1 2013-12-05 00:35:41 <HaltingState> how do i calculate target from "bits"?
  2 2013-12-05 00:44:32 <agath> Today I had a talk with an ASIC producer (Switzerland). He said that they are going to build an huge mining center in Asia...
  3 2013-12-05 00:44:46 <agath> and this is the trend that many will follow.
  4 2013-12-05 00:44:59 <agath> Isn't it too dangerous to concentrate so much mining power in such a few places?
  5 2013-12-05 00:46:06 <jakov> how much mining power will he repersent?
  6 2013-12-05 00:46:29 <hno> agath, yes there is dangers in large concentrated patches of hashing power, just as there is dangers in too large pools.
  7 2013-12-05 00:46:42 <agath> he said around 200K chips for the first production and then they would continue to produce more in the future...
  8 2013-12-05 00:46:46 <jakov> mining is very competative
  9 2013-12-05 00:47:18 <jakov> the only foreseeable problems afaik are centralised mining pools and bandwidth usage, which might make miners want to be located physically near
 10 2013-12-05 00:47:28 <jakov> if you have a miner, use p2pool
 11 2013-12-05 00:47:47 <agath> Probably there will be a point where less than a dozen of ASIC producers will manage huge mining datacenters, and most of them will be located in ASIA / China
 12 2013-12-05 00:47:58 <gmaxwell> agath: yes. Perhaps we'll commit suicide that way, who knows.
 13 2013-12-05 00:48:40 <agath> I raised this issue to him, and he said he knows that it poses a risk, but raised his shoulders and said "all the others are going that way, what could I do?"
 14 2013-12-05 00:48:44 <jakov> do large miners have economy of scale advantages over small ones?
 15 2013-12-05 00:48:48 <hno> jakov, bandwidth is not so much an issue in mining today. latency is somewhat, but not much. Reliability of the connection is.
 16 2013-12-05 00:48:49 <gmaxwell> agath: unfortunately a lot of these operations are now self funding with money they got from previously selling hardware in preorder, so they don't even have to raise money anymore and so don't get investor sanity checks.
 17 2013-12-05 00:49:10 <agath> gmaxwell: indeed.
 18 2013-12-05 00:49:30 <gmaxwell> jakov: no, not really. at least not beyond a moderate point. There are scaling disadvantages too. (e.g. waste heat usually becomes more costly to deal with at larger operating scales)
 19 2013-12-05 00:49:40 <upb> agath: you tried to persuade him not to produce those chips with the argument that 'it will hurt the coin' ?:P
 20 2013-12-05 00:50:12 <hno> there is several of these patches of hashing power being built.
 21 2013-12-05 00:50:13 <agath> upb: he knows about the risks, he is not stupid... but it seems that the only thing they are looking at is money/greedy
 22 2013-12-05 00:50:33 <gmaxwell> upb: the stronger argument is that your investment will be worthless if the centeralization undermines bitcoin. But at least the people I've talked to don't care about risks beyond a few months.
 23 2013-12-05 00:50:54 <gmaxwell> and they're also convinced that they'll be compensated if a state takes over their farm.
 24 2013-12-05 00:50:55 <agath> And of course this method (upb) wouldn't persuade him. As much as it wouldn't persuade you if you were in his skin.
 25 2013-12-05 00:52:49 <jakov> most centralisations in other sectors i can think of come from economy of scale
 26 2013-12-05 00:52:54 <jakov> which mining doesnt have
 27 2013-12-05 00:53:20 <hno> jakov, it does have clear scale benefits.
 28 2013-12-05 00:53:42 <jakov> what are they?
 29 2013-12-05 00:53:49 <gmaxwell> jakov: there are substantial economies of scale in production of equipment though, which are now being leveraged in a way that produces operating consolidations.
