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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: wanna_bmw who wrote (153863)	1/5/2002 9:40:24 AM
From: Charles Gryba	Read Replies (1) \| Respond to of 186894

wbmw, very interesting. Not to burst your bubble or anything but won't one prestonia cost as much as 2 Athlon XP 1800+? C p.s. The MFlops increase is tremendous.

To: wanna_bmw who wrote (153863)	1/5/2002 9:46:35 AM
From: Charles Gryba	Read Replies (1) \| Respond to of 186894

wbmw, I just read the whole article. The guys seems to know nothing about multi-threading. He was suggesting that the sandra guys may have control over the scheduling of threads to a specific physical or logical CPU. Basically, when you create a thread you ask the OS to give you one and the OS decides how to optimally assign that thread to a given CPU. Now, I don't know if Intel has provided MSFT with any additional APIS to add to the Windows kernel to provide finer control but who would want to code like that? The scheduling of threads should be left to the OS. C

To: wanna_bmw who wrote (153863)	1/5/2002 11:40:26 AM
From: Dan3	Read Replies (2) \| Respond to of 186894

Re: Prestonia - Complete with HYPERTHREADING benchmarks Those benchmarks are pretty ugly, at least now we know why they've delayed activating SMT. Re: With only a single 1.8GHz Prestonia CPU enabled Go back an re-read that post, what he did was limit the number of threads for the two processors - to reduce contention conflicts caused by SMT. The Athlon at 1.5GHZ (1800+) is faster than 1.8GHZ P4 Xeon (prestonia) with hyperthreading - at least in the test you posted. As I said, at least now we know why they've delayed letting anyone see the fabled SMT. Maybe they can get more benefits by optimizing compilers, but the numbers you posted, if real, are ugly for Prestonia and not encouraging for Intel's SMT.