00:00:46 grr 00:00:54 bsmntbombdood: wut 00:02:28 hmm 00:02:34 is there a bsd with binary pkgs? 00:03:33 all of them 00:03:42 really? Even Open? 00:03:49 do you still have to use ports, though? 00:03:55 I suppose there may be frontends 00:04:02 http://www.openbsd.org/cgi-bin/man.cgi?query=pkg_add&apropos=0&sektion=0&manpath=OpenBSD%20Current&arch=i386&format=html 00:04:14 hmm. 00:04:19 Does bsd have anything similar to lvm? 00:04:20 ewwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww 00:04:27 Gracenotes: wha 00:04:28 t 00:04:38 yes, it was total skill that my eww was exactly as long as your comment 00:04:43 w at 00:06:12 nothing. btw mah raytracer. http://67.223.225.106/tasd/ 00:06:29 On the twelfth day, the LORD created cheap VM creation. And it was good. <<< No, that was on the zeroth day. You don't seriously think he would run us without sandboxing? 00:07:57 there has been discovered a new and wonderful form of poetry! :o 00:08:10 its esoteric in its delightfulness :x 00:08:13 wat 00:08:43 what, you dont like my inelegant use of the there+passive? dont give me your grammar nazism! 00:09:04 apostrophe. DISQUALIFIED 00:09:06 What's the poetry. 00:09:11 no, he meant angkor wat. that's where it was discovered, right? 00:09:20 exactly 00:09:38 the discoverers were eating soup from that very area when the reporters arrived 00:09:40 they were like 00:09:42 ... 00:09:45 anyway 00:10:05 take a text in language A. spell check it as if it's language B. then translate the text from language B into language A. 00:10:36 Thhhe, were eating poetry 00:10:36 drinking tea from that 00:10:37 discover 00:10:39 e 00:10:41 r 00:10:43 s 00:10:45 area where it was discover- 00:10:47 -ed 00:10:49 00:10:51 AVANT GARDE 00:10:58 c-c-c-c-c-c-poetry breaker 00:11:01 oh very, ehird 00:11:04 very avant garde :p 00:11:06 :o 00:11:09 Together lord. 00:11:09 The oil problem is, lead away I switch languages in Misheard 2000 (wedge mouthguard does automatically hen I switch keyboard lay-outs), mouthguard bugs me bolt the facet that dinner table slow is note supporter Bi the game crazier, Belause {name} mouthguard haircut bone install. And let me count ouch, that is far point of attachment. 00:11:28 psygnisfive: that's far too coherent for that process surely 00:11:34 well 00:11:44 it was English-as-Dutch -> English 00:11:54 ah 00:11:56 so similarish words 00:11:57 so there were preexisting similarities 00:12:35 I KILL THE 00:12:35 MAYBE SHOULD JAPANESE 00:12:36 you try 00:12:38 (view monospaced) 00:13:16 dont be racist ehird 00:13:27 you left out koreans and other slanteyes 00:13:27 psygnisfive: hey that's just one strand of reading! 00:13:29 try other swoops 00:13:39 MAYBE you SHOULD KILL THE JAPANESE 00:13:42 MAYBE I SHOULD try JAPANESE 00:13:42 Oh, I love having a monospaced IRC client 00:13:48 see? 00:13:50 but :\ reading it horizontally doesn't make sense 00:13:54 MAYBE you SHOULd try JAPANESE 00:13:55 it should have a double meaning 00:14:01 Gracenotes: yeah I tried that at first 00:14:18 We're gonna have to slap the dirty little Jap 00:14:37 hey I used to be a 00:14:37 >>>>>>>>>>>>>>>>>SPIRAL< 00:14:38 then I 00:14:46 i'm all about meaningless furor as art. 00:15:27 spiral, you say? http://www.onemanga.com/Uzumaki/ 00:15:27 meaningless führer 00:15:34 (start at chapter 1) 00:15:59 mean english führer 00:16:21 hahaha 00:16:31 the same guy who made Enigma of Amigara Fault 00:19:55 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 00:20:34 -!- bsmntbombdood has joined. 00:25:01 has anyone ever tried to read finnegans wake? 00:26:25 first few pages 00:27:10 its less comprehensible than the dutch-english thing from above 00:27:28 :-D 00:27:34 riverrun, past Eve and Adam's 00:27:57 bababadalgharaghtakamminarronnkonnbronntonnerronntuonnthunntrovarrhounawnskawntoohoohoordenenthurnuk! 00:28:06 three quarks for muster mark 00:28:51 heaven hell murray gell-man 00:28:59 who i would guess has read the thing full through 00:30:13 Fwomp. 00:30:46 Brékkek Kékkek Kékkek Kékkek! Kóax Kóax Kóax! Ualu Ualu Ualu! Quaouauh! 00:32:50 I've flipped through it 00:38:35 Finnegan's Wake wiki, by the way: http://www.finnegansweb.com/wiki/index.php/Page_3 00:39:02 annotated with wiki tools 00:39:12 Hah 00:39:25 Analyzing it destroys it. 00:40:15 ehird: beat you by two and a half years on that point :P http://slashdot.org/comments.pl?sid=202042&cid=16540372 00:40:57 wow... it's been a long time, internet. 00:41:09 2006 ain't long ago 00:41:21 feels like it 00:41:28 true. I was an idiot in 2006. 00:41:53 started netting in '98. so... about 9 years to become unstupid 00:46:27 im listening to a reading of finnegans wake 00:46:29 grrrrr 00:46:33 which is read with an irish accent 00:46:35 rrrrrrrrrrrrrrrrrrrrrrg 00:46:55 and im almost willing to believe that if this was read with a very thick accent, itd actually be readable 00:47:22 Idea: Finnegans Wake... sung! 00:48:11 *70s sort of rock, think Led Zepplin* Riverrun, past Eve and Adam's yeah/From swerve of shore to bend/Of bay, brings us by/A commodius vicus/Of recirulation back to Howth Castle/aaand Enviro-ons 00:48:13 *solo* 00:48:20 maybe joyce was just illiterate and couldnt spell :o 00:51:06 finnegans wake is very reminiscent of the output of a markov chain model of english sentences 00:52:02 we could construct some sort of grammatical model of english and have it generate random sentences that are wake-esque :o 00:52:47 maybe a CxG model 00:54:18 -!- FireFly has quit ("Later"). 01:02:34 http://toastytech.com/guis/winvistawin101.png 01:03:01 the fuck 01:03:15 psygnisfive: Windows 1.01 Vista! 01:03:23 *1.0 01:03:32 er no 1.01 01:07:02 -!- puzzlet_ has quit (Remote closed the connection). 01:07:06 -!- puzzlet has joined. 01:07:21 -!- coppro has joined. 01:07:47 file:///home/gregor/codu/public_html/imgs/win3plusplus.png 01:07:53 GregorR: FAIL 01:07:54 -!- puzzlet has quit (Read error: 104 (Connection reset by peer)). 01:08:20 D-8 01:09:41 -!- Sgeo has joined. 01:09:50 Yeah, so that was stupid. 01:10:03 * ehird considers if you could use compiz-fusion + dmenu for a minimalist environment 01:10:16 GregorR, what wsa stupid? 01:10:19 http://codu.org/imgs/win3plusplus.png // what I meant to link. 01:10:23 Sgeo: file:///home/gregor/codu/public_html/imgs/win3plusplus.png 01:10:24 01:07 GregorR: file:///home/gregor/codu/public_html/imgs/win3plusplus.png 01:10:49 ehehe 01:11:00 Anyway, go look at http://codu.org/imgs/win3plusplus.png :P 01:11:04 aw. clicky link clicks nowhere 01:11:05 That was relevant to http://toastytech.com/guis/winvistawin101.png 01:11:11 toasty tech has shit like that too, yeah 01:11:17 /var/www/ ftw 01:11:33 Gracenotes: This is just my hg clone 01:12:07 -!- puzzlet has joined. 01:12:11 There's a Moka5 thingy for Win 3.1 01:12:12 Err, that is, that file:// path is just an hg clone 01:12:16 Don't know how legal it is 01:12:23 "Moka5"? 01:12:35 MokaFive LivePC Engine makes use of VMware Player as a virtualization platform, but includes additional features such as automatic updates of virtual machines, streaming and caching of virtual machine images, integrated backup, hardware-accelerated 3D graphics support, and zero install when running from a USB drive.[1] 01:12:36 lame 01:13:19 The thing that people are supposed to see as weird in my screenshot is that it's Firefox and bash running on Windows 3 :P 01:13:37 Yes. 01:13:41 It's not too interesting :P 01:13:43 Been there done that 01:13:43 And being in NT351 01:13:44 Ok, see nthat. 01:13:45 *seen that 01:13:56 Bleh, you people are too smart :P 01:17:42 -!- puzzlet has quit (Remote closed the connection). 01:17:46 -!- puzzlet has joined. 01:18:16 eh, it seems likely that Arch is the distro for m 01:18:17 e 01:18:30 ORLY? I've never tried Arch. 01:18:38 Summarize its advantages in one IRC line :P 01:19:13 (Using Debian as a base for comparison if necessary) 01:19:37 GregorR: it doesn't mod upstreams much, large amount of user-contributed packages, friendly & big community, it's minimalist, and yet you can still install big glob suites with one command, the package manager seems alright, and it has 64-bit binaries and all that 01:20:04 compared with debian: mods upstreams less, seems to be generally more oriented to fixing things at upstream instead of patching downstream 01:20:22 and is generally more minimal 01:20:43 Yeah, Debian makes sometimes-significant modifications to upstream, meaning that sometimes the Debian version is actually quite a bit different than upstream. 01:20:54 yes 01:21:01 I don't mind value-add stuff, I just don't want it by default 01:21:02 Example: netpbm sucks on Debian, I always install it manually :P 01:21:12 Oh, my example is a value-subtract. 01:21:17 Yeah 01:21:24 Upstream's for me, and arch is similar enough to debian that I can probably just use it as a lean, mean Linux universe machine 01:21:31 Okidoke. 01:21:48 try it in a vm if you want 01:22:10 GregorR: oh, and rolling release, which I don't actually care about 01:22:18 rolling release works better when you don't mod upstream ofc 01:22:28 Right 01:22:44 Out of curiosity, does Arch work on any non-x86alike platforms? 01:22:57 GregorR: i686 and x86_64 only, it seems. 01:23:20 GregorR: The only things I'd consider running a non-x86 on is embedded stuff, though. 01:23:29 And for embedded stuff I wouldn't use a full linux distro :-) 01:23:33 Sure, just curious. 01:23:37 Angstrom = Awestrom 01:25:20 The Chrome 3D button disappeared 01:25:21 GregorR: Oh, and, Arch uses a BSD-style init. 01:25:21 ! 01:25:25 NOOOOOOOOOOOOOO!!!!!!!!!!!!!!!!!!! 01:25:30 GregorR: Runlevels, you understand. And whatnot. 01:25:38 /etc/rc.conf pointing to shit in /etc/rc.d/. 01:25:43 Hmm, that's interesting. 01:25:44 It's simple. I like that. init.d is confusing. 01:25:55 http://wiki.archlinux.org/index.php/FAQ#Q.29_What_exactly_is_this_.27BSD-style.27_init_framework_I_keep_hearing_about.3F 01:27:16 Oh, and Arch's community seems to generally prefer throwing text around instead of big anti-unix environments. 01:27:21 I can work with either, but minimalism is nice. 01:27:29 Even if I _do_ need something to fill up that 12GB ram. 01:27:50 i has a 12 gb of ram 01:28:02 bsmntbombdood: mine's faster ;-P 01:28:19 bsmntbombdood: competing against my own rig is hard, don't make me :P 01:30:31 i'm not sure what to put on this new box 01:30:45 bsmntbombdood: Arch? :-P 01:30:49 Or a BSD. 01:30:58 bsmntbombdood: Well, a linux helps, for the LVM alignment stuff. 01:31:01 I don't know how to do that on BSD. 01:32:45 !slashes /.\0/,\,0,\,1//.\1/,\,1,\,0//,\,/.//.//.0.1.1.0 01:32:45 bsmntbombdood: I'm not sure where you'd put the mechanical HD, though. /home would put dotfiles on it defeating the point. ~/stuff as the mechanical HD is a bit inelegant. 01:32:45 01101001 01:33:33 !slashes /.\0/,\,0,\,1//.\1/,\,1,\,0//,\,/./.0.1.1.0 01:33:34 .0.1.1.0.1.0.0.1 01:33:36 ehird: ~/media, with symlinks at ~/music, ~/porn, ~/movies 01:33:51 -!- puzzlet has quit (Remote closed the connection). 01:33:55 -!- puzzlet has joined. 01:34:04 bsmntbombdood: what about source code? it's not media, but do you want to hammer the SSD with those writes? I guess it's just me being paranoid :-P 01:34:16 But yeah, I'm generally suspicious of symlinks, it may be the best solution in that case. 01:34:20 *while I'm 01:34:47 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,1,\\,0\/\/,\\,\/.\//***/.//.0 01:34:47 01101001 01:35:03 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,1,\\,0\/\/,\\,\/.\//*******/.//.0 01:35:03 01101001100101101001011001101001100101100110100101101001100101101001011001101001011010011001011001101001100101101001011001101001 01:35:19 Anyway. Bye. bsmntbombdood: if you want to ask me something or whatever, do it in /msg so it sticks out tomorrow :P 01:35:45 you mean you don't have a script to show you highlights in a different window? 01:43:34 You need a script? 01:50:28 yes 01:52:27 * coppro boggles 01:54:17 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\//***/.//.0 01:54:18 01001 01:54:23 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\//****/.//.0 01:54:23 01001010 01:59:58 !slashes /*/\/+\\+\/=\\=.\\0-\/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\//****/.//++.0.1 01:59:58 ==01001010-0100101001001 02:01:19 What's the quickest way from 0 to Sha1 sum in C? 02:01:49 !slashes /*/\/+\\+\/=\\=.\\0-\/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/=\\=\/+\\+\//****/.///-/\\\\///0/1//1/\*/++.0.1 02:01:49 1/++01,,==0101001001010010100100101001001 02:01:57 erp 02:02:50 -!- puzzlet has quit (Read error: 60 (Operation timed out)). 02:02:58 -!- puzzlet has joined. 02:03:08 !slashes /*/\/+\\+\/=\\=.\\0-\/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/=\\=\/+\\+\//****/.//++.0.1 02:03:09 ++01-010-01001-01001010-0100101001001 02:04:44 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\0-\/\/=\\=\/+\\+\//****/.//++ 02:04:45 ++0-01-010-01001- 02:04:57 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//****/.//++ 02:04:57 ++1-0-01-010- 02:06:05 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//****/.///+\+///-/\//++ 02:06:18 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//****/.///+\+///-/\/\//++ 02:06:19 10010 02:06:32 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//****/.///+\+///-/=/++ 02:06:33 1=0=01=010= 02:06:57 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//****/.///+\+///-/=/++.1 02:06:58 1=0=01=010=01001 02:07:04 What are you trying to do? 02:07:08 fibonacci 02:07:28 O 02:08:16 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//****/.///+\+///-/\\\\/++.1 02:08:16 100101001001 02:08:30 !slashes /*/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//****/.///+\+///-/\\\//++.1 02:08:30 1/0/01/010/01001 02:08:34 ah 02:09:25 !slashes /!/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//!!!!/.///+\+///-/\\\///0/1//1/*/++.1 02:09:26 */*/**/***/***** 02:09:51 !slashes /!/\/.\\0\/,\\,0,\\,1\/\/.\\1\/,\\,0\/\/,\\,\/.\/\/+\\+\/=\\=.\\1-\/\/=\\=\/+\\+\//!!!!!!!!!!/.///+\+///-/\\\///0/1//1/*/++.1 02:09:51 */*/**/***/*****/********/*************/*********************/**********************************/*******************************************************/***************************************************************************************** 02:10:49 GregorR: huh? 02:13:03 bsmntbombdood: What's the least-overhead SHA-1 library. 02:13:36 (I want to toss one into something I'm writing, without also including MD5, SHA-bleh, MD4, MD-insecure, ...) 02:13:43 copy-paste the code from the SHA submission 02:14:29 Yeah, that's what I ended up doing :P 02:14:31 's lame though :P 02:25:48 -!- bsmntbombdood has quit (Success). 02:26:23 -!- bsmntbombdood has joined. 02:33:46 although we don't know how to do infinite loops in ///, it does seem like it is much easier to do computations involving bounded looping 02:34:29 MAIN B BUS UNDERVOLT 02:36:34 unless there are other obstacles, that should be enough to allow the equivalent of primitive recursion 02:38:30 I'm making a programming language that is literally impossible to program in. 02:39:09 we already have those 02:39:23 astrosteve: 700zm means that a spellbook is either cancellation or 02:39:25 err 02:39:31 stupid c-p 02:39:33 http://dangermouse.net/esoteric/ 02:42:13 -!- WangZeDong has quit (Read error: 60 (Operation timed out)). 02:43:27 -!- Slereah has joined. 02:45:46 You know, I'm not sure that even in real life I could talk the way ehird types. I don't speak in series of phrases; I say entire sentences at once. 02:45:58 Or maybe not. You know how unreliable introspection is. 02:46:21 not at all, _my_ introspection is utterly flawless. 02:49:59 So you're a self-improving artificial intelligence. 02:50:15 * kerlo locks oerjan in a Yudkowsky Box and submerges him under ten feet of seawater. 02:50:35 Hmm, this won't block sound all that well. 02:50:36 *GACK* *SPUTTER* ... 02:51:15 Sorry. I'm not allowed to let you out until this emergency meeting I've called concludes that it's safe to do so. 02:51:36 Everything will be very expeditious; you should be on dry land within 15 minutes if everything goes well. 02:53:14 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 02:53:47 -!- bsmntbombdood has joined. 03:02:27 GregorR, details? 03:02:36 Sgeo: I'm putting it up on the wiki now. 03:02:41 ooh 03:06:31 http://esoteric.voxelperfect.net/wiki/ShaFuck 03:07:02 It's so stupid it's stupid! :P 03:08:08 Anyone want to bruteforce a Hello world? 03:08:22 Actually, some programs might not even be writable in that 03:08:27 Why not? 03:08:35 Is SHA-1 160 bits? 03:08:37 Yes. 03:08:56 Sgeo: Oh, because the SHA-1 sums of 1024 bytes may or may not cover every possibility, right. 03:09:08 Sgeo: I chose such a large number in hopes that it would be "enough" *shrugs* 03:09:38 A lot of sha-1 sums might result in code that would be valid if it were not for comments 03:09:45 Easing the comment restriction might help 03:10:29 Easing the comment restriction would make it trivial. 03:10:35 You would just have to find eight sequences. 03:14:22 (8/256)^20 ~ 8e-31 03:16:00 oerjan: Yes, that IS an interesting statistic :P 03:16:31 so, something like several million years to find a single legal program with a current CPU? 03:16:40 Probably. 03:16:44 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 03:17:21 -!- bsmntbombdood has joined. 03:17:47 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 03:18:22 -!- bsmntbombdood has joined. 03:19:58 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 03:20:32 -!- bsmntbombdood has joined. 03:28:47 Would it be easier if I interpreted every 3 bits as a BF operation? That way everything would hash to valid BF, and it would be a problem of finding isomorphic BF ... which is probably still totally impossible :P 03:31:21 what are we talking about? 03:31:41 http://esoteric.voxelperfect.net/wiki/ShaFuck 03:32:12 at least it should then be possible to find actually syntactically correct programs 03:33:00 oerjan: Well, then it would be /im/possible to find /non/-syntactically-correct programs :P 03:33:13 you still have to match [] 03:33:24 Touché 03:33:30 hmmmm 03:33:42 i don't think it's _that_ difficult to write in 03:33:55 Poor, deluded bsmntbombdood :P 03:34:12 bsmntbombdood: as currently written, the main problem is the comment prohibition 03:34:25 oerjan: there are still other ways of nopping 03:34:39 um no 03:34:57 There are other ways of nopping, but even they would be nearly impossible to find. 03:35:16 the problem is it takes millions of years to find a program whose SHA-1 contains only bf commands 03:35:35 Only (8/256)^20 sequences are valid /at all/. Even finding one that's valid would take millenia. 03:35:48 Like he said :P 03:35:51 if you use the ascii-encoded brainfuck, sure 03:36:08 thus the new 3 bit suggestion 03:36:28 Which would still be nigh-on impossible, because those nopping isomorphisms are still pretty darn rare :P 03:36:58 i have a feeling you will be able to get it down to ~60 bits of complexity per block 03:37:10 bsmntbombdood: Feel free :P 03:37:25 so you will be able to produce shafuck at a rate of like 1 instruction per week 03:37:34 i think any program which contains [] has a hard time being a nop 03:38:58 now 20 bytes = 160 bits, which isn't divisible by 3 03:39:39 but discarding one bit, 53 commands 03:39:59 Right, yes, discarding the last bit. 03:40:11 Mind you, I didn't agree to that, I'm just presenting it as a thought experiment :P 03:40:43 (6/8)^53 ~ 2.4e-7, so it's not that hard to find something not containing [] 03:40:58 GregorR: it would probably be more interesting to use something like md5 or even md4 03:41:13 More interesting as in more possible to write. 03:41:20 yes 03:41:21 But I wasn't aiming for something that could actually be written :P 03:41:33 if you use sha1, your only option is brute force 03:41:45 (That we know of ;) ) 03:41:55 hm you probably want to avoid ., as well, for most building blocks 03:42:21 Definitely. 03:42:37 It's a PITA to make nops out of +-<>, it's almost impossible to make nops out of .,[] 03:42:46 so how many different combinations of 21 +'s and 21 -'s are there? 03:42:49 (Well, it is impossible to make nops out of ,.) 03:42:50 (4/8)^53 ~ 1.1e-16 03:43:17 with +-<> the whole becomes a kind of vector problem 03:43:46 add something to a number of cells, shift some amount 03:44:04 two with opposite shifts can nullify 03:44:40 and then linear algebra to find something which +- only one cell 03:45:00 no need to make it complicated 03:45:02 which should be possible with enough of those 03:45:11 um, this is not making it complicated 03:45:33 oerjan: Actually it is, you can just /run it/ to see what it modifies :P 03:45:49 er i'm not talking about single programs 03:45:57 OH 03:46:08 i'm talking about using linear algebra to combine blocks into something with a simpler effect 03:46:09 I thought you wanted them to be clean /within/ a single chunk. 03:46:10 I see. 03:46:32 grrr 03:46:52 how many combinations of 21 +'s and 21 -'s are there? 03:47:10 Oh, I thought that was rhetorical. 03:47:12 if you don't have <> then those cancel, you know 03:47:50 GregorR: no, given that you have an easy upper bound for complexity 03:48:05 42 over 21, or 42!/(21!*21!) iirc 03:48:07 Something like 21! I'd guess, but I don't know. 03:48:16 Apparently oerjan knows :P 03:48:30 oerjan: Uh, 42/21 is 2 :P 03:48:32 what's the log_2 of that? 03:48:37 ! is factorial 03:49:00 over is not division, but binomial coefficient 03:49:07 Ah 03:49:28 ok, 38.9 03:51:05 160 - log((2 * 42!/(21!*21!))) = 120 bits 03:51:27 + 3 bits for the instruction 03:51:39 and given control over +-<>, you can probably find a program that contains only those plus a single .,[ or ] 03:52:23 so i think this might be doable 03:56:04 (4^52 * 53)/(8^53) ~ 1.47e-15, so that is somehow slightly easier than a pure +-<> program - this cannot be right 03:57:56 oh, maybe it is 03:59:37 I think the long story short is "you're screwed" 03:59:45 hm? 04:01:03 well that is a bit of programs to search for 04:03:31 That's 680,272,108,843,537 sequences. 04:04:08 That's only 7 days at 1 BILLION sequences per second :P 04:04:35 At a more reasonable 1 million sequences per second, that's 22 years. 