00:00:00 --- log: started forth/10.09.14 00:41:22 * ams wishes he could do: here 32 dump > file 00:41:31 where > would redirect all to file... mm... 01:15:48 --- quit: crc (Ping timeout: 240 seconds) 01:33:02 --- quit: ygrek (Ping timeout: 245 seconds) 01:55:47 * ASau` can do: here 32 ' dump s" file" with-output-file 01:56:01 i suppose that works 01:57:06 2007-10-09 25.3 : \ ' build-table s" ~/tmp/xcp-o" with-output-file 01:57:26 I'm doing it at least for 3 years already. 02:04:29 here 32 dump of course isn't very portable ;-) 02:05:53 Neither are files. 02:07:19 files are standardised 02:07:34 More or less. 02:07:38 yeah 02:07:59 "Sometimes more, sometimes less." 02:08:06 Which is annoying. 02:08:28 never had a real problem with it 02:08:43 there are many other things that hurt more 02:08:51 for me, anyway 02:09:30 That too. 02:09:44 there is only one thing in the iso standard that i refuse to implement. guess? 02:10:00 and it's in CORE btw 02:10:53 I'd say there're enough things in CORE you might wish not to implement. 02:11:39 hrm not really 02:11:52 hex output in uppercase. i refuse to do that. 02:12:01 totally unreadable 02:13:00 oh, and signed number output when base is not decimal is retarded as well, heh 02:14:00 what are your gripes? 02:15:49 I don't like the design in whole. 02:16:09 CORE itself is unstructured. 02:16:44 E.g. you have quite high-level constructs like LEAVE with convoluted semantics, and LEAVE is high-level? huh 02:18:33 and there is no Pff. 02:19:15 really :-) 02:19:36 Yes, LEAVE is quite high-level. 02:19:39 and yeah LEAVE is weird. but not evil, as such 02:19:59 Instead of DO LEAVE LOOP UNLOOP, AGAIN should be in CORE. 02:20:04 well, DO ... LOOP is high level, by that same account 02:20:06 yeah 02:21:03 i never liked DO ... LOOP , too much overhead etc -- but it's convenient to use, and in an optimising compiler it compiles to almost nothing :-) 02:21:21 I agree. 02:21:41 Still the design of ANS is really chaotic. 02:22:23 that's because it wasn't designed 02:22:47 The self-proclaimed committee is a bunch of vendors each one pushing his own libraries into the standard. 02:23:20 in the current standard? hardly. i do see a lot of that with the 200x stuff though :-( 02:23:45 I mean 200x. 02:24:10 gotcha 02:25:00 It looks like they have not a sign of sense what the design is. 02:25:30 As if neither Ada, nor Common Lisp, nor Scheme never existed. 02:25:53 the design is (or should be) that there _is_ no design :-) 02:26:38 And that's quite annoying. 02:27:46 imho not terribly much should be codified 02:27:59 that's on of the strengths of Forth 02:28:42 IMO the design should be fixed first 02:28:44 before any codification. 02:29:10 Forth is still pretty chaotic, there's too much ad-hockery. 02:29:25 i find it very coherent actually 02:31:31 we'll just have to disagree. that's fine, right :-) 02:36:39 --- quit: tgunr_ (Remote host closed the connection) 03:10:16 I wonder if I find enough typing information after fixing FICL. 03:11:42 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 03:30:20 --- join: crc (~charlesch@184.77.185.20) joined #forth 04:38:27 --- quit: crc (Ping timeout: 245 seconds) 05:12:37 --- quit: ygrek (Ping timeout: 245 seconds) 05:16:31 --- join: crc (~charlesch@184.77.185.20) joined #forth 05:44:43 --- quit: martinhex (Remote host closed the connection) 05:48:15 --- join: martinhex (~mjc@93-97-29-243.zone5.bethere.co.uk) joined #forth 06:39:26 --- quit: crc (Ping timeout: 264 seconds) 07:09:46 --- join: crc (~charlesch@184.77.185.20) joined #forth 07:15:51 --- nick: KipIngram-zzz -> KipIngram 08:56:38 --- join: kar8nga (~kar8nga@i-84.vc-graz.ac.at) joined #forth 09:14:22 --- quit: kar8nga (Remote host closed the connection) 09:50:57 --- join: qFox (~C00K13S@5356B263.cable.casema.nl) joined #forth 10:05:34 --- quit: gogonkt (Read error: Connection reset by peer) 10:38:23 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 11:00:09 --- join: alexshendi (~alexshend@178.2.210.147) joined #forth 11:47:22 200x is a huge improvement over ANS 11:47:41 here in 2010, ANS doesn't standardize nearly enough 11:56:07 What do you see as huge improvement? 