`________________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`________________________
`
`KINGSTON TECHNOLOGY COMPANY, INC.,
`Petitioner,
`
`v.
`
`POLARIS INNOVATIONS LTD.,
`Patent Owner.
`
`____________
`
`Case IPR2016-01622
`U.S. Patent 6,850,414 B2
`____________
`
`
`
`PETITIONER’S OPPOSITION TO PATENT OWNER’S MOTION TO
`AMEND
`
`
`
`
`
`Mail Stop Patent Trial and Appeal Board
`Patent Trial and Appeal Board
`U.S. Patent and Trademark Office
`P.O. Box 1450
`Alexandria, VA 22313-1450
`
`
`
`TABLE OF CONTENTS
`
`INTRODUCTION ......................................................................................... 1
`I.
`STATEMENT OF PRECISE RELIEF REQUESTED ............................. 2
`II.
`III. ARGUMENT ................................................................................................. 2
`A. Simpson in view of the Intel Specification Discloses “said printed circuit
`board . . . has a height of 1 to 1.2 inches perpendicular to said contact
`strip,” as Recited in Substitute Claim 9 and Claim 4. ................................ 4
`1. It would have been both within the skill level of a POSITA and
`physically possible to apply Simpson’s chip arrangement on Intel’s
`5.25” by 1.2” PCB. ............................................................................................ 4
`2. There was a strong motivation to apply Simpson’s chip arrangement
`on a 5.25” by 1.2” PCB. ................................................................................. 11
`3. Patent Owner’s assertions that claim 9 (and claim 4) are patentable are
`disingenuous at best......................................................................................... 14
`B. Simpson in view of Karabatsos Discloses “said printed circuit board . . .
`has a height of 1 to 1.2 inches perpendicular to said contact strip,” as
`Recited in Substitute Claim 9 and Claim 4. ............................................. 15
`C. Bechtolsheim in view of Tokunaga and Karabatsos Discloses a Memory
`Module as Recited in Claim 1 where “said printed circuit board . . . has a
`height of 1 to 1.2 inches perpendicular to said contact strip,” as Recited
`in Substitute Claim 9 and Claim 4. ........................................................... 17
`1. Bechtolsheim in view of Tokunaga and Karabatsos Discloses a
`Memory Module as Recited in Claim 1. .................................................... 17
`2. Bechtolsheim in view of Tokunaga and Karabatsos Discloses a
`Memory Module as Recited in Substitute Claim 9 (and Original Claim
`4). ......................................................................................................................... 22
`D. The Claimed Circuit Board Height is Not Entitled to Independent
`Patentable Weight. .................................................................................... 23
`IV. CONCLUSION ............................................................................................ 25
`
`
`
`i
`
`
`
`
`
`Exhibit
`
`EXHIBIT LIST
`
`Description
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`
`1001 U.S. Patent 6,850,414 to Benisek (’414 patent)
`
`1002 UK Patent Application GB 2 289 573 A to Simpson
`
`1003
`
`PC SDRAM Unbuffered DIMM Specification, Version 1.0
`
`1004 U.S. Patent Application Publication No. 2002/0196612 to Gall
`
`1005
`
`PC133 SDRAM Unbuffered DIMM Specification, Version 1.0
`
`1006 Declaration of Professor Vivek Subramanian (“Subramanian”)
`
`1007
`
`ʼ414 Patent File History
`
`1008
`
`File History for U.S. Patent No. 6,332,183
`
`1009 District Court Complaint
`
`1010
`
`Professor Vivek Subramanian’s Curriculum Vitae
`
`1011
`
`Intel Small Outline Package Guide
`
`1012 Micron 64Mb: x32 SDRAM Features
`
`1013 U.S. Patent 4,954,088 to Fujizaki
`
`1014 Deposition of Dr. Joseph Bernstein
`
`1015
`
`IPR2017-00974 Paper 2: Petition for Inter Partes Review
`
`1016
`
`IPR2017-00974 Ex. 1006: Declaration of Professor Vivek
`
`Subramanian
`
`ii
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`1017 U.S. Patent Application Publication No. 2002/0006032 to
`
`Karabatsos
`
`1018 U.S. Patent No. 5,973,951 to Bechtolsheim
`
`1019 U.S. Patent No. 6,038,132 to Tokunaga
`
`1020 German Publication No. DE 101 24 361 A1 (“Kiehl”)
`
`1021
`
`English Translation of German Publication No.
