00:00:00 --- log: started retro/06.07.27 03:03:50 --- join: Raystm2 (n=NanRay@adsl-68-95-248-185.dsl.rcsntx.swbell.net) joined #retro 03:11:50 --- quit: Raystm2- (Read error: 145 (Connection timed out)) 03:19:52 --- quit: Raystm2 (Read error: 104 (Connection reset by peer)) 03:20:05 --- join: Raystm2 (n=NanRay@adsl-69-149-47-61.dsl.rcsntx.swbell.net) joined #retro 03:42:13 --- quit: tathi ("leaving") 04:09:39 --- quit: crc (Read error: 110 (Connection timed out)) 06:02:22 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #retro 06:18:46 --- join: timlarson_ (n=timlarso@65.116.199.19) joined #retro 06:45:59 --- join: nighty_ (n=nighty@66-163-28-100.ip.tor.radiant.net) joined #retro 07:32:58 haha! my new version of my forth is now better than the old version, horray! 07:33:05 :) 07:33:12 well done 07:35:09 cool! 07:35:31 so, now I can hopefully start to include useful things like if, then and so on.. 07:35:49 perhaps, xell won't be a vapoware thing. 07:36:39 virl, I started working with retro IA-32 assembler. 07:37:03 ok, so you write a cool assembler? 07:37:07 yep. ;) 07:37:36 I gues I use the BSD lisence, if somebody else wants it as well. 07:37:45 hopefully it will bang, and then it's possible to go away from x86 specific code. 07:38:06 I propably get it bang, I've very big hopes for that. 07:38:55 Intel just made a fun puzzle. But it is not preventing me from building a cool assembler. 07:39:05 that would be nice to see, when my favorite text-forth comes to new levels =) 07:40:16 virl: I've very big plans ahead, so your favorite text-forth may turn upside down yet. 07:41:52 you've been helped me a lot, it is nice that I can at last do something back for a change. 07:42:46 Also happened to learn one thing. Forth syntax kicks lisp's butt! 07:42:51 :) 07:48:31 syntax? what syntax? :) 07:49:20 one kind of you end up with accidently by following forth semantics. ;) 07:58:33 hehe 07:59:30 did you see my paste yesterday that lets you use arithmetic words as prefixes for variables? 08:00:53 nope, show it. 08:01:05 http://retroforth.net/paste/?id=69 08:01:19 too long to show in irc 08:01:38 desperately needs some cleaning up, but it works :) 08:01:59 show example about use. 08:02:21 10 +foo @foo +foo 08:02:32 lol, cool! 08:02:34 :) 08:03:33 well, could you do sth. like this? hex 10 !FFFF with your thing. 08:03:35 the possibilities of customizing the "word?" word are really amazing... 08:04:37 yeah stuff like hex 10 +foo 08:05:26 or just use the hex prefix: $10 !foo @foo 08:06:08 @ and ! prefixes are already in the current repo, they're crc's work. 08:06:17 ah. 08:06:26 (as is the $ hex prefix) 08:06:55 hey! maybe you could be able to make a command which allows you to write easily more prefixes. 08:07:16 * docl has been considering this :) 08:07:37 like: prefix: c@ and here what is happening for the word, plus ; or ;prefix 08:07:55 prefix: ( n "define" -- ) 08:08:38 yeah ; would terminate most likely 08:08:51 just like the new class-defining words 08:09:19 '# prefix: @ # ; 08:10:25 :) 08:10:37 maybe you could make " prefix as well. 08:10:55 so people could write: ."blah blah" instead of ." blah blah" 08:11:04 there's a thought 08:14:36 docl, does which -word still work? 08:16:02 as in 10 -foo ? sure 08:16:27 I mean 'which' -word 08:16:40 hmm not sure how that works 08:16:40 or am I just trying to use it wrong way? 08:16:47 which :name unpack type 08:16:51 gives garble. 