| 00:00:06 | phantomcircuit: | there's little reason for servers not to generate their own dh prime in the background when they're started though |
| 00:00:14 | gmaxwell: | nsh: fortunately it seems that there doesn't have to be a _use_ for number theory for people to go on thinking it up. :) |
| 00:00:23 | nsh: | * nsh smiles |
| 00:00:25 | nsh: | quite |
| 00:01:20 | nsh: | recalls a discussion i was having last night based on this reporting: https://www.quantamagazine.org/20150519-will-computers-redefine-the-roots-of-math/ |
| 00:01:23 | gmaxwell: | phantomcircuit: well the reason would be whatever unknown cause there was for non-safe DH numbers on thousands of hosts. :) |
| 00:02:28 | nsh: | (precis: widespread use of Coq might be able to achieve part of the early 20th century overambitious project of systematizing mathematics mechanically) |
| 00:02:46 | nsh: | but with less hubris this time around, hopefully |
| 00:02:47 | phantomcircuit: | gmaxwell, true |
| 00:03:37 | nsh: | Coq is a deduction framework, and as Poincare succinctly said: "it is by deduction that we prove, but it is by intuition that we discover" |
| 00:04:17 | moa: | nsh: http://people.brandeis.edu/~cwe/pdfs/primes_and_riemann.pdf might be good place to start re: riemann => primes |
| 00:04:27 | nsh: | ty moa |
| 00:05:27 | nsh: | there's also: http://empslocal.ex.ac.uk/people/staff/mrwatkin/zeta/riemannhyp.htm |
| 00:05:29 | gmaxwell: | nsh: deductive systems can help greatly in discovery too. Because you can take a guess and ask the system if it holds-- given all you've taught it so far, and get a yes/no/maybe right away... and this can greatly speed up your work. |
| 00:05:39 | nsh: | right |
| 00:06:16 | moa: | zeoes of zeta and distribution of primes is just so weird there is something deep in there ... i think it is what drives most of the interest in riemann |
| 00:06:22 | nsh: | ideally, we want mathematics to be almost entirely creative and intuitive, leaving the drudgery to the machines of loving grace |
| 00:06:38 | moa: | zeroes of zeta |
| 00:06:44 | nsh: | * nsh nods |
| 00:07:42 | nsh: | oh that was it, hyperelliptic curve cryptography has a relation to riemann / zeta function |
| 00:08:31 | nsh: | http://en.wikipedia.org/wiki/Hyperelliptic_curve_cryptography#Order_of_the_Jacobian |
| 00:08:35 | moa: | i've found some clues but nothing definitive |
| 00:09:00 | moa: | a link between bernoulli and euler numbers e.g. |
| 00:11:34 | moa: | a closed form for apery's and a general solution for zeta(2n+1) family would be a good start |
| 00:12:01 | nsh: | * nsh nods |
| 00:13:52 | moa: | but wait there's more! ;) |
| 00:42:48 | frankenmint: | how big can a transaction hash be in digits? |
| 00:45:55 | nsh: | binary or binascii? |
| 00:46:08 | frankenmint: | uh i think binascii |
| 00:46:26 | nsh: | -- |
| 00:46:26 | nsh: | A sha256 is 256 bits long -- as its name indicates. If you are using an hexadecimal representation, each digit codes for 4 bits ; so you need 64 digits to represent 256 bits -- so, you need a varchar(64) , or a char(64) , as the length is always the same, not varying at all.10 Feb 2010 |
| 00:46:32 | frankenmint: | not compessed down |
| 00:46:35 | nsh: | -- google inlines stackoverflow answers now... |
| 00:46:43 | nsh: | (in search results, heh) |
| 00:47:06 | nsh: | more details: http://bitcoin.stackexchange.com/questions/2859/how-are-transaction-hashes-calculated |
| 00:49:53 | frankenmint: | great answer to read |
| 00:50:06 | frankenmint: | dumb related question anyone know why its using little endian? |
| 00:50:16 | frankenmint: | err you answered I mean, do you know |
| 00:50:54 | nsh: | speculations of this sort are inevitably apocryphal |
| 00:51:11 | nsh: | either satoshi had different ideas about endianness at different times when he was designing bitcoin |
| 00:51:26 | nsh: | or the satoshis didn't quite form a coherent consensus on the matter |
| 00:51:37 | nsh: | * nsh is happy not knowing |
| 00:53:45 | frankenmint: | my path towards compsci enlightenment improves just a little bit more today, thank you nsh |
| 00:53:47 | frankenmint: | http://en.wikipedia.org/wiki/Bitwise_operation |
| 00:54:03 | nsh: | * nsh smiles |
