00:20:39 -!- ZeroOne_ has changed nick to ZeroOne.
00:26:06 -!- calamari has joined.
00:48:14 -!- Flannel has joined.
00:49:34 -!- Flannel has left (?).
00:53:00 -!- kipple has quit (Read error: 104 (Connection reset by peer)).
00:58:47 -!- kipple has joined.
01:16:27 -!- wildhalcyon has quit ("Whoops, There I go").
01:28:28 -!- Aardwolf has quit ("Ik zen der is mee weg").
01:30:58 -!- calamari has quit ("Leaving").
03:10:37 -!- wildhalcyon has joined.
03:41:47 -!- kipple has quit (Read error: 110 (Connection timed out)).
04:10:29 -!- wildhalcyon has quit (Read error: 104 (Connection reset by peer)).
04:22:18 -!- calamari has joined.
05:48:38 -!- wildhalcyon has joined.
06:03:57 -!- wildhalcyon has quit ("Whoops, There I go").
06:32:48 -!- nooga has joined.
06:39:44 <calamari> well, EsoShell doesn't work with gcj.. does compile tho! :)
06:43:52 <nooga> i don't believe in such things
06:44:50 <nooga> java was designed to produce it's own bytecode
06:46:02 <nooga> i just saw the jsux, do you know what's it?
06:47:36 <calamari> I have no idea what you're talking about, I feel okay about that in this channel, tho :)
06:48:23 <nooga> jsuix is a javascript unix implementation :)
06:48:45 <nooga> UNIX-like OS written in JavaScript
06:48:48 <nooga> http://www.masswerk.at/jsuix/
06:51:03 <nooga> click "open terminal"
06:51:19 <nooga> it's amazing, they even have ported vi
06:53:52 -!- Anulka has joined.
06:54:51 -!- Anulka has changed nick to nooga_.
07:01:51 -!- nooga__ has joined.
07:07:41 <lament> nooga__: what you meant to say was
07:07:45 <lament> "java was designed to be slow"
07:07:51 <lament> a fact which i can't really agree with :)
07:11:17 -!- nooga has quit (Read error: 110 (Connection timed out)).
07:19:04 -!- nooga_ has quit (Read error: 110 (Connection timed out)).
07:25:21 -!- nooga__ has quit (Read error: 110 (Connection timed out)).
07:47:54 -!- calamari has quit (Read error: 110 (Connection timed out)).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:19:55 -!- nooga has joined.
08:26:30 -!- Keymaker has joined.
08:26:54 <Keymaker> the gammaplex inventor is not here, right? :(
08:27:40 <Keymaker> anyways, perhaps someone else could help me; how do it store the stack's top value to something register?
08:56:25 * nooga is trying to write a SADOL compiler for x86
09:58:45 -!- Aardwolf has joined.
10:01:19 <Keymaker> you were the gammaplex author?
10:01:30 -!- jix has joined.
10:01:49 <Keymaker> i just 10 seconds ago completed my first ever gammaplex program:
10:02:02 <Aardwolf> I'll check out what it does :)
10:02:17 <Keymaker> it's nothing special but it was my first :)
10:02:25 <Aardwolf> btw jix, Brainloller can't have infinite loops, the IP rotarors are reversible, and the IP starts at top left ;)
10:03:46 <jix> Aardwolf: ahh
10:03:53 <nooga> im' writing the "hunt the wumpus" game in SADOL :D
10:04:25 <Aardwolf> gotta recompile the gammaplex interpreter
10:04:37 <jix> Keymaker: cool
10:04:56 <jix> it does more than my first gammaplex program
10:04:59 <jix> it works...
10:05:56 <jix> who has write access to the esolang file archive?
10:06:41 <Aardwolf> it makes the screen blue if you click on it? :)
10:08:04 <Keymaker> who will help me with my 3d engine project in gammaplex?
10:08:27 <Aardwolf> that's what deltaplex is for, if I ever manage to finish it :)
10:08:45 <Aardwolf> drawing textured 3D triangles with only a few commands
10:09:11 <Aardwolf> there's even a terrain engine in it which is already finished :)
10:09:30 <Keymaker> woah.. it'll be something never-seen-before
10:09:32 <Aardwolf> you just give a heightmap, a texture, a colormap, some coordinates, and the drawterrain command :)
10:10:58 -!- nooga has quit.
