🚀 go-pugleaf

In article <1992Sep26.192710.13739@microsoft.com>, markcha@microsoft.com (Mark Chace) writes:
> In article <1992Sep25.021321.26351@julian.uwo.ca> wlsmith@valve.heart.rri.uwo.ca (Wayne Smith) writes:

>>Secondly, how would a 386-40 compare with a (286-25 with a 287-25)?
>>(with regards to Quick Basic interpreted and executibles).
> 
> The 386-40 is a Significantly faster machine than the 286-25.  Unless you
> are doing a whole lot of heavy duty floating point arithmethic, the 386
> will be MUCH faster.  Even if you are doing a lot of math, I am not convinced
> the 286+Coprocessor will be much faster. 

I recently wrote a program that runs MUCH faster on an old, 6 MHz (!) 286
with 287, than on a 386-20 with no coprocessor.  I think this is because
the program does a few trig operations.  From this and previous experiences
it seems that the coprocessor helps significantly, even if there is only 
a small fraction of floating-point code in the program. 
________________________________________________________________________
Raymond W. Arritt                     |  Dept. of Physics and Astronomy  
Assistant Professor                   |  University of Kansas 
arritt@kuhub.cc.ukans.edu             |  Lawrence, KS  66045  
Apr  1 12:29:21 timed[92]: THIS MACHINE IS A SLAVE
Apr  1 12:29:27 timed[92]: THIS MACHINE IS A CANDIDATE
Apr  1 12:29:31 timed[92]: This machine is master