00:00:00 --- log: started forth/03.09.21 00:00:04 --- quit: Herkamire ("bedtime") 00:12:37 * arke is back (gone 07:23:18) 00:51:56 * arke is away: sw33t sl33p, sw33t l4di3s in my dr34ms, st4y h3r3 with m3, l3ts h4v3 fun... 01:57:11 --- join: mur (murr@baana-62-165-186-174.phnet.fi) joined #forth 02:06:55 --- quit: arke (zelazny.freenode.net irc.freenode.net) 02:06:55 --- quit: Robert (zelazny.freenode.net irc.freenode.net) 02:06:55 --- quit: ChanServ (zelazny.freenode.net irc.freenode.net) 02:06:55 --- quit: I440r_ (zelazny.freenode.net irc.freenode.net) 02:06:55 --- quit: jstahuman (zelazny.freenode.net irc.freenode.net) 02:06:55 --- quit: mur (zelazny.freenode.net irc.freenode.net) 02:06:55 --- quit: SDO (zelazny.freenode.net irc.freenode.net) 02:06:56 --- quit: ianP (zelazny.freenode.net irc.freenode.net) 02:06:56 --- quit: onetom_ (zelazny.freenode.net irc.freenode.net) 02:06:57 --- quit: I440r-wrk (Read error: 104 (Connection reset by peer)) 02:08:07 --- join: ChanServ (ChanServ@services.) joined #forth 02:08:07 --- join: mmanning (~mmanning@64.238.237.215) joined #forth 02:08:07 --- join: mur (murr@baana-62-165-186-174.phnet.fi) joined #forth 02:08:07 --- join: I440r_ (~I440r@sdn-ap-005txhousP0299.dialsprint.net) joined #forth 02:08:07 --- join: arke (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth 02:08:07 --- join: SDO (~SDO@co-trinidad1a-42.clspco.adelphia.net) joined #forth 02:08:07 --- join: onetom_ (~tom@cab.bio.u-szeged.hu) joined #forth 02:08:07 --- join: jstahuman (dan@pcp053555pcs.brlngt01.nj.comcast.net) joined #forth 02:08:07 --- join: Robert (~snofs@h31n2fls31o965.telia.com) joined #forth 02:08:07 --- join: ianP (ian@inpuj.net) joined #forth 02:08:07 --- mode: zelazny.freenode.net set +o ChanServ 03:27:18 --- quit: I440r_ (Ping timeout: 14400 seconds) 04:16:45 --- join: TheBlueWizard (TheBlueWiz@pc25dn1d.ppp.FCC.NET) joined #forth 04:16:46 --- mode: ChanServ set +o TheBlueWizard 04:18:16 hiya all 04:19:18 terve :) 04:19:41 mitä kuuluu? 04:19:45 terve...feeling better? 04:20:43 me...nothing new other that I get on early to get done with Internet stuff before doing household chores today :) 04:21:13 :) 04:23:13 um...are you feeling better now? 04:27:35 hmm 04:27:38 physically? 04:28:52 last night you said you are not feeling well 04:29:04 and I said, "Go to bed" :) 04:29:14 Hi mur & TheBlueWizard 04:29:44 hiya Robert 04:29:54 oh 04:30:03 well i was feeling tired in morning but i'm oknow 04:49:17 good :) 04:49:35 well, gotta go now...bye all 04:49:39 cya 04:49:55 bye mur :) 04:49:59 --- part: TheBlueWizard left #forth 04:51:45 --- quit: jstahuman (zelazny.freenode.net irc.freenode.net) 04:52:43 --- join: jstahuman (dan@pcp053555pcs.brlngt01.nj.comcast.net) joined #forth 08:08:22 * arke is back (gone 07:16:24) 08:39:44 --- join: I440r (~I440r@sdn-ap-005txhousP0299.dialsprint.net) joined #forth 08:42:52 --- quit: arke (Read error: 110 (Connection timed out)) 08:46:31 Hey. 08:49:11 hi 09:00:19 --- join: I440r_ (~I440r@sdn-ap-019txhousP0392.dialsprint.net) joined #forth 09:18:26 --- join: XeF4 (~xef4@isozyme106.gprs.suomen2g.fi) joined #forth 09:19:19 hi 09:19:23 hi 09:20:05 xexexexexff 09:23:25 --- quit: I440r (No route to host) 09:46:57 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth 09:46:58 --- mode: ChanServ set +o kc5tja 09:51:04 hoi 10:37:39 --- join: Speuler (~Speuler@mnch-d9ba4801.pool.mediaWays.net) joined #forth 10:38:02 'day 10:38:15 --- quit: mur ("Murr.") 10:38:24 gdays 10:38:57 bongo :) 10:39:03 my x86/linux compiler is taking on shape 10:39:06 hi mark ! 10:39:21 can compile a number of things already 10:39:33 cool 10:39:42 creates stand-alone x86 code, ELF 10:39:56 u used isforth as a reference ? 10:40:04 reference ? 10:40:27 didn't know isforth can compile stand-alone code 10:40:28 as in "now how would i do that".... 