00:00:00 --- log: started forth/03.12.31 01:05:28 --- join: schihei (~schihei@p5085DBA9.dip.t-dialin.net) joined #forth 01:25:26 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 01:34:50 --- quit: schihei (Client Quit) 01:40:07 --- join: schihei (~schihei@pD9E5CB53.dip.t-dialin.net) joined #forth 02:47:56 --- join: karingo (karingo@88.portland-04-05rs.or.dial-access.att.net) joined #forth 02:54:19 --- quit: onetom ("Lost terminal") 02:54:39 --- join: onetom (~tom@cab.bio.u-szeged.hu) joined #forth 03:07:43 --- join: Klaw (~anonymous@ip68-225-235-97.oc.oc.cox.net) joined #forth 03:07:57 any awake? 03:17:27 Klaw - do you have a question? 03:17:40 err, no, sorry, I sleep at the moment. 03:27:26 no, it's alright, figured it out 03:35:39 --- join: aktnot (ident@233.80-202-65.nextgentel.com) joined #forth 03:39:22 --- join: crc (~crc@ACAFE53F.ipt.aol.com) joined #forth 04:01:54 --- join: networm (~networm@L0632P27.dipool.highway.telekom.at) joined #forth 04:38:57 --- quit: karingo () 05:00:26 --- join: aktnot_ (ident@233.80-202-65.nextgentel.com) joined #forth 05:06:49 --- quit: aktnot (Read error: 60 (Operation timed out)) 05:08:22 --- quit: hovil ("Leaving") 05:14:06 --- quit: crc ("I was using TinyIRC! Visit http://www.tinyirc.net/ for more information.") 05:57:55 --- quit: warpzero (Read error: 110 (Connection timed out)) 06:05:47 --- join: networm_ (~networm@L0651P10.dipool.highway.telekom.at) joined #forth 06:07:56 --- quit: networm (Read error: 60 (Operation timed out)) 06:25:32 --- quit: schihei (Client Quit) 06:30:19 --- join: fridge (~fridge@dsl-203-33-163-107.NSW.netspace.net.au) joined #forth 06:42:38 year @ 1+ year ! 06:43:34 --- quit: aktnot_ ("leaving") 07:03:41 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth 07:57:42 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 07:57:45 hello 07:58:02 y0 08:19:45 hey 08:19:57 fridge: :) 08:26:37 hey peeeps 08:26:54 What the forth is going on here?! 08:27:02 ?? 08:27:20 water run brb 08:28:03 I got a present from my bank for new years. They stole 28.00 from me 08:28:17 ;) 08:28:43 idiot bill payment program takes checks out 2 days BEFORE the date you say you want them paid. 08:29:02 I say pay on the 1st because that is when I get payed. 08:29:10 they pay on 30th and charge me NSF Fee 08:29:18 idiots 08:30:31 is this a bug or a feature? 08:31:12 back to silent running mode now.... 08:32:42 now I have to take out all of the payments I schedules and put them in to be paid 2 days later 08:33:24 which means I need to write down all of the account numbers telephone numbers and crap 08:34:28 Heh. 08:34:54 Not a very good program.. 09:02:07 well I talked to my account manager and she says it is not going to cost me a fee 09:02:10 after all. 09:02:29 they have simply held the check for 24 hours to see if it going to be NSF 09:03:11 but the statement output is plain wrong showng a deposit of 2308 and then a item for 316.26 that is nsf? :) 09:03:31 cybercrud 09:04:23 banks must use the hitchhikers guide to the galaxy mathematical system :) 09:06:07 gastronics? no matter what you order it comes to exactly the amount of money you have in your pocket. 09:32:08 :D 09:41:03 Bring little and order alot then. 09:42:46 hi all 09:47:08 * chrisrw is away: brb 10:13:34 * chrisrw is back (gone 00:26:27) 10:18:19 --- nick: chrisrw -> arke 10:21:11 hey arke 10:25:47 hiya' 10:30:24 --- join: wossname (wossname@HSE-QuebecCity-ppp81161.qc.sympatico.ca) joined #forth 10:31:05 --- quit: proteusguy (Read error: 54 (Connection reset by peer)) 10:42:53 --- quit: snowrichard ("Leaving") 10:46:55 --- quit: networm_ ("Client exiting") 10:51:22 --- nick: arke -> chrisrw 10:52:36 --- nick: chrisrw -> arke 13:16:46 --- quit: wossname ("i'll try f in a bit, be back on tonight") 13:34:42 * arke is away: I'm busy 14:12:59 --- join: wossname (wossname@HSE-QuebecCity-ppp80918.qc.sympatico.ca) joined #forth 14:16:09 --- join: TheBlueWizard (TheBlueWiz@pc40dn1d.ppp.FCC.NET) joined #forth 14:16:09 --- mode: ChanServ set +o TheBlueWizard 14:16:18 hiya all 14:18:11 hey 14:18:18 year changed here 18 mins ago 14:18:27 i wish i coudlnt hear all this noise :/ 14:18:29 hiya mur...Happy New Year 14:18:42 so terrible mess 14:18:46 hehe :) 14:18:48 all insticts target me 14:18:52 visual mess 14:18:53 noise mess 14:19:09 boom booms, music (old hits) and this weird IDM 14:19:20 turned of IDM 14:19:23 IDM? 14:19:23 for peace of mind 14:19:42 music genre 14:19:46 "intelligent dance music" 14:19:58 i thought i described it to you :) 14:20:03 never heard of it 14:20:06 derrived from ambient 14:20:20 ambient, yeah 14:20:23 the room is full of stuff everywhere 14:20:27 material mess 14:20:43 and i'm targetted all media (radio, explositions, this and amiga game i am playing) 14:20:44 * TheBlueWizard shrugs 14:20:46 grrrr 14:20:57 i woudl like to get rid of all 14:21:06 heh 14:21:07 just black room with nothing 14:21:27 but turning off lights woudnt give me peace of mind, because i know that it is stil terrible mess here 14:22:00 black? you sound awfully mad...as if you hate the beginning of a new year already 14:24:40 its very hot here 14:24:44 and terible mess 14:26:18 black is terrible colour 14:26:20 hot? hmm....I thought it'd be rather chilly up in Helsinki 14:28:09 i'm in lahti 14:28:13 the computer is hot 14:28:21 and we have heating 14:28:23 and it's hot 14:28:34 and cna't open doors so that dog will be normal because of the explositions 14:28:38 (although he isn't) 14:28:43 open windows 14:28:45 bad air and hot 14:28:46 ugh 14:29:03 sounds like you got a lousy place to live in 14:30:12 Hey mur 14:30:21 Is it 2004 there? 