00:00:00 --- log: started forth/18.11.21
00:01:35 --- nick: jedb_ -> jedb
00:38:24 <john_cephalopoda> I got a strange bug in my program...
00:38:38 <john_cephalopoda> https://bpaste.net/show/8e5d38090995
00:38:50 <john_cephalopoda> matrices.fth:51: Invalid memory address
00:40:33 <john_cephalopoda> Oh, wait...
00:40:54 <john_cephalopoda> Yeah, I'm stupid. Forgot an RDROP
00:51:23 --- quit: rpcope (Ping timeout: 244 seconds)
00:58:45 --- quit: john_cephalopoda (Quit: Trees can see into your soul.)
01:04:06 --- join: ncv (~neceve@unaffiliated/neceve) joined #forth
01:10:12 --- quit: rdrop-exit (Quit: Lost terminal)
01:13:02 --- quit: ncv (Remote host closed the connection)
01:39:26 --- quit: pierpal (Ping timeout: 244 seconds)
01:41:06 --- join: pierpal (~pierpal@host91-236-dynamic.22-79-r.retail.telecomitalia.it) joined #forth
01:46:39 --- quit: pierpal (Quit: Poof)
01:46:55 --- join: pierpal (~pierpal@host91-236-dynamic.22-79-r.retail.telecomitalia.it) joined #forth
02:03:06 --- quit: ashirase (Ping timeout: 268 seconds)
02:07:59 --- quit: smokeink (Remote host closed the connection)
02:08:05 --- join: ashirase (~ashirase@modemcable098.166-22-96.mc.videotron.ca) joined #forth
02:08:20 --- join: smokeink (~smokeink@118.131.144.142) joined #forth
02:16:47 --- join: nighty- (~nighty@s229123.ppp.asahi-net.or.jp) joined #forth
03:17:03 --- join: john_cephalopoda (~john@unaffiliated/john-cephalopoda/x-6407167) joined #forth
03:20:09 <john_cephalopoda> https://thecutecuttlefish.org/tmp/mobile.png <-- pforth on android
03:39:49 --- quit: xek (Ping timeout: 244 seconds)
03:45:22 --- join: dddddd (~dddddd@unaffiliated/dddddd) joined #forth
04:05:00 --- join: xek (~xek@apn-37-248-138-80.dynamic.gprs.plus.pl) joined #forth
04:12:19 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
04:36:01 --- quit: john_cephalopoda (Quit: Trees can see into your soul. They lurk everywhere.)
05:00:30 --- join: john_cephalopoda (~john@unaffiliated/john-cephalopoda/x-6407167) joined #forth
05:01:23 --- quit: rpcope (Ping timeout: 252 seconds)
05:06:15 --- quit: pierpal (Quit: Poof)
05:06:33 --- join: pierpal (~pierpal@host91-236-dynamic.22-79-r.retail.telecomitalia.it) joined #forth
05:08:21 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
05:09:41 --- quit: dddddd (Ping timeout: 246 seconds)
05:16:12 --- join: proteusguy (~proteus-g@cm-134-196-84-98.revip18.asianet.co.th) joined #forth
05:16:12 --- mode: ChanServ set +v proteusguy
05:20:24 --- quit: proteusguy (Client Quit)
05:20:42 --- join: proteusguy (~proteus-g@cm-134-196-84-98.revip18.asianet.co.th) joined #forth
05:20:42 --- mode: ChanServ set +v proteusguy
05:21:53 --- join: dddddd (~dddddd@unaffiliated/dddddd) joined #forth
05:34:44 --- join: Zarutian (~zarutian@173-133-17-89.fiber.hringdu.is) joined #forth
05:37:05 --- quit: gravicappa (Ping timeout: 268 seconds)
06:05:18 --- quit: rpcope (Ping timeout: 250 seconds)
06:21:19 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
06:25:48 --- quit: rpcope (Ping timeout: 268 seconds)
06:33:49 <WilhelmVonWeiner> there is a gforth app by the way john_cephalopoda 
06:35:50 <WilhelmVonWeiner> also you can RDROP out of the loop at any point if you put those MATRIX...THEN checks in their own words
06:37:37 <WilhelmVonWeiner> not loop, "create" word. Someone said loop in a talk I'm listening to.