10:40:47 I440r_, you seem to underestimate me :) 10:40:50 it can create turnkey applications that will run on any machine even if isforht isnt installed there 10:41:06 nono - im sure you couold figure it all out heh 10:41:15 small programs compile to about 100 bytes 10:41:19 FS/Forth for Linux will eventually be able to do the same. 10:41:33 I've been considering going that route myself. 10:41:34 100 bytes including ELF header? 10:41:38 yes 10:42:00 "program" is then trivial of course 10:42:06 XeF4: Look at some of the Linux assembly programs out there -- they totally bastardize and rape the ELF header by putting executable code into unused fields in the header. 10:42:18 i didn't 10:42:28 you mean the teensy executable files 10:42:33 Yes. 10:42:39 Like, file copy in 64 bytes or something. 10:42:40 yes - thats dumb 10:42:43 but is used it as ref to build my elf header 10:42:44 * kc5tja nods 10:42:48 that doesnt help anyone learn the elf stuff at all 10:42:56 kc5tja: I've written several such stubs myself. 10:43:12 And it's downright destructive when you need symbol relocation information. :-) 10:43:22 (which I think I'll be needing) 10:43:24 i don't do any optimizing right now 10:44:09 The only optimization I do in FS/Forth's target compiler is tail-call optimization. That's so I can get away with the reduced set of control-flow words. 10:44:19 an example of code is on http://www.forthfreak.net/index?NativeCodeCompiler 10:44:35 I will be adding support for literals and literal optimization shortly. 10:44:43 almost outdated, as it can compile colon words too now 10:44:47 That way, + can expand to add eax,n where appropriate, among other primitives. 10:45:06 : count 1+ dup 1- c@ ; : foo >r count + r> ; and the like 10:45:29 i keep the optimizing stuff for later 10:45:58 Since all my primitives are immediate words, it makes sense to exploit the x86 architecture up-front. 10:46:05 wondered whether anyone wants to join into the development 10:46:30 source is still pretty dirty 10:46:58 will have to clean it up a bit, once i can resolve from libraries 10:47:20 'll go the same route as fthcmp 10:47:44 Never heard of fthcmp. 10:47:45 lib included at the end of main prog, resolves all forward refs 10:47:52 Ahh 10:48:07 words included from lib may create new fwd refs 10:48:08 Which references would be unresolved? 10:48:23 the ones used in main prog, to be resolved from library 10:48:45 to make sure only required code goes init the executable 10:48:59 what is not referenced, is skipped 10:49:02 That's a circular definition if I'd ever heard one. 10:49:09 What references would those be? :) 10:49:15 : main 5 . ; 10:49:17 Forth primitives? OS-specific bindings? 10:49:17 . is unresolved 10:49:27 library included, resolves . 10:49:28 OK, you're going a totally different route than I am then. 10:49:37 . creates new fwd ref, for example to emit 10:49:52 I'm following the Chuck Moore cmForth/PygmyForth approach -- every Forth program has at least the core kernel inside it. 10:50:01 none here 10:50:21 an empty program contains just the code to exit it 10:50:39 no threaded code in the executable 10:50:48 plain machine code 10:50:53 I don't have any threaded code either. 10:51:27 I'm just saying, when I "save" a forth program as a turnkey, it saves the entire Forth environment into that ELF. 10:51:39 The Forth kernel is so small that I can afford to do that. 10:51:41 yes, you do. 10:51:42 i don't 10:51:59 i don't want a hello world program which is 2 k big 10:52:09 could use c if i wanted that :( 10:52:33 Speuler: how does the turnkey compiler know what to copy from the dictionary? 10:52:52 it just resolves what hasn't been resolved yet 10:52:56 from the lib 10:53:42 every reasonably complex word creates new fwd refs 10:53:52 all of them need to be resolved 10:54:05 that's the code compiled into the executable 10:54:15 so it is a two-pass system? 10:54:18 no 10:55:17 : foo 5 . ; ... ( lib: ) u: . 