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: Scumbria who wrote (74956)	3/1/1999 1:00:00 PM
From: Tony Viola	Read Replies (2) \| Respond to of 186894

S, >>>Nobody knows how to write a compiler which can take advantage of IA-64's theoretical processing power.<<< Well then what are IBM, SGI and all the rest screwing around with it for? TV

To: Scumbria who wrote (74956)	3/1/1999 1:14:00 PM
From: L. Adam Latham	Read Replies (1) \| Respond to of 186894

Scumbria: Re: It is much simpler to write software which partitions the compute tasks at the granularity of the CPU. Attempting to multi-thread a single instruction stream for the APIC architecture, is a ridiculously complex problem. I disagree. I think you're underestimating the advances that have been made in compiler technology. In addition, there are libraries available now (some for free) that have highly tuned, multithreaded routines that just have to be linked into your application. Writing "software which partitions the compute tasks at the granularity of the CPU" is still quite tedious, and is certainly no simpler than writing multithreaded code, IMO. Adam

To: Scumbria who wrote (74956)	3/1/1999 2:39:00 PM
From: Tenchusatsu	Read Replies (1) \| Respond to of 186894

<Nobody knows how to write a compiler which can take advantage of IA-64's theoretical processing power.> You sure? I believe there is a lot of communication between compiler writers and Merced-land. <It is much simpler to write software which partitions the compute tasks at the granularity of the CPU. Attempting to multi-thread a single instruction stream for the APIC architecture, is a ridiculously complex problem.> Just as ridiculously complex as trying to squeeze out more ILP using a traditional out-of-order RISC-like architecture? Remember, it's always better to have one processor working twice as fast than to have two processors working at normal speed. Tenchusatsu