10:12:48 <Keymaker> aardwolf; how do i get random numbers for example between 0 and 50?
10:17:01 <Keymaker> i think we should run a gammaplex demo competition
10:18:51 <jix> well why not?
10:20:29 <Keymaker> should there be something theme, limitation or other?
10:20:42 <Aardwolf> hehe yeah, a limitation in code size maybe
10:21:19 <Keymaker> and probably non-interactive demo as well?
10:21:25 <Aardwolf> can someone do better than tron3k's pong? :)
10:22:01 <Aardwolf> non interactive demos would be cool but maybe a bit slow on gammaplex
10:22:14 <Aardwolf> after all, drawing a single mandelbrot is already taking a while
10:22:22 <Keymaker> probably not slower than intereactive
10:22:37 <Aardwolf> yeah but in a non interactive the graphics are a bit more important :)
10:22:48 <Aardwolf> in interactive you can get away with two paddles and a ball :D
10:23:10 <Keymaker> i meant with "interactive" stuff that user can click buttons and something happens
10:23:24 <Keymaker> and with "non-interactive" a demo that just runs and ignores all clickings
10:23:46 <Keymaker> but in every case the demo should be not a still picture
10:23:59 <Aardwolf> something like those demos (Unreal and Second Reality and such) would be too slow :(
10:24:24 <Keymaker> it won't need to be anything very awesome
10:24:41 <Keymaker> i mean for example a bouncing ball would be neat already
10:25:56 <Aardwolf> Has anyone ideas how I can optimize the interpreter? Also it has a huge switch case because there are so many commands, I wonder if that is taking away a lot of speed?
10:26:57 <Keymaker> btw; it will take about two or three weeks until i have properly spare time for this
10:27:05 <Keymaker> but then i guess we could start a competition
10:27:47 <Keymaker> just wondering how we would get 'nuff entries
10:28:29 <Aardwolf> hmm I wonder if the demo scene can be attracted to gammaplex
10:28:41 <Aardwolf> You know, Deltaplex would be so much better for this sort of stuff with it's 3D graphics :)
10:29:04 <Aardwolf> except it's harder to code in it
10:29:09 <Aardwolf> because the commands are on an image
10:30:23 <Keymaker> i've got one idea you could possibly do
10:30:45 <Keymaker> for example make all the deltaplex files to be some dpx packages
10:31:00 <Keymaker> and some program to stuff all the textures, code and images into a dpx package
10:31:09 <Keymaker> then the interpreter reads the package and so on
10:31:20 <Keymaker> and the code could be 2d txt like in gammaplex
10:31:22 -!- kipple has joined.
10:31:35 <Keymaker> and it'd be a lot easier to handle all the textures and stuff if they were separate files
10:31:43 <Aardwolf> hmm but that would remove something I find so nice, the textures and code being together on an image
10:31:59 <Aardwolf> I have another idea, make a sort of IDE that will put it all on the image
10:32:32 <Keymaker> otherwise it can be rather difficult to do anything
10:32:42 <Aardwolf> apart from paint shop pro, no painting program is good enough to do it
10:32:50 <Aardwolf> and PSP doesnt work in linux :cries:
10:33:07 <Aardwolf> (only PSP shows the RGB color if you go with the mouse over the pixel)
10:33:31 <Aardwolf> and that's important to know which command the pixel is
10:33:56 <Aardwolf> I've been thinking about making a simple painting program that would do this, but I'm stuck in QT
10:36:52 <Aardwolf> I could make the painting program in OpenGL tho, using my OpenGL gui :)
10:36:52 <Aardwolf> but that would be weird, not? A IDE/painting program in OpenGL?
10:36:52 <Aardwolf> have you got any suggestions for new gammaplex commands?
10:37:42 <Keymaker> i was thinking about deltaplex
10:40:13 -!- J|x has joined.
10:42:31 <Aardwolf> it's an infinite loop to make the program stop, a sort of fake way to end
10:42:46 <Keymaker> when does the program stop, by the way?
10:43:11 <Aardwolf> there's a close button and a menu in the interpreter to quit
10:43:28 <Keymaker> but no "terminate program" command?