14:30:31 30 minutes left here :) 14:30:43 Time for some fireworks. Happy new year! 14:39:26 yes 15:08:18 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 15:08:22 hello 15:08:46 looking at a PDF version of F83 Draft ANS standard. 200 + pages 15:08:51 hey 15:08:59 hairy 15:09:03 hiya snowrichard 15:09:24 I'm sure it will be good treatment for my insomnia 15:09:46 like IBM mainframe programming manuals :) 15:10:44 220 pages 15:10:56 tbe 15:11:39 there is an OS project called unununium 15:11:41 what a name 15:12:11 interesting name :) 15:12:37 I think its an element that there has not been enough of it ever made to measuer 15:12:40 measure 15:12:56 high end of periodic table 15:13:20 heh 15:13:27 * mur knows uuu :) 15:13:34 i'm its channel even 15:13:55 they have a channel? I got the system disk to boot under bochs once 15:14:13 bochs cool 15:15:00 oh now I am in #uuu on freenode as an op :) 15:15:38 they moved fromm FN to OFTC 15:16:07 --- quit: warpzero ("Hey guys, I think my internet is about to discon--") 15:16:17 ok 15:17:02 though it would be cool to make a forth to run under the uuu kernel 15:17:24 lighter than linux kernel 15:18:00 but I don't know x86 assembler that well. 15:18:09 lots of other ones though 15:18:22 url to isforth? 15:18:47 isforth.clss.net 15:18:51 tks 15:18:51 --- join: warpzero (~warpzero@dsl.142.mt.onewest.net) joined #forth 15:18:59 --- quit: snowrichard ("Leaving") 15:22:46 --- quit: meme (Read error: 104 (Connection reset by peer)) 15:26:19 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 15:30:44 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 15:30:52 ok that wasn't too bad got it built 15:30:59 the isforth that is 15:31:09 love this new 2Ghz processor :) 15:32:10 back in the old days I used a forth that was all caps. ugly 15:32:24 FPC? not sure 15:33:03 In the desert you can't remember your name la la la 15:34:00 now to read the source for isforth it has been awhile since I looked at it. 15:36:16 open office can open the .f files that is cool. 15:37:41 guess I would start by looking at isforth.asm 15:40:32 int 80 is a linux syscall? 15:42:19 yep 15:43:25 on ancient IBM 360 instruction was called SVC for supervisor call and SVC 14 was EOJ 15:44:09 useless little bits of fluff stuck in my brain :) 15:44:44 A941 6502 code for load A with "A" 15:45:26 I wrote an assembler in machine language 15:45:35 I think I memorized all 156 instructions 15:46:01 sold a copy to the Woz and Jobs 15:46:21 when they were still working in their garage 15:46:32 * TheBlueWizard yanks that useless fluff out of snowrichard's brain and burn it to ashes :) 15:46:58 but I stopped smoking :) 15:47:33 quit in August when I quit drinking :) 16:18:55 * warpzero is away: Happy new years everyone, I hope your year sucks. 16:19:25 not as much as yours will 16:20:20 * TheBlueWizard sends warpzero to /dev/null 16:23:06 You need to have a diskfile in DF0 to use the Kickstart replacement! 16:23:09 anyone knows UAE? 16:25:29 * TheBlueWizard has seen UAE in action, but isn't familiar with it 16:31:07 i can't understand how it works 16:41:16 hello 16:41:41 independence bowl is on so I am back on computer I don't like sports much 16:42:05 I can't understand how it works (football that is) 16:42:55 snowrichard: it is easy to understand football...it is HUMAN WAR! ;) 16:43:36 ceremonial war yes that is kind of what I thought 16:44:21 like that star trek episode where the computer lauched a simulatted nuke and if you draw a bad number you have to walk into a disintegrator 16:45:04 a little worse that getting drafted :) 16:45:57 strawberry fieldsss forever ...... 16:46:14 nothing is real 16:46:16 nah...it is just fun watching football 16:46:42 beatle lyrics 16:46:50 let me take you down 16:47:28 more entertaining than fball 16:47:56 I got that on tape 16:48:01 just now 16:48:44 cheap music library run my sound card out to my tape deck, make cassettes, pick the ones I like and change them back to oggs 16:49:03 audacity 16:49:52 which is corrct 80x25 or 80x24 16:50:15 standard pc console had 25 I think 16:50:24 but? 16:50:48 have been a while 16:51:11 my first terminal was 64x16 (tvt-2) 16:51:49 had about 20 TTL chips in it :) 16:52:20 that was 1976 16:53:07 I got a tty (ASR-33) later. paper tape 16:53:40 blazing 10 CPS (110 baud) 16:53:59 and cpu clock was 1mhz 16:54:01 8 bit 16:54:07 a real speed demon :) 16:54:11 the good old days 16:54:22 pit 16:54:24 out 16:54:56 damn roman candles going off now 16:55:06 hope they dont set the neighborhoood on fire. 16:56:44 the peope at the crack house are setting off the fireworks 16:57:36 crack house closer than the bank some neighborhood. :)) 16:58:45 --- quit: snowrichard ("Leaving") 17:03:21 got it working now 17:03:23 cool 17:04:59 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 17:05:46 I am going to take a stab at documenting the isforth so that I can understand it. I can give you ftp access to get the files after I get the web page for it done. ok? 17:06:13 ARGH! 17:06:36 so cruel. i cant click one, because cursor comes out of window before the program cursor :/ 17:07:45 I am volunteeering to do a web page for isforth. I will get back to you when it done about where to send the files if you want them ok? 17:07:59 phew it woked after all 17:10:46 what worked? 17:10:49 I was out 17:11:12 heard it through the grapevine ...... 17:11:27 not too longer would you be mine .... 17:12:30 I need to download intels x86 reference manuals so I can read the assembler code. 17:13:44 get 'em to fedex you a set 17:13:51 they're free 17:17:08 already downloaded love this DSL 17:17:24 all three volumes 17:18:16 never tried to call them they would not hire me 17:18:32 ? 