00:04:48 -!- cheater2 has quit (Ping timeout: 240 seconds).
00:06:44 -!- augur has quit (Ping timeout: 260 seconds).
00:18:08 -!- Oranjer has left (?).
00:23:04 -!- poiuy_qwert has quit (Ping timeout: 276 seconds).
00:26:14 -!- poiuy_qwert has joined.
00:49:33 -!- cheater2 has joined.
01:00:09 -!- mre has joined.
01:02:43 -!- mre has changed nick to aox.
01:29:01 -!- Oranjer has joined.
02:54:34 -!- aox has quit (Quit: I'll be back).
02:55:42 -!- oerjan has quit (Quit: Reboot).
03:06:28 -!- oerjan has joined.
03:14:12 -!- oerjan has quit (Quit: leaving).
03:19:22 -!- augur has joined.
03:20:02 -!- Gregor-L has joined.
03:52:35 -!- lament_ has joined.
04:02:24 -!- pikhq_ has joined.
04:02:40 -!- wareya has quit (Ping timeout: 240 seconds).
04:03:01 -!- jcp has quit (Quit: Later).
04:03:41 -!- wareya has joined.
04:04:04 -!- yiyus has quit (Ping timeout: 248 seconds).
04:04:10 -!- pikhq has quit (Ping timeout: 240 seconds).
04:04:40 -!- fizzie has quit (Ping timeout: 240 seconds).
04:04:58 -!- yiyus has joined.
04:05:15 -!- CT|P4 has joined.
04:05:39 -!- P4 has quit (Ping timeout: 246 seconds).
04:06:48 -!- Mathnerd314_ has joined.
04:07:08 -!- fizzie has joined.
04:07:10 -!- jcp has joined.
04:09:10 -!- Mathnerd314 has quit (Ping timeout: 240 seconds).
04:09:19 -!- Mathnerd314_ has changed nick to Mathnerd314.
04:12:20 -!- yiyus has quit (Ping timeout: 260 seconds).
04:16:09 -!- SimonRC has quit (Ping timeout: 246 seconds).
04:18:15 -!- SimonRC has joined.
04:18:30 -!- yiyus has joined.
04:20:35 -!- augur has quit (Ping timeout: 265 seconds).
04:29:49 -!- pikhq_ has changed nick to pikhq.
04:39:12 -!- pikhq has quit (Read error: Connection reset by peer).
04:47:39 -!- Sgeo has quit (Ping timeout: 265 seconds).
04:48:45 -!- coppro has joined.
05:03:03 -!- Sgeo has joined.
05:04:05 -!- augur has joined.
05:12:07 -!- pikhq has joined.
05:14:52 -!- poiuy_qwert has quit (Quit: Leaving).
05:29:20 -!- soupdragon has quit (Quit: soupdragon).
06:05:24 -!- pineapple has quit (Read error: Operation timed out).
06:05:26 -!- pineapple has joined.
06:17:18 -!- lament_ has changed nick to lament.
06:21:42 -!- Gracenotes has joined.
06:22:52 -!- augur has quit (Ping timeout: 265 seconds).
06:38:27 -!- FireFly has joined.
06:38:51 -!- augur has joined.
06:41:59 -!- Sgeo has quit (Remote host closed the connection).
06:56:52 -!- tombom has joined.
07:08:47 <AnMaster> <oerjan> there's always uncyclopedia. <-- my attempts were somewhat less absurd than it, mine could actually pass for valid entries at a quick glance, something that really isn't true for uncyclopedia
07:29:10 -!- cheater2 has quit (Ping timeout: 240 seconds).
07:30:12 -!- BeholdMyGlory has joined.
07:33:22 -!- lament has quit (Ping timeout: 276 seconds).
07:42:01 -!- Oranjer1 has joined.
07:43:46 -!- Oranjer has quit (Ping timeout: 276 seconds).
07:50:54 -!- Vegabondmx has joined.
07:51:20 -!- tombom has quit (Quit: Leaving).
07:56:44 -!- coppro has quit (Remote host closed the connection).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:03:46 -!- Oranjer1 has left (?).
08:18:58 -!- Tritonio_GR has quit (Read error: Connection reset by peer).
08:25:15 -!- Vegabondmx has quit (Ping timeout: 240 seconds).
08:32:51 -!- cheater2 has joined.
08:33:11 -!- marchdown has joined.
09:33:11 -!- marchdown has quit (Read error: Connection reset by peer).
09:33:36 -!- marchdown has joined.
09:40:39 -!- marchdown has quit (Remote host closed the connection).
10:30:57 -!- augur has quit (Remote host closed the connection).
10:57:48 -!- Phantom_Hoover has joined.
11:07:37 -!- hiato has joined.
11:15:05 -!- MizardX- has joined.
11:18:46 -!- MizardX has quit (Ping timeout: 276 seconds).
11:19:05 -!- MizardX- has changed nick to MizardX.
11:56:47 -!- BeholdMyGlory_ has joined.
11:59:10 -!- BeholdMyGlory has quit (Ping timeout: 240 seconds).
11:59:23 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
12:00:15 -!- BeholdMyGlory_ has quit (Changing host).
12:00:15 -!- BeholdMyGlory_ has joined.
12:00:21 -!- BeholdMyGlory_ has changed nick to BeholdMyGlory.
12:22:06 -!- Tritonio_GR has joined.
12:31:37 -!- oerjan has joined.
12:56:44 -!- augur has joined.
13:27:16 -!- CT|P4 has quit (Changing host).
13:27:16 -!- CT|P4 has joined.
13:27:18 -!- CT|P4 has changed nick to P4.
14:04:16 -!- oerjan has quit (Quit: Good night).
14:06:57 -!- marchdown has joined.
14:07:00 -!- marchdown has quit (Excess Flood).
14:07:33 -!- marchdown has joined.
14:20:55 -!- augur has quit (Ping timeout: 276 seconds).
14:34:01 -!- sebbu2 has quit (Ping timeout: 268 seconds).
14:35:30 -!- sebbu has joined.
14:42:32 -!- Rugxulo has joined.
14:46:59 <Rugxulo> two links for alise (assuming he logreads):
14:47:00 <Rugxulo> http://www.staticramlinux.com/
14:47:08 <Rugxulo> http://www.forthfreak.net/index.cgi?QBFORTH
14:47:59 <Rugxulo> oh, and fizzie, have you heard of llfunge? http://www.tilk.eu/
14:48:43 <fizzie> "Almost correct - it disallows program mutation."
14:48:49 <fizzie> That's a cop-out if I ever saw one. :p
14:48:57 <Rugxulo> I know, just wanted you to be aware of it
14:49:19 <fizzie> But no, hadn't heard of it before.
14:49:39 -!- Gregor-L has quit (Ping timeout: 252 seconds).
14:51:01 -!- poiuy_qwert has joined.
14:54:08 <Deewiant> Befunge without program mutation is like C without variables
14:56:47 <fizzie> Deewiant: I don't do any program mutation in fungot, so it's not *quite* that awful.
14:56:48 <fungot> fizzie: these unique items make us invincible! but you are true heros. the world, tee, hee! it's not the only one thing we need to defeat you, lavos. the ocean palace?
14:57:50 <Deewiant> Well, like without functions then
14:59:20 <fizzie> I'm a bit surprised it's a JIT compiler and not an ahead-of-time compiler, though I guess it's understandable; it's a bit hard to predict what &&x will do.
15:02:04 <fizzie> It is at least partially.
15:02:17 <fizzie> There's a stack-stack, and a hashmap-based fungespace with three dimensions.
15:02:51 <Deewiant> Aye, I guess that'd make it -98
15:03:46 <fizzie> No sign of fingerprints, though.
15:04:36 <Rugxulo> unimplemented: ( ) ; = i j k o q t x y
15:04:59 <Deewiant> x unimplemented! Should be ahead-of-time :-P
15:05:06 <fizzie> Yes, I just found that bit.
15:05:47 <fizzie> It's also a bit strange at places.
15:06:11 <fizzie> The 'Stack' struct has a "void * mem" and then #define stack_top(stack) ((stack)->mem + (((stack)->ptr - 1) * (stack)->elsize))
15:07:39 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
15:20:46 -!- Gregor-L has joined.
15:28:37 -!- kar8nga has joined.
15:30:23 -!- poiuy_qwert has joined.
15:53:58 -!- Tritonio_GR has quit (Read error: Connection reset by peer).
15:54:16 -!- lament has joined.
15:55:29 -!- Rugxulo has quit (Quit: ERC Version 5.3 (IRC client for Emacs)).
16:08:43 <pikhq> Deewiant: It's perfectly feasible to do Befunge without *program* mutation but with g and p. :P
16:09:09 <Deewiant> Yes, but it's sooooooo boring and lame
16:09:12 <pikhq> Though arguably this isn't Befunge, but rather a simpler, compilable language.
16:09:46 <AnMaster> pikhq, still not one that is trivial to compile
16:09:54 <AnMaster> you need to trace through every possible path
16:10:21 <AnMaster> pikhq, and if you ever use x it may not be compilable, depending on if you can figure out all possible ways to hit that x
16:10:41 <AnMaster> (and there the x itself may mess it up)
16:11:08 <AnMaster> in such cases best you could do would be threaded-code kind of thing, with one jump between each befunge instruction
16:11:32 <AnMaster> which is arguably still compiling, but not very interesting
16:11:43 <pikhq> AnMaster: *Possible*, not trivial. ;)
16:11:52 <pikhq> Also, what was x again?
16:12:18 <AnMaster> pikhq, pop y and x. Set delta from that
16:12:26 <AnMaster> pikhq, in trefunge it would be pop z, y, x
16:12:53 <pikhq> That can't be determined except by executing the code.
16:13:17 <AnMaster> pikhq, well, if you can figure out that it can only be hit with constant parameters ever...
16:13:41 <AnMaster> yes you still need to trace the program
16:13:56 <AnMaster> pikhq, but you would have to trace the program in any case if you want to compile it
16:14:07 <AnMaster> pikhq, and if you JIT then you can just as well handle g and p properly
16:14:29 <AnMaster> Deewiant, well that is one case you can't determine it
16:15:22 <AnMaster> Deewiant, but if you had >11x right at the start of the program for example, and then was able to see it was not reachable from above/below/right...
16:15:29 <AnMaster> of course you also need to check all the other x
16:16:28 <pikhq> Can't be determined in general without (partial?) execution.
