Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!gem.mps.ohio-state.edu!apple!bloom-beacon!eru!luth!sunic!draken!d88-eli
From: d88-eli@nada.kth.se (Erik Liljencrantz)
Newsgroups: comp.sys.ibm.pc
Subject: Re: 286,386sx,386?
Summary: 386SX is way beyond 286
Message-ID: <1914@draken.nada.kth.se>
Date: 8 Oct 89 21:24:33 GMT
References: <110200011@uxe.cso.uiuc.edu> <1989Oct6.202428.3240@uncecs.edu>
Reply-To: d88-eli@nada.kth.se (Erik Liljencrantz)
Organization: Royal Institute of Technology, Stockholm, Sweden
Lines: 27

In article <1989Oct6.202428.3240@uncecs.edu> mvolo@uncecs.edu (Michael R. Volow) writes:
>One consideration which I have not seen mentioned, is that the cost
>of a medium speed 80286 machine (e.g. 16Hz) plus the cost of an
>add on EMS card with memory (usually $400-$550) roughly equals the
>cost of a 386SX, where EMS can be defined and managed by software.
>An 80286 with only motherboard memory and with a chipset that includes
>EMS firmware (e.g., NEAT?) should be cheaper theoretically. However,
>does the EMS firmware in the newer 80286 chipsets work and do all the
>LIM 4.0 tricks?

Yes, the NEAT chipset do handle EMS 4.0. But the memory management features
found in the 386, including the 386SX, can do more than providing EMS support
through software. In my system I have 96Kb of high memory (above EGA memory,
below BIOS) where memory resident programs can reside. I use 386-MAX to
accomplish this and its easy to rearrange the memory into EMS or EXTended
memory. Just small changes in CONFIG.SYS.

As for the speed of the 386SX compared to the 286 I can only confirm that
it is a bit slower at equal MHz's, but just a little.

PC Magazine recently tested a number of 286 systems. Their conclusion was:
By at least a 386SX! Just what I did...

>M Volow, VA Medical Center, Durham, NC 27705
>mvolo@ecsvax.UUCP           919 286 0411
-- 
Erik Liljencrantz     | "No silly quotes!!"
d88-eli@nada.kth.se   |  Embraquel D. Tuta