`
`DE 101 42 361 A1 (“Kiehl”)
`
`1022
`
`Second Declaration of Professor Vivek Subramanian
`
`1023
`
`Translation Certification for German Publication No.
`
`DE 101 42 361 A1 (“Kiehl”)
`
`1024
`
`IPR 016-01623 Paper 16: Patent Owner’s Response to
`
`Petition for Inter Partes Review
`
`
`
`
`
`
`iii
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`
`I.
`
`INTRODUCTION
`
`Kingston Technology Company, Inc. (“Petitioner”) respectfully opposes the
`
`Patent Owner’s Motion to Amend (Paper 18). In its motion, Patent Owner has
`
`elected to reinsert the subject matter of non-instituted claim 4 (a height limitation)
`
`back into the present proceeding. Patent Owner’s only stated basis for the
`
`patentability of the height limitation is that claim 4 has “already been found to not
`
`have been shown by Petitioner in this case to be disclosed or suggested by the prior
`
`art.” Paper 18 at 4. This statement is both a mischaracterization of the Board’s
`
`decision and factually incorrect. See Paper 7 at 18; see also Paper 16 at 6. As set
`
`forth below, the addition of a height limitation to claim 8 does not make new claim
`
`9 patentable, as building memory modules with a height between 1 inch and 1.2
`
`inches was well known in the art and there were numerous reasons why a person of
`
`ordinary skill would use that height with a module meeting the requirements of
`
`claim 8. See Ex. 1006 at ¶¶95-96; Ex. 1016 at ¶¶35-40, 50-51, 104.
`
`The reasons why proposed claim 9 is invalid are largely set forth in pending
`
`IPR2017-00974, which addresses the patentability of claim 4 of the ’414 patent.
`
`As claim 4 includes the very height limitation that Patent Owner seeks to add
`
`through amendment (and also depends on claim 1), the art and arguments set forth
`
`in IPR2017-00974 largely apply here. In fact, in evaluating the merit of Patent
`
`Owner’s motion to amend, the Board will be required to consider the same subject
`
`1
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`matter as in IPR2017-00974. As such, Petitioner asserts that it would be most
`
`efficient for both the Board and the parties for IPR2017-00974 to be instituted and
`
`aligned in schedule with this IPR.1
`
`II.
`
`STATEMENT OF PRECISE RELIEF REQUESTED
`
`Petitioner requests that the Board find Patent Owner’s proposed amendments
`
`to claim 8 that would incorporate features of claim 4 unpatentable and deny Patent
`
`Owner’s motion to amend.
`
`III. ARGUMENT
`
`Patent Owner asks the Board to hold that the addition of the limitation “and
`
`has a height of 1 to 1.2 inches perpendicular to said contact strip” renders now
`
`cancelled claim 8 patentable. By virtue of the fact that this limitation appears in
`
`
`1 By itself requesting consideration of the patentability of the height limitation,
`
`Patent Owner has waived any argument that proceeding with IPR 2017-000974
`
`would be inefficient. Moreover, as proposed claim 9 includes every limitation of
`
`claim 4, a finding that proposed claim 9 is not patentable would mean that claim 4
`
`is by definition also not patentable. It would be against the interest of justice to
`
`allow subject matter that has been found unpatentable to remain within the patent
`
`monopoly – further warranting institution of IPR2017-00974 to ensure consistent
`
`application of the law.
