1 2011-07-16 00:00:02 <freewil> ok does bitcoin have a config setting for that or do you just configure it through the OS?
  2 2011-07-16 00:00:13 <gmaxwell> Bitcoin has a setting for it.
  3 2011-07-16 00:00:22 <freewil> ok great thanks
  4 2011-07-16 00:00:29 <gmaxwell> -proxy=<ip:port>   Connect through socks4 proxy
  5 2011-07-16 00:00:53 <freewil> ah genius
  6 2011-07-16 00:46:48 <freewil> what does 'keypoololdest' refer to in the getinfo rpc call
  7 2011-07-16 00:51:41 <b4epoche> prolly the oldest key in the pool
  8 2011-07-16 00:52:32 <freewil> thats strange
  9 2011-07-16 00:52:59 <freewil> im not that familiar with the bitcoin network so i dont know what that refers to exactly
 10 2011-07-16 00:53:50 <b4epoche> bitcoin makes a pool or ~100 keys to use after you request the first new one (I think)
 11 2011-07-16 00:54:23 <freewil> so does this have to do with your public addresses in your wallet
 12 2011-07-16 00:54:28 <b4epoche> thus, you're wallet will have like 100 keys in it at any time.  i.e. it doesn't generate a new one 'on demand'
 13 2011-07-16 00:54:54 <b4epoche> your private keys are in wallet
 14 2011-07-16 00:55:27 <gmaxwell> b4epoche: no it will have a pool of 100 unused keys, and it generates new ones to keep it topped off.
 15 2011-07-16 00:55:56 <b4epoche> there's some reason for it doing that but I can't recall it
 16 2011-07-16 00:55:56 <freewil> i dont think this has anything to do with that
 17 2011-07-16 00:55:58 <gmaxwell> The keypoololdest is the oldest unused key. A backup older than that is about to become lossy.
 18 2011-07-16 00:56:15 <gmaxwell> b4epoche: so that backups actually work. :)
 19 2011-07-16 00:56:39 <b4epoche> well, so the irregular backups work
 20 2011-07-16 00:56:44 <b4epoche> s/the/that
 21 2011-07-16 00:57:04 <freewil> gmaxwell: the value im getting back for keypoololdest doesnt look like a public address hex value though
 22 2011-07-16 00:57:04 <gmaxwell> Not just irregular. If not for the pool, you could backup every day, and still lose most of your coins if you lost your wallet.
 23 2011-07-16 00:57:11 <freewil> its just an integer
 24 2011-07-16 00:57:32 <b4epoche> prolly an index
 25 2011-07-16 00:57:37 <gmaxwell> freewil: its unix time, number of seconds since 1970. date +s%
 26 2011-07-16 00:57:59 <b4epoche> gmaxwell:  I mean backup every time a key is generated...
 27 2011-07-16 00:58:06 <gmaxwell> date +%s
 28 2011-07-16 00:58:13 <freewil> oh i get it
 29 2011-07-16 00:58:28 <b4epoche> ah, time since the epoche
 30 2011-07-16 00:58:28 <gmaxwell> b4epoche: so every time you send bitcoin, yes that would work... but thats not much of a backup scheme. :)
 31 2011-07-16 00:58:48 <b4epoche> a one file raid
 32 2011-07-16 00:58:59 <gmaxwell> Yes, and Raid is not a backup.
 33 2011-07-16 00:59:10 <gmaxwell> rm -rf /   ooops.. my data is now redundantly deleted. ;)
 34 2011-07-16 00:59:21 <b4epoche> depends on the type of raid
 35 2011-07-16 01:00:01 <gmaxwell> all types of raid will replicate your deletion,  because raid (that provides redundancy at all) is high availablity, not backup. :)
 36 2011-07-16 01:01:01 <b4epoche> I thought there was one raid setup you could use as effectively a backup
 37 2011-07-16 01:01:08 <dsockwell> lol no
 38 2011-07-16 01:01:11 <dsockwell> maybe raidz
 39 2011-07-16 01:01:14 <dsockwell> maybe
 40 2011-07-16 01:01:26 <dsockwell> but that's zfs and no
 41 2011-07-16 01:01:48 <gmaxwell> no.. there are filesystems with snapshots, which if used right are a kind of weak online backup. But raid isn't a backup.
 42 2011-07-16 01:02:59 <b4epoche> it's drive swapping I'm thinking of but that's really not a good bitcoin wallet solution
 43 2011-07-16 01:03:22 <dsockwell> if you kept swapping out drives on a raid1 that's like a backup
 44 2011-07-16 01:03:26 <gmaxwell> hah, okay, yea, you could raid 1 and swap out drives..
 45 2011-07-16 01:03:34 <dsockwell> an expensive one
 46 2011-07-16 01:03:50 <b4epoche> well, you could rotate through
 47 2011-07-16 01:03:50 <gmaxwell> sadly drives are about the cheapest backup medium per gigabyte now.
 48 2011-07-16 01:04:20 <gmaxwell> b4epoche: just make sure your raid implementation is smart enough to not decide the old drive is the one you're keeping when you rotate it back in!
 49 2011-07-16 01:05:12 <b4epoche> keeping?  oldest drive (backup) gets overwritten
 50 2011-07-16 01:05:45 <senseles> just stick a backup of your wallet on an SSD
 51 2011-07-16 01:05:48 <senseles> and shove it in a safe
 52 2011-07-16 01:05:53 <dsockwell> ...
 53 2011-07-16 01:06:24 <gmaxwell> b4epoche: right, but you've got a mirror, you pull a drive. Insert a old one. System decides to rebuild the old one onto the new one. 0_o (because it's not like swapping in old drives is going to be a well tested case for a raid system)
 54 2011-07-16 01:06:38 <senseles> r1soft would be a good backup solution
 55 2011-07-16 01:06:42 <gmaxwell> In any case, backups are why there is a keypool.
 56 2011-07-16 01:06:47 <b4epoche> gmaxwell:  oh, I see what you're saying...
 57 2011-07-16 01:06:47 <senseles> take hourly snapshots encrypted on a remote server
 58 2011-07-16 01:06:58 <b4epoche> yea, you need a master/slave setup
 59 2011-07-16 01:07:15 <gmaxwell> senseles: http://duplicity.nongnu.org/
 60 2011-07-16 01:07:22 <bliket_> a bag for me, a bag for you, let's get wrecked on bags of glue
 61 2011-07-16 01:07:29 <b4epoche> gmaxwell:  was that you're mining setup you posted a picture of earlier?
 62 2011-07-16 01:07:40 <gmaxwell> I just have a simple cron job for my online wallet that uses the backupwallet rpc + xz + gpg.
 63 2011-07-16 01:07:44 <gmaxwell> b4epoche: yes.
 64 2011-07-16 01:07:52 <bliket_> link??
 65 2011-07-16 01:07:53 <b4epoche> very nice
 66 2011-07-16 01:08:03 <b4epoche> bliket_:  he posted it for you
 67 2011-07-16 01:08:12 <bliket_> where @?
 68 2011-07-16 01:08:19 <gmaxwell> You commented in fact.
 69 2011-07-16 01:08:23 <gmaxwell> https://people.xiph.org/~greg/v1.jpg
 70 2011-07-16 01:08:24 <bliket_> yeah
 71 2011-07-16 01:08:45 <bliket_> are those x16 risers?
 72 2011-07-16 01:09:09 <gmaxwell> bliket_: yes, except for the 1x ones you can't see.
 73 2011-07-16 01:09:19 <b4epoche> 12 cards?
 74 2011-07-16 01:09:30 <gmaxwell> On those two systems.
 75 2011-07-16 01:10:05 <Akiron> question for those familiar with bitcoin source--where is the block checking code located?
 76 2011-07-16 01:10:17 <b4epoche> running open air, eh?
 77 2011-07-16 01:10:21 <gmaxwell> Akiron: most of it in main.cpp / main.h
 78 2011-07-16 01:10:24 <bliket_> what kind of puppies are those?
 79 2011-07-16 01:10:31 <gmaxwell> b4epoche: yes, in my basement.
 80 2011-07-16 01:10:34 <bliket_> gmaxwell: what kind of puppies???
 81 2011-07-16 01:11:03 <b4epoche> gmaxwell:  I was thinking about doing the same but I'm not much of a hw guy...  but I have have very cool basement
 82 2011-07-16 01:11:08 <bliket_> btw i am laughing my ass off to myself because I am using the word "puppies" instead of "graphics cards"
 83 2011-07-16 01:11:09 <gmaxwell> those machines have 5830s.
 84 2011-07-16 01:11:23 <bliket_> nice puppies
 85 2011-07-16 01:11:23 <gmaxwell> The funny thing is that the more ghetto systems run a lot cooler.
 86 2011-07-16 01:11:33 <bliket_> those are like german shepards
 87 2011-07-16 01:11:42 <gmaxwell> https://people.xiph.org/~greg/v0.jpg  (will get rearranged to the v1 style this weekend)
 88 2011-07-16 01:11:53 <b4epoche> yea, figured open air would be better
 89 2011-07-16 01:11:57 <bliket_> the 5830 is like a german shepard, the 6990 is like a full grown pitbull
 90 2011-07-16 01:12:10 <bliket_> those nvidia cards are like poodles
 91 2011-07-16 01:12:14 <bliket_> lol
 92 2011-07-16 01:12:37 <b4epoche> gmaxwell:  are those just two motherboards?  with six pcie slots each?
 93 2011-07-16 01:12:51 <gmaxwell> the v0 cards are all 53.50 to 62.50 degress right now...
 94 2011-07-16 01:13:05 <b4epoche> C I assume, not F
 95 2011-07-16 01:13:32 <bliket_> yes of course C
 96 2011-07-16 01:13:32 <gmaxwell> yes, C.   the v1 cards are 60-67 right now
 97 2011-07-16 01:13:37 <b4epoche> well, of course, unless you're in southern hemisphere
 98 2011-07-16 01:13:49 <bliket_> there is no way in hell those are F, unless you live on the north pole
 99 2011-07-16 01:13:52 <gmaxwell> b4epoche: yes,
100 2011-07-16 01:14:22 <gmaxwell> b4epoche: The boards have 5x 16 slots (wired 4x 8, one by 4x) and one 1x slot.
101 2011-07-16 01:14:41 <bliket_> 890-fxa gd-65?
102 2011-07-16 01:14:49 <bliket_> the refined taste of wine motherboard
103 2011-07-16 01:14:53 <b4epoche> a low-power cpu?
