00:00:00 --- log: started forth/04.12.13
00:53:39 --- quit: OrngeTide ("nite")
04:19:36 --- quit: saon ("Leaving")
05:20:51 --- quit: raystm22 ("User pushed the X - because it's Xtra, baby")
07:29:29 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth
08:22:49 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth
09:01:13 --- quit: tgunr (Remote closed the connection)
09:03:53 --- join: qFox (C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth
09:21:53 --- quit: I440r_ ("Leaving")
09:39:29 --- join: Raystm2 (~Ray@adsl-68-95-255-125.dsl.rcsntx.swbell.net) joined #forth
09:46:35 --- join: snowrichard (~richard@adsl-64-217-183-211.dsl.lgvwtx.swbell.net) joined #forth
09:48:27 --- quit: snowrichard (Remote closed the connection)
10:06:55 --- quit: Raystm2 ("User pushed the X - because it's Xtra, baby")
10:12:02 --- join: raystm22 (~color4th@adsl-68-95-255-125.dsl.rcsntx.swbell.net) joined #forth
12:09:17 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
12:09:33 <slava> what x86 registers can one freely clobber without push/pop? i know eax is one.
12:09:58 --- join: Raystm2 (~Ray@adsl-68-95-255-125.dsl.rcsntx.swbell.net) joined #forth
12:10:05 <slava> Raystm2, hi
12:10:08 <slava> Raystm2, what x86 registers can one freely clobber without push/pop? i know eax is one.
12:11:00 <Raystm2> Hi slava :)0
12:12:51 <Raystm2> hmm I've see that somewhere, lets see if EAX is one then edx is one also right? ah I've seen it in Randal Hydes High Level Assembler site Webster.com. 
12:13:10 <slava> edx? k.
12:13:46 <Raystm2> Unfortunately i've just reinstalled this system and don't have the HLA docs yet. 
12:14:05 <Raystm2> I'm going to he's site for you to find out for sure :)
12:14:18 <slava> nah its ok
12:14:21 <slava> i'll do it myself
12:16:06 <Raystm2> ok -- but know I have to know also :)
12:18:20 <Raystm2> it's not webster.com it's webster.cs.ucr.edu
12:19:17 <slava> i have to redo my compiler to store the stack pointer in a global var not in a register, because of windows compiler crapiness
12:21:00 --- join: divgrad (~wer@leased-line-224-207.telecom.by) joined #forth
12:27:07 <Raystm2> really.
12:27:22 <Raystm2> is there no room for you to use a register?
12:28:49 <slava> oh, its not that
12:29:02 <Raystm2> it's the compiler?
12:29:18 <slava> my interpreter is written in C, and the datastack is in a C global. with gcc, I can use an asm declaration to put it in a registe.r not so with MS VC++.
12:29:49 <Raystm2> oh.
12:29:55 <Raystm2> I love learning :)
12:31:48 --- join: fca (1469-ident@h229n2fls31o815.telia.com) joined #forth
12:35:26 --- part: divgrad left #forth
12:58:39 --- join: saon (1000@c-24-129-90-197.se.client2.attbi.com) joined #forth
13:28:16 --- join: arke (apache@11.198.216.81.dre.siw.siwnet.net) joined #forth
13:30:28 --- quit: saon ("Leaving")
13:31:23 --- join: I440r (~mark4@216-110-82-203.gen.twtelecom.net) joined #forth
13:31:23 --- quit: I440r (Client Quit)
13:37:41 --- quit: raystm22 ("User pushed the X - because it's Xtra, baby")
13:39:52 --- quit: arke ("CGI:IRC (EOF)")
13:39:52 --- join: guest142 (apache@11.198.216.81.dre.siw.siwnet.net) joined #forth
13:40:03 --- nick: guest142 -> arke
13:43:04 --- join: I440r (~mark4@216-110-82-203.gen.twtelecom.net) joined #forth
13:49:55 --- join: raystm22 (~color4th@adsl-68-95-255-125.dsl.rcsntx.swbell.net) joined #forth
13:55:36 --- part: slava left #forth
14:01:17 --- quit: arke ("CGI:IRC (EOF)")
14:07:03 --- quit: Raystm2 ("User pushed the X - because it's Xtra, baby")
14:09:50 --- join: saon (1000@c-24-129-90-197.se.client2.attbi.com) joined #forth
14:16:20 --- join: Raystm2 (~Ray@adsl-68-95-255-125.dsl.rcsntx.swbell.net) joined #forth
14:18:08 --- quit: raystm22 ("User pushed the X - because it's Xtra, baby")
14:57:43 --- quit: qFox ("this quit is sponsored by somebody!")
