gt1342a@prism.gatech.EDU (PLACENTA) (01/17/90)
When I said that SCO sez "Don't use 386SX machines for 386DX specific software
because it doesn't work properly," I was quoting Byte magazine of last month.
Anyone with any doubts may check for his or herself. Thanks for the snotty
comments in the mail, though, /dev/null appreciates them. Let me explain that
the article in Byte, which I was advising people to check out, says this far
mor eloquently than I. The issue is NOT, I repeat, NOT incompatibility within
the 386/386SX processor. The key word is BUS. Yes, the 386DX and 386SX are
the same inside, but the 386SX is on a 16 bit bus. THIS IS THE POINT. SCO
claims that their software does not work properly in a 16 bit 386SX machine,
and since it is their product, one could probably trust them. Byte also throws
some cold water on the hope of using a 386SX machine as a unix box all on their
own. I'm not about to retype the contents of the article - anybody who is
interested for any reason in what Byte has to say on the subject can read the
magazine. Any problems with what they say need not be addressed to me; address
it to Byte. Any more premature, Nasty mail sent to me will be forwarded to
/dev/null. Next time, before you flame, why don't you check out the facts?
--
Charles LeDuc | gt1342a@prism.gatech.edu | GT PoBox 31342a Atl GA 30332-1001 USA
NetMail 7301/203or516 | BIDET?? Try washing your ENTIRE body, France - Trust me.
...!{decvax,hplabs,ncar,rutgers,purdue}!gatech!prism!gt1342a |Please apply all
...!{allegra,amd,hplabs,ut-ngp}!gatech!prism!gt1342a | Standard Disclamers.md@sco.COM (Michael Davidson) (01/18/90)
gt1342a@prism.gatech.EDU (PLACENTA) writes: > SCO claims that their software does not work properly in a 16 bit 386SX > machine, and since it is their product, one could probably trust them. Sorry, but this statement is not correct. If anyone at SCO made this claim they were wrong. The 386SX is fully supported by SCO software.
kelly@uts.amdahl.com (Kelly Goen) (01/19/90)
In article <4512@scolex.sco.COM> md@sco.COM (Michael Davidson) writes: > >gt1342a@prism.gatech.EDU (PLACENTA) writes: > > >> SCO claims that their software does not work properly in a 16 bit 386SX >> machine, and since it is their product, one could probably trust them. > >Sorry, but this statement is not correct. If anyone at SCO made this >claim they were wrong. The 386SX is fully supported by SCO software. OK ALL lay off the poor guy already... this subject has been thrashed to death .... The article in question referred to usage of certain 16 bit memory cards being used in the 32 bit environment... what it actually relates to is inadequate hardware designs of CERTAIN vendors memory cards... what actually causes these panics is insufficient wait states for the access to the card.... without sufficient wait states the address latches on the card are not in the proper state when the memory read or write cycle starts...result is that the OS(doesnt matter if it is UNIX XENIX or 32 bit vcpi under dos...) no memory appears present to the processor then the panic results...btw if you have a 386 with the Chips and technology chip set and an AMI bios ...you can vary the number of wait states and produces this panic at will(kids dont try this on a PRODUCTION box).... hope this will clarify matters to the readers of this newsgroup... by the way does anyone know how to route from the internet to bix??? I really want to set the byte authors hair on fire over this one... SCO if you have anything further to add I can be emailed at kelly@uts.amdahl.com...or simply publish to this group... cheers kelly