DCT

1:18-cv-01173

Realtime Adaptive Streaming LLC v. Advanced Micro Devices Inc

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 1:18-cv-01173, D. Colo., 05/15/2018
  • Venue Allegations: Venue is alleged based on Defendant AMD having regular and established places of business within the District of Colorado and having committed acts of infringement in the district.
  • Core Dispute: Plaintiff alleges that Defendant’s Graphics Processing Units (GPUs) and Accelerated Processing Units (APUs), which incorporate its Video Coding Engine, infringe patents related to adaptive data and video compression systems.
  • Technical Context: The technology involves systems that dynamically select different data compression algorithms to optimize performance based on system conditions such as throughput or bandwidth, a key function in modern video encoding and streaming.
  • Key Procedural History: Post-issuance proceedings for the asserted patents raise significant questions regarding the viability of certain claims. Inter Partes Review (IPR) certificates attached to U.S. Patent No. 8,934,535 indicate that asserted Claim 15 has been disclaimed. Similarly, an IPR certificate for U.S. Patent No. 9,769,477 indicates that asserted Claim 1 has been cancelled. The complaint does not address these post-grant outcomes.

Case Timeline

Date Event
2001-02-13 Priority Date for ’046, ’535, and ’477 Patents
2008-06-10 ’046 Patent Issued
2015-01-13 ’535 Patent Issued
2017-09-19 ’477 Patent Issued
2018-05-15 Complaint Filing Date

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 7,386,046 - “Bandwidth Sensitive Data Compression and Decompression”

The Invention Explained

  • Problem Addressed: The patent identifies a fundamental trade-off in data compression between execution speed and compression efficiency, noting that algorithms achieving high compression ratios are often slow, and vice versa. It also describes how the limited data retrieval bandwidth of mass storage devices can create performance bottlenecks in computer systems (Compl. ¶1; ’046 Patent, col. 1:26-54).
  • The Patented Solution: The invention proposes a system with a controller that monitors the "throughput" of a data compression system. When the controller determines that throughput has fallen below a set threshold, indicating a bottleneck, it commands the system to switch to a different, faster compression algorithm to increase throughput and clear the backlog (’046 Patent, Abstract; Fig. 1).
  • Technical Importance: This adaptive approach allows a system to dynamically balance the need for speed against the desire for storage efficiency, prioritizing throughput when the system is under load and compression ratio when it is not (’046 Patent, col. 8:1-4).

Key Claims at a Glance

  • The complaint asserts independent claim 40 (Compl. ¶12).
  • The essential elements of Claim 40 are:
    • A data compression system for compressing and decompressing data.
    • A plurality of selectively utilized compression routines, including at least a first and second compression algorithm.
    • A controller for tracking throughput and selecting a compression routine based on that throughput.
    • The tracking of throughput comprises tracking a number of pending access requests to a storage device.
    • When the controller determines throughput falls below a predetermined threshold, it commands the data compression engine to use a routine providing a faster rate of compression to increase the throughput.
  • The complaint reserves the right to assert other claims of the patent (Compl. ¶20).

U.S. Patent No. 8,934,535 - “Systems and methods for video and audio data storage and distribution”

The Invention Explained

  • Problem Addressed: As a continuation of the ’046 patent, the ’535 patent addresses the same core issues of managing the trade-off between compression speed and efficiency, particularly in the context of data storage and retrieval systems where bandwidth can be a limiting factor (’535 Patent, col. 1:45-67).
  • The Patented Solution: The invention describes a method where a "parameter or attribute" of a data block (e.g., data type, expected access frequency) is determined. Based on this parameter, one or more "asymmetric compressors"—defined as compressors where compression and decompression times differ significantly—are selected from a plurality of available options to compress and then store the data block (’535 Patent, Abstract; col. 9:4-24).
  • Technical Importance: This allows a system to apply different compression strategies tailored to the data's characteristics and intended use, such as using a slow, high-ratio compression for archival data that is written once but read many times (’535 Patent, col. 14:12-24).

Key Claims at a Glance

  • The complaint asserts independent claim 15 (Compl. ¶30). An associated Inter Partes Review Certificate (US 8,934,535 K1) indicates that claims 15-30 of the patent have been disclaimed.
  • The essential elements of Claim 15 are:
    • Determining a parameter of at least a portion of a data block.
    • Selecting one or more asymmetric compressors from a plurality of compressors based on the determined parameter.
    • Compressing the portion of the data block with the selected compressor(s) to create compressed data blocks.
    • Storing at least a portion of the compressed data blocks.
  • The complaint reserves the right to assert other claims of the patent (Compl. ¶38).

