Comments Locked

42 Comments

Back to Article

  • azfacea - Friday, March 6, 2020 - link

    something something APU something something
  • extide - Saturday, March 7, 2020 - link

    They wouldn't need this though, AMD clearly have their own HBM memory controllers.
  • JasonMZW20 - Saturday, March 7, 2020 - link

    AMD licenses Synopsys HBM2 PHYs and controllers. At least, they did for Vega 10/20.

    https://www.synopsys.com/dw/ipdir.php?ds=dwc_hbm2
  • Kurosaki - Friday, March 6, 2020 - link

    Sue them!!!
  • Sivar - Friday, March 6, 2020 - link

    Which JEDEC meeting did they steal this from?
  • rocky12345 - Friday, March 6, 2020 - link

    You beat me to it because I was thinking the same thing.
  • Spunjji - Thursday, March 12, 2020 - link

    100%
  • rocky12345 - Friday, March 6, 2020 - link

    My question is this since they are working with the HBM tech or at least making a controller to work with it have they also licensed themselves to be able to make products work with the HBM2e memory. I only ask because of this companies shading dealings in the past where they sat in the JEDEC meetings years ago then left quickly put out a whole bunch of patents without actually making a product and then sat and waited until memory companies started making and releasing memory products based on the JEDEC spec and Rumpbus then quickly sued pretty much every company they could to suck money from something they basically stole.

    To me at least this company is the scum of the scum and I would personally never buy anything they had anything to do with. END OF RANT Ok one more even scum of the earth types most likely think this company is lower than scum...lol
  • Yojimbo - Friday, March 6, 2020 - link

    They made their witchcraft with newts eyes and toad tongues and puppy dog tails. Beware!
  • DeeDee - Sunday, May 31, 2020 - link

    rocky12345, you're wrong. A little knowledge can be a dangerous thing. Here are the findings
    of facts:

    rocky12345, you don't have the slightest idea what you're talking about:

    << In sum, substantial evidence does not support the jury’s verdict that Rambus
    breached its duties under the EIA/JEDEC policy. Infineon did not show the first element of
    a Virginia fraud action and therefore did not prove fraud associated with the SDRAM
    standard. No reasonable jury could find otherwise. The district court erred in denying
    JMOL of no fraud on the SDRAM verdict. Because of these holdings, the new trial and
    injunction issues are moot. >>

    <<Because Infineon did not show that Rambus had a duty to disclose before the DDRSDRAM standard-setting process formally began, the district court properly granted JMOL
    of no fraud in Rambus’s favor on the DDR-SDRAM verdict. >>

    http://media.corporate-ir.net/media_files/NSD/RMBS...
  • Yojimbo - Friday, March 6, 2020 - link

    Rambus shot my dog.
  • surt - Friday, March 6, 2020 - link

    And by doing so, decreased the average level of evil of their actions.
  • Sivar - Monday, March 9, 2020 - link

    "And by doing so, decreased the average level of evil of their actions."

    This is mathematically sound even though shooting someone's dog is, in fact, evil.
  • PeachNCream - Saturday, March 7, 2020 - link

    They probably did you a favor. Dogs are like...yuck!
  • drexnx - Friday, March 6, 2020 - link

    the rambus chicanery with SDRAM was literally 20 years ago, time to move on
  • jeremyshaw - Saturday, March 7, 2020 - link

    Yes, it's time to move on and never use RAMBUS ever again.
  • evilpaul666 - Saturday, March 7, 2020 - link

    So... they're veteran patent trolls?
  • Strunf - Monday, March 9, 2020 - link

    It wasn't just SDR, it was DDR too and the last case they lost was in 2011.
    After that they settled with everyone for "pocket money".
  • DeeDee - Sunday, May 31, 2020 - link

    Too bad that there's some misinformed zealots here who have bitter feelings based on erroneous
    facts:

