`FOR THE DISTRICT OF TEXAS
`WACO DIVISION
`
`FG SRC LLC ,
`
`Plaintiff,
`
`v.
`
`INTEL CORPORATION,
`
`Defendant.
`
`Case No. 6:20-cv-00315-ADA
`
`JURY TRIAL DEMANDED
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT
`
`Plaintiff FG SRC LLC (“SRC”) files this First Amended Complaint for Patent
`
`Infringement (“Complaint”) against Defendant Intel Corporation (“Defendant” or “Intel”).
`
`Plaintiff alleges as follows:
`
`I. NATURE OF THE ACTION
`
`1. This is an action for infringement of U.S. Patent No. 7,149,867 (the “’867 patent”).
`
`2. SRC is a limited liability company incorporated in Delaware and is the successor to SRC
`
`Computers, LLC (“SRC Computers”).
`
`3. Defendant Intel is a corporation duly organized and existing under the laws of the State
`
`of Delaware, having a regular and established place of business in the Western District of
`
`Texas, including at 1300 S. Mopac Expressway, Austin, Texas 78746.
`
`II. JURISDICTION
`
`4. This action arises under the Patent Laws of the United States, 35 U.S.C. § 1, et seq.,
`
`including 35 U.S.C. §§ 271, 281, 283, 284, and 285. This is a patent infringement lawsuit over
`
`which this Court has subject matter jurisdiction under 28 U.S.C. §§ 1331 and 1338(a).
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 1
`
`
`
`5. This United States District Court for the Western District of Texas has general and
`
`specific personal jurisdiction over Defendant because Defendant is present in and transacts and
`
`conducts business in and with residents of this District and the State of Texas. Defendant has
`
`also purposefully and voluntarily availed itself of the privileges of conducting business in the
`
`United States, the State of Texas, and the Western District of Texas by continuously and
`
`systematically placing goods into the stream of commerce through an established distribution
`
`channel with the expectation that they will be purchased by consumers in Texas and this
`
`District.
`
`6. Defendant maintains regular and established places of business in the State of Texas and
`
`in the Western District of Texas.
`
`7. Plaintiff’s causes of action arise, at least in part, from Defendant’s contacts with and
`
`activities in the State of Texas and this District. Upon information and belief, Defendant
`
`committed acts of infringement in this District giving rise to this action and does business in
`
`this District, including making sales and/or providing services and support for its customers in
`
`this District. Defendant purposefully and voluntarily sold one or more of its infringing products
`
`with the expectation that they would be purchased by consumers in this District. These
`
`infringing products have been and continue to be purchased by consumers in this District.
`
`Defendant committed acts of patent infringement within the United States, the State of Texas,
`
`and the Western District of Texas.
`
`8. Defendant, directly and/or through intermediaries, uses, sells, offers for sale, ships,
`
`distributes, advertises, and/or otherwise promotes products in this District and the State of
`
`Texas. Defendant regularly conducts and solicits business in, engages in other persistent
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 2
`
`
`
`courses of conduct in, and/or derives substantial revenue from goods and services provided to
`
`residents of this District and the State of Texas.
`
`III. VENUE
`
`9. Venue is proper in this District under 35 U.S.C. § 1400(b) because: (1) Defendant has a
`
`physical place located in this District, (2) it is a regular and established place of business, and
`
`(3) it belongs to Defendant. See In re Cray Inc., 871 F.3d 1355, 1360 (Fed. Cir. 2017).
`
`10. Defendant maintains several facilities, which it refers to as campuses, in this District.
`
`https://www.intel.com/content/www/us/en/location/usa.html.
`
`11. Defendant maintains a campus at 1300 S. Mopac Expressway, Austin, Texas 78746.
`
`This is a regular and established place of business belonging to Defendant.
`
`12. Defendant maintains a campus at 6500 River Place Blvd, Bldg. 7, Austin, Texas 78730.
`
`This is a regular and established place of business belonging to Defendant.
`
`13. Defendant maintains a campus at 5113 Southwest Parkway, Austin, Texas 78735. This
`
`is a regular and established place of business belonging to Defendant.
