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: John Walliker who wrote (81680)	3/24/2002 11:00:44 PM
From: Ali Chen	Read Replies (1) \| Respond to of 93625

"claims that Rambus use "current steering drivers" which were copied from SCI. Actually they use "constant current drivers" which are something very different." Could you spell the difference please? - Ali

To: John Walliker who wrote (81680)	3/25/2002 8:00:22 PM
From: cordob	Read Replies (1) \| Respond to of 93625

I think the reference is to the software/firmware protocol involved in SCI (SCISI) The investor_in_red message you quoted claims that Rambus use "current steering drivers" which were copied from SCI. Actually they use "constant current drivers" which are something very different. Looking at the protocol used in rdram it is nothing very special, which was not used at least 10 years before the patent date, but I always understood that a different application of that to the memory chip environment was a valid "newness thing" for a patent. Cheers Cor

To: John Walliker who wrote (81680)	3/29/2002 8:59:57 PM
From: SBHX	Respond to of 93625

John, That post was interesting to me from at least a 1st order analysis in that it matched my initial thoughts when I first heard about the rambus patents a long time ago. It seemed to me that I heard these ideas long before and was surprised that Farmwald and Horowitz could squeeze a few patents based on this. But as we know they are both honorable men --- smart and honorable men <g> and I shouldn't even ask these questions. However, not having read the patents themselves, other than generalities, I didn't think I was qualified to comment either way. As far as I know, with the patent filing process, if the inventors knew at any point that there was any prior art that directly addresses the novel ideas of the patent, then they have a responsibility to abandon or modify the patent in such a way that there is still novelty in that. However, if there was any question of similarities, and if the patent filers knew about such an item then they had to add that as a reference in the patent filing itself. Anyone who could prove that the filers (horowitz, farmwald et al) knew of the SCI ideas and intentionally did not add them during the patent filing has a strong case of making those patents questionable. But while I would be surprised if it was that simple, it also seems this is almost impossible to prove. All that Farmwald and Horowitz have to do is take the stand and swear under oath that they did know of the SCI ideas or they didn't think they are relevant to the rambus patents and it would be hard to proceed without someone contradicting this. However, if they modified the patents specifically because of something they knew in the SCI idea, then the SCI idea has to be referenced as well or the patents immediately become questionable. Why doesn't anyone pursue this? This is where the patent filing process gets interesting. Patent clerks/examiners in the USPTO are not infallible. In many cases, if the patent clerks catch a potential prior art, an office action is issued and the patent filers have to deal with them. Sometimes claims get modified/dropped and the filing continues. At other times, the entire application is abandoned and closed. It is very hard to challenge validity of a patent once granted, but it now appears it is easier to do something else in the courts (as we saw with IFX). But if rmbs was as pure as the driven snow (who is?), then RMBS vs IFX would have been a slam dunk. I won't be surprised if many patents granted in the past may have prior art that the filers in good faith did not know about. I haven't looked. Meanwhile, something that continues to bug me is this question. Are Farmwald and Horowitz the Einstein and Hawkings of semiconductor physics? I don't really have an answer to that, however, we do have the following list of patents assigned to the various players in this drama. RMBS : 150 patft.uspto.gov IFX : 444 patft.uspto.gov MU : 6699 patft.uspto.gov Basically, the rambus patents have to be much more compelling than the IFX and MU to deserve a 2.5% royalty rate on DDR controllers and/or a 5% royalty rate on any ICs with a built-in SDR/DDR controller. If so, Farmwald and Horowitz certainly deserve a place among the giant intellects of this century. Horowitz seems pretty low key. But Farmwald has a neat web page : farmwald.com Farmwald is 1/2 of the pair that gave the world rambus and Rambus is Farmwald's greatest creation by far. Some of his other efforts while profitable (eg : chromatic research) had some questionable execution issues that were always an interesting topic of discussion in the valley. But there is no question is that this is a smart guy. Interestingly enough, if you looked at his favourite companies invested through skymoonventures.com RMBS is not in that list. Can't blame him, he already has tons of rmbs stock. And rmbs is already public. But it would be nice if his baby was on his list of favourite companies. It would add a touch of human drama to this. SbH