00:00:00 --- log: started forth/17.04.28 00:13:51 --- quit: ACE_Recliner (Ping timeout: 255 seconds) 00:15:03 --- join: ACE_Recliner (~ACE_Recli@c-50-165-178-74.hsd1.in.comcast.net) joined #forth 01:34:04 --- join: John[Lisbeth] (~user@2601:601:8f01:a6a0:d0c6:610b:a9d2:4eec) joined #forth 02:10:26 --- quit: nighty-- (Quit: Disappears in a puff of smoke) 02:58:02 --- join: GeDaMo (~GeDaMo@212.225.127.213) joined #forth 03:32:58 --- quit: ACE_Recliner (Remote host closed the connection) 03:38:40 I made a presentation about my Forth: 03:38:43 https://gitpitch.com/larsbrinkhoff/lbForth/pitch 03:42:51 interesting stuff 03:43:19 how far from realization is the AVR version? i have a project which would benefit from this a lot 03:46:15 --- join: true-grue (~true-grue@176.14.219.178) joined #forth 03:57:41 Quite far, sorry. 03:59:27 I have to split code and data into separate memory spaces to support the Harvard architecture. 04:06:57 i may have a chance to take a look at it, depending on how much of a disappointment amforth will be 04:16:14 Cool. You could try running the sample AVR assembly language program. 05:11:53 --- join: nighty-- (~nighty@s229123.ppp.asahi-net.or.jp) joined #forth 05:44:21 --- quit: wa5qjh (Remote host closed the connection) 06:28:28 --- join: roundsf`` (~user@2a00:2381:1a72:10:4d0:a9fd:4224:d728) joined #forth 07:34:59 indeed 08:25:45 --- join: Keshl_ (~Purple@24.115.185.149.res-cmts.gld.ptd.net) joined #forth 08:26:06 --- quit: Keshl (Ping timeout: 260 seconds) 08:33:21 --- join: neceve (~ncv@86.125.247.109) joined #forth 08:33:21 --- quit: neceve (Changing host) 08:33:21 --- join: neceve (~ncv@unaffiliated/neceve) joined #forth 08:44:36 --- join: kkjliio (50b26e9b@gateway/web/freenode/ip.80.178.110.155) joined #forth 09:10:44 --- quit: Keshl_ (Read error: Connection reset by peer) 09:11:05 --- join: Keshl_ (~Purple@24.115.185.149.res-cmts.gld.ptd.net) joined #forth 09:14:20 * crc has been reading comp.lang.forth again. It's even worse now than it used to be. :( 09:15:08 at least we got dimensions and jfar .... ;) 09:26:02 : hope ( hour -- ) darkest? if dawn then ; 09:28:08 --- join: John[Lis` (~user@2601:601:8f01:a6a0:d0c6:610b:a9d2:4eec) joined #forth 09:31:53 --- quit: John[Lisbeth] (Ping timeout: 246 seconds) 09:51:36 It's a shame about comp.lang.forth. There are many good programmers there, but it's like many of them prefer to engange trolls rather than talk about Forth 09:54:01 Well, the problem is that they see their opponents as trolls. The internet was much better place to discuss when people did not call each other "trolls". 09:55:11 but there were also fewer of those who posted something just to annoy others 09:56:36 larsb: was your presentation made with org-mode + reveal? 09:59:20 No, I made it with a new thing callled gitpitch. 10:02:50 kkjliio: you asked the other day about setting bits within a word. https://pastebin.com/SETZHYBt 10:26:10 bleh? 10:29:10 --- join: gravicappa (~gravicapp@ppp83-237-167-221.pppoe.mtu-net.ru) joined #forth 10:34:55 DocPlatypus: ? 10:35:09 I got a message with just "bleh" in it 10:35:30 from John[Lis` ? 10:35:37 ya 10:36:19 he's not able to speak on this channel due to trolling behavior and has been sending random bits like that to me since then 10:49:26 https://pastebin.com/raw/0V7B4Xa0 - not sure if 10 times, didn't get my measure technique down yet 10:50:42 10 times? 10:51:51 --- join: Zarutian (~zarutian@168-110-22-46.fiber.hringdu.is) joined #forth 10:52:10 i thought your version was 10 times faster, but it's inconclusive, its much faster can't tell by how much 10:55:14 ah 10:55:44 it could be made faster if one really wanted 11:05:14 --- quit: proteusguy (Ping timeout: 260 seconds) 11:14:38 --- join: proteusguy (~proteus-g@2405:9800:b408:bc31:5e96:9dff:fe27:a5a1) joined #forth 11:14:38 --- mode: ChanServ set +v proteusguy 11:41:44 apparently I haven't been keeping up with things 11:50:26 kkjliio: Here's a simpler possible definition of mask: 11:50:26 : lshift-by swap lshift ; : mask 1 lshift-by 1- lshift-by ; 11:54:48 pointfree: pointfree: nice, thanks. 