00:00:00 --- log: started forth/03.12.15 00:28:44 --- join: schihei (~schihei@blueice1x.de.ibm.com) joined #forth 01:10:15 --- join: Robert (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 01:37:35 --- join: njd (~junk@njd.paradise.net.nz) joined #forth 01:46:19 Hi 02:24:42 --- quit: njd ("Client exiting") 02:45:26 --- join: giuppi (~giuppi@socks1.fastwebnet.it) joined #forth 02:57:35 --- quit: giuppi (Remote closed the connection) 05:08:10 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 05:41:46 --- quit: I440r ("work is thataway -->") 06:24:30 --- join: I440r (~mark4@saturn.vcsd.com) joined #forth 06:47:52 --- join: rO| (rO_@p508077EE.dip.t-dialin.net) joined #forth 06:48:38 * rO| greets 06:50:23 would some1 from .us be so kind to give me "his" www.google.com + news.google.com ip-address? 06:55:10 ok, just let me install bind first :) 06:56:40 thx. 06:57:25 I440r: still got that job btw? 06:58:53 ; <<>> DiG 9.2.2 <<>> www.google.com 06:58:54 ;; global options: printcmd 06:58:54 ;; Got answer: 06:58:54 ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 60020 06:58:54 ;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0 06:59:00 ;; QUESTION SECTION: 06:59:02 ;www.google.com. IN A 06:59:06 ;; ANSWER SECTION: 06:59:08 www.google.com. 2859 IN CNAME www.google.akadns.net. 06:59:12 www.google.akadns.net. 114 IN A 216.239.41.99 06:59:18 ;; Query time: 20 msec 06:59:20 ;; SERVER: 192.168.128.13#53(192.168.128.13) 06:59:22 ;; WHEN: Sun Dec 14 17:02:00 2003 06:59:26 ;; MSG SIZE rcvd: 83 06:59:30 contract is going to be coming to a close soon 06:59:42 did all of that paste make it ? 06:59:48 wow. I had forgotton how verbose dig is. 07:00:01 lol 07:01:18 ; ANSWER SECTION: 07:01:18 news.google.com. 300 IN A 216.239.51.104 07:01:23 thers news. :) 07:01:28 I440r: ok, now i know i will need an us proxy :-( thx anyway 08:10:11 hmm...I'd have to say Jeff has been going a bit over the top on c.l.f. lately... 08:11:50 --- quit: schihei (Client Quit) 08:25:52 jeff fox ? 08:26:43 yah 08:27:36 I mean, I don't follow clf, so I don't know all what provocation he has been given, but... 08:28:53 well jeff is a radical forther for sure but i trust HIS views more than i trust the ans crowd 08:29:23 well yeah, definitely. 08:29:37 but his message from yesterday titled "FHG"...? 08:29:54 --- join: aktnot (ident@181.80-202-66.nextgentel.com) joined #forth 08:30:43 --- quit: aktnot (Client Quit) 08:30:46 I mean, he _has_ to know there are people on there who aren't going think it's funny. 08:32:59 dunno, just seems like his posts from the last couple days have little to no useful forth content and are just aimed at provoking flames. 08:40:03 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth 08:40:14 I440r: Are you at keyboard? 08:40:21 --- mode: ChanServ set +o kc5tja 08:40:37 I440r: Are you at keyboard? 08:40:55 he was around a few minutes ago... 08:40:57 I'm fairly confused as to what the problem is with my ELF header at the moment. 08:41:14 I have a proper string section, I have proper program header sections, and I have a proper set of section headers. 08:41:36 Yet, running my a.out file will produce only a stack segment; no .text image is loaded into the process space. 08:42:11 Well, I have to get ready for work, but I'll idle here in case he responds. 08:42:27 * kc5tja was kind of hoping to get a moderately quick response. 08:42:52 (note I got my info from interpreting /proc/*/maps) 08:43:10 objdump can see all the sections and program headers perfectly. 08:43:29 As far as binutils are concerned, it's a well-formed, perfectly valid ELF executable. 08:43:55 hmm, sounds like Linux is much more picky about ELFs on x86 08:43:57 --- join: hovil (~matt@CommSecureAustPtyLtd.sb1.optus.net.au) joined #forth 08:44:20 I would imagine that'd extend towards other architectures, not just x86. 08:44:31 well, I just looked at my simple ELF header 08:44:33 But yeah, it's clearly not compliant to the complete ELF standards. 08:44:47 I have no section headers, and no string table 08:44:58 Just the ELF file header and one program header 08:45:15 I have to include the section headers and string table to get meaningful output from binutils. 08:45:22 They're ignored by the Linux loader. 08:45:31 But I have one program header too, and it is utterly ignored. 08:45:34 weird 08:45:44 Do we need a dummy program header as the first header? 08:46:23 not according to the standard... 08:46:55 Program Header: 08:46:55 LOAD off 0x00000034 vaddr 0x08000000 paddr 0x08000000 align 2**2 08:46:55 filesz 0x00000015 memsz 0x00004000 flags rwx 08:46:59 That's what objdump produces. 08:48:10 The only differences I can see are that the code alignment is not on a page boundary (nor is its alignment set to be 2**12), and it's a writable code segment. 08:48:40 Program Header: 08:48:41 LOAD off 0x00000054 vaddr 0x10000054 paddr 0x00000000 align 2**16 08:48:41 filesz 0x0000000c memsz 0x00100000 flags rwx 08:48:44 I440r says the 'w' flag is ignored by Linux for loadable segments, so I didn't muck with that. 08:48:49 that's mine... 08:49:26 Hmm... Let me try hacking it to be page aligned... 08:49:30 See if that helps. 08:50:33 one thing...Linux on PPC pretty much just loads it into memory and then executes it, so the code has to start at an address past where the headers are loaded in memory. 08:50:48 hence "off 0x00000054 vaddr 0x10000054" 08:54:13 but I think that's only because the PPC ABI supplement says that all sections have to be aligned in the file the same way as in memory (mod 64KB) grr... 08:54:31 So in other words, Linux thoroughly disregards the ELF standards. 08:54:38 I noticed it does this on x86 too. 08:54:56 PPC seems to more or less follow them. 08:54:58 mod 4K for x86. 08:55:13 But there are a couple of things like that which you have to know which aren't in the general SysV standard 08:55:21 only in the appropriate processor supplement 08:55:28 (which took me forever to find). 08:55:30 The ELF spec says that each section/segment has its own virtual address, starting at the first byte of the section IN THE FILE, not in mmap()'ed memory. 08:57:26 I can't imagine why they'd do something so freakin' stupid. 08:57:42 OK, I have a pretty interesting challenge for me then. 08:57:44 makes paging simpler 09:00:08 "...delaying physical reads frequently obviates them, improving system performance. To obtain this efficiency in practice, executable and shared object files must have segment images whose offsets and virtual addresses are congruent, modulo the page size." 09:01:12 OK, that's what did it -- the virtual address and file offset addresses. 09:01:17 My code now works! :) 09:01:23 kc cool :) 09:01:32 * I440r was tracking this conversation hehe 09:01:34 :) 09:01:39 <-- hardly working! 09:01:53 I440r: No, that is more properly called `supervising.' :) 09:02:02 ooh yea :) 09:02:18 see i knew if i supervised you would get it fixed ;) 09:02:29 * I440r steals the credit lol 09:02:34 * kc5tja notes that the section/segment offsets can still be file-related offsets too; it's just that virtual addresses have to be set accordingly too. 09:02:52 : foo [entry-point] foo ; 09:02:59 properly hangs the process. :D 09:03:41 My next step is to get it to print "Hello World" to the screen. 09:03:44 I'll try to do that after I get back from work. 09:04:35 --- join: Serg_Penguin (~z@212.34.52.140) joined #forth 09:05:05 Thanks guys. OK, I'm off to work now! 09:05:13 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 09:05:52 hi 09:06:19 * Serg_Penguin is restoring WinXP what fell miserably on secretary comp ;(( 09:06:51 serg i released new isforth WITH kernel.com included 09:07:09 yeahh ! that rulez ! 09:07:29 and isforth works in fbsd if you have linux emulation set up right 09:07:45 but you prolly have to brandelf -f 03 kernelcom 09:08:07 * Serg_Penguin boots to Linux so rare what i even forgot r00t passwd and set it blanc to avoid lotsa damns 09:08:18 lol 09:08:48 * Serg_Penguin is getting a life ;)) 09:09:01 and a gf ? 