Rambus Could Reap Millions In Patent Settlements

← Back to Stories (view on slashdot.org)

Rambus Could Reap Millions In Patent Settlements

Posted by kdawson on Tuesday July 27, 2010 @03:30PM from the what-innovation-looks-like-now dept.

RedEaredSlider writes "Rambus, a designer of semiconductor chips, won a long-running patent battle with NVIDIA, but that dispute is not the only one the company is involved in — and the upcoming decisions could mean millions in additional revenue. Besides the NVIDIA decision, Rambus is involved in a suit with Hynix Semiconductor that will be heard in October. In that case, Hynix had originally sued Rambus in 2000, but Rambus counter-sued. Hynix lost, and appealed. The parties will appear before the US Court of Appeals for the Federal Circuit in October. A judgment in Rambus's favor would be worth at least $397 million, according to the company's general counsel, Tom Lavelle."

40 of 82 comments (clear)

Min score:

Reason:

Sort:

To keep up by Lord_of_the_nerf · 2010-07-27 15:48 · Score: 3, Funny

ATI is going to want to get sued in about six months for $500 million.
1. Re:To keep up by ScrewMaster · 2010-07-27 15:50 · Score: 3, Interesting
  
  ATI is going to want to get sued in about six months for $500 million.
  Maybe they'll make the mistake of suing IBM.
  
  --
  The higher the technology, the sharper that two-edged sword.
Rambus... by Sol+Rosinberg · 2010-07-27 15:52 · Score: 5, Insightful

Wow, there's a company I haven't heard of in years. Didn't they make some hideously expensive RAM that was supposed to perform twice as well as normal RAM, but never lived up to the hype? Basically, they patented the design, couldn't get it to work right, and now they're suing the companies who did.
1. Re:Rambus... by obeythefist · 2010-07-27 16:00 · Score: 5, Informative
  
  Those are the guys! It was many years ago. They apparently sat on some big RAM council board, took a bunch of notes, and sprinted to the patent office as soon as the meeting was over. BAM! Patent infringement lawsuits all over the place.
  Unlike other patent trolls, however, they did manage to release a horrible implementation of a quasi-serial memory bus that had atrocious real world performance. And managed to persuade Intel to get in bed with them, which for many was a big reason that AMD was doing so well at the time.
  
  --
  I am government man, come from the government. The government has sent me. -- G.I.R.
2. Re:Rambus... by Bacon+Bits · 2010-07-27 16:06 · Score: 5, Informative
  
  No, no, they got it to work right. It worked four times as fast (at the time) but it cost more than four times as much. And they didn't allow anybody else to produce the chips by refusing to sell license their patent to other manufacturers. And you had to buy in pairs (and have those stupid blanks in unpopulated slots, and who wants to buy those stupid things?). And they didn't reduce the price at all even though the competition began to close the gap in speed and surpassed them in capacity.
  RDRAM failed because it was more expensive due to price fixing monopoly, short in supply due to monopoly, more difficult to use and install, and advanced more slowly than the competition (due to a smaller set of researchers working on improving it). And Intel priced their processors far and above the price of AMD's DDR-only, better-performing competition. RDRAM is a good example of customer demand shaping the market.
  
  --
  The road to tyranny has always been paved with claims of necessity.
3. Re:Rambus... by tlhIngan · 2010-07-27 16:35 · Score: 4, Informative
  
  Wow, there's a company I haven't heard of in years. Didn't they make some hideously expensive RAM that was supposed to perform twice as well as normal RAM, but never lived up to the hype? Basically, they patented the design, couldn't get it to work right, and now they're suing the companies who did.
  Actually, they're quite big and raking in the dough. They're in one of the best selling consoles, and in this generation's worst-selling console.
  Yes, the PS2 has 32MB of RAMBUS RDRAM in it. It was one of their big announcements after thair failed foray into PCs.
  And yes, there's also 256MB of XDR DRAM in the PS3 (the other 256 is GDDR3 for the nVidia RSX). XDR technology is owned by RAMBUS.
  Despite not selling well in the PC world, they're doing good enough in other markets. Hell, many /.'ers probably own a PS3 and thus have contributed to this patent trolling. Probably some of those complaining about patent trolls in this article too will extol the virtues of the PS3 (or PS2) in other threads without realizing that they're really just helping RAMBUS.
4. Re:Rambus... by obeythefist · 2010-07-27 16:52 · Score: 4, Funny
  
