To: Elmer Phud who wrote (154977 ) 4/6/2005 2:08:27 AM From: Joe NYC Read Replies (1) | Respond to of 275872 Elmer,This isn't an either or world. The theory has not been disproven, it just leaves open the exact definition of "bad" If you say they are at capacity (due to bad yields), but the silicon output doubles, 1.33 (units) * 1.5 (die size) = 2, it is pretty clear they were not at capacity, wouldn't you say?I'd also like to see you back up your claim of average die up 50%. AMD went from thoroughbred (~80 @130nm) to a mix of throughbred equivalent (~80), Barton (~100), Newcastle (~150) Clawhammer (~200), Sledgehammer (~200). While I don't know th breakdown, the average die size has to be somwhere between 100 and 150.The only way for AMD to show they have good yields is to ship the expected volumes for multiple quarters to show they aren't just selling off inventory. I think we have had 2 quarters with 8M+ units, and this will likely be 3rd one. You can't go one like this for several quarters in a row unless your capacity is north of 8M. I think AMD can still go up from here, 25% or more. Now, we can go back and forth, but let me present to you an argument of someone much more in the know, and that is Intel Corporation itself. If Intel believed AMD was capacity limited, it would be behaving entirely differntly than what we have seen. What would be the point of all the mobilization of sales forces, what would be the point of strong arm tactics, no point of cheating on benchmarks, if all your competitor could produce was what they were producing, and everybody else wanting a PC would have to come to you? All these actions of Intel are there to reduce AMD's demand. All the actions would be unnecessary if AMD were production limited. Joe