17:19:02 I applied to work there after I got my second college dgree 17:19:12 I have 3 pdfs 17:19:12 accept that, trust me. i use it instead of the printed manuals usually 17:19:22 it says ".doc" but it's plain text 17:19:30 pdfs are fine, but bloody slow and oversized 17:19:38 you can grep your way right through it, or whatever 17:19:51 but how do I accept it? 17:19:54 in xchat 17:20:56 not an idea 17:21:03 here, wait a second 17:21:28 I found where to accept now 17:21:34 says waiting 17:21:47 done 17:21:52 cool 17:22:06 * arke is back (gone 03:47:24) 17:26:33 [root@localhost richard]# updatedb 17:26:33 [root@localhost richard]# locate intel.doc 17:26:33 /home/richard/.xchat2/downloads/intel.doc 17:26:33 [root@localhost richard]# 17:26:57 would have been nice if the damn thing told me where the heck it put the file. 17:28:36 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth 17:28:36 --- mode: ChanServ set +o kc5tja 17:29:26 hiya kc5tja 17:29:45 THANK you for the file now that I have found it :) 17:30:05 intel.doc 17:31:04 xchat accepts a file and throws it in subdirectory of a hidden directory. 17:31:17 kind of hard to figure out where the heck it went to 17:31:51 irssi gives you a '/set dcc' which will show all the settings that start with 'dcc', including the one that tells irssi where to put 'dcc get'd files. 17:32:13 hi kc5tja 17:33:01 ok I just set it to something more reasonable 17:33:40 /home/richard/downloads 17:33:52 for completed files download to /tmp 17:34:36 dcc offer test.txt 17:36:46 testomg 17:36:52 testing..... 17:37:10 happy new year guys... 17:37:11 :) 17:39:16 --- quit: snowrichard ("Leaving") 17:39:52 whatever timezone you're in.. 17:44:24 Howdy 17:44:25 re TheBlueWizard 17:44:29 re arke 17:44:56 * kc5tja has been diagnosed with a most annoying illness: The Crud. 17:45:08 meaning.? 17:45:58 Imagine pneumonia, but without the fluid in the lungs. 17:46:52 hmmm pneumonia, was that "long infection" ? 17:47:00 hmm...never heard of that name...so it causes dry coughing? 17:47:14 i know the word, but not sure what disease is tied to it... (dutch word) 17:47:33 or like throath ache or something.. 17:47:52 nm i'll webster it 17:48:24 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 17:48:38 well sux... but its treatable i suppose? 17:48:48 you can turnkey app -- that is cool. just read some example code. 17:49:12 TheBlueWizard: Dry coughing, and the distinct inability to sleep because of so much coughing. Literally, you don't stop for as long as you're lying down. 17:49:28 qFox: I don't know yet. I need to call the doctors tomorrow. 17:49:34 sux :\ 17:49:34 that is like wht I was doing when I smoked 17:49:43 I'm sure that it's treatable via anti-biotics, since it's pneumonia-like. 17:49:44 constant cough 17:49:54 on new years day... you're probably lucky they have time for you.. 17:50:02 snowrichard: Was it a dry, unproductive cough, or was there fluid involved? 17:50:14 lugies sometimes 17:50:20 snowrichard: I have nothing. 17:50:21 brown 17:50:26 It's pure, utter, total, 100% dry. 17:50:34 No amount of coughing produces anything. 17:50:35 * qFox wonders what the death/wounded count is on fireworkrelated injuries this year, and in % compared to last year... 17:50:46 i'd hate to be a paramedic at nye 17:51:38 I suppose they want to do the turnover of iraq back to the iraqis on july 4th just to add insult to injury.. 17:52:10 hehe 17:52:12 just a prediction 17:52:24 that would actually be... evil 17:53:22 I am not really happy with our wonderful government I got injured working for them and they have fucked me over ever since 17:53:35 15 years to get full disability 17:53:54 7 years of living on 72 a month 17:55:01 --- join: randolm (wossname@HSE-QuebecCity-ppp81260.qc.sympatico.ca) joined #forth 17:56:06 hi woss 17:56:12 and they say they have NO RECORD of me having the HEAD INJURY that caused this mess 17:56:12 hi arke 17:56:14 snowrichard, thats disgusting 17:56:25 it seems i just missed a very important section of chat :l 17:57:06 which? 17:57:33 ah, it's just that i pop in after a reboot and you're finishing up a story about a head injury 17:57:48 your line almost looks like something out of fortunes-zippy 17:58:08 my dentist looked at what it has done to my teeth and said I could have died 17:59:37 one of my bottom teeth is crooked and it used to be straight 18:02:28 what is fortunes? 18:03:22 type "fortune" in console, i'm sure you have a copy :0 18:03:29 !fortune 18:03:31 or wait, running windows? 18:03:48 cygwin? :) 18:03:50 not installed 18:03:52 !!! 18:03:54 OMG 18:03:54 OMG 18:03:55 OMG 18:03:59 NO FORTUNE!? 18:04:01 get FORTUNE 18:04:07 life is empty without a cookie now and then 18:04:10 apt-get install fortune 18:04:17 mandrake urpmi 18:04:23 but my mirrors are borked 18:05:09 if you've got a distro cd, i _guarantee_ you have fortune on it 18:05:48 snowrichard: i can't believe you don't have fortune... 18:06:12 root@localhost richard]# urpmi fortune-murphy 18:06:12 Some package requested cannot be installed: 18:06:12 fortune-murphy-1.0-3mdk.noarch (due to unsatisfied fortune-mod) (Y/n) n 18:06:12 [root@localhost richard]# urpmi fortune-mod fortune-murphy 18:06:12 no package named fortune-mod 18:06:13 [root@localhost richard]# 18:06:21 mirrors are fucked 18:06:55 I can do it manually 18:08:26 did net install 18:08:46 --- quit: wossname (Connection timed out) 18:08:54 --- nick: randolm -> wossname 18:10:44 well I have a fortune data file at least. but not the program itself yet 18:10:52 futurama 18:11:18 I can write the program as python script in a few minutes I am sure. 18:11:21 :) 18:11:27 Fortune is pretty simple. 