  Ohh I didn't realise that if you patent troll an illegal oligopoly, it's no longer patent trolling, it is in fact awesome patenting for great justice?
  No, patent trolls are patent trolls.
  I didn't know Rambus employees were able to get onto slashdot to post as AC's, though, I thought the RAM in their desktops was too slow to allow them to log on.
  
  --
  I am government man, come from the government. The government has sent me. -- G.I.R.
5. Re:Rambus... by Nikker · 2010-07-27 16:53 · Score: 3, Interesting
  
  As someone who bought said ram as part of a Dell in 2003 I can attest I would like to give Rambus a collective kick in the nuts for doing this. At the time coated me approx $600CDN for the upgrade from pc233 I believe? For 512mb. 7 years later 512mb pc 1066 is still $200. The performance was actually pretty good though giving approx 4gb/s performance, which at the time was pretty good. But that coupled with a Prescott I truly understood the value of building your own.
  
  --
  A loop, by its nature, continues. If that didn't make sense, start reading this sentence again.
6. Re:Rambus... by JackieBrown · 2010-07-27 17:06 · Score: 5, Insightful
  
  If we PS3 owners can get past the fact that we are helping Sony, I doubt helping RAMBUS will cause much worry.
7. Re:Rambus... by YesIAmAScript · 2010-07-27 17:28 · Score: 5, Informative
  
  It was only faster for a very short period.
  When DDR came out and was adopted, it had higher throughput than RDRAM. And both SDRAM and DDR had lower latency. So SDRAM actually was higher performing in many cases and DDR was in all cases.
  In no time, you had to buy the fastest (most expensive) RDRAM 800 and in pairs in order to come even close to matching DDR on performance.
  A bank of RDRAM 800 had 1600MB/sec bandwidth, while a single bank of 133MHz (DDR-266) had 2166MB/sec bandwidth. A single bank of PC-133 SDRAM had 1066MB/sec bandwidth and much lower latency. Most RDRAM systems (before the very end) had RDRAM 600 in them, for a bandwidth of 1200MB/sec.
  RDRAM's advantage the whole time was the ability to put in a lot more RAM than with DDR because since each DIMM regenerated the signal before passing it on, you could daisy chain RAM as far away as you'd like, unlike where with SDRAM or DDR the total bus length was limited by the capacitance placed on by all the DIMMs on the bus.
  In essence, RDRAM had the same advantages and disadvantages of FB-DIMMs, which was also driven out of the market by cheaper and faster DDR3.
  
  --
  http://lkml.org/lkml/2005/8/20/95
8. Re:Rambus... by amorsen · 2010-07-27 18:26 · Score: 4, Informative
  
  It worked four times as fast (at the time) but it cost more than four times as much.
  Only in bandwidth. Latency sucked big time, and processors back then weren't very good at hiding latency, so real-world non-benchmark performance wasn't any better than decent SDRAM.
  
  --
  Finally! A year of moderation! Ready for 2019?
9. Re:Rambus... by meerling · 2010-07-27 18:48 · Score: 2, Interesting
  
  I know most people define patent trolls as never releasing real products, but I don't think it should be that strict. Any company that engages in significant patent lawsuits for cash settlements (not patent portfolio license exchanges) should be classed a patent troll. Obviously any company that doesn't produce product is automatically in that category, but there are others that should be as well. Rambus is one of them IMO.
10. Re:Rambus... by Dionysus · 2010-07-27 19:27 · Score: 2, Insightful
  
  Any company that engages in significant patent lawsuits for cash settlements (not patent portfolio license exchanges) should be classed a patent troll.
  
