00:00:00 --- log: started retro/13.01.08
03:30:58 --- quit: obobo (Read error: Connection reset by peer)
03:31:12 --- join: obobo (~chatzilla@dyn-dsl-pt-98-124-47-36.nexicom.net) joined #retro
04:05:09 <crc> http://rx-core.org/retroImage is a stable link to the latest image in my repo (updated automatically after every build)
05:02:19 --- quit: erider (Changing host)
05:02:19 --- join: erider (~chatzilla@unaffiliated/erider) joined #retro
05:03:31 <erider> crc: I have a ngaro vm that I have written in tcl but it is a partial vm
05:04:21 <erider> it is a port from your lua and perl version but I am working on fix a one off bug 
05:07:00 <erider> well I would say it is more of a bound checking bug 
05:07:49 <erider> the stack pointer is increment pass the stack's bounds 
06:40:28 <crc> erider: is the source online?
06:58:24 --- quit: karswell (Remote host closed the connection)
07:08:39 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
08:28:29 --- join: kumul (~kumul@173.215.194.228) joined #retro
10:05:01 --- nick: sabren -> tangentstorm
10:25:37 --- join: ncv (~quassel@unaffiliated/neceve) joined #retro
10:32:53 --- quit: tangentstorm (Quit: WeeChat 0.3.2)
11:00:30 --- quit: kumul (Read error: Connection reset by peer)
11:08:02 --- join: impomatic (~digital_w@46.208.50.63) joined #retro
11:19:26 --- join: tangentstorm (~michal@108-218-151-22.lightspeed.rcsntx.sbcglobal.net) joined #retro
11:36:31 --- quit: karswell (Remote host closed the connection)
11:46:47 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
13:01:09 --- join: Mat2 (~claude@91-65-144-133-dynip.superkabel.de) joined #retro
13:01:15 <Mat2> hello !
13:01:58 <tangentstorm> hey Mat2 :)
13:02:47 <Mat2> hi tangentstorm !
13:03:00 <tangentstorm> How's the new year treating you?
13:03:36 <Mat2> fine, lot of work
13:04:45 <tangentstorm> is your new video device online?
13:05:35 <Mat2> no time for finishing it, but next week it should be online
13:06:15 <tangentstorm> cool :)
13:06:43 <tangentstorm> i've been working on a little text editor in pascal
13:07:29 <Mat2> I think you can experiment with it together with the compiler because it's the graphic subengine of the vm
13:07:30 <tangentstorm> i'm going to try and make it scriptable in retro
13:08:47 <Mat2> have you program the editor in retro ?
13:09:09 <tangentstorm> no, it's in pascal, but i was thinking i'd make it a retro device
13:09:37 <tangentstorm> so like "move cursor up" would be a message you send to port X
13:09:46 <tangentstorm> same with cut, copy, paste, etc
13:10:22 <tangentstorm> i'm pretty sure it's possible to create pascal data structures directly inside the RAM array for the vm
13:10:28 <tangentstorm> so the two can be shared
13:11:00 <tangentstorm> (you just make a pointer to an entry in the array, and then cast it to whatever data type you want)
13:11:15 <Mat2> that would be no problem
13:11:58 <tangentstorm> And you can override the memory management functions to do this behind the scenes for you too.
13:12:27 <tangentstorm> I'm kind of hoping I can even expose the existing forth dictionary as a pascal structure.
13:13:09 <tangentstorm> like without necessarily writing a parser or anything, just taking a pointer and mapping it to a record
13:13:36 <Mat2> metro is near finished, and I have choosen a complete open-source, ARM based platform for the tablet project
13:13:46 <tangentstorm> oh yeah?
13:13:59 <Mat2> the costs are between 100 and 120 euro
13:14:08 <tangentstorm> is it the odroid?
13:14:32 <Mat2> it's an ARM7 based board with 7" touch panel
13:14:36 <tangentstorm> oh nice
13:15:07 <tangentstorm> http://www.hardkernel.com/renewal_2011/products/prdt_info.php?g_code=G135235611947   <- someone was offering to give one away to anyone who would port lazarus for free pascal to it :)
13:15:12 <tangentstorm> do you have a link?
