00:00:00 --- log: started forth/03.10.04 00:02:50 No, not even that. 00:03:08 Just related sets of blocks under a single conceptual entity. Like a "module" in Oberon. 00:03:37 Also, each block would be 2KB in size, implicitly, with only the first 1KB used to read source data. The second 1KB would be implicitly assumed to hold commentary. 00:03:49 (yet another experiment) 00:05:24 doing the odd-comment-even-source thing like cf? 00:06:08 Yes. 00:06:16 My VIBE editor is optimized for that. 00:06:30 It also seems to make sense in systems which lack block files, like FS/Forth. 00:10:21 topic is 'no response from Chuck Moore yet'? 00:10:43 We had virtually everyone here asking about when Chuck Moore was going to be coming. 00:11:00 So we changed the topic because they'd ask day after day after day. :) 00:11:50 thin invited him to the channel in an attempt to try and produce an online Fireside Chat (partially in response to my stated desire for such a thing) 00:12:25 and mine :) 00:12:53 hi jeff! 00:13:03 im guessing chuck is busy :) 00:13:36 be right back, phone 00:13:53 The Fireside Chat is a tradition from FIG either at the Combined FIG Forth Day or from a FORML at Asilomar. 00:14:10 These traditionally take place in November. 00:16:06 I would think that you could get Chuc to do an online chat, check in November or December. 00:17:11 Hmm 00:17:45 That'd be nice if he could give one. 00:18:00 Speaking of which, what's he been up to lately? 00:18:06 And you, for that matter? 00:18:12 Haven't heard from either of you folks in a long while. 00:18:35 I was here last week about this time for a few hours. 00:20:16 Yeah, I noticed you left a while before I got back to the keyboard. 00:20:25 I was pretty busy that day. 00:20:33 * arke had too much trouble with his connection at the time 00:20:51 Lots of homework and studying, job searching, and the like. Plus, I was still moving some stuff out of my old apartment into my new room here at the house. 00:23:51 kc5tja trying to clear the upper byte of a number on the stack .. how would I do that with the and logic you were talking about earlier? 00:24:30 $00FF AND 00:26:02 ok thanks 00:26:15 * arke really doesnt get bitwise anything :( 00:26:39 crazy backwards hardware? 00:27:30 To this day, I continue to insist, the best way to learn how to program anything, regardless of language, is to learn how computers actually work. 00:27:59 foxchip x86 00:28:09 Name me ONE computer book in print today that actually teaches how binary numbers really work, how computers talk to devices, how memory is actually organized (memory maps of objects in a run-time environment doesn't count, though those are useful too), etc. 00:28:25 It's pathetic. 00:28:37 This is why I'm leaving the computer industry, and going to electronics kits. 00:28:44 and you dont find much on the net on it either, i guess 00:28:56 Besides higher profit margins, it seems I have more creative freedom in that market. 00:29:05 arke: Some day, I'll write a book on it all. 00:29:10 maybe i should be looking on google for "bitwise operations for dummies" 00:30:12 bitwise is a nice sound byte. cheerful and upbeat, hip and educated. 00:31:41 ok, crazy backwards hardware, nuf said 00:32:22 Long live the 6502 and the 65816!! 00:32:41 maybe Robert wrote something about the topic .. theres not many things he hasnt written something about or of or for :) 00:33:20 When a CPU at 8MHz with a lousy 8-bit data-bus can achieve 85% the performance of an overwhelmingly more complex and expensive 68000 running at 8MHz with a full-width 16-bit bus, you just know it's gotta be good. :) 00:33:21 I worked on a nice computerized pipette a few years ago. The chip was a 4-bit cmove w/ lcd contoller and 6502 instruction compatibility. 00:33:53 Writing the test software for the testjig was like playing with an old apple II. 00:34:09 What do you mean by cmove? 