0 <# #s #> type space ; u: <# ... ; u: # ... ; .... 10:55:55 lowest level code is at the end 10:55:56 u: means "compile if unresolved" 10:56:59 Interesting -- this means your Forth compiler directly supports writing software in a top-down fashion, being READ in a top-down fashion. 10:57:10 so the interpreter just emits some filler bytes when it encounters an unresolved word and resolves them when compiling the lib? 10:57:27 that's a characteristic of fwd refs in general. not typical for my compiler 10:57:34 right 10:58:01 oldish interpreters of mine supported fwd refs already 10:58:09 was experimenting with those 10:58:30 I guess you don't care that makes redefinitions work in a totally nonintuitive way 10:58:51 it is a compiler, not an interpreter 10:59:00 you would use redefs in interpreters mainly 10:59:14 and no, i don't care 10:59:46 if i don't want a word resolved from lib, i define it in main program, thats akin to redefining it 11:00:07 only, that the code from lib won't be inlcuded 11:00:31 how is that non-intuitive ? 11:01:06 but eg : foo 5 . ; : . bar faz ; : quux 42 . ; 11:01:49 . is defined by the time the compiler gets to u: . but not when foo is compiled 11:02:15 no, . is defined there : : . bar faz ; 11:02:32 that resolves the : foo 5 . ; 11:02:47 and : quux 42 . ; won't create a fwd ref 11:02:58 after all, . has already been defined 11:03:04 --- quit: XeF4 ("Lost terminal") 11:03:47 that works well with cforth and fthcmp. 11:04:04 i don't see a reason why it shouldn't fork for my compiler 11:04:09 work 11:05:47 mmanning, there is an isforth page on the forth wiki btw 11:05:57 but it could need some polish/more details 11:06:17 --- join: XeF4 (~xef4@62.78.107.106) joined #forth 11:06:31 hmm .. somebody has already added some text to it .. 11:07:05 anyway, that means that plain colon redefinitions affect the behaviour of words defined before the redefinitions occur 11:07:19 that works well with cforth and fthcmp. 11:07:22 i don't see a reason why it shouldn't fork for my compiler 11:07:27 work 11:07:47 i used that kind of resolving from lib for about 8 years 11:08:16 there is no redefinition 11:08:38 it is either, resolving from lib, or form your own definition 11:08:44 from 11:09:20 by definiting the word yourself, you prevent the lib word to be compiled 11:09:34 as only unresolved words get compiled 11:09:44 (more or less) 11:10:08 contructs like undef foo : w1 .. ; : w2 ... ; : foo w1 w2 ; are possible 11:10:17 w1 and w1 are not fwd ref'd 11:10:23 --- join: XeF4_ (~xef4@thymine227.gprs.suomen2g.fi) joined #forth 11:10:48 rrrkele 11:11:09 Hi 11:11:23 too 11:11:43 --- quit: XeF4 (Nick collision from services.) 11:11:46 --- nick: XeF4_ -> XeF4 11:18:14 i've seperated the code generator from the opcodes 11:18:17 * XeF4 shouldn't say this in here, but has a compulsive need to do so: static compilers defeat the entire purpose of forth 11:18:43 how ? i'm still free to run an interpreter 11:18:57 and use the compiler to generate stand-alone code 11:19:11 because the program no longer has a live system it can abuse where needed 11:19:26 depends on what i compile into it 11:19:34 : main quit ; 11:19:36 :) 11:20:26 and fitting dictionaries + a tiny inner interpreter/compiler does _not_ lead to 2K hello worlds 11:20:47 : main ." hello world" ; 11:21:33 yes, and? 11:21:34 what do i need dictionaries for, when outputting a string ? 11:21:44 to "abuse" the string ?? 11:22:07 you don't, but that is quite an uncommon case where a program does nothing but output a string and leave 11:23:35 i could easily contruct a slightly more complex example which still does not need vocabularies, but does something useful 11:24:00 say, a filter 11:24:41 * Speuler has the impression XeF4 argues for the sake of arguing 11:27:28 mmanning: from the isforth docs: "any forth which can't compile its own source code is in your eyes not worth using" , and later "you need nasm and gnu make to build isforth", later "for each platform, a new kernel will be written from scratch" 11:27:55 what are you trying to say by that ? 