State of the art

[Chuck Guzis]

 On 9/23/2005,  Paul Koning  wrote:

>20 GFlops with a 3 GHz processor?  How can that be -- that would
>require 3 FP functional units per core, each issuing one instruction
>per clock cycle.  Did you mean 2 GFlops?

Could be smoke-and-mirror type of benchmarking, but here's the quote:

"However, said University of Virginia assistant professor David Luebke, a 3-GHz, dual-core Pentium 4 Extreme Edition's arithmetic units will theoretically run as much as 24.6 Gflops."

Admittedly, there are some weasel words there like "theoretically" and "as much as", but it might also be that University of Virginia assistant professor David Luebke moved a decimal point somewhere.

Cheers,
Chuck