00:00:00 --- log: started forth/09.08.22
00:09:13 --- join: GeDaMo (n=gedamo@212.225.115.96) joined #forth
00:27:41 --- quit: xjrn (Read error: 60 (Operation timed out))
00:43:42 --- join: kar8nga (n=kar8nga@e-9.vc-graz.ac.at) joined #forth
01:22:32 --- quit: segher (Read error: 110 (Connection timed out))
02:13:51 --- quit: kar8nga (Remote closed the connection)
02:20:40 --- join: DrunkTomato (n=DEDULO@ext-gw.wellcom.tomsk.ru) joined #forth
03:46:13 --- join: segher (n=rehges@84-105-60-153.cable.quicknet.nl) joined #forth
03:55:36 --- quit: madwork (Read error: 104 (Connection reset by peer))
03:55:49 --- join: madwork (n=madgarde@204.138.110.15) joined #forth
04:01:58 --- quit: madwork (Read error: 104 (Connection reset by peer))
04:02:00 --- join: madwork_ (n=madgarde@204.138.110.15) joined #forth
04:11:52 --- quit: GeDaMo ("Leaving.")
07:16:29 --- join: nighty__ (n=nighty@210.188.173.245) joined #forth
07:28:57 --- join: impomatic (n=John@nat67.mia.three.co.uk) joined #forth
07:29:02 <impomatic> Hi :-)
07:48:33 --- join: GeDaMo (n=gedamo@212.225.115.96) joined #forth
07:51:47 <impomatic> : 2over >r >r over over r> r> rot >r rot r> ;   seems like the simplest implementation in Forth.
07:52:22 <impomatic> Without using non-standard words or other double words.
07:54:49 <GeDaMo> : 2over 3 pick 3 pick ;
07:58:00 <impomatic> Thanks, forgot that. I wonder how many Forths have a fast implementation of pick
08:02:37 <GeDaMo> Probably easier to make pick efficient than roll
08:03:30 <GeDaMo> On x86 it would be a single instruction
08:13:12 <segher> : 2over  2 pick 2 pick ;
08:13:40 <segher> some systems have an "under" that you can define as  : under  2 pick ;\
08:14:50 <segher> it's easy to make  roll  efficient, if your stack resides in memory, btw
08:15:21 <segher> well, sort-of efficient, eh.  100x faster than the "stupid" implementation
08:15:36 <segher> you still shouldn't use roll ;-)
08:17:08 <impomatic> So I shouldn't implement swap as   : swap 1 roll ;  ? ;-)
08:17:14 <segher> heh
08:17:22 <segher> : swap  over >r nip r> ;
08:17:34 <impomatic> Don't worry, I'm using : swap tuck drop ;
08:17:43 <segher> or have your compiler track the stack, heh
08:17:59 <impomatic> That does cause me I slight problem, because I'm also using : tuck swap over ;
08:18:24 <segher> how many stack elements do you cache in registers?
08:18:30 <impomatic> Just one.
08:19:04 <segher> don't use swap for anything then ;-)
08:20:17 <impomatic> Hopefully it shouldn't be a problem. I'm on a fast computer. If the Cray can resolve an infinite loop in 6 seconds, mine shouldn't be too far behind.
08:20:26 <GeDaMo> You don't have swap as a primitive?
08:20:38 <segher> i don't.
08:21:08 <segher> all my primitives are one load or store maximum
08:21:21 <impomatic> Actually, I'm just joking. Nearly everything is a primative at the moment.
08:21:24 <segher> (well, the "regular" primitives.  i have  move  as a primitive as well)
08:21:56 <impomatic> I'm just wondering what I can reimplement in Forth without too much of a performance hit.
08:22:08 <segher> it depends
08:22:37 <segher> if your compiler does renaming, you need very different stacj movement primitives
08:23:45 <impomatic> renaming?
08:23:51 <segher> "register" renaming
08:23:55 <segher> stack slot really
08:23:57 <impomatic> Ah okay.
08:24:14 <segher> i cache 0, 1, 2, or "2 swapped" stack entries in register
08:24:52 <segher> it's experimental, i don't yet know how well that works :-)
08:25:21 <impomatic> Does that mean you need up to four version of each primative?
08:25:27 <segher> yeah
08:25:42 <segher> not of every primitive though
08:25:59 <GeDaMo> Only the ones that use the stack? :P
08:26:06 <segher> ha no
08:26:18 <segher> the compiler adjust the stack for many
08:27:55 <segher> another example of primitives you don't need...  "or" is used not much, you don't need it
08:28:30 <GeDaMo> You've been reading Chuck :P
08:28:58 <segher> of course
08:29:08 <segher> but things like paf/maf as well
08:29:45 <segher> i also do * and / in software on most systems
08:30:00 <segher> much easier to write it portable between systems that way :-)
08:30:18 <segher> if your applications needs lots of muls, that not such a great idea of course
08:34:23 <segher> what else...  i don't have DOCON DOVAL DODOES etc anymore
08:34:34 <segher> everything is DOCOL, and it's implicit
08:36:36 <impomatic> Hmmm...
