>>>> I am not so sure about the forecast nor the assumption that the industry has not adjusted to design rules >>>
I agree with you completely. It is not really an issue of selling more DRAMs for a good price and having a bankroll, but one of keeping up with technology. It is pretty well understood that during a slowdown is the time to buy machines and spend money on capital improvements. ASML, who make the best stepper in the world, had no lead time for delivery three months ago, now it is out almost a year. Samsung, Hyundai, Micron, LSI, and other non-volatile memory producers (ASIC, EEPROM, etc.) are buying euipment. You cannot produce a part .25um or below without two tools now; DUV Steppers and CMP (who would of thought 7 years ago a process that uses basically sand and salt would be used to manufacture DRAMs?)
Those who are not "upgrading" now will be out of business. Three - four years ago MU was the only DRAM company keeping up. The Koreans' learned from that and have not only caught up, but also over taken them in most cases. There was tactical error by MU three years ago: if they would have shrunk their 16 MEG production and kept the process at .25, and yield constant, they could have buried the rest of the competition. At the same time instead of doing a .21 they could have put all R&D resources into the .18 device and been a world leader. Those calls are tricky and without a real technician on the BOD it can be missed.
Anyway good call.
|
