Clark, BB, All, <<A technical question>>
Re: "Second, by the same reasoning, there should never been any new plants built."
I meant to reply to this last week, but was side-tracked or something. But I don't see the logic in the above statement. I was saying that chip capacity would essentially increase ~ 75% with the shrink from .35 to .25. This "new supply" of chips would only "keep up" for ~ 1 to 1.5 years, or until enough time passes that demand can catch up.
Re: "Shrinking alone can't keep up - it keeps up only with the number of transistors on any given chip."
Please read the story below.
-------------------------
FOLSOM, Calif. -- Trying to regain lost market share and boost sales for its flash memories, Intel Corp. this week will announce plans to spend $1 billion over the next two years to convert its Fab 9 flash memory plant to run a 0.25-micron process using 8-inch wafers. The process, which will be almost identical to the one used on the company's Pentium II lines, will enable Intel to eventually scrap its EPROM-based process.
Intel expects to start shipping a 16-Mbit flash based on the new process in the second quarter, followed shortly by 8-Mbit boot-block and 64-Mbit Strataflash chips. By 1999, Intel will use the process to produce 1.8-volt boot-block parts and 3-V Strataflash devices, company officials said.
The new process will yield four times as many chips per wafer as Intel's current process, and will increase production capacity and improve margins.
techweb.cmp.com
------------------------
Re: "The new process will yield four times as many chips per wafer as Intel's current process, and will increase production capacity and improve margins."
How are these two statements reconciled? Restated, my question is, from the Intel story, my take says the number of chips are increased by "four times", not just the number of transistors per chip.
Thanks for any insight on this,
Teri |