09:09:06 i even busted all da gamezz out of the box ;)) 09:09:07 why would you want to go and do something crazy like that for 09:09:19 lol 09:09:39 no gf by now ;)) 09:09:53 * I440r is working on it 09:09:59 ;)) 09:10:05 me 'll work soon too 09:10:06 * I440r has fallen in lust 4 or 5 times since starting this contract 09:10:15 lust ?? 09:10:19 lol 09:10:21 Texas honeys. 09:10:26 use dictionary :) 09:10:29 yea 09:10:34 drool! 09:10:35 Ride the wild bronc. 09:12:05 ;)) 09:12:48 i ditched one for trying to 'milk' me for money... 09:13:10 and another for teaching me how to live too much ;)))) 09:13:31 Serg_Penguin, I just made it so I had no money to milk 09:13:37 that seemed to get rid of all those types 09:14:34 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 09:14:40 i decided to re-tread one TV adv star ;)) 09:16:26 i'll start from a nice gift: big skipper's smoking pipe (she smokes) 10:00:27 --- quit: Serg_Penguin () 10:03:33 --- join: schihei (~schihei@pD9E5C566.dip.t-dialin.net) joined #forth 10:27:42 --- quit: tgunr (Read error: 104 (Connection reset by peer)) 11:05:39 --- quit: rO| () 11:28:10 hey 12:04:06 --- join: rO| (rO_@p508077EE.dip.t-dialin.net) joined #forth 12:05:26 hey guys 12:08:17 hi cykotic 12:40:16 hah. caldera.com is back up. 12:40:45 * tathi wishes people wouldn't DOS them when I've lost my copy of the SysV ABI. 12:51:26 --- quit: rO| () 13:03:29 --- quit: tathi ("leaving") 13:12:53 --- quit: hovil ("Leaving") 13:19:49 --- quit: schihei (Client Quit) 14:12:47 --- join: fridge (~fridge@dsl-203-33-164-75.NSW.netspace.net.au) joined #forth 15:20:18 --- quit: I440r ("bbl") 15:57:21 --- join: I440r (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth 17:05:06 how do you read from the mic port on linux? just cat < /dev/dsp? 17:05:33 Herkamire: umm... no 17:05:35 Herkamire: it depends 17:05:45 Herkamire: the older sound drivers would let you cat /dev/audio or something to a file 17:05:49 what sort of device names should I look for? 17:06:00 but the newer ones don't do that ... newer ones require some sort of program to actually record from the device 17:07:15 newer soundcards are basically winmodem style - they arent HARDWARE, they are SOFWARE 17:07:21 thats what ac97 BS is 17:07:33 no, it's not that 17:07:45 it's just that the drivers are different now 17:07:54 they don't do that raw interface type deal anymore due to more advanced features of a lot of cards 17:08:06 i.e., you used to be able to cat a wav file to /dev/sound or something 17:08:07 and it'd work 17:08:10 not anymore though 17:09:21 so I should look for some software to record? 17:10:12 audacity 17:10:14 or sox 17:10:17 sox for commandline 17:10:19 audacity for X 17:11:06 got a guy running isforth with a monochrome monitor - gets blinking text etc etc - not sure how to handle that 17:11:23 thats a pain 17:11:25 thanks 17:14:07 i love monochrome displays. :) 17:14:30 does isforth require color? 17:17:53 --- quit: fridge (orwell.freenode.net irc.freenode.net) 17:17:54 --- quit: zardon (orwell.freenode.net irc.freenode.net) 17:17:54 --- quit: MysticOne (orwell.freenode.net irc.freenode.net) 17:17:54 --- quit: TreyB (orwell.freenode.net irc.freenode.net) 17:18:00 --- quit: I440r (Remote closed the connection) 17:18:01 --- join: I440r_ (~mark4@12-160.lctv-a5.cablelynx.com) joined #forth 17:18:18 --- join: fridge (~fridge@dsl-203-33-164-75.NSW.netspace.net.au) joined #forth 17:18:18 --- join: zardon (~zardon@d216-232-233-5.bchsia.telus.net) joined #forth 17:18:18 --- join: TreyB (~trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 17:18:18 --- join: MysticOne (mysticone@mysticone.usercloak.freenode) joined #forth 17:26:18 sox: Failed reading /dev/dsp: Unable to set the sample size to 16 17:26:29 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 17:28:08 sox is just reading /dev/dsp 17:28:12 I can do that myself. 17:28:35 problem is that the only way I can get that to read from the mic is if I set the mixer to play the mic over my speakers, and I don't want that 17:29:01 --- quit: TreyB (orwell.freenode.net irc.freenode.net) 17:29:01 --- quit: MysticOne (orwell.freenode.net irc.freenode.net) 17:29:02 --- quit: fridge (orwell.freenode.net irc.freenode.net) 17:29:03 --- quit: zardon (orwell.freenode.net irc.freenode.net) 17:29:42 I was wondering if there could be a seperate device for the mic so I can record it without hearing it. 17:29:59 --- join: fridge (~fridge@dsl-203-33-164-75.NSW.netspace.net.au) joined #forth 17:29:59 --- join: zardon (~zardon@d216-232-233-5.bchsia.telus.net) joined #forth 17:29:59 --- join: MysticOne (mysticone@mysticone.usercloak.freenode) joined #forth 17:29:59 --- join: TreyB (~trey@cpe-66-87-192-27.tx.sprintbbd.net) joined #forth 17:30:09 I was wondering if there could be a seperate device for the mic so I can record it without hearing it. 17:34:06 http://cgi.tripod.com/lvgandhi/cgi-bin/addfile.pl?lnxmmhnt.txt 17:34:09 search for dsp 17:34:48 dunno if that helps 17:34:53 hehe 17:43:17 recording /dev/dsp only gets the mic if I set (in the mixer) the mic to be played over the speakers 17:43:26 what is the disadvantage to just writing everything in machine forth? 17:43:40 sorry. let me give that some context. 17:44:09 what is the disadvantage to writing a forth environment for/in machine forth and then rolling some 100 byte emulator for MF 17:44:50 speed 17:45:10 yea. i don't believe in speed. 17:46:45 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576567.sympatico.ca) joined #forth 17:47:00 Herkamire: huh? what do you mean set it to be played over the speakers? 17:47:29 --- quit: madgarden (Client Quit) 17:47:37 simplicity is something i feel that is important though. 17:47:44 the only way I was able to record from my mic was to go into aumix, and set the Mic line to P (for play) 17:47:49 this makes it play on my speakers 17:48:16 Herkamire: I think that sets it to record 17:48:24 Herkamire: you'll hear yourself over your speakers 17:48:31 Herkamire: you can turn down the speakers, though, to make it so you don't hear yourself 17:48:37 the options are R for recorod, and P for play. 17:48:42 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576567.sympatico.ca) joined #forth 17:48:43 maybe MF is the right choice for me. but i still have this gut instinct that i don't like it. and i'm not sure why. 17:48:45 it only records (and plays on the speakers) on P 17:49:07 --- quit: madgarden (Client Quit) 17:49:10 maybe because MF really doesn't solve anything. 17:49:14 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576567.sympatico.ca) joined #forth 17:49:17 --- quit: madgarden (Remote closed the connection) 17:49:24 just shifts the problem around. 17:49:26 * OrngeTide shrugs 17:49:29 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3576567.sympatico.ca) joined #forth 17:49:42 brb 17:49:48 Herkamire: what kind of sound card? 17:49:51 burbleflies 17:50:39 zardon: Heh, I actually managed to watch all of Zardoz last night. 17:52:17 PowerMac Screamer DMA sound driver rev 016 : 17:52:59 I've hooked a radio to my mic port and I would like to be able to record shows without hearing them 17:53:26 because I would like to have the computer automatically start recording at the right time, and I don't want to hear the radio whenever my speakers are on 17:56:16 madgarden: heh, cool. was it on space? 17:56:55 OrngeTide: I think there is value in having a simple forth environment. I don't knom machine forth but I think it fits the description 17:57:06 Yep. Heh, I love that channel. I went to the website just to watch those freaky little mutant baby movies. 17:57:41 those Ryerson shorts? 