18:11:34 * kc5tja remembers writing one for AmigaOS back in the day. 18:12:18 * kc5tja needs to look into converting his ham radio station into a rack-based system. 18:15:22 gotta go...bye all 18:15:28 --- quit: qFox ("if at first you dont succeed, quit again") 18:15:35 --- part: TheBlueWizard left #forth 18:15:44 computer geeks has some neat mini computer cases styles like stereo equip 18:16:46 snowrichard: Well, I'm looking for something that I can slip into (preferably) a 19" rack system. The rack can also serve as the common ground between all the pieces of gear, including power supply, antenna tuner, and radio itself. 18:18:25 need to reboot to install a printer 18:18:28 later 18:18:31 bye 18:18:34 --- quit: snowrichard ("Leaving") 18:46:18 brb 19:03:52 back 19:18:01 * MysticOne pokes kc5tja with a pork rib 19:18:22 * arke pokes MysticOne with a beef rib 19:18:58 * MysticOne eats both 19:19:07 heh 19:19:13 BEEFCAKE!! 19:19:29 no, rib 19:19:30 not cake 19:19:31 :) 19:19:34 * MysticOne is making ribs on the grill 19:25:03 --- quit: wossname () 19:25:50 * kc5tja is pretending to be able to play along with Pink Floyd's Live at Pompeii. 19:32:24 Heh. 19:32:40 I'm just downloading some videos from there :) 19:32:46 * warpzero is back (gone 03:13:47) 19:34:52 --- join: wossname (wossname@HSE-QuebecCity-ppp81260.qc.sympatico.ca) joined #forth 19:35:38 Damn I suck at this. :) 19:37:06 kc5tja: join THE CHANNEL ... 19:39:42 --- quit: wossname ("keep on haqing in the free world") 20:17:56 I can barely play the tunes in my piano book designed for 5 year olds 20:33:26 Heheh :) 20:33:33 Well, it's a nice distraction for me. 20:33:44 I haven't played with any of my CDs in quite a few years. 20:34:06 My left hand is starting to cramp up a bit though. 20:34:29 Gripping the neck of the guitar for some rhythm chords while playing along with Echoes is quite . . . a challenge. :) 20:36:32 ooh, i get to do massive optimizations now. 20:36:49 replacing rep movsw with some muuuuch faster stuff. 20:37:01 no MMX though, 386 code. 20:40:43 sup kc5tja 20:40:49 Hey what mail client do you guys use? 20:43:35 Currently, K-mail. 20:43:37 re warpzero 20:43:48 Is that KDE app? 20:45:17 Yep. 20:45:30 Wont be using that then. 20:45:36 I intend on ditching it as soon as I write a suitable replacement. 20:45:50 Not that KMail is bad. But it doesn't fit in with the rest of my desktop environment. 20:46:08 What is the rest of your desktop environment? 20:46:19 ROX-Filer and Ion window manager. 20:46:42 OIC. 20:47:03 I use sylpheed, but I'm going to try out evolution. 20:47:05 i use evolution. 20:47:10 evolution is teh shit 20:47:20 hrm, an OT question follows. 20:47:31 As soon as evolution gets done compiling... 20:47:40 I'm trying to create a super-fast 386 memory copy. 20:47:45 here's what I got so far: 20:48:31 rep movsd 20:48:34 mov eax, ecx 20:48:34 shr ecx, 1 20:48:34 and eax, 3 ; EAX is now offset for initial jump. 20:48:34 shl eax, 2 20:48:36 jmp cs:[.jmp+eax] 20:48:41 .jmp: dd .it4 20:48:43 dd .it3 20:48:44 dd .it2 20:48:47 dd .it1 20:48:50 .it1: movsd 20:48:52 dec cx 20:48:56 .it2: movsd 20:48:58 dec cx 20:49:01 .it3: movsd 20:49:02 dec cx 20:49:04 .it4: movsd 20:49:07 dec cx 20:49:08 jnz .it1 20:49:10 eek 20:49:12 sorry, thought that was less lines... 20:49:15 no, something faster than rep movsd even. 20:49:17 the first three lines are supposed to recalculate CX from 2-byte to 4-byte moves, and the remainder in EAX 20:49:29 hrm, those should be ECX 20:50:04 what do shr and shl do? 20:50:15 shift left or right 20:50:22 0 fill through carry 20:50:24 oh okay 20:50:37 dd? 20:50:58 just create a dword (4 bytes) with content. 20:51:11 in this case, the addresses of the iteration locations. 20:51:25 okay 20:51:48 is dec ecx faster than add ecx, -1 ? 20:51:54 prolly 20:52:01 i would assume so 20:52:03 I would guess so. 20:52:26 i'm thinking of doing it on a q-word boundary 20:52:31 * warpzero doesnt understand x86 assembly 20:52:34 such that each iteration would look like 20:52:37 movsd 20:52:39 movsd 20:52:43 sub ecx, 2 20:52:50 but that might not be faster. 20:52:53 actualy, the speed of the above is also dependant on the alignment of the source AND the target 20:52:53 i'm not sure. 20:52:55 how does movsd work? 20:53:04 and your neglecting moves of an odd number of bytes 20:53:18 I440r: how so? wouldn't the remainder take care of it? 20:53:38 movsd moves the dword pointed to by esi to the address pointed to by edi. it advances both pointers 20:54:05 your code will overrun a 5 byte buffer by 3 bytes 20:54:07 so then how is his thing faster than rep movsd? 20:54:20 and btw, it'll always be even, i think.. 20:54:21 it probably isnt 20:54:42 80x25 = 400x5 = 2000 20:54:55 2000*2 bytes = 4000 bytes 20:54:56 and why wouldnt that thing you suggested be addi? 20:54:57 your limiting it to display memory ? 20:55:09 erm, yes, it is for a display routine. 20:55:18 because addi is NOT an x86 instruction :) 20:55:41 arke you will NEED to be able to move WORDS bit DWORDS 20:55:41 :) 20:55:45 Okay. its a MIPS one tho. 20:55:50 in fact im not sure you can move dwords 20:55:54 im assuming a DOS text mode 20:56:04 nope 20:56:07 pmode OS 20:56:10 forth OS :) 20:56:22 well text mode yes ? 20:56:25 it needs to be super duper lighting fast. 20:56:30 yes, text mode. 20:56:31 a single character cell is TWO bytes 20:56:39 one for the character and one for the attribute 20:56:41 but it needs to work with any text mode defined 20:56:52 I wonder what glibc's memcpy() looks like? 20:56:56 but that won't be a problem, as each character is represented by 2 bytes :) 20:57:19 yes but if you move 5 characters thats 10 bytes and 10 does not divide by 4 20:57:35 it gives a remainder of 2 in EAX 20:57:39 so if your move routine used nothing but movsd you would overwrite beyond where your supposed to 20:57:48 so it jumps to .it2 the first time. 20:57:50 you should use movsw 20:57:59 but those are all MOVSD !!!!! 