15:55:47 --- quit: I440r ("Leaving")
16:08:21 --- quit: saon (Remote closed the connection)
16:09:35 --- join: saon (1000@c-24-129-90-197.se.client2.attbi.com) joined #forth
16:10:52 --- quit: saon (Remote closed the connection)
16:11:47 --- join: Sonarman (~matt@adsl-67-113-234-182.dsl.snfc21.pacbell.net) joined #forth
16:19:33 --- join: raystm22 (~color4th@adsl-68-95-255-125.dsl.rcsntx.swbell.net) joined #forth
16:35:25 <crc> hello all
16:40:11 --- quit: crc ("Leaving")
16:44:11 --- join: crc (~crc@pool-70-20-138-220.phil.east.verizon.net) joined #forth
16:44:16 --- mode: ChanServ set +o crc
17:07:18 --- join: forthweasel (~purple@c-4d5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
17:07:38 <crc> hi forthweasel
17:07:53 <forthweasel> Hi, crc 
17:08:18 * forthweasel is working at his new Forth. I hope it gets as fast as I want it to be!
17:08:30 <crc> cool
17:08:47 <forthweasel> NO high-level Forth code in the compiler.
17:08:51 <forthweasel> Except macros like IF
17:08:56 <crc> :)
17:09:16 <forthweasel> And I use a machieforthish set of core words, to base the optimizer on.
17:09:23 <crc> cool
17:09:24 <forthweasel> Planning to code it, too, in asm.
17:09:34 <crc> what assembler?
17:09:36 <forthweasel> I hope the extra complexity pays off.
17:09:37 <forthweasel> FASM
17:09:45 * crc likes FASM
17:09:46 <Sonarman> 5B
17:09:48 <forthweasel> Since it can do the hashing for me.
17:10:15 * crc has a pretty slow compiler, but it's *clean* and easy to understand
17:10:22 <forthweasel> Did I tell you that hashed dictionaries (even with dicts >100 words) only gave a ~30% speed improvement on my 386?
17:10:33 <crc> nope
17:10:49 <forthweasel> crc: This is actually a few lines of asm source, it's not that bad (hey, it's forth after all)
17:11:00 --- join: I440r (~mark4@216-110-82-59.gen.twtelecom.net) joined #forth
17:11:00 <crc> I don't used hashed dictionaries in retroforth
17:11:07 <crc> hi I440r
17:11:10 <forthweasel> OK, since that I'm trying to check what other things I can optimize.
17:11:19 <forthweasel> So I'm doing as much as possible in low-level code.
17:11:20 <I440r> hi
17:11:21 <forthweasel> Hi I440r 
17:11:43 <I440r> hi forthweasel 
17:12:00 <I440r> u new in here forthweasel ?
17:12:03 <forthweasel> No.
17:12:05 <I440r> or do i just not know that nick lol
17:12:10 <forthweasel> I'm Robert!
17:12:13 <I440r> lol
17:12:14 <crc> I use a lot of hand-assembled words in several programs
17:12:28 <forthweasel> I440r: Did you measure the speed improvement in isForth after introducing hashed dictionaries?
17:12:35 <I440r> yes
17:12:44 <forthweasel> Remember the numbers you got?
17:12:56 <I440r> a significant improvement but not a HUGE improvement
17:12:59 <I440r> no not really
17:13:04 <forthweasel> I got like 30%
17:13:17 <I440r> because soon after that i added the mmap of files and that made me HUNDREDS of times faster
17:13:20 <forthweasel> Which was less than expected, but 6s -> 4s on the 386 is good
17:13:33 <forthweasel> What did it do before?