00:34:17 There is 'backwards' and there is 00:34:44 'BACKWARDS' as far that goes. ;-) 00:34:54 OK, I'm still really not understanding. 00:39:08 are you asking about the high-word in low address characteristic of someone's hardware or software? 00:41:53 No. Ok. Never mind. So $00FF AND was actually clear, zero out, the high byte in a 16-bit data word. OK. 00:42:16 heh, you typed it faster than me :) 00:42:38 feed forward? 00:43:29 hrm? 00:47:43 Oh, sorry, I typed 'cmove' back there.... 00:48:23 Yeah. :) 00:48:40 That's one of the things that confused me. :) 00:48:52 that was suppose to be cmos, cmos 4-bit 6502 variant, very tiny and low power, nice for inside of pen like device 00:48:55 I guess I should have been more explicit -- cmove to me is a verb. :) 00:50:55 I didn't realize that I had been the one who introduced 'cmove' into the discussion? That was before I came in, I was talking about complimentary symetry metal ozide semiconductor circuits of course. Wasn't that the context? 00:51:01 ;-) 00:52:01 oxide 00:52:32 heh finally ... found a website with some nice bit things. i think i get it now :) 00:53:17 foxchip: I forget honestly. It's 1AM. :) 00:53:43 * kc5tja cooks up some lasagna from Trader Joe's... afk for a few seconds 00:54:18 Which city in the Pacific Timezone? 00:55:04 One with a Trader Joe's... 00:55:14 University City 00:55:29 A small suburb of San Diego City, just a minute or two from UCSD. 00:55:51 Thank goodness for microwave ovens. :D 00:58:03 where would humanity be without microwaves and refrigerators? :) 00:58:35 Hehe, probably utopia. :) 00:59:01 * kc5tja is reading up on ball lightning. Fascinating stuff! 00:59:49 Ever seen it? 01:00:59 Nope. 01:01:09 Not sure if I want to get close enough to one to actually see it though. 01:01:18 That sort of thing sounds awfully dangerous. 01:01:55 I can take being close to a tornado, I can take earthquakes, aurora, and I absolutely love thunderstorms. 01:02:06 But ball lighting frightens the bloody hell out of me. 01:02:48 Containing the power of a massive amount of energy, confined to the size of a grapefruit, which can explosively discharge at any moment, as a soap bubble bursts . . . not my idea of fun. Too much of a gamble for me. 01:04:00 Nothing like a ball of plasma, five times hotter than the surface of the sun, to lighten up your day. :D 01:04:17 (especially one that can go through objects, metal or otherwise, like they weren't there) 01:08:51 Have you ever seen it? 01:12:46 yeah, but I was not in control of it. I saw a nice 'science fair experiment' with iron plasma one time, well read about it. 01:14:17 It injected iron filings into a flame, added a high voltage arc and got some warm plasma. Then by charging up and discharging a big cap they heating up the accelerated a pulse of much hotter plasma. 01:15:26 I saw real ball lightning in a big midwestern thunderstorm one time. Wierd stuff, lightning, jets and sprites and all that 01:17:15 * kc5tja nods 01:17:35 I don't know what I'd do if I were face to face with a ball of plasma. 01:17:40 I'd probably cry. 01:17:42 THen run. 01:17:43 :) 01:18:26 Rule #1 of Aikido: Harmonize with the Earth, and the universe at large. 01:18:38 Rule #2 of Aikido: Don't piss off the Earth or the universe. :D 01:19:15 As with lightning, ball lightning is probably one of those things that I'd probably like to see from a distance. A LONG distance. 01:19:53 I hold a great deal of respect for megawatts (probably even gigawatts) of power confined to the size of a basketball. 01:31:56 I noticed little aerogel capaciters for sale the catalog recently. I wonder how short of a pulse they can pull and what the effective series resistance is below 1ms? 01:32:47 Interesting. 01:33:06 I've never heard of them before. What is their capacitance range? 01:34:09 Does 50J at 2.5V sound right? 