`
`2
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`non-instituted claim 4, Patent Owner asserts that it makes claim 8 patentable. First,
`
`as set forth below, the Board denied institution of claim 4 not because the Board
`
`found the limitations of claim 4 to be patent eligible, but because Petitioner has not
`
`explained adequately “how and/or why a person of ordinary skill in the art would
`
`have” combined this feature with Simpson’s design. Paper 7 at 18; see also Paper
`
`16 at 6. The fact remains, however, that each and every limitation of Patent
`
`Owner’s substitute claim 9 (and by extension original claim 4) is disclosed within
`
`the prior art. As set forth below (with additional explanation of both how and
`
`why), it would have been obvious to a person of ordinary skill to design a memory
`
`module using known prior art components that would meet all of the limitations of
`
`substitute claim 9 (and by extension original claim 4) on a PCB with a height of 1
`
`to 1.2 inches as specifically called out in industry design specifications.2
`
`
`2 Petitioner also notes that although the Board in the present case provided Patent
`
`Owner with detailed guidelines for its motion to amend, Paper17, Patent Owner
`
`did not follow them. For instance, Patent Owner failed to “show patentability [of
`
`the substitute claim] over the prior art that is relevant to the substitute claim[],
`
`including prior art of record and prior art known the Patent Owner,” which would
`
`include the art in IPR2017-00974. Id. at 4.
`
`3
`
`
`
`A.
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`Simpson in view of the Intel Specification Discloses “said printed
`circuit board . . . has a height of 1 to 1.2 inches perpendicular to
`said contact strip,” as Recited in Substitute Claim 9 and Claim 4.
`The Board has already recognized that there is a reasonable likelihood that
`
`all the features of substitute claim 9 are unpatentable in view of Simpson, except
`
`for Patent Owner’s amendment to incorporate a height limitation of 1 to 1.2 inches.
`
`Paper 7 (Instituting review of claims 1 and 8). Indeed, Patent Owner chose to
`
`cancel claims 1 and 8 in view of the Board’s previous finding. Moreover, the
`
`Board has also recognized that the Intel Specification teaches “a height in the range
`
`of 1 to 1.2 inches” for a printed circuit board. Paper 7 at 18. However, the Board
`
`decided not to institute IPR proceedings as to the PCB board height, not because
`
`this feature is necessarily patentable, but rather because Petitioner has not
`
`explained sufficiently “how and/or why a person of ordinary skill in the art would
`
`have” combined this feature with Simpson’s design as insufficient. Paper 7 at 18;
`
`see also Paper 16 at 6. Petitioner welcomes the opportunity granted by Patent
`
`Owner’s having reinjected this issue into the proceeding to further explain the facts
`
`supporting its position.
`
`1.
`
`It would have been both within the skill level of a POSITA
`and physically possible to apply Simpson’s chip
`arrangement on Intel’s 5.25” by 1.2” PCB.
`First, it would have been possible for one of skill in the art to implement
`
`Simpson’s memory chip layout on PCB having a height of 1 to 1.2 inches. Ex.
`
`4
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`1022 at ¶¶17-30. As Patent Owner makes explicitly clear, Simpson’s “goal is to
`
`provide flexibility by adding sockets that allow a user to add memory . . . to the
`
`existing PCB.” Paper 6 at 42. Simpson does not in any way limit the dimensions of
`
`the PCB, the memory chips, or the sockets used to permit memory expansion. See
`
`e.g., Ex. 1016 at ¶¶52, 104; see also Ex. 1002 generally. Indeed, patent drawings
`
`cannot be relied upon to define precise proportions of elements if not designated as
`
`being drawn to scale. See Hockerson-Halberstadt, Inc. v. Avia Group Int’l, Inc.,
`
`222 F.3d 951, 956 (Fed. Cir. 2000). There is no recitation in Simpson of the
`
`drawings being to scale, in spite of Patent Owner’s attempts to inject scale where it
`
`does not exist. See Paper 6 at 40-42.
`
`What Simpson does teach is a layout of memory chips on a memory module
`
`that may be implemented with any size chips on any size board. See Ex. 1002 at
`
`13:3-6; see also Ex.1022 at ¶¶17-18. Simpson explicitly states that while, “[t]he
`
`first example is based on a standard 72 terminal DRAM memory module but . . .
`
`this technique can equally be applied to . . . 30, 144 and 168 terminal modules, and
`
`also to other types of memory.” Id. Towards that end, the Intel Specification
`
`complements Simpson by disclosing a known design standard for the height and
`
`width of a 168 terminal memory module. See Ex. 1003 at 7; Paper 2 at 36-37, 42-
`
`43; Ex. 1022 at ¶¶17-18.