17:58:06 Herkamire: you don't know machine forth? it's just the basic instruction set for MuP21 or F21, usually tweaked slightly to make it less work to use. (it's not an emulator/simulator) 17:58:16 Ryerson? I dunno... they are usually shown just before a new show starts. They're rather weird. 17:58:26 yeah 17:58:30 film students 17:58:39 OrngeTide: ok, I am somewhat farmilliar with the instruction set of the 25x thing 17:58:41 You sure? They all have the same character in them. 17:58:52 so you define like 24-30 opcodes and then your entire system can be done, without the use of any embedded machine-code in your words. 17:59:38 i wish i could find this damn link. someone wrote a machine forth that fit into an x86 boot sector. 18:00:00 it didn't do much. but it was a small asm MF and a bit of forth. kinda cool :) 18:00:20 smaller code, I think, typically implies simplicity. 18:00:31 at least statisticaly it's less code to possibly have bugs:) 18:00:44 maan, I want to discuss this, but dinner's ready. I'll catch up when I get back 18:01:04 ehhe. 18:01:11 i gotta go anyways. i'm still at work. ttyl 18:11:33 --- join: kc5tja (~kc5tja@66-91-231-74.san.rr.com) joined #forth 18:11:33 --- mode: ChanServ set +o kc5tja 18:23:51 --- quit: fridge (Remote closed the connection) 18:35:07 hey! kc is here! 18:35:14 * MysticOne thinks about what thing to say sucks today 18:35:15 j/k 18:35:16 :) 18:39:56 Funny you say that because I'm getting hungry now. 18:40:13 and hunger... sucks? 18:40:19 * kc5tja is thinking of heading out to grab something to eat. 18:40:23 Yes, hunger sucks. 18:40:38 you could eat... 18:40:40 * MysticOne thinks 18:40:55 how long has it been since you had a tuna, steak, and cheese sammich? :) 18:41:32 * chandler is eating tomato soup and a package of club & cheddar cracker sandwiches 18:41:56 chandler: what does club taste like? 18:41:59 like golf club? 18:42:01 night club? 18:42:07 no, club crackers 18:42:19 MysticOne: Those saltine clue bats, you know? 18:42:23 kc5tja: hehehe 18:42:27 MysticOne: Not too long, actually. 18:42:35 kc5tja: oh, I hadn't heard you talk about one in a while 18:42:38 kc5tja: always sounded good 18:42:41 * kc5tja nods 18:42:59 I normally get tuna, turkey, and cheese now though -- just to reduce red meat intake. 18:43:09 but red meat is so red :( 18:43:22 I know, and I love red meat. But, irritable bowel syndrome and all that.. 18:43:35 wouldn't all veggies be better? 18:43:48 Perhaps, but all-veggies doesn't fill me up. 18:43:55 It doesn't satisfy me like meat does. 18:44:07 I'm still trying more stuff with tofu 18:44:10 that's a hard thing to cook 18:44:23 Yeah, you have to marinate it for like 24 hours or somehting, I heard. 18:44:58 well, Alton Brown on Good Eats said as long as you squeeze all the moisture out first, you can marinate it for like an hour 18:45:05 (firm tofu) 18:45:13 I made some and put it in chili 18:45:16 which cooked for about 4 hours 18:45:19 Well, how do you squeeze the moisture out and still have it be in neat little cubes? 18:45:21 I figured that'd be enough time for it to absorb some flavor 18:45:31 kc5tja: you wrap it in paper towels and put a weight on top of it 18:45:38 like a cookie sheet with something heavy on it 18:45:44 it gently squeezes the water out 18:45:46 But that still deforms it though. 18:46:08 I made some pickled eggs on the weekend. 18:46:33 kc5tja: no it doesn't 18:46:41 kc5tja: (this is before you cut it into cubes, btw) 18:46:43 its' when you have a block 18:46:49 MysticOne: Ahh. 18:47:03 the worst problem, though, wasn't flavor 18:47:05 it was the texture 18:47:11 I'm going to see if I can get some TVP next time 18:47:17 TVP? 18:48:02 textured vegetable protein ... basically granulated soy 18:48:05 makes it feel like hamburger 18:48:11 Ahh 18:48:20 * kc5tja was thinking transvestite pickles or something. 18:48:28 no, I don't put those in chili 18:48:31 those go in something else 18:48:40 * MysticOne grins 18:48:43 I'm afraid to know what though. :) 18:49:15 like, on a plate with a salad 18:49:18 err, sandwich 18:49:19 and chps 18:49:23 >:) 18:49:24 btw ... 18:49:30 did you know the US produces 2/3s of the world's soy? 18:49:39 * MysticOne did research 18:49:43 No. 18:49:53 actually produce 2/3s of it ... and we use it as animal feed 18:50:03 seems counterproductive :) 18:50:08 Ahh 18:52:02 I'd rather use the majority of it for fuel production purposes, and kiss those oil producers overseas goodbye. 18:52:19 But that will never happen. 18:52:30 And I digress. I'll be going for a bit -- food time. 18:52:37 * kc5tja is away: dinner 19:03:44 --- quit: tathi ("leaving") 19:06:00 MysticOne: still at work? 19:06:25 me? 19:06:31 I haven't been at work for like 4 1/2 hours 19:06:45 21:01 < OrngeTide> i gotta go anyways. i'm still at work. ttyl 19:07:13 oops, that is somebody else :) 19:07:20 yeah, that's not me :) 19:07:21 hehehe 19:27:31 sup dudes 19:36:06 not much, dudette 19:45:06 i am not a female 19:45:30 warp0x00 nads drop 19:45:34 now you are. ;) 19:45:56 warp0x00: hey baby! 19:46:00 * MysticOne chuckles 20:03:18 * kc5tja is back (gone 01:10:41) 20:03:39 I'm back 20:03:52 I thought you were kc5tja 20:04:22 Soon I'll be satiated. 20:05:24 wow 20:05:26 you change names a lot 20:05:36 one of our salespeople wants to sell me a Yaesu HT 20:05:43 After I become satiated, I'll probably become tired. 20:05:44 says it's brand new and does 6m, 2m, 70cm, 23cm .... 20:05:48 for $250 20:05:49 And then probably 'sleeping.' 20:06:03 What model? 20:06:17 I'm not sure exactly 20:06:19 he said, but I don't remember 20:06:24 * kc5tja heard Yaesu's made fairly good HTs, and rugged. 20:06:33 oh yeah 20:06:36 this one is water resistent 20:06:46 I think it's submersible 20:07:15 * kc5tja chooses Kenwood for the base station. I'd choose Icom for automotive or marine mobile. And Yaesu for HTs, I think. 20:07:22 Each company has found their niche. 20:07:36 Though, Yaesu also makes a kick-butt base station too -- the 1000MP. :) 20:08:33 I like my little Alinco 20:08:36 I need to put it in my car 20:08:49 I know one thing is for sure: one of the first things I am going to code in FS/Forth is a serial port controller for my rig. :) 20:09:00 hehehe 20:10:18 I've heard decent things about Alinco rigs too. 20:10:27 It's just that I don't know if they've cornered a niche yet. 20:10:41 I dunno 20:10:51 it's pretty easy to use, small, easy to read ... 20:10:56 and uses RJ-45 for the Mic plug :) 20:11:12 Nice. 20:11:14 Cheap. 20:11:14 :) 20:11:22 And thus, easy to hack/replace if damaged. 20:11:47 I still think ham radio gear ought to start using XLR connectors though. 20:11:48 yeah 20:12:12 While more expensive ($8 for an adapter at Radio Shack!), they *rock*. And the audio signal is balanced, so it's less affected by stray RF. 20:12:59 And they're quick (dis)connect too -- faster than BNC connectors, I think. 20:13:21 w00t 20:14:52 this is not funny 20:15:13 warp0x00: yes it is 20:15:14 I mean ... 20:15:15 what? 20:15:48 ill rape your face\ 20:15:55 bring it on! 20:16:10 Now THAT would be funny. 20:16:21 * kc5tja can sell that on the Internet. 20:16:25 hehehe 20:16:27 it'd be short 20:16:27 :) 20:16:29 yes kc5tja you can 20:16:40 Sell anything on the Internet it seems. 20:17:08 HIIIIYAH! TAKE THIS! *thrusts forward* 20:17:20 --- join: jdrake (~GodlessHo@CPE00045afdd0e8-CM014410113717.cpe.net.cable.rogers.com) joined #forth 20:17:20 --- part: jdrake left #forth 20:17:25 That'll teach you to try and rape my face 20:17:47 MysticOne: fine... i'll rape you in the ass then 20:17:50 Wait a minute... 20:18:01 you'll feel my hearbeat against your rectal wall untill you bleed 20:18:14 Since MysticOne said "Bring it on!", that implies consent between the parties involved. Hence, it's no longer rape. 20:18:22 Sorry MysticOne -- you're on your own on this one. 20:18:25 kc5tja: it's because I'm teaching him a lesson :) 20:18:38 warp0x00: too late, you tried the face rape, and I defended myself 20:18:42 so there's no ass rape now 20:19:05 warp0x00: Yeah -- it imploded. :) 20:19:14 It made a nice *foop!