IPR2022-00996 (US11,016,918)
`IPR2022-00999 (US11,232,054)
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 1
`
`
`IPR2022-00999 (US11,232,054)
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 1
`
`
`
`TOPICS
`
`GROUNDS 1-3
`Harris Does Not Receive Power Via Edge Connections
`Harris Does Not Receive The Recited Signals From the Host
`A POSITA Would Not Have Used the Required # of (Buck) Converters
`Ground 2 ASSUMES Separate Converters To Supply Each FBDIMM-Required Voltage
`A POSITA Would Not Have Replaced Harris’ Redundant Power
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`GROUNDS 4-5
`The Recited “Memory Module” Means A Main Memory Module
`Spiers’ PCI Card Is Not a Main Memory Module
`POSITA Would Not Use DDR2/DDR3 or Recited # of (Buck) Converters
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 2
`
`
`
`The Challenged Claims Recite PCB Interface to the Host System Memory Slot
`
`
`
`The ’918/’054 PatentsThe ’918/’054 Patents
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 3
`
`
`
`The PCB Interface (to Host Memory Slot) Has Edge Connections
`Coupling Power/Data/Address/Control From Host to the Memory Module
`
`
`
`The ’918/’054 PatentsThe ’918/’054 Patents
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 4
`
`
`
`Harris’s Memory Module
`
`EX1023 (Harris), FIG. 1A.
`
`Patent Owner Response (918) at 4.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 5
`
`
`
`Harris Does Not Provide Power to the Memory Module
`Via Memory Slot Edge Connections
`
`Harris’ external voltage is provided to the
`side of the module, not the edge connections
`
`EX1023 (Harris), FIG. 1A (modified/annotated).
`
`Patent Owner Response (918) at 5.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 6
`
`
`
`Petitioner Admits Location of Control/Data/Address Edge Connections on Harris
`
`Reply, at 3-4
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 7
`
`
`
`Harris Expressly Replaces The Power Supply Pins
`On The Edge Connection Interface
`
`EX1023 (Harris), [0012]
`
`Patent Owner Response at (918) 4, 6, 8.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 8
`
`
`
`Harris Expressly Replaces The Power Supply Interface Pins
`On The Edge Connections
`
`EX1023 (Harris), [0012]
`
`Not “interface” pins
`
`Patent Owner Response at (918) 4, 6, 8.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 9
`
`
`
`Harris Expressly Eliminates System Board Power Supply
`
`EX1023 (Harris), [0019]
`
`If 12V were supplied from the motherboard, it
`would also be “system-board-specific power supply”
`
`Patent Owner Response at (918) 4, 8, 10-11.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 10
`
`
`
`12V is “Unregulated”; Unregulated Voltage Never Routed Through Interface
`
`Dr. Mangione-Smith:
`• Unregulated voltage never provided
`through interface pins or system
`board
`• Harris does not describe alternative
`locations for unregulated or
`regulated voltage pins
`Ex. 2031, ¶ 64
`
`EX1023 (Harris), [0010]
`
`Patent Owner Response at (918) at 6.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 11
`
`
`
`12V is “Unregulated”; Unregulated Voltage Never Routed Through Interface
`
`Dr. Mangione-Smith:
`• Unregulated voltage never provided
`through interface pins or system
`board
`• Harris does not describe alternative
`locations for unregulated or
`regulated voltage pins
`Ex. 2031, ¶ 64
`
`EX1023 (Harris), [0016]
`
`Patent Owner Response at (918) at 6.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 12
`
`
`
`Harris Does Not Provide Power to the Memory Module
`Via Memory Slot Edge Connections
`
`Memory
`Slot Edge
`Connection
`
`EX1023 (Harris), Fig. 3 (annotated).
`
`
`
`Wolfe: “Figure 3 is a block diagram and it does not show the power supplies”.Wolfe: “Figure 3 is a block diagram and it does not show the power supplies”.
`
`Patent Owner Sur-Reply (918) at 4.
`EX2030, 130:19-23 (cited at POR, 8)
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 13
`
`
`
`If Interface Pins Supplied Power in Harris, Figure 3 Would Depict
`
`Patent Owner Response at (918) at 8.
