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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: grok who wrote (67225)	8/1/1999 12:57:00 PM
From: Charles R	Read Replies (2) \| Respond to of 1573864

KZ, <I expect that rambus usage will grow pretty rapidly and reach penetration levels of 30-50% in a few years. > Now, technologically I like what Rambus has to offer but there seems to be an assumption that the volumes are certain or near-certain. I wonder if that is a true assumption. Would you have this view, for example, if Timna were not there or if Timna were not a success? It could be argued that Timna could be squeezed between Playstation II and basic PC (with Celeron or Coppermine-128 or Cyrix or National or AMD low-end offerings) and will not EVER become a volume product. Timna may be too old a core to provide some of the gaming performance that would be necessary for this class of devices. So, I personally would not bet on Timna's overwhelming success. Possible but far from a sure thing. Even on the much touted graphics advantages, I can see graphics/chipset guys integrating memory at 1-T SRAMs or as outside DRAM modules or straight DRAM integration like some people have done to avoid Rambus premiums. Why put expensive memory for the entire PC if one can integrate a much cheaper version with the graphics controller? In high-tech industry, being near certain about anything that is a year away, let alone years away, is uncomfortably risky for me. Any thoughts are appreciated because I would love to hear this angle debated. Chuck

To: grok who wrote (67225)	8/1/1999 1:16:00 PM
From: Bill Jackson	Read Replies (1) \| Respond to of 1573864

KZNerd, Rambus will succeed if it works and the cost premium is not excessive. From what I read Rambus costs 3-4 times as much as what is currently the volume leader. True those prices have been beaten down by cutthroat competition so it is an artificial price point and we should not expect Rambus to be that cheap, but 3-4 times as much?, that will keep it a niche player, used in a few critical places. Are Rambus parts hard to make?, hard to test?,slow to test? low in yield? The ansers to these questions will tell the tale on Rambus. From what I read they are in fact hard and slow to test. They should be as makeable as common memory and the yield should also be comparable...and of course the Rambus fee on top of all this, and the smaller number of makers sho have signed on to make them. It would not surprise me to find standard memory closing the speed gap to a degree and keeping the majority of lower end systems, and that lower end will migrate upwards in performance as we have seen over the years, relegating Rambus to a permanent high end profitable niche. Look at APple, a niche player and profitable(now). Bill