Something wrong with this test

[deweycd]

I have just started using seti@home and looked at my computer tested compared to several others. It seems I have a fast comptuer.

AuthenticAMD
AMD Athlon(tm) 64 Processor 3000+
Measured floating point speed 2020.52 million ops/sec
Measured integer speed 3770.86 million ops/sec

Compared to:
AuthenticAMD
AMD Athlon(tm) 64 Processor 3700+
Measured floating point speed 2038.22 million ops/sec
Measured integer speed 3782.67 million ops/sec

OR:
AuthenticAMD
AMD Athlon(tm) 64 X2 Dual Core Processor 3800+
Measured floating point speed 2243.19 million ops/sec
Measured integer speed 4160.01 million ops/sec

CORRESPONDING SITES
http://setiathome.berkeley.edu/show_host_detail.php?hostid=2185230
http://setiathome.berkeley.edu/show_host_detail.php?hostid=2152537
http://setiathome.berkeley.edu/show_host_detail.php?hostid=2052615

Anyone have any ideas why mine would be so high for a computer oced by 10%

Any ideas would be great
Thanks

 

[Scout]

Well sure, by clocking your chip to 2.2, that's the regular speed of a 3700+ isn't it? So you should be able to match the speed of that faster chip.

The X2 3800 runs at 2.0 GHz. so your single core is faster than either one of the 3800's dual cores. It doesn't appear that the benchmark takes into account the fact you probably can run two instances of SETI on that chip.

 

[deweycd]

But over clocking a 3000+ shouldn't give me the same results as a base 3700+ though should it? I would have to estimate that I should be getting 3300+ style scores.

 

[Scout]

AMD's model numbers are not linear like that. They give "extra credit" in their model number scheme for things like extra cache and so forth which may make no difference in a particular benchmark.

Your overclock puts you at the same 2.2 GHz. core speed as the 3700+ so you're going to see similar performance in many benchmarks, particularly if the benchmark doesn't take advantage of the large 1meg cache on the 3700+

 

[deweycd]

Cool, thanks, I didn't realize that.