00:00:00 --- log: started forth/05.05.26 00:57:10 --- join: Philbert (~philh@CPE-203-45-190-162.qld.bigpond.net.au) joined #forth 01:01:35 --- join: snoopy_1711 (snoopy_161@dsl-084-058-022-096.arcor-ip.net) joined #forth 01:01:44 --- nick: Philbert -> LuckyPhil 01:03:39 --- part: LuckyPhil left #forth 01:03:57 --- join: LuckyPhil (~philh@CPE-203-45-190-162.qld.bigpond.net.au) joined #forth 01:04:08 hello 01:06:46 Hi. 01:07:15 hey there Robert 01:09:15 How are things, lucky phil from down there? 01:09:52 Pretty good.. 01:10:03 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 01:10:07 how are things up there? 01:10:26 --- nick: snoopy_1711 -> Snoopy42 01:11:03 Uhm, not too bad. Rainy and cold. 01:15:51 all quiet on the forth front... 01:16:33 Yeah, the 70s somehow wandered away into the history books. 01:16:53 i miss the 70s 01:16:58 bring em back i say 01:18:03 * Robert does'nt even remember the 80s ;) 01:55:34 forth needs killerapps! 01:56:03 Yeah... write a game that's very addictive, and can't be coded in non-Forth languages. 01:56:06 --- join: Raystm2 (~vircuser@adsl-69-149-32-57.dsl.rcsntx.swbell.net) joined #forth 01:58:10 Hi Ray. 01:58:11 C is probably more difficult to recode a game written in Forth, lisp and languages like lisp are more easy in that point. 01:59:51 --- quit: nanstm (Read error: 54 (Connection reset by peer)) 02:33:07 why does a killerapp for forth be incompatable with other languages? 02:39:23 Otherwise Microsoft will port it to BASIC and make millions of it. 02:39:29 Which they use to kill Forth. 02:42:19 so what you are saying is that there will never be a killerapp for forth then 02:42:54 Right ;) 02:43:08 Unless you make a Forth computer everyone usese. 02:43:09 uses 02:43:50 i find forth to be very implementtion specific 03:27:37 yes, that's right 03:35:51 a forth computer would be a nice idea, but how would that be possible, when nowhere is a real fast processor for forth? 03:37:25 even the theoretical processor(iirc 25x) of chuck would be too slow 03:38:54 Well... start soldering. ;) 03:56:30 for the equipment I would need money, much money and they I don't have. fuck money. 04:00:36 Yeah... better stick to Java ;) 04:01:38 urg, Java. 04:01:43 java... yech! 04:02:58 Java is broken by it's design. Forth has a design 04:04:42 which is probably better than most of the mainsteam computer languages. 04:05:06 i like forths design 04:05:09 its rather elegant 04:06:05 and small 04:07:06 I had to do Java for school this year :( 04:08:47 I don't know any other programming language except brainfuck which can be comfortable in one line screen like a calculator LCD. 04:08:57 written 04:09:01 sorry 04:09:50 brainfuck.. now that is a language! :) 04:11:31 one nibble language, ideal for prefetching. 04:12:00 My Brainfuck computer has 8 output LEDs. 04:12:07 And 4 input switches. 04:12:28 a brainfuck computer? how did you build it? 04:12:44 Just a PIC and some LEDs/switches. 04:12:51 robos.org/bfcomp/ 04:13:23 cool 04:13:35 btw, I found out my TI82 can be programmed in machine language with some hack... now all I need is a stereo 2.5mm connector (only have mono now, but some day...) 04:14:09 Robert: it don't work any good 04:14:51 Frek: Aw... Why not? 04:15:07 Robert: asm shells are unstable on TI82 04:15:14 :( 04:59:27 --- join: Herkamire (~jason@c-24-218-95-147.hsd1.ma.comcast.net) joined #forth 04:59:27 --- mode: ChanServ set +o Herkamire 04:59:37 Hi. 05:04:00 --- join: qFox (~C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth 05:04:57 --- join: tathi (~josh@pcp01375108pcs.milfrd01.pa.comcast.net) joined #forth 05:09:28 hi 05:09:58 just went