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: MileHigh who wrote (19927)	5/11/1999 2:32:00 PM
From: Tom Warren	Read Replies (1) \| Respond to of 93625

From Electronic News--May 3, 1999 Tilting at Windmills Jonathan Cassell "To dream the impossible dream To fight the unbeatable foe..." "The Impossible Dream" From "Man of La Mancha" Two weeks ago in this section, I wrote a column describing how money and power can triumph over technological merit. This week, Electronic News is running the first in a series of articles that present technological issues related to Rambus. Electronics News' decision to present these issues raises the question: if money and power always triumph, what is the purpose of discussing the technological merits and shortcomings of Rambus? In the fight to control the next-generation of DRAM technology, the money and power clearly is in Rambus' corner. Rambus' chief backer is Intel, which reigns supreme in the world of PC hardware. Because of this, the fight led by those who oppose Rambus has about as much chance of success as Don Quixote did when he tilted against windmills. But while Don Quixote failed in his fight against the windmill, he achieved a victory on a much higher level: a triumph of the will and a vindication of the values he upheld. Don Quixote's victory lay not in the outcome of the fight, but in the nobility and righteousness of the struggle. By opening up the Rambus debate, Electronic News is aiming for a similar higher-level victory: to inform interested parties about the issues they face and the choices they really have as they enter the Rambus era. With Rambus having silenced such discussion by its licensees and by the press as much as it is able to, Electronic News has become the only forum where a desperately needed debate and discussion of the issues can take place. One of the chief issues in the Rambus debate is cost. Little hard information has been available on what the real cost premium of Direct Rambus DRAM (D-RDRAM) will be relative to current SDRAM. The D-RDRAM cost penalty is as high as 60 percent compared to PC-100 SDRAM, according to an analysis performed by Bill Gervasi, vice-chairman of the JEDEC memory timing subcommittee. Another issue is performance. When making a comparison between single channel D-RDRAM 800 and 266 megahertz DDR, Rambus trails 266MHz DDR in the key parameters of latency and peak bandwidth, according to Gervasi's analysis. Still another issue comes in the area of board layout and design. D-RDRAM pushes memory clock speeds to 400MHz, a massive increase compared to the 100MHz clocks used today. To accommodate such speeds, PC boards will have to comply with much tighter specifications for impedence and will face increased problems with EMI, Gervasi says. Rambus disputes Gervasi's analysis. But if he's correct, the upshot of all these factors--cost, performance, board layout and others--may slow acceptance of Rambus in desktops. Due to cost and board layout issues, D-RDRAM could be relegated to a high-end niche longer than Rambus anticipates, decreasing volume and excluding the technology from the fastest growing segment of the PC market. Due to its relative performance shortcomings, D-RDRAM may not turn out to be the best memory for graphics, an area where the technology is supposed to shine. Because of these issues, Rambus may not reach its ambitious prediction that it would take a majority of the market by 2001. If the problems turn out to be show-stoppers for the technology, Intel potentially could reduce or end its support for D-RDRAM, taking away Rambus' ace in the hole. This would put Rambus on a more equal footing with other next-generation DRAMs. OEMs and chipmakers should be aware of these possibilities and plan accordingly. Alternatives, such as DDR, should be given serious consideration. Other sources of supply for DRAMs, modules, core logic and even microprocessors may be required. However, OEMs and chipmakers could not come to this conclusion and take this course of action without a frank and open discussion of the technical issues of D-RDRAM and of other next-generation DRAM technologies. This is a discussion that has not taken place in a public forum until now. This is the ideal that Electronic News is aspiring in its coverage of D-RDRAM: To uphold the value of openness, to champion freedom of choice, and to affirm the value of discussion and debate. Jonathan Cassell is executive editor for Electronic News. He can be reached at jcassell@cahners.com