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: Bilow who wrote (30671)	9/26/1999 6:35:00 AM
From: John Walliker	Read Replies (3) \| Respond to of 93625

Carl, The problem with RDRAM is that it just doesn't leave much room for error on the parts of too many different parts. The technology was over done to the point where there is too little margin left for deviance from specification. The problem may well be incorrect logic in the chip-set and have nothing to do with whether Rambus provides adequate margins or not. ...through six socket connectors (i.e. two per RIMM), you are not just asking for trouble, you've married its daughter. I do agree, connectors are the invention of the devil and should be used as sparingly as possible. However, there are far fewer Rambus signals, than (DDR) PC100/133, so there is probably little real difference. Many applications of Rambus will involve chips soldered to the motherboard, which brings many advantages. I suspect that the low impedance chosen for RIMM implementations of Rambus is to minimise the perturbations caused by connector parasitic impedances. However, RIMMs do use gold plated contacts, unlike many SIMs which attempt to achieve reliable connections with solder coated contacts. I believe that the biggest disservice Intel did to the computer industry some years ago was to introduce chip-sets that did not require parity or error checking and correcting memory. The box-makers of course adopted the cheaper unprotected memory. This means that single-bit soft errors can quietly corrupt data without anyone noticing until it is too late. With parity protection the computer stops when an error is detected. With ECC, single-bit soft errors are automatically corrected and multi-bit or uncorrectable errors (which are much rarer) stop the system. I always try to use ECC, despite the higher cost. Errors are very rare indeed, but just occasionally a particular memory chip may have a high error rate. Its good to know. John