*nix performance

[William E. Davidsen Jr]

In article <1428 at draak.cs.vu.nl> blom at cs.vu.nl (Blom M L) writes:

| One of my main interests is whether a fast (25MHz with cache etc) 386 with
| some users attached to it can compare itself with a VAX/750 or other minis.

  Benchmarks (both mine and others) seem to show that a 20MHz 386 is
about 3:1 faster than an 11/780. I guess that compares to a 750
somehow... the F.P. performance is more like 5:4 faster, but very few
program are as F.P. intensive as they seem.

  You will probably want to use a "smart" serial card to take the load
off the CPU when interrupts come in, and Xenix comes with device drivers
for several.

  When I measured the performance of programs compiled on Xenix386 and
IX/386, I found no cases where Xenix was slower. I found a few cases in
which the Xenix compiler got into a loop and required hand
simplification of an expression (out of hunderds of programs) and about
15 cases where the ix/386 compiler either core dumped of tried to talk
the assembler into using "register 25". Both have been upgraded since
then, so I don't know if this is still representative.
-- 
	bill davidsen		(wedu at ge-crd.arpa)
  {uunet | philabs}!steinmetz!crdos1!davidsen
"Stupidity, like virtue, is its own reward" -me