00:00:00 --- log: started retro/12.12.12
02:08:07 --- quit: karswell (Remote host closed the connection)
02:18:23 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
02:40:28 --- quit: karswell (Read error: Connection reset by peer)
02:50:45 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
04:49:29 --- nick: tangentsleep -> tangentstorm
05:12:05 --- quit: karswell (Remote host closed the connection)
05:22:21 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
05:55:52 <crc> [ &slice.OFFSET @ [ slice.fetch/retreat ] timesRepeat ] 'pl0.display' define
05:56:34 <crc> that takes care of being able to display the output, so all that remains is recursion which parable doesn't support yet
05:57:35 <tangentstorm> :)
05:58:00 <tangentstorm> hmm
05:58:26 <tangentstorm> isn't the y-combinator supposed to allow recursion in languages that don't support recursion?
06:00:50 <crc> actually, since redefinitions are allowed, this works:
06:00:50 <crc> [ ] 'foo' define
06:00:50 <crc> [ #1 - dup #0 > [ dup foo ] ifTrue ] 'foo' define
06:00:50 <crc> #10 foo
06:09:21 <tangentstorm> oh i see.. because it's all postfix.. huh
09:10:47 --- quit: karswell (Remote host closed the connection)
09:21:00 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
09:24:59 --- join: intothev01d (~intothev0@unaffiliated/intothev01d) joined #retro
09:41:07 --- quit: karswell (Remote host closed the connection)
09:51:21 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
10:32:54 <crc>  tangentstorm: it works, because redefintions replace the original, rather than leaving it (as retro does)
10:35:55 <tangentstorm> yeah, i get it. i didn't understand what was hard about recursion until i remembered your everything is postfix rule
10:36:03 <tangentstorm> hard to call a name that hasn't been defined yet :)
10:50:34 <crc> yup. but the redefinition model takes care of that :)
11:27:27 --- join: Mat2 (~Claude@91-65-144-133-dynip.superkabel.de) joined #retro
11:27:31 <Mat2> hello
11:28:49 <crc> hi Mat2 
11:29:03 <Mat2> hi crc !
11:30:10 --- join: Jarod- (~huh@31.37.36.192) joined #retro
11:30:18 <Jarod-> hi
11:31:13 <Mat2> hi Jarod
11:31:21 <Jarod-> I read a lot of good things about Forth and I wanted to give it a try with Retro
11:31:48 <Mat2> nice !
11:32:14 <Jarod-> I'm reading "An introduction to Retro" , it says there's a binary for C implementation
11:32:21 <Jarod-> but I can't find it
11:33:13 <Mat2> do you mean the retro-image file or an executable for your OS ?
11:33:55 <Jarod-> the executable, this is how i read it : "Windows users can download a binary distribution which includes the F# (for .NET) and C implementations. These are confirmed to work on XP, Vista, and Windows 7."
11:34:46 <Mat2> ok, I'm using linux and compiled retro from the sources, please ask crc about the windows binarys
11:36:00 <Jarod-> k, thanks, you mean I should text him in pv ?
11:37:05 <Mat2> or write him an e-mail, otherwise post in the mailing list
11:37:29 <Mat2> crc: here is someone who need help with retro 
11:38:15 <tangentstorm> Jarod-: i've never tried to run either on windows, but Mat2 and I have written a pascal version that runs on windows :)
11:38:33 <tangentstorm> let me bundle up a binary for you
11:38:39 <Jarod-> thanks Mat2, will do
11:39:09 <Mat2> hi tangentstorm
11:39:12 <Jarod-> tangentstorm, is it more efficient that the python package ?
11:39:14 <tangentstorm> heya :)
11:39:21 <tangentstorm> Jarod-: way faster than the python version.
11:39:29 <Jarod-> nice ^^
11:39:54 <tangentstorm> c, go, and pascal are probably the fastest... although mat2 is working on some crazy assembly stuff that's going to be even faster :)
11:40:01 <Jarod-> I wanted to avoid installing python
11:40:55 <Mat2> Jarod: The pascal version have a very fast SDL terminal emulator with support for 256 colours and independent colour attributes for each character
11:40:59 <tangentstorm> do you already have the retro distribution? 
