00:00:00 --- log: started forth/19.11.30 00:05:39 Yeah it was expendy 00:05:52 Mostly because it was so large 00:06:00 how large ? 00:06:11 And yeah, I just shoved a DC barrel jack on it, wall wart problem 00:06:31 its like, I dunno.. 308 odd mm.. or something 00:06:48 I guess it doesnt have a clock case or enclosure ? 00:06:54 Nope 00:07:15 yeah thats pretty big 00:07:37 Its going to have a holden logo mounted on the top, with the hours leds going through holes of the logo, and the seconds/outer leds being exposed and not covered by the log 00:07:38 logo 00:07:48 Id have hand wired it onto threeply :) 00:07:58 or plastic 00:08:31 but them I'm a tech and Ive made so much gear by hand over the last 47 years 00:09:12 Im currently experimenting with making copper rivets for protype work 00:09:28 I am totally over making more than I need to myself 00:09:50 kiwis ... no staying power! 00:10:08 Once I get my man cave 5000 operational I can dick about with home made pcb's and different shit 00:10:15 tp, lol phhhhhht 00:10:22 I'm totally over depending on *anything* from China 00:10:35 hahah, Man Cave 5000! 00:11:23 gees I'm annoyed 00:11:49 for the first time since 2014, I'm actually stuck on a Forth program bug 00:12:07 it's been days and Ive got nowhere 00:12:18 tp, all right step aside 00:12:29 :P 00:12:45 hahah, it's true, Im a tech not a real programmer 00:12:56 but I do make great copper rivets! 00:13:04 got the bug code somewhere ? 00:13:15 it's too complex dude 00:13:35 it's my own piece of hell to master 00:14:08 and it's slow to test incolving code on the mcu interacting with code on the pc 00:14:13 involving 00:15:04 thanks anyway, besides if you solved it, Id have to kill myself or my ausie mates would when they heard a kiwi was smarter than a ausie! 00:16:25 whats more, I suspect the bug is looking at me and screaming "im here ya wanker" and I cant see it yet 00:16:31 lol 00:17:01 :) 00:17:22 Its annoying how obvious some bugs are after you find them 00:17:44 to add insult to injury, my code is fairly pretty, it SHOULD be fine, it passes my tests 00:17:57 only UGLY Forth code fails 00:18:41 yeah, plus it's partly because Ive added to code written by someone else and I havent refactored their code to suit me 00:20:09 jsoft, I'll happily show the code on my doc site when it's working :) 00:20:50 jsoft, you may even use it one day if you stay with Mecrisp-Stellaris and cortex-m 00:21:53 it's just a binary extractor. It obtains the dictionary in a working project, straight from the mcu, and turns it into a bootable binary file on the pc 00:21:57 --- quit: dddddd (Remote host closed the connection) 00:22:37 that way one can just flash the binary into a virgin mcu and will boot at the exact state of your breadboarded project 00:23:42 I've been using it for several months but Ive just added iHex 04 type extended addressing up to 1mB of flash, and thats my problem area 00:25:13 it's working to 1mB of flash right now using my OLD nasty, ugly brute forced table but I wanted a nice neat algorithm that would generate iHex up to 2 gB of flash 00:25:33 and thats where i ran into problems 00:25:44 Why don't you just fix it ? 00:25:45 :p 00:25:57 dang! why didnt I think of that! 00:26:13 I've been trying the last 2 days, hence the reason Im annoyed 00:26:32 that dang bug is very hard to catch 00:27:14 my algo works fine, I can generate extended addresses all the way to 2gB using a address generator 00:27:33 as I said, it's a stupid bug .. bugs happen! 