matt@srs.UUCP (Matt Goheen) (01/17/89)
I have heard that the difference between the 80386 and the 386SX is a
matter of a couple of interal pin connections and a retail price change
of something like $200. Now, all we have to do is buy up a bunch of
386SX chips, open them up and rewire the necessary pins -- think it
can be done for $200 per chip? (might not be too hard for ceramic,
dunno about plastic)
--
- uucp: {rutgers,ames}!rochester!srs!matt Matt Goheen -
- internet: matt@srs.uucp OR matt%srs.uucp@harvard.harvard.edu -
- "We had some good machines, but they don't work no more." -mslater@cup.portal.com (Michael Z Slater) (01/19/89)
> I have heard that the difference between the 80386 and the 386SX is a > matter of a couple of interal pin connections and a retail price change > of something like $200. I don't believe this is true. Reliable sources have told me that it is indeed a different design. And by the way, the difference in price isn't that much -- as of July, the 1000-pc prices were $207 for the 386 and $165 for the SX, both at 16 MHz. The lower-cost package of the SX is in part responsible. A better example of pricing for perceived value rather than by mfr cost is the 386SX vs the 376. The 376 is basically the same as the SX, but lacks paging, real mode, and V86. It is priced at $99 in 100s -- dramatically less than the SX. Even though the chip size is possibly a bit smaller it certainly doesn't make that much difference. This is a clear indication that the 386 margins are very high, and that the basis of the price is DOS compatibility. You can think of the 376 as a "De-DOSed" 386SX, priced lower to appeal to the embedded control market but crippled so it won't compete with the SX. Michael Slater, Microprocessor Report mslater@cup.portal.com