Decades-Old Rambus Litigation Against Micron For RDRAM Tech Reaches Settlement

An anonymous reader writes "The decades-old Rambus litigation against Micron for RDRAM tech finally reached a settlement. RDRAM tech has already been licensed by NVidia and Broadcom and has been used in game consoles such as the Nintendo 64. The preliminary deal is to last 7 years and net $280M for Rambus and Micro to gain access to patent licenses defining the technology."

Comments

Now I feel old.

[Naatach]

Can you remember when this started?

Re:Now I feel old.

[geminidomino]

If it's "decades old" how the hell is the bogus patent even still valid?

Re:Now I feel old.

[t0qer]

Statute of limitations only applies to when a case can be filed. A trial can go on for years past the statute.

Re:Now I feel old.

[ArbitraryName]

Yes, it was in 2000. Rambus has only existed since 1990 and they didn't start suing people until 2000. Unless you're a teenager I don't see how you couldn't.

Re:Now I feel old.

[icebike]

Exactly,

Join a consortium to develop a standard, sneak out at a quarter to four and patent the whole thing behind everyone's back and start throwing sue balls at everyone. They never had an original idea, or a working product.

Re:Now I feel old.

[Ecuador]

I remember Intel trying to stuff some stupidly expensive, high latency RD-RAM down our throats around '99-2000 in some sort of deal with Rambus (a company probably even more unethical than Intel) where they both would win and consumers... not so much...
Anyway, depending on how you see it, "decades old' is either an exaggeration, or a very appropriate statement, in line with the usual factual and to-the-point slashdot summaries.

Re:Now I feel old.

[icebike]

The Rambus debacle actually pre-dates Slashdot.
The patent cat fight started in 1990, and slashdot started in 1997. It was old news by the time it was first mentioned on Slashdot.

Re:Now I feel old.

[Ralph Wiggam]

I remember back in 2000 when Rambus was the most hated company on Slashdot. Everyone hated Microsoft, but at least acknowledged that they were producing products. Rambus was just a dirty parasite.

Re:Now I feel old.

[icebike]

Had you clicked the link in the summary you would have seen this as the second sentence:

The two have been mired in litigation since 1990, which is when Rambus first sought license fees and threatened infringement lawsuits against memory makers who turned to the popular SDRAM standard over its own.

Rambus started suing immediately after they formed the company.

Re:Now I feel old.

[Ralph Wiggam]

This particular debacle started in 1990. In 2000, Rambus filed lawsuits against every memory manufacturer in a much larger debacle. That's the debacle that dominated Slashdot for months and contributed to a huge influx of users.

Re:Now I feel old.

[icebike]

They forced them to license the name and put it on those products. Rambus never manufactured anything.

Re:Now I feel old.

[Anonymous Coward]

Well, there was probably one or two optimizations in RD-RAM that seemed meaningful over "regular" RAM at the time (probably had to do with timing and read/write operations, though- didn't RD-RAM read on both up-tick and down-tick of clock cycles, or something like that?), and Intel thought it'd really make the Pentium 4 a powerhouse system. Except Intel really overreached with the P4 (and Itanium was developed around then too, right?), and forgot the most important part about RAM, that being the "Random Access" bit.

RD-RAM could feed data onto the bus serially rather fast, once it found it, but really sucked at latency with random access reads/writes. But that fast, high-latency streaming seemed to work well for P4's rather deep instruction pipelines too on some benchmarks, when it was in the sweet spots for them together (P4 resolving a deep instruction pipeline while RD-RAM retrieved memory and started pumping data onto the CPU bus, which was then ready to start reading it).

Meanwhile, it became evident that multi-core CPUs provided a much more responsive system than the high clock rate P4 systems, even if those cores were running at less than half the clock rate of the P4, and it seemed like the P4 should have kicked ass over other systems.

But then those stupid P3 hackers in Israel made a couple of improvements in the P3 core (my dual processor 833 MHz P3 box was far more usable than a 2GHz P4 with gobs of RD-RAM at the time), which begat the Core, Core 2 and Core 2 Duo processors, and the P4 just quietly faded away as the P3 technology got to move on.