04:04:37 that's only 50 bits 04:04:46 that's easy sauce with dedicated hardware 04:05:04 AND you only have to do it 8 times, and you can then just compose blocks 04:05:46 i'm not sure about that 8 times 04:06:01 GregorR: you need to add some sort of chaining of blocks to prevent that 04:06:31 bsmntbombdood: Note that this concept is only pseudotractable IF I make it use 3-bit operations, which I'm not going to. 04:06:32 ouch 04:06:39 bsmntbombdood: With 8-bit operations, it's not even remotely tractable. 04:06:45 agreed 04:07:01 but it's kinda dumb not to allow comments 04:07:15 If I did allow comments, it would be trivial! The whole idea is for it to be IMPOSSIBLE :P 04:07:20 yeah 04:07:32 alas, if you allow comments it becomes trivial yeah 04:08:35 (248^19 * 20)/(256^20) ~ 0.042 04:08:48 it's stupid to make it completely impossible 04:08:52 that's to search for something equivalent to a single command 04:09:02 much more interesting just to require an immense, but practical computational load to write 04:09:39 bsmntbombdood: And you think that the 3-bit variety is immense but practical? 04:09:50 now if you used 3 bits plus chaining, it might become just about right? 04:09:58 "Chaining"? 04:10:19 bsmntbombdood's word 04:10:23 GregorR: so the same block has a different meaning depending on where you put it 04:10:37 i assume it means so that you cannot append found blocks easily 04:10:54 And I would do that by ...? 04:11:16 if you have a block that hashes to '+nopnopnopnop...', '-nopnopnopp...' etc, you can just put them together without any work 04:11:45 bsmntbombdood: Yeah, but I still find it EXTREMELY unlikely that you'll ever find those :P 04:12:07 you can't find such a block 04:12:12 every nop in BF takes two operations 04:12:41 coppro: We've already been over that, they're claiming it's (just barely) tractable, I'm not convinced. 04:13:11 the ascii version isn't 04:13:23 (Right, yes, the 3-bit version) 04:13:41 well, actually, you could get an odd-numbered nop, but it will erase some part of the memory 04:13:55 How about this: The hash of each 1024-byte block is taken to be the hash of everything in the file up to and including that block? 04:14:18 yeah, something like that 04:14:19 that's more fun 04:14:44 If I made it 4-bit, with the fourth bit allowing comments, that would be waaaaaaaaaay too easy, right? 04:15:20 do that and only use 60 or so bits of the hash, and you have system that trivial to interpret but very hard (but still possible) to write 04:15:23 kind of like hash cash 04:16:16 Bleh, the whole joy was that it was just plain impossible to write X-P 04:16:18 Joy for me X-P 04:17:10 maybe if the fourth bit _disallowed_ comments instead? 04:17:30 it's not interesting if it's impossible 04:17:47 Well, I'm going to sleep one way or another. 04:17:49 SO TAKE THAT 04:24:23 -!- Slereah has quit (Read error: 60 (Operation timed out)). 04:25:10 -!- Slereah has joined. 04:44:07 -!- Gracenotes has quit (Remote closed the connection). 04:44:27 -!- puzzlet has quit (Remote closed the connection). 04:44:34 -!- puzzlet has joined. 04:49:32 -!- Gracenotes has joined. 05:29:57 -!- oerjan has quit ("leaving"). 06:05:08 So, about that sonic computer. 06:05:28 A computer that uses Sonic for computation? 06:05:32 Spiffy. 06:05:42 g'night... time to think up a new precedence system for Agora 06:05:56 coppro: No precedence. 06:05:59 NONE! 06:06:10 No, a computer that uses sound for computation. 06:06:13 Everyone's a winner! 06:06:13 (I'm not sure what the game state would devolve to after that) 06:06:27 lots and lots of Paradox wins, I expect 06:06:59 Oh, I'd get Canada to win by paradox somehow. 06:07:12 :) 06:07:18 Am I a member of Canada's basis? 06:07:36 Are you Canadian? 06:07:39 yes 06:07:50 * coppro looks up the Wikipedia article on phonons... time to boggle 06:07:50 Then yes, you are. 06:08:03 Moot point unless you can get Canada to join Agora, though. 06:08:09 * kerlo ponders /// code to escape stuff 06:08:29 Replace / and \ with a and b, replace a and b with \/ and \\. 06:08:51 I guess you'll want to have guards as well. 06:10:03 !slashes /#\//a//#\\/b//a/\///b/\\//#//it's #/#/#\#\ 06:10:04 it's 06:10:14 * kerlo applauds 06:10:20 pikhq: it's not required to be a player to win 06:10:30 though getting Canada to be a person would be tough 06:10:48 !slashes /#\//a/d/#\\/b/d/a/\//d/b/\\/d/#//it's #/#/#\#\ 06:10:49 ddddit's 06:10:55 Actually, we could probably get the definition of "player" changed. 06:11:01 Person, though? 06:11:16 It's already a non-first-class person, right? 06:12:47 !slashes /##\//a/d/##\\/b/d/a/\//d/b/\\/d/#//it's ##/##/##\##\ 06:12:47 ddddit's 06:12:56 Grr. 06:16:00 !slashes /##\//a/d/##\\/b/d/a/\//d/b/\\/d/#//it's ##/##/##\##\ abba 06:31:08 -!- puzzlet has quit (Read error: 60 (Operation timed out)). 06:42:05 -!- Gracenotes has quit (Remote closed the connection). 06:43:35 -!- Gracenotes has joined. 07:01:43 -!- puzzlet has joined. 07:38:14 ohnoslashes 07:40:18 Sonic computers reminded me of http://en.wikipedia.org/wiki/Delay_line_memory#Acoustic_delay_lines which are an awesome idea... I mean, just compare http://en.wikipedia.org/wiki/File:Mercury_memory.jpg with a boring DIMM memory chip. 07:43:49 i just got back from Star Trek 07:49:33 -!- WangZeDong has joined. 07:51:49 -!- Slereah has quit (Read error: 110 (Connection timed out)). 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:06:55 -!- Sgeo has quit (Read error: 110 (Connection timed out)). 08:33:04 -!- jix has joined. 08:46:34 -!- puzzlet has quit (Read error: 60 (Operation timed out)). 08:48:43 -!- puzzlet has joined. 08:59:30 -!- coppro has quit (Remote closed the connection). 09:06:34 -!- Slereah has joined. 09:18:40 -!- WangZeDong has quit (Read error: 110 (Connection timed out)). 09:29:48 http://arxiv.org/abs/0905.0740 09:56:53 http://james-iry.blogspot.com/2009/05/brief-incomplete-and-mostly-wrong.html 10:04:22 -!- M0ny has joined. 10:06:03 hey 10:07:09 Hello mona 10:08:20 spèce de scélérat 10:08:39 Deewiant: rotfl 10:08:47 Toi-même 10:08:58 c'est pas vrai ! 10:38:13 -!- psygnisfive has quit (Remote closed the connection). 10:39:35 -!- psygnisfive has joined. 10:43:53 i just got back from Star Trek 10:43:59 yes. 10:44:02 psygnisfive: And is it as godawful as the commercials make it out to be? 10:44:06 no 10:44:10 it was wonderful 10:44:12 in some respects 10:44:15 I figured it couldn't be that bad. 10:44:19 the plot was thin 10:44:26 and it takes place in an alternate universe 10:44:34 Whaaaaaaaaaaaaaaaaaa 10:44:36 yes 10:44:38 but that aside 10:44:47 the vision of the star trek universe is quite nice 10:50:36 -!- impomatic has joined. 10:50:44 Hi :-) 10:50:57 Anyone know which was the first 2D programming language? 10:51:17 befunge i suppose 10:51:18 I think it was Befunge 10:52:08 Thanks 10:56:11 Strange that Befunge wasn't mentioned on usenet until 1996 10:57:06 There was an eso mailing list 10:57:17 Which probably saw most of the discussion 10:57:35 Heh. Mailing lists. 10:58:03 it was invented in 1996 10:58:17 "Later the Befunge Mailing List evolved into the Esoteric Topics Mailing List, and the word began to be used to describe this kind of language. " 10:58:30 -- http://esolangs.org/wiki/Esoteric_programming_language 10:59:54 http://esoteric.sange.fi/archive/ goes only as far back as 2001 11:01:08 http://sourceforge.net/mailarchive/forum.php?forum_name=funge-list has a tiny amount of stuff 11:01:17 Now reading about Biota from 1991 11:01:56 I just wanted to check if Corelife (April 1993) was predated by any other 2D languages. 11:02:05 Like the wiki puts it: "Befunge was preceded in 1991 by a similar but less featureful language Biota, which was designed for experiments in self-reproduction. It was followed soon after, in 1994, by another similar language, Orthagonal, the design of which was spurred by a discussion on alt.folklore.computers." 11:02:56 So it was predated by Biota and possibly Befunge, depending on the month Befunge was released. 11:03:33 "Original document September, 1993" says the Befunge-93 spec in Catseye. 11:05:37 Well, that's pretty recent then 11:05:38 http://esolangs.org/wiki/User_talk:Calamari has a broken link to a partial archive of the befunge mailing list 11:05:45 Because this september is still going on 11:05:50 *rimshot* 11:05:50 But that's all I can find 11:06:40 esolangs archives should be on the web. 11:07:31 Thanks :-) 11:08:09 So I guess Corelife was the second 2D programming language :-) 11:08:14 There's the forum, of course, but that's recent: http://esoteric.voxelperfect.net/forum/subback.html 11:08:42 Deewiant: The broken link is (was?) a pkzipfixed former broken zip; it was talked here on http://tunes.org/~nef//logs/esoteric/05.11.17 -- I guess it's gone dead again. 11:08:49 People use that forum? 11:08:53 fizzie: Yes, that's how I found it. 11:09:15 well, i can upload those maillist archives, if somebody need them. 11:09:17 I guess we can poke calamari about it if he shows up. 11:09:38 mtve: What do you have? 11:09:59 well, i was a subscriber since almost the begining :) 11:10:22 Then please do :-) 11:10:34 ok, noted, will do it in a few days. 11:10:52 I'd rather have it on more machines than just yours, it'd be a shame to lose them to a drive crash or something ;-) 11:12:08 yep, a bitrot becomes a real problem. i can't find many interesting things on the web which were there, i'm sure. 11:12:31 Heh, http://esoteric.sange.fi/archive/current is a collection of spam. :p 11:12:52 Actually hmm, I've got some FBBI patches which are older than the stuff on sange.fi, I wonder where I got them from 11:12:55 Lots of offers for millions of dollars. 11:13:36 Scratch that, they were from 2003, not the 2000 I thought. 11:13:42 most of interesting things are already on the wiki, but archives do have their value. 11:14:15 Archives give you a linear view of history, wikis give you a flat view of the current state. 11:14:34 I hate bitrot :-( 11:14:40 fizzie: Looks like both 2006 and 2007 are also all spam. 11:14:52 WTF COMMERCIAL!! Why is there a commercial for Emerald Nuts that implies a grizzly death of the three main characters shortly after the commercial ends >_< 11:15:23 I've been archiving stuff I find interesting so it isn't lost forever. 11:20:19 http://www.esolangs.org/wiki/ShaFuck 12:03:05 -!- MizardX has quit ("Proclamation of invalidity!"). 12:09:36 Aw, mooz's Befunge pages are gone 12:11:59 Are they in the web archive? 12:12:05 Yes 12:12:15 But no pics/downloads, of course 12:13:06 Hmmm... some pages have pics / zips archived too. 12:14:07 mtve: Exactly how slow is http://frox25.no-ip.org/~mtve/code/eso/bef/chess/ supposed to be? (Or, what interpreter can run it) 12:14:37 pretty fast, any sane interpreter should run it well. 12:15:01 less then a second to response afair 12:15:09 CCBI and FBBI are both going at 45 minutes on the first move, tkbf93 segfaulted :-P 12:15:19 I don't have other -93 interpreters. 12:15:38 I think I asked mooz about that stuff, and he said he has an archive in some non-public place. 12:15:42 eh, maybe i forgot something :) i have my own interpreter somewhere near to that 12:16:15 Ah, so you do 12:16:34 actually a link from that page 12:16:49 Okay, the C one offers fast response 12:16:53 Let's see how it breaks the standard ;-) 12:17:05 -!- MizardX has joined. 12:17:13 i was trying to write it as strict as possible afair 12:18:02 Welp, Mycology likes it 12:18:24 Wonder what's going on... will have to look into this at some point 12:18:29 -!- Judofyr has joined. 12:34:32 -!- pikhq has quit (Read error: 104 (Connection reset by peer)). 12:34:52 ehird: http://www.phoronix.com/scan.php?page=article&item=amd_r600_700_guide, usable 3D for ATIs "should be just a matter of weeks" 13:12:44 -!- oklopol has joined. 13:12:58 okokokokokokokokokokokokokokokokokokokoko 13:13:01 okokokokokokokokokokoko 13:13:06 okokokokokokokokokokokokokokkokokokokokokokokokokokoko 13:13:12 okokokokokokokokokokokokokokokokokoko 13:13:16 oh 13:13:18 i have failed 13:13:33 insert harakiri 13:14:10 -!- MizardX- has joined. 13:16:13 Koko kokko. 13:16:29 -!- MizardX has quit (Nick collision from services.). 13:16:31 -!- MizardX- has changed nick to MizardX. 13:19:29 koko kokkoko? 13:19:49 kokki kokkaa kakkaa 13:20:16 Kokoo kokko kokoon. 13:20:29 I guess we did the kokko-thing here already, though. 13:20:39 i guess we somewhat did. 13:21:25 ^cho koko kokko 13:21:25 koko kokkooko kokkoko kokkoo kokko kokkokokkookkokkokoo 13:21:38 ^help 13:21:38 ^ ; ^def ; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^style [style]; ^bool 13:21:41 ^show cho 13:21:42 >,[>,]<[<]>[[.>]<[<]>[-]>] 13:22:09 ^choo koko kokko 13:22:10 koko kokko oko kokko ko kokko o kokko kokko kokko okko kko ko o 13:22:18 There were those two functions of dubious usefulness. 13:22:21 ^show choo 13:22:21 >,[>,]+32[<]>[[.>]<[<]>[-]>] 13:22:40 That's pretty much the same except with a space added between repetitions. 13:22:48 using syntax elements from both C++ and Haskell feels a bit weird. 13:23:21 i just love how C++ uses << and >> for streams 13:23:26 i guess not strictly syntax thing 13:23:36 Bitshift the output stream, yes. 13:25:17 well true, i guess "hello world" >> cout would be more logical, because right argument tells how much ~ where to shift 13:25:37 (how much) ~ (where), that is 13:26:38 but cin >> variable makes sense at least 13:26:58 anyway strongly typed operators are a pretty stupid idea anyway 13:27:03 anywaywayawyw 13:27:17 I don't suppose the streams overload <<= and >>= with something fancy? 13:27:32 like what? 13:27:43 i doubt it, not that i've checked 13:28:01 well 13:28:11 actually << and >> already have those semantics 13:28:21 so they could just be the same thing 13:29:11 Yes. I don't think they are, though. 13:30:29 i'm not sure why they would be 13:32:07 !cho ososo 13:32:11 ^cho ososo 13:32:12 ososososoososoo 13:32:15 ^choo ososo 13:32:16 ososo soso oso so o 13:32:57 ^choo ososo soso oso so o 13:32:57 ososo soso oso so o soso soso oso so o oso soso oso so o so soso oso so o o soso oso so o soso oso so o soso oso so o oso oso so o so oso so o o oso so o oso so o oso so o so so o o so o so o so o o o o ... 13:33:02 Meaningful! 13:37:00 if there was a reverse choo, you could say "o so", and your "meaningful" would be meaningful 13:37:47 ^cho >,[>,]<[<]>[[.>]<[<]>[-]>] 13:37:47 >,[>,]<[<]>[[.>]<[<]>[-]>],[>,]<[<]>[[.>]<[<]>[-]>][>,]<[<]>[[.>]<[<]>[-]>]>,]<[<]>[[.>]<[<]>[-]>],]<[<]>[[.>]<[<]>[-]>]]<[<]>[[.>]<[<]>[-]>]<[<]>[[.>]<[<]>[-]>][<]>[[.>]<[<]>[-]>]<]>[[.>]<[<]>[-]>]]>[[.>]<[ ... 13:38:45 Deewiant: what does that do 13:38:55 Which 13:39:00 the program 13:39:05 Which program 13:39:11 *the* progarm 13:39:12 I don't know about fungot's one 13:39:12 *program 13:39:13 Deewiant: meet is simple enough.) indent all of the tail of the list. :p 13:39:27 The one I gave to ^cho is cho's source 13:39:39 What came out; who knows 13:39:51 It indents all of the tail of the list, I guess. 13:39:51 i know it's cho's source, i just saw it 13:39:57 I'm sure fungot knows what it's speaking of. 13:39:58 fizzie: on dit fnord, en kerro miten, mutta liittyy fnord ja fnord ilman siihen fnord fnord fnord" 13:40:07 D: 13:40:14 fungot: Stop with the Finnish gibberish. 13:40:14 fizzie: nice job on ooe on c.l.s. btw this is called " robotron" 13:40:16 ^style 13:40:16 Available: agora alice darwin discworld europarl ff7 fisher ic irc* jargon lovecraft nethack pa speeches ss wp 13:40:40 That's from #douglasadams@ircnet; we have occasionally lapsed into Finnish there. 13:41:38 suomi fnord fnord kieli 13:41:50 fungot: fnord fnord 13:41:51 Deewiant: but i don't really know enough about s48's internals to see where i can 13:42:15 To fnord a fnord you need intimate knowledge of s48's internals. 13:43:01 fungot: How is fnord formed 13:43:01 Deewiant: i'll go going to work on srfi 44 seems too complicated for too little gain. 13:43:18 fungot: How fnord get fnord 13:43:19 Deewiant: pretty much. although regular continuations don't do fnord? ( although i find sdl a bit distasteful. but not enough to actually volunteer to write it 13:43:30 I really should make it take the trailing parts of the comment it's replying to as a context, but that would mean I'd have to map back from strings to tokens in a reasonable time. 13:52:06 -!- tombom has joined. 13:55:24 -!- jix has quit ("bye"). 13:58:06 -!- impomatic has left (?). 14:33:25 -!- jix has joined. 14:35:36 -!- okloduk has joined. 14:48:46 ^show cho 14:48:47 >,[>,]<[<]>[[.>]<[<]>[-]>] 14:48:55 ^show choo 14:48:55 >,[>,]+32[<]>[[.>]<[<]>[-]>] 14:49:32 fizzie, you don't turn [-] into set 0? 14:49:34 mhm 14:50:20 lifthrasiir, I think esotope-bfc needs a plugin architecture! ;P 14:50:38 that way you could easily write new optimisations, possibly in some DSL. 14:50:47 hm maybe pattern based... 14:51:23 No. I only run-length-encode consecutive <>+- (and there's that bug). I don't have a "set" instruction in the bytecode, either, although I guess I could easily have one. Should probably fix at least the bug. 14:51:38 yeah 14:52:12 fizzie, I'm not saying you should implement esotope-bfc in befunge. Just that [-] is so common and easy to do 14:52:29 possibly [-]+ into set 1 and such 14:52:59 * AnMaster thinks he should write a semi-optimising one in scheme. Just doing the basic stuff. 14:53:19 lifthrasiir, why the name "esotope"? 14:55:00 lifthrasiir, by the way, I have some other ideas: What if some code snippet is reused a lot. Maybe add an "size optimising" option that tries to "factor out" common code into separate functions 14:55:18 if the code snippet is long enough that is 14:55:39 -!- dbc has quit (wolfe.freenode.net irc.freenode.net). 14:55:50 this also opens the questions of intrinsics in an extended mode of course. But maybe that is a bad idea... 14:56:15 -!- oklopol has quit (Read error: 110 (Connection timed out)). 14:56:23 -!- bsmntbombdood has quit (Connection reset by peer). 14:57:02 -!- bsmntbombdood has joined. 15:03:31 -!- tombom has quit. 15:03:55 -!- okloduk has quit (Read error: 113 (No route to host)). 15:10:32 -!- dbc has joined. 15:15:43 fizzie, what do you think about a DSL like this for an optimising bf compiler: 15:15:45 match: { loop: { '-', move : (store myvar count), '+', move: (is_equal? -myvar count) } } replace: { set: { offset: (var myvar), value: {get_cell 0}}, set: { offset: 0, value: 0}} 15:16:00 to match [->>>+<<<] into a "move and set old to 0" 15:16:13 s/into/and change into/ 15:16:33 it needs some work 15:16:59 (like: the syntax there is a horrible mix of python dicts and scheme) 15:19:35 AnMaster: Thanks to mtve, some old messages from the Befunge mailing list, includes info on -96 and -97: http://frox25.no-ip.org/~mtve/tmp/bef_maillist_0_520.txt 15:20:23 ooh 15:20:25 tmp 15:20:26 hm 15:20:28 * AnMaster saves 15:20:44 I asked him about it and he said it saves stuff forever :-P 15:20:53 But more copies is always a good thing :-) 15:21:24 Deewiant, does it contain 97 versikn? 15:21:27 version* 15:21:59 AnMaster: It has messages from -96 and -97 where they discuss Befunge. 15:22:03 ok 15:22:05 And the -96 and -97 standards. 15:22:10 the specs are in there? 15:22:19 Probably not since the last message is from June. 15:22:22 ah... 15:22:26 I don't even know if -97 ever had a finalized spec. 15:22:49 But that also has stuff on BeGlad and the like, which is interesting. 15:22:50 "I also got most of Visual Befunge for Windows actually working." 15:22:51 wow 15:22:53 just wow 15:22:56 impomatic might be interested in that stuff, too. 15:23:05 Deewiant, um why? 15:23:11 did he implement some interpreter? 15:23:19 No, but it's similar to Core Wars. 15:23:22 oh ok 15:23:29 I think, at least. 15:23:34 I haven't read any of that yet. 15:23:35 Deewiant, does it explain the team id thing? 15:23:48 That was the first thing I grepped for, actually :-) 15:23:50 And no, it doesn't 15:23:53 ouch 15:23:59 Maybe it does, just not under that name 15:24:03 Like said, I haven't read any of it. 15:24:42 Chronefunge: Subject-singularity Time-Travel Befunge! 15:24:48 Chris Pressey's idea, originally. 15:24:57 "PS. Competition: smallest self-listing prog (but not f****n' emptiness)" 15:24:58 :D 15:25:11 Looks very similar to TRDS. 15:25:46 ooooh 15:25:53 sine in befunge-93? 96? 15:25:59 Sine? 15:26:17 Oh, an impl of it 15:26:19 2. sine function 15:26:19 yes 15:26:25 1. postfix calculator 15:26:26 Looks like -93. 15:26:28 seems interesting too 15:26:36 Did you see chess.bef? 15:26:49 http://frox25.no-ip.org/~mtve/code/eso/bef/chess/ 15:27:14 I was wondering why FBBI and CCBI both infinite-loop on it but his Bef93 interp doesn't 15:27:37 Deewiant, might be due to reflecting on something in 98? 15:27:47 AnMaster: In 93 compatibility mode, of course :-P 15:27:55 Deewiant, don't know then 15:28:47 Yes, I don't either. 15:28:56 Haven't looked into it in detail yet. 15:31:15 In Befunge-96 all IPs had the same stack 15:31:20 I.e. shared 15:31:47 g and p were deprecated, that's interesting 15:40:45 Real life needs lazy evaluation... 15:41:18 AnMaster: Visual Befunge - http://members.home.nl/wimrijnders/bef.htm 15:42:32 no screenshots? 15:42:42 "I just thought of something: Are you sure that you want to have longs as 15:42:42 the coordinates? I can't imagine anyone having a befunge program more than 15:42:43 65000 x 65000." 15:42:48 I don't have any vm set up atm. So testing it will be highly messy 15:42:53 Does fungot exceed that? 15:42:54 Deewiant: that pesky exponential growth!! i'm having too much fun with this guy 15:43:00 Deewiant, the stack? yes 15:43:03 for ul 15:43:14 AnMaster: No, Funge-Space. 15:43:27 Deewiant, yes the ul stack... in funge-space 15:43:41 Ah, it's stored in Funge-Space? 15:43:53 yes 15:44:00 and it grows into negative funge-space 15:44:07 far into it 15:44:12 Heh 15:44:13 I'm not quite sure what I set the ^ul stack limit to, it might go below -32768. 15:45:03 Other than that I don't use many far-off coordinates. The ^bf tape is at some row, columns [0, 999] or something equally small. 15:46:31 !help 15:46:32 Supported commands: addinterp bf_txtgen daemons delinterp fyb help info kill userinterps 1l 2l adjust asm axo bch befunge befunge98 bf bf16 bf32 bf8 bfbignum boolfuck c chiqrsx9p cintercal clcintercal cxx dimensifuck echo forth glass glypho hello kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor rot13 sadol sceql sh slashes test trigger udage01 underload unlambda whirl yodawg 15:46:40 !bf_txtgen a 15:46:43 39 +++++++++++[>+++++++++>+>><<<<-]>--.