`
`5
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`Given that Simpson explicitly states that its technique that can be applied to
`
`a variety of module and memory types (including a 168 pin module like the one
`
`shown in the Intel Specification), proposed claim 9 cannot be valid if one of
`
`ordinary skill would have been able to combine the references and would have had
`
`a reason to do so. The answer to both questions is undoubtedly, yes. 3
`
`First, with regard to ability to combine, as noted above, Simpson’s own
`
`disclosure states that his invention could be applied to different size modules not
`
`shown in the patent (including a 168 pin DIMM such as described in the Intel
`
`Specification) – clearly indicating that it was adaptable by a person of ordinary
`
`skill. See Ex. 1002 at 13:3-6, 14:10-12; see also Ex. 1022 at ¶18. Moreover, in the
`
`context of another IPR, Patent Owner’s own expert agreed that one of skill in the
`
`art would know how to design a memory module to meet a known design
`
`
`3 Petitioner also notes that “[t]he test for obviousness is not whether the features of
`
`a secondary reference may be bodily incorporated into the structure of the primary
`
`reference . . . Rather, the test is what the combined teachings of those references
`
`would have suggested to those of ordinary skill in the art.” In re Keller, 642 F.2d
`
`413, 425 (C.C.P.A. 1981). See also In re Sneed, 710 F.2d 1544, 1550 (Fed. Cir.
`
`1983) (“it is not necessary that the inventions of the references be physically
`
`combinable to render obvious the invention under review.”).
`
`6
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`specification. Ex. 1014 at 170:7-8 (“Again, you can design anything according to
`
`any specification you’re given.”); see also Ex. 1006 at ¶¶95-96 (Dr. Subramanian
`
`agrees with Dr. Bernstein).
`
`Moreover, any such designs would have been well within the capabilities of
`
`one of skill in the art. Ex. 1006 at ¶38; Ex. 1022 at ¶30. As Simpson explains, “a
`
`wiring configuration can easily be designed by a person skilled in the art” to wire a
`
`memory chips on a PCB memory module. Ex. 1002 at 14:27-30; see also Ex. 1006
`
`at ¶38. Consequently, one of skill in the art would not have encountered technical
`
`difficulties in designing a memory module to both meet the design specifications
`
`of the Intel Specification and incorporate Simpson’s chip layout, but instead would
`
`have enjoyed a very reasonable expectation of success. Ex. 1006 at ¶¶37-41; Ex.
`
`1022 at ¶30.
`
`To the extent that Patent Owner may argue that the components described in
`
`Simpson could not fit on a module with a height of between 1 and 1.2 inches, a
`
`reference known to Patent Owner from copending IPR2016-01623 very clearly
`
`proves otherwise. The Kiehl reference, a published German patent application,
`
`clearly demonstrates that it was possible to fit two rows of memory chips in an
`
`arrangement as shown in Simpson on a 5.25” x 1.2” PCB. FIG. 1 of Kiehl
`
`(reproduced below) illustrates two rows of memory chips, the top row oriented
`
`vertically and the second row oriented horizontally. Ex. 1022 at ¶20. Kiehl
`
`7
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`explicitly states that the memory module has a width of 5.25 in (13.33 cm)” and a
`
`“height of less than 1.2 in (3.048 cm), or preferably less than 1.125 in (2.85 cm).”
`
`Ex. 1021 at [0002]; see also [0024]. Kiehl’s memory chips fit within a height of
`
`1.125 inches even with one memory chip oriented vertically and lying above a
`
`horizontally oriented memory chip. Kiehl’s memory chips would clearly fit within
`
`the same height if oriented with Simpson’s chip layout. Ex. 1022 at ¶¶20-22.
`
`
`
`To remove any doubt, Kiehl states that the memory chips shown in FIG. 1
`
`have a “width b of about 8mm” (0.315 inches) and a “length l of about 14 to
`
`16mm” (0.557 - 630 inches). Ex. 1021 at [0024]. Stacked two high, as in Simpson,
`
`the total height would be only 0.630 inches leaving over half an inch (0.570
`
`inches) for memory module pins, memory sockets, and spacing between chips. Ex.
