RAMBUS back in the legal limelight
Thrax
🐌Austin, TX Icrontian
The JEDEC Solid State Technology Association, composed of more than 300 high-profile companies, is perhaps best known for being the steering group behind DRAM technologies. On an invitation basis, influential semiconductor companies are invited to participate in the development or ratification of de facto standards. At the conclusion of these motions, such as in the case of SDRAM or DDR SDRAM, the technology is pushed as agreed upon by every member of JEDEC to such an extent that it becomes the technology to use in its given application.
Enter RAMBUS, to whom an invitation was extended in 1992. RAMBUS' entry could not have come at a more fortuitous time for the firm, as it helped Hynix, Qimonda (Infineon), Micron and Samsung bring SDRAM into the world. It was also during their four year tenure that RAMBUS played a large role in the development of DDR1. However, in 1995/1996 it became increasingly evident that RAMBUS was not prepared to comply with a "Reasonable and non-discriminatory" licensing policy that any firm who contributes technology to the standard must agree to. Therefore after four years of membership, RAMBUS stepped down from the organization to blaze its own (Legal) trail.
The development, ratification and standardization process at JEDEC is a long and involved process that requires the contribution of existing patents, and the development of new technologies to complete the picture. In addition, the meetings are not conducted in secret, as minutes and whitepapers are published on JEDEC's website. As a result of these insecurities, several key events played a part in RAMBUS' vast legal entanglement:
2000
RAMBUS begins filing lawsuits against the largest memory manufacturers at the time, with the claim that RAMBUS is the legal owner of SDRAM and DDR-SDRAM technologies. Seven memory firms, including Samsung, quickly settle with RAMBUS and agreed to pay royalties. RAMBUS next moves to Infineon, which is joined by Micron and Hynix in the fight. Countersuing with claims of fraud, RAMBUS v. Infineon goes to court under trial by jury. Amidst the proceedings, it is revealed that RAMBUS had been patenting the new technologies brainstormed in 1992-1996 JEDEC meetings used to "Complete the picture" of the emerging SDRAM and DDR-SDRAM standards. Found guilty of fraud by the jury, with the notion that the patents were filed in extremely bad faith, the claims against Infineon are dismissed.
2002
The United States FTC files charges against RAMBUS, alleging anti-trust violations. The FTC filing essentially declared that RAMBUs had employed underhanded "Patent Ambush." A patent ambush is a ploy by which a company deceives a standards body by neglecting to reveal it owns the patents to several key elements being employed in the developing standard. In the case of RAMBUS, these key elements were actually patented by RAMBUS while they were being brainstormed by the JEDEC body.
2003
FTC v. RAMBUS goes to trial, and is eventually dismissed in 2004 by the Chief Judge of Administrative Law for the FTC, Stephen J. McGuire. In McGuire's ruling, it is put forth that memory manufacturers had no viable alternatives to RAMBUS technologies. Similarly, McGuire asserts that the firms involved in the JEDEC fiasco were aware of RAMBUS' patent portfolio when talks over SDRAM and DDR-SDRAM were under way.
Also in this year, the 2001 ruling of fraud for the 2000 RAMBUS v Infineon trial is overturned by the Federal Court of Appeals. The US Supreme Court refuses to hear the new trial, and thusly the case is remanded back to Virginia.
2004
Infineon pleads guilty to price-fixing in an attempt to force RAMBUS out of the market. Hynix, Samsung, Elpida and Infineon are fined a collective sum of USD$730m for collusion.
2005
Unsatisfied with the results of their previous legal endeavors, RAMBUS files suit against Hynix, Nanya, Inotera, Micron and Infineon claiming that DDR 2, GDDR 2 and GDDR 3 chips employ elements of RAMBUS' patent portfolio. In March of this year, RAMBUS had its new and year 2000 claims against Infineon dismissed. At the settlement table, Infineon agrees to pay USD$5.9m quarterly licensing fees. In addition, legal wrangling between Infineon and RAMBUS ceases. The cases with Micron and Hynix remain in court.
2006
On the heels of McGuire's 2002 ruling which validated RAMBUS' legal maneuvering, the FTC files an appeal. On 2 August, 2006, the appeal is accepted and McGuire's ruling is overturned. The new FTC ruling states that section 2 of the Sherman Antitrust Act was violated in RAMBUS' attempt to illegaly monopolize the memory industry, and that RAMBUS deliberately attempted to deceive JEDEC which violates the FTC Act.
5 February, 2007
The US FTC limits the maximum royalties that RAMBUS may collect on its portfolio. As of 2010, RAMBUS can no longer collect on DDR-SDRAM or SDRAM. The ruling also limits the amount that can be obtained for memory controllers or components involved in the operation of DDR/SDRAM chips. RAMBUS has appealed this ruling and awaits a court date.
Which brings us to August of this year which has seen two significant developments in the long and sordid history:
Firstly, the FTC has ruled that RAMBUS has in fact deceived a standards-setting committee in a fraudulent fashion. The new FTC order stops RAMBUS from collecting royalties on US and foreign patents on goods being imported to or from the US. This maneuver has spurred EU regulators.
