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: JBoyd who wrote (34307)	7/11/1998 6:28:00 PM
From: Maxwell	Read Replies (3) \| Respond to of 1572033

JBoyd: <<I'M DESPERATELY LOOKING FOR EXCUSES TO HOLD THIS STOCK!!!>> Hold on to the stock. This stock is pretty much bottom out. I will get to the details a little later. <<Do you think that AMD has some (very few I'm sure) competitive advantages over Intel at this time. Is AMD losing more ground? Will they have a product to sell when Dresden is completed?>> AMD does have competitive advantage over Intel. Here are some 1) K6/K6-2/K6-3 die size at 0.25um are all SMALLER than Intel 0.25um PII. Smaller silicon are means cheaper to manufacture if the yield is good. 2) K6-2-300MHz can do 1.2GFLOPs FPU versus PII-300MHz of 0.3 GFLOPs of FPU. This shows up when you play games such as QuakeII with 3DNow drivers. K6-2-300 will beat PII-300 when softwares are written to access K6 3D engine. 3DNow is the FUTURE and Intel is behind. 3) K6-2 is selling cheaper than PII. 4) AMD GETS MORE dice per wafer than Intel gets with their PII. 5) Cost of manufacturing of socket 7 chip is cheaper than slot 1. <<K-6 design. You indicated that it was a great design, but they were overly aggressive in implementation. Yousef made the same point and says AMD is two to three years behind Intel in process technology. Is process technology basically manufacturing or is it design? I'm really struggling to understand this aspect of the business. >> Don't listen to Yousef. He doesn't know what he is talking about. His process knowledge is limited. He understand little of IC design and none on semiconductor business. For your information, IBM/AMD process is the best and most aggressive in the industry today. With local interconnect AMD is able to shrink the chip smaller that what Intel could achieve. The K6-2 with integrated 256KB cache has 21.3M transistors with only 117 sq. mm in die size on 0.25um process whereas the PII on 0.25um has 7.5M transistors with die size 131sq. mm! Smaller die means you can put more functionality on a chip without sacrificing the yield and this is a COMPETITIVE ADVANTAGE. Intel is more conservative in their design because they want their chips to be MORE MANUFACTURABLE. This in turns lose out on the competitive advantage which results in a LARGE DIE SIZE. Yousef only premise is that "because Intel achieves higher clock speed Intel has a superior process". This is a fallacy. Higher clock speed comes with design for a given process. There is only so much you can do with process and the rest must rely on design. Digital is able to achieve 550MHz on 0.35um while Intel can only squeeze out 300MHz. Is it because Digital has superior design? None at all. Digital has better IC design. AMD gambled on a complicated process to get the competitive advantage. The gamble so far has really paid off. Not only did they mastered the local interconnect, bumps, and 5 metal layers they get great yield. AMD is penetrating the market with the K6-2 very well. It is a two thumbs up product. This is one unique product that Intel cannot block. The reason is that Intel HAS NO 3D CPU! By Christmas 3DNow should be written in stone. What about AMD making money? Watch for Q3 and Q4. AMD should be making money. The reason their fixed cost is so high is that they are building Fab30. This drains their cash. By Q3 and Q4 their sales should outpace their cost and investment. AMD needs Fab30 to compete. It takes a critical mass to effectively compete against Intel. The future for AMD is much brighter than the K5 days when the stock was at $13. Back then AMD didn't have a future. Now AMD has a whole line of products come down the pipeline such as K6-3 and K7. There are customers (CPQ, HWP, Acer, IBM, GTW, Toshiba, NEC, Packard Bell, CTX, Polywell, etc.) want AMD parts and they want AMD to succeed. Maxwell

To: JBoyd who wrote (34307)	7/11/1998 7:15:00 PM
From: Elmer	Read Replies (2) \| Respond to of 1572033

