Advanced Micro Devices Inc (AMD)

[kpf]

tecate

Theoretically, you could be right - even with on-die memory controller(s!) (because I'd think it would be essential for a state-of-the-art design): If Intel started developing the project for a product to challenge Opteron three years ago at least and has working silicon already (iaw it would be what they demonstrated at IDF without explaining what it is to protect current Xeons from Osborne-effects if they did - and Itanium as well).

OTOH, it would be advisable to speak up for Intel in order to prevent the Opteron-snowball from gaining speed in its roll. I would expect Intel to get vocal about it before New Years Eve in this respect. Probably in a way we would not be able to judge whether it is kludge or not though.

K.