U.S. Patent No. 9,769,477 - “Video data compression systems”

  • Patent Identification: U.S. Patent No. 9,769,477, titled “Video data compression systems,” issued on September 19, 2017 (Compl. ¶9).
  • Technology Synopsis: The ’477 patent, also in the same family, discloses a system with multiple different asymmetric data compression encoders. A processor is configured to determine one or more data parameters related to the throughput of a communications channel and, based on these parameters, select one or more of the encoders to compress video or image data (’477 Patent, Abstract).
  • Asserted Claims: The complaint asserts independent claim 1 (Compl. ¶48). An associated Inter Partes Review Certificate (US 9,769,477 K1) indicates that claims 1-29 of the patent have been cancelled.
  • Accused Features: The accused features are AMD’s Video Coding Engine and its alleged use of the H.264 standard and Scalable Video Coding to select between different encoders (e.g., CAVLC or CABAC) based on parameters like bitrate or resolution to adapt to communication channel throughput (Compl. ¶¶49, 52).

III. The Accused Instrumentality

Product Identification

  • The accused products are AMD's Video Coding Engine (VCE), which the complaint alleges is a hardware accelerator integrated into all of AMD’s GPUs and APUs, including the AMD Radeon R9, R7, R5, and HD series graphics cards (Compl. ¶¶11, 29, 47).

Functionality and Market Context

  • The complaint alleges that the VCE is a fixed-function hardware accelerator that supports the H.264/AVC video compression standard, including its Scalable Video Coding (SVC) extension (Compl. ¶11). The system is alleged to operate by determining parameters such as bitrate or resolution, which correspond to H.264 "profiles" and "levels," and then selecting between different compression encoders, such as Context-Adaptive Variable Length Coding (CAVLC) and Context-Adaptive Binary Arithmetic Coding (CABAC), based on those parameters (Compl. ¶¶15, 17, 19). The complaint includes a system overview diagram from an AMD user guide illustrating a "Video Compression Engine" performing "H.264 AVC encoding" (Compl. p. 7, Fig. 6.1).

IV. Analysis of Infringement Allegations

7,386,046 Patent Infringement Allegations

Claim Element (from Independent Claim 40) Alleged Infringing Functionality Complaint Citation Patent Citation
a data compression system for compressing and decompressing data input AMD’s Video Coding Engine, which allegedly performs H.264 video compression. ¶14 col. 8:57-59
a plurality of compression routines selectively utilized by the data compression system, wherein a first one...includes a first compression algorithm and a second one...includes a second compression algorithm The system's alleged use of different H.264 profiles and associated encoders, such as CAVLC and CABAC. ¶15 col. 8:60-62
a controller for tracking throughput and generating a control signal to select a compression routine based on the throughput A controller alleged to exist in the accused products that decides which compression technique (e.g., CABAC, CAVLC) to use based on channel bandwidth. ¶19 col. 8:63-65
wherein said tracking throughput comprises tracking a number of pending access requests to a storage device The complaint recites this limitation but factually alleges the controller makes decisions based on "current or anticipated throughput" such as "channel bandwidth." ¶19 col. 8:66-67
wherein when the controller determines that the throughput falls below a predetermined throughput threshold, the controller commands the data compression engine to use one of the plurality of compression routines to provide a faster rate of compression so as to increase the throughput The complaint alleges that the system selects lower or higher quality streams based on low or high bandwidth situations, respectively, to adapt to throughput. ¶19 col. 28:4-8
  • Identified Points of Contention:
    • Scope Questions: A central dispute may arise over the term "tracking throughput." Claim 40 explicitly defines this term as "tracking a number of pending access requests to a storage device." The infringement theory, however, is based on adapting to "channel bandwidth" (Compl. ¶19). This raises the question of whether the accused functionality meets the specific definition of the key term as recited in the claim itself.
    • Technical Questions: What evidence does the complaint provide that the accused system's alleged logic—selecting a lower quality stream for lower bandwidth—performs the specific function of commanding a "faster rate of compression" for the express purpose of increasing throughput, as required by the claim?