`Patent Owner Sur-Reply at 3, 19.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 14
`
`EX1023 (Harris), [0017]
`
`
`
`Harris Touts the Benefits of Eliminating System-Board Power Supply
`(i.e., an Internal Voltage Source)
`
`EX1023 (Harris), [0019]
`
`EX1023 (Harris), [0020]
`
`Patent Owner Response (918) at 8.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 15
`
`
`
`Avoiding Interface Pins for Voltage Allows System Board to Host
`Evolving Technology
`
`Patent Owner Response at 3, 8, 35
`Patent Owner Sur-Reply at 5.
`
`EX1023 (Harris), [0002]
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 16
`
`
`
`Providing Power to Memory Modules Via Non-Edge Connections Was Known
`
`Backup
`Power
`Source
`
`Edge connections
`
`EX2036, 42.
`
`EX2035, 3
`
`Edge connections
`
`Backup Power Source
`
`EX2035, 3; EX2036, 42.
`Patent Owner Response (918) at 5.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 17
`
`
`
`The Claims Are Not Satisfied By Merely Supplying Power from the Host
`
`Patent Owner Sur-Reply (918) at 6.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 18
`
`
`
`TOPICS
`
`GROUNDS 1-3
`Harris Does Not Receive Power Via Edge Connections
`Harris Does Not Receive The Recited Signals From the Host
`A POSITA Would Not Have Used the Required # of (Buck) Converters
`Ground 2 ASSUMES Separate Converters To Supply Each FBDIMM-Required Voltage
`A POSITA Would Not Have Replaced Harris’ Redundant Power
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`GROUNDS 4-5
`The Recited “Memory Module” Means A Main Memory Module
`Spiers’ PCI Card Is Not a Main Memory Module
`POSITA Would Not Use DDR2/DDR3 or Recited # of (Buck) Converters
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 19
`
`
`
`Claimed PCB Interface Receives Data, Address, and Control Signals From the Host
`
`
`
`The ’918/’054 PatentsThe ’918/’054 Patents
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 20
`
`
`
`FBDIMM – Advanced Memory Buffer (AMB)
`
`EX1027 at 4 (annotated).
`
`Patent Owner Response (918) at 11-12.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 21
`
`
`
`The Petition Relies on Signals Generated On-Module By The AMB,
`Not Signals Received From The Host
`
`The Petition
`
`Petition (918) at 21.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 22
`
`
`
`The Petition Relies on Signals Generated On-Module By The AMB,
`Not Signals Received From The Host
`
`The Petition
`
`Alleged “control signals”
`
`Alleged “address signals”
`Alleged “control signals”
`
`Petition (918) at 22-23.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 23
`
`
`
`AMB Specification Distinguishes Data from Signals
`
`“Data” from
`controller to
`interface
`
`“Signal” from
`AMB to DRAM
`
`Patent Owner Response 918) at 12.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 24
`
`
`
`Harris+FBDIMM Module Receives Serialized Encoded Information
`
`No pins on
`FBDIMM
`module to
`receive address
`or control
`signals from
`host
`
`EX1027, 4.
`
`Patent Owner Response (918) at 13.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 25
`
`
`
`Harris+FBDIMM Module Generates The Relied-on Address
`and Control Signals On-Module
`
`Address and
`Control Signals
`Relied On By
`Petitioner Are
`Generated by
`AMB
`
`Patent Owner Response (918) at 13.
`
`EX1027, 4.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 26
`
`
`
`Harris+FBDIMM Module Receives Serialized Encoded Information
`
`EX2040 at 1.
`
`Patent Owner Response (918) at 12.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 27
`
`
`
`Dr. Wolfe Testified That The AMB Receives Address/Control Information
`
`Andrew Wolfe
`Samsung’s Expert
`
`EX2030, 10:7-11.