out to Sylvesters for breakfast 05:10:07 just after they opened 05:10:20 I don't think I've ever been there when there wasn't a huge line to get in 05:12:00 :) 05:51:02 found a neat free software game called enigma 05:51:10 has an ebuild 05:51:16 I'm going to go take a nap 05:58:08 Hrm, nice 05:58:26 Herkamire: btw, just added Herkforth to Dotquote .... I forgot last time for some reason :) 05:59:00 * arke has gotten one interested response from c.l.f. about it so far, and that was astrobe in a private email 06:39:27 --- quit: Frek (Read error: 104 (Connection reset by peer)) 06:40:39 --- join: Frek (~anvil@h254n2fls31o815.telia.com) joined #forth 06:47:41 herkforth... hmm interesting 07:02:21 --- quit: LuckyPhil ("leaving") 08:00:34 cool. I finally got isforth ppc up to date with all I440r's changes. 08:01:04 Nice :) 08:02:14 I think he's just about done too, so there might actually be an official release soon. 08:02:36 Well, I still have a few things to argue with him about first. :P 08:56:46 --- join: vitaminmoo (~vitaminmo@dsl-95-113.peak.org) joined #forth 09:03:29 --- quit: vitaminmoo ("Leaving") 09:03:44 --- join: vitaminmoo (~vitaminmo@dsl-95-113.peak.org) joined #forth 09:03:49 --- quit: vitaminmoo (Read error: 104 (Connection reset by peer)) 09:03:50 --- join: vitaminmoo (~vitaminmo@dsl-95-113.peak.org) joined #forth 11:08:07 --- join: Topaz (~top@sown-85.ecs.soton.ac.uk) joined #forth 11:11:30 Hi 11:32:27 teh hies 11:36:05 teh hies 11:36:42 Heisser Tee. 11:36:57 Yay 11:44:53 --- quit: Topaz ("Leaving") 12:19:07 --- join: warpzero (~warpzero@wza.us) joined #forth 12:19:07 Hi. 12:19:17 teh hies 12:19:27 HO\T THAIS 12:19:59 Group drooling over PS3. Maybe you should port RF to it when it's released ;) 12:20:44 --- nick: Raystm2 -> nanstm 12:21:03 yeah RF on the ps3, not easy probably it would only run after a lot of hacks because of its DRM system 12:21:11 hi arke 12:21:14 DRM? 12:21:27 Digital Rights Managment 12:22:08 new gaming consoles aren't enthusiast friendly 12:22:39 Hm, I read some (possibly not based on reality) rumor about them changing their attitude on this one. 12:26:41 http://www.eetimes.com/news/latest/showArticle.jhtml?articleID=163106213 12:26:46 That's the one I was thinking about. 12:27:07 I guess it's mostly about the Cell, though. 12:27:30 cell looks very interesting 12:31:12 --- quit: warpzero (clarke.freenode.net irc.freenode.net) 12:31:12 --- quit: Frek (clarke.freenode.net irc.freenode.net) 12:31:12 --- quit: tathi (clarke.freenode.net irc.freenode.net) 12:31:12 --- quit: Herkamire (clarke.freenode.net irc.freenode.net) 12:31:12 --- quit: OrngeTide (clarke.freenode.net irc.freenode.net) 12:31:13 --- quit: Robert (clarke.freenode.net irc.freenode.net) 12:31:13 --- quit: cmeme (clarke.freenode.net irc.freenode.net) 12:31:26 --- join: Herkamire (~jason@c-24-218-95-147.hsd1.ma.comcast.net) joined #forth 12:31:26 --- join: warpzero (~warpzero@wza.us) joined #forth 12:31:26 --- join: Frek (~anvil@h254n2fls31o815.telia.com) joined #forth 12:31:26 --- join: tathi (~josh@pcp01375108pcs.milfrd01.pa.comcast.net) joined #forth 12:31:26 --- join: OrngeTide (~orange@rm-f.net) joined #forth 12:31:26 --- join: Robert (~snofs@c-f778e055.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 12:31:26 --- join: cmeme (~cmeme@216.184.11.2) joined #forth 12:31:26 --- mode: irc.freenode.net set +o Herkamire 12:31:51 Automatic purges... that's a nice feature of Freenode. 12:44:56 --- join: Topaz (~top@sown-85.ecs.soton.ac.uk) joined #forth 13:15:15 one problem of the cell is that DRM is inbuild into its design. :( 13:16:50 Ouch. 13:17:38 yep 13:17:44 like a VISA card 13:17:48 has an internal key 13:17:56 if you try to read it, it self destructs 13:17:59 so, you're screwed 13:18:00 :) 13:21:48 anyone know what "vme" means in the cpuinfo on linux ? 