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: pgerassi who wrote (41394)	5/29/2001 8:44:21 AM
From: that_crazy_doug	Read Replies (1) \| Respond to of 275872

<< Even by Anand's benchmarks in DB serving, a single Tbird beats 2 way P3, thus, a dual 1.2 Tbird beats a 4 way P3. Therefore, companies can buy dual Athlon servers that outperform 4 way P3 servers and with 1.4 Tbirds on the way, even 4 way P4 servers. The bottleneck for the 2 and 4 way P3s and P4s is that they share the bus bandwidth where 2 way Athlons do not. Besides, most DBs are sped up by Beowulf type clusters in most over SMP boxes due to these being IO bound mostly. Again, just look at Anand's server cluster. >> How fast were the p3s? I'd have a hard time believing that a tbird 1.2 would beat dual 1 ghz p3s in anything that really made use of both processors.

To: pgerassi who wrote (41394)	5/29/2001 12:53:46 PM
From: Ali Chen	Read Replies (2) \| Respond to of 275872

Pete, "a single Tbird beats 2 way P3, thus, a dual 1.2 Tbird beats a 4 way P3." This logic contains a serious leap. We don't know much about interprocessor communication, context switching, and memory sharing overhead for dual AMD systems to jump into such conclusion. It is very possible that the long delay in dual-Athlon systems was related not to basic hardware functionality, but rather to some vague 2-P performance bottlenecks. "The bottleneck for the 2 and 4 way P3s and P4s is that they share the bus bandwidth where 2 way Athlons do not." This statement also may be not true. To my recollection, the bus bandwidth was never fully utilized on any application, so the actual bottleneck remains to be the memory bus. Which is still the same for single or dual systems, and latency limited. "AMD will take the dual workstation market away from Intel since they were already taking market share with single CPUs already." I find it hard to believe. Being lately at several local shops, the typical shelf spectrum is: Intel P4 1.7GHz Intel P4 1.3GHz AMD Athlon 1.2GHz Intel P-III 1.0GHz Intel P-III 866MHz Intel P-III 800MHz Intel Celeron 700MHz This is for HP and Compaq machines, e-Machines are Intel-only. Looks like a total disaster for me. Sorry. - Ali