11:41:13 <Jarod-> yes i downloaded it
11:42:33 <Mat2> by the way: I have found the complete source code for Lords Of Midnight inclusive the graphic engine and all shape graphics
11:42:42 <tangentstorm> whoah! nice!
11:42:48 <Mat2> http://www.icemark.com/downloads/
11:42:48 <tangentstorm> they open sourced it or something?
11:43:12 <Mat2> I do not know, we should read the licence carefully
11:44:16 <tangentstorm> Jarod-: you'll be the first user for our sdl version for windows here, so... don't be hatin' on retro if this doesn't work :)
11:44:28 <Jarod-> ok :)
11:47:14 <Mat2> bugs will be resolved if you commit them as soon as possible
11:47:58 <Jarod-> do you have a bug tracker for the pascal version ?
11:48:19 <tangentstorm> https://github.com/sabren/b4/
11:48:28 <tangentstorm> the source code is in ng
11:48:34 <tangentstorm> but not the sdl part
11:48:54 <tangentstorm> or... just use the retro bug tracker?
11:49:21 <tangentstorm> version with the sdl device is here: https://www.assembla.com/code/avm/git/nodes/master/metro/backport
11:50:46 <Mat2> for bug reports and feature requests about the SDL stuff, ask me (here or write me an e-mail)
11:51:34 <Mat2> for vm related bugs ask tangentstorm
11:51:53 <Mat2> and for retro related questions and requests ask crc
11:52:12 <Jarod-> ok, can I use retro documentation or does b4 too different ?
11:52:38 <Jarod-> oups sorry for my english
11:52:54 <crc> Jarod-: http://rx-core.org/retro_windows_11.5-wip.zip for windows binaries + image
11:53:15 <Jarod-> Thanks crc :)
11:54:29 <crc> np
11:54:45 <Mat2> jarod: generally, if you want a feature to be implemented, that should not be the problem
11:55:21 <tangentstorm> nice.. well, i'll package this up anyway :)
11:55:33 <tangentstorm> i'm having git issues
11:57:09 <Mat2> beside very time-consuming requests like a web browser of course
11:59:40 <tangentstorm> something's screwy with my git install... it just hangs instead of fetching submodules
11:59:50 <Mat2> tangentstorm: commercial usage of the sources is not allowed and for all other usages we shall give him credit and write an e-mail to him
12:00:03 <tangentstorm> works for me
12:00:08 <tangentstorm> what's it written in? :)
12:00:41 <Mat2> 100% IA16 assembly :)
12:00:54 <tangentstorm> great :)
12:00:57 <Mat2> should run under MSDOS
12:01:08 <Mat2> or Dosbox
12:02:28 <Mat2> uh, this code is not very optimizated, seems like a direct port from z80 sources
12:02:44 <Jarod-> is it suppose to rewrite all my input ?
12:02:53 <tangentstorm> if you use --with yes
12:03:06 <Jarod-> I don't 
12:03:12 <tangentstorm> it's google code that's causing the git problem i think
12:03:34 <tangentstorm> what are you doing?
12:03:51 <Jarod-> my screen look like that :
12:03:58 <Jarod-> ok 11 2 + putn
12:04:00 <Jarod-> 11
12:04:02 <Jarod-> ok 2
12:04:05 <Jarod-> ok +
12:04:08 <Jarod-> ok putn 13
12:04:12 <Jarod-> ok
12:04:36 <crc> Jarod-: that is normal
12:04:52 <crc> retro's parser evaluates as input is typed, not at the end of a lne
12:04:54 <crc> *line
12:04:55 <tangentstorm> did you cut and paste?
12:05:29 <crc> what vm are you using?
12:05:52 <Jarod-> i launched retro.exe
12:07:10 <tangentstorm> jarrod: did you type all that in and press enter, and then it did that?
