SI
SI
discoversearch

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.
Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?


To: Tenchusatsu who wrote (93578)2/16/2000 1:20:00 PM
From: kash johal  Read Replies (1) | Respond to of 1572391
 
Tench,

re:"But I can tell you that the Willamette guys are confident they'll smoke Athlon's floating-point performance."

Really!!!

Is that once software is optimized for the new SIMD instructions using a special compiler.

Or just using a std. compiler without specific optimization.

If its the latter I am IMPRESSED.

If the former then its a CON.

regards,

Kash.



To: Tenchusatsu who wrote (93578)2/16/2000 2:36:00 PM
From: Daniel Schuh  Read Replies (3) | Respond to of 1572391
 
The thing that still puzzles me most about the (limited) Willy info is the Rambus angle. First of all on the Playstation angle, yes, the Playstation II? III? will use Rambus. 32 meg. 1 chip, maybe 2. A one chip memory may have enough advantages in terms of packaging there to be worth the cost. In the PC market?

Then there's the Willy WILL use Rambus angle (or for real cheapskates, a really crappy, low perf SDRAM interface), versus Foster will using SDRAM/DDR/whatever, because "that's what the customers want". Again, what has this got to do with the CPU? Unless Willy's also got an integrated north bridge equivalent, or at least an integrated memory controller instead of the so-called FSB interface. It would make sense for Timna to use Rambus, except for the expense, but why Willy?

Aside from cumine production, which will no doubt get ironed out sooner or later, I'd say Intel's biggest problem now is they don't have a decent cumine chipset, or anything that people are willing to give up the BX for. And the message I get from the Willy/Rambus angle is that it's going to stay that way through next year. This just doesn't make sense. What customers wants to pay 5x as much for memory, with marginal at best performance advantage? Who would even want to pay 2x as much? Again, in special circumstances, like the Playstation, where integration is important, maybe. In a PC, where you got to have expandability, why?

It's not like AMD has been stellar with chipsets, in fact, I'd say that's the weak link for Athlon right now. But Intel seems to be squandering what was always a strength for them, and I can't figure out why, except for maybe some "business model" thing, as in this recent Rambus bluster piece:

Created earlier this year at the urging of Intel Corp., the alliance hopes to field a low-cost PC main memory for the 2003 market. While a number of leading DRAM vendors were asked to join the committee, Rambus, whose Direct Rambus DRAM architecture is now making its way into high-end PCs, did not receive an invitation.

That fact doesn't concern Avo Kanadjian, newly appointed vice president of worldwide marketing at Mountain View, Calif.-based Rambus. ?When ADT is done analyzing their DRAM requirements, they will be looking at a Rambus-compatible device. They'll need our participation,? Kanadjian said. . . .

Asked how Rambus would reconcile its roprietary, royalty-based business model with the ADT's insistence on a royalty-free, open architecture, Kanadjian said, ?There must be some accommodation [by ADT] with the Rambus business model or a derivative of that model.
ebnews.com

Offhand, I'd say Rambus is in no position to enforce the tough talk there, except with Intel behind them. I can't see where the benefit to Intel is worth the risk, but Intel seems to have chosen that path. Sure doesn't seem to have much to do with "what the customers want" in the PC sector, though.

Cheers, Dan.