`
`14. Defendant operates its Programmable Solutions Group (“PSG”) in this District.
`
`15. Members of Defendant’s PSG work in this District.
`
`16. Members of Defendant’s PSG work at one or more of its campuses in Austin, Texas.
`
`17. Defendant acquired Altera Corporation (“Altera”) in December 2015.
`
`18. Defendant purchased Altera for approximately $16.7 billion.
`
`19. Defendant acquired Altera at least in part because it was a “leading provider of field-
`
`programmable gate array (FPGA) technology.”
`
`https://www.sec.gov/Archives/edgar/data/50863/000119312515414642/d105836dex991.
`
`htm.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 3
`
`
`
`20. Altera is now part of Intel. Id.
`
`21. PSG was formed after Intel’s acquisition of Altera. Id.
`
`22. Altera was headquartered within the State of Texas at 3400 Waterview Parkway, #300,
`
`Richardson, Texas 75080.
`
`23. Altera maintained a regular and established place of business at 9442 N. Capital of
`
`Texas Hwy, #1-850, Austin, Texas 78759.
`
`24. Altera maintained a regular and established place of business at 5113 Southwest Pkwy,
`
`Austin, Texas 78735.
`
`25. As part of Intel, PSG creates programmable logic devices, including FPGAs.
`
`https://jobsearcher.com/j/system-validation-engineering-intern-at-intel-in-austin-tx-
`
`AZGgGE?utm_campaign=google_jobs_apply&utm_source=google_jobs_apply&utm_mediu
`
`m=organic.
`
`26. Intel was actively recruiting a System Validation Engineering Intern (the “Intern”), in
`
`this District, as of March 16, 2019. https://jobsearcher.com/j/system-validation-engineering-
`
`intern-at-intel-in-austin-tx-AZGgGE?utm_campaign=google_jobs_apply&utm_source=
`
`google_jobs_apply&utm_medium=organic.
`
`27. Intel was or is recruiting the Intern to work in its PSG group. Id.
`
`28. Intel was or is recruiting the Intern to work in its PSG group in Austin, Texas. Id.
`
`29. Intel was actively recruiting a Network Performance and Analytics Engineer (the
`
`“Network Engineer”) to work in this District as of April 11, 2020.
`
`https://jobsearcher.com/j/jr0132726-network-performance-and-analytics-engineer-at-intel-
`
`corporation-in-austin-texas-DDQVad.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 4
`
`
`
`30. One of the posted requirements for the Analytics Engineer is that said engineer has
`
`“Familiarity with Intel Processors, FPGA & NICs within a data center context.”
`
`https://jobsearcher.com/j/jr0132726-network-performance-and-analytics-engineer-at-intel-
`
`corporation-in-austin-texas-DDQVad.
`
`31. Intel was actively recruiting a Systems Performance and Analytics Engineer (the
`
`“Systems Engineer”) to work in this District as of April 11, 2020.
`
`https://jobsearcher.com/j/jr0132727-systems-performance-and-analytics-engineer-at-intel-
`
`corporation-in-austin-texas-LW6Vdl.
`
`32. One of the posted requirements for the Systems Engineer is that said engineer has
`
`“Familiarity with Intel Processors, FPGA & NICs within a data center context.”
`
`33. Intel was actively recruiting a Firmware Engineering Manager (the “Engineering
`
`Manager”), in this District, as of February 29, 2020. https://jobsearcher.com/j/firmware-
`
`engineering-manager-at-intel-corporation-in-austin-texas-
`
`2dBgdV?utm_campaign=google_jobs_apply&utm_source=google_jobs_apply&utm_medium
`
`=organic.
`
`34. Intel was or is recruiting the Engineering Manager to work in its PSG group. Id.
`
`35. Intel was or is recruiting the Engineering Manager to work in its PSG group in Austin,
`
`Texas. Id.
`
`IV. FG SRC LLC AND DEFENDANT’S PRODUCTS
`
`A. FG SRC LLC
`
`36. SRC Computers was co-founded by Seymour R. Cray, Jim Guzy, and Jon Huppenthal
`
`in 1996 to produce unique high-performance computer systems using Intel’s Merced
`
`microprocessor.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 5
`
`
`
`37. SRC is the successor to SRC Computers.