`
`1022 at ¶¶21-22.
`
`In addition, there were numerous other memory chips available as early as
`
`1999 which would fit comfortably on a 5.25” x 1.2” PCB in the pattern of
`
`Simpson. See e.g., Ex. 1011 at 3-1, 3-2; Ex. 1012 at 12; Ex. 1002 at Fig. 6, [0023].
`
`For example, the Intel TSOP memory chips shown on page 3-1 of Ex. 1011 have a
`
`8
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`height (“Package Body Width E”) of 0.315 inches and a length (“Terminal
`
`Dimension D”) of 0.787 inches. Stacked two high, as in Simpson, the total height
`
`would be only 0.630 inches leaving over half an inch (0.570 inches) for memory
`
`module pins and spacing between chips. Ex. 1022 at ¶¶23-24. Furthermore, if
`
`arranged along the length of the PCB as shown in Simpson (four chips lengthwise
`
`and one vertical) the total length of the PCB that the chips would occupy is only
`
`3.463 inches (4 * 0.787 + 0.315) leaving 1.787 inches of space for chip spacing
`
`and other board components. Ex. 1022 at ¶¶23-24.
`
`Similarly, the Micron SDRAM chips of Ex. 1012 could be arranged
`
`according to Simpson’s layout to achieve a total height of 0.926 inches, thereby,
`
`leaving 0.274 inches of PCB space for memory module pins and chip spacing, and
`
`a total length of 3.936 inches, leaving 1.287 inches of PCB space for chip spacing
`
`and other board components. Ex. 1022 at ¶¶25-26.
`
`Lastly, Karabatsos discloses SDRAM chips having a height of 22.62 mm
`
`(0.891 inches) and width of 11.76 mm (0.463 inches). See Ex. 1017 at Fig. 6,
`
`[0023]. Similar to the Micron chips, Karabatsos’ chips could be arranged according
`
`to Simpson’s layout to achieve a total height of 0.926 inches, thereby, leaving
`
`0.274 inches of PCB space for memory module pins and chip spacing, and a total
`
`9
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`length of 4.027 inches, leaving 1.223 inches of PCB space for chip spacing and
`
`other board components.4 Ex. 1022 at ¶¶27-29.
`
`To the extent that the Patent Owner claims that the Simpson’s sockets would
`
`also somehow make a height of 1.2 inches impossible, sockets at the time were
`
`available that were no wider than the memory chip packages and which were
`
`specifically designed to permit “close mounting of the package[s] on [a] printed
`
`board.” Ex. 1013 at col. 2:31-32; see also Figs. 2, 5, 8, and col. 2:28-32, col. 2:47-
`
`31, col. 5:18-24; Ex. 1022 at ¶30. Accordingly, neither chip size nor socket size
`
`would have prevented a person of ordinary skill who was knowledgeable about
`
`Simpson from implementing its teachings on a standard sized Intel memory
`
`module using known components before the ’414 Patent was filed. Id.
`
`Beyond even this, however, the teachings of Simpson is not the particular
`
`embodiment shown in its figures, id., but the design of a memory module that
`
`permits expansion by adding additional chips. Ex. 1002 at 11:14-16 (“the invention
`
`
`4 To the extent that Patent Owner might argue that 0.274 inches of remaining
`
`vertical space is insufficient, Karabatsos discloses a minimum PCB height of 1.125
`
`inches, leaving only 0.234 inches of space even above one vertical memory chip
`
`(1.125 - 0.891 = 0.234), which must be sufficient space according to Karabatsos.
`
`See also Ex. 1022 at ¶29.
`
`10
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`can be said to provide [a memory module] to which individual memory devices
`
`may be added to increase the memory capacity by discrete amounts.”). In fact,
`
`Simpson even invites modifications of its example embodiment. Id. at 14:10-12
`
`(“The quantity, position and type [of memory chips and sockets] are dependent
`
`upon the design preferences of the module designer.”) (emphasis added); Ex.
`
`1022 at ¶17.
`
`2.
`
`There was a strong motivation to apply Simpson’s chip
`arrangement on a 5.25” by 1.2” PCB.
