DCT
0:24-cv-62093
T52 Ltd v. Citrix Systems Inc
Key Events
Amended Complaint
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: T52 Ltd. (England and Wales)
- Defendant: Citrix Systems, Inc. (Delaware)
- Plaintiff’s Counsel: The Concept Law Group, P.A.; DiNovo Price LLP
- Case Identification: 0:24-cv-62093, S.D. Fla., 12/08/2025
- Venue Allegations: Venue is alleged to be proper in the Southern District of Florida because Defendant maintains a regular and established place of business in the district, has transacted business there, and has allegedly committed acts of patent infringement within the district.
- Core Dispute: Plaintiff alleges that Defendant’s virtualization products, which enable remote access to applications and desktops, infringe a portfolio of eight U.S. patents related to efficient remote graphics processing, video compression, and Graphics Processing Unit (GPU) sharing.
- Technical Context: The technology at issue addresses performance bottlenecks in delivering graphically intensive applications from centralized servers to remote users, a foundational challenge in the fields of virtual desktop infrastructure (VDI), cloud gaming, and remote work.
- Key Procedural History: This is a Second Amended Complaint. Plaintiff alleges Defendant had knowledge of the asserted patent family as early as April 2011, when the family’s PCT application was cited during the prosecution of one of Defendant’s own European patent applications. The complaint also details multiple pre-suit notice letters sent to Defendant in August 2013, July 2016, and October 2022, which may be relevant to allegations of willful infringement.
Case Timeline
| Date | Event |
|---|---|
| 2002-03-01 | Earliest Priority Date for all Patents-in-Suit |
| 2011-03-29 | U.S. Patent No. 7,916,147 Issues |
| 2011-04-04 | Plaintiff alleges Citrix learned of patent family via citation in its own patent prosecution |
| 2011-12-20 | U.S. Patent No. 8,081,192 Issues |
| 2012-06-19 | U.S. Patent No. 8,203,568 Issues |
| 2013-06-18 | U.S. Patent No. 8,466,922 Issues |
| 2013-08-14 | Citrix allegedly receives first notice letter regarding patents |
| 2015-08-18 | U.S. Patent No. 9,113,146 Issues |
| 2015-08-25 | U.S. Patent No. 9,117,285 Issues |
| 2016-07-29 | Citrix allegedly receives second notice letter regarding patents |
| 2016-08-23 | U.S. Patent No. 9,424,621 Issues |
| 2017-12-26 | U.S. Patent No. 9,852,490 Issues |
| 2022-10-31 | Citrix allegedly receives third notice letter regarding patents |
| 2025-12-08 | Second Amended Complaint Filed |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 7,916,147 - "Centralised Interactive Graphical Application Server"
The Invention Explained
- Problem Addressed: The patent addresses the technical challenges of running interactive graphics applications (like video games) on a central server for remote users. Specifically, it identifies the high computational cost and latency associated with traditional video compression techniques, which rely on "brute force" pixel-by-pixel comparisons to detect changes between frames. (Compl. ¶¶58, 60; ’147 Patent, col. 2:39-49). Prior art solutions were also noted to require specialized hardware or specially written applications, limiting their commercial viability. (Compl. ¶¶62-64).
- The Patented Solution: The invention proposes a method to make compression more efficient by analyzing the underlying graphics instructions before an image is rendered, rather than analyzing the final pixels. (Compl. ¶67). The system intercepts graphics instructions, generates a second, modified set of instructions, and processes this second set to create "compression assistance data" (e.g., motion vectors) that informs the encoder how to compress the final rendered image more efficiently. (’147 Patent, Abstract; Compl. ¶69). This allows the system to leverage knowledge of the synthetic image's composition to bypass computationally intensive pixel analysis. (’147 Patent, col. 21:10-17).
- Technical Importance: This approach sought to reduce latency and processing overhead in remote graphics systems, making them more commercially viable with standard hardware components like GPUs. (Compl. ¶¶65, 70).
Key Claims at a Glance
- The complaint asserts independent claim 1 and apparatus claim 21.
- Claim 1 (Method):
- intercepting a first set of instructions for a graphics processor module;
- generating a second set of instructions for the module;
- processing the first and/or second set of instructions to generate first graphics data (image frames);
- processing the second set of instructions to generate second graphics data;
- processing the second graphics data to generate compression assistance data;
- processing the first graphics data using the compression assistance data to generate a compressed video signal;
- wherein generating the second set involves analyzing the first set to determine which instructions are useful for generating the compression assistance data. (Compl. ¶105).
- The complaint reserves the right to assert additional dependent claims. (Compl. ¶106).