  Nah, still too strict. How about anybody Slashdotters don't like who also holds any patens are patent troll?
  
  --
  Je ne parle pas francais.
11. Re:Rambus... by hardboiled.tequila · 2010-07-27 19:31 · Score: 2, Funny
  
  RDRAM's advantage the whole time was the ability to put in a lot more RAM than with DDR because since each DIMM regenerated the signal before passing it on, you could daisy chain RAM as far away as you'd like, unlike where with SDRAM or DDR the total bus length was limited by the capacitance placed on by all the DIMMs on the bus.
  I read the above paragraph with the voice of Doc Emmett Brown... "No no no! You're not thinking fourth-dimensionally!"
12. Re:Rambus... by asc99c · 2010-07-27 20:10 · Score: 4, Insightful
  
  I think the problem people have with RAMBUS is specifically that they sat in on the JEDEC working groups to watch the development of new standards, and then when everyone had decided on the design and invested huge amounts of money, they piped up that they had a patent on it. This was completely against the principles of the JEDEC working groups - where the meetings are specifically aimed at ensuring they are developing an open standard free from their member's patents which might otherwise block the technology - not necessarily 100% unpatented, but where they know what the patent licensing agreements are. Generally all the members agree to cross-licence the patents between themselves - this is the 'cost' of joining the JEDEC standards group.
  The RAMBUS technology (and to a leser extent, it's implementation) is actually rather good, so they aren't a classic patent troll, but their submarine patent on DDR technology makes them clearly some form of patent troll.
13. Re:Rambus... by RMH101 · 2010-07-28 00:25 · Score: 4, Interesting
  
  I remember working 2nd line support for a company with a load of RAMBUS Dells. Caused a bit of a commotion when the usual upgrade-by-slapping-in-more-RAM option was costed. I can remember it costing more than a new Optiplex at the time. Much hilarity with management.
14. Re:Rambus... by stevew · 2010-07-28 02:38 · Score: 3, Informative
  
  You guys have the story slightly wrong - they didn't just take notes. They revealed technology that they got everyone else to agree to put into the standards THEN announced patents on same.
  It's a slight difference - but makes them even nastier.
  
  --
  Have you compiled your kernel today??
15. Re:Rambus... by _avs_007 · 2010-07-28 09:02 · Score: 2, Interesting
  
  In most standard bodies that I participate in, there is a clause in the member's agreement that says that anything you disclose/discuss to/with the committee, constitutes IP submission to the committee's patent pool of any IP associated with the topic in discussion. And there is usually a mandatory IP check before the spec is ratified. Any members that fail to declare IP that they wish not to license according to the member's license agreement, lose rights to said IP. I've seen a few cases where IP was later declared before the spec was ratified. In most cases, the company agrees to the licensing terms mandated in the member's agreement. In the once case I've seen where the member refused the licensing terms, their membership in the standard's body was revoked. (Basically they said, agree to the terms you agreed to in the member's agreement, or leave the standard's body. It was then pointed out if they left the standard's body, they would LOSE ALL RIGHTS to any patents that were submitted to the patent pool by other members, because the license agreement is only between the members)
  
  Not sure how JEDEC works...
Re:"a designer of semiconductor chips" ???? by ScrewMaster · 2010-07-27 15:53 · Score: 4, Insightful

No. NO. Rambus is an IP troll, at best they specify how chips talk to each other on a motherboard. Smarter people then design chips.
Yeah. Rambus has a lot in common with SCO ... parasitic beasts at best.

