00:00:00 --- log: started forth/10.03.26 00:47:12 --- join: crc_ (~charlesch@71.23.210.149) joined #forth 00:50:29 --- quit: crc (Ping timeout: 258 seconds) 01:16:43 --- join: ASau` (~user@77.246.230.180) joined #forth 01:28:25 --- quit: ygrek (Ping timeout: 245 seconds) 01:42:59 --- quit: proteusguy (Ping timeout: 260 seconds) 01:56:16 --- join: proteusguy (~proteusgu@zeppelin.proteus-tech.com) joined #forth 03:00:47 --- join: DavidC99|x (~DavidC99@bas2-windsor12-1088927738.dsl.bell.ca) joined #forth 03:02:27 --- quit: DavidC99 (Ping timeout: 265 seconds) 03:05:08 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 03:08:29 --- nick: DavidC99|x -> DavidC99 03:19:57 --- quit: ygrek (Remote host closed the connection) 03:20:25 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 03:44:35 --- quit: cataska (Quit: leaving) 04:39:26 Now I have 64-bit pForth on 32-bit system. :) 05:33:08 --- join: dinya_ (~Denis@94.51.213.250) joined #forth 05:48:26 --- quit: dinya_ (Ping timeout: 248 seconds) 06:05:00 --- join: dinya_ (~Denis@94.51.213.250) joined #forth 06:06:50 I'm adding conditional capability at the "nesting" level to my processor design. So I can conditionally nest or not nest into colan definitions. Since I also added that bit for tail recursion earlier that gives me another conditional jump mechanism, which is better than the opcode level ones. 06:07:08 So my conditional jump opcodes will go away; there's no reason to keep them. 06:07:49 I use two bits for this, so I can have 0) unconditional, 1) if zero, 2) if negative, and 3) if ??? 06:07:54 Any suggestions for the ??? 06:14:14 Make it "if positive" and Fortran comes to you. :) 06:22:23 Oh, now that I look more carefully it is "if positive," not "if negative." Did you mean include them both, or did you mean "if positive instead"? 06:23:54 The C18 has opcodes for "if" (which basically becomes "jump if zero") and "-if" (which becomes "jump if positive"). That's how I wound up with those conditionals in the first place. 06:25:04 But I see what you mean now - you're referring to that FORTRAN construct that let you select among <0, =0, and >0 cases. 06:25:18 That does seem like a logically clean way to round it out. 06:26:44 I'd say that there're two reasonable candidates: 06:26:51 positive and non-zero. 06:27:00 Yes, I agree. 06:27:05 You should have some frequency data for those. 06:27:26 Not in my own archives, but I can find something I'm sure. 06:27:33 Good - thanks. 06:27:45 Without these the decision is arbitrary. 06:28:32 Yep, though I think <0, =0, >0 has a better logical structure than =0, !=0, and a choice of >0 or <0. 06:28:43 if-zero has its usage in "until", if-non-zero is "while". 06:30:11 Very true. 06:32:47 Ok - that last is good enough for me until I check frequency data. For now =0, !=0, and >0. 06:36:02 It is better to test for negative rather than positive. 06:37:59 Maybe so - I just took the C18 instruction set "at its word," and it tests for positive. I'm not wedded to that at all. 07:06:10 --- quit: Deformative (Ping timeout: 246 seconds) 08:11:02 --- join: Deformative (~joe@67-194-9-24.wireless.umnet.umich.edu) joined #forth 08:14:54 --- quit: ASau` (Quit: day is over.) 09:10:31 --- join: kar8nga (~kar8nga@jol13-1-82-66-176-74.fbx.proxad.net) joined #forth 09:56:33 --- quit: Deformative (Ping timeout: 240 seconds) 10:46:51 --- join: forther (~forther@c-98-210-250-202.hsd1.ca.comcast.net) joined #forth 10:47:35 hi all 10:50:08 Hi forther. 10:50:55 Ok ASau - I went with unconditional, =0, !=0, and <0. 11:23:00 you mean jump if? 11:26:02 Yes. 11:26:23 ?brach, "0= ?branch", and "0< ?branch" 11:28:59 ASau, are you official maintainer of netbsd gforth package or just eventual commiter? 11:29:15 The latter. 11:30:37 (There's no privacy in the Internet...) 11:30:44 anyway I appreciate someone care about gforth on netbsd. and surprised 11:31:45 there's not so many forthers around with nickname "ASau" so it was not that hard to guess 11:32:56 I rather suspect there you've found another bug in gforth. 11:33:23 nope 11:33:35 works like a charm for me 11:34:09 That's luck. 11:34:59 I don't think so ... 