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: Ali Chen who wrote (40064)	4/16/2000 7:28:00 AM
From: John Walliker	Respond to of 93625

Ali and Carl, re: controlled impedance There is a reason quite apart from yields for the autocalibration of the Rambus interface. Fundamental to Rambus is the use of constant current drive. This allows bits from one device to pass the output stage of another active device without being affected and without reflections occurring. It is this that allows multiple clock domains on a single bus, which will become increasingly important as speeds increase. The required drive current depends on the bus impedance. The system designer can choose the most suitable impedance and the Rambus components will all optimise themselves accordingly. A side effect of this is that any variation in drive characteristics will also be calibrated out. The reason for allowing various impedances is to allow different power consumption tradeoffs. I would guess that the Playstation 2 would use about 50 or 60 Ohm tracks and terminators, because with no connectors it would be easy to maintain constant bus impedance. In contrast, a PC with RIMMS uses 28 Ohms so that the impedance discontinuity associated with the connectors is small enough to be negligible. In contrast, the reason for varying the drive current with PC100/133 is almost certainly to control ringing. Ringing is a resonance of the bus signal between two or more impedance discontinuities, such as the stubs leading off to each DIMM. It is rather like water slopping from end to end in a bathtub. When this happens it can damage the inputs of the chips on the bus, degrade the integrity of the signal and increase radio interference. It does this because the signal level at the resonance, or ringing frequency is much higher that that just associated with transmitting the data. Hence, at that frequency the radiated signal is higher and it becomes harder to stay below the regulatory limit. As for printed circuit materials not being characterised at high frequencies, you are obviously going to the wrong suppliers. There are plenty who will do this. The relevant test equipment is perfectly standard, has been around for many years and is not at all expensive compared with the laminating presses. John