00:34:02 --- join: rdrop-exit joined #forth 00:37:44 tp, another reason to move to forth for arm: https://www.youtube.com/watch?v=7MIWFHsF7Jw 00:37:56 Look at that hideous amount of farting around to get the environment set up 00:38:50 I know what you mean 00:39:40 all that eclipse bloatware with massive C libraries was too much for me 00:39:48 c[] zzz hey Forthians 00:39:57 heyhey Zen Forth Guru! 00:40:36 I was watching the new Forth Day videos on youtube and dozed off 00:40:58 rdrop-exit, tried out your example code the other day, works good :) 00:40:58 jsoft, I tested attilic 'truestudio' and was impressed but the dual celeron intel NUC thats my quick spare ran like a slug 00:41:13 rdrop-exit, that exciting huh ? 00:41:23 glad to hear it jsoft! :) 00:42:29 rdrop-exit, had to use c! instead of b! 00:42:36 I think it was my lunch more than the content that caused me to doze off 00:42:51 rdrop-exit, too many carbs? 00:43:04 rdrop-exit, too much bread or potatoe ? 00:43:14 jsoft, ya I mentioned that you probably would need to use c! instead of b! on your Forth 00:43:35 rdrop-exit, I do indeed remember that 00:43:45 Spinach and cheese ravioli with pesto sauce 00:43:56 rdrop-exit, there ya go, carbs for africa 00:44:06 rdrop-exit, we dont even have c, on the cortex-m0 00:44:28 rdrop-exit, we have to emulate it 00:45:36 I wouldn't expect a chip to provide c, 00:46:16 I mean a CPU to provide c, it's usually a high level definition 00:46:51 I have b, 16, 32, and plain old , 00:47:16 we have the same kind of data widths 00:47:34 only the names have been changed to protect the guilty 00:47:48 : b, ( x -- ) dp b+, ;inline 00:48:18 : 16, ( x -- ) dp 16+, ;inline 00:48:18 : 32, ( x -- ) dp 32+, ;inline 00:48:18 : , ( x -- ) dp +, ;inline 00:50:51 thats too easy! 00:51:34 oh - oh ! massive lightning storm heading this way 00:51:49 ill probably drop out soon 00:52:00 stay safe 00:52:42 hehe, I'll try 00:52:53 it's the damaging winds Im not fond of 00:53:05 but we may get some rain! yehaa! 00:53:45 I factored them so that I could easily make comma-ing words for memory buffers not just the dictionary pointer 00:56:35 rain has been light and infrequent the last few days here 00:57:33 i recall you saying 00:58:45 when it pours it really pours here, some streets start to flood within minutes 00:59:05 it's similar here in summer 00:59:14 being semi tropical 00:59:28 but not as bad as where you are 00:59:59 the current Forth Day speaker works at a company that makes robots for space work 01:00:09 cool job 01:00:10 nice 01:00:19 I hope so 01:01:21 I guess space and Forth are a good mix 01:01:24 a recycler for 3d printing material for a space station 01:01:56 people cant just stick in a Arduino and select the "mars landing library for spacecraft" 01:02:24 wow, rain is pounding atm, first time in at least 3 months I can recall 01:02:49 they have to remember to enable Javascript first 01:03:02 everything in space seems to be new and original, perfect for Forth 01:03:04 hahahah 01:03:23 not to mention radiation hardened 01:03:46 thi Ive heard that Microchip has made some rad hardened PIC's in recent years 01:04:05 but they wont run Forth I'm fairly sure 01:05:44 FPGAs are a good fit for space I imagine 01:06:05 because of the reconfigurability 01:06:13 youd think so, but space doesnt care about the latest tech 01:06:28 it only cares about 100% reliability 01:06:59 losing a 4 billion spacecraft because some new hitech FPGA failed is their biggest worry 01:07:20 I have a video in my to-watch queue related to space computing... 01:07:29 thats why Forth just keeps being used, it has a track record of reliability 01:08:07 searching for it... 01:08:29 just getting certification for space use would be very hard now 01:09:06 there were SEVEN Forth controlled experiments on the Philae Lander in 2011 01:09:29 all ran the rad hardened Forth Cpus 01:10:20 "Uptime 15,364 days - The Computers of Voyager" by Aaron Cummings 01:10:31 https://www.youtube.com/watch?v=H62hZJVqs2o 01:10:47 The Voyager 1 and Voyager 2 space probes, both launched in 1977, each had a primary objective to explore Jupiter and Saturn. This goal was achieved by 1981. Yet Voyager, NASA's longest running mission, has continued to this day. Both Voyager probes are still operating, and returning scientific data from outside our solar system. 