01:37:32 50 Joules in a little battery sized (lightweight) cap seems nice. Rev up the high voltage generator connect the bank of ultra-gel-capacitors 01:38:04 So, we're talking 150F? 01:41:38 It would be really fun to build digital cicuits on aerogel too. Its dialectric is about the same as air so they can build these nice silicon aerogel circuits that can contain photoelectrics, micromems, and faster thansistors than heavy silicon on something close to the weight of air and with excellent thermal insulation properties as well 01:43:39 yeah, check out the latest digikey catalog on aerogel capacitors, neat parts 01:43:47 over and out 01:47:16 --- quit: foxchip () 01:51:48 hmmm 01:51:52 * kc5tja will have to check. 01:51:58 But I think it's bedtime for me too. 01:57:39 was that "the" Fox? 01:59:47 Yes, it was. 02:00:12 that's the most stable I've ever seen him 02:00:35 I guess things are looking up.. 02:02:24 I've found most of the things he says to be quite true, but you have to see things as he sees them. 02:02:38 10x the code does mean 10x the cost of storing it, 10x the cost of debugging, etc. 02:02:42 Those DO add up. 02:02:45 BUT... 02:02:47 :)' 02:03:22 The amount of overhead involved with such code might be sufficiently small for most organizations to just absorb without too much of a hassle. 02:05:09 10x the code can mean a lot more than 10x the debugging since the coder can't hold the whole structure of the sytem in his head anymore 02:06:34 * kc5tja nods 02:06:38 BUT :-) 02:06:49 Complexity goes up with the square of the design space. 02:09:46 that's not what I take issue with, the reapeating of the same claims without substantiating them (maybe he thinks they're self-evident? why even speak then?) 02:09:54 + is. 02:12:35 In a way, though, how do you substantiate something as highly subjective as Forth code? 02:13:13 I'm not sticking up for him; I'm just considering both sides. It's true that it's nearly impossible to compare a Forth program even to another Forth program, let alone to something written in C. 02:14:26 well, he could release some of this supertiny stuff he claims to have developed 02:15:18 True. 02:15:44 Hmm...OK, I think I need to hit the sack, man. The room is starting to take on weird, neo-cubist dimensions for me. :D 02:15:47 somehow I suspect he has a huge set of old test routines and can't be bother to finish any of it 02:15:49 night 02:16:05 hehe -- laters. :) 02:16:12 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 04:00:57 --- join: mur (murr@baana-62-165-187-43.dsl.phnet.fi) joined #forth 04:01:39 morning 04:02:04 can you give me some natural written like forth snipplets (in irc)? 04:02:54 ? 04:03:04 which woudl have table-kind simplicity 04:03:11 but readability of natural texts 04:03:12 What should they do? 04:03:14 Oh. 04:03:28 all the same what they do :) 04:04:34 : run init load begin exec halted until save ; 04:05:00 That's how a virtual machine I know looks like. 04:05:02 The main loop 04:05:08 not bad, but it looks a bit clumsy in natural sense of languages 04:05:34 Of course. 04:05:38 But so does finnish! 04:06:43 >;) 04:28:46 --- join: schihei (~schihei@web.kst.fh-albsig.de) joined #forth 04:29:04 --- quit: schihei (Read error: 104 (Connection reset by peer)) 05:37:45 robert lol thats a realy cool answer you gave there :) 06:12:47 The "run" word? 06:13:08 I skimmed through 1000 lines of code to find one that is readable. :P 06:13:22 hehe 06:13:36 your answer about finnish was cool :) 06:13:48 lol 06:14:01 you need lots of intelligence to learn finnish 06:14:02 Well, it's true :P 06:14:27 you would know better than me there:P 06:30:43 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 06:37:28 nonsense, you just need to drink lots of piimä 06:37:49 eh ? 