29 Nov
2016
29 Nov
'16
3:29 p.m.
On 2016-Nov-29, at 1:10 PM, Paul Berger wrote:
On 2016-11-29 5:00 PM, Eric Smith wrote:
Those are the statements I was recalling the sentiment of.
I used the phrase proto-DSP in the sense of not-quite a DSP but directed towards high
data/IO throughput, or precursor to DSPs.
I remember the device from going through the Osborne book, and had wondered where it might
have been actually used, thinking of medical or mil/avionics, so it's interesting to
hear in this thread it found use in more commonplace/accessible products.
On Tue, Nov 29, 2016 at 12:11 PM, Brent Hilpert
<hilpert at cs.ubc.ca> wrote:
I just went and got my Osborne book...
" The 8X300 is described by its manufacturer as a "microcontroller" rather
than a "microprocessor". This distinction draws attention to the unique
capabilities of the 8X300 which make it the most remarkable device described in this
book"
"The 8X300 is designed to serve as a signal processor or logic controller, operating
at very high speed." is the 8X300 the 'unusual', proto-DSP
device discussed in a chapter of
the Osborne book?
I don't know whether Osborne described it that way, but since the 8X300 and
8X305 don't have a multiplier, only support 8-bit addition, and can only
address 512 bytes of data address space, they wouldn't make very good DSP
processors. However, since all instructions execute in 250 ns, they'd be
able to perform some DSP operations faster than contemporary 8-bit
microprocessors, which usually took at least 2 us for the simplest
instructions, and even longer for fancier instructions.