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: Daniel Schuh who wrote (98865)	3/22/2000 8:35:00 AM
From: Hans de Vries	Read Replies (1) \| Respond to of 1571483

Daniel: Shifts are relatively much more expensive in Willamette family processors, and multiplies are somewhat more expensive. My own reading of that little tidbit is that slick Willy doesn't have a barrel shifter, and that you're perhaps being quite optimistic about the cost of shift instructions there. I'd guess it's really a cycle per bit shifted, i.e. a<<3 is three cycles at least. Just a guess, though. Thank you for the information Daniel. This is remarkable indeed. It might be that the 3 GHz ALU is restricted to a few functions while others are performed by "neighbour hardware" like the MMX units which would cause extra arbitration and transport delay. I wonder then if there is any feedback from ALU stages after 0.5 and 1.0 cycles at all.... Regards Hans.