EHP.. Any comments on micro Throughput

Nobody commented on the math Throughput benchmarks I did on the PC and a microcontroller.
I was thinking that someone would show me that high math throughput is not
required. I know old time CNC had 8080 CPU's and it did this stuff. I'm sure
floating point math took 100's of us or Milliseconds on those CPU's. So I guess
they used integer math and "picked" some good assumptions for their controls.


Stepper system first
They must have been creative and used integer math. I guess that 32bit integers
So here I go with a simple attempt. Say we want
.0001" step resolution so that Math round off errors will be within +/-.001"
( one order of magnitude). We set the lsb to be .0001" ( english screws)
we will then have 10000 codes per inch of machine travel. With 32 bit signed
integers we can express 2.147 billion codes or 214,000 inches of travel ( way
more than enough) We could even go finer than this if we want. Say .000010
(10 millioths of an
inch) would still give 21,400 inches of travel in a 32 bit word.

So Maybe we just forget about floating point and try integer math in 32 bit
words. Then a micro might have a chance. I'll
benchmark the 8051 at 24Mhz and see how fast it can do a 32 add
multiply and divide.


dave