[Coco] Some notes on ROM Paks

[Arthur Flexser]

I agree with all these points.  I have a very vague recollection that
Cyrus Chess may not have worked unmodified on the CoCo 3 due to a checksum
test that was thrown off by the CoCo 3 $FE00 RAM--can anyone confirm that?
Or maybe the reason that there was a second version of Downland was this
sort of CoCo 3 incompatibility?

Art

On Mon, 26 Apr 2004, Boisy G. Pitre wrote:

> As Art posted a few days ago, we've been having a private thread about 
> getting to the last 512 bytes of a ROM Pak (either 16K or 32K), or even 
> the 32K of CoCo 3 ROM.
> 
> Here are some conclusions that I've drawn from the discussion.  Feel 
> free to prove or disprove them.
> 
> (1) On a CoCo 2, it is not possible to access the last 256 bytes of a 
> ROM Pak ($FFXX).  Therefore, 16K ROM Paks do not have valid code in the 
> last 256 bytes.
> 
> (2) On a CoCo 3, it IS possible to access the last 512 bytes of a ROM 
> Pak ($FE00-$FFFF) through use of the MMU register trick as Art 
> outlined.
> 
> (3) Even so, it is unlikely that CoCo 3 ROM Paks use the last 512 bytes 
> of the ROM Pak, unless the game uses the MMU register trick to get at 
> those last 512 bytes.
> 
> Assuming these hold true, then:
> 
> (1) It is safe to copy all but the last 256 bytes of a 16K ROM Pak into 
> a blank EPROM, and that game should run fine on a CoCo 1/2.
> 
> (2) It *may*, and in all likelihood *IS* safe to copy all but the last 
> 512 bytes of a 16K or 32K ROM Pak from a CoCo 3 into a blank EPROM, and 
> that game should run fine on a CoCo 3.
> 
> (3) Since 16K ROM Paks designed for the CoCo 2 could possibly use the 
> second to last 256 bytes ($FEXX), such a PAK would NOT work properly on 
> a CoCo 3 since the CoCo 1/2 era Pak wouldn't know about the MMU or the 
> CoCo 3 (does anyone have an example of such a Pak?)
> 
> Boisy
> 
> 
> -- 
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco
>