`
`Patent Owner Sur-Reply (918) at 8.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 28
`
`
`
`Harris+FBDIMM Module Received Data, Address, and Control Information
`
`Patent Owner Sur-Reply (918) at 8.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 29
`
`
`
`TOPICS
`
`GROUNDS 1-3
`Harris Does Not Receive Power Via Edge Connections
`Harris Does Not Receive The Recited Signals From the Host
`A POSITA Would Not Have Used the Required # of (Buck) Converters
`Ground 2 ASSUMES Separate Converters To Supply Each FBDIMM-Required Voltage
`A POSITA Would Not Have Replaced Harris’ Redundant Power
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`GROUNDS 4-5
`The Recited “Memory Module” Means A Main Memory Module
`Spiers’ PCI Card Is Not a Main Memory Module
`POSITA Would Not Use DDR2/DDR3 or Recited # of (Buck) Converters
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 30
`
`
`
`Petitioner’s Proposed Voltage Mappings for ’918 Patent
`Based on Harris+FBDIMM
`Recited Voltages of Claim 1
`
`Petition (918) at 26.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 31
`
`
`
`Petitioner’s Proposed Voltage Mappings for ’054 Patent
`Based on Harris+FBDIMM
`Recited Voltages of Claim 1
`
`Petition (054) at 27.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 32
`
`
`
`Harris Proposed Only Two Voltages Generated On-Module In Face of DDR
`Specifications, Including FBDIMMs
`
`“Harris further discloses that when implemented as the on-
`board VRM 102, the single “high-frequency switching voltage
`converter” converts the externally supplied voltage “into
`appropriate local voltage levels that power … the Vdd and Vcc
`paths 108, 106, respectively.”
`
`EX2031 (Mangione-Smith Declaration), ¶79.
`
`Vdd on board
`
`Targets DDR-
`Standardized
`Devices
`
`Vcc on board
`
`Petition at 10, 16.
`Patent Owner Response (918) at 17.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`EX1023 (Harris), [0009]
`
`Ex. 2064, p. 33
`
`
`
`Petitioner’s Expert Confirms That a POSA Reading Harris Would Understand And
`Comply With JEDEC Standards
`
`EX1003 (Wolfe), ¶161
`
`EX1003 (Wolfe), ¶158
`
`Petition at 16.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 34
`
`
`
`In FBDIMM Specification, One Source For VCC/VCCFD;
`One Source For VDD, VDDQ, VDDL
`
`EX2031 (Mangione-Smith Declaration), ¶88.
`
`EX2031 (Mangione-Smith Declaration), ¶88.
`
`Patent Owner Response (918) at 23.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 35
`
`
`
`FBDIMM Uses Single Source For VDD/VDDQ/VDDL and VCC/VCCFBD
`
`FBDIMM Standard provides a single
`set of VCC pins for VCC/VCCFBD
`
`FBDIMM Standard provides a single
`set of VDD pins for VDD/VDDQ/VDDL
`
`Patent Owner Response at (918) 22.
`
`EX1028 (JEDEC FBDIMM Specification), 11
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 36
`
`
`
`The FBDIMM Standard Uses A Single Converter For VDD/VDDQ/VDDL
`
`FBDIMM Standard provides for a single
`converter for VDD/VDDQ/VDDL
`
`EX1026 (JEDEC DDR2 SDRAM Specification), 9
`
`Patent Owner Response (918) at 25.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 37
`
`
`
`Petitioner’s Reliance on an Alleged “Option 2” Fails To Motivate
`Using Separate Buck Converters For VDD/VDDQ/VDDL
`
`Option 1
`
`Option 2
`
`EX1026 (JEDEC DDR2 SDRAM Specification), 9
`
`Patent Owner Response (918) at 25.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 38
`
`
`
`Petitioner’s Alleged “Option 2” Are Actually Alternatives to Vref Tracking
`
`“Options” end in
`periods, not “ANDs”
`
`EX1026 (JEDEC DDR2 SDRAM Specification), 9
`
`Patent Owner Response (918) at 25.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 39
`
`
`
`Turning Voltages On Or Off Separately Does Not Motivate Separate Converters
`
`EX1026 (JEDEC DDR2 SDRAM Specification), 9
`
`Patent Owner Response (918) at 25.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 40
`
`
`
`The Experts Agree: The Industry Rejected Using Multiple
`Converters in FBDIMMs for VDD/VDDQ/VDDL
`
`Andrew Wolfe
`Samsung’s Expert
`
`EX2030 (Wolfe Deposition), 133:5-9
`
`EX2031 (Mangione-Smith Declaration), ¶90.
`
`William Mangione-Smith
`Netlist’s Expert
`
`Patent Owner Response (918) at 25.
`Patent Owner Sur-Reply (918) at 17.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 41
`
`
`
`The Industry Rejected Using Multiple Converters To Supply
`VDD/VDDQ/VDDL or VCC/VCCFBD For Good Reasons
`
`EX2006 (Micron Technical Note), 4
`
`Patent Owner Response (918) at 23-24.
`Patent Owner Sur-Reply (918) at 17.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 42
`
`
`
`The Industry Rejected Using Multiple Converters To Supply
`VDD/VDDQ/VDDL or VCC/VCCFBD For Good Reasons
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`EX2031 (Mangione-Smith Declaration), ¶88.
