24 <EvanR> is there a link to that distribution site on github?
 337 2011-02-04 03:30:36 <jgarzik> yyz.us/bitcoin is the official download site for cpuminer
 338 2011-02-04 03:30:44 <EvanR> tell google that
 339 2011-02-04 03:31:01 <jgarzik> EvanR: I don't know how to set that up on github. If that's possible, I should do that
 340 2011-02-04 03:31:23 <EvanR> i dont know github
 341 2011-02-04 03:31:50 <lfm> blind leading blind eh
 342 2011-02-04 03:31:56 <andrew12> heh
 343 2011-02-04 03:32:37 <andrew12> jgarzik: are you trying to figure out if it's possible to make a site with github?
 344 2011-02-04 03:32:54 <andrew12> or are you trying to figure out how to upload tars and shit on github
 345 2011-02-04 03:32:55 <EvanR> a link to the project homepage on the github page
 346 2011-02-04 03:33:01 <andrew12> because both are possible
 347 2011-02-04 03:33:08 <andrew12> oh
 348 2011-02-04 03:33:31 <andrew12> sec
 349 2011-02-04 03:33:31 <lfm> andrew12: they just want people who stumble on github to find the official downloads
 350 2011-02-04 03:33:53 <andrew12> jgarzik: on https://github.com/jgarzik/cpuminer, click admin,
 351 2011-02-04 03:33:58 <andrew12> ...damn you enter key
 352 2011-02-04 03:34:25 <andrew12> er
 353 2011-02-04 03:34:40 <EvanR> so, as this 3+ hour long block on the pool continues, my expected payout is growing. was 0, then .1 .2 now .3
 354 2011-02-04 03:34:43 <andrew12> jgarzik: on https://github.com/jgarzik/cpuminer, hover over the empty space under the description. it's pretty obvious from there
 355 2011-02-04 03:34:49 <EvanR> is it going to continue to grow?
 356 2011-02-04 03:34:56 <andrew12> EvanR: maybe!
 357 2011-02-04 03:35:01 <EvanR> well it cant go over 50
 358 2011-02-04 03:35:09 <andrew12> what if it does?! :p
 359 2011-02-04 03:35:16 <EvanR> ._.
 360 2011-02-04 03:36:41 <lfm> whats your khash/s?
 361 2011-02-04 03:37:55 <hacim> when you run a cpu miner, do you want to run one thread per physical CPU/
 362 2011-02-04 03:38:07 <luke-jr> hacim: one thread per core, like with anything
 363 2011-02-04 03:38:33 <hacim> luke-jr: so if you have 2 cpus hyperthreaded  you launch 4
 364 2011-02-04 03:38:43 <luke-jr> no
 365 2011-02-04 03:38:48 <luke-jr> hyperthreading is not cores
 366 2011-02-04 03:38:59 <lfm> oh you have to try it, hyperthreaded cpus may not count as full cpus
 367 2011-02-04 03:39:16 <luke-jr> hacim: you have 2 single-core CPUs?
 368 2011-02-04 03:39:17 <lfm> sometimes they work and sometimes they dont
 369 2011-02-04 03:39:42 devrandom1 has quit (Ping timeout: 240 seconds)
 370 2011-02-04 03:39:43 <luke-jr> lfm: spawning cores + 1 may be of minute benefit, but that's about it really
 371 2011-02-04 03:39:53 <lfm> depends
 372 2011-02-04 03:39:56 <EvanR> lfm: 1.2M on the cpu 620 on the gpu
 373 2011-02-04 03:40:39 <lfm> ;;bc,calcd 621200 1
 374 2011-02-04 03:40:39 <gribble> The average time to generate a block at 621200 Khps, given the supplied difficulty of 1, is 6 seconds
 375 2011-02-04 03:40:48 <luke-jr> lol
 376 2011-02-04 03:40:54 <EvanR> $_$
 377 2011-02-04 03:40:56 <hacim> i'm getting 475.52 per thread
 378 2011-02-04 03:41:12 <luke-jr> can you imagine, if ArtForz had his GPU miner at the start? XD
 379 2011-02-04 03:41:28 <presence> the difficulty would have went up fast
 380 2011-02-04 03:41:29 <luke-jr> he'd be insane rich
 381 2011-02-04 03:41:29 <EvanR> hed have 21 million!
 382 2011-02-04 03:41:34 <presence> nah
 383 2011-02-04 03:41:39 <presence> it would auto-scale
 384 2011-02-04 03:41:39 <luke-jr> 51k bitcoins off the bat
 385 2011-02-04 03:41:41 <lfm> luke well difficulty would have been a lot higher
 386 2011-02-04 03:41:49 <luke-jr> lfm: only after 1020 blocks
 387 2011-02-04 03:41:55 <luke-jr> the first 1020 would have been easy money
 388 2011-02-04 03:42:04 <lfm> 2016 block you mean?
 389 2011-02-04 03:42:07 <luke-jr> yeah
 390 2011-02-04 03:42:19 <luke-jr> $100k if he saved it until now
 391 2011-02-04 03:42:31 <luke-jr> in under 4 hours
 392 2011-02-04 03:42:32 <lfm> luke-jr:  ya I saw what happened when he turned on testnet
 393 2011-02-04 03:42:42 <luke-jr> $25k/hr
 394 2011-02-04 03:42:47 <luke-jr> wish I could get paid that
 395 2011-02-04 03:42:49 <luke-jr> XD
 396 2011-02-04 03:43:08 <lfm> hindsight
 397 2011-02-04 03:43:10 <EvanR> 100k meh
 398 2011-02-04 03:43:18 <EvanR> few years salary lol
 399 2011-02-04 03:43:25 <EvanR> chump change
 400 2011-02-04 03:43:44 <luke-jr> Warning: overflow encountered in uint_scalars
 401 2011-02-04 03:43:46 <luke-jr> so anyone else get that?
 402 2011-02-04 03:43:57 <luke-jr> I'd really like some peace of mind that my mining is working :x
 403 2011-02-04 03:44:09 <lfm> if he had done that it might have ruined bitcoin from the start and it might never have been worth anything, you cant say for sure
 404 2011-02-04 03:44:11 <EvanR> you got coins from it?
 405 2011-02-04 03:44:16 <luke-jr> EvanR: not so far
 406 2011-02-04 03:44:17 <EvanR> its working xD
 407 2011-02-04 03:44:19 <EvanR> oh
 408 2011-02-04 03:44:47 <EvanR> lfm: like satoshi?
 409 2011-02-04 03:44:48 <EvanR> lol
 410 2011-02-04 03:44:48 <epicurus> i'll wager 500 quatloos on the newcomer
 411 2011-02-04 03:44:51 <luke-jr> but I'm only supposed to find a blokc every 4 days or so
 412 2011-02-04 03:44:58 <luke-jr> wtf are quatloos
 413 2011-02-04 03:45:12 <luke-jr> is that a new bitcoin unit?
 414 2011-02-04 03:45:15 <lfm> ditto
 415 2011-02-04 03:45:18 <EvanR> luke-jr: yeah so it might be 8 days, maybe even 16!
 416 2011-02-04 03:45:21 <EvanR> :S
 417 2011-02-04 03:45:32 <hacim> i get pretty much the same hashrate with cryptopp as SHA256
 418 2011-02-04 03:45:48 <EvanR> gotta use cyptopp_asm32
 419 2011-02-04 03:46:05 <lfm> i use via
 420 2011-02-04 03:46:21 <luke-jr> does anyone here by chance have Deadly Games?
 421 2011-02-04 03:46:25 <hacim> via gives me an illegal instruction
 422 2011-02-04 03:46:36 <hacim> i dont see any improvement with cyptopp_asm32
 423 2011-02-04 03:46:39 <luke-jr> hacim: pretty sure via requires special hw
 424 2011-02-04 03:46:43 <lfm> well you have to have a VIA cpu, not intel or amd
 425 2011-02-04 03:46:59 <luke-jr> lfm: you should add it to the wiki
 426 2011-02-04 03:47:06 <luke-jr> under CPUs/Nvidia
 427 2011-02-04 03:47:18 <lfm> nvidia isnt via
 428 2011-02-04 03:47:33 <luke-jr> no?
 429 2011-02-04 03:47:37 <lfm> what wiki?
 430 2011-02-04 03:47:42 <luke-jr> bitocin wiki
 431 2011-02-04 03:47:42 <hacim> ;;bc,calcd 2743
 432 2011-02-04 03:47:42 <gribble> (bc,calcd <an alias, 2 arguments>) -- Alias for "echo The average time to generate a block at $1 Khps, given the supplied difficulty of $2, is [time elapsed [math calc 1/((2**224-1)/$2*$1*1000/2**256)]]".
 433 2011-02-04 03:47:57 <luke-jr> nvidia didn't buy via?
 434 2011-02-04 03:48:10 <lfm> dont think so
 435 2011-02-04 03:48:37 <luke-jr> O.O
 436 2011-02-04 03:48:43 <luke-jr> weird maths
 437 2011-02-04 03:48:58 <luke-jr> lfm: https://en.bitcoin.it/wiki/Mining_Hardware_Comparison
 438 2011-02-04 03:49:46 <luke-jr> I guess put it under VIA then
 439 2011-02-04 03:50:03 <presence> I wanna know why teslas suck ass
 440 2011-02-04 03:50:08 <presence> 1.x Tflops each
 441 2011-02-04 03:50:19 <presence> the drivers have to be balls
 442 2011-02-04 03:50:24 <EvanR> lol
 443 2011-02-04 03:50:27 <presence> which is lame since it doesnt drive video
 444 2011-02-04 03:50:31 <presence> it JUST DOES MATH
 445 2011-02-04 03:50:31 <luke-jr> presence: because it's nvidia
 446 2011-02-04 03:50:39 <luke-jr> all nvidia stuff sucks
 447 2011-02-04 03:50:59 <lfm> teslas are good for double precision fp and also good for ecc memory
 448 2011-02-04 03:52:01 <lfm> altho ati DP may be has caught up now
 449 2011-02-04 03:52:01 devrandom1 has joined
 450 2011-02-04 03:52:40 <ArtForz> they did
 451 2011-02-04 03:52:52 <ArtForz> 6970 is faster than C20xx for DP
 452 2011-02-04 03:53:11 <ArtForz> well, C/M/whatever 20xx
 453 2011-02-04 03:53:32 <epicurus> http://www.youtube.com/watch?v=umbqfa3XyTA#t=4m54s
 454 2011-02-04 03:53:45 <epicurus> to learn about quatloos
 455 2011-02-04 04:00:19 andrew12 has quit ()
 456 2011-02-04 04:15:29 sabalaba has joined
 457 2011-02-04 04:38:31 Cusipzzz has quit (Quit: KVIrc 4.0.2 Insomnia http://www.kvirc.net/)
 458 2011-02-04 05:23:37 Vladimir has joined
 459 2011-02-04 05:28:40 lfm has quit (Quit: bye)
 460 2011-02-04 05:31:44 hackNstuff has quit (Ping timeout: 240 seconds)
 461 2011-02-04 05:36:07 doublec has quit (Quit: Leaving)
 462 2011-02-04 05:51:34 <frewsxcv> if there is no central bitcoin server, where does the data come to generate coins?
 463 2011-02-04 05:51:56 <luke-jr> frewsxcv: from the person who generates it
 464 2011-02-04 05:52:04 <luke-jr> frewsxcv: it's just a bruteforce mechanism
 465 2011-02-04 05:52:34 <luke-jr> the SHA256 hash of the block must be less than the difficulty
 466 2011-02-04 05:52:58 <luke-jr> so miners are just bruteforcing different data to try to find a hash that fits
 467 2011-02-04 05:53:50 andrew12 has joined
 468 2011-02-04 05:54:44 andrew12 has quit (Changing host)
 469 2011-02-04 05:54:44 andrew12 has joined
 470 2011-02-04 05:55:56 <andrew12> was just about to ask why BCBot sends its message after you identify, then i remembered that the server makes it look like you reconnected
 471 2011-02-04 05:57:37 <luke-jr> it has to
 472 2011-02-04 06:00:48 noagendamarket has quit (Ping timeout: 245 seconds)
 473 2011-02-04 06:01:03 <frewsxcv> luke-jr: how does the person who sends the data know where to send the data?
 474 2011-02-04 06:01:23 <luke-jr> frewsxcv: it sends it everywhere
 475 2011-02-04 06:01:46 <draginx> its a… torrent.. network lol =x
 476 2011-02-04 06:02:11 <draginx> Upload/Download
 477 2011-02-04 06:03:48 <frewsxcv> draginx: torrents have a tracker
 478 2011-02-04 06:07:10 <luke-jr> do clients care what the Timestamp is for blocks?
 479 2011-02-04 06:07:25 <luke-jr> eg, will they notice if a miner is off by several years?
 480 2011-02-04 06:10:30 <nanotube> yes, there is a validity window.
 481 2011-02-04 06:11:42 <luke-jr> how big is it?
 482 2011-02-04 06:11:58 <luke-jr> what if it just takes a long time to cross the network?
 483 2011-02-04 06:12:08 <luke-jr> what if someone's timezone is just set wrong?
 484 2011-02-04 06:12:30 <draginx> isn't the irc a torrent tracker? doesnt bitcoind look for irc if it's not found then to a common list? Isn't that common list open sourced?
 485 2011-02-04 06:13:01 <luke-jr> draginx: that list cannot be copyrighted
 486 2011-02-04 06:13:12 <luke-jr> and Bitcoin shares no connection to bittorrent
 487 2011-02-04 06:13:51 <draginx> luke-jr: sorry, i meant open sourced as in anyone can see where it goes too through the source of your code (you can do MD5 checksums or just git it :P)
 488 2011-02-04 06:14:10 Diablo-D3 has joined
 489 2011-02-04 06:14:43 <draginx> https://github.com/bitcoin/bitcoin here you go :)
 490 2011-02-04 06:20:49 <Diablo-D3> wee
 491 2011-02-04 06:20:57 <Diablo-D3> I've found 3 blocks on BPM
 492 2011-02-04 06:22:49 <luke-jr> heh, I was just pondering pre-processing block hashes, but it looks like bitcoind already does that ☺
 493 2011-02-04 06:23:06 <luke-jr> can the GPU miners use the midstate, or do they start from scratch every hash?
 494 2011-02-04 06:23:15 <Diablo-D3> luke-jr: they use the midstate already
 495 2011-02-04 06:23:29 <Diablo-D3> mine just happens to double check before submit by fully hashing
 496 2011-02-04 06:23:59 <luke-jr> no offense, but I'm not using yours :P
 497 2011-02-04 06:24:07 <Diablo-D3> well, you said gpu miners
 498 2011-02-04 06:24:11 <Diablo-D3> mine is a gpu miner
 499 2011-02-04 06:24:16 <luke-jr> yeah
 500 2011-02-04 06:24:23 <luke-jr> there are others ☺
 501 2011-02-04 06:24:31 <luke-jr> Diablo-D3: I did try yours on my MBP, but it didn't owrk
 502 2011-02-04 06:24:39 <luke-jr> probably becuase I don't have *Snow* Leopard
 503 2011-02-04 06:24:45 <Diablo-D3> Yes, thats why
 504 2011-02-04 06:24:49 <Diablo-D3> no GPU miner will work on yours
 505 2011-02-04 06:25:06 <luke-jr> oh well
 506 2011-02-04 06:25:17 <Diablo-D3> and if yours has hybrid chips (which it probably dosent because it didnt some with 10.6), mine wont work either
 507 2011-02-04 06:32:17 <jgarzik> does the diablo miner check more than high 32 bits of target?
 508 2011-02-04 06:32:28 <Diablo-D3> in the kernel? no
 509 2011-02-04 06:32:31 <Diablo-D3> in the host? yes
 510 2011-02-04 06:32:36 <jgarzik> And it appears that m0mchil checks high 64 bits of target, if I'm reading the code correctly.
 511 2011-02-04 06:32:54 <Diablo-D3> m0's kernel checks H in the kernel, G in the host
 512 2011-02-04 06:33:24 kiba has quit (Ping timeout: 240 seconds)
 513 2011-02-04 06:33:24 <Diablo-D3> I check H in the kernel, H and G in the host, and also check if the whole hash actually works in the host
 514 2011-02-04 06:39:25 <luke-jr> does m0's miner tell me if it finds a block? :P
 515 2011-02-04 06:40:53 <luke-jr> hmmm
 516 2011-02-04 06:42:53 <Diablo-D3> yes
 517 2011-02-04 06:45:13 <luke-jr> it's cute how all the possible cheats are secured properly :p
 518 2011-02-04 06:45:22 <luke-jr> and/or implemented
 519 2011-02-04 06:45:43 <luke-jr> 'midstate' is already done
 520 2011-02-04 06:46:25 <luke-jr> setting my clock 1.5 hours into the future won't cause others' blocks to be rejected for that long because they're "old", since the clients use a minimum of lastblock-timestamp + 1 for their own timestamps…
 521 2011-02-04 06:47:54 Slix` has quit (Read error: Connection reset by peer)
 522 2011-02-04 06:49:44 <luke-jr> Diablo-D3: since I was on ignore when I implicitly asked the first time: do you think there's a chance your miner could beat 265 MH/s on a 5850?
 523 2011-02-04 06:50:43 lfm has joined
 524 2011-02-04 06:51:18 alystair has quit (Ping timeout: 272 seconds)
 525 2011-02-04 06:51:51 <Diablo-D3> have you tried it recently?
 526 2011-02-04 06:52:03 <luke-jr> never.
 527 2011-02-04 06:52:13 <luke-jr> would be an awful pain to get some kind of Java setup
 528 2011-02-04 06:52:33 <Diablo-D3> what, apt-get install openjdk6-jre?
 529 2011-02-04 06:53:00 <luke-jr> my VM is Gentoo :p
 530 2011-02-04 06:53:59 <Diablo-D3> then... emerge... dongs. I dont know.
 531 2011-02-04 06:54:01 <Diablo-D3> whatever the gentoo does.
 532 2011-02-04 06:54:15 <luke-jr> it probably fails.
 533 2011-02-04 06:54:45 * luke-jr doesn't have too much confidence in Gentoo developers when it comes to non-standard stuff
 534 2011-02-04 06:54:57 <Diablo-D3> ... but its open source GPL'ed software!
 535 2011-02-04 06:55:06 <Diablo-D3> you run one command to build it!
 536 2011-02-04 06:55:08 <luke-jr> is it? :p
 537 2011-02-04 06:55:11 <Diablo-D3> how hard can this be!
 538 2011-02-04 06:55:20 <luke-jr> dunno, Java doesn't work like most software
 539 2011-02-04 06:55:31 <luke-jr> and Gentoo tries to support multiple JDKs at once
 540 2011-02-04 06:55:33 TheAncientGoat has joined
 541 2011-02-04 06:55:44 <Diablo-D3> well you dont need MULTIPLE jdks
 542 2011-02-04 06:55:46 <Diablo-D3> you need one
 543 2011-02-04 06:55:55 <luke-jr> but Gentoo still does the same crap for one
 544 2011-02-04 06:56:24 <luke-jr> maybe I'll clone my VM someday to try it…
 545 2011-02-04 06:56:48 <luke-jr> wanna setup another VM to try PCSX2 in anyway
 546 2011-02-04 06:58:42 <OneFixt> who here knows m0m's miner inside-out?
 547 2011-02-04 07:00:25 <lfm> ;;seen m0mchil
 548 2011-02-04 07:00:25 <gribble> m0mchil was last seen in #bitcoin-dev 1 week, 0 days, 13 hours, 55 minutes, and 55 seconds ago: <m0mchil> pool is diff = 1
 549 2011-02-04 07:00:42 <lfm> you missed him
 550 2011-02-04 07:08:44 epicurus has quit (Quit: Leaving)
 551 2011-02-04 07:21:02 <OneFixt> Diablo-D3: ping
 552 2011-02-04 07:27:34 RazielZ has joined
 553 2011-02-04 07:30:28 <Diablo-D3> what
 554 2011-02-04 07:31:51 devrandom1 has quit (Quit: Ex-Chat)
 555 2011-02-04 07:37:45 <andrew12> heh
 556 2011-02-04 07:38:16 <Diablo-D3> OneFixt: what
 557 2011-02-04 07:39:47 <OneFixt> Diablo-D3: I have a question about your miner
 558 2011-02-04 07:40:03 <Diablo-D3> it helps if you ask it
 559 2011-02-04 07:40:18 <OneFixt> you said it does H==0 checks in the kernel + winning hash check in the program?
 560 2011-02-04 07:40:25 <Diablo-D3> yes
 561 2011-02-04 07:40:47 <OneFixt> can a h==0 hash overwrite another h==0 hash, like in m0m's?
 562 2011-02-04 07:41:02 <OneFixt> ie, do you loop through many threads before checking whether the hash is a winner?
 563 2011-02-04 07:41:11 <Diablo-D3> yes, but the chances of seeing two interesting h==0 at the same time is incredibly low
 564 2011-02-04 07:41:29 <OneFixt> how many threads run before you check the h==0?
 565 2011-02-04 07:41:42 <Diablo-D3> whatever the current global run length is
 566 2011-02-04 07:42:01 <OneFixt> i'll take m0m's as an example, and use a 350MH card
 567 2011-02-04 07:42:11 <Diablo-D3> yeah, but its a function of time
 568 2011-02-04 07:42:27 <OneFixt> let's take 1 sec kernel run which is default i think
 569 2011-02-04 07:42:35 <Diablo-D3> -f 60 is the default
 570 2011-02-04 07:42:50 <ArtForz> back
 571 2011-02-04 07:42:51 <Diablo-D3> but on -f 1, its either 350 million or 350/3
 572 2011-02-04 07:42:52 <Diablo-D3> I forget
 573 2011-02-04 07:42:56 <OneFixt> ah, art!
 574 2011-02-04 07:42:56 <luke-jr> I use -f 1
 575 2011-02-04 07:42:58 <ArtForz> OneFixt: checked the math, you're right
 576 2011-02-04 07:43:02 <OneFixt> phew
 577 2011-02-04 07:43:07 <Diablo-D3> math on what?
 578 2011-02-04 07:43:09 <ArtForz> about 3.5% loss with 350M trys/run
 579 2011-02-04 07:43:11 <OneFixt> yeah the sim confirmed it, but no point in looking at it
 580 2011-02-04 07:43:36 <Diablo-D3> OneFixt: art's miner has large output buffers to store this
 581 2011-02-04 07:44:06 <ArtForz> my original kernel used one "space" per 1024 nonces
 582 2011-02-04 07:44:16 <ArtForz> so chances of a dupe were pretty damn low
 583 2011-02-04 07:44:27 <Diablo-D3> I probably should switch to something similar for one reason
 584 2011-02-04 07:44:32 <OneFixt> yep, true
 585 2011-02-04 07:44:38 <Diablo-D3> pool share stuffing.
 586 2011-02-04 07:45:38 <ArtForz> and even with a 2* H==0 in one spot I don't lose nonces
 587 2011-02-04 07:46:01 <OneFixt> yeah, 7 weeks of runtime to get one h==0 at 1024
 588 2011-02-04 07:46:04 <ArtForz> I recheck all 1024 in that area on the host to find the H==0 ones
 589 2011-02-04 07:46:16 <OneFixt> ah
 590 2011-02-04 07:46:34 <Diablo-D3> pool share stuffing should be hilarious
 591 2011-02-04 07:46:36 <ArtForz> becasuse for some reason flag=1 is faster than flag=cur_nonce
 592 2011-02-04 07:46:58 * luke-jr notes that even if you DID get two interesting at the same time, they would each invalidate the other anyway…
 593 2011-02-04 07:47:06 <Diablo-D3> ArtForz: because there are two magical registers, one holds 0, one holds 1
 594 2011-02-04 07:47:07 <OneFixt> funny, but that does remove the inefficiency
 595 2011-02-04 07:47:11 <Diablo-D3> since they are so fucking common
 596 2011-02-04 07:47:26 <Diablo-D3> luke-jr: yes
 597 2011-02-04 07:47:29 <Diablo-D3> luke-jr: but not as shares
 598 2011-02-04 07:47:33 <luke-jr> oh, I suppose
 599 2011-02-04 07:47:34 <ArtForz> rechecking 1024 out of 2**32 nonces on the host is less slowdown than the hit from flag=nonce
 600 2011-02-04 07:48:13 <OneFixt> yeah that makes sense somewhat
 601 2011-02-04 07:48:16 <ArtForz> 1 CPU hash for every 4M GPU hashes
 602 2011-02-04 07:48:54 <OneFixt> with the miners currently being used by most people though, this thing is a problem
 603 2011-02-04 07:49:00 <ArtForz> yep
 604 2011-02-04 07:49:03 <ArtForz> especially m0s
 605 2011-02-04 07:49:04 <OneFixt> i've noticed slush's pool underperforming lately
 606 2011-02-04 07:49:05 <OneFixt> yep
 607 2011-02-04 07:49:07 <ArtForz> diablos is less affected
 608 2011-02-04 07:49:10 <luke-jr> this radeon is heating up my PC :/
 609 2011-02-04 07:49:22 <Diablo-D3> hmm
 610 2011-02-04 07:49:27 <Diablo-D3> a 8mb buffer could do like
 611 2011-02-04 07:49:28 <luke-jr> watch it screw up my motherboard's SATA sockets ☹
 612 2011-02-04 07:49:32 <ArtForz> with -f 60 its only like 6M nonces/kernel
 613 2011-02-04 07:49:44 <Diablo-D3> 8/4*1024 hashes?
 614 2011-02-04 07:50:25 <OneFixt> so .058%
 615 2011-02-04 07:50:41 <ArtForz> I use 1MB buffer size
 616 2011-02-04 07:50:52 <Diablo-D3> ArtForz: yeah, but I use variably sized kernels
 617 2011-02-04 07:50:58 <Diablo-D3> so the cap has to be big enough
 618 2011-02-04 07:51:21 <OneFixt> still, we should fix the code up for those who run -f 1
 619 2011-02-04 07:51:26 <ArtForz> 1MB = 256M nonces per kernel run
 620 2011-02-04 07:51:41 <Diablo-D3> heh
 621 2011-02-04 07:51:54 <Diablo-D3> well wait
 622 2011-02-04 07:51:57 <Diablo-D3> thats still awfully low
 623 2011-02-04 07:52:00 <ArtForz> 1024 nonces = 4 byte, so 1 byte = 256 nonces
 624 2011-02-04 07:52:00 <Diablo-D3> since
 625 2011-02-04 07:52:19 <Diablo-D3> -f 1 on a 256 mhash card maxes that out
 626 2011-02-04 07:52:25 <ArtForz> yes, thats about -f 1.5 on a 5970
 627 2011-02-04 07:52:49 <Diablo-D3> so I could easily do a 4 MB buffer
 628 2011-02-04 07:52:54 <ArtForz> yep
 629 2011-02-04 07:52:54 <Diablo-D3> and future proof it for the next few years
 630 2011-02-04 07:53:09 <Diablo-D3> I cant remember what the max is on 4xxx
 631 2011-02-04 07:53:15 <Diablo-D3> because I ran into that on my split kernel shit
 632 2011-02-04 07:53:19 <Diablo-D3> I think its 16
 633 2011-02-04 07:53:29 <ArtForz> that should be good for up to 2**30 nonces/clenqueue
 634 2011-02-04 07:54:04 <ArtForz> = about 1Gh/s at -f 1
 635 2011-02-04 07:54:15 <ArtForz> yeah, should be good enough for a while
 636 2011-02-04 07:54:57 <ArtForz> I'd limit max workitems just in case
 637 2011-02-04 07:55:24 larsivi has joined
 638 2011-02-04 07:55:35 <ArtForz> and btw, I dont like dynamically adjusting #workitems, seems to like to cause weird oscillations
 639 2011-02-04 07:57:28 <ArtForz> = tried it, caused some issues - "fuck this, just use a cmdline param"
 640 2011-02-04 07:58:54 <OneFixt> agreed
 641 2011-02-04 08:03:02 <prax> is freenode like an all linux thing?
