Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Intel Businesses Patents

Intel Told To Pay $2.18 Billion After Losing Texas Patent Trial (bloomberg.com) 101

Posted by msmash from the closer-look dept.
Intel was told to pay $2.18 billion after losing a patent-infringement trial over technology related to chip-making. From a report: Intel infringed two patents owned by closely held VLSI Technology, a federal jury in Waco, Texas, said. The jury found $1.5 billion for infringement of one patent and $675 million for infringement of the second. Intel, the world's largest chipmaker, denied infringing either of the patents and said one was invalid because it claimed to cover work done by Intel engineers, but the jury rejected those arguments. The patents had been owned by Dutch chipmaker NXP Semiconductors, which would get a cut of any damage award, Intel lawyer William Lee of WilmerHale told jurors in closing arguments Monday. VLSI, founded four years ago, has no products and its only potential revenue is this lawsuit, he said. VLSI "took two patents off the shelf that hadn't been used for 10 years and said, 'We'd like $2 billion,"' Lee told the jury. The "outrageous" demand by VLSI "would tax the true innovators."
This discussion has been archived. No new comments can be posted.

Intel Told To Pay $2.18 Billion After Losing Texas Patent Trial More

Intel Told To Pay $2.18 Billion After Losing Texas Patent Trial

Comments Filter:

  • Good (Score:5, Insightful)

    by Carewolf ( 581105 ) on Tuesday March 02, 2021 @04:13PM (#61117182) Homepage

    Now, please lobby to fix the stupid US patent system.

    • 10 years? There's your problem, right there.

    • But what would East Texas do without all that lucrative patent-troll money coming in? Won't someone think of the East Texans?

    • Re: (Score:2)

      by stikves ( 127823 )

      Ah... My poor child. They make much more from patents compared to occasional peanuts they lose on these lawsuits.

    • This isn't so much a problem with the patent system, it is Intel error and patent trolls. The patents themselves are fine. The ability to abuse them in this way is not.

  • The "VLSI Technologies" I remember was (before Philips/NXP bought them) the company that made most of the early ARM CPUs for Acorn, including the original ARM1.

    Is this new VLSI a shambling corpse wearing the mask of the old companies name; or are Intel's lawyers being economical with the truth, and this new VLSI is the remains of the original company (or close to it) which has been spun back off into a standalone company again?

    • I didn't think VLSI as a brand was still being used. But if Bloomberg's own links in the article are worth anything it seems this is the same VLSI that Phillips/NXP bought in the late 90's. Not sure what they actually do anymore. Is NXP just using them as a place to store their patents?

      • Re:Not the real VLSI then? (Score:5, Interesting)

        by sjames ( 1099 ) on Tuesday March 02, 2021 @04:47PM (#61117298) Homepage Journal

        It may be the same in name, but not in function or intent. Perhaps it has been gutted and remains only in name and lineage.

        But TFA also says VLSI Technology LLC (the plaintiff) was founded 4 years ago. It sounds like a spin-off in name only that is meant to safely go "bankrupt" with no actual assets should something go wrong with the patent troll game. The fact that it was given the 10 year old patents in 2019 bolsters that conclusion.

        It checks all the boxes. They have no actual production or development capability, few assets, and a treasure chest full of dusty patents they have no use for other than court.

        It's hard to come up with any sympathy for Intel given all their shenanigans over the years, but they may be right about the nature of their opponent.

        • TFA also said NXP gets a cut of the award. Koninklijke Philips acquired VLSI in the 90s and spun off NXP in the 00s. It looks very much like NXP spun off a company for the express purpose of fighting this patent battle.

          On the surface VLSI looks like a dictionary definition of a patent troll. Underneath it looks like a subsidiary of a very real high volume chip manufacturer.

          I honestly wonder why NXP attempted to make this look as dodgy as possible. They may have garnered quite a lot more sympathy and more fi

          • Re: (Score:2)

            by sjames ( 1099 )

            So NXP now owns a patent troll. Basically legit business wants to maintain an arms length relationship with shady practices. Kinda like when a respected doctor has an arms length relationship with a pill mill.