06:37:50 lol 06:37:58 makes any language sound natural ?? :) 06:38:15 piimä = soured almost-bt-not-quite buttermilk 06:38:48 yukk 06:42:50 it's quite ok and after leaving it sit for a few weeks, it's just wonderful 06:43:09 --- quit: tathi ("leaving") 06:44:03 lol your sic :P 06:44:57 * XeF4 has drunk US buttermilk, it's disgusting, this is different :-) 06:51:54 piimä = skimmed milk 06:51:55 afaik 06:52:28 skipped milk=rasvaton maito 06:52:32 er skimmed 06:52:50 oh 06:52:52 who cares 06:52:58 piimä is not good anyways 06:53:12 XeF4 do yuo eat piimä and mämmi every day? :) 06:54:21 * XeF4 too lazy to make mämmi and no shops sell it this time of year 06:55:28 it's not good even 06:56:26 --- join: yasam (~sam@210-55-45-248.dialup.xtra.co.nz) joined #forth 07:03:22 * XeF4 feels he has had this conversation before 07:05:00 heh 07:05:15 piimä straight from a shop is not good, but ~2 weeks after the expiry date it's excellent. 07:05:36 * XeF4 not sure if this is intentional 07:08:48 ugh 07:08:56 hope you dont get sick from that :) 07:24:25 piimä = filmjölk= 07:24:27 piimä = filmjölk? 07:24:55 no idea 07:24:59 surmjölk kanske 07:26:18 surmjölk joo 07:27:32 Hmm.. 07:36:07 --- join: semtex (~semtex@forthfreak.net) joined #forth 07:36:55 he was reading email recently (topic - chuck) 07:37:03 he replied to mine 07:37:15 * semtex aka Speuler 07:37:18 Hi :) 07:37:22 hi rob 07:37:54 er, were you the semtex on ircnet #coders around 1997? 07:38:01 nope 07:38:31 semtex is my 2nd nick if i'm logged in as Speuler form somewhere else 07:38:49 * semtex aka CaffeineJunkie 07:39:02 --- join: Speuler (~Speuler@mnch-d9ba58b2.pool.mediaWays.net) joined #forth 07:39:10 therefore 07:39:22 what did chuck have to say? 07:39:34 that he has no quarrel with the info 07:39:49 and that it is a good thing 07:40:06 what is a good thing? 07:40:59 ah. sent him a standard request to check pages concerning his stuff for possible copyright infringements and correctness in general 07:41:09 wiki pages 07:41:25 wasn't quite sure because of 25x and x18 info 07:41:57 25x pages ar repostedonto the wiki? 07:42:27 also added the x18 data. copied pages to server, to not have to rely on external links 07:43:17 and the schematics of a novix clone 07:43:56 did he say anything about fabricatoin plans for the 25x? 07:44:01 so i thought i better get me chuck's "ok" for that stuff 07:44:11 nope. 07:45:53 output of /whois semtex looks nice 07:46:13 Hehe, yes,. 07:46:24 No doubt which is your favourite language :) 07:47:29 i might be fan of south-east scotland as well 07:48:01 * XeF4 wonders if speuler's nick wouldn't fix problems with modern computing faster than his hostname :-) 07:48:26 i had a drink in the forth pub in forth road 07:49:00 not related to Forth language at all 07:50:42 think that was in north berwick 07:51:36 there's a village not very far from where i am now, with the name Forth 07:52:29 would like an mailbox address there :) 07:52:47 forthfreak, forth street x, nnnn Forth 07:53:52 that address would rock :) 08:07:22 --- join: ASau (~asau@158.250.48.200) joined #forth 08:07:43 Good evening! 08:07:52 Dobryj veczer! 08:07:52 privet aleksej 08:07:59 hi 08:08:08 ili eto bylo alex? 08:08:10 hmm 08:08:16 bylo eto 08:08:47 Hm. What does channel's topic mean? 08:09:25 ne snaju 08:09:30 :) 08:09:47 Well, no matter. 08:10:02 da. eto harasho 08:10:10 What do you think about Glibc? 08:10:30 I've dug into it recently. 08:11:42 I need faster float numeric processing, so I want to use FPU directly. 08:20:32 Does anyone have word set ready to deal FPU? 08:21:16 Or maybe, there is somebody interested. 08:21:33 Privet, ASau :) 08:21:43 Privet! 08:21:48 Kak dela, Robert i ASau? 08:22:02 I'm fine, mur 08:22:12 ;) Kak sazha bela. 08:22:31 ASau: I'm thinking about implementing FPU words in my Forth at some point.. for example when it's running. ;) 08:22:47 hmm 08:22:48 grrr 08:22:50 now i realise 08:22:54 Forth is suitable for latin 08:23:03 since there's one custom where you add verb as last word 08:23:12 but no one codes in latin! :( 08:24:06 Tatarcha? ;) 08:26:20 Forth is not suitable for English and German, Russian allow adding verb anywhere. 