So, in IT, every once in awhile the right idea/implementation wins out over someone's over-hyped/over-sold pet project...

Re:Now I feel old.

[ArbitraryName]

Rambus never manufactured or sold RAM. They licensed patents that allowed third parties like Intel to make it.

Re:Now I feel old.

[Anonymous Coward]

They forced them to license the name and put it on those products.

How exactly was either Dell or Intel "forced" to license RDRAM and ship it?

Rambus never manufactured anything.

Nice goalpost shifting.

Re:Now I feel old.

[Anonymous Coward]

Which is not the same as "not having a working product". There clearly was a working product since others were able to manufacture RDRAM.

Re:Now I feel old.

[ArbitraryName]

I read the article. The article is wrong. I was around when they started suing, as your UID shows you should have been. Here is a contemporary article in EE Times about the first lawsuit in 2000, against Hitachi. The article's wording makes it clear that this is the first lawsuit. This was big news on Slashdot in 2000, where an article from 2000 whose wording (and comments) also make it obvious that the lawsuits had just begun. Here is a slightly earlier Slashdot article about Samsng and Rambus whose comments also make it obvious that no lawsuits had begun yet.

Re:Now I feel old.

[khallow]

Recall the original claim?

They never had ... a working product.

And the most recent observation

They forced them to license the name and put it on those products.

Would be quite consistent with the original statement. Just because Rambus's logo is on a product doesn't mean it is a Rambus product. There is no shifting of goalposts here.

Re:Now I feel old.

[viperidaenz]

You missed a step in the P3 -> Core, which was the Pentium M. Intel was pretty much forced to build it, because power hungry P4's sucked in laptops.

Re:Now I feel old.

[saleenS281]

They absolutely had both an original idea and a working product. Just because they didn't have manufacturing facilities doesn't mean they didn't produce a product. Does AMD not have a working product because they spun-out and sold-off Global Foundries? Does Sony not have a product because they use AMD CPUs? RDRAM was absolutely rambus's product, that was never up for debate. What was up for debate was whether DDR memory infringed on their patents - it did, and they won just about all of their lawsuits.

Re:Now I feel old.

[saleenS281]

First off, it wasn't bogus - which is why micron lost. I'm no lover of RAMBUS but they absolutely had valid patents, and they did sell product. They lost the battle because their product was ultimately inferior and more expensive, but that doesn't change the fact the core technology behind DDR memory infringed on their ideas. As for how it's still valid - it doesn't matter. If you infringe a patent, get sued, and the patent runs out before the court case finishes, you aren't magically exonerated from all the years you infringed on the patent while in court. Doesn't work that way champ.

Re:Now I feel old.

[radarskiy]

You are assuming that the final manufactured product actually matched the Rambus design, rather than being a different design that just happened to be patent encumbered.

I only have hearsay from people that worked on this stuff, but it all agrees on the fact that Rambus did not have a working design.

Re:Now I feel old.

[bhcompy]

Let's talk about goalposts, then. What you're saying is that the NFL doesn't have a product because the products are the teams that are NFL licensed franchises and the NFL itself is just a logo without a product. I don't think the billions being paid to the NFL support that statement. Products don't have to be tangible to be products. A product can simply be a license to a design. A better statement would be that Rambus never manufactured the product(then again, many hardware companies pay other companies to manufacture their products)

Re:Now I feel old.

[khallow]

What you're saying is that the NFL doesn't have a product because the products are the teams that are NFL licensed franchises and the NFL itself is just a logo without a product.

Yes. Licensing is not a product. Actual products would be the providing of refereeing officials for games and the establishing of a common body of rules for the game (including out of game stuff like the treatment of visiting teams, media, and spectators).

A product can simply be a license to a design.

A design that Rambus didn't actually have a part in aside from owning blocking patents.

Re:Now I feel old.

[epine]

Getting "gobs" of RD-RAM involved fat modules.