            • No it's stranger than that. NXP owned the patent and actively is using the patent. Calling this patent trolling is a bit strange considering NXP could instead have simply sued Intel and then it would be a perfectly normal patent disagreement so there's little reason to keep "arms length from a shady business" if your business isn't actually shady.

              Honestly I think it's a PR move. Intel killed VLSI in the 90s with Intel's shady practices. Maybe they were hoping by bringing back the name that they'd win some b

    • Re: (Score:3, Informative)

      by gabebear ( 251933 )
      It is the same company https://en.wikipedia.org/wiki/... [wikipedia.org]

      I have mixed feelings about this verdict since Intel muscled VLSI out of the PC market in the 90s using gangster monopoly tactics(along with a lot of other companies).

      • Re: (Score:1)

        by jabuzz ( 182671 )

        To be precise they used to make the chipsets that where used in PC's. Then Intel decided to muscle in on that market. They are a real company, they have real patents and Intel infringed them. Intel get to pay damages.

        • They were a real company, then they stopped existing as an independent entity after being acquired by NXP. This new VLSI is not the same company, does not have the same employees, is not holding the same patents, etc.

    • Re: (Score:3, Informative)

      by Anonymous Coward
      This is definitely not the same company. This is "VLSI Technology LLC" and currently has 21 patent infringement fights going on across five active cases. This is a patent troll banking on the confusion created by a using a similarly named company to one that used to make great products. "VLSI Technology LLC" is a front company for Fortress [lexology.com].

  • was where Intel's lawyers admitted that VLSI held the patent in question.

    Genius.

    • I'm not sure how you could not admit a fact. What you're going to disagree with the name on top of the patent filing? Most patent trolls hold the patents they are trolling over. That is very rarely contested in court.

    • Re: (Score:2)

      by Sebby ( 238625 )

      Not an "open", public company? Obviously the word 'private' should've been used instead.

      • Re: (Score:3)

        by bws111 ( 1216812 )

        No, the word 'private' should not have been used. A closely held company IS a public company, but with a very limited number of shareholders.

  • "Invention" (Score:5, Informative)

    by Anonymous Coward on Tuesday March 02, 2021 @04:51PM (#61117322)

    It's worth noting that the patent was issued to Freescale Semiconductor, that was bought by NXP Semiconductor, and NXP sold the patent off to the patent troll farm VLSI.

    That aside, the patent in question is US7247552B2
    "Integrated circuit having structural support for a flip-chip interconnect pad and method therefor"
    https://patents.google.com/patent/US7247552B2/en [google.com]

    For context, most people remember DIP chips (Dual inline pins)
    Flip chip is an IC package type very similar to ball grid arrays you find on modern CPU chips.
    The difference? One has solder balls applied to the chip, the other has the solder paste applied to the board. Then in both cases, you run it through an oven to adhere the solder.

    What is the patent about? Adding support legs to the bottom of the chip, so the balls of solder aren't squished out.

    That's the "invention" worth 2+ billion dollars. Adding legs.

    So in this case when Intel says the patent isn't valid because everyone in the world, including Intel prior to the patent being issued, uses this method, it's because it's true.
    Flip chip packages can't be soldered in any other way without supports since the end result literally will not function.

    Prior art of almost 10 years of independent solutions to the problem was rejected by the jury.

    Imagine trying to hammer two boards together with a nail just by holding them. It doesn't take a genius to notice that won't work, and perhaps you should put them down on the floor or something for support first.
    Only to find out "putting them on the floor for support" is a patented invention.

    • That's the "invention" worth 2+ billion dollars. Adding legs.

    • Re: (Score:1)

      by HiThere ( 15173 )

      Well, it's a Texas patent court, so I accept that it's wrong. OTOH, the infringer is Intel, so I accept that they actually infringed.

      OTOH, "legs on a chip" shouldn't be patentable. But why did the jury say it was? Perhaps we're missing something.

