AMD:News, Press Releases and Information Only! Message Board

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!

AMD 222.50

+7.1%

3:43 PM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: GW who wrote (694)	9/24/1997 1:53:00 AM
From: Patient Engineer	of 6843

Nice article today in EETimes speculating on AMD's plans for Socket 7. Sorry no link, I read the paper. Over the last few weeks I've heard that K6+ will fully pipeline the FP unit, add a 2nd MMX unit, add floating point instructions to MMX to speed 3D (MMX is 9 parts marketing to one part reality when it comes to graphics currently), increase the size of the L2 cache by 2x to 4x depending on the source of the rumor and today Atiq indicates that it might support a backside L2 cache. Any and all of these would be good architectural enhancements and all of them are practical to do in .25 micron. If AMD does all of them, then it will be a very significant competitor for Deschutes and Katmai next year. We'll find out at the Microprocessor Forum. The big question is when will AMD be able to ship this part in volume? If they can ship it in volume before Q4'98 (which is questionable given AMD's difficulties at Fab 25), then it will have a big positive impact on ASPs beyond what LB and Brown Bros are projecting. Certainly Intel will not sit on their hands, but one thing about Intel is they don't tend to surprise. They are so big, that they have to telegraph their punches. Deschutes seems to be little more than a clock tweak with perhaps added assistance for MP. Not an adequate threat to K6+ if the rumors are reasonably accurate. Katmai is more of a threat because it will extend the x86 instruction set and it isn't clear that AMD will be able to maintain absolute compatibility in a timely manner. Merced will create a marketing headache for AMD, but it is unlikely to be relevant to the mainstream PC business until 2000. Overall the picture is muddled (which is why I continue to scale back my holdings of AMD). In the short run, AMD should get a boost with a good conference call after earnings (though the fact that many of the analysts covering AMD are asleep and haven't changed their earnings estimates will assure that AMD gets another slug of bad press with the announced loss), and Microprocessor Forum should be good (though Intel will steal the limelite with details of Merced). My biggest short term worry is that AMD will not even make their 1M chip estimate for Q3. With street prices increasing in the face of claimed 150K chips/week being fabricated, I feel we are at risk for a negative suprise here. AMD gave the 1M chip forecast when there were still 5 weeks left in the quarter. It is still possible for them to miss even revised number for Q3. Finally, after researching FLASH a bit more, I maintain my position that AMD is foolish to be building another fab to serve this market. Intel is the low cost producer now and the margins are already very thin. AMD should try to sell their half of FASL to Fujitsu and concentrate of the MPU business. They should sell Vantis to Altera. That would put AMD in a position to either concentrate on MPUs or sell out to IBM or Motorola. Such decisive action would surely move the stock and increase shareholder value.

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next