13:22:12 err, no 13:27:29 --- quit: saon ("leaving") 13:27:39 --- join: saon (1000@c-66-177-224-130.hsd1.fl.comcast.net) joined #forth 13:30:42 Hi again, saon 13:36:03 teh hies saon 13:40:25 --- quit: qFox ("this quit is sponsored by somebody!") 13:42:26 arke thanks 13:42:46 Herkamire: :) 13:51:03 hey all 13:51:25 had to move some cabling around, bumped my connection to the cable modem 13:52:37 :) 13:59:00 but now i can play 2 video games at once, so, i think it's worth it 14:01:50 HAha :) 14:21:19 --- quit: Frek (Read error: 104 (Connection reset by peer)) 14:21:45 --- join: Frek (~anvil@h254n2fls31o815.telia.com) joined #forth 14:47:15 wow. 14:47:23 I have enough trouble playing one video game 14:52:32 :) 14:56:43 --- join: danniken\ (CapStone@ppp-70-247-75-6.dsl.ltrkar.swbell.net) joined #forth 14:56:48 --- part: danniken\ left #forth 14:57:31 --- nick: nanstm -> tiff 14:58:11 --- join: Al2O3 (~Al2O3@c-24-1-126-202.hsd1.tx.comcast.net) joined #forth 15:05:49 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth 15:05:58 on powerpc, what's the best way to set the low 3 bits of a register to zero? 15:07:42 i have to go now ,but if somebody provides an answer, i'll read the logs 15:07:48 --- quit: slava ("USE: factor") 15:10:59 slava, probably to and with 0xfffffff8 15:23:47 --- quit: arke (Remote closed the connection) 15:23:47 --- join: arke (f2@bespin.org) joined #forth 15:43:46 --- quit: Topaz (Remote closed the connection) 15:53:59 on ARM it's easy. since you can do small positive and negative numbers 15:54:24 hi all 15:54:29 Hi. 15:54:30 hi 15:56:30 --- quit: KB1FYR (Read error: 110 (Connection timed out)) 16:01:55 slava: clrrwi rD,rA,3 is probably best. 16:02:06 That's CLeaR Right bits of Word (Immediate) 16:02:23 and is an extended mnemonic for rlwinm rD,rA,29,31. 16:02:51 rlwinm = cool instruction 16:05:49 more generally, clrrwi rD,rA,N = rlwinm rD,rA,32-N,31. 16:05:55 arke: damn right :) 16:06:01 rlwimi is cool too 16:16:26 whats that? 16:16:36 rtate left with immediate mask immediate? 16:23:40 rotate left word and insert with mask, something like that. 16:23:59 rlwimi rD,rA,shift,mask_hi,mask_lo. 16:24:08 rotates rA left by `shift' bits 16:24:24 then inserts bits of rA from mask-hi to mask-lo into rD. 16:25:19 andi or rlwimi both works; the first is better if one care about performance on all PowerPCs 16:27:12 andi. won't work if you want all of the high bits. 16:27:18 it only uses a 16-bit immediate. 16:27:23 right? 16:47:28 --- quit: virl ("Verlassend") 16:49:14 tathi: yes 16:50:05 have to use and then and and a precomputed register constant 16:50:17 however rlwimi would be "neater" 16:50:51 problem is just if you're targeting the PowerPC 970 rlwimi will preform badly as it's one of the cracked instructions 16:51:28 so it uses atleast 2 cycles to executed; rlwimi on g4 is 1 cycle, just like and/andi 16:51:40 -d 16:52:46 oh, ok. 16:53:27 still, I wouldn't expect it to be slower than lis; addi; and; ... is it? 16:55:56 no 16:56:59 and and a precomputed register constant would be faster; but it would mean you have to sacrifice one register (or load the constant a few instructions ahead). 16:57:24 but on the G4 it doesn't matter; and rlwimi would look better 16:57:42 G4/G3 I mean 16:59:48 gotch 16:59:50 a 17:20:23 --- nick: tiff -> nanstm 17:20:28 Frek, what do you mean "it's one of the cracked instructions" ? 17:20:43 * OrngeTide didn't know 970 would run things slower than older procs... 