      • There must be millions of inventions that make up a modern chip. Of which thousands are patented by trolls.

        I am actually surprised that technology manages to function at all. In the early days of Software Patents we predicted doom. Yet it is merely awful, not unworkable.

        How is that possible?

      • The courts are selected precisely for being patent troll friendly, and Texas is a great old state full of people to stick on juries that don't care about no fancy electronic gizmos but are upset about giant corporations beating up on little guys that pay for a P.O. Box in their county. I can guarantee you that most jurors in Waco were predisposed to hate Intel before they were ever called up to serve. Never mind that literally every flip chip maker did this before the patent was filed, these jurors honest

    • I just read the patent.

      The actual innovation is not legs to support the chip. It's adding extra metal to the inner layers of the circuit board to make it more rigid.

      This is obvious, and I'm willing to bet there's prior art.

    • Re:"Invention" (Score:4, Informative)

      by Anonymous Coward on Tuesday March 02, 2021 @05:40PM (#61117498)

      Same poster here, bearing two corrections to my last post.

      1- Regarding why the flip chip support is needed, I was incorrect by saying the balls of solder get squished.
      2- Support legs are not the only solution to the problem

      I work at an electronics manufacturing company, and "we" do this exact type of SMD assembly.
      By "we", I mean not myself, I do IT.
      So I asked a process engineer here what they thought.

      For #1, it is BGA's that use balls of solder. The problem with BGA is only a small "cylinder" through the ball is needed to make electrical contact. But the ball is a ball, surface tension forms a sphere.
      So they have to account for extra clearance between balls, limiting how close each can be.

      Flip chip is the advancement. NO balls of solder involved. They use solder paste, applied through a stencil with cutouts where the connections are to be made, and a machine applies a drop of paste on the pad on the circuit board before another machine inserts the chip on top.

      For #2, it is still the same fundamental problem, but it is when the pick-and-place machine puts the chip in place, not during oven reflow.

      Our pick-and-place machine uses a tool tip and suction to pick up the chip and place it on the board.
      If it pushes down too hard before releasing the suction however, the same thing happens, the solder paste squishes out like a pancake, shorting with other pads near by.

      Support legs prevent that from happening in the first place, and is how our single head chip shooter machine works.
      Our dual head lightning chip shooter however can measure the force it is applying and the force feedback. It is apparently sensitive enough to tell if and by how much it is squishing the solder paste, and is able to stop before doing so too much.

      Additionally there are multi-stage reflow ovens available (that we don't have) that can control the temperature along the belt in 5+ zones. The solder paste is mixed specially to increase its melting point a few degrees, and the oven can be set to raise the temp to that amount in just one zone, and 30 seconds later in the next zone drop back down to normal.

      It is still decades old technology, just used more often for chips with hundreds to thousands of connections in order to keep the chips small physically.

      Intel just got unlucky by independently inventing the one particular method of using flip chip that a patent troll got their hands on a patent for...

    • That's why you never, ever play the lawyers' game. You kill them with heavy ordinance from orbit.

    • It's worth noting that the patent was issued to Freescale Semiconductor, that was bought by NXP Semiconductor, and NXP sold the patent off to the patent troll farm VLSI.

      Actually it's more complicated than that. Koninklijke Philips N.V. acquired VLSI which was a very real not patent trolling company in 1999. Philips spun of NXP Semiconductor along with all technology acquired from VLSI in 2006.

      NXP still exists and makes products. For some reason NXP chose to re-register VLSI Technologies and transfer them the patents in 2016 almost exclusively for this lawsuit. So while VLSI itself my very much look like a dictionary patent troll on the surface it serves a corporate master

      • The classic case on obviousness was the shaving cream can.

        In litigation to invalidate the patent as obvious, the judge commented to the effect that, "sure, now that you've seen it, it's obvious--but you spent millions unsuccessfully trying to figure it out before you saw it."

        hawk, an attorney, but not giving legal advice and not shaving, either

        • Not shaving (COVID beard) or not giving shaving advice to clients? ;-)

          • Re: (Score:2)

            by hawk ( 1151 )

            the last time I was fearless was a good thirty years ago, and only lasted a few days . . .