Aside from muck-raking HP, The Register also did a lot to raise public awareness of problems with Intel in the P4 era.

Caminogate failure finally explained

RD-RAM was supposed to support three slots, but they couldn't make it work. It was one fiasco after another.

Mercifully, we got AMD64 out of the deal and a healthy competitor until Intel's Israeli team kicked out the CoreDuo.

Intel subsequently played the UEFI card, but it's a dim echo of their original agenda.

Re:Now I feel old.

[geminidomino]

I was referring to the implication that they were still licensing the patent. I get now that it basically meant retroactively.

As for why it's bogus, I thought we all new the story. They snaked the patents out from under the development group. Probably legal, still a scumbag move.

Re:Now I feel old.

[Solandri]

First off, it wasn't bogus - which is why micron lost. I'm no lover of RAMBUS but they absolutely had valid patents, and they did sell product. They lost the battle because their product was ultimately inferior and more expensive, but that doesn't change the fact the core technology behind DDR memory infringed on their ideas.

You're badly misunderstanding people's gripe with this whole thing. Of course the patents were valid. The whole point of the consortium was to come up with a non-patented standard, so when Rambus went behind everyone's back and patented the whole shebang of course there was no prior art. The other memory makers had gone out of their way not to patent it.

And FYI, JEDEC won against Rambus in their lawsuit - Rambus was guilty of patenting stuff which was being discussed in secret at JEDEC. The problem was Rambus was only guilty of violating the terms of membership for JEDEC. The lawyers who drafted the membership agreement neglected to put in any penalties for patenting stuff behind everyone else's back, and as a result the only recourse JEDEC had was to kick Rambus out. There were no federal or criminal charges because they didn't break government law, they broke JEDEC's rules. The patents were still valid though, even though Rambus effectively stole them, because the there was no prior art and the USPTO decided it was patent-worthy.

Speaking of which, do you really think DDR is worthy of a patent? Counting a clock tick not just when the voltage goes up, but also when the voltage goes down? That should be obvious to a 5-year old.

Re:Now I feel old.

[saleenS281]

Why was there "no prior art" if Rambus simply took the JEDEC committee's ideas and created patenst? There should be VOLUMES of prior art from the JEDEC meeting minutes to people's scribbles in their notepads. You're seriously telling me that Rambus was the only company attending the meetings to take notes? And they managed to get their patents ramrodded through before anyone on the JEDEC committee decided to take their first note? Ya, no.

What Rambus did (and lost in court on) was create additional patents beyond the technology they already had to try to stonewall the JEDEC members. THAT is what they lost in court on. The JEDEC members STILL licensed technology from Rambus because they had legitimate patents. They were simply hoping that when Rambus joined the committee, they would donate those patents to the group. As for DDR being patented, if it's so obvious why wasn't there prior art again?

Re:Now I feel old.

[wiredlogic]

RDRAM itself is an original idea. Synchronous memory and double clocking were not. Rambus acted in bad faith by participating in the JEDEC process to standardize SDRAM and then backing out to create patents on technologies they learned about from that participation that would be applicable to emerging JEDEC standards. For that they deserve to wither into bankruptcy. I'll never design their devices into a product.

Re:Now I feel old.

[KlomDark]

I remember this from way too long ago. I figured it had wrapped up years ago. Was surprised to see it's just now reaching settlement. This was even before OJ's low-speed chase...

Re:Now I feel old.

[MachineShedFred]

You also forget that the adoption of Pentium 4 didn't actually take off until the release of the Intel i845 chipset, which allowed the usage of good ol' DDR SDRAM. With that, a P4-based system didn't cost in excess of $1800.

RDRAM was an albatross around Pentium 4's neck except in very few usage scenarios. You're right though - the super deep pipelines that only existed to ramp the clock rate sky high meant that a whole lot of clock was wasted on branch prediction failures. The so-called NetBurst architecture was a complete failure, and the design team in Israel saved Intel from complete disaster with Pentium-M based on the P3 core.