--
The higher the technology, the sharper that two-edged sword.
Re:and all the compsci want to know by ScrewMaster · 2010-07-27 15:54 · Score: 2, Funny

what algorithm were they using? And how were they load balancing? You'd almost have to be doing a bucket sort initially? And there was no detail whatsoever as to the data being sorted. Were they just a crapton of 64 bit words, sentences, database records, what was the data being sorted? I can sort a petabyte in O(1) time if it consists of two records each 1/2pb in size. None of what they've said means anything without getting into the context.
Oh, sorry ... this is abuse. Argument is down the hall.

--
The higher the technology, the sharper that two-edged sword.
Re:World Cup of data sorting by ScrewMaster · 2010-07-27 15:55 · Score: 2, Funny

How many parallel predicting octopuses were required to predict their victory?
You mean "predictopi".

--
The higher the technology, the sharper that two-edged sword.
The SCO outlook by coldmist · 2010-07-27 16:01 · Score: 4, Insightful

Although their chances are better than SCO's (debatable, but I'd give it to them), this story sounds as rosy as an SCO fanboy writing their weekly column.
"could mean millions" Could. Could.
I really wish we had a news service that posted honest stories.
Rambus has sued the world, and finally one of them stuck. nVidia is the loser this time. If only Rambus would die, then we could all move on in life.
See how much nicer that would be! ;)

--
Don't steal. The government hates competition.
1. Re:The SCO outlook by Lawbeefaroni · 2010-07-27 16:42 · Score: 4, Informative
  
  "One of them stuck?" Of the top of my head, Samsung recently (2010) settled with them for something like $200M plus $100/year for 5 years for licensing, plus shares for a grand total of $900M.
  Hate to break the news to you but Rambus isn't going to die. Unless someone like Samsung buys them.
  
  --
  "When it rains, it pours." --Morton's Salt
Re:and all the compsci want to know by the_other_one · 2010-07-27 16:05 · Score: 2, Funny

Where do I go for sexual harassment. I'm not getting any,

--
134340: I am not a number. I am a free planet!
Re:"could reap"!? by Klinky · 2010-07-27 16:31 · Score: 3, Insightful

Haters Gonna Hate,
Patent Trolls Gonna Litigate
Lawful Evil by mmmmbeer · 2010-07-27 16:33 · Score: 3, Funny

Because crime doesn't pay, but exploiting flawed laws pays big time!
1. Re:Lawful Evil by Anonymous Coward · 2010-07-27 18:04 · Score: 2, Insightful
  
  Naming the character alignment for r/l entities sounds like a fun nerd-game.
  I claim that the pirate-bay guys are Chaotic Neutral!
2. Re:Lawful Evil by noidentity · 2010-07-27 19:08 · Score: 2, Insightful
  
  You joke, but it's exactly true. If you commit a crime, you only have your own resources and power to carry it out. But if you get the state to do the work for you, then you have virtually unlimited power. The more you can have laws made which restrict the free market, the more power you have against the other guy (sure, the other guy might be able to wield said power, but if you got the laws made the right way, you will have better control of their power).
Number Bias? by Revotron · 2010-07-27 17:41 · Score: 2, Insightful

Reporting possible settlements as stated by the complainant's GC is like buying a diamond ring from a jewelry store at the sticker price. It's going to be a horribly, horribly overblown number way above reason and logic, because their goal is to squeeze every penny out of you. It's just like buying a car - only chumps pay the sticker price.
Even if Rambus can prove every single violation, there's a great deal of doubt in my mind that the judge will actually award them what they think it's worth.
Octopi/octopuses by tebee · 2010-07-27 17:55 · Score: 2, Informative

No he doesn't, see http://www.youtube.com/watch?v=wFyY2mK8pxk - and as an added bonus you get a very nice piece of thinking man's crumpet as well.