* sound. :) 20:19:40 fine 20:19:44 That's why MysticOne's so up-tight. *ducks* 20:19:58 ill overcome your soul with love and happyness untill you kill yourself 20:20:10 * kc5tja rotflmfo! 20:20:14 warp0x00: wouldn't I feel loved and happy and *not* want to kill myself? 20:20:22 yes 20:20:58 so ... 20:20:58 okay 20:21:00 umm... 20:21:02 love me? 20:21:16 Oh gawd... 20:21:21 wait, lemme check with my wife if it's okay if a stranger on IRC fills my soul with love and happiness 20:21:25 * kc5tja throws tomatoes at MysticOne for that one. 20:21:59 she said no 20:22:03 oh well 20:22:06 sorry to disappoint, warp0x00 :( 20:22:30 there is nothing you can do about it 20:22:54 MysticOne: Your wife will be your downfall. 20:23:06 yeah 20:23:15 she's a total cockblock 20:23:29 Whoa...things are gettin' steamy in here. 20:23:44 well, I'm so hot, and warp wants me :( so I guess he's frustrated 20:24:02 actually 20:24:07 ill just rape your wife 20:24:16 and any children you may have 20:24:21 warp0x00: may she have mercy on your soul 20:24:27 lol 20:25:28 alt.sex.forth :) 20:25:38 I440r_: hehe 20:26:04 or alt.warp0x00.needs.to.get.laid.so.he.can.stop.hitting.on.people.on.irc.and.saying.he.wants.to.fuck.their.wives.and.children 20:26:10 or get a life 20:26:11 one or the other 20:26:41 okay 20:26:45 you are on a chat about forth 20:26:49 dont tell me to get a life 20:26:59 warp0x00: you're on IRC, pot calling the kettle black 20:27:09 (and also in said channel) 20:27:11 10 POT 20:27:14 20 KETTLE 20:27:18 30 BLACK 20:27:23 40 GOTO 10 20:27:26 hehe 20:27:31 that's for #basic 20:27:39 is there a #basic ? 20:27:49 heck if I know 20:43:05 - /join #basic will tell you :) 20:48:35 On the subject of Alincos, I've had a couple. The DJ-610T was a decent radio, but the front end was easily overloaded. And it had a UI only a Japanese person could love. However, the DJ-235T is very usable (220 rig) 20:49:23 I have, umm... a DR-605? 20:49:24 I think that's it 20:49:25 2m/70cm 20:49:38 * warp0x00 wishes he was smart enough to be a ham 20:49:56 I don't see a quad-bander in Yaesus current line up. 20:51:40 They have a triple bander, the VX-7R, but no 23cm on it. 20:51:51 And I think the 220 output on the VX-7R is like 200mw. 20:54:18 I think I threw away my HRO catalog. I picked up a Kenwood TH-F6A at the last hamfest. I later looked at another HT, and was really glad I hadn't gotten it, because of the low power output on 220. 20:56:37 Utilizing a reliable FET power amplifier circuit, the VX-7R provides a full 5 Watts of power output on the 50, 144, and 430 MHz Amateur bands, with bonus coverage of the 222 MHz band at 300 mW (USA version only). And for 6-meter AM enthusiasts, you also get 1 Watt of carrier power on the 50 MHz band. Four power levels may be selected, each offering its own degree of battery conservation. 20:56:40 VX-7R. 20:57:07 That must be the one your buddy is trying to sell you. $250 is a decent price, if it's in like new condition. I think they're about $325 new. 20:57:31 $309 from HRO. 20:58:07 * MysticOne jams out to LTE 20:58:39 yeah, I think that's right 20:58:43 I can't buy it though 20:58:45 don't have $250 :) 20:58:48 :( 21:00:42 the dude in #basic is totally nuts guys 21:01:46 yeah, I can imagine 21:02:02 Heh 21:02:21 no seriously 21:02:29 we had some quick conversation 21:02:34 and then he just went nuts 21:03:15 No in this channel ever goes nuts. 21:03:22 (python sucks!) 21:03:42 * jc ducks into a spider hole 21:04:03 Fucking moron... *grumbles* 21:04:04 ;D 21:04:18 j/k :) 21:04:37 Hey, your opinion on perl is worth no more and no less than my opinion on Python. So I don't wanna hear about it. 21:04:43 oh yeah? 21:04:45 well forth sucks 21:04:49 know why? 21:04:51 because I don't know it 21:04:53 that's why 21:04:54 beats perl 21:05:05 I'll raise your perl with a forth and a fortran 21:05:10 Uhhhh, just out of curiosity, MysticOne, why are you here, then? 21:05:13 * ianp COBOLs all over MysticOne 21:05:25 jc: well, a while back, I was going to work on learning forth 21:05:28 and kc is a good friend of mine 21:05:30 so I figured I'd hang out 21:05:32 ah, cool. 21:05:37 * kc5tja shuts all your asses up by raising your COBOL with two INTERCALs and a BrainFuck. 21:05:37 then I saw that forth gets talked about 2% of the time in this channel 21:05:39 and I kinda forgot 21:05:41 :) 21:05:54 [22:04:43] --- ryno sets mode +Q #basic 21:05:54 * jc countattacks with XPL0 compiled for a 6800 21:06:25 * MysticOne counters with holarith cards! 21:06:31 I think that's spelled right 21:06:37 Nope. 21:06:44 DENIED!!! 21:06:46 oh 21:06:49 * MysticOne is the weakest link 21:06:52 good bye :( 21:06:59 Hollerith 21:07:00 That's 200 Monopoly dollars from MysticOne. 21:07:14 Or, Microsoft Money. 21:07:34 Just like I trust MS with my PC security, I'd trust them with my money. 21:08:07 jc: yeah, they'll probably use it to put some quality software company out of business 21:08:21 * jc likes Quicken 21:11:19 You know, I basically like Zilog. But sometimes... I'm playing with the ez80 Acclaim! board and OS + TCP/IP stack. The OS is actually Xinu. But they kindly don't tell me what the idle task does. Does it spin? Does it put the chip in low power mode? Does it do seti@home? Who knows! 21:16:01 Zilog is the shit 21:16:09 * warp0x00 still runs OASIS/THEOS 21:16:31 * kc5tja hates Zilog. 21:16:49 The Z-80 is the biggest screwup of a processor, second only to the 8086. 21:16:55 (and it clearly inspired the 8086) 21:17:25 the Z-80 is da bomb 21:17:27 Is there anything you *do* like? 21:17:32 forth 21:17:46 * kc5tja is a fan of the 6502 personally, though it isn't a perfect CPU either. 21:17:57 But I find it to be easier to program than the Z-80 by a long shot. Certainly more convenint. 21:18:01 convenient. 21:18:27 I'll give it it's 0 page operators. I sure as hell won't give it anything for it's stack. 21:18:32 * Herkamire eliminates a distracting blinking 12:00 by unplugging his new shitty sterio. 21:18:59 While true, I find it is rarely an issue. 21:19:15 In fact, in all the years I've programmed a 6502 box of some kind, I've never had an issue with a 256-byte stack. 21:19:20 I want simple hardware 21:19:27 (though, I note the 65816 has a 64K stack size) 21:19:40 0 page 21:19:47 i remember someone telling me about 0 page 21:20:00 i dont remember them saying nice things about it 21:20:08 But the 6502 is faster than the Z-80 anyway, clock for clock, even without its zero page. 21:20:16 I *love* zero-page. 21:20:44 im a fan of MOVE 21:20:45 Sorry, I'll trade a little speed for the convience of real 16 bit index registers. 21:20:45 i want one 21:22:02 Not me. I'll trade 16-bit index registers for simpler, faster, cheaper hardware any day. 21:23:09 Not sure where there's a "simpler" and "cheaper" difference. Having just bought Z80s and 6502s, they're about the same price. And there's little difference in the amount of glue logic required to build a system out of either. 21:23:28 On his aha page Jeff Fox ways saying that one could manufacture his F21 chip in low quantities for about $1/chip. anybody know what "low quantities" might be? 21:23:49 prolly ~1000 21:24:18 In fact, 6522 VIAs are more expensive than Z80 PIOs. 21:24:30 jc: When the Z-80 was introduced, it cost $250. When the 6502 was introduced, it cost $25. 21:24:43 BFD. Everything was cheaper 2 years later. 21:24:45 And it's got a much, much simpler bus architecture. 21:24:52 With the 8080 came out, it was $800 21:24:52 * I440r_ loves 6502 21:25:12 a9 01 85 02 20 12 34 6c 01 ff 21:25:17 anyone see the bug ??? :) 21:25:25 I440r_: Yes. 21:25:35 I440r_: However, it's fixed in the 65816. 21:25:47 they ruined it !!!!!!!!!!!!! 21:25:48 lol 21:26:08 did they also fix the x2 crash ? 