17:13:35 <I440r> for how large a file
17:13:42 <I440r> for a 1k file i would say that was pityfull hehe
17:13:46 <forthweasel> I440r: For self-compiling
17:13:57 <forthweasel> Don't know exactly how much
17:14:00 <forthweasel> I'll check
17:14:27 <forthweasel> About 40kB
17:14:37 <I440r> thats respectable
17:14:50 <forthweasel> It's OK, I guess.
17:14:57 <I440r> isforth does 100 times that in the blink of an eye
17:15:05 <I440r> load your whole source first
17:15:08 <forthweasel> On a 386? :)
17:15:11 <I440r> yes
17:15:21 <forthweasel> Cheater. *grumble*
17:15:22 <I440r> i compile 800k per second on a k6-2/550
17:15:28 <I440r> how fast is your 386
17:15:49 <forthweasel> Not very fast. Probably 1% the speed of your box
17:16:01 <I440r> cat /proc/cpuinfo heh
17:16:06 <forthweasel> Low efficiency and iirc a clock speed of 25MHz
17:16:08 <forthweasel> Naaah
17:16:12 <forthweasel> I run MINIX/DOS on it
17:16:27 <forthweasel> btw, what did you do before mmaping?
17:16:37 <I440r> read one line of file at a time
17:16:39 <forthweasel> read() with 1 character? ;)
17:16:44 <I440r> into tib
17:16:50 <I440r> well yes
17:16:50 <forthweasel> No buffering?
17:16:55 <I440r> no buffering
17:16:58 <forthweasel> Whoa
17:17:10 <forthweasel> I'm using a 16kB internal buffer now.
17:17:19 <forthweasel> Hopefully that'll reduce the I/O overhead
17:17:59 <forthweasel> I'm going to do some benchmarking later
17:18:04 <forthweasel> And compare to isForth ;)
17:18:38 <I440r> make a gentoo live cd with isforth on it and see how long a ./extend takes :)
17:19:01 <forthweasel> You know, last time I did that mistake was when I was 13 and learning asm
17:19:19 <forthweasel> Copying a file with 1-char read/writes in MS-DOS on a floppy...that hurts
17:19:36 <I440r> hehe
17:19:36 * forthweasel teases I440r some ;)
17:19:50 <forthweasel> Hmm
17:19:55 <I440r> it wasnt a mistake
17:19:58 <forthweasel> I just noticed my tforth is 184kB!
17:20:06 <I440r> it was a stepping stone 
17:20:08 <forthweasel> (sources)
17:20:19 <forthweasel> Bah, I guess I'll accept that
17:20:31 <I440r> and how much of tforth is really isforth hehehe
17:20:35 <I440r> thief!
17:20:38 <forthweasel> :D
17:20:44 <I440r> :)
17:20:50 * forthweasel gets a little worried
17:20:55 <I440r> lol
17:20:58 --- join: crc2 (~crc2@pool-70-20-138-220.phil.east.verizon.net) joined #forth
17:21:05 <forthweasel> Waaaait a second
17:21:12 <forthweasel> We never had this conversation, OK?
17:21:18 <crc> hmm
17:21:24 * forthweasel discovers how tforth reads from files
17:21:41 <I440r> well i gotta go afk
17:21:45 <forthweasel> Bye
17:21:56 --- quit: tathi ("bed")
17:21:58 <I440r> actually im gona hit the hay reall soon
17:22:09 <I440r> was up till 4ish last nite 
17:22:10 <forthweasel> OK, sleep well and forget all about this :)
17:22:11 <I440r> bleh
17:22:14 <forthweasel> Hehe
17:22:17 <I440r> about what ?
17:22:22 <forthweasel> That's the spirit!
17:22:27 <I440r> about you STEALING isforth ???