`
`Patent Owner Response (918) at 25.
`Patent Owner Sur-Reply (918) at 17.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 43
`
`
`
`The Industry Rejected Using Multiple Converters To Supply
`VDD/VDDQ/VDDL or VCC/VCCFBD For Good Reasons
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`EX2031 (Mangione-Smith Declaration), ¶90.
`
`Patent Owner Response (918) at 25.
`Patent Owner Sur-Reply (918) at 17.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 44
`
`
`
`Petitioner’s Evidence of Alleged Motivations to Use
`Separate Converters Are Untethered to Harris
`
`No evidence
`of analog
`voltages
`
`FBDIMM
`does not run
`at hundreds
`of amperes
`
`EX1062 (Power-Supply Management—Principles, Problems, and Parts) at 13.
`
`VDD, VDDL, and VDDQ are each
`supplied to the same devices
`
`Petition (918) at 31.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 45
`
`
`
`Petitioner Has Not Established A POSITA Would Have Used Separate
`Buck Converters For VDD/VDDQ/VDDL or VCC/VCCFBD
`
`System-
`Board-
`Specific
`Voltage
`Level
`
`EX1028, 11
`
`Patent Owner Sur-Reply (918) at 16.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 46
`
`
`
`Petitioner Has Not Established A POSITA Would Have Further Modified
`Harris+FBDIMM With Separate Buck Converters Given Space Constraints
`
`EX1026 (Harris), [0010]
`
`X 4
`
`• EX1078, 23
`
`EX2042, 7.
`
`Patent Owner Response (918) at 20
`Patent Owner Sur-Reply (918) at 12
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 47
`
`
`
`Petitioner Has Not Established A POSITA Would Have Used Separate
`Buck Converters For VDD/VDDQ/VDDL or VCC/VCCFBD
`Voltage mappings that require
`using separate converters for
`the same voltage level under
`Petitioner’s theory
`
`
`
`The 918 PetitionThe 918 Petition
`
`Petition at (918) 26.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 48
`
`
`
`Petitioner Has Not Established A POSITA Would Have Used Separate
`Buck Converters For VDD/VDDQ/VDDL or VCC/VCCFBD
`
`
`
`The 054 PetitionThe 054 Petition
`
`Voltage mappings that require
`using separate converters for
`the same voltage level under
`Petitioner’s theory
`
`Petition (054) at 27.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 49
`
`
`
`Petitioner’s Own Evidence Shows that Single Buck Converter
`Outputting Multiple Voltages Were Known
`
`Single buck converter
`receiving only Vin (1) and
`outputting different Vout2
`and Vout3
`
`EX1048, 2 (annotated)
`
`Patent Owner Sur-Reply (918) at 14.
`Patent Owner Response (918) at 18.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 50
`
`
`
`Single Buck Converters Can Generate Multiple Voltages
`
`Patent Owner Sur-Reply (918) at 14.
`Patent Owner Response (918) at 18.
`
`EX2003 at 2.
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`EX2004 at 99 (pdf page 11)
`
`Ex. 2064, p. 51
`
`
`
`Harris Expressly Discloses Using a Single Converter to Provide Multiple Voltages
`
`EX1023 (Harris), [0010]
`
`Patent Owner Response (918) at 17.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 52
`
`
`
`Harris’ Claims Confirm a Single Module Can Supply Multiple Voltages
`
`EX1023 (Harris), Claim 1
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 53
`
`
`
`Absent Further Modification, Harris+FBDIMM Would Not
`Use Separate Converters For Each Voltage
`
`
`
`’918 IPR’918 IPR
`
`Harris’ existing system
`
`At most only 2 more
`regulators needed
`under Petition’s
`theory
`
`Patent Owner Response (918) at 18.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 54
`
`
`
`Absent Further Modification, Harris+FBDIMM Would Not
`Use Separate Converters For Each Voltage
`
`
`
`’054 IPR’054 IPR
`
`Harris’ existing system
`
`At most only 1 more
`regulator needed
`under Petition’s
`theory
`
`Patent Owner Response (054) at 21.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 55
`
`
`
`Voltage Mapping C Fails: Harris Does Not Generate VTT On-Module
`And A POSA Would Not Be Motivated to Modify Harris To Do So
`
`The 918 Petition
`
`Petition (918) at 26.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 56
`
`
`
`Voltage Mapping C Fails: Harris Does Not Generate VTT On-Module
`And A POSA Would Not Be Motivated to Modify Harris To Do So
`
`The 054 Petition
`
`Petition (054) at 27.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 57
`
`
`
`Harris Does Not Teach Or Suggest Generating VTT On-Module
`
`EX1023 (Harris), [0012]
`
`Patent Owner Sur-Reply (918) at 17-18.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 58
`
`EX102 (Harris), FIG. 1A.
