| |
Hook
Account closed
Registered: Jun 2003
Posts: 1 |
code..
hey all... how do i makle a demo.. somboddy.. plz help me.. i am a frind of perff the webmaster.. |
|
... 5 posts hidden. Click here to view all posts....
|
| |
Graham
Account closed
Registered: Dec 2002
Posts: 990 |
where 2^266 is the number of all atoms in the entire universe... |
| |
T.M.R
Account closed
Registered: Dec 2001
Posts: 749 |
i think that's where i've been going wrong; i guessed the number of atoms...? |
| |
HCL
Registered: Feb 2003
Posts: 728 |
Hmm.. Does that mean there are more demos than atoms in the universe!? That's cool ;), just start diggin!! |
| |
Perff
Administrator
Posts: 1680 |
Hmm. It should be easy enough to make a small script-thingy that would create all the possible disk one by one, and then test it one some emulator or something to see if it would run anything. If yes it could store it for some humans to look at it later, else just trow it away.
Now all we need is an 9999999999999999999 mhz (or faster) machine to run the script on.
(Perhaps a new idea for a distributed network? "Search For C64 Demos")
;) |
| |
MagerValp
Registered: Dec 2001
Posts: 1078 |
I've actually thought about making random code generators for small algorithms (say 10 bytes or less). With a table of legal instructions, and a small, fast 6502 simulator that you run for 1000 cycles or so, you should be able to go through a lot of combinations :) Then you need a way of specifying the initial reg values, and some sort of goal, and just let it crunch. Bound to end up with some interesting results :)
|
| |
Cruzer
Registered: Dec 2001
Posts: 1048 |
What we need is an artificial intelligence program that can look at the emulator output each time a piece of random code runs, and see if it looks like a kewl effect. Then the human user only needs to look at the most interesting routines, instead of a lot of junk that might run without crashing, but is bound to be utter crap in 99.999...% of the cases. |
| |
WVL
Registered: Mar 2002
Posts: 902 |
I'm wondering how many of those 'demos' will actually turn out to be 'SEUCK'-games ;) |
| |
HCL
Registered: Feb 2003
Posts: 728 |
Autoboy/HT was already into this topic some time ago. I think he went into that darkness while studying the AI-course at the University here in Linköping (I had the same course some years ago, but it didn't affect me that badly ;). Then i started to 'teach' him some c64-coding, and hoped to see some mutating demo-thingie.. What happened since then, i don't know. |
| |
Cruzer
Registered: Dec 2001
Posts: 1048 |
@HCL: Yeah, I could imagine it would be quite a black hole to fall into to start implementing something like that with current technology. I guess we need computers that are alot faster than today (like quantum computers) and some almost human level AI, before it's realistic. So it will probably take a couple of decades before us human c64 coders are obsolete. ;) |
| |
HCL
Registered: Feb 2003
Posts: 728 |
@Cruzer: Ok then, i'll go on coding my demo effects until someone comes up with that tool.. ;) |
| Previous - 1 | 2 - Next |
