Xref: utzoo comp.sys.ibm.pc:10191 comp.lang.c:6040
Path: utzoo!mnetor!uunet!husc6!mit-eddie!nathan
From: nathan@eddie.MIT.EDU (Nathan Glasser)
Newsgroups: comp.sys.ibm.pc,comp.lang.c
Subject: Re: MSC5.0 and ATI Wonder Card
Message-ID: <7812@eddie.MIT.EDU>
Date: 6 Jan 88 21:54:56 GMT
References: <700@brandx.rutgers.edu>
Reply-To: nathan@eddie.MIT.EDU (Nathan Glasser)
Distribution: na
Organization: MIT, EE/CS Computer Facilities, Cambridge, MA
Lines: 17
Keywords: Conflict between MSC5.0 and ATI Wonder Card?

In article <700@brandx.rutgers.edu> kgordon@klinzhai.RUTGERS.EDU (KGORDON) writes:
>I have a C program...
>This program runs correctly most of the time, but occasionally gives
>a runtime Math Floating Point Error.  By simply changing modes (i.e. monochr-
>ome to EGA) the program will once again (without recompiling) run properly.
>Has anyone out there experienced anythine similar?

In either the readme.doc file or elsewhere in the MSC5.0 material there
is a mention of a known Dos bug for some particular version (3.2?), and
it includes a patch to Dos for this bug. The bug apparently has something
to do with floating point errors. Check this out...it might solve your
problem.
-- 
				Nathan Glasser
fnord				nathan@{mit-eddie.uucp, xx.lcs.mit.edu}
"A tribble is the only love that money can buy."	    
Presently there is insufficient evidence to conclude that tribbles spread AIDS.