8,934,535 Patent Infringement Allegations

Claim Element (from Independent Claim 15) Alleged Infringing Functionality Complaint Citation Patent Citation
determining a parameter of at least a portion of a data block The system's alleged determination of bitrate, resolution, or Group of Pictures (GOP) structure. ¶32, ¶33 col. 21:55-56
selecting one or more asymmetric compressors from among a plurality of compressors based upon the determined parameter or attribute The alleged selection between CAVLC and CABAC encoders based on the H.264 profile that corresponds with the determined parameter. ¶34 col. 21:57-60
compressing the at least the portion of the data block with the selected one or more asymmetric compressors to provide one or more compressed data blocks The act of using the selected encoder (CAVLC or CABAC) to compress video data into compressed data blocks. ¶36 col. 21:61-64
storing at least a portion of the one or more compressed data blocks Storing the resulting compressed data in memory, such as buffers or a hard disk. ¶37 col. 21:65-66
  • Identified Points of Contention:
    • Scope Questions: A threshold legal question is whether the assertion of Claim 15 can proceed, given that the provided IPR certificate indicates it has been disclaimed.
    • Technical Questions: Does the accused system's alleged selection between standard-defined encoders (CAVLC and CABAC) based on pre-set H.264 profiles constitute the claimed act of "selecting... asymmetric compressors based upon the determined parameter," or is it merely an implementation of a pre-existing industry standard?

V. Key Claim Terms for Construction

For the ’046 Patent:

  • The Term: "tracking throughput"
  • Context and Importance: The construction of this term will be critical. The infringement allegation rests on the accused system monitoring "channel bandwidth," while Claim 40 explicitly defines "tracking throughput" as "tracking a number of pending access requests to a storage device." Practitioners may focus on this term because the defendant will likely argue that this explicit definition in the claim language is controlling and that the accused functionality does not meet it.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The patent’s background section refers more generally to compressing data based on the "throughput (bandwidth) of a system" (’046 Patent, col. 1:17-19).
    • Evidence for a Narrower Interpretation: Claim 40 itself contains the limiting definition: "wherein said tracking throughput comprises tracking a number of pending access requests to a storage device" (’046 Patent, col. 28:1-3). Such a definitional statement within the body of a claim provides strong evidence for a narrower construction.

For the ’535 Patent:

  • The Term: "asymmetric compressors"
  • Context and Importance: This term is central to defining the invention of the ’535 patent. The complaint alleges that standard H.264 encoders like CAVLC and CABAC are "asymmetric compressors" (Compl. ¶34). The dispute may turn on whether these standard components, selected as part of a pre-defined H.264 profile, function as the claimed "plurality of compressors" from which a selection is made based on a determined parameter.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The specification defines an asymmetric algorithm as one where "the execution time for the compression and decompression routines differ significantly," a definition that could potentially encompass CAVLC and CABAC (’535 Patent, col. 9:4-14).
    • Evidence for a Narrower Interpretation: The claim structure implies an active selection process ("selecting one or more...from among a plurality") based on a determined parameter, which may suggest a more dynamic or bespoke selection than simply implementing a profile from an industry standard.

VI. Other Allegations

  • Indirect Infringement: For all asserted patents, the complaint alleges induced infringement. The basis for inducement includes AMD’s alleged affirmative acts of providing "training, demonstrations, brochures, installation and user guides," which allegedly instruct customers to use the accused products in their "normal and customary way" to infringe the patents, particularly by using the H.264 standard and its Scalable Video Coding features (Compl. ¶¶24, 42, 60).
  • Willful Infringement: The complaint alleges that AMD has had knowledge of the patents and their infringement since at least the filing of the lawsuit. It further alleges that AMD acted with "willful blindness to the probability, that the induced acts would constitute infringement," which may form a basis for alleging willful infringement, particularly for post-suit conduct (Compl. ¶¶23-24, 41-42, 59-60).

VII. Analyst’s Conclusion: Key Questions for the Case

  • A threshold issue for both the ’535 and ’477 patents will be one of procedural viability: can infringement counts based on asserted claims that have been disclaimed (Claim 15 of the ’535 patent) or cancelled (Claim 1 of the ’477 patent) in post-grant proceedings be maintained?
  • For the ’046 patent, a core issue will be one of definitional scope: can the term "tracking throughput," explicitly defined in Claim 40 with reference to "pending access requests to a storage device," be construed to cover the complaint's factual allegations, which center on adapting to "channel bandwidth"?
  • A key evidentiary question across all asserted patents will be one of inventive distinction: does the accused products' alleged implementation of the H.264 video compression standard, including the selection of pre-defined profiles and their associated encoders like CAVLC and CABAC, constitute the specific, adaptive selection methods recited in the claims, or is it simply the practice of a pre-existing industry standard?