Apple sued by University of Wisconsin over A7 chip at heart of iPhone, iPad
The Wisconsin Alumni Research Foundation, the patent licensing arm of the University of Wisconsin, alleges in a newly-filed lawsuit that Apple's A7 processor infringes a university-developed patent that improves "the efficiency and performance of contemporary computer processors."
At issue is United States Patent No. 5,781,752 entitled "Table Based Data Speculation Circuit for Parallel Processing Computer." Computer scientists Andreas Moshovos, Scott Breach, Terani Vijaykumar, and Gurindar Sohi invented the technology while at the University of Wisconsin -?Madison.
The patent's abstract details a process for allowing quicker execution of processor instructions:
Apple is accused of implementing the patent's technology in the company's new 64-bit A7 processor that powers the flagship iPhone 5s, iPad Air, and iPad mini with Retina display. The complaint alleges that Apple was aware of the patent's existence because it is cited in several newer patents issued to the Cupertino, Calif. company.
The lawsuit was necessary, the foundation alleges, because "Apple has stated that it is the policy of the company not to accept or consider proposals regarding licensing from outside entities like WARF for any purpose."
At issue is United States Patent No. 5,781,752 entitled "Table Based Data Speculation Circuit for Parallel Processing Computer." Computer scientists Andreas Moshovos, Scott Breach, Terani Vijaykumar, and Gurindar Sohi invented the technology while at the University of Wisconsin -?Madison.
The patent's abstract details a process for allowing quicker execution of processor instructions:
According to the foundation's complaint, the researchers' work "has been recognized as a major milestone in the field of computer microprocessor architecture/design." Sohi, the group's leader, was given the Eckert-Mauchly Award --?"the computer architecture community's most prestigious award" -- for the work contained in the patent, the filing says.A predictor circuit permits advanced execution of instructions depending for their data on previous instructions by predicting such dependencies based on previous mis-speculations detected at the final stages of processing.
Apple is accused of implementing the patent's technology in the company's new 64-bit A7 processor that powers the flagship iPhone 5s, iPad Air, and iPad mini with Retina display. The complaint alleges that Apple was aware of the patent's existence because it is cited in several newer patents issued to the Cupertino, Calif. company.
The lawsuit was necessary, the foundation alleges, because "Apple has stated that it is the policy of the company not to accept or consider proposals regarding licensing from outside entities like WARF for any purpose."
The foundation is seeking an injunction preventing Apple from continuing to use the technology without a license as well as monetary damages --?including interest --?and legal fees.WARF is informed and believes, and on this basis alleges, that Apple has incorporated the technology of WARF's '752 patent into the A7 processor to achieve enhanced efficiency and performance. WARF now asks this Court to prevent Apple's unauthorized use of the '752 patent.
Comments
Is there any chance that there was parallel development of this technology?
Is there any chance that there was parallel development of this technology?
It wouldn't matter. If the University patented it first under those circumstances, they win. Apple either needs to either:
1. Show they didn't use the patent; or
2. Have the patent invalidated through prior art, non-uniqueness, etc.
This will be another one of those trails that in the end you really won't know who is right. This especially the case if tried by jury as there is no assurance of technical credibility in a jury trial.
Beyond all of that I would imagine that there was an attempt at an agreement before this resulted in a legal action. If so Apple has already read the patent and declined to license the technology. In other words Apple believes it has a strong case that will stand up in court even with a jury trial.
One more thing, this could be a fishing expedition by the professors involved to expose more information about Apple A7 series. It will be interesting to see what Apple and the courts do to protect Apples IP in this case.
Is there any chance that there was parallel development of this technology?
It wouldn't matter. Obviously the devil is in the details about whether 2 implementations are the same, but if you and I invent the same thing (in secret, so it's not in the public domain), and you patent it, then I'm out of luck. Fortunately or unfortunately, "copying" is not required to prove patent infringement.
Alternatively, if we invent different methods for doing the same thing, then there is no conflict.
There's a reason why intellectual property law is the most lucrative specialty.
Of course there is a chance. Just like there is a chance of snow today. That doesn't mean a lot though.
from Patently Apple: "Since the issuance of the '752 patent, Apple has filed one or more patent applications that cite the '752 patent as relevant prior art."