11:29:43 * Speuler couldn't help grinning when he read that 11:32:17 --- join: XeF4_ (~xef4@anneal34.gprs.suomen2g.fi) joined #forth 11:32:44 Speuler: from time to time, but I really do think static forth compilers are useless 11:33:35 they maybe for you. but they may be useful for others 11:33:57 every thought about distributing an executable ? 11:34:55 an, you may compile a forth interpreter into the executable if you desire 11:35:22 even a slimmed-down version of one 11:35:30 I've thought about distributing an executable 11:36:05 and the basic runtime stuff can be fit into <1KB which is at least a lot less than the C runtime stubs 11:36:17 actually.. now I am arguing for the sake of arguing 11:37:14 do you have the option to remove it if desired ? no. can you compile that stuff into the executable with a compiler? yes. 11:37:36 so the compiler gives you a superset of possibilities 11:37:57 if you don't need it, just don't use it 11:38:22 but arguing "they are unnecessary" from that, documents a narrow horizon 11:39:04 which i seem to have noticed before with you 11:39:06 does it give me a superset of possibilities? It prevents free redefinitions to start with 11:39:50 no, it doesn't 11:40:24 : foo 5 . ; : . dot1 ; ( resolves ) : . dot2 ; : bar . ; ( using dot2 ) 11:41:03 or even: : . dot1 ; : foo . ; : . dot2 ; : bar . ; 11:41:55 then I suppose it's somewhat usable 11:42:02 library def applies ONLY if the word is UNRESOLVED 11:42:21 and if it is unresolved, there can't be a redefinition of the word 11:43:14 so to use the library . for awhile before other redefs one must : . . ; ? 11:43:42 no, you just leave it UNRESOLVED , like in : foo 5 . ; 11:44:14 your example WOULD work but is unnecessary 11:46:11 same way you could argue "to be able to use . , i have to define it likes this : . . ; first 11:46:47 (in a normal interpreter, i mean to say) 11:50:35 --- join: XeF4__ (~xef4@thymine116.gprs.suomen2g.fi) joined #forth 11:50:49 --- quit: XeF4 (Read error: 110 (Connection timed out)) 11:51:45 21:45 < XeF4_> so to use the library . for awhile before other redefs one must : . . ; ? 11:52:07 no, you just leave it UNRESOLVED , like in : foo 5 . ; 11:52:09 your example WOULD work but is unnecessary 11:52:14 same way you could argue "to be able to use . , i have to define it likes this : . . ; first 11:52:22 (in a normal interpreter, i mean to say) 11:52:54 ou seemed to imply that : foo 5 . ; : . dot1 ; would compile foo against dot1 11:53:36 right. foo creates the fwd ref of . , which is resolved by : . dot1 ; 11:53:56 : . dot1 ; : foo . ; would be the same 11:54:11 so how do I 11:54:13 oops 11:54:36 so how do I compile foo against the library . and make dot1 apply to subsequent definitions 11:55:10 you can't 11:55:37 this is what I was saying 11:55:46 ? 11:56:04 how do you compile a forth program normally if you don't define . ? 11:56:23 . is defined before the first reference is encountered 11:56:32 the lib is for resolving what you haven't got yet 11:56:38 i.e. there are only backward references to the lib . 11:56:54 what you do already have, does not need resolving form lib 11:57:00 well, you can 11:57:27 : foo . ; include lib : . mydot ; : bar . ; 11:57:38 yes it does, because I want to use the lib defs most of the time and my own debug-info-printing ones in some spots 11:59:18 with fthcmp/cforth i have arranged the lib so that it can be included several times 11:59:26 was necessary for tsrs 11:59:50 resolve after resident part, then resolve after the non-resident part 12:00:28 so that no open fwd refs in the tsr part were left over, with code only in the transient part 12:00:58 so first include resolved al refs in the tsr part. 12:01:14 and later, the transient remaining fwd refs were resolved too 12:01:50 result was that the tsr part contained only the code required by the tsr part. 12:02:08 and the transient parts additionally the throw-away stuff 12:06:29 *nod* 12:06:43 --- quit: XeF4_ (Read error: 110 (Connection timed out)) 12:15:09 btw, it is not a good example you gave: redefining . for testing purposes implies your still testing/debugging. 