the truth about the performance of the 2.8Ghz P4

It appears that the "rumoured" prefetch bug that was floating around on the net, the one about a bug affecting all p4's, and p4 xeons, seems to have been fixed on the 2.8 p4. It appears that eradication of this bug has resulted in the mysterious, but slight, performance improvement in the 2.8 p4. The 2.8 got the usual performance increase from the increased clock rate, but also on top of that, got an extra few percent from the prefetch bug being fixed. Also, there's a possibility that the bug fix accounts for the fact that the 2.8 is OC'ing slightly better than alot of people expected. If you look at <A HREF="http://www.geek.com/news/geeknews/2002Aug/bch20020830016143.htm" target="_new"> this link</A>, then you can see that OCed to 3.917, the 2.8 P4 is still stable, according to the OC'ers, and going from 2.8 to 3.9 is about a 40% overclock, which is IMO, impressive (even though they are using LN2), considering that NW's life is ending soon (after Prescott takes over).

Here is the link for the prefetch bug:

<A HREF="http://www.geek.com/news/geeknews/2002Aug/bch20020830016138.htm" target="_new">http://www.geek.com/news/geeknews/2002Aug/bch20020830016138.htm</A>



- - -
All good things must come to an end … so they can be replaced by better things!