Apple Inc. Message Board

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 : Apple Inc.

AAPL 255.48

-1.1%

Jan 16 3:59 PM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: Alomex who wrote (31886)	1/17/2002 1:39:58 AM
From: Dave	Read Replies (1) of 213182

I take it from your response that you cannot think of a single common task that "(1) is slow enough that you have to wait for it, and (2) is processor-bound (not e.g. bandwidth-bound), and (3) cannot be accelerated by AltiVec. That's surprising. I would have thought there would be one or two. I can't think of any either though, so I guess we're in agreement. By the way, there are some other classes of problems that are not normally vector intensive, but can be reformulated to be accelerated to use vectors. For example (from Apple's Velocity Engine Executive Summary at developer.apple.com Audio: Floating-point Inverse FFT: 3.6x Video: Floating-point Forward DCT: 3.6x Video: 16-bit Integer Forward DCT: 11.4x Graphics: Floating-point Matrix Cross-Multiply: 6.2x Small Kernels: 8-bit Integer Sorting: 14x Linear Systems: Floating-point gaussian elimination: 1.43x These accelerations are comparisons of AltiVec speeds vs. native G4 speeds on the same chip. Dave

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next