U.S. Patent No. 8,081,192 - "Centralised Interactive Graphical Application Server"
The Invention Explained
- Problem Addressed: The ’192 Patent, part of the same family as the ’147 Patent, addresses the same technical problem of inefficient compression in remote graphics delivery systems that rely on computationally expensive pixel-level analysis. (Compl. ¶¶58, 60).
- The Patented Solution: The solution is conceptually similar to the ’147 Patent, focusing on analyzing graphics commands to aid compression. The method involves intercepting graphics instructions from software, analyzing them to determine which are useful for generating compression assistance data, generating a second set of instructions based on that analysis, and then using that data to generate the final compressed video signal. (’192 Patent, Claim 1; Compl. ¶140). This method avoids the "brute force" approach of comparing rendered pixels frame-by-frame. (Compl. ¶68).
- Technical Importance: The invention aims to improve the performance and reduce the cost of centralized systems for delivering interactive graphics applications. (Compl. ¶65).
Key Claims at a Glance
- The complaint asserts independent claim 1.
- Claim 1 (Method):
- intercepting graphics instructions outputted by graphics generating software;
- analyzing the instructions to determine which are useful for generating compression assistance data;
- generating a second set of instructions responsive to the analysis;
- processing the original or second set of instructions to generate graphics data (image elements);
- processing the second set of instructions to generate compression assistance data;
- processing the graphics data using the compression assistance data to generate the compressed video signal. (Compl. ¶140).
- The complaint reserves the right to assert additional dependent claims. (Compl. ¶141).
U.S. Patent No. 8,203,568 - "Sharing a Graphical Processing Unit Between a Plurality of Programs"
- Technology Synopsis: This patent addresses the inefficiency of dedicating a single GPU to a single remote graphics session. It discloses a method for a single GPU to be shared concurrently by multiple programs, processing instructions for each, storing their respective output frames in different memory locations, and compressing and transmitting each frame to its respective remote device. (Compl. ¶¶74-75, 165).
- Asserted Claims: Independent claim 1 is asserted. (Compl. ¶165).
- Accused Features: The complaint accuses the GPU-sharing and virtualization features of Defendant's Citrix Hypervisor product. (Compl. ¶¶79, 86).
U.S. Patent No. 8,466,922 - "Centralised Interactive Graphical Application Server"
- Technology Synopsis: This patent discloses a method for generating compressed image signals for multiple programs by determining a difference between the graphics instructions output by those programs. This difference is then used to generate compression assistance data, which facilitates the separate compression and transmission of image data for each program to different remote devices. (Compl. ¶190).
- Asserted Claims: Independent claim 1 is asserted. (Compl. ¶190).
- Accused Features: The complaint accuses Defendant’s compression assistance and processing functionalities, such as those in HDX and Thinwire technologies. (Compl. ¶¶79, 82-83).
U.S. Patent No. 9,113,146 - "Centralised Interactive Graphical Application Server"
- Technology Synopsis: This patent claims a method focused on identifying a specific graphics instruction that has an "influence on a co-ordinate" of the image data. Responsive to this identification, the system generates compression assistance data containing information about the pixel at that co-ordinate, which is then used to generate the compressed image data. (Compl. ¶215).
- Asserted Claims: Independent claim 1 is asserted. (Compl. ¶215).
- Accused Features: The complaint accuses Defendant's compression assistance and processing functionalities, particularly the "Selective H.264" feature that identifies and encodes "actively changing regions." (Compl. ¶¶79, 83).
U.S. Patent No. 9,117,285 - "Centralised Interactive Graphical Application Server"
- Technology Synopsis: This patent claims a system (rather than a method) that performs a similar function to the ’146 patent. The system comprises an "instruction interception module" and a graphics processor, and operates by identifying an intercepted instruction that influences a co-ordinate and generating compression assistance data (CAD) or instructions (CAI) related to that pixel co-ordinate to produce compressed image data. (Compl. ¶240).
- Asserted Claims: Independent claim 1 is asserted. (Compl. ¶240).
- Accused Features: The complaint accuses Defendant’s overall virtualization systems incorporating Hypervisor and HDX/Thinwire technologies. (Compl. ¶79).
U.S. Patent No. 9,424,621 - "Centralised Interactive Graphical Application Server"
- Technology Synopsis: This patent claims a method where graphics instructions from multiple programs are actively modified (e.g., by adding, removing, or replacing instructions). These modifications are performed "to facilitate the generation of the... compressed image data transmissions," which are then processed by a GPU and sent to separate remote devices. (Compl. ¶274).
- Asserted Claims: Independent claim 1 is asserted. (Compl. ¶274).
- Accused Features: The complaint accuses Defendant's compression, processing, and GPU-sharing functionalities. (Compl. ¶79).