 642 2011-02-04 08:04:02 <prax> errr no huh
 643 2011-02-04 08:06:53 <luke-jr> just majority
 644 2011-02-04 08:07:09 <prax> k
 645 2011-02-04 08:07:22 <prax> I used some flavor of linux only once heh
 646 2011-02-04 08:07:35 <prax> thats what most of you guys here are running eh?
 647 2011-02-04 08:09:00 <Diablo-D3> ArtForz: it doesnt seem to oscillate much for me
 648 2011-02-04 08:10:16 <ArtForz> try taunting it
 649 2011-02-04 08:10:30 <Diablo-D3> hurr
 650 2011-02-04 08:11:55 <ArtForz> run at lowish -f, then drag windows around, run flash, ..., you know, the usual candidates for stressing X
 651 2011-02-04 08:12:42 <Diablo-D3> yeah which is why I use -f 1000
 652 2011-02-04 08:12:54 <Diablo-D3> it wildly jumps all over the place to fill in any extra time
 653 2011-02-04 08:13:57 joe_1 has joined
 654 2011-02-04 08:23:33 sabalaba has quit (Ping timeout: 250 seconds)
 655 2011-02-04 08:23:38 ArtForz has quit (Read error: Operation timed out)
 656 2011-02-04 08:27:31 ArtForz has joined
 657 2011-02-04 08:37:28 sabalaba has joined
 658 2011-02-04 08:45:36 joe_1 has quit (Ping timeout: 246 seconds)
 659 2011-02-04 08:48:49 * Diablo-D3 just makes a double khash meter.
 660 2011-02-04 08:53:01 genjix has joined
 661 2011-02-04 08:53:36 xelister has quit (Read error: Connection reset by peer)
 662 2011-02-04 08:53:51 xelister has joined
 663 2011-02-04 08:53:51 xelister has quit (Changing host)
 664 2011-02-04 08:53:51 xelister has joined
 665 2011-02-04 08:56:54 <Diablo-D3> huh
 666 2011-02-04 08:58:02 altamic has joined
 667 2011-02-04 08:58:02 altamic has quit (Changing host)
 668 2011-02-04 08:58:02 altamic has joined
 669 2011-02-04 09:01:54 chaord has quit (Ping timeout: 255 seconds)
 670 2011-02-04 09:01:58 <CIA-98> DiabloMiner: Patrick McFarland master * ra7d67f2 / src/main/java/com/diablominer/DiabloMiner/DiabloMiner.java :
 671 2011-02-04 09:01:58 <CIA-98> DiabloMiner: Changed to 15 sec avg/forever double meter, further decreased run size
 672 2011-02-04 09:01:58 <CIA-98> DiabloMiner: floor, allow very maximum nonce before saturation flush - http://bit.ly/g9rYvD
 673 2011-02-04 09:06:01 MattJD has quit (Ping timeout: 250 seconds)
 674 2011-02-04 09:06:31 <genjix> im sending my laptop in for servicing... im paranoid of them setting up a rootkit or something
 675 2011-02-04 09:06:47 <genjix> should i hash my entire hardrive to check for changes before and after?
 676 2011-02-04 09:06:49 <Diablo-D3> genjix: how do you know they didnt already?
 677 2011-02-04 09:06:51 <genjix> how can i do that?
 678 2011-02-04 09:07:03 <genjix> i haven't sent it yet... i will do it.
 679 2011-02-04 09:07:06 <Diablo-D3> and no, you should just make a duplicate of your entire hd
 680 2011-02-04 09:07:31 <Diablo-D3> since they could accidently destroy your original drive
 681 2011-02-04 09:07:58 <genjix> don't think so... it's just a simple fix (solder new power jack)
 682 2011-02-04 09:08:19 <Diablo-D3> yeah but you know how stupid techs are
 683 2011-02-04 09:08:35 <Diablo-D3> copy the entire drive and remove any sensitive data from the drive first.
 684 2011-02-04 09:08:56 <genjix> ok thanks
 685 2011-02-04 09:08:57 MJD has joined
 686 2011-02-04 09:09:27 <genjix> ill copy sentitive info, delete them then reinstall once it's back
 687 2011-02-04 09:09:53 <xelister> hey genjix I have a solution for you, want to hear it?
 688 2011-02-04 09:10:23 <genjix> tell tell
 689 2011-02-04 09:11:59 <Diablo-D3> heh good going slush
 690 2011-02-04 09:12:21 <genjix> xelister: what for? :)
 691 2011-02-04 09:12:35 <Diablo-D3> oh right I still have him on ignore
 692 2011-02-04 09:13:01 <Diablo-D3> xelister: you may now speak
 693 2011-02-04 09:15:19 <xelister> genjix: ok
 694 2011-02-04 09:15:21 <xelister> They're climbin in your windows, They're snatchin your income tax
 695 2011-02-04 09:15:27 <xelister> tryna enslave you so y'all need to
 696 2011-02-04 09:15:45 <xelister> 1) hide your kids, hide your wife, buy a gun , buy a knife
 697 2011-02-04 09:15:47 <xelister> 2) watch my show, buy your gold, and buy your silver cuz they enslavin errbody out here
 698 2011-02-04 09:15:48 <xelister> 3) We don't have to look for no facts, we know this is true ...
 699 2011-02-04 09:15:50 <xelister> 4) they gon chip you, they gon tag you
 700 2011-02-04 09:16:02 * xelister but we can run and yell that, run and yell that, run and yell that, New World, New world order!
 701 2011-02-04 09:16:03 <Diablo-D3> zomg xel has turned black
 702 2011-02-04 09:16:07 <genjix> ahh you think i'm overly paranoid
 703 2011-02-04 09:16:13 <xelister> :>
 704 2011-02-04 09:16:35 <genjix> except i read on 4chan a techie posting that they copy everybody's harddrives and read all their stuff
 705 2011-02-04 09:16:51 <genjix> and tbh it seems like something i'd do
 706 2011-02-04 09:17:03 <xelister> lolz..
 707 2011-02-04 09:17:04 <xelister> :)
 708 2011-02-04 09:17:10 <genjix> out of curiousity, not malice
 709 2011-02-04 09:17:27 <genjix> (look for nudey pics, gossip, email passwords .etc)
 710 2011-02-04 09:17:46 <xelister> ok and seriously though.  Right cleaning hdd before is fine I guess. Also you could want to have laptop fixed in front of you, depending on type of shop-service and so on
 711 2011-02-04 09:17:53 <genjix> esp if you're a bored techie sitting doing nothing for hours
 712 2011-02-04 09:18:09 <xelister> "out of curiousity, not malice (look for nudey pics, gossip, email passwords .etc)" .... lololol
 713 2011-02-04 09:18:14 <genjix> kk, i'll do that too
 714 2011-02-04 09:18:54 <xelister> then what is malice, installing some plutonium there to give the owner a cancer? ;)
 715 2011-02-04 09:19:44 <genjix> stealing their credit cards, uploading child porn
 716 2011-02-04 09:20:33 <genjix> those actually cause real harm (loss of fundz, prison time)
 717 2011-02-04 09:20:34 <Diablo-D3> xelister: wernt you asking for a double meter anyhow?
 718 2011-02-04 09:20:34 <xelister> a techie "just" browsing through my hdd without my conesnt would find himself quickly missing left right, and would hight his left, arm, in his anus area
 719 2011-02-04 09:20:55 <xelister> woah typos. need coffey
 720 2011-02-04 09:21:02 <xelister> a techie "just" browsing through my hdd without my conesnt would find himself quickly missing left, and would find his right, arm, in his anus area
 721 2011-02-04 09:21:10 <genjix> everything else is just "oh noes my reputation is damaged" or "oh noes my secret plans are exposed"
 722 2011-02-04 09:21:14 <Diablo-D3> typos continue.
 723 2011-02-04 09:21:16 <genjix> i.e nobody cares
 724 2011-02-04 09:21:45 <xelister> speaking of reputation ruined
 725 2011-02-04 09:22:03 <xelister> anyone knows if Ati/AMD plans to release a non-shit drivers for 5xxx in near future?
 726 2011-02-04 09:22:28 <genjix> xelister: expect big changes in near future
 727 2011-02-04 09:22:34 <genjix> X is being dumped now
 728 2011-02-04 09:22:45 <genjix> new minimal interface is coming in the kernel :)
 729 2011-02-04 09:22:46 <xelister> like, ones that do not crash in 75% of cases of switching virtual displays / terminals while using 2 video outputs (2 X desktops) on both 5970 and 5770 (wtf ati)
 730 2011-02-04 09:23:01 <genjix> it'll be sweet. performance++
 731 2011-02-04 09:23:08 <xelister> ot that do not 100% of the case show garbled mouse cursor on 2nd display
 732 2011-02-04 09:23:13 <genjix> at last we'll see input redirection
 733 2011-02-04 09:23:15 <xelister> and that do not hang each 3-5 days
 734 2011-02-04 09:23:27 <xelister> genjix: hm??
 735 2011-02-04 09:23:38 <xelister> Diablo-D3: what double meter?
 736 2011-02-04 09:23:42 jav has joined
 737 2011-02-04 09:23:48 <Diablo-D3> xelister: short avg/forever meter
 738 2011-02-04 09:23:51 <genjix> xelister: X is being deprecated as a project
 739 2011-02-04 09:23:54 <xelister> ah yea
 740 2011-02-04 09:23:58 <genjix> *has been
 741 2011-02-04 09:24:18 <xelister> well I heavly modified it anyway to have various stats too.. is there anything very important to merge sine 2010.11's versions?
 742 2011-02-04 09:24:28 <xelister> like important security / stability / performance (>1%) fix?
 743 2011-02-04 09:24:30 <Diablo-D3> xelister: yes
 744 2011-02-04 09:24:38 <Diablo-D3> xelister: I added art's kernel, for one
 745 2011-02-04 09:24:43 <Diablo-D3> 3% increase
 746 2011-02-04 09:24:46 <xelister> oh
 747 2011-02-04 09:25:02 <Diablo-D3> and I made it more slush-proof
 748 2011-02-04 09:25:13 <xelister> what was needed for slush?
 749 2011-02-04 09:25:21 <Diablo-D3> nothing much
 750 2011-02-04 09:25:27 <Diablo-D3> he just managed to often crash my miner
 751 2011-02-04 09:25:39 <Diablo-D3> since he fed it shit
 752 2011-02-04 09:25:59 <xelister> how often? mine doesnt ;)   unless you mean occassionall cant allocate queue but that is probably due to desktop/X usage at same time as miner runs
 753 2011-02-04 09:26:37 <Diablo-D3> basically whenever he took the pool down
 754 2011-02-04 09:26:43 <Diablo-D3> it wouldnt handle it correctly in rare cases
 755 2011-02-04 09:26:57 <xelister> ah
 756 2011-02-04 09:27:08 <Diablo-D3> oh, and I added optional compressed traffic stuff on his request
 757 2011-02-04 09:27:12 <Diablo-D3> so it doesnt eat all his bandwidth
 758 2011-02-04 09:27:39 <xelister> genjix: hmm... I don't get how X is deprecated... anyway, radeons 5xxx are NOT usable on linux for desktop it seems, crashes x10 more often then nvidia, literaly
 759 2011-02-04 09:28:32 <Diablo-D3> radeons are fine on linux
 760 2011-02-04 09:28:41 <Diablo-D3> almost everyone here owns one
 761 2011-02-04 09:28:43 <Diablo-D3> and it works fine
 762 2011-02-04 09:28:45 <xelister> miner <---> bitcoind/pool is basically just asking for sha256 for base of current calculation and thats all?
 763 2011-02-04 09:29:00 <Diablo-D3> xelister: yes, it just does getwork
 764 2011-02-04 09:29:18 <jav> quick question: in a transaction the total amount of bitcoins of all IN addresses are moved to the OUT addresses, right? .. and everything that isn't allocated to a new OUT address is considered a fee? is that correct?
 765 2011-02-04 09:29:30 <xelister> Diablo-D3: but do you use radeon on desktop?  try switching X alt-ctrl-del -f6- f7 -f8 etc few times forth and back. do it 10 times - crash,  do it 2 times with dualdisaply - guaranteed crash
 766 2011-02-04 09:29:37 <xelister> erm. no -del
 767 2011-02-04 09:31:04 <Diablo-D3> xelister: it'll onl do that if KMS is on
 768 2011-02-04 09:31:10 <ArtForz> fglrx + fbcon + kms = bad idea
 769 2011-02-04 09:31:23 <Diablo-D3> fbcon alone doesnt do it from what I can tell
 770 2011-02-04 09:31:24 <Diablo-D3> just kms
 771 2011-02-04 09:32:10 <xelister> how to check if kms is active for me?
 772 2011-02-04 09:32:18 <Diablo-D3> if you didnt turn it off, its active
 773 2011-02-04 09:32:31 <xelister> perhaps ubuntu did or something... is it an boot option?
 774 2011-02-04 09:33:07 slush has joined
 775 2011-02-04 09:33:14 TheAncientGoat has quit (Ping timeout: 276 seconds)
 776 2011-02-04 09:33:42 <Diablo-D3> yes its a boot option, and no ubuntu hasnt disabled it
 777 2011-02-04 09:34:25 <xelister> btw, I hear someone has 200 USD/card radeon's 5970 ;)  you think this could be possible if the cards would come from factory that perhaps produced a bit more then they agreed with Ati overloard or that coworker borrowed some chips home or something?
 778 2011-02-04 09:34:33 <Diablo-D3> iirc its radeon.modeset=0
 779 2011-02-04 09:34:55 joe_1 has joined
 780 2011-02-04 09:35:21 <Diablo-D3> oh wait, its just nomodeset
 781 2011-02-04 09:36:24 <joe_1> when a bitcoin client receives a pair of double-spend transactions A then B, it refuses the B transaction no matter how many more blocks B has over A?
 782 2011-02-04 09:37:50 <edcba> transactions don't have blocks
 783 2011-02-04 09:38:04 <edcba> blocks have transactions
 784 2011-02-04 09:38:23 <joe_1> yeah but for any transaction i can say how many blocks of confirmation it has. it belongs to a block, and that block is X blocks deep.
 785 2011-02-04 09:38:25 <edcba> while not acknowledged transactions aren't tied to any block
 786 2011-02-04 09:38:57 <edcba> you can't have same transaction in 2 different blocks in same chain
 787 2011-02-04 09:40:15 <xelister> Kernel mode-setting (KMS)  [...]  promises to make booting faster, more graphical, and less flickery.
 788 2011-02-04 09:40:22 reQunix has joined
 789 2011-02-04 09:40:31 <xelister> ^---- wtf more graphical boot? sound's like "I'm an iFag" to me ;)
 790 2011-02-04 09:40:46 <Diablo-D3> xelister: no
 791 2011-02-04 09:40:54 <Diablo-D3> also, you do realize its ALREADY graphical, right?
 792 2011-02-04 09:40:56 <Diablo-D3> its fbcon
 793 2011-02-04 09:41:00 <xelister> I assumed KMS will be a good idea (one day?) because it makes the linux guys takes care of mode setting instead of nvidia/radeon noobs
 794 2011-02-04 09:41:08 <Diablo-D3> you get little tux in corner of fbcon
 795 2011-02-04 09:41:10 <Diablo-D3> ergo graphical
 796 2011-02-04 09:41:20 <Diablo-D3> xelister: KMS is literally what it says
 797 2011-02-04 09:41:27 <Diablo-D3> the kernel inits the hardware in full
 798 2011-02-04 09:41:35 <Diablo-D3> no more X initing hardware
 799 2011-02-04 09:41:48 <Diablo-D3> which also will lead to no more X as root
 800 2011-02-04 09:41:49 <xelister> Diablo-D3: when I seen ubuntu's faggotry of showing me grahpical mode bootup console (aka "scrolling black-white text thing") I switched it off..  actually, just changing to propertiary driver changes that It seemed
 801 2011-02-04 09:41:56 <Diablo-D3> ubuntu is not linux
 802 2011-02-04 09:42:02 <Diablo-D3> please stop spreading that myth
 803 2011-02-04 09:42:04 <joe_1> it's not?
 804 2011-02-04 09:42:10 <Diablo-D3> no, its a microsoft product
 805 2011-02-04 09:42:16 <Diablo-D3> it has nothing to do with linux
 806 2011-02-04 09:42:25 <Diablo-D3> no more fud in #bitcoin-dev plzkthx
 807 2011-02-04 09:42:40 <xelister> you are overdoing it ;)
 808 2011-02-04 09:43:03 <xelister> we know it sucks and all, but still better then people running windows
 809 2011-02-04 09:43:23 <Diablo-D3> dude
 810 2011-02-04 09:43:24 <joe_1> ok if ubuntu is gay then what linux distribution is real linux
 811 2011-02-04 09:43:27 <Diablo-D3> if I wanted to have a graphical boot
 812 2011-02-04 09:43:29 <Diablo-D3> I'd run osx
 813 2011-02-04 09:43:34 <Diablo-D3> joe_1: debian
 814 2011-02-04 09:43:40 <Diablo-D3> although in the next year or so Im switching to alpine
 815 2011-02-04 09:43:50 <xelister> wtf is alpine
 816 2011-02-04 09:43:54 <xelister> joe_1: gentoo?
 817 2011-02-04 09:44:30 <Diablo-D3> xelister: a new linux distro that is meant to be easy to use, fast, bloat free, and 9000% less faggotry than ubuntu
 818 2011-02-04 09:45:11 <ArtForz> well, making a distro better than "let's take sid and randomly break crap" isn't exactly rocket surgery
 819 2011-02-04 09:45:31 <xelister> what is appealing in ubuntu
 820 2011-02-04 09:45:40 <xelister> other then graphicall faggotry (sadly)
 821 2011-02-04 09:45:56 <xelister> is that so much stuff Just Works,  like lots of firmware/drivers/auto install/auto detect/auto whatever
 822 2011-02-04 09:46:05 <Diablo-D3> xelister: ITS EXTRA AFRICAN
 823 2011-02-04 09:46:18 <Diablo-D3> there. I said it. Ubuntu is the distro for black people.
 824 2011-02-04 09:46:23 <xelister> :O
 825 2011-02-04 09:46:24 <Diablo-D3> Linux is finally racist.
 826 2011-02-04 09:46:26 <ArtForz> errr... the same shit generally works on sid once you enable the nonfree repo
 827 2011-02-04 09:46:26 <Diablo-D3> I hope you're happy.
 828 2011-02-04 09:46:39 <xelister> ArtForz: perhaps
 829 2011-02-04 09:46:45 <joe_1> i felt like a thousand pound weight was lifted off my chest. i knew it was african
 830 2011-02-04 09:46:48 <joe_1> that drum beat when it loads
 831 2011-02-04 09:46:58 <xelister> ArtForz: debian pissed me off long ago, when its sofware was 100 years old behind upstream
 832 2011-02-04 09:47:00 <ArtForz> and sid at least doesn't ship a fucking RELEASE with alpha-level fglrx
 833 2011-02-04 09:47:01 <xelister> (actually 1.5 year old)
 834 2011-02-04 09:47:25 <xelister> do they /finally/ improved on that?
 835 2011-02-04 09:47:51 <ArtForz> ?
 836 2011-02-04 09:48:09 poseidon_ has joined
 837 2011-02-04 09:48:14 <ArtForz> if you run stable, no
 838 2011-02-04 09:48:15 <xelister> in like 2007, some libs like afair.. libboost? where 1 .. 1.5 years older then the upstream version was
 839 2011-02-04 09:48:22 <xelister> 1.5 years is *really* old
 840 2011-02-04 09:48:27 <ArtForz> because stable is meant to be... you know. stable.
 841 2011-02-04 09:49:13 <xelister> but waiting till say 2013 to use things the developers made (e.g. on my requests/bugreports) today, or that I read about today, is meh
 842 2011-02-04 09:49:16 <ArtForz> if you want current, use unstable aka sid
 843 2011-02-04 09:49:26 <Diablo-D3> also, you do realize libboost is matched to the version of gcc, right?
 844 2011-02-04 09:49:29 draginx has quit (Quit: Leaving.)
 845 2011-02-04 09:49:34 <Diablo-D3> its not recommended to mismatch libboost and gcc
 846 2011-02-04 09:49:41 <Diablo-D3> random shit goes bad
 847 2011-02-04 09:49:45 <Diablo-D3> because libboost coders suck dick
 848 2011-02-04 09:50:10 <xelister> well then ship also not superold version of gcc
 849 2011-02-04 09:50:13 <ArtForz> not to mention debian stable is pretty much what other distros would consider a LTS release
 850 2011-02-04 09:50:16 <xelister> so you tell me now this is possible on SID?
 851 2011-02-04 09:50:22 <xelister> My debian experience was like
 852 2011-02-04 09:50:49 <Diablo-D3> xelister: you are coming very close to me putting you back on ignore
 853 2011-02-04 09:51:32 <xelister> wow, linux ^_^ ... wait,  oh shit thats old   -- "update to SID, dude" ---> ok. yey new versions ^_^ ... wait, everything broken on update, cant reinstall or unsintall or upgrade now X-( what to do? --- "we told you to not use SID, fuck you" --> fuuuuuuuuuuu
 854 2011-02-04 09:51:47 <sipa> haha
 855 2011-02-04 09:52:01 <Diablo-D3> And there he goes.
 856 2011-02-04 09:52:22 <ArtForz> yes, lets use ubuntu, where upgrading from one *release* to the next is pretty much guaranteed to cause even more issues
 857 2011-02-04 09:52:25 <xelister> adn I didnt tried since, so is it nowdays possible to have more cutting endge Debian, but that will overall not break horribly?
 858 2011-02-04 09:52:55 <ArtForz> and btw, I didnt have ONE SINGLE PROBLEM with sid since the ia32-libs fiasco
 859 2011-02-04 09:53:02 <edcba> try it and don't annoy us with that xelister ;)
 860 2011-02-04 09:53:26 <xelister> hm perahaps perhaps
 861 2011-02-04 09:54:39 <xelister> so again, which version of debian makes sense... so I have rather recent versions, but also apt will not get into unusable state where basically only way to fix it, without dicking all day with it, is to reinstall
 862 2011-02-04 09:55:35 <edcba> just ask in #debian
 863 2011-02-04 09:57:23 <ArtForz> so, what version of ubuntu should I use that doesnt have broken sound support, NetworkMangler and fglrx drivers that arent like 2 years old or ubuntu-exclusive preview aka alpha releases?
 864 2011-02-04 09:57:30 Teppy1 has quit (Quit: Leaving.)
 865 2011-02-04 09:57:35 Teppy has joined
 866 2011-02-04 09:57:53 <joe_1> Alpine Linux began as a fork of the LEAF Project
 867 2011-02-04 09:58:25 <Diablo-D3> joe_1: yeah, but thats a pretty bad way of describing it
 868 2011-02-04 09:58:34 <Diablo-D3> it really is a debian-like distro without the faggotry
 869 2011-02-04 09:58:42 <Diablo-D3> like, they got rid of glibc
 870 2011-02-04 09:58:49 <Diablo-D3> no more libc bloat fest
 871 2011-02-04 09:58:54 <xelister> ArtForz: Im not saying that ubuntu is better ;)
 872 2011-02-04 09:59:03 <xelister> it has different kind of problems
 873 2011-02-04 09:59:07 Teppy has quit (Read error: No route to host)
 874 2011-02-04 09:59:11 Teppy has joined
 875 2011-02-04 09:59:28 <joe_1> whats wrong with libc, don't you need that for.. everything?
 876 2011-02-04 09:59:30 <xelister> hm... perhaps source based is the answer. Anyway then we can have cutting-edge & working-with-upstream-developers versions (more naturally)
 877 2011-02-04 09:59:59 <ArtForz> aka gentoo
 878 2011-02-04 10:00:26 <ArtForz> aka lets recompile everything because libc changed again
 879 2011-02-04 10:00:26 <Diablo-D3> joe_1: yes, but they swapped it with uclibc
 880 2011-02-04 10:00:43 <Diablo-D3> ArtForz: yes, it has the unfortunate side effect of breaking existing binaries
 881 2011-02-04 10:00:46 <Diablo-D3> they dont care.
 882 2011-02-04 10:00:51 <Diablo-D3> and honestly, I dont care either
 883 2011-02-04 10:00:56 <Diablo-D3> its all legacy software
 884 2011-02-04 10:01:14 <Diablo-D3> most linux devices dont even use glibc
 885 2011-02-04 10:01:20 <Diablo-D3> its too bloated and slow
 886 2011-02-04 10:03:22 <xelister> ArtForz: then what you use?
 887 2011-02-04 10:03:45 <ArtForz> sid on my main workstation/dev box
 888 2011-02-04 10:04:07 <ArtForz> stable+security on my servers
 889 2011-02-04 10:04:09 <Diablo-D3> ArtForz: and no, its not like gentoo
 890 2011-02-04 10:04:13 <Diablo-D3> the libc doesnt change all the time
 891 2011-02-04 10:04:15 <Diablo-D3> its a one time change
 892 2011-02-04 10:04:24 <Diablo-D3> its no different than how everyone had to deal with libc5->6 all those years ago
 893 2011-02-04 10:05:11 <joe_1> i thought binaries don't change unless the headers change. why would the c standard library headers change. i can understand implementation changes
 894 2011-02-04 10:05:17 <ArtForz> my notebook is also sid
 895 2011-02-04 10:05:18 <genjix> ubuntu switched because the owner of libc is an idiot
 896 2011-02-04 10:05:27 <xelister> :)
 897 2011-02-04 10:05:39 <Diablo-D3> genjix: no
 898 2011-02-04 10:05:47 <Diablo-D3> ubuntu switched to eglibc because debian did
 899 2011-02-04 10:05:51 <Diablo-D3> its STILL glibc
 900 2011-02-04 10:06:02 <genjix> oh right
 901 2011-02-04 10:06:14 <genjix> but it's a different project (different maintainer) afaik
 902 2011-02-04 10:06:58 <xelister> ArtForz: but can you even get the dependencies to build bitcoind on stable, and bitcoin gui on sid
 903 2011-02-04 10:07:06 <ArtForz> yes
 904 2011-02-04 10:07:24 <ArtForz> well, you still have to compile the custom wx for gui
 905 2011-02-04 10:07:25 <xelister> was it not a problem back in 2010 end
 906 2011-02-04 10:07:33 <xelister> yeah that is what I ment, wx 2.9
 907 2011-02-04 10:07:54 <Diablo-D3> different maintainer, essentially same project
 908 2011-02-04 10:07:58 <Diablo-D3> its not a perminent fork
 909 2011-02-04 10:07:59 <xelister> ubuntu did shiped 2.9
 910 2011-02-04 10:08:06 <Diablo-D3> they constantly update from glibc upstream
 911 2011-02-04 10:08:27 <ArtForz> except iirc ubuntus 2.9 is configured differently than what bitcoin expects
 912 2011-02-04 10:08:53 <xelister> ArtForz: bitcoin gui does build on current ubuntu with no manuall building on dependencies
 913 2011-02-04 10:08:56 <genjix> tbh ubuntu is still the best linux around
 914 2011-02-04 10:08:58 <xelister> *of
 915 2011-02-04 10:08:59 <genjix> linux done right
 916 2011-02-04 10:09:06 * xelister gets popcorn
 917 2011-02-04 10:09:27 <genjix> gentoo is pita
 918 2011-02-04 10:09:46 <genjix> arch is wip
 919 2011-02-04 10:09:50 <Diablo-D3> ArtForz: ehehehe on -f 1000 Im getting 75.6 long count
 920 2011-02-04 10:10:01 <genjix> fedora is 2nd rate
 921 2011-02-04 10:10:06 <xelister> ArtForz: this is some advantage. Building X to build Y to build Z to finally build the project you want to co-develop / cut-edge, is not so good sometimes
 922 2011-02-04 10:10:24 <joe_1> that's where windows comes in
 923 2011-02-04 10:10:37 <xelister> overall I like building and all but not when I have no time to play around
 924 2011-02-04 10:10:51 <genjix> lol u like building program
 925 2011-02-04 10:10:58 <genjix> strange person
 926 2011-02-04 10:11:10 <xelister> actually, on what satoshi develops that he used wx 2.9 ... but perhaps 2.9 was needed to correctly handle unicode?