12:07:11 * Mat2 implementing disassembler
12:07:18 <tangentstorm> er Jarod- .. sorry
12:07:35 --- nick: Mat2 -> Mat2-coding
12:07:46 <Jarod-> after the first ok, I typed 11 2 + putn, then enter and it wrote everything else
12:09:24 <tangentstorm> hrm. it's really meant to execute immediately on the space bar
12:09:46 <tangentstorm> i guess that windows version buffers the whole line first
12:10:12 <tangentstorm> but yeah, that's how it works :)
12:10:55 <crc> the c version under windows doesn
12:11:06 <crc> the c version under windows doesn't set buffering mode correctly. the .net varients do IIRC
12:12:02 <Jarod-> yeah ok the csharp version execute at space
12:12:27 <Mat2-coding> the pascal version should work also ok
12:21:25 <tangentstorm> grr... i think github removed the ability to put a downloadable file on there :/
12:21:34 <tangentstorm> i'll just put it on my website
12:23:36 <tangentstorm> http://games.tangentcode.com/retro/retro-ngaropas-sdl-win32-2012-12-12.zip
12:24:07 <Jarod-> thanks i got it too :)
12:24:13 <tangentstorm> i included both the plain console and sdl versions...
12:24:44 <tangentstorm> the image is from retro 11.4 but they're interchangable
12:25:12 <Mat2-coding> please share your thoughs, bugs and possible ideas with us !
12:25:37 <Jarod-> will do ^^
12:25:40 <tangentstorm> besides mat2's sdl console, the pascal version also has an external debugger if you use  -d on the command line
12:26:17 <Jarod-> first, a cursor would be nice for the sdl version :)
12:27:12 <tangentstorm> :)
12:30:39 <Mat2-coding> delete something and you will see one !
12:30:54 <Jarod-> yeah I saw that
12:31:28 <Mat2-coding> I will include a nice cursor
12:32:25 <Mat2-coding> next week you have a new version, not much work
12:32:39 <Jarod-> :)
12:32:55 * Mat2-coding write a how-to for the SDL code
12:37:35 <Jarod-> ok I'm starting to get the hang of it, I can't find where it says how to save and load my definitions
12:38:41 <Mat2-coding> from he command-line (cmd): start retro the following way:
12:38:52 <Mat2-coding> retro --with <file name>
12:39:32 <Mat2-coding> where <file name> is the file with your definations
12:39:53 <Jarod-> and it will save automatically with bye ?
12:40:20 <Mat2-coding> no you must save your changes into the retro-image
12:43:41 <tangentstorm> i think "save" ?
12:43:51 <tangentstorm> images are kind of disposable
12:43:59 <Jarod-> save, yes
12:44:08 <tangentstorm> crc has a document about how he uses it under articles on the site
12:44:23 <Mat2-coding> that is the equivalent of save-system in fig forth
12:44:26 <Jarod-> so I should make a copy of retroImage
12:44:44 <Mat2-coding> that would be a good idea
12:45:02 <tangentstorm> one per project probably
12:45:14 <Jarod-> just saved a : foo + putn ; inside
12:45:42 <Jarod-> without backup, that would become a mess :)
12:46:28 <tangentstorm> well you can also use --with whatever.rx
12:46:33 <Mat2-coding> its typical o use files for your definations and build up an image by demand
12:46:45 <tangentstorm> yeah
12:46:54 <Mat2-coding> ^to (I use a bit broken Apple keyboard)
12:47:04 <tangentstorm> i use a text editor, save to an rx file, and then use --with 
12:47:33 <tangentstorm> there are retro setups for emacs and vim under "etc"
12:47:49 <Mat2-coding> you can always test your definations out if you let a terminal session open beside your editor window
12:48:27 <Mat2-coding> another way would be to use the blockeditor (that's really forth style) from within retro
12:48:39 <Mat2-coding> crc have written
12:49:32 <Mat2-coding> for windows I suggest you to use notepad++ for editing
12:49:54 <Mat2-coding> it's free and used by many programmers word wide
12:50:05 <Jarod-> this is wah i use :)
12:50:07 <Jarod-> what
12:50:11 <Mat2-coding> :)
12:51:03 --- quit: karswell (Remote host closed the connection)
12:51:41 <Mat2-coding> then you can use retro like every other compiler from within notepad++ et voilá: you have an IDE
12:52:04 <Jarod-> so in test.rx, I can just type : foo dup * putn ;
12:52:20 <Jarod-> and launch retro --with test.rx
12:52:32 <Jarod-> because that crashed
12:52:43 <tangentstorm> which vm did it crash on?