 30 2013-12-05 00:54:30 <jakov> i guess the hope is the producers will sell their asics rather than mine with them
 31 2013-12-05 00:54:41 <gmaxwell> hno: in operating? there aren't really. (and I say this as someone who has been 1% of the network hashrate in not so distant memory)
 32 2013-12-05 00:55:03 <gmaxwell> jakov: even when they're selling them now, they're selling them at prices that make them very risky purchases. I am not adding hardware.
 33 2013-12-05 00:55:09 <hno> jakov, low latency, good leverage for negotiation power agreements. Reduced labor costs in maintaining the equipment. Reduced costs in building the equipment.
 34 2013-12-05 00:58:09 <gmaxwell> hno: latency is a ~non issue. Good industrial power prices in the US are often only 20% lower than regular residential rates. Maintenance is hard to say, even with dozens of miners failure rates are low enough they're too low for me to estimate.
 35 2013-12-05 00:59:25 <jakov> the people operating the miners dont have to be those who built them
 36 2013-12-05 01:00:09 <gmaxwell> On the diseconomies of scale, at large scale you have to pay to remove heat where at small scale the waste he is actually useful (e.g. my home is miner heated). Likewise, they can fit into cheap excess residential space.
 37 2013-12-05 01:00:21 <cfields> michagogo|cloud: yep, in addition to the umask fun, there were a few places where date was stored without time
 38 2013-12-05 01:00:40 <gmaxwell> yea, on the mfg side there is a need to operate at a considerable scale.
 39 2013-12-05 01:00:59 <hno> gmaxwell, we are not talking dozens miners, it's many thousands.
 40 2013-12-05 01:01:16 <michagogo> cloud|Aha, so that explains why (again, this was a few months ago when I was first looking at gitian) I was getting non-deterministic deps that were identical run to run
 41 2013-12-05 01:01:32 <jakov> gmaxwell which latitude do you live at? i mean does it ever get warm enough in the summer that miner heat is bad
 42 2013-12-05 01:01:34 <michagogo> cloud|(I mean, there was no determinism necessary, but...)
 43 2013-12-05 01:02:37 <gmaxwell> jakov: at reasonably low densities (e.g. under 20kw) its easy, at least in mild climates, cool passively with just outside air.
 44 2013-12-05 01:03:16 <cfields> michagogo|cloud: yea. most are harmless doc files that can just be deleted from the tarball. The biggie was qmake itself.
 45 2013-12-05 01:03:49 <cfields> so you can test all day and it looks fine, but tomorrow it would change
 46 2013-12-05 01:03:51 <gmaxwell> jakov: 37deg.
 47 2013-12-05 01:05:35 <gmaxwell> re: scale, see those pictures of the asicminer's owners private farm?  they're using $100/liter novec liquid cooling.
 48 2013-12-05 01:06:22 <gmaxwell> just because air cooling would resuit in a loss of density that would drive their realestate costs sky high.
 49 2013-12-05 01:09:04 <hno> gmaxwell, there is other places where space is not such an issue.
 50 2013-12-05 01:09:27 <jakov> they need a place where land and energy are simultaniously cheap
 51 2013-12-05 01:09:49 <jakov> doesnt happen often, usually land gets more expensive as power lines are built to it
 52 2013-12-05 01:09:57 <gmaxwell> This is just what people are actually building.
 53 2013-12-05 01:36:48 <saracen> Right, fixed directory.io. Getting old now, but I also leaked all electrum passphrases. http://electrum.directory.io/
 54 2013-12-05 01:38:47 <michagogo> cloud|saracen: I thought electrum passphrases were 12 words?
 55 2013-12-05 01:39:49 <saracen> michagogo|cloud: Yeah, that's what they say. I realised brainwallet.org worked with smaller combinations though, that you can manually enter. I've never used electrum, so I dont know if thats also the case
 56 2013-12-05 01:40:14 <michagogo> cloud|Electrum won't let you do that, I'm pretty sure
 57 2013-12-05 01:41:39 <saracen> Well, I'm off to bed. Tomorrow I'll try to find out what the min/max seed is from electrum and fix :)
 58 2013-12-05 01:41:49 <saracen> nn!
