In VLSI Technology LLC v. Intel Corporation, No. 22-1906 (Fed. Cir. 2023), VLSI sued Intel for infringement of U.S. Patent Nos. 7,523,373 (the “’373 patent”) and U.S. Patent No. 7,725,759 (the “’759 patent”). After a jury trial in 2021, VLSI was awarded $1.5 billion for literal infringement of the ’373 patent and an additional $675 million for infringement of the ’759 patent under the doctrine of equivalents. Prior to trial, Intel sought to amend its answer to assert a licensing defense. Intel argued that due to a recent change in ownership of Finjan, Inc., Intel was covered under Finjan’s license with VSLI because both Intel and Finjan were under the control of Fortress Investment Group LLC. The district court denied Intel’s motion. Intel appealed the infringement verdicts as well as the district court’s denial of the motion to amend.

Issues

The Federal Circuit considered several issues on appeal:

1. Whether substantial evidence supported that Intel literally infringed the ’373 patent?
2. Whether substantial evidence supported that Intel infringed the ’759 patent under the doctrine of equivalents?
3. Whether the district court abused its discretion by denying Intel’s challenge that VLSI’s damages expert departed from an economically sound methodology?
4. Whether the district court abused its discretion by denying Intel’s late-addition of a licensing defense?

Holdings and Reasoning

1. Substantial evidence supported the Jury’s verdict that Intel literally infringed the ’373 patent.

The Federal Circuit dispensed with Intel’s arguments with respect to infringement of the ’373 patent in relatively short order. Citing to both expert testimony and internal Intel documents, the Federal Circuit explained that there is substantial evidence for the jury to have found that Intel’s RING_RETENTION_VOLTAGE is the minimum operating voltage of the C6 SRAM, and thus Intel’s accused products infringe the ’373 patent.

Intel also argued that the claims of the ’373 patent require that falling below the minimum operating voltage be the causal trigger for switching from one voltage source to another. However, the Federal Circuit noted that this is an argument for claim construction, and when a claim phrase is not construed, a court may defer to the jury’s view of the claim phrase unless that view is contrary to the evidence presented. Finding no such contrary evidence presented by Intel and adequate support by VLSI’s expert to support the jury’s findings, substantial evidence existed to support the jury’s verdict of infringement.

2. VLSI’s evidence supporting its theory of infringement under the doctrine of equivalents fails as a matter of law.

Intel argued that VLSI’s evidence with respect to the doctrine of equivalents was legally insufficient to find infringement of the ’759 patent. The doctrine of equivalents expands the scope of liability for infringement where an accused product falls outside the literal scope of the claim elements, but nonetheless matches the function, way, and result of the asserted claim limitations. That is, where there are insubstantial differences between the accused product and the asserted claim, a finding of infringement under the doctrine of equivalents may be proper.

The ’759 patent claims a “first master device” that sends a “request to change clock frequency” to a “programmable clock controller.” The Federal Circuit explained that in order for the doctrine of equivalents to apply, VLSI had to show that Intel’s accused product, “the core and certain code which resides on the power control unit, together perform substantially the same function, in substantially the same way, to achieve substantially the same result as the claimed ‘first master device’ and that certain other instructions on the same power control unit perform substantially the same function, in substantially the same way, to achieve substantially the same result as the claimed ‘programable clock controller.’”

However, VLSI’s expert testimony on the issue did not contain any “meaningful explanation of why the way in which the request is made by Intel’s accused products is substantially the same as what the claim prescribes.” It is not enough for VLSI’s expert to chalk up any difference between the accused product and the asserted claim to a simple “design choice.” As explained by the Federal circuit, “[t]hat label does not indicate whether, or begin to explain why, the options in the choice are substantially different or substantially the same.” Here, the Federal Circuit is looking for particularized evidence “linking arguments as to the insubstantiality of the differences between the claimed invention and the accused device.” Finding none, the Federal Circuit found that VLSI’s doctrine of equivalents theory failed as a matter of law.

3. The district court abused its discretion by denying Intel’s challenge that VLSI’s damages expert departed from an economically sound methodology.

VLSI’s damages expert undertook a reasonable royalty analysis based on a hypothetical negotiation which took into consideration the incremental technical benefit attributable to Intel’s infringement, i.e., the value proposition of VLSI’s technology. However, in calculating the power savings results attributable to the asserted patents–the alleged technical benefit–VLSI’s expert included products that did not use the infringing functionality of the asserted patents. The Federal Circuit reasoned that such an error could not be considered harmless with respect to the damages calculation and remanded the damages case back to the district court for further adjudication.

4. The district court abused its discretion by denying Intel’s motion for leave to amend to add a license defense.

The district court denied Intel’s motion for leave to amend based on untimeliness of the motion, prejudice to VLSI of allowing the amendment, and futility of the proposed defense. With respect to timeliness, the Federal Circuit reasoned that a delay of approximately three months between the time Fortress acquired Finjan and the time Intel filed its motion for leave to amend was not sufficient undue delay to require denial of the motion, particularly in light of Intel’s diligence in complying with the notice requirements of the license.

Similarly, with respect to prejudice, the Federal Circuit found no basis on which prejudice alone could support denial of the motion for leave to amend. For example, the district court did not explain how or why having a district court judgment sooner rather than later could help VLSI secure licenses. Without a meaningful explanation of the prejudice to VLSI of allowing amendment, or why denying the motion would not prejudice Intel, the district court abused its discretion in denying the motion for leave to amend predicated on prejudice.

Finally, while the district court found that the license defense was untenable under Delaware law (the applicable law under the choice of law provision in the license agreement), the Federal Circuit noted that Delaware case law does not present a sufficiently clear answer to the issue of whether Intel falls under the license, particularly based on the limited analysis of the issue provided in the briefings on the motion for leave to amend. Based on that limited analysis, and the district court’s misinterpretation of Delaware law, the Federal Circuit held that it was error to deny the motion to add the license defense to the case.