> 1) 2 cartridges for each game. One with the game code and the other for
> the
> vector ROMs
>
True, or a cartridge for program/special functions, and a board with ROM
sockets for VROMs for each game...
> 2) I can build a universal vector ROM that contains things that would be
> used by most games (i.e.- characters), and all cartridge games would use
> that ROM. This would mean that all other shapes would have to be copied
> to,
> or built in, vector RAM. (that's how I'm doing space wars, the ship shapes
> are in normal program ROM space)
>
I did about the same thing for Vector Breakout-- just used the "stock"
Tempest ROM for characters a few shapes and then loaded VRAM from program
code.
> 3) Run a cable from the VROM socket to the cartridge so the vector
> generator
> would be accessing the vector ROMs on the cartridge. (this might not work
> because of delay)
>
This is probably OK, but shielding and crosstalk on a ribbon cable of any
length can be a problem without special wiring. Not my first choice...
> 4) Have a static RAM chip in the daughter card and copy the appropriate
> data
> down when the game is selected. There's a couple ways to do this, but the
> easiest would be for each game to copy the data down during
> initialization.
> This option also might be cool because it would allow shapes to be altered
> in what was previously only static data.
>
I think this is the best way. I don't know Space Duel hardware super-well
though-- is there a 74LS245 or something similar on the VROM databus (so
it's bidirectional). I *think* so 'cause the VRAM is in the same address
decoder range. The menu system would just load the VROM image into the
"shadow" VROM space prior to jumping to the game program code. I also like
that idea 'cause I can stick a 32K*8 SRAM there and we can add a
bank-selection register somewhere in memory. Then, since the VGGO is under
program control we can bank-switch the shadow VROM before every redraw to
achieve something like character-set animation without much of any CPU
use...
-Clay
Received on Mon Nov 1 13:31:25 1999
This archive was generated by hypermail 2.1.8 : Fri Aug 01 2003 - 00:32:26 EDT