16:17:10 -!- Geekthras has joined.
16:18:15 <Gregor-L> Your face can't be determined in general without (partial?) execution.
16:18:16 <AnMaster> pikhq, indeed. But don't you need that to compile befunge without g/p anyway?
16:19:02 <pikhq> AnMaster: Uh, that's called "parsing".
16:19:04 <AnMaster> pikhq, otherwise what would you compile? push(1); goto *next; ?
16:19:27 <AnMaster> pikhq, yes but you still need to know which directions the IP travels in if you compile it
16:19:38 <pikhq> Yes. That's called *parsing*.
16:19:40 <AnMaster> pikhq, or are you suggesting generating code for all possible deltas and such?
16:19:50 <AnMaster> which might be feasible for befunge93
16:20:09 <pikhq> Isn't that what *you* were talking about?
16:20:56 <AnMaster> pikhq, no? I was talking about tracing from start, and when you hit stuff like | then make a branch in the generated code (generate code for each separately). You will need to take care of merging paths too of course.
16:21:16 <AnMaster> but that seems like the obvious way to do befunge93-without-selfmod to me
16:21:36 <pikhq> ... Yes, that's called parsing.
16:21:39 -!- mostermand has joined.
16:21:58 <AnMaster> pikhq, you need tracing to do that, otherwise you would generate code for never-reached bits too
16:22:05 <AnMaster> plus you couldn't merge constants and so on
16:22:39 <pikhq> You seem to fail horribly at what the word "parsing" means.
16:22:45 <pikhq> And what "tracing" means.
16:22:50 <AnMaster> pikhq, how do you mean that: 99+ v Comment never executed v other code <
16:23:00 <AnMaster> would be codegened in your solution
16:23:25 <AnMaster> there may be paths from above/below, but in this case all will pass through the spaces between those words
16:23:28 <pikhq> That will be *parsed* and then *compiled* into push(9);push(9);add(); /...
16:23:38 <AnMaster> pikhq, yes and what about the comment
16:24:03 <pikhq> The parser notes the "v" and goes down to continue parsing.
16:24:10 <AnMaster> pikhq, yep, it traces the program
16:24:11 <pikhq> *Like any sane parser*.
16:24:22 <pikhq> THATS NOT WHAT TRACING MEANS.
16:24:32 <AnMaster> pikhq, it is what tracing it means in jitfunge at least..
16:25:26 <pikhq> And it'd be wrong.
16:26:23 <AnMaster> pikhq, it is a form of partial execution still. You need to follow all paths in the code to figure out what can be reached. In something like C that wouldn't be the task of the parser (rather, some optimisation like DCE or such probably)
16:27:04 <pikhq> It is still the task of the parser to ignore comments.
16:27:22 <Deewiant> Actually that's often done by the lexer
16:27:23 <AnMaster> pikhq, actually for C it is the pre-processor, not the compiler.
16:27:32 <pikhq> Deewiant: Fair enough.
16:27:33 <AnMaster> and that is a different lexer or parser
16:27:56 <pikhq> Oh, yeah. The C preprocessor nixes comments, not the compiler.
16:28:49 <AnMaster> pikhq, but what about code like (from start of program): 1#v_ Comment otherwise unreachable from elsewhere ^
16:28:54 <AnMaster> pikhq, would that be the parser still?
16:29:22 <pikhq> It would obviously parse both possible paths there.
16:29:36 <pikhq> A DCE pass could then see that one of the branches won't be taken.
16:30:25 -!- marchdown has quit (Quit: marchdown).
16:30:41 <AnMaster> pikhq, agreed. But I still think parser seems like the wrong word. A parser generally doesn't need to do stuff like tracking if it has already parsed the same code (which this one would need to do)
16:31:09 <pikhq> It's a 2D parser, rather than a 1D one.
16:31:11 <Deewiant> Languages generally are executed in the same direction from start to finish
16:31:13 -!- mostermand has quit (Quit: Leaving.).
16:31:28 -!- marchdown has joined.
16:31:29 <AnMaster> pikhq, and it does trace the program.
16:31:36 <Deewiant> pikhq: Even a Unefunge one might have to parse something from left to right and later right to left
16:31:42 <AnMaster> pikhq, in this context what would you call a tracer?
16:31:48 -!- mostermand has joined.
16:31:51 <pikhq> Yes, but a "tracer" is a type of debugger.
16:31:54 <AnMaster> oh and do tell fizzie to fix that in jitfunge then
16:31:59 <pikhq> Or a type of JIT that notes which paths are fast.
16:32:05 <pikhq> Which paths should be fast.
16:32:43 <AnMaster> pikhq, in jitfunge it is iirc the bit that follows a code path before handing it of to the codegen.
16:32:49 <pikhq> Most languages don't go backwards. :)
16:33:05 <Deewiant> Just saying that "2D" isn't the defining quality
16:33:07 <AnMaster> also marking that part of the path as traced and so on
16:33:09 <pikhq> AnMaster: Oh, that's *very* bloody confusing when dealing with a JIT.
16:33:35 <AnMaster> pikhq, the marking as used is so a g/p to it properly discards the previous compiled trace
16:34:33 <pikhq> AnMaster: He really should call it a "pather" or something.
16:34:50 <pikhq> It'll be an absolute headache if he decides to make his JIT trace.
16:35:10 <AnMaster> pikhq, it does that, but what meaning of trace do you have in mind? not the same presumably?
16:35:22 <AnMaster> I'm far from an expert on JITs
16:35:28 <AnMaster> I have no idea what you mean by tracing jit
16:35:43 <pikhq> A tracing JIT notes which code-paths are often-used so that it can optimise them.
16:35:56 <AnMaster> mhm, why is it called "tracing"?
16:36:01 <AnMaster> it has NOTHING to do with debuggers does it?
16:36:04 <Deewiant> Because it traces the execution.
16:36:08 <AnMaster> so all such should be renamed!
16:36:48 <AnMaster> pikhq, well, I just applied your logic to it. Same reasoning as above
16:37:05 <pikhq> The "tracer" as used in Jitfunge is not tracing execution. Merely discovering all possible paths for execution.
16:37:18 <AnMaster> pikhq, for why to call it parser, not tracer. This should be called hot-spot-JIT instead of tracing JIT.
16:37:37 <pikhq> HotSpot is a trademark of Oracle.
16:37:54 <pikhq> And refers to the HotSpot JVM, which traces.
16:38:14 <pikhq> Anyways, allow me to sum up this conversation.
16:38:20 <pikhq> You're an idiot, and your mother's a whore.
16:38:21 <AnMaster> pikhq, what? why does it slow down by outputting a debug trace? ;P
16:38:55 <pikhq> AnMaster: Because the debugging type of tracing takes much more work than just noting which branches are taken often. ;)
16:38:55 <AnMaster> pikhq, anyway that doesn't follow. Debugger and JIT are not the same field. Nor is parsers and debuggers.
16:39:16 <AnMaster> seems just fine to call it a tracer since it does trace the program. A tracing parser if you prefer
16:39:24 <pikhq> Would you call something a "tracer" if it went through some C code and noted every possible branch?
16:39:35 <pikhq> If you would, you're a fucking idiot.
16:40:29 <AnMaster> pikhq, depends. If it just marks every if/while/whatever with "here is a branch" then probably not. If it follows all the code paths, maybe.
16:40:53 <pikhq> You're a fucking idiot.
16:41:48 <AnMaster> pikhq, it depends on what sort of thing it is used for I would say. Rather context dependant thus.
16:42:38 <AnMaster> pikhq, but generally I do think tracer makes more sense in a 2D language
16:43:10 <AnMaster> probably in unefunge too, so make it "multi-directional"
16:43:29 <pikhq> PARSER PARSER PARSER PARSER PARSER
16:44:31 -!- marchdown has quit (Read error: Connection reset by peer).
16:45:26 <AnMaster> hm j might be possible to compile stil
16:45:51 <AnMaster> it would be like a switch case for every instruction forward/backward in the IP path
16:47:03 <AnMaster> there are certainly many other tricky instructions...
16:47:04 <Deewiant> That's still complexity explosion if there are too many
16:48:56 <AnMaster> Deewiant, hm? Not much unless you have many "go away from path" instructions during it. if you have stuff like j1234 you would get something like switch { case 1: push(1); case 2: push(2); ... }
16:49:57 <AnMaster> Deewiant, and of course it could also just join in into any other trace already done, say you have j blah blah ^ blah, where that ^ happens to be in the middle of another trace just going straight up through it
16:50:04 <Deewiant> If you have a weird enough delta that j can hit your whole program; but I guess you're only considering cardinal deltas, in which case it's not that bad
16:50:09 <AnMaster> it could goto into the middle of that trace in the suitable place
16:50:33 <Deewiant> For only cardinal deltas, it can hit anywhere on that line/column/3d-equivalent
16:50:45 <AnMaster> Deewiant, well I was ignoring x here yes since we concluded that in general it isn't feasible to statically compile if you have that in the program
16:51:16 <AnMaster> and yes "anywhere in the same line/col/3d-equiv" isn't too bad
16:52:12 <fungot> http://git.zem.fi/fungot/blob/HEAD:/fungot.b98
16:52:46 <AnMaster> Deewiant, not too bad still, plus a smart compiler could put a bound on the possible number in some cases.
16:53:11 <AnMaster> Deewiant, and i, that wouldn't be easy to handle if it was allowed to modify code, which we must presume it isn't
16:53:14 <fungot> AnMaster: we are looking to achieve a shorter life span... lavos will rule the world in a mere door that keeps us bound, hand, foot...and tongue kid? ...oh, it's you, isn't this morbid? the great adventurer toma levine rests in a grave to the north. it's a great place for a picnic! heard that magus's statue before my shift. i hate! ayla not like...
16:53:53 <Deewiant> Well you pretty much have to implement g/p anyway, and if you're going to assume that they never modify code you might as well assume the same for i
16:54:28 <AnMaster> Deewiant, wasn't that the context for this discussion? <pikhq> Deewiant: It's perfectly feasible to do Befunge without *program* mutation but with g and p. :P <pikhq> Though arguably this isn't Befunge, but rather a simpler, compilable language.
16:54:46 <Deewiant> Which is why I pointed out that i is no problem
16:55:08 <AnMaster> Deewiant, sorry, misread that line about i not being a problem
16:55:31 -!- lament has quit (Ping timeout: 245 seconds).
