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 : Micron Only Forum -- Ignore unavailable to you. Want to Upgrade?

To: A. A. LaFountain III who wrote (45588)	5/4/1999 7:29:00 PM
From: Fabeyes	Respond to of 53903

Thank you for the post. Several things have changed, which is the nature of this business. First, the use of a dual poly, two - three metal level processes are becoming more frequent in the traditional non-DRAM facilities. The need for Flash, EEPROM or EPROM while trying to focus on newer needs for communications has caused this marriage to mature more rapidly. A move in the direction of BiCmos with a non-traditional interface or silicon based process (SiGe or GAS) for speed further pushes the "newer" multi poly and multilevel metal processes. Damascene will be the next "DRAM" process to enter the more "quiet" memory arenas of ASIC and others. In addition, the trend of staying with a larger geometry associated with these is disappearing. Granted these companies are just now exploring the .35um arena, and will have no need to go immediately to below .25um. It is still a major step forward considering that most processes are based on 1um -.6um. As for doing Flash on a DRAM line is considered you can look to MQD for verification of just how hard it is. As the article states, "the best solution may come with the availability of Flash built in high volume DRAM fabrication facilities", is a true way to proceed. Processes become easier and more defined when large amount of silicon are directed to them. Part of the problems seen on the MQD, and the EMBEDDED projects are they fight for a small percentage of the total WIP in a DRAM fab, would go away. However, this is the Catch 22, to be done with large yields and in a true production mode, they need what the DRAM fabs offer. I often thought that if the Flash at Micron and, other companies I am acquainted with could have had a 50% ownership of the process line they would have succeeded. However, reality strikes too. How many Flash parts can you sell compared to a DRAM? Not enough to justify the needs. The length of life with a Flash part is something in question too; though Flash seems to be having a re-birth of sorts. With EMBEDDED DRAMs being the "next" hot commodity, will the need for Flash be there in five years? Designs are moving to incorporate bits of all DRAM based processes, and soon we could see a "chipset" within a chip doing away with some part types and processes; how many DRAMs are truly DRAMs? Not many, if any. They all incorporate some kind of other dimension for redundancy, access of different bits or wordlines, etc.