Rambus’ vs. Infineon
Introduction. Rambus has suffered three major losses in its patent infringement suit against Infineon:
- March 15, 2001. After the Markman hearing, the judge accepted Infineon’s interpretation of eight key words in Rambus’ patents;
- May 9, 2001. The jury returned a verdict that Rambus engaged in fraud.
- August 9, 2001. The judge denied Rambus’ motions for new trial and awarded attorneys’ fees to Infineon.
Questions Raised and Conclusions.
- What is Rambus’ maximum liability to Infineon
- Prediction: Rambus’ financial liability will remain around $7.5 million;
- What is the likelihood of reversal of the Markman ruling on appeal?
- Prediction: The court’s Markman decision will be upheld
- What does the outcome in Infineon portend for Rambus, long term?
- Prediction: Rambus will have short, medium and long-term difficulty enforcing its patents in accordance with its original business plan.
Damages. The Court awarded Infineon nearly $7.5 million:
1- $1.00 in nominal damages for Rambus’ fraud against JEDEC;
2- $350,000 in punitive damages (reduced from a jury verdict of $3.5 million)
3- $7.15 million for legal expenses.
Because the award is not based on Infineon’s antitrust nor its RICO counterclaims, the $7.5 million award is not subject trebling. The $7.5 million award is not likely to be reduced on appeal because:
- Legal expenses are awarded only when the judge makes a “discretionary” finding that the case is “exceptional”;
- To reverse the finding that the case is exceptional, the appeals court would have to reverse both the Markman ruling and the ruling that Rambus engaged in fraud;
- Because the jury verdict established fraud as a matter of “fact”, the fraud decision will not be reversed on appeal.
Markman Claim Interpretation. The judge interpreted eight terms appearing in 57 claims of four patents that Rambus charged Infineon with infringing. The terms are “bus,” “block size,” read request,” “write request,” “transaction request,” “first external clock signal,” “second external clock signal,” and “integrated circuit device”.
The judge adopted Infineon’s views in a thorough, well-reasoned, and legally supported memorandum opinion. The court appears to have carefully considered Rambus’ arguments and supporting legal authorities. It took pains to distinguish the conflicting cases and rationalize conflicting authorities. Most importantly, the court adhered to the correct hierarchy of intrinsic and extrinsic evidence prescribed by the court of appeals: the claims, the specification, the prosecution history and outside sources such as expert testimony. The court ultimately concluded that Rambus’ arguments were “sophistry.” The court went through this exercise with each of the contested terms. With each term, the court adopted an interpretation that made it virtually impossible for Infineon to infringe.
“De Novo” Review Markman rulings are subject to “de novo” review on appeal. This means that the appeals court owes no particular deference to the trial court’s determination and can re-interpret the claims from scratch. Although anything can happen on appeal, the trial court’s interpretation of the claims is on firmer ground to withstand appellate review than most cases, because:
- The appeals court is unlikely to reverse the interpretation of all eight terms;
- The jury’s fraud determination makes Rambus an unsympathetic appellant.
Trial Proceedings. As a result of the Markman ruling:
- Rambus withdrew one of the patents in suit;
- After Rambus presented its evidence, the judge ruled as a matter of law that Infineon did not infringe Rambus’ patents;
- Infineon withdrew its defenses of invalidity and unenforceability
- Infineon also withdrew
- its declaratory judgment counterclaims and
- its monopolization counterclaim.
• After Infineon’ evidence was introduced, the judge ruled in favor of Rambus as a matter of law,
-
- On Infineon’s breach of contract counterclaim
- On Infineon’s attempted monopolization counterclaim
On May 9, the jury returned a verdict:
- Finding actual and constructive fraud involving the JEDEC “SDRAM” and “DDR SDRAM” standards;
- Awarding nominal damages of $1.00; and
- Awarding punitive damages of $3.5 million, which the court reduced to $350,000.
No New Trial. On August 9, 2001, the court declined to grant Rambus’ motion for a new trial, overruled the jury’s determination of constructive fraud and upheld the juries determination of actual fraud. In so doing the court probably inflicted permanent damage to Rambus. The court’s ruling makes the following highly prejudicial, negative comments about Rambus’ business strategy, truthfulness and litigation strategy in support of the jury’s verdict:
Business Strategy: “Infineon presented evidence in the form of Rambus’ business plans and testimony from which the jury could have concluded that the patent infringement suit was part and parcel of Rambus’ fraud….Infineon proved that, because of Rambus’ fraudulent behavior while it was a member of JEDEC, Infineon designed the JEDEC compliant products that were alleged to have infringed the patents- in-suit.
“When Rambus instituted the final step in its fraudulent plan by charging that those patents were infringed by JEDEC-compliant SDRAM products, Rambus caused Infineon to expend substantial sums of money to investigate the merits of, and defend against, the infringement claims.”
“[T]he jury’s verdict leaves no doubt that Rambus committed intentional fraud, with the goal of inducing the JEDEC committee to create a standard which used Rambus’ patented technology so that Rambus could claim a royalty on the sales of products made to the specifications of the standard. Rambus also intended to sue to enforce these patents, believing that the cost of litigation and the risk-of-loss factor would pressure manufacturers into accepting licenses.”
Lack of Credibility. The Court found that Rambus’ CEO Tate, its in-house patent counsel, and an expert witness not only lacked credibility, but were in some cases downright deceitful and possibly perjurious.
Rambus’ Litigation Misconduct: “Rambus should have known… before jury selection began … that Infineon’s products could not be shown to infringe the remaining patents in suit and therefore, the suit was baseless. Rather than acknowledge the futility of its position, preserve its rights to appeal and proceed to defend the counterclaims, Rambus obdurately pressed claims it knew it could not win. That, and the fact that its claim construction from the outset was based on extrinsic evidence so basically at odds with the intrinsic evidence, makes this an exceptional case.”
“The litigation misconduct asserted by Infineon fits into four general categories: (1) the failure to reveal relevant documents during discovery; (2) inaccurate and, in some instances, outright untruthful testimony from Rambus representatives; (3) inaccurate and improper responses to interrogatories and requests for admissions (“RFAs”) during discovery; and (4) the destruction of litigation-related documents”.
“The tactics Rambus chose to employ made it very expensive for Infineon to uncover and prove the fraud…. Because of the document destruction (made, it should be remembered, in anticipation of litigation)… the documents produced omitted the documents that revealed, or pointed the way to, the truth. When the crime- fraud ruling forced disclosure of those documents, they told a story far different than reflected in Rambus’ initial discovery responses and depositions; and that, of course, necessitated another round of depositions where, with the aid of the new documents, the truth began to emerge”.
Rambus’ Future.
First, Rambus has admitted to the court in the Micron case that if the Infineon Markman ruling survives appeal, Rambus cannot sustain its infringement claim against Micron. Rambus has therefore requested a suspension of that case pending appeal of the Infineon decision.
Second, although Verdict does not know the terms of Rambus’ license agreements with other chip makes (which were negotiated under threatened and actual litigation), the fraud findings probably relieve most licensees of their obligation to pay royalties. Counsel for the licenses monitored the Infineon case, and the chip makers may now simply refuse to pay royalties.
Third, longer term, it is difficult to see a clear path for Rambus to overcome the complete discrediting of Rambus’ business plan, witnesses and litigation strategy. This puts the future of Rambus’ licensing program at significant risk.
Loose Ends.
On October 8, 2001, the court held an evidentiary hearing concerning the scope of the injunction. Oral argument on the injunction is scheduled for November 2, 2001. Verdict will attend the hearing if requested.