16:56:09 <AnMaster> anyway, what other instructions than x still pose a problem? j to some degree yes. but with x everything does more or less...
16:57:12 <Deewiant> Of the standard instructions, only x, I'm fairly sure
16:57:42 <AnMaster> well yes fingerprints might, but those are too numerous to consider.
16:58:10 <AnMaster> statically compiling the interleaving will most likely not work even if you don't optimise
16:58:23 <AnMaster> I mean, consider if the t is hit a number of time depending on a ?
16:59:25 <Deewiant> Programs with t tend to rely on self-modification anyway
16:59:38 <AnMaster> well, you could manage without it
17:00:02 -!- mostermand has left (?).
17:01:46 <AnMaster> sure the replacement for >< and other thread using p to remove one of those > or < would be a bit longer. something like >0# 0# g# _ code continues here
17:01:53 <AnMaster> should work to wait for something to change
17:02:00 <AnMaster> well for 0,0 it could be made shorter
17:02:20 <AnMaster> need a flag somewhere yes, but surely you can find some space in a 32-bit funge space easily even
17:09:25 <hiato> poiuy_qwert: Checked, yeah, used, no
17:15:53 <hiato> poiuy_qwert: "Using the q command(By poiuy_qwert): "
17:16:06 <hiato> so it sounds like you know more than you let on
17:18:59 <hiato> oh, lol, I got duped
17:19:12 <hiato> poiuy_qwert: I thought Lode hand made Zeta and you did only Gamma
17:20:24 <poiuy_qwert> hey theres no problem. you even knowing it exists makes me happy :P
17:21:54 <hiato> heh :) One of the first I toyed with due to the awesome looking befunge interp
17:22:39 -!- tombom has joined.
17:33:30 -!- augur has joined.
17:33:33 -!- augur has quit (Remote host closed the connection).
17:33:55 -!- augur has joined.
17:38:45 <poiuy_qwert> I don't think many people are familiar with it so don't feel left out ;P
17:39:21 <AnMaster> poiuy_qwert, "This language was designed to be very functional." <-- wait a second, not functional as in lisp or haskell is it?
17:40:27 <poiuy_qwert> functional as it in can do a lot of stuff, not as in functions/procedures/methods
17:41:14 <AnMaster> that befunge interpreter on the wikipage for zetaplex, 93 I assume?
17:42:02 <poiuy_qwert> I believe so. At the time I made that I didn't know there where variations *doh*
17:42:23 <AnMaster> when was it made? before 1998 perhaps?
17:42:59 <poiuy_qwert> when was what made, the befunge interpreter?
17:44:16 <poiuy_qwert> heh, well Zetaplex was made in 2007 :P so sometime after that.
17:46:52 -!- Tritonio_GR has joined.
17:54:49 <AnMaster> poiuy_qwert, is "This is a basic tic-tac-toe game (does not work with the current public interpreter, but the author will release the updated interpreter soon)" still valid?
17:56:14 <AnMaster> Deewiant, btw as far as I can tell fungot only uses x in comments
17:56:14 <fungot> AnMaster: that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alone can't stop! that sword alon
17:56:33 <fungot> AnMaster: but, we are far outnumbered! yes. well, i better! whoosh! i wonder how everyone! humans! they're my friends!
17:56:42 <AnMaster> fizzie, hey! saw that! A perfect one.
17:56:55 <AnMaster> (as in, not just part of the line with it, the entire line!)
17:59:01 <poiuy_qwert> AnMaster: in my old-new interpreter :P there are 3 interpreters, 1 public, an updated version of the public one that was never released, and now I'm working on the new one with a specification update. with a little tweaking it will work on the new interpreter though
17:59:25 <poiuy_qwert> so currently noone can run it unless i find that middle interpreter :P
18:03:39 <fizzie> Fix what in jitfunge? (There's too much context for me to read.)
18:03:55 <AnMaster> fizzie, that you call that part of it "tracing" iirc
18:08:44 <fizzie> I think it's a reasonable use of the word; I call "tracing" the part that interprets Funge-98 code, but also collects the instructions into a trace that can be JIT'd into code. It's quite a bit like what I understand is called a tracing JIT compiler.
18:08:57 <fizzie> It's not exactly a parser, since it also executes the code.
18:10:47 <fizzie> Admittedly I misuse the word a bit in that I use it also in the static compiler, where it's a lot more like a parser. I offer as an excuse that it does resemble the act of someone "tracing" the way the code would be executed with pen and paper, metaphoristically.
18:34:28 -!- poiuy_qwert has quit (Ping timeout: 240 seconds).
18:35:47 -!- poiuy_qwert has joined.
18:51:38 -!- kar8nga has quit (Remote host closed the connection).
19:01:40 -!- Tritonio_GR has quit (Read error: Connection reset by peer).
19:12:30 -!- tombom has quit (Ping timeout: 246 seconds).
19:14:07 -!- tombom has joined.
19:24:49 -!- Phantom_Hoover has joined.
19:25:06 -!- Phantom_Hoover has quit (Quit: Phantom_Hoover).
19:26:44 -!- calamari has joined.
19:28:59 -!- Gracenotes has quit (Quit: Leaving).
19:47:31 -!- KingOfKarlsruhe has joined.
19:47:43 <AnMaster> fizzie, I disagree about any excuse being needed
19:47:52 <AnMaster> it was pikhq who thought that it was badly named
19:55:36 -!- Phantom_Hoover has joined.
19:56:16 -!- Phantom_1oover has joined.
19:56:33 -!- Phantom_Hoover has quit (Client Quit).
19:56:35 -!- Phantom_1oover has quit (Client Quit).
20:01:42 <fizzie> Well, for the static compiler it is a bit less justified.
20:06:44 <hiato> anyone care to bash/grant constructive criticism to a new esolang of mine? or, at least the loose collection of ideas I had
20:07:08 <pikhq> hiato: Yes, it sucks and you should be ashamed.
20:07:28 <hiato> http://dpaste.com/193334/
20:08:11 <hiato> currently it's fugly, and I cant understand why my ideas aren't playing nicely together
20:16:00 <AnMaster> pikhq, btw you said parsers removed comments, I wish to differ. I'm pretty sure ick's parser does not
20:16:13 <AnMaster> they are after all compiled into runtime syntax errors
20:16:23 <AnMaster> (if they are executed that is)
20:17:01 <AnMaster> (and iirc that will only happen if you do strange things with reinstate)
20:22:12 <pikhq> AnMaster: Quite obviously, it depends upon the language in question.
20:26:06 <calamari> hiato: () [] {} don't match.. maybe that is making it look bad to you?
20:27:23 <hiato> calamari: Possible...hmmm, yeah. Now I just think it's not a great way of combining my ideas
20:28:09 <AnMaster> <hiato> currently it's fugly, and I cant understand why my ideas aren't playing nicely together <-- it reminds me of intercal
20:28:40 <hiato> but, erm, thanks :)
20:28:47 <AnMaster> hiato, mX.~7|$!)~:-3?4(;! however is extremely close to intercal, there is something missing but can't pinpoint it
20:29:06 <AnMaster> not completely sure about that though
20:29:10 <calamari> I've been trying to come up with a lang that permits maximum expression in minimum code space
20:29:22 <AnMaster> calamari, like, the reverse of intercal?
20:29:48 <AnMaster> also it sounds like a language perfect for golfing with
20:30:10 <calamari> AnMaster: yes exactly.. a golf language
20:30:16 <hiato> yeah, I guess the trouble is that case stuff, need to find a better encoding. As in, the current requires vars with names with at least three letters to express of possibilties
20:30:28 <AnMaster> calamari, isn't there one already?
20:30:38 <calamari> but perl wont fit in my watch.. I have the constraint of 900 byte states :)
20:31:20 <AnMaster> as in RAM usage being 900 bytes at most or what?
20:31:39 <calamari> golfscript looks like a winner
20:32:09 <calamari> rom is something like 24k.. but 900 bytes at a time, page swapped
20:32:29 <calamari> the program code will have to share with the ram
20:34:59 <AnMaster> calamari, what sort of device is this?
20:35:17 <calamari> Timex Data Link USB wristwatch
20:36:00 <AnMaster> it's quite insane to write programs for it
20:36:06 <AnMaster> with such a sucky architecture
20:36:38 <pikhq> calamari: I... Think that might be too little space for a Forth.
20:36:55 <AnMaster> calamari, I can't think of any good reason for 1025 bytes of address space...
20:37:04 <calamari> maybe some type of asm is the best I can do
20:37:08 <pikhq> AnMaster: 125 bytes.
20:37:21 <AnMaster> pikhq, 125 bytes of ram and 900 bytes of rom
20:37:36 <pikhq> The RAM is... Registers.
20:38:09 <AnMaster> pikhq, are they treated as memory mapped?
20:38:15 <pikhq> Not "is", but, good god. There's CPUs with more memory in its registers.
20:39:16 <AnMaster> pikhq, you see, I can't figure out a good reason to have 1025 bytes in total, it means you will need an extra bit to address the last byte.
20:39:28 <calamari> now that I think about it.. some machine code would be the smallest
20:39:30 <AnMaster> either you would use it fully or stop at 1024
20:39:55 <calamari> but not the native machine code, it's bloated
20:39:59 <pikhq> AnMaster: Seriously.
20:40:29 <calamari> anmaster there is more ram than 125 bytes
20:40:30 <AnMaster> pikhq, well, how do you address that ram
20:40:37 <calamari> but most of it is used by the watch os
20:41:03 <AnMaster> calamari, because 1025 bytes of address space made no sense indeed.
20:41:15 <calamari> I have no idea why the states are 900 bytes tho
20:41:31 <calamari> seems like some power of 2 would have made more sense
20:41:45 <AnMaster> calamari, how much ram is there in total?
20:42:01 <AnMaster> including the watch OS bit I mean
20:42:26 <calamari> 1 min, need to get out my calc
20:43:25 <calamari> actually nm, I don't feel like adding all that up
20:43:35 <calamari> 2048 bytes, plus some extra in the lcd
20:43:54 <AnMaster> calamari, well if when after you add it up you get 900 bytes left to the next power of two that is a good reason for 900 bytes
20:44:05 <AnMaster> it means a non-power-of-two ram size though
20:46:46 <calamari> oh wow.. the code is being copied into ram.. so you are absolutely right, I should be able to use any unused code space as extra ram
20:47:06 <calamari> dang, wish I'd realized this before
20:48:51 <AnMaster> calamari, hm if it is a rom you can't write it more than once
20:49:00 <AnMaster> presumably you mean eeprom or whatever
20:50:02 <AnMaster> calamari, and you don't want to rewrite some of those too often what with the limited cycle count (at least for flash, not completely sure for eeprom)
20:52:10 <AnMaster> or rather, flash is a subtype of eeprom
20:52:11 <calamari> It is a specific type of EEPROM (electrically-erasable programmable read-only memory) that is erased and programmed in large blocks; in early flash the entire chip had to be erased at once.