U.S. Patent No. 9,852,490 - "Centralised Interactive Graphical Application Server"
- Technology Synopsis: This patent claims a system featuring a "graphics instruction modification module (GIMM)." The system modifies graphics instructions and then processes the resulting pixels to determine differences between two points in time, thereby facilitating compression before transmission. (Compl. ¶299). This suggests an instruction-based method for generating motion-related data.
- Asserted Claims: Independent claim 11 is asserted. (Compl. ¶299).
- Accused Features: The complaint accuses Defendant’s overall virtualization systems incorporating Hypervisor and HDX/Thinwire technologies. (Compl. ¶79).
III. The Accused Instrumentality
Product Identification
- The accused instrumentalities include Citrix Hypervisor (also known as XenServer), Citrix DaaS (formerly Citrix Virtual Apps and Desktops), and products incorporating HDX and Thinwire technologies. (Compl. ¶79).
Functionality and Market Context
- The accused products provide virtual desktop infrastructure (VDI) and virtual application capabilities, allowing users to access and interact with applications running on remote servers. (Compl. ¶79). The complaint highlights two core functionalities:
- Compression Assistance: The HDX graphics engine, Thinwire, is alleged to use a "Selective H.264" mode that "sense[s] regions of transient content (fluid images or video) and encodes it." (Compl. ¶83). This functionality avoids re-encoding static portions of a display, thereby reducing bandwidth and processing load. The complaint provides a screenshot from Defendant's product documentation showing a policy setting to "Use video codec when preferred," which allegedly enables this infringing mode for "actively changing regions." (Compl. ¶85, p. 25).
- GPU Sharing: Citrix Hypervisor is alleged to enable GPU Virtualization, which "allows one physical GPU to be used by multiple VMs concurrently." (Compl. ¶86). This feature is marketed as enabling resource optimization and greater performance with fewer physical graphics cards. (Compl. ¶86). The complaint includes a screenshot from Defendant's documentation showing a user interface for creating "vGPU enabled VMs." (Compl. ¶87, p. 27).
- The complaint alleges these products are central to the market for remote work and cloud services, a field that has seen "enormous growth." (Compl. ¶57).
IV. Analysis of Infringement Allegations
’147 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| a) intercepting a first set of instructions for said graphics processor module, said first set of instructions relating to how to render image frames; | Defendant's HDX/Thinwire technology analyzes graphics content to categorize it for encoding, which necessitates intercepting the graphics instructions that produce the display. | ¶¶82, 83 | col. 9:1-5 |
| b) generating a second set of instructions for said graphics processor module; | The process of analyzing the display to sense "actively changing regions" is alleged to correspond to the generation and processing of a second set of instructions to identify areas for compression. | ¶83 | col. 4:45-50 |
| c) processing said first set of instructions ... to generate first graphics data, said first graphics data comprising image frames; | The Accused Instrumentalities render a virtual desktop or application display (the first graphics data) for the end-user. | ¶79 | col. 10:15-18 |
| d) processing said second set of instructions in said graphics processor module to generate second graphics data; | The "sensing" of transient content is alleged to be an intermediate analysis step corresponding to the generation of "second graphics data" from which compression information is derived. | ¶83 | col. 4:50-51 |
| e) processing said second graphics data to generate compression assistance data; | The output of the analysis is the identification of "selected" or "transient" regions, which constitutes the compression assistance data used to guide the encoder. | ¶83 | col. 4:52-54 |
| f) processing said first graphics data using said compression assistance data to generate a compressed video data signal; | The Thinwire encoder uses the identification of "actively changing regions" to encode only those specific regions with a video codec like H.264, generating the final compressed signal. | ¶83 | col. 4:55-58 |
| wherein the generation of said second set of instructions comprises analyzing said first set of instructions to determine which instructions of said first set of instructions are useful... | The "Selective H.264" feature analyzes the visual output of the graphics instructions to determine which parts of the screen are changing and therefore useful for generating compression data. | ¶83 | col. 21:38-44 |
’192 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| a) intercepting graphics instructions outputted by graphics generating software; | Defendant's HDX/Thinwire technology analyzes graphics content, which requires intercepting the instructions that create said content. | ¶¶82, 83 | col. 9:1-5 |
| b) analyzing the graphics instructions to determine which of the graphics instructions are useful for generation of compression assistance data; | The "Selective H.264" technology analyzes the display to "sense regions of transient content," thereby determining which underlying instructions are causing changes useful for compression. | ¶83 | col. 21:38-44 |
| c) generating a second set of instructions responsive to the results of the analysis; | The complaint does not provide sufficient detail for analysis of this specific element. | ||
| d) processing the graphic instructions ... to generate graphics data comprising one or more image elements; | The Accused Instrumentalities render the virtual desktop or application display for the user. | ¶79 | col. 10:15-18 |
| e) processing at least a portion of the second set of instructions to generate compression assistance data; | The analysis of "actively changing regions" results in data that guides the compression process. | ¶83 | col. 4:52-54 |
| f) processing at least a portion of the graphics data using at least a portion of the compression assistance data to generate the compressed video data signal. | The Thinwire encoder uses the knowledge of which regions are changing to apply video compression only to those areas, creating the final compressed signal. | ¶83 | col. 4:55-58 |
Identified Points of Contention
- Scope Questions: A central question may be whether Citrix’s alleged analysis of "display data" or "transient content" (Compl. ¶¶82-83) constitutes "analyzing ... graphics instructions" as required by the claims. A court may need to determine if analyzing a rendered frame buffer is distinct from analyzing the stream of API calls or commands that generated it, and whether the claims cover the former, the latter, or both.
