Re: New Math Coprocessor - who has a need for it???

[gudat@avalon.physik.unizh.ch (gudat henrik)]

In article <CA6z23.Ex6@wave.scar.utoronto.ca> 90taobri@wave.scar.utoronto.ca (Brian Tao) writes:
>In article <1993Jul13.194102.16667@ifi.unizh.ch>, gudat henrik writes...
>> 
>> The use of a math coprocessor? Get instant spreadsheets, calc mandelbrot
>> pictures in a fly (well, almost:). I think Mr. Disbrow once wrote a
>> mandelbrot program for the FPE...
>
>    I don't use spreadsheets (at least not ones large enough to justify
>getting a math copro) and I'm not going to spend $200+ so I can get
>instant Mandelbrots.  ;-)  What about toolbox routines that use math? 
>Will things like QuickDraw II, ACE (and other compression algorithms)
>and Pointless speed up?  I would buy it for those reasons.

QuickDraw and I guess ACe as well use integer math. Only floating point
stuff will be accelerated.

The NC is an interesting card for small businesses who have a lot to
calculate (I mentioned spreadshets already).

Please allow me to point out another detail I BET I have mentioned
already several times:
There are two "levels" of acceleration, namely:
1. Programs that call the SANE toolset will run faster because these
calls are processed on the card. ALL programs who depend on SANE are
affectd. You ned not recompile or rewrite them.
2. Programs that have been written for the FPE will be accelrated even
more. Problem here: ORCA/C requires you to specify the slot of card.
This is really bad. The programs will crash or do not work if th card is
in another slot. Other applications that support the FPE normally
locate th FPE/NC automatically.

Or: if a program does not support the NC directly, this doesn't man it
won't b accelerated.

Besides that:

- the NC does not give you the perfomance of a PowerPC
- the NC does not disconnect the main CPU in any way
- the NC comes with SANE patches, you need not write them yourself


Henrik {CNB}

-- 
BRIGHT SOFTWARE   *   P.O.Box 18   *   4153 Reinach 2   *   Switzerland
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
voice (0)61 2619454  gudat@avalon.physik.unizh.ch   Simple Solutions to
fax   (0)61 7115263  gudath@ezinfo.vmsmail.ethz.ch    Complex Problems.