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: jcholewa who wrote (21688)	12/6/2000 4:04:08 PM
From: dale_laroy	Read Replies (1) \| Respond to of 275872

Keeping the fp pipelines of the x87 flowing seems to take a minimum of one FXCH/FLD/FSTORE for every two calculations. While the Athlon is fully capable of executing this many FXCH/FLD/FSTORE operations without any performance hit due to its free FXCH and separate FLD/FSTORE pipe, decode becomes a real problem. Thus the peak sustainable x87 fp execution rate is four calculations every three cycles, versus less than four calculations every four cycles for P4 using scalar SSE2. The Athlon can exceed the per cycle fp performance of P4 by more than 33%, but nowhere near 2x.