>-. [22] 15:46:45 !bf_txtgen a 15:46:48 39 ++++++++++[>++++++++++>+>><<<<-]>---.>. [95] 15:46:50 hm 15:46:55 surely that could be much shorter? 15:47:02 !bf_txtgen a 15:47:05 39 ++++++++++[>++++++++++>+>><<<<-]>---.>. [18] 15:47:14 apart from the >><< bit I mean 15:47:20 !bf_txtgen a 15:47:23 39 ++++++++++++[>++++++++>+>><<<<-]>+.>--. [35] 15:47:24 !bf_txtgen A 15:47:27 35 ++++++++[>++++++++>+>><<<<-]>+.>++. [59] 15:47:30 !bf_txtgen A 15:47:32 35 +++++++++++[>++++++>+>><<<<-]>-.>-. [33] 15:47:37 hm 15:47:47 oh it adds a newline? 15:47:48 hm 15:48:36 Deewiant, "Get and Put Discouraged in favor of Get-Left, Get-Right, Put-Left and Put-Right" 15:48:43 aren't some of those in TOYS? 15:48:44 iirc 15:48:47 AnMaster: Yes, as I said, g and p deprecated. 15:48:51 yes 15:48:59 but you didn't say anything about TOYS 15:49:00 ... 15:49:04 which was my question 15:49:06 And yes, "get left/right hand" are stated to have been based on get-left / get-right in Befunge-96. 15:49:19 ah yes... 15:49:27 They may or may not be the exact same commands. 15:49:34 One-shot Stringmode - Looks useful. <-- is this ' I wonder... 15:49:46 Yes. 15:49:51 -!- asiekierka has joined. 15:49:54 Oh god 15:49:57 AnMaster: Grep for 'Edit #2 - Dec 19 1996 Chris Pressey, Cat's-Eye Technologies. 15:49:59 Virut attacked me AGAIN 15:50:01 AnMaster: You'll find a Befunge-96 spec. 15:50:04 but this time it should be dead for good 15:50:06 oooh 15:50:09 removed all EXE's and DLL's 15:50:12 AnMaster: Includes descriptions of directives. 15:50:15 and proceeding with safety to the remaining ones 15:50:27 Deewiant, \o/ \o/ \o/ 15:50:44 asiekierka, clean reinstall duh 15:50:47 AnMaster: Also those instructions. And yes, ' is "One-Shot Stringmode". 15:50:49 well, not exactly 15:50:53 "clean" reinstall is a full format 15:50:59 I would call it "half-clean" reinstall 15:51:29 asiekierka, full format was what I meant 15:51:43 only keep plain text files, stuff you can manually inspect 15:51:47 and 15:51:51 well, I can't do a full format 15:51:56 do not run as administrator 15:51:57 too much stuff on my hard disk to sort them 15:51:58 in the future 15:52:00 and 15:52:22 XP either offers omnipotent admin or cantdoanything guest 15:52:54 also, I have a better antivirus 15:52:56 err. I'm pretty sure it has a mode in between called "normal user" or such, instead of "limited user" 15:53:00 but maybe I misremember 15:53:14 there was that "power user" too 15:53:15 well, on XP SP2 there's either "Admin user" or "Normal/Limited user" 15:53:18 which was not very secure 15:53:28 AnMaster: Grep for "Edit #3", it's from 1996-12-31 and is the final Befunge-96 spec :-) 15:53:41 I've FINALIZED the Befunge-96 spec (because, uh... it's about 2 hours away 15:53:41 from 1997... and anything after that would have to be called "Befunge-97". 15:53:59 actually, XP says that there's either "Computer Administrator" mode (Admin mode basically) or "Limited mode" 15:54:05 asiekierka, I'm 100% sure "power user" mode 15:54:27 Hahaha, holistic deviancy 15:54:32 Best idea ever 15:54:35 check the user management thing found in one of them *.msc files 15:54:35 iirc 15:54:49 so either Power User or Can't-do-Anything-Useful-Other-Than-Use-Apps-Dammit User 15:54:52 Hahaha, holistic deviancy <-- in 96 draft? 15:54:57 AnMaster: In 96 final 15:55:03 Q: What's the FIRST Befunge spec 15:55:03 Wasn't in Edit #2. 15:55:04 ah 15:55:06 asiekierka: 93. 15:55:18 ...Drafts do count. 15:55:23 93. 15:55:25 :-P 15:55:32 ...There's no Befunge-92?! 15:55:36 Then I must go BACK IN TIME! 15:55:37 Not to my knowledge. 15:55:52 I think -93 came "out of the blue" when Chris released it. 15:55:54 I could be wrong. 15:56:36 Also, is there any esoteric language that's "top at the moment" here right now? 15:56:40 I think Befunge 15:56:43 ;$I rand6.bf 42 18 15:56:43 ; that was a compiler directive to include another .bf source 15:56:44 haha 15:56:51 Chris sez "I don't think there will be any more major changes" -- 1997-01-23 15:56:54 So much for that. 15:57:05 Deewiant, are you going to add a 96 compat mode to ccbi? 15:57:14 Maybe, if it's simple to do. 15:57:25 Haven't looked at that in enough detail to be able to say if it is :-) 15:57:28 Also, is there a tutorial on the mess of commands that's called Be(a)funge? 15:57:34 Maybe, if it's simple to do. <-- hey, that's my comment 15:57:36 No tutorial. 15:57:38 :P 15:57:51 asiekierka, what the hell is the "(a)" there for 15:58:08 i don't know 15:58:14 I know there is a Befunge-93 tutorial 15:58:27 asiekierka, you put the "(a)" there. What did you mean with it 15:58:27 No Befunge-9[678] tutorial. 15:59:23 * $C command constant 15:59:25 that is nice 15:59:41 ;$C @ 0 15:59:44 Oops! 15:59:45 :-P 15:59:58 "stack by a particular extentible null command" 16:00:06 not sure what an "extentible null command" is yet 16:00:14 but sound like one not defined by the spec? 16:00:18 AnMaster: I'm not telling! 16:00:19 Yes. 16:00:35 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 16:00:44 asiekierka, about what? 16:00:48 the (a) 16:00:51 Also, I want to implement BF for the C64 16:01:01 93 can be done. 16:01:03 I assume it was you being confused then 16:01:13 -!- bsmntbombdood has joined. 16:01:13 Nope 16:01:17 Deewiant, I'm not so sure, but I suspect 98 could be done. 16:01:17 The C64 doesn't have enough memory for 96. 16:01:18 Duh 16:01:22 Or 98. 16:01:29 You break my project. 16:01:34 Deewiant, um, limited memory of course 16:01:38 right 16:01:43 if you remove the 32 bit max it works 16:01:45 AnMaster: Whatever you assume, I added (a) to make a sentence that I DON'T EVEN KNOW WHAT MEANS 16:01:48 err 16:01:48 min 16:02:11 doh, there is already a BF interpreter for the C64 16:02:13 You lose fingerprints without 32, though. 16:02:22 Which is a bit of a shame. 16:02:25 Deewiant, actually you could do it 16:02:30 since funge-space would be sparse anyway 16:02:34 This is a birthday demo for Martin Wendt, a true C64 fanatic. It's a multicolour brainfuck interpreter, using the screen RAM for data and code, meaning that you see the code execute and modify itself! 16:02:37 you just will get OOM rather fast 16:02:39 that's the BF interpreter xD 16:02:48 Deewiant, you can manually do 32-bit numbers with 8 bit ones 16:02:51 if you have to 16:03:19 Hmm, 80*25 = 2000 * sizeof bytes for a Befunge-93-sized program 16:03:40 The C64 with BASIC FEATURES ON has ~38911 bytes of free memory 16:03:52 BASIC FEATURES == everything BASIC offers 16:04:02 And if you turn BASIC off you get even more memory 16:04:14 Deewiant, there is an OS with a webserver and web browser for C64 16:04:16 so 16:04:23 I don't think it is impossible 16:04:25 just hard. 16:04:32 Food time -> 16:04:35 AnMaster: Where to download it because I am wondering since forever 16:04:42 asiekierka, download what? 16:04:46 Contiki! 16:04:49 the C64 OS 16:04:50 no clue 16:04:53 google? 16:04:54 there's also GEOS but duh 16:05:00 AnMaster: Well, i don't even need it so nah 16:05:14 I don't really care a lot about C64 16:05:28 to me it is history, and not very interesting history 16:07:12 I do 16:07:25 I also own an Amiga but I don't own hardware to transmit data between PC and Amiga 16:07:31 and Amiga is just... too new for me 16:07:32 duh 16:07:35 not enough limits 16:09:17 Deewiant: DIdn't you work on a DOBELA interpreter? 16:09:25 asiekierka, he finished it iirc 16:09:26 ... 16:10:27 ...what??? 16:10:50 ...actually, yes, yes he did 16:11:08 Augh, not only for Linux only, but also 64-bit 16:11:12 there's NO WAI I can test it 16:11:41 ...did I use a very old meme just about now? 16:13:18 none that I know of 16:13:26 though "WAI" sounds meme-y 16:13:33 asiekierka, but why can't you test it 16:14:07 No Linux 16:14:18 asiekierka, you could install qemu and emulate 64-bit in case you only have a 32-bit CPU 16:14:20 and I'm not sure if my Core 2 Quad is 64-bit 16:14:23 and installing linux is easy 16:14:37 asiekierka, core 2 quad is most likely 64-bit 16:14:41 I already wasted 3 hours today installing XP 16:14:42 not 100% sure of intel cpus 16:14:52 Core 2 Quad IS an intel CPU, doh 16:14:54 asiekierka, agreed. installing windows is wasting time 16:15:03 asiekierka, yes... that was what I said 16:15:09 oh 16:15:11 and NO WAI is from the same meme as O RLY and YA RLY, thought that you should know 16:16:04 asiekierka, last time I checked installing linux usually took way less 16:16:12 16:14 asiekierka: Core 2 Quad IS an intel CPU, doh 16:16:15 something like, uh, 30 minutes? 16:16:18 stuuuupid 16:16:40 ehird, good evening 16:16:46 good morning! 16:16:56 ehird, not even in UK is it "morning" now 16:17:00 maybe "afternoon" 16:17:09 AnMaster: ehird time is objectively right. 16:17:11 11:34 Deewiant: ehird: http://www.phoronix.com/scan.php?page=article&item=amd_r600_700_guide, usable 3D for ATIs "should be just a matter of weeks" 16:17:13 i am aroused 16:17:26 open source hardware specifications are the best kind of porn! 16:17:35 AnMaster: Last time I checked installing my favourite distro of Linux usually took about the same time 16:18:15 asiekierka, which distro is that... 16:18:21 AnMaster: probably pclinuxos or something xD 16:18:25 Or Mandrakeiva! 16:18:27 ehird: Actually, it's SuSe. 16:18:28 asiekierka, and why aren't you dual booting then already 16:18:32 oh my 16:18:33 lawl suse 16:18:34 suse... 16:18:36 ...what? 16:18:40 I like it 16:18:42 I am used to it 16:18:42 It's like Fedora except crappier! 16:18:53 actually I used suse lots of years ago. And ehird's summary was perfect 16:19:08 I am used to it and I don't want to waste time to learn a bunch of different stuff 16:19:09 I went back to slackware after 16:19:52 i'm having a bit of cognitive dissonance here 16:20:01 my brain knows I shouldn't like Arch 16:20:05 but my mind knows I like it 16:20:11 asiekierka: It might work on BSDs too :-P 16:20:13 i'm pretty sure it's dangerous to hav ethem out of sync 16:21:06 In fact, it not only might but should work on BSDs and Solaris, though I haven't tested it 16:21:07 14:42 Deewiant: "I just thought of something: Are you sure that you want to have longs as 16:21:07 14:42 Deewiant: the coordinates? I can't imagine anyone having a befunge program more than 16:21:09 14:42 Deewiant: 65000 x 65000." 16:21:10 Hohohohohohoho 16:21:15 Did we find the 96/97 specs? 16:21:29 ehird, yes. The mind must be allowed to win, or you risk going mad. And not a good kind of mad (like esolang crazy/mad), but the gibbering lunatic kind of mad. 16:21:35 ehird: 2009-05-08 17:19:35 ( Deewiant) AnMaster: Thanks to mtve, some old messages from the Befunge mailing list, includes info on -96 and -97: http://frox25.no-ip.org/~mtve/tmp/bef_maillist_0_520.txt 16:21:35 ehird, we found 96 at least. 16:21:53 AnMaster: OTOH, I think that consciousness and the mind is merely a byproduct of the brain's process. 16:22:01 So the mind thinking something the brain doesn't might cause a metaphysical quantum rip ;-) 16:22:12 ehird, so quickly make the brain think the same 16:22:22 mtve said he's digging up more stuff so with any luck we'll have the full archives online sooner or later. 16:22:23 how? I'm the mind, I don't have control over it 16:22:35 ehird, oh hm. 16:23:05 Deewiant, great 16:23:46 The h "Holistic Deviancy" command pops a value off the stack and uses it as the 16:23:46 Holistic Deviancy value for the current program counter (and ONLY the current 16:23:46 PC), throughout the entire program. 16:23:47 err 16:23:53 what exactly is "Holistic Deviancy value" 16:23:56 "Forgive me, I've taken the drastic step of subscribing all of you to this 16:23:56 list, as you've been kind enough to take some interest in Befunge. " 16:23:58 Ouch. 16:23:58 ah right 16:24:00 AnMaster: Read the next paragraph. 16:24:01 it says a bit below 16:24:04 Not very nice. 16:24:49 Deewiant, like storage offset, except it offsets the value instead of position? 16:24:50 It wasn't a high-volume list when only that message was there, so I don't think it's very not-nice. 16:25:11 AnMaster: Kinda. 16:25:46 Storage offset only applies to commands that mess with funge-space; that, as I understand it, applies to execution as well. 16:26:45 (There is no reflect 180 degree 16:26:45 operator; but one can be simulated with T@.) 16:26:46 err 16:26:48 oh 16:26:51 like THAT 16:26:53 nasty 16:27:01 I.e. 1H1)G.@ prints 2. 16:27:20 Lol, it's fun watching people come up with inferior keyboard alternatives to a very mouse operation (clicking links) 16:27:24 Deewiant, err won't the G turn into a H by then? 16:27:32 or does it only affect fetches and such? 16:27:35 AnMaster: Yes, )G becomes 0H. 16:27:46 Which is exactly what I intended. 16:27:52 hm ok 16:27:55 right 16:28:03 Deewiant: Except that H puts the holistic deviancy on stack; you want 'h' instead? 16:28:12 Oops, right you are. 16:28:23 Gone for a few minutes -> 16:32:08 Speaking of limited-memory systems, the TI-86 has mooz's rather nice interpreter/debugger, and that thing has 96 KB of user-accessible RAM. Of course that's befunge93 only. 16:33:40 Befunge93 isn't much of a memory hog. 16:34:32 As if. 16:34:41 I had to buy 12GB of RAM to run my befunge-98 mission-critical programs 16:34:45 Still, CCBI manages to use 1484 K. 16:36:23 From srcc!crocus!pangea.ca!befunge-request Sat Jan 04 05:02:14 1997 16:36:25 hm 16:36:32 err 16:36:37 There's also rather little (can't really remember how little) of RAM that's really designated as "working RAM", since all programs the user installs (and other variables) are stored in the RAM too. Although you can add string variables to the VAT and use their contents as working RAM, and there's some free space you can use if you don't call the ROM function-plotting routines. 16:36:39 it isn't THAT old is it 16:36:43 the mailing list 16:36:48 AnMaster: 1997? 16:36:51 yes... 16:36:58 ehird, didn't they have MX records by then? 16:37:06 what's that got to do with anything 16:37:09 it's just an mbox 16:37:19 ehird, oh not !-bang path in routing 16:37:30 it's that too, I think 16:37:40 ehird, I'm pretty sure that is even older 16:37:48 I'm pretty sure there's a reason for it 16:37:54 well yes 16:38:31 "Before I make the commitment to design a fully Befung-ctional system for the Mac" 16:38:37 what is "Befung-ctional" I wonder 16:38:53 Functional befunge. 16:39:00 "a fully functional Befunge" 16:39:03 "a fully Befung-ctional" 16:39:09 they sound similar 16:39:15 hm ok 16:39:33 such things are easier for native speakers as usuak 16:39:35 usual* 16:39:40 "I got an idea for Befunge playing SimCity. Why not envision a Befunge program as a road map of a 16:39:40 city? Each cell is part of a line of cells in one row or one column, so why not structure them 16:39:40 that way?" 16:39:41 wow 16:39:46 awesome 16:40:02 oh my 16:40:08 that post uses pascal syntax 16:40:15 -!- FireFly has joined. 16:40:19 for describing certain things 16:40:45 AnMaster: unsurprising 16:40:51 Mac OS's official applications language was Pascal 16:40:54 until later on 16:41:18 actually 16:41:26 ehird, they were different posts 16:41:29 not related 16:41:38 AnMaster: still could have been a macci 16:41:38 e 16:41:44 yes 16:41:50 i mean, in 1997 you could use one of: 16:41:52 - windows 95 16:42:03 - a buggy, hell on earth to configure linux or some sort 16:42:03 ehird, anyway it seems to be describe some kind of run-length compression of funge-space using "streets" 16:42:05 *of 16:42:11 - a stable, even more hell to configure bsd 16:42:22 - a quite buggy but easy mac os 16:42:24 err no 16:42:29 pretty shit choice 16:42:32 ehird, it was talking about Modula-2 too 16:42:38 heh 16:42:42 don't even remember what type of language that was 16:42:47 AnMaster: first OO 16:43:03 ehird, there was NT too 16:43:05 I'm pretty sure 16:43:07 hm wait 16:43:10 modula-3 was the first oo 16:43:17 modula-2 was niklaus's pascal successor 16:43:19 AnMaster: well, sure 16:43:23 ehird, aha 16:43:31 AnMaster: but that was just windows 95 that's buggy in different ways 16:43:35 and has a rubbisher interface 16:43:39 ehird, what about Solaris? 16:43:44 err 16:43:45 SunOS 16:43:46 back then 16:43:48 iirc 16:43:56 AnMaster: you can't afford a Sun box! 16:44:00 those things cost *thousands* 16:44:09 true 16:44:42 Deewiant, found 97 spec yet? 16:44:45 also, I think NT was quite hard to get vs 95 16:44:50 AnMaster: Haven't looked. 16:45:01 so yeah, in 1997 your best choice was probably a BSD or MacOS 16:45:11 the former if you hate yourself :-P 16:45:30 ehird: In 1997, many would still use DOS. 16:45:31 bsmntbombdood: how's the order 16:45:46 Deewiant: True, but probably not by choice... 16:45:56 Hm, wait. 16:46:01 I think I used DOS more often than Windows in those days. 16:46:06 You could buy a NeXT box in 1997, couldn't you? 16:46:18 Well, those cost multi thousands too. But if you could afford it, that'd be sweet nectar. 16:46:27 yeah, you could, NeXT started in late 80s 16:46:33 ehird, wouldn't it have been even more obscure than SunOS? 16:46:41 In 1997 I was a slackwareist, and it wasn't all that bad, really. 16:46:46 I didn't even break it very many times. 16:46:48 AnMaster: Hey, TBL used it to make the web. 16:46:54 And it was more known then than now. 16:47:03 It'd be kind of like using a BSD today, probably. 16:47:10 (on the desktop that is) 16:47:13 ehird, somehow "then than now" just sounds funny 16:47:22 at least if you say it without context 16:47:28 1996: After NeXT 16:47:28 Apple Computer announced an intention to acquire NeXT on December 20, 1996 16:47:42 BeGlad: short for Befunge Gladiators 16:47:44 The _asm_exec_ram memory range on TI-86 (which is the main place for both code and data for the currently running application) is 9000 bytes; that's rather luxurious, given that the 80x25-byte Befunge93 space only needs 2000. 16:47:46 ok, forget that 16:47:51 Deewiant, spec of it? 16:48:04 Was at http://www.cats-eye.com/funge/doc/97/beglad-97.html, according to Google. 16:48:29 Cell = Record 16:48:29 Command : Whatever 16:48:29 Next, Prev : ^Cell; 16:48:29 End; 16:48:36 * AnMaster tries to remember what ^ was in pascal 16:48:42 pointer? 16:48:42 I haven't coded pascal for *YEARS* 16:48:43 Pointer? 16:48:48 ehird, that might be it 16:49:01 It's more pointy than a *. 16:49:05 y'know, Apple were gonna buy BeOS instead of OS X etc. But the CEO said something like "I have Apple by the balls and I am going to squeeze", iirc, so uh that didn't work out :-) 16:49:07 fizzie, hah 16:49:21 may be apocryphal I think i heard it on reddit 16:49:48 Didn't that alternative C++ syntax (SPECS) adopt ^ for pointers because it's more pointy than the old *? 16:49:56 Yeah. 16:49:57 SPECS is nice. 16:51:00 I guess the rationale ("^ is more pointy") is not in the specification, though. At least explicitly. 16:51:11 :-D 16:51:11 Deewiant, err "cabbagepatch variable"? I'm pretty sure I saw it in the -96 spec, but I don't want to loose the location I'm at in the file atm 16:51:40 wow 16:51:50 AnMaster: Err, are you on a single-tasking OS or something? 16:51:55 I found the first trefunge mention 16:51:58 Deewiant: It's called emacs 16:52:05 I don't think you can open the same file twice in emacs 16:52:05 Deewiant, I don't know what that term means 16:52:06 iirc 16:52:18 AnMaster: whoosh! 16:52:26 -!- MigoMipo has joined. 16:52:39 No results found for "cabbagepatch variable". 16:52:40 meh 16:52:59 Did you mean: "cabbage patch variable" -> No results found for "cabbage patch variable". 16:53:10 -!- jix_ has joined. 16:53:13 -!- jix has quit (Read error: 54 (Connection reset by peer)). 16:53:13 oh you mean like that 16:53:40 right you can of course get different views of same file in emacs 16:53:49 right found what it meant 16:53:58 ehird, Deewiant: fact is however I was using less 16:54:02 so emacs is irrelevant here 16:54:15 I wasn't talking about emacs, I was talking about your OS 16:54:18 ^Zup enter /cabbage 16:54:23 Deewiant, more work 16:54:24 AFAIK less doesn't delete the file while it's viewing it 16:54:24 :P 16:54:32 Yeah, so you'd rather have me do it :-P 16:54:52 Deewiant, you forgot gnu less supports --self-destruct 16:55:04 so you don't have to eat the paper 16:55:05 ;P 16:55:38 Deewiant, I was hoping you would happen to remember it 16:55:59 I like the idea of an alcoholistic delta command. I can see it now: not 16:55:59 only does the actual number added to each command change, but the program 16:55:59 counter weaves around, occasionally hitting things and saying "ow" and 16:55:59 "sorry, policeman, I'm drot nunk and I spasn't weeding!" 16:56:21 spasn't weeding? I wasn't having spasms while on weed. 16:56:25 Right? 16:56:33 ok this is not trefunge 16:56:38 but 3D funge 16:56:48 with collision detection??? 16:58:04 Deewiant, you could add that to a fingerprint in 98 16:58:19 All of -96 could be a fingerprint in 98. 16:58:25 yeah 16:58:40 everything could be a fingerprint in 98 16:58:42 ooh yes this is fun 3D funge 16:58:54 Each pane can have a different instruction on each side. The Program Counter 16:58:54 would follow in a straight line along the surface of the pane. If a pane was 16:58:54 blocking its way in front of it, it would turn upwards and follow it. If there 16:58:54 was no pane to walk onto, it would 'fall' over the edge and either go downwards 16:58:54 round the corner, or flip upside-down onto the other side of the pane. 16:58:57 :D 16:59:10 "To help you visualise this, imagine a rubik's cube with Befunge instructions drawn on the edges." 17:00:01 -!- oerjan has joined. 17:00:52 oerjan, hi. I don't get IWC today. Care to provide "IWC explained"? 17:01:08 once i've read it 17:03:10 He's worried about Death of Being Stared At Angrily By A Giant Frog 17:03:26 wow so he didn't disappear for good after all 17:03:45 Deewiant, yes. But it seems to be referencing something from popular culture 17:03:49 I just have that feeling 17:03:55 but I have no idea what it is referencing 17:03:56 Your feelings suck 17:04:22 Deewiant: Head Paradox, too, i expect 17:04:27 -!- asiekierka has quit. 17:04:34 heh 17:06:07 -!- oerjan has quit ("Reboot"). 17:10:18 -!- oerjan has joined. 