Re: AMD didn't beat Intel to copper, and the reason is because there was never a race. Intel chose to delay copper to their 130nm process, and this decision has yet to be shown as a bad idea
Following their move to copper, AMD market share went from 12% to 22%-23%.
One the things you have conveniently ignored is Intel's 130nm process, which has arrived months before AMD's, and has stopped AMD's mobile plans dead in their tracks
I'd suggest you reconsider your version of reality. Try walking into the notebook PC section of a Best Buy, SAM's Club, Circuit City, or CompUSA (a bastion of Intel products) and see whose mobile plans are dead. Compare that to 6 months ago.
You have also forgotten Intel's research in 300mm wafers, which offer 2.5x more die per wafer (over 8" and 180nm), and 30% lower cost per wafer
Unless die sizes are very large (like the 217mm P4) bare die costs are low enough that aren't particularly significant cost drivers. Even here, AMD's smaller die saves it more money than Intel's larger wafers save - but the cost difference is on the order of a few dollars per die, advantage AMD - not really a big deal. |
