00:04:10 <calamari> /usr/lib/gcc-lib/i386-redhat-linux/3.3.3/../../../crt1.o(.text+0x18): In function `_start':
00:04:10 <calamari> : undefined reference to `main'
00:04:10 <calamari> /tmp/ccACjpyI.o(.text+0x4e): In function `Application::main(JArray<java::lang::String*>*)':
00:04:10 <calamari> : undefined reference to `EsoShellAPI::EsoShell::class$'
00:04:10 <calamari> /tmp/ccACjpyI.o(.text+0x61): In function `Application::main(JArray<java::lang::String*>*)':
00:04:10 <calamari> : undefined reference to `EsoShellAPI::EsoShell::EsoShell[in-charge]()'
00:04:12 <calamari> collect2: ld returned 1 exit status
00:05:55 <graue> I think you need to specify which class has "main" in it, somehow, otherwise gcj doesn't know which main to use
00:08:15 <calamari> thanks, that eliminated the first error, but the 2nd two remain
00:08:33 -!- calamari has quit ("Leaving").
00:10:49 <graue> the OpenBSD package for gcj doesn't seem to work, it wants a "libgcj.spec" file that doesn't exist
00:21:23 -!- pgimeno has quit (Read error: 104 (Connection reset by peer)).
00:37:46 -!- pgimeno has joined.
00:54:39 -!- Aardwolf has quit (Read error: 113 (No route to host)).
01:03:50 -!- naughty_Alice has quit (Read error: 110 (Connection timed out)).
02:12:27 -!- calamari has joined.
02:17:26 <calamari> I guess there's a gcj-4.0 out now.. downloading that :)
02:20:49 <calamari> graue: hmm weird, I have that same spec error here (3.4)
02:21:37 -!- calamari has quit ("Leaving").
02:42:59 -!- cmeme has quit (Read error: 104 (Connection reset by peer)).
02:57:41 -!- wildhalcyon has joined.
03:23:17 -!- cmeme has joined.
04:06:37 -!- wildhalcyon has quit ("leafChat IRC client: http://www.leafdigital.com/Software/leafChat/").
05:34:48 -!- lament has quit (Connection timed out).
06:28:12 -!- lament has joined.
06:36:43 -!- graue has quit ("leaving").
06:47:12 -!- lament has quit (Read error: 110 (Connection timed out)).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:47:15 -!- ZeroOne has quit (Read error: 110 (Connection timed out)).
09:00:35 -!- GregorR has quit (Read error: 110 (Connection timed out)).
09:47:55 -!- lindi- has joined.
11:09:53 -!- ZeroOne_ has joined.
11:41:39 -!- Aardwolf has joined.
11:41:51 <Aardwolf> Someone seen this? http://jonripley.com/i-fiction/games/LostKingdomBF.html
11:43:33 <Aardwolf> The cool thing is that it actually works
11:44:07 <Aardwolf> You are in an unkempt yard overlooking a stagnant pond.
11:44:18 <grim_> nice to see some risc os developers too ;)
11:47:55 <Aardwolf> I'm gonna convert Lost Kingdom to brainloller :D
12:30:50 <Aardwolf> Haha, Lost Kingdom in Brainloller is working, but my interpreter is so slow that you have to wait every time until the texts appear and it responds to your input
12:36:02 <Aardwolf> http://www.student.kuleuven.ac.be/~m0216922/brainloller/lk.png
13:09:31 -!- kipple_ has joined.
13:43:54 -!- jix has joined.
13:52:22 <Aardwolf> How to post a zoomed in version of an image on the wiki?
13:52:34 <Aardwolf> (I mean, a resized version of the image)
13:52:34 <jix> zoomed in?
13:52:53 <jix> you just post the image and the wiki creates small preview images for you (afaik)
13:53:37 <Aardwolf> I mean, here: http://esoteric.voxelperfect.net/wiki/Brainloller is a very small hello world source code image. I'd like to post an enlarged version of it as well (like the width / height tags do in html)
13:56:59 <jix> [[Image:Wikipedesketch1.png|100px]]
13:58:30 <jix> doesn't work...
13:58:38 <jix> http://en.wikipedia.org/wiki/Wikipedia:Picture_tutorial
13:59:27 <jix> i'd upload a scaled version because some brothers uses interpolation for scaling images
13:59:35 <jix> (safari is one of them)
14:04:35 <kipple_> Aardwolf: are you the designer of Brainloller?