08:17:13 hmm -foo isn't a normal word 08:18:38 x' -foo shows you that, by returning a string and a false value 08:19:13 what word? does is handle non-defined words. if there is a defined word that matches -foo, it won't use the prefix 08:22:40 '" case: name pad swap copy true break <-- permits "hello" 08:23:30 hmm not as good as I thought, can't use multiple words 08:25:19 I would find it cool when the assembler isn't so 'assemblerish' 08:25:30 eh wouldn't 08:26:56 virl, shoot out what kind of syntax are you thinking of, I've already quite good ideas in mind, but more wouldn't be bad. :) 08:27:32 '" case: name pad swap copy 1+ dup allot 32 pad 1- c! x: " nip over negate allot + true break <-- messy hack, does the job :) 08:30:09 hmm, perhaps you have variables called the registers, so EAX and so on. on them you could work with @ or ! 08:30:53 uhm, no, that's not a good one. 08:31:21 Cheery, well, it seems that I said it only to have really ideas, sry :-| 08:31:30 an encoding checker that sees if the operand fits the instruction 08:31:33 btw. Cheery what are your ideas about it? 08:31:45 My first goal is to make an encoder which allows me to write instruction definitions in simpler form than what hexes are. 08:32:21 so an encoding-maker 08:32:36 second goal is to make some cool macro stuff over that, to spice up. 08:32:59 the instructions can all be stored in constants 08:33:06 so say instruction blah, opcode: blah ; 08:33:58 I think it is slightly more complicated. 08:34:08 which appeared in reaching the goal 0 08:34:24 goal 0: find out how IA-32 instruction format works (been there) 08:34:34 But not too much. ;) 08:34:41 * docl too :) 08:35:19 er I've been trying to figure out 386. not sure about IA-32 08:36:06 same thing I guess 08:36:28 y. 08:37:21 with some nasty extensions like MMX and so on.. 08:37:27 3dnow 08:37:47 and another one which plays on the newer ones. 08:42:17 What is MMX and 3dnow, I've not studied them yet much. 08:42:18 ? 08:43:46 MMX is a vector extension for the intel cpus and 3dnow is exactly the same with some added stuff, but for AMD cpus. 08:44:29 How do I detect if they are on? 08:44:33 so for multimedia apps and so 08:44:55 I've heard that requires some special support from program to use them, am I right? 08:45:17 on? they are built into the cpu, that are opcodes extensions. and probably you can detect them with cpuid 08:45:18 And they can speed up code a lot with some things. 08:45:24 ah, ok. 08:45:39 yeah, wisely used. 08:45:43 I just meant if they are at all in the cpu. 08:45:43 (if) 08:45:47 ok. ;) 08:45:54 I study them later. 08:46:16 when you have an intel cpu say a pentium III then you have 100% the MMX extension. 08:46:17 Now it is the most important for me to make this encoder. 08:46:40 but not the 3dnow, because that's only for AMD cpus ;-) 08:47:12 and iirc, AMD moved in their newer cpu line away from 3dnow and only supports MMX 08:48:05 So it is the MMX which is more important. Anyway, it doesn't matter, I believe. 08:48:24 yep MMX is more important 08:49:43 or SSE(that's the newer MMX)(from intel) 08:50:14 is simply an extension of MMX 08:51:38 or the floating point stuff is also sick. I mean I can do tan in hardware. it's something I say 'wow' on one side and on the other 'what a bloat' 08:52:31 it's like lisps: 'wow nice features' on one side and on the other 'bah, such a painfully syntax' 08:52:40 :D 08:53:00 Well, I think I'll have fun with it at some point. 08:53:22 Damn, this encoder's code will look nice. 08:54:57 and I'll be happy when I see it 09:28:18 Now I've simplest form of opcode field solved in my system. I think next I should make it able handle simplest of registers. 