As of today (23 August, 2007), the European Union has mirrored the FTC's motion which grants relief to EU member states. The EU's Statement of Objections charges the memory firm with patent ambush, deception and fraud; the SO gives RAMBUS nine weeks to respond before significant fines are levied on the firm.
Enter RAMBUS, to whom an invitation was extended in 1992. RAMBUS' entry could not have come at a more fortuitous time for the firm, as it helped Hynix, Qimonda (Infineon), Micron and Samsung bring SDRAM into the world. It was also during their four year tenure that RAMBUS played a large role in the development of DDR1. However, in 1995/1996 it became increasingly evident that RAMBUS was not prepared to comply with a "Reasonable and non-discriminatory" licensing policy that any firm who contributes technology to the standard must agree to. Therefore after four years of membership, RAMBUS stepped down from the organization to blaze its own (Legal) trail.
The development, ratification and standardization process at JEDEC is a long and involved process that requires the contribution of existing patents, and the development of new technologies to complete the picture. In addition, the meetings are not conducted in secret, as minutes and whitepapers are published on JEDEC's website. As a result of these insecurities, several key events played a part in RAMBUS' vast legal entanglement:
2000
RAMBUS begins filing lawsuits against the largest memory manufacturers at the time, with the claim that RAMBUS is the legal owner of SDRAM and DDR-SDRAM technologies. Seven memory firms, including Samsung, quickly settle with RAMBUS and agreed to pay royalties. RAMBUS next moves to Infineon, which is joined by Micron and Hynix in the fight. Countersuing with claims of fraud, RAMBUS v. Infineon goes to court under trial by jury. Amidst the proceedings, it is revealed that RAMBUS had been patenting the new technologies brainstormed in 1992-1996 JEDEC meetings used to "Complete the picture" of the emerging SDRAM and DDR-SDRAM standards. Found guilty of fraud by the jury, with the notion that the patents were filed in extremely bad faith, the claims against Infineon are dismissed.
2002
The United States FTC files charges against RAMBUS, alleging anti-trust violations. The FTC filing essentially declared that RAMBUs had employed underhanded "Patent Ambush." A patent ambush is a ploy by which a company deceives a standards body by neglecting to reveal it owns the patents to several key elements being employed in the developing standard. In the case of RAMBUS, these key elements were actually patented by RAMBUS while they were being brainstormed by the JEDEC body.
2003
FTC v. RAMBUS goes to trial, and is eventually dismissed in 2004 by the Chief Judge of Administrative Law for the FTC, Stephen J. McGuire. In McGuire's ruling, it is put forth that memory manufacturers had no viable alternatives to RAMBUS technologies. Similarly, McGuire asserts that the firms involved in the JEDEC fiasco were aware of RAMBUS' patent portfolio when talks over SDRAM and DDR-SDRAM were under way.
Also in this year, the 2001 ruling of fraud for the 2000 RAMBUS v Infineon trial is overturned by the Federal Court of Appeals. The US Supreme Court refuses to hear the new trial, and thusly the case is remanded back to Virginia.
2004
Infineon pleads guilty to price-fixing in an attempt to force RAMBUS out of the market. Hynix, Samsung, Elpida and Infineon are fined a collective sum of USD$730m for collusion.
2005
Unsatisfied with the results of their previous legal endeavors, RAMBUS files suit against Hynix, Nanya, Inotera, Micron and Infineon claiming that DDR 2, GDDR 2 and GDDR 3 chips employ elements of RAMBUS' patent portfolio. In March of this year, RAMBUS had its new and year 2000 claims against Infineon dismissed. At the settlement table, Infineon agrees to pay USD$5.9m quarterly licensing fees. In addition, legal wrangling between Infineon and RAMBUS ceases. The cases with Micron and Hynix remain in court.
2006
On the heels of McGuire's 2002 ruling which validated RAMBUS' legal maneuvering, the FTC files an appeal. On 2 August, 2006, the appeal is accepted and McGuire's ruling is overturned. The new FTC ruling states that section 2 of the Sherman Antitrust Act was violated in RAMBUS' attempt to illegaly monopolize the memory industry, and that RAMBUS deliberately attempted to deceive JEDEC which violates the FTC Act.
5 February, 2007
The US FTC limits the maximum royalties that RAMBUS may collect on its portfolio. As of 2010, RAMBUS can no longer collect on DDR-SDRAM or SDRAM. The ruling also limits the amount that can be obtained for memory controllers or components involved in the operation of DDR/SDRAM chips. RAMBUS has appealed this ruling and awaits a court date.
Which brings us to August of this year which has seen two significant developments in the long and sordid history:
Firstly, the FTC has ruled that RAMBUS has in fact deceived a standards-setting committee in a fraudulent fashion. The new FTC order stops RAMBUS from collecting royalties on US and foreign patents on goods being imported to or from the US. This maneuver has spurred EU regulators.
As of today (23 August, 2007), the European Union has mirrored the FTC's motion which grants relief to EU member states. The EU's Statement of Objections charges the memory firm with patent ambush, deception and fraud; the SO gives RAMBUS nine weeks to respond before significant fines are levied on the firm.