14:05:31 <kipple_> I'm doing some small edits to the article, and was wondering when this language was created (I'm guessing this year)
14:06:05 <kipple_> I like to have the year of creation on each language article
14:07:10 <Aardwolf> It was created this year in august
14:07:35 <Aardwolf> did you see this neat brainloller code already? http://www.student.kuleuven.ac.be/~m0216922/brainloller/lk.png :)
14:08:22 <kipple_> was about to ask if it was Lost Kingdom or something, then I saw the text at the bottom :D
14:09:00 <jix> heh maybe i'll write a optimizing bf2bl converter
14:09:06 <jix> that tries to save space
14:09:38 <kipple_> have you uploaded that one to the wiki?
14:10:05 <Aardwolf> no, Lost Kingdom is copyrighted by Jon Kipley
14:10:17 <Aardwolf> I mailed him to ask if I can keep it on my site though
14:10:52 <Aardwolf> By the was I should try to make the brainloller optimizer a lot faster
14:12:52 -!- {^Raven^} has joined.
14:13:13 <jix> Aardwolf: there is a bug in the interpereter ( i think so)
14:13:24 <jix> you check bounds at + and - but not at < and >
14:14:09 <jix> moin {^Raven^}
14:14:59 <Aardwolf> jix, I'll study it, maybe I'll completely rewrite it to make it much faster, inspired by other bf interpreters
14:15:27 <jix> i'm modifying it to dump the brainfuck commands
14:15:53 <jix> then i'm going to use it with bf2a (my bf2c compiler (very fast executables))
14:17:07 <Aardwolf> for a bf2bl converter, check the file in the bftobmp folder
14:17:29 <jix> Aardwolf: the output doesn't look optimzied
14:17:33 <Aardwolf> oh I just reuploaded the file because there was a bug
14:18:00 <Aardwolf> jix, what do you mean? there are as much pixels as brainfuck commands, except two colums of rotate commands on both sides, how can it be optimized?
14:18:12 <jix> Aardwolf: reusing space
14:18:28 <Aardwolf> yeah that is indeed possible, but it looks like a really hard problem to me
14:18:36 <jix> i have some ideas
14:18:38 <Aardwolf> you want to do things like wire crossing to reuse pixels?
14:19:15 <Aardwolf> I had that in mind when I designed it, and secretly hoped someone would actually try using that :)
14:22:20 <jix> hm there is a problem with dumping
14:22:24 <jix> if there is an infinity loop
14:23:32 <Aardwolf> By the way, should I remove the memory size constraint (of 30000 bytes), and would it be better wrapping or non-wrapping?
14:23:44 <jix> reallocating
14:24:11 <jix> i don't know c++ but i think yes
14:24:34 <Aardwolf> Then the contraint would be removed from the brainloller specification and it would be turing complete
14:25:23 <{^Raven^}> hehe, I'm gonna play with brainloller
14:26:03 <Aardwolf> I'm gonna make a new interpreter :)
14:26:11 <jix> i need 128 times the same instruction for doing some optimisations...
14:26:39 <Aardwolf> In the Lost Kingdom code are huge series of red pixels (>'s), maybe you can do something there
14:29:14 <{^Raven^}> run-length encoding of the loaded program will speed up execution dramatically
14:29:27 <jix> Aardwolf: if you want a good bf2high-level optimizer you can try http://esolangs.org/files/brainfuck/impl/bf2a-0.2.rb
14:29:50 <jix> but you have to catch infinity loops if you do optimizations
14:30:18 <Aardwolf> Ok. I think I'm going to learn a lot by trying this.
14:31:03 <Aardwolf> By the way I'm also writing a Deltaplex interpreter. It's extremely hard and complex, I got most of the 3D and audio engine working though.
14:32:35 <jix> are there beta specs/interpreter?