21:26:15 I440r_: Huh? 21:26:21 Now, what I have a line on is some of the original Rockwell 6502 Forth chips. A friend of mine in Sri Lanka has about 40 of them, and has set aside 10 of them for me. 21:26:33 thers only one X2 opcode and that $a2 21:26:43 every other x2 instruction would CRASH the processor 21:26:45 HARD 21:27:16 jc mjy dad has a 68cc11f :) 21:27:20 I440r_: All 256 opcodes are used in the 65816, though one is just a place holder for multi-byte opcodes in the future, and one is for triggering a hardware coprocessor. 21:27:22 or is ut a 65fhc11 ? 21:27:39 er 68 i mean 21:27:45 68hc11, maybe? 21:27:45 68hc11f i think 21:27:52 with an F on the end 21:27:55 for "FORTH" 21:28:07 it has a built in rom programmed with forth 21:28:23 my forth is native code, because speed is important to me (and it seemed straightforeward.) 21:28:24 I don't think it's the same part, but I have a 68HC11 card I picked up from the guy at the Microship.com project that's Forth based. I think it has MPE's forth in it. 21:28:32 I440r_: There is also a 65F11 too. Same thing basically; 6502 with an internal Forth ROM. 21:28:42 now I'm thinking that to get real fast code (easily) you need an optomizer 21:28:42 That's the one i"ve got a line on. 21:28:54 coola 21:28:56 It's in the zig-zag package, as I recall. 21:29:05 Like the Rockwell modem chip. 21:29:06 and I'm thinking that the easies thing to optomize is probably not going to be native code. 21:29:35 Herkamire: Optimizers are anything but trivial. 21:29:55 Herkamire: You need to build dataflow graphs and such from the Forth code, examine how code is used to optimize register usage, et. al. 21:30:05 Herkamire: It's doable (RAFTS), but not easy. 21:30:19 kc5tja: no shit optimization is an excercise 21:30:57 But I'm quite happy with FS/Forth's performance -- almost exactly 1/2 that of optimized GCC output (GCC 3.2.2) 21:31:04 I'm thinking that I could just make it so that optomizition could be added later 21:31:20 kc5tja: very good :) 21:31:29 :) 21:32:12 I'm thinking that probably about 1% of the code is going to be run enough that optomizing would be nice. 21:32:30 I want a system that automatically optomizes things that run a lot 21:32:54 My philosophy on optimizers is the same as for Lisp: don't bother with integrating a super-duper compiler into the core environment. 21:33:17 Profile first, then when you do need to optimize, you can give hints to the optimizer (in Forth terms, you use a totally different compiler). 21:34:14 I'm thinking normally I'll use a very simple, unoptomized aproach. and when something needs to be optomized, it will. 21:34:42 * kc5tja nods 21:34:48 I do only basic peephole optimization in my output. 21:34:54 E.g., if I have 45 + 21:35:14 instead of creating: sub esi,4; mov [esi],eax; mov eax,45; add eax,[esi]; add esi,4 21:35:22 it'll create add eax,45 instead. 21:35:29 That's where much of my performance comes from. 21:35:32 I was thinking about doing that. 21:35:45 It isn't terribly hard to implement either. 21:35:47 for + - * / anyway 21:36:13 When FS/Forth is officially 'done' enough for public dissemination, I'll show you the sources to see how it's done. 21:36:16 no, I can check the last 2 instructions to look for pushing a 16 bit literal onto the stack 21:36:21 You should be able to handle left and right shift, too. 21:36:33 jc: Yep. 21:36:59 Herkamire: That's how I do it. But because of x86's variable length instructions, I maintain a pseudo-queue of "most recently generated instructions," so to speak. 21:37:17 I don't know much about how to do a good optomizer. I don't feel it is even enough to sortof leave room for it. 21:37:29 kc5tja: neat :) 21:37:48 (FORTHCMP compiled programs are claimed to be faster than C). 21:37:55 Herkamire: I have little to no desire to include a speed optimizer in my Forth. 21:37:57 so does gcc have a forth frontend then? 21:37:57 from emarcel hendrix pag 21:37:58 argh 21:38:12 from marcel hendrix's page 21:38:17 I can just check 6 of the previous 8 bytes to see that they match push load-immediat xxx 21:38:35 warp0x00: No. Not officially made by GNU at least. FORTHCMP is a research tool, intended to study ways of making Forth run faster. 21:38:46 oic 21:38:57 Herkamire: I do other kinds of optimizations too; constant folding is just one of the things I do. 21:39:19 kc5tja: that's the path I'm headed for now 21:39:19 Another is DUP DROP optimization, DROP optimization, etc. 21:40:11 dont you mean DROP ? 21:40:20 a lot of optimizations can be done just by looking at the last few instructions 21:40:57 warp0x00: no he doesn't 21:41:08 oic 21:41:09 okay 21:41:10 then 21:41:18 well i wish he would give some better examples than just a cheezy lame hello world 21:41:23 warp0x00: No, though that's another possibility, but it's so rare, that I probably won't bother with it. 21:42:07 kc5tja: do you want to not talk about that thing anymore? 21:42:16 sorry if i am boring you 21:42:28 Nahh; my mind is kind of reeling right now. 21:42:58 well i wouldnt want to do anything thats been done before lol 21:43:39 * kc5tja nods 21:44:00 Oh, yea I remember. It wasn't MPE on that 68HC11, I think it was New Micros. 21:44:14 prolly the same one cuz that sounds familiar 21:44:30 i never messed with it, i dont like the hc11 line of controllers 21:44:54 i hate the way they automatically push all registers onto the stack on interrupt 21:45:03 i still want to own a MOVE chip 21:45:08 thats DUMB - what if your interrupt doesnt USE thosee registers 21:45:16 you just wasted time and stack space 21:45:34 I440r_: Then use those registers, for something, then you won't have an excuse. :D 21:45:40 lol 21:45:47 grrr :P 21:45:54 I think that is the basic behavior fo the 6809 is it not? 21:45:56 i like the 8051 much better :) 21:46:08 thats motorola for ya 21:46:16 I440r_: 68000 doesn't do that. :) 21:46:20 motorola processors are better 21:46:32 kc5 motorola processors rock. their controllers SUCK ass 21:46:38 motorola is a shitty company 21:46:39 filp that statement for intel 21:46:45 But the 6811 is the same core as the 6809. 21:47:01 I'm thinking the 6809 also saves all registers on interrupt. 21:47:02 I can't remember for sure though. 21:47:17 its not a big deal really - it just buggs me lol 21:47:25 All I know is that i never really liked the 6809 -- too much like a hodgepodge of cool and useless stuff. 21:47:48 The 68010, OTOH, is, as far as I'm concerned, the world's best CISC processor ever designed. 21:48:08 i like the 020 21:48:18 erm did the VBR come in on the 010 or the 020 ? 21:48:21 010 21:48:40 once they gave it a vbr it rocked 21:48:40 lol macintoshes used those 21:48:43 life begins at 68040 - senility begins at 86 21:48:48 Well, even ignorign that, the 68010 had better instruction timings for some things, and its loop mode was quite nice. 21:49:12 Macintoshes used 68000s all the way through to 68040s. 21:49:20 Amigas used 68000 through to 68060s though. :) 21:49:23 my amiga 1200 used the 020 21:49:35 i loved that box :( 21:49:39 * kc5tja nods 21:49:39 * I440r_ misses his amiga 21:50:05 * kc5tja still has his Amiga 500, with its monsterous 120MB harddrive, and egregrious 7.15909MHz 68000 processor. :D 21:50:12 lol i made a cable to allow me to put a full sized hard drive in it - but i could never get the box closed after taht heh 21:50:21 so anywone else in here wanna see if they can grok my insanity? 21:50:47 68K couldn't afford to stack all the registers in an interrupt. Too many, and it was too damn slow as it was. 21:51:36 job im doing now uses the coldfire which is a 68k derivative - i dont know much about it tho 21:51:39 At the time, it was a royal powerhouse. 