104 2011-07-16 01:15:14 <bliket_> amd seperaon 140, like a nice toblerone
105 2011-07-16 01:15:41 <bliket_> a green memory slot
106 2011-07-16 01:15:42 <bliket_> ?
107 2011-07-16 01:15:49 <bliket_> that can't be a 890fxa-gd65
108 2011-07-16 01:15:53 <bliket_> or a gd70
109 2011-07-16 01:16:05 <b4epoche> 890-fxa gd-65 only has 2 pcie
110 2011-07-16 01:16:10 <gmaxwell> It's a GD70.
111 2011-07-16 01:16:30 <gmaxwell> Kinda pricy, but it doesn't matter with six cards on it.
112 2011-07-16 01:16:33 <bliket_> do you wish you bought the gd65?
113 2011-07-16 01:16:40 <bliket_> like i do?
114 2011-07-16 01:17:22 <b4epoche> 890FXA-GD70
115 2011-07-16 01:17:26 <b4epoche> ah, yep
116 2011-07-16 01:17:59 <gmaxwell> bliket_: no not really, the cost difference is only $7 per slot or something like that.
117 2011-07-16 01:18:14 <bliket_> every penny saved is a penny earned
118 2011-07-16 01:18:22 <gmaxwell> Same reason I wired up >1x, in case I do something else on them that can benefit from more bus bandwidth.
119 2011-07-16 01:18:35 <bliket_> which will never be the case
120 2011-07-16 01:18:54 <b4epoche> 65 only has 2 pcie, no?
121 2011-07-16 01:18:57 <gmaxwell> Well, I've made a fair amount of money doing non-mining on my gpu systems.
122 2011-07-16 01:19:17 <gmaxwell> (and in fact, I only started mining because I had gpus from other business)
123 2011-07-16 01:19:21 <IO-> wait, you mean real work?
124 2011-07-16 01:19:28 <gmaxwell> Yes.
125 2011-07-16 01:19:30 <IO-> wow
126 2011-07-16 01:19:33 <IO-> get outa here
127 2011-07-16 01:19:34 <b4epoche> gmaxwell:  yea, what do you do?
128 2011-07-16 01:19:49 <CheapScotsman> Anyone familiar with jumping PSUs?
129 2011-07-16 01:20:00 <bliket_> he edits porn movies for bangbros
130 2011-07-16 01:20:07 <bliket_> he doesn't like talking about it
131 2011-07-16 01:20:17 <IO-> SR had a 6month BB account for sale, lol
132 2011-07-16 01:20:25 <gmaxwell> hah. Custom numerical optimization stuff, training for machine classification, signal processing stuff. ::shrugs::
133 2011-07-16 01:20:39 <bliket_> porn.
134 2011-07-16 01:20:45 <b4epoche> so, you test on your machines?
135 2011-07-16 01:21:00 <bliket_> if jacking off means testing
136 2011-07-16 01:21:05 <CheapScotsman> That would become a depressing job
137 2011-07-16 01:21:07 <b4epoche> or you actually getting outsourced computations?
138 2011-07-16 01:21:15 <gmaxwell> Both.
139 2011-07-16 01:21:28 <gmaxwell> I also have 160 opteron cores in the basement too. :)
140 2011-07-16 01:21:42 <b4epoche> you're own biz?
141 2011-07-16 01:21:57 <gmaxwell> More like net-profitable hobbies.
142 2011-07-16 01:23:08 <bliket_> we should start mining in the Dominican Republic
143 2011-07-16 01:23:14 <bliket_> anyone know why?
144 2011-07-16 01:23:19 <gmaxwell> Power is cheap there?
145 2011-07-16 01:23:27 <bliket_> yes US$0.006 per KWh
146 2011-07-16 01:23:35 <gmaxwell> what? ?!
147 2011-07-16 01:23:37 <bliket_> yep
148 2011-07-16 01:24:01 <gmaxwell> damn, in utah (not where I am) you can apparently get to 0.045, but I think thats about the lowest in the US.
149 2011-07-16 01:24:08 <gmaxwell> Are you sure thats the complete price there?
150 2011-07-16 01:24:12 <bliket_> yes sir
151 2011-07-16 01:24:47 <gmaxwell> but can we build flood proof miners?
152 2011-07-16 01:25:50 <bliket_> yes
153 2011-07-16 01:25:59 <bliket_> i guess you could call it that
154 2011-07-16 01:26:16 <bliket_> so best to buy the cheapest $/mhash machine and move to the dominican republic
155 2011-07-16 01:26:25 <bliket_> and start a datacenter there with all bitcoin miners
156 2011-07-16 01:26:31 <bliket_> run AC all day
157 2011-07-16 01:26:33 <bliket_> :D
158 2011-07-16 01:26:33 <senseles> was considering doing that
159 2011-07-16 01:26:37 <senseles> got 0.06 in my area of the US
160 2011-07-16 01:26:38 <bliket_> AC ALL DAY
161 2011-07-16 01:26:44 <senseles> then just run a small fiber line
162 2011-07-16 01:26:49 <senseles> since it doesn't need much bandwidth
163 2011-07-16 01:26:56 <gmaxwell> http://en.wikipedia.org/wiki/Communications_in_the_Dominican_Republic#Broadband_Internet_access
164 2011-07-16 01:27:04 <bliket_> you could probably do it with 56k
165 2011-07-16 01:27:13 <bliket_> how many miners can one run on 56k modem?
166 2011-07-16 01:27:19 <bliket_> lets say 1ghash
167 2011-07-16 01:27:20 <gmaxwell> senseles: 0.05(5?) here in the winter, 0.075 in the summer (now).
168 2011-07-16 01:27:30 <senseles> sad dominican republic has better infrastructure than philippines
169 2011-07-16 01:27:39 <gmaxwell> the latency on 56k modems is kinda terrible.
170 2011-07-16 01:27:49 <gmaxwell> >200ms rtt just from the modem itself.
171 2011-07-16 01:28:14 <gmaxwell> it appears that broadband is cheap in the Dominican Republic though.
172 2011-07-16 01:28:17 <bliket_> ok you could probably get a residential wimax connection
173 2011-07-16 01:28:36 <b4epoche> PA has pretty cheap rates too...
174 2011-07-16 01:28:52 <b4epoche> hence my mining interest
175 2011-07-16 01:29:23 <bliket_> pensilvania?
176 2011-07-16 01:29:38 <bliket_> pennsylvania
177 2011-07-16 01:30:12 <b4epoche> yea
178 2011-07-16 01:30:16 <IO-> i think i could live there
179 2011-07-16 01:30:28 <b4epoche> keep on frackin'
180 2011-07-16 01:30:47 <senseles> best part about being offshore is you can give your machines to someone else ~
181 2011-07-16 01:30:54 <senseles> for instance my wife own my bitcoin machines in the US
182 2011-07-16 01:30:58 <senseles> and she's not an american
183 2011-07-16 01:31:00 <senseles> so uh, no taxes
184 2011-07-16 01:31:02 <senseles> lol.
185 2011-07-16 01:31:12 <gmaxwell> eh. Tax games are for idiots, IMO.
186 2011-07-16 01:31:40 <senseles> giving 25% of my money to some corporate controlled jackasses who are going to use it to further supress society is not my idea of money well spent
187 2011-07-16 01:32:06 <gmaxwell> Taxes are just a cost of doing business, you factor them in, and live with them. The cost of just getting investigated is enormous, and the tax man usually comes out ahead.
188 2011-07-16 01:32:07 <b4epoche> and you'd spend it on?  pussy and beer?
189 2011-07-16 01:32:07 <senseles> while they fuck prostitutes and show their cocks off on twitter
190 2011-07-16 01:32:20 <IO-> i vote republican because i have too much money and want to give more to rich people
191 2011-07-16 01:32:41 <bliket_> i vote republican because I am a jackass
192 2011-07-16 01:32:58 <b4epoche> Bachmann/Palin ftw!
193 2011-07-16 01:33:02 <senseles> i'm trying to find a new home so i can give up my citizenship
194 2011-07-16 01:33:05 <phantomcircuit> id_dsa
195 2011-07-16 01:33:08 <phantomcircuit> <p>The key ??????id_rsa.pub?????? has been authorized.</p>
196 2011-07-16 01:33:13 <phantomcircuit> STILL NOT AUTHORIZED
197 2011-07-16 01:33:18 <phantomcircuit> i was wrong
198 2011-07-16 01:33:24 <phantomcircuit> cpanel is unhackable
199 2011-07-16 01:33:30 <senseles> the only thing good about being an american is the passport
200 2011-07-16 01:33:35 <phantomcircuit> it is so shitty it doesn't work even when it says you succeeded
201 2011-07-16 01:33:35 <senseles> you can get into anywhere without issue
202 2011-07-16 01:33:53 <IO-> i vote republican because we need more TSA, more wars. I wasn't using my rights anyways
203 2011-07-16 01:34:10 <senseles> as if democrats arent going to do the same thing
204 2011-07-16 01:34:23 <phantomcircuit> IO-, i vote democrat because i didn't need that money anyways
205 2011-07-16 01:34:48 <b4epoche> phantomcircuit:  cpanel?
206 2011-07-16 01:34:51 <senseles> we got a black democrat as president and the DEA is still raiding dispenseries in california :/
207 2011-07-16 01:35:11 <IO-> dep or rep, they're taking your money
208 2011-07-16 01:35:16 <senseles> 10th amendment what's that
209 2011-07-16 01:35:20 <senseles> state right who?
210 2011-07-16 01:35:31 <gmaxwell> #politics <----
211 2011-07-16 01:35:40 <senseles> ya, sorry :/
212 2011-07-16 01:35:46 <phantomcircuit> lol
213 2011-07-16 01:36:11 <phantomcircuit> IO-, at least the republicans are paying lip service to reducing taxes, the democrats have hardons for WW2 era tax levels
214 2011-07-16 01:36:16 <phantomcircuit> b4epoche, yes
215 2011-07-16 01:36:29 <phantomcircuit> b4epoche, it *is* vulnerable to a csrf attack
216 2011-07-16 01:36:29 <quellhorst> senseles: half white
217 2011-07-16 01:36:43 <phantomcircuit> but it's such a piece of shit software that i cant get the attack to actually run
218 2011-07-16 01:36:44 <gmaxwell> Why should y'all care about taxes when you're just planning on cheating them? lol.