12:15:27 a compiler you tend to use after you're finished with that 12:15:54 I'm aware of that, but there are bugs that don't always behave in the same way in 2 different environments 12:17:09 those would be bugs which don't show with the interpreter, but with the compiled stand-alone program 12:17:24 (chances are that those are compiler bugs :) 12:18:23 if they just behave differently, you neglected to fix the bug in the interpreter version 12:19:04 if not compiler bugs, then bugs in the library code 12:19:24 or faulty assumptions that happen to be true in the interpreter 12:21:16 i'd probably change . to .. or whatever at those place i want a debug version of dot, instead of a redefinition (and suffer from that one through the rest of the program) 12:22:00 indeed, i try to avoid redefinitions, as those are the cause of a whole new set of problems 12:22:22 that's what you don't want, especially when debugging 12:22:37 I seldom have problems with them 12:22:49 seldom != never 12:23:23 by avoiding them, i can never have problems with them 12:23:39 true, but it's quite seldom indeed 12:23:59 often enough to not expect the redef to be the cause of the problem :) 12:25:46 actually, I have never encountered a problem using redefs in that way but I don't want to say "I never have problems" because that is asking for trouble. 13:02:40 --- join: fridge (~fridge@dsl-203-33-160-145.NSW.netspace.net.au) joined #forth 13:03:09 I just recieved an amiga 2000.... the guy who owned it previously was a forth nut, I think he has no less than 5 seperate forth environments ;) 13:07:13 nice =) 13:09:24 back 13:09:32 OOoohhh...A2000s are good work-horse machines. 13:11:11 I need to buy a scart adaptor first 13:11:16 no video! 13:11:30 but all the other signals say it is in good healthy 13:11:36 condition 13:11:59 No video? 13:12:11 Or, you just don't have an Amiga-compatible monitor? 13:12:15 * fridge nods 13:12:19 it has scart-out 13:12:21 (well, an NTSC monitor that has an Amiga video port on it) 13:12:23 I have nothing with scart-in 13:12:26 * kc5tja isn't familiar with SCART. 13:13:09 I wonder if amigas were converted to PAL when released in .au 13:13:21 kc5tja: SCART is a european AV interconnect standard 13:13:25 Amigas have always been NTSC and PAL. 13:13:55 once I get it working 13:14:05 I am taking it down into the bomb shelter 13:14:07 no internet 13:14:11 no disruptions ;) 13:15:07 Heh :) 13:15:22 No Internet. No Disruptions. Just Forth. And AmigaOS. :D 13:15:25 fridge: you can make a scart->rca cable pretty easily 13:15:26 Probably AmigaOS 1.3 13:16:07 XeF4__: can you find me a wiring diagram... I tried pushing the compsite cable onto a few cables and got some signal.... but not a picture 13:16:39 not having a multimeter at the time, it was the best I could do to test if there was actually signal coming out the cable 13:17:01 I'm surprised that a2000 has scart out though 13:17:13 I tried to search for one, but I only found people trying to sell me things 13:17:44 http://www.drdish.com/knowledge/0005/000504.shtml 13:17:58 though I don't know if there is composite in the A2000's scart port 13:19:51 If it's anything like the Amiga 500 (the 500 and 2000 are virtually the same computer), it'll likely be monochrome. 13:20:49 500 has rgb out+monocrhome, but there is are also rgb pins in a scart plug 13:21:04 there is a seperate video out 13:21:07 mono video 13:21:20 I'm saying, if it does have composite, it'll probably be just composite monochrome. 