Wisconsin-Madison sues Intel for patent infringement

Correction, 12:50 p.m. PST: This blog initially misstated the name of the group that filed suit against Intel. The group is called the Wisconsin Alumni Research Foundation.

Forget the patent troll. Bring on the patent badger!

The Wisconsin Alumni Research Foundation sued Intel on Wednesday for patent infringement, charging that the Core 2 Duo infringes on a patent granted to University of Wisconsin at Madison researchers in 1998 for a processor design that can break instructions into separate strands for more efficient processing. (Thanks, Engadget.)

Bucky Badger's used to shaking up the Big Ten, but now he's going after Intel.

(Credit: University of Wisconsin-Madison)

The patent covers a method in which certain instructions that would normally have to wait for other instructions to finish processing before they can move forward can be processed in part while waiting for the other tasks to finish. It's like if you're waiting for someone to finish a report because you need their data to complete your own, but you go ahead and get started because you have a pretty good idea of what their data is going to conclude. It's a bit more complicated than that, of course, but prediction works very well in modern microprocessing, especially in processors with multiple cores.

In a press release, WARF said it tried to get Intel to license the technology back in 2001, but the company wasn't interested. Intel has also "aggressively marketed" this type of technology with its Core 2 Duo promotions, according to WARF.

I think that's referring to what Intel calls Intel Wide Dynamic Execution (click for PDF), which Intel says "improves execution speed and efficiency, delivering more instructions per clock cycle. Each core can complete up to four full instructions simultaneously."

An Intel representative did not immediately return a call seeking comment on the suit, which was filed in the U.S. District Court for the Western District of Wisconsin.

Wisconsin-Madison claims the fierce Bucky Badger as its mascot.

[dascha1]

Trademarks Used Illegally by Universities

You know, it's interesting how often Schools, Universities and <br />Institutions use US-protected Trademarks, that in many instances <br />are Registered and Property of private companies. Apparently they <br />seem to see value from it somehow. Even without paying, or at the <br />least, asking permission to use it on their web sites, published <br />papers, their 'own' research projects from the rightful owner.<br /><br />Can't have it both ways, now can we?

[binaryspiral75]

Same laws apply.

It's called plagiarism and it's a serious offense in any school. <br /><br />I don't think your argument is valid. Simply because some people plagiarize homework or research papers that Intel has no responsibility to license technology developed by a university.

[henry999]

WARF

WARF is the Wisconsin Alumni Research _Foundation_, not 'Fund'.<br /><br />It's main claim to fame is Warfarin, a blood thinner for heart <br />patients also used in rat poison. It's made them a fortune over the <br />years.

[Tom Krazit]

Thanks

We'll tweak that.

[jwfowble]

Vitamin D $$

Maybe the name Warfarin/coumadin is the most recognizable but UV light to do Vitamin D is where the money is. Does Steenbock ring any bells UW alums?<br /><br />Go Badgers!<br /><br /><a class="jive-link-external" href="http://en.wikipedia.org/wiki/Wisconsin_Alumni_Research_Foundation" target="_newWindow">http://en.wikipedia.org/wiki/Wisconsin_Alumni_Research_Foundation</a>

[TomboSlicko]

Intel should pay

Intel should give the university 10% all the profits if they infringed.

[ittesi259]

Problem with Patents

This what I hate about patents and wish the laws were more like trademarks as far as enforcing your patent. With a trademark you have to aggressivly fight an infringer or risk losing the trademark. However with this patent (which should have been called out when Intel first released Core2, not years after) we're seeing nothing but pure greed. Patent owners should be required to pull out their patents at the first suggestion of infringement, not just waiting to sue the people who make money. Don't tell me the researchers didn't pick up on this back when it was released....oh wait, I mean when the troll of a lawyer decided money could be made from it.

[juser_bogus]

check your facts before posting out your butt

"Falk said WARF contacted Intel in 2001 and made repeated attempts, including meeting face-to-face with company representatives, to offer legal licensing opportunities for the technology.<br /><br />In the suit, WARF claims that while its application for a patent on the technology was pending, one of its inventors met with Intel and offered to discuss licensing it for use in future Intel products. Instead, the company incorporated the patented technology into its products, including the Core 2 Duo processor, and has refused to obtain a license from WARF. The research foundation also claims that Intel has aggressively marketed the benefits of this invention as a feature of its Core 2 technology."<br /><br /><a class="jive-link-external" href="http://wistechnology.com/article.php?id=4526" target="_newWindow">http://wistechnology.com/article.php?id=4526</a>

[Flenser]

Intel licenses only under duress

The WARF story mirrors my experience with a university patent <br />infringment by Intel. Intel referenced the university patent in <br />later patents they filed covering various minor implementation <br />details deriving from the earlier university patent, and used the <br />university patent language in chip documentation. The <br />university tech transfer office was shocked by Intel's response to <br />licensing discussions ... basically six years of threats and <br />stalling. In the end the patent was purchased by a well-known <br />patent "front corporation" which licenses IP to its industry <br />investors (many big names, including Intel) without royalties. <br />Very disillusioning experience. I applaud WARF for taking on a <br />very tough and unscrupulous legal department; most academic <br />entities don't have the bankroll or fear the collateral fallout of a <br />straight-up patent lawsuit.

[gabeheim]

Intel Wide Dynamic Execution...

has been a feature of Intel CPU's since P6 (pentium pro through pIII and pentium M and predecessor to core architecture), released in 1995. So I doubt that is what WARF is suing them for. If it is, then intel has a lot of prior art, including IBM's research in the 60's for out of order execution and speculative execution. Since this does sound similar to the WARF patent, perhaps they are suing them for what might be considered an obvious improvement on OOE and Spec Exec? Time (and lots of lawyers) will tell.

[paulej]

Highly suspect

Patents are a tricky thing. It is entirely possible that Intel did develop this technology before WARF. Maybe they did not. It all boils down to exactly what the steps are.<br /><br />Simply executing instructions in parallel that can be executed in parallel is not new-- I did those sorts of thing as an undergraduate years ago, though it was certainly simpler than what Intel is doing in its processors, I am sure. Nonetheless, parallel execution has been studied by lots of people over the years.<br /><br />The real question is whether Intel did exactly what the patent claims. I would certainly not want to see Intel pay for a patent that it does not infringe at every step of the way, given the millions of dollars the company spends on its own research on the topic.<br /><br />Is WARF somewhat of a patent troll? I'd hate to think that a school would be a patent troll, but it certainly looks like it in this case. But, who knows?

[juser_bogus]

Explain.

it looks to me that you are being the troll by claiming "Highly suspect" or that it's possible Intel developed the tech without any evidence that suggests either. Put your money where your mouth is! It's clear from the available documents on the internet that WARF owns the patent and tried to negotiate in good faith for 6 years. <br /><br />here let me try some unfounded crazy claims.... it's entirely possible that ALL of Intel's technology was gathered unscrupulously just as this may have been.

[dsarokin]

Some nice patent searching

There's an awfully nice piece of patent searching detective work at this Google Answers site:<br /><br /><a class="jive-link-external" href="http://answers.google.com/answers/threadview?id=450081" target="_newWindow">http://answers.google.com/answers/threadview?id=450081</a><br /><a href="http://answers.google.com/answers/threadview?id=450081">Finding a patent by date</a><br /><br />The researcher tracks down an old patent with a minimum of information. I thought folks would appreciate being able to check it out.<br /><br />David