`
`
`
`A POSA Would Have Had Specific Reasons Not to Generate VTT On-Module
`
`“[S]upplying VTT from the motherboard would ensure that all DIMMs
`connected to the same memory controller would have the same
`termination voltages …. This would also eliminate undesirable ground
`loops between the DIMMs, on the one hand, and between the DIMM
`and the motherboard, on the other hand.”
`
`EX2031 (Mangione-Smith Declaration), ¶96
`
`“VTT supply must sink and source current…. The design complexity
`would favor having a single regulator for a group of DIMMs rather
`than one regulator per DIMM.”
`EX2031 (Mangione-Smith Declaration), ¶97
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`Patent Owner Response (918) at 28.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 59
`
`
`
`A POSA Would Not Generate VTT Using A Buck Converter
`
`VTT was typically generated by an LDO
`
`EX2006 (Micron Technical Note), at 7.
`
`
`
`See also EX2007-2010, EX2050 (linear VTT regulators that generate VTT from VDDQ).See also EX2007-2010, EX2050 (linear VTT regulators that generate VTT from VDDQ).
`
`Patent Owner Response (918) at 29-30.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 60
`
`
`
`There Are Specific Reasons To Use An LDO To Generate VTT
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`EX2031 (Mangione-Smith Declaration), ¶99
`
`Patent Owner Response (918) at 29-30.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 61
`
`
`
`Petitioner’s VTT Examples are for DDR1, not DDR2
`
`VDDQ of 2.5V
`is for DDR1*
`
`Patent Owner Response (918) at 40
`Patent Owner Sur-Reply (918) at 30.
`
`EX1040, p. 18
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`*EX1003 Wolfe Decl., ¶139
`
`Ex. 2064, p. 62
`
`
`
`Petitioner’s VTT Examples are for DDR1, not DDR2
`
`VDDQ of 2.5V
`is for DDR1*
`
`Patent Owner Response (918) at 40
`Patent Owner Sur-Reply (918) at 30.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`*EX1003 Wolfe Decl., ¶139
`
`Ex. 2064, p. 63
`
`EX1041, p. 2
`
`
`
`EX1040 and EX1041 Do Not Suggest Using Buck Converters in FBDIMMs
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`EX2031 (Mangione-Smith Declaration), ¶100
`
`Patent Owner Sur-Reply (918) at 18.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 64
`
`
`
`A POSA Would Not Use A Buck Converter To Generate
`VDDSPD, As Required By All Voltage Mappings
`
`The 918 Petition
`
`Petition (918) at 26.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 65
`
`
`
`Harris Does Not Teach Or Suggest Generating VDDSPD On-Module
`
`EX1023 (Harris), [0012]
`
`EX102 (Harris), FIG. 1A.
`
`Patent Owner Sur-Reply (918) at 18.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 66
`
`
`
`A POSA Would Not Use An On-Module Buck Converter To
`Generate VDDSPD, As Required By All Voltage Mappings
`• Petitioner fails to explain why a POSA would add an on-module converter for 3.3V
`VDDSPD when there is already a 3.3V power rail available on the motherboard.
`‒ “3.3V is a common power rail on the motherboard such that it can be supplied to
`FBDIMMs without the increased cost and problems associated with providing a regulator
`for each of Harris’ modified FBDIMM memory boards” EX2031, ¶102.
`• A POSA would understand there are reasons not to have multiple 3.3V regulators
`because the SMBus controller on the motherboard uses the same SMBus to
`communicate with the SPD of each FBDIMM.
`‒ “[I]f VDDSPD voltages are supplied by different regulators to the SMBus controller on the
`motherboard and individual DIMMs, they cannot track each other and would lead to
`communication problems when they are actually at different voltage potentials.” Id.
`• A POSA would not use a buck converter which is less efficient than an LDO.
`‒ “[T]he low current level required by components using VDDSPD means that buck
`converters would be very inefficient.” EX2031, ¶103.
