Hi wstera 02; Re tinkershaw's comments on prior art at the Infineon trial. Infineon did put a lot of prior art exhibits into evidence. But this was for the purpose of invalidating the Rambus patents, (as opposed to the purpose of showing that Infineon didn't infringe), and since they already had Judge Payne rule for prima facie fraud, and that was upheld by a 3 judge appeals court, they concentrated instead on the fraud case. (It also had the effect of adding a worse taint to Rambus than a simple invalidation.)
Prior art was not "put forward at trial", but it was put forward at evidence. The way I understand it, the rocket docket doesn't allow the lawyers to spend 9 months presenting a case in all its multitudinous aspects, like the OJ case went. If both sides had presented all the evidence they had, the case would be in trial for months. Here's prior art on using a DLL for synchronization of clocks:
A variable delay line PLL for CPUcoprocessor synchronization
citeseer.nj.nec.com
Re: "At trial it came out this was because Infineon largely lacked any in-house IP regarding the creation of SDRAM, they were desperate for the help."
Infineon didn't invent SDRAM, nor did they claim that they did. All they did was to implement the JEDEC published standard, and that was simple and obvious. In order for JEDEC to prove that they, for instance, knew about DLL, they would have to show that any one of the couple dozen companies present at JEDEC meetings had read the above linked article (which dates to 1988). Since it's an IEEE published paper, one would guess that lots of engineers at every one of those companies would have read it, but it's a sure bet that some employee of at least one of them was aware of it.
I should note that the idea of synchronizing chips using DLLs is a very general idea. It encompasses the case of using a DLL to synchronize between DRAM chips, and consequently, after the general idea is out, it is not possible to patent the specific idea. In other words, the IP is already either patented or in the public domain, so you can't patent it.
This is why Rambus had to add the "multiplexed bus" to all their claims, which is what excluded them from SDRAM. If they'd tried to patent the idea of using DLLs to synchronize memory, they'd have lost the case because it was already known how to synchronize memory chips (or any chips of any type at all), simply use DLLs.
Do you understand why it is that a general idea prevents a specific "instantiation" of that idea from being patented? If the specific idea was patentable, then the holder of the general patent will lose that part of his intellectual property rights. Since all the synchronizing uses of a DLL are specific, (i.e. to synchronize between the CPU and an FPU, for instance), there wouldn't be any actual use that the owner of the patent for "DLL chip synchronizing" would be able to build.
In the rest of your post you talk about how Rambus' patents were unfairly restricted to the specification or the preferred embodiment. Why don't you use the same logic when looking at the legal position of the person who holds the patent for using a DLL for synchronization? Maybe there is no such person. My belief is that there very likely is a patent for using DLLs to synchronize chips, but that it is so old that it ran out long before 1988. On the other hand, it is possible that the idea is so obvious that it has always been public domain. In either case, for Rambus to claim that IP, it has to be taken from somebody who already had it.
Re: "it is well established that the scope of a patent cannot be limited either by the "preferred embodiment" of the patent" I believe this is true, but this is not: "nor can it be additinally limited from making inferences from the specifications of the patent." Read the judge's memorandum for the details.
The basic problem here is that in order for Rambus to get DDR royalties, they have to reverse both the fraud verdict and the Markman ruling.
Re: " In regard to the fraud element I'm still a bit perplexed with the decision the jury came to. I honestly don't exactly know what they based their decision on."
The rest of the world seems to have no problem understanding the fraud element. Judge Payne, the 3 judge appeals court and the 7 jurors in particular. In addition to that, I didn't have any difficulty, nor did the trade press. But since you don't believe that prior art exists, I guess you don't see it that way. This is normal human behavior. It is easier to see someone else's sins than one's own. Some people are so blind that in addition to being blind to their own crimes, they are unable to see that of their friends. Get used to it.
-- Carl |