 59 2013-12-05 01:43:58 <gmaxwell> michagogo|cloud: no, it doesn't force you to behave securely unfortunately.
 60 2013-12-05 01:44:39 <michagogo> cloud|gmaxwell: Electrum will allow you to "recover" with a <12-word string?
 61 2013-12-05 01:45:53 <gmaxwell> michagogo|cloud: yes.
 62 2013-12-05 01:46:24 <michagogo> cloud|:-/
 63 2013-12-05 01:46:51 <michagogo> cloud|wait, how did it get to be 3:45?
 64 2013-12-05 01:48:45 <lianj> "It took a lot of computing power to generate this database." obvious troll
 65 2013-12-05 02:20:07 <rweichler> how much bandwidth does bitcoind typically use?
 66 2013-12-05 02:20:11 <rweichler> per month
 67 2013-12-05 02:28:35 <Neozonz> Disc|what does ProcessBlock: ACCEPTED mean and SetBestChain?
 68 2013-12-05 02:28:40 <Neozonz> Disc|mean
 69 2013-12-05 02:30:22 <andytoshi> Neozonz|Disc: ProcessBlock processes new blocks as they are published; ACCEPTED means that it passed the validation checks (i.e. the hash is computed correctly and the transactions are valid)
 70 2013-12-05 02:30:37 <andytoshi> SetBestChain means that this latest block is the new head of the longest chain
 71 2013-12-05 02:30:55 <Neozonz> Disc|thank you
 72 2013-12-05 02:31:14 <Neozonz> Disc|if a miner friends a block and submits it to litecoind, would it show something else?
 73 2013-12-05 02:31:24 <Neozonz> Disc|or the same info as above
 74 2013-12-05 02:31:32 <Neozonz> Disc|*friends/finds
 75 2013-12-05 02:33:00 <Neozonz> Disc|what does getblocks -1 to 0000000000000000000000000000000000000000000000000000000000000000 limit 500
 76 2013-12-05 02:33:01 <Neozonz> Disc| mean?
 77 2013-12-05 02:33:26 <andytoshi> i'd assume it's the same, litecoind did not change much
 78 2013-12-05 02:34:11 <Neozonz> Disc|i mean if a local miner finds a block and submits it to bitcoind, would it show something else ;p
 79 2013-12-05 02:36:36 <michagogo> cloud|4:32:59 <Neozonz|Disc> what does getblocks -1 to 0000000000000000000000000000000000000000000000000000000000000000 limit 500 mean?
 80 2013-12-05 02:37:14 <michagogo> cloud|If I'm not mistaken, it's a request for the first 500 blocks in the chain
 81 2013-12-05 02:37:51 <Neozonz> Disc|michagogo|cloud, what for? is it for getwork for miners to work on?
 82 2013-12-05 02:38:01 <michagogo> cloud|No, IBD
 83 2013-12-05 02:38:01 <Neozonz> Disc|(sorry it's hard to find this info anywhere...)
 84 2013-12-05 02:38:07 <Neozonz> Disc|IBD?
 85 2013-12-05 02:38:11 <michagogo> cloud|Neozonz|Disc: it's on the woki
 86 2013-12-05 02:38:13 <michagogo> cloud|Wiki
 87 2013-12-05 02:38:24 <andytoshi> Neozonz|Disc: IBD means "initial block download"
 88 2013-12-05 02:38:25 <michagogo> cloud|;;bc,wiki getblocks
 89 2013-12-05 02:38:26 <gribble> http://en.bitcoin.it/wiki/protocol_specification | Oct 31, 2013 ... 3.1 version; 3.2 verack; 3.3 addr; 3.4 inv; 3.5 getdata; 3.6 notfound; 3.7 getblocks; 3.8 getheaders; 3.9 tx; 3.10 block; 3.11 headers; 3.12 getaddr ...