            But I give simple beard shaving advice to clients: don't!

            Never trust a man who runs a piece of sharp metal across his face before he fully wakes up in the morning . . .

  • ... a useful idiot. :D

  • Intel has to know the technology has been patented. Regardless if VLSI is a patent-troll or not, the fact remains they own that patent. It's their right to not use it in production. If Intel wants to use their patent, why wouldn't they approach VLSI and discuss a royalty program?

    • Ever read one? They are unitelligiible, and there are millions of them. Most of them are not relevant. And it is not possible to know how a Jury of mums and dads would interpret one in deciding it was or was not valid. And if you think a patent is invalid having found it, and a Jury disagrees, then triple damanges for wilful infringement. Oh, and there is no real way to challenge a patent, you just have to wait to be sued in which case you have everything to lose (this last one has changed a little bit

      • I don't agree with you call the jury "mums and dads". They are valid jurors selected by both sides of the trial. They have the right to expect that their findings in the trial is just as good as any other 12 jurors.

        • You are asking ordinary people, most of whom do not even have a STEM degree, to make subtle distinctions about a highly technical subject of VLSI manufacture.

          In practice, they judge who is wearing the best suite, which expert witness has the best voice, and they very much like the big ribbon attached to the official document blessed by the USPTO, regardless of whether that is valid.

          • ... they very much like the big ribbon attached to the official document blessed by the USPTO, regardless of whether that is valid.

            My ribbons are only 37 mm long. It's $25 to get a Presentation Patent copy; they don't even send one when the patent is issued. Apart from the ribbons, it's a lot like a kindergarten Best Naptime achievement seal. Assignments can be strange: my patents ended up assigned to Google, even though I have never worked for them, and have only been to an office of theirs once, for an unrelated assignment.

            Unless something has changed since my ancestry there, that part of Texas is full of Baptists. If an attor

      • Re: (Score:1)

        by neoRUR ( 674398 )

        It has always been on the ownness for the patent holder to enforce the patent. But if you are developing something then its up to you to be careful to not infringe in the first place.
        I dont like patents, but I understand their need. It was meant to give the person with the idea a leverage to making a product before someone else, or some large company, comes in and overruns you. It was supposed to expire, but new laws and companies have changed the system into their favor and messed it up for others.
        They ar

        • Re:Patents are unknowable (Score:5, Insightful)

          by aberglas ( 991072 ) on Tuesday March 02, 2021 @07:11PM (#61117800)

          You cannot be careful not to infringe one of the millions of patents out there in practice. There are too many, they are badly classified, and they are unintelligible even with a careful reading.

          And if you find a patent that looks like it might be relevant there is no way to test that in court before taking the risk of pursuing the technology. And if you avoid all such possible infringements you do nothing at all.

          In theory the patent system is supposed to encourage invention. In practice (certainly for software) the "Inventive Step" requirement is a very low bar. So developers independently invent the same fairly obvious solutions to the same problems. And the patent trolls hit.

          Patents are mainly about patenting problems rather than solutions. You think of a problem that might arise in a few years time, an then patent an obvious solution to it.

    • Re: (Score:2)

      by tepples ( 727027 )

      How does recognizing "their right to not use it in production" "promote the Progress of Science and useful Arts"?

  • EU and Asia need to agree to some more sane rules for Patents and copyright and cut US out. With the lobbying organizations in place, US is a lost cause. Patents need to have much tougher checks against prior art and much higher innovation content. Bad faith patent applications need to be expensive to make. Some things should not be patentable at all and patent life needs to be reconsidered since the pace of the world has increased. There should be mandated licensing requirements too. Copyright needs to b

Slashdot Top Deals

Arithmetic is being able to count up to twenty without taking off your shoes. -- Mickey Mouse

Close