219 2011-07-16 01:36:48 <IO-> they only want to reduce taxes, more so, for millionairs. and they want college kids and old people to pay for it
220 2011-07-16 01:36:55 <senseles> i still have to pay taxes
221 2011-07-16 01:36:59 <b4epoche> http://www.cpanel.net/ ?
222 2011-07-16 01:37:03 <phantomcircuit> IO-, lolololol that's literally not true
223 2011-07-16 01:37:06 <phantomcircuit> b4epoche, yes
224 2011-07-16 01:37:58 <genis1> very interesting times ahead, if they dont raise the debt ceiling, we should see some unstable USD prices
225 2011-07-16 01:38:08 <phantomcircuit> IO-, you realize that if you calculate the effective federal tax rate people in the bottom 60% of incomes pay zero to -8% taxes right?
226 2011-07-16 01:38:17 <b4epoche> genis1:  like btc prices?
227 2011-07-16 01:38:38 <genis1> well, btc might go up as USD goes down
228 2011-07-16 01:39:09 <genis1> normally money chases items that store value or are deflationary. like gold, and possibly btc
229 2011-07-16 01:39:25 <b4epoche> I mean the volatility
230 2011-07-16 01:39:57 <senseles> even with all my bitching US is still one of the lowest tax rates in the world
231 2011-07-16 01:39:59 <gmaxwell> phantomcircuit: even if you had a flat tax rate at 35% people in the bottom 60% wouldn't account for crap because income is so unequal in the US.
232 2011-07-16 01:40:13 <senseles> I'm paying about 25%
233 2011-07-16 01:40:18 <senseles> as a self-employed individual
234 2011-07-16 01:40:20 <phantomcircuit> gmaxwell, yes but i mean they pay negative taxes
235 2011-07-16 01:40:30 <bliket_> #politics channel exists
236 2011-07-16 01:40:33 <genis1> gmaxwell: with inflation that all changes.  many people will bump to higher tax brackets
237 2011-07-16 01:40:50 <phantomcircuit> gmaxwell, so it is literally impossible to lower taxes for anybody except the top 40% since nobody else pays federal taxes
238 2011-07-16 01:41:03 <senseles> i wonder how many of the miners are in the US
239 2011-07-16 01:41:06 <phantomcircuit> see what im saying?
240 2011-07-16 01:41:12 <senseles> and i wonder what economic hardship would do to the mining rate
241 2011-07-16 01:41:18 <senseles> as people have to sell off hardware to make ends meet
242 2011-07-16 01:41:48 <gmaxwell> genis1: if you inflation correct the brackets that hasn't been that case, as they've been continually revised downwards effectively. Also, people with greater income have many more options to avoid taxes.
243 2011-07-16 01:42:16 <senseles> ya, like starting a foundation
244 2011-07-16 01:42:18 <gmaxwell> senseles: if you're on the boundary of economic hardship you probably shouldn't have that much invested. "don't risk what you can't lose"
245 2011-07-16 01:42:22 <senseles> then donating all your income to your own foundation
246 2011-07-16 01:42:26 <senseles> then use that foundation to pay your bills
247 2011-07-16 01:42:29 <senseles> == 0 taxes
248 2011-07-16 01:42:55 <phantomcircuit> senseles, if you get caught with your foundation paying your bills like that you go to jail
249 2011-07-16 01:43:00 <sacarlson> what is this value? txNew.vout[0].scriptPubKey  I assume it's just some big random number that was used to create the root of the block chain
250 2011-07-16 01:43:13 <gmaxwell> senseles: an income from long term investments being taxed at a much lower rate... (until you hit AMT at least)
251 2011-07-16 01:43:15 <phantomcircuit> sacarlson, read it xD
252 2011-07-16 01:43:15 <senseles> ya but you have to "travel" for the foundation
253 2011-07-16 01:43:20 <phantomcircuit> satoshi_, not cool
254 2011-07-16 01:43:29 <senseles> which of course entails first class tickets and 5 star hotels
255 2011-07-16 01:43:35 <bliket_> ban satoshi that guy owes me a bitcoin
256 2011-07-16 01:43:46 <JFK911> he owes me two bitcoins
257 2011-07-16 01:43:46 <sacarlson> phantomcircuit:  txNew.vout[0].scriptPubKey = CScript() << ParseHex("04678afdb0fe5548271967f1a67130b7105cd6a828e03909a67962e0ea1f61deb649f6bc3f4cef38c4f35504e51ec112de5c384df7ba0b8d578a4c702b6bf11d5f") << OP_CHECKSIG;
258 2011-07-16 01:43:47 <genis1> senseles: I know an accountant in jail for 10 years that did that for lots of people with off short accounts/foundations trusts
259 2011-07-16 01:43:51 <satoshi_> phantomcircuit: 'phantom' + circuit?
260 2011-07-16 01:43:59 <satoshi_> cool?
261 2011-07-16 01:44:01 <phantomcircuit> sacarlson, convert to ascii
262 2011-07-16 01:44:02 <senseles> that sucks
263 2011-07-16 01:44:12 <IO-> hi satoshi!
264 2011-07-16 01:44:13 <phantomcircuit> will the real satoshi please stand up
265 2011-07-16 01:44:29 <gmaxwell> someone care to ban the satoshi impersonator?
266 2011-07-16 01:44:31 <satoshi_> or the real phantom.
267 2011-07-16 01:44:48 <phantomcircuit> <-- is the real phantom
268 2011-07-16 01:45:11 <denisx> Iam Brian and my wife is also Brian!
269 2011-07-16 01:45:20 <phantomcircuit> wait wait wait
270 2011-07-16 01:45:21 <weidai> I am the real satoshi
271 2011-07-16 01:45:32 <phantomcircuit> satoshi_, you are the real satoshi aren't you
272 2011-07-16 01:45:51 <satoshi_> i'd answer that...but i'd have to kill ya.
273 2011-07-16 01:46:00 <phantomcircuit> oh well that's a no
274 2011-07-16 01:46:02 <phantomcircuit> thanks for playing
275 2011-07-16 01:48:04 <bliket_> The Times 03/Jan/2009 Chancellor on brink of second bailout for banks
276 2011-07-16 01:48:40 <bliket_> http://business.timesonline.co.uk/tol/business/industry_sectors/banking_and_finance/article5434660.ece
277 2011-07-16 01:48:51 <bliket_> i think satoshi expected us all to read this
278 2011-07-16 01:48:59 <bliket_> TLDR
279 2011-07-16 01:50:42 <bliket_> It's just meant for proof that the blockchain didn't start until jan 13th
280 2011-07-16 01:50:58 <bliket_> 2009
281 2011-07-16 01:51:09 <bliket_> jan 3rd*
282 2011-07-16 01:51:19 <zeph97> he could choose other article, but why that was chosen?
283 2011-07-16 01:51:28 <zeph97> yeah, he want us to read it :)
284 2011-07-16 01:51:43 <gmaxwell> Because its a snark about the current economic systems in use.
285 2011-07-16 01:53:23 <bliket_> maybe he had no political motivations
286 2011-07-16 01:53:42 <Akiron> gmaxwell: any sense where the special transaction that indicates where new mined bitcoins are to be sent is made and included in a block?
287 2011-07-16 01:53:50 <bliket_> it's just a coincidence he used some random financial new story
288 2011-07-16 01:54:43 <Akiron> bliket_: he choose a British newspaper to get everyone off the scent that he lives in the U.S.
289 2011-07-16 01:55:14 <gmaxwell> Akiron: hm? it's called the coinbase transaction, it's the first one included in the block.
290 2011-07-16 01:55:26 <unclemantis> i am looking at Casascius's paper wallet and I see a Private key and a hex code. When I try and convert the private key (which i think is base58) to hex (base16) I get an error saying that the key is not a valid base58 string. I then do the reverse and try and change the hex to base58 and I get something that completely does not resemble the private key at all. What am i doing wrong?
291 2011-07-16 01:55:36 <gmaxwell> Akiron: he pretty reliably used UK english.
292 2011-07-16 01:55:37 <Akiron> gmaxwell: oh duh, that makes so much sense
293 2011-07-16 01:56:02 <Akiron> gmaxwell: yeah, but he sleeps like he's in the Western Hemisphere
294 2011-07-16 01:56:39 <bliket_> when have we last heard from satoshi?
295 2011-07-16 01:56:56 <gmaxwell> Akiron: so do many UK hackers, night shifted...
296 2011-07-16 01:57:18 <Akiron> http://forum.bitcoin.org/index.php?action=profile;u=3;sa=statPanel
297 2011-07-16 01:57:24 <hwolf> Akiron: are you saying programers work 9-5pm?  Most guys i know stay up all night long
298 2011-07-16 01:57:28 <Akiron> i guess satoshi deleted his posts
299 2011-07-16 01:58:48 <Akiron> hwolf: well, it's possible, but absent evidence to suggest so, i'm not moved
300 2011-07-16 01:59:06 <gmaxwell> Who knows? Who cares. It's not important.
301 2011-07-16 01:59:15 <Akiron> satoshi deleted his posting time activity
302 2011-07-16 01:59:21 <Akiron> good thing i have a backup
303 2011-07-16 01:59:34 <gmaxwell> Akiron: or it doesn't keep it for inactive users.
304 2011-07-16 01:59:47 <Akiron> possible
305 2011-07-16 02:00:07 <Akiron> in any event, fun to speculate
306 2011-07-16 02:00:09 <gmaxwell> the posts are still there.
307 2011-07-16 02:00:39 <hwolf> I like the wacky theories, like he is from the future.
308 2011-07-16 02:00:50 <hwolf> or from outer space
309 2011-07-16 02:01:15 <unclemantis> or from the ground
310 2011-07-16 02:02:10 <bliket_> he is a plant
311 2011-07-16 02:02:14 <gmaxwell> He's actually a woman in south america.
312 2011-07-16 02:02:41 <bliket_> http://www.youtube.com/watch?v=T-sxSd1uwoU I found him
313 2011-07-16 02:02:43 <gmaxwell> anibalayles@yahoo.com.ar
314 2011-07-16 02:02:51 <gmaxwell> ^ Satoshi
315 2011-07-16 02:03:58 <hwolf> he is living and functioning in 6th dimensional space with bee's
316 2011-07-16 02:04:21 <senseles> kind of not surprising that it was invented in argentina
317 2011-07-16 02:04:22 <hwolf> you see the reason all of the bee population is dieing is because BTC was created
318 2011-07-16 02:04:25 <Akiron> satoshi = deep throat
319 2011-07-16 02:06:01 <gmaxwell> yea, a portal to the bee dimension was opened by the tevatron, and satoshi came through to teach use the method that the hyper-bees use for consensus currency.