 90 2013-12-05 02:38:34 <michagogo> cloud|Neozonz|Disc: ^
 91 2013-12-05 02:40:30 <Neozonz> Disc|thank u
 92 2013-12-05 02:43:03 <Neozonz> Disc|What about CreateNewBlock()
 93 2013-12-05 04:20:53 <wizkid057> should probably use a DHT
 94 2013-12-05 04:20:59 <wizkid057> just saying
 95 2013-12-05 04:33:45 <Plinker> http://money.cnn.com/2013/12/04/technology/security/passwords-stolen/index.html
 96 2013-12-05 06:04:24 <AlexLibman> I am having a problem with bitcoin-qt crashing.  It originally complained about block-chain corruption, so I tried to reindex, but it crashed some way through.  I tried removing ~/.bitcoin and re-download it, but it still crashes.
 97 2013-12-05 06:04:39 <AlexLibman> Last ~/.bitcoin/debug.log line: "SetBestChain: new best=[snip lots of hex] height=177016 log2_work=68.036957 tx=2902603 date=2012-04-24 10:01:28 progress=0.049531"
 98 2013-12-05 06:05:05 <AlexLibman> Ubuntu 13.10; 3.11.0-14-generic x86_64; bitcoin-qt v0.8.5.0-gef14a26-beta
 99 2013-12-05 06:06:22 <AlexLibman> I tried a bunch of stuff, including reinstalling bitcoin-qt and switching to http://ppa.launchpad.net/bitcoin/bitcoin/ubuntu - didn't help.
100 2013-12-05 06:08:24 <AlexLibman> It ran while redownloading the chain for several hours, then crashed.  Now it crashes upon start-up.  Two lines of output, the final being "Bus error", and the one before...
101 2013-12-05 06:08:34 <AlexLibman> "sni-qt/17811" WARN  01:07:34.702 void StatusNotifierItemFactory::connectToSnw() Invalid interface to SNW_SERVICE
102 2013-12-05 06:15:01 <phantomcircuit> AlexLibman, what's the failure?
103 2013-12-05 06:15:04 <phantomcircuit> hmm weird
104 2013-12-05 06:15:17 <phantomcircuit> that's some weird qt stuff
105 2013-12-05 06:17:13 <kjj> try -reindex
106 2013-12-05 06:18:21 <kjj> my guess is that your index is corrupt and is pointing outside the block file.
107 2013-12-05 06:18:42 <kjj> ahh, shit.  you get this while trying to reindex
108 2013-12-05 06:20:16 <kjj> mv your index files out and use -loadblock
109 2013-12-05 06:21:13 <kjj> also, double check your hardware.  make sure your CPU isn't overheating or overclocked, and run memtest86
110 2013-12-05 06:22:32 <fooledbyprimes> quick newbie question:  I just want to download the current block-chain for fun.  possible? thanks.
111 2013-12-05 06:23:07 <kjj> fooledbyprimes: grab the torrent if you don't need to be totally current
112 2013-12-05 06:23:59 <fooledbyprimes> so is that going to grab a binary file… compressed probably, eh?
113 2013-12-05 06:24:49 <kjj> not compressed, no.  but yes, a binary file.  format is simple, and google-able
114 2013-12-05 06:25:27 <fooledbyprimes> righteous…!   I'll look around tomorrow.  Thanks.
115 2013-12-05 06:32:09 <AlexLibman> I've tried -reindex, tried deleting the whole ~/.bitcoin, still crashes before becoming usable.
116 2013-12-05 06:33:17 <kjj> pull the previous release, or check your hardware
117 2013-12-05 06:33:28 <AlexLibman> Now bitcoin-qt crashes immediately without arguments, or after a while without arguments.