20:52:36 <AnMaster> but yes eeprom have limited cycle count too
20:53:18 <calamari> I have an eprom writer and eraser also .. I have no idea how many cycles you get there but it's fun to erase with the uv :)
20:53:23 <AnMaster> http://en.wikipedia.org/wiki/File:DSCN0411.JPG <-- the wasted space makes me sad
20:53:42 <AnMaster> calamari, do you still use it?
20:53:50 <AnMaster> I can't see any good reason for it any more
20:54:13 <AnMaster> calamari, computer? or something else?
20:54:33 <calamari> yeah it's an early 80's gaming system
20:54:38 <AnMaster> if a computer I don't see why it doesn't use something like floppies...
20:55:17 <calamari> it's actually very similar to the atari 800, which does have an optional floppy drive.. but the game system only accepts cartridges
20:56:25 <AnMaster> calamari, "EPROMs had a limited but large number of erase cycles; the silicon dioxide around the gates would accumulate damage from each cycle, making the chip unreliable after several thousand cycles. EPROM programming is slow compared to other forms of memory. Because higher-density parts have little exposed oxide between the layers of interconnects and gate, ultraviolet erasing becomes less practical for very large memories. Even dust insid
20:56:25 <AnMaster> e the package can prevent some cells from being erased."
20:56:49 <AnMaster> so be careful with it unless you can easily find replacements
20:57:18 <AnMaster> calamari, how often do you reprogram it?
20:57:24 <AnMaster> and what do you reprogram it for
20:57:36 <calamari> not often, most development is via emulator
20:58:08 <calamari> for testing out on the actual system, mostly for joystick and color issues
20:58:35 <AnMaster> calamari, I can't imagine why anyone still develop for it :D
20:58:49 <AnMaster> calamari, besides, most people will only have emulator
20:59:04 <calamari> people prgram for the atari 2600 still too
20:59:31 <AnMaster> calamari, the most I have done has been porting ick to Mac OS classic
20:59:36 <calamari> I have a mostly finished game for the 2600
20:59:50 <AnMaster> it kind of works, but only on real old macs, not in emulators
21:00:02 <AnMaster> get a system crash when compiling one of the files for ick in sheepshaver
21:00:09 <AnMaster> and well, I doubt it would work on 68k
21:00:19 <AnMaster> I only used sheepshaver and my old first-model ibook for it
21:00:37 <AnMaster> calamari, ick sometimes generate legal C code that mpw can't compile too
21:01:21 <AnMaster> calamari, but apart from that one file (I think it was some yacc or bison file) mpw under sheepshaver can compile it
21:01:46 <AnMaster> calamari, it might work for 68k with some changes to the makfile
21:02:04 <AnMaster> but since sheepshaver crashes with 68k binaries most of the time I haven't tried
21:02:30 <calamari> so I wouldn't be able to test it for you, sorry
21:02:33 <AnMaster> calamari, I think it is no-cost
21:02:43 <AnMaster> as in, you can download some *.img from apple
21:02:59 <AnMaster> calamari, never looked for old 68k versions. Try that site... what was the name
21:03:06 <calamari> I think I have os 8.6 or whatever the last was that worked on 68k
21:03:10 <AnMaster> a lot of old abandonware for mac
21:03:30 <calamari> I'll have to pull out my disks and retrieve an image
21:03:45 <AnMaster> calamari, ah http://www.macintoshgarden.org/ might have it
21:04:14 <calamari> that was something cool that apple did.. made its oses freeware when they became obsolete
21:04:25 <AnMaster> http://macintoshgarden.org/apps/macintosh-programmer%E2%80%99s-workshop
21:04:40 <AnMaster> calamari, well I don't think OS 9 ever became freeware
21:04:47 <AnMaster> it is the one I have in sheepshaver
21:04:54 <AnMaster> but I had an old legal copy of it
21:04:57 <calamari> doesn't matter.. os 9 doesn't run on 68k iirc
21:05:13 <AnMaster> calamari, well yes, but I wanted to play ppc games, not 68k ones!
21:05:25 <AnMaster> calamari, to be specific, the old avernum series
21:06:11 <AnMaster> you might have played exile? Avernum was basically the same with upgraded graphics (isomeric instead of top-down/sideways mix) and such
21:06:29 <calamari> I don't recall, what was exile
21:06:48 <AnMaster> calamari, http://en.wikipedia.org/wiki/Exile_%28video_game_series%29
21:06:53 <calamari> and do you remember a game where you ran around inside a cave collecting jewels?
21:06:58 -!- Oranjer has joined.
21:07:06 <AnMaster> calamari, no, but I played lots of modern such games
21:07:32 <AnMaster> it isn't quite as common as remakes of snake, pacman, space invaders and so on
21:08:03 <AnMaster> boulderdash was one I played iirc
21:08:05 <calamari> nope haven't played exile, looks neat
21:08:20 <AnMaster> calamari, also: http://en.wikipedia.org/wiki/Avernum_%28series%29
21:09:14 <AnMaster> calamari, which is basically same as exile but 1) more well balanced (exile had some issues with being very easy in parts and way too hard in others), 2) better graphics 3) a nice NPC dialog UI, especially Exile 1 had a bloody stupid UI for it
21:09:48 <calamari> I liked not actually having graphics, as silly as that sounds
21:10:01 <AnMaster> well I love nethack if that is what you mean
21:10:07 <AnMaster> calamari, don't think I ever played moria
21:11:07 <AnMaster> ah I know of that, tested it too iirc
21:11:49 <calamari> angband had too many races and objects, better to keep it simple sometimes
21:12:24 <AnMaster> calamari, oh iirc mpw came on segmented *.img
21:12:28 <AnMaster> so that may be annoying to download
21:12:35 <AnMaster> also atm I'm trying to located the port
21:13:30 <AnMaster> iirc the changes consisted in: build system for mpw + a few small patches
21:13:41 <AnMaster> ick is written in relatively portable C mostly
21:14:03 <AnMaster> calamari, oh and it won't compile for you since MPW tools can't invoke other MPW tools
21:14:11 <AnMaster> so it will print the command into the mpw shell
21:16:29 <AnMaster> calamari, oh and you have to convert any INTERCAL files to use CR for line ending
21:16:46 -!- augur has quit (Remote host closed the connection).
21:17:06 -!- augur has joined.
21:18:10 <AnMaster> calamari, I can make a patch against ick-0.29 if you want
21:19:28 <AnMaster> calamari, most of the patch is handling the crazy path stuff
21:20:02 <AnMaster> you see, ick generated paths like foo//bar.c instead of foo/bar.c (harmless) but foo:bar.c != foo::bar.c
21:20:08 <AnMaster> the latter is like foo/../bar.c
21:20:37 <AnMaster> also absolute path is messy to figure out
21:20:55 <AnMaster> :foo <-- also file in current dir
21:21:18 <AnMaster> or rather, that last one should be /mnt/foo/bar
21:21:44 <calamari> still copying the hard drive image off cd
21:24:20 <AnMaster> calamari, oh and yes you will need to change build system for pre-PPC. the ppc C compiler and the 68k ones had completely different command line options
21:24:57 <AnMaster> afraid I can't help much there. but iirc it was simple to get help in mpw
21:27:32 <AnMaster> calamari, I can't pastebin this file because line ending must not be converted in the patch, it is very important that the files that have CR continue to do so
21:27:40 <AnMaster> so I'm going to upload it elsewhere
21:28:27 <calamari> okay looks like I do have mpw installed
21:28:49 <AnMaster> calamari, does it work? try some example included
21:28:55 <AnMaster> also helps you figure out how it works
21:29:03 <AnMaster> because I haven't written a readme or anything
21:31:48 <AnMaster> calamari, ok remembered an old shell I have, putting it up there
21:32:28 <AnMaster> note this won't work with ipv6 (they have an AAAA entry but it seems broken atm, mentioned it to the guy who owns it, he is asleep atm though): http://pubacc.wilcox-tech.com/~anmaster/ick_classic_macppc.diff
21:32:40 <AnMaster> calamari, needs to be applied on ick 0.29
21:33:15 <AnMaster> calamari, the build system used will be in the dir macppc, and yes you will need to do something about the makefile :(
21:33:19 <calamari> btw stupid question.. what is ick?
21:33:57 <AnMaster> calamari, what else would one port to classic mac os
21:34:20 <AnMaster> ick at least offers a challenge
21:34:35 <calamari> okay having trouble getting my networking going in the emulator
21:34:41 <AnMaster> and ais523 (ick maintainer) mentioned several times it even works on DJGPP
21:34:49 <AnMaster> so thus I had to take the challenge
21:34:55 <AnMaster> calamari, I used the shared disk thingy
21:35:05 <AnMaster> calamari, isn't there a volume called "unix" or such on the desktop?
21:35:22 <AnMaster> calamari, well where did you set it to point on the real file system?
21:35:27 <calamari> but I was to get netowrking going anyhow lol
21:35:33 <AnMaster> just put something in there and it will be visible
21:35:51 <AnMaster> calamari, iirc that was quite easy hm...
21:36:46 <AnMaster> calamari, oh the mac port uses a hand written config.h :)
21:36:56 <AnMaster> it might need to be fixed to work on 68k
21:37:05 <AnMaster> I wouldn't be surprised if it needed even
21:37:27 <AnMaster> calamari, oh and the makefile is in macroman :)
21:37:35 <AnMaster> I do hope patch handles that properly
21:37:39 <AnMaster> I created the diff with bzr diff
21:38:20 <AnMaster> calamari, huh? what did you do?
21:38:35 <calamari> it think it might just be trying to active dhcp
21:38:53 <AnMaster> calamari, if you want I could .img the entire build dir (.img is likely to be most likely to preserve the encoding and everything else) but the object files will be ppc of course
21:39:02 <AnMaster> still, it might work better than that diff
21:39:32 <calamari> 403 forbidden on http://pubacc.wilcox-tech.com/~anmaster/ick_classic_macppc.diff
21:40:06 -!- poiuy_qwert has quit (Ping timeout: 245 seconds).
