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: jim kelley who wrote (43879)	6/9/2000 4:34:00 PM
From: Scumbria	Read Replies (1) \| Respond to of 93625

Jim, INTC thinks. DDR is a dead end technology! That explains why Intel is designing DDR into their entire server lineup. Scumbria

To: jim kelley who wrote (43879)	6/9/2000 4:38:00 PM
From: gnuman	Read Replies (1) \| Respond to of 93625

Jim, re: <RDRAM is still a new technology. The OS drivers have not been written/rewritten to take advantage of its architectural features.> What are you talking about? The Register referenced a new ATA storage driver for the 820 I/O controller hub. Tell me about drivers for the memory controller hub. <g> Are you telling me there's software that will make RIMM's run faster? <g>

To: jim kelley who wrote (43879)	6/10/2000 2:24:00 AM
From: Ali Chen	Read Replies (1) \| Respond to of 93625

<Also, it is clear that many of the criticisms of RDRAM RSL bus lodged by Carl apply also to DDR-II, i.e., Carl does not like the transmission line bus that RAMBUS uses. However, DDR-II will have the same problem of a very high clock rate together with a large number of data traces. DDR-II will require transmission line designed busses similar to RDRAM.> The observation that any high-frequency connection would require a proper design is a very smart observation for a person like you. However, at least two major things have escaped your attention. The first fallacy is that you think that "busses" are "similar to RDRAM". They are not. The DDR will not hook up 32 spatially distributed pins to the same bus wire as the ram-bs does. Driving a waveguide at 30+ points with inherently stochastic signals as all PC data are was the stupidest idea I ever heard of. Even in old time of thick Ethernet coax cabling, real engineers knew that there are always standing waves, so the cable was marked at proper intervals where the hookup should be done. I seems that Rambs engineers are somewhat ignorant in the field, and the RIMM specifications are plain dumb with this regard. Second, you forgot that the DDR data frequency is about 3 times lower than that on Rambs. Even assuming the linear scale of difficulties in signal integrity associated with ultra-high frequencies, it does make some relief in manufacturing tolerances... Sorry jim, you probably will never recover your investments in Ram-bs. Face it.