`
`38. Jim Guzy is a co-founder of Intel Corporation and served on Intel’s board for 38 years.
`
`39. Mr. Guzy was named to Forbes Midas List, which surveys the top tech deal makers in
`
`the world, in 2006 and 2007.
`
`40. Seymour Cray was an American electrical engineer and supercomputer architect who
`
`designed a series of computers that were the fastest in the world for decades.
`
`41. Mr. Cray has been credited with creating the supercomputing industry.
`
`42. Unfortunately, Mr. Cray died shortly after founding SRC Computers.
`
`43. But his legacy was carried on by Jon Huppenthal and a talented team of engineers that
`
`worked with Mr. Cray and Mr. Huppenthal for decades.
`
`44. SRC Computers’ focus was creating easy-to-program, general-purpose reconfigurable
`
`computing systems.
`
`45. In early 1997, Mr. Huppenthal and his team realized that the microprocessors of the
`
`day had many shortcomings relative to the custom processing engines that they were used to.
`
`46. As a result, they decided to incorporate dedicated processing elements built from Field
`
`Programmable Gate Arrays (“FPGAs”) and that idea quickly evolved into a novel system
`
`combining reconfigurable processors and CPUs.
`
`47. SRC Computers’ heterogenous system had 100x performance, 1/50th of the operating
`
`expense, 1/100th of the power usage, and required 1/500th of the space of more traditional
`
`computer systems.
`
`48. SRC Computers’ proven systems are used for some of the most demanding military
`
`and intelligence applications, including the simultaneous real-time processing and analysis of
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 6
`
`
`
`radar, flight and mission data collected from a variety of aerial vehicles in over 1,000 successful
`
`counter-terrorism and counter-insurgency missions for the U.S. Department of Defense.
`
`49. SRC Computers offered its first commercial product in 2015 called the Saturn 1 server.
`
`50. The Saturn 1 was 100 times faster than a server with standard Intel microprocessors
`
`while using one percent of the power.
`
`51. The Saturn 1 was designed to be used in HP’s Moonshot server chassis for data centers.
`
`52. SRC Computers has had over 30 U.S. patents issued for its innovative technology.
`
`53. SRC Computers’ patent portfolio covers numerous aspects of reconfigurable computing
`
`and has more than 2,090 forward citations.
`
`54. In February 2016, SRC Computers restructured into three new entities: a corporate
`
`parent FG SRC LLC, an operating company DirectStream, LLC (“DirectStream”), and a
`
`licensing entity SRC Labs, LLC.
`
`B. Accused Products
`
`55. In this complaint, Plaintiff accuses the following Intel products (collectively “Accused
`
`Products”) of infringing the ’867 patent:
`
`(a) Intel FPGA PACs (Programmable Acceleration Cards) including the Intel
`
`FPGA PAC D5005; Intel FPGA PAC N3000, and Intel FPGA PAC with
`
`Arria 10 GX FPGA;
`
`(b) Intel FPGA development kits including the Intel Stratix SoC Development
`
`Kit;
`
`(c) Agilex F-Series FPGA and SoC FPGA products including the AGF 004,
`
`AGF 006, AGF 008, AGF 012, AGF 014, AGF 022, and AGF 027;
`
`(d) Agilex I-Series SoC FPGA products including the AGI 022 and AGI 027;
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 7
`
`
`
`(e) Stratix 10 GX FPGA products including the GX 400, GX 500, GX 650, GX
`
`850, GX 1100, GX 1650, GX 2100, GX 2500, GX 2800, GX 1660, GX
`
`2110, GX 10M, GX 4500, and GX 5500;
`