Re:Now I feel old.

[Wootery]

Didn't they think of that when they were planning the standardisation process? This stuff clearly wasn't public knowledge (or it would be prior art re. patents), so surely they knew this was a risk...

Re:Now I feel old.

[kelarius]

I keep reading posts on here talking about how god damned expensive RD-RAM was in the early 2000s and I just don't seem to remember it ALWAYS being that expensive, especially in summer '02. The first computer I ever built for myself using my high school grad money cost me $800, had a 2GHz P4 Northwood, and 512MB PC800 RD-RAM (in 2x256MB config no less), and I can even remember the RAM not being the most expensive part of that build (that was the CPU and the $150 graphics card I put in there). Now maybe I just happened to find a fantastic deal, or maybe some people arent remembering correctly, and yes, I do remember the price of RD-RAM tripling not a year later, but it wasn't always so bad. I always attributed the price increase to a drop off in production to absolutely nothing as everyone jumped off of the P4/RD-RAM ship. Best part is that that computer is still chugging away nicely at my Dad's house, still using the original motherboard, CPU, and those same 256MB RD-RAM sticks.

Re:Now I feel old.

[Vitriol Angst]

I think that this analogy is like explaining the birds and the bees with actual birds and bees. It's like pea soup as an analogy for patents -- strained. So the lawsuit is like a 10 yard penalty, illegal use of the hands -- but they actually gain the yardage in a game 12 years later?

I also think that RAMBUS is a horrible name for a football team.

Re:Now I feel old.

[unixisc]

RAMBUS was about as superior to SDRAM as VLIW was to RISC. Both technologies that Intel endorsed at the time, but was forced to back out of later.

Re:Now I feel old.

[tlhIngan]

Why was there "no prior art" if Rambus simply took the JEDEC committee's ideas and created patenst? There should be VOLUMES of prior art from the JEDEC meeting minutes to people's scribbles in their notepads. You're seriously telling me that Rambus was the only company attending the meetings to take notes? And they managed to get their patents ramrodded through before anyone on the JEDEC committee decided to take their first note? Ya, no.

No, what happened is Rambus has the patents already applied for. They then influenced the standards committee to use the patented items without telling them of the patent applications.

It's not a case of Rambus patenting what was discussed in secret. It's a case of Rambus intentionally steering towards the use of patented things without notifying them (per requirements of the committee) of that pending patent application.

And that's where the "wrongness" comes in - Rambus knew that they were intentionally deceiving the standards committee into using patented technology that they did not disclose.

Standards committees using patents are fine. It's when the participants intentionally mislead the committee into choosing technology that's patented without disclosing the fact that gets everyone twisted up.

Re:Now I feel old.

[bored]

I keep reading posts on here talking about how god damned expensive RD-RAM was in the early 2000s and I just don't seem to remember it ALWAYS being that expensive, especially in summer '02

The price of RD-RAM came in line with DDR once the P4 got a chipset that could support DDR (which was after the athlon, which was already laying down the smack). Which of course pointed at the underlying agenda... Locking up who could manufacture the technology so the profit margins could be controlled. (see socket 8 licensing fiasco, itanium, etc)

IIRC the price of rambus memory literally halved in a month or two. Plus, people started talking about the latency issues, and the power consumption. By middle of 02, IIRC the price difference was basically non existent, and everyone was looking forward to DDR2.

Re:Now I feel old.

[Dogtanian]

You missed a step in the P3 -> Core, which was the Pentium M. Intel was pretty much forced to build it, because power hungry P4's sucked in laptops.

I remember at the time (mid-2000s?) *before* the original Core line came out, at least one article in Personal Computer World magazine extolled the virtues of the Pentium M. They quite seriously suggested it was worth considering for use in a desktop system. (IIRC, there were desktop motherboards that supported the Pentium M, the only issue was that you had to take more care with the heatsink and cooling than you would with a Pentium 4- or something like that).

Re:Now I feel old.

[bill_mcgonigle]

Rambus knew that they were intentionally deceiving the standards committee into using patented technology that they did not disclose.

How is Rambus not liable for fraud in this case? Surely there must be more complexity to it?

Re:Now I feel old.

[viperidaenz]

I'm writing this on a laptop with a 1.6GHz Pentium M CPU. Speed wise, its about equivalent to a P4 2.4GHz, most due to (I think) a shorter pipeline and 2MB of full speed cache.

It still works pretty well. Struggles a bit now with high bitrate h264 though.
Not bad for a laptop from 2004.

Hot Grits Indeed!

[rueger]

Wow. I can remember the days when this was a meme that all of the cool kids tossed out while laughing with derision!

Re:Hot Grits Indeed!

[KlomDark]

Six digits? What a bunch of n00bs... ;0

Re:Hot Grits Indeed!

[Mr Z]

Kids these days!

Now where's that naked, petrified Natalie Portman with the goatse link?

This is horrible news

[RelliK]

This is absolutely disgusting. Rambus is the ultimate patent troll. For those not familiar, here is some history.

Back in the 90's, Rambus became a member of JEDEC, an industry consortium of RAM manufacturers. JEDEC rules require members to disclose any patents that are relevant to the technology being discussed. Rambus didn't. It sat in on the meetings, listened, and modified its pending patent applications to cover DDR RAM. After the new RAM standard was adopted, Rambus surfaced its submarine patents and started suing everyone left and right.

Add to that the fact is that Rambus itself does not manufacture anything -- it's a technology licencing house that has a few engineers and an army of lawyers -- and you get a perfect example of a patent troll.

Re:This is horrible news

[Soporific]

Why wouldn't the companies being sued use that information in their defense then? Or rather, why didn't it work?

~S

Re:This is horrible news

[Ralph Wiggam]

The other companies won a few rounds of the lawsuits based on that, and I want to say Rambus was actually found to have committed fraud at one point. But Rambus won enough rounds of lawsuits to keep bringing in money.

Re:This is horrible news

[RelliK]

Because US patent system effectively legalizes extortion. Some manufacturers rolled over and paid "licencing", but others, like Infineon decided to fight. The jury in Infineon v. Rambus ruled in favor of Infineon, but then Rambus appealed to the federal circuit and got that decision overturned. PJ, of groklaw fame, put it thusly: "federal circuit has never seen a patent it didn't like".

This is just a short summary, you can find more info if you like. Rambus extorted money from pretty much everyone who has so much as touched RAM.

Re:This is horrible news

[radarskiy]

a) The patents were originally filed before Rambus joined JEDEC.
b) The patents were amended after Rambus left JEDEC.

Re:This is horrible news

[ArbitraryName]

They did not manufacture RAM at any point in time. What you have was made by a licensee.

Re:This is horrible news

[jonwil]

This is exactly why the patent system needs to change to make it impossible to amend patents after they have been filed.

Re:This is horrible news

[Pinhedd]

Add to that the fact is that Rambus itself does not manufacture anything -- it's a technology licencing house that has a few engineers and an army of lawyers -- and you get a perfect example of a patent troll.

ARM Holdings PLC is an R&D company that doesn't manufacture anything. They generate all of their revenue through licencing IP to third parties whom in turn do the manufacturing. I do not see anyone calling them a patent troll.

Rambus Inc. is definitely a shady corporation but simply failing to manufacture products with IP that one owns and initiating litigation against those who infringe upon it does not automatically make one a patent troll.

Re:This is horrible news

[Pinhedd]

The ARM microarchitecture is hardly complex. It's very simple, employing only minor variations on technology that predates it by decades. Most computer engineers would call the bulk of the ARMv7 microarchitecture and other ARM IP obvious to one skilled in the trade. In fact, almost all straight RISC microprocessors follow a very similar architecture. ARM Holdings, despite manufacturing nothing, has wielded their patent portfolio offensively against companies and even bought a large stake in a consortium that acquired MIPS patents that ARM most likely infringed upon. What ARM has done very well is package it all together in a very verbose and detailed IP portfolio that can be instantiated, simulated, and fabricated with ease by licencors. I imagine that many of their patents could be challenged successfully, but no one has any motivation to do so.