17:21:06 OrngeTide: it means the 970 don't natively implement it; so it translates it into two microcode instructions 17:21:35 ouch. 17:21:50 was this to keep the size/complexity down? 17:21:58 probably 17:22:21 it seems like a useful instruction and something that wouldn't be that hard for a compiler to generate. 17:22:30 same goes for virtually any instruction that does post update 17:22:33 like lwzu etc 17:22:42 oh. I see. 17:22:56 yes rlwimi is very useful 17:23:10 if not one of the most powerful PPC instructions at all 17:23:45 makes me wonder if i wouldn't be better off with a G4 with a fast bus rather than a G5. 17:24:11 well problem is the G4s can't handle a fast bus ;) 17:24:59 but honestly the G5 isn't really faster than the G4 on a mhz per mhz basis 17:25:12 but it can deal with a fast bus and it can be clocked higher 17:25:18 yea. it's too bad. it could be really competive with intels junk if ibm or motorola would make them with 533Mhz fsb or something 17:25:24 and the FPU is very strong on the G5 17:25:31 yea. that's true. 17:25:51 but we won't see G5 on a laptop. which is why I wish G4 could do things that it can't 17:26:11 yeah unfortunately Apple is stuck inbetween the companies 17:26:42 Motorola pays most attention to the embedded industry; and IBM pays the most attention at (ironically) the embedded industry as well (although game consoles in this case) 17:27:29 however last I heard Motorola got cold feet when apple turned to IBM for their G3 and G5 cpus (apple don't use the G3 anylonger though) 17:27:47 what about POWER, does ibm not really care about it's servers? 17:27:59 I thought PWER = PowerPC? 17:28:02 OrngeTide: of course they do; but I'm talking about PowerPC now 17:28:17 arke: POWER has the PowerPC ISA these days; but it's not the same CPU really 17:28:32 I thought G5 was just a POWER chopping in half? 17:28:37 PowerPC 970 is a POWER4 lite sort of 17:28:38 s/chopping/chopped 17:28:49 but then they patched it up with some "PowerPC" capabilities and so on 17:29:05 I see. and Apple is basically the only customer for that... 17:29:24 however strictly speaking the PowerPC 970 shouldn't be considered as a PowerPC as it don't follow the PowerPC specification 17:29:43 most noticeable is the drop of support for little endian 17:29:56 ah. just as well. 17:29:56 but then again it can be argued that it was hardly used at all 17:30:21 but it's not really an optional feature of the PowerPC platform; it's defined to be biendian 17:31:10 mips uses little endian a lot in smaller embedded systems, and big endian in higher end servers. the chip can go either way, I think it's just easier to talk to PCI when you're little endian. 17:31:24 yeah PCI is little endian 17:32:02 * OrngeTide did some mips r4kc work and some PowerQUIIC work (850&860). 17:32:04 but I can't recall it being used a lot on the PowerPC; now I don't know about the numerous embedded systems that uses PowerPC though. 17:32:36 but for desktop use I just know two things that used it; one being the now discontinued Windows NT for PowerPC, and the other being Connectix (now Microsofts) VirtualPC 17:32:48 well if the offerings of the embedded linux companies is any indication, little endian powerpc is extremely unpopular. 17:33:06 I see 17:33:09 NT for powerpc was little endian? odd. 17:33:18 yep NT for PPC was little endian 17:33:32 it probably eased porting too 17:33:47 NT for Sparc was big endian... 17:33:56 but the PowerPC was never truly little endian though. 