2 sitcks of 400mhz do not equal 800 mhz. This used to be how rambus would run their numbers and they were full of it. with a configuration like that, you get 800mb/s of effective bandwidth, but your ram is still running at 400mhz. Think of it like a river. If your river is wider (mb/s of bandwidth) it doesn't have to flow as fast (mhz=speed) to carry the same amount of information. The bottleneck usually comes in when you do not have enough memory bandwidth between your processor and your memory controller. Improving one usually hides the deficiency of the other, but they are not the same.
Dual channel ram is to the memory bus like dimms are to simms. Dual channel ram just means that the information meant for two separate sticks is not competing for the same pipe.
The ideal world is to go with how the athlon 64's do it and have the memory controller integrated directly into the processor. That way, there is less latency as the processor does not have to "look up" where the info. its looking for is in the ram. the integrated hypertransport inks in the 64 bit cpu's from athlon is one of the best things they've ever done.
If I\'m curt with you, it\'s because time is a factor. I think fast. I talk fast. And, I need you guys to act fast if you wanna get out of this. So, pretty please with sugar on top, clean the fu**ing car- Winston Wolf, Pulp Fiction