01:11:07 Haven't watched it yet, but sounds interesting 01:11:26 it does, wonder if it will have any real tech info 01:11:54 me too 01:13:57 i get sick of the usual recycled superficial tech palava 01:14:58 it's the bane of you being a Master Technician (tm) 01:15:19 26,87 C / 80,48 F the temp has dropped a few degrees already :) 01:15:36 yeah, and 47 years of industry experience 01:15:40 no space tho 01:15:56 only 28C here 01:15:58 (this being Australia ) 01:16:08 Hmm 01:16:26 why do you use a comma for decimal points 01:16:57 New Zealand has some space industry, and theyre planning to upgrade their propulsion from gunpowder to LOX in the next decade 01:17:29 jsoft, you use commas for 'fixed point' also :) 01:17:50 oh right, yeah in forth you mean 01:17:58 I've not actually had to do fixed point stuff yet 01:18:21 jsoft, try this " 9,12 7,13 f* f. " 01:18:39 thats 9.12 * 7.13 01:19:20 Mecrisp-Stellaris uses the European standard for fixed point, which is a comma instead of a decimal point 01:19:56 me, I think it's silly, but I'm a tech, I don't get a say 01:29:52 silly tech, Trix are for kids 01:30:45 (paraphrasing old US TV commercial) 01:40:06 --- quit: Lambdajack (Remote host closed the connection) 01:43:22 i never saw that, being in Australia 01:46:30 https://www.youtube.com/watch?v=Z3ApTwWRQNw 01:48:09 they did a ton of them, all about a silly tech who wants his decimal points 01:48:46 hahah 01:50:42 funny the things that come back to mind from childhood 01:51:29 reliability + redundancy + reconfigurability 01:51:33 yeah, the things we remember 01:51:55 (watching that Voyager video) 01:52:58 that trix is a old video, how CGI has improved since those days 01:53:51 it wasn't CGI back then 01:54:04 I'm happy to say, no power loss, no close lightning strikes out the the several hundred in this area, rain and lower temps! 01:54:12 no thats true 01:54:28 enjoy 01:54:40 it's not a lot of rain and the lightning has probably started a LOT of new fires 01:54:43 --- join: dave0 joined #forth 01:54:59 but at least it's *some* rain 01:55:21 in some areas west of me, they have no water at all 01:55:29 so this will at least bring some 01:56:50 speaking of pre-CGI, this is from 1945 (!) 01:56:56 https://www.youtube.com/watch?v=2msq6H2HI-Y 01:58:00 yeah I remember those days ... no nostalgia for me tho 01:59:00 must have been a ton of work 01:59:34 it must have been, but cartoonists were probably all set up for it back then 02:02:43 --- quit: smokeink (Read error: Connection reset by peer) 02:02:58 --- join: smokeink joined #forth 02:13:29 lol something interesting happened yesterday, a dude told me the numbers to a safe, ie, 1 2 3 4 he said. I somehow accidentally read it back to him has 4 3 2 1 02:13:38 well not read it back, but said it back 02:13:49 I wonder if thats some random forth thing 02:13:57 hahah 02:14:24 severe stackbrainitus 02:14:55 Yeah that is odd eh 03:10:34 --- join: DKordic joined #forth 03:32:58 --- quit: iyzsong (Quit: ZNC 1.7.1 - https://znc.in) 05:16:06 --- quit: smokeink (Ping timeout: 265 seconds) 05:40:10 --- quit: dave0 (Quit: dave's not here) 05:58:21 --- quit: proteus-guy (Ping timeout: 268 seconds) 06:25:13 --- quit: ryke (Ping timeout: 265 seconds) 06:31:43 --- join: Jookia joined #forth 06:54:25 * DKordic oO(when JavaScript hits the Browser.) 