17:34:06 OrngeTide: ah I see; yeah I donno why 17:34:39 although the sparc port was just the kernel and some minor apps and was never released. it was an R&D project that got canceled after a successful demo 17:35:31 i see 17:35:40 probably because Sun, being the random company that it is, tried to setup a deal where they'd use NT on Sun workstations. but then changed their mind. 17:35:54 but that's just a guess on my part. 17:36:16 yeah but there's lots of odd ball decisions in this industry :) 17:36:40 * OrngeTide wishes he didn't break his ARM dev kit. 17:37:27 it will be interesting to see what happens in the next few years though 17:37:30 if someone had a powerpc dev kit that was cheap as these arm kits, I'd totally buy that to make up for my broken ARM :) 17:37:40 I'm confident the x86 will go away within 5-7 years 17:38:03 you don't think amd64 will allow x86 to hold out for another couple decades? 17:38:25 intel tried to make a replacement for x86, at HP's expense. and it failed miserably... 17:38:26 OrngeTide: not unless they plan to drop backwards compatibility at some point 17:38:44 yeah the problem is always backwards compatibility 17:38:54 just microcode it. "legacy" apps can just run slow. 17:39:04 yeah 17:39:13 or it could even be handled in a software emulator 17:39:20 like Apple did when they went from 68k to PowerPC 17:39:26 yup. 17:39:52 but its probably better the earlier they implement it 17:39:53 --- nick: nanstm -> Raystm2 17:39:55 although x86 has been pretty rotten at emulation. PPC could run 68k apps quickly though. 17:40:14 OrngeTide: yeah; but I'm mostly talking about a successor to the x86 platform 17:40:18 Frek, I think the industry will hold out for as long as possible and change over at the very last moment. kind of like IPv6 17:40:32 probably 17:40:43 Frek, maybe Intel will license powerpc. :) 17:40:51 but it feels like x86 been stomping on the same spot for a while 17:40:56 if Sony Cell takes off, we might have powerpc everythign. 17:41:13 OrngeTide: in exchange for some of their patents I'm sure motorola and IBM/Apple could consider that 17:41:15 well people have been saying x86 should be dead back when 486 was new. 17:41:27 OrngeTide: yeah I'm aware of that 17:41:52 if IBM and Motorola were mainly interested in embedded, they probably wouldn't care if apple started buying chips from intel. *shrug* 17:41:55 and it's no doubt the x86 engineers did an amazing job keeping it alive and "cutting edge" as well 17:42:09 yea. they've basically made a pig fly. 17:42:14 yep 17:42:33 it would be like towing a trailer with a 300hp moped. 17:42:40 also lots of interesting processor technologies been initiated as optimizations for the x86 17:43:07 yea. I suppose that's the advance of starting with a less than ideal architecture. 17:43:07 OrngeTide: no but they would probably care if Intel start pumping out low price PowerPCs 17:43:24 yea 17:43:35 yea. intel is big in the embedded market. it would really hurt ibm and motorola. 17:43:38 however at the sametime I'm surprised how early the PowerPC "hit the roof" 17:44:11 basically there isn't enough money in making a screaming desktop powerpc. 17:44:26 the market is for low-power or for scalable servers. nothing in the middle. 17:44:30 you're probably right 17:44:40 also the scalability is a problem 17:45:13 so far most PowerPCs been useable in everything from embedded to desktops just by adjusting the clock etc 17:45:15 doesn't it have hyper transport? should scale to 8-way NUMA easily. 17:45:43 ya but I meant more in terms of powerconsumtion/heatgeneration :) 17:46:12 ah. yea. well that's managable in a high-end cluster. 