10:43:50 <Keymaker> ok. i was already wondering why i couldn't find one
10:44:10 <Aardwolf> when I look back at the spec, the { and } commands are really weird, and I should have let it store the working color and working coordinates in the registers (and should have called the registers the memory instead :))
10:44:24 <Aardwolf> and I should have allowed memory of arbitrary size
10:47:58 <Aardwolf> I'm going to try to improve the gammaplex interpreter later, and will remove the memory limit then (but not now yet)
10:48:09 <Aardwolf> I mean, backwards compatibility isn't lost by removing the limit :)
10:49:21 <Aardwolf> by the way I really like the mandelbrot renderer code because it has a subroutine using gosub and return
10:50:12 -!- jix has quit (Read error: 110 (Connection timed out)).
10:50:24 -!- J|x has changed nick to jix.
10:51:26 <Keymaker> i'll write a maze game in gammaplex
10:51:52 <jix> i'll write something cool in gammaplex
10:52:09 <jix> but first i'll complete the subskin wiki page
10:52:13 <Aardwolf> I'll write tetris or breakout when I have the time :)
10:52:45 <Aardwolf> tetris is in the suggestion list of gammaplex programs if I remember correctly :)
10:53:51 <Keymaker> snake/worm game would be nice too
10:54:40 <Aardwolf> or something like Elite!!!! :)
10:54:51 <Aardwolf> but that would probably be too slow :)
10:55:47 <Aardwolf> one of the first 3D space shooters, featuring wireframe ships, trading goods to earn money and buy better ships, pirates and so on :)
10:56:33 <jix> i'll port the TI 92+?/89/v200 pheonix (platinum)? game
10:56:51 <jix> its written in m68k asm
10:59:04 <Aardwolf> looks like a neat game http://www.ticalc.org/archives/files/ss/48/4815.gif
11:00:03 <jix> hmm but i can't read m68k asm..
11:01:16 <Keymaker> then you have to make a totally new one
11:01:17 <jix> but the game is the best game for ti calcs
11:01:19 <Aardwolf> but you can try to copy the gameplay
11:01:34 <jix> Aardwolf: yeah
11:02:23 <jix> but first i've to finish this stupid subskin wiki article
11:03:13 <jix> haha my friends always ask me what i'm going to do today and i always tell them i'm going to write programs the whole day..
11:03:56 <Aardwolf> I say something similar to my friends often too
11:05:13 <jix> i think they would be shocked if i say "i'm going somewhere to play football because the weather is nice" ;)
11:06:34 <Aardwolf> in brainfuck, if you type input and you type something of multiple letters, will each next , command in brainfuck use these next letters? (I'm just trying to figure out how the C getchar() command works in fact ;))
11:07:09 <jix> everything you write to STDIN is written to a buffer
11:07:19 <jix> its pushed to it
11:07:25 <jix> and getc shifts one char out
11:07:33 <jix> (queue not stack)
11:33:46 <Keymaker> by the way, is gammaplex file extension .txt?
11:33:56 <Keymaker> and, if you pop an empty stack, it returns 0?
11:39:34 <Aardwolf> hmm let me check the code for that last question
11:40:10 <Keymaker> nope, but i mean should i use .gpx on my site..
11:41:00 <Aardwolf> it's only a very little bit harder to open with a text editor :)
11:42:18 <Aardwolf> I implemented the gammaplex stack in a sort of weird way, but it'll probably return 0 unless it's memory is full
11:42:34 <Aardwolf> I hadn't discovered std::vectors back then hence the dumb memory limits and weird stack :)
11:47:56 <jix> done: http://esolangs.org/wiki/Subskin
11:52:40 <jix> Aardwolf: yes
11:52:55 <jix> but thats nothing special.. there were some OISC implementations before subskin
11:53:08 <jix> but i don't know any that uses my instruction
11:53:19 <Aardwolf> this certainly is short ruby code for an interpreter :)
11:54:22 <Aardwolf> but didn't my professor Object Oriented Programming warn us that one letter variable names are bad? :D
11:54:26 <jix> i had the subskin idea (and name!) a long time ago but never implemented it.. yesterday i searched a language for writing a super short interpreter for...
11:54:39 <jix> Aardwolf: haha
11:55:17 <Keymaker> i don't understand it fully however.. :\
11:55:38 <jix> i use them often in ruby... but only in a scope that has a length of... 1 line... ["h4","37"...].map{|i|i.hex} for example..
11:56:40 <Aardwolf> I use them often too, except in tasks for this professor heh
11:57:32 <jix> the only other name for i that would fit is iterator_variable... and thats too long because you need such kind of variables every 10th line
11:57:35 <Keymaker> aardwolf: the second rgb is 0 0 0 in the beginning in the gammaplex..?