| 00:54:36 | moa: | big, little what's the difference it's still endian |
| 00:54:58 | frankenmint: | yea I know what they are and have known for a long time |
| 00:55:18 | frankenmint: | but I'm not grasping why one is used over another or what point it came up that two equally systems work |
| 00:55:30 | frankenmint: | like working with + or - but its just using absolute value |
| 00:55:33 | nsh: | oh, there are good pages on this |
| 00:55:47 | moa: | bi-endian |
| 00:56:06 | moa: | interoperability is always good |
| 00:56:25 | nsh: | the short answer is: legacy. the longer answer has to do with architectural quirks of processors and network interfaces |
| 00:56:32 | moa: | keeps them on their toes |
| 00:57:08 | nsh: | the answers here are enlightening: http://programmers.stackexchange.com/questions/95556/what-is-the-advantage-of-little-endian-format |
| 00:57:19 | frankenmint: | * frankenmint read that as he reads "On simple low-cost processors, typically, bitwise operations are substantially faster than division, several times faster than multiplication, and sometimes significantly faster than addition. While modern processors usually perform addition and multiplication just as fast as bitwise operations due to their longer instruction pipelines and other architectural design choices, |
| 00:57:19 | frankenmint: | * frankenmint bitwise operations do commonly use less power because of the reduced use of resources." |
| 01:09:09 | frankenmint: | frankenmint has left #bitcoin-wizards |
| 01:18:52 | dgenr8: | middle-out-endian has the best Weissman score |
| 01:19:16 | nsh: | .wik Weissman score |
| 01:19:17 | yoleaux: | "Silicon Valley is an American television sitcom created by Mike Judge, John Altschuler and Dave Krinsky. The series focuses on six young men who found a startup company in Silicon Valley. The series premiered on April 6, 2014, on HBO. The first season consisted of eight episodes." — http://en.wikipedia.org/wiki/Silicon_Valley_(TV_series) |
| 01:20:10 | nsh: | dgenr8, gobbledegook metric? |
| 01:36:52 | phantomcircuit: | nsh, still generating the 8192 bit primes |
| 01:41:19 | nsh: | hehe |
| 01:41:28 | nsh: | phantomcircuit, using openssl, or? |
| 01:43:13 | phantomcircuit: | yeah |
| 01:44:35 | phantomcircuit: | 105:31.00 |
| 01:44:38 | phantomcircuit: | dat cpu time |
| 01:57:36 | moa: | phantomcircuit: primecoin` |
| 02:13:32 | phantomcircuit: | moa, DH safe prime generator |
| 02:14:36 | moa: | sounds like a new mining algo for yet another alt |
| 02:15:36 | moa: | suitably half-baked too |
| 02:18:32 | phantomcircuit: | moa, openssl gendh 8192 |
| 02:21:50 | moa: | Generating DH parameters, 8192 bit long safe prime, generator 2 |
| 02:21:51 | moa: | This is going to take a long time |
| 02:32:51 | nsh: | * nsh wonders why, if all generators are provably equivalent, there are two conventional ones |
| 02:32:53 | nsh: | (2, and 5) |
| 02:32:59 | nsh: | and how they were selected |
| 03:20:54 | phantomcircuit: | nsh, still running |
| 06:55:00 | phantomcircuit: | nsh, real 396m55.804s |
| 06:55:22 | phantomcircuit: | http://0bin.net/paste/Vk0K0uN31ZA43QCd#wSKKTU2bWLkcJIRaKysjkBqANNe58ssc5zzQ782wf2Q |
| 08:05:18 | holmes.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 | holmes.freenode.net: | Users on #bitcoin-wizards: andy-logbot NewLiberty_ shesek b_lumenkraft Relos dEBRUYNE d1ggy CoinMuncher gill3s mountaingoat jonasschnelli KuDeTa hktud0 antanst CryptoGoon wizkid057 mappum Luke-Jr hulkhogan_ berndj go1111111 leakypat_ ttttemp priidu p15 p15x Mably gavmatic pollux-bts jtimon felipelalli TheSeven weex roconnor dc17523be3 bosma Dr-G jrayhawk akrmn Emcy platinuum metamarc Oizopower jmcn_ sparetire_ jbenet HM Tiraspol dasource Starduster_ hashtag_ Giszmo |
| 08:05:18 | holmes.freenode.net: | Users on #bitcoin-wizards: arubi_ Quanttek nuke1989 mkarrer_ LeMiner waxwing btcdrak tucenaber kyuupichan helo superobserver bassguitarman Taek ebfull hashtagg Madars rustyn gmaxwell PaulCape_ lnovy crowleyman bedeho cdecker iddo OneFixt justanotheruser alferz mengine GreenIsMyPepper amiller prodatalab_ sneak epscy adams__ wiz michagogo mikolalysenko artifexd grandmaster dansmith_btc SubCreative airbreather lmacken Pan0ram1x dgenr8 c0rw1n Cory larraboj brand0 nickler |
| 08:05:18 | holmes.freenode.net: | Users on #bitcoin-wizards: PRab yrashk vonzipper cfields nsh Krellan_ stevenroose coryfields Meeh catlasshrugged cryptowest_ Alanius null_radix kanzure gavinand1esen Logicwax copumpkin bliljerk_ melvster azariah harrow tromp_ isis scoria andytoshi warptangent harrigan sparetire davout comboy TD-Linux yorick crescend1 veox mm_1 theymos Zouppen huseby _whitelogger wumpus binaryatrocity heath BananaLotus maaku face_ Apocalyptic [d__d] optimator Eliel gnusha tromp narwh4l |
| 08:05:18 | holmes.freenode.net: | Users on #bitcoin-wizards: lmatteis koshii mr_burdell throughnothing_ elastoma fluffypony Fistful_of_Coins yoleaux Jaamg se3000 xabbix mariorz catcow a5m0_ smooth dignork runeks Sqt poggy livegnik K1773R petertodd richardus nephyrin phedny so phantomcircuit afdudley pigeons SwedFTP guruvan ajweiss nanotube forrestv Muis warren Xzibit17 sdaftuar eric roasbeef BrainOverfl0w @ChanServ gwillen kinlo sl01 STRML espes__ AdrianG luigi1111 Anduck BlueMatt midnightmagic otoburb |
| 08:05:18 | holmes.freenode.net: | Users on #bitcoin-wizards: kumavis starsoccer d9b4bef9 gribble jessepollak ryan-c Keefe indolering Graet jaromil sturles [ace] merlincorey Iriez EasyAt morcos CryptOprah s1w |
| 08:05:18 | holmes.freenode.net: | [freenode-info] channel trolls and no channel staff around to help? please check with freenode support: http://freenode.net/faq.shtml#gettinghelp |
| 08:13:28 | stonecoldpat: | CoinMuncher: from what you said yesterday, IIRC there is some work here http://www.jbonneau.com/doc/BFGKN14-bitcoin_bribery.pdf that is a varaint of what you spoke about - so you said that you want to convince a miner by putting your own money on the line that a header is valid - instead this paper tries to "bribe" other miners to build on a block by sending them bitcoins in a block, |
| 08:13:28 | stonecoldpat: | so no its not "bullocks" what your thinking :P |
| 08:41:25 | CryptoGoon: | CryptoGoon is now known as TrollGoon |
| 08:41:29 | TrollGoon: | TrollGoon is now known as CryptoGoon |
| 08:46:14 | CoinMuncher: | yoohoo! :) |
| 08:46:48 | CoinMuncher: | stonecoldpat: thanks dude. Was just trying to read back the logs. |
| 08:48:41 | CoinMuncher: | I was just trying to come up with weird things you can do with transactions/mining and then think of an application for which it might be useful. Logically it's sort of the wrong way around, but I guess one might stumble upon something. On the other hand I'm sure thousands have gone before me. |
| 09:03:38 | Dr-G: | Dr-G is now known as Dr-G2 |
| 09:14:57 | PRab_: | PRab_ is now known as PRab |
| 13:15:21 | Dr-G2: | Dr-G2 is now known as Dr-G |
| 15:04:28 | KuDeTa_: | KuDeTa_ is now known as KuDeTa |
| 16:03:25 | lnovy: | lnovy is now known as zz_lnovy |
| 16:08:00 | zz_lnovy: | zz_lnovy is now known as lnovy |
| 16:42:58 | NewLiberty_: | NewLiberty_ is now known as NewLiberty |
| 17:28:11 | dgenr8: | phantomcircuit: 4096 for me is 27m real. what's your processor? |
| 17:30:16 | dgenr8: | this is Intel i7 4770 @3.4GHz |
| 18:45:04 | CryptoGoon: | CryptoGoon is now known as DogeBuffet3 |
| 18:45:52 | DogeBuffet3: | DogeBuffet3 is now known as CryptoGoon |
| 18:58:44 | NewLiberty_: | NewLiberty_ is now known as NewLiberty |
| 19:25:07 | wallet421: | wallet421 is now known as wallet42 |
| 21:47:53 | KuDeTa_: | KuDeTa_ is now known as KuDeTa |
| 21:52:52 | OneFixt_: | OneFixt_ is now known as OneFixt |
| 22:11:26 | wallet421: | wallet421 is now known as wallet42 |
| 22:38:40 | phantomcircuit: | dgenr8, i7-4700MQ |
| 22:38:41 | phantomcircuit: | the secret is to be running havege so you dont get stalls waiting for entropy |
| 22:38:48 | phantomcircuit: | note: that might mean the parameters aren't entirely random... but it doesn't matter! |
| 22:38:54 | phantomcircuit: | im sure it would be faster on my 4790K |
| 22:46:29 | c0rw1n: | c0rw1n is now known as c0rw|zZz |
| 23:05:18 | nsh: | phantomcircuit, is havege using any hardware sources? |
| 23:05:49 | nsh: | not sure i understand what you mean by some parameters not being entirely random |
| 23:06:39 | PRab_: | PRab_ is now known as PRab |
