LONDON – A Pittsburgh jury has found that chip company Marvell infringed two disk drive patents owned by Carnegie Mellon University and awarded damages totaling $1.169 billion. Marvell denied it infringed the patents and said it will appeal the decision.
The jury award against Marvell is reportedly the third largest ever in a patent case. Since the infringement was ruled by a judge to be "willful," the jury had the option of awarding up to three times the original damage amount.
The jury verdict was delivered on Dec. 26, 2012, as part of a patent infringement lawsuit brought by Carnegie Mellon University against Marvell Technology Group and subsidiary Marvell Semiconductor Inc. U.S. Patent Nos. 6,201,839 and 6,438,180 and the claims of infringement extend back to 2003.
Marvell said in a statement that the company expects no disruption to its business as a result of the ruling and that it would be challenge the judgment through "all appropriate post-trial motions and appeal processes." The Pittsburgh district court has yet to rule on Marvell's defenses and motions, including a request for a mistrial.
Marvell said the patents protect a technique related to read-channel signal detection that is not used in any Marvell chips. In addition, Marvell asserted that the patents protect a theoretical technique that cannot be implemented in silicon, even using the most advanced techniques available now or as far back as 2003. Marvell acknowledged that the jury disagreed with Marvell's position.
In a statement, Pittsburgh-based Carnegie Mellon University said: "We felt the evidence we submitted was compelling, and the jury agreed." The university said the disputed patents protected systems and methods developed by Jose Moura, an engineering professor, and Aleksandar Kavcic, a former PhD student who is now a engineering professor at the University of Hawaii.
The work that led to the inventions patented by Moura and Kavcic was supported by the university's Data Storage Systems Center.
Marvell has yet to set aside any liability in its financial statement as a result of the jury award, the company said.
If Marvell simply responded to reasonable request by CMU to pay a justified, reasonable royalties of $200K, this would not have happened. They pay the price of arrogance! Industry beware of stealing university property!
It isn't the dollars per patent that should guide the damages, it is the losses suffered by the inventor and the undeserved profits earned by the infringer. What volume of sales did Marvel make using the CMU intellectual property? I'd think this would be a central part of the lawsuit and the basis upon which the award would be made.
As companies spend their resources prosecuting and defending patent lawsuits, or paying the Danegeld (tribute) to avoid lawsuits, they have less and less money to fund engineering. As the engineering jobs vanish, there won't be any need for more STEM students. Problem solved!
Evenif Marvel infringed the patents, 1.2B for 2 semi patents is too high. The Pittsburgh jury may be smoking crack before the verdict. The numbers are totally unheard of in similar cases,even if you add a reasonable onetime payment and royalties for all the chips Marvel sold with this. Oh wait, if jury can manage to get CMU a free 1.2B, they can provide free education and scholarships to their family, friends and relatives! Why cant patent fights be in neutral locations instead of someone's homeground?
You didn't count the negative impact to the Marvell employees, including those CMU graduates hired by the Marvell. Patent war is a zero-sum game. It wont create wealth, it just redistribute the wealth. The settlement of the lawsuit very likely will be in the unit of M, not B.
Agree, this certainly calls for an expert to examine the claims. I find it rather unusual to get a patent on an obscure theoretical idea if that is what really happened (as Marvell claims). It would be great to hear from the two professors comment on this BUT they may have a gag order from the university.
I second the motion that we get a little more info on the tech issues.
BTW: If the $1.2B goes to the endowment and starts earning at 3%, it would yield $36M/year.
This would fund 720 $50K full-ride scholarships.
So, about 180 new frosh a year could be on full ride. Might ripple through and change the entire competitive college landscape. Given the country's desperate need for STEM students, wouldn't, couldn't, shouldn't, this be nice?
Might take 10 years, but is likely to be 100x as important as Solyndra, etc for our future.
After reading the article I am not sure what really happened. If I read Marvell's position in the article correctly, the disputed patent was not technically feasible nor used. If I understand (reading into the ruling) the judgement against Marvell they both used the patent and implemented it. I would have very much liked some investigative reporting showing at least a high level of the disputed technology/patent details.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.