| 00:25:03 | zooko: | Wow, this is an interesting paper arguing that if you're a miner on the bitcoin network today, you can increase revenue by 2% if you omit all txns. |
| 00:25:27 | justanotheruser: | zooko: what, because you have a 2% higher orphan rate? |
| 00:25:53 | justanotheruser: | also, that sounds like a failing in properly choosing tx based on fee :P |
| 00:25:57 | sipa: | zooko: 2% seems high |
| 00:27:39 | zooko: | justanotheruser: I think that is what it is arguing. |
| 00:28:15 | justanotheruser: | 25btc per mb*2% = 0.0005btc per kb fee needed in addition to make it profitable |
| 00:30:01 | zooko: | https://mail.google.com/mail/u/0/#inbox?compose=14c81166fb498f97 |
| 00:30:03 | zooko: | oops |
| 00:30:08 | zooko: | ftp://ftp.gate.cnrs.fr/RePEc/2014/1412.pdf |
| 00:30:16 | zooko: | HT Marie Vasek |
| 00:30:43 | kanzure: | you expect me to believe that zooko uses gmail? |
| 00:32:14 | justanotheruser: | how else is he supposed to have a human-meaningful identity that is secure? |
| 00:32:21 | phantomcircuit: | zooko, they're just arguing that larger blocks have a higher orphan rate right? |
| 00:48:31 | zooko: | phantomcircuit: I'm not 100% sure what the argument is... |
| 00:50:44 | phantomcircuit: | zooko, yeah it is |
| 00:50:52 | phantomcircuit: | "When pro cessing transactions in a blo ck, a miner increases his reward but alsodecreases his probability to earn any reward b ecause the time needed for his block to reach consensus dep ends on its size. " |
| 00:51:33 | phantomcircuit: | zooko, that's old and boring and there are solutions which roughly fix this |
| 00:51:44 | phantomcircuit: | see https://github.com/TheBlueMatt/RelayNode |
| 01:16:38 | bramc: | Even sending off to the relay network takes some amount of time. Transaction fees being effectively zero is not a good thing. |
| 01:17:47 | instagibbs: | worst case is that fees get bumped to match orphan risk. Not a crisis. |
| 01:18:51 | gmaxwell: | zooko: the numbers in that paper are pure BS. |
| 01:19:15 | gmaxwell: | it's based off the "decker" numbers which are from measurements in 2012 and don't reflect the current state of technology. |
| 01:20:21 | gmaxwell: | bramc: relay network sending a block, if txn are already preforwarded takes sending one or two packets. |
| 01:20:25 | phantomcircuit: | bramc, if you run a relay node locally it doesn't |
| 01:20:42 | bramc: | gmaxwell, Oh right, that does mostly fix it |
| 01:20:50 | gmaxwell: | relay node client. |
| 01:21:20 | phantomcircuit: | right just the client is enough |
| 01:22:34 | gmaxwell: | bramc: in any case, the decker numbers in that paper are prior to signature caching, prior to ultra prune, prior to removing several arbritary 100ms sleeps in the bitcoin networking code, and a dozen other performance improvements. So even without the relaynetwork client, they don't reflect the actual network behavior. |
| 01:24:08 | phantomcircuit: | gmaxwell, 10ms but yeah |
| 01:24:39 | gmaxwell: | phantomcircuit: hm. You sure? |
| 01:24:51 | phantomcircuit: | pretty sure |
| 01:25:37 | gmaxwell: | speaking of that, we still have one related; I think the person working on it (Ashley has abandoned it) |
| 01:26:10 | sipa: | there were 100ms and 10ms ones |
| 01:26:13 | phantomcircuit: | no you're right it was 100 |
| 01:26:13 | sipa: | long ago |
| 01:26:15 | gmaxwell: | we nitpicked it to death, alas (well the patch Ashley made a ton of very difficult to audit locking changes which I think are avoidable) |
| 01:26:18 | sipa: | not sure which are left, but at least one |
| 01:26:21 | phantomcircuit: | https://github.com/bitcoin/bitcoin/commit/97f844dd95c54b0fe2f2a1bb006c74ff544ff125 |
| 01:29:22 | phantomcircuit: | sipa, i think the only way the networking thread sleeps is if it has nothing to do |
| 01:29:52 | phantomcircuit: | so that means average 50ms latency added to processing any incoming message |
| 01:30:55 | gmaxwell: | phantomcircuit: yes, see the pullreq by Ashley. replaces that control with a semaphore but it adds a bunch of prods to make it spin all over the code. |