14:33:18 <Aardwolf> There are beta specs, but not online yet
14:33:34 <Aardwolf> the interpreter isn't really an interpreter yet, only the code to support it
14:33:52 <Aardwolf> I'll improve the beta specs and put them online
14:35:45 <{^Raven^}> Aardwolf: take a look at Erik Bosman's interpreter that comes with lost kingdom, maybe you can steal (convert to C++) the optimisation code
14:36:43 <Aardwolf> {^Raven^}, I was planning to do so, as that interpreter runs LK very fast :)
14:37:39 <Aardwolf> I'll start by reading the code from the png image and storing it in a 1D array (following the IP rotate commands), and then use that 1D command for the rest of the BF interpretion
14:37:57 <jix> Aardwolf: uhoh.. infinity loops
14:38:16 <jix> you have to add 2 instructions
14:38:21 <jix> { and } for start and stop infinity loop
14:38:38 <jix> and have to track all ip-positions/directions
14:38:51 <Aardwolf> that's easy, done that for gammaplex already :)
14:39:02 <jix> or better make a new buffer with the size of the image
14:39:21 <jix> that makes the "was i here before?" check faster
14:39:54 <Aardwolf> I also thought about storing the positions of corresponding [ and ]'s to jump to them faster
14:40:07 <Aardwolf> it would have to be stored 4 times, once for each direction
14:40:49 <jix> no use one buffer and use 0b0001 for direction 1 0b010 for 2......
14:42:25 <jix> gcc is too slow...
14:43:19 <jix> and makes to much use of recursion
14:43:23 <jix> Out of stack space.
14:43:23 <jix> Try running 'ulimit -S -s unlimited' in the shell to raise its limit.
14:43:55 <jix> and the code i'm compiling has only 1 function
14:44:23 <Aardwolf> wow I never had problems with g++ heh
14:44:42 <{^Raven^}> Aardwolf: I can't get brainloller to compiler :(
14:45:17 <Aardwolf> Try with g++ *.cpp, it works for me
14:45:32 <Aardwolf> it has no library dependencies at all!
14:45:44 <jix> i'm not talking about BL
14:45:57 <jix> i'm talking about LostKng.c
14:46:05 <jix> autoconverted from LostKng.c
14:46:40 <{^Raven^}> jix: you realise how long gcc will take to compile it ;)
14:47:06 <Aardwolf> So I guess there must be created some very demanding C code? :D
14:47:28 <jix> hmm it has less nested loops than straight bf2c
14:49:01 <Aardwolf> Hey how about this as a .b to .c compiler: the generated .c code is the source code of a .b interpreter and at the start of a code a char buffer being defined with in {, , , ... } symbols the whole .b code. Then the interpreter uses this char buffer (instead of a loaded file).
14:49:43 <{^Raven^}> that only offers the same speed as the bundled interpreter
14:49:59 <jix> this is called pseudo-compiler (used for befunge compiling)
14:50:47 <jix> ruby has no problem with the loop level in LostKng.b
14:50:59 <jix> and i make heavy use of recursion
14:54:59 <Aardwolf> {^Raven^}, did you manage to compile brainloller already? If not, what are you compiling it with?
14:56:47 <jix> hmm what is the correct c label syntax?
14:57:37 <Aardwolf> ask in #c, but afaik it's goto blah; and the label is blah:
14:57:46 <jix> hmm i'm using that
14:59:37 <{^Raven^}> Aardwolf: it turns out that both .cpp files were corrupted on download, it compiles now but segfaults
15:00:23 <Aardwolf> try redownloading them, I recently reuploaded them
15:02:31 <Aardwolf> if you're on linux you can also try downloading a.out and running that
15:08:11 <Aardwolf> I'll reupload the files again, maybe I forgot changing one and they aren't compatible? Because the version on my HD is most certainly working witout segfaults and can run both hello.png and lk.png
15:15:39 <kipple_> I just downloaded it and it worked fine (debian)
15:17:40 <kipple_> though lost kingdoms doens't seem to work properly
15:19:26 <kipple_> perhaps it only works on linux...
15:19:32 <{^Raven^}> jix: goto is the spawn of the devil
15:20:04 <{^Raven^}> kipple_: segfaults on WHEL3 and Windows
15:20:09 <jix> {^Raven^}: yes
15:20:15 <jix> but i have to use it
15:20:27 <jix> because gcc is too crappy to handle a few hundred nested loops
15:20:32 <{^Raven^}> jix: ahh, someone has a gun to your head?
15:20:35 <jix> and it's auto-generated code anyway
15:21:04 <lindi-> jix: how does gcc fail to handle that?
15:21:04 <jix> it looks awful and unreadable without gotos too
15:21:18 <jix> lindi-: catched stack overflow
15:21:32 <jix> aRGH still out of stack space...!!!
15:21:56 <lindi-> jix: ah, when you try to run or compile it?