13:16:38 <Mat2> https://www.olimex.com/Products/OLinuXino/A13/A13-OLinuXino-WIFI/
13:16:42 <Mat2> this is the board
13:17:01 <Mat2> https://www.olimex.com/Products/OLinuXino/A13/A13-LCD7-TS/
13:17:10 <Mat2> and these is the touch panel
13:18:57 <Mat2> my problem was the need of an open-source graphic driver, because it's common for ARM vendors only to release there SoC's with closed binaries as firmware
13:19:00 <tangentstorm> i'd be intimidated trying to put the hardware together but it looks like a pretty cool system
13:20:06 <Mat2> but the used SoC assembled a MALI based GPU for which an open-source driver exist
13:20:23 <tangentstorm> huh
13:20:51 <tangentstorm> it kind of comes down to how good you want the graphics to be
13:21:02 <tangentstorm> er..
13:21:09 <tangentstorm> sorry, my mind was drifting there :)
13:21:14 <Mat2> *g*
13:21:18 <tangentstorm> I was thinking about the videos from the latest forth day
13:21:59 <tangentstorm> chuck moore was showing off the ga144 again... he's got a 1024x768 running off it now
13:22:13 <tangentstorm> he just uses it to draw 4-5 colors though :)
13:23:06 <tangentstorm> in last year's talk, he was generating the signal sort of on the fly. no internal memory mapped pixels or anything
13:23:35 <tangentstorm> but for his circuit design program he had to use bitmapped graphics
13:24:02 <tangentstorm> so he's actually got it running like a very old/primitive graphics card
13:24:32 <tangentstorm> Anyway, I was just thinking about that, and about some of the FPGA things that implement their own graphics card too.
13:25:00 <Mat2> his way generating the video signal is comparable to the interupt and software driven approach prefered for the first video generator chips avariable, like the TIA (Atari 2600)
13:25:02 <tangentstorm> You can do it, but you don't get nearly the quality of these high production closed source chips.
13:25:36 <tangentstorm> haha. that's exactly what it looks like. his circuit program looks like an atari 2600 game :D
13:26:27 <tangentstorm> https://www.youtube.com/watch?v=bB4pbjgvZoo  
13:26:30 <tangentstorm> it's about an hour
13:26:36 <Mat2> that is understandable because the GA144 is so memory restricted and more important - memory access seem to be very expensive (somewhat about 250 ns for access)
13:26:46 <tangentstorm> <chuck moore's newest fireside chat>
13:26:56 <tangentstorm> yeah
13:27:56 <Mat2> I think some day there realize that offering a good synchron memory interface is perfered for there markets (whatever these are..)
13:28:34 <tangentstorm> I'm not sure they know. :)
13:28:45 <tangentstorm> I haven't found/watched greg's talk yet
13:28:57 <tangentstorm> he seems to be the one that describes the business side of things.
13:29:27 <tangentstorm> I don't know him or anything. I just can't remember his last name at the moment. :)
13:29:48 <Mat2> the GA144 would be an interesting alternative to CPLD's or simple FPGA's probably with more internal ressources avariable
13:29:55 <tangentstorm> But I think that chip has some serious potential for hobbyists.
13:30:13 --- quit: karswell (Remote host closed the connection)
13:30:32 <tangentstorm> They basically said they could put 100 times as many cores on the chip for the same price if they had a reason/demand for it.
13:31:09 <tangentstorm> They have to design the chip but the manufacturing process can handle it.
13:31:35 <tangentstorm> at least that's how i understood it a year ago :)
13:31:47 <Mat2> hmm, it's all about the price
13:32:05 <tangentstorm> Sooner or later, niklaus wirth is going to deliver Oberon on his RISC chip.
13:32:48 <tangentstorm> it's not really a chip. it's an fpga. 
13:33:00 <Mat2> look, there exist a cheap FPGA based board with 1 mB ram which can easily be used for a MISC multicore with hundreds of cores inclusive memory and video generators
13:33:03 <tangentstorm> or it runs on an fpga. however you say it :)
13:33:37 <Mat2> and the best is: All open-sourced
13:33:56 <tangentstorm> someone has to show people how to put the pieces together
13:33:58 <tangentstorm> like
13:34:21 <tangentstorm> i'm intimidated by soldering and the electronics part. that's not me. i'm good at software.