17:22:30 <forthweasel> Nah
17:22:31 <I440r> }:)
17:22:43 <forthweasel> About tforth using read() with a 1 character buffer. :P
17:23:02 * forthweasel hides
17:23:04 <I440r> first get file size. then allocate a buffer exactly that big
17:23:08 <I440r> then read file into that buffer
17:23:21 <forthweasel> I should do that.
17:23:23 <I440r> then make that buffer your TIB and set #tib equal to the file size
17:23:38 <I440r> the vector refill to a close/unmap file function
17:24:05 <forthweasel> I can't believe how I overlooked this@$#!
17:24:11 * forthweasel screams in (mental) pain
17:24:21 --- nick: forthweasel -> dumbweasel
17:24:28 <I440r> cool. i inflicted great agony on robert :)
17:24:28 <dumbweasel> Oh, well, good night ;)
17:24:32 <I440r> my day is complete :)
17:24:32 <dumbweasel> Hah
17:24:51 <dumbweasel> Glad to help
17:40:55 --- quit: raystm22 (Read error: 104 (Connection reset by peer))
18:02:43 <Sonarman> hey dumbweasel
18:02:56 <Sonarman> did you still want those stats on letter frequency in the gforth dictionary?
18:03:22 <Sonarman> well, i've gotta go
18:03:35 <Sonarman> i'll msg the top 15 to you (with percentages) in case you want them
18:03:38 --- quit: crc2 (Read error: 104 (Connection reset by peer))
18:03:56 --- join: crc2 (~crc2@pool-70-20-138-220.phil.east.verizon.net) joined #forth
18:04:15 --- quit: crc2 (Read error: 104 (Connection reset by peer))
18:06:25 --- join: crc2 (~crc2@pool-70-20-138-220.phil.east.verizon.net) joined #forth
18:33:28 --- join: saon (1000@c-24-129-90-197.se.client2.attbi.com) joined #forth
18:42:36 --- quit: crc2 (Remote closed the connection)
18:43:48 <crc> Hi saon
18:44:31 <saon> hey crc
18:45:31 <crc> I'm going to bed in ~10 minutes :)
18:50:25 <crc> goodnight
19:09:17 <dumbweasel> Night
19:09:31 <dumbweasel> And I, the dumb weasel Swede, is still up
19:39:00 --- join: arke (f2@bespin.org) joined #forth
19:40:56 <dumbweasel> Hi
19:41:24 <arke> hi
19:41:37 <Herkamire> hi
19:43:32 <arke> :)
20:18:58 --- quit: arke ("Lost terminal")
20:24:45 <Herkamire> hehe :) this artical was listing some of the benifits of pear programming then said: "An additional benefit of pairing is that accumulated wisdom and best practices can be rapidly disseminated throughout an
20:24:49 <Herkamire> organization through successive pairings. This is, incidentally, the same benefit that sexual reproduction brought to the
20:24:52 <Herkamire> genome."
20:25:06 <dumbweasel> Pear programming - sounds tasty
20:25:15 <Sonarman> hahaha... damn you, weasel
20:25:20 * dumbweasel eats Sonarman 
20:25:50 * Sonarman lodges himself in dumbweasel's windpipe
20:26:56 --- join: raystm22 (~color4th@adsl-68-95-255-125.dsl.rcsntx.swbell.net) joined #forth
20:27:20 <dumbweasel> Hi Ray
20:28:57 <raystm22> Hi dumbweasel :0
20:29:14 <raystm22> what are working on? anything?
20:29:50 <dumbweasel> Yeah
20:30:05 <dumbweasel> This Forth compiler for Linux, trying to get is as fast as possible.
20:30:16 <dumbweasel> But I'm also planning an optimizer to make it fast.
20:30:27 <dumbweasel> We'll see how well that works out.
20:30:28 <raystm22> optimizing, is fun :)
20:30:45 <dumbweasel> Well...it certainly makes things a bit more interesting!
20:30:52 <Herkamire> hehe:
20:30:52 <Herkamire> When you build a prototype, there is always the risk that someone will say "that's good enough, ship it". One way to minimize
20:30:55 <Herkamire> the risk of a prototype being put into production is to write the prototype in using a language or tool that you couldn't
20:30:58 <Herkamire> possible use for a production version of your product.