0
Comments
Why can't news reporting always be so clear?
Oh yes, it's very interesting as well.
Thanks, I enjoyed the article.
I'd say that these are companies run by investors who hire lawyers. They have no product or service to speak of, but damned if they're not going to get their money's worth out of their company...
Rambus taught the industry how to make synchronous DRAM work under non disclosure agreements, and was invited to join JEDEC. On at least two occasions Rambus asked to present their technology for standardization, but were refused.
Instead, Rambus could see that JEDEC was ripping off their IP for use in sdram, and attended their last JEDEC meeting in 1995, and sent a withdrawal letter in June 1996. JEDEC didn't start working on DDR until December 1996 (from JEDEC's own witnesses in the FTC trial).
JEDEC only required presenters to disclose patents, and possibly applications. Rambus did disclose their first patent anyway, which had the same specifications as the public european patent in 1993. Member companies such as Mitsubishi and Hynix revied those patents, and determined that Rambus could, and probably would file additional patent claims that would cover SDRAM. Rambus told the SLDRAM committee and member companies knew about Ramubus' DDR patent claims, but chose to incorporate it into the standard without asking Rambus.
You talk about legal wranglings, but fail to point out that Rambus had a technology that the JEDEC companies said wouldn't work (high speed synchronous DRAM) yet now they claim that the inventions were obvious.
The fact of the matter is Rambus designed revolutionary products, and like a company that spends millions to develop a new drug, wants to be paid for it. You ignore that the real story is that a small company is being ripped off by large companies that want to use the technology for free.
The Court of Appeals - Federal Circuit said that Rambus did nothing wrong, and disclosed MORE than they needed to. They also said that the JEDEC disclosure policy had had a staggering lack of defining details, noting that JEDEC could have had a written a better policy, they simply did not.
The FTC's Chief Administrative law judge, Judge McGuire found entirely in Rambus' favor, noting that many of the FTC and JEDEC witnesses were at best lying, because written documents contradicted what they said the JEDEC disclosure policy and actions were.
The full FTC commission overturned Judge McGuire, stating essentially that you had to believe the witnesses, even if there was written proof that they were lying.
There is a story here, but you are completely missing it. Rambus figured out how to make synchronous memory transfer work, how to make double date rate transfer work, and the memory manufacturers want the technology for free. In fact, there seems to be plenty of evidence that they illegally colluded to control memory prices to drive Rambus out of business.
To those that say Rambus never invented anything, why does JEDEC continue to incorporate MORE of Rambus' patented technologies, like on-die termination, flex-phase, fly-by-wire, fully buffered dimms, etc. The simple answer is Rambus has always been ahead of the curve, but the DRAM companies don;t want Rambus (and Intel) telling them what to make.
With a court battle going for so long, with so much at stake, so much hard research, and so many thoroughly documented JEDEC meetings, surely the preponderance of factual truth has come out or will come out.
If RAMBUS was in the clear, surely it would've come out by now, no? I just want the best product in my PC produced from legitimate patents. Thus far, RAMBUS has failed to prove their legitimacy. However, that has not influenced my piece. I presented the rulings, and nothing more.
She and justmyopinion~ are part of a PR campaign/damage control. I bet both of them, probably part of team, are cruising the Internet looking for places to drop the opinions. Bah, talking heads with sole purpose of propping up their business (or the business they've contracted to).
And the icing on the cake is that I'm suppose to go read RAMBUS.com analysis of events, despite being accused of consulting competing press releases. This is a rich, rich irony.
Fine, but who is she?
She is either a PR hack or someone trying to prop up RAMBUS stock. If the former, OK, but disclosure would be the decent thing to do.
(READ: Another RAMBUS fangirl from InvestorVillage)
//Edit: It should be noted that someone copied/pasted the entire news article (quite rudely and illegally) onto InvestorVillage with a link to our main page. That's how several have found it.
It's an influence campaign. The stockholders: if they couldn't read the writing on the wall for last six years, too bad! If they were speculating on the stock, that's the way it goes. Something about eggs and a basket.
Bottom line is that "the FTC has ruled that RAMBUS has in fact deceived a standards-setting committee in a fraudulent fashion". I'm now more interested in the evidence that the FTC found that made them make that ruling.
Rob, you've done nothing wrong in providing the top level actions. The parties involved are going to be sensitive to the public reaction.
My comment in my previous post here was based on the fact that I haven't heard on the streets of RAMBUS doing anything much in the industry in quite a while- nothing major since RDRRAM and nothing much since they worked on some possibilities with AMD (over a year ago and I'm not sure where that went). But this is like anything else- it's not in the mainstream and I'll have to dig deeper to get the facts and get caught up. However, I'm now hearing of this.
And that's the trap here. We don't have the details presented. We don't know the reasons why the FTC ruled like they did- just that they did. We also know it is being challenged and we don't know what evidence is being brought to bear there either.
Until we do- it's going to be subjected to opinion based on what we do know- mostly in the mainstream.