`There are several reasons why a person of ordinary skill in the art would be
`
`motivated to implement Simpson’s invention on a circuit board that is between 1
`
`inch and 1.2 inches in height as described in the Intel Specification. First, Simpson
`
`acknowledges that various aspects of its memory module are “dependent upon
`
`the design preferences of the module designer.” Ex. 1002 at 14:10-12 (emphasis
`
`added). With “constraints for the mechanical design of printed circuit boards
`
`[continuing to mature] in the years leading up to the priority date of the ’414
`
`Patent,” PCB board size would clearly be one such design preference. Ex. 1006 at
`
`¶39; Ex. 1017. at [0020]-[0022]. For instance, Simpson was an example of a
`
`printed circuit board design circa 1995 whereas the Intel Specification outlined
`
`“the mechanical design and tolerance constraints at the turn of the millennium.”
`
`Ex. 1006 at ¶39. Moreover, “one of ordinary skill in the art would recognize that
`
`modifying Simpson to comply with the physical constraints supplied by the Intel
`
`11
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`Specification would be a way to modernize Simpson to work with current
`
`motherboards and contemporary technology.” Id. at ¶40. Indeed, Simpson
`
`explicitly states that his memory design can be applied to various sized memory
`
`modules including a “168 terminal module” such as that disclosed in the Intel
`
`Specification. Ex. 1002 at 13:3-6; Ex. 1003 at 7 (The Intel Specification “defines . .
`
`. requirements for 168-pin . . . Dual In-Line Memory Modules.”). Thus, a person of
`
`ordinary skill in the art would want to combine Simpson with the Intel
`
`Specification and adopt the “physical constraints” consistent with circa 2000
`
`memory modules. Ex. 1006 at ¶38.
`
`Second, by “mid-2001, the dimensions of printed circuit boards had been
`
`standardized for some time.” Paper 2 at 36; see also Ex. 1006 at ¶95. In fact,
`
`Exhibit 1003 design specification for a 168-pin SDRAM DIMM, which indicates
`
`that those of skill in the art knew of design requirements to produce a DIMM that
`
`was standardized at 1 inch to 1.5 inches in height (i.e., the size sellable for use in
`
`PCs). See Ex. 1006 at ¶96; see also Ex. 1003 at 11 & 13. As such, the Intel
`
`Specification describes the circa 2000 modern standards for printed circuit board
`
`designs such as those described in Simpson. Ex. 1006 at ¶95. Given that DIMM
`
`sizes were standardized by 2000, one of ordinary skill would know that the
`
`invention of Simpson would need to be placed on a standard size memory module
`
`to fit consumer PCs, which use “standard” parts. Id. at ¶¶95-96. And, as
`
`12
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`demonstrated above, appropriately sized memory chips existed at the time that
`
`permitted those of skill in the art to achieve such industry design standards.
`
`Third, there was a recognized market need for “low profile” memory
`
`modules. As explained in the Petition in IPR2017-00974, “Karabatsos describes
`
`that the market demand for higher capacity memory at the time resulted in
`
`manufacturers producing many memory modules (e.g., DIMMs) that had ‘profiles
`
`of 1.5 inches or greater.’ Ex. 1015 at 21 (citing Ex. 1017 at [0004]-[0020]).
`
`Moreover, Karabatsos had recognized that the “unavailability of DIMMs with
`
`lower profiles [had] had a serious, negative effect on many computer designs.” Ex.
`
`1017 at [0020]. Karabatsos further explains the negative effect of tall DIMMs on
`
`computer designs lead to server manufacturers designing enclosures around the
`
`height of the memory modules or taking “extraordinary steps” to “slant[] . . .
`
`DIMM sockets at 22½ degrees.” Ex. 1017 at [0021], [0022]. One of ordinary skill
`
`would be aware of this discussion and the solution being provided by Karabatsos
`
`and would, thus, recognize that the problem of DIMM profile height applied
`
`generally to memory modules, including the ones described in Simpson. See Ex.