320 2011-07-16 02:06:19 <hwolf> I knew it!
321 2011-07-16 02:06:29 <gmaxwell> The whole hive heritage made them really good at decenteralized protocols.
322 2011-07-16 02:07:24 <bliket_> that isnt satoshi
323 2011-07-16 02:07:36 <bliket_> that is just some guy that happened to be called satoshi
324 2011-07-16 02:07:54 <gmaxwell> And isn't satoshi some guy that happened to be called satoshi ??
325 2011-07-16 02:07:56 <gmaxwell> ;)
326 2011-07-16 02:08:03 <bliket_> [21:02] <gmaxwell> anibalayles@yahoo.com.ar
327 2011-07-16 02:08:50 <gmaxwell> well of course not, does that look like the name of a cross dimensional hyper-bee?
328 2011-07-16 02:10:01 <bliket_> https://www.bitcoinmarket.com/market/members/view/?market_memberID=1357
329 2011-07-16 02:10:18 <bliket_> apparently this guy has only traded 4 times on the bitcoin market
330 2011-07-16 02:11:36 <CIA-103> bitcoin: Eric Hosmer master * r88cab1e / src/makefile.vc : Added crypter to makefile.vc. - http://bit.ly/mSylms https://github.com/bitcoin/bitcoin/commit/88cab1e4a32febe9cd4633f03cdaff830110f2d1
331 2011-07-16 02:11:37 <CIA-103> bitcoin: Jeff Garzik master * r133ccbe / src/makefile.vc : Merge pull request #418 from EricJ2190/master ... https://github.com/bitcoin/bitcoin/commit/133ccbe4087514501dec1f7496c62489437f0db8
332 2011-07-16 02:13:49 <hwolf> bee's had one purpose, and that was to explain and teach the only human that could understand 6th dimensional thought how to create a decentralized form of money.  the creation of bitcoin coincides with the drastic decline in bee populations.  the bee's are checking out of planet earth now that bitcoin was created
333 2011-07-16 02:14:13 <IO-> nono, satoshi was the fifth beatle
334 2011-07-16 02:14:49 <hwolf> satoshi is actually a bee's that figures out how to communicate over the internet
335 2011-07-16 02:14:59 <hwolf> I mean he is a group of bee's
336 2011-07-16 02:14:59 <sacarlson> does the coinDD of 30% mean that bitcoin has only had 36.2% of it's total coins ever in circulation?  as seen in http://abe.john-edwin-tobey.org/
337 2011-07-16 02:15:40 <quellhorst> bees have moved their brains to the internet
338 2011-07-16 02:15:48 <quellhorst> in amazon ec2 and s3
339 2011-07-16 02:16:09 <hwolf> the bee's started the whole p2p cloud revolution and we were just now catching on
340 2011-07-16 02:16:34 <b4epoche> damn africanized bees
341 2011-07-16 02:17:13 <IO-> africanInternetized bees?
342 2011-07-16 02:17:59 <gmaxwell> sacarlson: thats not what coinDD is measuring.
343 2011-07-16 02:18:07 <sacarlson> everyone know that satoshi is an extraterstrial that never even came to earth just taped into the Internet and provided bitcoin tech.
344 2011-07-16 02:18:14 <sacarlson> gmaxwell: oh what is it?
345 2011-07-16 02:18:49 <gmaxwell> sacarlson: if you hold onto 10 btc for 1 day and send it to another address thats 10 bitcoin days destroyed.
346 2011-07-16 02:19:20 <gmaxwell> That char calculates what percentage of the total possible number of BDD have happened.
347 2011-07-16 02:19:52 <gmaxwell> So e.g. if the first block were spent 24 hours after it was first created, then that would destroy just one day (a very small number from that count)
348 2011-07-16 02:20:27 <gmaxwell> effectively it's a measure of how much of the economic activity is coins that have been sitting still for a while.
349 2011-07-16 02:21:23 <beppu> has satoshi resurfaced or is he still MIA?
350 2011-07-16 02:21:26 <sacarlson> gmaxwell: I see my number as only 7% because my economic activity on weeds is small
351 2011-07-16 02:21:55 <sacarlson> beppu: I told you guys he's an ET
352 2011-07-16 02:22:06 <sacarlson> beppu: had to back to the big ship
353 2011-07-16 02:23:33 <beppu> will bitcoin be enough to wrest control of the money supply from the Rotschilds  ;)
354 2011-07-16 02:23:34 <sacarlson> gmaxwell: so I wonder how would this coinDD compare if seen against say the USD or ERO?
355 2011-07-16 02:24:31 <CheapScotsman> So I got a second PSU set up but now it has both of my core clocks stuck at 157MHz for two 5830s and I can't change it...
356 2011-07-16 02:24:41 <CheapScotsman> Anyone have an idea of what that may be?
357 2011-07-16 02:36:41 <BlueMattBot> Project Bitcoin-Test build #10: STILL FAILING in 18 sec: http://www.bluematt.me/jenkins/job/Bitcoin-Test/10/
358 2011-07-16 03:39:59 <devrandom> ;;later tell BlueMatt applied WM patch at https://github.com/devrandom/wxWidgets-release and new bitcoin gitian config at https://github.com/devrandom/bitcoin/blob/master/contrib/gitian.yml
359 2011-07-16 03:39:59 <gribble> The operation succeeded.
360 2011-07-16 03:41:19 <devrandom> ;;later tell BlueMatt not fully tested yet
361 2011-07-16 03:41:19 <gribble> The operation succeeded.
362 2011-07-16 03:44:53 <genjix> any channel admins around>
363 2011-07-16 03:44:54 <genjix> ?
364 2011-07-16 03:45:01 <genjix> nanotube: ?
365 2011-07-16 03:45:58 <genjix> brocktice: hey
366 2011-07-16 03:46:37 <genjix> Keefe: hey
367 2011-07-16 03:46:47 <genjix> someone who has chanoops on #bitcoin-otc
368 2011-07-16 03:48:03 <genjix> cmon, you can stop people getting potentially scammed
369 2011-07-16 04:22:55 <bliket_> Early adopter selling all his bitcoins: http://blockexplorer.com/address/12higDjoCCNXSA95xZMWUdPvXNmkAduhWv
370 2011-07-16 04:23:01 <bliket_> on 6-15
371 2011-07-16 04:23:11 <bliket_> exactly 1 month ago
372 2011-07-16 04:24:18 <bliket_> 77624BTC
373 2011-07-16 04:25:33 <wirehead> he can retire now
374 2011-07-16 04:26:22 <wirehead> his magnum opus was setting up a bunch of rigs and letting them run all day
375 2011-07-16 04:50:30 <infinitev> so can someone fill me in is hdtv chips the newest thing in bitcoin mining?
376 2011-07-16 04:50:35 <infinitev> or is there an even better one now
377 2011-07-16 04:51:17 <[Tycho]> What is "hdtv chips" ?
378 2011-07-16 04:57:06 <infinitev> people were ripping the chips from hdtv and using them to calculate the SHA hashes at quite a high speed anyone here know anything about it
379 2011-07-16 04:58:22 <infinitev> lol
380 2011-07-16 04:58:49 <wirehead> that sounds so impractical
381 2011-07-16 04:58:51 <forrestv> infinitev, i'd guess that those are just fpgas
382 2011-07-16 04:59:00 <forrestv> there's no reason for them to have a sha-256 asic ...
383 2011-07-16 04:59:16 <wirehead> oh my gaw let's butcher hdtvs for their chips instead of directly ordering them
384 2011-07-16 04:59:18 <infinitev> ahh
385 2011-07-16 04:59:32 <infinitev> well i haven't done it
386 2011-07-16 04:59:44 <infinitev> and those things die soo quick, they just replace them if they go
387 2011-07-16 05:00:10 <infinitev> so i don't think anyone is going out anf getting hdtv just for chips but using old ones for parts etc
388 2011-07-16 05:01:58 <infinitev> trying to find the calculation but they were managing to get 4.0 Ghash/s they reckoned
389 2011-07-16 05:02:50 <infinitev> just wanted to know if anyone had messed with this tech
390 2011-07-16 05:03:51 <infinitev> forrestv: thats a pretty good hash rate yeah, or is that slow compared to some new technique i'm not aware of yet?
391 2011-07-16 05:04:40 <doublec> it sounds like a myth
392 2011-07-16 05:05:04 <infinitev> well i'll get the calculation
393 2011-07-16 05:05:09 <infinitev> and you can see for yourself
394 2011-07-16 05:05:20 <infinitev> the maths is solid
395 2011-07-16 05:06:08 <infinitev> i'm not trying to make out like I know, but really just trying to get your opinions
396 2011-07-16 05:06:46 <doublec> my opinion is it is a clever marketing campaign by hdtv manufacturers to sell more tv's
397 2011-07-16 05:06:59 <infinitev> lol @ wirehead --> thats it send me your hdtv i'm gonna rip out the chips :P
398 2011-07-16 05:07:15 <doublec> obviously they are jealous of the number of ati gpu's being sold for mining purposes
399 2011-07-16 05:07:23 <infinitev> doublec.. ok
400 2011-07-16 05:07:32 <infinitev> lol
401 2011-07-16 05:09:30 <infinitev> FPGA mining is different to GPU mining right?
402 2011-07-16 05:10:27 <infinitev> doublec: cause what you said would only make sense if they were :P
403 2011-07-16 05:10:49 <infinitev> well it really wouldn't make sense either way
404 2011-07-16 05:10:51 <infinitev> but hey
405 2011-07-16 05:11:25 <infinitev> ***weren't
406 2011-07-16 05:11:39 <infinitev> grammar out the window time for coffee
407 2011-07-16 05:12:35 <infinitev> How likely is it that running Bitcoin mining on FPGA mining will come to replace GPU mining? If it is likely, how long will it be?
408 2011-07-16 05:12:36 <infinitev> Consider the following:
409 2011-07-16 05:12:36 <infinitev> FPGA is more energy (and space) efficient than GPUs.
410 2011-07-16 05:12:37 <infinitev> optimizations which are currently kept secret (see #1).
