Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Rink who wrote (219659)	12/8/2006 1:44:29 PM
From: Elmer Phud	Respond to of 275872

Rink I'd actually be more interested in hearing your remark about my main conclusion. Do you agree with it? Your conclusion makes sense, hypothetically. Intel could have poor yield like TWY believes, or Intel could have idle capacity like I offered as a possible alternative explanation. Either one fits the facts. You decide for yourself which one is more likely.

To: Rink who wrote (219659)	12/8/2006 3:58:09 PM
From: pgerassi	Read Replies (2) \| Respond to of 275872

Dear Rink: If Intel was idling fabs, we would see "low utilization" charges in their earnings reports. Since there were none all the way through Q3/06, they were not idling any fabs to any major extent through Q3/06. Thus revenue per wafer for C2Ds must be either lower than that for either P4 or CD (Yonah). That is good die per wafer * ASP(good die) = $ per wafer. Given the higher ASP(good die) for C2D than either P4 or CD, the only way the revenue per wafer can be lower is that good die per wafer is lower for C2D than either of the others by a larger ratio than ASPs. So ASPs must make up for the lower numbers of good die per wafer. Thus C2D must go after the high margin high performance end in notebooks, desktops and servers. This is exactly opposite of how AMD is doing 65nm K8. It is putting them out in the high volume sweet spot. This can only be done when revenue per wafer is higher with 65nm K8 than 90nm K8. The older generation needs the higher margin to maintain higher revenue per wafer. So with ASP(good die) about the same or slightly lower, good die per wafer must be relatively higher, likely much higher. Given the lead times, server will be the last place for 90nm K8. And it will switch when Fab 30 is fully converted to 65nm/45nm 300mm Fab 38. The above shows why AMD will quickly switch to 65nm production in Fab 36. The above fully fits the public facts as seen by us for Intel's and AMD's actions. So it is more likely to be true than any other explanation. Pete