11:57:26 zounds. i just noticed that both our definitions of mask fail when the mask size is the word size 12:09:29 lshift-by - nice, it all so a tiny bit faster 12:09:36 also* 12:13:28 --- quit: neceve (Quit: Konversation terminated!) 12:14:51 good day, thanks for the talk. i'll cya around. bb. 12:15:38 --- quit: kkjliio (Quit: Page closed) 12:19:38 koisoke: : lshift-by swap lshift ; : mask 1 lshift-by dup dup 1- lshift-by or ; 12:25:48 whoops there might be a bug in that too... 12:26:35 yeah. could do it easily with conditionals or double numbers, but that would be gross 12:44:59 * crc has rewritten the core string-to-number conversion code in his forth. No more variables and quite a bit shorter. :) 12:56:10 koisoke: : mask ( start count -- mask) 1- dup 1 lshift-by -rot 1 lshift-by 1- lshift-by or ; \ 0 &64 mask . -1 0 &4 mask u. 1111 13:05:34 pointfree: that ensaddens on the zero-length case. not sure if that matters in practice though 13:06:01 right now i have : mask dup 1 lshift-by tuck > if drop true else 1- then lshift-by ; but that is a bit gross 13:08:27 koisoke: There's one other bug in my latest definition. 13:09:07 koisoke: There's a whole book on this topic that I'd like to get through http://www.jjj.de/fxt/ 13:11:11 http://www.jjj.de/fxt/fxtbook.pdf 13:11:36 this looks worth a look-over. thanks 13:16:26 --- quit: bedah (Quit: Ex-Chat) 13:31:10 * koisoke leaves it at : mask 1 over lshift tuck > if drop true else 1- then swap lshift ; lest the whole day vanish into this 13:35:10 gforth on intel does lshift modulo the word size. not sure if that holds on other architectures. 13:36:35 rshift too 13:37:24 koisoke: probably because shifting is usually implemented via mux-es iirc 13:37:37 so true 64 rshift -> true, true 127 rshift -> 1 13:51:52 --- quit: true-grue (Read error: Connection reset by peer) 13:53:07 --- quit: GeDaMo (Remote host closed the connection) 13:55:47 Zarutian: *nod* i guess checking if any of the higher bits are set and giving 0 (or -1 for signed rightshift) for the shift result would waste die space supporting something programs shouldn't do anyway 13:56:46 one some archs shifting is not implemented as instructions. Only a single bit rotate operation is. 13:57:45 that too 13:58:15 x86 has shl/shr/sal/sar though 13:59:33 x86 has BCD mode conversion instructions too so there is that 14:23:51 --- join: hellcode (~sigma@189.180.110.142) joined #forth 14:26:35 --- quit: gravicappa (Ping timeout: 268 seconds) 15:29:51 --- nick: hellcode -> GNU-hellcode 15:31:21 --- nick: GNU-hellcode -> not-yakuza 16:04:06 --- join: ACE_Recliner (~ACE_Recli@c-50-165-178-74.hsd1.in.comcast.net) joined #forth 16:10:22 --- join: wa5qjh (~Thunderbi@121.54.90.138) joined #forth 16:12:39 --- quit: ACE_Recliner (Ping timeout: 268 seconds) 16:13:30 --- part: not-yakuza left #forth 16:14:20 --- join: ACE_Recliner (~ACE_Recli@c-50-165-178-74.hsd1.in.comcast.net) joined #forth 16:27:46 --- quit: DocPlatypus (Ping timeout: 260 seconds) 16:38:54 --- quit: John[Lis` (Ping timeout: 258 seconds) 16:42:08 --- join: DocPlatypus (~skquinn@2601:2c2:c300:ff70:c1b:9a7:c6f6:d56) joined #forth 17:41:51 --- quit: Zarutian (Quit: Zarutian) 19:46:16 --- join: neceve (~ncv@unaffiliated/neceve) joined #forth 20:52:54 --- mode: ChanServ set -o koisoke 20:53:25 ?! just identified to nickesrv and chanserv deops me 21:11:31 pwn3d 21:21:30 --- quit: ACE_Recliner (Ping timeout: 240 seconds) 21:24:29 --- join: ACE_Recliner (~ACE_Recli@c-50-165-178-74.hsd1.in.comcast.net) joined #forth 21:33:50 --- quit: coelebs (Ping timeout: 260 seconds) 21:49:06 --- join: coelebs (viileppi@gateway/shell/blinkenshell.org/x-uunfoayuwalkcytg) joined #forth 22:39:29 --- quit: wa5qjh (Remote host closed the connection) 23:42:17 --- join: bedah (~bedah@85.16.91.37) joined #forth 23:59:59 --- log: ended forth/17.04.28