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: Tenchusatsu who wrote (61017)	6/8/1999 7:10:00 PM
From: fyo	Read Replies (2) \| Respond to of 1573807

Tench: With AMD banking their hopes on a killer K7 FPU, I wonder how useful 3DNow! would be on K7. That's a very good question. However, the K7 has zero overhead when switching to (and from) 3DNow!, as opposed to the K6-2 which had a substantial overhead (can't remember the exact latency, sorry). The latencies of several of the operations have been significantly improved. A matrix-x-vector transform loop takes about 20 cycles on the K6-2 (3DNow!), 18 on the PIII (SSE) and 14 on the K7 (3DNow!). Note that this is according to AMD, so take it with a grain of salt if you will. Anyway, my point is that there seem to be some 'tweaks' of 3DNow! in the K7. Supposedly, there is also an extra instruction or two... --fyodor