13:21:28 but that doesn't have any signal.... if its not coming through the scart, I'm in trouble =( 13:22:13 you can get a composite modulator second-hand easily enough, look for a 2nd-hand A500 13:23:15 http://tinyurl.com/f3pd 13:23:19 I was just going to get that 13:23:37 So the Amiga's all had SCART-compatible video ports? The Amiga 500 that I have has a 23-pin video adapter on the back. I'm pretty confident it's not SCART. But according to the page posted above, the SCART port has 21 pins. 13:23:40 fridge, http://www.hardwarebook.net/connector/av/amigavideo.html 13:24:02 I didn't think it was. Thanks for the confirmation. 13:24:33 fridge, http://www.hardwarebook.net/connector/bus/videoexpansionamiga.html 13:25:00 fridge: Does the Amiga have both a SCART and an Amiga video port? 13:25:08 fridge, http://www.hardwarebook.net/connector/power/amiga2000power.html 13:25:37 fridge, http://www.hardwarebook.net/connector/userinput/keyboardamiga5.html 13:25:51 fridge, http://www.hardwarebook.net/connector/userinput/amigamousejoy.html 13:26:31 fridge, http://www.hardwarebook.net/cable/av/amigascart.html 13:26:36 kc5tja: it came with a cable that is amiga video port one end, scart on the other 13:26:45 Speuler: thanks! 13:26:53 your welcome 13:27:02 fridge: Gotcha. 13:28:11 fridge: oh.. OH.. I thought you meant there was a scart port in the back of the Amiga 13:29:01 XeF4__: whoops, sorry I wasn't clear 13:29:49 :) 13:30:11 Well, I hope it works out for you. 13:30:22 fridge: then you "just" need an rgb monitor that works with 15khz sync 13:30:27 I, in the mean time, will be showering, and after that, riding on my bike. 13:30:45 * kc5tja strongly recommends a Commodore 1084 or 1084S. 13:31:29 If you can find one in working condition, those monitors are ideal for use with the Amiga. You do need a compatible cable between them though -- the 1084(s) doesn't have a 23-pin video port on the back. It ought to be relatively simple to make a cable for it though. 13:31:37 ebay can't help me 13:32:02 iirc, they have the same 9-pin connection as the 1083 13:32:32 * kc5tja isn't familiar with the 1083. 13:33:09 but if I buy a scart-> composite adaptor a normal tv should suffice? 13:33:22 not sure of the history of it,but this 1083S has the RGB input, no composite and is in a somewhat modern-looking case 13:33:27 Video quality might suffer a bit, but that ought to work. 13:33:41 oooh goody. 13:33:44 there are identations in the plastic where composite ought to be but they are not driled through 13:34:08 fridge: erhm, scart has pin assignments for both rgb and composite 13:34:19 so you need an rgb->composite converter 13:34:25 XeF4__: My 1084S also lacks composite input through RCA plugs. It does support composite in through one of its DIN ports though. 13:34:53 fridge: there are little dongles for the 23-pin amiga video port with composite output fwiw 13:34:55 I'm not using that port though -- I'm using the analog RGB input. 13:35:04 kc5tja: this has no DIN ports, just a single db9 13:35:16 Oh, my monitor has two DIN ports on the back. 13:35:17 and a toggle switch between analog/digital rgb 13:35:50 Yep. I have an analog RGB DIN port, and TTL RGB DIN port, plus said switch. I think composite signals can go into the TTL RGB port, IIRC. 13:35:54 It's been a few years since I looked. 13:36:54 Well, I'm off to take a shower. I'm feeling pretty icky. 13:37:04 * kc5tja is away: shower, then bicycling... 