15:22:00 <Aardwolf> Deltaplex spec and early screenshot are here: http://www.student.kuleuven.ac.be/~m0216922/deltaplex/
15:22:11 <jix> lindi-: yes
15:25:04 <kipple_> aardwolf: looks interesting. though I'm not sure if I'll take the time to read it all ;)
15:29:09 <jix> lindi-: the output of bf2a.rb LostKng.b
15:30:50 <jix> Aardwolf: has gammaplex support for pixel shaders?
15:31:39 <jix> yes delta...
15:31:46 <jix> hmm i like pixel shaders...
15:31:55 <jix> i wrote a julia fractal renderer using pixel shaders
15:32:03 <jix> but i can't use looping
15:32:14 <jix> so the program is too big for my graphic card with 30 iterations
15:32:57 <jix> my card is ok
15:33:11 <jix> ati radeon 9800 pro with 128mb vram
15:33:29 <jix> Aardwolf: has deltaplex support for AA?
15:34:20 -!- kipple_ has quit (niven.freenode.net irc.freenode.net).
15:34:22 -!- {^Raven^} has quit (niven.freenode.net irc.freenode.net).
15:34:27 -!- puzzlet has quit (niven.freenode.net irc.freenode.net).
15:34:28 -!- pgimeno has quit (niven.freenode.net irc.freenode.net).
15:34:28 -!- CXII has quit (niven.freenode.net irc.freenode.net).
15:34:28 -!- cmeme has quit (niven.freenode.net irc.freenode.net).
15:34:28 -!- grim_ has quit (niven.freenode.net irc.freenode.net).
15:34:29 -!- cpressey has quit (niven.freenode.net irc.freenode.net).
15:34:29 -!- ChanServ has quit (niven.freenode.net irc.freenode.net).
15:34:29 -!- jix has quit (niven.freenode.net irc.freenode.net).
15:34:29 -!- ZeroOne_ has quit (niven.freenode.net irc.freenode.net).
15:34:29 -!- lindi- has quit (niven.freenode.net irc.freenode.net).
15:34:29 -!- mtve has quit (niven.freenode.net irc.freenode.net).
15:34:29 -!- Aardwolf has quit (niven.freenode.net irc.freenode.net).
15:36:01 -!- ChanServ has joined.
15:36:01 -!- irc.freenode.net has set channel mode: +o ChanServ.
15:36:03 -!- Aardwolf has joined.
15:36:18 -!- mtve has joined.
15:36:33 <Aardwolf> jix, do you mean full screen anti aliasing?
15:36:57 -!- jix has joined.
15:37:00 <Aardwolf> jix, do you mean full screen anti aliasing?
15:37:23 <Aardwolf> isn't that an option you can turn on in the settings of your video card drivers?
15:37:26 <jix> i could enable it by setting an opengl overwrite on deltaplex
15:37:34 <jix> but native support would be cool too
15:37:37 <Aardwolf> if so, it could screw up the readability of fonts though
15:38:01 <jix> not full screen anti aliasing
15:38:06 -!- kipple_ has joined.
15:38:08 -!- {^Raven^} has joined.
15:38:13 -!- pgimeno has joined.
15:38:13 -!- CXII has joined.
15:38:14 -!- CXII has quit (niven.freenode.net irc.freenode.net).
15:38:14 -!- pgimeno has quit (niven.freenode.net irc.freenode.net).
15:38:15 <jix> that's something inexistent afaik
15:38:19 -!- puzzlet has joined.
15:38:22 <jix> because font aa is taks of the font system
15:38:29 -!- cpressey has joined.
15:38:39 <jix> different from the things SDLgfx aa functions
15:38:46 <jix> so it's part of software for most things
15:38:52 -!- pgimeno has joined.
15:38:59 <jix> but 3d graphics are hardware...
15:39:04 -!- CXII has joined.
15:39:17 <jix> and you can draw triangles with a smooth AA border
15:39:34 <Aardwolf> actually there aren't even plans for lighting at the moment :)
15:39:34 <jix> and they look nicer than those without AA
15:39:44 <jix> and thats something you can toggle using OGL
15:40:21 <Aardwolf> if it's a simple OGL command I could consider it, I'm avoiding extensions at the moment though
15:41:19 -!- cmeme has joined.
15:41:24 <jix> i think it's an extension...
15:42:28 <Aardwolf> one thing I fear is that the interpreter will be too slow to actually allow making something cool
15:43:36 -!- grim_ has joined.
15:50:39 <jix> Aardwolf: deltaplex's constants list is missing a very useless value
15:51:36 -!- ZeroOne_ has joined.