11:56:20 %these% substitutions? 11:56:21 ASau, DocPlatypus, segher: speaking of 200x, is there a place where an interested but uninvolved party could read up on what's been happening on that front? I haven't been keeping track of it, but I'm very curious to know how it's going. 11:56:38 forth200x.org or something. 11:58:16 yeah that's it 11:58:52 Hmm. There's no mailing list? 11:58:53 and ASau ... a lot of it is the same. it's mostly stuff that's been added, if you don't like the new features most ANS Forth programs will still work as is 11:59:28 Which stuff is improvement? 11:59:51 Thanks very much for the pointer, guys. 12:00:06 These awful record support? 12:01:14 Or readded "forget"? 12:03:03 I can't remember any improvement brought by 200x, 12:03:16 you seem to understand nothing on that. 12:33:54 --- join: CW__ (~ClearWave@cpc5-lewi15-2-0-cust24.2-4.cable.virginmedia.com) joined #forth 12:35:14 Greetings! forth programmers 12:35:53 would you have any recommendations for a beginner learning? recommended books or resources please? 12:36:23 http://www.mpeltd.demon.co.uk/arena/ProgramForth.pdf ? 12:36:50 Anti-recomendations: use with extreme caution "Starting Forth" and "Thinking Forth" 12:36:55 or don't use them at all. 12:37:05 elaborate 12:37:20 They are obsolete. 12:37:32 They talk about obsolete dialect and obsolete techniques. 12:37:47 The world has changed much in recent 25 years. 12:37:48 yes please elaborate , and thank you kindly for the link, im lookijng at the pdf now, cheers!! ;-) 12:37:57 I see 12:38:36 No doubt now it's time for some old fart to pop up and call me idiot or anything. 12:39:48 I really like the look of this system of commands, its different to anything i've seen and looks logic and not as abstract as some of the other languages and prompts 12:39:50 Unfortunatly Forth community is mostly such old farts or old farts to be. 12:40:05 is forth only used in embedded applications generally? 12:40:15 If you like logic, you've chosen wrong language. 12:40:31 It is hard to tell where it is used nowadays. 12:40:45 well im 27 and really enthusiastic and inexperienced and want to build things... 12:41:07 About 5 years ago one of the most practical applications were nnCron and nnBackup. 12:41:16 I met their users in wild. 12:41:24 yes, I see 12:41:35 That is, I met users who weren't Forth programmers themselves. 12:42:06 soo cool, nncronn looks great! 12:43:13 --- part: alexshendi left #forth 12:43:18 im hoping theres a wealth of code already out there available to download as although I want to acquire the fundamentals I don't want to try and reinvent the wheel 12:43:47 Don't hope. 12:43:55 i want to get straight to planning and learn as i build, the proper way...i guess :-) 12:43:56 There's very little code. 12:44:28 thats another thing that appeals to me, its very minimal and concise it appears 12:44:53 It's illusion. 12:45:07 please explain 12:45:27 It's still conceptually complex. 12:45:51 You'll meet many things that are driven by tradition rather than logic. 12:46:03 yes, understood. 12:46:27 Former mock up implementations first became tradition, then they codified them. 12:47:05 Im rather surprised, forgive me but i had the impression it was mostly a logical development sequence, not full of abstract terms and lingo..? 12:47:13 Then there was huge effort in early 90s to modernise Forth, 12:47:13 and make it more suitabe for real world. 12:47:20 yes 12:47:34 Now this almost reversed and driven sideways. 12:49:29 from what i see, one could use the native unix kernel in an off the shelf Apple computer and embed ones own forth program to run in a dedicated fashion, is this presumption at all possible? 12:49:51 or am i jumping the gun? 12:49:57 why would you want to? 12:49:58 if you keep in mind that Starting Forth and Thinking Forth were written during the earlier years of Forth and don't take them too closely to heart, you're fine 12:50:20 there is a reason that Thinking Forth was released under a free documentation license not that long ago 12:50:50 Yeah. It's become obsolete and almost useless. 