11:58:28 <Aardwolf> You know what's sort of stupid, the gammaplex specification is based on the interpreter sometimes instead of the opposite
11:58:57 <jix> subskin is spec first...
12:00:02 <jix> the Hello, world! subskin example is self modifying...
12:01:12 <jix> every a-bit-more-complex subskin program has to use self modification
12:01:48 <Aardwolf> yeah the second color is initially 0 0 0 appearantly
12:02:21 <Aardwolf> that must make it pretty hard to program in :)
12:03:49 <jix> with a text editor it is
12:04:04 <jix> with a assembler like thing it's easier
12:04:55 <Aardwolf> hehe this is always fun, first write the complete interpreter in C++, when done, compile it for the first time and see how many errors there are
12:06:06 <Aardwolf> image based, and about as complex as befunge
12:06:22 <jix> http://en.wikipedia.org/wiki/Mycelium
12:06:24 <Aardwolf> mycelium has something to do with mushrooms too btw :)
12:06:44 <Aardwolf> all befunge clones should have a mushroom like name ;)
12:07:27 <Aardwolf> now I gotta make some mycelium programs to see if the interpreter works correctly
12:25:25 -!- grim_bed has changed nick to grim_.
13:13:26 <Keymaker> aardwolf, does the 'i' command print the number to locations in x and y?
13:19:27 <Aardwolf> it's good that you ask those questions, it'll allow me to improve the spec some time
13:47:04 <pgimeno> jix, I have access to the repository
13:47:23 <Aardwolf> When typing something in console and pressing enter, is this enter character 10 or 13?
13:48:07 <pgimeno> 10 when getchar() receives it, but that happens in Unix only
13:48:30 <Aardwolf> I receive it with getchar(). How to make it platform independent?
13:48:34 <pgimeno> in Windows getchar() receives a 13 then a 10
13:49:01 <pgimeno> I think so, maybe jix can give a definitive answer
13:49:34 <Aardwolf> I wonder if Lost Kingdom checks for both 10's and 13's
13:50:10 <pgimeno> I think that the language specifications free you of those problems
13:51:42 <pgimeno> so even in Windows the language gets always 10 after a carriage return
13:51:48 <Aardwolf> I wonder if they did it on purpose, each choosing a different way to represent enter (mac, win and unix)
13:52:07 <Aardwolf> Now all that's missing is an OS that does 10 first and then 13 :rolleyes:
13:52:08 <pgimeno> I suppose there are historic reasons
13:52:36 <Aardwolf> is the reason that ascii was not well defined? I mean maybe the carriage return and newline thing was too confusing?
13:54:00 <pgimeno> yeah, that's probably the reason
14:02:07 <pgimeno> in some old typewriters a carriage return was to move the carriage horizontally and then you needed to press line feed, which rotated the drum
14:02:47 <pgimeno> there's not an end-of-line character well-defined in ASCII
14:03:14 -!- Keymaker has quit ("I've seen this dj vu before..").
14:03:37 <pgimeno> the ZX Spectrum (a Z80-based micro) took CR as end-of-line too
14:04:26 <pgimeno> it's CP/M the one which took the CR/LF combination, later inherited by DOS
14:08:03 <jix> mac os 9 (oooold) uses 13 as return
14:08:11 <jix> mac os x uses 10
14:09:45 <pgimeno> which is like asking, would it be painful if Windows switched to 10 too?
14:10:33 <jix> pgimeno: mac os x is another system .. not the next version of mac os 9
14:10:42 <jix> so you can't compare it with windows
14:11:05 <jix> mac os x has some compatibility libs and binary loaders to allow the user to run some old apps
14:12:10 <jix> but mac os 9 was a really old kernel.. there was no command line no STDIN no STDOUT no posix at all cooperative multi tasking resource forks... mac os x is based on NextStep and BSD
14:13:47 <jix> mac os x uses the Mach kernel (mach started as a fork of the bsd kernel but got rewritten) used by Next. Most cli programs and parts of c libs are from BSD afaik and the whole Objc thing + Cocoa is just a newer version of NextStep
14:14:26 <jix> all cocoa classes start with NS for NextStep (NSWindows NSArray NSString NSDictionary....)