15:51:39 <jix> hmm the jix-useless-constant *g*
15:52:44 <jix> the solution real solution for x for x*|x|^(1/4)+Cos(x^2)-sqrt(5+Sin(9+x))=0
15:54:09 -!- wildhalcyon has joined.
15:54:46 <jix> it's ~ 2.056964219114166926775407552183364454808154333338709547536597358325935202358797284252482781728389933848895663035350228179996536714334290
15:55:57 <jix> i can't imagine a problem that needs this value
15:57:17 <jix> and Mathematica's solve wasn't able to solve the equation ^^
15:57:36 <jix> no i don't have maple
15:58:13 <jix> it said there is no algebraic way to solve it
15:58:37 <jix> there is no algebraic way to solve all equations of 5th order or higher.. so it's nothing special
15:58:56 <Aardwolf> gmail is suddenly bugged, I can't send emails anymore with it :s
15:58:58 <wildhalcyon> Some +5th order equations are algebraically solvable
15:59:17 <jix> wildhalcyon: i said there is no way to solve ALL of them some are solvable but not all
15:59:30 <jix> add a '.' anywhere in my last msg
15:59:51 <jix> i used Newton's Method to near the result
16:00:23 <wildhalcyon> That'll work, how long did it take to get your answer so.. erm.. extremely close?
16:00:51 <jix> i don't know i did every step by hand
16:01:08 <jix> but i don't see mathematica calculating the steps
16:01:14 <wildhalcyon> Aardwolf, what about adding Polya's random walk constants?
16:02:03 <Aardwolf> What could they be used for? :D
16:02:36 <Aardwolf> The physical constants are truly useful if you want to make a space shooter in deltaplex :)
16:03:05 <wildhalcyon> Umm.. well, I guess they're not TERRIBLY useful, but polya's constant for n=1 is 1, so that could be a quick way to get "1"
16:10:40 <wildhalcyon> well. the higher ones aren't very practical for many purposes
16:11:10 <jix> i have a new cool constant!
16:11:19 <jix> srand('deltaplex'.sum);puts "0."+(0..100).map{|e|rand(10)}.join
16:11:26 <jix> replace 100 with a higher value for more digits
16:11:39 <jix> oh it's ruby code
16:17:41 <jix> brainfuck beats them all! http://www.googlefight.com/index.php?lang=en_GB&word1=brainfuck&word2=c+lisp+scheme+perl+php+python+ruby+bash
16:19:29 <jix> <GregorR> Also, anyone who uses the term "pwnd" in my MUD engine will find their character mysteriously dead.
16:25:07 -!- CXI has joined.
16:26:07 -!- wildhalcyon has quit (Read error: 104 (Connection reset by peer)).
16:26:28 -!- CXII has quit (Read error: 54 (Connection reset by peer)).
17:00:24 -!- sp3tt has joined.
17:28:56 <jix> hah 175 bytes code!
17:29:03 <jix> my shortest brainfuck interpreter
17:30:41 <jix> argh eof handling...
17:30:48 <jix> should i use 0 or -1?
17:40:37 <jix> 177 bytes code
17:41:15 <Aardwolf> 177 bytes code in what language?
17:41:31 <jix> 178 for EOF == -1
17:42:06 <jix> $>.sync=m="\0";j=0;eval$<.read.tr(x='^[]+><,.-',"").gsub(/./){%w{while\ m[\
17:42:06 <jix> j]>0 end m[j]+=1 (j+=1)>=m.size&&m<<0 j-=1 m[j]=STDIN.getc||0 putc\ m[j]
17:42:06 <jix> m[j]-=1}[x.index($&)-1]+";"}#esoteric@irc.freenode.net Jannis Harder 2005#
17:43:16 <jix> it's a bit faster than obi.rb
17:44:28 <mtve> jix: http://www.kernelpanic.pl/perlgolf-view.mx?id=34
17:44:44 <jix> eek perl...
17:46:03 <mtve> and http://www.perlmonks.org/?node_id=132368
17:47:31 <jix> but that's perl
17:47:47 <jix> perl is more cryptic than ruby
17:47:54 <jix> normal ruby is clean and readable
17:49:04 <jix> and there are more perl than ruby golfers!
18:23:02 -!- clog has joined.
18:23:02 -!- clog has joined.
18:23:03 -!- cpressey has joined.
18:23:03 -!- tokigun has joined.