08:37:11 <GeDaMo> "When you've seen one Forth, you've seen one Forth" :P
08:37:21 <impomatic> :-)
08:37:24 <segher> yeah exactly :-)
08:37:51 <segher> although, it's not hard to read other forth's code as soon as you are fluent in one
08:38:05 <segher> well-written code, of course
08:39:57 <impomatic> My Forth by comparison is unremarkable :-)
08:40:05 <segher> ITC?
08:40:27 <impomatic> Or it will be when I've finished it. Still 14 words to finish. I haven't worked on it this week, thanks to being ill.
08:40:38 <impomatic> Yes, ITC, but shouldn't be hard to change.
08:41:25 <segher> ITC is a good choice, really
08:42:01 <segher> better than DTC, anyway
08:42:24 <segher> and much better than "stupid" STC
08:43:29 <impomatic> I thought the advantage of STC is to have the compiler automatically inline words below a certain size.
08:43:40 <impomatic> Anyway, I don't like STC
08:46:32 <segher> "stupid" STC in that definition only inline primitives
08:46:37 <segher> inlines
08:46:47 <segher> you know, 80% of the STC systems
08:46:54 <segher> (totally made up statistic, heh)
08:47:16 <segher> it results in a) big code; and b) slow code, on most CPUs
08:47:40 <segher> while people who do it think the results are the opposite.  sad.  :-P
08:48:52 <impomatic> :-(
09:38:56 --- quit: GeDaMo ("Leaving.")
10:02:56 --- quit: crc (Client Quit)
10:03:27 --- join: crc (n=charlesc@c-68-80-139-0.hsd1.pa.comcast.net) joined #forth
10:36:44 --- quit: nighty__ (Client Quit)
10:42:37 --- quit: DrunkTomato ()
10:47:32 --- join: DrunkTomato (n=DEDULO@ext-gw.wellcom.tomsk.ru) joined #forth
11:35:57 <impomatic> Are Neil Bawd and Wil Baden the same person? The names are anagrams.
11:39:00 <segher> they are
11:39:03 <segher> he is?
11:41:01 --- join: jauaor (n=araujo@gentoo/developer/araujo) joined #forth
11:41:54 <jauaor> hello
11:42:51 <impomatic> Hi jauaor
11:42:59 <jauaor> hello impomatic 
11:43:00 <jauaor> :)
11:55:01 <crc> good afternoon
12:04:33 <impomatic> Sounds like me: "By doing just a little every day you can gradually let the task completely overcome you." (Wil Baden)
13:08:36 --- join: aguai_ (i=aguai@114-44-4-62.dynamic.hinet.net) joined #forth
13:14:14 --- quit: aguaithefreak (Read error: 104 (Connection reset by peer))
13:40:58 * jauaor needs to find a name for his tiny language
14:17:14 --- quit: madgarden (Remote closed the connection)
14:36:30 --- join: madgarden (n=madgarde@CPE001d7e527f89-CM00159a65a870.cpe.net.cable.rogers.com) joined #forth
14:39:32 --- quit: DrunkTomato (Read error: 60 (Operation timed out))
16:33:26 <segher> jauaor: h1n1?
16:35:00 <segher> preon?
18:37:33 <jauaor> segher: hah
18:37:35 <jauaor> h1n1
18:37:38 <jauaor> what does it mean?
18:37:38 <jauaor> :)
18:44:07 <segher> pig flue
18:44:09 <segher> flu
19:15:07 --- quit: madgarden ("Leaving")
19:19:59 --- join: madgarden (n=madgarde@CPE001d7e527f89-CM00159a65a870.cpe.net.cable.rogers.com) joined #forth
19:32:10 --- quit: madgarden ("Leaving")
19:42:05 --- join: madgarden (n=madgarde@99.236.137.241) joined #forth
20:15:35 <jauaor> oh
20:15:36 <jauaor> hah
20:32:10 --- join: crc_ (n=charlesc@c-68-80-139-0.hsd1.pa.comcast.net) joined #forth
20:32:11 --- quit: crc (Read error: 104 (Connection reset by peer))
22:35:52 --- quit: impomatic ("mov.i #1,1")
22:41:48 --- quit: I440r (Read error: 104 (Connection reset by peer))
22:42:09 --- join: I440r (n=mark4__@c-69-136-171-118.hsd1.in.comcast.net) joined #forth
22:42:09 <gogonkt> lo
22:59:54 --- quit: I440r ("Leaving")
23:46:41 --- quit: madgarden (Read error: 145 (Connection timed out))
23:59:59 --- log: ended forth/09.08.22