12:50:53 ok, in that case what would you recommend as far as an up to date reference of the language? 12:51:02 and btw ASau I have yet to see these supposed bugs in GNU Forth you mentioned 12:51:07 what is obselete? 12:51:13 "Thinking Forth" 12:51:18 ok 12:51:40 ASau: no... it's been updated and is still very useful. I don't know what you're smoking dude 12:51:45 DocPlatypus: there's "marker" bug still present in released version, 12:52:19 okay what exactly is the "marker" bug? I don't really use "marker" often, I have maybe one piece of code that uses it 12:52:19 DocPlatypus: plus it always had problems with "order" when you used "wordlist" directly rather than via "vocabulary" 12:52:40 again a feature I use rarely if at all 12:52:41 DocPlatypus: create this file: 12:52:53 marker task .( here) cr task 12:53:04 and include it 1000 times. 12:53:12 If you succede, report. 12:53:36 --- join: koisoke_ (~kwbrad@linux.utu.fi) joined #forth 12:53:43 --- quit: koisoke (Quit: leaving) 12:54:01 --- nick: koisoke_ -> koisoke 12:54:24 ASau: what's supposed to happen? 12:54:35 I have yet to have it error our 12:54:37 out* 12:55:02 Or even better. 12:55:09 GNU Forth..what does one run it on,? please excuse the noob questions im a forth learning foetus.. 12:55:22 It mentions emacs editor.. 12:55:27 Execute "marker task task" 1000 times. 12:55:27 CW__: there's a version for most major operating systems 12:55:36 cool 12:56:44 ASau: worked fine for me. is it supposed to fail? 12:56:49 this is on 0.7.0 12:57:07 Try including some file. 12:57:19 Maybe it requires interaction with include/require. 12:58:25 does anyone have some inspiring stories of real world useful implementations of forth in systems, embedded or otherwise? 12:58:26 1000 times should trigger bug? 12:58:33 Maybe more. 12:58:48 It depends on number of included files. 12:58:52 CW__: FedEx uses handheld computers programmed in Forth or at least did at one time 12:59:02 ASau: no crash here 12:59:13 i see, thats interesting 12:59:17 and you're sure this was on 0.7.0 that you tried this? 12:59:18 included-files @ marker tested tested included-files @ <> . 12:59:37 Yes, I'm sure. 12:59:47 what operating system and hardware? 12:59:56 It doesn't depend on operating system. 13:00:03 Nor on hardware. 13:00:09 It's straightforward bug. 13:00:19 You can find description and fix on comp.lang.forth 13:00:23 so you have tried this on every OS and hardware GNU Forth runs on and it crashes on every single one of them? 13:00:27 because it works fine for me here. 13:00:45 included-files @ marker tested tested included-files @ <> . 13:00:46 Ubuntu 10.04 LTS, rather unremarkable i386 PC 13:00:59 It doesn't matter, because the code is shit. 13:01:22 Do you use vendor package or build it yourself? 13:01:48 ASau: vendor package 13:01:54 that line you just pasted 13:01:55 Check vendor patches then. 13:01:58 should print 0, correct? 13:02:11 Yes, if the bug is fixed. 13:02:26 okay 13:02:27 I get 0 13:02:36 Perhaps vendor patch. 13:02:43 this is why I'm asking what OS and hardware 13:02:44 Build it yourself and check unmodified gforth. 13:02:54 basic software troubleshooting 13:03:01 or are you not familiar with that? 13:03:17 I assure you, that if I come to ubunto and install gforth myself, 13:03:18 I'll reproduce it. 13:08:42 no bug here, gforth 0.7.0 on archlinux, upstream version 13:08:51 does forth allow the execution of multiple simultaneous strings of calculations running in parallel?, is it possible to code that on anabling hardware? 13:09:13 enabling* sorry 13:09:29 What do you mean exactly? 13:09:32 Does C? 13:09:33 thank you scj 13:09:45 scj: what do you understand under "upstream"? 13:09:59 no vendor patches 13:10:05 scj: have you built it yourself or again vendor package? 13:10:44 ok i just built from the tarball at http://www.complang.tuwien.ac.at/forth/gforth/ 13:10:47 http://aur.archlinux.org/packages/gforth/gforth/PKGBUILD 13:11:20 for instance could one program an application which has several separate operations with calculations running in parallel, yet separately? 