18:23:16 -!- mtve has joined.
18:23:59 -!- Aardwolf has joined.
18:26:39 -!- jix has joined.
19:21:27 -!- lindi- has joined.
19:29:41 -!- {^Raven^} has quit (Read error: 145 (Connection timed out)).
19:33:33 <kipple_> I've made a graph of the size of the wiki based on the database dumps: http://rune.krokodille.com/lang/wikisize.php
19:34:29 <kipple_> it has actually become smaller lately...
19:37:44 <jix> i need a shorter esolang interpreter
19:37:50 <jix> 177 bytes is too large
19:38:19 <Aardwolf> what's the shortest one every written?
19:39:01 <jix> Aardwolf: for brainfuck there are shorter ones than mine
19:39:21 <kipple_> there are some very short C bf interpreters
19:39:25 <jix> i'm talking about a "suitable for computation" language
19:39:30 <jix> kipple_: shorter than 177 bytes?
19:39:38 <jix> there are shorter perl interpreters for it
19:40:12 <kipple_> perhaps not. haven't counted the bytes
19:40:37 <kipple_> a higher level language like perl should be able to do it shorter than C
19:40:52 <jix> there is a ~90 byte perl version
19:41:32 <jix> but i think a ruby version that is shorter than 177 bytes (i can do it in 176 bytes if i wouldn't limit source lines to 75chars)
19:41:40 <jix> is very very hard
19:42:57 <kipple_> why limit source line length?
19:43:12 <jix> because i like to use small ruby programs as my sig
19:43:28 <kipple_> do you have a link to the perl version?
19:43:37 <jix> 18:34:34<mtve>and http://www.perlmonks.org/?node_id=132368
19:43:55 <jix> its 88 bytes
19:43:57 <jix> $_.=qw(++ }while$ -- ++$ $$p=shift; --$ print$ ${)[$p%60%8].'$p+0;'while$p=ord getc;eval
19:44:35 <jix> i try to write the shortest befunge 93 interpreter possible (in ruby)
19:47:52 <Aardwolf> how can that be a bf interpreter :o
19:48:12 <kipple_> I just tried it. doesn't seem to work
19:50:26 <kipple_> on helloworld.b it outputs 72101108108111-18703-3-110
19:50:42 <jix> maybe it needs an older / newer perl version
19:50:52 <jix> but my interpreter works
19:51:40 <kipple_> at first it looks like just the ASCII values for hello world, but then something happens...
19:52:33 <jix> that's the problem with perl.. as soon as your program that worked for ages stops working you're lost and have to rewrite it because you can't read the source code... that wouldn't happen with BF
19:52:42 <jix> because BF doesn't change
19:53:01 <Aardwolf> C and C++ programs will work forever too
19:53:09 <kipple_> except for EOF, cell size and wrapping
19:53:41 <Aardwolf> The std:: thing wasn't there at the beginnings of C++
19:53:52 <jix> kipple_: yes but that are things you think about when you write your program...
19:54:02 <jix> you don't think about operator precedence if you write a perl program
19:54:24 <kipple_> but how do you write a bf-program that takes into account all variations of EOF?
19:54:35 <kipple_> jix: I don't write perl programs ;)
19:54:41 <Aardwolf> hey, what is the EOF issue about? End Of File? Does it mean end of program here?
19:54:51 <jix> Aardwolf: end of stdin
19:55:01 <jix> kipple_: i stopped writing perl programs a long time ago
19:55:27 <Aardwolf> hmm, what is the end of stdin things you type in a linux console?
19:56:02 <jix> Aardwolf: that's something special not an ascii value..
19:56:31 <jix> in c you have feof for checking for eof... because getc's return value isn't usable if you work with binary data
19:56:51 <jix> in BF you don't have feof.. and you have to use one of the characters as EOF marker
19:56:52 <Aardwolf> I've never had the need to check for eof in anything I programmed :s
19:57:03 <Aardwolf> what kind of EOF does Lost Kingdom use?
19:57:14 <jix> does Lost Kingdom use EOF?
19:57:47 <jix> programs that read files
19:58:32 <kipple_> there are 3 versions of cat in the Brainfuck article on the wiki. one for each common EOF method
19:58:46 <Aardwolf> I have actually no idea which EOF brainloller uses, I never thought about it while making the interpreter lol
19:59:02 <jix> if you use getc it probably uses -1
19:59:11 <jix> == 0xFF (mod 256)
19:59:29 <kipple_> I read the C spec and it seems like EOF can be ANY negative value...
