To: Ali Chen who wrote (42760 ) 5/17/2001 5:49:53 PM From: tinkershaw Read Replies (3) | Respond to of 54805 Today we know that the Rambus-1 has failed in PC market. Only after 7-year-long efforts from real manufacturers - Intel and others - the Rambus memory was able to meet the PC requirements. That is the equivalent of saying that CDMA in 1996 was not able to meet the needs in the handset market. Which is ridiculous. Of course, in 1996, CDMA could have met the needs of the marketplace. It simply wasn't necessary at the time given the state of wireless technology at the time. Similarly RDRAM, even in its initial 500 Mhz format was not really necessary prior to reaching these Ghz speeds which are not becoming mainstream. Heck, I'm typing on a Pentium II, 266 Mhz computer. RDRAM and high-bandwidth, and Ghz processors are still not necessary. Doesn't mean it won't be in everyone's box within a year. In regard to the other theory: To fit it into mass-manufactured PCs, the memory must meet certain requirements: the channel must be long enough to accommodate manufacturable packaging, modules must be socketed, etc Then how do you explain DDR which has so far proven as unstable as the initial 820 mislaunches and also due to inherent stability problems (which I'm sure will eventually be worked out, but not at the moment for mass market and certainly not for the corporate market this year anyways. Much less trying to get the DDR 266 working). Micron has certainly not seemed to have any problem trying to establish a non-standardized technology, for which compatible parts vary from manufacturer to manufacturer as the standard, and stuffing them in a box. In fact Micron had a monopoly on the DDR PC market for the past year, they ended up practically paying someone to take over their PC business and their DDR monopoly. Tinker