Re: Your chart forgets that the chipset covers two sockets and puts one FSB to each socket. There aren't enough pins for Intel to put two FSBs into one Xeon socket.
And your model confuses the issue further by assuming uniform memory access in an Opteron system, which also isn't the case. If you want to assume that Opteron has more memory bandwidth because it can access remote memory, then you have to take remote latencies into account, which then retard performance back to a low latency chipset connection.
Re: TDPs set to same standard
Where'd the extra 40W come from? If you're going to claim that QC Opteron will be sufficiently below 140W while running real world apps, then you should be subtracting from the AMD TDP, not adding to the Intel TDP. And before you starting running away with your imagination, you better wait and see how power measurements correlate with the NGA TDP, before assuming it will be just like Netburst.
Market Data 
Markets 
AI