21:51:49 Coldfire rocks too, though not as much. 21:51:54 ive only made like 8 code defs 21:52:10 Take a 68000, strip it of all CISC-like features, leaving only its RISC-like features. Optimize hardware so instructions run in single cycle time. 21:52:18 this coldfier is like 28465278364578235 times more processing power than this app needs 21:52:29 which is good cuz they are going to need it cuz they have almost ZERO design lol 21:52:45 The interesting thing is, each time they introduce a new ColdFire, they re-introduce one aspect of the original 68000 line. :) 21:52:53 :) 21:53:12 so they are saying ... if it aint broke why did we fix it ?> 21:53:14 lol 21:53:18 I think it's Motorola's attempt to redesign the 68000 slowly over time, so that it can keep up with competitor's products. 21:53:54 No, the 68000 core was more or less broke, in that it tried to do too much (shades of 6809). It was basically a PDP-11 on a chip, only it wasn't compatible with the PDP-11. :D 21:53:56 its them saying "do we make something new or recycle something old" 21:54:08 The ColdFire is more along the lines of, "OK, we made a mistake in the past, now we're fixing it." 21:54:19 ok :) 21:54:55 I think modern ColdFire's have something like 80 to 90% compatibility with user-mode 68000 software now-a-days. 21:55:02 And nearly all the instructions are single-cycle in execution time. 21:55:02 hrm if someone doing seti@home FINDS something do they get a prize ?? :) 21:55:08 no im not doing seti at home lol 21:55:18 Yes. 21:55:31 yes 21:55:35 You get world recognition for having helped discover ETI. 21:55:36 I should do Seti@home 21:55:37 they get their name on something 21:55:43 use up some spare cycles on this big ol' G5 21:55:45 i do seti 21:55:49 it sounds too much like "train spotting" 21:55:51 lol 21:55:54 I440r_: that movie was great 21:55:59 in a ... weird way 21:56:03 i belive if we find ETI it could save the human race 21:56:03 err no 21:56:05 They problem with the Coldfire, at least what I remember from last time I looked at, is like the PPC. Requires too much support hardware. 21:56:10 train spotters are lame 21:56:10 Well, Arecibo is no more, so no new data is being collected. 21:56:14 Perhaps they have a single chip solution now, I don't know. 21:56:19 kc5tja: wha? why? 21:56:24 when did it go away? 21:56:25 MysticOne: No money. 21:56:29 MysticOne: Last year. 21:56:30 grrrr 21:56:35 stupid government 21:56:46 it might be they get a new array however 21:56:55 yeah 21:56:57 I swear 21:57:08 governments are never going to understand anything important 21:57:09 jc: Well, when you have caches of any kind, you need a custom chipset to handle the burst-mode on the bus. The higher the bandwidth, the more esoteric the bus. 21:57:12 nothing above money 21:57:26 could do RC5 or Folding@Home or something ... 21:57:43 * kc5tja was thinking about Folding@Home, but not right now. 21:58:02 There's also a Mersenne Prime at home type thing too. 21:58:18 But I'd be more inclined to work with Folding@Home. 21:58:19 yeah 21:58:20 does seti@home use a lot of bandwidth? 21:58:25 Herkamire: Nope. 21:58:25 Herkamire: no 21:58:32 My dad uses it over a telephone dial-up modem at 33.6kbps. 21:58:32 like none 21:58:40 Negligable impact. 21:58:47 yeah, it's not much at all 21:58:52 it transfers like 30k or something every 4 hours 21:58:54 d/l a small packet and crunch for several hours 21:58:55 wow. what does it do then? 21:59:12 never mind 21:59:14 analyzes data, looking for specific things 21:59:20 30k every 4 hours works for me 21:59:33 well less hours if your system is faster 21:59:34 Mmmm, eggnog. 21:59:40 afaik, looking for kinda non-naturally occuring signals 21:59:42 the packets aren't large 21:59:46 Well, I think you can configure how big a chunk you want it to process at a time. 21:59:47 and they're plaintext 21:59:54 kc5tja: no 22:00:04 jc: i want 22:00:29 I don't know if they're plaintext, but the are signed. And periodically the send finished packets to another user for checking. 22:01:04 Somewhere, there's a working 'y' key on this keyboard. 22:01:41 I wish there was a network like that where once you had contributed enough, you could send a program out to be run on lots of computers and get the results back 22:02:07 --- join: chrisrw (~chris@ca-cmrilo-cuda1-c3b-124.vnnyca.adelphia.net) joined #forth 22:02:07 --- quit: chrisrw (Remote closed the connection) 22:03:18 problem is most of the things I'd want to do with it would use more bandwidth 22:03:19 Cool. Fixed the page-alignment stuff, so the generated ELF file for : foo [entry-point] foo ; is now only 242 bytes. :D 22:03:26 All but five of which is boiler-plate. 22:05:28 wtf is an FX51 cpu? 22:05:37 one better than an FX50? 22:05:43 who 22:05:46 makes them 22:05:47 I suspect it's an 8051 clone. 22:05:52 * MysticOne desn't know anything about them 22:06:11 they are the fastest CPU for seti@home aparently 22:06:27 OK, definitely NOT an 8051 clone then. 22:06:29 :) 22:07:05 According to Google, the FX51 is a specific generation of AMD Athlon. 22:07:24 maybe before I spend too much time thinking about getting a network of computers to compute something for me when they're not busy, I should make something where my computer does something for me when it's not busy. 22:07:25 aparently 22:07:27 It's an Athlon64 CPU. 22:07:54 Herkamire: :) 22:08:12 Herkamire: You're into fractals; maybe you can do distributed fractal generation. :) 22:08:13 they are also aparetnlty fucking fast 22:08:16 You're thinking of the XA 8051. 22:08:32 --- join: imaginator (~gps@166.70.196.201) joined #forth 22:08:36 A freaky little part if there ever was one. 22:08:37 warp0x00: Athlons have always been major speed demons. 22:09:05 kc5tja: yeah :) that's one of my thoughts. it's very very little data (even if you send the algorithm) but returning large images or movies would be more bandwidth 22:09:06 My 800MHz Athlon is more than sufficient for my Neverwinter Nights game, despite them recommending a 1GHz Intel CPU. :) 22:09:09 kc5tja: yeah im runnnig a barton and 2 tbreds 22:09:11 * jc wonders if he should mistake kc5tja's comment as meaning he actually likes something. 22:10:12 I also want to do some crazy raytracing movies, but I can't be distributing huge source files 22:10:21 jc: I love Athlons. They're all I will ever use at the moment. 22:10:38 Herkamire: Yeah, raytracing is another good distributed application. 22:10:55 I generally prefer Athlons. Although the very latest Intels are looking decent. Still expensive. 22:11:02 Yep. 22:11:04 What I really want is a dual MP system. 22:11:06 my only problem with Athlons is heat 22:11:11 they seem to get too hot ... 22:11:13 I recommended an Intel P4 for my roommate's recording computer. 22:11:14 quite often too 22:11:26 MysticOne: solution: water cooling 22:11:29 MysticOne: I haven't had a problem with that. 22:11:33 I've been thinking about how to do raytracing really really realistically, and it's going to slow it down quite a bit 22:11:53 What I find a little disturbing yet satisfying is that when I upgraded the BIOS on my laptop, there was an M4 microcode upgrade. 22:11:54 kc5tja: tbirds were evil hot running at 90+W 22:12:02 warp0x00: true, true ... 22:12:10 we got the most interesting heatsink today for a special application 22:12:17 instead of just normal cooling fins ... 22:12:18 do tell 22:12:21 it has tons and tons of screws 22:12:31 or, well, I say they're screws ... that's what they look like 22:12:31 oh yes 22:12:33 to increase the surface area 22:12:40 ive seen those 22:12:50 * kc5tja ought to go into business selling fractal heatsinks. :) 22:13:02 I figure, if they have fractal antennas, fractal heatsinks ought to work, right? :) 22:13:06 infinite surface area! 22:13:17 they have fractal antennas? 22:13:20 warp0x00: Yep. 22:13:26 link plz 22:13:27 * kc5tja has always wanted to try one too. 22:13:29 Maybe someday... 22:13:45 warp0x00: Google, basically. I don't have any off hand at the moment. 