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: Tenchusatsu who wrote (33563)	11/2/1999 1:04:00 AM
From: Dan3	Read Replies (1) \| Respond to of 93625

Re: Anyone can play that game.. Latency is a setting in the bios. You are setting the VC133 to run slow and then complaining about its performance. Would you clock a Coppermine 733 at 600MHZ and then benchmark it against an Athlon 700 running at 700MHZ? >>>>>no one promised that DDR-II would have the same or better latency than DDR. As PCXXX memory gets faster, its latency (in general)improves because the core runs faster as the speed goes up. Rambus uses the same speed core in all grades. There is a reason for this: The logic that has to serialize the data is pushed quite hard to accommodate even 100MHZ cells. This is a fundamental challenge due to the architecture of rambus, and its current performance is barely achievable on the .18 processes just now becoming available. Like any technology, it can and will (if it is still around) be improved, but rambus is maxed out until .13 is cheap enough to be appropriate for commodity memory production. There is still considerable headroom available to PCXXX/DDRXXX at .18 They really should have gone with 300/600 on a 32 bit bus for .18 - but the latency is still lousy. The 840 uses pairs of RIMMs on pairs of busses at a significant cost increase to achieve bandwidth competitive with DDR but still lags badly in latency. 333MHZ DDR is already in production for use in video cards. Rambus is a clever, though complicated and expensive, solution to a problem that has vanished: the "brick wall at 100MHZ" for DRAM. It just doesn't make sense to try to make slow DRAM run fast using that elaborate architecture when it is so much cheaper and easier to just make fast DRAM. This isn't FUD or Hype, it's just the way things ended up working out. Dan

To: Tenchusatsu who wrote (33563)	11/2/1999 1:59:00 AM
From: dumbmoney	Read Replies (1) \| Respond to of 93625

Anyone can play that game. We see too much of that on the AMD thread. You know what that's called? Partisanship. Correct. I don't like it either. But the 820 delay means that DDR now has a fighting chance, not because DDR is inherently better (it isn't, despite your arguments to the contrary), but because those on the Rambus side fumbled the ball. I think the effect of the delay is overstated, actually. Intel had already decided to do PC133, so it's not clear what changes. Thinking long term (2+ years) for a minute...I don't know what the "next next" generation of PC memory will be, but I'm quite sure it won't be 16 bits wide. 64+ bits is easy, pins get cheaper, and bandwidth needs grow. It's a mystery to me why Intel opted for a narrow memory bus. Multiple narrow channels on multiple RIMMs are not cost effective compared with a single wide channel on a single module (where a single chip doesn't have to provide the full width, but it could).