 927 2011-02-04 10:11:18 <Diablo-D3> ArtForz: I used to have to -f 1 to get that
 928 2011-02-04 10:11:22 <genjix> weee look at that incomprehensible text scrolling quickly past...
 929 2011-02-04 10:11:30 <Diablo-D3> ArtForz: which leads me to believe, on my box, I lose 3% by keeping interactivity
 930 2011-02-04 10:12:26 <Diablo-D3> which, meh, fine
 931 2011-02-04 10:12:43 <Diablo-D3> the floor is now so low after the newest modifications I think I'll be able to game without shutting the miner off
 932 2011-02-04 10:12:57 Zarutian has joined
 933 2011-02-04 10:12:58 <slush> ;;bc,stats
 934 2011-02-04 10:13:00 <gribble> Current Blocks: 106131 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 716 blocks | Next Difficulty In About: 4 days, 9 hours, 12 minutes, and 44 seconds | Next Difficulty Estimate: 24615.15663506
 935 2011-02-04 10:13:27 <ArtForz> also, debian is at least mostly cross-platform
 936 2011-02-04 10:14:07 <da2ce7> ;;lastseen noagendamarket
 937 2011-02-04 10:14:07 <gribble> Error: "lastseen" is not a valid command.
 938 2011-02-04 10:14:11 <da2ce7> ;;last,seen noagendamarket
 939 2011-02-04 10:14:12 <gribble> Error: "last,seen" is not a valid command.
 940 2011-02-04 10:16:17 <genjix> ;;seen noagendamarket
 941 2011-02-04 10:16:17 <gribble> noagendamarket was last seen in #bitcoin-dev 19 hours, 37 minutes, and 19 seconds ago: * noagendamarket is on the nod
 942 2011-02-04 10:21:27 <joe_1> what do i need to do to start compiling bitcoin under mingw
 943 2011-02-04 10:22:09 <genjix> boost mainly
 944 2011-02-04 10:22:17 <genjix> libdb (i think)
 945 2011-02-04 10:22:37 <genjix> best look in makefiles or do a test run on linux
 946 2011-02-04 10:22:44 <genjix> make -f makefile.unix bitcoind
 947 2011-02-04 10:23:41 <joe_1> oh ok thanks
 948 2011-02-04 10:23:46 <joe_1> found makefile.mingw
 949 2011-02-04 10:24:37 nevezen has joined
 950 2011-02-04 10:34:05 m0mchil has joined
 951 2011-02-04 10:36:39 <m0mchil> OneFixt, ArtForz: just to clear this out - there is no loss whatsoever returning just one H per kernel run
 952 2011-02-04 10:36:55 <m0mchil> regardless of using pool or not
 953 2011-02-04 10:37:34 <ArtForz> yes there is
 954 2011-02-04 10:37:40 <Diablo-D3> m0mchil: maximum loss would be ~3% on -f 1 with a 5970
 955 2011-02-04 10:37:48 <Diablo-D3> because you're going to hit H==0 conflicts
 956 2011-02-04 10:37:49 <ArtForz> actually 5870 is worse
 957 2011-02-04 10:37:56 <Diablo-D3> and it'll only really matter on the pool
 958 2011-02-04 10:38:01 <m0mchil> 5970 - 4 GPUs combined or?
 959 2011-02-04 10:38:34 <ArtForz> you run enough globalworkitems to hit 1 sec per clenqueuendrangekernel, right?
 960 2011-02-04 10:38:43 <m0mchil> right
 961 2011-02-04 10:38:57 <ArtForz> and you only return one H==0 result max from that
 962 2011-02-04 10:39:04 <m0mchil> go on
 963 2011-02-04 10:39:06 <ArtForz> a 5870 can get like 350Mh/s
 964 2011-02-04 10:39:17 <ArtForz> = 350M nonces tested in one call
 965 2011-02-04 10:39:50 <Diablo-D3> yeah but 2**32 is 4294 million
 966 2011-02-04 10:39:58 <ArtForz> you lose about 3.5% of H==0 solutions due to runs containing 2 or more H==0 nonces
 967 2011-02-04 10:40:25 <m0mchil> but there are runs with no Hs in them, right?
 968 2011-02-04 10:40:49 <Diablo-D3> yes
 969 2011-02-04 10:41:01 <m0mchil> what do you lose of empty runs?
 970 2011-02-04 10:41:04 <Diablo-D3> I bet 3.5% of those runs would have no H==0
 971 2011-02-04 10:41:28 <Diablo-D3> it seems like it should align that way
 972 2011-02-04 10:41:49 x6763 has quit (Ping timeout: 246 seconds)
 973 2011-02-04 10:43:19 x6763 has joined
 974 2011-02-04 10:43:40 <ArtForz> ?
 975 2011-02-04 10:43:50 <ArtForz> you dont lose anything of empty runs, tzhey're still empty
 976 2011-02-04 10:44:00 <Diablo-D3> ArtForz: I think he means, how many runs turn up empty
 977 2011-02-04 10:44:08 <ArtForz> that doesn't change
 978 2011-02-04 10:44:24 <Diablo-D3> then... how many runs turn up empty?
 979 2011-02-04 10:44:42 <m0mchil> let me do some statistics here... moment
 980 2011-02-04 10:44:44 <ArtForz> about 88% of runs iirc
 981 2011-02-04 10:45:40 <ArtForz> nm, 91.6%
 982 2011-02-04 10:47:41 <ArtForz> of the remaining 8.whatever% of total runs that have at least one H==0, something like 96.6% have exactly one H==0
 983 2011-02-04 10:48:40 <ArtForz> = total you only have like 0.28% runs with 2 or more H==0, which doesnt look like much
 984 2011-02-04 10:49:24 <ArtForz> but *of the runs that do have a H==0* it's something like 3.4%
 985 2011-02-04 10:50:54 <Diablo-D3> ahh so
 986 2011-02-04 10:50:55 <Diablo-D3> it
 987 2011-02-04 10:50:56 <Diablo-D3> doesnt matter
 988 2011-02-04 10:51:00 <ArtForz> ?
 989 2011-02-04 10:51:21 <ArtForz> = you *ignore* like 3.5% of H==0 you'd normally find
 990 2011-02-04 10:51:34 <ArtForz> thats quite significant
 991 2011-02-04 10:51:38 <m0mchil> 3.4% of 0.28% = 0.009%?
 992 2011-02-04 10:51:52 <Diablo-D3> m0mchil: no
 993 2011-02-04 10:52:00 <Diablo-D3> 0.28% of ALL runs, 3.4% of runs with at least H==0
 994 2011-02-04 10:52:34 <ArtForz> 3.4% of what would be your H==0 count if you didnt overwrite a bunch of em
 995 2011-02-04 10:52:55 <ArtForz> = you're cutting your effective hashrate by those same 3.whatever%
 996 2011-02-04 10:52:59 <ArtForz> less with slower cards of course
 997 2011-02-04 10:53:42 <ArtForz> as those do less trys/run = lower chances of >1 H==0 in a run = less lost H==0s
 998 2011-02-04 10:54:34 <m0mchil> ok, let me check something... brb
 999 2011-02-04 10:54:40 <ArtForz> so it is a pretty serious issue for pool clients
1000 2011-02-04 10:56:14 <ArtForz> Diablo-D3: shouldnt your miner show the same behaviour at -f 1 ?
1001 2011-02-04 10:57:41 sabalaba has quit (Read error: Operation timed out)
1002 2011-02-04 10:58:59 <sipa> couldn't you use multiple output buffers for different kernels running in parallel, to decrease the chance of overwritten found H==0's?
1003 2011-02-04 10:59:15 <ArtForz> thats what my original kernel was doing...
1004 2011-02-04 10:59:21 <sipa> i see
1005 2011-02-04 10:59:23 <ArtForz> used one output dword per 1024 tries
1006 2011-02-04 10:59:53 <ArtForz> as the chance of two H==0 solutions in 1024 tries is pretty damn small
1007 2011-02-04 11:00:01 <slush> can anybody shortly tell me what are you talking about, guys? I see 'pool' ,'serious issue', 'cutting hashrate', which makes me little nervous
1008 2011-02-04 11:00:42 <ArtForz> OneFixt found a problem with how m0s mienr is handing back "good" nonces to host
1009 2011-02-04 11:01:04 <ArtForz> it can only hand back one solution per kernel run, which is about 1 second worth of tries
1010 2011-02-04 11:01:26 <m0mchil> well, diablo's too
1011 2011-02-04 11:01:32 <ArtForz> so if you have a fast card, whenever it finds two or more sultions in one second = one run, it just ignores one
1012 2011-02-04 11:01:36 <ArtForz> diablos with -f 1
1013 2011-02-04 11:02:28 <sipa> seems not that hard to fix
1014 2011-02-04 11:02:50 <ArtForz> with default -f 60 diablos effectively uses 1/60th your run size = a LOT less chance of 2+ H==0 in one run
1015 2011-02-04 11:02:56 echelon has quit (Ping timeout: 240 seconds)
1016 2011-02-04 11:03:24 <tcatm> code to check for nonces within a range: http://pastebin.com/bgh0x6Zi
1017 2011-02-04 11:03:29 reQunix has quit (Quit: Computer has gone to sleep.)
1018 2011-02-04 11:04:02 <m0mchil> mine's -f 30, until recently it was -f 60
1019 2011-02-04 11:04:04 <ArtForz> doesnt help
1020 2011-02-04 11:04:14 <ArtForz> *checks*
1021 2011-02-04 11:06:37 <ArtForz> hmmm... yeah
1022 2011-02-04 11:06:52 <ArtForz> still a problem at low -f though
1023 2011-02-04 11:07:49 <ArtForz> whats the point of running low -f if you get a few % higher hashrate but start throwing away a few % valid solutions?
1024 2011-02-04 11:08:20 <m0mchil> what's the expectation of 340 MHash @ x1 difficulty?
1025 2011-02-04 11:08:50 devon_hillard_ has joined
1026 2011-02-04 11:09:26 <lfm> ;;bc,calcd 340000 1
1027 2011-02-04 11:09:27 <gribble> The average time to generate a block at 340000 Khps, given the supplied difficulty of 1, is 12 seconds
1028 2011-02-04 11:09:34 <ArtForz> mean of 0.0791624 H==0 per run
1029 2011-02-04 11:09:36 <m0mchil> thanks
1030 2011-02-04 11:10:03 <sipa> so, let's find the optimum
1031 2011-02-04 11:10:04 <ArtForz> 12.63 sec or so
1032 2011-02-04 11:10:53 <ArtForz> thanks to -f 1 lossage it should be above 13 sec
1033 2011-02-04 11:11:19 <m0mchil> ok, again, why it is a problem only against pool?
1034 2011-02-04 11:11:35 <tcatm> http://pastebin.com/ar8jGTtG my miner that doesn't have the problem and *should* work with art's kernel. It just doesn't suppport getwork
1035 2011-02-04 11:12:01 <ArtForz> oh wait, you dont check G in kernel anymore
1036 2011-02-04 11:12:11 <ArtForz> so problem should appear always
1037 2011-02-04 11:12:34 devon_hillard has quit (Ping timeout: 264 seconds)
1038 2011-02-04 11:13:22 <ArtForz> well, most obvious with high hashrate per GPU and really low -f
1039 2011-02-04 11:13:33 <slush> is
1040 2011-02-04 11:13:46 <slush> is 'really low -f' also 5 for 1.1Ghash miner?
1041 2011-02-04 11:13:54 <ArtForz> 1.1GH for a single card?
1042 2011-02-04 11:14:01 <slush> 2x5970
1043 2011-02-04 11:14:03 <slush> so 4 gpus
1044 2011-02-04 11:14:05 <ArtForz> thats 4 GPUs
1045 2011-02-04 11:14:10 <Diablo-D3> really low -f is 1.
1046 2011-02-04 11:14:12 <slush> oh, this is related to gpu
1047 2011-02-04 11:14:51 <ArtForz> basically problem is, with the current kernels each gpu can only pass one H==0 solution per "job"
1048 2011-02-04 11:15:00 <ArtForz> and one job = 1/framerate seconds
1049 2011-02-04 11:15:14 <sipa> on one gpu
1050 2011-02-04 11:15:15 <ArtForz> = that GPUs hashpersec/framerate
1051 2011-02-04 11:16:29 <slush> oh, see
1052 2011-02-04 11:16:49 <slush> so bigger framerate solve this problem, right? Not clearly, but the probability is lower
1053 2011-02-04 11:16:54 <ArtForz> yes
1054 2011-02-04 11:17:00 <ArtForz> probability is a LOT lower
1055 2011-02-04 11:17:12 mtgox has joined
1056 2011-02-04 11:17:21 <sipa> at 340Mh/s and -f 1, i get 3.9% loss
1057 2011-02-04 11:17:25 <ArtForz> iirc reduced by more than the same factor you increase framerate
1058 2011-02-04 11:17:29 <sipa> correct, ArtForz?
1059 2011-02-04 11:17:39 <ArtForz> looks right
1060 2011-02-04 11:18:02 <sipa> at -f 5, that becomes 0.79%
1061 2011-02-04 11:18:19 <ArtForz> yep
1062 2011-02-04 11:18:32 <sipa> and -f 60, 0.066%
1063 2011-02-04 11:19:14 echelon has joined
1064 2011-02-04 11:19:53 <ArtForz> yup
1065 2011-02-04 11:20:40 <sipa> so, if you'd change the output buffer to let's say 256 entries, and have each kernel modify output[nonce & 0xFF], you effectively get only 0.015% loss at -f 1
1066 2011-02-04 11:20:52 <ArtForz> yup
1067 2011-02-04 11:21:23 <tcatm> you could even check nonce..nonce+0xFF in hostcode and have 0% loss at -f 1
1068 2011-02-04 11:21:23 <m0mchil> I am still not convinced... does someone tried a long run at diff x1? I checked with my results for last 14 hours to pool, seeing 10% worse than expected results
1069 2011-02-04 11:21:37 <ArtForz> I did a few simulations
1070 2011-02-04 11:21:39 <sipa> there is a lot of random variation still too
1071 2011-02-04 11:21:42 <m0mchil> but this could be because of network overhead
1072 2011-02-04 11:22:02 <slush> m0mchil: 10% is more than expected network overhead
1073 2011-02-04 11:22:06 <slush> _much_ more
1074 2011-02-04 11:22:06 <ArtForz> you need a huge sample size to get anything close to being conclusive for a few % difference
1075 2011-02-04 11:22:06 <sipa> network overhead shouldn't be more than a 2% loss
1076 2011-02-04 11:23:03 <m0mchil> well, it's only ~4000 samples... I'd like to see something like 100 000
1077 2011-02-04 11:24:01 <Diablo-D3> loss in what
1078 2011-02-04 11:24:02 <Diablo-D3> performance?
1079 2011-02-04 11:24:25 <ArtForz> H==0 results vs. time*hashrate
1080 2011-02-04 11:24:49 <sipa> Diablo-D3: loss of submitted shares
1081 2011-02-04 11:24:57 <ArtForz> it's pretty obvious when you simulate with REALLY low -f
1082 2011-02-04 11:25:16 <ArtForz> like, 0.09something
1083 2011-02-04 11:25:16 <sipa> submitted valid counted shares
1084 2011-02-04 11:27:32 <ArtForz> basically, what happens when you have a globalworksize of 2**32?
1085 2011-02-04 11:27:45 <ArtForz> you get a *mean* of 1 H==0 per run
1086 2011-02-04 11:28:07 <ArtForz> = you have quite a lot of runs with more than 1 H==0
1087 2011-02-04 11:29:06 <ArtForz> = you lose a lot of H==0 results because you just ignore any > 1 in a run containing multiple results
1088 2011-02-04 11:30:48 <sipa> 41.8% of them
1089 2011-02-04 11:32:02 <ArtForz> should be a binomial distribution, right?
1090 2011-02-04 11:32:35 <sipa> yes
1091 2011-02-04 11:32:41 <ArtForz> yep
1092 2011-02-04 11:33:36 <sipa> i calculate 1-(P(#(H==0) == 1)/(1-P(#(H==0) == 0)
1093 2011-02-04 11:33:41 <sipa> )
1094 2011-02-04 11:36:04 jav has quit (Remote host closed the connection)
1095 2011-02-04 11:39:23 TheAncientGoat has joined
1096 2011-02-04 11:40:22 reQunix has joined
1097 2011-02-04 11:43:48 reQunix has quit (Remote host closed the connection)
1098 2011-02-04 11:55:05 davout has joined
1099 2011-02-04 11:55:21 <davout> hey all
1100 2011-02-04 11:58:35 <tcatm> hey davout
1101 2011-02-04 11:58:44 <tcatm> you broke the trade history api
1102 2011-02-04 12:09:21 skeledrew has joined
1103 2011-02-04 12:10:01 skeledrew1 has quit (Ping timeout: 276 seconds)
1104 2011-02-04 12:10:46 <davout> tcatm: yea, i warned about it a while ago :D
1105 2011-02-04 12:10:54 <tcatm> fix it :P
1106 2011-02-04 12:10:58 <davout> you fix it !
1107 2011-02-04 12:11:04 <tcatm> I never received a warning
1108 2011-02-04 12:11:13 <davout> i did not "break" it, I improved it
1109 2011-02-04 12:11:22 <tcatm> how do I fix it?
1110 2011-02-04 12:11:25 <davout> yea, I mentioned it on the forums
1111 2011-02-04 12:11:34 <davout> true i didn't tell you explicitly :)
1112 2011-02-04 12:11:37 <davout> hmm
1113 2011-02-04 12:11:49 <davout> the paths have changed a little
1114 2011-02-04 12:12:03 <davout> view-source:https://bitcoin-central.net/trades.xml
1115 2011-02-04 12:12:10 <davout> or json for the matter
1116 2011-02-04 12:12:21 <tcatm> same format?
1117 2011-02-04 12:12:29 <davout> hmm
1118 2011-02-04 12:12:32 <davout> not sure
1119 2011-02-04 12:12:45 <davout> ticker is at the same place
1120 2011-02-04 12:12:55 <davout> bc.net/ticker.(xml|json)
1121 2011-02-04 12:13:08 <davout> lemme check the commit i'll tell you if the format changed
1122 2011-02-04 12:13:18 <tcatm> looks like it fetched new trades
1123 2011-02-04 12:15:03 <davout> looks like i broke the currency filter
1124 2011-02-04 12:15:08 <davout> were you using it ?
1125 2011-02-04 12:15:16 <tcatm> Not sure... :)
1126 2011-02-04 12:15:25 <davout> just a get parameter
1127 2011-02-04 12:15:28 <davout> brb food
1128 2011-02-04 12:17:43 <tcatm> I'm filtering currencies for tradehistory again so that's not a problem
1129 2011-02-04 12:27:52 <m0mchil> I guess poclbm should be fixed... what % loss is acceptable, is 0.25% ok (with output buffer of 32)?
1130 2011-02-04 12:27:59 <m0mchil> or to go with 256?
1131 2011-02-04 12:30:02 <sipa> what is the smallest worksize you use?
1132 2011-02-04 12:30:55 <m0mchil> how is worksize related? or you mean threads per run?
1133 2011-02-04 12:31:09 <tcatm> m0mchil: why don't you use one dword for every thread?
1134 2011-02-04 12:31:43 <sipa> thread?
1135 2011-02-04 12:31:55 <tcatm> gpu thread
1136 2011-02-04 12:32:01 <tcatm> e.g. globalthreadsize
1137 2011-02-04 12:32:12 <sipa> can't that be millions?
1138 2011-02-04 12:32:28 <m0mchil> it is indeed, with -f 1
1139 2011-02-04 12:32:41 <tcatm> is that a problem?
1140 2011-02-04 12:32:47 <m0mchil> 285 mln on single 5970 core
1141 2011-02-04 12:33:00 <sipa> = >1GiB of memory ...
1142 2011-02-04 12:33:08 <tcatm> reduce globalthreadsize
1143 2011-02-04 12:33:18 <sipa> yeah, by increasing -f :D
1144 2011-02-04 12:33:29 <tcatm> or increase the loopsize from 1024 to 8192
1145 2011-02-04 12:33:34 <m0mchil> sipa, back to your question - what is worksize?
1146 2011-02-04 12:33:53 <sipa> m0mchil: i meant global thread size
1147 2011-02-04 12:34:31 <m0mchil> ah, minimum in poclbm is 0x4000 (assuming minimum 'w' of 64 * 256)
1148 2011-02-04 12:34:46 <tcatm> btw: on a 5870 at 350Mhash/s with -f 1 it takes 1.4 MByte
1149 2011-02-04 12:35:04 <sipa> huh
1150 2011-02-04 12:35:20 <sipa> that's way less than 4 bytes per global thread
1151 2011-02-04 12:35:21 <tcatm> 350e6 hashes, / 1024 (kernel loop) * 4 byte
1152 2011-02-04 12:35:28 <sipa> oh, you're using the loop
1153 2011-02-04 12:35:29 <sipa> right
1154 2011-02-04 12:36:15 <tcatm> so lots of headroom for faster GPUs
1155 2011-02-04 12:36:22 <tcatm> or even smaller -f values
1156 2011-02-04 12:38:25 <sipa> m0mchil: i'd say use one output per 'unit'
1157 2011-02-04 12:38:33 <sipa> assuming that's a power of two
1158 2011-02-04 12:38:36 <andrew12> of course, this channel is full of people at 7:32
1159 2011-02-04 12:38:51 skeledrew1 has joined
1160 2011-02-04 12:39:29 <sipa> m0mchil: at 340MH/s, -f 1, and 4096 outputs, you lose 0.0000024% :)
1161 2011-02-04 12:39:42 <sipa> sorry, 16384
1162 2011-02-04 12:39:48 <tcatm> why risk losing anything at all?
1163 2011-02-04 12:39:54 <edcba> 13:06 <@tcatm> same format?
1164 2011-02-04 12:39:54 <edcba> 13:06 < davout> hmm
1165 2011-02-04 12:39:54 <edcba> 13:06 < davout> not sure
1166 2011-02-04 12:40:06 <edcba> if you are using xml you'd better validate it...
1167 2011-02-04 12:40:09 <m0mchil> output[64] 0.12%, output[128] 0.06 and so on
1168 2011-02-04 12:40:32 <m0mchil> I think even 0.12 is quite acceptable
1169 2011-02-04 12:40:37 skeledrew has quit (Ping timeout: 255 seconds)
1170 2011-02-04 12:40:39 <tcatm> there's plenty of time for checking every 1024 hashes in CPU while the next kernel is running
1171 2011-02-04 12:40:39 <sipa> 64 outputs -> 0.06% loss
1172 2011-02-04 12:40:42 <m0mchil> but, yeah, why loose anything at all
1173 2011-02-04 12:41:11 <Diablo-D3> what about just 4 outputs?
1174 2011-02-04 12:41:22 <m0mchil> 2%
1175 2011-02-04 12:41:26 <sipa> 0.986%
1176 2011-02-04 12:41:36 <sipa> (at 340MH/s, -f 1)
1177 2011-02-04 12:41:42 <m0mchil> sipa, why are we ofset :)
1178 2011-02-04 12:41:53 <sipa> quite sure my formula is right
1179 2011-02-04 12:42:05 <tcatm> a semprom 140 (common miner CPU) can do 2880e3 hashes/s so checking 1024 every now and then is no problem
1180 2011-02-04 12:42:40 <sipa> tcatm: so you say, create 1024 outputs, and check all nonces that map to a given output if that is set to non-zero?
1181 2011-02-04 12:42:50 <tcatm> no
1182 2011-02-04 12:42:59 <tcatm> create outputs equal to globalthreadsize
1183 2011-02-04 12:43:05 <m0mchil> sipa, my mistake
1184 2011-02-04 12:43:09 <tcatm> every thread checks 1024 nonces
1185 2011-02-04 12:43:12 <Diablo-D3> welp
1186 2011-02-04 12:43:13 <Diablo-D3> its official
1187 2011-02-04 12:43:14 <Diablo-D3> I just gamed
1188 2011-02-04 12:43:17 <Diablo-D3> with the miner running
1189 2011-02-04 12:43:19 <Diablo-D3> and saw no change
1190 2011-02-04 12:43:25 <tcatm> if any of them is H=0 it sets output[threadid]
1191 2011-02-04 12:43:25 <sipa> yeah, but that only works if you use the loop
1192 2011-02-04 12:43:57 <tcatm> then you check those 1024 nonces again in hostcode to find the one with smallest G
1193 2011-02-04 12:44:15 <tcatm> yeah, well. just use the loop?
1194 2011-02-04 12:44:16 <sipa> or just find all
1195 2011-02-04 12:44:23 <Diablo-D3> actually
1196 2011-02-04 12:44:26 <tcatm> why all? smallest G is enough
1197 2011-02-04 12:44:29 <Diablo-D3> hrm
1198 2011-02-04 12:44:33 <Diablo-D3> yeah smallest G would be enough
1199 2011-02-04 12:44:38 <sipa> when you're using the pool, that's not enough
1200 2011-02-04 12:44:46 <tcatm> sure
1201 2011-02-04 12:44:48 <Diablo-D3> sipa: yes it is
1202 2011-02-04 12:44:57 <Diablo-D3> because any H == 0 is enough for the pool
1203 2011-02-04 12:45:00 <sipa> yes
1204 2011-02-04 12:45:15 <sipa> but you want to submit as much shares as possible
1205 2011-02-04 12:45:21 <sipa> not just the best one
1206 2011-02-04 12:45:30 <Diablo-D3> yes, but I said that like 2 hours ago
1207 2011-02-04 12:45:33 <Diablo-D3> and no one cared
1208 2011-02-04 12:45:43 <tcatm> don't you submit one share per getwork?
1209 2011-02-04 12:46:00 <Diablo-D3> tcatm: not with the pool
1210 2011-02-04 12:46:04 <Diablo-D3> you can edge it out
1211 2011-02-04 12:46:08 <sipa> well, i didn't read that Diablo-D3, but i agree
1212 2011-02-04 12:46:31 <sipa> if you're using the pool, this "bug" causes some % less shares to be submitted to pool
1213 2011-02-04 12:46:49 <sipa> fix: send all
1214 2011-02-04 12:46:56 andrew12 has quit (Quit: parting is such sweet sorrow...)