411 2011-07-16 05:13:56 <infinitev> ^^^^^ read that doublec
412 2011-07-16 05:15:59 <doublec> infinitev: I'm not at all interested, sorry
413 2011-07-16 05:16:21 <doublec> infinitev: My comment was only to indicate my complete disbelief that anything involving hdtv's is going on
414 2011-07-16 05:18:08 <infinitev> they used hdtv chips to make the fpga array
415 2011-07-16 05:18:10 <mtrlt> yeah sounds like bullshit
416 2011-07-16 05:18:16 <infinitev> okay
417 2011-07-16 05:18:20 <infinitev> its bullshit
418 2011-07-16 05:18:26 <infinitev> keep mining with your gpu's
419 2011-07-16 05:19:10 <mtrlt> well so far i've only heard one person tell an anecdotal story
420 2011-07-16 05:19:23 <senseles> it's already not profitable on most systems
421 2011-07-16 05:19:34 <senseles> but that doesnt stop people from trying to buy 6990s
422 2011-07-16 05:19:44 <senseles> human beings are dumb
423 2011-07-16 05:19:49 <infinitev> mtrlt: constructive all i heard from you was thats bullshit
424 2011-07-16 05:19:59 <mtrlt> infinitev: that's all one can say to an anecdotal story
425 2011-07-16 05:20:05 <doublec> infinitev: I'm not saying fpga's are or are not useful, or gpu's are or are not dead
426 2011-07-16 05:20:15 <mtrlt> infinitev: anecdotal story that sounds very far-fetched, i might ad.
427 2011-07-16 05:20:17 <mtrlt> add
428 2011-07-16 05:20:20 <infinitev> doublec: ahh
429 2011-07-16 05:20:22 <doublec> infinitev: like mtrlt, there's nothing showing anything about hdtv's other than you
430 2011-07-16 05:20:35 <doublec> infinitev: so I'm saying "I don't believe this hdtv stuff"
431 2011-07-16 05:20:36 <infinitev> mtrlt: how about do some research then answer the wuestion when you know what your talking about
432 2011-07-16 05:20:40 <mtrlt> yea i was talking about the hdtv stuff
433 2011-07-16 05:20:44 <infinitev> i am not saying this is true thats why i am fucking asking
434 2011-07-16 05:20:49 <mtrlt> i know fpgas can be used for mining.
435 2011-07-16 05:21:28 <doublec> infinitev: thanks for asking. the answer to your query is "highly unlikely"
436 2011-07-16 05:21:42 <infinitev> what is an fpga array's made up of
437 2011-07-16 05:21:43 <mtrlt> i concur
438 2011-07-16 05:21:44 <infinitev> tell me
439 2011-07-16 05:21:50 <doublec> infinitev: if your query is "are people ripping apart hdtv's for mining"
440 2011-07-16 05:22:05 <infinitev> well i fucking read an article on bitcoin forum that said that dude
441 2011-07-16 05:22:09 <doublec> infinitev: I don't care what fpga array's are made of
442 2011-07-16 05:22:14 <mtrlt> infinitev: you'd better give a link then :P
443 2011-07-16 05:22:15 <Eliel_> well, if he hdtv stuff has fpga:s in them that can be removed intact, there are some people positioned so they could get them reasonably cheaply that way.
444 2011-07-16 05:22:26 <infinitev> thanks Eliel_
445 2011-07-16 05:22:37 <mtrlt> Eliel_: but i find that highly unlikely
446 2011-07-16 05:22:40 <mtrlt> Eliel_: the first part that is.
447 2011-07-16 05:22:53 <infinitev> well you'd be wrong mtrlt according to my research
448 2011-07-16 05:22:54 <mtrlt> the second part is possible
449 2011-07-16 05:23:06 <mtrlt> infinitev: so far you've only talked and not given a single link anywhere
450 2011-07-16 05:23:12 <infinitev> one sec
451 2011-07-16 05:23:20 <doublec> infinitev: so why ask, if you've done all the research already?
452 2011-07-16 05:23:26 <mtrlt> yeah
453 2011-07-16 05:23:33 <infinitev> cause i am asking for someone ho knows about it
454 2011-07-16 05:23:37 <infinitev> not people who don't
455 2011-07-16 05:23:53 <infinitev> who are gonna say i don't know but its unlikely
456 2011-07-16 05:23:53 <mtrlt> >_>
457 2011-07-16 05:24:11 <mtrlt> where's the link at
458 2011-07-16 05:24:18 <doublec> he's busy writing it
459 2011-07-16 05:24:21 <infinitev> fuck ya if i had the link
460 2011-07-16 05:24:27 <infinitev> i would have pasted it by now
461 2011-07-16 05:24:30 <infinitev> douchebag
462 2011-07-16 05:24:37 <infinitev> i'm trying to find it
463 2011-07-16 05:24:39 <mtrlt> lol
464 2011-07-16 05:25:01 <doublec> if I had shares in hdtv companies I'd be writing it right now
465 2011-07-16 05:26:28 <mtrlt> okay just did a google search for +hdtv fpga bitcoin
466 2011-07-16 05:26:39 <mtrlt> was it that hard ,really :P
467 2011-07-16 05:27:49 <infinitev> ahah
468 2011-07-16 05:28:01 <infinitev> http://hackaday.com/2007/08/31/nsahome-diy-shared-fpga-cracker/
469 2011-07-16 05:28:07 <infinitev> https://forum.bitcoin.org/index.php?topic=16918.0
470 2011-07-16 05:28:38 <infinitev> so yeah
471 2011-07-16 05:29:03 <mtrlt> well, now i believe :P
472 2011-07-16 05:29:06 <infinitev> can someone who knows what they're talking about fill me in whether this technique is the best to go forward with or if there is something better to do with
473 2011-07-16 05:29:34 <mtrlt> if you can get cheap fpgas, go for it.
474 2011-07-16 05:29:47 <Eliel_> yep, this won't work for everyone
475 2011-07-16 05:30:40 <Eliel_> good thing I've got flat rate electricity :)
476 2011-07-16 05:31:04 <Eliel_> so, other than dealing with the heat, I don't need to worry about energy efficiency :P
477 2011-07-16 05:31:12 <infinitev> really flat rate electricity
478 2011-07-16 05:31:15 <infinitev> thats awesome
479 2011-07-16 05:31:29 <infinitev> what country is that
480 2011-07-16 05:31:44 <infinitev> Eliel_: ?
481 2011-07-16 05:31:54 <senseles> they did that stuff in crappy apartments in college
482 2011-07-16 05:32:08 <Eliel_> Finland, student apartment :P
483 2011-07-16 05:32:08 <senseles> flat rate electricity i mean
484 2011-07-16 05:32:13 <mtrlt> Eliel_: heat and the noise :-P
485 2011-07-16 05:32:32 <mtrlt> i couldn't have had 4x6990 in the student box i lived in a year ago.
486 2011-07-16 05:33:00 <Eliel_> yeah, if I'm going to get that much heating going on, I do need to figure out some way to get the heat out
487 2011-07-16 05:33:17 <Eliel_> so far I'm only doing 550Mh/s though, so...
488 2011-07-16 05:33:42 <cjdelisle> fpgas are interesting but I figure at some point someone will show up with asics for sale cheap.
489 2011-07-16 05:38:28 <Eliel_> well, at least that fpga thing (advertised as cheap) is ... well... expensive.
490 2011-07-16 05:38:34 <senseles> lol
491 2011-07-16 05:38:36 <Eliel_> $20k for one board
492 2011-07-16 05:39:29 <senseles> i'd pay that if it were capable of like 40ghash/s
493 2011-07-16 05:39:40 <Eliel_> yeah, it only does around a tenth of that.
494 2011-07-16 05:39:50 <Eliel_> at least with current software
495 2011-07-16 05:39:59 <senseles> i guess the biggest step is production?
496 2011-07-16 05:40:08 <senseles> once its designed, etc?
497 2011-07-16 05:40:42 <Eliel_> yep, think so.
498 2011-07-16 05:41:15 <[Tycho]> The biggest stop is mask creation.
499 2011-07-16 05:41:35 <senseles> http://asicminer.net/?p=22
500 2011-07-16 05:41:52 <[Tycho]> Mask set for modern real ASIC should cost about $1 000 000
501 2011-07-16 05:42:01 <senseles> wonder what thats about
502 2011-07-16 05:42:05 <Eliel_> but if the hdtv systems use those same fpga chips, someone able to put together a board like this and then make it usable on a system... well... I'm finding myself doubting the chips from hdtv:s would be so useful
503 2011-07-16 05:42:51 <Eliel_> [Tycho]: the mask, that's something physical?
504 2011-07-16 05:42:56 <[Tycho]> Yes.
505 2011-07-16 05:43:21 <pixglen> hey guys, help test my bitcoin exchange http://forum.bitcoin.org/index.php?topic=29057.0
506 2011-07-16 05:44:31 <[Tycho]> Structured ASICs are like 50 times cheaper.
507 2011-07-16 05:45:18 <cjdelisle> Yea those links don't look promising
508 2011-07-16 05:46:37 <cjdelisle> I was imagining someone who worked at a foundry that did small production runs would build them but I suppose it would be more investement than any individual could pony up.
509 2011-07-16 05:51:44 <senseles> arent there existing chips you can just plaster on pcb ?
510 2011-07-16 05:52:09 <senseles> when you're talking about production runs you mean for a specialized processor?
511 2011-07-16 05:52:14 <Eliel_> cjdelisle: aside from those with huge numbers of bitcoins stacked up.
512 2011-07-16 05:53:38 <senseles> so fpga chips you're looking at like 200mhash/s for 8 watts?
513 2011-07-16 05:53:47 <senseles> there was a thread on the forum where some guy was making his own boards
514 2011-07-16 05:53:50 <[Tycho]> senseles, check wikipedia for "FPGA" and "ASIC" terms to see the difference.
515 2011-07-16 05:54:21 <senseles> i understand the difference but i wasnt sure if you were talking about the production of PCB, etc to be expensive
516 2011-07-16 05:54:30 <senseles> or if it was the chip fabrication that was the most expensive
517 2011-07-16 05:54:37 <senseles> seems FPGA would be a good short cut
518 2011-07-16 05:54:56 <[Tycho]> PCBs are cheap
519 2011-07-16 05:55:24 <senseles> so if you can get your hands on some cheap FPGA chips you could slap them on PCB and at least have a more cost effective way than GPU's to mine
520 2011-07-16 05:55:41 <senseles> im assuming thats what the guy was talking about with HDTV
521 2011-07-16 05:55:43 <senseles> chips
522 2011-07-16 05:55:48 <[Tycho]> FPGAs are expensive and a little bit slower, but you can use different firmware and upgrade them as you wish.
