Posted 03/28/2008 at 12:28:08pm

Not quite... I assume you're alluding to FSB saturation and/or bottleneck due to the "slow" speeds of DDR2? Which is complete garbage, DDR2 1066 is every bit as fast as DDR3 1333, proof -> http://www.xbitlabs.com/articles/memory/display/ddr3_6.html#sect0. Hypothetical Intel build: Q6600 for $190 @ Frys B&M(was actually $180 for a brief stint online @ frys) Patriot Viper 2x2GB DDR2 1066 kit for $98 after $20MIR at newegg DFI DK P35-T2RS for $140 @ newegg A comparable AMD build would at least cost the same if not more. Not only that, but getting a Q6600 to 3.6 ghz with a TRUE(thermalright ultra 120 Extreme) should be a breeze. Performance/$ and even more so Performance/watt(X4 vs C2Q) are both in Intels favor... Back in the K9/netburst days I was AMD all the way. Hypothetically, if VIA's upcoming Isaiah and S3's 430GT offered superior performance I'd be all over that no doubt(actually if they offer superior Perf/watt I still might buy em). Undying fanboyism = inferior hardware.

Posted 03/27/2008 at 05:49:26pm

I read your post... and now my head hurts :P. AMD roadmaps have shown that DDR3 support will only be available for both 45nm CPUs and AM3 platforms. The lack of DDR3 support on AM2+ platforms is no ones fault, other than AMD. Thus their inability to compete with DDR3 platforms is moot, though I think you would find little difference if say a DFI LT X48-T2R(DDR2) were used in place of the 790i board. Perhaps the performance ∆ would actually grow larger in that scenario?

Posted 03/27/2008 at 12:31:04pm

"For the most part, the Q9300 owned the night. It generally outpaced the old Q6600 quad core and the new Phenom quad core in most of our tests. In a clock-for-clock test, we would declare the Q9300 the winner with the Q6600 as the runner up." How did you come to this conclusion??? Q6600 -> 2400mhz = 266FSB(1066QDR) x 9 Q9300 -> 2500mhz = 333FSB(1333QDR) x 7.5 The Q9300 has both a higher operating frequency and FSB. It would have been interesting if you lowered the multi on the q6600 to x8 and increased fsb to 312-313. That would have provided a more accurate "clock vs clock" comparison.