17:10:52 Deewiant, seems asiekierka disliked your interpreter btw? 17:11:07 asie is a fickle person. 17:11:08 If he can't run it I'm not surprised :-P 17:11:39 Deewiant, you could run it even on a sparc I think. Just use qemu's support for emulating other architectures. 17:12:07 assuming qemu works on non-x86 (which iirc it does) 17:12:18 Of course you can run it on anything if you emulate, that's the whole point of an emulator. But then you're not really running it on the environment the emulator is running in. 17:13:32 I think I'll write my own browser. Why? Indeed, why. 17:14:02 Befunge is NOT good for: 17:14:03 Writing an OS. 17:14:03 Writing a reasonably-sized program. 17:14:03 Speed-sensitive applications. 17:14:05 Just about anything. 17:14:13 Existing. 17:14:21 AnMaster: i don't think there is a specific reference anyway, diaries are everywhere in fiction 17:14:30 Befunge-96 is a bizarre, half-formed, misfeatured, brain-damaged 17:14:30 standard that appears to have been written at 3:00 in the morning. The 17:14:30 standard itself was written by Chris Pressey, and appears on his site 17:14:30 (see question 5). Do not write in it! Do not bother to implement it! 17:14:35 oerjan, right. But it seemed more than that 17:14:51 AnMaster: it's probably a trope 17:14:57 btw, do any of you write a diary in real life? Or know someone who does. 17:15:15 nope 17:15:15 Given that it took 4 months for people to completely dismiss -96... I'm not going to implement it :-P 17:15:23 it seems rather uncommon outside fiction... 17:15:28 AnMaster: Blogs. 17:15:39 Have pretty much replaced diaries for TEH NU GEN. 17:15:45 AnMaster: Anne Frank made one, i hear 17:15:56 ehird, that is less private. a diary is not public like a blong 17:15:58 oerjan: pfft, like a dirty jew could be clever enough to write 17:15:59 pull the other one 17:15:59 blog* 17:16:04 they couldn't even figure out vowels 17:16:16 though they do have a similar use 17:16:20 AnMaster: kids are less private. there are probably blogs with details that would be considered only suitable for a diary years ago 17:16:45 hm... that's probably true 17:19:20 #if __BORLANDC__ 17:19:31 Ah, the memories. 17:20:08 C compilers back then were a pile of incompatibilities 17:20:16 I 17:20:16 tried PCC (a shareware compiler), but then I realized it is 17:20:16 missing STRING.H. I tried Microsoft Quick C, but it does not 17:20:17 seem to support srand() and rand() commands. 17:20:25 PCC was shareware? 17:21:12 Not the pcc you're thinking of. 17:21:27 http://www.desmet-c.com/ 17:21:36 lol@not supporting rand 17:22:01 I support Ayn Rand all the way 17:22:50 Deewiant, so there is some stuff discussing 97 but where is the 97 spec? 17:22:58 not on the list as far as I can find 17:22:58 * ehird stabs Slereah. 17:22:59 Hard. 17:23:08 Why do you ask me? If it's not there it's not there. 17:23:25 Deewiant, maybe you found it? 17:24:12 FUNGINOMY: The practice of fully naming fungoids (Cartitoribefungoid is the 17:24:13 full name for Befunge-93, and it looks like Befunge-97 will be 17:24:13 Cartispheribefungoid. Both implementations are Cartefungoids - Cartesian 17:24:13 grid space.) 17:24:13 FUNGINOMINOMY: The process of coming up with words like "funginomy." 17:24:42 "well, this is really sick, but I had a thought.. 17:24:42 it could be a form of encapsulation, making sure that any PC that enters that 17:24:42 block of code has to do it through the right port (which will global-shift 17:24:42 it correctly) or get fucked up. Otherwise someone could hit it slightly off 17:24:42 and end up skipping vital steps of your process." 17:24:46 I'm not sure of the context 17:24:47 yet 17:24:54 but this sounds extremely weird 17:25:05 Cartitoribefungoid? 17:25:08 That's an awesome name for -93 17:25:20 What's -98? :P 17:25:24 Laheybefungoid 17:25:29 :-D 17:25:38 That's not very funge-soundin 17:25:38 g 17:25:45 Maybe cartilaheybefungoid 17:25:54 :D 17:25:55 Since I guess it is Cartesian. 17:27:11 actually 17:27:14 AnMaster: Grep for `Here's the list of "what it all means."`, list of very short names/descriptions for Befunge-97 instructions at the time 17:27:17 it's a shark programming language 17:27:32 I think this might be the start of the idea of storage offset 17:27:38 I'm not 100% sure yet 17:27:40 80,25-Cartefinitoribefunge-93. 17:28:16 AnMaster: Storage offset-related ideas were hashed quite early there, I think. 17:28:35 Somebody said that a new thread should access relative to where it was born 17:28:49 Deewiant, any idea of what the thread where you can find this in: 17:28:50 Can we leave it for "Befunge-97" or perhaps "Befunge-97: The Director's Cut" 17:28:50 ("The Truly Twisted Compile?") :-) 17:28:54 17:29:01 so what is that thread about. I can't figure it out 17:29:19 >Goddess, that's sick. But I think it's sufficiently befunged. 17:29:23 Is the context 17:29:25 Deewiant, yes 17:29:31 You can work your way back can't you? 17:29:42 Deewiant, I tried but I can't find the start of that thread 17:30:10 Well, if befunge is to be multiple-PC'ed, you could have the "global" 17:30:10 attributes affect only particular PC's (eg, have them as fields in 17:30:10 PC structs). That way, a "global increment" would only affect a particular 17:30:10 PC. 17:30:33 can't you lot import it into a mail client? 17:30:34 it's an mbox... 17:30:41 -!- tombom has joined. 17:30:58 ehird, hm I guess I could try to open it in mutt 17:31:21 -!- pikhq has joined. 17:31:26 except this mutt is set to work on maildir style mailboxes by default 17:31:46 * AnMaster reads the docs 17:33:03 Hahaha, a Java Befunge-93 interpreter which uses strings as indices into a hash table for infinite space 17:33:10 /*NEED AN OBJECT FOR A KEY ANYWAYS..WHY NOT A STRING?*/ 17:33:19 return new String(String.valueOf(_x) +","+ String.valueOf(_y)); 17:33:24 NEED UPPERCASE COMMENTS..WHY NOT USE TWO-DOT ELLIPSES? 17:33:27 Maybe because it's a bad idea? :-D 17:33:42 I love the way he creates 6 strings in one line 17:33:48 In order to index a hash table 17:34:20 ehird: From the mouth of Christopher Lahey himself: cartinfinilahebefunge 17:34:28 Deewiant: Beautiful. 17:34:36 http://www.christopherlahey.com/ ey 17:34:46 "ASP.NET Development" 17:34:47 ":-( 17:34:53 Christopher James Lahey 17:35:10 ah, googling suggests it's not the same person 17:35:28 Interesting, his name is pronounced /lehi/ 17:35:36 ! 17:35:38 I wonder where it's from 17:35:38 Laheyyyyyyyyy. 17:35:39 :-( 17:35:40 :P 17:35:44 Stress on the first syllable, too. 17:35:52 Hawaii, maybe? 17:35:56 Or thereabouts 17:36:06 Hawai'i, y'mean? 17:36:34 No, Hawaiʻi 17:36:57 But anyway: "As prescribed in the Hawaii Admission Act that granted Hawaiian statehood, the federal government recognizes Hawaii to be the official state name." 17:37:11 Off to sauna -> 17:37:42 crazy Finns ;) 17:37:44 So the glottal stop is implicit. Got it. :p 17:38:06 Actually, before that, pasting this: 17:38:07 Another thought on Quantefungoids (Quantum Mechanics): 17:38:07 How about T is like a subatomic particle decay? That is, the PC is split 17:38:07 into two PC's (or ip's if that's your cup of tea), one going (relative) 17:38:07 left, one going right. All delta-commands either PC further encounters are 17:38:10 listened to IN MIRROR SYMMERTRY BY BOTH PC's. Yes, Quantefunge could be 17:38:12 the first programming language ever to implement spooky action at a 17:38:15 distance!!! 17:38:37 From: "Chris Pressey" 17:38:38 X-Mailer: Microsoft Internet Mail 4.70.1155 17:38:40 oh god 17:38:47 :-D 17:39:28 Deewiant, hah indeed 17:39:36 Watercooling uses such creepy terms. "Fill and bleed". 17:40:05 Anybody written a program in ShaFuck yet? :P 17:40:38 * oerjan swats GregorR -----### 17:40:52 GregorR, what is ShaFuck? 17:40:54 Hey, don't get mad at me for YOUR incompetence. 17:40:55 GregorR: describe? 17:40:59 http://www.esolangs.org/wiki/ShaFuck 17:41:03 find sha1(x)=y(bf prog) 17:41:05 program is x? 17:41:11 when y is interpreted as binary 17:41:13 data 17:41:23 yep 17:41:41 But I wrote an interpreter and everything! 17:41:48 And nobody's written any code! :( 17:41:51 (Not even me :P ) 17:42:23 the syntax is a little awkward to get just right, you see 17:42:30 GregorR, err does it mean interpreter tries to calculate the corresponding program for a sha1 sum? 17:42:57 AnMaster: If by "calculate" you mean "just run", then yes. 17:43:47 GregorR, I meant finding a string corresponding to the sum 17:44:05 AnMaster: your 1024 byte program is a string 17:44:06 now 17:44:08 the interpreter calculates 17:44:11 sha1(prog) = bf 17:44:14 aha 17:44:15 and runs bf as brainfuck, comments not allowed 17:44:17 Yeah, there's no math to that. You take the sum, and interpret it as a string. 17:44:25 in case you haven't realized yet, it's a bit hard to write a program for 17:44:26 right 17:44:31 ehird, yes that way it is 17:44:41 If by "hard" you mean "quite literally impossible within the lifespan of the universe" :P 17:44:41 I just didn't figure out which direction it was done 17:44:47 GregorR: http://codu.org/eso/ this page is funny 17:45:09 It used to work great, but I can never align it right now >_< 17:45:25 it looks like a geocities page 17:45:42 So does your FACE. 17:45:45 It's 2L :P 17:46:17 GregorR, why are you using two layers, it would be easier to just put it inline 17:47:00 It's a different color. It would actually be a fair PITA to put it inline. Plus that makes copy/paste useless. 17:47:31 GregorR, you need an extra newline before FYB, and then one before ORK 17:47:38 and several before Glass 17:47:58 other than that it looks fine in konqueror 17:48:15 AnMaster: Yeah, the problem is that certain browsers were displaying the boldness weird. I'm fixing it now to just not use bold (why did I not do that in the first place? :P ) 17:48:36 * ehird configures the most expensive Mac possible: $87,247. 17:48:38 Free shipping! 17:48:47 There, fixed. 17:48:51 ehird: Lawl 17:48:57 ehird: That'd better come with free blowjobs. 17:49:20 GregorR, ah you added the missing newlines 17:49:26 Admittedly, it's an Xserve with $49,995 "Alliance" Apple support, a service part kit, a $18,999 Promise VTrak E-Class 16x 450GB SAS RAID subsystem (7.2TB), an unlimited managed systems remote desktop kit, 17:49:28 bold worked just fine though 17:49:29 all the display adaptors, 17:49:31 TWO power supplies, 17:49:31 as far as I could see 17:49:38 a dual-channel 4gb fibre card 17:49:41 and a quad-channel of the same 17:49:43 AnMaster: No, the problem is that /some browsers/ made bold lines take up more space, so things got wonky. 17:49:51 3 1TB 7.2k rpm ones with a raid card 17:49:52 a 128GB ssd 17:49:54 GregorR, those are buggy IMO :P 17:49:59 24GB (6x4GB) ddr3 ecc ram 17:50:00 Yarly X_X 17:50:08 and 2 x 2.93ghz quad core Nehalem xeons 17:50:18 All for the LOW LOW LOW price of $87,247! 17:51:05 It's practically free. 17:51:43 Minus all that price :P 17:51:48 Yeah. 17:51:52 Just a bit of price from free. 17:54:08 "This Auction is for 3 x Apple power Mac G3 400Mhz, 128 Ram" for £1.04 on eBay. 17:54:16 Damn selling shit on ebay is stupid :-P 17:55:08 (I was trying to find the most expensive mac possible and the least expensive mac possible) 17:55:32 $1.56 to $87,247. Apple's covered all the market. 17:56:14 -!- impomatic has joined. 17:58:12 -!- BeholdMyGlory has joined. 17:59:16 "Best quality organic fertilizer, bovine based" 18:00:21 or "dung" 18:00:34 Slereah: sssh 18:00:58 ehird: Oh, you can buy far more expensive computers than that. 18:01:08 GregorR: Not macs 18:01:21 ehird: My point being that they haven't "covered all the market" 18:01:22 That is the singlemost expensive computer you can buy from apple.com 18:01:27 True. 18:01:40 GregorR: They've covered about 99% of it, then, at least :P 18:01:51 Have some info about CoreLife now, the second 2 dimensional programming language: http://corewar.co.uk/corelife/ 18:02:02 I'm not sure supercomputers like the RoadRunner can count as (a) one computer (b) a market of any kind 18:02:03 Predates Befunge by a few months 18:02:10 Somebody should reimplement that in JS. 18:04:39 CoreLife in JS, that'd be great :-) 18:05:18 Just submitted it to reddit to see what kind of comments it gets. 18:06:12 impomatic: http://esoteric.voxelperfect.net/wiki/Biota 18:06:24 also, http://esoteric.voxelperfect.net/wiki/Timeline_of_esoteric_programming_languages 18:06:38 he knows 18:06:45 Power Mac G3 for $1.04? Hot damn. 18:06:49 Sorry, 3x. 18:06:54 pikhq: Three of them! 18:06:59 * pikhq needs to upgrade his PII router. 18:07:01 :p 18:07:01 pikhq: And yeah, it only has one bid, but 22 hours to go 18:07:05 I suspect one of 18:07:13 (a) seller refuses to give 18:07:17 (b) price shoots up 18:07:19 (c) miracle happens 18:07:47 -!- puzzlet_ has joined. 18:07:48 Hmm. Allow me to try to make an absurdly expensive PC... 18:07:55 Sorting ebay sections by "Price + P&P: lowest first" is a nice way to prey on people's innate disability to recognize what things are worth 18:08:21 * pikhq wants 4^2 cores 18:08:21 "2x MAC OSX CDROMS V10.2" 18:08:23 TWO of them! WOW! 18:08:49 pikhq: With 2 x Nehalems, just enable hyperthreading = 16 threads at once 18:09:03 You definitely want to work in multiples of Xeon Nehalem 3.2ghzs 18:09:05 I said cores. 18:09:09 Not threads. 18:09:11 ;) 18:09:12 $1.6k per CPU 18:09:22 and that's from newegg 18:09:24 Well, yeah. If we're talking expensive, we're going with Intel. 18:09:29 and xeon 18:09:36 (unless AMD decided to make chips out of gold?) 18:09:37 the i7 of same speed is just $1k 18:09:38 well i fell asleep too early and wake up right now (2:10 AM here). :| 18:09:40 so... 18:09:45 Deewiant: (as a reply that should be done 6 hours ago) oh well, you know tkbf93? :p 18:09:50 pikhq: http://www.newegg.com/Product/Product.aspx?Item=N82E16819117179 18:10:01 pikhq: Get 4 of them. 18:10:07 = 16 cores 18:10:13 lifthrasiir: I have a copy of it :-P 18:10:15 I'll get more if there's a motherboard that can handle it. 18:10:24 pikhq: Also, for each one, get a Megahalems from Prolimatech: http://www.prolimatech.com/products/megahalems.html 18:10:32 lifthrasiir: And it segfaulted on chess.bef :-P 18:10:32 pikhq: It's not on newegg, but it costs about $80 iirc 18:10:33 LMAO 18:10:39 Deewiant: yes, that sucks :p 18:10:46 pikhq: So yeah, pretty expensive heatsink 18:10:53 maybe that was my first attempt to golf 18:11:01 $320 for 16 cores and you still need to buy fans :-) 18:11:07 lifthrasiir: Is it 64-bit clean? I didn't try 32-bit, maybe that's the reason? 18:11:22 pikhq: I'll help you find some megaexpensive ram, ey? 18:11:27 Deewiant: i cannot remember all the detail now. 18:11:29 Yeah. 18:11:35 pikhq: /msg 18:13:17 -!- tombom has quit ("Peace and Protection 4.22.2"). 18:13:44 lifthrasiir: Looks like it uses just int so it should be safe 18:13:54 Deewiant: hmm it seems to have growable (singly-linked) stack at least 18:14:24 lifthrasiir: Ah, I was wondering what Z was :-) 18:14:55 -!- jix_ has quit (Read error: 110 (Connection timed out)). 18:15:05 Since you're closing the file only when you exit you might as well have made that exit(0)... 18:15:36 Deewiant: anyway that was a holy crap. i could write better thing now... in terms of length of program. ;) 18:15:41 :-D 18:15:56 i see lots of rooms of improvement now 18:17:06 Deewiant: if you want to see more horrible thing, see http://hg.mearie.org/angolmois/file/angolmois-1.0/angolmois.c 18:18:07 lifthrasiir: 18:18:08 00manifest.i@angolmois-1.0: no match found 18:18:14 oh well 18:18:27 try this: http://hg.mearie.org/angolmois/file/22549e0cee7a/angolmois.c 18:18:28 http://hg.mearie.org/angolmois/file/a49bf45b49ef/angolmois.c 18:18:36 Ah, 1.0 18:18:37 :-) 18:18:46 more recent de-obfuscated version is quite readable. but 1.0 was horrible. 18:18:55 [CRC: 72fb6a0c] 18:18:56 :-D 18:19:27 AnMaster: the name esotope is a portmanteau of esoteric and isotope, popped from my head someday. 18:20:04 and DSL idea is good, but that could need heavy canonicalization that is yet not good in current phase. anyway thanks for suggestion. 18:20:10 -!- puzzlet has quit (Read error: 110 (Connection timed out)). 18:20:51 lifthrasiir, I got the idea from the optimiser DSL "oil" found in ick 18:23:04 Most "standard" computer 18:23:04 languages (like, C and Perl) really are Unefungoids. 18:24:05 Deewiant: with "x" set-delta commands? :p 18:24:21 More like j 18:24:29 ah yes 18:24:30 AnMaster: Befunge-97 spec is there 18:24:40 AnMaster: "Edit #2 (rough) - Jun 5 Chris Pressey" 18:24:59 yay 18:25:06 err 18:25:07 As I thought, directives contain = 18:25:09 Deewiant, same file? 18:25:13 Yes. 18:25:24 ah 18:25:37 I knew I wasn't thinking just about Mini-Funge :-P 18:26:22 ok, don't have time to read it right away, going afk for a bit. I guess I will put them cut out specs (and the original file) up somewhere, with permission from the person you got the file from 18:26:29 Some nice multithreaded instructions there 18:26:54 Funky blocking is provided by Y "Merge Threads". 18:26:58 Funky blocking :-D 18:27:00 Deewiant: back to the angolmois, it has an embedded font; it parses game data file itself and generates all the surfaces on the fly. i wasted my time a lot 5 years ago, and never want to do such thing more XD 18:27:08 :-D 18:27:09 Hmmm... time to go, it's a quiz night :-( 18:27:19 -!- impomatic has left (?). 18:27:29 In Befunge-97, any existing command (see Appendix A) can be redefined in 18:27:29 one of three ways: rebinding it to another existing Befunge command with B 18:27:29 "Bind Command", defining a routine with D "Define Routine", or defining it 18:27:29 to a runtime library call with L "Bind Library Call". 18:28:34 Library calls look like just implementation-dependant FFI 18:28:38 Heh, five different types of wrapping you can select from. 18:30:41 *Monefungoid: like a Unefungoid but only capable of going one direction 18:30:41 (the IP can't move backwards.) 6502 Assembly language is a monefungoid. 18:30:41 (I considered Polariziunefungoid but discarded it for obvious reasons.) 18:30:42 -!- tombom has joined. 18:32:06 AnMaster: FWIW that's not the latest version, he updates it later but only links to his now-defunct web site. 18:42:14 Hah, queuemode got added to Befunge-97 18:43:17 -!- oerjan has quit ("leaving"). 18:50:31 Deewiant: Me and pikhq's megarig is currently at $22,790.56 for CPUs, mobo, RAM, CPU coolers, thermal compound and graphics cards 18:50:34 *mega-rig 18:50:38 We're nowhere near done :-) 18:50:51 Is that mostly RAM? 18:51:00 Deewiant: $6,239 of RAM. 18:51:01 128GB. 18:51:06 ... only DDR2, though, but... 18:51:14 8 CPUs of 4 cores each. 18:51:15 But it's 128GB. 18:51:23 We had to buy a $553 expansion card for the mobo to support more CPUs and RAM 18:51:24 Where'd you find a mobo with 8 sockets? O_o 18:51:27 The RAM is in 32 sticks 18:51:30 Deewiant: Nope, 4 sockets 18:51:34 And an expansion card using two pci x16 18:51:37 Ah. 18:51:39 That gives more RAM and CPU sockets 18:51:41 Xeon? 18:51:43 As hypertransport slots. 18:51:44 Nope 18:51:46 AMD's more expensive 18:51:47 Opteron. 18:51:49 :-D 18:51:50 Because Intel only do dual motherboards 18:51:52 -!- MizardX has quit ("back in 30 minutes or so"). 18:52:20 Most Expensive CPUs As Power Goes Upwards: Intel AMD 18:52:37 :-P 18:54:17 Deewiant: we just spend $141.96 on RAM fans. 18:54:26 Four of them, to cool our 32 RAM sticks. 18:54:37 :-) 18:54:47 How many decibels do you estimate that thing is at? 18:54:49 400? 18:55:04 No, but I bet it's going to hit 100. 18:55:05 Deewiant: So far, probably about 60dB or thereabouts 18:55:17 (the air can only handle 180dB or so) 18:55:28 That's 8 CPU fans, 4 RAM fans, and 2 stock graphics card fans 18:55:35 Apparently the RAM fans are horribly noisy from a review 18:55:41 Deewiant: We have 8 of these: http://www.newegg.com/Product/Product.aspx?Item=N82E16835103056 18:55:42 :D 18:55:48 I'm not sure they'd fit, but we can just mod them if not, right? 18:55:54 Stock GPU fans? Go custom 18:55:59 Deewiant: Can't 18:56:05 Whaddya mean can't 18:56:09 "Money can buy". 18:56:09 Deewiant: It's a GTX 285 18:56:12 It's only been out since december 18:56:17 So what 18:56:19 And also, VGA coolers top out at around $50 on newegg 18:56:22 Just mod them 18:56:22 Deewiant: nothing supports it 18:56:23 I looked 18:56:26 You don't need "support" 18:56:33 My cooler supports up to the X1300 or so 18:56:33 True enough. 18:56:36 And we don't care about $50. 18:56:39 But yeah./ 18:56:42 $50 isn't worth it. 18:56:43 It fit just fine, after removing a bit of the heatsink 18:56:45 There are bigger fish. 18:56:46 We're spending more on SATA cables. 18:56:47 ;) 18:56:53 And stock GPU coolers are really fucking loud anyway ;-) 18:57:08 ehird, who are these "we"? 18:57:09 (I assume we'll max out the onboard SATA and the RAID card, right?) 18:57:14 ehird and I. 18:57:18 AnMaster: Me and pikhq trying to build the most expensive computer, ever. 18:57:23 We're spending $6k just on RAM. 18:57:23 ah 18:57:27 ouch! 18:57:37 128GB. 18:57:38 OK, so pikhq, we have processors, motherboard, cpu expansion card, RAM, ram fans, CPU fans/heatsinks, thermal compound and video cards. 18:57:52 I think, I think we should go buy some expensive harddrives 18:57:55 I'll look at hard drives, you look at RAID cards? 18:57:56 IODrives 18:58:01 ehird, not the loudest one? 18:58:04 I'll look at hard drives 18:58:05 For RAID cards I recommend Areca 18:58:15 Deewiant: We don't have spare PCI slots any more. 18:58:17 AnMaster: comes with the deal 18:58:17 hm 18:58:18 I think they're on the high end of things 18:58:20 Darn 18:58:22 pikhq: PCI expansion cards? 18:58:24 pikhq: THe ramsans go into pci 18:58:25 ehird, AND just do IBM RoadRunner but with twice as many CPUs. That will be more expensive I bet 18:58:31 AnMaster: That's not a computer 18:58:34 That's a cluster 18:58:42 Anyway 18:58:46 Our 8 CPUs are using two PCIe, the GPUs are using two more, and we have another PCI slot for the RAID. 18:58:47 I'll find the most expensive SATA harddrives 18:58:47 true 18:58:58 Probably an SSD, mayb 18:58:58 e 18:59:08 ehird, SSD with ram cache? 18:59:16 LOL WAT 18:59:18 pikhq: 18:59:19 Or some absurdly speedy 1.5TB drive. 18:59:19 147GB hd 18:59:22 15k RPM 18:59:25 LOL WAT? 18:59:27 *SCSI* 18:59:28 $359.99 18:59:32 * pikhq <3 18:59:35 The most expensive single harddrive on newegg 18:59:38 Well, internal. 