523 2011-07-16 05:56:30 <[Tycho]> ASICs are cheap, but they aren't reprogrammable and you need very expensive masks to start a production run of real chips.
524 2011-07-16 05:56:46 <senseles> mask to etch onto the wafer?
525 2011-07-16 05:56:50 <[Tycho]> Yes.
526 2011-07-16 05:57:12 <[Tycho]> (Cheap if thousands+ ordered, of course)
527 2011-07-16 05:57:27 <senseles> once you have a mask is it reusable or does it degrade over time?
528 2011-07-16 05:57:35 <senseles> like, is there a maximum production run per mask?
529 2011-07-16 05:57:37 <[Tycho]> It will last long enough.
530 2011-07-16 05:59:58 <senseles> can one chip design for an asic be better than another?
531 2011-07-16 06:00:13 <senseles> or would 2 completely different people who set out to make an asic for mining end up at the same result
532 2011-07-16 06:00:44 <senseles> would spacing matter? like, 32nm etching vs 65nm?
533 2011-07-16 06:00:50 <[Tycho]> Yes.
534 2011-07-16 06:01:03 <[Tycho]> It's very important.
535 2011-07-16 06:01:29 <[Tycho]> Chip packaging is also important.
536 2011-07-16 06:02:25 <senseles> so even if someone did set out to make an asic
537 2011-07-16 06:02:32 <senseles> there could be an overlooked flaw when they go to make the mask
538 2011-07-16 06:02:39 <senseles> and then they end up printing 10,000 nonfunctional chips
539 2011-07-16 06:02:43 <senseles> that would suck
540 2011-07-16 06:03:32 <[Tycho]> Well, they should do appropriate design tests.
541 2011-07-16 06:17:48 <Joric> starting from today google code supports git
542 2011-07-16 07:07:22 <CheapScotsman> I'm having an issue where GUIMiner will open to my taskbar but won't pull up into a GUI window
543 2011-07-16 07:07:34 <CheapScotsman> and no double clicking or exiting/opening changes such, anyone experience this?
544 2011-07-16 07:14:59 <rlifchitz> ;;bc,stats
545 2011-07-16 07:15:01 <gribble> Current Blocks: 136543 | Current Difficulty: 1564057.4508376 | Next Difficulty At Block: 137087 | Next Difficulty In: 544 blocks | Next Difficulty In About: 3 days, 10 hours, 57 minutes, and 36 seconds | Next Difficulty Estimate: 1679374.45194400
546 2011-07-16 07:22:56 <CIA-103> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r63ec45e / (10 files in 2 dirs): Renamed Accounting to KeyIndex. ... https://github.com/bitcoinjs/node-bitcoin-p2p/commit/63ec45eb9d6e83a43f004bd263e9eb359945a994
547 2011-07-16 07:27:13 <md2k7> I limit the connections on my only port-forwarded bitcoind
548 2011-07-16 07:27:49 <md2k7> I now have the problem that I can't connect to it via -addnode or -connect, because all connections are used up
549 2011-07-16 07:27:58 <md2k7> is there a way to "reserve" a connection slot for my other machines?
550 2011-07-16 07:29:35 <infinitev> [tycho], senseles thanks guys for that info
551 2011-07-16 07:30:18 <infinitev> and Eliel_
552 2011-07-16 07:32:28 <infinitev> from what i can see is that ROI on fpga arrays is okay atm, but as the difficulty of bitcoin goes up it will need to get much cheaper to be worthwhile
553 2011-07-16 07:44:59 <CIA-103> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r2793e3d / lib/blockchain.js : Insert all transactions for a block in a single query. - http://bit.ly/oWpt7f https://github.com/bitcoinjs/node-bitcoin-p2p/commit/2793e3d80454c20a2106dbe065711e31b2a60ca8
554 2011-07-16 08:13:31 <lfm> md2k7 I dont think there would be any way to reserve a connection port
555 2011-07-16 08:31:06 <prof7bit> i cant believe it. Google mail is giving me COBOL related advertisements now. I never told them my age.
556 2011-07-16 08:31:23 <BlueMatt> lol
557 2011-07-16 08:32:15 <Joric> well you're using pascal thats why
558 2011-07-16 08:32:20 <prof7bit> lol
559 2011-07-16 08:32:32 <dsockwell> reminds me of when google kept trying to sell me gene sequencers
560 2011-07-16 08:33:28 <Joric> germans landed http://bit-coin-trading.com/forum/index.php?topic=36.msg49#new
561 2011-07-16 08:33:28 <sipa> i used to play a online html game called 'the last knights' - it was remarkable how often i got an ad about "Find army singles now!"
562 2011-07-16 08:33:57 <Joric> "Because I had video proof of sending the BitCoins, PayPal actually reversed some of the chargebacks"
563 2011-07-16 08:35:09 <dsockwell> i think if that adwords campaign had lasted a little longer i'd be the only homeless man in the state who had his own genetics lab
564 2011-07-16 08:43:23 <prof7bit> i need a name for my bitcoin implementation
565 2011-07-16 08:43:40 <prof7bit> fpBitCoin maybe?
566 2011-07-16 08:43:54 <mtrlt> unimaginative :(
567 2011-07-16 08:44:08 <mtrlt> like, many pools have "bitcoin pool" in the name. equally unimaginative :(
568 2011-07-16 08:44:22 <mtrlt> or that diablo made a miner called diablominer
569 2011-07-16 08:47:30 <prof7bit> maybe fpBitCoin only for the non-gui core units that implement the protocol in the same tradition that many free pascal cross platform reimplementations of former delphi/windows only things start with fp in their name.
570 2011-07-16 08:48:57 <prof7bit> i could then postpone the decision how to name the full blown GUI app for a while.
571 2011-07-16 08:52:20 <Joric> i wrote a sound engine back in 2002, was inspired by the 'Useless Sound Player' kindof a nice name
572 2011-07-16 08:52:21 <prof7bit> or maybe something with "Turbo" in its name might also trigger some associations (in people who are older than 35)
573 2011-07-16 08:52:30 <Joric> name it useless bitcoin client
574 2011-07-16 08:53:43 <prof7bit> it should not contain words with negative associations.
575 2011-07-16 08:53:55 <Joric> http://www.hornet.org/cgi-bin/scene-search.cgi?search=FreddyV
576 2011-07-16 08:54:28 <mtrlt> well, how about "useless piece of shit". double negative cancels out
577 2011-07-16 08:55:23 <prof7bit> no
578 2011-07-16 09:11:37 <Joric> http://i54.tinypic.com/14ahpg3.jpg
579 2011-07-16 09:35:51 <prof7bit> how big are the openssl libraries on windows if i bundle my app with only the minimum needed binaries from openssl?
580 2011-07-16 09:36:40 <prof7bit> (too lazy now to start windows and download and install it)
581 2011-07-16 09:36:50 <BlueMatt> openssl binaries are really small
582 2011-07-16 09:36:53 <BlueMatt> compared to boost, etc
583 2011-07-16 09:38:55 <prof7bit> openssl will be the only dependency (and i needed 2 days to convince myself to use it instead of some small native pascal implementation)
584 2011-07-16 09:39:28 <sipa> how far are you with the implementation
585 2011-07-16 09:39:29 <sipa> ?
586 2011-07-16 09:41:33 <prof7bit> i spent the whole day (only my free time) with evaluating how to link to openssl yesterday and find out which parts of the headers to port, i did nothing on the p2p stuff yesterday.
587 2011-07-16 09:43:01 <prof7bit> now i have all stuff and all needed libs together and can start working on the application again.
588 2011-07-16 09:43:51 <prof7bit> the crypto stuff was the only big question mark remaining until yesterday
589 2011-07-16 09:44:55 <Zagitta> Is there any information about how to construct the merkle root on the wiki?
590 2011-07-16 09:45:32 <justmoon> Zagitta, BitcoinJ has a very well commented implementation
591 2011-07-16 09:46:29 <Zagitta> justmoon: splendid, thanks :)
592 2011-07-16 09:47:19 <prof7bit> there exists a native pascal implementation of ecdsa (along with its own unit to do big integer arithmetics) but i did not like it, the code looked not very elegant and very 1980's like (without making use of modern language features like some OOP and operator overloading for the bignum stuff)
593 2011-07-16 09:49:42 <prof7bit> and also i suspected endian problems in its bignum library and I want a completely portable implementation.
594 2011-07-16 09:51:01 <prof7bit> so i finally decided to use the well tested openssl instead.
595 2011-07-16 09:52:11 <Zagitta> :< i have a feeling all this block downloading and transaction fetching is going to be way over my head
596 2011-07-16 09:55:46 <mtrlt> Zagitta: just do it :P
597 2011-07-16 09:55:57 <mtrlt> don't think how hard it's gonna be
598 2011-07-16 09:59:50 <Zagitta> mtrlt: bu bu but :<
599 2011-07-16 10:00:54 <sipa> Zagitta: take a list of all transaction hashes, then iterate over it, in each step calculating sha256(sha256(hash1 | hash2))
600 2011-07-16 10:01:05 <sipa> reducing the length of the list in each step by two
601 2011-07-16 10:01:28 <sipa> and if there is an odd number of hashes in the list, replace the last one with sha256(sha256(hash | hash))
602 2011-07-16 10:01:48 <sipa> until there is only one hash left
603 2011-07-16 10:01:51 <sipa> that's the merkle root
604 2011-07-16 10:04:52 <Zagitta> sipa: That was easier than expected... thanks once again :)
605 2011-07-16 10:08:34 <Zagitta> does it matter in what order i hash things? Do i for example need to know all the transactions before hashing so they can be iterated or can i simply add them one by one like currentHash = sha256(sha256(hash1 | hash2)) and then add(newHash): currentHash sha256(sha256(currentHash | newHash))  ?
606 2011-07-16 10:13:13 <sipa> you don't add them one by one
607 2011-07-16 10:13:19 <sipa> you halve the number in each step
608 2011-07-16 10:18:54 <Eliel_> merkle tree it's called I believe.