| 01:32:12 | phantomcircuit: | gmaxwell, you know what the pr # is? |
| 01:33:30 | fanquake: | 4230 |
| 01:34:05 | sipa: | #bitcoin-dev |
| 01:35:28 | gmaxwell: | ^ |
| 02:03:49 | gmaxwell: | andytoshi: so now that rust 1.0 beta is out, have you updated all your rust code yet? |
| 02:05:17 | phantomcircuit: | gmaxwell, lol |
| 05:06:25 | phantomcircuit: | bramc, for reference |
| 05:06:26 | phantomcircuit: | [2015-04-04 05:05:13+00] 0000000000000000083b0cb560b11e15716c62b0ae37ac826166923db4e2fcad recv'd, size 350675 with 1361 bytes on the wire |
| 05:07:09 | phantomcircuit: | that's just 2 ethernet frames |
| 08:05:18 | hitchcock.freenode.net: | topic is: This channel is not about short-term Bitcoin development | http://bitcoin.ninja/ | This channel is logged. | For logs and more information, visit http://bitcoin.ninja |
| 08:05:18 | hitchcock.freenode.net: | Users on #bitcoin-wizards: andy-logbot Mably weex hktud0 damethos RoboTeddy zooko koshii moa Pan0ram1x p15 dgenr8 TheSeven helo gonedrk ryanxcharles sadoshi justanotheruser DougieBot5000 rusty fanquake GAit SwedFTP cluckj PaulCapestany adam3us Hunger- lmacken dc17523be3 yorick nubbins` sqt face antgreen` prodatalab HM Dr-G Luke-Jr luny eordano gribble yrashk Xzibit17 artifexd kumavis adams_ platinuum pollux-bts otoburb bsm117532 c0rw1n a5m0 vonzipper catlasshrugged |
| 08:05:18 | hitchcock.freenode.net: | Users on #bitcoin-wizards: jgarzik spinza Madars huseby midnightmagic BlueMatt throughnothing kefkius GreenIsMyPepper dasource warren TD-Linux c-cex-yuriy mariorz binaryatrocity devrando1 hguux___ fenn wizkid057 jaekwon arubi Starduster rustyn ebfull ajweiss LeMiner waxwing airbreather_1 maaku thrasher` Tiraspol SubCreative starsoccer Adlai jaromil Anduck melvster iddo aakselrod MoALTz_ alferz bosma sneak deepcore poggy amincd PRab go1111111 bliljerk101 amiller raizor |
| 08:05:18 | hitchcock.freenode.net: | Users on #bitcoin-wizards: Chillum wiz nsh luigi1111w JonTitor szumen kyuupichan Logicwax d9b4bef9 tromp runeks phantomcircuit EasyAt hashtag_ lechuga_ luigi11111 isis nanotube yoleaux gmaxwell andytoshi berndj gavinandresen dignork AdrianG s1w livegnik optimator fluffypony Meeh cursive dansmith_btc morcos guruvan BananaLotus bedeho heath roasbeef_ Fistful_of_Coins dardasaba jonasschnelli comboy_ tromp_ stonecoldpat nuke_ afdudley espes__ pigeons eric jessepollak sipa |
| 08:05:18 | hitchcock.freenode.net: | Users on #bitcoin-wizards: warptangent ryan-c mkarrer_ phedny so nuke1989 STRML harrow` forrestv michagogo null sl01 larraboj lnovy Iriez Apocalyptic jcorgan [d__d] petertodd kanzure catcow Muis cfields Zouppen coryfields_ cryptowest_ kinlo wumpus gwillen nickler Alanius sdaftuar null_radix epscy Taek Krellan Oizopower Keefe mappum jbenet gnusha Graet Eliel veox indolering K1773R leakypat CryptOprah BrainOverfl0w MRL-Relay azariah btc___ @ChanServ brand0 davout |
| 08:05:18 | hitchcock.freenode.net: | Users on #bitcoin-wizards: NeatBasis mr_burdell |
| 08:25:27 | fanquake: | fanquake has left #bitcoin-wizards |
| 11:55:43 | airbreather_1: | airbreather_1 is now known as airbreather |
| 13:08:54 | sadoshi: | sadoshi has left #bitcoin-wizards |
| 15:52:19 | kanzure: | "A distributed system is one in which the failure of a computer you didn't even know existed can render your own computer unusable." |
| 17:00:50 | andytoshi: | gmaxwell: been working on it over the last couple weeks, it's been tough because i need all my dependencies to be compilable at the same time, if any one breaks then i'm stalled :) |
| 17:01:15 | andytoshi: | and the web stuff that rust-jsonrpc depends on has totally changed, i'm gonna have to rewrite that lib i think before i can get wizards-wallet to work |
| 17:01:41 | andytoshi: | but rust-bitcoin is in pretty good shape and bitcoin-secp256k1-rs is totally usable (except i should rename it :P) |
| 17:02:17 | zooko: | Sweet. ☺ |
| 19:02:24 | belcher_: | belcher_ is now known as belcher |
| 20:34:17 | buster: | buster is now known as Guest66176 |
| 20:35:11 | Guest66176: | Guest66176 has left #bitcoin-wizards |