1215 2011-02-04 12:47:17 <Diablo-D3> doing all outputs sucks dick really
1216 2011-02-04 12:47:21 <sipa> why?
1217 2011-02-04 12:47:34 <Diablo-D3> because you have to allocate large output buffers
1218 2011-02-04 12:47:46 <tcatm> 1.4MB is small
1219 2011-02-04 12:47:49 <sipa> 13:28:52 <@tcatm> btw: on a 5870 at 350Mhash/s with -f 1 it takes 1.4 MByte
1220 2011-02-04 12:47:57 <Diablo-D3> no, 4 bytes is small
1221 2011-02-04 12:48:32 <tcatm> looks like I have to port my miner to support getwork :P
1222 2011-02-04 12:48:41 <sipa> then use my idea and use 1 KiB of output buffer, risk losing 0.015%
1223 2011-02-04 12:49:28 <Diablo-D3> well the important part is that just writing 1 to the output instead of the nonce is a theoretical speed win
1224 2011-02-04 12:49:55 <Diablo-D3> so if I output 64... thats uh...
1225 2011-02-04 12:50:26 <tcatm> well then compress those 1.4 MB into a bitfield?
1226 2011-02-04 12:50:32 <Diablo-D3> no
1227 2011-02-04 12:50:39 <Diablo-D3> the action of using 1 itself is the speed win
1228 2011-02-04 12:50:42 <tcatm> But I really don't think bandwidth is a problem at all, even with PCI-E x1
1229 2011-02-04 12:50:54 <Diablo-D3> its a problem because no DMA
1230 2011-02-04 12:50:58 <tcatm> why?
1231 2011-02-04 12:51:12 <Diablo-D3> Im not going to explain to you how computers work.
1232 2011-02-04 12:51:26 <Diablo-D3> but without DMA, no one wants to shove a 4mb page of shit 1000 times a second
1233 2011-02-04 12:51:49 <tcatm> 1000 times? we're talking about once a second
1234 2011-02-04 12:51:55 <Diablo-D3> yes if you use -f 1.
1235 2011-02-04 12:52:05 <sipa> then use a smaller buffer when you have a larger -f
1236 2011-02-04 12:52:09 <tcatm> if you use higher -f the buffer ceomes smaller...
1237 2011-02-04 12:52:14 <tcatm> becomes*
1238 2011-02-04 12:52:17 <Diablo-D3> tcatm: bzt.
1239 2011-02-04 12:52:22 <Diablo-D3> no one wants magically size changing buffers.
1240 2011-02-04 12:52:29 andrew12 has joined
1241 2011-02-04 12:52:39 <tcatm> magically?
1242 2011-02-04 12:52:55 <sipa> you're already "magically" changing worksizes
1243 2011-02-04 12:53:01 <andrew12> still got about 15 mins before i gotta actually go to class
1244 2011-02-04 12:53:02 <Diablo-D3> sipa: not at all.
1245 2011-02-04 12:53:05 <sipa> *globalthreadsizes
1246 2011-02-04 12:53:25 <andrew12> programmers love the word magic
1247 2011-02-04 12:53:32 <Diablo-D3> allocating and destroying buffers is expensive
1248 2011-02-04 12:53:34 <Diablo-D3> no one wants that.
1249 2011-02-04 12:53:48 <tcatm> you can tell elEnqueueReadBuffer exactly how many bytes to copy from GPU->host
1250 2011-02-04 12:54:04 <Diablo-D3> yes, if I want to, sure.
1251 2011-02-04 12:54:04 <tcatm> so allocate say 8MB and only copy what you want
1252 2011-02-04 12:54:35 <Diablo-D3> but I could also punch people in the dick until khash come out.
1253 2011-02-04 12:55:09 <slush> I'm going to full restart of server after more than a month. I'm curious how many miners will completely crash :)
1254 2011-02-04 12:55:24 <Diablo-D3> slush: dude, they crash half the time anyhow
1255 2011-02-04 12:56:04 <slush> Diablo-D3: really? My/your miner is quite stable for many weeks...
1256 2011-02-04 12:56:20 <afed> my copy of poclbm has survived
1257 2011-02-04 12:56:24 <Diablo-D3> slush: yes, except your pool went down earlier and my miner went kaput
1258 2011-02-04 12:56:27 <afed> it got p. quiet in here though :)
1259 2011-02-04 12:56:39 <andrew12> hm
1260 2011-02-04 12:56:58 andrew12 has quit (Client Quit)
1261 2011-02-04 12:59:56 <slush> Diablo-D3: my miner reconnected after restart without any problem
1262 2011-02-04 13:00:04 <slush> Diablo-D3: plus points for you :)
1263 2011-02-04 13:00:40 <Diablo-D3> heh
1264 2011-02-04 13:00:43 <tcatm> slush: is a miner supposed to submit multiple results per getwork?
1265 2011-02-04 13:01:17 <slush> tcatm: yes, no problem with this
1266 2011-02-04 13:01:22 <slush> pool accept it
1267 2011-02-04 13:01:36 <tcatm> k
1268 2011-02-04 13:01:47 <tcatm> so my miner won't support pools
1269 2011-02-04 13:02:23 <slush> tcatm: why not?
1270 2011-02-04 13:02:39 <slush> no problem when he won't do multiple submits
1271 2011-02-04 13:02:51 <tcatm> won't he get less shares?
1272 2011-02-04 13:02:56 <Diablo-D3> no
1273 2011-02-04 13:03:07 <slush> tcatm: he will get so much shares how many unique and valid share you upload :)
1274 2011-02-04 13:03:54 <tcatm> so if it only submits one result per getwork instead of two (on avrage) I get only half the bitcoins?
1275 2011-02-04 13:04:06 <slush> tcatm: no, why
1276 2011-02-04 13:04:14 <slush> if you crunch some nonces all the time,
1277 2011-02-04 13:04:24 <slush> there is no reason why you should get lower reward
1278 2011-02-04 13:04:25 <sipa> ?
1279 2011-02-04 13:05:04 <slush> tcatm: in average, there is one valid share per getwork
1280 2011-02-04 13:05:50 <tcatm> looks more like 2..3 (if you search the whole 2^32nonces)
1281 2011-02-04 13:05:50 <slush> well said: per 2*32 nonce tries
1282 2011-02-04 13:06:04 <slush> how so?
1283 2011-02-04 13:06:31 <tcatm> I got that figure form watching my miners H=0 debug output
1284 2011-02-04 13:06:41 <slush> for how long time?
1285 2011-02-04 13:06:48 <tcatm> not long
1286 2011-02-04 13:06:53 <slush> ;)
1287 2011-02-04 13:07:05 [Noodles] has quit (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
1288 2011-02-04 13:08:05 <davout> yay bc is working on bitcoin charts
1289 2011-02-04 13:08:06 <sipa> you lose 41% of shares if you search the full nonce space of every getwork received, and only submit at most one result
1290 2011-02-04 13:08:12 <davout> tcatm: thank you !
1291 2011-02-04 13:08:21 <slush> sipa: yes, that's true
1292 2011-02-04 13:08:41 <sipa> that is tcatms question i think
1293 2011-02-04 13:08:41 <slush> but you don't lose anything if you drop getwork once you find a share and start crunching new one
1294 2011-02-04 13:08:50 <slush> oh, then I misunderstood, sorry
1295 2011-02-04 13:08:56 <sipa> of course
1296 2011-02-04 13:09:23 * sipa looks at tcatm
1297 2011-02-04 13:18:11 ducki2p has quit (Remote host closed the connection)
1298 2011-02-04 13:18:40 ducki2p has joined
1299 2011-02-04 13:22:22 <tcatm> slush: Can I request multiple getworks per user?
1300 2011-02-04 13:22:58 <tcatm> i.e. request new work when noncespace is near end but still return results for old getwork
1301 2011-02-04 13:23:01 [Noodles] has joined
1302 2011-02-04 13:23:31 <slush> tcatm: yes
1303 2011-02-04 13:23:39 <slush> it's preferred
1304 2011-02-04 13:23:55 <tcatm> great
1305 2011-02-04 13:24:10 <slush> by default, there is register of 12 jobs per worker on pool
1306 2011-02-04 13:24:11 <tcatm> Maybe I'll add pool mode later
1307 2011-02-04 13:24:37 <slush> and I changed it manually for some large mining rigs where it was not enough
1308 2011-02-04 13:25:07 <slush> tcatm: is your miner public?
1309 2011-02-04 13:26:03 <tcatm> the old one is, but doesn't support getwork
1310 2011-02-04 13:26:10 <tcatm> I'm rewriting it for getwork now
1311 2011-02-04 13:26:13 davex__ has joined
1312 2011-02-04 13:26:32 <slush> which language?
1313 2011-02-04 13:26:34 <tcatm> plain c
1314 2011-02-04 13:38:43 grondilu has joined
1315 2011-02-04 13:38:49 <grondilu> Very last day for the gold coin auction!!  http://www.biddingpond.com/item.php?id=246
1316 2011-02-04 13:45:23 <genjix> 1 aĉetanto
1317 2011-02-04 13:45:36 ApertureScience has joined
1318 2011-02-04 13:45:37 <genjix> kial oni aĉetus oron?
1319 2011-02-04 13:47:50 <tcatm> anyone know how to use autotools?
1320 2011-02-04 13:48:44 <genjix> no but you could try cmake if you're looking for a build system
1321 2011-02-04 13:49:44 <tcatm> that's even more ugly
1322 2011-02-04 13:50:03 * UukGoblin heard waf was nice
1323 2011-02-04 13:50:07 <ArtForz> ;;bc,mtgox
1324 2011-02-04 13:50:07 <gribble> {"ticker":{"high":0.88,"low":0.6642,"vol":30155,"buy":0.7051,"sell":0.879,"last":0.88}}
1325 2011-02-04 13:50:08 <grondilu> genjix: same tial, kial oni acxetus bitcoinojn
1326 2011-02-04 13:50:14 <UukGoblin> but yes, I've used the dreaded autotools before many times
1327 2011-02-04 13:50:32 rli has joined
1328 2011-02-04 13:50:46 <Diablo-D3> so guys
1329 2011-02-04 13:50:52 <Diablo-D3> what did we decide the nonce lost on 1024 was?
1330 2011-02-04 13:50:53 <genjix> oni elspezas bitcoinojn, sed oni malfacile spezas oron
1331 2011-02-04 13:51:00 <davex__> ArtForz, So I tried the 10.11 and 10.12 drivers, and same thing.  although now I'm noticing that DiabloMiner is going 300% CPU usage.  getting about 18Mh/S still
1332 2011-02-04 13:51:12 <Diablo-D3> davex__: what sdk?
1333 2011-02-04 13:51:16 <ArtForz> that sounds like you're using sdk 2.2 or 2.3
1334 2011-02-04 13:51:28 <davex__> java sdk?
1335 2011-02-04 13:51:36 <Diablo-D3> no stream sdk
1336 2011-02-04 13:51:37 <UukGoblin> ati stream sdk
1337 2011-02-04 13:51:47 <davex__> mmm lemme check
1338 2011-02-04 13:51:54 <davex__> yup 2.3
1339 2011-02-04 13:52:04 <UukGoblin> need. 2.1.
1340 2011-02-04 13:52:04 <ArtForz> because thats EXACTLY what I observed with my miner and Stream/APP 2.2 or 2.3
1341 2011-02-04 13:52:04 <Diablo-D3> use 2.1
1342 2011-02-04 13:52:09 <davex__> oh
1343 2011-02-04 13:52:18 <rli> ;bc,stats
1344 2011-02-04 13:52:26 <rli> ;;bc,stats
1345 2011-02-04 13:52:28 <gribble> Current Blocks: 106157 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 690 blocks | Next Difficulty In About: 4 days, 5 hours, 12 minutes, and 0 seconds | Next Difficulty Estimate: 24641.70745980
1346 2011-02-04 13:52:47 <ArtForz> looks like next diff increase wont be too bad
1347 2011-02-04 13:52:56 <UukGoblin> indeed
1348 2011-02-04 13:53:01 <rli> hopefully:)
1349 2011-02-04 13:53:05 <UukGoblin> *launches new cluster*
1350 2011-02-04 13:53:11 <ArtForz> only 690 blocks to go and < 12% up
1351 2011-02-04 13:53:41 <davex__> it's not keeping up with the market price at all, it seems like
1352 2011-02-04 13:53:53 <ArtForz> or maybe the market price is catching up :P
1353 2011-02-04 13:54:12 <davex__> hah
1354 2011-02-04 13:54:36 <UukGoblin> yeah, about time
1355 2011-02-04 13:54:46 sotto has joined
1356 2011-02-04 13:54:54 <UukGoblin> there've been huge difficulty increases without relevant price adjustments
1357 2011-02-04 13:55:05 <ArtForz> yup
1358 2011-02-04 13:55:17 grondilu has quit (Quit: leaving)
1359 2011-02-04 13:55:27 <ArtForz> I expect another difficulty surge when 6990 is available in quantity
1360 2011-02-04 13:55:51 <davex__> but, generating costs are what like $ .1 per bitcoin?
1361 2011-02-04 13:55:56 <slush> UukGoblin: Mainly because building rigs take some time ;)
1362 2011-02-04 13:55:59 <UukGoblin> btw, how are your ASICs ArtForz? :-)
1363 2011-02-04 13:56:03 <ArtForz> less
1364 2011-02-04 13:56:10 <nevezen> isn't it already difficult once people started using gpu miners?
1365 2011-02-04 13:56:20 <ArtForz> still waiting for chips to ship
1366 2011-02-04 13:56:39 <UukGoblin> have you got PCBs for them ready to rock? :-]
1367 2011-02-04 13:56:43 <ArtForz> yep
1368 2011-02-04 13:56:45 <UukGoblin> :-)
1369 2011-02-04 13:56:48 <slush> ArtForz: 6990 will be multigpu?
1370 2011-02-04 13:56:52 <ArtForz> yep
1371 2011-02-04 13:57:03 <ArtForz> 6990 = dual (downclocked) 6970
1372 2011-02-04 13:57:08 <ArtForz> already got all PCBs and support components
1373 2011-02-04 13:57:23 <lfm> davex__: no, ati opencl sdk
1374 2011-02-04 13:57:24 <UukGoblin> well done
1375 2011-02-04 13:57:48 <UukGoblin> you must have been working on that full time
1376 2011-02-04 13:57:50 <davex__> lfm: yeah got it
1377 2011-02-04 13:57:54 <ArtForz> not really
1378 2011-02-04 13:58:02 <slush> I thought 6970 is equivalent of 5980 in mining...
1379 2011-02-04 13:58:10 <ArtForz> 6970 is about == 5870
1380 2011-02-04 13:58:18 <ArtForz> actually a bit slower
1381 2011-02-04 13:58:19 <slush> oh, of course
1382 2011-02-04 13:58:32 <ArtForz> and uses about 20W more power
1383 2011-02-04 13:58:46 <slush> so what's the point to wait to 6990
1384 2011-02-04 13:58:54 <ArtForz> hopefully 6990 will be running at lower voltage like 5970 vs 5870
1385 2011-02-04 13:59:12 <ArtForz> and cheaper per Mhash
1386 2011-02-04 13:59:40 <UukGoblin> and hopefully they'll fix their sdk ;-)
1387 2011-02-04 13:59:42 <ArtForz> board+PSU+case drives $/Mh up quite a bit if you have to use single GPU cards
1388 2011-02-04 13:59:48 dishwara has joined
1389 2011-02-04 14:00:22 <slush> so if I understand, it will be still worse than 5970 for big rigs
1390 2011-02-04 14:00:29 <ArtForz> yes
1391 2011-02-04 14:00:33 <lfm> bah no connect
1392 2011-02-04 14:00:33 lfm has quit (Quit: brb)
1393 2011-02-04 14:00:40 <ArtForz> but all signs point to there wont be no more 5970s
1394 2011-02-04 14:00:58 lfm has joined
1395 2011-02-04 14:01:22 <ArtForz> the $500 sapphires in early dec were pretty much AIBs clearing stock
1396 2011-02-04 14:01:42 <nevezen> is that the fastest card on the market at the moment?
1397 2011-02-04 14:01:51 <ArtForz> what?
1398 2011-02-04 14:01:52 <UukGoblin> best Mhash/$
1399 2011-02-04 14:02:04 <ArtForz> best Mh/$ for card alone is still 5770
1400 2011-02-04 14:02:10 <nevezen> the 6990
1401 2011-02-04 14:02:23 <ArtForz> nobody knows
1402 2011-02-04 14:02:24 <UukGoblin> really?
1403 2011-02-04 14:02:36 <ArtForz> probably 2-3 AMD execs
1404 2011-02-04 14:02:46 <nevezen> what's the fastest then, regardless of the price?
1405 2011-02-04 14:03:03 <ArtForz> 4GB 850MHz 5970s
1406 2011-02-04 14:03:07 <ArtForz> aka 5870x2
1407 2011-02-04 14:03:12 <Diablo-D3> ArtForz: well
1408 2011-02-04 14:03:13 <Diablo-D3> to be fair
1409 2011-02-04 14:03:16 <ArtForz> about $1k a pop
1410 2011-02-04 14:03:19 <UukGoblin> ArtForz, according to https://en.bitcoin.it/wiki/Mining_Hardware_Comparison 5970 is best
1411 2011-02-04 14:03:20 <Diablo-D3> 5870 is just as fast at that rate
1412 2011-02-04 14:03:30 jon______ has joined
1413 2011-02-04 14:03:38 <ArtForz> and how do you fit 8 of those in one box? :P
1414 2011-02-04 14:03:55 <Diablo-D3> the same way you fit 8 5970s in a box
1415 2011-02-04 14:03:57 <UukGoblin> oh are you counting the card's fixed cost?
1416 2011-02-04 14:04:06 jon______ has quit (Client Quit)
1417 2011-02-04 14:04:16 <ArtForz> you mean 4 5970s
1418 2011-02-04 14:04:38 <ArtForz> 4*5870x2 == 8*5870
1419 2011-02-04 14:04:43 <Keefe> the drivers can't handle more than 8 chips, right?
1420 2011-02-04 14:04:47 <ArtForz> yeah
1421 2011-02-04 14:04:54 <UukGoblin> oh really?
1422 2011-02-04 14:04:56 <UukGoblin> shit
1423 2011-02-04 14:05:03 <ArtForz> you can get around that with a ugly hack though
1424 2011-02-04 14:05:06 * UukGoblin was thinking about putting 5 of them on one mobo
1425 2011-02-04 14:05:19 smark_ has joined
1426 2011-02-04 14:05:30 <UukGoblin> can you not simply run 2 x servers?
1427 2011-02-04 14:05:36 <UukGoblin> X I mean
1428 2011-02-04 14:05:36 <ArtForz> yep
1429 2011-02-04 14:05:47 <ArtForz> well, "simply"
1430 2011-02-04 14:05:56 <UukGoblin> ah, right
1431 2011-02-04 14:06:23 <UukGoblin> actually, I just had a crazy idea this morning... basically... pci-express x16 they call it... so it should be able to take 16 x1 cards, right? :-]
1432 2011-02-04 14:06:47 <ArtForz> no
1433 2011-02-04 14:06:53 <Diablo-D3> no
1434 2011-02-04 14:07:01 gejohn has joined
1435 2011-02-04 14:07:02 <UukGoblin> well
1436 2011-02-04 14:07:11 <UukGoblin> it has 16 x1 serial data links
1437 2011-02-04 14:07:14 <ArtForz> some PCIe bridges actually *can* do that
1438 2011-02-04 14:07:21 <Diablo-D3> yeah some bridges can
1439 2011-02-04 14:07:22 <Keefe> how expensive?
1440 2011-02-04 14:07:28 <Diablo-D3> expensive.
1441 2011-02-04 14:07:28 <ArtForz> if you have to ask...
1442 2011-02-04 14:07:33 <Keefe> heh
1443 2011-02-04 14:07:37 <UukGoblin> ah! :-) so it /is/ possible :-)
1444 2011-02-04 14:07:53 * UukGoblin is not as stupid as /me thought :-]
1445 2011-02-04 14:08:02 <ArtForz> another way: VMs + PCI passthrough
1446 2011-02-04 14:08:37 <UukGoblin> ArtForz, you mean for the 'put 10 chips in a box' problem?
1447 2011-02-04 14:08:40 <ArtForz> yep
1448 2011-02-04 14:09:20 <UukGoblin> or perhaps do a binary search for an '8' in ati drivers and change it to '128' ;-]
1449 2011-02-04 14:09:27 <ArtForz> you could probably have something like 60 GPUs hanging off a single box that way
1450 2011-02-04 14:10:14 <ArtForz> expensive as hell, but possible
1451 2011-02-04 14:10:55 <UukGoblin> well surely comparable to the ASIC investment? ;-)
1452 2011-02-04 14:11:32 <ArtForz> yep
1453 2011-02-04 14:11:39 <ArtForz> just with 10x the power use
1454 2011-02-04 14:11:46 <UukGoblin> :-]
1455 2011-02-04 14:11:49 smark_ has quit (Quit: Page closed)
1456 2011-02-04 14:11:58 <ArtForz> well, hopefully it'll be 10x
1457 2011-02-04 14:12:09 <ArtForz> I guess I'll be happy with 8x, too
1458 2011-02-04 14:12:14 <UukGoblin> :-)
1459 2011-02-04 14:12:21 <UukGoblin> I can't wait for your results :-)
1460 2011-02-04 14:12:36 <davex__> ok, 2.1 same thing.  deleted 2.3 stream sdk and pointing at 2.1
1461 2011-02-04 14:12:59 <ArtForz> are you sure 2.3 is really gone?
1462 2011-02-04 14:13:10 <ArtForz> I've never seen diablominer use 100% cpu with 2.1
1463 2011-02-04 14:13:13 <davex__> not entirely
1464 2011-02-04 14:13:29 <UukGoblin> try rebooting
1465 2011-02-04 14:13:41 <UukGoblin> I know it's linux, but it sometimes helps ;-)
1466 2011-02-04 14:13:55 <Diablo-D3> my miner uses 100% cpu for the first 30 seconds.
1467 2011-02-04 14:13:55 <ArtForz> check what libOpenCL it's loading
1468 2011-02-04 14:14:00 <ArtForz> it does? why?
1469 2011-02-04 14:14:11 <UukGoblin> kernel compilation?
1470 2011-02-04 14:15:08 <davex__> oh fuck.  stupid
1471 2011-02-04 14:15:21 <davex__> i had copied the .so's to the diablominer target natives folder
1472 2011-02-04 14:15:23 <sipa> slush: by the way, do you have data on how much BTC a particular account and/or miner has produced (in total, ever)?
1473 2011-02-04 14:15:26 <ArtForz> yes, fucking stupid is fun.
1474 2011-02-04 14:15:38 <davex__> forgot that i did that
1475 2011-02-04 14:15:55 <Diablo-D3> ArtForz: it doesnt sleep.
1476 2011-02-04 14:16:07 <Diablo-D3> it obsessively runs the calibration code repeatedly
1477 2011-02-04 14:16:11 <ArtForz> why copy em at all?
1478 2011-02-04 14:16:22 <UukGoblin> fucking? that small town in austria?
1479 2011-02-04 14:16:30 <davex__> no good reason.  just didn't want to set LD_LIBRARY_PATH
1480 2011-02-04 14:16:31 <ArtForz> yup
1481 2011-02-04 14:16:41 <ArtForz> you know, there's symlinks too
1482 2011-02-04 14:16:52 <davex__> yeah that would have been smarter
1483 2011-02-04 14:17:09 <slush> sipa: yes
1484 2011-02-04 14:17:18 <sipa> possible to show that? ;)
1485 2011-02-04 14:17:22 <slush> on the account level, not by worker
1486 2011-02-04 14:17:30 <slush> sipa: basically it is in 'dailyreward' graph
1487 2011-02-04 14:17:39 <sipa> oh of course
1488 2011-02-04 14:17:57 <slush> but I can show total BTC also, no prob
1489 2011-02-04 14:19:32 <UukGoblin> oh, I have a question too - did anything improve in terms of flood/spam resistance? i.e. if I created a 1000 accounts and started repeatedly doing microtransfers between them?
1490 2011-02-04 14:20:07 <ArtForz> yes
1491 2011-02-04 14:21:07 <UukGoblin> damn, me and my questions sometimes - "and if so, what is it?" ;-)
1492 2011-02-04 14:22:23 <UukGoblin> I mean, the 'attack' I'm thinking about would be continuously bloat up all blocks so that every transaction would require a fee
1493 2011-02-04 14:22:35 <ArtForz> nodes will just ignore transactions that dont have enough fees to get into a block legitimately, there can be at most 14 or so transactions with 0/unconfirmed inputs per block, block size limits before fees are required was lowered significantly
1494 2011-02-04 14:23:41 <ArtForz> oh, and fee-lees transactions are also scored on sum of input values
1495 2011-02-04 14:24:47 <ArtForz> = low value spammish transactions pretty much only get a few kB per block
1496 2011-02-04 14:26:12 <ArtForz> imo still missing code to drop "spammish" transactions when transaction cache grows beyond a certain size
1497 2011-02-04 14:27:21 <UukGoblin> oh, so well-established accounts (with well-confirmed balances) will get higher priority for fee-less transactions, cool
1498 2011-02-04 14:27:35 <ArtForz> yep
1499 2011-02-04 14:28:23 <davex__> well shit...
1500 2011-02-04 14:28:23 <ArtForz> iirc TX score = sum of (input_value * #confirmations) for all inputs
1501 2011-02-04 14:28:38 <ArtForz> so 0-conf inputs always have a score of 0
1502 2011-02-04 14:29:42 <UukGoblin> awesome
1503 2011-02-04 14:29:47 <UukGoblin> so it's ready for production! :->
1504 2011-02-04 14:29:49 <ArtForz> while a well-confirmed 50btc or so tx with a few 1000 confs pretty much guarantees priority over other free transactions
1505 2011-02-04 14:42:49 <tcatm> Does the pool abuse strategy Raulo describes really work?
1506 2011-02-04 14:43:34 <ArtForz> dunno, never bothered simulating it
1507 2011-02-04 14:43:35 <Diablo-D3> [2/4/11 9:37:30 AM] DEBUG: Attempt 5 found on ATI RV770 (#1)
1508 2011-02-04 14:43:43 <Diablo-D3> [2/4/11 9:37:30 AM] DEBUG: Attempt 6 found on ATI RV770 (#1)
1509 2011-02-04 14:43:45 <Diablo-D3> wee
1510 2011-02-04 14:43:52 <sipa> tcatm: link?
1511 2011-02-04 14:44:02 <tcatm> http://www.bitcoin.org/smf/index.php?topic=3165.0
1512 2011-02-04 14:44:12 * tcatm writes a RPC multiplexer
1513 2011-02-04 14:44:36 molecular has quit (Ping timeout: 276 seconds)
1514 2011-02-04 14:44:51 <ArtForz> I suspect it works
1515 2011-02-04 14:45:30 <tcatm> If I understand it right, it doesn't matter when I submit those 43.5% to the pool, does it?
