Accelerator boards - no future? Bad business?
echristopherson at gmail.com
Fri Apr 22 13:51:37 CDT 2016
On Fri, Apr 22, 2016 at 1:25 PM, Guy Sotomayor <ggs at shiresoft.com> wrote:
> > On Apr 22, 2016, at 11:03 AM, Swift Griggs <swiftgriggs at gmail.com>
> > Remember all the accelerator boards for the Mac, Amiga, and even PCs in
> > 90's ? I've often wished that I could get something similar on my older
> > systems. For example, fitting an R16k into an O2 or doing dynamic
> > translation on a 4.0Ghz i7.
> The main reason for a lot of the “accelerator” boards was to be able to run
> non-native code on the same platform. With the advent of performance
> micros (latest x86 and Power) there is little need to do that because
> (or dynamic translation) is fast enough and with the various virtualization
> capabilities, it’s not unusual to have multiple different OS’s running on
> same HW.
> Apple did this with some success in it’s various CPU transitions. When
> switched from 68K to PPC, the PPC emulated the 68K code. The same
> happened when they switched from PPC to x86, again the PPC code was
> emulated (actually in that case it was dynamically translated).
I've heard about PPC Amigas from time to time. What was the state of
68K->PPC emulation or dynamic translation on those?
> > So, here's the question. Is my dream likely to ever be possible enough
> > a boutique shop could pull it off and not lose their shirt on the
> > costs and R&D to do it ? I'm encouraged by things floppy emulators that
> > produced for these old machines. However, that's probably significantly
> > easier to make than a CPU accel board.
> It’s also keeping in character with the old machines. It’s not adding new
> capabilities but more replacing old peripherals with something a bit more
> convenient. Adding in a new accelerator means not only developing the
> HW but also writing a boatload of *new* SW in order to be able to take
> advantage of it.
> Most of what you see (and what I’m mainly doing these days in terms of
> “hobby”) is producing parts to a system in order to keep it running rather
> than adding completely new capabilities. Most companies would rather
> spend their time and budget doing things for a high ROI and for large and
> growing markets.
I like the new types of peripherals but it makes me a little uncomfortable
knowing that e.g. in the case of the uIEC-SD for Commodores, the clock
speed of the peripheral is 16 to 20 times that of the original host CPU. I
keep hatching little schemes of perhaps putting a tiny OS kernel in the
thing, but at that point *it* would become the computer and the 128 would
be just sort of sitting there. The same is true of the CosmosEx device I've
been thinking of getting for my Atari STs; it has a Raspberry Pi inside.
More information about the cctalk