20:58:01 hr,. 20:58:03 oh yeah 20:58:04 Whats movsw? 20:58:04 you're right. 20:58:05 :) 20:58:12 same thing single word? 20:58:14 warpzero: the 2-byte equivalent of movsd 20:58:20 right 20:58:24 besides - most of the time is going to be spent calulating the corrdinates of the block your going to move 20:58:29 erm the addres FROM the corrdinates 20:58:34 arke: What hardware is this for? 20:58:36 you wanna see MY code to do this ? 20:58:37 x86 20:58:48 i guess i might as well just use rep movsw then.. :( 20:58:50 Yeah i got that... what specific hardware? 20:58:57 386-up. 20:59:09 its gotta be 386 compatible :) 20:59:23 All but really old hardware is going to outpace the memory controller anyway. 20:59:39 thats exactly why :) 21:00:09 i think what I'm going to do is use MMX, and just set a flag in the compiled binary. 21:00:16 I hope Robert doesn't mind :) 21:00:20 arke how are you calculating cursor address based on x/y coordinates ? 21:00:20 hrm 21:00:29 the problem here is, i;ve never used mmx ... lol 21:00:39 mmx wont be faster 21:00:45 ok, nvm then 21:00:50 I440r: thats rob's code, lemme see 21:00:52 rep movsd is faster than using mmx and if you use mmx you cant also use fp 21:00:57 Why not: use mov? (single byte) or movsw to do the remainders at both edges and then use aligned rep movsd for the middle? 21:01:12 thats the best way 21:01:14 BUT!!! 21:01:17 thers a danger 21:01:19 Sorry to burst everyone's bubbles here, but rep movs* is one of the slowest instructions to use on x86 architectures. 21:01:31 if the block you are moving is aligned different to the destination 21:01:31 crt_set_cursor: 21:01:31 mov byte[ebx+screen.curx],cl 21:01:31 mov byte[ebx+screcrt_set_cursor: 21:01:32 mov byte[ebx+screen.curx],cl 21:01:34 mov byte[ebx+screen.cury],ch 21:01:36 call crt_set_cursor_physical 21:01:40 ret 21:01:42 en.cury],ch 21:01:44 call crt_set_cursor_physical 21:01:48 ret 21:01:50 oops, pasted twice :( 21:01:52 kc5tja: yes, thats why i wanted to go around it. 21:01:54 i thought MMX would help there... 21:01:55 looks horrible heh 21:01:55 kc5tja: Bubble not bursted. I don't know much about x86. 21:02:03 crt_set_cursor: 21:02:03 mov byte[ebx+screen.curx],cl 21:02:04 mov byte[ebx+screen.cury],ch 21:02:04 no mmx wouldnt help 21:02:06 call crt_set_cursor_physical 21:02:10 ret 21:02:12 here thats it. 21:02:21 Since MMX uses the floating point registers, you're better off just using the floating point load/store instructions. 21:02:24 I440r: how come? its got that cache-go around instruction too, and can stream at 128 bits. 21:02:25 theres where you are goinbg to be blowing MOST of your time 21:02:53 kc5tja: thats a move by intel i never will understand... 21:02:55 successive loads to and stores from a single fpu reg is fast 21:03:11 but i BELIEVE that on some pentiums its faster to just use rep movsw 21:03:26 We could always just use C. 21:03:30 on a 386 rep movs* is very fast. 21:03:34 * I440r thwaps warpzero 21:03:36 heh 21:03:42 but 486 introduced pipelining, making it the slower approach 21:03:58 warpzero: forth, if not assembly! 21:04:12 I440r: but I do believe Robert wants to stay 386 compatible 21:04:42 if he wasnt asleep, i'd ask him... 21:05:06 * warpzero is going to need to learn both ppc and x86 assembly at some point. 21:05:34 I'm not porting my shit to anything else. :P 21:06:00 rep movsw will be your best choice 21:06:08 :(( 21:06:18 How does rep work? 21:06:20 because it will move single characters without your needing to worry about alignment issues on the source or destination 21:06:20 (not to doubt you, I440r, but..) 21:06:25 can I have a second opinion on that? 21:06:48 rep repeats SOME instructions by the number in the ecx reg 21:06:49 * arke reads AMD optimization manual 21:07:02 Okay. 21:07:04 so if ecx is 10 then rep movsw executes TEN movsw's 21:07:14 or rather it executes THAT movsw 10 times heh 21:07:25 its like an auto loop back onto the same instruciton 21:07:35 but it only works with string instructions. 21:07:42 which makes it very inflexible :( 21:07:43 you cant do stupidness like rep mov eax, [blah] 21:07:44 tho 21:07:52 which sucks! 21:07:54 it only works on string instructions i believe 21:07:59 yeah 21:08:02 only string instructions. 21:08:04 REP only works on string instructions. 21:08:24 kc5tja: you think as well that rep movsw will be my best choice here? 21:08:38 what are your goals? Maximum performance or minimum size? 21:08:44 will an explicit loop be faster? 21:08:52 no 21:08:54 er, more performance. 21:08:54 An explicit loop will always be faster. 21:09:07 blah: 21:09:09 movsd 21:09:11 well, it wont on a 386, but i think it would on 486+ 21:09:12 loop blah 21:09:15 is slower than 21:09:17 I440r vs. kc5tja ... ROUND 1 ... FIGHT 21:09:17 loop is bad!!! 21:09:17 rep movsw 21:09:24 The problem with REP MOVSW is that it sits in ONE pipeline of the computer, which effectively seizes the remaining pipes -- nothing else can execute while the MOVSx instruction is running. 21:09:31 yes, but they're both slower than 21:09:35 blah: 21:09:38 movsd 21:09:40 dec cx 21:09:42 jz blah 21:09:48 Isn't there just a memory range copy instruction? 21:09:50 Depends. 21:09:55 warpzero: That's movsw 21:10:03 But it's faster to just do the loop explicitly. 21:10:04 arke that dec ecx jNZ blah is only faster on SOME verions of the pentium 21:10:05 but it only moves 1 word? 21:10:10 you should use LOOP 21:10:11 (i think at least.. 