`‒ Buck converter: “efficiency less than 20% at 10mA current” Id.
`‒ LDO: “An LDO with a12V-input-3.3V-output would be 26% efficient.” Id.
`
`Patent Owner Response (918) at 30-31.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 67
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`
`
`Use of Buck Converters Is Entirely Hindsight
`
`EX1023 (Harris), [0010].
`
`Patent Owner Response (918) at 4, 8, 16, 17.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 68
`
`
`
`Petitioner Argues That General References to Step Converters Does Not
`Inherently Disclose Buck Converters
`
`Petition (918) at 4.
`
`EX. 1003 (Wolfe)
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 69
`
`
`
`TOPICS
`
`GROUNDS 1-3
`Harris Does Not Receive Power Via Edge Connections
`Harris Does Not Receive The Recited Signals From the Host
`A POSITA Would Not Have Used the Required # of (Buck) Converters
`Ground 2 ASSUMES Separate Converters To Supply Each FBDIMM-Required Voltage
`A POSITA Would Not Have Replaced Harris’ Redundant Power
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`GROUNDS 4-5
`The Recited “Memory Module” Means A Main Memory Module
`Spiers’ PCI Card Is Not a Main Memory Module
`POSITA Would Not Use DDR2/DDR3 or Recited # of (Buck) Converters
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 70
`
`
`
`Amidi Does Not Use Separate On Board Converters for DDR2 Required Voltages:
`VDD, VCC, VTT
`
`EX1024 (Amidi) at 7:52-59.
`
`EX1024 (Amidi) at FIG. 6.
`
`Petition at 6, 79 (citing EX1003 at ¶196).
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 71
`
`
`
`Ground 2 Still Assumes Motivation to Use
`Separate Converters For Each FBDIMM Required Voltage
`
`Petition at 75.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 72
`
`
`
`TOPICS
`
`GROUNDS 1-3
`Harris Does Not Receive Power Via Edge Connections
`Harris Does Not Receive The Recited Signals From the Host
`A POSITA Would Not Have Used the Required # of (Buck) Converters
`Ground 2 ASSUMES Separate Converters To Supply Each FBDIMM-Required Voltage
`A POSITA Would Not Have Replaced Harris’ Redundant Power
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`GROUNDS 4-5
`The Recited “Memory Module” Means A Main Memory Module
`Spiers’ PCI Card Is Not a Main Memory Module
`POSITA Would Not Use DDR2/DDR3 or Recited # of (Buck) Converters
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 73
`
`
`
`Harris Already Provides Redundant Power to the Memory Module
`
`EX1023 (Harris), [0014]
`
`EX1023 (Harris), FIG. 1B.
`
`Patent Owner Response (918) at 32.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 74
`
`
`
`Harris’ Redundancy Implementation Provides Backup Power Using Redundant
`Input Voltages
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`EX2031 (Mangione-Smith Declaration), ¶107
`
`Patent Owner Response (918) at 34.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 75
`
`
`
`Dr. Wolfe Agrees: Harris’ Existing Redundancy Implementation
`is Functionally Similar to Amidi’s Backup Power Supply
`
`Harris
`
`Andrew Wolfe
`Samsung’s Expert
`
`EX1003 (Wolfe Declaration), ¶170.
`
`Patent Owner Response (918) at 34.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 76
`
`
`
`Petition: Amidi’s Backup Power Supply Is Functioning
`“In the Same Way” As Harris’ Redundant Implementation
`
`
`
`The PetitionThe Petition
`
`Petition (918) at 55.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 77
`
`
`
`A POSA Would Not Have Preferred a Battery Backup
`Solution To Harris’ Existing Redundancy Implementation
`
`• “[U]sing Amidi’s battery only results in additional and unnecessary complexity, with
`Harris’s 12V external source having to be stepped-down with a buck converter to 5V
`for Amidi’s battery, only to then be stepped back up with a boost converter to 12V.
`EX2031, ¶108.
`
`• The ’918 patent notes that “batteries may require maintenance, may need to be
`replaced, are not environmentally friendly, and the status of batteries can be difficult
`to monitor.” EX1001, 4:56-58; Id.
`
`• “Amidi’s battery backup solution would require a substantial amount of on-board
`space ….” EX2031, ¶108.
`
`• “[A] battery backup solution would also be able to provide backup power for only a
`finite period of time based on the battery’s capacity. This would represent an
`inadequate solution for power interrupts which last longer.” Id., ¶109.
