In message <[log in to unmask]>, Mark Ranft
<[log in to unmask]> writes
>Yes, but the Whetstone and Dhrystone benchmarks are industry standard.  I am
>pretty sure that is the point of Craig's request.
>


Here's some results to get the graph started. Interesting that the 947
came out worse on the whetstone test than the 917, I would have expected
them to be virtually identical. Maybe extra memory in the 917 or the
older version of MPE?


HPCPUNAME = SERIES 917LX                     MPE 6.0

Dhrystone(1.1) time for 10000000 passes = 151
This machine benchmarks at 66225 dhrystones/second

Loops: 100, Iterations: 50, Duration: 907 sec.
C Converted Double Precision Whetstones: 551.3 KIPS


HPCPUNAME = SERIES 947LX                     MPE 6.5

Dhrystone(1.1) time for 10000000 passes = 151
This machine benchmarks at 66225 dhrystones/second

Loops: 100, Iterations: 50, Duration: 921 sec.
C Converted Double Precision Whetstones: 542.9 KIPS


HPCPUNAME = SERIES 918LX                     MPE 6.5

Dhrystone(1.1) time for 10000000 passes = 137
This machine benchmarks at 72992 dhrystones/second

Loops: 100, Iterations: 50, Duration: 12 sec.
C Converted Double Precision Whetstones: 41.7 MIPS



HPCPUNAME = SERIES e3000/A400-100-11         MPE 7.0

Dhrystone(1.1) time for 10000000 passes = 45
This machine benchmarks at 222222 dhrystones/second

Loops: 100, Iterations: 50, Duration: 9 sec.
C Converted Double Precision Whetstones: 55.6 MIPS



--
Alan Yeo
[log in to unmask]    Just because you're paranoid
Phone +44 1684 291710   it doesn't mean someone isn't!.
Fax   +44 1684 291712

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *