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.

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

To: Amy J who wrote (107518)	4/24/2000 9:06:00 AM
From: chic_hearne	Read Replies (2) \| Respond to of 1573930

Amy J, How much AMD do you own? chic

To: Amy J who wrote (107518)	4/24/2000 9:32:00 AM
From: steve harris	Read Replies (1) \| Respond to of 1573930

Amy, It's a long history why I have the opinion iNteL is having problems. I don't think Grove is running the day to day operations of the company, and Barrett, Otellini, and Gelsinger do not install confidence. As long as the current spin continues, the problems will continue. That aside, did not intel say that the supply problems would be resolved in Q1? Was that not due to the new B0 stepping? Now intel says the supply problem will be resolved in 2H00? Willy is the answer to 2H00? Or is Otellini trying one more P3 tweak? I think intel gambled on a P3 core tweak and failed. Wasn't the "notched gates" suppose to fix the high end? Intel has exceed earnings guidance in order to make their estimates the last two quarters. Did you see the delays posted by DELL on their website concerning the high end P3s? I think there is a 800-1gig P3 problem, and it's not demand. Yields may be fine, bin splits may be the problem. I believe Grove made mistakes, quickly recognized them, and became the great businessman he is. I think the current management has not quickly recognized the problem yet. A previous post outlined what I saw and I believe it remains a people problem near the top. The employees busting their butts are not the problem. Message 11910572 steve

To: Amy J who wrote (107518)	4/24/2000 9:38:00 AM
From: that_crazy_doug	Read Replies (1) \| Respond to of 1573930

<< It sounds to me like there are two different views: the camp that believes Intel's yields are poor vs the camp that believes Intel is faced with high demand. What are the arguments for proving Intel's yields are poor? My question (which no one has answered yet) is: if the processes improve, why would raw materials be expected to increase? (this is a sincere question - I'm not a process girl :) >> My main thought on why yields aren't too good comes down to this: They had 8%/13% growth YOY for Q4/Q1. That is the huge growth they've been facing. I haven't followed Intel for the past 10 years, so I'm taking a stab in the dark at this, but I bet that's some of the slowest growth core business revenues in the past 10 years. So I don't think it being a big surprise is much of an excuse. This growth also happened in the middle of a process shrink, so capacity should have been doubled in every fab that went through the shrink if yields were good at .18. I'm not sure how much .18 capacity Intel has right now, but even if it's 1/4th, they should have been able to handle 25% growth without any new fab plants. Basically for me to believe yields are really good, it would imply only 13% of the capacity is at .18 micron and that they were running at complete full capacity in the previous year. (I believe they have much more than 13% .18 micron capacity, and Elmer keeps talking about the big semiconductor recession, so it doesn't seem likely to me they were running full tilt a year ago, though I really have no idea on the second point, maybe you do?) Many people keep pointing out that there was no way they could forecast demand 2 years ago to start building fabs. That may very well be true, however, it's my opinion that if the ramp to .18 was really going well they wouldn't need any new fabs, because they'd double capacity without adding a single fab. They're expected to go to .13 in another year and a half and thus double capcity again. I don't think that they're going to grow by a factor of 4 from when they were entirely .25 micron. As for your last question, I would expect raw materials to decrease as process improves. (at least decrease for the same amount of processors made). The only reason I could think of them increasing is if yields weren't very good (so you had to run more wafers to get fewer parts), or there was a lot of verification that temporarily inflated it (wafers running through that weren't making real parts). I guess the other possibility is if you are just producing a heck of a lot more chips, but then it'd be misleading to say that raw materials were up (since you're really just making a whole lot more product).

To: Amy J who wrote (107518)	4/24/2000 10:39:00 AM
From: pgerassi	Read Replies (1) \| Respond to of 1573930

Dear Amy: The answer is simple. To get a somewhat better process, different materials need to be used. Since these materials are generally not currently in high production, they tend to be more expensive. The easy increases have been already used. Now they must use more exotic materials and techniques, (read expensive). For an example, one of the things limiting performance is the capacitance between layers and at the MOSFET gates. Higher capacitance, requires more current and time to fill and empty. When you can get a lower capacitance, the speed increases. The easiest material, (and still most used), is Silicon Dioxide. You can get it by simply diffusing oxygen into the wafer. To use a different material, the silicon must be etched away and filled with the new material. This adds costs and some lower capacitance producing materials are quite expensive. Also a lower capacitance material allows the thicknesses to be reduced which is another way to increase speed. Any of the posters who know the materials end, such as PB, could expand on this in further detail. Thus you can see how the same number of wafers, can cause a higher raw material cost when exotic materials and processes are used. Pete