`(f) Stratix 10 SX SoC FPGA products including the SX 400, SX 500, SX 650,
`
`SX 850, SX 1100, SX 1650, SX 2100, SX 2500, SX 2800, SX 4500, and SX
`
`5500;
`
`(g) Stratix 10 TX SoC FPGA products including the TX 400, TX 850, TX 1100,
`
`TX 1650, TX 2100, TX 2500, and TX 2800;
`
`(h) Stratix 10 MX FPGA products including the MX 1650 and MX 2100;
`
`(i) Stratix 10 DX SoC FPGA products including the DX 1100, DX 2100, and
`
`DX 2800;
`
`(j) Arria 10 GT FPGA products including the GT 900 and GT 1150;
`
`(k) Arria 10 GX FPGA products including the GX 160, GX 220, GX 270, GX
`
`320, GX 480, GX 570, GX 660, GX 900, and GX 1150;
`
`(l) Arria 10 SX SoC FPGA products including the SX 160, SX 220, SX 270, SX
`
`320, SX 480, SX 570, and SX 660;
`
`(m) Cyclone 10 GX FPGA products including the 10CX085, 10CX105,
`
`10CX150, and 10CX220;
`
`(n) Arria V GX FPGA products including the 5AGXA1, 5AGXA3, 5AGXA5,
`
`5AGXA7, 5AGXB1, 5AGXB3, 5AGXB5, and 5AGXB7;
`
`(o) Arria V GT FPGA products including the 5AGTC3, 5AGTC7, 5AGTD3,
`
`and 5AGTD7;
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 8
`
`
`
`(p) Arria V GZ FPGA products including the 5AGZE1, 5AGZE3, 5AGZE5,
`
`and 5AGZE7;
`
`(q) Arria V SX SoC FPGA products including the 5ASXB3 and 5ASXB5;
`
`(r) Arria V ST SoC FPGA products including the 5ASTD3 and 5ASTD5;
`
`(s) Cyclone 10 LP FPGA products including the 10CL006, 10CL010,
`
`10CL016, 10CL025, 10CL040, 10CL055, 10CL080, and 10CL120;
`
`(t) Cyclone V E FPGA products including the 5CEA2, 5CEA4, 5CEA5,
`
`5CEA7, and 5CEA9;
`
`(u) Cyclone V GX FPGA products including the 5CGXC3, 5CGXC4,
`
`5CGXC5, 5CGXC7, and 5CGXC9;
`
`(v) Cyclone V GT FPGA products including the 5CGTD5, 5CGTD7, and
`
`5CGTD9;
`
`(w) Cyclone V SE SoC FPGA products including the 5CSEA2, 5CSEA4,
`
`5CSEA5, and 5CSEA6;
`
`(x) Cyclone V SX SoC FPGA products including the 5CSXC2, 5CSXC4,
`
`5CSXC5, and 5CSXC6;
`
`(y) Cyclone V ST SoC FPGA products including the 5CSTD5 and 5CSTD6;
`
`and
`
`(z) Max 10 FPGA products including the 10M02, 10M04, 10M08, 10M16,
`
`10M25, 10M40, and 10M50.
`
`56. Each of the Accused Products includes an FPGA.
`
`57. In contrast to a purpose-built chip, which is designed with a single function in mind and
`
`then hardwired to implement it, an FPGA is more flexible.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 9
`
`
`
`58. With an FPGA, a large majority of the electrical functionality can be changed; more
`
`specifically, said functionality can be changed by the design engineer, changed during the PCB
`
`assembly process, or even changed after the equipment has been shipped to customers out in
`
`the field.
`
`59. FPGAs provide off-load and acceleration functions to CPUs, effectively speeding up
`
`the entire system performance.
`
`60. FPGAs provide benefits to designers of many types of electronic equipment, ranging
`
`from smart energy grids, aircraft navigation, automotive driver’s assistance, medical
`
`ultrasounds and data center search engines – just to name a few.
`
`61. Today’s FPGAs include on-die processors, transceiver I/O’s at 28 Gbps (or faster),
`
`RAM blocks, DSP engines, and more. More functions within the FPGA mean fewer devices
`
`on the circuit board, increasing reliability by reducing the number of device failures.
`
`62. FPGA functionality can change upon every power-up of the device.
`
`63. Programming an FPGA is a matter of connecting them up to create the desired logical
`
`functions (AND, OR, XOR, and so forth) or storage elements (flip-flops and shift registers).
