🚀 go-pugleaf

In article <26701@cup.portal.com>, mmm@cup.portal.com (Mark Robert Thorson) writes:
> Several days ago, amoss@morgan.com said he bought a Cyrix 387-clone FPU
> for less than the price of a 387 from Intel.  (The Cyrix chip is pin
> compatible, but faster;  Cyrix claims 4-10 times faster.)
> 
> I spoke to Cyrix today, and I was told that the list price for their part
> is EXACTLY the same as for Intel.  I was also told that there were chips
> from the previous stepping still in the distribution channels, and that 
> these chips may be the ones being sold for "bargain basement" prices
> (actual quote).  Caveat Emptor.

Well, I didn't say I bought one at the lower prices. We had to get ours
before we found out the lower quotes. It happens that we get the '387
chips at less of a discount than the Cyrix. Given that our chip is 
a bit older, and we regressed it against the '387 in most ways we 
could think of to no great ill effect, I'm not sure the earlier step
chips are so dangerous, if that's what we have. What is the bug that
pushed them to another step? We could look for it if it's something
specific. One of our groups did an evaluation which has prompted a
evaluation for a Firm-wide hardware support group, and I'm sure they
would be most interested in any known vulnerability. I am guessing 
that running one of these chips for a four hundred hours isn't the same
as running ten of them for a week, so it's quite possible that 
someone with more copies of the chip would be in a better position
to say if quality has changed recently.

Later,
Andrew Mullhaupt