13:11:26 code to reproduce? 13:13:31 koisoke: ASau gave several. first try: marker task .( here) cr task 13:13:38 about 1000 times or so 13:14:10 then try: included-files @ marker tested tested included-files @ <> . 13:14:20 should print 0 13:16:00 Ah, the bug is in interaction with include. 13:16:09 Put that into file and do smth like 13:16:13 :noname 10000 0 do s" /tmp/test.fs" included loop ; execute 13:16:27 Yes, it should print "0". :p 13:17:02 for example if one was building a robot, and the legs and manoeuvring was a set of commands, balance operations are another, gyroscopic compensation formulas and commands running on another and so, on, all different executions but contributing to each other to achieve an overall end result with information being shared across relevant areas, is this something out of forths scope or ideally suited to it...? I need something t 13:17:02 can share data and act on it in a modular fashion... 13:19:17 across modules (applications) 13:19:18 I repeat question: does C allow calculations running in parallel? 13:19:51 I should think so 13:20:42 Tried it? 13:20:53 not at all 13:21:10 Parallel programming is quite subtle issue. 13:21:29 C doesn't allow parallel calculations. 13:21:39 It isn't supported in language. 13:22:07 But you have special support from operating system. 13:22:19 E.g. you have clone(2) and pthreads. 13:22:33 And other specially crafted libraries. 13:23:35 I see, how odd and interesting..when I heard a story about some of the techs writing forth for the moon landers it was mentioned that many calculations were running in parallel.. 13:24:04 okay, finally confirmed the bug. 13:24:19 clone(2) and pthreads fascinating. like fork(2) ? 13:24:26 i still haven't been able to reproduce it 13:24:40 and whatever it is, is really bad. because Gforth came back and said :noname was undefined 13:24:50 at least it did last time 13:24:57 4096 iterations of marker task .( here) cr task here 13:25:05 :noname 10000 0 do s" /tmp/test.fs" included loop ; execute ok 13:25:11 with an empty file in /tmp/test.fs 13:25:23 included-files @ marker tested tested included-files @ <> . 0 ok 13:25:56 oh right 13:26:00 zounds 13:27:16 er 4069 iterations of marker task .( here) cr task rather 13:29:33 koisoke: you should put that comparison into the file. 13:31:04 has anyone used fuzzy logic in their work with forth? 13:31:55 * ASau sighs. 13:32:12 CW__: this kind of questions is wrong in Forth community. 13:32:18 is it possible? 13:32:49 Is it possible in C? 13:32:50 anything is possible, but only if you're willing to do the work yourself 13:32:52 sorry, how is it inappropriate, im unaware forgive me 13:33:14 how is c related to forth ? 13:33:27 soory im really new to all this 13:33:29 Alright, take any other programming language. 13:33:40 What language do you know better? 13:33:50 Java? 13:33:52 Scheme? 13:33:54 Haskell? 13:34:03 c and forth are both programming languages, but not related 13:34:18 none, I want to start with something useful because i want to create something in particular 13:34:44 I'd say that the choice is wrong. 13:35:00 i researched and found forth to be on a superficial enquiry potentially quite suitable for the task.. 13:35:10 If you choose Forth you decide to do almost everything yourself. 13:35:14 Think about it. 13:35:22 Then think about it once more. 13:35:39 when you say do everything youself , what do you mean? 13:35:50 Are you sure that at any time you have to do almost everything yourself. 13:36:01 I mean exactly what I write: 13:36:11 you have to write almost everything yourself: 13:36:16 sure, I expect that, thats fine 13:36:41 Im prepared to take that time to make this thing i have in my mind a working reality 13:36:47 starting from such elementary libraries like string and math. 13:37:05 Almost with anything else you could save your time. 13:37:30 really? like what do you suggest? and why? 13:37:33 modern languages come with batteries included, with forth you gotta smelt the iron to make the axe to chop the wood to make the waterwheel etc etc 13:37:55 scj: don't repeat this shit. 13:38:01 scj: it's plain wrong. 