`
`1016 at ¶¶50-52, 56. When constructing a memory module to implement
`
`Simpson’s invention on a circuit board that is between 1 inch and 1.2 inches in
`
`height as stated by the Intel Specification, one of ordinary skill in the art would be
`
`13
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`motivated to apply the height of Karabatsos to gain the advantages of a lower
`
`profile height in an upgradable memory module according to Simpson. See id.
`
`Lastly, the Supreme Court has recognized that “[w]hen there is a design
`
`need or market pressure to solve a problem and there are a finite number of
`
`identified, predictable solutions, a person of ordinary skill has good reason to
`
`pursue the known options within his or her technical grasp,” and such solutions are
`
`“likely the product not of innovation but of ordinary skill and common sense.”
`
`KSR v. Teleflex, 550 U.S. 398, 421 (2007); see also MPEP §2143 I(F). In addition,
`
`“when a patent ‘simply arranges old elements with each performing the same
`
`function it had been known to perform’ and yields no more than one would expect
`
`from such an arrangement, the combination is obvious.” KSR at 550 US at 417
`
`(quoting Sakraida v. Ag Pro, Inc., 425 U.S. 273 (1976)).
`
`Therefore, substitute claim 9 (and indeed original claim 4) of the ’414 Patent
`
`are not patentable under 35 § U.S.C. 103.
`
`3.
`
`Patent Owner’s assertions that claim 9 (and claim 4) are
`patentable are disingenuous at best.
`In its Motion to Amend, Patent Owner brazenly asserts that “[t]he ’414
`
`Patent’s novel recognition and utilization of those features permitted the PCB
`
`height to be reduced to 1.0–1.2 inches, resulting in a combination of layout and
`
`height that is absent in the prior art.” Paper 18 at 2. Yet, despite its arguments to
`
`the contrary, see Paper 6 at 43, Patent Owner demonstrably knew that it was
`
`14
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`possible to fit two rows of memory chips on a PCB having a height of between 1.0
`
`and 1.2 inches. In Patent Owner’s Response to the Petition in IPR2016-01623,
`
`Patent Owner quotes Kiehl’s teaching that “existing types of DIMMs … should
`
`have a width and/or height of less than 1.2 in (3.048 cm), or preferably less than
`
`1.125 in (2.85 cm). The goal is to preserve these dimensions while
`
`accommodating 36 DRAMs on one PCB.” Ex. 1024 at 55 (quoting Kiehl at
`
`[0002]) (emphasis in original).
`
`Yet, having this knowledge, Patent Owner failed to both disclose Kiehl and
`
`show patentability over it in the Motion to Amend. This is in direct violation of
`
`both the Board’s explicit instructions and the law. See Paper 17 at 4 (citing
`
`Microsoft v. Proxyconn, 789 F.3d at 1306–08; MasterImage 3D, Inc. v. RealD,
`
`Inc., Case IPR2015-00040, slip op. at 1–2 (PTAB July 15, 2015) (Paper 42)
`
`(precedential)). As demonstrated above, Patent Owner clearly both knew about the
`
`Kiehl reference and should have recognized its materiality to its amendments.
`
`B.
`
`Simpson in view of Karabatsos Discloses “said printed circuit
`board . . . has a height of 1 to 1.2 inches perpendicular to said
`contact strip,” as Recited in Substitute Claim 9 and Claim 4.
`Petitioner explained in its IPR2017-00974 Petition the details of how
`
`Simpson in view of Karabatsos disclose and render obvious the limitations of
`
`claim 4 which Patent Owner has proposed as an amendment to claim 8. See
`
`Ex. 1015 at 15-39. Moreover, as noted above, the Board has already recognized
`
`15
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`that there is a reasonable likelihood that all the features of substitute claim 9 are
`
`unpatentable in view of Simpson, except for Patent Owner’s amendment to
`
`incorporate a height limitation of 1 to 1.2 inches. Paper 7 (Instituting review of
`
`claims 1 and 8.). Karabatsos discloses “[a] low profile, registered DIMM [that] has
`
`a height of about 1.2 inches, and a width of about 5.25 inches.” Ex. 1017 at
`
`Abstract, FIG. 1C. Karabatsos specifically discloses a memory module having a
`
`“height of between 1.125 and 1.250 inches.” Ex. 1017 at [0027].