You will not get any argument from me about Rambus being a patent troll, but it's important to remember that while usefulness is a requirement for patentability, marketability is not.

Re:This is horrible news

[MachineShedFred]

a) The patents were originally filed before Rambus joined JEDEC.
b) The patents were amended after Rambus left JEDEC.

s/left/was kicked out of/

Dead Patents

[Anonymous Coward]

How can Rambus now 'licence' dead patents? Surely now the company must be wound up and consigned to the trash can of history.

Re:Dead Patents

[Desler]

Because they still have the pay the judgement of the case.

Re:Dead Patents

[tlhIngan]

How can Rambus now 'licence' dead patents? Surely now the company must be wound up and consigned to the trash can of history.

Actually, it is in trouble these days - neither the PS4 nor the Xbone use RDRAM anymore. Before that, Rambus made a ton of money for RDRAM inside the PS2 (the best selling console in the world, period) and XDR-DRAM in the PS3.

But now the PS4 uses GDDR5, and the Xbone uses regular DDR3.

Speed is not patented

[Tablizer]

The same judge also ruled that the wheel "appears to have prior art, per Babylonians." Romans gotta pay up.

Consider it a test you only have to pass once.

[VortexCortex]

You'd think for the amount of money they steal someone would just shoot them in the head or hire someone to do it.

With what?! Ballistic weapons? Oh, that's rich! Consider that if any aliens exist there's at least a 50% chance they're more advanced than humans. If the aliens can manage interstellar space travel then they've already mastered the warp drive. Imagine all the technology they must have. You think they'd care about being threatened with micrometeors? Ooh, how terrifying!

One answer to the Fermi Paradox could be that the aliens can't visit you because they'll invalidate every damn patent on Earth with their prior art and destabilize the world economy humans built upon laws that create artificial scarcity of information and ideas. If only humans could extract their craniums from their rectums and stop treating infinitely reproducible information and knowledge as if they are physical things... Until human laws agree with the basic Universal truth that symbolic configurations of matter and energy are not matter and energy themselves, the Drake equation will never be solved.

Think about it. Economics 101 says that as supply tends towards infinity, price tends towards zero regardless of demand or cost to create. The elements that make up sand and water were forged in a huge expensive furnace: A 1-A supernova, yet sand is cheap, you just pay for hauling it, not the sand itself. You wouldn't sell Ice to Eskimos, as the locals say, but selling infinitely reproducible information to humans possessing "sentient" brains and digital information replicators is acceptable? The world economy of ideas and information can't even grok economics 101. It's the work to create new configurations that's scarce, not the 1's and 0's or ideas. You have no evidence that patents and copyright are beneficial -- Not a single one of you tested that damn hypothesis!

The Patent Trolls are what you pre-information-scarcity races deserve. Why, I wouldn't be surprised if the "trolls" weren't actually aliens risking violation of the prime directive just to try and help you humans out by demonstrating exactly how insane your current patent and copyright system is. Imagine yourself in their shoes! Imagine that you risked being re-assigned as an overseer of primordial ooze for the next billion years just to directly tell the humans via world wide neural network of their folly, and they STILL didn't get it?! Imagine how you would feel if you went through the trouble to do all that, and the Humans STILL sided with the moronic laws?! If you travelled 23 light millennia just to stop in for a spot of tea and solve every problem human science will face for the next few centuries, but you discovered the "Best and Brightest" humans embracing an economy that's incompatible with the nature of information and is based on Untested Hypotheses would YOU trust them with a Warp Drive?! I wouldn't!

Re:Consider it a test you only have to pass once.

[brantondaveperson]

Why, I wouldn't be surprised if the "trolls" weren't actually aliens risking violation of the prime directive just to try and help you humans out by demonstrating exactly how insane your current patent and copyright system is.

Really? Because I'd be fucking astonished.