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 : AMD:News, Press Releases and Information Only! -- Ignore unavailable to you. Want to Upgrade?

To: Pravin Kamdar who wrote (4987)	3/17/1998 7:13:00 PM
From: Petz	Read Replies (1) \| Respond to of 6843

Pravin, re:(2) "the more the merrier" ...legitimized the C6+ -- Yes, but the deal also legitimizes the AMD-3D extensions. ...C6+ should yield better than the K6-3D -- Maybe, AMD K6-3D is only 80 mm die size. The C6 (0.35) is 88 mm, but the C6+ (0.25) may not be any smaller because of silicon for 2 FPU's and MMX units. ...IBM won't start cranking out chips for AMD and IDTI until the fourth quarter The IDT announcement did NOT say when the first wafer starts would occur, so, possibly the IBM C6+ will be out in the 3rd quarter. ...keep severe pricing pressure on AMD. I don't think Centaur (IDT) could reasonably take more than 10% of the x86 market by the end of 1998, no matter how good or what price their products will be sold at. This is because they are an unknown. Right now a Cyrix PR233 chip is much cheaper than a K6-233, but outperforms it (except games) by 5-10%. I'd be worried about Centaur Winchips cannibalizing the Socket 7 market if IBM labels them as IBM chips, but thats not implied by the words "foundry agreement." BTW, I do own more IDTI than AMD. Petz

To: Pravin Kamdar who wrote (4987)	3/18/1998 1:15:00 AM
From: Maxwell	Respond to of 6843

Pravin: Here is my take on the MXi Cyrix chip. In theory this chip sounds very good but in reality it will be difficult to implement. To integrate the graphic engine into the CPU there are 2 ways of doing it. 1. Using 1 engine and share the time between graphic and other execution instructions. 2. Building graphic engine on board next to the CPU on the same die. The option 2) would give you a faster performance improvement over current architecture convention (grahic card on PCI/AGP bus). To implement 2) you must build a specialized high transfer bus for graphic output and work out all the CPU interrupts involved. This can be done and protocols must be standardized. It is very difficult these days to get anyone to agree on anything. If no one buy in then Cyrix is on its own making its own motherboard. The other disadvantage is that the die size is now quite large and thus more difficult to manufacture. A 686MX Cyrix chip consists of about 8M transistors and a good graphic engine can take up to 4M transistors. Including the controllers the chip will be at least in the 10s of millions of transistors. In my opinion the MXi will be on the drawing board for sometime. Maxwell