bareta@ihuxv.ATT.COM (Benyukhis) (01/10/89)
Can anyone on the net describe what kind of an animal the 386/SX chip is, as well as its price and availability if possible. Thanks much, Edward Benyukhis
pjh@mccc.UUCP (Pete Holsberg) (01/11/89)
In article <3114@ihuxv.ATT.COM> bareta@ihuxv.ATT.COM (Benyukhis) writes:
=
=Can anyone on the net describe what kind of an animal the 386/SX
=chip is, as well as its price and availability if possible.
=
The 386sx is essentially a 386 with an external 16-bit data bus. Very
much like the 8088/8086 relationship.
--
Pete Holsberg UUCP: {...!rutgers!}princeton!mccc!pjh
Mercer College CompuServe: 70240,334
1200 Old Trenton Road GEnie: PJHOLSBERG
Trenton, NJ 08690 Voice: 1-609-586-4800
smith@iwblsys.UUCP (Mickey Smith) (01/11/89)
In article <3114@ihuxv.ATT.COM>, bareta@ihuxv.ATT.COM (Benyukhis) writes: > > Can anyone on the net describe what kind of an animal the 386/SX > chip is, as well as its price and availability if possible. > The 386SX is a full implementation of the 386DX part and supports all 386DX software. Its is not an upgrade to a 80286. The 386SX runs at 16 MHz and has a bus bandwidth of 16 Megabytes/Sec. The external bus is a non multiplexed 16 bit data bus and 24 bit address bus. The part is supportted with a 80387SX for Numerics. The part is available now and pricing is available through an Intel Field Sales Office. -- Mickey Smith, Regional Software Specialist | Opinions are my own and not Intel Corporation | that of my employer. 7071 Orchard Lake Road | W. Bloomfield, MI 48332 (313) 851-8096 | So there!!
aris@altger.UUCP (aris) (01/12/89)
In article <569@mccc.UUCP> pjh@mccc.UUCP (Pete Holsberg) writes: >The 386sx is essentially a 386 with an external 16-bit data bus. Very >much like the 8088/8086 relationship. Will the 386SX run 386 software such as SCO xenix 2.3.1 (And VPIX under that..)? Or will there be a new version of SCO Xenix that will support the 386SX chip? Thanks in advance, Aris Stathakis ..!uunet!mcvax!olnl1!olgb1!olsa99!aris and/or ..!uunet!mcvax!unido!altger!aris
john@data.UUCP (John Mantey) (01/14/89)
The 386SX is basically a subset of the 386 processor. The SX does not have as large an address space as the 386 (ie 24 bit address bus instead of a 32 bit address bus). This gives the 386sx an address capability of 16M of physical ram as opposed to the 4G of the 386. The SX only has a 16 bit data bus, similar to the 286, instead of a 32 bit address bus. This just means it takes 2 data cycles to get data that a 386 would be able to get in one. Other than that the two processors are the same. Software written for the sx will run unmodified on the 386, and software for the 386 that never tries to access physical memory above 16M will work on the SX. As to availability and cost, don't ask me, cause I don't know. johnm
sjb@dalek.UUCP (Seth J. Bradley) (01/14/89)
In article <1304@altger.UUCP> aris@altger.UUCP (aris) writes: > Will the 386SX run 386 software such as SCO xenix 2.3.1 (And > VPIX under that..)? Or will there be a new version of SCO Xenix > that will support the 386SX chip? The 386SX has been fully certified under UNIX. There should be no need for a special version. -- Seth J. Bradley UUCP: uunet!{lll-winken|ubvax}!dalek!sjb Internet: lll-winken.llnl.gov!dalek!sjb
akm@uoregon.uoregon.edu (Anant Kartik Mithal) (01/14/89)
I recently had the opportunity of using a NEC powermate with a 386SX, 16 MHz and 2 MB of memory. On running Norton's SI (ver 3.00) on it (the only software I could get: the machine uses 3.5" drives, and my software is all 5.25") produced some strange results. Approximately half the time it gave speeds of 15.7 +/- .1 and the (approximately) other half it gave speeds of 18.3 +/- .1 Can anyone explain this? I also felt (gut feeling, coloured by the speed of the hard disk in the machine, and I have no idea how fast that was) that the machine was not particularly faster than my 12 MHz, 0 Wait State 286.
tif@cpe.UUCP (01/14/89)
Written 7:43 am Jan 12, 1989 by altger.UUCP!aris in cpe:comp.unix.xenix >In article <569@mccc.UUCP> pjh@mccc.UUCP (Pete Holsberg) writes: >>The 386sx is essentially a 386 with an external 16-bit data bus. Very >>much like the 8088/8086 relationship. > Will the 386SX run 386 software such as SCO xenix 2.3.1 (And > VPIX under that..)? Or will there be a new version of SCO Xenix > that will support the 386SX chip? It does require a different version of SCO XENIX 386. I suppose their are some subtle differences that make the regular version non-functional. I have no information on the availability of the different version.
dyer@spdcc.COM (Steve Dyer) (01/15/89)
In article <6800062@cpe> tif@cpe.UUCP writes: >It does require a different version of SCO XENIX 386. I suppose >their are some subtle differences that make the regular version >non-functional. Huh? Where'd you get THAT idea?? -- Steve Dyer dyer@ursa-major.spdcc.com aka {ima,harvard,rayssd,linus,m2c}!spdcc!dyer dyer@arktouros.mit.edu
tif@cpe.UUCP (01/17/89)
Written 2:14 am Jan 14, 1989 by uoregon.UUCP!akm in cpe:comp.unix.xenix >I recently had the opportunity of using a NEC powermate with a 386SX, 16 MHz >and 2 MB of memory. On running Norton's SI (ver 3.00) on it (the only >software I could get: the machine uses 3.5" drives, and my software is >all 5.25") produced some strange results. Approximately half the time it >gave speeds of > 15.7 +/- .1 >and the (approximately) other half it gave speeds of > 18.3 +/- .1 Could this be related to something like the load address of the program being on (or not on) a 4-byte boundary and requiring something extra for a fetch? Paul Chamberlain Computer Product Engineering, Tandy Corp. {killer | texbell}!cpe!tif
mslater@cup.portal.com (Michael Z Slater) (01/17/89)
> 386SX
Pricing for the 386SX at the time of its introduction last June was
$219 in 100s and $165 in 1000s. Note that these are "book" prices, and
negotiated contract prices will vary considerably.
For comparison, here are some other 100-piece prices:
Intel 286 12.5 MHz $ 84
AMD 286 16 MHz 126
Intel 386SX 16 MHz 219
Intel 386 16 MHz 299
Since these prices are about 6 months old, they are probably 5-10% high.
Michael Slater, Microprocessor Report mslater@cup.portal.com
wagner@gpu.utcs.toronto.edu (Michael Wagner) (01/24/89)
I think the funny results from Norton SI are a problem with the utility and not the 386SX chip itself. I've experienced similar anomolies on my Zenith SupersPORT (which is a 286 chip). Some days it's got a speed of 11.7; other days, 13.4. I can't seem to corelate it with anything (although I'm open to suggestions). Michael