14:15:11 <jix> apple buyed next
14:15:23 <jix> some people say it was the other way around
14:16:44 <Aardwolf> I tried windows longhorn beta, and was wondering what was actually changed except a nerf in the file searching system
14:16:47 <jix> because steve jobs (one of the apple founders and ceo at some time and today) got fired by apple and founded Next.. Next had no money and apple buyed it... but steave jobs is now apple ceo and apples software is based on next...
14:16:54 <pgimeno> nice job the Subskin page, btw
14:17:06 <jix> so it was like NeXT buying apple
14:17:20 <jix> pgimeno: ah about the file archive
14:17:32 <jix> i'd like to submit my bfx.rb brainfuck interpreter
14:18:09 <jix> and i'd like to have write access too because i often write interpreters for languages
14:18:19 <pgimeno> except I'm having lunch right now
14:18:37 <pgimeno> do you know how to use svn?
14:21:23 <jix> i used it once to download source code.. seems to be pretty ease (not like cvs)
14:21:23 <pgimeno> it's very CVSish, just properly done
14:21:23 <jix> i know it works like cvs.. but cvs is not as easy as svn because of cryptic commands and and and...
14:21:23 <jix> if i see svn commands i can guess what they do
14:21:23 <jix> if i see cvs commands i just hope they don't act like rm -rf /
14:21:23 <pgimeno> anyway the cvs/svn up / cvs/svn commit cycle is similar in both
14:21:52 <jix> i want to learn how to use svn anyway
14:22:26 <pgimeno> it's easy enough for you, I guess
14:23:36 <pgimeno> meaning, I don't expect you having the trouble many people have with it
14:25:18 <jix> one thing i don't like about svn is that under osx it detects the UI language instead of don't finding a local and using english as default
14:25:24 <jix> i want gui in german
14:25:26 <jix> but cli in english
14:27:51 <jix> graue was active on the wiki yesterday but wasn't here yesterday...
14:28:31 <pgimeno> what are your locale settings? I use English CLI
14:28:41 <jix> i have no locale settings
14:28:54 <jix> it isn't set on mac os x by default
14:29:27 <jix> but svn is the first cli program that uses the gui lang if it doesn't find a locale
14:29:27 <pgimeno> then you can write a wrapprt
14:30:30 <jix> edited my environment.plist
14:35:13 <pgimeno> I'm ready, do you want me to upload the file?
14:47:02 <pgimeno> I've taken the code from the wiki (removing the LF) and I'm ready to commit
14:49:31 -!- wildhalcyon has joined.
14:50:18 <pgimeno> added cat.subskin and hello.subskin, ready to commit too
15:00:45 <{^Raven^}> Aardwolf: Lost Kingdom checks for and accepts both CR and LF for line endings
15:03:05 <{^Raven^}> Aardwolf: If the console returns CRLF or LFCR as a line ending the second byte is treated as a null command
15:11:47 <{^Raven^}> Aardwolf: RISC OS return 10 for Enter
16:18:34 -!- nooga has joined.
17:03:13 <Aardwolf> I'm busy testing and debugging the Mycelium interpreter
17:08:41 <Aardwolf> It's doing weird though, there are no floats and no doubles in my interpreter, yet I get a "Floating point exception"
17:17:39 -!- lindi- has quit (Read error: 110 (Connection timed out)).
17:19:29 <nooga> anyone from London here?
17:27:37 * {^Raven^} is in Sheffield - not *that* far away
17:28:11 <jix> pgimeno: bfx.rb isn't on the wiki
17:28:15 <jix> but thanks for adding subskin
17:28:27 <nooga> i just saw the Battersea Power Station on Pink Floyd's cover
17:28:50 <jix> pgimeno: can i dcc bfx.rb to you?
17:28:54 <nooga> this building looks just awesome
17:32:34 <nooga> http://upload.wikimedia.org/wikipedia/en/thumb/9/9e/Battersea_Powerstation_-_Across_Thames_-_London_-_020504.jpg/800px-Battersea_Powerstation_-_Across_Thames_-_London_-_020504.jpg
17:41:17 <nooga> maximally surrealistic eh?
17:41:30 <nooga> i'd love to go there and make some photos
17:41:43 -!- nooga has quit.
17:48:23 <jix> i've written skasm (SubsKin ASseMbler).. it's a big ruby hack but it works
18:23:09 <pgimeno> jix: is subskin.rb an appropriate name?
