albert, Re: "let me take you way way way way back when new pr rating was introduced. It was postulated that unsuspected buyer may confuse pr with frequence and assume that the performance is equal to that of p4 with the same clock and that would be misleading. Now following the same logic path you want that unsuspected customer, as Dan pointed out, to assume that athlon xp 1900+ performs on par with p4-1.6GHz."
Exactly, albert. You may not have realized it, but you are making my point.
You just have to ask yourself what AMD has been trying to do based on their actions - come up with a reliable way of educating the consumer on performance, or just assuming that the consumer doesn't know any better, and copy Intel's game of feeding off that ignorance.
A solution based on the former wouldn't need to compare against anything that Intel currently has. It would stand as a testament of its own merits, because it's correct, and the consumer can correlate the performance that they need. The latter presumes that the consumer can't tell the difference between performance and a big, arbitrary number. Since performance can sometimes seem smaller than a big, arbitrary number, stating the true performance can sometimes produce a negative effect. Therefore, presuming that consumers cannot adapt to a True Performance Rating, and Big Arbitrary Numbers are Good and Pleasing to the Ignorant Consumer, thus QuantiSpeed is born.
wbmw |