12:53:20 <Jarod-> ah ok, only csharp and fsharp
12:54:30 <tangentstorm> i don't know for sure but i think csharp may be unfinished.
12:55:05 <tangentstorm> i have c# express installed but i've never tried either version
12:55:32 <Mat2-coding> sorry I can't help you with the NET vm too
12:56:03 <Jarod-> the c version is ok, the sdl too
12:57:02 <Mat2-coding> (I'm not even sure if this is an error from NET'S JIT compiler, but ... ok, it's a Microsoft product so this can be)
12:57:12 <Jarod-> sdl is weird, one time, it doesn't print the definition at start, and the second time it does
12:57:43 <Jarod-> when it doesn't, it doesn't print the first character typed either
12:58:05 <Mat2-coding> I will check this out, thanks
12:58:20 <crc> tangentstorm: c# should be mostly complete, f# is complete for sure
12:59:02 <tangentstorm> ah
12:59:03 <Jarod-> looks like a color coding issue, because when it does print the definition at start, the first character typed is green like the definition, but the second one is yellow
12:59:43 <Jarod-> so my guess would be that the first time it's writing the definition with the background color
13:00:01 <tangentstorm> i'm not entirely clear what's up with the color changes in SDL... it goes orange -> yellow for me
13:00:10 <tangentstorm> pretty sure it's intentional :)
13:00:57 <Mat2-coding> ok, I see the problem. Will fix that next week too
13:01:15 <Mat2-coding> (it's an SDL internal problem)
13:01:18 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
13:03:32 <Mat2-coding> Jarod: You can see my progress here: https://www.assembla.com/spaces/avm/messages
13:03:51 <Jarod-> yes
13:03:53 <Mat2-coding> I will post if one problem is fixed
13:04:00 <Jarod-> ok
13:04:29 <Jarod-> oops , retro doesn't catch empty stack
13:04:55 <Jarod-> should save often
13:05:50 <Mat2-coding> tangentstorm: I can invite you as team-member. this way you can circumvent your Google related problems and upload files to my space (beside direct synchronisation of our both code basis)
13:06:13 <Mat2-coding> if this help you
13:06:45 * Mat2-coding finish disassembler
13:08:52 <crc> hmm, --with seems to be broken on the f#. I'll look at this in a couple of days
13:10:00 <Jarod-> must be difficult to keep up with so much implementations
13:10:38 <Mat2-coding> crc: beware that two NET variants exist: The original Microsoft one and Mono, it can be that both handle files slightly different
13:12:26 --- quit: karswell (Remote host closed the connection)
13:15:39 <crc> Mat2-coding: I have same problems under both mono and ms.net
13:15:58 <crc> Jarod-: it's managable, since things seldom change in the vm implementation
13:16:12 <tangentstorm> Jarod-: c and javascript are probably the main ones that people use. the pascal one is brand new... the others... i don't know where the people who made them ran off to
13:16:14 <crc> some (like f#) were written entirely by others, so take me more time to debug
13:17:19 <tangentstorm> Mat2-coding: sure. i also plan to have a download section at retropascal.org eventually... but i don't even have a site up yet :)
13:20:42 <crc> tangentstorm: IIRC, last time I did a survey of usage, C and Python were the most common
13:21:31 <tangentstorm> haha... I'm just going to make up random facts based on my biased opinions, an then let crc correct me :D 
13:21:40 <tangentstorm> you and your actual data
13:21:40 <crc> :)
13:22:10 * Mat2-coding finished disassembler
13:22:22 <crc> pypy group and others are/were using the vm and a subset of corpse and rancid for benchmarking purposes earlier this year
13:22:31 <tangentstorm> oh interesting!
