What this really says to me though is that Intel's CSI is not up to snuff and/or ready for production. Which really begs the question, why not just use Hypertransport as opposed to developing (re-inventing?!) a chip interconnect? Oh, that's right, Intel can't use the same interconnect (HT) as AMD, because when Intel begins using an IMC, the only difference between Intel and AMD will be marketing and product taglines.
The FB-DIMM's are still killing me, I don't care how much proc cache or how high they pump the fsb. I still say an enthusiast's platform should use non-ECC DIMMs. At least they got right by using procs with an unlocked multiplier. Going with desktop Yorkfields die'd to Skt 771 was an interesting move as well, I wonder if they'll market them as "The V8's" and in pairs like AMD does with the FX's.
IMO, the more I read about the V8 project, the more I just keep thinking that it's a knee jerk reaction to a compelling platform (QuadFX) idea and to "compete" in a highly niche market. Overall though, I have to agree with Anand when the articles states, "They (V8 and QuadFX) may be perfect for workstation users who also want to game on their machines, but other than that specific market we don't see much of a need for even the most demanding enthusiasts. Stressing four cores is tough enough most of the time, making the argument for eight at the sacrifice of price, power and noise is a difficult one."