19:59:44 <jix> kipple_: but probably it's -1
20:13:38 -!- kipple_ has quit (Read error: 104 (Connection reset by peer)).
20:14:02 -!- kipple has joined.
20:23:04 -!- sp3tt has quit (Client Quit).
20:26:54 -!- wildhalcyon has joined.
20:47:20 -!- jix has quit ("Bitte waehlen Sie eine Beerdigungnachricht").
20:48:59 <Aardwolf> what is brainfuck supposed to do if you do < so many times that the pointer will get a negative value?
20:50:39 -!- jix has joined.
20:54:40 <wildhalcyon> I believe the original spec left that behavior undefined
20:57:34 <jix> most interpreters give an error (a wanted error msg or an unwanted segfault)
21:00:05 <jix> there are not enough programming language that use sound files as source code
21:01:15 <wildhalcyon> I don't disagree, but the ones that do don't seem to do so in a particularly novel way...
21:01:57 <jix> one could try to detect frequencies...
21:02:54 <wildhalcyon> I was thinking more along the lines of beats
21:03:27 <wildhalcyon> I dont supose that would work well with violin or guitar for instance
21:04:19 <wildhalcyon> If you were to use frequencies, how would you.. differentiate them?
21:05:22 <wildhalcyon> Well, right, but would you bin them, or have a fixed-window FFT?
21:05:41 <jix> i know NOTHING about FFTs... so i don't know..
21:06:30 <wildhalcyon> Well, the size of the FFT window is related to the resolution of the frequencies that it can observe
21:06:41 <jix> it's possible with wavelets but i think you need some special transforms for that.. because with a normal transform you can only differentiate between octaves
21:07:40 <wildhalcyon> Really? AFAIK wavelet transforms can be used for frequency detection as well, especially with a very discretized (2~4-bit) wave
21:08:12 <jix> yes but i only looked at wavelet transformations for compressing
21:08:56 <jix> and the high-pass data and low-pass data are spitted at a frequency that is half the sampling rate
21:11:51 <wildhalcyon> It would be interesting to see a language that implemented loops as actual sound loops
21:14:13 <wildhalcyon> You could model it after genetic code - tonal patterns would mark different commands
21:19:39 -!- GregorR has joined.
21:20:57 <jix> ah continuous wavelet transform seems to be the thing needed for this...
21:21:36 <wildhalcyon> In other news, I think I solved the topology problem that was plaguing my funge-variant.
21:21:53 <jix> i only used discrete wavelet transforms
21:22:59 <wildhalcyon> Well, a discrete Haar wavelet transform should work for CD data
21:37:42 -!- jix has quit ("Bitte waehlen Sie eine Beerdigungnachricht").
22:15:44 <Aardwolf> Hey, a turing complete language is never able to generate a true random number, while a quantum random number generator can. Is a program that uses such a thing then "higher" than turing complete?
22:19:53 <lindi-> what's true random number exactly?
22:23:22 -!- {^Raven^} has joined.
22:31:57 -!- wildhalcyon has quit (Read error: 104 (Connection reset by peer)).
22:40:30 <{^Raven^}> Aardwolf: Is any particular version of C++ required to compile brainloller? (I have GCC 3.23)
23:02:40 <Aardwolf> {^Raven^}, no not really, it should be all standard C++. I uploaded a very new changed version though, maybe it works now??
23:05:01 -!- grim_ has changed nick to grim_bed.
23:25:19 -!- wildhalcyon has joined.
23:48:18 <cpressey> it would not be difficult to define a language which is both turing-complete and can generate truly random numbers... take a turing machine and add a "randomly move head left or right w/equal probability" code, for example.
23:54:02 <cpressey> that machine is turing-complete, meaning, it can do everything a regular turing machine can do. however, if you mean to say that a regular turing machine cannot do everything that that machine can do - that would be correct (modulo the definition of "truly random")
23:55:25 <cpressey> or... another, more succinct way to say what i mean:
23:56:00 <cpressey> unless you believe that *everything* can be reduced to computation, then there are *lots* of things that are "higher than turing-complete" - just look around you :)
23:59:39 <Aardwolf> Wow, Lost Kingdom has a very nasty HUGE loop (the [ and ] very far apart) and now that I set some optimizer setting in my interpreter to let it work with such huge loops, LK suddely goes a lot faster