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: gnuman who wrote (42797)	5/22/2000 7:18:00 PM
From: Bilow	Read Replies (3) \| Respond to of 93625

Hi Gene Parrott; Re Willamette and SDRAM. Every product that Intel came out with that used RDRAM to SDRAM translator technology was a disaster. (Recalls, bad performance.) Why would they risk repeating that with Willamette? I think they're stuck with RDRAM until they get their DDR chipsets out. That's why they are repeatedly stating that Rambus is the only Willy solution. It is, if you want to use an Intel chipset. The fact is that even though Intel has had pretty nasty problems with RDRAM, they have had even worse problems with their RDRAM to SDRAM kluge. I mean, how would you like to be the guy who has to explain to Compaq that you are going to support Willy with a memory translator hub? The technology is slow, and worse yet, they've not got it to work reliably yet. Basically, it takes all the worst features of RDRAM and combines it with all the worst features of SDRAM. You have the lack of robustness, noise sensitivity, (most of the) motherboard layout issues and bad latency of the RAC interface, along with the noise generation of the (relatively) high voltage, badly terminated SDRAM scheme. SDRAM works fine by itself, it's immune to it's own noise, but I have no doubt that the i820 SDRAM problem was noise from the SDRAM bus coupling back onto the Rambus channel (possibly any number of mechanisms). The whole thing just bites. I know that Intel will, in the end, abandon Rambus. In fact, I'm sure that they are right now in development on DDR chipset solutions across the board, from the low end to the high, and not simply working on server DDR. It will take them a long time to get those solutions out. Until then, they are very tightly attached to Rambus. It is a horrible place to be in, and the rest of the industry knows it. VIA is expecting to pick up 50% of the chipset market by the end of the year because Intel doesn't have modern, high volume, solutions that avoid the Rambus. Even Intel is having to use other company's chipsets. In the chipset arena, they are temporarily bankrupt due to bad management decisions from the past two years. A memory translator hub isn't going to save them, I think. They will have to wait for real Intel support for DDR. I know that what I am saying that Rambus will continue (temporarily) to be supported by Intel. In the long run, Rambus is dead, dead, dead. But it takes a long time to get new chipsets out, and until Intel gives itself enough of a Heimlich maneuver to get the Rambus out of its windpipe, you will continue to see Intel perform CPR on the twitching body of RDRAM. They're attached to each other. -- Carl