1516 2011-02-04 14:46:42 molecular has joined
1517 2011-02-04 14:51:16 <sipa> tcatm: should still be valid
1518 2011-02-04 14:51:21 <sipa> the shares you submit
1519 2011-02-04 14:53:41 <slush> tcatm: no, it matter
1520 2011-02-04 14:54:13 <slush> It's the most effective to start mining immediately when pool find new block
1521 2011-02-04 14:54:27 <slush> then the share has the biggest value, for short rounds, for example
1522 2011-02-04 14:54:38 <tcatm> hm true
1523 2011-02-04 14:54:47 <ArtForz> I'm still not 100% convinced
1524 2011-02-04 14:54:53 <slush> I'm not happy with it, but I will have to publish my new pool feature - delayed stats
1525 2011-02-04 14:55:03 <slush> which effectively solve all this stuff
1526 2011-02-04 14:55:04 <Diablo-D3> rutrow
1527 2011-02-04 14:55:13 <Diablo-D3> I have found a wonderful new bug in my program
1528 2011-02-04 14:55:19 yebyen has joined
1529 2011-02-04 14:55:24 <tcatm> slush: how could that solve the problem?
1530 2011-02-04 14:55:46 <Diablo-D3> I cant return 1 and do nonce checking
1531 2011-02-04 14:55:56 <Diablo-D3> I have to get rid of hardware bitching if I do
1532 2011-02-04 14:55:57 <slush> tcatm: basically, not showing status of the current round :)
1533 2011-02-04 14:55:58 * yebyen has officially doubled his money at mtgox.com
1534 2011-02-04 14:56:00 <yebyen> woohoo
1535 2011-02-04 14:56:06 <yebyen> (only doubled)
1536 2011-02-04 14:56:14 <sipa> Diablo-D3: how do you mean?
1537 2011-02-04 14:56:25 <tcatm> slush: not worth it. I'll just connect to your bitcoind and check the blocks it outputs
1538 2011-02-04 14:56:28 <Diablo-D3> [2/4/11 9:49:29 AM] DEBUG: Attempt 1235330 found on ATI RV770 (#1)
1539 2011-02-04 14:56:28 <Diablo-D3> [2/4/11 9:49:29 AM] ERROR: Invalid block found on ATI RV770 (#1), possible driver or hardware issue
1540 2011-02-04 14:56:31 <Diablo-D3> thats how I mean.
1541 2011-02-04 14:56:35 <Diablo-D3> and thats after several seconds.
1542 2011-02-04 14:56:43 <slush> tcatm: how?
1543 2011-02-04 14:56:44 <tcatm> when I receive a block I didn't get from anyone else before I assume you found it
1544 2011-02-04 14:56:54 <tcatm> i.e. => new round
1545 2011-02-04 14:57:14 <slush> it's far from exact
1546 2011-02-04 14:57:21 Zarutian has quit (Quit: Zarutian)
1547 2011-02-04 14:57:58 <UukGoblin> hmm if it gives fixed 28% gain it's not exactly cheating ;-]
1548 2011-02-04 14:58:08 <UukGoblin> just give the extra 28% chance to everyone and it's fair :-]
1549 2011-02-04 14:58:10 <Diablo-D3> so I'll just go for less loss
1550 2011-02-04 14:58:34 <sipa> it is cheating because if everyone did it, the pool would die :)
1551 2011-02-04 14:58:40 <slush> yes
1552 2011-02-04 14:58:55 <slush> it works only when small group will do it
1553 2011-02-04 14:59:02 <UukGoblin> ah
1554 2011-02-04 14:59:16 <slush> because when everybody disconnect @ 43%, pool never find new block ;)
1555 2011-02-04 14:59:45 <UukGoblin> well it might find the next one later
1556 2011-02-04 14:59:45 * sipa still thinks only counting shares from the current block (instead of whole round) is the best solution
1557 2011-02-04 15:00:08 <sipa> you'll have larger variations, but way smaller than when mining on your own
1558 2011-02-04 15:00:29 <slush> sipa: but when you contribute only 43% time between two bitcoin blocks, the gain will be very similar
1559 2011-02-04 15:00:42 <Diablo-D3> so
1560 2011-02-04 15:00:45 <Diablo-D3> if I use 64 outputs
1561 2011-02-04 15:01:00 <Diablo-D3> instead of losing 3.5% of winning shares, I lose 0.05%
1562 2011-02-04 15:01:10 Xunie has quit (Read error: Connection reset by peer)
1563 2011-02-04 15:01:24 <sipa> 0.06% i get (at 340MH/s, -f 1)
1564 2011-02-04 15:01:26 <sipa> but yes
1565 2011-02-04 15:01:43 <Diablo-D3> well 0.0546%
1566 2011-02-04 15:01:46 <Diablo-D3> so my rounding is superior
1567 2011-02-04 15:01:49 devon_hillard_ has quit (Quit: Leaving)
1568 2011-02-04 15:01:58 devon_hillard has joined
1569 2011-02-04 15:02:06 <sipa> 0.061832878483238%
1570 2011-02-04 15:02:13 <Diablo-D3> feh
1571 2011-02-04 15:02:23 <UukGoblin> kinda reminds me of the Monty Hall Problem :-]
1572 2011-02-04 15:02:24 <sipa> anyway, something like that
1573 2011-02-04 15:02:33 <Diablo-D3> I gotta see how much damage multiple outputs do
1574 2011-02-04 15:02:54 <molecular> Diablo-D3, what's the problem?
1575 2011-02-04 15:03:05 <Diablo-D3> molecular: well, see, your mother has been hitting on me again
1576 2011-02-04 15:03:23 <Diablo-D3> I keep telling her shes not my type, but she doesnt get the hint
1577 2011-02-04 15:03:27 <molecular> Diablo-D3, oh damn. has she been drinking again?
1578 2011-02-04 15:03:56 <Diablo-D3> I think so
1579 2011-02-04 15:04:08 <molecular> Diablo-D3, that explains why she's telling a different story
1580 2011-02-04 15:04:23 <devon_hillard> At what stage does a dual 16x pci-express bus begin to matter for gaming, compared to a dual 8x (considering SLI or CrossfireX)
1581 2011-02-04 15:04:39 <Diablo-D3> devon_hillard: nothing.
1582 2011-02-04 15:04:57 <molecular> Diablo-D3, the question was: is it really a hardware issue?
1583 2011-02-04 15:04:59 <Diablo-D3> running them both at x1 1.0 may not actually have a real effect.
1584 2011-02-04 15:05:17 <Keefe> for gaming?
1585 2011-02-04 15:05:27 <devon_hillard> diablo-d3, not speaking about mining, but gaming, or other graphics-intensive applications
1586 2011-02-04 15:05:28 <Diablo-D3> molecular: no, I was testing host side nonce block checking
1587 2011-02-04 15:05:42 <Diablo-D3> devon_hillard: even on gaming, it probably wont have a huge effect
1588 2011-02-04 15:05:58 <Diablo-D3> devon_hillard: x4 1.0 or x2 2.0 seems to be about enough for gaming
1589 2011-02-04 15:06:04 <devon_hillard> what is the point of 2x16x then?
1590 2011-02-04 15:06:08 <ArtForz> err... no
1591 2011-02-04 15:06:18 <ArtForz> it was about 2 years ago
1592 2011-02-04 15:06:50 <EvanR-work> question about mining pool, seems like im getting about 50 every two days, and this rate is much closer to actually 50 in two days than mining alone. is this me miscalculating, coinsidence, or expected
1593 2011-02-04 15:06:52 <ArtForz> smth like a 5970 already takes a benchmarkable hit at 2.0 x8
1594 2011-02-04 15:06:56 <slush> tcatm: and I think that I solved problem with listening on blocks from the pool
1595 2011-02-04 15:07:10 <Diablo-D3> ArtForz: really?
1596 2011-02-04 15:07:12 <ArtForz> yep
1597 2011-02-04 15:07:15 <Diablo-D3> I find that hard to believe
1598 2011-02-04 15:07:25 <Diablo-D3> the only way that'd be true is if most of the textures dont fit in texture ram
1599 2011-02-04 15:07:28 <sipa> EvanR-work: the whole point of the pool is to give much more frequent small payouts
1600 2011-02-04 15:07:28 <ArtForz> don't remember where it was
1601 2011-02-04 15:07:30 <Diablo-D3> if it does, you have OTHER problems
1602 2011-02-04 15:07:41 <tcatm> slush: better deploy the fix then. My multiplexer is running :)
1603 2011-02-04 15:07:46 <ArtForz> but a pretty big/legit review sites
1604 2011-02-04 15:07:57 <slush> tcatm: multiplexer?
1605 2011-02-04 15:08:01 <sipa> tcatm: multiplexer = cheater?
1606 2011-02-04 15:08:07 <EvanR-work> sipa: right. and seems like over the course of two days, my payout is much closer to 50
1607 2011-02-04 15:08:20 <EvanR-work> rather than zero, 100, 150 sometimes
1608 2011-02-04 15:08:22 <tcatm> a small daemon to connect to both pool and bitcoind and getwork from both depending on roundstatus
1609 2011-02-04 15:08:23 <sipa> if you mine alone you'll have the expected rate
1610 2011-02-04 15:08:31 <sipa> but over much larger periods of time
1611 2011-02-04 15:08:31 <EvanR-work> in the long term yes
1612 2011-02-04 15:08:44 <EvanR-work> but is it expected to see that rate over the short term on the pool
1613 2011-02-04 15:08:45 <slush> tcatm: I'll run bitcoind on unknown IP ;)
1614 2011-02-04 15:08:51 <sipa> since there is a very real chance of not finding a block in a week time
1615 2011-02-04 15:09:01 <ArtForz> iirc while x8/x16 was like no diff for older games, it really sucked down perf for new stuff
1616 2011-02-04 15:09:01 <sipa> and also a real chance of finding a few on a single day
1617 2011-02-04 15:09:21 <EvanR-work> yeah i said screw that last night ;)
1618 2011-02-04 15:09:32 <EvanR-work> long live slush
1619 2011-02-04 15:09:47 <tcatm> slush: I could watch for getwork results and compare previoushash with blockchain
1620 2011-02-04 15:09:50 <ArtForz> wild guess: DMAing textures to card over slow bus causes contention somewhere
1621 2011-02-04 15:10:17 <sipa> tcatm: that won't tell you whether a block was found by the pool, only whether a block was found, no?
1622 2011-02-04 15:10:17 <slush> tcatm: so?
1623 2011-02-04 15:10:33 <slush> tcatm: this will work only if YOU find the block for the pool ;)
1624 2011-02-04 15:10:40 <ArtForz> maybe because driver has to run quite a bit of "housekeeping" over the bus for xfire?
1625 2011-02-04 15:10:51 <tcatm> well assuming I have a few well connected nodes I can tell whether the network knows about the prevhash or not
1626 2011-02-04 15:10:54 davout has quit ()
1627 2011-02-04 15:11:19 <sipa> the network might still know sooner than your next getwork
1628 2011-02-04 15:11:30 <sipa> but yes, it's one source of information
1629 2011-02-04 15:11:48 <slush> tcatm: you mean from the next getwork after pool find a block?
1630 2011-02-04 15:11:54 <tcatm> yep
1631 2011-02-04 15:12:01 <slush> tcatm: prevhash is in getwork somewhere?
1632 2011-02-04 15:12:04 <sipa> yes
1633 2011-02-04 15:12:08 <slush> I don't know all pieces
1634 2011-02-04 15:12:38 <molecular> is prevhash necessary in getwork? if not, slush could ommit it in his push proto
1635 2011-02-04 15:12:59 <sipa> getwork contains redundant information
1636 2011-02-04 15:13:20 <sipa> so if the miner only uses the correct parts, certain pieces could be overwritten with nonsense
1637 2011-02-04 15:13:51 <tcatm> probably easier to only count shares from current block
1638 2011-02-04 15:14:01 <sipa> indeed
1639 2011-02-04 15:14:28 <slush> I don't like it
1640 2011-02-04 15:14:42 <molecular> it's bad for low-performance miners
1641 2011-02-04 15:14:45 <slush> btw pool is also connected to 300+ nodes
1642 2011-02-04 15:14:51 <sipa> it's not bad, it's just less good
1643 2011-02-04 15:15:06 <sipa> but yes, slow miners will notice more
1644 2011-02-04 15:15:06 <slush> so there is big probability you will be noticed about new hash before you get new work to miner
1645 2011-02-04 15:20:16 <afed> ;;bc,diff
1646 2011-02-04 15:20:17 <gribble> 22012.4941572
1647 2011-02-04 15:20:40 <afed> ;;bc,help
1648 2011-02-04 15:20:40 <gribble> Alias bc,bcm, Alias bc,blocks, Alias bc,btcex, Alias bc,calc, Alias bc,calcd, Alias bc,diff, Alias bc,estimate, Alias bc,help, Alias bc,hextarget, Alias bc,labs, Alias bc,lbs, Alias bc,markets, Alias bc,mtgox, Alias bc,nexttarget, Alias bc,poolstats, Alias bc,prob, Alias bc,stats, Alias bc,timetonext, Alias bc,totalbc, and Alias bc,wiki
1649 2011-02-04 15:20:51 <afed> ;;bc,estimate
1650 2011-02-04 15:20:51 <gribble> 24617.69625825
1651 2011-02-04 15:21:04 <afed> ;;bc,stats
1652 2011-02-04 15:21:07 <gribble> Current Blocks: 106166 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 681 blocks | Next Difficulty In About: 4 days, 3 hours, 52 minutes, and 48 seconds | Next Difficulty Estimate: 24617.69625825
1653 2011-02-04 15:21:37 Xunie has joined
1654 2011-02-04 15:26:33 <dishwara> hi
1655 2011-02-04 15:26:37 <presence> ups--
1656 2011-02-04 15:27:09 <dishwara> i a have 9400GT thinking about buying 5870 ati
1657 2011-02-04 15:27:36 <presence> if the rate of processing goes up at this rate, the gpu's will be as the cpu's are today
1658 2011-02-04 15:27:48 <dishwara> so far i saw chat abt 5870, meaning ati stopped producing 5870?
1659 2011-02-04 15:28:27 <dishwara> anyone know please tell
1660 2011-02-04 15:28:27 <presence> I see lots of vendors selling the 5870
1661 2011-02-04 15:28:34 <presence> so I wouldnt worry
1662 2011-02-04 15:28:40 <presence> 5970 is another matter
1663 2011-02-04 15:29:15 <dishwara> oh, i confused 5870 & 5970
1664 2011-02-04 15:29:17 larsivi has quit (Read error: Connection reset by peer)
1665 2011-02-04 15:33:01 davex__ has quit (Remote host closed the connection)
1666 2011-02-04 15:33:29 davex__ has joined
1667 2011-02-04 15:33:48 <m0mchil> pushed fixed poclbm to github
1668 2011-02-04 15:35:40 <rli> m0mchil: what is fixed?
1669 2011-02-04 15:36:26 <m0mchil> some results lost due single output per kernel run
1670 2011-02-04 15:37:47 <molecular> m0mchil, allright, seems to make it faster herer
1671 2011-02-04 15:38:05 <sipa> no way you can have noticed that by now
1672 2011-02-04 15:38:07 <molecular> did you do anything else since about 3 days ago?
1673 2011-02-04 15:38:08 <m0mchil> it won't be faster visibly
1674 2011-02-04 15:38:20 <molecular> it shouldn't affect hashrate, should it?
1675 2011-02-04 15:38:23 <sipa> no
1676 2011-02-04 15:38:32 <sipa> only number of submitted shares
1677 2011-02-04 15:38:39 <m0mchil> just will return slightly more 'shares'
1678 2011-02-04 15:38:39 <molecular> did anything else get changed the last 2 or 3 days?
1679 2011-02-04 15:38:58 <m0mchil> no
1680 2011-02-04 15:39:10 <molecular> because it got faster, definitely. hashrate is now roughly equal to Diablo's. was lower before
1681 2011-02-04 15:39:12 <molecular> weird
1682 2011-02-04 15:39:35 <sipa> stochastics :)
1683 2011-02-04 15:39:44 <molecular> on the hashrate?
1684 2011-02-04 15:39:53 <sipa> hmmm
1685 2011-02-04 15:41:05 <molecular> what happens if one uses the same worker account on slush's pool for 2 miners?
1686 2011-02-04 15:41:29 <slush> molecular: in the worst case, some shares will be rejected
1687 2011-02-04 15:42:04 <molecular> the overflow warnings in m0mchill's miner are harmless, right? think I'll remove these...
1688 2011-02-04 15:43:02 <m0mchil> molecular, what's your setup?
1689 2011-02-04 15:44:39 <Diablo-D3> Ooob'[k
1690 2011-02-04 15:45:04 <m0mchil> cat?
1691 2011-02-04 15:45:23 <presence> which github has the poclbm updates?
1692 2011-02-04 15:46:01 <m0mchil> git://github.com/m0mchil/poclbm.git
1693 2011-02-04 15:46:01 dishwara has quit (Read error: Connection reset by peer)
1694 2011-02-04 15:46:32 dishwara has joined
1695 2011-02-04 15:46:44 frewsxcv has quit (Ping timeout: 240 seconds)
1696 2011-02-04 15:46:56 dishwara has left ()
1697 2011-02-04 15:47:15 <presence> m0m: any special process needed to run on windows?
1698 2011-02-04 15:48:21 frewsxcv has joined
1699 2011-02-04 15:48:29 <m0mchil> latest binary not released yet
1700 2011-02-04 15:48:51 <presence> ah ok
1701 2011-02-04 15:49:33 <m0mchil> if you've run pure python poclbm on win previously - no
1702 2011-02-04 15:49:45 <presence> I have actual python
1703 2011-02-04 15:49:50 <presence> but I havent tried to run it
1704 2011-02-04 15:49:52 skeledrew1 has quit (Read error: Connection reset by peer)
1705 2011-02-04 15:50:58 skeledrew has joined
1706 2011-02-04 15:52:25 <presence> are those three files the only three needed?
1707 2011-02-04 15:52:45 <Diablo-D3> grr
1708 2011-02-04 15:52:48 genjix has quit (Remote host closed the connection)
1709 2011-02-04 15:54:29 <m0mchil> better wait for binary
1710 2011-02-04 15:54:34 <sipa> slush: i haven't done the calculation to see whether you still gain from using the 43% technique when only-shares-for-current-block would be in effect, but at least it won't kill the pool if everyone did it
1711 2011-02-04 15:54:52 <presence> heh
1712 2011-02-04 15:57:10 <presence> 1 more block cleared and Ill have 100BTC :D
1713 2011-02-04 16:01:40 lolcat has joined
1714 2011-02-04 16:01:58 <lolcat> What is the price of bitcoins and are the API fixed?
1715 2011-02-04 16:02:30 <sipa> the price varies, but currently a lot of trades around 0.8 USD/BTC
1716 2011-02-04 16:02:39 <presence> my daily reward has been falling :(
1717 2011-02-04 16:02:43 bulletbill has joined
1718 2011-02-04 16:04:41 <lolcat> sipa: WHAT? I bougth my bitcoins at 0.252 or something
1719 2011-02-04 16:05:00 <EvanR-work> well good
1720 2011-02-04 16:05:12 <EvanR-work> you can sell now if you think its a peak ;)
1721 2011-02-04 16:05:48 <slush> EvanR-work: if you don't think this is a peak, please buy 110 BTC right now :)
1722 2011-02-04 16:05:59 <EvanR-work> haha
1723 2011-02-04 16:06:14 <EvanR-work> another reason i like your pool is the feedback
1724 2011-02-04 16:06:27 <EvanR-work> bitcoind and the miners i tried dont tell me enough
1725 2011-02-04 16:06:50 <EvanR-work> <120 confirmations the generation doesnt even show up
1726 2011-02-04 16:07:12 <slush> yes, bitcoind has troubles with generated blocks
1727 2011-02-04 16:09:28 <lolcat> Does anyone know how many bitcoins I have purchased?
1728 2011-02-04 16:09:38 <EvanR-work> you do?
1729 2011-02-04 16:10:14 <slush> lolcat: on mtgox?
1730 2011-02-04 16:12:00 <lolcat> Nope, in the bazar channel
1731 2011-02-04 16:12:15 <lolcat> I think I backed mt wallet up to my gmail I hope
1732 2011-02-04 16:13:21 <lolcat> I backed it up to somewhere clever I remember, can't remember where though
1733 2011-02-04 16:13:32 <EvanR-work> gmail doesnt sound too safe
1734 2011-02-04 16:13:37 <EvanR-work> unless you encrypted it?
1735 2011-02-04 16:13:55 <lolcat> Don't think so, but if I can't find it, neither can you
1736 2011-02-04 16:15:00 <EvanR-work> google can
1737 2011-02-04 16:15:23 <lolcat> They allready have my credit card information
1738 2011-02-04 16:15:25 <afed> ;;bc,stats
1739 2011-02-04 16:15:27 <gribble> Current Blocks: 106175 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 672 blocks | Next Difficulty In About: 4 days, 2 hours, 33 minutes, and 36 seconds | Next Difficulty Estimate: 24674.40961031
1740 2011-02-04 16:15:41 <EvanR-work> lolcat: slippery slope eh?
1741 2011-02-04 16:15:50 bulletbill has left ()
1742 2011-02-04 16:15:53 <lolcat> EvanR-work: Yep
1743 2011-02-04 16:15:58 bulletbill1 has joined
1744 2011-02-04 16:16:49 bulletbill1 is now known as bulletbill
1745 2011-02-04 16:16:52 <lolcat> Wonder where my backup really is
1746 2011-02-04 16:17:07 <EvanR-work> when uploading your wallet to any server you dont have control of, better encrypt it
1747 2011-02-04 16:18:22 <lolcat> I guess I will have to dig the harddrive out of my busted laptop
1748 2011-02-04 16:18:28 <lolcat> But it is encrypted...
1749 2011-02-04 16:19:29 <lolcat> I have 35 bitcoins on mtgox
1750 2011-02-04 16:20:08 <EvanR-work> is in your wallet or on mtgox
1751 2011-02-04 16:20:16 bulletbill has left ()
1752 2011-02-04 16:20:17 <EvanR-work> cant be both
1753 2011-02-04 16:20:38 <lolcat> I can have 35 bitcoins in my wallet, and 35 bitcoins on mtgox
1754 2011-02-04 16:20:47 bulletbill has joined
1755 2011-02-04 16:21:04 <lolcat> I think I have around 80 in my wallet though, and 35 on mtgox
1756 2011-02-04 16:21:34 <EvanR-work> you have 80 coins in the wallet and you didnt take carefully make a backup
1757 2011-02-04 16:21:41 <EvanR-work> s/take //
1758 2011-02-04 16:22:03 bulletbill has left ()
1759 2011-02-04 16:22:15 <lolcat> Well, now they are actually worth something
1760 2011-02-04 16:22:38 <lolcat> And I know I have a backup, I just dont remember where
1761 2011-02-04 16:22:40 <EvanR-work> 80 at 25 cents would be worth something to me xD
1762 2011-02-04 16:23:11 <lolcat> Still, it is the kind of money I would spend on a beer.
1763 2011-02-04 16:23:24 <lolcat> I don't backup my beer
1764 2011-02-04 16:23:24 <EvanR-work> ah, that explains that
1765 2011-02-04 16:23:42 <EvanR-work> when you got lots of money you tend to misplace it
1766 2011-02-04 16:23:49 <EvanR-work> lol
1767 2011-02-04 16:23:57 <lolcat> But now, with 120btc at 0,8 I wouldn't mind selling
1768 2011-02-04 16:23:59 andrew12 has joined
1769 2011-02-04 16:25:42 Kiba has joined
1770 2011-02-04 16:26:34 <lolcat> I need to finance my VPS'ing needs
1771 2011-02-04 16:27:01 jon_smark has joined
1772 2011-02-04 16:29:00 andrew12 has quit (school!d8241fa2@gateway/web/freenode/ip.216.36.31.162|Quit: Page closed)
1773 2011-02-04 16:30:52 <lolcat> How weird, on my vps I could download bitcoin in 1 second, on my home connection it takes 15
1774 2011-02-04 16:30:54 <Vladimir> nice,,, had a little nap, wake up and hey BTC is trading 10-15% higher, should sleep more.... and more often... lol
1775 2011-02-04 16:30:55 <gribble> Error: "," is not a valid command.
1776 2011-02-04 16:31:47 <Kiba> Vladimir: isn't it awesome? My saving jumped 10-15% in value overnight!
1777 2011-02-04 16:33:17 jon_smark has quit (Quit: Page closed)
1778 2011-02-04 16:33:23 <luke-jr> Market is actually down today
1779 2011-02-04 16:33:35 <Vladimir> yea, good stuff... not worth losing any sleep though over day to day moves up and down.... let's's see if it hits 10$ in 12 month time...
1780 2011-02-04 16:34:24 <Kiba> luke-jr: down today?
1781 2011-02-04 16:34:41 * Kiba finding it hard to accumlate bitcoin
1782 2011-02-04 16:35:00 <Vladimir> he's probably measuring high of the day or something like that
1783 2011-02-04 16:35:38 <luke-jr> Kiba: yeah, it was .88, now it's .74
1784 2011-02-04 16:36:43 <Kiba> I am reading .8
1785 2011-02-04 16:37:28 <Kiba> the market is very wild these days
1786 2011-02-04 16:37:55 <tcatm> same for traffic to bitcoincharts so there might a lot of new users
1787 2011-02-04 16:38:13 <Kiba> reddit, tcatm
1788 2011-02-04 16:38:20 <luke-jr> Kiba: where are you reading .8?
1789 2011-02-04 16:38:25 <Kiba> bitcoincharts
1790 2011-02-04 16:38:36 <luke-jr> PayPalUSD Bid=0.6520, Ask=0.7400, Price=0.7500, Volume=200
1791 2011-02-04 16:38:46 <luke-jr> well it's wrong XD
1792 2011-02-04 16:38:56 <Kiba> what market are you watching?
1793 2011-02-04 16:38:56 <tcatm> ;;bc,markets
1794 2011-02-04 16:38:57 <gribble> bcEUR: 0.6 (165 BTC) 0.40001/0.6 | bcLREUR: 0.3 (113.067 BTC) 0.3/0.6 | bcLRUSD: 0.78 (700 BTC) 0.75/0.88 | bcmLRUSD: 0.41 (500 BTC) 0.51/1 | bcmPPUSD: 0.5102 (100 BTC) 0.652/0.74 | bcmPXGAU: 0.0065 (1000 BTC) 0.0046/None | btcexJPY: 2 (10 BTC) None/None | btcexRUB: 24.31 (0.95 BTC) None/None | btcexWMR: 24 (0.5 BTC) None/None | mtgoxUSD: 0.8 (29245.5 BTC) 0.795/0.8
1795 2011-02-04 16:38:58 <luke-jr> I'm reading https://www.bitcoinmarket.com/home/markets/
1796 2011-02-04 16:38:59 <Kiba> I am watching mtgox
1797 2011-02-04 16:39:16 <luke-jr> [11:27:28] <luke-jr> Market is actually down today
1798 2011-02-04 16:39:21 <luke-jr> I said Market, not MtGox
1799 2011-02-04 16:39:41 <Kiba> tcatm: the data on bitcoin market seem to be stale
1800 2011-02-04 16:40:03 <tcatm> it is. I'm still not using the realtime api
1801 2011-02-04 16:40:51 <Kiba> did bitcoinmarket moved to the next version of their site or something?
1802 2011-02-04 16:41:17 <tcatm> still seems to be the old site
1803 2011-02-04 16:42:13 <Kiba> it seem that Euro is trailing mtgox
1804 2011-02-04 16:42:32 <Kiba> USD
1805 2011-02-04 16:43:38 bulletbill has joined
1806 2011-02-04 16:46:05 <Vladimir> http://www.taters.net/cgi-bin/btc/matrix.pl?axisinc=0.01
1807 2011-02-04 16:51:11 m0mchil has quit ()
1808 2011-02-04 17:00:27 <tcatm> anyone want to earn 500 BTC?
