No Subject
- To: misc
- From: Wayne Morellini <waynemm@xxxxxxxxxxx>
- Date: Wed, 07 Jul 1999 15:44:36 EST
From: theFox@UltraTechnology.com (Jeff Fox)
>>F21 chips that
>>nobody really wants to buy because there ill suited to the market. With
>>them being as usefull as a backside scratcher.
>Can you explain what market you were refering to?
A hypothetical situation that might happen to a Misc processor that is not
as cutting edge as it's origional planned release date and with not much
marketing focuss, not refering specifically to the F 21 of course.
The more I look at the F21 the more cool it looks, if you can attach other
chiups to it to do extra things. Of course 20 bit interfaces make using
32-bit devices (and memory) a bit more costly to interface to and pdram is
a bit slow for it's speed. Of course a p32 with twin sdram P133 ports (or
ddr 100, giving 200Mhz equivalent) would solve everything, leave tongues on
the ground etc ad, not quiet, infinitum ad nauseum (actually I think I am
just having the same effecrt) but at this stage way to
expensive(unrealistic). Why dual port, well as chuck stated sdram has a 1-2
cycle hit for sdram so unless you cache video data the video processor would
drag the memory interface down to around 43mhz.
What market? Any market that allows you ship lots of 10, 000 chips and
become well off for your efforts. The embedded market might be good for
1000 lots or with a deal 10,000 lots. The set top box market was a good
idea when it was suggested a year por so before ITV was formed. Starting
out quickly, rapid developement with gate arrays and standard parts to ship
within 1-2 years would have provided proof of concept, marketable products
(little competition) and cash for custom silicon research (to replace those
gate array parts). What else maybe make it ussuable as a cell in cell
libaries for other designs. Pick a niche (say under performing handheld
games systems that can't pack in the normal power circuitry, the Atari
Jagura handheld system had a 10 mintue batrtery life). Anything that is
going to add performance, convience and save money, that has little
competition (management saves, engineers enjoy, and marketability).
Origionally the piont was about us needing the discussion on volume markets
here because a limited hobbyist market can't support it.
>>7. work out your own new designs in FPGA or VLSI
Hey didn't somebody mention a 500Mhz prograqmmable part here before. Would
that make a good basis for a cheap hobbyist version of the F21 (trying not
to say 32) until custom silicon comes.
I still think it is an excellent job though, and commend you for it.
Thanks.
Wayne.
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com