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: TigerPaw who wrote (18974)	4/22/1999 12:39:00 PM
From: Dave B	Read Replies (2) \| Respond to of 93625

TigerPaw, Thanks for jumping in. I really want to understand this technology, unfortunately the one 'C' I got in college was in the introductory EE course (and they were being generous <g>). I have to admit that I understood very little of your response -- could you take a step back and explain it in simpler terms? And it sounds as if this would be another difference between RDRAM and DDR DRAM -- is that correct? Thanks mucho, Dave

To: TigerPaw who wrote (18974)	4/24/1999 6:56:00 PM
From: Alan Hume	Read Replies (1) \| Respond to of 93625

Hi Tiger, "Very few electrons are involved." Hmmmmmmmmmm, Why does it get so hot then? Alan

To: TigerPaw who wrote (18974)	4/27/1999 1:34:00 AM
From: Dave B	Read Replies (3) \| Respond to of 93625

TigerPaw, I had a chance to go over your responses and Tenchsatsu's with my technical friend and he disagreed with some of your statements. Hopefully I can relay them accurately and you can tell me where you disagree with his disagreements (and hopefully this won't get disagreeable <g>). In his view, there are no radio signals traveling around the system on top of which are encoded the data or control signals. We were assuming that you intended your explanation to mean that information was encoded onto some underlying carrier signal like AM/FM radio signals are encoded on top of a carrier either by modulating the amplitude or frequency (so this may be our first misunderstanding of your statements). As he explained it, the Rambus systems are still digital systems. A certain voltage level indicates a one and another voltage level indicates a zero -- standard digital technology. Where I think he may have agreed with you is in the explanation of the saturation of the driver transistors. In regular SDRAM, the levels for the logical 1 and 0 are the same as the supply voltage and ground (Vdd and Gnd). To get the signals to hit this level, you have to saturate the transistors you mentioned to some degree. The benefit of this is the noise immunity at the cost of speed, because to change the signal to the other extreme, you have to unsaturate the transistor (it sounded as if he was describing a process similar to draining a capacitor) and that takes some time. In the end, you're only going to be able to transition the signal at a certain speed, and he thinks that 133Mhz SDRAM/DDRDRAM are hitting that limit, short of reducing the difference between Vdd and Gnd to some negligible amount, which then makes you very susceptible to noise. He said that the way Rambus works is that Vdd and Gnd are at 2.5V and 0V, while the data and control signals transition between 1.0V and 1.8V so the transistors don't get saturated and the signal can change much faster. The trade-off, however, is that the system is now much more susceptible to noise. Rambus gets around this with their pseudo-differential technology where each signal is compared to the Reference Voltage signal (in normal differential technology, he says, each signal would have it's own paired-off signal for comparison) providing greater noise immunity. He also added that Tenchusatsu's comment about the memory chips only having to drive the signal to 1/2 the transition difference (.4 volts) when sending data to the chipset seems correct since the reflection bumps it up to the full .8 volt difference, and that maybe this is what you were referring to. Any thoughts on all of this? Thanks for your time. Dave