06:38:38 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
06:43:07 --- quit: smokeink (Remote host closed the connection)
06:56:42 --- quit: rpcope (Ping timeout: 276 seconds)
07:15:51 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
07:16:28 --- quit: jedb (Remote host closed the connection)
07:19:24 --- join: MrMobius (~default@c-73-134-82-217.hsd1.va.comcast.net) joined #forth
07:22:30 --- quit: rpcope (Ping timeout: 245 seconds)
07:36:10 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
07:38:06 --- join: jedb (~jedb@199.66.90.113) joined #forth
07:38:20 <KipIngram> Conditional returns for the win...
07:42:03 --- quit: Zarutian (Read error: Connection reset by peer)
07:42:09 --- join: Zarutian_2 (~zarutian@173-133-17-89.fiber.hringdu.is) joined #forth
07:48:55 --- quit: tabemann (Ping timeout: 250 seconds)
08:14:46 --- nick: Zarutian_2 -> Zarutian
08:54:23 --- join: xek_ (~xek@apn-37-248-138-83.dynamic.gprs.plus.pl) joined #forth
08:58:02 --- quit: xek (Ping timeout: 264 seconds)
09:01:11 --- quit: rpcope (Ping timeout: 244 seconds)
09:06:49 --- quit: moony (Write error: Connection reset by peer)
09:08:41 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
09:13:09 --- quit: rpcope (Ping timeout: 240 seconds)
09:16:33 --- join: gravicappa (~gravicapp@ppp83-237-167-87.pppoe.mtu-net.ru) joined #forth
09:18:59 --- nick: xek_ -> xek
09:45:09 --- join: lonjil2 (~quassel@2a02:418:6050:ed15:ed15:ed15:e741:32d6) joined #forth
09:45:55 --- quit: lonjil (Ping timeout: 250 seconds)
09:47:09 --- join: probablymoony (moony@unaffiliated/moonythedwarf) joined #forth
10:45:26 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
10:50:12 --- quit: rpcope (Ping timeout: 244 seconds)
10:54:46 --- quit: dys (Ping timeout: 252 seconds)
11:00:52 --- quit: xek (Ping timeout: 264 seconds)
11:01:44 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
11:02:52 --- join: dys (~dys@tmo-107-89.customers.d1-online.com) joined #forth
11:09:03 --- join: xek (~xek@apn-37-248-138-80.dynamic.gprs.plus.pl) joined #forth
11:13:44 --- quit: rpcope (Ping timeout: 252 seconds)
11:24:35 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
12:26:22 * crc is going to implement something similar to the dcpu16 hardware model in the next release of retro. (thanks to prompting by Zarutian)
12:35:15 <WilhelmVonWeiner> Are you adding I/O to Nga, Rx, or RetroForth?
12:36:16 <WilhelmVonWeiner> (This DCPU-16 style thing.) 
12:36:59 <crc> Retroforth, specifically. Not the minimal nga spec or the basic kernel part (rx).
12:38:05 <crc> It'll replace the entire current set of additional opcodes with three new ones for enumerating, identifying, and interacting with the I/O
12:39:50 <crc> This mostly affects the internals, but will make it easier to adapt and reuse the IO code across the different interfaces.
12:41:16 <crc> It'll also make it considerably easier for me to add new I/O in the future as this will internally use a jump table similar to the opcode processor instead of a bunch of giant switch() statements
12:42:05 <WilhelmVonWeiner> I thought the opcode processor was a jump table?
