Gene,
I would like to see the industry and press agree on an industry standard set of BM's that de-politicizes the whole issue.
You realize, of course, that this will never happen. Each benchmark developer thinks that their benchmark is the "best". Just as people develop new products to be "better" than whatever exists, people will continue to develop new benchmarks and update the benchmarks they already use. And you have to keep developing new tests to benchmark new features in the technology (who would've guessed 15 years ago that we'd ever need a benchmark measuring the frames-per-second performance of PCs?). A "committee" would never be able to keep up with advances in technology.
My one wish is that the benchmark publishers publish the details of their tests completely. Then, at least, you could determine whether the test applied to the real world environment in which you expect to be working. If so, use the result. If not, don't.
In the absence of complete disclosure, I had been waiting anxiously for the PC World and PC Magazine results. They are, IMO, the least biased testers of all the testers I've seen and appear to have a minimal "personal agenda" (except to sell magazines). Also, as widely read publications, their benchmarks have the power to sway the most opinions. Obviously I was pleased with both sets of results <G>.
Dave |