609 2011-07-16 10:19:17 <sipa> yes, and the root of that tree is called the merkle root :)
610 2011-07-16 10:20:11 <Zagitta> sipa: i don't think you got my example but nvm it :)
611 2011-07-16 10:22:44 <sacarlson> with the new code for aux chain but after reading https://github.com/vinced/namecoin/blob/mergedmine/doc/README_merged-mining.md I think I could just run off aux on an alternate chain and hopefully only as mostly a listener to the bitcoin main net as aux. as a rule only one block per each bitcoin block added to the parent block chain
612 2011-07-16 10:22:48 <sacarlson> would that work?
613 2011-07-16 10:23:10 <erus`> http://www.youtube.com/watch?v=jvKUqmEiA3E
614 2011-07-16 10:25:45 <sacarlson> so a miner on this new alternate chain would just use bitcoin as a time stamp and first miner to use the validated stamp would mine the block in the alternate chain
615 2011-07-16 10:27:05 <sacarlson> this I would think would eliminated the need for powerfull mining on alternate chains
616 2011-07-16 10:31:19 <sacarlson> if this works it might eliminate the need for my licenced miner project I am happy to say
617 2011-07-16 10:32:56 <ar4s> how the hell do I talk in #bitcoin-otc :(
618 2011-07-16 10:33:14 <BlueMatt> gribble pms you instructions when you join -otc
619 2011-07-16 10:36:28 <ar4s> Ty, didn't think to check there
620 2011-07-16 10:36:29 <sacarlson> and I'm hoping wouldn't even require any changes in the bitcoin miners to implement as we only need to add changes to the  listener miners
621 2011-07-16 10:37:59 <cjdelisle> what exactly is the problem being solved by merged mining?
622 2011-07-16 10:38:29 <sacarlson> cjdelisle: in the namecoin or in this new alternate chain?
623 2011-07-16 10:38:58 <cjdelisle> In your chain
624 2011-07-16 10:39:33 <cjdelisle> I don't know what it solves in namecoin either but I assume namecoin has different requirements.
625 2011-07-16 10:39:34 <sacarlson> cjdelisle: in my chain it would reduce the power needed to keep security for unlimited aux chains
626 2011-07-16 10:40:14 <cjdelisle> hm
627 2011-07-16 10:40:43 <sacarlson> cjdelisle: as we see all these new chains are created that require miners and more power, all that we need is timestamps from the original bitcoin
628 2011-07-16 10:41:30 <cjdelisle> Ahh, so you are talking about creating an alt chain which tracks the mainline chain.
629 2011-07-16 10:42:23 <sacarlson> cjdelisle: yes as what is really done in bitcoin it to create validated time stamps to imprint transactions
630 2011-07-16 10:42:57 <cjdelisle> So you do some tiny pointless transaction and add to that transaction some piece of information which your chain depends on?
631 2011-07-16 10:43:30 <BlueMatt> sacarlson: you do realize that if you do this in transactions it actually provides no dicernable increase in security?
632 2011-07-16 10:43:38 <BlueMatt> only if its done in the coinbase
633 2011-07-16 10:44:06 <sacarlson> BlueMatt: no not in a transaction but in a mined block in the alternate chain
634 2011-07-16 10:44:24 <BlueMatt> that also provides nothing
635 2011-07-16 10:44:47 <BlueMatt> the only way to add security to an alternate chain through bitcoin's mining power is to add block hashes of the alternate chain to coinbase in bitcoin
636 2011-07-16 10:45:04 <sacarlson> BlueMatt: you could have multiple listener miners that monitor bitcoin and the first to broadcast there added solution would add to the alternate chain
637 2011-07-16 10:45:41 <sacarlson> BlueMatt: yes and that's what the new namecoin software does
638 2011-07-16 10:45:54 <BlueMatt> wait, you are doing what?
639 2011-07-16 10:46:02 <marvin__> ;bc,stats
640 2011-07-16 10:46:03 <BlueMatt> you have bitcoin block hashes stored in alt chain blocks?
641 2011-07-16 10:46:14 <sacarlson> BlueMatt: yes
642 2011-07-16 10:46:25 <BlueMatt> what advantage does that possibly give you?
643 2011-07-16 10:46:27 <marvin__> ;bc,stats
644 2011-07-16 10:46:38 <BlueMatt> ;;bc,stats
645 2011-07-16 10:46:40 <gribble> Current Blocks: 136568 | Current Difficulty: 1564057.4508376 | Next Difficulty At Block: 137087 | Next Difficulty In: 519 blocks | Next Difficulty In About: 3 days, 6 hours, 25 minutes, and 36 seconds | Next Difficulty Estimate: 1680539.80897239
646 2011-07-16 10:46:42 <marvin__> Hey, what happened to gribble?
647 2011-07-16 10:46:44 <marvin__> nvm
648 2011-07-16 10:46:58 <sacarlson> BlueMatt: I thought with the rule that you could only have one added block per bitcoin block you could pigy back a new chain
649 2011-07-16 10:47:24 <BlueMatt> that wouldnt work and wouldnt provide additional security
650 2011-07-16 10:47:41 <BlueMatt> a. many nodes throughout the network would have a different idea of the chain
651 2011-07-16 10:47:45 <sacarlson> BlueMatt: so how is namecoin method working then?
652 2011-07-16 10:48:01 <BlueMatt> no, ok it would work, but wouldnt provide more security
653 2011-07-16 10:48:03 <sacarlson> BlueMatt: https://github.com/vinced/namecoin/blob/mergedmine/doc/README_merged-mining.md
654 2011-07-16 10:48:28 <BlueMatt> how does that add security?
655 2011-07-16 10:48:29 <sacarlson> BlueMatt: wouldn't provide more security than what bitcoin?
656 2011-07-16 10:48:41 <BlueMatt> than just a standard separate chain
657 2011-07-16 10:48:58 <sacarlson> BlueMatt: a separate chain with only 300kh/sec?
658 2011-07-16 10:49:13 <BlueMatt> yes, this mining method adds no security to that
659 2011-07-16 10:49:46 <cjdelisle> sacarlson: you're talking about for example: mining a block then taking the hash of that block and donating a millionth of a coin to an imaginary address equal to the hash of that block?
660 2011-07-16 10:49:59 <BlueMatt> unless you are actually storing information in bitcoin blocks, you are adding nothing
661 2011-07-16 10:50:34 <cjdelisle> Because that would sort of work but then your wallet id would be hardcoded into the software and it would trust you asif you were god.
662 2011-07-16 10:50:51 <sacarlson> BlueMatt: you are adding all the transaction that happend in the time between the bitcoin event of a block mineing event
663 2011-07-16 10:51:30 <BlueMatt> sacarlson: the whole point of alternate chains is: you make not only the alternate chain miners verify the validity of their blocks, but you also make bitcoin miners do it as well.  In other words miners are mining on both chains at once with the same hashes
664 2011-07-16 10:51:41 <BlueMatt> what you are doing here is clearly not that, so you gain nothing
665 2011-07-16 10:51:57 <BlueMatt> the point is to get more mining power
666 2011-07-16 10:52:56 <sacarlson> BlueMatt: well I assume the linked mining of namecoin will add security to there net?
667 2011-07-16 10:53:53 <sacarlson> BlueMatt: but I thought the real goal of mining was to create timestamps but I must be wrong
668 2011-07-16 10:54:17 <BlueMatt> ah that link you just posted, no that is correct as it adds the current merkle hash to the coinbase of bitcoin blocks
669 2011-07-16 10:54:21 <BlueMatt> no that is not the point of mining
670 2011-07-16 10:55:25 <sacarlson> BlueMatt: so to take advantage of this new version I would still require mods to bitcoin miners or do a patch to some namecoin pools to acheave added security
671 2011-07-16 10:55:37 <BlueMatt> yep
672 2011-07-16 10:55:53 <BlueMatt> that is the *only* way to get more security in a second chain
673 2011-07-16 10:55:59 <BlueMatt> aside from adding more mining power, that is
674 2011-07-16 10:57:22 <sacarlson> BlueMatt: well I'm going to incorportate the namecoin method into MultiCoin
675 2011-07-16 10:57:37 <BlueMatt> well based on that doc, they did it right
676 2011-07-16 10:57:43 <BlueMatt> which requires bitcoin miners to change...
677 2011-07-16 10:57:57 <sacarlson> BlueMatt: or namecoin miners
678 2011-07-16 10:58:01 <BlueMatt> no
679 2011-07-16 10:58:06 <sacarlson> yes
680 2011-07-16 10:58:07 <BlueMatt> it requires bitcoin miners to change
681 2011-07-16 10:58:21 <sacarlson> not if namecoin is my aux and
682 2011-07-16 10:58:28 <CIA-103> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r68c2e7c / (6 files in 3 dirs): Ported RPC functions from jgarzik's getblock branch. - http://bit.ly/qmKmE5 https://github.com/bitcoinjs/node-bitcoin-p2p/commit/68c2e7c07f8fbf9187f13b94058ab0dcbec8bf18
683 2011-07-16 10:58:30 <BlueMatt> oh, I see what you are saying
684 2011-07-16 10:58:38 <BlueMatt> yea well yea if you use namecoin as the parent chain
685 2011-07-16 10:58:49 <BlueMatt> but why would you do that?
686 2011-07-16 10:58:53 <sacarlson> BlueMatt: yes namecoin as parant
687 2011-07-16 10:59:01 <BlueMatt> namecoin doesnt have nearly the same power as bitcoin
688 2011-07-16 10:59:32 <sacarlson> BlueMatt: because I would expect they require less power to produce I could use a smaller pool to get results
689 2011-07-16 11:00:06 <BlueMatt> well, the doc you posted isnt quite done right, its close, but not quite
690 2011-07-16 11:00:25 <BlueMatt> in the parent block, instead of storing a merkle root for the subchain
691 2011-07-16 11:00:40 <BlueMatt> you should store a merkle root for a merkle tree containing the merkle roots of subchains
692 2011-07-16 11:00:44 <sacarlson> BlueMatt: I guess I could create a good single linked to bitcoin and that could become the parant to many if it was that easy
693 2011-07-16 11:00:51 <BlueMatt> that way you can have unlimited subchains gaining the same security from bitcoin
694 2011-07-16 11:01:00 <BlueMatt> instead of crazy trees of sub and super chains
695 2011-07-16 11:02:02 <sacarlson> BlueMatt: I just asume a big net like bitcoin will take too long to get things done,  I want instant changes without much interaction
696 2011-07-16 11:02:22 <BlueMatt> no, you just need one miner who gets blocks
697 2011-07-16 11:02:40 <BlueMatt> you dont need to get every miner to change
698 2011-07-16 11:02:47 <BlueMatt> say, just get one pool
699 2011-07-16 11:02:59 <BlueMatt> wont provide as much security, but if that pool is trustworthy enough, it would work
700 2011-07-16 11:03:07 <sacarlson> BlueMatt: so with one pool I'll get blocks every hour?
