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: DDB_WO who wrote (211068)	9/18/2006 7:42:29 AM
From: Hans de Vries	Read Replies (2) \| Respond to of 275872

Re:...In that mentioned die estimation thread I also asked Hans de Vries about blue prototype core vs. K8L core changes, because once he made a floorplan based on the photo and the published die plot: chip-architect.com His answer (I'm sure, it's him and he just forgot to login or so): That wasn't indeed me. It's the same guy who sometimes posts at RWT under the same name. I did had to move and strecth things a bit to get the "blue core" background correctly alligned with the QC layout graphics. There are also differences in macro cells. There is a lot of confusion on the Rev.G / Rev.H naming. The only information from AMD itself came from here: pc.watch.impress.co.jp babelfish.altavista.com From which I understand: 1) The "Blue 65 nm die" is not Rev.G 2) Rev.G is "almost an optical shrink" to 65 nm from Rev.F 3) The improved core is Rev.H (The Green QC with 128 bitFP) The "Blue 65 nm die" seems to be an early Rev.H prototype without the 128 FP. To do an "almost optical shrink" of Rev.F to 65nm is not at all obvious. This would give resolution problems at the finest metal layers. It probably could not be done at the end of 2005: ftp://download.intel.com/technology/silicon/Yan_Borodovsky_SPIE_2006.pdf From page 18 on you can see how to failure of 157 nm lithography promted Intel to scale the first metal layer by only 5% while a scaling of 30% is required for an optical shrink. Maybe now, with newer, high NA, lithography equipment it has become possible to do "almost optical shrinks" and the Rev.G cores will look very much like Rev.F cores. The only example of such an optical shrink may well be the Mobile Core shown in recent precentations. Rev.G core: epscontest.com Rev.H core: epscontest.com Regards, Hans