Ignorant Grinch, Re: "you don't. No matter what your excuses are,
the fact remains that on a decode-limited application
the P4 running at "marketing frequency" of 2000MHz
has the same performance as P-III running at "true"
1000MHz. I just don't see how the internal details of
how many "decoders", 2 or 1.5, or else, can obscure
the simple conclusion: A decoder is a functional
unit, it is designed to deliver certain performance.
As the Khapstone benchmark shows, when decoders
are exposed to heavy decoding job,
2000MHz of p4 are equal to 1000MHz of P-III.
Therefore this points more and more to the conclusion
that majority of P4 die area is running at MarketHertz,
not MegaHertz."
Sorry, Grinch, your conclusion is not correct. In fact, it completely contradicts other known facts that you are refusing to acknowledge.
Yes, Kap's piece of code runs approx twice as slow on a Pentium 4 with large data sets.
And, yes, Kap's program happens to stress the decoder unit.
However, this does not mean that the clock that feeds the decoder units runs at half the frequency.
It's both Plain and Simple. End of discussion. Get used to it.
wbmw
P.S. I don't remember what the argument was about the TLBs, but I can see you are still unwilling to admit that you were wrong about Intel not listing their stock repurchasing on their balance sheet. I clearly show that they did include it, so once again, Grinchy, you are wrong. Live with it. |
