Not sure if that's an insult, but
http://voodoopc.blogspot.com/2006/03/if-only-they-had-time-machine.html
~~Mad Mod Mike, pimpin' the world 1 rig at a time
No insult at all. You put out alot of good useful information. I am suprised, that so many of the Intel Fanboys are so uninformed on how computers actually work.
I had read Rahul Sood's blog earlier. Made alot of sense. I have put CPUs in boxes, where I didnt have the latest bios without the proper microcode fixes for newer CPUs. They almost never run to their capability, if they even run at all. So his blog shows there is a high potential to producing invalid benchmark results from the AMD side.
And using a customized ATI driver, on the AMD machine, is just wrong. So that brings into question, were they using some optimized for Intel version of windows too? That brings back memories of multiple exes for different CPU optimizations. It doesn't make a fair comparison between platforms, when one has it, and the other doesn't.
Noone can prove outright, the benchmarks were false. But there is a lot of evidence, that they cannot be taken as gospel. And a lot says that it was not a valid comparison. I just wish we had more info concerning their server comparisons. It does look like Intel did let people play with their rigs too much, because enough info was gathered to show that the testing was not on the level.
Maybe Intel does plan to make improvements to Conroe, to back up the performance gains they are showing. But I am not gonna hold my breath. And I am not gonna build another system, that I cant upgrade steadily over a 2 year period.
On the server standpoint, AMD has a big thing in their Socket F, if it is going to support even quad cores, when they come out. Companies dont like to replace servers yearly. Its nice to have options to upgrade.