1809 2011-02-04 17:00:42 <sipa> for?
1810 2011-02-04 17:00:58 <tcatm> fixing the audio driver for my computer so it works with pulseaudio
1811 2011-02-04 17:00:59 <luke-jr> "
1812 2011-02-04 17:01:12 <luke-jr> meh, that'd require more than 500 BTC for me
1813 2011-02-04 17:01:13 <rli> tcatm: for sure -> 1J1cz9HUkaitMvHfAFVWYZN31YZtsQw2jc thx :)
1814 2011-02-04 17:01:26 <luke-jr> tcatm: pulseaudio is crap
1815 2011-02-04 17:01:37 <luke-jr> I won't use it.
1816 2011-02-04 17:01:54 <tcatm> I want to play audio over wifi from my laptop
1817 2011-02-04 17:02:24 mtgox has quit (Ping timeout: 240 seconds)
1818 2011-02-04 17:02:40 <luke-jr> um, so?
1819 2011-02-04 17:03:04 <tcatm> pulseaudio can do that
1820 2011-02-04 17:03:07 <luke-jr> so can mpd
1821 2011-02-04 17:03:18 <tcatm> ?
1822 2011-02-04 17:03:21 <presence> new rpcminer is much more consistent in hash processing
1823 2011-02-04 17:03:30 <tcatm> mpd's already running on the box
1824 2011-02-04 17:03:53 <luke-jr> so use it
1825 2011-02-04 17:04:03 <tcatm> how do I redirect audio to it?
1826 2011-02-04 17:04:08 <sipa> mplayer -ao pcm:wavheader:file=/dev/stdout | ssh laptop 'mplayer -ac /dev/stdin'
1827 2011-02-04 17:04:12 <sipa> :p
1828 2011-02-04 17:04:37 <sipa> eh
1829 2011-02-04 17:04:52 <luke-jr> tcatm: you add the files and play them
1830 2011-02-04 17:05:12 <tcatm> doesn't really work for movies or youtube videos
1831 2011-02-04 17:06:21 <luke-jr> then you use sipa's method
1832 2011-02-04 17:06:22 <luke-jr> :p
1833 2011-02-04 17:06:43 <Kiba> tcatm: hmm, I don't know how to fix pulseaudio
1834 2011-02-04 17:06:56 <tcatm> either pulseaudio or my driver is bad
1835 2011-02-04 17:07:50 <Kiba> I don't know how to fix scary jobs like that
1836 2011-02-04 17:07:55 <Kiba> however, I can program
1837 2011-02-04 17:07:59 <Kiba> sites
1838 2011-02-04 17:08:05 <Kiba> let me know if you have a job for me
1839 2011-02-04 17:12:58 sotto has quit (Quit: Quitte)
1840 2011-02-04 17:13:47 devrandom has quit (Remote host closed the connection)
1841 2011-02-04 17:15:17 rli has quit (Quit: leaving)
1842 2011-02-04 17:17:12 <luke-jr> tcatm: 500 BTC is like 8-10 hours of my time; I have serious doubts I could figure out your problem in that time. sorry.
1843 2011-02-04 17:17:22 <luke-jr> maybe someone familiar with pulseaudio could do it that quick
1844 2011-02-04 17:18:08 lumos has joined
1845 2011-02-04 17:18:18 mtgox has joined
1846 2011-02-04 17:18:42 <lolcat> I wonder what service bitcoin needs
1847 2011-02-04 17:22:41 alystair has joined
1848 2011-02-04 17:24:51 <luke-jr> lolcat: apparently pulseaudio fixing
1849 2011-02-04 17:25:07 <tcatm> or fixing the driver
1850 2011-02-04 17:25:27 <tcatm> I *think* I messed something about while allocating the audio buffer but it works fine with MPD
1851 2011-02-04 17:26:04 bitanarchy has joined
1852 2011-02-04 17:32:57 dukeleto has quit (Excess Flood)
1853 2011-02-04 17:33:28 dukeleto has joined
1854 2011-02-04 17:34:04 <lolcat> What is the two other somewhat interessting bitcoin channels?
1855 2011-02-04 17:35:09 <ducki2p> you found them :)
1856 2011-02-04 17:37:17 ApertureScience has quit (Remote host closed the connection)
1857 2011-02-04 17:37:34 ApertureScience has joined
1858 2011-02-04 17:41:30 chaord has joined
1859 2011-02-04 17:44:45 alystair has quit (Ping timeout: 240 seconds)
1860 2011-02-04 17:47:10 Kiba has quit (Remote host closed the connection)
1861 2011-02-04 17:47:47 <slush> does bitcoin generate new receiving address for every mined block?
1862 2011-02-04 17:47:54 <slush> I think so
1863 2011-02-04 17:49:35 Zarutian has joined
1864 2011-02-04 17:51:28 <slush> tcatm: ---^
1865 2011-02-04 17:52:17 <EvanR-work> er slush
1866 2011-02-04 17:53:56 <slush> EvanR-work: ?
1867 2011-02-04 17:54:09 <EvanR-work> im not getting any shares
1868 2011-02-04 17:54:19 <EvanR-work> and the expected reward is not showing
1869 2011-02-04 17:54:23 <EvanR-work> all my shit is running...
1870 2011-02-04 17:54:44 <slush> yes, I introduced first anti-cheating things
1871 2011-02-04 17:54:49 <slush> counting shares will be back soon
1872 2011-02-04 17:54:58 <EvanR-work> interesting
1873 2011-02-04 17:55:15 <slush> as I fixed counting from round-oriented to time-oriented
1874 2011-02-04 17:55:34 <slush> sorry for temporary troubles, I'd like to protect honest pool users
1875 2011-02-04 17:56:27 <Vladimir> slush, isn't every new block generation reward goes to the same bitcoin address of yours?
1876 2011-02-04 17:56:39 <luke-jr> hey
1877 2011-02-04 17:56:47 <luke-jr> who made that real-physical bitcoin thing?
1878 2011-02-04 17:56:52 <nanotube> luke-jr: tcatm
1879 2011-02-04 17:56:53 <Vladimir> or it is just after 120 blocks?
1880 2011-02-04 17:56:58 <luke-jr> tcatm: ping
1881 2011-02-04 17:57:19 <nanotube> luke-jr: what you want to put in an order for another one? :)
1882 2011-02-04 17:57:28 <luke-jr> nanotube: I want to put in an order for bulk
1883 2011-02-04 17:57:38 <nanotube> ah cool... whatcha gonna do with them?
1884 2011-02-04 17:57:44 <luke-jr> teach my children
1885 2011-02-04 17:58:09 <nanotube> nice :)
1886 2011-02-04 17:58:19 <luke-jr> this way, I can use different coins for decimal currency (USD) and tonal currency (bitcoin)
1887 2011-02-04 17:58:43 <luke-jr> not sure how practical it would be for tcatm to make them though
1888 2011-02-04 17:58:49 <EvanR-work> uhm written notation doesnt change the amount of currency
1889 2011-02-04 17:58:55 <EvanR-work> you are still using the same currency
1890 2011-02-04 17:59:22 <EvanR-work> oh wait
1891 2011-02-04 17:59:24 <luke-jr> EvanR-work: there are 100 (256 decimal) divisions of a TBC
1892 2011-02-04 17:59:30 <luke-jr> and only 100 decimal divisions of a USD
1893 2011-02-04 17:59:40 <EvanR-work> you are indoctrinating your children with this shit?
1894 2011-02-04 17:59:50 <luke-jr> it's called teaching
1895 2011-02-04 18:00:01 <EvanR-work> i pray for their future
1896 2011-02-04 18:00:05 <luke-jr> thanks
1897 2011-02-04 18:00:07 <EvanR-work> lol
1898 2011-02-04 18:00:48 <lolcat> bitcoind getnewadress <account>, how do I make the account?
1899 2011-02-04 18:01:14 <luke-jr> tcatm: would it be practical to make 16 TBC coins, 32 TBCᵗ coins, and 64 TBCˢ coins? (total 112 coins)
1900 2011-02-04 18:01:21 <luke-jr> lolcat: like that
1901 2011-02-04 18:02:03 <lolcat> lolcat@server11:~/bitcoin-0.3.19/bin/32$ ./bitcoind getnewadress 1
1902 2011-02-04 18:02:03 <lolcat> error: {"code":-32601,"message":"Method not found"}
1903 2011-02-04 18:02:17 lumos has quit (Quit: Leaving)
1904 2011-02-04 18:04:00 <slush> Vladimir: afaik, bitcoin generates new address for every block
1905 2011-02-04 18:04:28 <newsham> i'm not sure how anyone could use BTC for normal purchases rigt now with so much volatility and deflation in btc value
1906 2011-02-04 18:05:05 <EvanR-work> yesterday we discussed a way for a retailer to stabilize prices
1907 2011-02-04 18:05:08 <newsham> if you listed a tshirt at 20btc today ($15) and sold it next week, you'd only get $10
1908 2011-02-04 18:05:43 <EvanR-work> if someone offers you rights to sell btc at a fixed price for a predetermine time period, then you can base your prices on that and simplify your books
1909 2011-02-04 18:05:53 <EvanR-work> that someone deals with the market and shit
1910 2011-02-04 18:05:55 <EvanR-work> for you
1911 2011-02-04 18:05:59 <presence> why would someone do that?
1912 2011-02-04 18:06:03 <EvanR-work> to make a profit
1913 2011-02-04 18:06:05 <newsham> so you want futures on $/btc market?
1914 2011-02-04 18:06:12 <EvanR-work> right
1915 2011-02-04 18:06:39 <newsham> i bet that would drive up speculative bubble even more :)
1916 2011-02-04 18:06:40 <EvanR-work> this is for retailers who will mostly want to just accept btc and convert straight to dollars for whatever reason
1917 2011-02-04 18:06:42 <MacRohard> the risk is probably unmanageable right now
1918 2011-02-04 18:06:58 <MacRohard> the futures traders could become bitcoin insolvent very easily
1919 2011-02-04 18:07:19 <EvanR-work> it has some risk
1920 2011-02-04 18:07:38 <newsham> also there's barely enough liquidity in $/btc to sustain a market right now.  i doubt there'd be enough in futures, especially if you had markets for several different time periods
1921 2011-02-04 18:07:55 <newsham> the bid/ask spreads would be huge
1922 2011-02-04 18:08:11 <EvanR-work> we suggested to an online retail guy 1 week intervals
1923 2011-02-04 18:08:30 <EvanR-work> since the price is generally increasing, it seems like a good idea
1924 2011-02-04 18:09:08 <EvanR-work> it would certainly have an effect on the market ;)
1925 2011-02-04 18:09:10 <newsham> right but you need to pick fixed expiration dates, so you need to keep starting up new markets for future weeks
1926 2011-02-04 18:09:28 <newsham> and you prob want at least two markets (soon to expire, and the week after that) so there's not a gap in coverage
1927 2011-02-04 18:09:29 <MacRohard> is there a mybitcoin or mtgox merchant thing that can autoconvert bitcoins to dollars?
1928 2011-02-04 18:09:30 <EvanR-work> id just renegotiate next week
1929 2011-02-04 18:09:38 <MacRohard> if they convert at the time of sale it doesn't matter to them
1930 2011-02-04 18:09:45 <EvanR-work> MacRohard: no, but theres a merchant service
1931 2011-02-04 18:09:52 <EvanR-work> and you can convert yourself at the desired time
1932 2011-02-04 18:09:56 <newsham> bitcoin design worried a lot about inflation but not too much about deflation :(
1933 2011-02-04 18:10:13 <newsham> anyway the current deflation is probably not fundamental, its probably all speculative
1934 2011-02-04 18:10:21 <newsham> maybe things will be better after a pop?
1935 2011-02-04 18:10:45 <MacRohard> well it shouldn't be a big deal.. if people are buying bitcoins in order to make purchases then the roundtrip in and out of bitcoin will happen fast enough (within like 30min) to hopefully not be that badly affected by unstable excahnge rates
1936 2011-02-04 18:10:49 <newsham> i guess i wouldnt mind selling some lnog term futures on expectations of a pop :)
1937 2011-02-04 18:11:10 <EvanR-work> MacRohard: i dont think the retails should be immediately converting
1938 2011-02-04 18:11:29 <newsham> how fast do most people get into and out of $ and btc?   i used an mtgox exchanger which took 2 weeks
1939 2011-02-04 18:11:39 <newsham> which is quite slow, especially with such deflation
1940 2011-02-04 18:11:44 <MacRohard> EvanR-work, well if their costs are in dollars they probably should convert enough to cover that at least
1941 2011-02-04 18:11:48 <presence> thats impossibly slow
1942 2011-02-04 18:12:02 <EvanR-work> MacRohard: eventually yes
1943 2011-02-04 18:12:07 <MacRohard> also if noone is selling their coins then there's nothing for people to buy
1944 2011-02-04 18:12:08 <EvanR-work> you need some buffer
1945 2011-02-04 18:12:23 <EvanR-work> no one selling coins is very unlikely
1946 2011-02-04 18:12:51 <newsham> btw, why does bitcoin use one of those silly web page forums for discussion?  cant we grow up and use the internet?
1947 2011-02-04 18:13:04 <EvanR-work> yeah there really should be an irc channel or something
1948 2011-02-04 18:13:19 draginx has joined
1949 2011-02-04 18:13:27 <newsham> i mean for persistent long term conversation
1950 2011-02-04 18:13:29 <MacRohard> i was thinking of trying to newgrp an alt.bitcoin usenet grup
1951 2011-02-04 18:13:30 <newsham> not ephemeral chat
1952 2011-02-04 18:13:34 <MacRohard> but i doubt anyone would use it
1953 2011-02-04 18:13:47 <EvanR-work> forums are the established way to do that, newsgroups are old school these days
1954 2011-02-04 18:13:55 <newsham> i would prefer mailing list.  usenet is dead, although i personally would subscribe to usenet over a web forum
1955 2011-02-04 18:14:00 <newsham> last thing i need is to join another web forum
1956 2011-02-04 18:14:16 <MacRohard> email is pretty fucked these days too
1957 2011-02-04 18:14:18 <newsham> ps: google groups == mailing list that is also a web forum
1958 2011-02-04 18:14:28 <EvanR-work> this particular channel is publically logged, so its not that ephemeral
1959 2011-02-04 18:14:29 <newsham> my email is definitely not fucked these days
1960 2011-02-04 18:14:38 <MacRohard> mine is
1961 2011-02-04 18:14:54 <EvanR-work> email is shit, if thats 'the real internet'
1962 2011-02-04 18:14:56 <EvanR-work> then shoot me
1963 2011-02-04 18:15:34 <presence> Im with newsh here
1964 2011-02-04 18:15:38 <presence> I hate the web
1965 2011-02-04 18:15:48 <EvanR-work> the web is also shit
1966 2011-02-04 18:15:51 <presence> if I could go back to archie and have the www discarded, I would be way happy
1967 2011-02-04 18:16:09 <newsham> the web is fine for some things
1968 2011-02-04 18:16:15 <EvanR-work> oh?
1969 2011-02-04 18:16:17 <newsham> but a login and password just to talk about bitcoins?
1970 2011-02-04 18:16:19 <presence> its fine for placing an order for something
1971 2011-02-04 18:16:35 <EvanR-work> wait, what is the web fine for? i havent figured that one out yet
1972 2011-02-04 18:16:44 <newsham> evanr: news.bbc.co.uk
1973 2011-02-04 18:16:47 <EvanR-work> it doesnt do anything right
1974 2011-02-04 18:17:01 <newsham> citeseer
1975 2011-02-04 18:17:21 <newsham> spec sheets and ordering pcbs and parts
1976 2011-02-04 18:17:30 <newsham> programming documentation
1977 2011-02-04 18:17:33 <newsham> need i go on?
1978 2011-02-04 18:17:34 <EvanR-work> downloading documents
1979 2011-02-04 18:17:43 <EvanR-work> what does the web have to do with that besides the http download mechanism
1980 2011-02-04 18:18:05 <presence> its ok for real time streaming if you are one of the few that doesnt get hosed by your ISP
1981 2011-02-04 18:18:18 <newsham> evanr: indexed search, hyperlinks, renderable catalogs, etc
1982 2011-02-04 18:18:41 <EvanR-work> all of that would be better on a different platform
1983 2011-02-04 18:18:45 <newsham> you dont even need to "download" much of the docs, you can render directly in your web client and read
1984 2011-02-04 18:18:55 <newsham> "would be better" is besides the point
1985 2011-02-04 18:19:09 <EvanR-work> it works doesnt mean its not shitty
1986 2011-02-04 18:19:09 <presence> better on what different platform?
1987 2011-02-04 18:19:10 <newsham> this is a silly argument..  nobody is saying the web is perfect, but you're deluded if you think its not useful
1988 2011-02-04 18:19:37 <presence> I would love to have a car that runs on fuelcells, but its not viable yet
1989 2011-02-04 18:19:42 <newsham> in many ways the web is a horrible piece of crap of a design
1990 2011-02-04 18:19:57 <presence> its horrible because web designers are horrible
1991 2011-02-04 18:20:03 <EvanR-work> not just web design
1992 2011-02-04 18:20:20 <newsham> but people dont use the web because of its design.  they use it because of its content
1993 2011-02-04 18:21:40 <EvanR-work> its just a classic case of worse is better, because all dignity was left at the door, it could be implemented in less than a decade and take over the world
1994 2011-02-04 18:21:52 <afed> slush:
1995 2011-02-04 18:22:03 <EvanR-work> how it still operates i dont know
1996 2011-02-04 18:22:11 <newsham> the perfect is the enemy of the good
1997 2011-02-04 18:22:33 <EvanR-work> another example is the php system
1998 2011-02-04 18:22:42 <EvanR-work> s/good/absolutely horrible/
1999 2011-02-04 18:23:06 <slush> afed: what
2000 2011-02-04 18:23:07 <presence> you have to shit or get off of the port
2001 2011-02-04 18:23:14 <presence> perfect never gets implemented
2002 2011-02-04 18:23:19 bulletbill has left ()
2003 2011-02-04 18:23:19 <presence> because its waiting for the fixes
2004 2011-02-04 18:23:24 <afed> slush: unhappy with lack of shares data but i scrolled up and see that that is temporary
2005 2011-02-04 18:23:31 <EvanR-work> haskell was implemented
2006 2011-02-04 18:23:35 <EvanR-work> ;)
2007 2011-02-04 18:23:41 <EvanR-work> and reimplemented
2008 2011-02-04 18:24:01 <newsham> evanr: go write code.
2009 2011-02-04 18:24:19 * EvanR-work looks at the other window with php code in it
2010 2011-02-04 18:24:20 <EvanR-work> :(
2011 2011-02-04 18:25:02 <newsham> i'm all for people making better systems.
2012 2011-02-04 18:25:33 <EvanR-work> i wouldnt mind if everyone agreed with newsham, about pragmatic matters necesitating shitifying of designs
2013 2011-02-04 18:25:51 <EvanR-work> however, most web developers think everything is amazing and the greatest thing since bread
2014 2011-02-04 18:26:14 <EvanR-work> things are the way they are because its the right way
2015 2011-02-04 18:27:00 <EvanR-work> theres economic inefficiency associated with getting these systems working and keeping them working. like friction keeping a wagon from being pulled
2016 2011-02-04 18:27:32 <EvanR-work> despite the validity of newtons first law of motion
2017 2011-02-04 18:28:09 <luke-jr> EvanR-work: re fixed BTC prices, retailers could just mark up a lot and get some kind of currency devaluation insurance :p
2018 2011-02-04 18:28:17 devrandom has joined
2019 2011-02-04 18:28:37 <EvanR-work> yes, but this would mean lower prices for customers
2020 2011-02-04 18:28:41 <luke-jr> offering bitcoin devaluation service can then be yet another way for people to invest in bitcoin's future
2021 2011-02-04 18:28:56 <presence> everything thats bad about the web in a single package?  Joomla.
2022 2011-02-04 18:28:56 <luke-jr> s/service/insurance/
2023 2011-02-04 18:29:04 <EvanR-work> insurance for this sounds like a scam really
2024 2011-02-04 18:29:07 <newsham> lukejr: futures trading
2025 2011-02-04 18:29:18 <luke-jr> EvanR-work: all insurance sounds like a scam :P
2026 2011-02-04 18:29:20 <newsham> not enough volume for a good market imo.
2027 2011-02-04 18:29:21 <EvanR-work> yeah
2028 2011-02-04 18:29:28 <EvanR-work> especially when they dont pay out
2029 2011-02-04 18:29:52 <luke-jr> well, the real problem is, bitcoin devaluation insurance would have to payout in a non-bitcoin currency to be worth anything
2030 2011-02-04 18:30:09 <newsham> luke: nah..
2031 2011-02-04 18:30:12 <EvanR-work> newsham: well if someone is selling stuff for bitcoins and periodically converting to usd, that would obviously increase the volume
2032 2011-02-04 18:31:26 <newsham> luke: i'll sell you march 1st 2011 BTC for $5 a pop right now
2033 2011-02-04 18:31:49 <EvanR-work> newsham: note that i wasnt talking about futures
2034 2011-02-04 18:31:54 <EvanR-work> more like put options
2035 2011-02-04 18:32:09 <newsham> did you want futures (buy march 1st 2011 btc) or did you want options (buy the right to buy btc by march 1st)?
2036 2011-02-04 18:32:23 <newsham> futures would be easier for retailers
2037 2011-02-04 18:32:29 <newsham> and options would require way more options
2038 2011-02-04 18:32:30 <EvanR-work> why?
2039 2011-02-04 18:32:31 <newsham> way more markets
2040 2011-02-04 18:32:51 <newsham> because you'll need a put market and a call market for each of the strikes and for each of the expiration months you're interested in
2041 2011-02-04 18:32:52 <EvanR-work> theres no markets, i just offer you 75 cents per coin for the next week
2042 2011-02-04 18:32:59 <newsham> vs. one market for each expiration month for futures
2043 2011-02-04 18:33:18 TheAncientGoat has quit (Ping timeout: 276 seconds)
2044 2011-02-04 18:33:39 <newsham> ie "the right to buy btc for $1.10 on march 1st 2011" would be one market
2045 2011-02-04 18:33:41 <lolcat> I sell options if anyone is interessted!
2046 2011-02-04 18:33:53 <newsham> "the right to buy btc for $1.20 on march 1st 2011" would be another market
2047 2011-02-04 18:34:05 <EvanR-work> sounds pretty complex
2048 2011-02-04 18:34:21 <newsham> i guess if you just wanted to do OTC trades you could just offer bespoke contracts
2049 2011-02-04 18:34:36 <EvanR-work> yes
2050 2011-02-04 18:34:39 <newsham> but you'd pay a higher premium unless you found a sucker.
2051 2011-02-04 18:35:06 <lolcat> I could sell anyone an option to buy my bitcoins for a certain price march 10th, so that if they don't buy at said price I get 1% if they do buy I get said price
2052 2011-02-04 18:35:13 <newsham> futures work very well for retailers, btw..  they know they will convert some btc to $ on march 1st, so they base their prices on march 1st btc futures
2053 2011-02-04 18:35:30 <newsham> the futures market was designed just for that purpose (ie. selling my corn when it comes to harvest)
2054 2011-02-04 18:35:49 <EvanR-work> well i was told he wanted money *now8
2055 2011-02-04 18:35:50 <EvanR-work> *
2056 2011-02-04 18:36:04 <EvanR-work> theres no corn
2057 2011-02-04 18:36:18 <newsham> well you want price stability over a reasonable period (like a month, or at least a week) right?
2058 2011-02-04 18:36:28 <EvanR-work> who?
2059 2011-02-04 18:36:31 <newsham> retailer
2060 2011-02-04 18:36:35 <EvanR-work> yes
2061 2011-02-04 18:36:49 <EvanR-work> and you want the usd as soon as youre paid
2062 2011-02-04 18:36:58 * ducki2p loves porkbelly futures
2063 2011-02-04 18:37:04 <EvanR-work> or whenever you want
2064 2011-02-04 18:37:20 <EvanR-work> before the deadline
2065 2011-02-04 18:39:33 <newsham> yah i guess options are the best fit
2066 2011-02-04 18:39:45 <newsham> but you'd need way more volume in $/btc to support options on it
2067 2011-02-04 18:40:08 <lolcat> Why?
2068 2011-02-04 18:40:12 <newsham> look at how disconnected and volatile the current $/btc market is http://www.thenewsh.com/~newsham/x/mtgox/
2069 2011-02-04 18:40:39 <newsham> the bid ask spread is huge (2c on 78c price right now, often much bigger)
2070 2011-02-04 18:41:02 <EvanR-work> whats this graph, spread?
2071 2011-02-04 18:41:14 <newsham> now pick a single expiration date and a single strike price, how many participants will be in that market?
2072 2011-02-04 18:41:41 <newsham> thats the mtgox $/btc market.  its just showing the trades in the graph
2073 2011-02-04 18:41:51 <newsham> the black bars show the min to max price for that time period
2074 2011-02-04 18:42:05 <newsham> the blue tick shows the first price and the red tick shows the last price for that time period
2075 2011-02-04 18:42:27 <newsham> the light grey lines just connect the trading periods, and cover time periods where there were no trades
2076 2011-02-04 18:42:28 <EvanR-work> the grey lines are annoying and wrong ;)
2077 2011-02-04 18:42:44 <EvanR-work> they travel back in time sometimes
2078 2011-02-04 18:42:45 <newsham> the grey lines are periods where there is no market price
2079 2011-02-04 18:43:16 <Diablo-D3> [01:27:44] <newsham> ie "the right to buy btc for $1.10 on march 1st 2011" would be one market
2080 2011-02-04 18:43:16 <Diablo-D3> [01:27:46] <lolcat> I sell options if anyone is interessted!
2081 2011-02-04 18:43:16 <Diablo-D3> [01:27:58] <newsham> "the right to buy btc for $1.20 on march 1st 2011" would be another market
2082 2011-02-04 18:43:17 <Diablo-D3> erm
2083 2011-02-04 18:43:24 <Diablo-D3> all you're doing is futures.
2084 2011-02-04 18:43:39 <newsham> diablo: yah i was suggesting futures
2085 2011-02-04 18:43:44 <newsham> others said they'd prefer options
2086 2011-02-04 18:43:45 <Diablo-D3> futures suck
2087 2011-02-04 18:43:52 <Diablo-D3> most people do NOT understand how to use them
2088 2011-02-04 18:43:55 <Diablo-D3> including most of wall street
2089 2011-02-04 18:44:01 <EvanR-work> heh
2090 2011-02-04 18:44:09 <newsham> evanr: i should point out that todays market was more active than normal.. many days have had much fewer trades
2091 2011-02-04 18:44:18 <newsham> diablo: same for options.
2092 2011-02-04 18:44:23 <Diablo-D3> I agree
2093 2011-02-04 18:44:30 <lolcat> Diablo-D3: I though an option was the rigth to buy something at a later time
2094 2011-02-04 18:44:55 <Diablo-D3> lolcat: yes, but he described a future, not an option
2095 2011-02-04 18:44:55 davout has joined
2096 2011-02-04 18:44:56 <newsham> lolcat: yup.  or the right to sell something at a later time.
2097 2011-02-04 18:45:07 <newsham> it can also be the right to buy or sell something up to and including the later time
2098 2011-02-04 18:45:34 <EvanR-work> newsham: so what youre saying is youd expect such a business model to fail because they would cause the price to drop just by operating?