    <<But that is not how this litigation, broadly conceived, begun. Micron sued Rambus for declaratory judgment on August 28, 2000 in Delaware. Hynix sued Rambus for declaratory judgment the next day here in San Jose, California. Persuasive circumstantial evidence suggests that Micron and Hynix coordinated their filing of the declaratory judgment lawsuits. See Conduct Trial Tr. 4006:8-4008:12 (Mar. 5, 2008) (former Hynix employee Farhad Tabrizi admitting to his belief in 2000 prior to litigation with Rambus that Hynix would litigate "to the end of Rambus company"); 5131:1-25; 5135:1-16 (Mar. 18, 2008) (testimony of Micron CEO Steve Appleton that he met with Hynix personnel in early August of 2000 and that he cancelled licensing meetings with Rambus to sue for declaratory judgment); 5633:16-5634:1 (Mar. 20, 2008) (testimony from Hynix employee D.S. Chung about licensing meetings Hynix scheduled with Rambus, then cancelled by suing for declaratory judgment); cf. Licensing Trial Tr. 420:24-423:6 (Sept. 24, 2008) (testimony of Samsung employee Jay Shim regarding joint defense agreement between Samsung, Hynix, Micron, and Infineon enteredinto in August 2000).>>

    scribd.com
  • DeeDee - Sunday, May 31, 2020 - link

    Better link:

    https://www.scribd.com/document/11619032/3897
  • DeeDee - Sunday, May 31, 2020 - link

    Rambus Receives Decision in SK Hynix Case

    Rambus entitled to compensation for infringement based on fair and reasonable royalty rates

    https://www.rambus.com/rambus-receives-decision-in...

    SUNNYVALE, CALIFORNIA, UNITED STATES – 09/23/2012 – Rambus Inc. (Nasdaq:RMBS), one of the world’s premier technology licensing companies, today announced that the judge for the Northern District of California (NDCA) has issued his decision in the matter with SK Hynix. The Honorable Ronald M. Whyte has found that the Rambus patents in this case are valid and infringed by SK Hynix and Rambus is entitled to receive royalty payments for past infringement based on reasonable and non-discriminatory (RAND) rates.

    “This is a positive result as it is consistent with what we’ve been seeking all along – reasonable compensation for the use of our patented inventions,” said Thomas Lavelle, senior vice president and general counsel at Rambus. “We appreciate the Court’s extensive efforts in working through years of complex arguments. While this decision does not provide SK Hynix with a going-forward license, we are hopeful it will lead to putting this matter behind us completely and allow us to reach reasonable agreements.”

    In his ruling, Judge Whyte found that Rambus executed its document retention practices during a time when it reasonably anticipated litigation, and thus willfully spoliated evidence, but also found that Rambus did not deliberately destroy documents it knew to be damaging. The parties have been ordered to provide briefs on the issue of the damages SK Hynix will need to pay Rambus.

    Rambus management will discuss this decision during a special conference call on Monday, September 24, 2012 at 6:00 a.m. PT. The call will be webcast and can be accessed through the Rambus website. A replay will be available following the call on Rambus’ Investor Relations website or for one week at the following numbers: (855) 859-2056 (domestic) or (404) 537-3406 (international) with the ID# 34575979.

    Background of the Matter

    This case was originally filed by SK Hynix against Rambus in August 2000. The case was split into three separate phases with Rambus prevailing in all three phases. The first phase considered SK Hynix’s allegations that certain Rambus patents should be unenforceable under the doctrine of unclean hands and spoliation
  • DeeDee - Sunday, May 31, 2020 - link

    drenxnx, very good point. But what's scary about some people is that their methodology
    of thought can be dangerous. I'm not saying just because of this but for any other
    important reason:

    It Ain’t What You Don’t Know That Gets You Into Trouble. It’s What You Know for Sure That Just Ain’t So - Mark Twain.
  • Sahrin - Friday, March 6, 2020 - link

    AMD put 16GB of HBM on the $700 Radeon VII.

    They're crazy if they don't think there's a market for this tech on a CPU.
  • MrEcho - Friday, March 6, 2020 - link

    Im just waiting for this to happen. And now with the memory controller on its on chip...
  • close - Monday, March 9, 2020 - link

    Would be nice for highly integrated systems (laptops, AIOs, and such, assuming the power consumption is kept in check), or if you could still expand the memory using regular DDR DIMMs. But this implies having more complexity on the IMC side. Also trying to fit another 4 dies (for 64GB of RAM) today on something like a 3700X would make for a crowded package.
  • MDD1963 - Saturday, March 7, 2020 - link

    TIme for RAMBUS to file a lawsuit over other vendors perhaps using bits and bytes and stuff....and transferring them...to and from!
  • Lord of the Bored - Saturday, March 7, 2020 - link

    Rambus is still around? Why aren't they dead yet?
  • FreckledTrout - Saturday, March 7, 2020 - link

    Because they keep coming up with new ideas and selling IP. They are a hybrid patent troll and actual make new IP.
  • electronicschlong - Saturday, March 7, 2020 - link