- Technical Questions: The complaint alleges that identifying "actively changing regions" meets the claim limitation of generating "compression assistance data." (Compl. ¶83). The patents' specifications, however, provide examples of such data that include "motion vector data." (’147 Patent, col. 4:40-41). A factual dispute may arise over whether simply identifying a "dirty region" is technically equivalent to the more detailed "compression assistance data" described in the patent specifications.
V. Key Claim Terms for Construction
The Term: "compression assistance data" (’147 Patent, Claim 1(e))
- Context and Importance: This term defines the output of the patented analytical process and the key input to the improved compression step. Its scope is critical; if construed broadly, it could cover any data that aids compression (e.g., a list of changed screen regions), while a narrow construction might require more specific information (e.g., motion vectors). Practitioners may focus on this term because the infringement allegation hinges on equating Citrix's identification of "actively changing regions" (Compl. ¶83) with this claimed data.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim language itself does not explicitly limit the term, defining it functionally as data generated from "second graphics data" that is then used to "generate a compressed video data signal." (’147 Patent, Claim 1).
- Evidence for a Narrower Interpretation: The patent specification provides specific, and potentially limiting, examples. It states that "compression assistance data includes motion vector data" and lists other examples such as "what form of encoding to use on the objects" and "whether or not to use overlapped motion estimation." (’147 Patent, col. 4:38-55; Compl. ¶71). The abstract also explicitly mentions extracting "motion vectors."
The Term: "analyzing said first set of instructions" (’147 Patent, Claim 1)
- Context and Importance: This term is the core of the inventive concept, distinguishing it from prior art "brute force" pixel comparison. (Compl. ¶¶60, 68). The dispute may turn on what is being analyzed. Is it the stream of API calls from an application, or the rendered output that results from those calls? The infringement case appears to depend on this analysis covering Citrix's alleged analysis of "display data." (Compl. ¶82).
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The term "analyzing" is not qualified, and one could argue that analyzing the direct output of instructions is a form of analyzing the instructions themselves.
- Evidence for a Narrower Interpretation: The patent describes an "instruction interception module" that intercepts instructions sent to a graphics processing unit. (’147 Patent, Fig. 6; Abstract). This architecture suggests an analysis of the instructions in transit before or parallel to rendering, which could support a narrower interpretation that excludes post-rendering framebuffer analysis.
VI. Other Allegations
Indirect Infringement
- The complaint alleges inducement of infringement based on Defendant’s marketing materials, product documentation, and user interfaces that allegedly instruct and encourage customers to enable and use the accused functionalities, such as the "Selective H.264" and GPU-sharing features. (Compl. ¶¶81, 84-87, 119). Contributory infringement is alleged on the basis that the accused features are specially made for an infringing purpose and have no substantial non-infringing use. (Compl. ¶¶99, 127).
Willful Infringement
- Willfulness is alleged based on Defendant’s purported knowledge of the patents. The complaint pleads specific instances of alleged pre-suit knowledge, including three separate notice letters sent in 2013, 2016, and 2022, as well as knowledge derived from Defendant's own patent prosecution activities where the patents' parent application was cited as prior art as early as 2011. (Compl. ¶¶40-47, 50).
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of technical implementation: Does Citrix's method of identifying "actively changing regions" by analyzing "display data" operate by "analyzing... graphics instructions" as required by the claims, or does it analyze a post-rendering artifact in a way that is technically distinct from the patented invention? The case may depend on evidence revealing the precise point in the graphics pipeline where Citrix's analysis occurs.
- A second central question will be one of definitional scope: Can the term "compression assistance data," which is exemplified in the patent specification with specific data types like "motion vectors," be construed to cover the identification of screen regions for selective encoding, or is there a fundamental mismatch in the type and granularity of the data being claimed versus the data being used?
- For the patents related to GPU sharing (e.g., the ’568 patent), a key question will be whether Citrix's virtualization architecture, which manages GPU access for multiple virtual machines, meets the specific claim limitations regarding how frames for different programs are stored in "different GPU accessible memory locations" and subsequently compressed and transmitted to distinct remote devices.