00:00:00 --- log: started retro/13.08.03
00:02:00 --- join: sirdancealo2 (~sirdancea@cst-prg-100-24.cust.vodafone.cz) joined #retro
01:16:29 --- quit: Endeg (Ping timeout: 264 seconds)
01:34:45 --- quit: sirdancealo2 (Ping timeout: 240 seconds)
01:47:15 --- join: sirdancealo2 (~sirdancea@cst-prg-2-3.cust.vodafone.cz) joined #retro
02:47:30 --- join: ncv (~quassel@unaffiliated/neceve) joined #retro
03:49:30 --- quit: sirdancealo2 (Read error: Connection reset by peer)
05:01:37 --- join: sirdancealo2 (~sirdancea@98.82.broadband5.iol.cz) joined #retro
10:14:33 --- join: Mat2 (~claude@91-65-144-133-dynip.superkabel.de) joined #retro
10:14:36 <Mat2> hello
10:15:02 --- join: Kumul (~nmz@adsl-64-237-236-72.prtc.net) joined #retro
13:47:22 <tangentstorm> Mat2, crc ...re: "15:05:12 Mat2 a document to specify port addresses and there associated device interfaces"
13:48:01 <tangentstorm> do you think it's worth setting up a register of ports?
13:48:11 <tangentstorm> I'm a bit hesitant...
13:48:28 <Mat2> hi tangentstorm
13:48:52 <tangentstorm> hi :)
13:49:01 <Mat2> short answer: yes
13:49:19 <tangentstorm> how many ports are there anyway? i think i put 64 in the pascal version..
13:49:41 <Mat2> I think we get somewhere about 20
13:50:12 <tangentstorm> i'm planning a websocket port
13:50:25 <Mat2> used out of (insert word size here)
13:50:47 <Mat2> ok, 21
13:51:02 <tangentstorm> i have some weird ideas about using them for things like queues too.
13:51:39 <tangentstorm> like like extra, external stacks/queues/rings
13:51:49 <tangentstorm> all kinds of stuff really.
13:52:58 <Mat2> the problem is: Retro should be platform independent. This way the environment can be used for application and system-programming in an efficient way
13:53:00 <tangentstorm> but on the other hand, i'm also probably going to move a little bit away from pure ngaro, for simple reasons like wanting floats.
13:53:05 <tangentstorm> yeah.
13:53:50 <tangentstorm> i don't really have a better idea anyway. i almost want to treat the ports like objects with a plug and play interface, but i know that complicates the code.
13:55:56 <Mat2> Parabel support different data-types. For a MISC style VM it would be probably better to support floating-point arithmetic though integer approximation
13:56:14 <Mat2> (that is also more precise)
13:57:13 <tangentstorm> i guess i'm happy with anything as long as it runs.
13:57:56 <tangentstorm> have you ever worked with the "ascii coded decimal"? (i kept seeing it in the intel instruction set)
13:59:00 <Mat2> yes, you compile for IA32 architectures (32 bit) ?
14:07:24 <tangentstorm> i'm not doing anything with it yet... i just saw them and was curious. :)
14:08:10 <Mat2> you can see it as some kind of BCD conversion
14:08:44 <Mat2> one useful usage is converting binary numbers to there ASCII equivalent
14:09:23 <tangentstorm> like a string?
14:10:07 <Mat2> however, for AMD64 in 64-bit native mode, these instructions are invalid opcodes and no longer supported (as I know)
14:10:22 <Mat2> ^yes
14:10:39 <tangentstorm> ah.
14:11:00 <tangentstorm> can you switch back and forth between ia32 and amd64 modes? i got the impression you couldn't.
14:11:19 <Mat2> you can
14:11:31 <Mat2> (but you wouldn't)
14:12:02 <tangentstorm> :D
14:12:47 <Mat2> there exist even some kind of 64-bit unreal mode for accessing memory above 4 gB within real-mode
14:14:20 <tangentstorm> i think i've seen that.
14:14:45 <Mat2> but these kind of hack is not portable because it depends on some system-registers
14:15:00 <Mat2> onyl avariable for AMD cpu'S
14:15:04 <Mat2> ^only
14:19:54 --- quit: Kumul (Quit: Divided by 0)
14:22:14 <Mat2> I would avoid them
15:25:22 <Mat2> ^would
15:25:29 <Mat2> get some sleep, ciao
15:25:36 --- quit: Mat2 (Quit: Verlassend)
17:26:37 --- quit: ncv (Remote host closed the connection)
17:37:34 --- join: Kumul (~nmz@adsl-64-237-236-72.prtc.net) joined #retro
21:16:50 --- quit: Kumul (Quit: Divided by 0)
23:59:59 --- log: ended retro/13.08.03