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`Patent Owner Response (918) at 30-31.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 78
`
`
`
`TOPICS
`
`GROUNDS 1-3
`Harris Does Not Receive Power Via Edge Connections
`Harris Does Not Receive The Recited Signals From the Host
`A POSITA Would Not Have Used the Required # of (Buck) Converters
`Ground 2 ASSUMES Separate Converters To Supply Each FBDIMM-Required Voltage
`A POSITA Would Not Have Replaced Harris’ Redundant Power
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`GROUNDS 4-5
`The Recited “Memory Module” Means A Main Memory Module
`Spiers’ PCI Card Is Not a Main Memory Module
`POSITA Would Not Use DDR2/DDR3 or Recited # of (Buck) Converters
`Additional Reasons Why Dependent Claims Are Not Obvious
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 79
`
`
`
`’918 Claims 5-7, 9-13, 16-22, 24-27:
`Trigger Signal Produced In Response to Detecting Over-voltage
`
`
`
`The ’918 PatentThe ’918 Patent
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 80
`
`
`
`’054 Claims 6-7, 11-12, 17, 29:
`Trigger Signal Produced In Response to Detecting Over-voltage
`
`
`
`The ’054 PatentThe ’054 Patent
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 81
`
`
`
`Petition: Harris Allegedly Teaches Detecting Over/Under Voltages
`
`
`
`The PetitionThe Petition
`
`Petition (918) at 62
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 82
`
`
`
`Harris’ ± 15% Is a Design Tolerance, Not Detection of Over/Under Voltages
`
`EX1023 (Harris), [0013]
`
`EX2031 (Mangione-Smith Declaration), ¶116
`
`
`
`And no trigger signal is produced in Harris if supply voltage exceeds design tolerance And no trigger signal is produced in Harris if supply voltage exceeds design tolerance
`
`Patent Owner Response (918) at 39.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 83
`
`
`
`Petition: Allegedly Well-Known to Detect Over/Under Voltages
`
`The Petition
`
`Petition (918) at 62
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 84
`
`
`
`Petitioner’s Evidence Shows That Allegedly Suitable
`Regulators Did Not Detect Over-Voltage
`Undervoltage, but no
`Overvoltage detection
`
`EX1078, 1, 3.
`
`Patent Owner Sur-Reply (918) at 20.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 85
`
`
`
`Wide Input Voltage Tolerance Obviates the Need for Over-Voltage Detection
`
`EX1040
`
`EX1041
`
`Patent Owner Response (918) at 40.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 86
`
`
`
`High Upper Limit on Input Voltage
`Obviates the Need for Over-Voltage Detection
`
`117. … [V]oltage regulators that Samsung alleges are suitable for use
`with Harris do not detect input overvoltage, as Dr. Wolfe confirmed.
`See EX2030, 58:24-61:12; EX1040; EX1041…. An unregulated 12V
`voltage source with ±15% wide tolerance has an expected voltage
`range of 10.2 – 13.8V. The 28V upper limit for the converters in
`EX1040 is well above the 13.8V upper range. So is the 16V upper limit
`for the converters in EX1041. Indeed, even if the +/-15% tolerance
`represents only a 3-sigma process, the upper limit for the converters
`in EX1040 would be 26.7 sigmas away; and the upper limit for the
`converters in EX1041 would be 6.7 sigmas away. That means the
`probability that the input voltage would be above the permissible
`operating range of these converters is extremely low (6 sigmas means
`one in half a billion chance). As such, a POSITA would have no need to
`include input overvoltage detection for these regulators.
`
`EX2031 (Mangione-Smith Declaration), ¶117.
`
`Patent Owner Response (918) at 40.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 87
`
`William
`Mangione-Smith
`Netlist’s Expert
`
`
`
`Petitioner Has Not Established Any Reason to Detect
`Over-Voltage in the Harris+Amidi Combination
`
`Patent Owner Sur-Reply (918) at 22.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 88
`
`
`
`Petition: A POSA Would Allegedly Be Motivated to Switch to Amidi’s Backup
`Power When Over-Voltage Is Detected Based On Hajeck
`
`
`
`The PetitionThe Petition
`
`Petition (918) at 77
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 89
`
`
`
`Hajeck Does Not Suggest Switching To Backup Power In
`Response To Detecting An Over-Voltage
`Hajeck’s voltage detection 48 circuit sends a “busy signal” 38 back to the host 32 to prevent the
`host 32 from performing write operations to the memory subsystem 30 when voltage anomalies
`are detected. EX1038, 3:30-57
`
`Patent Owner Response (918) at 41.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 90
`
`EX1038 (Hajeck), FiG. 2.
