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: Tony Viola who wrote (24345)	7/8/1999 11:33:00 PM
From: J_W	Respond to of 93625

Tony, (warning technical discussion follows) Thanks for the response to my post. Who knows how many people are out there building boxes which are never checked for EMI. As long as no one complains or the FCC doesn't know about them, they just go merrily along churning out boxes that may or may not meet FCC EMI specs. Certifying all the parts that go into a box would help all boxmakers to meet specs without undergoing testing on each configuration. That would make a whole lot of sense. Just a few comments on your post. To the Rambus design, with differential pair wires for clock (and signal?), and the terminated transmission lines, and, that other gizmo where the clocks are purposely skewed, these all help minimize emi a lot. Clocks are a primary source of EMI. Rambus uses a differential clock, which is by its nature self canceling in regards to EMI. This is due to fact that the two clock lines are complementary to each other. The Rambus Channel is designed to minimize skew. That is clock and data do not get out of sync with each other when they arrive at their destination. Both clock and data travel down separate equal length transmission lines in parallel. This also allows pipelined operation. That last one, I forget the name of it, where they skew clocks, only at some portions of the cycle, or period...sound like purposeful clock jitter generation. Hope they know what they're doing. Clock jitter is a long time enemy in high speed systems. Okay. This is where spread-spectrum clocking comes in. By varying the clock frequency by small amounts, the total energy radiated at any particular frequency is reduced. In other words instead of have all EMI occur at one frequency, the EMI is spread over a range of frequencies. Since clock and data propagate down the same bus in parallel, they will still arrive in sync (no skew). Rambus has specifications on how much the clock can be frequency shifted without causing problems to the operation of Rambus Channel. Tench, Scumbria please jump in and correct me if I have mis-stated any of these concepts. Now as to how EMI would be handled in an SDRAM system is an area which I don't know much about. Would someone care to comment on this? Regards, Jim