Re: The case the memory manufacturers have brought against Rambus is largely the result of an industry that is resisting a long-overdue abandonment of an outdated model built upon cross licensing. [from fred hager]
It's hard to believe that RAMBus is genuinely arguing for the elimination of the idea of shared, cross-corporate research in semiconductors, as well as pushing the future of the industry as a few ip houses that come up with semiconductor designs and a few fab companies that produce these companies' designs. Yet, Tate baldly stated this at the conference call. I suggest people listen to the conference call: It basically consolidates RAMBus's big-picture arguments in an extremely naked, unnuanced way. The contradictions just scream at you. Skip the first few minutes, which are fluff. At 14:40, there is the funny party of the conference call-- the one minute of no questions, followed by exactly one flunkee questioner, who goes on for about four minutes. Some of the highlights:
"The revolutionary aspect of our company and its success has resulted in a great deal of envy among some companies locked into older business models in the highly competitive, commodity DRAM market... [who are fighting us] so that they would not have to change the way they do business. ...[it is clear that there is] a group effort by a few of these companies to defeat RAMBus, and a fear of RAMBus, so that they would not have to change the way they do business. They will not be successful in these efforts." He then talks about these the success of RDRAM.
"RAMBus taught the DRAM memory industry a wide range of inventions including burst transfers, timing compensation using delay lock loops, low swing signaling, transfer on both edges of the clock, source synchronous clocking, and much, much more."
Here is an industry representing what I consider to be the pinnacle of human engineering progress-- 50 years of sustained, monumental improvement, and RAMBus says the business model and means of engineering is "outdated" and needs to be replaced, all the while offering a chief ware that has not been empirically shown to accelerate or even maintain the existing performance curve.
-Eric |
