Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uunet!crdgw1!crdos1!davidsen
From: davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr)
Newsgroups: comp.sys.ibm.pc
Subject: Re: 286 vs 386SX vs 386 again
Keywords: 80386sx DOS
Message-ID: <1460@crdos1.crd.ge.COM>
Date: 25 Oct 89 16:10:58 GMT
References: <904@gumby.cc.wmich.edu>
Reply-To: davidsen@crdos1.UUCP (bill davidsen)
Organization: GE Corp R&D Center
Lines: 20


  The 286 executes a few instructions in one less clock cycle than the
386. It is technically "faster" but I think you would be hard pressed to
even measure it, much less notice. I believe that some benchmarks which
heavily use these instructions may run 1-2% soler on a 386.

  Programs which use the 386 instructions will run MUCH faster than the
same program using the 286 instructions. I measured 4:1 faster for some
benchmarks I just posted, and I see 2:1 on production programs. I
believe Phil Katz mentioned ZIP runs 40% faster on a 386.

  The bottom line is that you might lose up to 2% on some programs with
a 386, and gain up to 200% (real programs, not benchmarks). You will be
able to run new software as it comes out.

-- 
bill davidsen	(davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen)
"The world is filled with fools. They blindly follow their so-called
'reason' in the face of the church and common sense. Any fool can see
that the world is flat!" - anon