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: grok who wrote (28881)	9/7/1999 2:04:00 PM
From: Bilow	Read Replies (1) \| Respond to of 93625

HI KZNerd; Any reason for your prediction re Sony's playstation? There is no doubt that rambus is most naturally suited to play stations, which is where it has been most successful to date. The reason for this is that rambus provides a higher bandwidth per chip. This is a problem not otherwise easily solved. In addition, this means that the designers don't have to deal with that silly idea of putting 24 (or 32) rambus chips onto the same channel. Man are those systems going to be sensitive! My bet is that if you give the motherboard a good shake, you're going to mess up a couple pins on those controlled impedance connectors, and discover that your memory is dropping bits. This reminds me of the disaster back at the Wafer Scale Integration company, or whatever their name was, that wanted to use the whole silicon wafer. The basic problem with rambus is that not enough margin has been applied to the design. One of the problems that sometimes gets digital designers is a tendency to believe their simulations just a little too much. People think that just because a design works at slow, nominal, and fast simulations, it must therefore work in real life. For synchronous designs, this is likely to be okay. But in designs where a whole bunch of propagation delay figures have to cancel in order to leave an eye in the data path, it is the deviations from simulation that add up to a failure. The big advantage of rambus is the smaller granularity at a particular memory bandwidth. This is of great advantage for small machines (like playstations) but is of no advantage at the high end. Prediction: If a memory maker manages to get rambus into production, they will be more loath to migrate to a different (i.e. smaller) process, as compared to previous and competing memory interface technologies. The reason is that rambus is so sensitive. No wonder the memory people hate this. Because this ramp-up time will not improve with succeeding memory generations, eventually you will notice that the available (i.e. not merely sampled, but cheap and with a high yield,) rambus chips are lagging behind other chips in density. For the playstation market, this will not be a big deal, rambus is perfect for them. But the high end market will be impacted. -- Carl