`
`64. Unlike a CPU, which is essentially serial (with a few parallel elements) and has fixed-
`
`size instructions and data paths (typically 32 or 64 bit), an FPGA can be programmed to
`
`perform many operations in parallel, and the operations themselves can be of almost any
`
`width, large or small.
`
`65. The highly parallelized model in FPGAs is ideal for building custom accelerators to
`
`process computer-intensive problems.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 10
`
`
`
`66. Properly programmed, an FPGA has the potential to provide a 30x or greater speedup
`
`to many types of genomics, seismic analysis, financial risk analysis, big data search, and
`
`encryption algorithms and applications.
`
`67. Defendant’s customers can use FPGAs to accelerate its applications more than 30x
`
`when compared with servers that use CPUs alone.
`
`68. The speed increase is a result of the FPGAs handling computer-intensive, deeply
`
`pipelined, hardware-accelerated operations, which also allows for highly parallelized
`
`computing.
`
`V. MARKING AND NOTICE
`
`A. Marking and Constructive Notice to Defendant.
`
`69. SRC Computers complied with 35 U.S.C. § 287 by (i) placing the required notice on all,
`
`or substantially all, of its products made, offered for sale, sold, or imported into the United
`
`States, or (ii) providing actual notice to Defendant.
`
`70. For example, SRC Computers placed notices such as the following on all, or
`
`substantially all, of its products since at least February 19, 2013:1
`
`
`1 E.g., https://web.archive.org/web/20100930014237/http://www.srccomp.com/techpubs/
`patentedtech.asp.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 11
`
`
`
`
`71. The website listed in the notice, WWW.SRCCOMP.COM/
`
`TECHPUBS/PATENTEDTECH.ASP, stated the following:
`
`
`
`
`
`72. The website also listed at least the following patents since September 30, 2010. The ’867
`
`patent, asserted in this case, is highlighted:
`
`
`
`
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 12
`
`
`
`6,026,459
`
`6,076,152
`
`6,247,110
`
`6,295,598
`
`Patent # Patent Title
`System and method for dynamic priority conflict resolution in a multi-processor
`computer system having shared memory resources
`Multiprocessor computer architecture incorporating a plurality of memory
`algorithm processors in the memory subsystem
`Multiprocessor computer architecture incorporating a plurality of memory
`algorithm processors in the memory subsystem
`Split directory-based cache coherency technique for a multi-processor computer
`system
`Multiprocessor with each processor element accessing operands in loaded input
`buffer and forwarding results to FIFO output buffer
`
`6,339,819
`
`6,434,687
`
`System and method for accelerating web site access and processing utilizing a
`computer system incorporating reconfigurable processors operating under a single
`operating system image
`
`6,356,983
`
`6,594,736
`
`6,627,985
`
`6,781,226
`
`System and method providing cache coherency and atomic memory operations in
`a multiprocessor computer architecture
`System and method for semaphore and atomic operation management in a
`multiprocessor
`Reconfigurable processor module comprising hybrid stacked integrated circuit die
`elements
`Reconfigurable processor module comprising hybrid stacked integrated circuit die
`elements
`6,836,823 Bandwidth enhancement for uncached devices
`6,941,539 Efficiency of reconfigurable hardware
`Multiprocessor computer architecture incorporating a plurality of memory
`algorithm processors in the memory subsystem
`6,964,029 System and method for partitioning control-dataflow graph representations
`Process for converting programs in high-level programming languages to a
`unified executable for hybrid computing platforms
`System and method for providing an arbitrated memory bus in a hybrid
`computing system
`Computer system architecture and memory controller for close-coupling within a
`hybrid processing system utilizing an adaptive processor interface port
`System and method for explicit communication of messages between processes
`running on different nodes in a clustered multiprocessor system
`Reconfigurable processor module comprising hybrid stacked integrated circuit die
`elements
`7,134,120 Map compiler pipelined loop structure
`
`6,961,841
`
`6,983,456
`
`6,996,656
`
`7,003,593
`
`7,124,211
`
`7,126,214
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 13
`
`
`
`7,149,867
`
`7,155,602
