00:00:00 --- log: started retro/11.03.01 00:38:10 --- quit: docl (Ping timeout: 240 seconds) 00:40:11 --- join: docl (~Luke@unaffiliated/docl) joined #retro 06:25:22 --- join: aisa (~aisa@173-10-243-253-Albuquerque.hfc.comcastbusiness.net) joined #retro 08:56:34 --- quit: aisa (Quit: aisa) 09:58:29 --- join: jdp (~jdp@24.238.32.162.res-cmts.segr.ptd.net) joined #retro 10:10:45 --- join: aisa (~aisa@173-10-243-253-Albuquerque.hfc.comcastbusiness.net) joined #retro 11:37:45 --- quit: aisa (Read error: Connection reset by peer) 11:54:05 --- join: aisa (~aisa@173-10-243-253-Albuquerque.hfc.comcastbusiness.net) joined #retro 12:33:53 the python implementation now supports "include", "needs", and ":include" and also the --with command line parameter 12:34:11 --- mode: ChanServ set +o crc_ 12:34:21 --- nick: crc_ -> crc 12:59:27 I suppose that ruby will be the next vm to get updated to be more feature complete 12:59:42 I should probably try to at least get saving of images working 13:28:13 --- quit: aisa (Quit: aisa) 13:31:03 crc: any sudden desire to code a more ruby-like forth? :P 16:55:30 no 17:05:49 --- join: roarde (~roarde@pdpc/supporter/active/sixforty) joined #retro 17:09:44 hi roarde 17:12:28 hi, crc 17:12:54 "brutally minimal": You had to know that'd get my attention. 17:33:57 brutally minimal? 17:34:31 from the commentary? 17:34:43 comment at the end of metacompiling in last couple of pulls I did of core.rx 17:34:53 ok 17:35:44 So I'm in the process of "cut here" :P 17:38:51 Haven't had a chance to look at more than the first page of the commentary; will be reading it thoroughly. 17:39:46 let me know if I can clarify anything in it 17:41:08 I'm saving such to do all at once unless it's an error. Not online enough to do otherwise or follow up. 17:44:13 understood 17:50:36 foucist: the ruby implementation can now save images. Still no support for including files at runtime yet, or access to the host environment 17:51:09 the code for saving images is ugly: http://sprunge.us/RedM 17:55:06 --- quit: jdp (Read error: Connection reset by peer) 17:55:22 --- join: jdp (~jdp@24.238.32.162.res-cmts.segr.ptd.net) joined #retro 18:30:46 --- join: Quartus (~neal@CPE0022b0b24a15-CM001947482b20.cpe.net.cable.rogers.com) joined #retro 18:35:19 crc: can you give me an example value for @memory[i] ? 18:36:02 the @memory[] array is a set of little endian, signed 32-bit integers 18:37:10 anything in the range of -2,147,483,648 to 2,147,483,647 18:43:30 crc: f.write(@memory[i].pack("N")) should replace all the byte1-4 vars & putcs 18:43:34 i think that'll work 18:44:42 [1230123123].pack("N") #=> "IR0s" $stdout.putc(1230123123 % 256) #=> s $stdout.putc(1230123123 / 256) #=> 0 18:46:30 I get: 18:46:30 vm/retro.rb:94:in `handle_devices': undefined method `pack' for 8:Fixnum (NoMethodError) 18:46:49 crc: you need to call pack on an array object 18:47:44 hmm 18:48:16 can't call it directly on ruby number (fixnum) 18:48:27 yeah so [@memory[i]].pack.. 18:48:42 bit weird i guess 18:49:49 ok, that works (with pack("i")); commited 18:50:52 crc: "N" is for 32 bit, what was i for? 18:51:28 signed int 18:51:38 "N" is unsigned 18:55:19 k 18:59:05 crc: what do you know about eforth? how complicated is it in comparison to retro 18:59:18 crc: currently the only forth implemented for the GA144 is eforth heh 18:59:53 it'd probably be trivial to make an ngaro vm on GA144 though 18:59:56 I've not touched eforth in years. The fundamental concept is similar, in that both retro and eforth are built from a fairly small set of primitive functions, but eforth would be closer to the machine generally 19:00:07 (sidenote: color/arrayforth isn't. go figure. or "figures") 19:00:38 roarde: yeah, mostlly cuz bill muench went ahead w/ eforth and chuck has been too busy to directly do that yet 19:04:13 The eForth Virtual Machine (e4VM) is implemented by a virtual CPU consisting of two nodes (105 and 106.) This CPU is one of the three principal clients of the external SRAM control cluster, consisting of four nodes (107, 007, 008 and 009.) These six nodes, constituting the minimal e4VM 19:04:26 node = f18a core 19:06:18 The virtual CPU may be extended with application specific functions that interface through (and may be housed in) nodes [adjacent nodes] 19:06:28 interesting 19:10:32 my guess is that about half the translations could be a simple exchange of one number for another, then I think there's an exec bit 19:16:02 * crc finished support for the "--with" command line argument, "needs", "include", etc in the ruby implementation 19:26:12 * crc finished support for "getEnv" in the ruby implementation. It's now on matching the python implementation in features and should be capable of running casket-based web apps now. 19:26:20 has Retro 11 always required a double return from colon definitions? 19:26:58 no; that's a fairly recent change 19:27:26 I'm using timeline to try to locate comments on it. 19:27:40 and strictly speaking, it can be disabled with a rebuild; it's only needed if you want to use the debugging tools like autopsy (for "see") 19:28:29 http://rx-core.org/rx.fossil/info/3e521e4850 19:28:42 thanks 19:46:58 * crc is off to bed; goodnight 19:47:40 nite, crc 20:22:06 --- quit: Quartus () 22:23:36 have there been efforts to get retro running on micros yet? 22:24:07 as in 8-bit or so? 22:24:39 8 bit might be tough due to cells having to be able to address anywhere in the image 22:24:45 but if it exists yeah, sure i'll check it out 22:24:53 i meant on more modern ones like avr's maybe 22:26:45 don't really know, but I'd bet such was at least attempted pre-retro10 22:27:04 wouldn't be applicable since, of course 22:28:19 are there C compilers and standard libs for that? 22:29:39 afaik, no one's done a retro 10+ in any assembler 22:32:06 Eventually I'll do one for z80 sim, but will use 32-bit cells and cheat with an EMM module. 23:26:26 --- quit: roarde (Quit: Leaving.) 23:26:52 --- join: roarde (~roarde@pdpc/supporter/active/sixforty) joined #retro 23:56:00 --- quit: roarde (Quit: Leaving.) 23:59:59 --- log: ended retro/11.03.01