`
`
`
`Hajeck Reduces The Risk Of Data Loss By Inhibiting New Write Operations,
`Not Switching To A Different Power Source
`
`EX1038 (Hajeck), 4:3-9
`
`
`Hajeck expressly continues to provide regularHajeck expressly continues to provide regular
`
`power in the event of voltage anomaliespower in the event of voltage anomalies
`
`Patent Owner Response (918) at 41.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 91
`
`
`
`Hajeck Teaches to Use a Charge Pump To Protect Against Over-Voltages,
`Not To Switch to Backup Power
`
`EX1038 (Hajeck), 4:3-9
`
`Patent Owner Response (918) at 41.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 92
`
`
`
`Hajeck Teaches Continued Operation With Same Power Source in Over-Voltage
`Condition
`
`EX1038 (Hajeck), 3:6-15
`
`Patent Owner Response at 41, 44.
`Patent Owner Sur-Reply at 34.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 93
`
`
`
`Dr. Wolfe Agrees: Hajeck’s Charge Pump Would Absorb Over-Voltages
`
`“So something like a spike would be diluted in that
`reservoir and would not be passed through.”
`
`Andrew Wolfe
`Samsung’s Expert
`
`EX2030 (Wolfe Deposition), 229:9-11.
`
`“There will be some set of spikes and surges that can
`be absorbed by the charge pump.”
`
`EX2030 (Wolfe Deposition), 230:11-13.
`
`Patent Owner Sur-Reply (918) at 21.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 94
`
`
`
`None Of The References Motivate Switching To Backup
`Power In Response To An Over-Voltage
`
`• Harris
`‒ discloses only a design tolerance of +/- 15% for input voltage.
`‒ does not disclose detecting over-voltages or switching to backup power .
`• Amidi
`‒ discloses switching to backup power only in response to power loss.
`• Hajeck
`‒ does not disclose/suggest switching to backup power in response to an over-
`voltage because, in that condition, it inhibits write operations while continuing to
`provide regular power from Hajeck’s charge pump.
`‒ discloses using a charge pump to protect against over-voltage.
`
`
`The alleged desire to switch to backup power supply to avoid data loss in an over-The alleged desire to switch to backup power supply to avoid data loss in an over-
`
`voltage condition is not found in any of prior art references, alone or in combination.voltage condition is not found in any of prior art references, alone or in combination.
`
`Patent Owner Response (918) at 40-42
`Patent Owner Sur-Reply (918) at 20-22.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 95
`
`
`
`Claims 10-11, 15, 22: Non-volatile Memory Within The Logic Element
`
`
`
`The ’918 PatentThe ’918 Patent
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 96
`
`
`
`The AMB in Harris+FBDIMM Is Alleged To Include Non-Volatile Memory
`
`
`
`The PetitionThe Petition
`
`Petition (918) at 67.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 97
`
`
`
`Petitioner Has Not Established That an AMB Includes Non-Volatile Memory
`
`Dr. Wolfe’s testimony confirms Petitioner has not provided
`competent evidence that claims 10-11, 15, 22 are obvious:
`
`Andrew Wolfe
`Samsung’s Expert
`
`EX2030 (Wolfe Deposition), 293:1-6.
`
`Patent Owner Response (918) at 43.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 98
`
`
`
`Petitioner Has Not Established That an AMB Includes Non-Volatile Memory
`
`Petitioner’s reference to SPD for claim 15 fails to bridge the gap
`because SPD and AMB are two different components:
`
`Andrew Wolfe
`Samsung’s Expert
`
`EX2030 (Wolfe Deposition), 292:21-25.
`
`Patent Owner Response (918) at 43.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Ex. 2064, p. 99
`
`
`
`Claims 11-12, 18-19, 25-26: Write Operation In Response To Over-Voltage
`
`
`
`The ’918 PatentThe ’918 Patent
`
`
`The Petition relies on the AMB as the “logic element.” Pet. 67. The Petition relies on the AMB as the “logic element.” Pet. 67.
`
`So the AMB must write the information into non-voltage So the AMB must write the information into no
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