118 2013-12-05 06:35:35 <AlexLibman> Errr, or after a while with "-reindex"
119 2013-12-05 06:35:57 <AlexLibman> Hmm, I think previous ubuntu version (raring) was 0.8.1-1, the current I just upgrated to (saucy) is 0.8.5-1, and that's what started crashing.  I then switched to the PPA version, same problem.  Should I go back to 0.8.1-1?
120 2013-12-05 06:36:19 <gmaxwell> when you run it from the commandline what message are you getting when it crashes?
121 2013-12-05 06:36:38 <AlexLibman> Sometimes just "Bus error".
122 2013-12-05 06:37:29 <gmaxwell> meh, I'd memtest86x the machine, I've not seen anyone else reporting similar symptoms.
123 2013-12-05 06:37:41 <AlexLibman> And sometimes that other Qt message I pasted above.  At first I've wondered if maybe it's because I'm using such a stripped-down desktop, running i3wm, with most optional stuff uninstalled, but that wouldn't explain the timing of the crashes.
124 2013-12-05 06:37:57 <AlexLibman> Thanks for the tip, I'll do that sometime soon.
125 2013-12-05 06:38:17 <kjj> bus error rarely means there is an actual bus error, but in your case...
126 2013-12-05 06:38:43 <justanotheruser> Is there a complete bitcoin library for c?
127 2013-12-05 06:38:56 <gmaxwell> bus error is usually caused by crazy vm problems like trying to dereference a nonsense pointer, or — outside of x86, unaligned access.
128 2013-12-05 06:39:12 <gmaxwell> justanotheruser: What are you trying to accomplish?
129 2013-12-05 06:40:12 <kjj> I presume that some other error would pop up if you tried to read outside of a memmapped file.
130 2013-12-05 06:40:57 <kjj> and I also sorta presume that bitcoind checks the bounds first, rather than just taking the index's word for it
131 2013-12-05 06:43:12 <gmaxwell> erp. running a salvage wallet on a tn wallet here took the balance from 193k to 150 btc. 0_o
132 2013-12-05 06:43:47 <kjj> well...  was the 193k real, or in error?
133 2013-12-05 06:46:22 <gmaxwell> Real.
134 2013-12-05 06:46:41 <gmaxwell> Amusing that it got a couple though.
135 2013-12-05 06:46:59 <kjj> did it miss a key, or many keys?  was the wallet actually corrupt?
136 2013-12-05 06:47:35 <gmaxwell> many keys, wasn't corrupt as far as I knew, I just wanted to relearn all the txn from the blockchain for testing.
137 2013-12-05 06:48:42 <kjj> odd.  salvage looked like a pretty simple brute-force operation.
138 2013-12-05 06:50:37 <gmaxwell> yea, well, it's reproducable in any case, so ... onto the list it goes.
139 2013-12-05 06:53:17 <imton> guys
140 2013-12-05 06:53:18 <imton> hope you are doing well
141 2013-12-05 06:53:19 <imton> what's the current state of tx replacement?
142 2013-12-05 07:02:22 <gmaxwell> ... and importwallet's rescan didn't work.
143 2013-12-05 07:02:29 <gmaxwell> 2013-12-05 06:55:35 Rescanning last 150263 blocks
144 2013-12-05 07:02:29 <gmaxwell> after the imports it says
145 2013-12-05 07:02:42 <gmaxwell> but then doesn't appear to rescan and shows a balance of 0.
146 2013-12-05 07:03:30 <kjj> doh.  bug in rescan then, and maybe not in salvage?
147 2013-12-05 07:04:56 <gmaxwell> yea, maybe... trying a manaul rescan now on the imported wallet, it seemed to be finding things.
148 2013-12-05 07:05:11 <gmaxwell> (manual rescan on the salvaged wallet did nothing useful)
149 2013-12-05 07:05:22 <kjj> just out of curiosity, on the salvage, how many items did the dbenv.Salvage call find?