20:31:12 <raystm22> :)
20:31:30 <dumbweasel> Herkamire: Then your boss will ask you why you can't ship this QBasic program.
20:31:40 <raystm22> then you can sell both systems as separate or together if you wish.
20:32:03 <Sonarman> Herkamire: where is this from? :)
20:32:04 <dumbweasel> Heh, selling a prototype in BASIC together with the real one?
20:51:11 <Herkamire> http://www.laputan.org/mud/mud.html
20:51:44 <Herkamire> dumbweasel: yeah, I was trying to think of how to make it uterly impossible :)
20:55:28 <Herkamire> writing in herkforth should work
20:58:22 <dumbweasel> Hehe
20:58:42 <dumbweasel> Or making up some very odd language, where the compiler inserts random bugs.
20:58:54 <dumbweasel> Or maybe that was C++
21:13:09 --- quit: raystm22 (Read error: 110 (Connection timed out))
21:13:17 --- quit: Raystm2 (Read error: 110 (Connection timed out))
21:19:07 <ASau> Labas ritas!
21:19:17 <dumbweasel> Privet, ASau 
21:22:56 <Herkamire> heh: We believe that a certain amount of up-front planning and design is not only important,
21:23:00 <Herkamire> but inevitable.
21:24:51 <dumbweasel> Well...a design is inevitable. Even BASIC programmers end up with one ;)
21:26:56 --- join: Raystm2 (~Ray@adsl-68-95-255-188.dsl.rcsntx.swbell.net) joined #forth
21:27:05 <dumbweasel> Hi again
21:39:17 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
21:39:37 --- join: slava2 (~slava2@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
21:39:44 <slava2> hello
21:39:59 --- part: slava left #forth
21:40:20 <slava2> dumbweasel, how's your compiler?
21:41:11 <dumbweasel> dumb.
21:41:24 <dumbweasel> Well, so far I just got the basic ITC compiler running.
21:41:50 <dumbweasel> Going to try to code an optimizer in asm for it.
21:41:56 <slava2> dumbweasel: optimizer in asm?
21:42:00 <dumbweasel> Right.
21:42:03 <slava2> why?
21:42:18 <slava2> write the optimizer in forth then have it optimize itself :)
21:42:18 <dumbweasel> Because I like to cry
21:42:36 <dumbweasel> Hmm... I guess that's reasonable.
21:42:40 <slava2> dumbweasel, i'm using an irc cleint written in factor right now
21:42:46 <dumbweasel> Neat
21:43:47 <slava2> dumbweasel: are you writing your own assembler?
21:44:01 <slava2> i'm trying to find a *readable* assembler in forth that i can port
21:44:10 <slava2> gforth's one is written in blocks and its like 100 screens long
21:44:13 <slava2> win32forth is... win32forth
21:44:43 <dumbweasel> No, not really...
21:45:12 <dumbweasel> It's a pretty small subset one actually uses, so it shouldn't be too hard. Except the addressing *shrug*
21:45:16 <slava2> you'll just have some hex literals here and there? :)
21:45:35 <slava2> the addressing is the main problem -- direct, indirect, displacements, etc.
21:45:45 <dumbweasel> Hehe, nah, the  isForth prototype had a small "assembler".
22:21:06 --- quit: slava2 (Remote closed the connection)
22:30:16 --- quit: Raystm2 ("User pushed the X - because it's Xtra, baby")
22:50:04 --- quit: Sonarman ("night")
22:55:09 <ASau> I have very productive day-offs recently.
22:55:13 <ASau> But not for Forth.
22:55:26 <ASau> I had...
23:05:07 --- join: cmeme (~cmeme@216.184.11.2) joined #forth
23:05:24 --- quit: cmeme (Remote closed the connection)
23:06:08 --- join: cmeme (~cmeme@216.184.11.2) joined #forth
23:59:59 --- log: ended forth/04.12.13