18:59:39 haha 18:59:40 Is it SAS? 18:59:40 LEt's look at the... 18:59:41 SOLID 18:59:42 STATE 18:59:44 DISKS 18:59:46 a 18:59:48 k 18:59:50 a 18:59:52 EXPENSIVE 18:59:52 ... 18:59:54 DISKS 18:59:56 pikhq: $799 18:59:58 Intel X25-E 19:00:00 pikhq: Something like 64GB 19:00:05 yep 19:00:06 64GB 19:00:10 pikhq: How many sata slots do we have? 19:00:12 why 19:00:13 so 19:00:14 many 19:00:17 n 19:00:18 e 19:00:18 w 19:00:19 l 19:00:20 i 19:00:20 AnMaster: Because this is epic. 19:00:21 n 19:00:21 8 on the mobo, dunno on the RAID. 19:00:21 e 19:00:23 s 19:00:26 AnMaster: fuck off 19:00:27 ? 19:00:38 pikhq: OK, so $6,392 of SSD madness. 19:00:49 pikhq: A whopping 512GB of storage. 19:01:01 Get an SATA RAID card and stick a 2TB drive on there. 19:01:08 Ah hm 19:01:09 Or something. 19:01:10 d 19:01:10 i 19:01:10 a 19:01:10 g 19:01:10 o 19:01:10 n 19:01:12 a 19:01:14 l 19:01:16 (For best viewing experience please use a monospace font.) 19:01:18 * AnMaster runs 19:01:25 pikhq: Maybe we should fill EVERY SLOT with a SATA RAID card. 19:01:29 And fill those cards with X25-Es 19:01:41 Y'mean, instead of the graphics card? 19:01:42 pikhq: y/n 19:01:46 I think I approve. 19:01:46 No no no 19:01:51 Oh. 19:01:53 Can't you get SATA raid things 19:01:53 As in 19:01:55 They go direct into SATA 19:02:02 *Oh*. 19:02:04 Hmm 19:02:10 You can right? 19:02:18 Ooh. 19:02:18 Dunno. 19:02:28 pikhq: the top single areca raid card is $1,599 19:02:31 takes up a pci x8 19:02:35 and I loathe to lose graphics cards 19:02:35 Spiffy. 19:02:39 But, um, we'll use onboard video, right? 19:02:44 Right. 19:02:49 So we want two of them. 19:02:53 Oh my god. 19:02:54 pikhq? 19:02:56 It has a fan. 19:02:59 The fucking RAID card. Has a fan. 19:03:00 :D 19:03:06 I approve. 19:03:08 This RAID card. Is the most powerful RAID card on the planet. 19:03:10 Gahs this library is very unfun to install. 19:03:16 # This file is part of SEP's set of GNU make rules for painless compilation 19:03:16 # This file overrides defaults that are incorrect for the SUN4 architecture 19:03:22 SYSLIB= /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/libg2c.a -lm 19:03:22 ehird: Interesting that your HDs are so cheap 19:03:33 That line is not making this very painless. 19:03:40 OK, pikhq, 19:03:42 I found some things that cost 2000 € 19:03:49 Internal Connectors6xSFF-8087 19:03:50 Maybe they're just rare 19:03:54 Are those the harddrive thingies 19:04:15 pikhq? 19:04:19 * pikhq Wikipedes 19:04:23 I wanna know how many HDs can go in one of these cards 19:04:29 We're still filling the sata ports with x25-E 19:04:29 s 19:04:34 and we'll fill these raids with them too 19:04:58 It's a SAS connector. 19:05:11 pikhq: So we can't connect SATA drives? 19:05:15 ... is there an adaptor? 19:05:19 We can buy 6 adaptors, you see. 19:05:21 ehird, so you are trying to make it as loud as well? 19:05:28 AnMaster: That's just a bonus. 19:05:28 as loud as possible as well* 19:05:31 ah 19:05:35 Cost takes precedent. 19:05:37 *predecence 19:05:40 *precedence 19:05:42 ehird, how much would replacing this with passive cooling cost 19:05:46 or water cooling 19:05:46 SATA drives can be hooked to SAS controllers. 19:05:50 AnMaster: $infinity for passive cooling. 19:05:51 Just not the other way around. 19:05:58 $-something for water cooling 19:06:04 ehird, ok, go for replacing it all with water cooling then! 19:06:07 No 19:06:08 Minus. 19:06:10 It would cost less. 19:06:15 You can't beat $1,119 on fans and heatsinks for the CPUs. 19:06:16 pikhq: Alright. 19:06:22 SFF-8087 is something you can connect four SATA drives into, I think. 19:06:23 pikhq: So one card holds 6 drives? 19:06:23 ehird, sure? 19:06:32 pikhq: So we need 12 more drives. 19:06:51 Well, at least there exists a four-port SFF-8087 connector; I'm not sure if they all are like that. 19:06:52 pikhq: $15,980 on storage. 19:06:58 ehird, sure the power supply can handle it? 19:07:06 AnMaster: we'll have two 19:07:06 We can get a PCI RAID card, right? 19:07:07 or four 19:07:12 pikhq: these are pci 19:07:17 http://www.newegg.com/Product/Product.aspx?Item=N82E16816151053 19:07:20 We're getting two 19:07:21 Pay attention 19:07:23 And each does 6 drives 19:07:24 You said "PCI x8". Which is PCIe. 19:07:27 Ah. 19:07:32 ehird, mhm, you can't fit them inside the case then right? Unless you are using some weird server case with a backup power supply 19:07:33 Wait 19:07:35 Normal PCI. 19:07:39 != 19:07:40 PCIe. 19:07:42 but in those cases it is for power supply 19:07:46 That is PCIe. 19:07:48 pikhq: RAID cards + SSDs currently total to $19,179.98 19:07:54 pikhq: I'll find the most expensive regular PCI raid card 19:08:13 pikhq: I think we may need to invent a new kind of mathematics to calculate this price 19:08:18 Omegad. 19:08:25 pikhq: how much IDE shit does our mobo have? 19:08:27 None? Some? 19:08:32 ehird: Yes, if it is that "ARC-1680IX-24 PCIe x8 SAS RAID Card" card you can connect 24 drives in it. Four to each SAS connector. 19:08:35 And you can buy SATA←→IDE adaptors, right? 19:08:35 1 IDE connector. 19:08:39 fizzie: WOW. 19:08:40 REALLY? 19:08:43 ... Yes. 19:08:44 EVEN MORE! 19:08:45 Yes, you can. 19:08:55 56 SSDs. 19:08:57 fizzie: OMG. 19:08:58 That's $44,744 19:09:12 HAHAHA 19:09:14 what OS will you use? I bet window would be the most expensive OS to use on it 19:09:19 AnMaster: No way man 19:09:22 Think enterprise 19:09:25 But pikhq. 19:09:26 OMG. 19:09:28 56 fucking drives. 19:09:29 AnMaster: RHEL. 19:09:37 pikhq, RHEL is expensive? 19:09:39 I have no clue 19:09:39 ehird: :D 19:09:39 Okay, pikhq? Find the most expensive SATA to IDE cable, please. 19:09:45 pikhq: Then we'll add one more SSD in to it, you see. 19:09:47 AnMaster: Thousand or so a seat. 19:09:57 ehird: Two SSDs, you mean. 19:10:02 pikhq, using windows would be cheaper indeed... 19:10:03 pikhq: You said one IDE port. 19:10:04 One IDE handles two drives. 19:10:06 Ah. 19:10:07 Okay. 19:10:08 I didn't know. 19:10:12 OK, find two adaptors, pikhq? Or one. 19:10:13 ehird: "Charged" 19:10:13 I don't know. 19:10:16 Just find the adaptors we need. 19:10:17 bsmntbombdood: Nice. 19:10:25 bsmntbombdood: Will you buy THIS rig for us? 19:10:27 It's currently at... 19:10:45 at what 19:10:48 bsmntbombdood: $70,016.52 19:10:53 schweet 19:10:54 And no way are we done. 19:10:59 pikhq: Go find that adaptor, boy. 19:11:03 that's gotta be mostly disks? 19:11:03 I'll find a regular PCI raid card. 19:11:07 Searching. 19:11:11 bsmntbombdood: $44k of SSDs 19:11:13 but the ram is $6k 19:11:15 bsmntbombdood, we're spending some $6,000 on RAM. 19:11:25 the processors are $14k, bsmntbombdood 19:11:29 eight four-core opterons 19:11:37 ehird: Those RAID cards each have an external SAS slot as well. 19:11:42 *Enclosures*. 19:11:43 :) 19:11:47 pikhq: let's figure that out later 19:11:51 GET THAT ADAPTOR BITCH :P 19:11:59 * pikhq is looking 19:12:12 ehird, what CPU 19:12:15 Hey pikhq, we only have one PCI slot right? 19:12:18 19:11 ehird: eight four-core opterons , AnMaster 19:12:20 Yeah. 19:12:23 ah 19:12:35 ehird, sure you can't do more? 19:12:44 AnMaster: yes 19:12:46 we calculated 19:12:51 pikhq: Hmm. Not worth it. $269.99 is the most expensive raid thing for PCI 19:12:58 I'm sure we can find a better use 19:12:58 I mean the most expensive mobo might not be the one that allows the most expensive equipment! 19:13:03 Oh wait 19:13:03 Maybe. 19:13:04 pikhq: it is worth it 19:13:07 since it adds another ssd 19:13:09 what mobo? 19:13:12 ehird, so a cheaper mobo might result in a higher total cost 19:13:15 bsmntbombdood: http://www.newegg.com/Product/Product.aspx?Item=N82E16813151085 19:13:18 AnMaster: We know, goddamn. 19:13:20 We've played with permutations 19:13:23 AnMaster: This mobo supports the most CPUs. 19:13:23 ah 19:13:25 And the most RAM. 19:13:30 pikhq, most PCI cards? 19:13:33 ehird: unnacceptable 19:13:36 We've got 32 freaking RAM slots. 19:13:37 bsmntbombdood: what 19:13:49 pikhq: there are no PCI raid cards 19:13:49 sry 19:13:51 pikhq: found that adaptor? 19:13:51 -!- cherez has joined. 19:13:56 In a motherboard that supports more than one CPU slot. 19:13:57 ehird: Not yet. 19:14:03 -!- cherez has quit (Client Quit). 19:14:03 i thought you were using xeons ? 19:14:08 bsmntbombdood: turned out cheaper 19:14:08 XD 19:14:11 due to only being able to have 2 procs 19:14:27 don't go for money...go for performance without hinderance from money 19:14:31 No no no. 19:14:32 We're going for money. 19:14:34 That's the goal. 19:14:45 We can do performance tomorrow :P 19:14:46 lame 19:14:58 ehird, what form factor is that mobo O_o 19:15:05 It is also very high performance. 19:15:12 AnMaster: We have a PCIe expansion card for the 4 more CPUs 19:15:12 and opterons are NUMA 19:15:20 We've got 4 GB per core. 19:15:23 bsmntbombdood, NUMA > SMP 19:15:35 ehird: It's a hypertransport card. 19:15:46 pikhq: that's it? 19:15:50 AMD has a freaky way of doing Hypertransport via PCIe slots. 19:15:51 bsmntbombdood: dude 19:15:53 that's 128GB in total 19:15:56 bsmntbombdood: Largest DDR2 stick. 19:15:56 pikhq, wow 19:16:06 only ddr2 memory? 19:16:06 128GB. 19:16:09 bsmntbombdood: yes 19:16:11 it's most expensive 19:16:15 4GB ddr2 is highly expensive 19:16:18 bbiab 19:16:19 one stick 19:16:21 that's retarded 19:16:22 $194.99 19:16:29 God, bsmntbombdood, we're going for PRICE. 19:16:30 -!- Hiato has joined. 19:16:41 it's trivial to go for price 19:16:51 We've spent over an hour on it. 19:16:52 Yes it is. But we're doing it, you globmonkey. 19:16:56 pikhq: found that adaptor? 19:17:08 ehird: I'm looking for it. 19:17:15 Gah, I'll look 19:17:40 Argh, they're all the wrong way around 19:17:48 pikhq: hey hey 19:17:49 pikhq: we can go 19:17:52 sata → usb → ide 19:17:57 i'm pretty sure 19:18:07 pikhq: :)))) 19:18:12 ehird: buy a ramsan 19:18:15 Uh, no... 19:18:18 pikhq: yep 19:18:22 i found sata→usb 19:18:24 and usb→ide 19:18:25 We can go USB -> SATA, though... 19:18:29 *What*?!? 19:18:31 bsmntbombdood: "contact for price" 19:18:33 Makes no sense. 19:18:33 pikhq: yep 19:18:35 http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=ide+sata+adapter&x=0&y=0 19:18:38 scroll down a bit 19:18:41 Nippon Labs USB-SATA USB to IDE/SATA Adapter w/ power - Retail 19:18:43 is the final connection 19:18:47 and 19:18:47 SABRENT USB-DSC5 Serial ATA (SATA) or IDE 2.5" and 3.5" to USB 2.0 Cable Converter Adapter - Retail 19:18:48 wait what? 19:18:49 voila 19:18:56 SATA→USB→IDE 19:18:59 why are you even considering usb or pata for anything? 19:19:04 bsmntbombdood: shut up 19:19:07 we're getting x25-Es 19:19:08 slow as shit 19:19:08 and they're expensive 19:19:12 and we want two more drives 19:19:16 so we need to put IDE shit 19:19:16 How much was that x25-E? 19:19:18 but the ssds only do sata 19:19:21 pikhq: $700 or so a pop 19:19:25 get 32 x25-es and raid 10 them 19:19:26 And we can do two more of them with these adaptors 19:19:30 Hard to beat. 19:19:36 bsmntbombdood: STFU, we're getting 56 of them already 19:19:40 with IDE we can get58 19:19:41 lawl 19:19:51 pikhq: shall I add those adapters? 19:20:05 Just a sec. 19:20:21 I think I found the normal IDE->SATA adaptors. 19:20:29 err 19:20:30 pikhq: wrong way around 19:20:37 we have a sata drive and want to plug it into ide 19:20:41 Yes. 19:20:44 we can do this by going via usb 19:20:46 IDE->SATA. 19:20:51 oh 19:20:54 that's the way you do the diagram? 19:20:55 heh 19:21:03 pikhq: how much is it? 19:21:06 via USB may be more expensive 19:21:15 ... 19:21:16 -!- cherez has joined. 19:21:16 -!- cherez has quit (Remote closed the connection). 19:21:16 http://www.newegg.com/Product/Product.aspx?Item=N82E16812200373 ? 19:21:23 But that's just plain silly. 19:21:26 pikhq: ours is more expensive 19:21:29 $39.98 for each drive 19:21:34 I mean, more so than what we've been doing previously. 19:21:35 -!- cherez has joined. 19:21:40 Oh, fine. 19:21:44 -!- cherez has left (?). 19:22:11 you can get external sata->sata raid cards can't you? 19:22:14 http://www.newegg.com/Product/Product.aspx?Item=N82E16812200196 Also, this one is $26.99 per drive. 19:22:22 we now have $46,342 in SSDs 19:22:23 pikhq: nice. 19:22:33 back 19:22:39 Erm. That's just $26.99. Oh well. 19:22:46 pikhq: want the current total? 19:23:06 Processors, motherboard, CPU card, RAM, RAM fans, CPU fans/heatsinks, thermal compound, RAID cards, IDE to SATA adapters, and drives 19:23:07 http://www.newegg.com/Product/Product.aspx?Item=N82E16812200156 This is $30 per drive, though. 19:23:09 ehird: Sure. 19:23:19 hmm I'll add that then total 19:23:30 OK 19:23:31 here goes 19:24:03 WITH STORAGE: $72,674.41 19:24:07 WITHOUT STORAGE: $26,332.41 19:24:13 (that's JUST the ssds missing) 19:24:14 ehird: how do you define "computer"? 19:24:20 ehird, how many GB 19:24:30 AnMaster: 3712GB 19:24:30 bsmntbombdood: Single system. 19:24:32 so about 3TB 19:24:32 ;p 19:24:40 that's if you RAID 0, of course 19:24:43 pikhq: go on 19:24:45 64GB per disk and 58 disks 19:24:55 ehird, RAID 1 them. More *wasted* money then 19:25:02 ;P 19:25:17 pikhq: OK, I think we need some optical drives now? 19:25:24 (RAID 1 does make sense, just not for *THAT* many drives) 19:25:29 To burn and rip the family photos / music albums. 19:25:33 You know, on this family PC. 19:25:34 Aren't there those SATA multiplexers? 19:25:39 In 3000 AD. 19:25:39 ehird: Definitely. 19:25:43 Bluray burner? 19:25:47 also, I think maxing out storage any further sux 19:25:50 it's $44k already 19:25:51 http://www.newegg.com/Product/Product.aspx?Item=N82E16816102107 19:25:53 it's almost like cheating 19:25:53 what about floppy? 19:25:57 I assume you get at least two 19:26:06 bsmntbombdood: what does that buy us 19:26:15 it looks good, I just don't know what it does :-) 19:26:19 also, it looks a bit too external 19:26:36 ehird, does the mobo actually have onboard video? 19:26:37 AnMaster: We'll get the best fucking floppies in the universe. 19:26:38 -!- psygnisfive has quit (Remote closed the connection). 19:26:41 And yes. 19:26:43 It's a server board. 19:26:43 ah 19:26:54 $329.99 for the most expensive bluray burner 19:26:58 pikhq: how many can we fit? 19:26:59 Oh wait. 19:27:00 you can get 8 of those with your mobo, so 96 drives 19:27:03 We've maxed out the slots, haven't we pikhq? 19:27:04 http://www.newegg.com/Product/Product.aspx?Item=N82E16827248014 Most expensive internal burner. 19:27:09 Yeah. 19:27:11 No 19:27:12 $329.99 19:27:13 http://www.newegg.com/Product/ShowImage.aspx?CurImage=13-151-085-S02&ISList=13-151-085-S01%2c13-151-085-S02%2c13-151-085-S03%2c13-151-085-S04%2c13-151-085-S05&S7ImageFlag=1&Item=N82E16813151085&Depa=0&WaterMark=1&Description=TYAN%20S4985G3NR%20SSI%20MEB%20footprint%20Server%20Motherboard%20-%20Retail <--- *three* ethernet ports? 19:27:14 There's PCI SATA slots. 19:27:22 ehird, what about case 19:27:23 Mmkay. 19:27:23 AnMaster: Yeah. 19:27:26 pikhq: Just tell me how much free we have ;-) 19:27:33 1 PCI slot. 19:27:42 pikhq: OK, so only one burner. 19:27:46 Do we need any adapter shit? 19:27:48 The external SAS slots on the RAID cards. 19:27:55 Ah. 19:27:56 So 3. 19:28:05 989.97 on burners 19:28:05 Not bad 19:28:23 Each SAS slot is 4 SCSI slots. 19:28:25 Erm. 19:28:28 Ooh. 19:28:29 SATA. 19:28:37 ;) 19:28:39 So 9 burners? 19:28:45 err 19:28:49 Do we need an adapter for the PCI slot thing, pikhq? 19:28:53 The non-RAID one. 19:28:56 hopefully you are considering how to package this stuff 19:28:59 Yeah. Just a sec. 19:29:06 you do need some sort of case 19:29:09 ehird, pikhq: case, what about case? 19:29:09 bsmntbombdood: One absurdly massive case. 19:29:12 yeah 19:29:16 They make large enough cases. 19:29:16 biggest case in the universe 19:29:17 bsmntbombdood, yes I said that about a minute ago before 19:29:19 s/case/a full rack/ 19:29:23 haha 19:29:26 :) 19:29:29 just put it in a cardboard box 19:29:34 http://www.newegg.com/Product/Product.aspx?Item=N82E16816102107 19:29:36 use that 19:29:38 it's perfect 19:29:39 MADE OF GOLD. 19:29:54 pikhq: have you got the adapter? 19:30:09 pikhq: I kind of feel like we should cut back on storage, you can grow the price there pretty much unbounded 19:30:11 :P 19:30:12 Not yet; will soon. 19:31:02 actually, i think you could arrange that external raid controller in a tree 19:31:15 bsmntbombdood: see, that's the thing 19:31:19 you can have pretty much infinite storage 19:31:23 for pretty much infinite dollars 19:31:26 where's the challenge? 19:31:28 ... 19:31:31 that's what i said 19:31:31 it's just free ... anti-money 19:31:32 http://www.newegg.com/Product/Product.aspx?Item=N82E16816102080 19:31:41 other components, now that takes skill to waste money 19:31:45 more interesting if you have to consider performance 19:31:51 Technically, it's a RAID card, but it can be used for JBOD. 19:32:40 Time for the most expensive floppy drive 19:33:13 2U is pretty good for 12 drives isn't it? 19:33:40 pikhq: do you think we should cut back on the storage? 19:33:43 It really is cheating 19:33:59 ehird: Replace those with those 2TB SATA drives. 19:34:11 2TB drives aren't expensive 19:34:11 We can at least have really silly stats while cutting back. 19:34:12 ;) 19:34:19 "Cutting back". 19:34:20 I know. 19:34:23 pikhq: I think we should avoid spending much on storage 19:34:30 $44k just on drives doesn't take any skill 19:34:33 you can get more expensive hard drives than the x25-e anyway 19:34:41 also, priciest internal floppy is $7.99 so meh 19:34:45 OK, let's cut back on storage. 19:34:47 We still want the RAID cards. 19:34:50 I think. 19:35:01 So we'll go for the most expensive non-solid-state drive. 19:35:01 -!- MizardX has joined. 19:35:14 Hopefully that should only add a few thousand. 19:35:18 2TB SAS card I think. 19:35:34 4 per RAID card, and the onboard SATA wouldn't be used... 19:35:38 * ehird picks regular SATA drives. 19:35:41 http://www.newegg.com/Product/Product.aspx?Item=N82E16816102113 19:35:47 XD 19:35:51 STOP LINKING TO IT bsmntbombdood 19:35:55 Storage is cheating. :P 19:35:58 that's a different one! 19:36:12 pikhq: 2TB regular speed drive, or 300GB 10k RPM drive. The latter is a velociraptor. 19:36:15 I say the latter, because: 19:36:18 It's still big. 19:36:21 10k fucking RPM is loud. 19:36:21 And. 19:36:24 We'll need. 19:36:28 A fucking. Lot. 19:36:29 Of... 19:36:32 HARD DRIVE FANS. 19:36:34 pikhq: y/n 19:36:37 $229.99 a pop 19:36:53 That's $13,339.42 in storage. 19:37:12 pikhq: I think we'll need a fan per every 2 harddrives. We are putting these under load, right? 19:37:22 Yeah. 19:37:30 300GB 10k drive. 19:37:33 :D 19:37:57 $29.99 for a hard drive enclosure 19:38:00 But we want fans too 19:38:38 http://www.newegg.com/Product/Product.aspx?Item=N82E16822332012 19:38:43 pikhq: $22.99 for the most expensive HD cooler with a fan 19:38:46 i thought you could find an hdd more expensive than that 19:38:51 $29.99 for the most expensive full stop, so not a big loss 19:38:54 and it has a FAN! 19:38:58 And it's ADJUSTABLE! 19:39:17 pikhq: $1,333.42 on harddrive cooling. 19:39:20 Whoo... 19:39:35 http://www.newegg.com/Product/Product.aspx?Item=N82E16817122110 19:39:57 pikhq: I'm going to find the most expensive, highest-voltage power supplies. 19:39:58 Two of them. 19:39:59 Or more. 19:40:04 Actually, yeah, more. 19:40:10 pikhq: Uh... issue./ 19:40:16 58 harddrives. They need power. 19:40:22 How many can one power supply do? 19:40:23 :-P 19:40:28 We will need an awful lot of power supplies. 19:40:29 Don't forget UPSs 19:40:33 pikhq: Wanna cut back on storage? 19:40:37 As in, actually less drives? :-P 19:40:45 So we can actually get this thing to POST 19:41:00 Oh. My. God. 19:41:03 iStarUSA IS-2000R4H1UP 2000W Redundant 1U Server Power Supply - Retail 19:41:03 lol, it's going to be hard to power 19:41:06 $1,285.99 19:41:14 You were saying? 19:41:25 pikhq: ... even so... 19:41:30 We need a bunch to power 58 drives. 19:41:33 As in, it won't fit in a case. 19:41:34 I guess we're getting the 10U rack case? :p 19:41:39 Fuck, I mean, 58 drives won't fit in a fucking case. 19:41:40 But. Um/ 19:41:41 you are going to have trouble getting the psu physically close enough to the drives 19:41:46 bsmntbombdood: Exactly! 19:41:51 We need less drives, pikhq. 19:42:00 you need one of the rackmount drive bays! 19:42:07 We need less drives too. 19:42:08 it's the only practical way to do it 19:42:12 http://www.newegg.com/Product/Product.aspx?Item=N82E16817147073 19:42:17 no, you don't need to cut down on drives 19:42:18 This thing doesn't seem to have many ports 19:42:26 bsmntbombdood: Bah. 19:42:31 bsmntbombdood: Give us a rackmount drive bay then. 19:42:35 None of that tricky RAID stuff. 19:42:37 We've handled that. 19:42:59 ehird: I'm searching for cases. 19:43:01 how much did you spend on it? 19:43:03 pikhq: noo! 19:43:04 i want to :-P 19:43:11 bsmntbombdood: $3,199 in RAID cards. 19:43:16 Well, $139.99 extra 19:43:20 For the bluray adapter RAID card thing 19:43:28 ehird: those raid bays are $4,500 each 19:43:29 pikhq: Srsly, I can find a case. :-P 19:43:34 and you need several of them 19:43:35 so yeah 19:43:41 more expensive 19:43:47 bsmntbombdood: ... hmm. We can do that later. 19:43:50 Huh. There's *actually* 9U cases. 19:44:10 One. 19:44:11 ehird, total cost so far? 19:44:15 and what about the case 19:44:16 AnMaster: $thousands 19:44:18 http://www.newegg.com/Product/Product.aspx?Item=N82E16811165161 19:44:19 Our case. 19:44:27 Does it come with case fans? 19:44:29 We need case fans. 19:44:40 ehird, sure the case is large enough? 19:44:46 for all those 58 harddrives 19:44:49 AnMaster: We're putting the drives in external RAID rack mounts. 19:44:53 ah 19:44:59 Apart from the 10 for the SATA/IDE. 19:45:03 ehird, that isn't "one unit" thing for one computer then 19:45:05 :P 19:45:07 I'm sure we can fit 10 + an adapter. 19:45:08 AnMaster: stfu 19:45:13 shouldn't it be a rackmount case? 19:45:19 bsmntbombdood: hell no! 19:45:21 4-post racks are expensive 19:45:22 think of how heavy this will be 19:46:21 ehird, the case is too small I think. And how would you mount the mobo in it 19:46:29 It's not too small! 19:46:42 A 4U rackmount case alone can handle 24 disks. 19:46:48 ehird, checked dimensions? 