22:14:03 fractenna.com 22:14:28 We had a heat sink design program at IBM that was pretty nifty. You know those heatsinks with real tall narrow spires? WORTHLESS. 22:15:19 yeah i know 22:15:25 There's a ratio of base cross section to height where it ceases to become an effective radiator. It's something like ~4x the crosssection of the base. 22:15:53 So if it's .25 x .25 at the bottom of the spire, any taller than ~1" becomes wasted material. 22:16:02 i love how HUGE CPU cache is 22:16:16 Yea. 8x the memory of my original CP/M system. 22:16:25 no i mean 22:16:27 physically 22:17:44 You know that's the only reason the Slot-1 and Slot-A designs even existed. It was because they couldn't fit the cache onto the CPU die. And it's a lot more expensive to put multiple dies on a PGA. So the slot-x was developed, cache on the CPU die became practical, and the slot died. In fact, electrically, the slot is a very poor design. 22:17:53 Leads to assymetric trace lengths. 22:18:35 That isn't the issue. Stray capacitance and inductance were major electrical factors. Also termination. 22:18:58 Which is caused by... say it with me now... "assymetric trace lengths" 22:19:02 * kc5tja pets his Athlon -- slot A. :D 22:19:07 Nope. 22:19:12 It's caused by the contacts in the slot itself. 22:19:27 your both wrong 22:19:32 its caused by too much penis 22:19:45 oversexed archictures 22:19:54 That's part of it. Having signals arrive out of phase, particularly when multiple data or clocks are on the wire at the same time is even more of an issue. 22:20:21 jc: I thought it was a ploy to make it so AMD couldn't copy designs :) 22:20:59 Maybe in their dreams. 22:21:09 maybe 22:21:18 AMD ended up not copying them anyway; EV6 blew away Intel's best offering by at least an order of magnitude. :) 22:21:36 EV6? 22:21:44 Actually, the reality is the way reverse engineering works to avoid lawsuits, the engineers don't even look at an Intel part. 22:21:51 The front-side bus used by the Athlon; originally used on DEC Alpha RISC CPUs. 22:22:00 ah yes 22:22:20 thats cuz intell was using SDR like 33Mhz at that point still 22:22:34 For instance, at IBM, if you've ever seen a listing of someone elses BIOS (Anna books, disassemblies of Phoenix, etc), you can't work in their BIOS group at any position that allows you exposure to their source. 22:24:12 Makes sense. 22:24:26 But they can certainly consult though, yes? 22:24:39 How do you mean? 22:25:27 If I worked at Award, and have the source to a Phoenix BIOS listing, then I cannot work on Award BIOS directly. But someone who CAN can ask me about Phoenix's implementation of various features, right? 22:26:20 What IBM generally does is has 2 divisions in the group. One group plays with the BIOS, and documents the behavior, and writes a spec. The other group implements the BIOS from the spec. The first group is called the black or dirty group, the second are called the white or clean group. 22:26:30 I dunno about that situation. 22:26:43 IBM is like seriously phobic about contamination. 22:27:05 That's basically what I was describing, only with a cleaner degree of separation. 22:28:34 What I don't know is if the first group is allowed to disassembled the BIOS, or use a ICE on it, or if they can only work from published specs. 22:28:59 The guys I dealt with when we had them building the BIOS for the 4-way Xeon system were the clean group. 22:29:24 I dunno if you've seen the memory compression technology from IBM, but I worked on the first server that had that techology. 22:29:34 Nope. 22:29:40 im reading this review w/ benchies 22:29:46 the reviewer is kindof an idiot 22:29:50 i invented a compression method! 22:29:59 i compressed a 40k image down to 90 bytes 22:30:00 I440r_: /dev/null doesn't count. 22:30:04 no 22:30:05 It's pretty neat. Does real-time compression with just about a garuanteed 2x compression. 22:30:15 i could decompres it. and it was NOT lossy 22:30:38 I440r_: so why haven't you revolutionized the world then 22:30:40 They have Linux drivers that dynamic size the memory in the system. So 512MB looks like 1GB most of the time, and *zero* impact on system performance. 22:30:54 however. it requires alot of adjacent pixles of the same colour 22:30:58 It was a white screen. 22:31:00 lol 22:31:02 no 22:31:14 it uses horizontal and vertical compression 22:31:17 I440r_: Well, run-length encoding can get similar compression ratios. 22:31:29 with alot of pels of the same colour adjacent it compresses most of them downt to a single bit 22:31:39 it used RLE as the last step 22:31:59 take your xchat window for instance with all the text in it 22:32:10 it would do great compressing a snapshot of that 22:32:18 thats what gif does 22:32:41 no. gif isnt as good as this lol 22:32:42 well 22:32:52 however 22:32:59 same technique 22:33:00 gif works on a line by line basis, but within the line it compresses horizontally only. 22:33:04 if i feed it data with lots of colour changes you get negative compression 22:33:12 lol 22:33:17 I440r_: Most engines do. 22:33:18 make it adaptive then 22:33:26 it is adaptive 22:33:32 sort of 22:33:39 Wavelets and fractals are the only compression techniques that truely compress in both directions concurrently. 22:33:45 it compresses each line after the first using the horizontal methid and the vertical method 22:33:49 and picks the best 22:33:57 kc5tja: how do you use fractals for compression lol 22:34:29 warp0x00: http://links.uwaterloo.ca/fractals.home.html 22:34:30 warp i can send you a copy of the data compression handbook ;) 22:34:46 I440r_: please do 22:35:03 lol its sorta illegal :P 22:35:18 sorta why? 22:35:32 because its a book for sale. not for free distrobution :P 22:35:45 i have it in electronic and physical 22:35:55 * kc5tja still wants to read Thinking Forth; I've yet to find a copy of that book anywhere, online or otherwise. 22:35:57 well im unlikely to buy it anyway 22:36:12 ya 22:36:54 I'll tell you one that that really annoys me about Intel (and AMD may do this, I don't know). With Intel, the level of documentation you can get access to depends on your relationship with them. Even though IBM designed cutting edge servers with Intel techology, there were documents that we couldn't get that the BIOS guys could. And supposedly there was a level of documentation *they* couldn't get, that was reserved for people who designed chipsets. 22:37:06 And that's just complete bullshit. Especially when things don't work as advertised. 22:37:26 Which is so common with Intel design reference materials that it's not even remotely amusing. 22:37:40 jc: This is very common practice. 22:37:48 Try developing your own video driver for nVidia chipsets. 22:37:59 Unless you work for Microsoft directly, good luck. 22:38:23 I can see if they won't give it to Joe Blow, or a low volume company, but at the level IBM dealt with Intel, it caused a lot of resentment. 22:39:01 I like the fact that I can get almost any documentation I want about PowerPCs from IBM. 22:39:17 I got so many free data books on PowerPCs years ago, back when the 604 was king of the hill. 22:39:39 I think I still have my 603e book lying around. :D 22:43:00 jpeg uses wavelet right? 22:43:04 No 22:43:08 ? 22:43:10 Discrete Cosine Transform. 22:43:15 wtf is that 22:43:21 Almost like a Fourier Transform. 22:43:32 i am stupid 22:43:38 Discovers frequency components of an image, and compresses based on that. 22:43:41 i dont know anyhting about these maths 22:43:56 * warp0x00 crawls into the corner and hides 22:44:04 :) 22:44:20 you peole scare me 22:45:13 how do you know so much jesus 22:45:36 For every one thing I know, there is something that you know that I don't. 22:45:49 i seriously doubt it 22:45:54 You'd be surprised. 22:46:00 and/or i know the wrong stuff 22:46:25 I know more than both of you combined. 