`
`7,155,708
`
`7,167,976
`
`7,197,575
`
`7,225,324
`
`7,237,091
`
`7,282,951
`
`7,299,458
`
`7,373,440
`
`7,406,573
`
`7,421,524
`
`7,424,552
`
`7,565,461
`
`7,620,800
`
`System and method of enhancing efficiency and utilization of memory bandwidth
`in reconfigurable hardware
`Interface for integrating reconfigurable processors into a general purpose
`computing system
`Debugging and performance profiling using control-dataflow graph
`representations with reconfigurable hardware emulation
`Interface for integrating reconfigurable processors into a general purpose
`computing system
`Switch/network adapter port coupling a reconfigurable processing element to one
`or more microprocessors for use with interleaved memory controllers
`Multi-adaptive processing systems and techniques for enhancing parallelism and
`performance of computational functions
`Multiprocessor computer architecture incorporating a plurality of memory
`algorithm processors in the memory subsystem
`Reconfigurable processor module comprising hybrid stacked integrated circuit die
`elements
`System and method for converting control flow graph representations to control-
`dataflow graph representations
`Switch/network adapter port for clustered computers employing a chain of multi-
`adaptive processors in a dual in-line memory module format
`Reconfigurable processor element utilizing both coarse and fine grained
`reconfigurable elements
`Switch/network adapter port for clustered computers employing a chain of multi-
`adaptive processors in a dual in-line memory module format
`Switch/network adapter port incorporating shared memory resources selectively
`accessible by a direct execution logic element and one or more dense logic devices
`Switch/network adapter port coupling a reconfigurable processing element to one
`or more microprocessors for use with interleaved memory controllers
`Multi-adaptive processing systems and techniques for enhancing parallelism and
`performance of computational functions
`
`
`B. Actual Notice to Defendant.
`
`73. Intel learned of the ’867 patent and its infringement of said patent at least as a result of
`
`the filing and/or service of SRC’s Original Complaint for Patent Infringement, which included
`
`detailed claim charts illustrating how the Accused Products infringe the ’867 patent. This
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 14
`
`
`
`complaint additionally provides notice of the ’867 patent and infringement by the Accused
`
`Products.
`
`74. As discussed in more detail below, Intel knew of the ’867 patent before this lawsuit was
`
`filed.
`
`75. Between July 2015 and September 2015 SRC Computers and Intel met and
`
`corresponded regarding a potential acquisition by Intel of SRC Computers and/or its
`
`intellectual property (“IP”), including the ’867 patent. Persons involved on behalf of SRC
`
`Computers included SRC Computers’ chairman and controlling shareholder, Brandon
`
`Freeman, and Jon Huppenthal, President and CEO of SRC Computers. Persons involved on
`
`behalf of Intel included at least the following individuals: (1) Matthew Hulse, Associate
`
`General Counsel & Director, Patent Transactions Group, (2) Trina Van Pelt, Vice President
`
`and Managing Director, Strategic Transactions Group, (3) Steve Megli, Vice President and
`
`General Manager, Strategic Transactions Group, (4) Bob Nunn, Managing Director, Strategic
`
`Transactions Group, (5) Milan Djukic, Strategic Business Development, Strategic Transactions
`
`Group, (6) PK Gupta, Engineering Manager, Data Center Group, and (7) Sheldon Bernard,
`
`Strategic Business Development, Strategic Transactions Group.
`
`76. A third party, 3LP Advisors, LLC (“3LP”), assisted with discussions on behalf of SRC
`
`Computers.
`
`77. In order to assist Intel with reviewing SRC Computers’ patent portfolio, at SRC
`
`Computers’ request, 3LP prepared a presentation entitled “SRC IP Portfolio Overview,”
`
`which was provided to Intel on or around July 30, 2015.
`
`78. The presentation referenced the ’867 patent eight times. An excerpt from the
`
`presentation, in which highlighting has been added, follows.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 15
`
`
`
`
`
`79. Previously, at least between February 2006 and August 2013 SRC Computers and
`
`Altera met and/or corresponded on several occasions. During that time, on several occasions
`
`SRC Computers made Altera aware of its patent portfolio, which included the ’867 patent.
