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 : Rambus (RMBS) - Eagle or Penguin -- Ignore unavailable to you. Want to Upgrade?

To: Steve Lee who wrote (44455)	6/15/2000 7:25:00 PM
From: Bilow	Read Replies (1) \| Respond to of 93625

Hi Steve Lee; Re how it is that RDRAM had a survival chance before Nov/Dec of last year, even though it is technologically a bad idea. RDRAM is not an impossible technology, it is just an inefficient one. That is, with our current (and near future) manufacturing infrastructure, it is more expensive to deliver a given level of performance with RDRAM than it is with competing alternatives. Inferior technology wins out a substantial fraction of the time. Everybody sees this all the time, in many different places. For instance, I am typing this note on a "QWERTY" keyboard. Intel was the big pusher for RDRAM. The September fiasco of last year put us all into wait and watch mode. If Intel had immediately postponed the RDRAM ramp until February or March 200, we would have immediately known that DDR was going to kill RDRAM. Instead, they left us in suspense. When Intel missed the Christmas selling season, it was all over. Memory system designers the world over took RDRAM off their list of options for future memory designs. Designers that were partially complete got reevaluated. Some of them got redirected to DDR, and some continued on to make the design wins that RDRAM has announced since then. Saying that RDRAM is "bad" technology is a simplification. A more complete description is to say that the technology is too expensive. "Too expensive" means in relation to what is delivered. Not enough bang for the buck. If Intel had gotten the memory makers to shift to big volume production on RDRAM back late last year, there would have been a chance that DDR would have been aborted. That would have left RDRAM as the only viable next generation memory standard left standing. This would have forced market acceptance. It would still have been inferior technology, but it would have been sufficiently cheaper to overcome its technological disadvantages. Engineering is mostly about money. It just isn't as simple as "Either it is sound or it isn't. Either DDR is better or it ain't." If a technology is cheap enough, it can take design wins from a superior, but more costly, technology. Which technology wins depends on the details of just how superior, and just how much cheaper. These are not the kind of things that are subject to useful one-liner analysis. The current high prices of RDRAM are silly. The performance advantage that SDRAM provides over DRAM is far greater than the performance advantage of RDRAM over SDRAM, but SDRAM didn't get accepted by the PC market until it had achieved pricing parity. The memory makers helped this along by giving breakeven pricing to their big customers. Right now, (spot) PC800 pricing is somewhere around 300% higher than PC133. I know that this price difference is falling, and I fully and completely expect it to fall much further. In fact, I am on record as stating that it should get to around 50% by the end of this year. This isn't enough. In order to get crossover, the price disadvantage has to drop to 0%. The memory makers have most emphatically and repeatedly stated that this is not going to happen (for the long run, the spot market is crazy). Since they are the ones manufacturing and selling both types (note that Rambus makes nothing), it would require gonads of steel for a designer to choose RDRAM. Engineers are by nature a good bit more cautious than that. The situation hasn't changed since November of last year. DDR ramped sufficiently that it caught fire in the graphics market. Current production of DDR is approaching that of RDRAM (in terms of units shipped). This means that the fire can no longer be put out. Thus the better technology survives and wins. -- Carl