Re: Thoughts on 8-bit acceleration.

[mdj <mdj.mdj@gmail.com>]

On May 6, 11:17 pm, "sfahey" <sfa...@a2central.com.remove-mnb-this>
wrote:
>   To: mdj
>   Re: Re: Thoughts on 8-bit acceleration.
>   By: mdj to comp.sys.apple2 on Thu May 06 2010 01:02 am
>
>  > card. Even with all that it's still vastly inferior to the Zip design.
>
> Interesting, and I'm not disputing your analysis but I'll add that card-based
> accelerators seem to be more reliable than Zip Chips. With their fragile pins,
> internal cache failures (and a few other issues) they are hardly a panacea.

Absolutely, and from that perspective the Transwarp II is the finest 8
bit accelerator since it's ostensibly a RocketChip on a card.

Ironically my Transwarp I which I've used as a fallback accelerator
when without a working Zip has also developed RAM problems but it was
easily fixed.

> Tony suggested to me that the ZipGS might be able to work in a //e, because it
> lacks the voo-doo the TWGS has onboard. It would just take some tweaking of the
> ASIC possibly.
>
> What do you think of that?

Tweaking an ASIC sounds pretty difficult to me... A Transwarp II style
accelerator should be doable using the Carte Blanche. Should be
possible to go a lot faster too as the Zip design has quite a bit of
headroom for increasing the multiplier. I'd expect that at somewhere
between 16 and 32Mhz write-back would slowly start to bottleneck it,
but it'd be interesting to see where the limit is

Matt