I haven't read the patent but every patent cites other patents. I always wonder if these citings mean they aren't using the technique or are. As for parallel processing, that's been going on for a very long time and I'm getting tired of patents of processes like this that give everything to the one person forever. What have they done lately to improve on it, maybe nothing.
This is another possibility, P.A. Semi was in the low power world for a very very long time before being acquired by Apple. Who knows this patent might be based on stolen P.A. Semi tech. It wouldn't be the first time.
This patent is from 1996.
P.A. Semi was founded 2003.
Add to it that I just bought an iPhone 5s!
Look at that. Another patent holder waiting 18 years before suing “infringers”.
Their patent cites 24 different Apple patents & 1 PA Semi patent, not to mention dozens more. These Apple patents they reference were published as late as December 2013.
a publicly funded university which ostensibly used public monies to develop a technology. seems like it ought to be available to anyone to use ...
Universities regularly profit by licensing technologies their employees created. I think Stanford alone must get a significant part of their money this way. I know UW is a public institution, but they also receive private grants for research so it's not so clear cut. I agree in principle that research should be publicly available for no cost, but it is what it is.
Look at that. Another patent holder waiting 18 years before suing “infringers”.
I have no idea about the merits of UW's patent, but I don't think they can be labeled as "waiting 18 years" when the A7 just came out a few months ago.
Once again, the whole damn system of copyrights and patents needs to be utterly wiped away and started over from scratch. There's zero invention any more because everything has been patented, including things no one actually invented, and the few entities actually utilizing the ideas are sued by everyone who claims to have patented the ideas... And half of the filings are duplications of other ideas already filed and so much of it doesn't even make sense to patent. And copyright is way too long (thank Disney).
But so long as the lawyers are making millions in this business, it's not going to change. The only people benefitting from patent law these days are the wealthy, not the small inventor.
Interesting. Would this apply to Qualcomm, NVIDIA, and other licensees of ARM, or do they either not use the technology or have licenses? So far, the mainstream press hasn't picked up on this, so there's not much analysis out there.
How detailed are the patents when it comes to execution? Could it be that the description of what is being done simply sounds similar but when you get into the details there is only a superficial resemblance?
For example we each describe a means of travel based on a mechanical device using rotating parts to transfer torque created by the engine through a shaft such that the force generated is directed into vehicle motion and you build a car but I build a helicopter then clearly despite any similarities in transfer of torque we have not infringed on each other's inventions. (I know that is likely far to general and vague but it is just an example). Whereas if we each describe a method by which a linear motion of piston is connected to a crankshaft then we are getting specific enough that there could be infringement. Of course if there are a multitude of ways in which such a motor can be constructed then perhaps not.
Perhaps it is just a symptom of only seeing the headlines and high level details but it seems to me that many of these patent disputes are over things that are so broadly defined that a great deal of interpretation is REQUIRED to reach any sort of determination. It should be that if you can't explain it to a group of 4th graders and have them be able to tell the competing claims apart then you have infringement - but if the 4th graders can clearly distinguish between the competing claims then no infringement.
I wonder if anyone has collected data on how much money lawyers make off this sort of dispute - talking about long term all cases - not specific cases - or if any correlation at all between how much each side in a case spends on lawyers versus the outcome.
http://arstechnica.com/tech-policy/2008/02/intel-badgered-by-u-of-wisconsin-speculation-circuit-patent/
This patent is from 1996.
P.A. Semi was founded 2003.
How long to patents last? 17 years sounds familiar. Coincidentally 17 years from 1996 was 2013.
Update: the duration was upped from 17 years to 20 years back in 1995, and this patent was granted in 1998, so it still plenty of time left on it.
http://www.uspto.gov/inventors/patents.jsp#heading-5
In a sentence, they sue the A7 because it is so fast ?
No doubt that UW had already approached Apple about licensing the patent before filing suit. Seeing as they were unable to come to terms Apple may either be unwilling to pay the asking price, or may feel the patent isn't valid and therefor refuses to license it. A lot of larger companies appear to ignore royalty/licensing demands, preferring to settle it in a courtroom. I suppose it's cheaper overall?
Just toss this in with the hundreds of other IP claims made against big techs. Nothing new here. IMO Apple will probably settle for some dollar amount before it goes to trial.
EDIT: I noticed in the complaint that "Apple has stated that it is the policy of the company not to accept or consider proposals regarding licensing from outside entities like WARF for any purpose, making the initiation of this lawsuit a necessity."