Path: utzoo!attcan!uunet!cs.utexas.edu!news-server.csri.toronto.edu!utgpu!watserv1!maytag!watdragon!lion!sjorr
From: sjorr@lion.uwaterloo.ca (Stephen Orr)
Newsgroups: comp.sys.amiga
Subject: Re: Maple V news for Amiga
Message-ID: <1990Nov11.155852.3076@watdragon.waterloo.edu>
Date: 11 Nov 90 15:58:52 GMT
References: <3760@mindlink.UUCP> <15580@brahms.udel.edu> <7960@ucdavis.ucdavis.edu>
Sender: daemon@watdragon.waterloo.edu (Owner of Many System Processes)
Distribution: na
Organization: University of Waterloo
Lines: 29

In article <7960@ucdavis.ucdavis.edu> zerkle@iris.ucdavis.edu (Dan Zerkle) writes:
>No, that's not it.  The 68881 and 68882 have their own special
>instruction set.  When you have one of these math chips, you can mix
>these special instructions in with the regular 680x0 instructions in
>your machine language.  That way, the 680x0 instructions get executed
>by your CPU, and your 6888x instructions get executed by your FPU.

Remember that the software emulating math libraries, from 1.3 onward
will detect a math co-processor, and re-direct requests through them.
While this isn't as fast as inline code, it is more machine independant.
However, this doesn't solve the Maple problem, because Maple have their
own custom math libraries.

>If you don't have an FPU, you have to emulate the floating point
>instructions in software.  This works, but is not nearly as fast.  A
>certain graphics program I wrote runs about 10 times faster (no
>exaggeration) when I use the 68882.  In other words, getting math chip
>support is very important to those folks who own one and who do a
>large amount of number crunching.

And here we come to the whole point of it. It turns out that Maple, despite
being an advanced math package, does VERY LITTLE actual number crunching.
Maple deals with just about everything symbolically, and only does fp
substitutions as the last step. According to a friend of mine who is on the
Maple team here at Waterloo, the only place that a math co-pro is really
going to affect the speed is in the 3-D graphics stuff, and he's going
to talk to them about that part.

					Stephen Orr