21:40:41 <AnMaster> # Download C-INTERCAL from this server.
21:40:41 <AnMaster> # Download C-INTERCAL via gopher (IPv6 only).
21:41:15 <AnMaster> http://overload.intercal.org.uk/c/
21:41:21 <AnMaster> calamari, hope that helps anyway
21:42:47 -!- poiuy_qwert has joined.
21:43:18 <calamari> can't seem to connect to http://overload.intercal.org.uk/c/
21:43:42 <AnMaster> calamari, what about http://c.intercal.org.uk/
21:44:10 <AnMaster> calamari, ah the ipv6 works but not the ipv4
21:44:28 <AnMaster> calamari, get an ipv6 tunnel, or wait a few minutes for me to upload it on that shell as well
21:45:53 <AnMaster> calamari, okay it should be up on the same place as my patch
21:46:00 <Ilari> There are gopher clients that support IPv6?
21:46:53 <AnMaster> calamari, now I hope it applies to that one. If not try running autoreconf and then applying it. It seems I might have done that before starting working, and who knows what that might have done to config.h.in and such
21:47:03 <Ilari> Got direct link to download? :-)
21:47:08 <calamari> cool I think I will have to sign up and get a shell with them
21:47:08 <AnMaster> calamari, but even if it fails on any *.in files it should work well
21:47:26 <AnMaster> Ilari, and it is on http://c.intercal.org.uk/
21:47:33 <AnMaster> "# Download C-INTERCAL from this server.
21:47:33 <AnMaster> # Download C-INTERCAL via gopher (IPv6 only)."
21:47:49 <AnMaster> Ilari, note that very server seems broken over ipv4 but works like a charm over ipv6
21:47:53 -!- Rugxulo has joined.
21:47:59 <AnMaster> gopher://gopher.intercal.org.uk/1C-INTERCAL <-- works for me
21:48:13 <AnMaster> calamari, and who is it you have to get a shell with?
21:48:28 <AnMaster> calamari, pubacc.wilcox-tech.com ? or what?
21:49:11 * Rugxulo isn't nearly the Atari enthusiast he probably should be
21:49:47 <AnMaster> calamari, quota is 100 MB but it depends on a lot on all-around niceness. It is a bit like "we don't add rules until we see they are needed"
21:50:05 <AnMaster> calamari, very nice shell, though I have access to some more on that server (involved in running an irc network who that guy who owns that shell is also involved in)
21:50:14 <AnMaster> don't have root on that one though
21:50:30 * Rugxulo blindly assumes calamari is heavy into AtariAge forums
21:50:37 <calamari> I am hosting with another place and they have an insane amount of restrictions
21:50:50 <calamari> haven't been on there in a while
21:51:25 <AnMaster> calamari, the broken ipv6 annoys me on pubacc though. *goes to use one of the other domain names and hopes there aren't and weird vhosts*
21:52:32 <pikhq> calamari: Not primarily.
21:52:45 <AnMaster> but well, we will need it soon
21:52:47 <pikhq> However, it damned well *will* become common sometime soon.
21:52:57 <pikhq> We've got about a year left of IPv4 adress space.
21:52:57 <calamari> I wonder if I could even get an ipv6 address from cox
21:53:23 <AnMaster> calamari, I use a sixxs tunnel
21:53:42 <AnMaster> works well, is fast (the end point for the tunnel is quite near here)
21:53:47 <Rugxulo> and yes, Ick works with DJGPP, I tested it remember?
21:54:12 <AnMaster> calamari, http://pubacc.wilcox-tech.com lists the rules iirc
21:54:25 <pikhq> It'll take a bit longer for us to be "absolutely out", as that is when the IANA runs out of allocations to RIRs.
21:54:46 <AnMaster> calamari, and I will be *very* *very* annoyed if you do anything like resource hogging on it, since one of the ircds run on it.
21:54:50 <pikhq> And, as such, the RIRs will still be able to make allocations for at least a bit of time.
21:55:27 <AnMaster> calamari, not mine, but I know the owner personally and I'm involved in that irc network as I said above
21:55:29 <calamari> nah I just have a simple webpage
21:55:38 <AnMaster> <AnMaster> calamari, very nice shell, though I have access to some more on that server (involved in running an irc network who that guy who owns that shell is also involved in)
21:55:51 <calamari> traffic is probably next to none
21:56:00 <AnMaster> well sure, I'm sure he will be okay
21:56:06 <calamari> actually I bet the mnost traffic is via esolangs.org
21:56:18 <AnMaster> calamari, it is however in US, so US laws and such would apply
21:56:24 <AnMaster> to whatever extent that applies
21:56:50 <AnMaster> Rugxulo, well I was thinking about DMCA and so on
21:57:08 <AnMaster> and I very much doubt the owner would like that
21:57:37 * Rugxulo missed what exactly you two were trying to accomplish ... ??
21:57:54 <AnMaster> Rugxulo, read the patch in http://pubacc.wilcox-tech.com/~anmaster/ :)
21:57:56 <calamari> I am trying to compile ick on 68k
21:58:11 <Rugxulo> so how did Atari 2600 and 5200 come up?
21:58:13 <AnMaster> but it will need some modification for 68k mpw
21:59:04 <Rugxulo> P.S. (to calamari) AnMaster wasn't even alive when 5200 came out! (I think)
21:59:15 <AnMaster> Rugxulo, depends. when did it come out?
21:59:45 <Rugxulo> I forget, '82 or '84 ?? must've been '82 or such as 1984 brought about the crash, and 7800 came out two years late ('86) when NES got popular
21:59:46 -!- Phantom_Hoover has joined.
22:00:05 <AnMaster> well, 1986 was before my time too
22:00:23 <Rugxulo> I'm not exactly well-versed myself ('79), but I do have a JagCD and (broken) Lynx II
22:00:54 <Rugxulo> JagCD and Lynx? more Atari machines
22:00:56 <calamari> Atari Jaguar with the toilet attachment
22:01:11 <AnMaster> calamari, how would CD stand for toilet?
22:01:22 <Rugxulo> no, it looks like a toilet (finds Wikipedia pic)
22:01:38 <Rugxulo> http://en.wikipedia.org/wiki/Atari_Jaguar_CD
22:01:50 <Rugxulo> had a 68000, too (among others)!
22:02:10 <AnMaster> Rugxulo, that looks surprisingly modern
22:02:19 <AnMaster> I mean using a cd instead of a cartridge
22:02:21 <Rugxulo> it was their last machine before "reverse merger" (go bye bye)
22:02:36 <Rugxulo> 1995-ish for the CD part, 1993 for the base cart part
22:02:43 <calamari> was the first 64 bit gaming system (well mostly 64 bit)
22:02:45 <Rugxulo> same guy who did built-in VLM in JagCD did one for XBox 360
22:02:56 <Phantom_Hoover> There are extremely minimal imperative, stack-based and functional programming.
22:03:02 <Rugxulo> mostly, yes ... even though everybody disagrees (as if it matters)
22:03:32 <Rugxulo> read the article, silly, it explains it ;-)
22:04:29 <AnMaster> "and a Myst demo disc. " <-- what?
22:04:40 <Rugxulo> Myst runs on many machines
22:04:58 <Rugxulo> the demo disc was just a "demo" (incomplete game) that came with the hardware with a few other CDs too (Tempest 2k soundtrack)
22:05:02 <AnMaster> bundled with an old performa iirc
22:05:09 <calamari> lol for that matter, you could make myst in dhtml
22:05:28 <AnMaster> calamari, with html5 and the video stuff yes
22:05:36 <Rugxulo> Blue Lightning (CD game), Tempest 2000 soundtrack (CD audio for VLM), Myst "demo", oh and Vid Grid (CD music video game)
22:05:43 <AnMaster> it did include videos for moving parts as bit of the screen in some places
22:05:44 <Rugxulo> that's what mine came with
22:05:56 -!- KingOfKarlsruhe has quit (Remote host closed the connection).
22:06:00 <calamari> yeah it did, but videos are easy these days
22:06:11 <AnMaster> calamari, oh and myst was and is a good game
22:06:20 <Rugxulo> I never even beat it, got fairly close though
22:06:33 <Rugxulo> got stuck in some underground tunnel, couldn't figure out what the heck to do
22:06:40 <Rugxulo> probably need to just get a walkthrough one of these days ;-)
22:06:41 <AnMaster> calamari, and that safe code in the log house? I found it on first try by pure chance
22:06:59 <Rugxulo> kinda tedious game, but whatever
22:07:03 * Rugxulo likes arcade games a bit more
22:07:20 <AnMaster> Rugxulo, oh that underground tunnel, different sounds for north/south/east/west iirc
22:07:24 <Rugxulo> Lynx had many more good arcade ports, though
22:07:24 <calamari> I still have the cd.. maybe it works under wine lol
22:07:28 <AnMaster> and combinations for different directions
22:07:36 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
22:07:39 <AnMaster> so you could rotate for the right direction
22:07:46 <AnMaster> calamari, only if quicktime does I bet
22:07:51 <AnMaster> since it used that at least on mac
22:08:03 <AnMaster> wouldn't surprise me if it used qt for windows too
22:08:40 <AnMaster> Rugxulo, if you were in that tunnel in some vessel that is
22:08:57 <Rugxulo> heck, somebody even ported Tempest 2000 to DOS, but all I've ever found was a demo
22:09:21 <Rugxulo> although somebody else remade it as Tsunami 2010 for Windows (but heavily depended on gfx card for some stuff)
22:09:23 -!- Phantom_Hoover has joined.
22:09:31 <Rugxulo> shooting alien bug thingies coming out of a vortex / hole
22:09:43 <calamari> I knew a guy that ported Tempest to an enhanced dvd player
22:09:52 <Rugxulo> http://en.wikipedia.org/wiki/Tempest_2000
22:10:06 <Rugxulo> Minter did Tempest 2000 (Jag) and Tempest 3000 (Nuon DVD)
22:10:16 <AnMaster> with just different enemies I mean
22:10:37 <Rugxulo> same tube / hole idea, but with many different enemies and obstacles
22:10:38 <AnMaster> Phantom_Hoover, make an esoteric one
22:10:49 <calamari> http://en.wikipedia.org/wiki/Tempest_3000
22:11:00 <AnMaster> Rugxulo, I always preferred RPGs
22:11:03 <calamari> I should say I knew the guy working on nuon
22:11:17 <Rugxulo> "the guy" ... who did what?
