[sci.astro] Computer

schaper@pnet51.orb.mn.org (S Schaper) (10/28/89)

I'd opt for 80860's that are rad-hardened, or 88k's in parallel, I have a
friend in Mass, programming AI for a company already producing a machine with
508 88k's in parallel. And Laser Ram Disks to record data during those times
when there is too much gathered to radio back real-time.

Disclaimer: on this message AI means expert system type software, not Boar
Power   :-)

Steve
Schaper

UUCP: {amdahl!bungia, uunet!rosevax, chinet, killer}!orbit!pnet51!schaper
ARPA: crash!orbit!pnet51!schaper@nosc.mil
INET: schaper@pnet51.cts.com

stevo@uniblab.Jpl.Nasa.Gov (Steve Groom) (11/01/89)

In article <1368@orbit.UUCP> schaper@pnet51.orb.mn.org (S Schaper) writes:
[ referring to the future use of "modern" processors for flight hardware,
in particular a version of the 8086 ]
>I'd opt for 80860's that are rad-hardened, or 88k's in parallel...

Of course you would.  Believe me, the people responsible for flight
software would too.  The problem is that it's not a trivial task to
design and produce rad-hard parts.  You don't just whip up a special
version of a chip in a few months and stick it in a spacecraft.  That's
why a rad-hard version of the 8086 is just now being discussed for use,
because has taken this long to produce it.

Personally, I'd rather see them using a SPARC :-)

--
Steve Groom, Jet Propulsion Laboratory, Pasadena, CA
stevo@elroy.jpl.nasa.gov  {ames,usc}!elroy!stevo