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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: John Walliker who wrote (115214)	10/30/2000 12:59:56 PM
From: Tony Viola	Read Replies (2) \| Respond to of 186894

John, I'll look at that website when I get a chance. Maybe Rambus defined the electrical environment perfectly well. However, someone blew it on one side of the interface design or the other (or both). As an Intel stockholder, I do have the right to ask questions when a new product development goes awry, and costs me obvious bucks. Do you know what went wrong? I personally won't feel good about the Pentium 4 until there are alternative memory types available for it. By your posts, I can tell that you have an excellent background in high frequency (probably any frequency) environments. Me too, and I also have been around computer architecture and design for a few years. That's why I've always been leery of Intel going way down the Rambus road. Sure, the very narrow data bus is nice, but if you don't design in the transmission line environment perfectly, it's prone to many, many electrical problems. I sound like a broken record, I know, but one never knows when someone might be reading that has some influence. Then again, it is happening: DDR SDRAM...keep that design moving, please. Tony

To: John Walliker who wrote (115214)	10/30/2000 1:09:19 PM
From: Ali Chen	Read Replies (2) \| Respond to of 186894

<Are you seriously suggesting that Intel did not realise that Rambus uses transmission line techniques?> Judging from early Rambus publications and patent applications, I am seriously suggesting that Rambus had no clue about what they were dealing with. I believe that the only touch with reality was provided by real hardware guys from Intel, HP and Tektronix. Early waveforms paublished by Rambus clearly show that the combined probe and oscilloscope bandwidth was severely inadequate to observe intricacies of real signals, which may totally clogg the reality and mislead design engineers. Also, all original Rambus proposals were around 250MHz clock/500MHz data rate, with multiplexed address and data lines. Apparently this "solution" did not cope well with reality too, since later the bus was changed to separate commands and data. More, I also believe that the 800MHz specs were introduced because of competitive pressure from SDRAM technique, which created the whole new layer of problems with signal integrity fro Rambus. It took 10 years of Intel help for Rambus to arrive that transmission lines are all about impedance inhomogenieties, where every vias may be eventually a problem. So much for a robust and stable digital solution for mass produced products. BTW, is there a Rambus research center in England, or what? Just curious... - Ali