22:11:28 <Rugxulo> boring? no way, it's a classic, seriously!
22:11:31 <AnMaster> Rugxulo, <calamari> http://en.wikipedia.org/wiki/Tempest_3000
22:11:39 <AnMaster> Rugxulo, so is pacman, it is also boring
22:11:49 <Rugxulo> Pac-Man is very hard, that's why it's boring (to me)
22:12:01 <Rugxulo> Ms. Pac-Man is a bit more fun
22:12:02 <AnMaster> Rugxulo, I want a story in my games. Zelda (not zelda 1 though), secret of mana, games like that
22:12:26 <Phantom_Hoover> Except it has enough variation that if you don't die for long enough it gets entertaining.
22:12:27 <AnMaster> Phantom_Hoover, well, nethack is technically interesting and somewhat varied.
22:12:33 <Rugxulo> arcade games usually don't have stories, they are good mostly for quick sessions, trying to get high scores, completing levels, etc.
22:12:37 <AnMaster> plus there are lots of ways to die
22:12:42 <AnMaster> Phantom_Hoover, funny ones too
22:12:48 <AnMaster> Phantom_Hoover, but ascended a few times :)
22:12:52 <Rugxulo> you can't even die in Myst (although you might want to, heh)
22:13:03 <AnMaster> Phantom_Hoover, I mean outside debug mode
22:13:07 <AnMaster> but sure debug mode is interesting
22:13:14 <AnMaster> riding a black dragon for example
22:13:26 <AnMaster> which can happen otherwise too
22:13:44 <Phantom_Hoover> And then it was only because I found a /oD in a bones file.
22:13:45 <Rugxulo> BTW, worst . game . ever ... Presswurst (DON'T even bother reading about it)
22:13:51 <AnMaster> <Rugxulo> you can't even die in Myst (although you might want to, heh) <-- you can loose the game though
22:14:00 <AnMaster> Rugxulo, there are two ways to do it iirc
22:14:17 <Rugxulo> lose as in not doing such and such correctly at the end? (yeah, I know)
22:14:18 <AnMaster> Rugxulo, but myth is interesting, there is a story to it
22:14:30 <AnMaster> if you read the books in the library and so on
22:14:31 <Rugxulo> it's interesting, just somewhat easy to get lost
22:14:49 <Rugxulo> I don't like games that let you run around in circles, esp. since I always end up doing that out of confusion!
22:14:50 <AnMaster> Rugxulo, nah, I have a paper with the codes somewhere, wrote them down to make replay easier
22:14:59 <Rugxulo> me too, but it's still tough
22:15:09 <AnMaster> because yes rotating that tower to get the codes, that is boring
22:15:41 <Rugxulo> it's been at least four or five years since I've played it
22:16:16 <Rugxulo> I wonder if Tsunami 2010 works in WINE
22:16:37 <Rugxulo> "opengl accelerator recommended" ... probably a dealbreaker
22:16:51 <Rugxulo> I remember it liked one video card, the other not so much :-/
22:17:00 -!- hiato has quit (Quit: underflow).
22:17:30 <calamari> Rugxulo: did you get lost when you first played Doom? I did, it was great
22:18:07 <Rugxulo> can't remember, been a long time since I first played Doom
22:18:27 <Rugxulo> nowadays I'd rather play FreeDoom ;-)
22:21:12 <AnMaster> calamari, tell me if you get ick working, it should be possible to base the build system on the ppc one, if you get it to work please send me a patch and I shall try to integrate it so both works
22:21:31 <AnMaster> calamari, oh and that .pax.gz, it can be unpacked with tar
22:21:47 <AnMaster> I'll leave it to ais523 to explain what pax is
22:22:01 <AnMaster> or anyone else who knows the history of POSIX, tar and pax
22:22:33 <AnMaster> Phantom_Hoover, yes mostly, but it was done because there were so many incompatible tar versions out there
22:22:47 <AnMaster> iirc pax is mostly tar compatible too
22:23:22 <AnMaster> Phantom_Hoover, well pax was a compromise because no vendor wanted to switch to another vendors' tar format
22:23:40 <AnMaster> but yes iirc there is a cpio←→tar war
22:23:54 <AnMaster> don't think pax is involved in it
22:24:07 <AnMaster> calamari, iirc rpm are cpio-based too
22:24:08 <Rugxulo> pax involved in war? that would be irony ;-)
22:25:00 <AnMaster> http://en.wikipedia.org/wiki/Pax_%28Unix%29
22:25:05 <AnMaster> calamari, and I prefer arch linux
22:25:23 <Rugxulo> lzip is gaining in popularity, too
22:25:44 <AnMaster> "Rather than sort out the incompatible options that have crept up between tar and cpio, along with their implementations across various versions of UNIX, the IEEE designed a new archive utility."
22:25:52 <pikhq> calamari: xz is a compression format that uses the LZMA algorithm.
22:26:06 <AnMaster> though that ignores the tar-tar wars too
22:26:09 <pikhq> It's the defacto successor to the "lzma" program.
22:26:20 <AnMaster> pikhq, and incompatible with lzma
22:26:33 <Rugxulo> .7z is the official format, but it's not really 100% *nix friendly (he focuses a lot on Win32)
22:26:36 <AnMaster> pikhq, as in, lzma can't unpack xz, annoying on my jaunty laptop
22:26:36 <pikhq> Though the xz program supports lzma files just fine.
22:26:46 <Phantom_Hoover> Why do we need 20-odd archive-compression combinations?
22:26:57 <Rugxulo> p7zip should be able to unpack .xz if you use 9.04, but sadly that version is fairly old and buggy
22:26:59 <AnMaster> pikhq, what was wrong with .lzma?
22:27:09 <pikhq> AnMaster: I don't recall.
22:27:13 <calamari> no clue.. gz and bz2 seem fine
22:27:19 <Rugxulo> lzma isn't a real format, just a raw data dump (I think?)
22:27:33 <Rugxulo> .7z (LZMA) is much much stronger and faster than Bzip2
22:27:34 <pikhq> Rugxulo: No, it had headers and everything, I though.\
22:27:50 <Rugxulo> well it lacks something (permissions? Unicode? large files?)
22:28:07 <calamari> what does it strip another 100 bytes off? lol
22:28:21 <AnMaster> Rugxulo, .lzma isn't an archive format
22:28:32 <Rugxulo> the biggest block size in Bzip2 is 900k, I think 7-Zip can go into the GB for dictionaries
22:28:33 <AnMaster> after all both are used over tar
22:28:48 <pikhq> calamari: Significantly better compression than bzip2, and much faster decompression.
22:28:51 <Rugxulo> only because of *nix permissions, on Win32 nobody cares about .tar
22:29:18 <Rugxulo> 7-Zip can do Bzip2 compression in .7z, .Zip, or .bz2 formats
22:29:31 <Rugxulo> or use LZMA, LZMA2, PPMD, etc.
22:30:07 <pikhq> *Ah*. The .lzma format had stupid, arbitrary limits on the compression options for the LZMA algorithm.
22:30:13 <Rugxulo> in fact, I think the fastest way to compress with 7-Zip is "-mx1 -m0=bzip2"
22:30:52 <calamari> okay installed xz, so now I can do tar -Jcvf :)
22:30:52 <Rugxulo> (sees topic) "Esperanto is still bannable!" ... huh??
22:31:11 <pikhq> (such as limits on the dictionary size)
22:31:40 <pikhq> Rugxulo: *Surely* gzip is faster.
22:32:00 <Rugxulo> probably, it only uses 32k dictionaries (or whatever)
22:32:19 <Rugxulo> but 7-Zip has its own improved Deflate, even eeks out a smaller file ;-)
22:32:38 <pikhq> What with gzip being one of the faster compression formats in common use and all...
22:32:45 <pikhq> calamari: Strictly better, IIRC.
22:32:47 <Rugxulo> Rar is proprietary, only Unrar is semi-free
22:32:56 <Rugxulo> but it's okay, from what I hear
22:33:08 <calamari> yeah.. rar is used a lot in piracy
22:33:27 <Rugxulo> Rar is popular in Russia, Lha is still popular in Japan, Zip in U.S., etc. etc.
22:33:34 <AnMaster> <Rugxulo> only because of *nix permissions, on Win32 nobody cares about .tar <-- sure, and I don't care about win32
22:33:45 <AnMaster> though win64 is marginally more interesting
22:33:47 <Rugxulo> just saying, .7z by itself is fine for us ;-)
22:33:55 <AnMaster> Rugxulo, it isn't like .7z or .zip can do windows ACLs either
22:34:20 <pikhq> .tar also lets you archive things like block devices. :)
22:34:24 <AnMaster> Rugxulo, oh and bzip2 is very slow to compress
22:34:31 -!- sshc has quit (Quit: leaving).
22:34:39 <AnMaster> pikhq, not the data in those though
22:34:46 <Rugxulo> pretty slow, yes, but it claims to be faster than PPMD while still being fairly competitive
22:35:00 <AnMaster> but compared to gzip or whatever
22:35:04 <Ilari> What's that punycode name in /topic?
22:35:12 <pikhq> Rugxulo: PPMD is *extremely* slow.
22:35:15 <pikhq> Ilari: Just a Unicode string.
22:35:45 <Rugxulo> no, PAQ is extremely slow ;-)
22:35:57 <pikhq> IIRC, 僕は問題にユニコードが好きだ。
22:36:11 <pikhq> Might be a permutation thereof, though.
22:37:08 -!- impomatic has joined.
22:37:09 <calamari> as long as klingon isn't bannable
22:37:13 <AnMaster> <Rugxulo> Rar is popular in Russia, Lha is still popular in Japan, Zip in U.S., etc. etc. <-- the only files I run into on daily basis are .tar.{gz,xz,bz2,lzma}, and very rarely .zip. A few times (when messing with Mac OS classic simulators): .zip
22:37:36 <impomatic> Has anyone seen this on Amazon? "Befunge: Stack- oriented Programming Language, Reflection (computer science), Esoteric Programming Language, Programming Language, Brainfuck, INTERCAL, Whitespace (programming language), Malbolge (Paperback)" http://bit.ly/cYWyWE
22:37:37 <Rugxulo> .zip is still used very frequently
22:38:01 <impomatic> Is the Klingon programming language still online? It used to be on Geocities.