17:46:16 the PowerPC 970 is a very power hungry chip, still it's considerably less complex than a x86; which is quite odd 17:46:26 unlike my AMD Duron which just runs 90C all the time. 17:46:38 well my G5 usually lies around 50C 17:46:52 the G4 was around 30-35C 17:47:03 well a lot of it has to do with how you are driving your transistors. 17:47:11 it probably does 17:47:32 my mac mini is 45C, but I don't have air conditioning in my apartment. in the cooler months it's 35C 17:47:40 ah 17:48:25 however me personally I don't care if the processor would be 200C aslong as it performed well :D 17:48:37 and the cooling system didn't sound like an airplane 17:48:57 I personally wish to see cell based macs; but it's probably just a dream 17:49:40 but then again what makes the cell so strong is not it's powerpc heritage but all these SPUs 17:49:58 same as the Emotion Engine; the MIPS core was quite weak, but hell the vector units could rock 17:50:48 yup. 17:51:09 and really for games that's what you want. your game logic isn't a big hit. it's all the multimedia processing 17:52:26 ya 17:52:38 but that's what makes the cell so interesting 17:52:48 the Emotion Engine only had Vector DSPs 17:52:52 look at a really old console: a 6502 and then some custom engines to handle the multimedia. sure it was just some simple FM sounds and a Sprite engine. but the games could be pretty complex on a 2Mhz 8-bit processor. 17:52:59 the SPUs on the cell is general purpose processors 17:53:04 Frek, yea. the SPUs are very cool. 17:53:17 fat bus and generic enough to do lots of things. 17:53:28 the PowerPC seems to be more of a controller chip in the cell than a "master cpu" 17:53:38 as the MIPS was in the EE 17:53:58 yeah consoles always been cool like that 17:53:59 yea. the powerpc is just there to feed data and instructions 17:54:14 i think it's interesting that all the new consoles are powerpc based. 17:54:27 but then again the amiga pretty much showed that the same thing could be applied to a "general purpose computer" as well 17:54:47 well I donno if that's as interesting 17:54:57 I mean there's basically just two options x86 and PowerPC 17:54:59 amiga was a nice machine. but the company was terrible. 17:55:05 yes 17:55:09 and it still is 17:55:13 --- quit: tathi ("leaving") 17:55:18 Frek, you don't think mips is an option anymore? 17:55:18 just like that other amiga wannabe company genesi 17:55:27 OrngeTide: I consider MIPS pretty much dead 17:55:29 ARM is nice. but the speed isn't there. 17:55:47 Frek, utoh. cuz Cisco wants me to do some work on some dual-core mips:) 17:55:53 :) 17:56:13 well I could be wrong; but I got the feeling no major work is being done on the MIPS at all these days 17:56:13 they want to run Java on them... 17:56:32 mips is still around for embedded. 17:56:41 ah 17:57:04 mainly because licensing is relatively cheap. (versus ARM or PowerPC) 17:57:17 yeah I shouldn't talk alot about MIPS though I haven't really used anything but SPIM/SAL and r5900 when I did some PS2 stuff 17:57:30 yea. that's old stuff. 17:57:49 people are putting r12k's in networking appliances now. 17:58:06 but I never really liked the MIPS that much I donno why though :) 17:58:22 but one thing I react on was that you had to maintain "a link register" on your own 17:58:40 mips is the truest RISC. which is also why it hasn't really stayed in step with current advances in architectures. 17:58:51 i see 17:58:58 mips is simple and fast. being in #forth you should appreciate it:) 17:59:04 :) 17:59:06 :D 17:59:06 it's not simple to code for though. ehhe 17:59:10 hehe 17:59:14 branch delay slots are cool 17:59:23 instruction pokes are really fun 17:59:42 OrngeTide: speaking of which it's quite odd I'm here at all considering I hardly do any forth dev. 17:59:49 older mips forced you to manage your own tag ram for cache. and even fairly recent mips force you to do your own tlb cache management 18:00:03 Frek, I like forth but I don't get to do much in it. 18:00:19 I like to pester I440r about isforth though. 