    Once again Rambus isnt making a product or service, they're making a patent to sue innovators. Rambus is a parasite on the industry and the world will be better when they disappear.
  • Yojimbo - Saturday, March 7, 2020 - link

    They are making a design which they patent and license to anyone who wants to use it. If they didn't license the design from RAMBUS they'd have to license it from someone else or design it themselves. Designing it themselves would mean hiring people with the expertise to do it. Apparently, RAMBUS is a lower cost, or higher quality, designer. It's a standard service that is a key part of the tech industry. It is performed by many companies and relied on by many companies that make products.
  • webdoctors - Sunday, March 8, 2020 - link

    That's not entirely accurate. When you license a commercial IP block design from someone, its verified and tested. Its something tangible that is production quality. Its not just a block diagram and a patent. Otherwise what's to stop someone from printing out all the papers from ASPLOS/MICRO/ISCA and filing them all at the patent office and hoping it passes under the radar of the authors and selling it to 3rd parties as a "design" folks can license.

    When you get a block from Syopsys or Cadence or other 3rd party IP design companies its usually pretty mature and maybe even taped out on a test chip to ensure it works.

    tldr Avoid Rambus unless you want to spend more on lawyers than engineers
  • UltraWide - Monday, March 9, 2020 - link

    Thank you for this! It's nice to have some facts once in a while!
  • Yojimbo - Friday, March 13, 2020 - link

    What facts? The fact that Rambus hasn't sued anybody in something like 20 years? Why do you his assertions that make no sense to be "facts"? You believe that Rambus are forcing people to buy their PHYs?
  • Yojimbo - Friday, March 13, 2020 - link

    Please source me where you get the idea that Rambus's IP blocks are not verified and tested whereas others are. And even if so, it's not relevant, because if Rambus's blocks are not verified or tested then that is perfectly OK with their customers who choose them. I believe what I said was entirely accurate and what you said is just nonsense.
  • balle - Thursday, March 26, 2020 - link

    You might want to step into DesignCon and see a working demo of this technology that is tested, verified and is being used by multiple customers
  • DeeDee - Sunday, May 31, 2020 - link

    So you're saying that Rambus doesn't collaborate, verify and test?

    <<Samsung and eSilicon Taped Out 14nm Network Processor with Rambus 28G SerDes Solution>>

    https://news.samsung.com/global/samsung-and-esilic...
  • DeeDee - Sunday, May 31, 2020 - link

    You're saying that Rambus doesn't do tape outs?

    <<Rambus Announces Tapeout of GDDR6 Memory PHY on TSMC 7nm Process Technology
    Leading IP to support TSMC’s customers with AI, HPC, automotive and networking applications

    SUNNYVALE and SANTA CLARA, Calif. – Jan. 30, 2019 >>

    https://www.rambus.com/rambus-announces-tapeout-of...
  • DeeDee - Sunday, May 31, 2020 - link

    No tape outs? Why do you comment without understanding what you're commenting about?

    Rambus Tapes Out 112G XSR SerDes PHY on Leading-edge 7nm Process
    Highlights:

    Provides critical building block to deliver data for next-generation data center, networking, high-performance computing (HPC), artificial intelligence (AI) and machine learning (ML) applications
    Delivers superior power, performance and area (PPA) for extra short reach (XSR) links with innovative architecture designed for leading-edge 7nm process node
    Expands SerDes PHY portfolio for 112G OIF-CEI industry standard

    SUNNYVALE, Calif. – Sep. 25, 2019

    https://www.rambus.com/rambus-tapes-out-112g-xsr-s...
  • DeeDee - Sunday, May 31, 2020 - link

    Yojimbo, thank you for your informed and intelligent comment.

    I would think that individuals reading about technology would have high enough
    IQs to make informed and well thought out comments. But too bad that isn't the
    case in some instances. Some bring up old and misinformed biases that they
    just can't grow out of!
  • ravyne - Sunday, March 8, 2020 - link

    Appears to be a typo in the second paragraph, it says memory devices of up to 16Gbps, but I think you mean to say 16Gb.
  • master098 - Friday, May 8, 2020 - link

    Made us delighted with your article ,best regards from https://typhoon-tv.com/
  • legendsneverdie1455 - Monday, July 6, 2020 - link

    Thanks for the detailed article, Would love to see more of these. https://tricksground.com/typhoon-tv-apk-download/

Log in

Don't have an account? Sign up now