12:42:37 <WilhelmVonWeiner> or am I thinking of Nga.c
12:43:21 <crc> it is, for the core nga opcodes, but not the current i/o opcodes
13:02:13 --- quit: jedb (Read error: Connection reset by peer)
13:06:01 --- quit: rpcope (Ping timeout: 268 seconds)
13:13:25 --- quit: gravicappa (Ping timeout: 268 seconds)
13:13:51 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
13:22:55 --- quit: rpcope (Ping timeout: 245 seconds)
13:28:45 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
13:38:29 --- quit: rpcope (Ping timeout: 240 seconds)
13:41:13 --- quit: xek (Ping timeout: 252 seconds)
13:45:26 --- join: jedb (~jedb@2604:6880:c742:5a70::f3e) joined #forth
13:47:23 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
13:52:04 --- quit: rpcope (Ping timeout: 244 seconds)
14:01:01 --- quit: jedb (Ping timeout: 252 seconds)
14:07:38 --- join: jedb (~jedb@199.66.90.113) joined #forth
14:28:07 * pointfree just tried john_cephalopoda's sixel mandelbrot set visualization. Very cool!
14:46:27 --- join: smokeink (~smokeink@118.131.144.142) joined #forth
15:01:40 <john_cephalopoda> pointfree: Thanks!
15:03:15 <john_cephalopoda> pointfree: I am working on a sixel library and a matrices library right now, so eventually you'll be able to just write your data to a buffer matrix and let the sixel library do the work for you.
15:05:42 <john_cephalopoda> Off for the night~
15:05:42 --- quit: john_cephalopoda (Quit: Trees can see into your soul.)
15:06:12 <WilhelmVonWeiner> why not just write it straight to linear cells? That's what I'm confused about
15:16:24 --- join: rdrop-exit (~markwilli@112.201.164.82) joined #forth
16:06:25 --- quit: wa5qjh (Remote host closed the connection)
16:08:45 --- join: wa5qjh (~quassel@freebsd/user/wa5qjh) joined #forth
16:33:19 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
16:35:36 --- quit: jedb (Quit: Leaving)
16:35:47 --- join: jedb (~jedb@199.66.90.113) joined #forth
16:38:47 --- quit: rpcope (Ping timeout: 268 seconds)
16:39:02 --- join: dave0 (~dave0@47.44-27-211.dynamic.dsl.syd.iprimus.net.au) joined #forth
16:39:34 <dave0> hi
16:44:00 <WilhelmVonWeiner> hey dave
16:44:27 <rdrop-exit> Good morning Dave and Wilhelm
16:44:50 <WilhelmVonWeiner> it's
16:44:55 <WilhelmVonWeiner> 0044 here
16:45:07 <WilhelmVonWeiner> suppose that is kind-of "morning"
16:45:24 <rdrop-exit> 08:44 Thursday here
16:47:32 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
16:47:37 <rdrop-exit> Still on my first cup of coffee
16:50:15 <rdrop-exit> Started watching the Forth Day videos yesterday.
16:50:44 <WilhelmVonWeiner> I should get around to watching some
16:51:26 <WilhelmVonWeiner> i saw the first slab of one about GA and the new GA144, up until the description of the new OKAD
16:51:27 <rdrop-exit> I didn't get through much, too many distractions yesterday.
16:51:52 <rdrop-exit> I watched the first two morning sessions.
16:52:06 <rdrop-exit> Ting and Golding.
16:54:29 --- quit: rpcope (Ping timeout: 240 seconds)
16:55:13 <rdrop-exit> I'll try to finish the morning video sometime today.
16:55:28 <WilhelmVonWeiner> quite long too
16:56:54 <rdrop-exit> True, I might skip over some sessions, if the topic isn't interesting to me.
16:58:45 <rdrop-exit> There are 8 morning sessions on the agenda
17:00:38 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
17:01:37 <rdrop-exit> The afternoon video doesn't have an agenda list
17:05:49 --- quit: rpcope (Ping timeout: 244 seconds)
17:13:10 <pointfree> rdrop-exit: click "sort by -> newest first"
17:14:23 <pointfree> thumbs-up the comment so it becomes a top comment
17:14:31 <rdrop-exit> Thanks Pointfree, that worked. Wonder why it doesn't display unless you sort first.