18:24:30 <pgimeno> re bfx.rb: dcc is ok, what is it supposed to be?
18:26:01 <jix> bfx.rb is the shortest ruby brainfuck interpreter
18:26:08 <pgimeno> (dcc is ok as long as I'm not afk, of course)
18:26:09 <jix> and yes subskin.rb is the name i use
18:26:35 <pgimeno> so brainfuck/impl looks like a sensible place, right?
18:28:04 -!- Aardwolf has quit ("Ik zen der is mee weg").
18:34:17 <pgimeno> commited both - now you'll have to wait for the archive copy to be updated or look at the repository directly
18:38:10 <pgimeno> the direct link to the repository is http://esoteric.voxelperfect.net/svn/esofiles/ but graue discourages using it
19:16:54 -!- kipple has quit (Read error: 104 (Connection reset by peer)).
19:17:08 -!- kipple has joined.
20:20:21 <jix> my old hello world implementation was stupid...
20:22:42 <jix> graue was on the wiki today
20:22:47 <jix> why isn't he here?
20:23:56 <jix> he fixed my Subskin article
20:24:23 -!- Aardwolf has joined.
20:24:56 <jix> hrhr it's good to have someone fixing the spelling/grammar of the wiki articles
20:32:38 <jix> here is a trace of the new hello world program: http://www.harderweb.de/jix/hello-world-trace.html
20:37:26 <Aardwolf> wow, that really is a trace...
20:38:32 <jix> it's useful for debugging
20:38:40 <jix> but i wrote all programs without it
20:38:52 <jix> but for my next program i need a trace
20:39:10 <jix> and an assembler
20:51:03 <jix> i have a routine for printing a number in reverse
21:11:35 <wildhalcyon> I don't really understand it, but it looks nice
21:11:57 <jix> i'm working at 99 bottles of beer
21:37:32 -!- wildhalcyon has quit (Read error: 104 (Connection reset by peer)).
21:38:07 <{^Raven^}> jix: you know that ruby, bf to c compiler you made
21:38:23 <jix> i should know it.. i made it
21:38:25 <{^Raven^}> did you ever manage to compile lostkingdom.c
21:38:41 <jix> i have no other c compiler than gcc on my machine
21:38:43 <jix> and gcc fails
21:38:48 <jix> because the stack is too small
21:39:06 <jix> and i removed all non-hw stack limits and set the hw stack to the highest possible value
21:39:12 <{^Raven^}> after two hours of compiling the machine terminally crashes
21:39:37 <jix> i get a "Stack size to small" error after 2 mins
21:40:22 <jix> if you have access to a linux machine http://fabrice.bellard.free.fr/tcc/ may work for you
21:40:26 <{^Raven^}> mine takes 2 hours to mention the missing linefeed at the end of the file and soon after the machine explodes
21:41:21 <{^Raven^}> only the TCP/IP stack survives and I can't even shh in to kill gcc!
21:41:57 <{^Raven^}> i'm sure it works though, everything else has
21:44:36 <jix> port LK to subskin
21:45:04 <jix> without my assembler or tracer or interactive debugger
21:45:56 <jix> i'm done with the "N bottle[s] of beer" routine!
22:00:50 <jix> it works!!!
22:01:17 <jix> hmm it was a bad idea to trace all verses of 99 bottles of beer
22:02:06 <jix> 295 words 800 characters
22:02:50 <jix> it's 2 lines shorter than the output!
22:08:21 * {^Raven^} 's game is finished and uploaded to the competitions site :D
22:09:21 <{^Raven^}> The game is called 23:15 and the page is here: http://us.geocities.com/dunric/advcomp.html
22:10:13 <{^Raven^}> Aardwolf: (named after the first version of brainloller that compiled)
22:11:24 <jix> 99bob in subskin WORKS!
22:18:10 -!- calamari has joined.
22:31:18 <jix> moin calamari
22:31:27 <jix> i wrote 99bob in subskin
22:31:33 <jix> (you know subskin?)
22:31:45 <calamari> jix: your project yesterday got me thinking about a minimal tc implementation
22:32:00 <jix> subskin is my OISC implementation
22:32:08 <jix> the result of my project from yesterday
22:32:19 <jix> ahh there's a bug in 99bob
22:32:41 <calamari> how many bytes did it end up as?