--
N.B. this user is far too lazy to write a witty and intelligent sig.
1. Re:Octopi/octopuses by ScrewMaster · 2010-07-27 23:11 · Score: 2, Funny
  
  No he doesn't, see http://www.youtube.com/watch?v=wFyY2mK8pxk - and as an added bonus you get a very nice piece of thinking man's crumpet as well.
  "Ah say, ah say, it's a JOKE, son!" -- Foghorn Leghorn
  
  --
  The higher the technology, the sharper that two-edged sword.
2. Re:Octopi/octopuses by Dishevel · 2010-07-28 03:20 · Score: 2, Informative
  
  Semi cute I would give you. Cute I MIGHT give you. Though really cute I have to say I must callout as bullshit.
  
  --
  Why is it so hard to only have politicians for a few years, then have them go away?
Re:"could reap"!? by Dahamma · 2010-07-27 18:11 · Score: 2, Insightful

Hah, and the haters and trolls who work for Rambus (or have stock?) apparently have mod points today...
Re:"a designer of semiconductor chips" ???? by Anonymous Coward · 2010-07-27 18:42 · Score: 5, Informative

Actually Rambus is doing research and making test devices as well as ASICs on behalf of other companies.
Very much like ARM and IBM.
Rambus has a 2:1 lawyer to engineer ratio (they did 2-3 years ago), which skews them towards patent troll ;)
patent troll by Lead+Butthead · 2010-07-27 18:47 · Score: 4, Insightful

Rambus is one company that needs to be nuked from orbit while all of their lawyers and officers are locked in the building.

--
ELOI, ELOI, LAMA SABACHTHANI!?
Re:"a designer of semiconductor chips" ???? by hairyfeet · 2010-07-27 19:51 · Score: 4, Insightful

Nope, sorry, they are patent trolls. If you read the history of their involvement with JEDEC, they basically fucked EVERYBODY, including us consumers, by sitting in on the SDRAM meetings and secretly patenting everything that wasn't nailed down.
So while they may have invented a few cool things their actions at JEDEC were total patent trolling and any patents that arose from those meetings seriously needs to get tossed out. Instead what will happen is you Mr Anon Coward, Me, and everybody else will pay more for memory of ANY kind, be it for motherboards, GPUs, or frankly just about anything using modern memory tech, because those trolls were allowed to steal everything that wasn't nailed down. So unless some completely new memory tech comes down the pipe (and even then as broad as the patents are you might nor even escape then) we will ALL be paying a "RAMBUS tax" on anything that has memory for the next 20+ years. What we need is a company like IBM or Google to come along and buy them out and offer their patents with RAND pricing, instead of allowing these trolls to continue to exist.

--
ACs don't waste your time replying, your posts are never seen by me.
Not quite by arth1 · 2010-07-28 00:10 · Score: 3, Informative

Those are the guys! It was many years ago. They apparently sat on some big RAM council board, took a bunch of notes, and sprinted to the patent office as soon as the meeting was over. BAM! Patent infringement lawsuits all over the place.
Not quite... It was more like the council had a non-disclosure agreement, and Rambus showed them what they were working on, after which a couple of the other manufacturers turned around and blatantly copied the inventions. Unfortunately for them, the patents in question had been filed long before.
Which is why Rambus has been awarded compensation in several judgments.
While Rambus certainly has bad sides and skeletons in the closet (they're a corporation after all), it's not correct to call them a patent troll.
And their RDRAM failed for two major reasons: Price and Samsung
The price was MUCH higher, especially for 32-bit RDRAM. Yes, it was faster, but the benefits were not that high. Add that Samsung, the sole producer of 32-bit RDRAM, couldn't deliver in a timely manner, and the customers had a choice between buying a more expensive RDRAM motherboard with very expensive RDRAM that they would have to wait weeks for, or a marginally slower DDR motherboard with very cheap RAM that they could get straight away. Still, RDRAM found its way into many servers.
But RDRAM existed, and it was faster (if comparing apples to apples) -- it wasn't a patent troll, but patented technology they had actually made into products.
People (even here) were slamming Rambus because they wanted cheap, and if Rambus won, DDR2 prices were likely to go way up.
1. Re:Not quite by cheesybagel · 2010-07-28 03:01 · Score: 4, Insightful
  