`
`In addition, Karabatsos discusses a particular motivation for designing a
`
`memory module to have a lower profile of 1.2 inches. MPEP § 2134(I)(G) (citing
`
`DyStar Textilfarben GmbH & Co. Deutschland KG v. C.H. Patrick Co.,464 F.3d
`
`1356, 1360, (Fed. Cir. 2006)). As discussed above, Karabatsos recognized the
`
`negative effects on computer designs caused by the “unavailability of DIMMs with
`
`lower profiles” and sought to address this need by designing a lower profile
`
`memory module. Ex. 1017 at [0004]-[0020]. One of skill in the art, having made
`
`the same recognition as Karabatsos, would have been clearly have been motivated
`
`to design Simpson’s memory module to have a lower profile height of about 1.2
`
`inches to alleviate this need for lower profile memory modules while still
`
`providing a readily upgradeable memory module, as envisioned by Simpson. Ex.
`
`1016 at ¶¶50-52, 56.
`
`16
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`Furthermore, Petitioner demonstrates above that it was entirely possible for
`
`one of skill in the art to apply Simpson’s memory chip layout on a 5.25” x 1.2”
`
`PCB using components that were known at least as early as 1999, including chip
`
`sizes disclosed by Karabatsos. Ex. 1022 at ¶¶19-30.
`
`Therefore, for the reasons set forth herein and in Petitioner’s IPR2017-00974
`
`Petition, substitute claim 9 and original claim 4 of the ’414 Patent are not
`
`patentable under 35 § U.S.C. 103.
`
`C. Bechtolsheim in view of Tokunaga and Karabatsos Discloses a
`Memory Module as Recited in Claim 1 where “said printed circuit
`board . . . has a height of 1 to 1.2 inches perpendicular to said
`contact strip,” as Recited in Substitute Claim 9 and Claim 4.
`Petitioner explained in its IPR2017-00974 Petition the details of how
`
`Bechtolsheim in view of Tokunaga and Karabatsos disclose and render obvious the
`
`limitations of claim 4 which Patent Owner has proposed as an amendment to
`
`claim 8. See Ex. 1015 at 39-55.
`
`1.
`
`Bechtolsheim in view of Tokunaga and Karabatsos
`Discloses a Memory Module as Recited in Claim 1.
`For example, as shown in FIG. 2a of Bechtolsheim below and described in
`
`Ex. 1015 at 39-50, Bechtolsheim discloses:
`
`An electronic printed circuit board configuration [see Fig. 2A], comprising:
`
`an electronic printed circuit board having a contact strip for insertion into
`
`another electronic unit [annotated in green below]; and
`
`17
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`a memory module having at least nine identically designed integrated
`
`semiconductor memories [annotated in red below];
`
`each one of said semiconductor memories being encapsulated in a
`
`rectangular housing having a shorter dimension and a longer dimension;
`
`said housing of each one of said semiconductor memories being identically
`
`designed and being individually connected to said printed circuit board;
`
`. . .
`
`said longer dimension of said housing of each one of said semiconductor
`
`memories . . . being oriented parallel with said contact strip, as recited in original
`
`claim 1 of the ’414 Patent.
`
`
`
`Regarding the limitation of “one of said semiconductor memories being
`
`connected as an error correction chip,” Bechtolsheim discloses that the memory
`
`module (SIMM) can include error correction. See e.g., Ex. 1018 at 3:64-67, claims
`
`5 and 10 (“contact pads [of the contact strip] include contact pads dedicated for
`
`data comprising an error correction code.”). Furthermore, in the same filed of
`
`18
`
`
`
`Case IPR2016-01622
`Attorney Docket No. 37307-0007IP1
`semiconductor memory module design, Tokunaga demonstrates that it was well-
`
`known before the priority date of the ’414 patent to use typical semiconductor
`
`memory devices for “ECC function[s] and/or parity function[s]” to “check errors in
`
`input/output data” of the memory devices and ultimately the memory module. Ex.
`
`1019 at 4:62-67; 9:11-16.
`
`Moreover, because Bechtolsheim already anticipates including error
`
`checking capabilities in the disclosed memory module, evidenced by the “contact
`
`pa