13:38:17 thats why I wanted to use forth , because i could determine everything in a custom fashion from the foundations up 13:38:21 CW__: try Scheme or Common Lisp. 13:38:57 You can do that in Lisp as well. 13:39:19 Noone stops you from implementing LOOP or object system. 13:40:01 --- join: alexshendi (~alexshend@178.2.210.147) joined #forth 13:41:03 OTOH, when you need the work done fast, you just go and do it reusing someone's code. 13:41:04 ok, im checking out Lisp 13:41:10 looks interesting 13:41:24 ASau: how is that any different from what you said less than 10 lines previous? 13:42:02 It is different in everything. 13:42:16 It isn't "batteries included". 13:42:30 If you want to prove otherwise, prove it: 13:42:39 show me web client in SBCL distribution. 13:42:46 In ECL. 13:43:25 CLISP doesn't include even ASDF. 13:43:49 CL has a distinct lack of batteries 13:44:03 Sure. 13:44:16 It's the code that matters: 13:44:30 CL has it, Forth doesn't. 13:44:47 Im lost now, lol...ok im going to go away and do some research, plenty of food for thought, thank you very much for sharing! 13:44:48 If you need to come up with web site in a week, you can code it in CL, 13:44:51 but not in Forth. 13:45:28 Even when you don't have web server nor web framework in CL. 13:46:01 I wanted to create an hardware unit thta had onboard processing executed with forth, thta was robust and that noone could fuck with.. 13:46:22 you know dedicated that accepted 3 main variables 13:46:46 at least 13:47:09 on a custom FPGA 13:49:39 a custom fpga and forth implementation to cope with 3 (slowly changing?) inputs? 13:50:37 can't use a cheap microcontroller? 13:50:47 the 3 variable inputs would perform a calculation and then cease or be piped/passed on to another function 13:51:42 3 variables would be separate calculations 13:52:09 possibly a cheap micrcontroller, yes 13:52:37 --- quit: ygrek (Ping timeout: 245 seconds) 13:53:05 what environment would one use to program a micro controller with forth or similar language? 13:53:16 forth specifically? 13:53:50 can it be done over usb from another terminal? like a mac running gnu forth? 13:55:51 i haven't looked at them since 2005, but normally one compiles a program on the host system and flashes it on to the controller 13:56:15 and of course there are emulators 13:56:49 yes, I know , was wondering if you knew whch hosts are best and which interfaces are recommendable 13:56:54 maybe they exist with interactive debugging support in the hardware now, i don't know 13:57:34 CW__: not really, sorry. 13:57:51 no sweat, thanks 13:58:29 catcha later! 13:58:43 --- part: CW__ left #forth 14:09:23 --- quit: qFox (Read error: Connection reset by peer) 14:15:09 --- join: Nobody_1707 (~mdmorri1@adsl-240-132-37.msy.bellsouth.net) joined #forth 14:15:36 I found the oddest bug in PowerMops today... 14:19:59 http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTi%3DrhbBoGSnGqiD7roa1XGpkWGvP%2Bb4%3DDOVfzm%3Df%40mail.gmail.com&forum_name=powermops-users 14:20:47 Yes, you read that right. Find has been broken for months, and I'm the only one who's noticed. 14:27:11 --- quit: scj (Ping timeout: 240 seconds) 14:28:02 --- join: scj (syljo361@boneym.mtveurope.org) joined #forth 14:29:02 I think OS X users would actually be justified in writing their own Forth at this point… 14:33:06 --- quit: scj (Remote host closed the connection) 14:43:58 --- join: scj (syljo361@boneym.mtveurope.org) joined #forth 14:52:11 --- join: Nobody_1707_ (~mdmorri1@adsl-240-132-37.msy.bellsouth.net) joined #forth 14:52:11 --- quit: Nobody_1707 (Read error: Connection reset by peer) 14:52:11 --- nick: Nobody_1707_ -> Nobody_1707 15:00:58 --- part: alexshendi left #forth 15:11:25 nobody_1707: why? 15:11:55 just use gforth, heh 15:12:03 Apart from doing it for fun, I can't think of a really good reason. 15:12:05 it's not perfect, but it's wide-spread 15:12:38 Gforth can't do GUI's on OS X, and the OS X terminal sucks for interactive development. 15:12:42 In theory the latest XCode makes ObjC palatable. 