11:35:37 It was quite hard to convince them to release it. 11:36:04 well 0.6 was broken just before they released it 11:36:14 And I'm tired of proving that there should be two more releases already. 11:36:36 Bernd told me they commited the fix in few days, but it was too late 11:37:26 I'm fixing pforth nowadays. 11:37:50 If Phil doesn't apply my changes, there will be another fork. :( 11:37:55 pforth? is it faster, than gforth? 11:38:03 It isn't. 11:38:31 But it gives hope that you can deal with FFI insanity. 11:38:33 too bad. the speed is very critical to what I'm doing 11:38:45 Sound? 11:38:52 and I don't care about calling C at all 11:38:55 no 11:38:58 simulation 11:39:07 I run s40 simulation 11:39:18 Ah. 11:39:45 If you want to hire me, I'm seeking job. ;) 11:39:46 and 50% difference for me is I'm waiting one hour to see the results of the algorithm or 30 minutes 11:39:52 HUGE difference 11:40:13 Yes, I know what 5% is. 11:40:25 not know 11:40:44 but it's good you mention it 11:41:22 I'm not sure if you will fill comfortable programming multicores with 64 words of memory only 11:41:59 I need to get accustomed to that. 11:42:19 What kind of problems do you solve on this MPP systems? 11:42:42 On speed, in fact, I don't know what is faster. 11:43:00 It's such a subtle thing... 11:43:08 depends 11:43:38 it's very good for raw video processing 11:43:58 on embedded, where power consumtion is a key 11:44:20 it's very good in audio processing 11:44:22 I've never worked with video. 11:44:29 I worked with audio. 11:44:53 Do you mean multi-channel encoding/decoding? 11:45:06 our biggest app so far is hearing aid system 11:45:33 encode-process-decode it seems. 11:46:12 in few words it's multiband compressor 11:46:25 plus directional microphone 11:47:13 alot of calculations, consuming 55mwatts 11:47:13 If you allow telecommute temporary contract, I can work for lower wage. 11:47:35 I know 11:47:41 we use to do it 11:47:58 but right now it's transition time and we don't hire 11:48:42 My current contract is in suspense, and can "suddenly" finish. 11:49:27 Alright, you know how to find me. :) 11:49:35 (There's no privacy on the internet...) 11:50:53 BTW, do you really use NetBSD? 11:55:47 no 11:56:00 it just run on one of my boxes 11:57:01 I like the kepp it non bloated 11:57:17 it boots in 15 seconds 11:59:04 and run just few daemons 11:59:17 very minimalistic. I like it 12:22:12 --- quit: forther (Quit: Leaving) 12:35:23 --- quit: kar8nga (Remote host closed the connection) 13:23:00 --- quit: nottwo (Quit: leaving) 13:33:33 --- quit: proteusguy (Ping timeout: 240 seconds) 13:46:48 --- join: proteusguy (~proteusgu@zeppelin.proteus-tech.com) joined #forth 13:57:02 --- join: qFox (~C00K13S@5356B263.cable.casema.nl) joined #forth 13:57:53 --- join: nottwo (~trannie@designvox-gw.iserv.net) joined #forth 14:01:42 --- join: Deformative (~joe@bursley-185022.reshall.umich.edu) joined #forth 14:15:48 --- join: Guthur (~Michael@host81-156-239-195.range81-156.btcentralplus.com) joined #forth 14:38:30 --- quit: gnomon (Ping timeout: 252 seconds) 14:38:48 --- join: gnomon (~gnomon@CPE0022158a8221-CM000f9f776f96.cpe.net.cable.rogers.com) joined #forth 14:50:21 --- quit: Deformative (Ping timeout: 276 seconds) 14:51:36 --- part: X-Scale left #forth 15:02:32 --- join: forther (~forther@c-98-210-250-202.hsd1.ca.comcast.net) joined #forth 15:03:30 --- quit: qFox (Quit: Time for cookies!) 15:06:20 --- quit: ygrek (Ping timeout: 245 seconds) 15:28:59 --- join: ygrek (debian-tor@gateway/tor-sasl/ygrek) joined #forth 16:03:25 --- join: Quartus` (~Quartus`@74.198.8.60) joined #forth 16:19:24 --- quit: forther (Quit: Leaving) 16:21:20 --- quit: ygrek (Ping timeout: 245 seconds) 16:26:50 --- quit: Quartus` (Ping timeout: 246 seconds) 17:45:15 --- join: ConcreteGnome (~opera@210.11.146.14) joined #forth 17:47:07 --- join: Deformative (~joe@c-76-112-68-135.hsd1.mi.comcast.net) joined #forth 20:12:04 --- nick: madgarden -> F2Cgarden 22:18:11 --- quit: Guthur (Quit: Computer says no) 23:33:08 --- quit: segher (Quit: This computer has gone to sleep) 23:55:57 --- join: qFox (~C00K13S@5356B263.cable.casema.nl) joined #forth 23:59:59 --- log: ended forth/10.03.26