17:15:00 <rdrop-exit> There are 4 afternoon sesssions
17:15:35 <rdrop-exit> 12 sessions in all, they really packed the day
17:17:05 <rdrop-exit> I'll probably skip the gForth, lisp, and Cosy ones
17:29:07 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
17:33:09 --- quit: rpcope (Ping timeout: 240 seconds)
18:00:44 --- quit: pierpal (Ping timeout: 250 seconds)
18:07:33 --- join: tabemann (~tabemann@h193.235.138.40.static.ip.windstream.net) joined #forth
18:27:35 --- join: rpcope (~GOTZNC@muon.copesystems.com) joined #forth
18:43:28 --- quit: X-Scale (Ping timeout: 244 seconds)
18:46:28 --- quit: dddddd (Remote host closed the connection)
19:01:13 --- quit: tabemann (Ping timeout: 268 seconds)
19:06:17 --- join: X-Scale (~ARM@143.21.54.77.rev.vodafone.pt) joined #forth
19:28:38 --- quit: dave0 (Quit: dave's not here)
20:02:53 --- join: tabemann (~tabemann@2602:30a:c0d3:1890:2db1:4ef6:126b:d5ee) joined #forth
20:38:34 --- join: pierpal (~pierpal@95.239.223.85) joined #forth
21:25:12 --- nick: probablymoony -> moony
21:32:38 * tabemann is happy that he found a faster way to calculate e^x - with only 8 divisions to reach the full precision of double-precision floating point
21:34:34 --- join: dave0 (~dave0@47.44-27-211.dynamic.dsl.syd.iprimus.net.au) joined #forth
21:34:47 <tabemann> hey
21:35:01 <dave0> hi
21:35:09 <rdrop-exit> hello
21:35:43 <dave0> hi tabemann, rdrop-exit 
21:35:46 <dave0> sup?
21:36:08 <rdrop-exit> Just finished lunch, chillin'
21:36:35 <dave0> i've got some candy :-)
21:37:05 <rdrop-exit> I've got Nutella
21:37:08 <tabemann> working on FEXP
21:37:12 <rdrop-exit> :-)
21:37:23 <tabemann> apparently my FEXP works better for 1E than for 10E
21:37:36 <tabemann> so I've gotta increase the number of divisions
21:37:46 <rdrop-exit> I haven't used floating point in 30 years
21:38:21 <rdrop-exit> That paper in the ACM Computing Surveys killed it for me
21:40:06 <rdrop-exit> I still have the issue on my shelf
21:40:23 <rdrop-exit> Just in case
21:41:13 <tabemann> okay 2 + ((70 - 6) / 4) divisions gets me as much precision as I can get
21:41:32 <tabemann> doing everything with fixednums since then?
21:41:54 <rdrop-exit> Yes
21:42:28 <rdrop-exit> and binary scaling
21:42:40 <rdrop-exit> when possible
21:42:57 <tabemann> yeah, I implemented floating point in attoforth, complete with a separate floating point stack and being able to load and store to memory with both single-precision and double-precision
21:43:12 <rdrop-exit> https://en.wikipedia.org/wiki/Binary_scaling
21:44:07 <rdrop-exit>  pi Convergent rational approximation of pi.