15:12:52 --- nick: TreyB1 -> TreyB 15:13:19 huh what? i use terminal.app twenty hours per day, and it's almost perfect 15:13:51 and supposedly you can bridge cocoa from gforth. i never tried 15:13:57 Really? 15:14:07 I wonder if one could leverage iForth for code gen. 15:14:23 if you have an x86 mac, perhaps 15:14:26 Is there a Mac version of iForth? 15:14:40 it's not terribly could at optimised x86 code though 15:14:43 good 15:15:08 (not for modern machines, i mean) 15:15:15 That's a shame, a native x64 OS X Forth would be nice. 15:15:33 dunno if it can do amd64 at all 15:15:37 Mm... taste that libdispatchy goodness. 15:15:52 iForth generates fabulous x86 and x86_64 code last I heard. 15:16:19 that's not what i heard. but i could be wrong 15:17:10 Ok, I could have made the x86_64 part up, but the x86 code gen under WindowsXP rivals MSVC. 15:18:50 Now that LLVM supports custom calling conventions do you think LLForth will finally get a data stack? It would be a decent base to build on if it did. 15:19:00 treyb: that is hardly a recommendation! 15:19:38 I've not followed LLForth. 15:20:54 Well, it's on GITHub, so if it's possible to add it and the author doesn't want to then I could fork it. Of course, then I'd have to learn LLVM's C++ interface. shudders 15:22:10 i refuse to work on llvm code, heh 15:22:21 How does LLForth manage without a data stack now? 15:23:57 It stores a limited number of parameters in registers and refuses to let you have more parameters per word then it has registers allocated. 15:24:28 With something StrongForth-inspired you could know just exactly how far back the word references the data stack and build a stack frame appropriately. 15:25:25 Yeah, but I'm not really interested in StrongForth. 15:25:40 --- join: tathi (~josh@dsl-216-227-91-166.fairpoint.net) joined #forth 15:25:41 Close, but no cigar :-) 15:26:37 Ok, off to make dinner. G'day all. 15:27:51 Nobody_1707: er... `bl word find + . .` is *supposed* to cause a stack underflow. That's not a bug. 15:28:26 Oh, I thought it returned two cells of data, or is it word that's causing the problem? 15:28:53 If you type `bl word find` at the interpreter, bl word will parse "find" from the input stream and return the address of a string. 15:29:08 Then + tries to add that address to ... nothing ... and causes an underflow. 15:29:19 Oh... So it should be bl word no-op find? 15:29:29 yup. 15:29:52 At least the other bug is real... 15:29:59 Which other bug? 15:30:27 It returns xt 0 if it finds a user defined word. 15:30:43 That's what the first set of test cases was for. 15:31:09 Oh, I see. You didn't actually say what they *did* return, just what they should. 15:32:29 Oh, also note that the interpreter may use `word`, so it might be completely useless in the interpreter. 15:32:43 It's only guaranteed to work in a definition. 15:33:14 I just tested though and it does work in PowerMops. 15:33:19 Cool. 15:34:12 And it's returning the right value interactively now, so it's probably another facet of the buggy new optimizer in this version. 15:34:29 Ah. :( 15:45:35 Yeah, the worst part is that PowerMops can only be recompiled from the 68k version of Mops. 15:46:19 YOu'd think it would be self hosting by now, wouldn't you? 15:47:02 Is that a problem? 15:47:28 But yeah, you would. 15:49:49 Yes, because it means I have to run a virtual machine running a copy of classic MacOS in order to recompile PowerMops. I'm not even sure if I have a classic Mac emulator anymore... 15:50:25 Anyway, I sent in a revised test case: http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTimA5Svx2xMV4jGHZGAVMwiJH5AooeVZDD82xSBB%40mail.gmail.com&forum_name=powermops-users 15:51:41 Oh, I thought the PowerMacs came with built-in 68K emulation. 15:52:46 Well, I hope you get someone to fix it. 15:52:58 Can you turn the optimizer off for now? 15:56:46 I'm looking, at the very least I can rewrite [defined] as : [defined] bl word s" find" evaluate nip 0<> ; 15:58:24 Oy. :) 16:00:47 Okay, there are like 10 VALUEs that decide whether the optimizer will run, so I'll just set them all to FALSE. 