13:34:25 <Mat2> http://www.retroleum.co.uk/v6z80p/
13:35:11 <Mat2> you can by it from batch and the nice guy behind this project assemble the board for you
13:35:31 <tangentstorm> but if someone gave me a pre-assembled fpga with plenty of excess capacity hooked up to your touchscreen, or to a keyboard mouse video + ethernet... i could hack away on an open source cpu design
13:36:23 <tangentstorm> wow that's pretty nice
13:37:41 <Mat2> I think the hardware is not the problem anymore
13:38:00 <tangentstorm> yeah
13:38:30 <tangentstorm> there's a guy on the oberon list talking about his own oberon-based arduino-alike... trying to do a kickstarter
13:38:45 <Mat2> we need a good, platform independent operating system and applications, applications and applications...
13:39:37 <tangentstorm> and what people are saying is basically: things are already out there if you want to make your own computer. someone even pointed out olimex and how they can build and ship complete boards like that for less than what we'd pay for parts in other countries
13:40:01 <Mat2> yes, it's a nice situation
13:40:28 <Mat2> and much cheaper than (mis)reusing existing standard hardware
13:40:29 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
13:40:32 <tangentstorm> i'm not sure it was so nice for the guy wanting to do the kickstarter, but good for consumers probably :)
13:40:45 <Mat2> you say it
13:41:43 <tangentstorm> well i gotta go take a shower and get back to work here.
13:41:52 <Mat2> have a nice day !
13:42:05 <tangentstorm> i look forward to trying out some rainbow letters soon :)
13:42:15 <tangentstorm> you too. seeya ;)
13:42:18 --- nick: tangentstorm -> tangentwork
13:42:23 <Mat2> check the repro next week !
13:42:27 <Mat2> see you !
13:55:05 <erider> hi all 
13:55:29 <erider> crc: no but it can be 
14:07:34 <Mat2> hi erider !
14:07:42 <erider> hi Mat2 
14:07:50 <erider> how are you doing 
14:07:53 <Mat2> good new year, how you doing ?
14:08:08 <erider> I am good just debugging some code 
14:08:19 <erider> same to you
14:09:22 <Mat2> fine, I am working on my "self-build tablet based upon an retro dialect" project
14:09:41 <Mat2> beside work
14:09:50 <erider> that sounds like fun 
14:09:58 <Mat2> it is
14:10:16 <Mat2> (and will be ARM based)
14:10:43 <erider> I have been porting ngaro to tcl 
14:11:24 <erider> well I have a partial port or basic rewrite of the perl and lua version 
14:13:04 <erider> but I can't get the stack pointer to stay in its limits :-( tcl doesn't have arrays so I am using list 
14:14:15 <erider> so I am doing some debugging 
14:14:30 <Mat2> hmm, ok, I know the TCL vm a bit from work
14:14:52 <Mat2> why don't you map the stack to an string array ?
14:15:15 <erider> I can get the retro image loaded into the memory list but the opcode processor is failing 
14:17:25 <Mat2> I think you have implemented the interpreter in TCL and not embedded its vm in C, right ?
14:17:33 <erider> that gets hairy when you try to modify strings in place. list are easier to work with 
14:17:42 <Mat2> ah ok
14:18:03 <erider> I don't have to do a lot of split and join commands 
14:19:19 <erider> I think the issue is where I initialize the lists 
14:21:05 <Mat2> I had give out hacking in TCL for one job because of some hash bugs (mainly buffer overflows and and integer wrap-ups)
14:22:20 <Mat2> probably you problem is simply one of these
14:22:33 <erider> maybe 
14:23:39 <Mat2> TCL is very forth like in this aspect for a scripting language
14:24:24 <erider> its a programmable language which it neat 
14:27:16 <Mat2> like forth -> stacks, lisp (and dialects like scheme, logo) -> lists
14:27:21 <Mat2> TCL -> strings
14:27:38 <Mat2> some very similar conceptions
14:27:52 <Mat2> I would call these a language family
14:28:05 <erider> yup ;-)
14:30:39 <Mat2> how about bringing all features together ?