18:00:40 arke, ehhe those are the best! 18:00:45 OrngeTide: I think a started to hang out here when I did some openfirmware projects a few years ago, then I noticed there was some other interesting discussions at time so I believe that's why I stay 18:01:33 I'm conflicted. as much as I like forth. I haven't found anything better than C for doing device driver work. :( 18:01:40 which is sad because C really bites. 18:01:44 Heh. 18:01:46 but it has nice tools:) 18:02:03 I like C better I think; but that's probably because I never really got into forth way of thinking and so on 18:02:22 --- join: KB1FYR|LT (~Alex@196-220.suscom-maine.net) joined #forth 18:02:47 I would like some static type checking and other magic in forth. combine that with interactive forth and you have enough tools to beat C to a pulp. but most forthers think my ideas are perverse. 18:03:09 forth is awesome at testing and diagnosing code. 18:03:17 heh it's probably the strict typing I hate the most about C :P 18:03:32 C's typing is lazy/sloppy 18:03:50 compared to ObjC it's very strict 18:04:09 a good system would enforce strict typing. and you'd be able to write compile-time code (like a forth immediate) to deal with type conversion. 18:04:22 you do ObjC too? 18:04:24 OrngeTide: python ? 18:04:31 or that's relaxed typing ? 18:04:34 I don't remember 18:04:34 well I mean a real programming language :P 18:04:50 OrngeTide: that's what I do the most; as I'm Mac based 18:05:21 ObjC is actually strong dynamic typing. which is why it's so easy to use. 18:06:15 hmm I never heard that term before; just strict typing like C++ and dynamic/relaxed typing 18:06:38 Frek, i'm just learning Objc. I did a bunch of OpenGL stuff using AGL but I'm converting it to Cocoa. (my program actually compiles on X11/*nix and soon Windows as well. and just replaces the UI with whatever is appropriate for the platform) 18:06:51 ah 18:07:16 well you can have weak dynamic typing. strong dynamic typing. weak static typing(that's what C is). and strong static typing(that's sort of what C++ is) 18:07:32 oh dear god 18:07:40 C++ is almost as lazy as C 18:07:42 forth is typeless, as far as I can tell:) 18:07:48 arke, almost:) 18:07:49 yep 18:07:49 what's the difference between weak dynamic and weak static typing ? 18:07:53 that's why I said "sort of" 18:08:04 arke, C++ doesn't have a real generic pointer. so it's slightly stronger. 18:08:27 OrngeTide: thats what char* + typecasts are for 18:09:13 well typecasting makes it strongly typed. 18:09:46 Frek, well dynamic typing means the type exists at run time. and static means that it must be known at compile time. 18:10:05 OrngeTide: yeah 18:10:06 Frek, so a weak dynamic type would mean you can implicitly change type at run time. 18:10:21 okay 18:10:32 int a; unsigned b; long c; a=b=c; /* weak static typing */ 18:10:44 and a strong dynamic typing language would not really have any types at all ? 18:11:20 Frek, sure it would. every "object" in the language would have a type. and if you want to go from one type to another you have to call some sort of convertion function or method. 18:11:40 ok ok 18:11:51 struct { int type; union { ... }; }; ... if(a->type != b->type) abort(); 18:12:04 that would simulate a strong dynamic type system. 18:12:11 okay 18:12:32 i'd like a forth with strong static typing. and for dynamic you can just roll your own to suit your needs. 