21:44:10 <rdrop-exit>  : pi ( -- numerator denominator )
21:44:18 <rdrop-exit>    $ 24baf15fe1658f99  $ bb10cb777fb8137 ;
21:44:30 <rdrop-exit> for 64-bit Forths
21:46:05 <rdrop-exit> PDF
21:46:10 <rdrop-exit> http://perso.ens-lyon.fr/jean-michel.muller/goldberg.pdf
21:49:37 <tabemann> back
21:55:59 --- quit: proteusguy (Remote host closed the connection)
21:58:20 <tabemann> okay, I honestly did not know about catastrophic cancellation
21:58:55 <tabemann> but I don't see how binary scaling improves on it
21:59:21 <tabemann> because you still have to shift right, and any bits that are below a certain significance are lost
21:59:34 --- quit: pierpal (Read error: Connection reset by peer)
21:59:51 --- join: pierpal (~pierpal@95.239.223.85) joined #forth
22:00:50 --- quit: dave0 (Quit: dave's not here)
22:03:55 <rdrop-exit> For me it's about up front simplicity, you know what you're getting up front, don't have to ponder any weird corner cases
22:06:48 <rdrop-exit> Have you seen this:
22:06:50 <rdrop-exit> https://www.crcpress.com/The-End-of-Error-Unum-Computing/Gustafson/p/book/9781482239867
22:07:55 <tabemann> no I haven't
22:08:11 <rdrop-exit> It was available locally but at double the US price, so I passed on it. Will probably order from Amazon eventually.
22:09:16 <tabemann> I should probably implement fixednums with binary scaling in attoforth
22:09:32 <tabemann> since attoforth on here uses 64-bit cells, there's more than enough precision to go around
22:09:57 <tabemann> and fuck, attoforth supports 128-bit double cells, so potentially I could use those for fixednums
22:11:01 <rdrop-exit> I don't have doubles in my 64-bit Forths, but my scaling operators */mod and u*/mod internally use an 128-bit intermediary.
22:12:15 <rdrop-exit> Also have *hi and u*hi
22:12:48 <rdrop-exit> for the high-order 64 bits or a 128-bit product
22:13:42 <tabemann> yeah, I implemented a full set of double-cell operations
22:14:01 <tabemann> with better support than what Forth 2012 specifies
22:14:31 <tabemann> this is because I anticipate running attoforth on 32-bit systems, and it'd be useful to have 64-bit double cells on those
22:14:32 <rdrop-exit> I figured I'd never need them on a 64-bit Forth, those four operators I just mentioned are all I really need
22:15:29 <rdrop-exit> My 64-bit Forth is for the Host PC side of an umbilical Forth, it's not meant for targets
22:17:03 <tabemann> I'd probably rewrite attoforth if I wanted to run it on a sufficiently small system where I'd consider using an umbilical
22:17:44 <tabemann> I'm thinking more like a Raspberry Pi-type system - smaller than a modern tower or laptop, but way larger than the kind of board you'd run something like Mecrisp on
22:18:38 <tabemann> for starter, attoforth uses a IO manager system designed for running under a preemptive multitasking OS
22:19:14 <tabemann> (attoforth has its own preemptive multitasking system internally independent of and lighter-weight than OS multitasking)
22:19:33 <rdrop-exit> My Forth's are all cooperative.
22:23:30 <rdrop-exit> My host Forth has only two tasks, foreground and background, coroutines for any additional concurrency requirements.
22:23:44 --- join: proteusguy (~proteus-g@184.82.233.251) joined #forth
22:23:44 --- mode: ChanServ set +v proteusguy
22:24:33 <rdrop-exit> *two cooperative tasks
22:24:59 <rdrop-exit> No preemptive multitasking
22:25:58 --- quit: jedb (Read error: Connection reset by peer)
22:26:55 --- join: gravicappa (~gravicapp@ppp83-237-163-149.pppoe.mtu-net.ru) joined #forth
22:31:10 <tabemann> back
22:31:40 <tabemann> yeah, I opted for stuff like locks and condition variables and like
22:32:06 <tabemann> but I've gotta go to bed now
22:32:26 <rdrop-exit> In a cooperative Forth a simple variable can be used as a mutex
22:32:31 <rdrop-exit> good night
22:33:17 --- join: jedb (jedb@gateway/vpn/mullvad/x-iiaaicvyqqvnohnx) joined #forth
22:54:59 --- quit: dys (Ping timeout: 252 seconds)
23:59:59 --- log: ended forth/18.11.21