19:46:49 It's just 8 CPUs, some RAM, some RAM fans, some gigantic CPU fans and heatsinks and 10 drives. 19:46:59 And they hotswap. 19:46:59 AnMaster: 35" x 24" x 23" 19:47:02 AnMaster: Bitch. 19:47:04 ehird, and the board? 19:47:12 Dude. 19:47:15 How big do you think mobos are? 19:47:18 damn, the biggest rackmount cases newegg has are 4u 19:47:24 ehird, I don't know in inches 19:47:24 It's 16" x 13" 19:47:34 cm makes more sense 19:47:36 AnMaster: it's 19:47:48 AnMaster: the case is 88cm x 60cm x 58cm 19:48:05 ehird, I just can't estimate distances in inches 19:48:10 so that is why I asked 19:48:17 AnMaster: Yeah. But. That's big, okay? :P 19:48:24 yes it is 19:48:33 88 cm wide? deep? high? 19:48:41 ... 19:48:55 that case is too cheep 19:48:57 http://www.newegg.com/Product/Product.aspx?Item=N82E16811152124 19:49:01 even that is more expensive 19:49:01 http://www.newegg.com/Product/Product.aspx?Item=N82E16811165161 says "Dimensions 35.44" x 24.22" x 23.63""? 19:49:03 err 19:49:07 bsmntbombdood: We need something we can fit shit in. 19:49:07 hm 19:49:08 -!- oklopol has joined. 19:49:08 right 19:49:09 -!- Gracenotes has quit (Remote closed the connection). 19:49:16 I misread that 19:49:50 lol, that 9u case is _insulated_ 19:49:51 lofl 19:49:55 haha 19:50:03 bsmntbombdood: it doesn't matter if our computer burns 19:50:08 as long as it gives the POST beep, we've won 19:50:22 no 19:50:25 that's retarded 19:50:30 no it's not 19:50:31 taht's our goal 19:50:55 -!- okloduk has joined. 19:53:27 -!- okloduk has quit (Read error: 54 (Connection reset by peer)). 19:53:51 -!- okloduk has joined. 19:54:26 I think we're done here. 19:54:38 Well, yeah. We could go further but I've lost interest :P 19:54:47 $80,000 is enough. :p 19:55:44 and there is absolutely no practical use for this system 19:55:52 Tomorrow I think we should go for "most powerful gaming computer ever". Then post a sale for it on an overclocking forum... plus a markup of a few hundred dollars. 19:55:56 Watch the purchases roll in. 19:56:00 Be in the money. 19:56:19 LMAO 19:56:31 bsmntbombdood: Yes there is. 19:56:35 Stick Xen on it. 19:56:40 Simulate a cluster. 19:56:41 :p 19:56:46 lawl 19:56:47 nope 19:57:12 pikhq: cheaper to buy a cluster 19:57:18 True. 19:57:27 But significantly less impressive and noisy. 19:57:28 you wont' be able to utilize all those drives 19:57:45 bsmntbombdood, I beg to differ. 19:57:52 Blu-Ray disk *dumps*. 19:57:58 ;) 19:58:00 bsmntbombdood: Dude, it's just 16TB. 19:58:11 bsmntbombdood: You could get a bloody porn collection that big, easy. :P 19:58:14 Yes, bloody porn. 19:58:19 i don't mean the size 19:58:47 bsmntbombdood: watch 58 vids from the porn collection without buffering all at once. 19:58:48 duh 19:58:52 SPEED 19:59:02 ehird: Blu-ray vids. 19:59:07 Yes. 19:59:09 Blu-ray porn dumps. 19:59:30 Compile Gentoo in an hour. 19:59:32 you won't be able to get the full speed capable 19:59:44 with 4 processors and a single mobo 19:59:58 you need to split that up over multiple servers 19:59:59 8 chips. 20:00:11 And you underestimate how freaking speedy Hypertransport is. :p 20:00:12 I'd say the buses are the bottleneck 20:00:27 Though, yeah, at this point HT is actually a bottleneck. 20:00:35 (and that's actually damned impressive) 20:00:59 Hmm. 20:01:02 What are the drives attached to? 20:01:03 Do the Opterons do hyperthreading? 20:01:09 Deewiant: SATA, IDE, and various RAIDs. 20:01:10 No, hyperthreading is an Intel thing. 20:01:12 Two RAIDs, that is. 20:01:14 No. They do more CPUs. 20:01:34 ehird: What is "a RAID" and what is it attached to 20:01:35 I would guess you can't saturate the whole hypertransport bus with just two PCIe 8x cards. 20:01:44 A PCIe raid card, Deewiant. 20:01:46 Two of them. 20:01:51 Hmm. 20:01:54 That's 2*2000 MB/s in one direction. 20:02:05 Deewiant: around $3k in total I think 20:02:07 There's also the onboard SATA. 20:02:11 Which I doubt is in the northbridge. 20:02:45 How many drives per card? 20:03:04 Since they're SAS cards driving SATA cards... A lot. 20:03:15 16, I think? 20:03:16 Enough to saturate the PCI-E bus? 20:03:19 Deewiant: 24. 20:03:31 Deewiant: If I remember the model right. Six four-port interfaces. 20:03:43 Those PCIe cards are on a PCIe 4x slot. 20:03:50 (physically 16x). 20:03:57 Wasn't it 8x? 20:04:14 No, the motherboards only offer 4x on those two slots. 20:04:16 Well, I haven't been following very closely. 20:04:21 Oh, okay. 20:04:27 (the 16x slots are being used for hypertransport) 20:04:29 That's only a GB/s then. 20:04:40 24 drives will saturate that easily. 20:04:47 Well, then you definitely can't fill the whole hypertransport bus with disk activity. 20:05:08 Because the PCIe bus will be filled instead. 20:05:18 Yes. 20:05:23 :D 20:06:38 ./../include/mat_inc_d.H:13:22: error: iostream.h: No such file or directory 20:06:38 This is some annoyingly old code, I guess. Does removing "-pedantic" let it even a modern GCC find the silly " automagically imports std::" versions of C++ headers? 20:07:08 :-( 20:07:11 lvcreate's failing 20:07:34 Ah, wait 20:07:39 I seem not to have device-mapper. 20:07:45 I wonder if I need to modprobe something 20:07:53 modprobe device-mapper 20:08:02 pikhq: Apparently I don't have that module. 20:08:05 This is on a LiveCD. 20:08:09 Hmm. 20:08:12 I'm trying to set up an LVM setup to install to. 20:08:13 fizzie: Googling for that just finds people complaining about why their Hello world doesn't compile, and people telling them to use "using namespace std". 20:08:16 I wonder if it is in a package. 20:08:18 That's "dm_mod". 20:08:24 Ah. 20:08:25 That works. 20:08:35 LVM: so intuitive it's unintuitive 20:08:42 Didn't realise; I've got the DM modules compiled in. 20:09:02 fizzie: What do you need to do to have that loaded every time on bootup? I forget the file. 20:09:09 (easier than futzing with the initramfs to only have the modules I need to boot) 20:09:37 -!- oklopol has quit (Read error: 110 (Connection timed out)). 20:10:06 You can put "dm_mod" in /etc/modules, that works in many systems. Although I guess it should be automagically loaded by udev or something; I've certainly never had to manually modprobe it. (Though maybe it autoloads only if it finds LVM volumes or something.) 20:10:13 Yay, I have /dev/sda1 = /boot and /dev/sda2 = physical volume of /dev/osdrive with /dev/osdrive/root, /dev/osdrive/home 20:10:20 fizzie, my root filesystem is on LVM. 20:10:33 Ah. 20:10:33 pikhq: So is mine. 20:10:39 Although, I will probably ... hm. 20:10:39 Wait. 20:10:47 Then putting dm_mod in /etc/modules does nothing. 20:10:50 On my real SSD, I'll only have /boot and a physical volume of /dev/osdrive containing /dev/osdrive/root. 20:10:54 Do I actually need LVM for that? 20:11:02 The guide here suggests so 20:11:15 ehird: No, but LVM is sexy. 20:11:29 pikhq: http://thunk.org/tytso/blog/2009/02/20/aligning-filesystems-to-an-ssds-erase-block-size/ 20:11:37 It really seems to imply that LVM is needed to do the alignment. 20:11:38 Live migration of disks = Mmm. 20:11:56 Hmm. 20:12:11 This thing is going to be horribly slow; putting SSD-tuned settings into a VM writing to a mechanical HD :-) 20:12:41 * ehird creates ext4 filesystems on the two osdrive things. 20:12:55 pikhq: what do you think I should use for my 1gb /boot? 20:12:56 ext2? :-P 20:13:00 1 GB? 20:13:01 I don't exactly need journaling on it 20:13:06 Sure that's enough? 20:13:06 Deewiant: it's what the blog post has 20:13:16 Doesn't mean it's a good idea :-P 20:13:18 Good enough for Theodore Ts'o, good enough for me. ;-) 20:13:24 ext3. 20:13:34 pikhq: What does journaling buy me on /boot? 20:13:35 Don't need it, but it's a good idea. 20:13:36 *shrug* 20:13:37 Kernel developers probably have good reason to have a big /boot. 20:13:43 Deewiant: True 'dat. 20:13:51 Although I still think that's excessive. :-P 20:14:00 ehird: If you're writing to it when you crash, you haven't fucked it up. 20:14:04 ;) 20:14:24 Heck, mount that sucker with ordered; /boot is something you really don't want going. 20:14:40 /boot isn't very vital. 20:14:52 Can't boot? Stick in livecd, put kernel on there. 20:14:53 Kind of sucks to find it broken when you reboot. 20:15:03 Sure, I'm just saying it's not valuable 20:15:09 -!- okloduk has quit (Read error: 110 (Connection timed out)). 20:15:16 I wonder if the "stripe-width=32" will do anything on the olde-style-aligned /boot. 20:15:18 It's not performance-sensitive either 20:15:23 So there's no point in not journaling 20:15:28 Deewiant: Oh but there is. 20:15:30 Less writes to the SSD. 20:15:37 Admittedly, that's barely anything :-) 20:15:40 I wouldn't worry about that. 20:15:47 Thinking that way will just get you into trouble :-P 20:15:51 I'll omit stripe-width. It's just /boot. 20:15:54 Deewiant: Yeah, totally. 20:16:01 I need to keep reminding myself that it's a fuckin' drive, it'll last years dammit. 20:16:12 Worst case, SECURE ERASE the bugger and copy contents back on. 20:16:18 Just so long as you don't have swap on it. 20:16:29 pikhq: I'm gonna have 12GB of RAM. No swap for me. 20:16:36 Goodie. 20:16:44 ehird: Besides, SSDs are supposed to be better with SMART and such, knowing much more precisely when they're going to break. 20:16:55 I don't know if that's the case in practice yet, though. 20:16:58 It'd _slow_ me down... on the SSD, contribute to further wear and tear. On the mechanical HD? Iiiiiiiiiiiiiii'll geeeeeeeeeettt baccccccccccccck tooooooooooo yoooooooooooooouuuuuuuuuuuuuu. 20:17:05 And their failure mode consists of "Uh, you can't write there." 20:17:10 Yep. 20:17:14 SSDs are really resilient. 20:17:49 Depends on the firmware, I guess. 20:17:53 OK, /arch/setup time. 20:18:02 I have this on-a-compact-flash-card system, and I haven't even bothered to think about anything flash-specific on it, even though the write-cycle-handling thing is probably something really really primitive compared to an actual "SSD drive" type of product. (Well, except putting everything that's going to be written more often than daily into tmpfs.) 20:18:11 Deewiant: That's assuming that you're completely out of write cycles. 20:18:27 fizzie: Pfft, I'm going to have lots of stuff being written on /. 20:18:35 Everything but /tmp and /home/ehird/media, probably. 20:18:48 Probably, linux will have fully working TRIM support when I get this. 20:18:51 So it won't be a big deal. 20:20:45 ^_^, the Arch install program let me make the LVM setup just fine. 20:21:01 I'll have to manually edit fstab for /tmp, ofc. 20:21:10 But not on this install. 20:21:28 AnMaster: you said you need a guide for LVM? 20:21:31 shit was easy. 20:22:54 Heh, my HD isn't liking this alignment stuff it's doing. 20:23:00 Krrrrrrrrrrrrrrrnkcluckcluckcluck. 20:23:26 The alignment stuff is for SSDs. ;) 20:23:31 No duh. 20:23:32 (or RAIDs) 20:23:33 This is practice ;-) 20:23:38 In a VM. 20:23:45 ;) 20:23:48 But the HD backing the VM storage isn't liking its antics. 20:24:06 -!- oklopol has joined. 20:24:44 Unsurprisingly, doing SSD alignment on an HD isn't very smart :-P 20:25:00 It's not going to hurt anything, I imagine. Just be slow... which the VM already is, right? 20:25:19 I doubt my HD is the bottleneck 20:28:37 -!- jix has joined. 20:30:23 back 20:30:33 ehird, depends on what you do with it exactly 20:31:24 http://station-mir.ginnungagap.in.ua/fb2.png ← Thus proving that all tiling WM users really just want a framebuffer console. 20:32:32 ehird, if there was a good framebuffer web browser, I would use a framebuffer console. ;p 20:32:40 pikhq: elinks 20:32:49 pikhq: you can even make it display graphics iirc 20:33:10 elinks doesn't do framebuffer. 20:33:16 That would be Links2. 20:33:23 just run it as a normal console app? 20:33:23 Which doesn't even support CSS2. 20:33:34 Then it doesn't handle graphics. 20:33:48 Or complex Javascript. 20:33:51 pikhq: just add a handler to run a framebuffer graphcis display when you choose an image 20:34:00 and sheesh, that's not very framebuffer 20:34:05 What I run currently is based on Gecko. 20:34:11 WebKit ftw 20:34:19 "Not very framebuffer"... 20:34:19 Uh... 20:34:56 I guess it's not very "graphical" either. 20:35:16 For the longest time I used a framebuffer console for just about anything; I guess the web going so multimedia was the major issue. 20:35:58 There's also the bit about my graphics card not handling widescreen via VESA. 20:36:00 Hooray for WMs and graphics and mice. 20:36:10 I guess you could theoretically speaking run Firefox or something using the directfb GTK backend. 20:36:20 But then you're replacing X. 20:36:28 For no good reason. 20:36:54 pikhq: Apart from X sucking, you mean. 20:37:03 Well, only for the web-browser, and at least switching between the console and the "fake-X" would be faster. 20:37:41 Hmm. Arch seems to build its own kernel except with a binary. 20:37:42 Admittedly my framebuffer-console-using days were back when my display hardware was a Matrox G400 card, which worked pretty nicely with matroxfb; I certainly wouldn't want to be stuck with vesafb. 20:38:07 (Also mplayer had that G400-specific video acceleration, obviating the need for Xv.) 20:38:39 Well, Gxxx-specific, anyway. It wasn't just the 400. 20:38:45 Really, the main reason not to use the framebuffer aside from multimedia web stuff is vesa-fb sucks. 20:38:46 -!- Judofyr has quit (Read error: 60 (Operation timed out)). 20:38:48 ;) 20:39:00 And because non graphical environments suck. 20:39:12 ehird, you're just dumb. 20:39:17 Thanks. :) 20:40:02 Deewiant: Hey, they're bringing Scheme back: http://www.cs.hut.fi/~scheme/course-description.html 20:40:04 Also, a framebuffer is very much a graphical environment. 20:40:19 That's nice. 20:40:19 It's a bad graphical environment. 20:40:20 It's just not an exceptionally complex one. 20:40:24 That is mostly used for text. 20:40:36 Lilja's there, unsurprisingly; he's a Haskell guy 20:40:38 -!- Judofyr has joined. 20:41:04 OK, GNU/Arch/LVM/FakeSSD/BSDinit/ls/cp/RichardStallman'sBlood bootup is go. 20:41:05 Which might explain why lazy evaluation is explicitly a goal, too 20:41:19 Aaaand it work... 20:41:23 ... No. No it did not. 20:41:34 Hmm, I wonder if I'll register for that 20:41:35 It cannot find ze /dev/mapper/osdrive-root. 20:41:41 fizzie: What's the file thing? modprobe? 20:41:45 Deewiant: Well, laziness is in "latter part of SICP" too. 20:41:59 Ha. Ha. Ha. The kernel rescue ramfs console has no ls. 20:42:01 ehird: If your root is on LVM, it's the initrd's responsibility for getting things loaded. 20:42:05 Has cat though! 20:42:19 fizzie: And it failed. It don't know shit about no /dev/mapper/blah. 20:42:33 Well, I don't know. If it's some Arch thing, then I definitely don't know. 20:42:51 I don't even know about Debian's initrd, since mkinitramfs or whatever has always built me a working one. 20:42:53 I think it's just... Linux. :-P 20:43:00 pvscan 20:43:08 No, the initrd contents are very distro-specific. 20:43:19 I should probably boot up the live cd. 20:43:19 Or at least somewhat distro-specific. 20:43:31 fizzie: Debian's initrd has LVM stuff in it. 20:43:35 pikhq: no pvscan on the ramfs 20:43:39 maybe I should googol 20:43:45 And Gentoo requires you to use genkernel --do-lvm 20:43:47 pikhq: Yes, so I've assumed, since it works. 20:43:50 -!- oklopol has quit (Read error: 104 (Connection reset by peer)). 20:43:56 Ah. :) 20:43:58 ehird: Yeah. 20:44:23 pikhq: after reading that ssd alignment article, do you think LVM is needed? It seems to be but I can't figure out why 20:44:29 -!- oklopol has joined. 20:44:35 Not really. 20:44:43 But LVM is sexy. :p 20:44:54 I want to avoid it unless I need it for alignment :P 20:44:59 Maybe I'll ask tedts'o. 20:50:09 hmm 20:50:18 pikhq: the X25-M assumes 20GB of writes a day 20:50:28 if you go over that it enables some sort of slowing down thing to extend its life 20:50:37 but 20:50:42 i can't imagine many people actually write over 20gb/day... 20:51:07 And "slowing down" probably means "only 50 times faster than an avergae HD" 20:51:09 s/gae/age/ 20:51:14 Yeah :P 20:52:45 HI THERE YOU RIGHT ABOUT THE BLOCK SIZE I DONT KNOW ANY THING ANY THING ABOUT ERASING A BLOCK BECAUSE LESS BLOCKS SHORTANS THE SSD OR FLASH MEMORY SOO THE SSD AND FLASH MEMORY NOW ARE USEING 1 MB TOO 4 MB BLOCKS NOT THE OTHER WAY AROUD AND BESIDES THE SSD AND FLASH MEMORY CAN LAST OVER 150 YEARS AT 24 7 OPERATION THE FUTURE SSD AND FLASH MEMORY MAY HAVE INFINT READ WRITE 20:52:48 — a comment 20:53:46 First, there's a key just to the left of the "a" key. Hit it. Second, here's a 1st grade basic spelling book. 20:54:07 And, hell, just to cover everything, here's a grammar textbook, too. 20:54:08 ;) 20:54:41 pikhq: Try rapping that comment. It's designed for it, I mean "ANY THING ANY THING"? "SOO"? 20:54:44 That's so filler to fit beats. 20:54:51 LMAO 20:54:52 HI THERE! YOU RIGHT ABOUT THE BLOCK SIZE 20:54:53 I DONT 20:54:58 KNOW ANYTHING ANYTHING 20:55:05 Obviously, it's meant to be yelled. 20:55:06 ABOUT ERASING A BLOCK BECAUSE... 20:55:08 [chorus] 20:55:17 LESS BLOCKS SHORTANS THE SSD OR FLASH 20:55:21 MEMORY SOO THE SSD AND FLASH MEMORY 20:55:26 NOW, ARE USEING 1 MB TOO 20:55:31 4 MB BLOCKS -- NOT THE OTHER WAY 20:55:33 [verse] 20:55:36 AROUND, AND, BESIDES 20:55:39 THE SSD AND FLASH MEMORY CAN LAST 20:55:40 :) 20:55:43 OVER 150 YEARS AT 24 7 OPERATION 20:55:50 THE FUTURE SSD AND FLASH 20:55:53 MEMORY MAY HAVE INFINITE WRITE 20:55:56 [repeat chorus] 20:56:00 It's the SSD Cube Rap. 20:56:11 Like the time cube? 20:56:12 ehird: that's not INFINITE but INFINT. 20:56:15 Exactly, pikhq. 20:56:17 lifthrasiir: Right you are. 20:56:23 :) 20:57:46 INFINT is the int-typed constant for the +inf float value. 20:57:54 :D 20:57:58 So infint write life? 20:58:00 That's a lot of writes. 20:58:06 Well. 20:58:11 fizzie: how do you get +inf in C, again? 20:58:15 It's like MAXINT, only bigger. 20:58:23 as a float 20:58:27 -!- Judofyr has quit (Read error: 110 (Connection timed out)). 20:58:37 1.0f/0? 20:58:42 * pikhq notes that Gene Ray, Doctor of Cubic has gone racist of late 20:58:45 1e99999? 20:58:56 Deewiant: So int infint = (int)(1.0f/0); 20:58:57 Deewiant: That's NAN, not +inf. 20:58:58 pikhq: oh? 20:59:03 pikhq: you sure? 20:59:03 pikhq: Incorrect. 20:59:07 0/0 is NaN. 20:59:11 -!- tombom_ has joined. 20:59:13 There's the macro INFINITY in . 20:59:14 1/0 is +inf, -1/0 is -inf. 20:59:22 Deewiant: how about 1/-0? 20:59:22 Oh, really? 20:59:27 lifthrasiir: -inf. 20:59:30 * pikhq shakes a fist at floats 20:59:47 [ehird:~] % cc -x c /dev/stdin 20:59:47 int main(void){printf("%i\n",(int)(1.0f/0));return 0;} 20:59:49 /dev/stdin: In function ‘main’: 20:59:49 hmm, so +0 and -0 separates in that cases 20:59:51 those* 20:59:51 /dev/stdin:1: warning: incompatible implicit declaration of built-in function ‘printf’ 20:59:53 /dev/stdin:1: warning: division by zero 20:59:55 [ehird:~] % ./a.out 20:59:57 2147483647 20:59:59 So in the future, SSD & flash memory will be able to handle 2147483647 writes. 21:00:01 Assuming 32-bit. 21:00:14 :-P 21:00:18 Why signed? 21:00:27 INFINT. 21:00:29 Not INFUINT. 21:00:34 :-P 21:01:08 -!- tombom_ has quit (Client Quit). 21:01:48 M' sxat'. 21:01:52 Anyone used vmware? 21:01:59 Actually "INFINITE" is defined rather funnily in C99: it expands to (float-typed) constant expression "representing positive or unsigned infinity, if available"; if there is no infinity, it expands to a float constant that overflows at translation time, and therefore forces a diagnostic. 21:02:02 Yuh. 21:02:19 pikhq: How are its linux-guest video drivers etc? 21:02:30 Parallels' don't install, I've had flakiness with VirtualBox's. 21:02:37 Not bad, though I don't recall them handling 3D. 21:03:08 pikhq: Do you think they could handle the latest kernel and Xorg and etc? :P 21:03:08 IIRC, VMware's video drivers are part of X. 21:03:14 Yeah. 21:03:21 Part of X. ;) 21:03:36 Hey, that's nice. 21:03:45 pikhq: do you still have to pay for the version that lets you OMG CREATE YOUR OWN VM? 21:03:58 Yeah, but you can get a 30-day free trial for that. 21:04:09 And after that, you can use the player. 21:04:12 Hee. 21:04:25 pikhq: with Parallels you just had to give a unique email address and reactivate 21:04:29 Yes, I noticed Debian installing a xserver-xorg-video-vmware (as a part of xserver-xorg-video metapackage) recentlyish. 21:04:33 penguinofthegods+youreallfuckingmorons@gmail.com, yum. 21:05:14 ehird: :) 21:05:33 VMware server is also sort-of freewareish, and it can create VMs. I'm not quite sure what it lacks; probably quite a lot on the GUI and graphics-emulation side. 21:05:56 Uhhh. You can't get VMWare Workstation for OS X. 21:06:06 (I also have no idea whether you can run vmware-server-created VMs with the player.) 21:06:38 fizzie: vmware server wants 5 billion personal details from me >_< 21:06:42 Maybe I'll try qemu. 21:06:48 It can emulate regular vga 21:07:22 Qemu emulates the VMware SVGA "card". 21:07:43 "-vga type: "vmware" VMWare SVGA-II compatible adapter. Use it if you have sufficiently recent XFree86/XOrg server or Windows guest with a driver for this card." 21:08:03 It also emulates a Cirrus noname card, which I use. 21:08:08 Also, goddamn qemu is slow :-) 21:08:13 That, too. 21:08:22 qemu is actually doing emulation. 21:08:36 Is there kqemu for OS X? 21:08:59 "preliminary ports to FreeBSD and MS Windows". 21:09:00 Apparently not. 21:09:07 And no KVM either, of course. 21:09:14 Beh I 21:09:16 'll try virtualbox 21:09:33 User Mode Linux would be the fastest way to have a linux VM, I think. 21:09:43 Doesn't it just patch up the kernel so it can be run as a userspace executable? 21:09:52 At least that's what the name implies to me it should do. 21:10:48 Pretty much, yes. But it's a bit rough around the edges. I used UML for my virtual web-server before switching to Xen. 21:11:03 Why would you virtualize your own server? 21:11:19 Mm, VirtualBox is fast. 21:11:34 So that it's easier to play with snapshots and restoration and things like that. 21:11:44 On a webserver? :P 21:12:09 God damn VirtualBox is fast 21:12:29 Yes; the web server was a single LVM logical-volume snapshot, so I could just discard it if it went bad. 21:13:23 VirtualBox uses hardware virtualisation. 21:13:45 Of course it's fast. ;) 21:14:01 So does Parallels. 