22:33:11 <jix> http://esolangs.org/wiki/Subskin
22:34:05 <calamari> I'm curious what the smallest tc interpreter I can make for ms-dos x86
22:34:28 <calamari> maybe oisc is the way to go, but maybe not
22:34:49 <jix> imo subskin is the best oisc instruction
22:35:43 <calamari> I always remember subtract and branch if borrow
22:35:55 <jix> i have no branch in the instruction
22:36:02 <jix> but a memory mapped IP
22:36:15 <jix> you can do more calculations without stupid branching
22:36:57 <jix> if you don't know if the result is negative or not you just put a nop-like instruction after it
22:37:10 -!- jix has quit (Remote closed the connection).
22:37:22 -!- jix has joined.
22:38:00 <Aardwolf> {^Raven^}: cool, good luck with the competition
22:38:07 <jix> calamari: memory mapped
22:38:10 <jix> see the wiki page
22:40:24 <{^Raven^}> calamari: finally finished my game! it's on the competition site now
22:45:51 <calamari> for some reason that reminds me of monsters inc.. 23-19!! :)
22:46:00 <Aardwolf> 2315.exe in Wine sais: invalid channel at line 10, and every command I try is either Syntax Error or a Mistake
22:46:11 <Aardwolf> or is it part of the puzzle :)
22:49:09 <jix> i can't eat myself!
22:49:41 <jix> under my pc emulator it works
22:50:27 <Aardwolf> Wine must not be that good, I wonder if Dosbox could run it?
22:50:46 <Aardwolf> (actually wine has been having poor results on almost any program I tried :s)
22:51:35 <jix> you could try a risc-os emulator
22:51:49 <Aardwolf> This program can not be run in DOS mode
22:51:59 <jix> http://www.redsquirrel.fsnet.co.uk/
22:53:31 <calamari> I ran it from the command line rather than from the gui
22:54:12 <Aardwolf> ah maybe it didn't find the data file
22:54:41 <jix> i'm in this silly tube and can't do anything... :(
22:55:03 <Aardwolf> It sais I'm very hungry, but if I type eat it sais I'm not that hungry :)
22:55:16 <Aardwolf> by the way is the font only this screwed up in my wine?
22:55:28 <jix> calamari: ouch i typed eat tueb
22:55:57 <calamari> hmm weird.. wine lost focus and now I can't type.. need to restart
22:56:11 <Aardwolf> lol I have the choice between hungry or tired
22:56:25 <jix> Aardwolf: no...
22:56:47 <Aardwolf> the font is only barely readable in my wine, default settings afaik
22:56:48 <jix> The chamber door opens.
22:57:25 <Aardwolf> ouch wine lost focus on me too, after typing "rest", so don't type it
22:57:45 <calamari> sounds like whatever emulator he used wasn't up to the challenge :)
22:58:40 <Aardwolf> darn, can't go n, e, s, w, u, d
22:58:51 <jix> the door is open but i cant go through it
22:59:12 <calamari> whoohoo! got the door open, need to save :)
22:59:31 <jix> lol i type: b and it says: violence is not the answer
22:59:42 <{^Raven^}> Aardwolf: It's an issue with wine, the CSD is not set correctly and the program cannot find the data file
23:00:02 <Aardwolf> yeah it's solved if you open it from console :)
23:00:03 <calamari> oops, doesn't look like it can save under wine
23:00:12 <Aardwolf> wine isn't really too great :(
23:01:49 <jix> {^Raven^}: i opened the door but i can't walk through it
23:02:06 <jix> i don't know the command
23:02:30 <Aardwolf> it really understands a lot of commands :)
23:03:07 <jix> try break *
23:03:36 <calamari> aha, look = room description, examine = look at something in particular
23:03:43 <Aardwolf> hmm what was again this secret 6 letter word of Zork?
23:04:06 <calamari> might be helpful.. was trying "look at"
23:04:47 <jix> it was xyzzy
23:04:56 <{^Raven^}> When the game dies with a invalid channel error, type: *cd z:\home\<path to 2315.exe>
23:06:05 <jix> nothing happens if i type xyzzy BUT it understands the command
23:07:00 <calamari> hmm maybe I can steal a card off that dead guy
23:07:16 <jix> i get tons of Mistake msgs now
23:07:20 <Aardwolf> I think we found a clue to one of the secrets :)
23:07:51 <jix> i answered with >n to Play again
23:07:57 <jix> instead of just
23:08:18 <{^Raven^}> jix, hte game has exit, type RUN to begin again
23:08:33 <jix> {^Raven^}: i'm not using wine
23:08:43 <jix> {^Raven^}: i'm using a complete pc emulator with win xp
23:08:54 <jix> because wine needs a x86 cpu
23:09:09 <jix> virtual pc (comercial)
23:09:11 -!- wildhalcyon has joined.