18:13:19 i see 18:14:29 example. 5 7 + . ( good ) 5.1 6.9 + ( also good ) 5 6.9 + ( compilation error ) 5 ->f 6.9 + ( good, you converted the int to a float ) 18:15:04 a week type system would have just guessed at what types to use and probably upgraded them both to floats. this is what C does. 18:15:12 and it's sometimes a source of bugs. 18:15:13 how does forth deal with like "5 6.9 +" today ? 18:15:21 it doesn't. 18:15:24 i see 18:15:47 5.1 6.9 F+ ... so you have a completely different name for it. 18:15:57 okay 18:16:03 there is something to convert an int to a float. but I forget what it's called. 18:16:35 alright 18:16:47 the ugliness with a type system in forth is you need to have a bunch of different ways to specify literals. 18:17:22 yea i guess 18:17:28 because how would you end a double-cell number in versus a single-cell. you start getting all these ugly modifiers like lisp and scheme have. 18:17:40 s/you end/you specify 18:17:45 brain fart :) 18:17:55 I read it as "specify" anyway :) 18:18:02 haha 18:18:17 i may have originally ment "enter" 18:18:22 :) 18:18:51 well either-way it's starts getting really late; but it's been interesting :), good night. 18:18:57 anyways do you talk to Herkamire much? he does some PowerPC stuff. has some nice forth 18:19:07 yea. i gotta go too. ttyl 18:19:20 OrngeTide: nope I haven't talked to Herkamire (at all I think) 18:19:26 yep ttyl 18:19:44 search for HerkForth on the web. 18:25:34 --- quit: KB1FYR|LT (Read error: 110 (Connection timed out)) 18:26:16 who me? 18:27:26 http://herkamire.com/jason/herkforth 18:28:58 man, just too late 18:29:08 is Frek doing cool stuff on PPC? 18:31:36 --- join: I440r (~mark4@rrcs-24-242-160-169.sw.biz.rr.com) joined #forth 18:32:45 --- quit: madgarden (Read error: 54 (Connection reset by peer)) 18:33:14 --- join: madgarden (~madgarden@Toronto-HSE-ppp3708121.sympatico.ca) joined #forth 18:36:10 Herkamire, frek is a powerpc whiz. he answered a bunch of cool ppc asm questions :) 18:36:35 * OrngeTide wonders if Frek would like to work at Cisco with him. 18:36:37 --- quit: madgarden (Read error: 104 (Connection reset by peer)) 18:36:45 --- join: madgarden (~madgarden@Toronto-HSE-ppp3708121.sympatico.ca) joined #forth 18:36:49 --- quit: madgarden (Read error: 54 (Connection reset by peer)) 18:36:58 --- join: madgarden (~madgarden@Toronto-HSE-ppp3708121.sympatico.ca) joined #forth 18:37:37 --- join: KB1FYR|LT (~Alex@196-220.suscom-maine.net) joined #forth 19:14:40 --- join: KB1FYR (~Alex@196-220.suscom-maine.net) joined #forth 19:17:04 --- quit: saon|smgl (Read error: 60 (Operation timed out)) 19:29:15 --- join: I440r_ (~mark4@rrcs-24-242-160-169.sw.biz.rr.com) joined #forth 19:35:22 --- join: saon|smgl (~saon@c-66-177-224-130.hsd1.fl.comcast.net) joined #forth 19:40:24 --- quit: saon|smgl ("Lost terminal") 19:43:06 --- join: saon|smgl (~saon@c-66-177-224-130.hsd1.fl.comcast.net) joined #forth 19:46:55 http://fredsmith.us/mirror/funny/ch_ayb.jpg 19:52:18 --- quit: vitaminmoo ("2.6.12-rc4 -> rc5") 19:54:41 --- join: vitaminmoo (~vitaminmo@dsl-95-113.peak.org) joined #forth 20:31:18 --- quit: warpzero (Read error: 104 (Connection reset by peer)) 20:32:46 --- join: warpzero (~warpzero@wza.us) joined #forth 21:16:49 --- quit: I440r ("Leaving") 21:25:21 --- quit: KB1FYR|LT (Read error: 60 (Operation timed out)) 21:59:06 zoly around? 21:59:52 3 weeks ago :( 22:00:46 :/ 22:11:43 * Al2O3 wonders if zoly is working on the ruby port of some 4th system :) 22:11:51 saw a post of his online that mentioned his interests. 23:59:59 --- log: ended forth/05.05.26