Last week, the U.S. Court of Appeals for the Federal Circuit issued an opinion in Intel Corp. v. PACT XPP Schweiz AG, No. 22-1037 (Fed. Cir. 2023), reversing and remanding a PTAB finding that substantial evidence showed the prior art did not teach the segment-to-segment claimed limitation and there was no motivation to combine prior art references.
Intel Corporation (“Intel”) petitioned for inter partes review of claims 4 and 5 of PACT XPP Schweiz AG’s (“PACT”) U.S. Patent No. 9,250,908 (“’908 Patent”). Intel asserted that it would have been obvious to a person of ordinary skill in the art (“POSA”) to replace the Kabemoto reference’s secondary caches with the Bauman reference’s segmented global secondary cache and that the ’908 Patent would have been obvious in light of such a combination. The PTAB disagreed and found that the prior art failed to disclose the claimed segment-to-segment limitation of the ’908 Patent and that a POSA would not have been motivated to combine the teachings of Kabemoto and Bauman. Intel appealed.
The CAFC reversed and remanded. The CAFC noted that Intel had shown that the segment-to-segment limitation was present in the Bauman reference as Intel’s cited illustrations showed that each cache segment was clearly connected to its neighboring cache segment. As such, the CAFC held that the PTAB’s finding that the Bauman reference failed to teach the segment-to-segment limitation lacked substantial evidence.
Additionally, the CAFC held the PTAB’s rejection of Intel’s “known-technique” rational as the basis for a motivation to combine the Bauman and Kabemoto references lacked substantial evidence. The CAFC noted that the motivation-to-combine analysis is a flexible one and that if there’s a known technique to address a known problem using prior art elements according to their established functions, then there is a motivation to combine. As such, the CAFC found that using a global, segmented secondary cache (as in Bauman) has been used to improve cache coherency in multiprocessor systems, and a POSA would have recognized that such a system could be used to improve similar multiprocessor systems such as those found in Kabemoto. Importantly, the CAFC rejected the PTAB’s suggestion that the Bauman system must be an “improvement” to the Kabemoto system and held it was sufficient that the Bauman system was a “suitable option” to replace the Kabemoto system. Thus, the CAFC reversed and remanded.