23:09:14 <{^Raven^}> if you end up exiting the game without closing the window type RUN to begin again
23:10:05 <calamari> raven: cool game, I'll have to play it again when I have more time and from qemu where it should be more reliable :)
23:10:48 <Aardwolf> debugging Mycelium is very annoying, even tho the syntax is easy, writing code in a painting program is hellish :p
23:10:58 <jix> wildhalcyon: i wrote 99bob in subskin!
23:11:34 <jix> its 2 lines shorter than the output
23:12:09 <jix> output: 297 lines 11756 bytes, program: 296 lines 786 lines
23:12:20 <jix> there are many 1 char lines
23:12:28 <jix> most lines are 2 chars
23:12:33 <jix> and some are 3 chars
23:13:11 <jix> and i wrote an improved hello world version
23:13:27 <jix> and a trace program: http://www.harderweb.de/jix/hello-world-trace.html (sample output)
23:13:55 <jix> and an assembler (used in 99bob.. cat and hw are written 100% by hand)
23:16:12 <wildhalcyon> Im trying to wrestle with the topology issues in my lang before I write the spec & interpreter; then I'll write a roguelike
23:17:43 <wildhalcyon> see, I encourage everyone to write games in there languages
23:19:06 <jix> hmm but what game
23:19:47 <calamari> in a language without i/o, how is the result of the computation expressed?
23:23:21 <calamari> jix: btw, how does skip() work? is it possible to form an endless loop with it?
23:28:20 <jix> a,b,c,d,e,f,g,h,i .. if subskin a,b,c skips the next instruction is g,h,i if it doesn't skip the next instruction is d,e,f
23:28:48 <calamari> is it possible to get back to a,b,c?
23:28:59 <jix> of course.. IP is memory mapped
23:29:14 <jix> you can branch everywhere by setting the IP to any adress
23:30:59 <calamari> I wonder if a lang could be tc if it was allowed to contain only a single loop
23:31:36 <jix> i'm trying to trace the whole execution of 99bob
23:32:13 <jix> im at 92 and the trace is > 10mb
23:33:33 <jix> it's going to be > 100mb
23:33:47 <jix> 80 bottles left 32mv
23:34:12 <calamari> hmm, perhaps I can answer this by construction with a dfa.. if a dfa with only one loop is still a dfa, I'd bet all is ok
23:35:11 <calamari> need to define loop more precisely tho
23:35:53 <jix> there may be only one place where the IP is allowed to move in reverse direction
23:36:33 <calamari> jix: it'd be a state with more than one transition to it
23:37:23 <calamari> so states would only be allowed to have a single transition to them, except the start state which is allowed two
23:37:55 <jix> code_a;if(bla){code_b};code_c
23:37:59 <jix> no loop right?
23:38:24 <jix> there is a code_a => code_c transition and a code_b => code_c transition
23:38:59 <jix> a loop less code is a code where there is no way to enter a state after it was entered once
23:39:26 <jix> 104mb 38bottles left
23:42:09 <jix> most browser have problems with 160mb of html?
23:43:03 <calamari> ahh, drawing out a transition table helps
23:43:29 <calamari> each transition that doesn't result in a loop is a transition to a state that has not yet been seen
23:44:08 <jix> no... it's the same thing you said before...
23:44:56 <calamari> jix: the exit state of the if and else parts go to the same state but it has not yet been seen
23:45:12 <calamari> jix: however, the start state has been seen
23:45:28 <jix> but i think my definition is simpler
23:45:28 <calamari> so transitioning to it is a loop
23:45:46 <calamari> jix: neither definition is quite there yet..
23:46:07 <jix> i define a loop as a code that gets executed multiple times
23:46:22 <jix> and my definition of a loop less code makes that impossible
23:46:23 <calamari> there is no such thing as "code" in a dfa tho :)
23:46:32 <calamari> it's all states and transitions
23:46:37 <jix> replace code by state+transition
23:53:23 -!- jix has quit (Remote closed the connection).
23:54:16 -!- jix has joined.