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: combjelly who wrote (243071)	10/30/2007 11:54:11 PM
From: wbmw	Read Replies (1) \| Respond to of 275872

Re: Have you noticed that AMD went with a thicker gate oxide than Intel at 65nm? As I posted before, I think AMD emphasized low power at the expense of performance at 65nm. A relatively thick gate oxide would seem to track with that. That would be a legitimate argument if AMD kept their performance constant. Typically, you would expect a new process to offer either better performance at equal power, or lower power at equal performance. And if it were to come with lower performance, you would at least expect vastly lower power. But is 65nm really vastly lower power, or much higher performance at the same power level? As far as I can see, no. At 90nm, AMD could manufacture 2.0GHz parts @ 35W, and with 65nm, they can handle 2.3GHz @ 45W. At 90nm, AMD could manufacture 2.6GHz parts @ 65W, and with 65nm, they can handle 2.7GHz @ 65W. 90nm also has 89W nad 125W power levels, but there are no 65nm equivalents today, unless you go up to quad core with Barcelona. Overall, it appears to me that 65nm comes with marginal power benefits, but hardly as much as someone might expect from a full process shrink.

To: combjelly who wrote (243071)	10/31/2007 2:01:36 AM
From: graphicsguru	Read Replies (2) \| Respond to of 275872

Have you noticed that AMD went with a thicker gate oxide than Intel at 65nm? I saw Savantu claim this. Are there solid links indicating that it ended up thicker than Intel's and also thicker than originally planned? I'd be interested to see the actual #s, and also if anyone knows the usual relation between gate oxide thickness and transistor switching speed. TIA if you have this.