# … and we have a winner …

Way way back, long long ago, I used mdbnch benchmark to test machines. I was amazed when SGI’s R8k got this done in under 20 seconds. The sub 15 second R10/R12k results were awesome. The sub 10 second Alpha results were amazing.
For a while, Opterons and Xeons have been in the 2-3 second range. Some recent chips were in the 1.4 second range.
I always wondered when we would crack one second.

[landman@pegasus-i5482 mdbnch]\$ ./new.x
MDBNCH: A MOLECULAR DYNAMICS BENCHMARK, VERSION OF DECEMBER 17, 1988
*******************************************************************************
MD BENCHMARK FOR   256 PARTICLES, 1000 STEPS.
O(N**2) BRUTE FORCE LIST FORMATION EVERY 10 WITH SKIN = 1.00
PAIR CORRELATION FUNCTION NOT COMPUTED
...
*******************************************************************************
COMPLETE BENCHMARK EXECUTION TIME :    0.870000 CP SECONDS.


Intel 5482. Fast box. Yeah, I am impressed. This benchmark is mostly in cache these days, but I am still pretty impressed.

### 1 thought on “… and we have a winner …”

1. I got similar results on my Core 2 Duo (Intel T7500):
both second.f and mdbnch.f were compiled with ‘gfortran-4.1 -O3 -funroll-loops -march=nocona’:
“COMPLETE BENCHMARK EXECUTION TIME : 0.916056 CP SECONDS.”