21:14:04 Debian has a "virtualbox-ose-guest-x11" package if you want to run Linux+X guest in the VM; I'm not sure what other distributions include it too. I guess it does improve the integration a bit. 21:14:04 But virtualbox is faster. 21:14:14 Yes, it does improve it a lot 21:14:28 Why didn't I try Arch before? I love it 21:15:28 -!- tombom has quit (Read error: 110 (Connection timed out)). 21:17:05 Modern electronic-rock music, inaugurated in the early 1960s, is, and always has been, a joint enterprise of British military intelligence and Satanic cults. On the one side, the Satanists control the major rock groups through drugs, sex, threats of violence, and even murder. On the other side, publicity, tours, and recordings are financed by record companies connected to British military intelligence circles. Both sides are intimately entwined with the 21:17:08 biggest business in the world, the international drug trade. 21:20:06 ehird: Oooh, link? 21:20:12 http://www.jesus-is-savior.com/Miscellaneous/satanic_roots_of_rock.htm 21:20:15 Incidentally, can you do flock(2) advisory-locking on a filehandle that's actually an open(2)ed directory? I don't quite see why not, but I've been surprised before. 21:20:37 pikhq: "Clearly, the Beatles' album was dedicated to Satanist Aleister Crowley (pictured to left). It was released 20 years, nearly to the day, after Crowley's death in 1947, and its title song began with the lyrics, "It was twenty years ago today..." The album's cover featured a picture of Crowley." 21:20:46 http://station-mir.ginnungagap.in.ua/fb2.png ← Thus proving that all tiling WM users really just want a framebuffer console. <-- wow, url for download? 21:21:04 AnMaster: dvtm 21:21:06 * pikhq finds that hilarious 21:21:08 * AnMaster googles 21:21:09 http://www.brain-dump.org/projects/dvtm/ 21:21:18 AnMaster: to do the multi-split thing you have to put a dvtm in a dvtm 21:21:22 so you can tile while you tile 21:21:38 ... Beatles. Picture of innocence. 21:21:42 WARNING: MEME DETECTED! EVACUATE CHANNEL! 21:21:57 "I am the Walrus". Need I say more? 21:22:15 Ubuntu's new screen-profiles thing looks rather curious; ran into it when testing a diskless husk of a computer with Ubuntu live-cd-on-a-USB-stick. 21:22:19 I am the Walrus, whereby "am" I mean "take" and "the Walrus" I mean "heroin". 21:22:32 -!- Hiato has quit ("Leaving."). 21:22:37 Also, "simple little tunes". 21:22:44 *Someone's* only listened to one album. 21:23:29 And... Wow. Claiming that the Beatle's popularity was a hoax? 21:23:32 tomorrow never knows is the most simple tune ever 21:23:34 totally 21:23:35 Someone cranked the crazy up to 11. 21:24:09 Up to X11. (Because that's obviously the pinnacle of crazy.) 21:24:25 i agree 21:24:59 Also, they claim Sgt. Pepper's Lonely Hearts Club Band was dedicated to the promotion of psychedelic drugs. 21:25:20 Wrong album; that would be just about every late-career album *but* that one. 21:25:33 :-D 21:25:58 Sgt. Pepper's was merely dedicated to pschedelic rock. 21:26:07 s/psch/psych/ 21:26:47 dvtm is just wow 21:26:47 :D 21:27:50 Also, LSD was 100% legal at the time. :p 21:28:11 (made illegal in '67 or '68...) 21:28:46 Late '66, rather. 21:28:56 ( 21:28:59 '71 for the UK) 21:31:40 AnMaster: you should have forced me to get arch working, dammit :-D 21:31:50 ehird, didn't you get it working? 21:31:57 I did. 21:31:59 I meant earlier. 21:32:05 It's a ruddy good distro. 21:32:18 ehird, why? Then you would have hated it because I forced you to do it I bet ;P 21:33:10 True :P 21:33:24 But seriously, it's nice. 21:33:30 They should totally use apt though/ 21:33:34 err 21:33:35 s/\/$// 21:33:48 I only ever had one issue with it's package manager. 21:33:55 I haven't had any. 21:33:57 I just like apt ;-) 21:33:59 good :) 21:34:08 I think they fixed it 21:34:11 Although pacman's probably second-best. 21:34:12 Every binary distro should use apt; it's freaking awesome. 21:35:00 Especially with such features as debootstrap... 21:35:24 (so very easy to set up a Xen VM running Debian...) 21:35:30 I ran into some issue about pacman complaining that it couldn't upgrade bash because "/usr/share/man/man1/bash.1.gz already exists". I haven't seen that bug for months though 21:35:48 iirc they fixed it 21:36:27 ehird, I guess you will get the occasional issue with arch. since it is so bleeding edge usually. 21:36:43 I like bleeding edge. 21:36:46 usually easy enough to fix, and the irc channel is very friendly and helpful 21:36:50 I mean, I'm using ext4 for everything but /boot! 21:36:58 It was just declared stable in... december? 21:36:59 fair enough 21:37:18 Yay, the "AUR" thing has virtualbox-additions as a package. 21:37:35 ehird, I was thinking more about "gcc 4.3.3 in stable less than a week after it was released" 21:37:47 ehird, tip for AUR: use yauort 21:37:50 AnMaster: I think gcc releases are generally conservative in themselves 21:37:51 yaourt* 21:37:56 Catchy name. 21:38:01 Deewiant, I love tab complete :P 21:38:03 Oops, forgot to change hostname. 21:38:05 -!- MigoMipo has quit (Nick collision from services.). 21:38:11 ehird, ? 21:38:15 on the vm 21:38:17 Okay: change hostname, add user account, install sudo, passwd -l root. 21:38:19 oh 21:38:46 augh. What happened to man passwd 21:38:58 Magick. 21:38:59 oh locale issues 21:39:02 -_- 21:39:08 passwd - ändra användarlösenord 21:39:09 funny 21:39:19 no idea why I have the man page in Swedish at all... 21:39:38 pacman -Sy sudo, la la la la la 21:39:46 iwonder how safe it is raid0 ssds 21:39:57 ok now I changed things in a different way 21:40:01 and this is even crasier 21:40:04 crazier* 21:40:05 passwd - Axndra anvAxndarlA9|senord 21:40:13 I have no idea how that could happen 21:40:14 bsmntbombdood: It should be fine. You considering it? I'd just buy the $600 160GB one if so; it'll be cheaper overall probably 21:40:22 Also, it's not as if OS drives need an awful lot. 21:40:35 it should say: "passwd - Ändra användarlösenord" 21:40:54 or better, in English 21:40:56 ehird: just curios 21:41:07 and two raid0 drives are faster than 1 big one... 21:41:24 bsmntbombdood: I highly doubt it would be possible to notice the difference with an SSD> 21:41:28 s/>$/./ 21:41:30 bsmntbombdood: No seeking. 21:41:33 heh 21:41:38 there is a matching command to su 21:41:40 for group 21:41:44 strange 21:41:45 sg 21:41:45 sg 21:41:47 indeed 21:41:48 haha 21:41:50 ehird: read/write bandwidth is doubled 21:41:53 ehird, I just never noticed it before! 21:41:59 you don't use it a lot 21:42:09 of course I know about sgid binaries, like nethack 21:42:10 bsmntbombdood: it'd probably end up more or less unnoticable for the cost 21:42:14 -!- puzzlet_ has quit (Read error: 60 (Operation timed out)). 21:42:24 nethack is setgid? 21:42:25 why? 21:42:31 scoreboard files? 21:42:36 yes 21:42:36 YES 21:42:38 heh 21:42:41 sgid games 21:42:48 hmm 21:42:51 yaourt isn't in arch :-D 21:42:58 I guess I have to do what it does manually to get it 21:43:00 ehird, it is in AUR 21:43:25 ehird, so what you do is, get that manually and build it, then use yaourt for all future AUR package installs 21:43:29 yeah 21:43:31 including updating yaourt 21:43:41 I'll need base-devel won't I 21:43:59 don't remember what language yaourt is in 21:44:01 but 21:44:04 mm 21:44:06 you need makepkg binary 21:44:07 might as well do it 21:44:09 ehird: http://hothardware.com/News/Intel-SSDs-RAID-0-A-Case-Study-In-Speed-Take-2/ 21:44:11 year 21:44:14 *yeah 21:44:17 ehird, it is rather simple to build it anyway 21:44:27 as normal user create a directory 21:44:27 -!- puzzlet has joined. 21:44:31 cd into said directory 21:44:45 wget or otherwise download the PKGBUILD and any other patch files listed at the AUR web page for it 21:44:48 then makepkg -c 21:44:50 bsmntbombdood: The question is whether it is noticable in 90% of regular usage cases 21:44:55 *noticeable 21:45:02 ehird: who cares, it's fast as shit 21:45:16 bsmntbombdood: So's our $80k system 21:45:19 ehird, you will have a *i686.tar.gz (replace i686 with x86_64 or such if that is what you use) 21:45:28 I'm reading the wiki for it AnMaster 21:45:30 *wiki page 21:45:30 then you do sudo pacman -U thispackagefile 21:45:38 where that is that tar.gz 21:45:41 ehird, good :) 21:45:55 You don't even need to install it the first time? 21:45:59 Just ask it to install itself 21:46:05 Deewiant, I didn't know that 21:46:10 All you need to do is get the binary working 21:46:13 http://wiki.archlinux.org/index.php/Yaourt 21:46:15 Two ways 21:46:20 Easy way is to install it via pacman 21:46:28 But meh @ that 21:46:41 I recommend archlinuxfr 21:46:45 Do you? 21:46:45 Why. 21:46:51 It's easiest. 21:47:01 ... it's just a few commands to install yaourt 21:47:20 And it's a one-time thing; no sense cluttering the repo list 21:47:23 of course, installing a package from source on gentoo is *even* easier 21:47:23 :P 21:47:31 AUR updates are more painful than non-AUR. 21:47:39 bsmntbombdood: If you really have the cash, RAID 0 some of the X25-*E*s. 21:47:39 Deewiant, that's true 21:47:41 Deewiant: Bah, fine. 21:47:47 bsmntbombdood: That'll be superfast. 21:47:51 duh 21:48:04 If you really have the cash, we speced a $80,000 machine. 21:48:06 :D 21:48:13 a completely useless machine 21:48:31 pikhq, can I see the full component list of it please 21:48:32 It's useful for regular computing tasks as well as the crazy shit :P 21:48:39 AnMaster: No, but I can give you one. 21:48:39 Probably far better to get 4 $20,000 machines or, I dunno, 80 $1,000 machines. 21:48:51 ehird, err? 21:49:04 AnMaster: Here: 21:49:08 AnMaster: http://pastie.org/472570.txt?key=gzff3jen8qjjkn1gx7brlg 21:49:12 native language joke detected 21:49:16 Doesn't include a power supply or case or well a lot of things 21:49:16 and no 21:49:17 not a joke 21:49:20 But that's how far we got 21:49:27 Then we got bored since it was trivial to spill money on the rest 21:49:29 you didn't finish it? 21:49:32 nope 21:49:34 see last line 21:49:41 Deewiant: What does archlinuxfr contain apart from yaourt, then? 21:49:42 300GB VelociRaptor ($229.99) x 58 = $13,339.42 21:49:46 Got bored. 21:49:49 didn't you go SSD? 21:49:54 AnMaster: yes, but that came to $44k 21:49:57 and we decided it was stupid 21:49:59 ehird: I don't know how to query Pacman for that. I'm fairly sure I've got something else from it, too. 21:50:01 since you can grow pretty much unbounded with them 21:50:15 Deewiant: so how are AUR upgrades harder? 21:50:16 ehird, you can grow unbounded with disks too 21:50:18 ehird: Though we'd need a less beefy PSU for them. 21:50:28 Which would actually matter with such a system. 21:50:28 ;( 21:50:29 ;) 21:50:32 AnMaster: yes, but we just scaled down so it wasn't so much cheating 21:50:49 the case isn't there 21:50:55 nor is the power supply 21:50:58 oh 21:50:58 we. got. bored. 21:51:00 :P 21:51:01 ehird: For one thing, each AUR package is update-checked separately, instead of just checking a database timestamp (or hash? Not sure how it works) 21:51:01 We could in principle have infinite storage, daisy-chaining external SAS RAID units. 21:51:01 right 21:51:12 pikhq: yeah, exactly 21:51:14 although the latency would suck 21:51:21 but you can easily spend 5 gajillion dollars in storage 21:51:23 It would be horrendous. 21:51:27 it's harder to spend it on, say, CPUs 21:51:27 We could in principle have infinite storage, daisy-chaining external SAS RAID units. 21:51:27 But you could do it. 21:51:29 oooooh 21:51:32 and graphics cards 21:51:36 And get several exabytes. 21:51:43 UTM here I come! 21:51:47 pikhq: How to run a turing machine: Daisy chain them as you go. 21:51:54 ehird: :D 21:51:54 (in principle!) 21:51:59 The latency is so low that they'll have time to come in tomorrow and do it at their leisure. 21:52:01 Err 21:52:02 so high 21:52:06 ehird, I said that two seconds before! 21:52:09 (on this side) 21:52:12 LMAO 21:52:13 AnMaster: I thought of it before, though 21:52:17 but I wasn't at the keyboard 21:52:18 ehird, hard to know 21:52:32 AnMaster: about 3-4s brain-to-start-typing latency there 21:52:55 but I was thinking of it as soon as pikhq mentioned it. " We could in principle have infinite storage, daisy-chaining external SAS RAID units." 21:53:21 but that is completely pointless 21:53:28 ... 21:53:29 what 21:53:46 bus bandwidth and that 21:54:04 "error: could not register 'archlinuxfr' database (unexpected system error)" ← pretty cryptic error for "you didn't run me as root" 21:54:06 bsmntbombdood, read the context kay? 21:54:12 So, I'm now imagining a freaking massive data center, filled with nothing but disks and a single system, hooked up to it via 8 SAS cables. 21:54:14 AnMaster: How ironic 21:54:16 don't put more than 10tb/server 21:54:17 ehird, indeed. 21:54:21 and indeed again 21:54:23 ehird: yaourt uses sudo automatically. 21:54:29 Deewiant: EWWWWWWWWWWWWWWWWW 21:54:30 EWWWWWWWWWWWWWWWWWWW 21:54:30 bsmntbombdood: We realise this. 21:54:32 EEEEEEEEEEEEEWWWWWWWWWWWWWWWWWWWWWWWWWWWWW 21:54:34 ehird, um what 21:54:36 Deewiant: MAKE IT STOP!! 21:54:37 ehird: Or su, not sure which. 21:54:41 AnMaster: Automatically sudoing is the devil. 21:54:42 Deewiant, sudo 21:54:44 bsmntbombdood: We're joking about means to create a UTM, man. 21:54:44 ;) 21:54:51 ehird, it tells you in advance iirc 21:54:53 I think I read somewhere that it actually tries both. 21:54:58 sigh 21:55:00 Eeew. 21:55:04 ehird, it is very clear with what it does 21:55:14 It just shouldn't. 21:55:21 ehird, then it wouldn't work 21:55:27 since you shouldn't build packages as root 21:55:33 ... what? 21:55:37 Whyever not? 21:55:47 ehird, building as normal user is recommended 21:55:47 It's exactly what I intend to do. 21:55:51 o_O Why? 21:55:53 then installing as root 21:55:57 ehird, search me 21:56:02 ehird: Malicious PKGBUILDs. 21:56:03 it uses fakeroot anyway 21:56:08 so there is no obvious reason 21:56:12 Deewiant: Oh, so I only use all my home. Awesome. 21:56:18 Deewiant, but makepkg uses fakeroot! 21:56:20 It's not like all my valuable data is on there or anything. 21:56:25 I don't know anything about fakeroot. 21:56:32 I'm just going to use it as root. 21:56:38 Pretends to be root. 21:56:39 anyway IMO it should just su to a temp user 21:56:47 ehird, don't you agree? 21:56:52 Maybe. 21:56:52 Fails if it does anything actually unsafe. 21:56:55 Wow. 21:57:00 Yaourt is written in Bash. 21:57:06 like it internally does "su - package-builder" 21:57:08 or such 21:57:09 I ran "yaourt" and I watched each line of help scroll by individually, one by one. 21:57:13 That's how slow it is :-) 21:57:13 And I thought Portage was bad. 21:57:14 aha 21:57:16 ehird, it was fast for me 21:57:25 (Portage is written in Python, with attempts to rewrite it in C) 21:57:28 pikhq, um... portage does change to a different user. 21:57:31 "Runing [sic] yaourt as a non-privileged user requiers[sic] some entries in sudoers file:" 21:57:41 ehird, hey they were french! 21:57:41 AnMaster: I was describing how fakeroot works. 21:57:43 (tells you how to make pacman be sudoable by everyone, pretty much) 21:57:46 time yaourt --help >/dev/null -> 0.13user 0.38system 21:57:48 So by default it doesn't do that 21:57:56 Not what Portage does for its sandboxing. 21:57:59 Deewiant: VM, yo. 21:58:01 -!- BeholdMyGlory has quit (Remote closed the connection). 21:58:03 Deewiant, similar results on my P3 even 21:58:14 so ehird's computer suck badly we see, 21:58:15 ;P 21:58:18 My point was that that is a long time. 21:58:21 Yeah. My virtual computer sucks. 21:58:22 0.4 seconds to output a string? 21:58:24 pikhq, portage sandboxes in two ways 21:58:27 Deewiant: Jeeze. 21:58:32 pikhq, 1) change to different user 21:58:35 2) LD_PRELOAD 21:58:40 hmm, is there a command that decides between yaourt and pacman automagically? 21:58:40 just to be extra secure 21:58:49 AnMaster: should just do as macports: chroot 21:58:57 Basically. 21:59:15 ehird, well you could write a module to do that I think 21:59:26 Deewiant: Uh... Portage isn't much better. 21:59:28 you put the base libraries and all the packages' dependencies into a build directory, chroot in, su to low-powered user that still has rights to the whole chroot, build & install package 21:59:29 ehird, however that needs to mirror stuff in the chroot 21:59:31 sudo emerge --help 1.29s user 0.16s system 98% cpu 1.463 total 21:59:33 go out of chroot, copy files over 21:59:49 pikhq: I was not comparing to Portage. 21:59:57 * pikhq knows 22:00:04 * pikhq was 22:00:12 * pikhq was curious 22:00:14 21:58 ehird: hmm, is there a command that decides between yaourt and pacman automagically? 22:00:16 * pikhq uses /me 22:00:19 * pikhq ! 22:00:23 # time pmerge --help &>/dev/null 22:00:23 real 0m0.187s 22:00:23 user 0m0.148s 22:00:23 sys 0m0.034s 22:00:24 ;P 22:00:26 sudo pacman -S xorg 22:00:28 pkgcore for the win! 22:00:31 \o/ 22:00:36 ldc --help-hidden is over three times as long, and outputs in 0.00 seconds. 22:00:41 21:58 ehird: hmm, is there a command that decides between yaourt and pacman automagically? 22:00:56 ehird, you mean like install deps for AUR packages when needed from the normal repos? 22:01:02 yes yaourt can do that 22:01:08 AnMaster: that too, but also, 'foo -S packagethatsinmainrepo' 22:01:09 and also 22:01:13 'foo -S packgeinaur' 22:01:13 -!- KingOfKarlsruhe has joined. 22:01:14 both work 22:01:20 ehird: yaourt has a pacman-compatible interface. 22:01:23 ehird, again yaourt can already do that 22:01:24 With some additions. 22:01:34 More like slowterface. 22:01:45 So why isn't yaourt the official pacman? 22:01:45 The only thing I've found that pacman can do that yaourt can't is output a help string for a specific command. 22:01:48 ehird, what in "interface" implies fast. 22:02:12 ehird, interface is something slow. Like 200 baud serial. 22:02:13 :P 22:02:21 (as well) 22:02:39 "I think the big peaks and valleys are a result of high erase/re-write latency of MLC flash. You don't see that with SLC actually." 22:02:41 I wonder if that's true. 22:02:44 It's not what Anand told me :P 22:02:57 ehird, Anand? Nand? andand? 22:03:04 Anand. anandtech.com 22:03:13 * AnMaster tries to figure out what logical operation anand would be 22:03:17 like a nand + and? 22:03:19 An and, obviously. 22:03:26 see wut i did thur 22:03:40 MEME WARNING? 22:03:49 Nope. 22:03:50 Anand 22:03:51 An and 22:03:53 yes 22:03:54 but 22:03:58 " see wut i did thur" 22:04:02 that has meme warning? 22:04:02 That's not a meme. 22:04:05 That's just a cheesy phrase. 22:04:06 oh ok 22:04:11 -!- Sgeo has joined. 22:04:14 ehird, Cheddar? 22:04:25 See what I did cheddar? 22:04:28 No, that doesn't work. 22:04:40 ehird, true, cheddar is too dry I guess. 22:05:04 and I don't know the *English* names of other cheeses 22:05:08 mostly 22:05:27 bbiab 22:05:28 Not even American "cheese"? 22:05:38 American sleaze. 22:06:02 -!- nooga has joined. 22:06:04 hyh 22:06:11 idiotic problem 22:07:42 You know what creeps me out? Seeing CPUs. Too fucking small parts, dammit! 22:10:13 WTF? I have no /etc/X11/xorg.conf 22:11:34 brb 22:11:44 mmmm garlic 22:12:12 -!- oklopol has quit (Read error: 104 (Connection reset by peer)). 22:12:19 ehird, latest X11 doesn't need it 22:12:25 auto config stuff 22:12:27 -!- oklopol has joined. 22:12:40 anyway: strongly garlic flavoured bread + butter + cheddar 22:12:41 :D 22:12:44 ehird, agree? 22:12:51 I take that silence as a yes 22:14:03 How's about blackjack and hookers with that? Y'know what, screw the bread. And the blackjack! 22:15:11 "" <-- heuristics detected popular culture reference but not found in database. 22:15:12 -!- oklopol has quit (Read error: 104 (Connection reset by peer)). 22:15:25 -!- oklopol has joined. 22:15:26 Futurama. 22:15:33 aha 22:16:21 Another one of those shows that Fox thought "Oh, it's good! Cancel it!" 22:16:28 pikhq, what does the game blackjack have to do with it. 22:16:46 AnMaster: Quoting Bender. 22:16:51 true 22:16:57 Well, channeling, rather. 22:16:59 but what was the original context for it 22:17:14 Bender likes gambling. 22:17:18 ah 22:17:19 That's about it. 22:17:23 AnMaster: Consider reading curl -I slashdot.org on occasion. 22:17:23 what do you mean "channeling"? 22:17:34 It's not quite a quote. 22:17:43 Deewiant, what does -I do? 22:17:46 He never said anything about bread, specifically. ;) 22:17:53 AnMaster: Consider reading curl --help. 22:17:53 pikhq, true 22:18:06 (from what google tells me) 22:18:13 (when googling for the exact phrase) 22:18:32 -I/--head Show document info only 22:18:33 oh 22:18:42 Deewiant, right that one is old 22:18:50 X-Fry: Sweet justice! Sweet, juicy justice! 22:18:55 I have seen other ones for other clients 22:19:09 It doesn't depend on client, it varies by timestamp. 22:19:15 I think it changes every minute or something. 22:19:27 Deewiant, more than that 22:19:30 In any case, if you know of it, haven't you ever seen 'X-Bender'? 22:19:36 every single request 22:19:42 Deewiant, not that I remember 22:19:52 -!- puzzlet has quit (Remote closed the connection). 22:19:56 -!- puzzlet has joined. 22:20:02 Do it a couple of times, then, until you do, and you will have. 22:20:03 Deewiant, I knew they did strange things in their headers. But I never really cared about it 22:20:12 X-Bender: Nothing like a warm fire and a super-soaker of fine cognac. 22:20:14 ok 22:20:15 Hmm? 22:20:16 It's not a "strange thing". 22:20:37 They just serve up random Futurama quotes in X-. 22:20:41 X-Fry: That doesn't look like an "L", unless you count lower case. 22:20:45 that is futurama too? 22:20:46 Who does? 22:20:51 pikhq: /. 22:20:52 I never watched futurama 22:20:53 AnMaster: Yeah. 22:20:57 Deewiant: Ah. 22:21:07 AnMaster: You should; it's brilliant. 22:21:09 Deewiant, rebus time? 22:21:19 I didn't think it was that brilliant, actually. It was OK. 22:21:24 pikhq, well maybe, I don't have a tv currently 22:21:31 haven't had for over a year 22:21:40 Not currently aired. 22:21:42 nor a TV card 22:21:46 So, TV won't help. 22:21:50 pikhq, I don't do pirate copies. 