2099 2011-02-04 18:45:41 <EvanR-work> because no one is there to buy?
2100 2011-02-04 18:45:54 <Diablo-D3> yeah pretty much
2101 2011-02-04 18:46:04 <lolcat> But options can make bitcoins really expencive like tunips in Holland
2102 2011-02-04 18:46:10 <newsham> evanr: no, i would expect that you couldnt sustain a marketplace for options
2103 2011-02-04 18:46:12 <Diablo-D3> the speculation is keeping btc prices normal due to the low volume skewing it back down
2104 2011-02-04 18:46:15 <Diablo-D3> lolcat: I agree
2105 2011-02-04 18:46:20 <newsham> because individual options wouldnt have enough buyers and sellers to fix at a stable price
2106 2011-02-04 18:46:23 <Diablo-D3> we dont want options nor futures due to the low volume
2107 2011-02-04 18:46:34 <Diablo-D3> and we'll never have enough volume
2108 2011-02-04 18:46:40 <Diablo-D3> lets face it, bitcoin is too cool for normal people
2109 2011-02-04 18:46:44 <EvanR-work> newsham: i wasnt really suggesting a marketplace for options
2110 2011-02-04 18:46:52 <EvanR-work> i was talking about an individual agreement with one retailers
2111 2011-02-04 18:47:02 <afed> how do you all interpet the recent higher price?
2112 2011-02-04 18:47:06 <Diablo-D3> EvanR-work: private agreements between bitsnobs is their own business
2113 2011-02-04 18:47:12 <EvanR-work> right
2114 2011-02-04 18:47:17 <newsham> evanr: you could definitely do that, but without competition you're not likely to get good pricing
2115 2011-02-04 18:47:18 <Diablo-D3> heee bitsnobs
2116 2011-02-04 18:47:22 <Diablo-D3> I just coined a new word
2117 2011-02-04 18:47:23 <afed> i was suprised by it but i was happy to sell
2118 2011-02-04 18:47:24 kisom_dev has quit (Ping timeout: 240 seconds)
2119 2011-02-04 18:47:29 <EvanR-work> newsham: as a retailer?
2120 2011-02-04 18:47:49 <newsham> evanr: both as a buyer and a seller..  if buyers and sellers cant agree on a price, there is no transaction
2121 2011-02-04 18:47:56 <EvanR-work> ?
2122 2011-02-04 18:48:04 <tcatm> luke-jr: ping me in 2 hours (yes, I can manufacture more coins)
2123 2011-02-04 18:48:10 <EvanR-work> it cant be a bad price for both
2124 2011-02-04 18:48:12 <newsham> "i want to buy insurance" "ok give me $100M"  "oh, I didnt want it that badly, thanks anyway"
2125 2011-02-04 18:48:31 <EvanR-work> duh
2126 2011-02-04 18:48:56 <newsham> ok, i've explained my position, i'm done arguing.
2127 2011-02-04 18:51:57 <EvanR-work> without a market, im thinking one of the partys might get a bad deal in absolute terms
2128 2011-02-04 18:52:06 <EvanR-work> but if the retail is making enough
2129 2011-02-04 18:52:10 <newsham> btw, I will sell the right to sell 30 btc at $0.75 on march 1st for $1.25
2130 2011-02-04 18:52:11 <EvanR-work> dont matter ;)
2131 2011-02-04 18:52:14 <newsham> if anyone's interested
2132 2011-02-04 18:52:42 <slush> tcatm: We can play the game. Everytime you tip new pool round with precision to minute, I'll pay you 1btc
2133 2011-02-04 18:53:03 <lolcat> Diablo-D3: I would love to sell my bitcoin for hunreds of barrels of beer.
2134 2011-02-04 18:53:06 <slush> and if you don't tip it correctly, you will pay me :)
2135 2011-02-04 18:53:20 <Diablo-D3> so guys
2136 2011-02-04 18:53:28 <Diablo-D3> have we decided that losing 0.05% of hashes is fine?
2137 2011-02-04 18:53:32 <EvanR-work> lolcat: i want to make a barter site ;)
2138 2011-02-04 18:53:45 <Diablo-D3> 64 outputs has zero effect on miner performance, but 1024 does
2139 2011-02-04 18:54:09 <Diablo-D3> thats 2**6 vs 2**10
2140 2011-02-04 18:54:23 <Diablo-D3> so where it effects it is either, 7, 8, or 9
2141 2011-02-04 18:54:40 <lolcat> EvanR-work: Beer4bitcoin.com? :P
2142 2011-02-04 18:54:41 <slush> tcatm: I offer budget of, say, 20 BTC to this game
2143 2011-02-04 18:54:56 <EvanR-work> lolcat: xfory.com ;)
2144 2011-02-04 18:54:56 <Diablo-D3> and 2**6 is 0.05% lost vs 3.5%
2145 2011-02-04 18:55:03 * EvanR-work runs to register that
2146 2011-02-04 18:55:30 <lolcat> Is there anything Norwegian you guys want to buy?
2147 2011-02-04 18:55:31 <Diablo-D3> arent all 5 letter domains already registered?
2148 2011-02-04 18:55:40 <presence> losing .05% might mean loss of the winning hash
2149 2011-02-04 18:55:41 <lolcat> I have a 4 letter domain!
2150 2011-02-04 18:55:42 <EvanR-work> lolcat: do you have that weird candy?
2151 2011-02-04 18:55:45 <lolcat> b1ag.com
2152 2011-02-04 18:55:47 <lolcat> EvanR-work: Wich one?
2153 2011-02-04 18:55:52 <Diablo-D3> presence: vs the existing 3.5%?
2154 2011-02-04 18:55:58 <EvanR-work> it tastes horrible and its very popular in sweden
2155 2011-02-04 18:56:08 <presence> why the hell is there a 3.5% loss
2156 2011-02-04 18:56:28 <Diablo-D3> presence: very long story, and this only applies to people with fast cards and small -f numbers
2157 2011-02-04 18:56:46 <Diablo-D3> presence: the slower the card or the larger the -f, the less of the loss
2158 2011-02-04 18:56:54 <lolcat> EvanR-work: What colour? Form?
2159 2011-02-04 18:57:14 <EvanR-work> sort of like taffy i think
2160 2011-02-04 18:57:15 <Diablo-D3> presence: basically, you can have kernel runs that have more than one H==0
2161 2011-02-04 18:57:21 <lolcat> EvanR-work: Liquorice?
2162 2011-02-04 18:57:24 <Diablo-D3> presence: multiples arent returned.
2163 2011-02-04 18:57:26 <EvanR-work> no
2164 2011-02-04 18:57:52 <presence> so what is a "low f"
2165 2011-02-04 18:57:54 <lolcat> No idea what you mean, but I could go to sweeden and get you some if you are buying
2166 2011-02-04 18:57:58 <Diablo-D3> presence: 1.
2167 2011-02-04 18:58:12 <presence> ah, I think I was running yours as -f 1000
2168 2011-02-04 18:58:24 <EvanR-work> lolcat: maybelol
2169 2011-02-04 18:58:29 <EvanR-work> if i could remember what its called
2170 2011-02-04 18:58:56 <lolcat> Is it chocholaty or syntehtic?
2171 2011-02-04 18:58:59 <lolcat> Or natural?
2172 2011-02-04 18:59:01 <Diablo-D3> presence: most people are only losing like, 1%
2173 2011-02-04 18:59:33 <presence> if its a pool 1% could suck
2174 2011-02-04 18:59:45 sabalaba has joined
2175 2011-02-04 18:59:54 <EvanR-work> lolcat: its made with this chemical...
2176 2011-02-04 19:00:06 <EvanR-work> ill look up
2177 2011-02-04 19:00:11 <lolcat> So its not fish? it is not dairy based?
2178 2011-02-04 19:00:19 <EvanR-work> doubt it
2179 2011-02-04 19:00:54 <lolcat> EvanR-work: Does it contain nuts?
2180 2011-02-04 19:01:27 <EvanR-work> no
2181 2011-02-04 19:01:47 <presence> ludefisk?
2182 2011-02-04 19:02:01 <lolcat> presence: I belive that contains fish...
2183 2011-02-04 19:02:08 <presence> fish and lye :D
2184 2011-02-04 19:02:16 <lolcat> Want to buy some?
2185 2011-02-04 19:02:38 <presence> um, hell no, that crap is worse than kim-che
2186 2011-02-04 19:02:45 <presence> and kim-che can be pretty nasty :D
2187 2011-02-04 19:05:46 <citiz3n> this is  a very strange discussion
2188 2011-02-04 19:06:04 <lolcat> Yes, and sadly not to helpfull for me to start retailing
2189 2011-02-04 19:08:01 <EvanR-work> ill get back to you on that
2190 2011-02-04 19:10:21 <lolcat> Ok, I have to see Black Ducks on the movie theater now
2191 2011-02-04 19:10:47 Worf has joined
2192 2011-02-04 19:12:47 akem has quit (Read error: Connection reset by peer)
2193 2011-02-04 19:14:04 <citiz3n> has anyone lost a graphics card mining?
2194 2011-02-04 19:14:09 <citiz3n> fan died, card fried, etc?
2195 2011-02-04 19:14:16 <citiz3n> i would think not, but im curious
2196 2011-02-04 19:14:26 <citiz3n> the chips must have overheat protection
2197 2011-02-04 19:16:52 <draginx> water cooling :D
2198 2011-02-04 19:17:49 <lolcat> citiz3n: My laptop died, not sure if it is related to mining though
2199 2011-02-04 19:18:32 <nanotube> newsham: there's a bitcoin-discussion googlegroup...
2200 2011-02-04 19:24:43 <citiz3n> your laptop video died completely?
2201 2011-02-04 19:25:10 lolcat has quit (Ping timeout: 260 seconds)
2202 2011-02-04 19:25:25 <citiz3n> that almost answers that question ;)
2203 2011-02-04 19:32:07 larsivi has joined
2204 2011-02-04 19:35:25 <ArtForz> killed a 5770 and a 5970 so far
2205 2011-02-04 19:36:41 <nanotube> <Diablo-D3> lets face it, bitcoin is too cool for normal people  <-- so was the internet, just a couple short decades ago.
2206 2011-02-04 19:38:47 <slush> citiz3n: unless you are not overvoltaging, it should be safe
2207 2011-02-04 19:39:09 <ArtForz> yep
2208 2011-02-04 19:39:09 <slush> citiz3n: 5970 has temperature sensors and start throttling when VRM goes too high
2209 2011-02-04 19:39:17 * citiz3n bangs head against desk because UPS is so backed up and hasn't delivered his cards yet
2210 2011-02-04 19:39:40 <ArtForz> well, the 5970 was pretty much my own fault
2211 2011-02-04 19:40:26 <ArtForz> HD Accelero 5970 cooler, known to have really bad VRM cooling
2212 2011-02-04 19:41:40 <ArtForz> ran the card @ 1.15Vcore with VRM temps barely < 120°C 24/7 for a few weeks
2213 2011-02-04 19:43:12 <Worf> i can only read one tem sensor, which i think is GPU core - how do you know your vrm was that hot?
2214 2011-02-04 19:43:49 <ArtForz> wrote my own tool to read VRM temps on 58xx/59xx in linux
2215 2011-02-04 19:44:07 <molecular> can you opensource that?
2216 2011-02-04 19:45:03 <ArtForz> yeah, I'm still working on making something releasable
2217 2011-02-04 19:45:23 <molecular> cool
2218 2011-02-04 19:45:33 <ArtForz> it's really pretty simple
2219 2011-02-04 19:46:23 <ArtForz> theres a open source plugin for rivatuner that shows you the meaning of the various i2c registers on the volterra VRM controllers
2220 2011-02-04 19:47:39 <ArtForz> then "simply" use a low level i/o trace tool to capture which GPU registers rivatuner pokes to get at that internal I2C bus
2221 2011-02-04 19:47:59 <luke-jr> newsham: what?
2222 2011-02-04 19:48:25 <ArtForz> hint: ATI 5970 VoltageTools register accesses are a lot easier to make sense of
2223 2011-02-04 19:49:52 <ArtForz> (all it does is set core / i/o voltages to 5970 or 5870 levels)
2224 2011-02-04 19:50:09 <ArtForz> and yes, by directly bitbanging that internal i2c bus
2225 2011-02-04 19:52:17 satamusic has quit (Ping timeout: 255 seconds)
2226 2011-02-04 20:03:08 <presence> so with the overclock/voltage, what was your return on processing?
2227 2011-02-04 20:03:57 <ArtForz> what?
2228 2011-02-04 20:04:48 <presence> what increase in speed did you get with the overclock?
2229 2011-02-04 20:04:57 <ArtForz> 910Mhz average clock
2230 2011-02-04 20:05:07 <presence> so about  60Mhz
2231 2011-02-04 20:05:32 <ArtForz> = 25.5% above stock, 9.5% or so above my average OC @ stock V
2232 2011-02-04 20:06:20 <devon_hillard> http://www.youtube.com/watch?v=GOpBlYx2H1o
2233 2011-02-04 20:06:20 <ArtForz> decided not worth ruining efficiency and risk frying more cards for 10%
2234 2011-02-04 20:06:24 <presence> the 5970 is a bit on the edge just from a power usage basis
2235 2011-02-04 20:06:27 <Netsniper> whats the kwh/block average?
2236 2011-02-04 20:06:29 <devon_hillard> just replace the nvidia cards for AMD
2237 2011-02-04 20:06:51 <ArtForz> *yawn*
2238 2011-02-04 20:06:58 <ArtForz> sucky build
2239 2011-02-04 20:07:50 <presence> I have a few teslas, overrated :(
2240 2011-02-04 20:07:55 <presence> especially at the price
2241 2011-02-04 20:07:58 <ArtForz> yup
2242 2011-02-04 20:08:10 <ArtForz> that applies to nvidia in general
2243 2011-02-04 20:09:26 <presence> thats sad
2244 2011-02-04 20:09:43 <ArtForz> a 6970 can push more double precision flops than a 20xx tesla
2245 2011-02-04 20:09:56 <presence> I have 2 2070s in a few systems
2246 2011-02-04 20:10:21 kisom_dev has joined
2247 2011-02-04 20:10:35 <presence> Was using them for pyrit
2248 2011-02-04 20:10:49 <ArtForz> why?
2249 2011-02-04 20:11:25 <ArtForz> look at the speeds cpyrit-calpp is pushing
2250 2011-02-04 20:11:53 <ArtForz> and thats for double precision, for single precision FP and integer a HD6970 can push about 2.5x peak of a tesla 20xx
2251 2011-02-04 20:12:14 <presence> I got the teslas when they first released
2252 2011-02-04 20:12:49 <presence> 58xx and 69xx wasnt out
2253 2011-02-04 20:12:49 <ArtForz> for int/sp, 1351.7 Gop/s peak vs. 515.2
2254 2011-02-04 20:13:25 <ArtForz> huh?
2255 2011-02-04 20:13:34 <devon_hillard> those dudes crammed 7 top line gaming cards in a single box, a great achievement
2256 2011-02-04 20:13:35 <ArtForz> I think I have my timeline mixed up again
2257 2011-02-04 20:13:58 <ArtForz> yes, and they only had "cooling issues"
2258 2011-02-04 20:14:12 <devon_hillard> btw, aren't there PCI-express 'hubs' for connecting more devices to the same bus?
2259 2011-02-04 20:14:39 <ArtForz> yes
2260 2011-02-04 20:14:44 <ArtForz> PCIe bridges
2261 2011-02-04 20:15:02 <ArtForz> theres even ones in nice external enclosures with a cable and a x16 host adapter card
2262 2011-02-04 20:15:33 <ArtForz> thats for example how dells external 16*tesla box works
2263 2011-02-04 20:15:44 <devon_hillard> except tesla is a ripoff
2264 2011-02-04 20:15:53 <ArtForz> yep
2265 2011-02-04 20:15:58 <ArtForz> but they do have damn good marketing
2266 2011-02-04 20:16:05 <devon_hillard> what's a cheap ready-made enclosure? no cards included
2267 2011-02-04 20:16:27 <ArtForz> btw... pack cards together like that with pretty much zero external airflow and shit runs hot, how could've anyone guessed...
2268 2011-02-04 20:16:36 <ArtForz> </sarcasm>
2269 2011-02-04 20:17:29 <presence> http://www.cooldrives.com/1popcto4xpci.html
2270 2011-02-04 20:17:32 <ArtForz> sad part is, if they would've used the ~2" space left next to their custom carrier cage it probably would've worked
2271 2011-02-04 20:17:33 <devon_hillard> what do you mean zero airflow?
2272 2011-02-04 20:17:39 <devon_hillard> don't you see the desktop fan?
2273 2011-02-04 20:17:44 <ArtForz> yeah
2274 2011-02-04 20:18:02 <ArtForz> ghetto engineering at its finest
2275 2011-02-04 20:18:27 <devon_hillard> you should see my case, cut out a hole in the side for a 12cm fan
2276 2011-02-04 20:18:47 <devon_hillard> turns out, the hole was on the wrong side, so the fan couldn't fit :p
2277 2011-02-04 20:18:47 <ArtForz> I think you've seen my quad 5970 "cases"
2278 2011-02-04 20:19:29 <devon_hillard> ArtForz: no
2279 2011-02-04 20:19:47 <ArtForz> http://bayimg.com/KAAeaaAdp
2280 2011-02-04 20:20:01 <devon_hillard> hm... I suspect there's some crosstalk between all those pci-express cables
2281 2011-02-04 20:20:03 <ArtForz> that was a test fit in the prototype
2282 2011-02-04 20:20:33 <ArtForz> http://bayimg.com/eABDfaAdd < finished stack
2283 2011-02-04 20:20:57 <presence> http://www.magma.com/expressbox7x8RASG2.asp
2284 2011-02-04 20:21:08 <ArtForz> yep
2285 2011-02-04 20:21:09 <presence> that would be HOT with 8 GPUs in it
2286 2011-02-04 20:21:11 <devon_hillard> cool
2287 2011-02-04 20:21:24 <ArtForz> magmas shit is way too expensive :/
2288 2011-02-04 20:22:02 <ArtForz> dual 1kW PSUs, 3 120x38mm intake fans per box
2289 2011-02-04 20:22:17 <presence> and not designed for GPUs...since it has no cooling to speak of
2290 2011-02-04 20:22:26 <ArtForz> yep
2291 2011-02-04 20:22:38 <presence> the slots need to be 2x that distance apaer
2292 2011-02-04 20:23:04 <ArtForz> my cases probably also violate about two fucktons of FCC, electrical and fire safety regulations
2293 2011-02-04 20:23:11 <presence> hell you could only fit 4 5970s in there
2294 2011-02-04 20:23:15 genjix has joined
2295 2011-02-04 20:23:21 <slush> http://mining.bitcoin.cz/media/img/miner.jpg
2296 2011-02-04 20:23:39 <slush> yesterday I finally got PCIe risers, so I'll mount it somehow :)
2297 2011-02-04 20:23:55 <devon_hillard> ArtForz: what's an accessible pci-e bridge?
2298 2011-02-04 20:23:55 <presence> where did you get your risers from?
2299 2011-02-04 20:24:02 <ArtForz> yeah, I tried smth like that at the beginning, middle cards ran hot like hell
2300 2011-02-04 20:24:12 <ArtForz> china
2301 2011-02-04 20:24:25 <slush> ArtForz: yes, I'm not running it like this for now. It was too hot
2302 2011-02-04 20:24:31 <davex__> ArtForz, what's your favorite mainboard for 5970s?
2303 2011-02-04 20:24:54 <ArtForz> my fav used to be quad 5970 on MSI 790FX-GD70
2304 2011-02-04 20:25:03 <devon_hillard> or 'backplane' I think they're called
2305 2011-02-04 20:25:29 <slush> presence: http://www.dhgate.com/pci-e-express-x16-riser-card-1-slot-with/p-ff8080812c305fe5012c367decb86967.html
2306 2011-02-04 20:26:00 <ArtForz> yep, exactly that one
2307 2011-02-04 20:26:13 <ArtForz> it's a 1:1 clone of ADEX PE-FLEX16
2308 2011-02-04 20:26:14 <slush> that's because you sent me this link :)
2309 2011-02-04 20:26:39 <presence> so how do you expand to get more card slots?
2310 2011-02-04 20:26:53 <ArtForz> generally, you don't
2311 2011-02-04 20:27:11 <presence> so this is just to take a card out of the chassis?
2312 2011-02-04 20:27:20 <slush> yes
2313 2011-02-04 20:27:27 <slush> for more space between cards
2314 2011-02-04 20:27:49 <ArtForz> cheapest option nowadays is imo smth like MSI 870-G45
2315 2011-02-04 20:27:55 <ArtForz> and just run 2 cards per box
2316 2011-02-04 20:28:00 RazielZ has quit ()
2317 2011-02-04 20:28:11 akem has joined
2318 2011-02-04 20:28:30 <presence> unfortunately if you were going to run like 12 cards, you would have alot of power leeched just due to the other pc components
2319 2011-02-04 20:28:36 <ArtForz> not really
2320 2011-02-04 20:28:43 <ArtForz> rest of the PC is like 20W
2321 2011-02-04 20:30:30 <ArtForz> CPU is >90% idle, no need to keep a HD (if any) spinning
2322 2011-02-04 20:32:23 <ArtForz> also, 790FX uses a lot more power than 770 chipset
2323 2011-02-04 20:32:24 <davex__> do you just run archlinux or something on each MB?
2324 2011-02-04 20:32:36 <ArtForz> a pretty minimal debian
2325 2011-02-04 20:33:24 <ArtForz> just plain X, no graphical DE, no -dev stuff on miners
2326 2011-02-04 20:34:43 <ArtForz> / is < 1G
2327 2011-02-04 20:36:08 altamic has quit (Ping timeout: 246 seconds)
2328 2011-02-04 20:38:37 <presence> I guess if you went with usb/netboot you probably could pull it off
2329 2011-02-04 20:38:39 <presence> no drives
2330 2011-02-04 20:39:08 <presence> but you still need to buy more MBs/RAM/PSUs
2331 2011-02-04 20:39:14 <ArtForz> no
2332 2011-02-04 20:39:21 <ArtForz> you need 2 PSUs for 4*5970 anyways
2333 2011-02-04 20:39:44 <presence> I was thinking 5870
2334 2011-02-04 20:40:11 <presence> but yes, you need more than one PSU
2335 2011-02-04 20:40:24 <presence> even with 1000W you might not have enough A on the right rails
2336 2011-02-04 20:40:34 <ArtForz> quad 5870 with a 1kW should work
2337 2011-02-04 20:41:04 <presence> you just need to make sure you find the right psu
2338 2011-02-04 20:41:07 <ArtForz> 1.2kW+ PSUs are way too expensive
2339 2011-02-04 20:41:10 <ArtForz> yep
2340 2011-02-04 20:41:15 <presence> all those 5870s will be using 5v
2341 2011-02-04 20:41:19 <ArtForz> ?
2342 2011-02-04 20:41:46 <ArtForz> GPUs have been 12V-only for half a decade or so
2343 2011-02-04 20:41:56 <presence> sorry, I mean 12v
2344 2011-02-04 20:42:00 <ArtForz> yes
2345 2011-02-04 20:42:10 <presence> 3.3v and 5v wont be a problem
2346 2011-02-04 20:42:16 <presence> because nothing really would be using them
2347 2011-02-04 20:42:21 <ArtForz> and any decent modern PSU has 95%+ of available output on 12V
2348 2011-02-04 20:42:24 <presence> proc is pretty much all
2349 2011-02-04 20:42:32 <ArtForz> for 12V + DC-DC designs, 100%
2350 2011-02-04 20:42:38 <ArtForz> CPU also runs off 12V
2351 2011-02-04 20:43:07 <ArtForz> only stuff on +5/+3.3 nowadays is chipset, ram and drive controllers
2352 2011-02-04 20:43:25 <ArtForz> and thats simply to keep *something* on those voltages
2353 2011-02-04 20:43:55 <ArtForz> it'd be pretty trivial to only use +12 for a modern PC
2354 2011-02-04 20:43:55 dwdollar has left ()
2355 2011-02-04 20:44:40 draginx1 has joined
2356 2011-02-04 20:45:07 draginx has quit (Read error: Connection reset by peer)
2357 2011-02-04 20:45:08 <ArtForz> I really like SuperFlower 80+ gold design
2358 2011-02-04 20:47:03 <ArtForz> SF-800P14XE has 92% efficiency with 800W load on +12
2359 2011-02-04 20:48:08 <ArtForz> thats @ 230 VAC
2360 2011-02-04 20:49:37 <Diablo-D3> ArtForz: are you building shit again? :D
2361 2011-02-04 20:49:42 <ArtForz> ?
2362 2011-02-04 20:50:34 <hacim> slush: i'm not as harsh as luke-jr about the pool work you do, i appreciate it. i just was noticing yesterday I'm getting ~2.15 less from the pool than I expect, and was trying to figure out why
2363 2011-02-04 20:50:51 <hacim> sorry, off by 2.536/day
2364 2011-02-04 20:51:05 <slush> see daily pool reward
2365 2011-02-04 20:51:08 <slush> was quite low
2366 2011-02-04 20:51:10 <slush> ...unlucky
2367 2011-02-04 20:51:28 <midnightmagic> Hey how close are we to Folding@Home anyway?
2368 2011-02-04 20:51:33 <slush> you still can't take one day and make conclusions
2369 2011-02-04 20:52:05 <midnightmagic> Looks like they're at about 5203 native TFLOPS.
2370 2011-02-04 20:52:09 <ArtForz> still quite a bit away
2371 2011-02-04 20:52:20 <slush> hacim: follow 7day moving average
2372 2011-02-04 20:52:21 <midnightmagic> but by my very rough calculations, we're only at maybe 1.6 TFLOPS or so.
2373 2011-02-04 20:52:27 <ArtForz> ?
2374 2011-02-04 20:52:33 <midnightmagic> er..
2375 2011-02-04 20:52:34 <ArtForz> hat sounds way too low
2376 2011-02-04 20:52:37 <midnightmagic> sorry, 1600
2377 2011-02-04 20:52:45 <midnightmagic> 1.6PFLOPS. :)
2378 2011-02-04 20:52:46 <hacim> slush: thats an average over the last week
2379 2011-02-04 20:52:59 <slush> oh, so it sounds bad then
2380 2011-02-04 20:53:11 <ArtForz> ;;bc,stats
2381 2011-02-04 20:53:13 <gribble> Current Blocks: 106220 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 627 blocks | Next Difficulty In About: 3 days, 18 hours, 34 minutes, and 0 seconds | Next Difficulty Estimate: 24919.36130709
2382 2011-02-04 20:53:22 <hacim> if I look at my 7day moving average, its even less (10btc)
2383 2011-02-04 20:53:38 <slush> hacim: 10 btc daily from which expected amount?
2384 2011-02-04 20:53:46 <slush> Did you calculated change in difficulty?
2385 2011-02-04 20:53:52 <hacim> but i credit that to a few things: one was the maintainence you did, two was I turned down the hashing for a few hours while I did some other things
2386 2011-02-04 20:54:16 <hacim> slush: 15.06
2387 2011-02-04 20:54:17 <ArtForz> ... thats about 1.5PFLOPS muladd
2388 2011-02-04 20:54:36 <slush> so you earned only 5 btc instead of expected 15?
