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: CatLady who wrote (51072)	8/24/2000 6:21:40 PM
From: Bilow	Respond to of 93625

Hi CatLady; Re Newell's treatise on the problems with DDR. Okay, here is the quote of interest: "DDR-200 may or may not be practical in a PC. The DDR noise problem is not a fiction of Intel, it is fundamental physics. Specifically, Faraday's Law, one of Maxwell's equations, tells us that the inductive currents (noise) generated is proportional to the voltage swing. RDRAM uses voltage steps of 0.4 volts at a time; DDR uses 2.5 volts/step. This is why RDRAM can run faster and yet be cleaner. DDR-266 is quite improbable in a PC setting. Unfortunately for DDR, basic physics cannot be finessed." Much of the potential useful lifespan of DDR will be consumed before any platforms with serious volume potential are produced if they are ever produced. #reply-14262474 First of all, the guy is wrong about the voltage step in DDR. He says it is 2.5 volts, which is wrong. He must think that DDR uses 2.5 volt CMOS I/O. In fact, it uses SSTL-2, which has lower voltage levels. Look on Page 9 of this presentation on DDR from Samsung. You will note that the voltage swing is at most 2 volt, and is more typical 1 volt: via.com.tw By comparison, RDRAM's peak voltage swing is around 0.8 volts, much more comparable to DDR. He then writes: "Faraday's Law, one of Maxwell's equations, tells us that the inductive currents (noise) generated is proportional to the voltage swing." This is just not true, except perhaps if you are keeping a bunch of other things constant, in particular, the line impedance, edge rate termination type and frequency. In short it is at best an inaccurate simplification. DDR at 266MHz utilizes frequencies that are 1/3 of the frequencies used by RDRAM at 800MHz. The guy ignores this difference completely, and it dominates the difference between the technologies. In addition, RDRAM uses lower impedance circuit traces, as these are more resistant to noise, but that is not the same as saying that they are less likely to radiate noise. The basic fact is that the list of requirements for making a motherboard that accepts RSL levels at 800MHz is much longer than the list that allows SDRAM to reach 133MHz or DDR to reach 266MHz. This is probably the best indication that you can get that DDR is a more robust technology. As far as DDR being "improbable in a PC setting", what a joke. It really is no longer necessary for me to address this kind of silly statement. The list of companies that are bringing out DDR stuff is huge. The memory makers (i.e. Samsung, Micron, Hyundai, Infineon) say it is cheap to make. The chipset makers (in particular VIA, Sis, ALi, AMD, Nvidia, Radeon, ServerWorks and Intel) say they can make chipsets that use it. The motherboard makers are checking early chipsets against DDR DIMMs, and they like the stuff. Benchmarks are just now hitting the press. If you really think that motherboard makers are going to release motherboards that can't be made just to drop the price of RMBS, you are living in a hell of a dream world. -- Carl