07:06:07 --- quit: jn__ (Ping timeout: 246 seconds) 07:16:33 --- join: jn__ joined #forth 07:40:28 --- join: dddddd joined #forth 08:06:52 --- join: proteus-guy joined #forth 09:06:32 --- quit: rdrop-exit (Quit: Lost terminal) 09:26:01 hey guys 09:27:25 ∀Hi 09:44:20 --- join: ryke joined #forth 11:31:51 --- join: f-a joined #forth 12:00:22 --- quit: ryke (Ping timeout: 276 seconds) 12:42:53 --- join: inode joined #forth 12:53:47 --- join: WickedShell joined #forth 13:05:41 --- quit: tabemann (Ping timeout: 245 seconds) 13:22:51 --- quit: inode (Quit: ) 13:22:55 --- quit: gravicappa (Ping timeout: 276 seconds) 13:35:44 --- join: tabemann joined #forth 14:03:51 --- join: ryke joined #forth 14:23:09 jsoft, I found my bug :) 14:23:23 hi tab and DKordic 14:26:57 what was it, tp 14:27:54 f-a, whats more, I found it by accident 14:29:27 it was simply this. The part of the program written by Matthias put the BASE into hex, and my Word which was called from Matthias section used the decimal base .... 14:30:40 f-a, while refactoring I decided to make everything use base decimal, and suddenly it worked properly! 14:30:55 this one really caught me out 14:32:33 f-a, my 'new and better code' is actually longer than the 'table' driven solution I had been using for the last few months but my latest rewrite can handle generating iHex to it's 32 bit maximum, whereas the old code only went to 1 mB of Ihex 14:33:56 f-a, to complicate matters, my new code always worked perfectly in my separate tests which was confusing, bit of course in the separate tests it was all base decimal ! 14:34:58 glad to hear you came up victor 14:35:06 bugfixing is always a pleasure when it ends well :P 14:36:42 it was the fist bug that I have been truly stuck on since 2014 when I started learning Forth 14:37:05 normally I can work out the bugs easily enough 14:37:52 oddly I hadnt been happy with the original program switching to base hex on entry and restoring the base on exit 14:38:21 but I didnt see it as a issue or a reason for my bug 14:43:16 back 14:43:21 hey guys 14:43:32 hey tabemann ! 14:43:41 tabemann, I found my bug :) 14:44:07 cool 14:44:45 tp: do you have a testsuite? 14:44:51 rather, what do you use to run tests 14:46:48 * tabemann doesn't write tests with hashforth 14:47:11 f-a, in this case all I needed was to generate addresses from 0x0 to 0xFFFFFFFF 14:47:51 f-a, and my program would generate iHex extended addresses at each 64kB boundary 14:49:13 f-a, the problem occurred after my program inserted the first extended address into the code that generated the iHex dump 14:49:27 so why do all thia ? 14:49:29 this 14:50:06 the original program had a max ihex dump of 64kB and I needed 128kB 14:50:33 but iHex without "extended addressing" has a max of 64kB 14:50:43 yep 14:51:15 so I had to add "extended addressing", which now caters for a full 32 bits should I ever need to clone a MCU with 2GB flash ;-) 14:51:22 I don't know why mecrisp-stellaris, when you first build it, doesn't automatically have ihex files generated for it 14:51:37 --- quit: ryke (Ping timeout: 246 seconds) 14:51:43 it does iirc 14:51:55 it's distributed with ihex files 14:52:18 but when you build it yourself it doesn't generate them 14:52:22 ahh 14:52:38 italics so early in the morning ! 14:52:48 * tp reaches for his shades 14:52:57 I think I'm going to stick to mecrisp-stellaris until I get the hang of embedded 14:53:23 you couldnt do much better than Mecrisp-Stellaris in my experience 14:53:34 yeah 14:53:52 almost makes me wonder why do a half-assed job at porting another forth to cortex-m4 14:53:58 if it could be said to have any weaknesses, perhaps one is that Mecrisp-Stellaris does everything for you 14:54:35 Mecrisp-Stellaris is the Forth equivalent of Arduino ;-) 14:54:40 lol 14:55:11 tabemann, there are very few cortex-m forths 14:55:17 I only know of 2 14:55:21 why? 14:55:27 it's hard ? 14:55:48 2 forths that actually work and are usable I mean 14:56:13 I assume the other is riscy pygness? 