13:22:41 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
13:23:02 * Mat2-coding now on the way to finish the JIT compiler
13:23:13 <tangentstorm> i had guessed javascript because of the demo
13:23:17 <tangentstorm> and the downloads
13:23:44 <tangentstorm> and because i thought i remembered you saying you used it yourself
13:28:46 <Mat2-coding> tangentstorm: I have added you as team member
13:29:10 <tangentstorm> awesome. thanks :)
13:29:58 <Jarod-> ok I used NppExec in Notepad++ to launch retro with my rx file
13:30:05 <Jarod-> it works with sdl
13:30:25 <Jarod-> but with the c version, retro is launched but there is no console window
13:33:01 <Mat2-coding> that's because the C version need a terminal for output. You must start cmd with retro as executive and then it *can* work, but I guess the problem under Windows will be redirection of he output stream
13:36:37 <Mat2-coding> 0000000000000000: mov r8,    1AAAAAAAAAAAAAAA000000000000000A: mov r9,    1BBBBBBBBBBBBBBB
13:36:37 <Mat2-coding> 0000000000000014: shr r9,    000000000000001F
13:36:37 <Mat2-coding> 0000000000000018: mov r10,   1DDDDDDDDDDDDDDD
13:36:37 <Mat2-coding> 0000000000000022: mov r11,   1EEEEEEEEEEEEEEE
13:37:46 <Mat2-coding> yeah, the opimization works, it compiles stack based code into register based machine-code !
13:38:37 <Mat2-coding> that's much better code than LLVM will generate :)
13:38:42 <tangentstorm> Mat2-coding: not entirely sure what i'm looking at here, but it sounds cool :)
13:41:38 <Mat2-coding> tangentstorm: LLVM and similar envirionments inclusive all native-code generating forth's I know compile stack based VM code into either a list of calls to primitive routines (thats subroutine threading) or into machine-code which emulates a stack (or use one)
13:42:33 <Mat2-coding> this resulting in suboptimal code because stack accesses depending on much more memory accesses as register based one
13:42:58 <Mat2-coding> specially for architectures like ARM, that's critical
13:43:51 <tangentstorm> i hope you're going to publish this :)
13:44:19 <Mat2-coding> my compiler maps the data stack to a set of registers so the generated code is much more optimizated for most cpu's
13:44:39 <Mat2-coding> (and will perform much better)
13:44:55 <Mat2-coding> sure, if I'm finish
13:45:05 <Jarod-> next best thing after a stack based cpu :)
13:47:07 <Mat2-coding> the VM code sequence was:LIT #1AAAAAAAAAAAAAAA LIT #1BBBBBBBBBBBBBBB
13:47:25 <Mat2-coding> SHR
13:47:38 <Mat2-coding> LIT #1DDDDDDDDDDDDDDD
13:47:48 <Mat2-coding> LI #1EEEEEEEEEEEEEEEE
13:47:55 <Mat2-coding> or:
13:49:19 <Mat2-coding> liLI, shrLI, LI 1AAAAAAAAAAAAAAAA 1BBBBBBBBBBBBBBB 1DDDDDDDDDDDDDDD 1EEEEEEEEEEEEEEE (AVM mnemonic)
13:50:51 <Mat2-coding> sorry i forgo a LI instruction:
13:51:25 <Mat2-coding> liLI, liShr liLI 1AAAAAAAAAAAAAAAA 1BBBBBBBBBBBBBBB 1DDDDDDDDDDDDDDD 1EEEEEEEEEEEEEEE
13:52:26 <Mat2-coding> liLI, liShr liLI 1AAAAAAAAAAAAAAAA 1BBBBBBBBBBBBBBB 1F 1DDDDDDDDDDDDDDD 1EEEEEEEEEEEEEEE
13:52:38 <Mat2-coding> ^that's correct
13:54:01 <Mat2-coding> and hat's the generated machine code:
13:54:05 <Mat2-coding> 49B8AAAAAAAAAAAAAA1A49B9BBBBBBBBBBBBBB1B49C1E91F49BADDDDDDDDDDDDDD1D49BBEEEEEEEEEEEEEE1E
13:54:20 <tangentstorm> pretty
13:55:00 --- join: impomatic (~digital_w@46.208.131.194) joined #retro
13:59:26 <Jarod-> ok guys, thanks for your help, see you soon :)
14:00:39 --- quit: Jarod- ()
14:21:01 --- quit: karswell (Remote host closed the connection)
14:25:46 --- quit: impomatic (Quit: var kings=new Array(3); for (x in kings) { kings[x].origin='orient'; kings[x].bearingGift=true; kings[x].travelled='afar'; })
14:31:18 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
14:33:05 --- join: erider (~chatzilla@unaffiliated/erider) joined #retro
14:35:28 <Mat2-coding> hi erider
14:40:19 <erider> hi Mat2-coding 
14:40:49 --- quit: karswell (Read error: Connection reset by peer)
14:41:18 <Mat2-coding> what's new ?