150 2013-12-05 07:05:29 <gmaxwell> a kazillion
151 2013-12-05 07:05:33 <gmaxwell> 28,000 or something like that.
152 2013-12-05 07:05:52 <gmaxwell> 2013-12-05 06:14:18 Salvage(aggressive) found 23314 records
153 2013-12-05 07:06:34 <kjj> get any skipping messages?
154 2013-12-05 07:06:56 <gmaxwell> nope.
155 2013-12-05 07:07:15 <gmaxwell> kjj: if you actually want to screw with it, I can give you the wallet.
156 2013-12-05 07:07:36 <gmaxwell> (I'm chasing other things)
157 2013-12-05 07:07:45 <gmaxwell> well, lets see if the import has the same issue.
158 2013-12-05 07:08:16 <kjj> I don't have tools for a wallet dissection, and I'm going to bed soon (like 10 minutes ago soon)
159 2013-12-05 07:08:22 <gmaxwell> I don't think it does— the imports rescan appears to be finding a lot more stuff.
160 2013-12-05 07:08:39 <gmaxwell> yea, well, not a priorty, its on my list of freaky stuff.
161 2013-12-05 07:09:23 <kjj> well, good to know that the keys are ending up in the new wallet
162 2013-12-05 07:09:52 <kjj> having to work around missing transactions is a lot less stressful than working around missing keys
163 2013-12-05 07:10:21 <gmaxwell> well I think the keys were _not_ ending up in the wallet in the salvage.
164 2013-12-05 07:10:55 <gmaxwell> because rescan didn't fix it; but I did an import and though the import's own rescan didn't work a rescan after startup appears to be working
165 2013-12-05 07:11:05 <gmaxwell> (taking forever because its a huge wallet, and I'm running in valgrind)
166 2013-12-05 07:11:24 <kjj> oh, you actually imported the keys.  I thought you were importing a dummy to trigger that rescan
167 2013-12-05 07:12:31 <gmaxwell> yea no, I wanted to distinguish rescanning bugs from salvage bugs.
168 2013-12-05 07:19:31 <kjj> very strange.  by the time it gets around to iterating the old data, any problems should cause it to exit
169 2013-12-05 07:20:56 <JyZyXEL> http://i.imgur.com/YOmDuDV.jpg
170 2013-12-05 07:26:30 <gmaxwell> hurrah, the balance is right now.
171 2013-12-05 08:31:48 <null> I know this isn't really -dev related, but what's your guys opinion on zerocoin?
172 2013-12-05 08:32:17 <null> or is there maybe a discussion on bitcointalk you can point out?
173 2013-12-05 08:32:32 <_ingsoc> What aspect are you talking about?
174 2013-12-05 08:33:36 <null> _ingsoc: well, the whole concept of anonymite. do you think bitcoin will implement anonymite someday?
175 2013-12-05 08:33:48 <null> *anonymity
176 2013-12-05 08:35:45 <_ingsoc> Probably not the way Zerocoin wants to.
177 2013-12-05 08:35:52 <_ingsoc> But who knows? Anything is possible.
178 2013-12-05 08:36:19 <null> yeah, i don't really care about the way :) just the concept
179 2013-12-05 08:37:09 <null> _ingsoc: what's the flaw with the zerocoin way? i just read the authors blog post, not yet the paper (maybe i should come back later :)
180 2013-12-05 08:37:40 <null> they claimed to have cut the zero proof size dramatically... is bloat still an issue?
181 2013-12-05 08:38:05 <_ingsoc> I can't speak on behalf of current Bitcoin devs.
182 2013-12-05 08:38:43 <null> i'm sure all this has been asked before. so if there's some log of this sort of discussion (ML, bitcoitalk, irclogs) i'd be happy to go through it instead of asking all over again
183 2013-12-05 09:33:10 <BlueMattBot> Project Bitcoin build #495: FAILURE in 43 min: http://jenkins.bluematt.me/job/Bitcoin/495/