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: semiconeng who wrote (22667)	12/16/2000 9:36:18 PM
From: combjelly	Read Replies (1) \| Respond to of 275872

"Also, when intel extended 16 bit into 32, they had tons of resources, a huge installed customer base, and very little competition," Depends on what you mean by "very little competition". There was AMD and Harris at least competing for the 80286 business. In fact, an annoyance for Intel was that AMD was pushing the 80286 to 16MHz and Harris reached 20MHz with their 286's, while Intel had capped performance at 12.5MHz to leave headroom for the 80386. Since the 286 would whomp on a 386 at the same clock rate in real mode (especially if there were frequent interrupts), this was a major problem for Intel. Luckily the desktop market saw the 386 as the future.

To: semiconeng who wrote (22667)	12/16/2000 11:33:12 PM
From: porn_start878	Read Replies (1) \| Respond to of 275872

Wait a minute semiconeng (Part II)... ----Quite a bit more than a handfull. I could sweep around the net and repost all the vendor links, but frankly it's been done many times. Comparing Digital with intel for designing and marketing microprocessors, is apples and oranges. Digital aint no intel.... By a Long Shot I consider 77 apps or so to be a handfull. The size of the Itanium's market could be approximated by : Size = 77/(all existing apps) * (an app's usage expression) ... but then you have to consider that thos app doesnt run alone on the system so you find the probability that a user run it's setup of n applications on an Itanium by calculating the nth power of Size (size is always less than 1...). As for Digital, they were very powerful at the time... they were even bigger than Intel, and they came up with a chip that revolutioned the PC. They were an healty orange while Intel is closer to an enormous apple rooting from everywhere. Ibm is an other example of a company that unsuccessfully tried to BE the future. Also, when intel extended 16 bit into 32, they had tons of resources, a huge installed customer base, and very little competition, AMD is attempting to repeat history, with allot less money, and allot smaller base, not to mention competition from a 600lb Gorilla. Good Luck. My interpretation is that it's not Intel that is responsible of the successful move to 32 bit but rather the need of the market to have a backward compatible platform to smootly move to 32 bit. Max interesting debate I think :)

To: semiconeng who wrote (22667)	12/17/2000 11:52:56 AM
From: Dan3	Read Replies (1) \| Respond to of 275872

Re: I would rather have a forward looking company. When I first came into High Tech, as a lowly Test Technician, One of my first mentors said to me: "In High Tech, There's No Standing Still In The Present. If You're Not Moving Forward, Then You're Falling Behind". I realize that others have different strategies, go for it. Exactly. While AMD is moving to copper/SOI, Intel is still months away from having a plain copper process such as the one AMD has had in mass production for a year. While Intel is moving to 64 bits only for low volume server chips, AMD is moving its mainstream processors to 64 bit. Intel isn't falling behind, it's falling further behind. Intel enters survival mode starting next quarter. Their slashing of capital investment makes it abundantly clear that they are aware of this. The P3/Celeron series is simply not competitive with AMD's equivalent chips. Hennesssy and Patterson's "Computer Architecture A Quatitative Approach" has a discussion of wafer yield on pages 10-13. They work one example that almost perfectly mimics what Intel faces when wafer starts are P4 instead of P3. The H&P example compares a 225cm2 die with a 100cm2 die (P4 is 227cm2 and P3 is about 95cm2). Using the same process, the expected yield at 100cm2 would be 132 good die per wafer, while the expected yield of 225cm2 die is 26. Intel could probably produce 60 million P3's next quarter or 12 million P4s. The tradeoff is that steep. P4 is not going to be a high volume part on .18 (unless Intel wants to cede half the market to AMD/VIA). AMD has a major new layout of the Athlon coming out very soon that will exacerbate the MHZ disadvantage that P3 has, reduce the MHZ advantage P4 has, possibly improve the IPC advantage of Athlon, and may bring Athlon's die size down, as well. Intel has quarterly costs of about $6Billion. When they are selling 35 million processors at $180 ea, that's not a problem - even if the non IAG part of Intel's business loses over a half Billion per quarter. During the next few quarters, P3 will be falling further and further behind Athlon, AMD will be fielding notebook CPUs again, and MPU business and server CPUs for the first time. Meanwhile, P4 is too large to produce in volume, so Intel's ASPs are going to drop. With AMD selling increasing volumes of faster processor and targeting an ASP of $100, Intel will have a hard time maintaining an ASP of $125 - which will drop revenues by $1.9 Billion (about what they were forced to cut from capex, so far). At the same time investment earnings look to be dropping by hundreds of millions per quarter. Intel may actually lose money during some of the next 3 quarters. In Q4 of next year, if everything goes well and there are no delays, Intel's .13 process should be producing enough volume to let Intel get its head well above water for 1 to 2 quarters. But by then (if all goes well for AMD), AMD will be fielding a line of 64 bit processors, with a smaller die size than the 32 bit P4, a higher IPC than the 32 bit P4, and equal or higher clock speed thanks to SOI. At that point Intel heads back under water again, and this time there isn't anything to bring it back up again for quite a while. Dan