14:44:29 <erider> nothing really what about you 
14:48:05 <Mat2-coding> I'm working on a JIT compiler
14:48:13 --- nick: Mat2-coding -> Mat2
14:48:55 <erider> good deal 
14:51:05 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
14:51:18 <Mat2> my first try was using GNU lightning, then LLVM, but both compiles with disappointing results for stack based code
14:51:45 <Mat2> (beside lightning offer no optimization at all)
14:52:22 <erider> that is not friendly 
14:56:37 <Mat2> but there have a plan for it...
15:00:54 <erider> ok 
15:02:12 <Mat2> i is only, there have these plan though more than four years or so
15:02:21 <tangentstorm> haha
15:02:43 <erider> I guess we will see what happens 
15:07:23 <Mat2> it seems so
15:07:51 <Mat2> have you ever played Lords Of Midnight ?
15:08:26 <erider> is it a game? 
15:12:52 <Mat2> yes
15:14:25 <Mat2> http://www.icemark.com/tower/index.html
15:14:41 <Mat2> it was first published for the ZX Spectrum
15:14:48 <Mat2> later for PC
15:16:22 --- join: kumul (~kumul@66-50-75-241.prtc.net) joined #retro
15:17:54 <Mat2> I'm interested in its graphic engine
15:18:22 <Mat2> because we want to code some adventure game 
15:18:54 <erider> I have never played it, sorry 
15:20:48 <Mat2> its programmer have published the source code for the DOS version
15:21:26 <Mat2> very readable but seem to be a straight conversion from some z80 sources
15:23:07 <erider> you are converting the graphic engine into a more modern chip set :-) 
15:27:17 <Mat2> SDL will fit
15:27:56 --- join: beretta (~beretta@cpe-107-8-120-84.columbus.res.rr.com) joined #retro
15:29:51 <Mat2> hi beretta
15:30:06 <beretta> howdie
15:30:07 <erider> hi beretta 
15:40:17 <Mat2> see you 
15:40:19 <Mat2> ciao
15:40:23 --- quit: Mat2 (Quit: Verlassend)
15:40:29 <beretta> bye
16:22:49 --- nick: tangentstorm -> tangentaway
16:51:48 --- quit: karswell (Remote host closed the connection)
16:56:22 --- nick: tangentaway -> tangentstorm
17:02:03 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
17:17:08 --- quit: karswell (Remote host closed the connection)
17:27:24 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
17:40:26 --- quit: karswell (Remote host closed the connection)
17:50:39 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
17:56:32 <erider> see you guys later 
17:56:44 --- quit: erider (Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121129162756])
17:56:45 <beretta> bye bye
19:06:13 <beretta> crc: check this out: https://sites.google.com/site/cocoboot2/legs-vm
19:30:57 --- quit: tangentstorm (Quit: WeeChat 0.3.2)
20:58:13 --- quit: beretta (Read error: Connection reset by peer)
20:59:39 --- join: beretta (~beretta@cpe-107-8-120-84.columbus.res.rr.com) joined #retro
21:36:41 --- quit: intothev01d (Quit: intothev01d)
22:17:45 --- quit: kumul (Quit: WeeChat 0.3.9.2)
23:59:59 --- log: ended retro/12.12.12