00:00:00 --- log: started retro/10.01.21 01:25:56 --- quit: Raystm2 (Client Quit) 04:09:42 --- join: crcx (i=d8012b82@gateway/web/freenode/x-nnsxxsyfemjugbdi) joined #retro 05:29:38 --- quit: crcx (calvino.freenode.net irc.freenode.net) 05:29:43 --- quit: docl (calvino.freenode.net irc.freenode.net) 05:29:44 --- quit: SimonRC (calvino.freenode.net irc.freenode.net) 05:29:47 --- quit: ChanServ (calvino.freenode.net irc.freenode.net) 05:29:56 --- quit: probonono (calvino.freenode.net irc.freenode.net) 05:31:15 --- join: crcx (i=d8012b82@gateway/web/freenode/x-nnsxxsyfemjugbdi) joined #retro 05:31:39 --- join: SimonRC (n=sc@217.155.91.250) joined #retro 05:31:39 --- join: docl (n=luke@97-120-211-224.ptld.qwest.net) joined #retro 05:32:47 --- join: ChanServ (ChanServ@services.) joined #retro 05:32:47 --- mode: irc.freenode.net set +o ChanServ 05:33:41 --- join: probonono (n=User@unaffiliated/probonono) joined #retro 06:14:31 --- join: mike50265 (i=3fe0bf56@gateway/web/freenode/x-nsaernvhcxzcrlaq) joined #retro 06:14:41 --- part: mike50265 left #retro 06:34:38 --- quit: probonono (calvino.freenode.net irc.freenode.net) 06:34:45 --- quit: ChanServ (calvino.freenode.net irc.freenode.net) 06:34:45 --- quit: docl (calvino.freenode.net irc.freenode.net) 06:34:49 --- quit: SimonRC (calvino.freenode.net irc.freenode.net) 06:35:10 --- join: ChanServ (ChanServ@services.) joined #retro 06:35:10 --- mode: irc.freenode.net set +o ChanServ 06:36:22 --- join: SimonRC (n=sc@217.155.91.250) joined #retro 06:36:22 --- join: docl (n=luke@97-120-211-224.ptld.qwest.net) joined #retro 06:41:51 * crcx has begun working on a basic set of tests for the core 06:48:32 --- quit: docl (calvino.freenode.net irc.freenode.net) 06:51:09 --- join: probonono (n=User@unaffiliated/probonono) joined #retro 07:06:40 --- join: docl (n=luke@97-120-211-224.ptld.qwest.net) joined #retro 07:30:31 --- quit: virl (Remote closed the connection) 07:30:43 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 09:07:56 --- join: Mat2 (i=5b40e9d1@gateway/web/freenode/x-cscvjqwldkuyxedf) joined #retro 09:08:08 good evening 09:08:08 hi Mat 09:09:50 I want to ask about the editor your working on 09:09:56 is it usable ? 09:10:15 the block editor? 09:10:18 yes 09:10:23 it's usable 09:10:28 fine :) 09:13:09 my little vm is working now (but not all instruction combinations are finished) 09:13:31 the generated assembler sources from gcc looks quite good 09:15:15 so I think porting retro to it would result in a performance gain 09:18:36 for sure I will have questions, so it can be I spam up here a lot of them in the future 09:18:42 hope that is ok ? 09:20:17 yes 09:24:10 have you ever worked with jonesforth ? 09:26:18 *Mat2 had a lot of fun with this Fig/ANS forth mishmash yesterday* 09:28:07 have a nice day 09:28:09 ciao 09:28:18 --- quit: Mat2 ("Page closed") 11:05:21 --- quit: virl (Remote closed the connection) 13:14:08 http://gist.github.com/283202 - hex dump prototype 13:15:13 also: I fixed a bug in the number conversion code that was allowing some invalid numbers to be entered 14:04:33 --- quit: crcx ("Page closed") 15:35:20 --- join: erider (n=chatzill@pool-173-69-160-231.bltmmd.fios.verizon.net) joined #retro 15:35:39 hi 15:36:43 hi erider 15:37:02 crc: whats up how is it going 15:37:29 I've been working on adding some test code to help identify bugs 15:37:46 and exploring ideas for aiding in debugging and code analysis 15:39:06 a debugger would be cool 15:39:26 not a full debugger, but tools to help examine compiled code in more detail 15:39:31 step the program one instruction at a time 15:39:46 --- join: virl (n=virl__@chello062178085149.1.12.vie.surfer.at) joined #retro 15:40:23 that's not easy with the current ngaro design 15:40:26 you can learn a lot about a program by watching it one instruction at a time ;) 15:40:41 ./retro --trace :) 15:41:01 though that's brutally slow, and if you pipe it to a file, it'll eat a *lot* of space 15:42:29 but useful right? 