22:21:55 BitTorrent or DVDs. 22:22:04 No love for VHS? 22:22:17 Deewiant, actually I still have a VHS player somewhere. Just no TV 22:22:21 to use it with 22:22:48 Hrm. Actually, Comedy Central recently started a new season. So, never mind. 22:22:57 this VHS player is so old it has a *wired* remote 22:23:02 as in a cable for the remote :D 22:23:15 (it's from the 1980s iirc) 22:24:06 yay 22:24:10 newegg shipped my stuff 22:24:58 pikhq, anyway buying the DVD would require me to figure out how to use a non-data dvd in my computer 22:25:06 I guess mplayer could handle it 22:25:14 It's a matter of putting it in the drive and pointing a player at it. 22:25:30 Deewiant, but mplayer has arcane syntax for doing that sort of stuff 22:25:37 like mplayer cdda://1-2 22:25:42 or whatever it was for cd 22:25:42 Then use a GUI. 22:25:51 I prefer smplayer myself. 22:25:57 Deewiant, xine is messy. kmplayer is confusing. 22:25:58 22:12 AnMaster: ehird, latest X11 doesn't need it 22:25:58 22:12 AnMaster: auto config stuff 22:25:59 what 22:26:05 ehird, yes. it uses hal and dbus 22:26:12 I think he's right, X11 no longer needs a conf 22:26:19 You can still make one, of course. 22:26:21 yes 22:27:10 Well, that's nice. 22:27:18 I still use the config because 1) I already have a working one and I don't see a point in learning a different system 2) X auto config by hal had issues with my joystick before. 22:27:38 AnMaster: It's easy to do with mplay. 22:27:42 mplayer dvd:// 22:27:43 issues == refused to let me use it for anything 22:27:44 mplayer, rather. 22:28:07 * AnMaster wants to use axis 8 and 9 for mouse in X, and button 27 for click in X. The rest is left to the flight sim 22:28:17 that works just fine with the good old joystick driver in X 22:28:45 wait 22:28:49 axis 10 and 11 22:28:51 not 8 and 9 22:29:07 8 and 9 are... uh, the two wheels on the throttle I think 22:29:27 too many axises (axes?) to remember 22:29:40 have it in the config for the flightsim :) 22:30:52 wth is "Super Audio CD"? 22:31:02 * AnMaster googles 22:31:27 anyway I saw some crazy cds recently. You know those old mini-format cds? 22:31:36 I think they were discussed in here recently 22:31:54 while mentioning that they broke slot-style cd drives 22:32:06 anyway I saw one just last week 22:32:11 This autoconfig thing is so awesome that it doesn't let me move my mouse. 22:32:14 A SACD is not a CD. 22:32:17 How do I make it generate a skeleton xorg.conf? 22:33:00 It is physically a DVD, IIRC. 22:33:11 pikhq, really? I saw this phrase oh the cover of a music cd: "OBS! Denna Super Audio CD går även att spela på en vanlig CD-spelare." <-- "Observe! This Super Audio CD can also be played on a normal CD player" 22:33:26 * AnMaster wonders how that works then 22:33:52 "Hybrid: The most popular of the three types, hybrid discs include a "Red Book" layer compatible with most legacy Compact Disc players, dubbed the "CD layer," and a 4.7 GB SACD layer, dubbed the "HD layer." It is not uncommon for hybrid discs to carry the "Compact Disc Digital Audio" logo to show that the disc is CDDA-compliant."' 22:33:57 That's clever. 22:34:05 heh 22:34:07 nice 22:34:28 can a DVD player in a computer play the super audio layer in it? 22:34:37 Note that there is absolutely no Linux support for SACD. 22:34:41 damn 22:34:44 With software support, yes. 22:34:45 Deewiant: would you happen to know? 22:34:54 pikhq, really sure? 22:34:56 Software support is rare. 22:35:18 (and SACD uses 96 kHz to 192 kHz audio; most sound cards can't handle it) 22:35:18 Really sure. 22:35:22 ehird: xorgconfig, or whatever it's called? 22:35:33 pikhq, wouldn't it need a different laser type? 22:35:41 Alternatively, google for one 22:35:54 AnMaster: As said, the SACD layer is physically a DVD. 22:35:59 aha 22:36:19 But it doesn't seem to use PCM sound at all; it's a one-bit delta-sigma encoding thing with 2822.4 kHz sampling rate. 22:36:39 SACD is DRM'd, and it has yet to be cracked. 22:36:42 I highly doubt humans can distinguish better sampling than a CD. 22:36:43 aha 22:36:48 ehird, agreed 22:37:04 ehird, I was just interested in the music on it. that was all 22:37:08 mm 22:37:08 ehird: In some cases, humans can distinguish 48 kHz. 22:37:12 cd quality is good enough for me 22:37:16 But *that* is soley because of poor ADCs. 22:37:24 ADCs? 22:37:30 Erm. DAC. 22:37:32 Analog Digital converter? 22:37:34 aha 22:37:39 pikhq, that makes more sense 22:37:41 Heck... Most people can't distinguish 128kbps MP3s from CD quality. 22:37:43 Yes, that includes you 22:37:57 ehird: I can. 22:38:05 Infallopedia says that SACD's quality is comparable to a 20-bit PCM format and a 192 kHz sampling rate. Though it's not exactly the same thing. 22:38:08 Heck... Most people can't distinguish 128kbps MP3s from CD quality. <-- Depends on what music. For some types of music the difference is more noticable 22:38:10 pikhq: Show the double-blind ABX test results, please. 22:38:11 192 kbps MP3s are quite a bit more difficult to distinguish, though. 22:38:18 pikhq: Otherwise you're talking shit. 22:38:30 ehird: You first. 22:38:31 ;p 22:38:35 fizzie, "Your search - Infallopedia - did not match any documents." 22:38:38 pikhq: I fully admit I cannot distinguish them. 22:38:43 AnMaster: it was a portmaneuver 22:38:44 I guess you meant wikipedia 22:38:49 ehird, yes 22:38:59 I wasn't sure if it was an existing joke of wikipedia 22:39:04 AnMaster: Yes; it's the only infallible thing around here. 22:39:04 like uncyclopedia 22:39:15 or just a nick for wikipedia 22:39:16 Curious that it's not a term that's in use. 22:39:19 I seem to have no xorgconfig. 22:39:21 Maybe it's in sbin. 22:39:29 ehird, why do you want it 22:39:31 Nope. 22:39:33 it auto configures 22:39:35 AnMaster: to generate a skeleton xorg.conf 22:39:35 ... 22:39:38 so I can modify it 22:39:39 ehird, you don't need it. 22:39:47 My mouse doesn't work, AnMaster. 22:39:48 ehird, you do it by editing HAL config 22:39:52 iirc 22:39:53 Okay. How? 22:39:59 I need to use the vboxmouse driver. 22:40:08 ehird, it uses evdev 22:40:13 hm 22:40:14 That's nice :-P 22:40:16 with a different driver 22:40:16 Hmm 22:40:19 it might be hard 22:40:25 anyway: don't ask me. I compiled xorg without hal support 22:40:32 I'd be fine with evdev if I could use the mouse 22:40:36 becuase I hate hal. With a passion. 22:40:45 I love hal. 22:40:47 ehird, it wouldn't allow the nice integration thingy 22:40:53 ehird, why 22:40:54 on 22:40:54 It's so workitude. 22:40:55 earth 22:41:06 "workitude"...? 22:41:18 AnMaster: (1) I don't care about the integration thingy (2) Because it's never broken for me, ever, and it means I do a lot less configuring. 22:41:52 ehird, why do you want "vboxmouse" if you don't want the integration bit 22:41:58 -!- okloduk has joined. 22:41:59 hm 22:42:00 Hmm, I just got spam with the title "Stop sendingme this crap". 22:42:02 AnMaster: Because I Cannot Move My Mouse. 22:42:04 ehird, does lsusb show your mouse? 22:42:07 It just stays there. 22:42:14 And no, just two root hubs. 22:42:15 I wonder if spam senders take titles from messages they receive? 22:42:16 or is it PS/2? 22:42:22 ok PS/2 I guess hm 22:42:22 It's virtual. I don't know 22:42:33 ehird, well it has to fake it to the guest *somehow* 22:42:33 Should I dmesg? 22:42:42 ehird, no. Try gpm instead 22:42:46 yep 22:42:47 ehird, or even better 22:42:47 PS/2 mouse 22:42:53 from dmesg|grep -i ps/2 22:43:11 [ 8.075700] PNP: PS/2 Controller [PNP0303:PS2K] at 0x60,0x64 irq 1 22:43:11 [ 8.093979] PNP: PS/2 appears to have AUX port disabled, if this is incorrect please boot with i8042.nopnp 22:43:11 [ 8.132103] mice: PS/2 mouse device common for all mice 22:43:13 like that? 22:43:25 yes, but also: 22:43:29 (I don't have a PS/2 mouse, I do have a PS/2 keyboard though) 22:43:45 psmouse serio1: ID: 10 00 64<6>input: ImExPS/2 Generic Explorer Mouse as /devices/platform/i8042/serio1/input/input5 22:43:50 ah 22:43:50 I typed that out by hand :P 22:44:20 ehird, try #xorg 22:44:24 they deserve it 22:44:37 What did they ever do to you? 22:44:47 ehird, begin using HAL for auto config dammit! 22:44:53 that is what they did 22:44:58 I like that! 22:45:00 Fuck xorg.conf. 22:45:01 traitors to true UNIX-kind 22:45:09 Er... HAL is more unix than the old system. 22:45:16 Do one thing, do it well, use it everywhere. 22:45:27 ehird, traitors to true-"what UNIX was in practice"-kind 22:45:29 then 22:45:37 or 22:45:38 It's not UNIXy in any form. 22:45:53 traitors to true-ICCCP-kind? 22:45:58 (spelling?) 22:46:08 what 22:46:14 Woot, signed up for a Scheme course. 22:46:26 Deewiant: Have you read your SICP today? 22:46:29 oh 22:46:30 Nope. 22:46:31 Deewiant: Maybe I should do it too. 22:46:36 ICCC-kind 22:46:38 is what I meant 22:46:53 Deewiant: Read SICP 22:47:00 AnMaster: International Christian Chamber of Commerce? 22:47:00 Deewiant: The IRC channel they mention is kind of quiet. 22:47:02 ehird: It's the course book, so I might, partly. 22:47:07 nter- Client Communications Conventions 22:47:09 err 22:47:11 Inter- Client Communications Conventions 22:47:12 Deewiant: *insert /prog/snake* 22:47:13 of course 22:47:14 Only chapters 3-5 for the course, though. 22:47:17 fizzie: I've been meaning to learn a Lisp anyway, so I figured I might as well do it thus. :-) 22:47:27 Oh? 22:47:36 Well read 1-2 then! 22:47:38 You can never have enough 22:47:39 >>SICP<<. 22:47:40 ehird, it must have been mentioned in that "unix haters handbook" you love to refer to 22:48:00 AnMaster: I'm pretty sure UNIX-HATERS would be aligned with a HAL-like system more than duplicate, incompatible configurations 22:48:15 Deewiant: "the material to read for the exam are the chapters 3-5 of SICP"; I guess chapters 1+2 could be relevant for the homeworks, though. 22:48:17 ehird, only if HAL worked 22:48:26 the old system works better in practise 22:48:30 AnMaster: that's called bugs, not design deficiencies. And it works for me. 22:48:31 Deewiant: I did the old Scheme course, but this one seems to be the Extended Edition. 22:48:36 "You can bring the coursebook to the exam." 22:48:40 Ho-ho 22:48:43 ehird, so you can move your mouse now? 22:48:44 great 22:48:54 But then, SICP isn't very useful in that sense, I guess, so it makes sense. 22:49:08 AnMaster: In case you haven't realised, VMs are not the typical use case of Linux. 22:49:16 Besides, it'll just be flicking a switch somewhere 22:49:26 ehird, I have no idea where 22:49:28 -!- okloduk has quit (Read error: 60 (Operation timed out)). 22:50:27 Can xorg deal with a partial xorg.conf? 22:50:30 i.e., just specifying one thing? 22:50:53 I don't see why not 22:51:12 ehird, unknown 22:51:22 "procedural abstraction, data abstraction, functional programming, concurrency, streams and lazy evaluation, programming language interpreters, logic programming, register machines and machine language programming"; that's a rather wide-ranging list; though it's pretty directly from the book. 22:51:31 ehird, but you need to turn off HAL to define any input devices at all iirc 22:51:34 it is all or nothing 22:51:40 or maybe you can tell HAL to ignore the mouse 22:51:42 Now THAT is stupid. 22:52:01 I'm pretty sure you can tell HAL to ignore the mouse. 22:52:21 ehird, read (short) http://www.gentoo.org/proj/en/desktop/x/x11/xorg-server-1.5-upgrade-guide.xml 22:52:32 it describes the differences to between classical and HAL 22:54:21 Alas, my keyboard is dead too, it seems. 22:55:09 ehird, I only had troubles with HAL. 22:55:35 I've had no troubles with HAL. 22:55:44 Ditto. Apart from this minor VM-related one. 22:56:08 ehird, I would call "keyboard and mouse not working" rather "major" 22:56:21 ... in Yet Another Virtual Machine. 22:56:26 Not very major 22:58:15 minix ftw 22:58:30 it's smaller than my unix flavoured kernel 22:58:43 -!- okloduk has joined. 22:58:44 and does something... unlike mine kernel ;p 22:59:10 i've got stupid problem 23:00:30 i'm writing SADOL editor, which should highlight parameters called by function under cursor 23:00:59 but i've got problem finding proper AST node using cursor coordinates 23:01:57 * AnMaster so loves when non-Swedish websites misses åäö when mentioning Swedish things resulting in hilarious results 23:02:12 Resulting in results 23:02:12 -!- M0ny has quit ("When you get sad stop being sad and be awesome instead."). 23:02:30 insulting re insults 23:02:48 -!- oklopol has quit (Connection timed out). 23:03:10 like a piece of music called "Förklädd gud" (translation: God in Disguise) was written as "Forkladd gud" meaning something like "traveled grease god" 23:03:26 though the latter is gramatically incorrect. 23:03:28 :-D 23:03:47 beh 23:03:49 kladd isn't grease really hm. grease is "kladdig" though 23:04:00 kladd is more like sticky/greasy/messy 23:04:22 -!- oklopol has joined. 23:04:25 syrup is "kladdig" too 23:04:28 as an example 23:04:35 well anyway the result was hilarious 23:04:53 -!- GregorR has changed nick to TraveledGreaseGo. 23:05:00 :( 23:05:05 btw, ehird, can you at least support bayes becoming active so it doesn't get deregistered?> 23:05:11 -!- TraveledGreaseGo has changed nick to TravelGreaseGod. 23:05:26 maybe 23:05:31 it'd mess with quorum. 23:05:39 good thing 23:05:45 for you maybe 23:05:59 wow another one in same site: "öron" (translation: ears) was changed to "oron" (translation: the worry) 23:06:02 ehird, ^ 23:06:15 oron öron 23:06:17 the worry of ears 23:06:20 -!- jix has quit ("leaving"). 23:06:28 ehird, err 23:06:32 almost 23:06:36 superkabel ;D 23:06:38 "oron över öron" 23:06:43 would be that 23:06:52 moron over moron? 23:06:54 pah 23:06:59 över meaning over, that being the correct preposition to use in there in Swedish 23:07:21 even latin has such a phenomenon in only a few cases 23:07:32 comex, what phenomenon? 23:08:40 comex, ö is not o with dots in Swedish. The difference is as large as between v and w or, a and e in English 23:08:49 if that is what you meant 23:09:00 it is a completely different letter in the alphabet 23:09:12 pronounced completely differently 23:09:50 http://en.wikipedia.org/wiki/Ö 23:09:51 I see 23:09:59 http://en.wikipedia.org/wiki/Ö#Letter_.C3.96_in_Scandinavian_languages to be specific 23:10:14 comex, the same is true for a vs å and a vs ä 23:10:23 they are three totally different letters 23:10:27 with totally different sounds 23:10:51 what sounds? 23:10:54 http://en.wikipedia.org/wiki/Å#Scandinavian_languages 23:11:15 The short version represents IPA /ɔ/. 23:11:17 comex, wikipedia says: "In Swedish, Ö is pronounced [øː] (e.g. "öl"), [œ] (e.g. "kött") or [ɶ] (e.g. "dörr")." 23:11:18 well that's sure helpful 23:11:21 for example 23:11:38 comex, I can't make a recording atm. People sleeping in next room. 23:11:41 Maybe tomorrow 23:11:43 heh 23:12:08 see also http://en.wikipedia.org/wiki/Ä#As_an_independent_letter 23:12:24 -!- oklopol has quit (Read error: 60 (Operation timed out)). 23:12:40 -!- oklopol has joined. 23:13:18 I'm removing support for multiple players, it just makes everything 10 times more complicated 23:13:21 and ugly 23:13:43 comex, what esolang? 23:14:36 not an esolang 23:14:39 I'm just being off-topic :p 23:15:39 but this is the first real-w-orld thing I've written in haskell 23:15:41 (don't tell ehird) 23:15:51 comex, he reads logs you know 23:15:56 I'm also here. 23:15:57 and you already highlighted him ;P 23:16:00 ehird, and that 23:16:09 But I'm well aware of comex's haskell incompetence. 23:16:33 AnMaster: yep, it's a joke 23:16:47 -!- Judofyr has joined. 23:17:20 The configuration of dvtm is done by creating a custom config.h 23:17:20 and (re)compiling the source code. 23:17:21 great 23:17:33 How dwm. 23:17:43 same in Polish...: łącze -> link, lacze -> flat tires 23:17:47 But seriously, it's not hard to parse an ini. 23:17:48 ehird, I haven't used dwm 23:17:54 AnMaster: it does the exact same. 23:17:58 I see 23:17:58 Seems that dvtm = dwm for framebuffer. 23:18:26 so we need awesome vtm 23:18:27 clearly 23:18:30 robić łaskę -> to show mercy, robić laske -> to do a blowjob 23:18:35 ehird, and tmonad 23:18:54 http://grx.no/kb/2008/08/17/notes-on-setting-up-arch-linux-in-virtualbox-with-awesome-wm/ 23:18:58 This should be helpful. 23:19:02 AnMaster: now -that'd- be overkill. 23:19:21 ehird, with a plugin architecture of course :P 23:19:34 I'm probably going to use pekwm 23:19:58 -!- okloduk has quit (Read error: 110 (Connection timed out)). 23:23:14 ehird, you can mix layouts in one window? 23:23:24 pekwm isn't tiling 23:23:39 I mean in dvtm 23:23:46 hm 23:24:17 You just run dvtm in dvtm. 23:24:28 right 23:24:28 lawl 23:24:37 damn! 23:25:04 TravelGreaseGod it was a non-exact translation anyway 23:25:22 AnMaster: I just thought it sounded funny :P 23:25:35 guys, i've got such thing 23:25:37 dbg: parsing "+12" 23:25:38 dbg: ["+", (0,0), [1, (1,0)], [2, (2,0)]] 23:26:05 (x,y) are coordinates of node in text field 23:27:07 and i'd like to be able to find a node which has specified coordinates without traversing the tree every time 23:27:47 * ehird forgets how to unmount a loopback mount 23:28:04 sudo umount . says it's busy... 23:28:29 i need to map the tree, somehow 23:28:51 ehird, why doesn't my del key work in ghci 23:29:00 comex: because ur doin it rong 23:29:39 it inserts '~' 23:29:44 the correct way to say "traveled grease god" would be "berest smörjgud" 23:29:46 I think 23:29:53 though that is a lossy translation too 23:29:58 err 23:30:11 loosy? lossy? losy? I can never remember that 23:30:15 lossy 23:30:17 thanks 23:31:53 though actually "berest" is more like "well-traveled" 23:32:12 but "for" is more like "yesterday I traveled somewhere" 23:33:47 -!- oklopol has quit (Read error: 110 (Connection timed out)). 23:34:12 anyway important hint for the future: Should you ever need to write a Swedish name, song title or other word: Don't drop any dots. In the best case it will look rather silly. In the worst case it will either be incomprehensible or mean something totally different. 23:35:14 -!- oerjan has joined. 23:35:26 oerjan, hi 23:35:32 hi AnMaster 23:35:47 oerjan, you might want to read the scrollback over silly dropping dots and rings from Swedish words 23:36:00 it is the reason GregorR is now known as TravelGreaseGod 23:36:11 oops 23:36:22 oerjan, it is hilarious 23:37:26 ehhh 23:38:18 ;'pc 23:38:25 my cat walks on my keyboard 23:38:54 * oerjan still wonders how they managed to name SATA 23:39:06 it seems - a little too close to something :D 23:39:21 oerjan: am I missing something 23:39:48 ehird: an N, possibly 23:40:09 Santa? I see. :P 23:40:15 haha 23:40:22 * oerjan swats ehird -----### 23:40:23 ehird, I think he was thinking of satan 23:40:31 * oerjan swats AnMaster -----### 23:40:44 * AnMaster hits oerjan o=========E 23:40:53 oh no, AnMaster _is_ satan 23:40:58 :D 23:41:01 the trident proves it 23:41:05 oerjan, fire poker != trident 23:41:16 i see three prongs, q.e.d. 23:41:32 oerjan, err my fire poker dowstairs *does* have three prongs 23:41:41 yes there is a fireplace in this house 23:42:17 always good to have for tormenting souls, i guess 23:42:21 in fact two of them, and the fire poker at one of them has two prongs, and the fire poker at the other one has three prongs 23:42:25 so I guess it varies 23:42:38 -!- FireFly has quit ("Later"). 23:43:56 ahahaha 23:44:01 ehird, there still? 23:44:05 yes 23:44:08 this is even more hilarios at same site 23:44:15 *hilarious 23:44:21 "önskekonsert" -> "onskekonsert" 23:44:47 meaning? 23:44:55 the former means "wished for/dream concert", the second means "evil concert" 23:45:06 err concert or concerto in English? 23:45:22 :-D 23:45:30 also, both exist. 23:45:37 right 23:45:51 ehird, does both mean the same? 23:45:55 `no 23:45:56 AnMaster: how do you unmount a loop back? 23:46:07 ehird, depends on how you loop mounted it 23:46:20 mount -o loop 23:46:21 mount -o loop /file /directory 23:46:24 then just umount 23:46:25 a concerto is probably a concert that's been too much in italy 23:46:31 oh, I was in the dir 23:46:42 ehird, should work the same? 23:46:50 umount . obviously fails 23:46:59 ah like that 23:47:00 indeed 23:47:09 ehird, if you used losetup and then mount you would need to umount then use losetup to remove it 23:47:34 but with -o loop it is just umount 23:47:42 if you use SATA for solving SAT, then you're really on thin ice 23:47:51 oerjan, SAT being? 23:47:59 satisfaction problem 23:48:01 -!- olsner has quit ("Leaving"). 23:48:07 oerjan, is this a pun 23:48:16 also a school test in the US, i think 23:48:26 * AnMaster waits for "hell frozen over joke" 23:48:29 AnMaster: continuation of previous 23:48:30 'Tis. 23:48:43 The more common of the two college entrance exams. 23:48:56 hell is not frozen over, we have spring here in trøndelag now 23:48:58 school test in UK too 23:49:01 oerjan, ... 23:49:08 different though 23:49:18 ehird: Different test. 23:49:25 yes 23:49:32 -!- KingOfKarlsruhe has quit (Remote closed the connection). 23:49:33 http://en.wikipedia.org/wiki/Hell,_Norway 23:50:11 oerjan, hah 23:50:43 yeah 23:50:45 been there 23:51:01 also in A (with this tiny circle upon) 23:51:08 hilarious 23:51:16 it's less than an hour's drive from trondheim, and just before our airport 23:51:46 and also the north-south trains pass through it, obviously 23:51:47 ah, trondheim 23:51:51 oerjan, with godsexpeditionen too! 23:51:53 :D 23:51:54 with that stupid statue 23:52:04 what stupid statue? 23:52:11 i'm sure we have many :D 23:52:26 that one with this "viking" near the water 23:52:30 bsmntbombdood: has your comp arrived yet? 23:52:41 ehird: newegg says they shipped it 23:52:47 near the water? huh, i don't recall that 23:52:55 erm 23:52:58 sec 23:53:03 there is of course a viking on the central square 23:53:14 being a depiction of the city's founder 23:53:47 it's a bit over the top, really tall 23:54:28 http://www.trondheim-photos.com/?k=view_photo&u=statues-5 23:54:32 this one 23:54:51 does ups work on the weekends? 23:54:59 maybe i've got it on my photos 23:55:34 hm i'm not sure where that is 23:55:48 probably one of the newly built areas 23:55:58 (given the building) 23:56:57 The words "central square" reminded me of a thing that's not really especially funny, but I'll mention it anyway; there was at the central-most spot in Lieksa (a tiny "city" of about 13k people nowadays, pretty close to the Finland-Russia border) this: http://zem.fi/g2/d/1033-2/img_2707.jpg -- the website URL would be, if translated to English, "www.connectiontonature.com". 23:57:38 -!- amca has joined. 23:59:40 -!- amca has quit (Client Quit).