22:38:13 <AnMaster> but true, it is used in .jar, .odt and so on
22:38:24 <AnMaster> Rugxulo, I said which ones I ran into
22:38:24 <pikhq> AnMaster: "I love Unicode in my topics."
22:38:35 <pikhq> I had also put the same in English into the topic.
22:39:41 -!- AnMaster has set topic: Howsabout we put "esoteric programming languages" SOMEWHERE in the topic? | Esperanto is still bannable! | Je peux utiliser une langue étrangère aussi. | xn--v8jad0f7b6z4eoa6v0hk534a7hlwhnnl8s. | Jag älskar Unicode i mina topicar. | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
22:39:49 <AnMaster> need to find a Swedish word for topics
22:40:02 * Rugxulo confused about E-o reference
22:40:33 <AnMaster> * Rugxulo was kicked from #esoteric (bannable!)
22:40:56 <Rugxulo> pikhq, any comments? (seen you mention it before)
22:41:27 <Rugxulo> Swedish word? smorgasborg?
22:41:56 <AnMaster> Rugxulo, none of them are Swedish
22:42:13 <Rugxulo> I know gesundheit isn't, that was a (corny) joke, but the other either??
22:42:21 <AnMaster> the first one means <nonsense>-gas-fortress
22:42:37 <AnMaster> (gas as in "not solid" not as in "fuel")
22:42:58 <AnMaster> Rugxulo, but with some dots over things it may mean completely different things
22:43:25 * Rugxulo not Swedish, not know spell good thusly
22:43:34 <AnMaster> Rugxulo, that is why forgetting dots and rings in Swedish is such a bad idea. you end up with something meaning something completely different
22:43:36 <impomatic> That Esolang book is available on the German Amazon. 34 Euros :-(
22:43:48 <AnMaster> Rugxulo, as in gas = gas, gås = gose
22:44:14 <Rugxulo> I know I know, accents, diacritics, blah ... not something I use a lot (in English) ;-)
22:44:29 <AnMaster> Rugxulo, they are different letters in the alphabet
22:44:33 <Rugxulo> I know I know ... special characters
22:44:34 <AnMaster> Rugxulo, and well, borg = fortress, bord = table
22:44:48 <Rugxulo> 34 EU is too much for 84 pages!!
22:44:56 <AnMaster> Rugxulo, I was unable to guess what you meant though apart from "smör"
22:45:10 <Rugxulo> E-o has 28 letters, six of which have special marks over them (circumflex * 5, breve * 1)
22:45:16 <impomatic> Esolang book on German Amazon http://bit.ly/avuLtH
22:45:35 <Rugxulo> "Smorgasborg is spelled smrgsbord
22:45:36 <Rugxulo> if you refer to a buffet in swedish.
22:45:36 <Rugxulo> (smrgs is a sandwich and bord a table in swedish)"
22:46:22 <AnMaster> Rugxulo, but the original word had in part a sensible and different meaning
22:46:58 <Rugxulo> "Sprache: Englisch" ... and yet not available except on German section of Amazon??
22:48:47 <calamari> use lulu.com to make your esolang book :)
22:49:54 <impomatic> Actually I there are 23246 books by the same three authors. I think they must me scraping content from the net and publishing it as print on demand or something. http://bit.ly/bqvVWb
22:50:39 <impomatic> Shame, I thought it'd be quite cool if someone actually wrote an esolang book :-(
22:50:42 <Rugxulo> it certainly looks weird (long obscure titles, boring subjects)
22:52:23 <calamari> Chris Pressey would have to be the author of an esolang book anyhow
22:53:17 <uorygl> `translatefromto en es Happy birthday!
22:53:34 <Rugxulo> `translatefromto en eo Happy birthday!
22:54:05 <calamari> `translatefromto en bf Happy birthday!
22:54:08 * AnMaster hits xilinx with something very very heavy
22:54:08 <HackEgo> var sl_select, tl_select, web_sl_select, web_tl_select;var ctr, web_ctr, h;var tld = ".com";var sug_lab = "";var sug_thk = "";var sug_exp = "";var dhead = "Dictionary";var dmore = "View detailed dictionary";var tr_in = "Translating...";var isurl = "";var show_roman = "Show romanization";var hide_roman =
22:54:11 <Rugxulo> oh, BTW, found a recursive .ZIP earlier today ;-)
22:54:27 <AnMaster> and stupid thing is I need a single file from it probably
22:54:42 -!- BeholdMyGlory has quit (Remote host closed the connection).
22:54:47 * Rugxulo wonders if he accidentally broke HackEgo
22:54:57 <Rugxulo> oh, whew, you broke it, not me ;-)
22:55:00 <uorygl> Nah, HackEgo was already broken.
22:55:08 <calamari> `translatefromto en es Happy birthday!
22:55:11 <Rugxulo> there is a command called "textbf" or something
22:55:23 <uorygl> `translatefromto en bf Look how broken you are!
22:55:25 <HackEgo> var sl_select, tl_select, web_sl_select, web_tl_select;var ctr, web_ctr, h;var tld = ".com";var sug_lab = "";var sug_thk = "";var sug_exp = "";var dhead = "Dictionary";var dmore = "View detailed dictionary";var tr_in = "Translating...";var isurl = "";var show_roman = "Show romanization";var hide_roman =
22:55:28 <Rugxulo> well, I forget, but it does exist
22:55:48 <HackEgo> bin \ cube2.base64 \ cube2.jpg \ hack_gregor \ hello.txt \ help.txt \ huh \ netcat-0.7.1 \ netcat-0.7.1.tar.gz \ out.txt \ paste \ poetry.txt \ quotes \ share \ test.sh \ tmpdir.23634 \ wunderbar_emporium \ wunderbar_emporium-3.tgz \ wunderbar_emporium-3.tgz.1
22:55:49 <HackEgo> ? \ addquote \ calc \ commands \ creatures \ define \ esolang \ etymology \ fortune \ google \ helpme \ imdb \ karma \ marco \ minifind \ paste \ ping \ quote \ rec \ roll \ runfor \ sayhi \ strfile \ swedish \ toutf8 \ translate \ translatefromto \ translateto \ unstr \ url \ wolfram
22:56:17 <uorygl> `run translatefromto en es Happy birthday! | toutf8
22:56:17 <fizzie> It has "gen" in the name somewhere, and wasn't it an egobot thing?
22:56:22 <HackEgo> var sl_select, tl_select, web_sl_select, web_tl_select;var ctr, web_ctr, h;var tld = ".com";var sug_lab = "";var sug_thk = "";var sug_exp = "";var dhead = "Dictionary";var dmore = "View detailed dictionary";var tr_in = "Translating...";var isurl = "";var show_roman = "Show romanization";var hide_roman =
22:56:33 <uorygl> bftxtgen was an EgoBot thing, yeah.
22:56:42 <HackEgo> ?, addquote, calc, commands, creatures, define, esolang, etymology, fortune, \ google, helpme, imdb, karma, marco, minifind, paste, ping, quote, rec, roll, \ runfor, sayhi, strfile, swedish, toutf8, translate, translatefromto, \ translateto, unstr, url, wolfram
22:56:46 <AnMaster> `run rm -rf wunderbar_emporium*
22:56:46 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
22:56:53 <AnMaster> `run rm -rf */wunderbar_emporium*
22:57:01 <EgoBot> help: General commands: !help, !info, !bf_txtgen. See also !help languages, !help userinterps. You can get help on some commands by typing !help <command>.
22:57:12 <AnMaster> Rugxulo, that can be a bit slow
22:57:16 <EgoBot> 41 ++++++++[>+++++++++++++>+>><<<<-]>.+.>++. [61]
22:57:28 <AnMaster> Rugxulo, it isn't very efficient for short strings
22:57:33 <AnMaster> does it with genetic algorithms
22:57:48 <EgoBot> 102 ++++++++[>+>+++++++++++++>++++><<<<-]>>.---.+++++++..+++.>.<++++++++.--------.+++.------.--------.<++. [763]
22:57:50 <fizzie> See for example >><< there.
22:58:21 <fizzie> Since it's always 4 cells in the multiplication loop.
22:58:24 <EgoBot> languages: Esoteric: 1l 2l adjust asm axo bch befunge befunge98 bf bf8 bf16 bf32 boolfuck cintercal clcintercal dimensifuck glass glypho haskell kipple lambda lazyk linguine malbolge pbrain perl qbf rail rhotor sadol sceql trigger udage01 underload unlambda whirl. Competitive: bfjoust fyb. Other: asm c cxx forth sh.
22:58:25 <calamari> had completely forgotten he integrated it into egobot
22:58:53 <uorygl> Genetic algorithms? That seems like not the best way to do that.
22:59:29 <calamari> uorygl: I'd encourage you to find the best way to do it.. then be sure to write a paper on it :)
22:59:55 -!- poiuy_qwert has quit (Quit: Leaving).
23:00:00 <EgoBot> 94 +++++++++[>+++++++++++++>+++++++++>+++++>+<<<<-]>>.<.>++++++++++++++++++++++++.<-.>----.>+.>+. [928]
23:00:23 -!- tombom has quit (Quit: Leaving).
23:00:24 <uorygl> !bf_txtgen 1[aj[jaj1[[1jaj[a1j[1j1j[a[[j1jaja[1j1[aj
23:00:27 <EgoBot> 286 ++++++++++++[>++++>+++++++>++++++++>++++<<<<-]>+.>+++++++.>+.+++++++++.<.>.<++++++.>.<<.>------..<.>>.<++++++.>.---------------.<.<.>+++++++++.>.>+.<<.<.>.>.<<++++++++++++++++++++++++++++++++++++++++++++++++.>>..<.>>.<<.<.>.<.>>.>.<<.>>.<.<<.>.>>---------------------------------------. [572]
23:00:31 <calamari> since it is returning quickly he must have the interations low
23:01:27 <uorygl> !bf_txtgen yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy
23:01:31 <EgoBot> 228 +++++++++++++++[>++++++++>++++++++>++++++++>++++++++<<<<-]>+..>+.......>+.......<..........<.....>>.>+......<....>..---------------------------------------------------------------------------------------------------------------. [357]
23:01:33 <AnMaster> calamari, isn't the iteration count the number in []?