15:44:19 sometimes 15:44:33 if you have the patience and space to spare 15:47:20 you cannot hook into the program well it is running 15:52:46 not at this time 15:55:31 erider: part of the slowness is due to the number of instructions it takes to do even simple tasks. 15:55:57 just executing 'bye' immediately after startup takes 5188 instructions 15:56:22 'words bye' takes 43,304 instructions 15:59:04 wow!!! 16:00:44 that's the downside to misc-style instruction sets 16:00:52 takes a lot of cycles to do simple things 16:13:23 time for a new instruction sets :) 16:15:16 nope 16:15:22 I like the current one. 16:15:39 CISC isn't a design philosophy I want to pursue 16:16:50 mat2 is working on a more powerful vm with a different instruction set, inspired by VLIW architectures 16:18:26 it's likely time for a better implementation of the current vm though. The ones I produced aren't optimized for speed. 16:23:47 yeah but its not bad either 16:27:28 with a better vm implementation we could get 20-40% more performance with no alteration to the image 16:28:33 hmm maybe 16:28:47 mat did some experiments on this already 16:29:07 http://github.com/Mat2/extended-ngaro 16:29:15 ok 16:31:00 well it should be faster in written in C. the current version is written in JS right 16:31:51 there are implementations in C, C# (for .NET), Java, J2ME, Emacs Lisp, and JS 16:32:03 and docl checked in some early stubs for a possible TCL implementation 16:32:21 oh, and a Go implementation too, from yiyus 16:32:49 of the vm 16:32:52 yes 16:33:32 then what is the problem? are there performances issues with the other implementations 16:33:48 there's always room for performance 16:33:56 true 16:34:20 the approach I used in implementing things was geared towards simplicity. but it's seldom compiled in an efficient manner. 16:34:27 crc: but you are currently working on the Js implementation right 16:35:03 I work slowly on all of the implementations 16:35:15 I see 16:35:42 I thought JS was the front runner for the project 16:36:03 JS makes a nice tech demo, and lets me use Retro on a lot of mobile devices easily 16:36:20 but the main implementation in use is the C (console) implementation 16:37:07 I see 16:37:16 ok cool 16:37:33 erider: the C implementation is needed to allow for metacompiling new images from source 16:37:59 erider: I can do it in the javascript port, but that takes an hour or so each time I change something 16:38:10 erider: the C implementation does the job in a few seconds 16:38:35 I that is were I was lost 16:38:52 I thought JS was the implementation 16:38:57 but I get it now 16:41:09 brb 16:41:22 --- quit: erider (Remote closed the connection) 16:50:57 --- join: erider (n=chatzill@unaffiliated/erider) joined #retro 17:01:17 back 18:35:03 --- join: docl_ (n=luke@97-120-215-184.ptld.qwest.net) joined #retro 18:38:01 --- join: crc2 (n=charlesc@71.23.210.149) joined #retro 18:48:35 --- quit: docl (Read error: 113 (No route to host)) 18:56:54 --- nick: docl_ -> docl 19:00:26 --- quit: crc (Read error: 110 (Connection timed out)) 19:25:24 --- quit: erider (Read error: 113 (No route to host)) 23:24:39 --- join: Guest41563 (n=carlos@201.250.180.189) joined #retro 23:31:21 --- quit: virl (Remote closed the connection) 23:59:59 --- log: ended retro/10.01.21