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: Elmer Phud who wrote (151544)	2/23/2005 2:20:12 PM
From: dougSF30	Respond to of 275872

Notice how Elmer ignored the 3.33 vs. 3.6 issue. :)

To: Elmer Phud who wrote (151544)	2/23/2005 2:26:21 PM
From: Rink	Respond to of 275872

Elmer, re: The X3 is designed to support Dual-core as well. Intel hasn't announced a dual-core Xeon yet but they hadn't announced a P4EE version of Smithfield with HT either until recently. The only thing I can see that needs to be done is validation. It might be going on as we type.... You never know... Doesn't matter. In current 4 socket Intel systems the bottleneck is the FSB. Extra cache on cpu and chipset will only help to decrease that disadvantage but not solve it by any means. To solve it would mean each Xeon would be connected directly to the northbridge chipset as follows: X X \ / NB / \ X X And somehow I think that will be too expensive: The NB will need 4 times the single FSB pin count, plus pin count for dual channel memory controller, plus the pin count for the SB, etc... Hence I keep saying it's rather unlikely that the FSB problem is solved. A 4 sockets single core cpu system will hence have the same size FSB bottleneck as a 2 sockets dual core cpu system. An 8 sockets single core cpu system will have the same size FSB bottleneck as a 4 sockets dual core cpu system. In fact now I'm thinking about it Smithfield might make sense in a 2 socket dual core cpu system when combined with IBM's X3. Hence your speculation might have some merit, you'd say. Let's think that over. If your speculation would be true (i.e. Smithfield will be released in Xeon form together with X3) then it's likely to be for IBM only. That again would not make Dell say today it didn't see the need for AMD anymore. So either Smithfield will NOT become available to IBM in Xeon form, or IBM will produce X3 directly or indirectly for Dell and HP. As so much of this speculation hasn't even been rumored about let alone hinted to by the main parties I think this is utopia. An alternative possibility supporting your speculation is that Intel releases a X3-like chipset themselves to appease Dell and HP and keep Dell from going AMD. That too is possible, unprecedented, and not even rumored about either. Who knows. Until at least any of the main parties provides at least the tiniest hint I'll call it Utopia. BTW, I think you have noted that your higher baseline point is completely consistent with all my previous remarks. Hence Potomac is likely not to be used meaningfully in combination with X3. Main target cpus for X3 are Cranford (1MB Xeon MP) and dual core 2MB Xeon MP whenever that would become available. Regards, Rink