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 : CYPRESS Semiconductor (CY) -- Ignore unavailable to you. Want to Upgrade?

To: Bald Eagle who wrote (1945)	7/2/1998 2:43:00 PM
From: Mark Oliver	Read Replies (1) \| Respond to of 2694

TJ made such a foul of himself at the annual meeting it was pitiful. I should have sold from the pay phone across the street. I believe CY has missed too many opportunities. This company is a looser. I finally sold after holding for over 2 years despite the great benefits of selling in the money calls and having them expire worthless again and again. I attended 3 annual meeting for CY before selling and each time TJ had some issue to beat a pulpit on. Well, the only pulpit shareholders should beat on is getting a new CEO. All his diversions from the failure of CY have no value compared to the huge underperformce of this investment. This nun and her story was meaningless. Anyway, just read this article about the future of SRAM. Pretty grim. Maybe CY can fill the gap with with their own embedded SRAM ASIC's such as USB chips? If history is any indicator, CY will fail. Only chance is to sell the company and get better management. Regards, Mark From the June 29, 1998 Issue of Electronic News In-Stat In-Sights Roadblocks In SRAM Future By Jesse O. Huffman, Senior Analyst, Static RAM Services The In-Stat five-year forecast for Static RAMs has just been completed, and the research has revealed many interesting issues. It appears that over the forecast period (1998-2002), SRAM vendors will see dramatic competition from emerging technologies. The fact that SRAMs are under attack from substitution technologies is not really new. Every few years, DRAM technology advances in speed performance and with its inherent cost advantages, displaces the slower performance spectrum of SRAM products. This lasts for a short period of time, then the SRAM designers advance their speed performance dramatically over DRAM and they gain back market share. This is all about to change dramatically. There appear to be several technology icebergs on the SRAM horizon. 0.18-Micron Process, The Enabler ------------------------------------------------------------------------ Most of the major microprocessor and ASIC vendors have announced that they will have 0.15--0.18-micron product available in some quantities in 1999. Production ramp-up throughout the industry will take place in 2000 and 2001. At this feature-size level, the integration capabilities are dramatic. For example, NEC claims their 0.18 technology will provide 80,000 gates per square-millimeter. This compares to 40,000 gates per square-millimeter for their 0.25-micron process. NEC has also stated that their technology will allow a given design to accommodate up to 128 megabits of DRAM or 16 megabits of SRAM and still have 17 million gates available for logic. Memory speeds will be greatly enhanced due to the elimination of off-chip latencies. Future ASIC designs will integrate every function possible for added value. SRAM densities of 1Mbit and less will be embedded, and discrete parts will disappear. The integration of PC L2 cache onto the microprocessor will free up SRAM wafer capacity. In 1999, when Intel and others begin L2 integration, SRAM vendors such as Samsung, Mitsubishi, NEC and Motorola will have wafer capacity freed up. Couple this move with the advancement of 0.15 to 0.18-micron process and it is quite clear that there will be no shortage of capacity over this forecast period. Substitution Technologies ------------------------------------------------------------------------ The very largest of the semiconductor companies have always invested in pure research in hopes of developing the next "killer" memory technology. For decades, research has been conducted on new memory structures, in search of a cost-performance breakthrough. News about such efforts is limited. Occasionally we get to hear about a failed attempt or an abandoned effort. The successful efforts are kept very confidential until they are market-ready. At least one such effort is getting ready to hit the market by 2001, which is within this forecast period. Motorola and IBM are developing a memory technology referred to only as "MRAM." Very little detail has leaked out at this time, but what we know is this new technology is non-volatile and costs half of current DRAM technologies. Motorola and IBM expect to replace DRAM, FLASH and half the SRAM markets with this technology. This is truly dramatic stuff if they pull it off. The In-Stat five-year forecast model includes (but not limited to) the following: _ PC L2 cache integration will begin in 1999, will absorb 50 percent of SRAM cache in 2000 and 100 percent in 2001. _ ASIC logic designers are currently incorporating SRAM into 50 percent of all new designs; this percentage will increase with on-chip SRAM densities of 1Mbyte and more beginning in 2000. _ The "Specialty" SRAM category will grow dramatically due to a move to combine logic-SRAM designs, and more "pseudo" SRAMs in the form of DRAM cores with SRAM interface. _ At least one new substitution technology, such as "MRAM" will become successful by 2002. The resultant In-Stat forecast shows the 2002 worldwide dollar market for discrete SRAM products to be substantially below the current $4 Billion dollar per year run rate. For details see In-Stat report SR9806MF. This is a 58-page report containing 28 data tables and 32 graphs.