21:10:16 Because then multiple instructions execute in multiple pipelines, and it can be better parallelized by the hardware. 21:10:21 loop clogs the pipeline as well. 21:10:34 warpzero: REP MOVSx where x is either B=Byte, W=Word, D=Double-word 21:10:37 thats it 21:10:48 i'm using teh nasm preprocessor. 21:10:57 to do optimization levels. 21:10:57 :) 21:11:08 arke: I encourage you to try different approaches, and profile them. Find out empirically which one is faster. 21:11:16 I think you'll find explicit loops are faster. 21:11:29 he can only do that on the specific processor he is testing on 21:11:34 Yeah I think that doing profiles is a DAMN good idea in this case. 21:11:37 unless I use bochs 21:11:38 :) 21:11:41 different methods have different results on different processors 21:12:04 like - run the same code on a p4 and an athlon... 21:12:16 different optimizations work on those 21:12:29 I440r: Well, since both AMD *and* Intel say, "Thou shalt not use REP MOVSx, for it is slow and will make you sterile," I think it's a pretty safe bet that the profiling will result in benefits across the board. 21:12:32 things that are optimized for one CAN be negativly optimized for th eother 21:12:55 When I write my native-code compiler I'm just going to have it have a set of possible options and profile everything. 21:13:17 actually, for a TEXT mode who cares lol 21:13:20 Stick them all in loops, see which is faster, continue. 21:13:32 now - if this was a graphics mode it would make a difference heh 21:13:34 I440r: lol 21:13:42 I440r: well, see, theres the problem. 21:13:47 your going to be moving TENS of characters! 21:13:49 lol 21:13:53 I440r: I (we?) do plan to implement graphics eventually 21:14:07 I440r: actually, exactly 2000 of them in text mode. 21:14:17 You're only going to be moving text data around? 21:14:17 which makes 4000 bytes. 21:14:23 How often will this be done? 21:14:26 most of the time your not going to be moving that many in one go 21:14:29 kc5tja: for now, yes 21:14:38 screen scroll... 21:14:48 kc5tja: well, eventually, only when needed. 21:14:54 arke: I don't care if it's "for now" or not. Premature optimization will ruin you. It's a bad habit. Stop it. 21:15:02 kc5tja: Ok. 21:15:02 Make software that works, now. Solve the problem you have, now. 21:15:06 kc5tja: point taken. 21:15:20 i think its fun to optimize but thats its ONLY merit 21:15:25 If, and only if, things get slow in actual practice, THEN you optimize. 21:15:32 ugh, I don't want to, but if I'm gonna do what I plan to do, I'm gonna have to throw out lots of code :( 21:15:46 mostly optimize your METHODS, not your instrucitons 21:15:50 arke: Well, what are you planning to do? 21:15:55 You haven't told us anything yet. 21:16:01 kc5tja: I'm helping Robert with his project. 21:16:03 :) 21:16:09 robert.zizi.org/f.html 21:16:24 its a DTC right now, and its gonna be STC. 21:16:34 but he has produced lots of valuable code. 21:16:46 (but also stolen lots from his previous projects, lol) 21:19:18 Personally, and this isn't necessarily the case with everyone, I have found this observation to hold true: all the time I spend trying to figure out how to make a piece of code general is greater than the sum of the time it takes me to write the original code, chuck it, then re-write it from scratch when I have to. 21:20:12 Interestingly enough, in extreme programming, this is codified (albeit in a strange form): if you write code more than twice, factor it. 21:20:42 Basically, you won't know what to factor (e.g., what to generalize) until you have side-by-side comparisons to perform a mental `diff' with. 21:21:09 which is why ALOT of the design phase is BULLSHIT. 21:21:13 not all 21:21:14 but alot 21:21:31 your design phase is your initial "best guess" 21:21:44 Well, design phase is good in that it gives you a target to strive for. 21:21:48 you should code first, design secnd, code again :) 21:22:04 You have to have it -- blindly coding without a set goal in mind will result in failure every time. 21:22:06 :) 21:22:08 yes. but it should be very very general with NO specifics in it what so ever 21:22:18 kc5 well ya lol 21:22:19 No, it should be as specific as possible. 21:22:19 but i do feel bad 21:22:27 because robert has put ALOT of code and time into it. 21:22:27 im gona code something.. dunno what yet but.... heheh 21:22:37 and i'm most likely only gonna be using 25-30% of it. 21:23:11 kc5 the more specifics you have in the design the more you paint yourself into a corner - unless your allowed to RE-design later 21:23:27 i.e. some designs are cast in stone and are thus DOOMED to failure 21:23:39 or at least mediocrity 21:23:52 I think you're confusing customer requirements with software architecture. The two are wholesale different things. 21:24:22 The customer rarely has a need to dictate architecture (unless, obviously, part of HIS product depends on the architecture being a certain way). 21:24:37 kc5 no - im talking about where a company does a design, writes the specification down, hires a coder who now has to follow that totally useless design without any deviation what so ever 21:24:49 you "must" do it this way because thats what the design says 21:24:54 No, you're still confusing the two concepts. 21:25:04 If they write the design, why don't they just code it themselves? 21:25:15 err who knows 21:25:24 They had to hire a programmer for a reason: that reason is they have zero clue about how to architect a software package. 