2389 2011-02-04 20:54:36 <hacim> based on the difficulty and the khash/sec my card is doing
2390 2011-02-04 20:54:36 chaord has quit (Ping timeout: 265 seconds)
2391 2011-02-04 20:54:40 <ArtForz> assuming 58xx/59xx
2392 2011-02-04 20:55:22 <slush> oh, I see, you earned 10/day and expect 15
2393 2011-02-04 20:55:23 <hacim> slush: no, I got 12.524 on average, instead of 15.06.
2394 2011-02-04 20:55:39 <hacim> 12.524 if you use my calculation, 10 based on the 7 day moving average
2395 2011-02-04 20:56:13 <slush> it's quite big difference
2396 2011-02-04 20:57:16 <ArtForz> so we're now something like 1/3 the calculation rate of F@H
2397 2011-02-04 20:57:43 <Diablo-D3> http://kottke.org/11/02/clapping-music
2398 2011-02-04 20:57:44 <Diablo-D3> ffffffffffffff
2399 2011-02-04 20:58:15 <slush> hacim: I can imagine 10% when many factors come together, but you have almost 20%
2400 2011-02-04 20:58:28 <slush> I have no easy explanation. Check if everything is working
2401 2011-02-04 21:05:57 <hacim> slush: yeah, same. it seems something is sort of strange
2402 2011-02-04 21:08:13 <ArtForz> is it me or is claim of 12 TFLOPs for fastra 2... weird
2403 2011-02-04 21:08:20 <ArtForz> 6*GTX295 + GTX275
2404 2011-02-04 21:09:17 <ArtForz> even counting muladd as 2 FLOPS I only get 7.8 TFLOPs for that
2405 2011-02-04 21:09:46 <presence> maybe they are adding both sp, dp, and int into 1
2406 2011-02-04 21:09:48 <Diablo-D3> ArtForz: so I figured out something weird
2407 2011-02-04 21:09:49 phantomcircuit has quit (Quit: Leaving)
2408 2011-02-04 21:09:56 <Diablo-D3> ArtForz: multiple outputs slows the kernel down
2409 2011-02-04 21:10:06 <ArtForz> hurrm, weird
2410 2011-02-04 21:10:12 <luke-jr> slush: I earned only 6 instead of expected 12
2411 2011-02-04 21:10:49 <Diablo-D3> ArtForz: but small numbers of them work fine
2412 2011-02-04 21:11:20 <slush> luke-jr: weekly average?
2413 2011-02-04 21:11:32 <ArtForz> the exact kernel I pastebinned with one-ouput-per-1024-nonces got 556Mh/s on stock 5970
2414 2011-02-04 21:11:43 <ArtForz> maybe again 4xxx suckage?
2415 2011-02-04 21:11:51 <Diablo-D3> Im not doing it quite like yours.
2416 2011-02-04 21:11:52 <slush> luke-jr: and your hashrate in miner is correct and no common network errors here?
2417 2011-02-04 21:12:38 <ArtForz> ARGH
2418 2011-02-04 21:12:49 <ArtForz> they're counting muladdmul as 3 FLOPS
2419 2011-02-04 21:12:56 <luke-jr> slush: 24 hours after I got my miner working
2420 2011-02-04 21:13:02 <slush> oh, so don't worry
2421 2011-02-04 21:13:27 <luke-jr> the difficulty goes up in 4 days. my hashrate is 1 block per 4 days. I can't risk it.
2422 2011-02-04 21:14:18 <Diablo-D3> ArtForz: but like, if i use 2**16 outputs, I lose a few mhash/sec
2423 2011-02-04 21:14:38 phantomcircuit has joined
2424 2011-02-04 21:14:40 <ArtForz> how are you distributing results to outputs?
2425 2011-02-04 21:14:45 <luke-jr> plus, everyone gets a thrill from mining a block on their own, right? :P
2426 2011-02-04 21:15:02 <Diablo-D3> output[tnonce >> 16] = tnonce; or whatever
2427 2011-02-04 21:15:12 <ArtForz> yes, thats slow
2428 2011-02-04 21:15:54 <Diablo-D3> it cant be that slow
2429 2011-02-04 21:16:14 <ArtForz> wait, you're doing that *outside* of the loop?
2430 2011-02-04 21:16:43 <Diablo-D3> Im not using the loop on 4xxx
2431 2011-02-04 21:17:12 <ArtForz> then it should cause some slowdown
2432 2011-02-04 21:17:55 <ArtForz> try it "backwards"
2433 2011-02-04 21:18:03 <Diablo-D3> backwards?
2434 2011-02-04 21:18:07 <ArtForz> output[tnonce & 0xFFFF] = thonce >> 16
2435 2011-02-04 21:18:24 <OneFixt> ArtForz: did you see m0m's changes?
2436 2011-02-04 21:18:33 <ArtForz> nope
2437 2011-02-04 21:18:38 <Diablo-D3> ArtForz: how would that help?
2438 2011-02-04 21:18:42 <ArtForz> assuming you use simple incrementing nonce
2439 2011-02-04 21:18:57 <Diablo-D3> Im not cycling all the nonces in the client
2440 2011-02-04 21:19:06 <ArtForz> ?
2441 2011-02-04 21:19:09 <OneFixt> he's doing this now:
2442 2011-02-04 21:19:09 <OneFixt> output[OUTPUT_SIZE] = output[nonce & OUTPUT_MASK] = nonce;
2443 2011-02-04 21:19:32 <OneFixt> but we need to make sure that this works
2444 2011-02-04 21:19:37 <ArtForz> yeah, that should help a lot
2445 2011-02-04 21:19:57 MJD has quit (Read error: Connection reset by peer)
2446 2011-02-04 21:20:27 <OneFixt> he's using 256 elements
2447 2011-02-04 21:20:28 <Diablo-D3> ArtForz: how would your way even be able to be performed?
2448 2011-02-04 21:20:35 <OneFixt> cutting the loss by 256 times
2449 2011-02-04 21:20:59 <ArtForz> what way?
2450 2011-02-04 21:21:09 MJD has joined
2451 2011-02-04 21:21:29 <Diablo-D3> & 0xFFFF would still leave me with a huge output array index
2452 2011-02-04 21:21:29 <gribble> Error: "0xFFFF" is not a valid command.
2453 2011-02-04 21:21:43 <ArtForz> then use & 0xFF or whatever
2454 2011-02-04 21:21:49 <ArtForz> for 256 outputs
2455 2011-02-04 21:22:17 <ArtForz> just try using the *low* part for indexing
2456 2011-02-04 21:23:14 <ArtForz> dunno if it'll help at all, but it might
2457 2011-02-04 21:23:50 <Diablo-D3> bleh, whats 0xFFFF in dec
2458 2011-02-04 21:23:54 <Diablo-D3> Im too lazy to remember
2459 2011-02-04 21:24:02 <ArtForz> 65535
2460 2011-02-04 21:24:06 <Diablo-D3> oh
2461 2011-02-04 21:24:35 <ArtForz> so output would be a 65536-item array
2462 2011-02-04 21:24:54 <Diablo-D3> nope it doesnt help
2463 2011-02-04 21:24:58 <ArtForz> sucks
2464 2011-02-04 21:25:05 <Diablo-D3> and 64 and 256 already work fine because its so small
2465 2011-02-04 21:25:38 <ArtForz> guess my kernel with the one-consecutive-output-per-thread is hitting some kind of optimization
2466 2011-02-04 21:26:22 <Diablo-D3> 256 still leaves me with 76.x mhash/sec
2467 2011-02-04 21:26:23 <Diablo-D3> so meh
2468 2011-02-04 21:26:29 <ArtForz> sounds good
2469 2011-02-04 21:26:46 <Diablo-D3> [2/4/11 4:20:42 PM] Block 2 found on ATI RV770 (#1)
2470 2011-02-04 21:26:47 <Diablo-D3> [2/4/11 4:20:42 PM] Block 3 found on ATI RV770 (#1)
2471 2011-02-04 21:26:47 <Diablo-D3> hee
2472 2011-02-04 21:26:58 <ArtForz> lucky or dupe? :P
2473 2011-02-04 21:27:04 <Diablo-D3> lucky
2474 2011-02-04 21:27:26 <ArtForz> guess you got a lucky kernel there ;)
2475 2011-02-04 21:27:36 <hacim> slush: the only thing I can find is the RPC timeout slowing down hashing
2476 2011-02-04 21:27:51 <Diablo-D3> hacim: it may or may not
2477 2011-02-04 21:28:04 <Diablo-D3> on miners like mine, http traffic overhead is swollowed by multiple execution threads
2478 2011-02-04 21:28:40 <Diablo-D3> ArtForz: so I lose 0.02% of all hashes
2479 2011-02-04 21:28:41 <Diablo-D3> oh well
2480 2011-02-04 21:28:43 <Diablo-D3> good enough
2481 2011-02-04 21:28:47 <ArtForz> yup
2482 2011-02-04 21:29:04 <ArtForz> and thats at -f 1 on a fast card
2483 2011-02-04 21:29:29 <Diablo-D3> yeah
2484 2011-02-04 21:29:38 <hacim> Diablo-D3: i'm not talking about http overhead, I'm talking about difficulty communicating with the RPC instance in the pool
2485 2011-02-04 21:29:54 <ArtForz> so quite a bit less for your average card/system
2486 2011-02-04 21:30:13 <Diablo-D3> hacim: you mean it times out
2487 2011-02-04 21:30:14 <CIA-98> DiabloMiner: Patrick McFarland master * r2f6c70a / (2 files in 2 dirs): Use multiple buckets on output to reduce lost pool shares - http://bit.ly/goz7vR
2488 2011-02-04 21:30:21 <ArtForz> I think we can live with < 1e-4 lossage
2489 2011-02-04 21:30:22 <Diablo-D3> ArtForz: well I already saw one =O
2490 2011-02-04 21:30:23 <Diablo-D3> =P
2491 2011-02-04 21:30:38 <hacim> Diablo-D3: right
2492 2011-02-04 21:30:54 <Diablo-D3> hacim: a thread on mine could pause timing out and the other two keep crunching
2493 2011-02-04 21:31:20 <hacim> Diablo-D3: "mine" is a bit hard to parse in this context :)
2494 2011-02-04 21:31:38 <Diablo-D3> heee
2495 2011-02-04 21:31:41 <hacim> Diablo-D3: so a thread is not one per card, but you have multiple threads going at once for the mining
2496 2011-02-04 21:31:50 <Diablo-D3> 3 per gpu
2497 2011-02-04 21:31:58 <hacim> m0n doesn't do that?
2498 2011-02-04 21:31:59 <Diablo-D3> it completely masks even slush's crap
2499 2011-02-04 21:32:12 <Diablo-D3> m0 uses one per gpu, and you have to run m0 more than once
2500 2011-02-04 21:32:16 <hacim> haha, comon, no need to slag slush
2501 2011-02-04 21:32:28 <hacim> sorry, i had to say that... slag and slush together were too good to pass up :)
2502 2011-02-04 21:32:29 <Diablo-D3> if I dont, who will? :<
2503 2011-02-04 21:32:36 <Diablo-D3> also lol
2504 2011-02-04 21:32:45 <hacim> you only run m0 more than once if you have more than one gpu
2505 2011-02-04 21:33:06 <Diablo-D3> yeah
2506 2011-02-04 21:33:58 <hacim> i'll give it a shot once I get 7 days of this miner in and then do 7 of yours to see how it compares
2507 2011-02-04 21:34:09 <hacim> if I can figure out how to adjust for difficulty changes
2508 2011-02-04 21:34:13 <Diablo-D3> Im not saying mine is shit proof
2509 2011-02-04 21:34:21 <Diablo-D3> but its pretty shit resistent
2510 2011-02-04 21:34:37 <hacim> heheh
2511 2011-02-04 21:34:59 <midnightmagic> not much of a representative sample. if you're bothered by any of the mining pools, just mine on your own. run against testnet first, and then from there after you've got a few blocks and you know it's working, switch to non-test and you're good.
2512 2011-02-04 21:37:59 <slush> Is there any guide how to build m0m's miner in WIndows?
2513 2011-02-04 21:39:10 <Diablo-D3> slush: erm, its python
2514 2011-02-04 21:39:12 <Diablo-D3> you dont build it
2515 2011-02-04 21:39:32 <slush> hmm
2516 2011-02-04 21:39:45 <slush> but it's kinda hard to get pyopencl working correctly
2517 2011-02-04 21:39:46 <slush> afaik
2518 2011-02-04 21:40:27 <slush> well, pyopencl is the only one dependency, so I change my question
2519 2011-02-04 21:40:34 <slush> how to build pyopencl on windows? :)
2520 2011-02-04 21:41:28 <Diablo-D3> heh
2521 2011-02-04 21:41:29 <Diablo-D3> dude
2522 2011-02-04 21:41:32 <Diablo-D3> I dont even build it on linux
2523 2011-02-04 21:41:48 <slush> me too
2524 2011-02-04 21:42:00 <Diablo-D3> it has a somewhat broken build process due to python sucking entire bags of dongs when you involve wrapping C libs
2525 2011-02-04 21:42:10 <Diablo-D3> anyone who says otherwise is a liar
2526 2011-02-04 21:42:26 <Diablo-D3> also, pyopencl itself provides windows binaries for it
2527 2011-02-04 21:42:44 <Diablo-D3> so just go steal theirs
2528 2011-02-04 21:42:51 <Diablo-D3> anyhow, Im going to bed
2529 2011-02-04 21:42:51 <Diablo-D3> night all
2530 2011-02-04 21:43:06 <Diablo-D3> slush: I told your thread to update to my newest miner
2531 2011-02-04 21:43:14 <Diablo-D3> for less share loss
2532 2011-02-04 21:43:17 <Diablo-D3> night
2533 2011-02-04 21:45:35 <luke-jr> Python sucks period :p
2534 2011-02-04 21:46:37 <dsg> 4 out of 5 trolls agree :)
2535 2011-02-04 21:47:00 <presence> haha
2536 2011-02-04 21:47:02 <presence> python is fine
2537 2011-02-04 21:47:14 <presence> it just sucks when its used in inappropriate places
2538 2011-02-04 21:47:22 Diablo-D3 has quit (Ping timeout: 240 seconds)
2539 2011-02-04 22:04:07 altamic has joined
2540 2011-02-04 22:09:04 <tcatm> yay, my new miner (based on cpuminer) does already 552 Mhash/s on a stock 5970 and I haven't optimized anything yet
2541 2011-02-04 22:09:27 <sipa> nice
2542 2011-02-04 22:12:52 <jgarzik> tcatm: based on cpuminer?  That's interesting.
2543 2011-02-04 22:13:03 <tcatm> it uses opencl
2544 2011-02-04 22:13:17 * jgarzik had thought about adding GPU support
2545 2011-02-04 22:13:37 <tcatm> cpuminer does the getwork stuff already so I re-used the code
2546 2011-02-04 22:13:41 <davout> tcatm: nice
2547 2011-02-04 22:13:57 altamic has quit (Ping timeout: 255 seconds)
2548 2011-02-04 22:15:19 <tcatm> let's try it on the pool
2549 2011-02-04 22:16:43 Kiba has joined
2550 2011-02-04 22:17:33 <luke-jr> tcatm: so coins?
2551 2011-02-04 22:17:43 <luke-jr> would it be practical to make 16 TBC coins, 32 TBCᵗ coins, and 64 TBCˢ coins? (total 112 coins)
2552 2011-02-04 22:17:45 <Kiba> holy shit
2553 2011-02-04 22:17:46 <tcatm> luke-jr: what coins?
2554 2011-02-04 22:17:50 <tcatm> sure
2555 2011-02-04 22:17:57 <Kiba> somebody brought 10900 bitcoin at .85
2556 2011-02-04 22:18:06 <luke-jr> tcatm: how much would it cost me?
2557 2011-02-04 22:18:26 <Kiba> seem to me that the next signfigant resistance is at 1.00 BTC
2558 2011-02-04 22:19:05 <tcatm> luke-jr: what's the total BTC value of those coins? :P
2559 2011-02-04 22:19:20 <luke-jr> tcatm: is it relevant? :p
2560 2011-02-04 22:19:21 <Kiba> but the bid resistance is quite...weak
2561 2011-02-04 22:19:57 <Kiba> the situation is quite fragile
2562 2011-02-04 22:20:04 <tcatm> manufacturing cost is close to 50 BTC, without polishing the surface
2563 2011-02-04 22:20:12 <luke-jr> ea?
2564 2011-02-04 22:20:19 <tcatm> for all coins
2565 2011-02-04 22:20:34 <luke-jr> total value in BTC is 0.01278976
2566 2011-02-04 22:21:17 <tcatm> can you make a drawing of how they should look?
2567 2011-02-04 22:21:30 <luke-jr> I haven't given it any thought :P
2568 2011-02-04 22:22:17 <tcatm> converting that drawing into a suitable format is quite time consuming so it'll add, so total incl. shipping might be 200 BTC
2569 2011-02-04 22:25:46 altamic has joined
2570 2011-02-04 22:25:46 altamic has quit (Changing host)
2571 2011-02-04 22:25:46 altamic has joined
2572 2011-02-04 22:29:12 davex__ has quit (Read error: Connection reset by peer)
2573 2011-02-04 22:29:31 Syke has quit (Ping timeout: 276 seconds)
2574 2011-02-04 22:32:45 Worf has quit (Remote host closed the connection)
2575 2011-02-04 22:36:21 <tcatm> the pool likes my blocks
2576 2011-02-04 22:40:32 <sipa> using ArtForz' opencl kernel?
2577 2011-02-04 22:40:46 <tcatm> yep
2578 2011-02-04 22:41:08 <tcatm> including checking _every_ nonce in those 1024 blocks again in hostcode
2579 2011-02-04 22:41:22 Syke has joined
2580 2011-02-04 22:41:41 <sipa> nice
2581 2011-02-04 22:41:47 <tcatm> about 1.6 MB output buffer
2582 2011-02-04 22:41:59 <ArtForz> hint: have 2 kernels with their own in/out buffers, process/setup/queue one while the other is in flight ;)
2583 2011-02-04 22:42:17 <sipa> diablo uses 3
2584 2011-02-04 22:42:32 <sipa> not sure that's necessary :)
2585 2011-02-04 22:42:32 <tcatm> code already supports multiple kernels
2586 2011-02-04 22:42:37 <ArtForz> *shrug* 2 works fine here
2587 2011-02-04 22:43:09 <sipa> if it's worth doing, it's worth overdoing!
2588 2011-02-04 22:43:13 <ArtForz> it's just so the GPU has something to do while the host is checking outputs/setting up new input
2589 2011-02-04 22:43:34 <tcatm> I check/setup while the kernel runs
2590 2011-02-04 22:43:54 <sipa> if the network delay for transmitting/receiving can take a few seconds, you probably need more kernels
2591 2011-02-04 22:44:08 bittertea has joined
2592 2011-02-04 22:44:12 <tcatm> i.e. clFlush() checkwork() prepare() clFinish()
2593 2011-02-04 22:44:20 <sipa> although i'd personally go for a separate dedicated communication thread
2594 2011-02-04 22:44:24 <ArtForz> not use flush
2595 2011-02-04 22:44:29 <ArtForz> use waitforevents
2596 2011-02-04 22:44:45 <tcatm> what's wrong with flush?
2597 2011-02-04 22:45:18 <ArtForz> dunno, why, but 3 waitforevents was a few Mh/s faster than 1 flush() here
2598 2011-02-04 22:45:47 <tcatm> don't I need the flush to actually start the kernel?
2599 2011-02-04 22:45:52 <ArtForz> no
2600 2011-02-04 22:46:07 <ArtForz> waiting on it has the same effect
2601 2011-02-04 22:46:17 <ArtForz> check the cl spec
2602 2011-02-04 22:50:58 <slush> Hey, new DiabloMiner isn't working on my miner property
2603 2011-02-04 22:51:16 <slush> show hashrate, but no share found for long time
2604 2011-02-04 22:51:34 <slush> (10 min on 2x5970)
2605 2011-02-04 22:54:26 <EvanR-work> slush: whats the final version of the 'my account' page going to look like. will it eventually show us our shares?
2606 2011-02-04 22:54:32 theymos has joined
2607 2011-02-04 22:55:29 <sipa> ;;bc,prob 1300000 10m
2608 2011-02-04 22:55:29 <gribble> 0.00821627954748
2609 2011-02-04 22:55:44 <sipa> slush: becoming quite unlikely indeed
2610 2011-02-04 23:12:59 ducki2p has quit (Ping timeout: 240 seconds)
2611 2011-02-04 23:20:00 ducki2p has joined
2612 2011-02-04 23:24:51 <molecular> sipa: bc,prob - show the probability of getting at least one block at given Khps, in a given time period.
2613 2011-02-04 23:25:46 <molecular> I think slush was talking about diff1 blocks?
2614 2011-02-04 23:25:59 <molecular> this seems to calc probability based on current difficulty?
2615 2011-02-04 23:27:11 <sipa> molecular: you're right
2616 2011-02-04 23:27:14 <sipa> ;;bc,prob
2617 2011-02-04 23:27:14 <gribble> (bc,prob <an alias, at least 1 argument>) -- Alias for "math calc 1-exp(-$1*1000 * [seconds $*] / (2**32* [bc,diff]))".
2618 2011-02-04 23:28:08 <sipa> ;;math calc 1-exp(-1300000*1000 * 600 / (2**32))
2619 2011-02-04 23:28:08 <gribble> 1
2620 2011-02-04 23:28:15 <molecular> but still, no share im 10mins on 2x5970 seems highly unlikely. it should find share ever 3-4 secs
2621 2011-02-04 23:28:21 <sipa> indeed
2622 2011-02-04 23:28:28 * sipa needs sleep
2623 2011-02-04 23:29:22 <molecular> what changes did Diablo make to his miner? only the "share loss" fix or also changes to kernel?
2624 2011-02-04 23:29:33 <molecular> oh well, that probably involves changes to kernel ;)
2625 2011-02-04 23:30:01 <sipa> indeed
2626 2011-02-04 23:35:34 <tcatm> ;;bc,target
2627 2011-02-04 23:35:34 <gribble> Error: "bc,target" is not a valid command.
2628 2011-02-04 23:36:25 <theymos> ;;bc,hextarget
2629 2011-02-04 23:36:26 <gribble> 000000000002FA29000000000000000000000000000000000000000000000000
2630 2011-02-04 23:36:51 <sipa> ;;bc,stats
2631 2011-02-04 23:36:53 <gribble> Current Blocks: 106244 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 603 blocks | Next Difficulty In About: 3 days, 14 hours, 35 minutes, and 51 seconds | Next Difficulty Estimate: 25015.88215764
2632 2011-02-04 23:38:33 <LobsterMan> what happens if someone gets a transaction to an address that resides in a node that is currently offline?
2633 2011-02-04 23:38:55 <luke-jr> I'd like to write a pool server that pays out immediately from the same block mined :P
2634 2011-02-04 23:39:07 <theymos> LobsterMan: They'll get it when they come back.
2635 2011-02-04 23:40:12 <sipa> LobsterMan: since nobody in the network knows who that address belongs to, the result can't be different from when the node is online :)
2636 2011-02-04 23:40:16 bittertea has left ()
2637 2011-02-04 23:40:25 <LobsterMan> ah fair enough
2638 2011-02-04 23:40:29 <sipa> except for the node itself of course
2639 2011-02-04 23:40:45 <molecular> uuuh, slush did it, stats removed ;(
2640 2011-02-04 23:42:29 <molecular> MTG|     TRADE|                      10900 @ $0.85 <- jesus, again!
2641 2011-02-04 23:43:00 <echelon> ugh
2642 2011-02-04 23:43:13 <Syke> oh yeah, that was me. :)
2643 2011-02-04 23:43:26 Dex has joined
2644 2011-02-04 23:43:29 <molecular> sure, now give back our coins!
2645 2011-02-04 23:43:30 <echelon> >:(
2646 2011-02-04 23:43:31 <echelon> why
2647 2011-02-04 23:44:01 <molecular> why what?
2648 2011-02-04 23:44:18 <tcatm> ;;bc,cald 342500 1
2649 2011-02-04 23:44:18 <gribble> Error: "bc,cald" is not a valid command.
2650 2011-02-04 23:44:23 <tcatm> ;;bc,calcd 342500 1
2651 2011-02-04 23:44:23 <gribble> The average time to generate a block at 342500 Khps, given the supplied difficulty of 1, is 12 seconds
2652 2011-02-04 23:44:29 <echelon> why would he buy at a high mark-up
2653 2011-02-04 23:44:55 <luke-jr> ;;bc,calcd 265000 1000000000
2654 2011-02-04 23:44:55 <gribble> The average time to generate a block at 265000 Khps, given the supplied difficulty of 1000000000, is 513 years, 48 weeks, 4 days, 22 hours, 9 minutes, and 18 seconds
2655 2011-02-04 23:44:59 <luke-jr> :o
2656 2011-02-04 23:45:03 <cosurgi> slush: when I'll have my stats back?
2657 2011-02-04 23:45:16 <molecular> echelon, because he thinks it's unlikely for the rate to drop.
2658 2011-02-04 23:45:25 <molecular> cosurgi, never!
2659 2011-02-04 23:45:37 <echelon> well he's making it bad for other buyers
2660 2011-02-04 23:45:45 <echelon> so i hope it does drop
2661 2011-02-04 23:45:46 <luke-jr> echelon: and good for sellers
2662 2011-02-04 23:45:47 <molecular> echelon, he just wanted these BTC, now!
2663 2011-02-04 23:46:05 <theymos> I'm still selling at 0.8 on #bitcoin-otc.
2664 2011-02-04 23:46:06 <luke-jr> Syke: I've got 100 at .8999 on Market
2665 2011-02-04 23:46:19 <molecular> echelon, of course, that's how a market works. the demand is high and rising: buy quickly
2666 2011-02-04 23:46:54 <Syke> with a quarter of the coins minted, there's a lot of room for growth up
2667 2011-02-04 23:47:11 <cosurgi> omg, I just noticed the price
2668 2011-02-04 23:47:25 <molecular> how high was the gold-price when 25% of the gold where mined?
2669 2011-02-04 23:47:27 <cosurgi> it went up two times, in just two weeks??
2670 2011-02-04 23:47:50 <cosurgi> good I didn't sell any
2671 2011-02-04 23:48:04 <molecular> keep 'em, for a long time
2672 2011-02-04 23:48:29 <molecular> dont be like the guy that bought a pizza for 10000 BTC ;)
2673 2011-02-04 23:48:40 <citiz3n> ive never used crossfire/SLI before
2674 2011-02-04 23:48:51 <citiz3n> is it something that requires a reboot to turn on/off
2675 2011-02-04 23:49:07 <sipa> no idea, never had to touch it :)
2676 2011-02-04 23:49:17 <citiz3n> i was thinking of popping a second card into a friend's computer
2677 2011-02-04 23:49:25 <sipa> for mining?
2678 2011-02-04 23:49:28 <citiz3n> yeah
2679 2011-02-04 23:49:35 <citiz3n> but if he wants to use crossfire for gaming (while not mining)
2680 2011-02-04 23:49:38 altamic has quit (Ping timeout: 245 seconds)
2681 2011-02-04 23:49:40 <citiz3n> im wondering how that would work
2682 2011-02-04 23:49:42 <sipa> ah ic
2683 2011-02-04 23:54:25 <citiz3n> damn
2684 2011-02-04 23:54:36 <citiz3n> i just ate and had to loosen my belt and unbutton my pants