RE: Building my monster C64 accelerator

From: Oliver Achten (achten_at_gmx.de)
Date: 2003-08-19 20:40:46

Hallo Ruud!
 
> > > Remark for Oliver: when using the 65816, what about using its 16 MB
> > > capabilities? Can provide you with quite some nice ideas.
> >
> > I´m eager to hear them... :-)
> 
> The start is quite simple:
> - you need a 74LS245 as buffer between the data lines of the C64 and 65816
> - you need a 74LS541 to buffer the addresses A11..A15 and the R/W line
> between the C64 and 65816
> - you need pull-up resistors for the addressbus and R/W line
> - you need a 74ALS573 to generate A16..23.
> - you need a 74LS688 to find out if the firts 64 KB are addressed or not.
> - you need some small glue logic
> 
> When data is read or written to the first 64 KB area, the original C64
> part
> s
> are addressed. If not, the 688 will disable the 245 and the 541 which
> cause
> s
> the C64 to think that data is read from the $Fxxx area.
> 
> Now you are free to use the complete 16 MB :)
 
Yeah, but wouldnt this actually kill the i/o space in the other banks? Why
would i want that when i could switch it off using the PLA anyway? This would
also kill the idea of having the virtual C64s...
Hmm, perhaps all memory above 512KB could be made accessible using this way.
Then i could do what you describe: enabling full 17.72Mhz speed to rest of
the 16MB address space...

> > No need to do that. With the fast SRAM replacing the old
> > DRAMs, everything (except register writes) can run at full speed.
> 
> Don't forget the ROMs. (but you mentioned using the 29F010)
> 
> 
> > Why slow-down writes when the Drams are replaced anyway? Only
> > access to the colour ram has to be slowed down.
> 
> Why not replace it as well? You mention the use of 8 virtual C64s. Then
> thi
> s
> also meeans the need of 8 color RAMs. Other wise you will face the problem
> that as soon as one virtual C64 changes the backgroundcolors, this action
> will affect ALL C64s.

Hmmm, that´s actually a very good idea... Every 65816 bank would have its
seperate 1k color ram address space, and the use of an 8KB cache ram chip would
make the color ram as fast as the rest....

-- 
COMPUTERBILD 15/03: Premium-e-mail-Dienste im Test
--------------------------------------------------
1. GMX TopMail - Platz 1 und Testsieger!
2. GMX ProMail - Platz 2 und Preis-Qualitätssieger!
3. Arcor - 4. web.de - 5. T-Online - 6. freenet.de - 7. daybyday - 8. e-Post


       Message was sent through the cbm-hackers mailing list

Archive generated by hypermail pre-2.1.8.