23:01:41 <uorygl> That is probably a really stupid solution.
23:01:51 <AnMaster> calamari, then the number at the front?
23:01:57 <AnMaster> calamari, stripping those >><< would be a good idea btw
23:02:29 <AnMaster> uorygl, but it adds a newline at the end
23:02:31 <uorygl> Better: +++++++++++[>+++++++++++>>><<<<-]>..............................................>.
23:02:39 <calamari> let it run longer it does better :)
23:02:41 -!- FireFly has quit (Quit: null).
23:02:56 <AnMaster> but a human can do even better
23:03:20 <EgoBot> 20 +++++[>++>>><<<<-]>. [11]
23:03:43 <uorygl> Oh, I guess newline and null are not the same character. :P
23:03:44 <AnMaster> fizzie, too lazy to count those first pluses
23:04:05 <AnMaster> uorygl, what made you think they were?
23:04:09 <calamari> you could run the output through a cleaner to get rid of the ><
23:04:42 <AnMaster> should be done in genetic algo though
23:04:48 <AnMaster> to favour short solutions properly
23:04:49 <fizzie> BrainFuck text generation: not an especially active field of research, I guess.
23:04:52 <calamari> it's a pretty hard problem to solve tho
23:04:55 <AnMaster> I mean, if you only remove those at the end
23:05:12 <AnMaster> in advantage for a worse solution
23:05:39 <calamari> well the reason it has those is because you set the number of memory cells used as an option
23:05:40 <impomatic> http://en.wikipedia.org/wiki/VDM_Publishing is the scam company producing all the books with material scraped from the net.
23:05:43 <AnMaster> mathematica shows a license dialog at start
23:05:49 <AnMaster> are they limited to one year or something?
23:05:53 <uorygl> Suddenly, I want to make a programming language based on LZW.
23:06:10 <calamari> so they aren't impacting the result
23:06:52 <calamari> usually, you would experient with different numbers of cells to see which gives the best result
23:07:06 <AnMaster> oh I see, it only works when network is up how strange
23:07:26 <AnMaster> calamari, no, as in ethernet hardware running
23:07:32 -!- Mathnerd314 has quit (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.2/20100316074819]).
23:08:11 <AnMaster> I had it removed to save battery before (it does make a difference of about 0.25 W according to powertop)
23:08:24 <AnMaster> wlan saves even more when down of course
23:08:38 <fizzie> They do have that license server nonsense, but one would think it'd be enough to have a loopback interface up.
23:08:51 <AnMaster> fizzie, I have to assume it was up
23:09:27 <fizzie> Maybe they've keyed the license to the Ethernet MAC. Changed NICs? Buy a new copy.
23:10:24 -!- augur has quit (Ping timeout: 260 seconds).
23:10:28 <AnMaster> okay *kicks Xilinx*. This is the worst idea for download ever invented
23:10:46 <AnMaster> the main file: a 2.9 GB uncompressed tar
23:11:02 <Rugxulo> what were you downloading anyways?
23:11:09 <AnMaster> containing according to tar -tf amongst other things large uncompressed files but also:
23:11:14 <AnMaster> Xilinx_ISE_DS_Lin_12.1_M.53d.0.4/idata/drop_0122_ise.zip.xz
23:11:36 <AnMaster> Rugxulo, well, if it is Xilinx it should be easy to guess
23:12:34 <fizzie> I've seen quite many .rars inside .rars inside .rars.
23:12:44 <AnMaster> Rugxulo, anyway Xilinx makes FPGAs
23:12:54 <AnMaster> I think that should answer what I'm downloading
23:13:03 <Rugxulo> also, seems p7zip is *still* stuck at 9.04 beta (bah)
23:13:12 <AnMaster> Rugxulo, I don't see the point of zip
23:13:18 <Rugxulo> fizzie, have you seen recursive archives yet?
23:13:40 <AnMaster> at least it has both bin/lin and bin/lin64 in it
23:13:53 <AnMaster> then to install the thing in a chroot of course
23:14:04 <AnMaster> I wouldn't trust xilinx an inch
23:14:12 <fizzie> Rugxulo: Yes, if you mean that .zip/etc that extracts into a copy of itself.
23:14:27 <AnMaster> unless I can extract only the gate timing info somehow
23:15:06 <fizzie> AnMaster: Grep logs for r.zip here. I have just this phone now.
23:15:26 <AnMaster> fizzie, n900 is not "just this phone" it is "wow this amazing phone"
23:15:28 <fizzie> I'd like to see a chain version; a .zip that extracts to a .rar that extracts to a .tar.xz tha extracts to the original .zip.
23:15:44 <AnMaster> fizzie, make that yourself! :D
23:16:00 <fizzie> http://swtch.com/r.zip anyhow.
23:16:21 <fizzie> Extracts into r/r.zip with the same contents.
23:16:29 <Rugxulo> http://research.swtch.com/2010/03/zip-files-all-way-down.html
23:16:48 * calamari wants an n900, but can't afford it
23:17:40 * Rugxulo wants a calamari but can't palate it
23:21:57 <uorygl> Why do you want the impalatable?
23:23:23 <AnMaster> "The nice thing about r.gz is that even broken web browsers that ordinarily decompress downloaded gzip data before storing it to disk will handle this file correctly! "
23:23:32 <AnMaster> hm I hope none do it recursively
23:23:37 <AnMaster> probably aren't coded to do that
23:26:17 -!- oerjan has joined.
23:27:21 <fizzie> AnMaster: From the comments: "In Safari on Mac OS X, with the option 'open "safe" files after downloading' set (on by default), it keeps uncompressing and filling up my disk. "safe" indeed ..."
23:27:32 <fizzie> (That was for the .zip.)
23:28:32 -!- impomatic has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.9/20100315083431]).
23:29:37 -!- Sgeo has joined.
23:31:16 <uorygl> http://research.swtch.com/2010/03/zip-files-all-way-down.html
23:31:26 <uorygl> "Recursion for the win." Ironically, his solution uses no recursion.
23:31:27 -!- Sgeo_ has joined.
23:31:47 <Rugxulo> huh, GCC will include gccgo in future versions
23:32:23 * uorygl wonders for a moment whether that's a compiler that takes a Go move and compiles it to a Go board with a response.
23:32:28 -!- Sgeo has quit (Client Quit).
23:33:27 <Rugxulo> no, it's that new Google language
23:37:52 <Rugxulo> ever heard of this? http://github.com/bonzini/superopt
23:42:55 <fizzie> Unfortunately I'm already asleep. (Or at least close enough for practical purposes.)
23:43:52 <fizzie> (Also called "the engineer's asleep".)
23:44:57 <AnMaster> Rugxulo, I heard of superoptimisers before
23:45:03 <AnMaster> Rugxulo, but that is length not time
23:45:12 <AnMaster> some longer sequence might be faster
23:45:23 <AnMaster> (x87 vs. SSE comes to mind here)
23:45:46 <Rugxulo> x86 is a pain to optimize for, seems impossible o_O
23:46:13 <pikhq> Different things are faster on different architectures.
23:46:17 <AnMaster> Rugxulo, same goes for many other arches though... cache effects, some nops to align on cache boundary might make it faster and so on
23:46:32 <uorygl> You know, I now wonder why OS X uses x86 these days.
23:46:57 <pikhq> uorygl: Because Intel offers better production guarantees on CPUs than IBM.
23:46:59 <AnMaster> it isn't just x86 where you have cache effects and god know what else though
23:47:10 <pikhq> Erm. s/architectures/chips/
23:47:34 <pikhq> AnMaster: x86 is worse about it than others.
23:47:35 <uorygl> It's also true that different things are faster on different architectures, of course.
23:47:41 <Rugxulo> Apple want fastest available, and Intel finally caught up to PPC in their eyes, I guess (while being cheap enough)
23:47:45 <uorygl> A BF processor would not be very fast.
23:48:06 -!- augur has joined.
23:48:28 <uorygl> Its instructions are small, though. It's difficult to beat three bits.
23:48:43 <uorygl> Unless that's patted to 32 bits or something.
23:48:58 <pikhq> uorygl: Why would it?
23:49:09 <pikhq> Not like Brainfuck is a von neumann architecture.
23:49:11 <pikhq> AnMaster: Brainfuck.
23:49:24 <pikhq> Y'know, base 7? ;)
23:49:44 <AnMaster> pikhq, well sure, but I was thinking about x86 still
23:49:45 <Rugxulo> base7? does anybody use that (besides ETA)?
23:49:55 <pikhq> Rugxulo: Brainfuck does.
23:50:22 <pikhq> Its code space is base 7, and it's not von Neumann.
23:50:29 <pikhq> AnMaster: Well, no.
23:50:45 <AnMaster> Rugxulo, also sure, VHDL can use anything crazy
23:50:46 <pikhq> It would be *perfectly* feasible to have a base 7 Brainfuck.
23:50:52 <pikhq> (for which we kill you)
23:50:56 <AnMaster> 5 bits gray code? sure why not
23:51:15 <uorygl> Brainfuck's instructions are base 7?
23:51:16 <oerjan> pikhq: base 8, surely?
23:51:29 <pikhq> oerjan: 8 instructions. 0-7.
23:52:19 <uorygl> Hmm, calling it base 7 might be slightly better, as then every base can be written as a single digit in itself.
23:52:32 <uorygl> Now not everything is base 10 any more!
23:52:36 <oerjan> i think there are possibly several brainfuck derivatives on the wiki based on treating bf as base 8.
23:53:08 <uorygl> We can speak of base 1, base 7, base 9, base F, base /...
23:53:51 <calamari> bitchanger has 4 instructions, so you can pack 4 per byte
23:54:36 <oerjan> as you can plainly see, that doesn't work very well beyond the standardized digits (9, F or Z dependent on how geeky you are...)
23:54:41 <Rugxulo> ETA explicitly uses base7 for its numbers
23:55:27 <Rugxulo> except it maps them to the same letters as instructions (htaoins) and avoids "e", which is used as the number terminator
23:55:29 <uorygl> Hmm, how's this for a BF instruction set...
23:56:08 <uorygl> @ - flip the current bit and move right. & - move left and if the current bit is 1, jump to the beginning of the program.
23:56:12 <uorygl> Darn, that doesn't really work.
23:56:33 <uorygl> Unfortunately, 4 is an infinite nonrepeating decimal!
23:56:55 <uorygl> Aww, Wolfram Alpha doesn't do base pi.