`
`Altera employees, including Mike Strickland and Martin Langhammer, participated. On
`
`information and belief, Martin Langhammer is currently a Principal Engineer at Intel. See
`
`https://www.linkedin.com/in/martin-langhammer-8056b17/?originalSubdomain=uk. On
`
`information and belief, Mike Strickland is currently a Senior Manager at Altera, which was
`
`acquired by Intel. See https://www.linkedin.com/in/strickland-michael-2715924/.
`
`80. On or around February 22, 2013, counsel for SRC Computers sent a notice letter to
`
`Altera advising that “Our client has recently become aware of Altera’s SoC FGPAs which are
`
`stated to integrate an ARM-based hard processor system (HPS) consisting of processor,
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 16
`
`
`
`peripherals, and memory interfaces with the FPGA fabric using a high-bandwidth interconnect
`
`backbone. From the information presently available to us, these devices may possibly involve
`
`SRC Computers’ patented technology.”
`
`81. On information and belief, Intel also knew of the ’867 patent as a result of its
`
`prosecution activities. More specifically, on information and belief, Intel learned of the ’867
`
`patent when prosecuting U.S. Patent Application No. 11/314,229 (the “’229 application),
`
`which was published as US2007/0143546. During prosecution of the ’229 application, the
`
`Examiner and Intel discussed US2004/0260884 (the ’867 publication”)—the publication
`
`corresponding to the ’867 patent—during an interview, and the Examiner explicitly referenced
`
`the ’867 publication in a rejection dated April 23, 2013. On information and belief, these
`
`prosecution activities would have brought the ’867 patent to Intel’s attention, particularly in
`
`light of the fact that the ’867 patent issued well before the interview.
`
`A. The ’867 Patent is Owned by SRC.
`
`VI. THE ’867 PATENT
`
`82. On January 22, 2020, DirectStream assigned the ’867 patent to SRC. The assignment
`
`was recorded with the USPTO on January 24 at Reel/Frame 051615/0344.
`
`83. All maintenance fees have been paid to the USPTO to keep the ’867 patent enforceable
`
`for its full term.
`
`B. Description of the Asserted Patent.
`
`84. The ’867 patent is entitled “System and method of enhancing efficiency and utilization
`
`of memory bandwidth in reconfigurable hardware” and issued on December 12, 2006.
`
`85. A true and correct copy of the ’867 patent is attached as Exhibit A.
`
`86. The ’867 patent is valid and enforceable.
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 17
`
`
`
`VII. COUNT ONE: DIRECT INFRINGEMENT OF THE ’867 PATENT
`
`87. Plaintiff incorporates by reference all paragraphs above as though set forth herein.
`
`88. Defendant has at no time, either expressly or impliedly, been licensed under the ’867
`
`patent.
`
`89. Defendant has and continues to directly infringe the ’867 patent by making, using,
`
`offering for sale, selling, and/or importing in or into the United States in violation of 35 U.S.C.
`
`§ 271(a) the Accused Products. For example, on information and belief Defendant tests,
`
`manufactures, and uses each of the Accused Products in an infringing manner at least in order
`
`to (1) ensure that functionality such as that appearing in SRC’s claim charts attached hereto,
`
`including but not limited to those portions of the chart describing descriptors appearing in its
`
`Embedded Peripherals User Guide, works as described and (2) provide support to its
`
`customers and partners, such as members of the FPGA Partner Program, which provides
`
`“[t]echnical support for product optimization.” See
`
`https://www.intel.com/content/dam/www/public/us/en/documents/flyers/intel-fpga-
`
`partner-program-flyer.pdf.
`
`90. Defendant’s direct infringement of the ’867 patent by the Accused Products has caused,
`
`and will continue to cause, substantial and irreparable damage to Plaintiff. Plaintiff is therefore
`
`entitled to an award of damages adequate to compensate for Defendant’s infringement, but not
`
`less than a reasonable royalty, together with pre- and post-judgment interest and costs as fixed
`
`by the Court under 35 U.S.C. § 284.