14:56:42 the otehr is coreforth, but it's abandoned by Mitch Bradely now 14:57:03 no, riscy is buggy and incomplete, and abandoned 14:57:47 dammit, I probably won't port hashforth to cortex-m4... rather I will write an SRT/NCI forth, and write a converter for hashforth VM code 14:57:48 riscy is pretty cool, coreforth is very complete, I like it a lot but I cant compile it 14:58:24 because hashforth has design features that make it hard to implement on something like m4 14:58:37 yet a M4 is the biggest of the big 14:58:39 namely its token table 14:59:01 M4's have up to 1MB of flash 14:59:10 I could port hashforth to M4... but it'd have to mostly reside in RAM 14:59:15 it's why the micropython people use it 14:59:39 and that wouldn't be practical 14:59:52 yeah 15:00:25 like I figured out that my board has 96K of RAM not 128K of RAM 15:00:34 good decision tho, youve saved yourself months of work that wouldnt really be that useful in the end 15:00:41 so all of the VM code would have to reside in 96K of RAM 15:01:12 youd have to rename F4 hashforth as well 15:01:19 to "FrankenForth" 15:01:30 lol 15:02:00 also the performance impact would be more acutely felt on M4 than on x86-64 15:02:54 yeah, a watercooled 8 core i7 running at 5.6GHz makes everything feel fast 15:04:22 not quite that here (more like 4 core aircooled i7 running up to 3 GHz but preferably at more like 1 GHz) 15:06:09 --- quit: f-a (Quit: leaving) 15:06:39 not that here either, but a guy I know has a 12 core i7 'sandy bridge' he o/c's to 5.6GHz which is insane 15:06:57 he has a massive custom watercooler for it 15:07:46 he said if I tried to do the same to my other machine (same CPU as his) Id blow the tubes of my Intel watercooler in about 30 seconds! 15:08:13 I'm not into overclocking anyway 15:08:52 except on cortex-m0 where a stm32f051 rated at 48MHz easily overclocks to double that speed 15:09:54 a 6x increase in speed (over the std Mecrisp-Stellaris 8MHz) is nothing to sneeze at 15:10:19 oh, are you using the std clock on your Mecrisp-Stellaris board ? 15:13:26 back 15:13:33 tabemann, if youre not running a clock increase Word, then the board is running at 8MHz from it's internal RC clock 15:13:55 which is a LOT slower than the usual 160MHz M4 15:14:02 I haven't tried increasing the clock on my mecrisp-stellaris board 15:14:11 lol, you can go 20x faster 15:17:06 the docs say 80 MHz 15:18:07 which is still 10x faster 15:18:36 bbs 15:19:16 oh, bit of a slug eh ? 15:19:24 ;-) 15:19:43 I'm pretty sure they OC to at least 160 MHz 15:19:45 tp, nice! does anything crap out? 15:20:01 I heard the 50mhz pic32 can do 72mhz for example but the timers stop working 15:20:09 MrMobius, yeah, I cant write to Flash at 98 Mhz 15:20:10 right 15:20:24 er 96 MHz 15:20:41 well I can but it's flaky. however I can read from flash just fine 15:21:00 aread write ram just fine 15:21:23 MrMobius, I ran one for 12 months at 96 MHz and Im using it still 15:21:36 including local temp of up to 46C 15:26:25 --- quit: WickedShell (Remote host closed the connection) 15:28:01 back 15:30:35 see, I wouldn't want to not be able to write to flash 15:32:19 it's still handy to know if you just want to run some Forth program at double the speed 15:32:46 I used it to cut my primes benchmark by exactly half 15:33:28 and as I normally test all my new Words from Ram, it made no difference 15:34:29 basically a Forth that cant run Words from embedded ram is useless to me 15:35:01 tabemann, you should try coreforth, it has a TON of awesome features cor cortex-m 15:35:20 tabemann, and it's doc is english, has a large library 15:35:49 written by the guy behind the old OLPC project 15:35:50 I was annoyed that tthe demo.fth file that came with Mecrisp-Stellaris included a cornerstone word but did not use it, so to remove it I had to flash the whole thing, compile to flash just the cornerstone onto the board, then flash the demo code onto the board 15:36:41 why remove it ? 