09:49:46 --- quit: virl (Remote closed the connection) 10:37:05 Huh, the intel instruction format surprized me a little, I must think it a little. 11:59:25 --- quit: docl (Read error: 110 (Connection timed out)) 12:03:22 --- join: docl (n=docl@74-36-210-133.dr01.hmdl.id.frontiernet.net) joined #retro 12:33:37 --- join: snoopy_1611 (i=snoopy_1@dslb-084-058-175-252.pools.arcor-ip.net) joined #retro 12:36:22 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 12:36:31 --- nick: snoopy_1611 -> Snoopy42 13:09:54 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #retro 13:27:12 here's an option for those who want to copy and paste screens directly: http://retroforth.net/paste/?id=72 13:42:00 --- quit: neceve (Remote closed the connection) 13:49:05 --- quit: timlarson_ ("Leaving") 14:29:01 --- join: virl (n=virl@chello062178085149.1.12.vie.surfer.at) joined #retro 14:37:39 --- part: Cheery left #retro 15:04:49 --- quit: Ray_work ("User pushed the X - because it's Xtra, baby") 15:30:03 --- join: crc (n=crc@pool-70-110-183-96.phil.east.verizon.net) joined #retro 15:30:22 good evening 15:30:37 should I read the channel log? 15:30:44 --- mode: ChanServ set +o crc 15:31:04 sure 15:31:21 * docl has been playing with prefixes 15:31:38 written any new ones? 15:31:50 + - / * 15:31:57 I see that 15:32:13 * crc notes that the code in paste#69 is kind of messy.... 15:32:22 yupyup 15:33:25 http://retroforth.net/paste/?id=73 15:33:30 should work 15:35:05 * crc will attempt to allow runtime addition of prefixes 15:35:42 also, today I came up with a way to copy and paste screen views directly 15:37:04 http://retroforth.net/paste/?id=74 15:45:28 --- quit: nighty_ (Read error: 113 (No route to host)) 16:09:48 almost done 16:13:58 docl: new prefix system is now in the repo 16:14:08 this one allows for a form of: 16:14:16 'x prefix: .... ; 16:14:30 to define a prefix definition for 'x' 16:25:35 neat! 16:38:45 '_ prefix: entry as .data 0 , ; 16:47:44 nice to create variables on the fly in the middle of code. 16:48:45 i was gonna say something about that yesterday but didn't think it possible. 16:51:31 hmm, can't make them in the middle of a define though 16:52:55 that's because of memory issues though 16:56:58 I'll take your word. I'm not seeing it yet. :) 17:10:34 docl: maybe it can be done... 17:10:43 * crc thinks on this a bit 17:17:45 the new prefix stuff is cleaner and more flexible than the old approach 17:23:21 Would you have to include a memory area where there could be say, room for x variables and then have a prefix that uses that area to define variables on the fly? 17:23:57 maybe just put a jump over the allocated variable space :) 17:24:06 whoa! yes 17:27:09 wow, looks much cleaner :) 17:30:08 http://retroforth.net/paste/?id=75 17:30:24 can create a variable inside a definition 17:30:38 (visible outside as well, but that's life....) 17:32:53 I guess I'll have to push 9.2.3 out soon... 17:34:06 visible outside is a side benifit. 17:35:09 a feature :) 17:35:55 there's an xt left on the stack 17:36:44 I did that on purpose 17:37:04 ok, cool 17:37:22 remove the dup in the [[ r> .... ]] set to get rid of it if you want 17:38:00 * crc reasoned that you wouldn't want to say : foo _bar !bar ; as something like that is pretty redundant 17:38:28 good point 17:48:55 * crc has started a new editor tutorial: http://retroforth.org/learn.editor 17:49:33 --- part: crc left #retro 17:49:41 --- join: crc (n=crc@pdpc/supporter/active/crc) joined #retro 17:49:41 --- mode: ChanServ set +o crc 18:00:06 That link doesn't work. 18:00:33 yes it does 