08:27:31 Though you see very strange orders in case of last word being verb. 08:28:33 It looks like: "This is ..., ... and ... . DO IT!" 08:29:57 Robert: Sharing ideas? 08:34:45 asau: i found the oldish 8087 forthstuff i was doing long time ago 08:36:03 That's interesing. 08:36:15 uses emulation if turned off 08:36:26 can be turned on/off on-the-fly 08:36:49 made it for a 8087 extension board for sirius (8086) 08:37:10 written for a natvie-code forth compiler 08:37:13 native 08:37:42 contains asm 08:37:56 I'm interested in common math functions. 08:38:16 That's why I've started digging Glibc. 08:38:53 I.e., I want math functions using FPU. 08:39:05 just elementary operations coded 08:39:59 Do you mean algebraic operations? 08:40:12 basic operation. + - * / 08:40:22 some i/o, conversion and the like 08:41:47 was Fox's visit here a one off or does he stop by occasionally? 08:42:13 he comes in often 08:42:23 Hm. 08:42:51 I need more complex things. 08:43:16 Exponents and logarithms. 08:43:20 err no - you need SIMPLER things :) 08:44:07 Do you think I like Taylor series? 08:44:07 Speuler: I'd like to know how to convert ASCII to a FP number... got code for that? 08:44:34 Robert: Reverse is more complex. 08:44:58 Yes, but not always needed. 08:45:01 d>float is part of the words 08:45:15 f>d as well 08:45:15 I440r: cheers for the info 08:45:28 Speuler: Can't you convert, say, "1.2345e6"? 08:45:37 Hm. 08:45:45 nope 08:45:51 Hm, OK. 08:46:06 Well, with just f>d and d/f it'll be simple. 08:46:22 Robert, then how would you print your calculated results in case of unknown order of magnitude? 08:46:22 But not always enough. 08:46:28 Like, if you're working with large numbers. 08:46:45 Or small ones, when high precision is needed. 08:46:58 ASau: What do you mean? 08:46:59 range <> precision 08:47:15 ;) Robert, how many zeros here are: 300000000000000000000000000000000000000000000000000000 08:47:25 At first glance? 08:47:55 No idea, that's why you use scientific notation :P 08:48:15 Got idea? 08:48:27 No, I won't even bother to count. 08:48:36 But what is your point? 08:49:02 You have binary representation in FPU and you need decimal for human readable output. 08:49:37 So you have to convert FPN to scientific or engeneering notation. 08:49:46 Yes. 08:50:17 That's more complex than input, as you know order of magnitude of your numbers. 08:50:35 ... in case of input 08:52:26 For today I have no good idea how to print number from FPU. I'm too lazy to go straight way. 08:54:50 Also, you need "lg" for this. 08:55:24 ...and "exp". :) 08:59:35 Hmm.. 08:59:45 I'll steal someone else's routines, I guess. 09:01:05 Robert, in case you achieve in stealing, share with me. ;) 09:01:40 Or join my hacking Glibc. :) 09:03:30 Hehe. 09:03:36 Of course 09:03:36 :) 09:09:20 * mur is zen of crashing uncrashing systems 09:20:38 --- quit: yasam ("dumdumdumdum") 09:30:57 --- part: Speuler left #forth 09:40:35 --- quit: I440r () 10:32:57 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 10:51:10 y0 people 10:51:13 * arke is awake again 11:13:54 ok 11:13:56 terve 11:14:24 Spokojnoj noczi! 11:14:37 --- part: ASau left #forth 11:51:53 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth 11:51:53 --- mode: ChanServ set +o kc5tja 12:14:57 mur terve\ 12:15:04 moi 12:15:09 whats up? 12:15:15 going to the figuk meeting? 12:15:21 shuold go to sauna :) 12:15:24 Oh yeah, I forgot about that. 12:16:12 us.ircnet.org / #figuk 12:16:13 :) 12:16:22 --- join: I440r (I440r@12-147.lctv-a5.cablelynx.com) joined #forth 12:16:27 Hey 12:17:00 hi 12:17:26 I440r y0 12:17:51 This place is cooler than the other place; we have lurkers here. :) 12:18:51 ?? 