13:45:44 --- join: jamc (~user@as3-6-8.asp.s.bonet.se) joined #forth 13:46:57 --- join: XeF4 (~xef4@62.78.104.90) joined #forth 13:47:40 http://tinyurl.com/o5a8 14:04:03 fridge: is 2nd-hand Amiga stuff hard to find where you are? 14:04:24 yes 14:05:11 computer markets don't stock anything other than beige x86... trading post and ebay etc are the only real source 14:06:36 --- join: I440r (~I440r@sdn-ap-008txhousP0202.dialsprint.net) joined #forth 14:07:52 --- quit: XeF4__ (Read error: 110 (Connection timed out)) 14:20:03 --- join: i440R___________ (~I440r@sdn-ap-010txhousP0419.dialsprint.net) joined #forth 14:25:57 --- quit: I440r (No route to host) 14:26:16 --- quit: I440r_ (No route to host) 14:26:25 --- join: XeF4_ (~xef4@primer252.gprs.suomen2g.fi) joined #forth 14:30:24 --- quit: XeF4 (Read error: 110 (Connection timed out)) 14:48:02 --- part: jamc left #forth 14:53:52 --- part: Speuler left #forth 15:08:45 * kc5tja is back (gone 01:31:40) 15:12:56 --- join: I440r_ (~I440r@0-2pool71-21.nas3.longview1.tx.us.da.qwest.net) joined #forth 15:12:56 --- quit: I440r_ (Client Quit) 15:26:38 --- quit: i440R___________ (No route to host) 15:49:57 --- join: XeF4 (~xef4@vivo171.gprs.suomen2g.fi) joined #forth 16:13:59 --- quit: XeF4_ (Read error: 110 (Connection timed out)) 16:28:33 --- join: yasam (~sam@210-54-206-171.dialup.xtra.co.nz) joined #forth 16:34:45 --- join: arke (~rk@ca-cmrilo-docsis-cmtsj-b-36.vnnyca.adelphia.net) joined #forth 16:36:59 * arke is away: I'm busy 16:37:25 * arke is back (gone 00:00:02) 16:47:56 --- join: gilbertdeb (~gilbert@fl-nked-ubr2-c3a-37.miamfl.adelphia.net) joined #forth 16:49:25 y0 all!!!! 16:50:18 ytu[20~[21~0 16:50:19 y0 16:52:57 herm? 16:53:04 --- join: XeF4_ (~xef4@plasmid63.gprs.suomen2g.fi) joined #forth 16:53:31 --- quit: gilbertdeb ("Told you I could quit any time!") 16:56:00 can I call C functions from gforth? 16:56:08 that would be nice ..... 16:56:15 you will from rkforth 16:56:18 once its done :) 16:57:01 --- join: kc5tja_ (~kc5tja@66-91-231-74.san.rr.com) joined #forth 16:58:04 will rkforth be portable? 16:58:04 --- quit: kc5tja (Nick collision from services.) 16:58:12 --- nick: kc5tja_ -> kc5tja 16:58:17 --- mode: ChanServ set +o kc5tja 17:00:28 see details section for more information? you mean design section? 17:00:44 oops big assumption 17:04:28 arke: are you Chris Walton? 17:06:54 yasam: uum, yeS? 17:07:07 yasam: and yes again :) 17:07:24 first, though, ill have it spit out pure DOS .com binaries :) 17:07:44 yasam: heh, ignore the website, its outdated. 17:09:30 If it is to be a nice little portabl forth with the graphics, I say keep up the good work 17:10:15 heh, havent gotten far at all yet :) 17:10:36 --- quit: XeF4 (Read error: 110 (Connection timed out)) 17:21:44 hmm, perhaps Ill hook up pforth to the SDL 17:22:58 * kc5tja is writing FS/Forth, which will have a binding to SDL. 17:23:19 In fact, it's first user interface will probably be SDL, because I don't feel like putting up with the total bullcrap that is Linux console or X11. :D 17:23:43 cool 17:24:00 SDL makes coding for Linux half-way fun again. 17:24:03 and I agree 17:24:29 :) SDL is great. it needs a decent widget system tho 17:25:03 * kc5tja doesn't think so. I like SDL just the way it is. 17:25:52 There are a couple of projects I want to undertake with FS/Forth. 17:26:11 The first is a printed circuit board editor, which possibly will expand into a circuit simulator too. 17:26:35 The second is a programmable logic device programmer/simulator/HDL type system. 17:27:39 kc5tja: i mean, a library which povides simple buttons etc., paragui just isnt cutting it :) 17:28:09 Third, I want to model a Constantinesco torque converter with it. I want to build a bicycle with one installed, but machining parts are very, very expensive -- hence, I want to make sure I get things right the first (few) times instead of just "guessing." 17:30:55 --- join: junk (~junk@njd.paradise.net.nz) joined #forth 17:31:26 --- quit: junk (Client Quit) 17:31:29 And of course I have other projects too. but those come right to mind. 17:31:56 For example, one game I'd like to code up is a game of Hearts. But, I've been thinking of doing that for Chuck's ColorForth, actually. 