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: muzosi who wrote (239516)	8/29/2007 3:37:46 AM
From: graphicsguru	Respond to of 275872

I agree that the first attempt at parallelization often uses extensive mutexes. But the parallel scaling on code like that is usually so bad that people either give up, or do it right. I haven't seen any mutex-heavy code that performed well enough to be used in a real-world practical situation. In general, I'd be tempted to agree with you that performing well on bad code is the most important thing. But bad multithreaded code often ends up running slower than single-threaded. That forces people to understand the situation or give up. I've seen a lot of the latter . . . .