SI
SI
discoversearch

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 (75419)7/7/2001 9:52:52 PM
From: Bilow  Read Replies (1) | Respond to of 93625
 
Hi Gene Parrott; Re: "Strip or micro-strip transmission line is well known as an effective method of controlling EMI on a PCB." Such techniques can be used with DDR as well as RDRAM. Of course, since RDRAM has higher edge rates, currents and frequencies, the problems are worse for it.

Re: "The RSL trace is only 18 mil's wide." While this is true, the center to center distance is 40 mils, as I stated, and this is how it should be compared to DDR, which doesn't require ground isolation. The 40 mils is composed of 18 mils of RSL trace, 10 mils of ground trace, and 12 mils of spacing. For an explanation, go to page 36 of this Intel link:
developer.intel.com

Sometime ago on this thread I posted links to Gerbers for the Intel i820, and the very poor use of PCB by RDRAM was obvious. Poke around with the above link to see just how much space an RDRAM channel really requires. If you want to seriously debate it, you have to go and read the above link.

As far as EMI for DDR goes, the voltage was decreased for DDR, as compared to SDRAM and the signalling technology was changed from unterminated CMOS to series terminated SSTL-2. The combination of these two changes mean that DDR doesn't have a worse EMI problem than PC133 SDRAM, which is already shipping. You haven't heard of Compaq, HP, or NEC being unable to ship PC2100 systems because of EMI, have you? The EMI "issue" just doesn't seem to have created any real problems for DDR.

-- Carl

P.S. I should note that for me, the term "EMI" includes crosstalk between signals. In other words, EMI is the general term, "crosstalk" is a specific variety of it. This only seems natural to me because it is clear that crosstalk is (1) electromagnetic in nature, and (2) a form of interference, LOL. But if you want to use the terms the way everybody else does, be my guest. It just ain't worth arguing over. Okay, I'll try and be more specific.