12:19:42 tihs is a tset. hoeewvr, you usdrteannd me? 12:20:26 mur: Test of what? (For such a simple expression, I udnerstand you) 12:21:29 i lavee tihs btearainser for you, bcak laetr 12:22:27 He's obviously having fun with a Linux text filter. 12:22:31 :P 12:25:20 no 12:25:28 scientifical tset 12:30:21 hardly. 12:34:09 --- join: LOOP-HOG (TofuMonste@207.191.240.247) joined #forth 12:36:31 --- quit: LOOP-HOG (Client Quit) 13:30:53 --- quit: semtex (Remote closed the connection) 14:10:49 --- join: crc (~crc@AC93E1D7.ipt.aol.com) joined #forth 14:11:37 kc5tja i want to emit either '0' or '1' based on a condition .. any way to do so without IF .. i.e. using the logic thing you talked about yesterday? 14:11:57 A different way of thinking about things, for sure. 14:12:21 remember that a logic true is represented as -1, and false as 0. 14:12:35 my condition returns 0 or non-zero 14:12:40 :D 14:12:40 Not an issue. 14:12:41 --- quit: crc (Client Quit) 14:12:48 good :) 14:13:01 Use 0= to invert the results, so for 0 it returns -1, and for non-zero, it returns 0. 14:13:11 Here's how I'd do it: 14:13:30 CREATE charmap '1 c, '0 c, 14:13:46 : result 0= charmap + c@ emit ; 14:13:51 err, oops. 14:13:59 : result 0= charmap 1+ + c@ emit ; 14:14:23 Excercise for the reader: Tell me why that works. :) 14:14:50 0.0 14:15:13 oh 14:15:15 haha 14:15:17 i get it :) 14:15:33 thats nice 14:15:38 thats _very_ nice 14:15:39 :) 14:15:57 another question: can i define words within words? 14:16:18 Nope. 14:16:23 OK 14:17:13 The word ":" is an interpreted word (e.g., it's not a compiler word), so embedding ":" in a word will start a new definition when said word is executed, which is patently not what you want in most cases (but sometimes, actually, it comes in handy!). 14:20:27 when would it be handy? 14:22:15 If you're defining a defining word that defines code, for example. Not exactly the simplest thing to think about in the abstract. :D 14:23:41 * arke 's head gets woozy 14:23:44 lol 14:24:11 I gives counter inside FOR .. NEXT right? 14:24:35 Yes. 14:24:46 Err....actually....no. 14:24:54 I gives the counter for DO/LOOP. 14:25:06 Try it - the worst that can happen is it can crash. :) 14:25:07 what about for next then? 14:25:24 Make sure you FLUSH before you run your code, so that everything is saved to disk. 14:26:03 FOR, IIRC, loads the value into the CX register, and NEXT just decrements that register, and if non-zero, jumps back to the FOR point. 14:26:08 I don't know. 14:26:13 It's been too long for me to remember. 14:26:18 I never use FOR/NEXT anyway in my code. 14:26:32 pygmy doesnt have do loop afaik 14:26:34 I've always used DO/LOOP, and even then, VERY rarely. 14:27:02 Most of my loops are done with BEGIN WHILE REPEAT. 14:27:21 YAY! I works! ) 14:27:27 cool. :) 14:27:32 Thanks for the refresher. 14:29:05 :) 14:29:13 does it backwards tho 14:29:30 n I - fixes that :) 14:34:22 Yes, because it's much easier to implement a count-down loop than it is a count-up loop. 14:34:55 NEXT can be coded as a single x86 'loop' instruction 14:35:09 :) 14:35:16 Yep 14:35:23 If you didn't care much about speed on modern processors, that is. 14:35:39 Now-a-days, it's faster to write out all the explicit steps and let the CPU's pipeline deal with it. 14:35:59 wow 14:37:28 This is the whole purpose behind RISC. 14:37:31 And ultimately, behind MISC. 14:43:32 --- join: TheBlueWizard (TheBlueWiz@pc75dn1d.ppp.fcc.net) joined #forth 14:43:32 --- mode: ChanServ set +o TheBlueWizard 14:43:37 hiya all 14:44:20 TheBlueWizard gonna stay more than 2 minutes today? 14:44:21 :P 14:44:40 hiya arke...I suppose yes :) 14:44:47 :) 14:45:07 why? 14:46:13 tbtbtbtbtbttbtbtbwtbtwbtbwtwbtwww 14:46:21 howdy 14:46:24 iltaa sininen velho 14:47:06 hiya mur!!! (lol re: your greeting) 14:47:11 hiya kc5tja 14:47:33 * mur ajattelee tänään paljon 14:48:12