14:31:11 <Mat2> I think the result would be something lice factor
14:31:50 <Mat2> ah, I forget APL -> vectors
14:32:29 * erider is not a big features guy
14:32:40 <erider> kiss 
14:33:39 <Mat2> I'm too
14:34:09 <Mat2> but all these features can be implemented on top of one data type, the vector
14:34:31 <erider> too much stuff to look up and try to remember when you have a lot of features 
14:37:40 <Mat2> you are right if one thinks about a forth equivalent of PL/1
14:38:14 <Mat2> or even more feature fixated Ada
14:38:45 <Mat2> hmm, ok ANSI Fo
14:39:09 <Mat2> rth with all of its extensions
14:43:05 <Mat2> ok, your are right - it's not a good idea
14:52:17 <erider> whats not 
14:53:32 <Mat2> combining the features of APL, Lisp, TCL and Forth
14:53:42 <erider> :-)
14:54:10 <Mat2> APL is sufficient for all of them ;-)
14:55:00 <tangentwork> I just used my text editor to edit and save its own source code! :D
14:56:23 <Mat2> or simply use Emacs as OS *g*
14:57:12 <Mat2> by the way I found ARM a very challenging platform for a stack based vm
15:09:20 <erider> it maybe that the image is not align right in the list :-( 
15:16:54 --- quit: ncv (Remote host closed the connection)
15:19:20 --- join: kumul (~kumul@173.215.194.228) joined #retro
15:27:31 <erider> nope that is not it, the data is right in the list 
15:37:54 --- quit: impomatic (Quit: http://retroprogramming.com)
15:40:42 <Mat2> hmm
15:45:59 <erider> Mat2: do you want to take a look see, maybe your tcl skills can help me find something I maybe looking over 
15:50:33 <Mat2> ok, have you a link ?
15:50:56 <erider> hold one 
15:53:07 <erider> Mat2 http://pastebin.com/04Bmkhww
15:53:54 <erider> Mat2: do you have the retroImage file? 
15:55:30 <erider> I think I found the error 
15:57:04 <Mat2> for 11.4, yes
15:57:29 <Mat2> what was the error ?
15:57:52 <Mat2> as I see it your code do not reload opcode
15:59:20 <Mat2> set opcode [lindex $memory $ip]
15:59:27 <Mat2> then the swithc, ok
15:59:34 <Mat2> ^switch
15:59:53 <Mat2> ip is incremented, all ok
16:00:22 <Mat2> but where is the point the next instruction is loaded in the loop ?
16:00:27 <erider> I get index out of range on that stack list 
16:01:53 * erider is looking around for a single step debugger
16:02:38 <Mat2> I get an integer wrap-up
16:03:20 <erider> ?
16:06:22 <Mat2> the list indexes are out of bound
16:07:00 <erider> I know but I am not sure why 
16:08:17 <Mat2> because the interpreter do only support lists with a fixed max. element depth
16:08:55 <erider> just finished coding it up and my main thing was to make sure the data got parsed right
16:09:26 <erider> that seems to be ok but the extra byte 
16:11:45 <Mat2> no I mean the TCL interpreter is written in C
16:12:08 <erider> yeah like most interpreters 
16:13:07 <Mat2> lists are implemented in a very restricting way on the heap, so the maximal number of list elements supported is limited
16:14:05 <erider> I was looking into making it dynamic 
16:14:46 <erider> tcl 8.6 has that built in out of the box 
16:15:11 <Mat2> ah ok, I use an older interpreter
16:15:32 <erider> I use 8.5 
16:17:41 <Mat2> ok, I must get some sleep, if you do not find the error I trace the code (probably ip is not setted correctly at some point)
16:18:32 * erider just found a debugger (installing it now)
16:18:59 <Mat2> I'm off ciao
16:19:08 <erider> good night 
16:19:15 --- quit: Mat2 (Quit: Verlassend)
16:42:25 --- quit: kumul (Read error: Connection reset by peer)
16:43:16 --- join: kumul (~kumul@173.215.194.228) joined #retro
17:49:13 --- quit: karswell (Remote host closed the connection)
17:59:35 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
18:12:36 --- quit: karswell (Remote host closed the connection)
18:22:51 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
21:40:15 --- quit: karswell (Remote host closed the connection)
21:50:30 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
21:55:20 --- nick: tangentwork -> tangentstorm
22:02:35 --- quit: karswell (Remote host closed the connection)
22:12:50 --- join: karswell (~coat@93-97-29-243.zone5.bethere.co.uk) joined #retro
23:01:01 --- quit: kumul (Quit: :) good luck!)
23:59:59 --- log: ended retro/13.01.08