| 23:09:18 | gmaxwell: | no one but intel knows. |
| 23:09:59 | gmaxwell: | http://www.chronox.de/jent/doc/CPU-Jitter-NPTRNG.html has a writeup related to that approach (though not specific to the havege code) |
| 23:11:34 | gmaxwell: | basically there is some argument that there exists physical noise sources in cpu timing, mostly from the interactions of the different clock domains between the cpu and memory (PLL jitter). There is also a whole lot of 'seeming randomness' that is probably just properties of the microarchitecture and are not actually random. |
| 23:12:00 | nsh: | i recall dankami harking about this a fair bit last year |
| 23:12:41 | nsh: | i paraphrased some poetry for him: all these random bits we pick, are but the tocks within the tick |
| 23:13:58 | gmaxwell: | Its tricky. If your CPU and memory are driven off a straight up divider circut on a common clock then there may be basically zero entropy there; but all the bogus 'randomness' means that these measures will still look random, so you can't tell when they're useless. |
| 23:15:10 | gmaxwell: | beating two oscillators against each other is a reasonable hardware source of entropy, ... except for the fact that most oscillators are highly vulnerable to injection locking; which can cause it to fail to produce much entropy. |
| 23:16:00 | gmaxwell: | (there is some nice demos of defeating smartcard RNGs by just using a radio transmitter (e.g. a microwave oven magnatron) to injection lock their ring-oscillator based RNGs and make them put out a regular pattern.) |
| 23:20:05 | moa: | resonance ruined randomness |
| 23:22:25 | dgenr8: | * dgenr8 had no idea havege existed. very neat |
| 23:32:07 | nsh: | .wik Injection locking |
| 23:32:08 | yoleaux: | "Injection locking and injection pulling are the frequency effects that can occur when a harmonic oscillator is disturbed by a second oscillator operating at a nearby frequency." — http://en.wikipedia.org/wiki/Injection_locking |
| 23:32:12 | nsh: | neat |
| 23:32:40 | nsh: | moa, "resonance ruined randomness" is a good album title for post-modernist cyperpunk ambient music or something |
| 23:32:54 | nsh: | i assume that's a genre; everything's a genre |
| 23:35:17 | moa: | ordo ad chaos |
| 23:35:52 | moa: | ordo ab chaos |
| 23:56:20 | nsh: | .t https://www.reddit.com/r/Bitcoin/comments/32tcym/i_see_the_future_laseretching_the_blockchain_on/ |
| 23:56:20 | yoleaux: | nsh: Sorry, I don't know a timezone by that name. |
| 23:56:25 | nsh: | .title https://www.reddit.com/r/Bitcoin/comments/32tcym/i_see_the_future_laseretching_the_blockchain_on/ |
| 23:56:26 | yoleaux: | I see the future. Laser-etching the blockchain on the surface of the moon. It would be visible to everyone with a telescope. No internet needed. : Bitcoin |
| 23:56:51 | nsh: | -- |
| 23:56:52 | nsh: | <xentrac> So it turns out that the surface of the moon has space for several hundred petabytes of information encoded at a one-millimeter resolution that could be read from practically-sized terrestrial instruments. |
| 23:56:52 | nsh: | <xentrac> And laser-printing on the moon should require less than a watt of power. |
| 23:56:52 | nsh: | <xentrac> From Earth. |
| 23:57:28 | nsh: | -- i am skeptical re the spacial resolution maths, and how you'd actually cause a visible reaction without any oxidising agents using laser power |
| 23:57:30 | moa: | send buzz up there to play golf on it |
| 23:57:35 | nsh: | * nsh smiles |
| 23:57:39 | moa: | mess up your balance |
| 23:58:28 | Taek: | I was never comfortable with moon math |
| 23:58:31 | Taek: | * Taek ducks |
| 23:58:41 | moa: | a distance time source for timestamping is a thought but you need and incorruptible mechanism for everybody to interpret it ... |
| 23:59:06 | nsh: | .wa distance to moon from earth |
| 23:59:12 | yoleaux: | Moon: distance from Earth: Current result: 390139 km (kilometers); Unit conversions: 242421 miles; 3.901×10⁸ meters; Comparison as distance: ~1.01 × mean Moon-Earth distance (3.85×10⁸ m); Corresponding quantities: Light travel time t in vacuum from t = x/c:: 1.3 seconds; Light travel time t in an optical fiber t = 1.48x/c:: 1.9 seconds |
| 23:59:37 | moa: | what's that in AUs? |