Itanium 1.6Ghz 533Mhz FSB spec scores

<b>Itanium 1.6Ghz, 9MB, 400Mhz FSB: </b>
Price $4200
Spec_INT: 1590
Spec_FP: 2712

<b>Itanium 1.6Ghz, 3MB, 533Mhz FSB, DP:</b>
Price $1150
Spec_INT: 1452
Spec_FP: 2692

<b>Itanium 1.6Ghz, 3MB, 400Mhz FSB, DP:</b>
Price $850
Spec_INT: 1408
Spec_FP: 2553

What this all means: The top bin itanium costs 4x the DP Itanium with 533Mhz, but is less than 10% faster in int and within <b>1% of the same FP performance</b>. This definitely means something; now you can buy a processor capable of 2600+ scores in Spec_FP for a rather more buyable $1150. If you're after world-class floating point calculations, this wouldn't be a bad choice. So for a scientific-research workstation, this could be interesting. If only more software support were at hand.

In any case, it's pretty obvious that Madison 9M is actually more than a speed and cache bump. Even the scores for the 3MB cache, 1.6Ghz version are above 2500 for FP, while the 1.6Ghz 3MB cache previous madison core scored less than 2000.....

Maybe by the time things get to montecito, we'll be seeing a little better scenario, who knows...

Just get some 8 way Opterons.

Most of the increase come from the new compiler itself.

i need to change useur name.