  Other companies than Samsung tried to manufacture RDRAM and failed even more, because of poor quality control. The fact is the design was complex, and hard to manufacture, leading to all sorts of issues, including localized heat zones. If you remember, RDRAM was when we started seeing heat spreaders in memory modules.
  Sending a signal on both edges of the clock has been done since like forever. I learned it in class before Rambus was even founded. If DDR infringes on a patent because of that, the patent office needs to review its standards for accepting patent applications.
2. Re:Not quite by Solandri · 2010-07-28 09:53 · Score: 2, Interesting
  
  Not quite... It was more like the council had a non-disclosure agreement, and Rambus showed them what they were working on, after which a couple of the other manufacturers turned around and blatantly copied the inventions. Unfortunately for them, the patents in question had been filed long before.
  
  Which is why Rambus has been awarded compensation in several judgments.
  Holy rewriting of history. Lemme quote the original FTC Commission report on the case (pdf):
  
  JEDEC operated on a cooperative basis and required that its members participate in good faith. According to JEDEC policy and practice, members were expected to reveal the existence of patents and patent applications that later might be enforced against those practicing the JEDEC standards. In addition, JEDEC members were obligated to offer assurances to license patented technologies on RAND terms, before members voted to adopt a standard that would incorporate those technologies. The intent of JEDEC policy and practice was to prevent anticompetitive hold-up.
  
  Rambus, however, chose to disregard JEDEC's policy and practice, as well as the duty to act in good faith. Instead, Rambus deceived the other JEDEC members. Rambus capitalized on JEDEC's policy and practice - and also on the expectations of the JEDEC members - in several ways. Rambus refused to disclose the existence of its patents and applications, which deprived JEDEC members of critical information as they worked to evaluate potential standards. Rambus took additional actions that misled members to believe that Rambus was not seeking patents that would cover implementations of the standards under consideration by JEDEC. Rambus also went a step further: through its participation in JEDEC, Rambus gained information about the pending standard, and then amended its patent applications to ensure that subsequently-issued patents would cover the ultimate standard. Through its successful strategy, Rambus was able to conceal its patents and patent applications until after the standards were adopted and the market was locked in. Only then did Rambus reveal its patents - through patent infringement lawsuits against JEDEC members who practiced the standard.
  
  The Commission finds that Rambus violated Section 5 of the FTC Act by engaging in exclusionary conduct that contributed significantly to the acquisition of monopoly power in four relevant and related markets. We further find a sufficient causal link between Rambus's exclusionary conduct and JEDEC's adoption of the SDRAM and DDR-SDRAM standards (but not the subsequent DDR2-SDRAM standard). Questions remain, however, regarding how the Commission can best determine the appropriate remedy. Accordingly, the Commission orders additional briefing for further consideration of remedial issues.
  Yes Rambus had some patents pending before they approached JEDEC. But the whole point of being a member of JEDEC was to disclose those patents and negotiate licenses for the patents before the technologies were codified as a standard. That way the other members could decide, with all information in hand, which memory technology they would make the standard. Rambus deliberately failed to do this and did not disclose what they were patenting, essentially tricking the others into creating a memory standard which incorporated their patented technology so they could later threaten them with patent infringement lawsuits. They event went so far as to take ideas other JEDEC members revealed with the understanding that it wouldn't be patented so everyone would be free to use it, and incorporated them into their own patents.
  
  The reason Rambus "won" was because even though the FTC found that they violated JEDEC's policies, the courts still found that the patents were valid. Normally when you write a contract, you're supposed to include
Re:and all the compsci want to know by Pharmboy · 2010-07-28 00:30 · Score: 5, Informative

Where do I go for sexual harassment. I'm not getting any,
Prison.

--
Tequila: It's not just for breakfast anymore!