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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: chic_hearne who wrote (102465)	4/5/2000 11:12:00 PM
From: Jim Davis	Read Replies (1) \| Respond to of 1577883

Re: BRANCH MISS-PREDICTION PENALTY I haven't studied any Willamette data. I don't have any speculations to make. Sorry.

To: chic_hearne who wrote (102465)	4/5/2000 11:28:00 PM
From: Scumbria	Read Replies (1) \| Respond to of 1577883

chic, Branch misprediction is a favorite obsession of CPU architects. More stupid decisions have been made in the name of reducing the branch mispredict penalty, than you can possibly imagine. The Willy architects made the correct choice in using a very deep pipeline. The penalty on misprediction is somewhat larger than Athlon, but they have achieved a higher clock rate, and used very aggressive branch prediction techniques to reduce the likelihood of misprediction. Both Athlon and Willy use a trace cache to minimize the penalty as well. The JC post you referred to was fairly accurate, but I disagree with his assertion that the per clock performance of Willy will be low. I expect that Willy will be a very tough competitor. Scumbria