15:37:04 because I wanted to create a cornerstone that erased everything 15:37:15 including mecrisp ? 15:38:23 everything but mecrisp 15:38:28 aha 15:39:20 why not just remove the cornerstone word from the demo.fth file before flashing it ? 15:40:15 I had a similar issue recently when loading a prebuilt Mecrisp-Stellaris with dissasembler 15:40:54 because I didn't look at the code first 15:41:05 it also had the 'modified' cornerstone that wouldnt let me remove the dissasembler after 15:41:56 so I just used the default image and added the dissasembler manually after 15:42:32 I havent really taken to the modified cornerstone either 15:42:50 well it's not that it had the cornerstone word which created a new cornerstone but that it didn't run it 15:43:26 plus any new cornerstone you write will be used instead or the previously flashed one 15:44:05 with Forth there are always seven ways till Sunday to do everything 15:44:24 I looked at coreforth and it definitely seemed abandoned 15:45:59 it's also indirect threaded... which may be a plus or may be a minus 15:46:12 a plus is that it'd be more memory-efficient 15:46:22 well waitq4 15:46:23 no 15:46:31 the doc is massive and well written it seemd 15:46:33 it wouldn't be 15:47:31 one idea would be to ask the author about its status, and if they're not planning on developing it, going and forking it 15:47:55 only if one is a programmer ;-) 15:48:04 and I am one 15:48:09 exactly! 15:48:56 a MASSIVE amount of work went into Coreforts Id think, shame to waste it ? 15:49:14 so why was it abandoned? 15:49:28 Im already using Mitch's USB code that was modified for use on a F103 15:49:32 i dunno 15:49:39 hes not young anymore I guess 15:49:47 the OLPC project died 15:50:10 maybe he's in love with Haskell or Python now ;-) 15:50:24 or Rust 15:51:17 I was in love with Haskell lol, and still think of it as a neat language 15:52:15 seems youre not alone 15:52:50 part of me still wants to write my own Forth for Cortex-M, and is only interested in using another Forth for the purposes of familiarizing myself with the target platform 15:53:30 why not ? Im sure most forthers go that way 15:54:12 after all, much of what is attractive to me about Forth is the ability to create a Forth of my own, unlike many other languages, where building your own language implementation is much, much harder 15:54:23 absolutely 15:54:49 I'd think that many forths exist because of that one attribute 15:55:44 I'll probably end up dissecting mecrisp-stellaris to figure out how it does everything 15:56:56 id prefer that myself because I'd learn from your comments here 15:57:05 lol 15:57:18 okay, I've gotta go - dinner time - I'll bbl 15:57:32 oddly the parts of Mecrisp-Stellaris that I wanted to understand proved to be really easy 15:57:35 cya! 16:02:49 --- quit: tabemann (Ping timeout: 276 seconds) 17:25:19 --- join: tabemann joined #forth 18:26:05 --- join: smokeink joined #forth 18:43:02 I'm annoyed 18:43:13 me too 18:43:17 I want to license my code as BSD3 18:43:48 but there is so much good stuff in Matthias's code that I'd have to write from scratch because his code is GPL3 18:44:03 and then I'd have to license my code as GPL3 too 18:44:16 I stuffed the commit of my latest working code somehow and now have a new bug 18:44:33 but I don't want to license my code as GPL3 because I want as many people as possible to be able to use it 18:44:44 all 42 Forth users ? 