Re: "K-6 design. You indicated that it was a great design, but they were overly aggressive in implementation. Yousef made the same point and says AMD is two to three years behind Intel in process technology. Is process technology basically manufacturing or is it design? I'm really struggling to understand this aspect of the business. " It's manufacturing but a process has to be developed first by people like Yousef before it is turned over to manufacturing as being production ready. A product is the marrage of a design and a manufacturing process. The designers are given a set of constraints called design rules which define the reasonable limits of the process. Paul or Yousef can do a much better job than I describing them but a few are transistor channel lengths, metal line spacing, oxide thicknesses etc etc. Libraries of cells made up of such things as flipflops, latches, gates, muxes, inverters etc. are used by design to turn their logical design into a real circuit. All cells are designed based on the design rules defined by the process developers. The rules are defined to achieve the best combination of performance, yield, die area, throughput time etc. Any one of these rules can be optimized at the expense of another. Die size can be minimized at the expense of yield and throughput time, an example would be AMD's decision to use local interconnects although I don't know as a fact that this is the cause of their major yield problems. It is the concensus of many here that AMD simply tried to optimize too many process factors and the whole thing came falling down like a house of cards. Their design was implimented with a set of design rules that was simply too aggressive for their process capabilities. Their design rules were never proven in high volume manufacturing because they never had a high volume part on .35u. They couldn't get high speed AND high yield on minimul die area all at the same time. I don't believe any technically competent person would have made such a high risk decision. Jerry said he was betting the farm (yours and mine, certainly not Jerrys farm) on the K6. He's the one who had the final say of pushing things to the breaking point. He was undoubtedly warned of the risks but his goal was and is to fight Intel to the last cent he can beg borrow or otherwise con out of investors. Too bad his goal never was to enhance shareholder value. As for holding this stock, I'm still holding mine but really only as a hedge. My cost is ~$18. Why not sell some covered calls? EP

To: JBoyd who wrote (34307)	7/11/1998 8:42:00 PM
From: Xpiderman	Respond to of 1572033

>>LOOKING FOR EXCUSES TO HOLD THIS STOCK!!! << Well, the following statements are from the Second Opinion ( suretrade.com ) about AMD: Symbol: AMD Name: ADVANCED MICRO DEVIC Exchange: NYSE OPINION: AVOID , DOUBLE BOTTOM BREAKOUT!!! COMMENT: MACD indicates a BEARISH TREND Chart pattern indicates a STRONG DOWNWARD TREND Relative Strength is BEARISH Up/Down volume pattern indicates that the stock is under DISTRIBUTION The 50 day MOVING AVERAGE is falling which is BEARISH The 200 day MOVING AVERAGE is falling which is BEARISH STOCK broke DOUBLE BOTTOM (BEARISH ) Point & Figure Chart Formation Last Week RECOMMENDATION AVOID THE STOCK IF YOU ARE LONG; CLOSE POSITION OR MONITOR STOCK CLOSELY

To: JBoyd who wrote (34307)	7/12/1998 1:02:00 AM
From: Dale J.	Read Replies (1) \| Respond to of 1572033

JBoyd, I'M DESPERATELY LOOKING FOR EXCUSES TO HOLD THIS STOCK!!! Well, I see you already have had input from some very smart people. I can only add, that if you do decide to hold, make sure its part of a diversified portfolio. Because these stocks can and do go to zero. I'm not suggesting that will happen to AMD, but just make sure you diversify. A case in point is Zenith. I never owned it myself, but I remember back about a couple years ago it was making a resurgence over the cable modem craze. It has since went out of favor with the Street. In fact it traded last at 3/8 before it expired just a month or so ago. Again I am not comparing Zenith to AMD, only citing the justification for diversification. Dale Below is the discussion heading from the Zenith SI thread on May, 1996. (Just two years ago) This gentlemen gave the Zenith thread his insights (in astounding detail I might add) as to what will happen: Zenith has found it's support level. A major majority of the money that moved into Zenith has moved out. Over the next couple of days Zenith will trade in a narrow range, most likely on the downside, possibly touching $15. Beginning around Tuesday 14th, Zenith will move higher, to approx. $21. It will hold this pattern through the week. There will be another small sell-off possibly on Friday 17th, most likely Monday or Tuesday the 20th-21st. This will be followed by a strong surge in the stock price on May 22nd-23rd to approx. $28. This strong surge will put the money back into Zenith. It will hold this pattern for a few days with some potential sell-off.