SPECmarks for RS/6000 systems - lies???

[jim frost]

CCHD at lure.latrobe.edu.au (Huw Davies - La Trobe University Computer Centre) writes:
>I have just got a copy of the September 1990 SPECwatch and I am
>a bit concerned about the following paragraph:

I was wondering when that would come out.  Given the stability of the
system when IBM announced the SPECmarks, and given that the
performance today is still not too good, I doubted the results.  On
pure benchmarks my system hits what IBM says it should right on the
nose or close enough not to matter.  But when it comes to running
actual applications the thing is much slower than expected, usually
slower feeling than a SparcStation and in compiles it's often slower
than our microVAX.

I'd like to know what kind of tuning they did to get their numbers so
I could do it to my machine!

jim frost
saber software
jimf at saber.com