18:00:35 http://retroforth.net/paste/?id=76 <- lets you page through the tutorial easier 18:00:46 the file is a blockfile for loading into the editor 18:01:06 Raystm2: it's a block file, so you'll want to dl it and load it in rf 18:01:53 * docl is loving the new d: 18:02:09 ' n is nothing 18:02:11 ~n is nothing 18:02:26 it's my irc client, still broken and I just created this one yesterday. I'm gonna have to do it again, cuz now i cant' click links 18:02:46 oh yeah, ~n 18:02:56 * crc uses opera as his irc client 18:03:12 for windows? 18:03:42 windows, linux, and freebsd :) 18:04:34 cool, think i'll grow up and get a big boys irc. 18:04:58 Cuz this toy i'm using is losing. 18:05:21 what are you currently using? 18:05:36 MeGALiTH's Visual IRC 2.0 18:05:50 ok 18:05:51 sounds ominous 18:06:05 it was free, i found it first. 18:06:08 on windows, the only tolerable IRC clients I have found are mirc, chatzilla, and opera 18:06:27 you like opera best? 18:06:31 irssi doesn't work on windows? 18:06:36 or just testing it? 18:06:44 docl: I don't know 18:06:48 I've never tried irssi on anything but linux. 18:06:51 Raystm2: I like it best overall 18:07:10 thanks for rec. will get an opera client. 18:07:22 besides, where else can I browse the web, chat, read/write email, all in one nice small package? 18:08:06 oh wait maybe my problem is with FireFox. It just downloaded an update but I havn't dissmissed the window i've been using. 18:08:19 ahh, that could do it :) 18:09:15 no I definately can not bring up a right-click menu in the irc 18:09:28 that's a tad annoying. 18:10:25 still I could double click the link and it's getting thru to the browser but i'm not getting any editor information, just a black screen, where as the pastes are loading just fine 18:11:57 does paste 75 create a variable named _name or name ? 18:12:02 name 18:12:15 very cool. 18:15:17 ut oh turns out I cain't right-click anything. 18:15:55 only one way to fix a windows box 18:16:01 treat it like it's crashed. 18:16:04 --- quit: Raystm2 ("Should have paid the bill.") 18:18:45 --- join: Raystm2 (n=NanRay@adsl-69-149-47-61.dsl.rcsntx.swbell.net) joined #retro 18:33:35 --- quit: docl ("leaving") 18:34:18 --- join: docl (n=docl@74-36-210-133.dr01.hmdl.id.frontiernet.net) joined #retro 18:34:44 * docl is using opera chat now 19:21:19 --- join: nighty_ (n=nighty@CPE00119576a9c5-CM0012c90d36fc.cpe.net.cable.rogers.com) joined #retro 19:21:42 --- quit: nighty_ (Remote closed the connection) 19:24:49 --- join: nighty_ (n=nighty@CPE00119576a9c5-CM0012c90d36fc.cpe.net.cable.rogers.com) joined #retro 19:25:02 --- quit: nighty_ (Remote closed the connection) 19:48:02 --- quit: ChanServ (Shutting Down) 19:48:40 --- join: ChanServ (ChanServ@services.) joined #retro 19:48:40 --- mode: irc.freenode.net set +o ChanServ 20:07:30 --- part: docl left #retro 20:09:10 --- join: docl (n=docl@74-36-210-133.dr01.hmdl.id.frontiernet.net) joined #retro 20:29:40 --- quit: ChanServ (Shutting Down) 20:30:15 --- join: ChanServ (ChanServ@services.) joined #retro 20:30:15 --- mode: irc.freenode.net set +o ChanServ 20:34:14 --- quit: ChanServ (Shutting Down) 20:34:48 --- join: ChanServ (ChanServ@services.) joined #retro 20:34:48 --- mode: irc.freenode.net set +o ChanServ 20:56:31 --- quit: Shain (Remote closed the connection) 20:59:23 --- join: Shain (i=steve@c-67-161-56-76.hsd1.ca.comcast.net) joined #retro 21:24:24 chess now has castleing and en passant. 22:00:28 http://retroforth.net/paste/?id=77 current snapshot 23:31:19 pawn promotion next. 23:59:59 --- log: ended retro/06.07.27