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: Gopher Broke who wrote (28534)	2/15/2001 12:25:25 PM
From: Joe NYC	Read Replies (1) \| Respond to of 275872

GB, Intel doesn't have to produce anything LDT to participate in this market. If, say Via, produces a northbridge to connect to Intel processor on one end and LDT on the onther, Intel can participate. THat brings up another question about the incorporation of on-chip memory controller, and if AMD is really going to do it, and if it will force Intel to follow. On chip memory controller will provide excellent latency, and overall speed boost, but it also eliminates from the picture a Mamba-like chipset with embedded DRAM L3, which can provide similar boost. Joe

To: Gopher Broke who wrote (28534)	2/15/2001 4:55:03 PM
From: Charles R	Respond to of 275872

<Interesting challenge! I notice you were not confident enough to leave out the multiprocessor factor, where Intel obviuously has 100% of the market currently. AMD would have to make huge inroads into the corporate market for this to come about.> LDT is a good move by AMD but I was pointing to the market realities. It is nothing do about my confidence and very much to do with the "100% of the market" that you mention. <So, assuming we put some time limit on this bet (2 years to allow Sledgehammer to get into volume production?) then it would still be unlikely that AMD could overtake Intel in volume of multiprocessor systems shipped.> Precisely. <However, if LDT does take off then Intel will be forced to support LDT it as well as SCP, otherwise they can't support all the LDT-capable devices people will be producing. The question then becomes how long it takes Intel to ship systems that implement LDT. Even if LDT is a runaway success it would probably take the Intel dinosaur a year to stop swimming against the tide.> LDT has many uses and multiprocessing happens to be a big one. I do not see any chance that Intel will LDT for multiprocessing. Intel will get most of the chipset work done internally and every chipset vendor will be begging Intel for rights to the bus (to the extent Intel enables third parties). <Therefore I will bet you $100 that in two years time there are higher volumes of x86-based multiprocessor systems shipping that support LDT than do not. > I accept this. Let's touch base and resolve it on 15 Feb 2003 ;-)