21:25:40 who cares lol - the point is they have a design and it MUST BE followed. down to the most minute of details 21:25:46 you are NOT allowed to change it 21:26:06 nevermind that when you actually go to code it you find part X of the design to be totally incompatible with part Y of the design 21:26:11 Details like, "The window must be XxY pixels in size, must have a title that reads FOO, et. al." are hardly something to be worried about. 21:26:14 which they had no way of knowing 21:31:27 I440r: I think what kc5tja is telling you is that the design of the program, the way it looks, is completely different from the way it's actually built 21:31:41 I440r: if they went so far as to define exactly how the program works, then they essentially wrote the program 21:32:54 yes 21:32:57 SHOULD be 21:39:34 brb 21:39:36 --- quit: I440r ("Leaving") 21:53:38 --- join: I440r (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth 22:07:53 happy new year from texas 22:08:10 Tejas! 22:08:36 I should probably go to bed soon. 22:08:39 I have to work. 22:09:05 what time do you guys open? 22:09:26 We open at 10:30AM, but I am scheduled for 11. 22:09:50 ahh 22:16:51 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 22:16:56 hey 22:17:07 what time is it? 22:17:22 (central US) 22:18:02 00:17 22:18:49 ok just reinstalled mandrake onto a 40G hd and it was set to eastern 22:21:18 is isforth yours? 22:21:33 yes 22:22:07 ok I want to understand how it works, and the way I do that is document everything to death so when I get the web page of my notes done you want a CD of the web site? 22:22:24 hell yes :) 22:22:33 volunteer of course 22:22:46 are you a good tech writer ? 22:22:51 fair 22:22:54 i can help you understand the kernel 22:23:07 and we could come up with some documentation for it in no time flat 22:23:54 I am just rusty on x86 assembly because I had never owned a PC and had time to learn it. 22:24:06 read a book about 8086 mode years ago 22:24:07 x86 assembly is easy 22:24:21 and i have LOTS of comments heh 22:24:24 well I can d/l the IA32 manuals form intel 22:25:00 go to the source for actual docs rather than guessing :) 22:26:02 actually the ia32 docs from intel go on and on and on and on and on... 22:26:07 test server name is richardsnow.is-a-geek.net 22:26:07 worse than the damned battery bunny 22:26:16 yeah over kill 22:26:21 3 volumes 22:26:21 best thing you can do to learn x86 is get the 8086 book from intel 22:26:34 is free? 22:26:43 limited money 22:28:18 see eventually I would like to write my own os kernel and want to base the thing on forth 22:28:35 or use the OSKIT one 22:30:02 need to reboot to get the damn clock set right I think 22:30:23 well you MIGHT be able to find a second hand copy of it in a book store 22:30:24 --- quit: snowrichard (Read error: 104 (Connection reset by peer)) 22:30:30 well you MIGHT be able to find a second hand copy of it in a book store 22:30:32 but dont count on it 22:30:42 oopts he left heh 22:31:59 --- join: snowrichard (~richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 22:32:05 you were saying? 22:32:16 now it is app 32 minutes into 2004 22:32:17 heh 22:32:30 Well, I'm getting tird. 22:32:32 tired even 22:32:34 me too 22:32:35 * kc5tja is going to bed. 22:32:39 i was saying you might be abel to get a second hand copy of th3e book 22:32:42 night kc 22:32:50 its the 8086 book, includes the 8088 22:33:01 any assembler book after that isnt worth a shit 22:33:08 think I had one a long time ago 22:33:08 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 22:33:14 erm any INTEL assembler book after that isnt wortha shit 22:33:18 its blue heh 22:33:25 no 22:33:35 not that one 22:34:16 but I have programmed 6502, 6809, Z80, IBM360, 68K so one more won't be hard 22:34:57 but it really is bedtime 22:35:27 --- nick: snowrichard -> zzzsnowzzz 22:36:01 :) 22:36:03 night 22:36:15 but the damn fireworks still popping 22:36:30 sound like ww3 22:36:53 I need coffee :) 22:39:14 i got some :) 22:39:23 in fact i got a whole pot of really good coffee 22:39:35 your just north of austin 22:39:39 im just east of dalas 22:39:52 actually I am in shreveport LA 22:39:59 im almost twice as far east of dalas as you are north of austin 22:40:12 east of dallas north orleans 22:40:13 ok LA is in california :P 22:40:19 Louisana 22:40:23 shreveport 22:40:34 I 20 and I 49 22:40:44 ooh 22:40:56 is about a mile from here 22:41:02 i was talking to someone else and got my wires crossed heh 22:41:14 your like 70 ish miles from me 22:41:40 really? 22:41:47 yes 22:41:53 73 i think is the distance 22:41:56 im in longview texas 22:42:09 OH damn I used to live in Marshall 22:42:14 heh 22:42:19 thats like spitting distance from here 22:42:21 I could have taken you to dinner 22:42:28 ya 22:42:45 would have prefered someone sexier tho :) 22:42:46 hehe 22:42:54 not like a date :) 22:42:58 lol heh 22:42:59 just to meet 22:43:02 yup 22:43:15 I visit my mom in Marshall quite a bit 22:43:16 i was just makin funny with it is all lol 22:43:19 cool 22:43:24 wish i could vizzit my moim 22:43:25 mom 22:43:28 she is in england:/ 22:43:32 oh 22:43:40 my grandma lives inBig Sandy 22:43:45 do you work tomorrow ? 22:43:48 91 yrs 22:43:49 wheres big sandy ? 22:43:55 wow thats cool! 22:44:00 just north gladwater 22:44:07 west I mean 22:44:08 in england if they make it to 100 they get a letter off the queen :) 22:44:22 I no work 22:44:27 i wanna stay in this area when this contract ends 22:44:30 disabled vet lots of time to write docs 22:44:35 aha 22:44:36 :)) 22:44:55 get paid to work at nothing all day :) 22:44:56 when this contract ends (possibly in a week) ill have lots too :/ 22:45:04 how much per hour ? 