17:31:58 --- quit: fridge ("http://lice.codehack.com") 17:32:28 I want to write some code for playing around with and visualising graphs 17:33:29 * kc5tja nods 17:33:35 --- join: XeF4 (~xef4@plasmid158.gprs.suomen2g.fi) joined #forth 17:33:46 for a research project, the members of which use various OSs so I need a portable forth with graphics 17:33:49 --- join: [Forth] (~Forth@njd.paradise.net.nz) joined #forth 17:34:01 My CTC simulator will try to render a simple wireframe model of the torque converter as it's "operating," letting the user change parameters in real-time while it is simulating. 17:34:05 [Forth]: greets 17:34:14 --- quit: [Forth] (Remote closed the connection) 17:34:19 heh 17:34:24 yasam: I fully understand. 17:34:30 arke: Potentially a bot? 17:34:44 it was 17:35:13 kc5tja: i found a reason: a new colorforth which uses real keyboard mapping and is portable to major os'es and itself (as in, can be its own OS as well) 17:35:15 njd.paradise.net.nz -- didn't see that. :) 17:35:22 njd? 17:35:30 h 17:35:30 heh 17:35:31 arke? 17:35:40 njd? 17:35:50 arke: ColorForth fits all those qualifications, except the keyboard. But I find the keyboard isn't THAT much of an issue in practice. 17:36:07 It does take some getting used to, but I find it to be rather comfortable. 17:36:16 i dont really :) 17:36:30 kc5tja: and you cant port colorforth to, lets say, ppc 17:36:39 Well, I am equally at home with Dvorak as I am with Qwerty (though I'm out of practice now), so it's not much an issue for me. 17:36:49 dvorak? 17:36:55 oh, nvm 17:36:56 arke: Portability is a myth. Rewrite it from scratch. It's not that hard. 17:37:03 njd: fellow nz forther? 17:37:10 arke: His keyboard mapping is a 27-key Dvorak mapping. 17:37:43 I agree that for my Forth, making use of the full keyboard makes more sense to me 17:38:18 But there is a great advantage to having only 27 keys -- the input task is made much simpler and the entire decision-tree for a program can be expressed as a single jump-table. That's really, really convenient. 17:38:33 * kc5tja wishes that were practical with the full 104-key keyboards. :) 17:39:44 heh 17:41:48 how can i tell xp to pretend to treat my keyboard as a dvorak? 17:43:09 kc5tja: how far have you progressed on your forth? 17:48:40 --- quit: jstahuman (Read error: 110 (Connection timed out)) 17:50:02 * arke is making dvorak stickers for his qwerty keyboard 17:50:08 --- quit: XeF4_ (Read error: 110 (Connection timed out)) 17:52:09 I'm still working on its target compiler. 17:52:20 The target compiler is written gforth. 17:52:46 But I have it compiling words and performing tail-call optimization. 17:52:53 I just need to implement numeric literals now. 17:53:42 I have a 16KB code space, and a 24KB Forth dictionary (no Compiler dictionary yet). 17:54:21 Once I'm pretty happy with things, I'm going to investigate the ELF header -- I'd like the target compiler to build a proper ELF executable for Linux. 17:55:28 Right now, it's just building a plain vanilla static image. 17:55:41 * kc5tja doesn't know much about the ELF header except the basic, basic stuff. 17:56:08 linuxassembly.org has elf info, i beliece 17:58:09 Yeah 17:58:15 I have the ELF specification on my other drive. 17:58:25 I'm going to read through it and cross-reference it with that site. 17:58:30 Great, GPL? 17:58:51 yasam: I'm thinking public domain or libpng license -- GPL is entirely too heavyweight for its own good. 18:00:42 I quite agree, PD would be my preference, if only because realistically the others are idle threats 18:00:55 from most people 18:01:43 * kc5tja nods 18:02:15 Besides, I don't like the thought of donating my hard work to RMS when he is violently opposed to Objective-C, let alone a bizarre language like Forth. 18:02:40 RMS? 18:03:03 (so opposed, in fact, that he proposes vociferiously to change the standard of the language to suit his personal political agenda -- don't use #import, use instead #ifndef/#include/#endif <-- precisely the type of thing that #import is designed *SPECIFICALLY* to avoid. 18:03:19