This is too juicy not to preserve for posterity:
Hook on Tom:
=============
Ahhhh, a new month, a new beginning.
Okay, about the Tom's Hardware stuff. I'm a little burned out when it
comes to getting in another pissing match over the stuff, so I'll just
stick to the facts:
- irrespective of legal issues pertaining to his possession of the IHVTEST,
it is not a valid benchmark program since the code in it is radically
different than the code used in Q3TEST. So if you want to know how fast
your card runs with IHVTEST, then use his numbers. If you want to know
how fast your card will run with Q3TEST, use ours.
- the reason we do benchmarking is because A.) we know our software, and
are thus aware of what likely bottlenecks will exist and how to properly
measure the contribution of those bottlenecks; B.) we have no advertising
affiliation when we post benchmarks, and do not need advertiser revenue
(from sources that pose a significant conflict of interest when it comes
to objectivity) to sustain ourselves; C.) we are engineers, and we understand
how the basic underlying principles of the scientific method work.
We don't like it when other sites run benchmarks in a suspect manner, and
in the past Tom's numbers, methodologies, and motives have all been
suspect.
That's all I have to say on the matter, I'll let others argue the finer
points.
================
I LOVE IT!
Chip
stockcharts.com
|
