Jim, <With large interests in ZDnet, Cnet and a huge advertising budget effecting other publications I'm afraid that AMD is SOL when it comes to getting a fair shake on comparative Athlon/Willy benchmarks.>
Actually, I agree with Keith Diefendorff of MDR when he says that "Benchmarks Are Bunk." But I don't (necessarily) agree that it's due to some grand conspiracy by Intel and the other Powers That Be.
The problem is that the performance picture has become so complicated that it's impossible to judge the strength of a processor on a few benchmarks. Computers are being put to very diverse uses, from 3D graphics development, to games, software compiling, database serving, web serving, multimedia, etc. Obviously one benchmark isn't going to be representative of all these uses. Nor is the processor the sole factor in all areas of performance.
Also, my personal pet peeve is the way people tend to exaggerate very small differences in performance. Enthusiast web sites like Tom's Hardware Guide makes a big deal over some comparison that shows a 5 frame-per-second difference between two processors or two platforms. When was the last time you were able to tell the difference between 60 and 65 FPS in a 3D animation?
So like you, I agree that the benchmark situation isn't going to get any better. Unlike you, I don't think it's an Intel-led conspiracy to muddy the waters. And as a rational engineer, I would hope it isn't, since basing future success on voodoo marketing is a real bad formula for business.
Tenchusatsu |