22:45:09 2308 a month 22:45:15 thats not too bad 22:45:23 i could live and save off that 22:45:34 well I got debts out the yazoo 22:45:38 except i cant now, i owe my dad alot of money 22:45:55 my coffee shoudl be done now brb 22:45:59 ive not worked in two years till this contract, 22:48:51 back.. only snack I could find in fridge was cake icing. have an icing sandwich 22:49:23 lol 22:49:30 i have mayo sw sometimes :)P 22:49:51 my blood sugar get low and I get nutty 22:50:54 I now have a 40G hd installed but I am supposed to give it back to the guy that runs this place. I offered to buy it from him. I wonder if he could tell the diff between the 40 G WD and a 6 G seagate? I bet not :) 22:51:35 lol 22:51:45 I have the seagate in the bag the WD was in with a win95 FAT32 part. tab. on it. 22:53:16 he is a psychologist not a compgeek 22:54:19 but your an honest guy right ??? :) 22:54:52 yes I will give it back just would be nice to use it. 22:55:08 ya 22:55:21 but they are only $50 bucks at compgeeks 22:55:22 i just got rid of some 1 gig drives, dismantled em 22:55:39 $85 for and 80 22:55:41 if i had any spare drives i would donate one 22:55:47 yes its about a buck a gig now 22:55:58 well I have a 6 and and a 3 22:56:19 and I have money in bank 22:56:41 think I'll check Ebay for wireless router 22:56:44 i just got a 200 gig drive heh 22:56:46 and laptop 22:56:52 i just got one of those too. a linksys 22:56:59 which model? 22:57:04 work with linux ok? 22:57:30 not sure of model - its a wireless g 2.4 ghz switch/router 22:57:43 yea linux doesnt need t5o know shit about it 22:57:56 its external 22:57:57 well yeah plumbing 22:58:07 rigth 22:58:12 a fat pipe :) 22:58:16 now - the wireless card in the laptop.. thats a problem 22:58:31 well im on cable here - its not dial up but its not exactly FAT either heh 22:58:42 the wireless hub works on the 2.4 ghz range 22:58:47 you want an OC3 22:58:56 hell yes! 22:59:00 you got one you can give me ? 22:59:02 heh 22:59:04 fuck no 22:59:15 I am on dsl 22:59:19 i have a copy of GR-CORE 253 :) 22:59:31 synchronous optical network specificaion 23:02:08 but really maybe in march we could meet and talk in person when I take a weekend trip. 23:02:23 318 934 1664 23:03:44 whussat ? 23:03:50 i prolly wont be here in march 23:03:55 what is what? feb? 23:03:57 very very unlikely that i will 23:04:02 no that number lol 23:04:10 phone 23:04:14 i live in indiana 23:04:21 im only here ona contract job and its ending real soon 23:04:30 I aint driving to indiana 23:04:33 lol 23:04:39 its only 15 hours from here! 23:04:54 takes 23 hours to drive to NoFuck VA 23:04:55 drove it for tnx giving and for xmas 23:06:16 my coffee cup empty 23:06:55 nofuck lol 23:07:15 btw, what is the nature of your disability if its ok to ask ? 23:08:43 in the nature of a major concussion caused me to have schizophrenic like symptoms 23:09:27 aha 23:10:43 and I have to now take Zyprexa and Zoloft 23:11:56 I was on a nuclear powered Aircraft Carrier and they did a flank reverse and then a hard right rudder. Knocked me into a metal locker and hit my head. 23:12:40 out cold, amnesia, relearn almost everything ... 23:12:54 clear core and start over :) 23:13:02 wow 23:13:14 had to relearn EVERYTHING ? 23:13:19 I was literally :born again: 23:13:25 wow 23:13:32 not eeverything, just forgot details like how to do my job 23:13:34 couldnt talk ? 23:13:38 understand english ? 23:13:43 for a while I could not talk 23:13:48 ! 23:13:49 but that was a drug interaction 23:13:55 ok 23:14:07 contact/thiothixeme 23:14:08 my sis just joined the navy 23:14:16 damn fw 23:14:21 fw ? 23:14:30 I had to learn how to play chess all over again 23:14:37 fire works 23:14:39 i never was very good at that 23:15:41 your sister in the Navy? cool wish her luck. 23:15:52 yea 23:16:00 she was gona go in as a nuke 23:16:20 but she decided not to because even if she trained as such theywere NO guarantee what so ever she woluld get that job 23:16:24 not unless they needed her 23:16:35 I could have been any rate I wanted except I am color blind 23:16:36 and if they decided they didnt they could put her ANYWHERE 23:16:54 so she decided to be a fire fighter instead 23:17:06 so DP is what I wanted anyway. I had to take the fire fighter training 23:17:12 once 23:17:18 a 2 day course 23:17:51 classroom and then a HOT lab! 23:18:02 buiding that looks like it on fire 23:18:17 looks like ? 23:18:32 has like burning oil and stuff in it but is contained 23:18:45 so it doesn't really burn the building 23:19:09 right 23:19:35 but make it hot :) 23:19:50 and a tear gas chamber 23:20:58 :) 23:21:19 to teach the proper use of breathing mask :) 23:21:35 hold breath :P 23:21:39 *snicker* 23:22:03 --- nick: zzzsnowzzz -> ZyprexasnowZolof 23:22:35 * ZyprexasnowZolof bored and damn fire works still exploding outside 23:23:06 --- nick: ZyprexasnowZolof -> snowrichard 23:23:14 if im tired i can still sleep with that going 23:23:28 but no matter HOW tired i am i cant sleep with a tv/radio going 23:23:36 in fact my sisters fish tank keeps me awake 23:23:43 sometimes I can't sleep unless I have radion going 23:24:03 I have tinnitis (ringing in ear) and music covers it up 23:24:17 too close to jets and 400 HZ motor generator 23:24:26 A flat forever :) 23:24:36 eeeeeeeeeeeeeeeeeeeeee 23:24:59 lol 23:25:09 you can use that to tune a guitar :) 23:25:14 this is anEBS test... 23:26:20 --- quit: snowrichard ("Leaving") 23:39:40 * arke is away: be back in about 15 minutes tops 23:39:49 --- quit: arke (Remote closed the connection) 23:53:03 --- part: ayrnieu left #forth 23:54:44 --- quit: cmeme (Remote closed the connection) 23:56:36 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 23:58:50 --- quit: cmeme (Read error: 104 (Connection reset by peer)) 23:59:18 --- join: cmeme (~cmeme@216.184.11.30.unused.swcp.com) joined #forth 23:59:59 --- log: ended forth/03.12.31