Replies to post #4814 on AMD strictly moderated (AMD)

[chipdesigner]

We're still waiting for your admission. Let's stick to the subject at hand, shall we? That is: The performance increase of the K8 relative to the K7. Most of it came from the IMC, as AMD says. You were WRONG to suggest otherwise. Show a little character, and admit it.

[mas]

You do realize that if you're right, it would mean that a Core 2 Duo with IMC (what we already know is going into Nahalem) is going to increase Intel's performance lead by your absurdly high estimates (and that's not including any other enhancements most likely going into the core).

Could be, however it has to be a conventional DDR controller and not an FB-Dimm controller which has no real benefit in latency. Also the already large cache and prefetching will tend to dampen the effect seen with K7-K8 but in some memory latency benches 10-20% could still be on the cards.