00:00:00 --- log: started retro/06.07.11 03:34:19 --- quit: crc (Read error: 110 (Connection timed out)) 03:59:14 --- join: crc (n=crc@pool-70-110-173-103.phil.east.verizon.net) joined #retro 03:59:21 --- mode: ChanServ set +o crc 04:18:23 --- join: nighty (n=nighty@66-163-28-100.ip.tor.radiant.net) joined #retro 05:57:19 hmm, a direct xbindings would be also nice to test. 05:57:38 so, going through the network arch of X 05:57:54 using netclient.f to open a socket and then talking directly to X 07:04:33 do you have netclient.f working under libc yet? 07:06:20 wait, that may not be necessary. you are planning to use rf-linux to talk to the X server? that is a cool idea. 07:51:49 it's just a matter of sending the right messages to the right port on the localhost? 08:22:56 yeah... 08:23:08 xlib does nothing different 08:23:24 here you call to X with an unix socket 08:24:29 but before I could do that, I think I need to get my C program working which does that ;-) 08:25:30 because X has a very ugly protocol 08:28:30 ahh, I see :-) 08:29:16 it would be nice to see the guts of the socket calling part in forth, just the same. 08:29:59 nevertheless it has a couple of nice features, for example you can request from the X server which endianess you want. 08:30:35 'the socket calling part'? 08:31:05 do you mean the socket syscall? and when yes, then we need to replace the linux kernel with a forth system ;-) 08:31:16 yeah :) 08:31:33 we'll get to it, eventually :) 08:32:03 well well... but I think then we need all those syscalls 08:32:27 to implement 08:33:15 we need the multitasking stuff, we need the driver interface and so on. or we simply say, fuckoff unix and create our own Forth OS. well then we have retro native. 08:33:40 so we need simply all those fancy drivers which exist for linux. 08:34:47 yeah. well, forth is great for reverse-engineering drivers and such, I hear 08:35:08 oh? from whom did you hear that? 08:35:10 the interactivity makes it simpler to probe devices and figure out how they work 08:35:15 can't remember where 08:37:46 * docl is using a nicer font now in xterm -- lynx actually looks good! :) 08:38:18 X is so crazy with it's bitmap fonts 08:38:51 yeah there's all sorts of UGLY ones. hard to find a good one 08:39:06 this one is -*-fixed-medium-r-*-*-*-*-100-100-*-90-iso10646-* 08:39:26 it's the 100-100 that does it, I'm thinking. that must be the resolution 08:40:06 docl, what are actually your feelings about reva? 08:40:28 it has some awesome newbie stuff on boot-up, that's for sure 08:40:44 what's newbie stuff? 08:40:48 help, see, etc. actually work. that's an improvement over the current retro. 08:41:26 also there's little "crutch" things like it shows the number of elements on the stack when you type .s 08:41:38 wow.. it has see, well, I haven't noticed that. 08:41:48 all optional from an expert's point of view, but really intuitive for teaching newbies 08:42:27 my feeling is, we need to back-port this stuff to retro, at least in an optional portion. 08:43:10 (I haven't played very much with reva, so this is just a surface evaluation) 08:44:16 it has some ans stuff in it 08:44:55 * docl has no problem with that as long as it's optional 08:44:58 personally I think the current development with classes could do it, so that crc, produces his 'doc' class 08:45:09 yeah, me too. 08:45:29 as long I don't need to write with it programs, it's fine. 08:45:51 :- 08:45:53 :-) 08:56:47 I get the idea that retro is more a theoretical stomping ground, and it's forks like reva are more suitable for end-users. 08:58:07 well, on this theoretical stomping ground we did some nice things. 08:58:32 true. once you get past the initial hurdles, retro is very nice. 09:05:52 --- join: neceve (n=claudiu@unaffiliated/neceve) joined #retro 09:20:42 hmm, reva is about twice as slow as a C program in printing 1000 times helloworld 09:34:35 --- join: Ray_work (n=Raystm2@199.227.227.26) joined #retro 12:51:17 --- quit: docl ("leaving") 12:53:12 --- join: docl (n=docl@70-101-145-1.br1.mcl.id.frontiernet.net) joined #retro 13:55:57 bah, a graphical game would be nice. 13:56:21 I'm asking myself how much of that you could do in X without much performance lose.. 14:12:53 hmm.. perhaps with the shared memory extension, well, it seems that there is a way to write directly to some memory 14:15:51 --- join: crc_work (n=cchilder@h-64-105-207-163.phlapafg.covad.net) joined #retro 14:17:13 about 45 more minutes... 14:17:36 crc, there is a way to talk more directly to the X server, there is the shared memory extension. 14:18:32 or even better the dga extension.. 