`
`91. Plaintiff adopts, and incorporates by reference, as if fully stated herein, Exhibits B
`
`through J, which are claim charts that describe and demonstrate how the Accused Products
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 18
`
`
`
`infringe exemplary claims of the ’867 patent. These charts collectively show that Intel infringes
`
`at least claims 1, 3, 4, 9, 11, and 12 of the ’867 patent.
`
`VIII. COUNT TWO: INDIRECT INFRINGEMENT OF THE ’867 PATENT
`
`92. Plaintiff incorporates by reference all paragraphs above as though set forth herein.
`
`93. Defendant induces infringement under 35 U.S.C. § 271(b) by actively and knowingly
`
`aiding and abetting direct infringement by its users.
`
`94. As discussed in § 5.B, Defendant received actual and constructive notice of the ’867
`
`patent.
`
`95. Defendant learned of its infringement of the ’867 patent at least as a result of the filing
`
`of the Original Complaint in this case as well as the filing of the instant complaint.
`
`96. Through at least the filing of the Original Complaint and the instant complaint, and the
`
`claim charts attached to both complaints, Defendant learned that its actions would result in
`
`users of the Accused Products infringing the ’867 patent.
`
`97. For example, the claim charts attached to both complaints show how Defendant’s
`
`Embedded Peripherals IP User Guide specifically provides users with instructions on using the
`
`Accused Devices in an infringing manner, such as by providing instructions regarding
`
`descriptors used with said products, and said guide and descriptors are explicitly illustrated in
`
`Plaintiff’s claim charts.
`
`98. Moreover, Defendant provides training and support to its customers, including through
`
`its FPGA Partner Program in which it provides “[t]echnical support for product optimization”
`
`such as that shown in Plaintiff’s claim charts. See
`
`https://www.intel.com/content/dam/www/public/us/en/documents/flyers/intel-fpga-
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 19
`
`
`
`partner-program-flyer.pdf; see also https://www.intel.com/content/www/us/en/partner/fpga-
`
`innovation-partners/overview.html (listing Intel partners).
`
`99. On information and belief Intel teaches users to optimize applications including
`
`optimizing usage of direct memory access, such as that shown by usage of the descriptors in
`
`SRC’s claim charts.
`
`100. For example, Intel taught its users how to use its FPGA acceleration cards with its
`
`OpenVINO toolkit to accelerate vision inference for facial recognition purposes. See
`
`https://www.youtube.com/watch?v=1GfZpubYm8s.
`
`101. Intel also taught in its Inference Engine Developer Guide how to perform direct
`
`memory access using an implementation of a vision processing unit (“VPU”). See
`
`https://docs.openvinotoolkit.org/latest/_docs_IE_DG_Extensibility_DG_VPU_Kernel.html.
`
`102. The VPU example referenced above teaches:
`
`(1) copying data from the DDR RAM memory (corresponding to the second memory of
`
`claim 1) to local (on chip) memory (corresponding to the first memory of claim 1) having 24x
`
`higher throughput (showing that the two memories each have a different “characteristic
`
`memory bandwidth” as required by claim 1),
`
`(2) “copying used sub-tensor by workgroup into local memory” and
`
`(3) matching the format of data for the calculation.
`
`103. For example, with respect to format matching, step 10 of the example discloses that
`
`“Local size is equal to (width of the input sensor, 1, 1) to define a large enough work group to
`
`get code automatically vectorized and unrolled, while global size is (width of the input tensor,
`
`height of the input sensor, 1).”
`
`PLAINTIFF’S FIRST AMENDED COMPLAINT FOR PATENT INFRINGEMENT – Page 20
`
`
`
`104. The example DMA is thus configured to prefetch data required by the algorithm and
`
`to match the format and location of data between the memories.
`
`105. Those teachings apply to both the referenced Neural Compute Stick 2 and also to
`
`Intel’s FPGA products.
`
`106. Intel actively provides support services for its products. An important part of Intel’s
`
`support services is the Intel Support Community. See https://community.intel.com/ Intel hosts
`
`Community Forums (including a forum for FPGAs and Programmable Solutions) where
`
`members can ask questions and receive support both from Intel engineers and fellow members.
`
`107. In one example Intel Community Forum thre