22:46:40 Dude, my roommates make me feel like I want to crawl under a rock and die because they are just so damn good with music, and I suck so bad.... 22:47:11 My sensei makes me feel the same way when he does some really gnarley aikido technique I ain't *ever* seen before. 22:47:12 etc. 22:47:12 dude, i am not good with music 22:47:24 or martial arts 22:48:19 Physical strength is for the weak of mind. 22:48:30 im with chandler on that one 22:48:36 Unless you need to open pickle jars. 22:48:46 jc: Even then. 22:48:51 Riiight. 22:48:53 chandler: lame excuse 22:48:56 Better to hit the jar with a hammer. 22:49:01 jc: I've opened jars that nobody else could open by employing basic thermodynamics. 22:49:10 you need good health to keep your brain running well 22:49:17 Friction makes heat, heat expands metal faster than glass, and voila -- jar is opened. 22:49:18 Herkamire: I am in good health 22:49:23 good health requires exersize and good eating. which build strength 22:49:24 kc5tja: i seriously doubt theres anything really usefull i know that you dont 22:49:50 Herkamire: I don't have a car. I walk everywhere. This means I get a non-zero amount of exercise 22:50:10 OK, a jar was a bad example. Next time you need to loosen the lugnut that the shop-monkey put on with an air-wrench set to 200 ftlbs, the hot water trick ain't help you. 22:50:23 chandler: Try biking -- uses a different set of muscles. And once you get that down, try aikido -- yet another set. Karate, still another. Etc. 22:50:32 kc5tja: I don't care that much 22:50:41 I have better things to be doing with my time 22:50:47 I was going to say something about biking 22:50:52 I bike to aikido most of the time 22:51:31 and when my army of robots takes over the world, your aikido won't help you 22:51:44 jc: Tool-making is one of mankind's best qualities. That requires intelligence, not strength. Ergo, my intelligence suggests to use same said airwrench to loosen said lugnut. :) 22:51:59 or some comparable tool. 22:52:00 Riiiight. Always keep one handy on the expressway? 22:52:02 kc5tja: is owning jc 22:52:12 No, he's special casing everything. 22:52:16 And actually getting damned annoying. 22:52:18 jc: So are you. 22:52:36 on both accounts. 22:52:41 ok cool. we agree. now on to bigger and better things 22:52:47 Gee, I wish *I* had answer for everything. 22:52:49 hmmmm 22:52:50 yeah like 22:52:52 sex 22:52:56 jc: So far, you haven't let us down. :) 22:53:33 * Herkamire reflects on his own advice, and decides he will actually go to bed now. 22:53:42 too late for sex. girlfriend is already asleep 22:53:51 Heh 22:53:57 I should get to bed myself. 22:54:05 yes 22:54:13 i hope my programming language gives you nightmares kc5tja 22:54:23 warp0x00: Nope. 22:55:03 If anything gives me nightmares, it's working at In-N-Out with every district manager in the store, all with clipboards and pens, all at the same time. 22:55:05 there should be a freenode channel for programming language theory 22:55:06 (like we had today) 22:55:58 kc5tja: that's nightmare material :) 22:56:11 wait 22:56:13 kc5tja 22:56:14 warp0x00: what channel do you think you're in? 22:56:18 you work at a burger place? 22:56:29 Yep. 22:56:30 Herkamire, wanna give it a shot? 22:56:37 you have all this expertise 22:56:42 and you work at a burger joint 22:56:45 I come to this channel to discuss programming language design and theory 22:56:53 ur my kind of person kc5tja 22:56:54 People are just full of surprises. :) 22:57:10 Herkamire k u ready mofo? 22:57:15 warp0x00: give what a shot? and I don't work at a burger joint 22:57:16 My former job was working for HIfn (http://www.hifn.com), where I was a semiconductor verification technician. 22:57:43 kc5tja: i just refuse to give away my creativity to some company 22:57:50 kc5tja: how did you go from semi tech to a burger joint? 22:57:55 Economy. 22:57:59 heh 22:58:02 Tech industry layoffs. 22:58:09 Herkamire: well i tried describing somethign to kc5tja and it mystified him 22:58:17 and semi industry is not in a good state right now 22:58:22 warp0x00: I missed it 22:58:27 Yep. 22:58:29 Herkamire: over /query 22:58:46 Herkamire: wanna give it a shot 22:59:15 warp0x00: give what a shot? understanding your language? 22:59:26 Before working for Hifn, I ran an ISP, but PacBell put us out of business. 22:59:32 (now SBC DSL) 22:59:34 Herkamire: yes 22:59:52 does it have a purpose beyond confusing people? 23:00:08 yeah 23:00:10 i like it 23:00:14 im going to try to implement it 23:00:15 cool. then yes 23:00:17 then I'll go to bed 23:00:23 okay 23:00:28 wanna do it in here or over /query 23:00:35 /query 23:00:53 hm, I wanted to see it to 23:00:54 er, too 23:00:56 so why not do it here? 23:01:09 whatever 23:01:23 okay 23:01:24 Who likes Lamba the Ultimate? 23:01:43 Lamba/Lambda 23:01:43 imaginator: never heard of it 23:02:27 http://lambda.weblogs.com/ 23:02:28 uh 23:02:44 imaginator: it's mostly harmless 23:02:47 Herkamire and chandler, lets go over to #codetunes to discuss it? 23:02:50 that's how I would describe it 23:02:53 It's a community where people introduce new programming languages, and discuss different ideas about them. 23:03:49 Why mostly harmless? What is a better place? 23:04:38 ACM? IEEE? 23:06:08 --- quit: jc ("Departure courtesy another successful alienation by kc5tja -- you want fries with that?") 23:06:33 Alienation? Hardly. jc brought it all on himself. 23:06:51 The logs hosted by nef, gathered by clog, will clearly show evidence thereof. 23:07:57 I don't understand what he was angry about 23:08:02 Nor I. 23:08:11 Never once did I insult, never once did I make a false accusation. 23:08:12 oh well 23:08:15 * kc5tja shrugs 23:08:17 kc5tja and jc were just being smartasses. if either got mad I would say it's their own fault. 23:08:24 yeah 23:08:29 as if kc hasn't gotten mad and left before 23:08:29 :) 23:08:34 so it's not like it's new 23:09:08 kc5tja: i hate you and everything you stand for so HA 23:09:14 Well, I'm off to do a radio show recording with my roommates. 23:09:22 warp0x00: HA? Well . . . HEH! 23:09:31 Then I'm off to bed. 23:09:48 And no, this is not a ham radio type radio show. 23:10:02 It's more a fun thing. It's not even broadcast. :) 23:10:04 (yet) 23:10:07 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 23:11:15 I'm heading to bed too 23:11:18 except I don't logoff 23:11:18 :) 23:12:30 MysticOne: it's OK to log off. this stuff is logged anyway :) 23:12:39 Herkamire: I never log off 23:12:45 screen! 23:12:50 Herkamire: shell + screen == joy! 23:13:00 * chandler hugs his server 23:13:05 * MysticOne hugs his too 23:13:14 I like to save electricty by turning off my computer when I sleep 23:13:22 Herkamire: they don't use that much 23:13:31 I play music while I sleep 23:13:36 but the laptop sleeps while I sleep, yes 23:13:43 the server is hosted at my alma mater though 23:13:54 Don't you mean shell screen + joy !? 23:14:10 oh, you ssh to a server, and run irc on that? 23:14:15 I do... 23:14:32 running irssi under screen 23:14:48 yeah 23:14:51 same thing here 23:14:54 except my desktop doesn't sleep 23:14:55 :) 23:15:00 nifty 23:15:16 so uh... why not run it on your local box? 23:15:34 Herkamire: because the remote box generally has much longer uptime and a much more persistent net connection 23:15:54 you get disconnected a lot? 23:15:59 dialup? 23:16:01 Herkamire: because the shell is more reliable usually 23:16:05 occasionally. no, broadband 23:16:14 university connection no less 23:17:01 actually, a lot of times I put this machine to sleep 23:17:04 unles I have something going on on it 23:17:23 I also have two IRC bots running on the desktop :-) 23:18:26 okay, I'm off 23:18:27 g'nyght! 23:18:59 goodnight 23:53:49 --- join: Robert_ (~snofs@c-305a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 23:53:49 --- quit: Robert (Connection reset by peer) 23:59:59 --- log: ended forth/03.12.15