14:18:45 if dga = direct graphics access 14:20:25 cool 14:20:35 hehe.. I think I do look at the dga thing 14:20:57 so we could get something which is ideal for forth 14:23:59 *grml* those stupid bastards 'gets deprecated' 14:26:42 * crc_work will be working on the help module tonight; I should be able to get it functional pretty quickly 14:34:30 hm.. there was also a postscript X Extension, hmm, some nice extensions, imho. 14:35:41 there was a postscript extension? that would have been handy 14:37:07 --- part: crc_work left #retro 15:01:27 --- quit: nighty (Read error: 113 (No route to host)) 15:02:42 --- quit: Ray_work (Read error: 104 (Connection reset by peer)) 15:11:56 home at last 15:12:14 --- join: tcn (n=user@pool-72-70-247-128.spfdma.east.verizon.net) joined #retro 15:17:35 hi tcn 15:23:07 hi crc 15:23:11 hi crc 15:23:27 well, yes it seems that there was/is a postscript extension 15:23:32 whatever that means... 15:23:36 * docl is experimenting with dtach 15:24:46 dtach is a cut down varient of screen? 15:24:55 yep 15:25:26 with a foreach loop, I have it paging through the dtach sessions 15:25:47 http://retroforth.net/paste/?id=23 15:44:45 --- join: crc2 (n=crc2@pool-70-110-173-103.phil.east.verizon.net) joined #retro 15:44:45 --- quit: crc2 (Broken pipe) 15:45:02 --- join: crc2 (n=crc2@pool-70-110-173-103.phil.east.verizon.net) joined #retro 15:45:02 --- quit: crc2 (Remote closed the connection) 15:49:47 ok now it's a set of bash scripts: http://retroforth.net/paste/?id=24 15:52:02 nice 16:00:57 ok.. I got a simple working program which does dga. 16:01:10 crc, which color depth do you run? 16:02:43 here's a much improved version of start.sh: http://retroforth.net/paste/?id=25 16:03:17 this prints the available dga modes, which one do you have? http://retroforth.net/paste/?id=26 16:03:30 it's C at the moment... 16:05:51 virl: 24 bit 16:07:15 ok... 16:07:37 does this c program give you a mode called here 102? 16:07:47 I'm still trying to compile it 16:09:34 this one should do it: gcc -o fileout -L /usr/X11/lib -lX11 -lXext -lXxf86dga 16:09:47 what library has XDGAQueryModes()? 16:10:01 Xxf86dga 16:11:33 and does it work? 16:12:05 yes 16:12:06 http://retroforth.net/paste/?id=27 16:13:07 wow.. much much fewer than here. 16:13:41 I have 176 modes ;-) 16:13:48 heh 16:13:56 I only use the 1280x1024 mode 16:13:59 nothing less 16:14:56 even the text consoles run at that res; so I may have a framebuffer already... 16:15:54 --- quit: neceve ("Leaving") 16:16:16 ok, hmm: this code should paint a rectangle you need to run it as root: http://retroforth.net/paste/?id=28 16:16:57 a red rectangle it should be. 16:21:47 no rectangle 16:21:59 the screen turns black, then pops back up to a normal desktop 16:23:40 got it 16:23:51 I had to remove the if around the code 16:25:02 and it is red 16:25:18 well, the if was for the framebuffer. 16:25:34 when it's not located then I'm asking myself how it works 16:25:52 I have no idea, but it does :) 16:26:16 even when you run it as root? 16:26:32 I didn't run it as root 16:26:39 I gave it root permissions though 16:28:01 well, interesting, perhaps it goes through the x protocol? 16:28:23 that is possible 16:29:46 because the if tests if the array of the framebuffer is here 16:30:21 it returns 0 on my system, but the display code works anyway 16:30:55 and it doesn't need root permissions at all 16:31:07 hey, let me try it 16:32:49 http://www.die.net/doc/linux/man/man3/xfree86-dga.3.html 16:33:14 I don't have DGA so it segfaults 16:34:20 seems to be not the way to go, ah, shit. 16:34:54 maybe I just need it in my xorg.conf 16:35:46 ah. ok, I know it, they go through the X protocol 16:36:30 the fill thing goes through the X protocol, that's why it works without root permissions 17:10:04 --- part: tcn left #retro 19:26:29 --- join: snoopy_1711 (i=snoopy_1@dslb-084-058-142-115.pools.arcor-ip.net) joined #retro 19:33:49 --- quit: Snoopy42 (Read error: 145 (Connection timed out)) 19:34:08 --- nick: snoopy_1711 -> Snoopy42 19:53:07 --- join: yoyofreeman (n=root@222.90.4.24) joined #retro 19:53:32 --- join: docl_ (n=docl@70-101-145-1.br1.mcl.id.frontiernet.net) joined #retro 19:55:42 --- quit: docl (Read error: 113 (No route to host)) 19:56:09 --- nick: docl_ -> docl 20:07:40 --- quit: yoyofreeman (Read error: 104 (Connection reset by peer)) 20:25:53 playtime. 20:42:32 there is something wrong with the way i'm using type. 20:42:43 does it take some kind of count? 20:43:00 addr count -- 20:57:35 fixed, didn't realize that ." automatically typed itself. Neat Code. 20:57:51 I removed all calles to type and the stack is fine now. 20:58:00 calls even 20:59:26 * Raystm2 playing around with the FEN chess notation format. 23:59:59 --- log: ended retro/06.07.11