SI
SI
discoversearch

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: NightOwl who wrote (68838)3/22/2001 1:31:49 AM
From: Dave B  Respond to of 93625
 
Good Evening, NightOwl,

I hope you're doing well. That was quite a long post, so I'll address just a couple of what I view as the key issues/assumptions of your message.

In this case you seem to place great weight on the notion that Samsung was contacted in 1992 about "SDRAM royalties," and that this was widely known, at least by JEDEC members.

I assume you are saying that what RMBS had sought, in 1992, were royalties on its 1992 and prior issued patents since those were the only patents it had at the time. And I further assume that what you and the IBM guy are saying is that RMBS was seeking royalties under those 1992 and prior patents against a IC product that Samsung was making under the SDRAM terminology.

In my view this simply means that RMBS has always "interpreted" its 1992 and prior patents as a very broad based IP. It is also apparent that many members of the DRAMURAI disagreed with this "interpretation."


My only purpose in bringing this up was to address the claim by many of the bears that Rambus did not think to try to get royalties for SDRAM until last year after it was clear that the RDRAM ramp was not going to be aggressive. The fact that in 1992 the JEDEC folks already knew that Rambus was saying that the SDRAM spec would infringe on the patents that Rambus had filed indicates that this is not the correct assumption. Note that I am not speaking to the validity of the patents one way or another, or whether or not they cover SDRAM; only that this was not an issue that suddenly appeared in 2000 as the bears claim -- JEDEC members knew in 1992 that Rambus might try to assert their claims.

It appears from the known facts that RMBS itself did not believe its 1992 and prior patents covered the initial iterations of SDRAM sufficient to enforce claims of royalties. It certainly did not act on those claims. JEDEC members had every right to disagree with RMBS' asserted coverage of "SDRAM" IP for its 1992 and prior patents.

Since the gentleman from IBM who was heading the JEDEC committee said that Rambus was considering strying to license Samsung for $10M for SDRAM, I'm not sure why you believe this.

As has been explained by Zeev (and I'm sure he can communicate it more effectively than I), Rambus was not able to push their claims for SDRAM royalties until the patent had actually been granted in 1999. They might have wanted to try to get SDRAM royalties earlier, but until a patent was in hand, why would anyone pay them? If someone approached me with that story, I'd simply ask them to come back after the Patent Office had indeed agreed that they had a valid patent (and I'd be happy to then pay them a reasonable price if I hadn't designed their patent out of my product by then), but until then don't bother knocking.

And perhaps most importantly, IFX is effectively claiming that SDRAM and DDR are free and clear of coverage by the 1992 and prior patents issued to RMBS and that since those concepts were "invented" through the JEDEC process [with or without RMBS' voluntary assistance] prior to the issuance and JEDEC knowledge of the subsequently issued RMBS patents, no royalties can be imposed.

Again, I'm sure Zeev can address this more effectively, but I'll bet that if I "invent" an idea that's already been patented, even without knowledge of the existing product, I would still have to pay royalties on the product I build to the patent holder. In fact, I'm sure of it. That's what patents are for; otherwise everyone would be claiming to have invented their idea independently, therefore they shouldn't have to pay royalties. As has also been discussed, the difference here is that you can't ask for royalties back to 1992 -- you have to start them from a reasonable time frame (after the date the patent has been granted as well as no earlier than the date that you first contacted them to inform them of the infringement, I would assume). That's why the announcements that are coming on the new licenses that begin from April of last year, or October -- all timeframes that begin after the patent was granted.

At this point it will be up to a jury to decide if the SDRAM and DDR specs infringe on Rambus patents. And if so, whether or not it was willful.

Good luck,

Dave



To: NightOwl who wrote (68838)3/22/2001 4:40:41 AM
From: John Walliker  Read Replies (1) | Respond to of 93625
 
Nightowl,

It is normal for NDAs to contain a clause requiring the party to whom the disclosure is being made to notify the discloser of any aspect of the information which is already known to the receiving party.

John