18:44:52 lol 18:46:39 I wonder if it's started snowing yet 18:47:15 yes, this is a problem you don't have in Australia 18:47:30 you guys just have to worry about A) fires and B) everything trying to kill you 18:48:30 B we get used to 18:48:47 A is always a drama 18:49:20 I read that 600 homes have been bunt to the ground in NSW in the last 3 months 18:49:33 6 people burnt to death 18:49:46 I assume insurance doesn't cover your house getting burnt down by wildfires 18:50:12 yeah it does dunno about the premiums tho 18:50:41 it's impossible to get flood insurance in certain flood prone areas tho 18:50:52 okay, I'm gonna head out - will be back on in a bit 18:51:02 no problemo 18:51:11 oh and I'm starting work on a new forth - zeptoforth 18:51:26 for cortex-M 18:51:40 written in assembly 18:52:16 wow 18:52:20 zeptoforth because the name's not taken, and because yocto is already being used as some large project's name 18:52:31 tabforth ? 18:52:36 and not attoforth because I already used that name 18:53:00 for my first attempt at a forth 18:53:06 ah yeah 18:53:13 clever 18:53:36 but anyways, I'll be back in a bit 18:54:58 tabemann, you could try 'cforth' thats mitch bradleys Forth and it's got a totally open licence afaics 18:58:26 --- quit: tabemann (Ping timeout: 268 seconds) 18:58:53 --- join: iyzsong joined #forth 19:12:49 tp, congrats on the bug squashing :D 19:14:15 jsoft, thanks, but after all that ... I didnt save the commit somehow and now I have a new bug and no working code to fall back on ...sigh 19:14:41 I remember committing it, I do that 3 pr 4 times a day, and yet there is no record of it! 19:15:00 ah well, thats why Im a tech and not a programmer :) 19:15:51 it's somewhere in 3 files, all are in the SCM, just a matter of finding which one 19:16:05 I've probably just made a typo somewhere 19:23:27 --- join: tabemann joined #forth 20:41:17 --- join: commonlisper9 joined #forth 20:43:24 Hello again :) Hope everyone is doing well. 20:45:54 I'd like to know what the general approach to making generic words is. If for example there is some record, like a linked list node - how would you make words that operate on nodes that are of different types? Would you first need to implement a type system, or perhaps feed a parameter on the stack representing the size of the record? 20:46:23 --- join: gravicappa joined #forth 20:50:32 commonlisper9: for my forth, hashforth, I implemented single-dispatch generic words by associating each class with an intmap, and assigning a unique integer to each generic word 20:50:49 where when called the generic word would look up its integer in the intmap for the argument's class 20:50:59 and if it found it, it would call it 20:51:45 otherwise it would look up the index for the unknown message handler in the intmap, and if present call that 20:51:56 and if that wasn't found, it would throw an exception 21:04:39 Ah, clever design 21:06:27 What about a template style? Is that... feasible or is it the wrong sort of approach? 21:08:22 Also, happy to say that my copy of Forth Programmer's handbook came :) 21:08:36 It appears to have been printed as soon as I ordered it 21:09:32 templates, if feasible, are probably not the right approach 21:18:01 --- join: ryke joined #forth 22:22:37 --- quit: smokeink (Remote host closed the connection) 22:22:57 --- join: smokeink joined #forth 22:35:38 --- quit: gravicappa (Ping timeout: 250 seconds) 23:39:17 --- part: commonlisper9 left #forth 23:54:51 wow, the f051's LSI is inaccurate 23:55:10 lsi ? 23:57:14 low speed internal oscillator, an RC beast 23:57:24 oh yes 23:57:51 Ive never used it, it's just for the adc or something ? 23:58:03 it can drive the RTC, wakeup timers, etc 23:58:05 and it cant be used as the system clock 23:58:18 err I think it can 23:58:46 I coundn't see how when I looked at the possibility 23:59:03 but I could have missed something 23:59:38 thats why I got a STM32L073 to play with, the M0+ class has everything 23:59:39 Ahh yeah nah 23:59:59 --- log: ended forth/19.11.30