Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!cs.uoregon.edu!ns.uoregon.edu!oregon!ngkong
From: ngkong@oregon.uoregon.edu
Newsgroups: comp.windows.ms
Subject: Hypothesis_Math Co-processor Speeds Up Win 3.0 !?
Message-ID: <20733.2809988e@oregon.uoregon.edu>
Date: 15 Apr 91 19:11:58 GMT
Distribution: usa
Organization: University of Oregon
Lines: 16

Hi,
	I noticed that one of the IBM 55SX runs window much faster, than the
other.  Both machines have the exact configuration, except the faster ones
has a co-processor installed.  I don't mean running programs that definitely
benefits from co-processor (eg, Excel), but the general window routines
(re-drawing screan, open/close files, etc).  I also know that in theory, 
Win 3 doesn't support a co-processor.  But on 55SX, which is a fairly slow 
machines, the difference in speed is fairly significant.
	The hypothesis we come up with is that the c-compiled code used in
Win 3.0 indirectly benefits from the presence of the co-processor, as it would
skip some instructions related to soft-emulation of co-processor.  Is there
any documented evidence that anyone knows?
	Comments/experience/discussion welcomes.


Casey