----- Original Message -----
From: <ncoplin@orbeng.com>
> For a while I've been assessing the feasibility of using a PC as a
> sort-of-SuperCPU, getting PC software to run an emulation of the CPU and a
> cable /adaptor to control the write back and I/O read functions necessary.
> With modern PCs some pretty amazing CPU emulation speeds are possible, but
> its all brought back to reality by the need to write to the real hardware.
>
Hello, Nick-
I checked out the 64HDD page and it does contain a lot of neat stuff and
interesting information. I checked out your 'V6510' project and it looks
good and looks well thought out. Apparently, you seem to be familar with the
SuperCPU and how accelleration should operate asynchronously for CBM
computers, how host memory should be mirrored, how i/o registers should be
handled, etc.
However, a 'fast' 6510 via the V6510 is just a 6510 in its stark essence. I
would favor using a 65816 core engine. There should be C language models of
the 65816 available and modern computers sport speeds enough to provide an
accellerated 65816 flavor.
I mean, all of this speed gained by the V6510 will still be bottled up by
the 64Kb memory limit. I'm aware of the REU and c128's memory banking
schemes. But such memory is *indirectly* addressed and limits the 6510/8502
core accordingly. The 65816 allows me to program and run geoZIP
(http://www.cs.tut.fi/~albert/Dev/gunzip-geos/), the WAVE, my Silicon Dreams
program (animation player), cool 65816 specific demos, etc.
Speed alone is very nice and I think you've struck reasonable compromises in
designing the hardware and software in providing a low cost solution for
Commodore based accelleration. But at the same time, I feel it should be
useful as well, by allowing the ability to emulate the extra opcodes, the
increased stack space, relocatable zero page space and sporting *directly*
addressable 16Mb of RAM. The 65816 emulation core will fit the bill very
nicely.
That said, I probably would not use it. I do own a SuperCPU, after all. :)
But for many Commodore owners, a low cost accelleration solution would be
very beneficial to those people. Moreover, having a 65816 engine readily
available at a low cost resulting in a wider reach within the Commodore
audience will spur more software development which can only benefit the real
McCoy, the SuperCPU, as it is kinda lacking in 65816-specific software.
Enjoy.
-Todd Elliott
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail 2.1.1.