Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!newstop!male!geraldo.Central.Sun.COM!texsun!digi!jgay
From: jgay@digi.lonestar.org (john gay)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: question about 486 co-processor
Message-ID: <1991Jun13.161205.884@digi.lonestar.org>
Date: 13 Jun 91 16:12:05 GMT
References: <1991Jun12.071753.2030@unixg.ubc.ca>
Distribution: na
Organization: none
Lines: 28

From article <1991Jun12.071753.2030@unixg.ubc.ca>, by wangf@unixg.ubc.ca (Frank Wang):

>      My friend bought a 486-25.  He said he did not order the 
> co-processor (something like Weitek co-processor).  I know that 486
> has the feature of built in math co-processor.  But what is 
> this co-processor for?  Why is there another co-processor appeared 
> after Intel integrated the math co with cpu?  My friend could not
> explain the function of this "Weitek" co-processor.  Can any one
> out there tell me what this co-processor is and why?

The Weitek co-processor is a memory-mapped math co-processor (unlike Intel's
interrupt driven co-processor).  The memory-mapped Weitek is supposed to
be faster than the Intel, but with the '486 I am not sure this is still
true.  The Weitek has to have compilers that will create code for it
(usually more expensive) and is itself expensive around $1000 I think.

Why - more floating point speed (at least as far as '386 version).


>      Thanks in advance.

welcome.

-- 
john gay.                      jgay@digi.lonestar.org
Like Thoreau, I rejoice that there are owls.  Part of the reason is that
owls eat mice, and thus make a certain number of cats unnecessary, which
is always a plus.  Jeff Duntemann, Dr. Dobb's June '91