16:02:05 BTW. 16:02:18 Yes? 16:02:28 TreyB, tathi, if you have time, try testing head pForth and head FICL. 16:02:37 Subversion and CVS correspondingly 16:02:58 If you can test history facility in pForth, esp. on NT, it would be nice. 16:03:15 I'm trying to push new release. 16:05:00 saper: have you checked FICL CVS HEAD? 16:10:12 Okay, I couldn't actually turn the optimizer completely off, so I evaluated find. On the bright side, my unit tests run now. 16:11:46 Now, where was I before this bug cropped up... 16:30:30 ASau: ok...pForth head builds ok here, coretest works, as do t_corex.fth and t_strings.fth. t_locals.fth and t_alloc.fth both fail; haven't looked into that yet. 16:31:38 Er...hmm..now they all seem to be working. 16:33:06 Hrm. And I don't know what exactly I did before. 16:37:16 I'm getting nonsensical errors from the C compiler when I try to build ficl. 16:37:32 e.g. ficlplatform/unix.h:25: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'ficlUnsigned8' 16:38:02 which is `typedef uint8_t ficlUnsigned8;` 16:42:19 Hrmph. My unistd.h doesn't seem to define the unsigned types with explicit bit widths. That's no good. 16:45:10 Well, other than that it seems to more or less work. I didn't test it extensively. 17:09:31 --- quit: gnomon (Ping timeout: 240 seconds) 17:13:01 Wow, pforth is slow. 17:13:27 Just ran one of my Project Euler solutions that takes about 5 seconds on gforth; took over 30 on pforth. 17:28:00 --- join: Nobody_1707_ (~mdmorri1@adsl-240-132-37.msy.bellsouth.net) joined #forth 17:28:00 --- quit: Nobody_1707 (Read error: Connection reset by peer) 17:28:00 --- nick: Nobody_1707_ -> Nobody_1707 17:33:34 --- join: gnomon (~gnomon@CPE0022158a8221-CM000f9f776f96.cpe.net.cable.rogers.com) joined #forth 17:40:03 --- part: Nobody_1707 left #forth 18:00:31 --- mode: ChanServ set +o crc 18:02:25 00000000000000000000000000000000- 18:02:29 cat 18:03:47 --- join: Nobody_1707 (~mdmorri1@adsl-240-132-37.msy.bellsouth.net) joined #forth 18:08:20 --- quit: Nobody_1707 (Ping timeout: 265 seconds) 18:34:19 --- join: Deformative (~Joseph@205-36.adsl.umnet.umich.edu) joined #forth 18:46:10 --- quit: tathi (Quit: leaving) 18:49:25 --- quit: Deformative (Ping timeout: 240 seconds) 18:49:54 --- join: Deformative (~Joseph@205-36.adsl.umnet.umich.edu) joined #forth 18:59:46 --- join: Deformati (~Joseph@205-36.adsl.umnet.umich.edu) joined #forth 19:03:26 --- quit: Deformative (Ping timeout: 264 seconds) 19:07:44 --- join: Deformative (~Joseph@205-36.adsl.umnet.umich.edu) joined #forth 19:10:58 --- quit: Deformati (Ping timeout: 272 seconds) 19:27:38 --- join: loop-hog (~jasondami@adsl-99-140-59-230.dsl.scrm01.sbcglobal.net) joined #forth 19:27:50 chat? 19:36:10 --- quit: Deformative (Read error: Connection reset by peer) 19:36:33 --- join: Deformative (~Joseph@205-36.adsl.umnet.umich.edu) joined #forth 19:36:43 chat? 19:37:09 I'm getting sleepy, getting ready to leave the office, but sure 19:37:11 for a few min 19:37:20 --- join: gogonkt (~info@113.105.204.48) joined #forth 19:37:21 sigh 19:38:19 it has been said that this bbs is pretty well chatted out 19:38:22 :/ 19:38:32 Maybe i can think of something clever to say tomorrow 19:38:35 :) 19:40:04 I used to own a pair of VR Goggles 19:40:18 They were only 8oz, but they still seemed heavy on my head 19:40:28 I found a pair that is only 2oz though 19:40:31 i want to try them out 19:42:47 I should look up the webpage and show you people 19:53:29 --- quit: Deformative (Ping timeout: 276 seconds) 19:53:31 later 19:53:35 --- quit: loop-hog () 20:43:32 --- join: Deformative (~Joseph@205-36.adsl.umnet.umich.edu) joined #forth 21:36:26 --- quit: crc (Ping timeout: 264 seconds) 21:36:58 --- join: crc (~charlesch@184.77.185.20) joined #forth 22:39:14 --- quit: Deformative (Ping timeout: 240 seconds) 23:08:19 tathi: yes pForth is slow (FICL is too), the focus is bug fixes for now. 23:08:39 --- join: dinya_ (~Denis@92.255.128.235) joined #forth 23:08:52 tathi, I didn't search the source of those warning, but the type should be defined in stdint.h 23:30:00 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 23:59:59 --- log: ended forth/10.09.14