701 2011-07-16 11:03:27 <BlueMatt> you dont have to have *every* block in the superchain
702 2011-07-16 11:03:39 <BlueMatt> blocks are tied to their previous blocks so you can just have one block a week
703 2011-07-16 11:03:58 <BlueMatt> not as good security, but it provides some
704 2011-07-16 11:04:39 <sacarlson> BlueMatt: that's why I thought I didn't really need any blocks at all if one a week is good why not one per year or 12 years is what my computer creates
705 2011-07-16 11:05:02 <BlueMatt> what?
706 2011-07-16 11:05:16 <sacarlson> BlueMatt: never mine
707 2011-07-16 11:05:30 <BlueMatt> did you mean never mind, or never mine?
708 2011-07-16 11:07:05 <sacarlson> BlueMatt: means I'm not sure what I was talking about need to read more
709 2011-07-16 11:07:20 <sacarlson> never mind
710 2011-07-16 11:07:38 <senseles> Does anyone know anything about FPGA that would mind answering a couple of basic questions?
711 2011-07-16 11:07:42 <sacarlson> sorry mining on my mind
712 2011-07-16 11:08:27 <[Tycho]> senseles, what questions ?
713 2011-07-16 11:08:44 <senseles> well i was going through and reading all the forum posts about FPGA
714 2011-07-16 11:08:52 <senseles> my dad is an EE and i wanted to get a board to play with
715 2011-07-16 11:09:01 <senseles> I was looking on ebay and i found this
716 2011-07-16 11:09:12 <senseles> According to what I read on the forum
717 2011-07-16 11:09:30 <senseles> you need 5000 logic gates roughly for each pipeline
718 2011-07-16 11:09:36 <sacarlson> senseles: I used to play with xlinx and altera but that was long ago
719 2011-07-16 11:09:47 <senseles> according to the stats on that chip it's 25000 logic gates @ a max clock speed of 50mhz
720 2011-07-16 11:09:57 <senseles> according to my math that would me each chip would be capable of 200mhash/s
721 2011-07-16 11:10:01 <senseles> which can't be right
722 2011-07-16 11:10:10 <[Tycho]> 5000 gates ? So it's not really unfolded...
723 2011-07-16 11:10:34 <senseles> what do you mean?
724 2011-07-16 11:10:56 <senseles> i was a CE so i didn't take many of these classes :(
725 2011-07-16 11:11:00 <senseles> mainly programming stuff
726 2011-07-16 11:11:04 <[Tycho]> I would expect more gates to full pipeline (1 clock per hash)
727 2011-07-16 11:11:30 <sacarlson> senseles: the code is much like programing mostly case stuf we played with
728 2011-07-16 11:11:31 <senseles> the way i was reading it, it needed 2x 2500 gates for 1 hash per clock
729 2011-07-16 11:12:20 <senseles> or did i misread and it was 10000 gates for 1 hash per clock?
730 2011-07-16 11:12:32 <[Tycho]> Of course I may be wrong.
731 2011-07-16 11:13:20 <senseles> even with 10000 gates for 1 hash per clock
732 2011-07-16 11:13:29 <senseles> that would still be 100mhash/s per chip
733 2011-07-16 11:13:36 <senseles> for a 20$ chip
734 2011-07-16 11:14:00 <BlueMatt> but whats the power usage compared to a gpu?
735 2011-07-16 11:14:05 <senseles> i guess what im wondering is, everyone says it's expensive yet on ebay it seems like fpga bonanza
736 2011-07-16 11:14:11 <sacarlson> senseles: I thought there was already open source project for  fpga in bitcoin
737 2011-07-16 11:14:16 <BlueMatt> some of the fgpas arent that great
738 2011-07-16 11:14:46 <b4epoche> an fpga bonanza?  on ebay you say?
739 2011-07-16 11:14:49 <senseles> i didnt check the power usage but im assuming it'd be less than 75watts
740 2011-07-16 11:15:00 <senseles> theres a FPGA blade server in there for 9999$
741 2011-07-16 11:15:02 <mosimo> how much would it cost for the controller boards to run 72 chips like that?
742 2011-07-16 11:15:08 <mosimo> ah :P
743 2011-07-16 11:15:26 <senseles> well the PCB would have to be custom designed i suppose
744 2011-07-16 11:15:51 <senseles> looking at that NSA@Home project it didn't look all too complex
745 2011-07-16 11:16:01 <senseles> the guy had like 10 chips on PCB inside a 1u chassis
746 2011-07-16 11:16:14 <b4epoche> link?
747 2011-07-16 11:16:26 <senseles> http://nsa.unaligned.org/
748 2011-07-16 11:16:31 <sacarlson> senseles: and power consumption?
749 2011-07-16 11:17:08 <senseles> if it's FPGA power usage should be < 10 watts for 1 chip
750 2011-07-16 11:17:14 <senseles> i didn't check
751 2011-07-16 11:20:35 <senseles> everyone got quiet, did you guys go buy up all the FPGA?
752 2011-07-16 11:20:36 <senseles> :(
753 2011-07-16 11:21:07 <b4epoche> we're all looking for ways to debunk your idea
754 2011-07-16 11:21:40 <senseles> kinda what i was looking for because 100mhash/s for 20$ cant be possible
755 2011-07-16 11:22:29 <b4epoche> http://www.xilinx.com/support/documentation/data_sheets/ds083.pdf
756 2011-07-16 11:22:37 <b4epoche> but I don't see any power information
757 2011-07-16 11:23:32 <b4epoche> and those aren't cheap
758 2011-07-16 11:23:56 <b4epoche> where's the $20 ones?
759 2011-07-16 11:24:13 <senseles> http://cgi.ebay.com/XC52156PQ160C-LOT-72pcs-Xilinx-XC5215-6PQ160C-FPGA-/190348530553?pt=LH_DefaultDomain_0&hash=item2c51a81379
760 2011-07-16 11:24:24 <senseles> its actually 6.75$ each
761 2011-07-16 11:24:26 <senseles> 72 chips
762 2011-07-16 11:24:53 <senseles> is that right though?
763 2011-07-16 11:25:21 <senseles> logicgates / 5000 or 10000 (unsure if its 2x2500 or 2x5000 that is needed)
764 2011-07-16 11:25:31 <senseles> * clock speed
765 2011-07-16 11:25:34 <senseles> = mhash/s ?
766 2011-07-16 11:27:19 <senseles> http://forum.bitcoin.org/index.php?topic=8543.0;all
767 2011-07-16 11:28:45 <senseles> "5000 gates per hash engine"
768 2011-07-16 11:30:14 <senseles> with 25k gates on the chip 5 pipelines per chip x 50mhz clock = 250mhash/s?
769 2011-07-16 11:30:23 <senseles> cant be right
770 2011-07-16 11:31:27 <b4epoche> what rates was the NSA guy getting?
771 2011-07-16 11:31:55 <senseles> unknown
772 2011-07-16 11:32:02 <senseles> i also don't know what chip he's using
773 2011-07-16 11:32:24 <b4epoche> he says:  http://nsa.unaligned.org/hw.php
774 2011-07-16 11:34:30 <sacarlson> senseles: well they did publish power using about 240W of power
775 2011-07-16 11:35:28 <sacarlson> senseles: also equivalent to over 1500 Athlon FX-60 CPUs  but I'm not sure hash rate that produces
776 2011-07-16 11:35:59 <senseles> probably not much
777 2011-07-16 11:36:02 <b4epoche> looks like he got a real job:  http://www.linkedin.com/profile/view?id=4275930&authType=NAME_SEARCH&authToken=rVt0&locale=en_US&srchid=47dd52df-3cd3-4c4b-aa28-006a14223b05-0&srchindex=1&srchtotal=2&goback=%2Efps_PBCK_*1_Stanislaw_Skowronek_*1_*1_*1_*1_*2_*1_Y_*1_*1_*1_false_1_R_true_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2_*2&pvs=ps&trk=pp_profile_name_link
778 2011-07-16 11:36:37 <senseles> looks like 14ghash/s on his board
779 2011-07-16 11:36:48 <senseles> using the stats from an AMD 64 x2 6000
780 2011-07-16 11:36:51 <b4epoche> wow, that's a ugly URL
781 2011-07-16 11:36:58 <senseles> which should be comperable to a fx-60
782 2011-07-16 11:37:17 <senseles> 14ghash/s sounds nice :/
783 2011-07-16 11:37:20 <senseles> in that tiny little package
784 2011-07-16 11:39:25 <b4epoche> that fellow is awesome:  http://fireuser.com/blog/amd_opencl_parallel_computing_demo_from_siggraph_asia_2008/
785 2011-07-16 11:39:51 <senseles> is that the NSA guy?
786 2011-07-16 11:40:03 <b4epoche> yep
787 2011-07-16 11:40:27 <senseles> 58mhash/s per watt
788 2011-07-16 11:40:30 <senseles> heh
789 2011-07-16 11:41:09 <senseles> machine i just bought last wensday with 4 5830 is 1.5mhash/watt :/
790 2011-07-16 11:43:12 <senseles> wow his machine would be pretty cheap to duplicate
791 2011-07-16 11:43:18 <senseles> those 16x processors he has are only 60$ a pop
792 2011-07-16 11:43:27 <senseles> er 15x
793 2011-07-16 11:43:46 <senseles> the other XC2S50 is 100$ for 20 of em on ebay
794 2011-07-16 11:44:43 <senseles> maybe 1500-2000$ for 5000+mhash/s
795 2011-07-16 11:44:51 <[Tycho]> You should also consider the amount of work to make FPGA-PC interface.
796 2011-07-16 11:45:07 <senseles> drivers and what not?
797 2011-07-16 11:45:17 <senseles> I guess i would need a custom miner too
798 2011-07-16 11:45:43 <[Tycho]> It's relatively easy with development kit, but with bare chips it may me more difficult.
799 2011-07-16 11:46:04 <senseles> is my math at least right in regards to mhash/s to cost?
800 2011-07-16 11:46:05 <[Tycho]> *be
801 2011-07-16 11:46:19 <senseles> ill buy 10