Eidos Display, LLC et al v. AU Optronics Corporation et al
UNSEALED VERSION OF 713 MEMORANDUM OPINION AND ORDER granting in part and denying in part 682 Motion to Exclude the Opinion and Testimony of Plaintiffs' Expert Arthur Cobb. Signed by Magistrate Judge John D. Love on 3/29/2017. (rlf)
IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF TEXAS
EIDOS DISPLAY, LLC,
CHI MEI INNOLUX CORPORATION,
CHI MEI OPTOELECTRONICS USA,
INC., CHUNGHWA PICTURE TUBES,
NORTH AMERICA, INC.,
CIVIL ACTION NO. 6:11-CV-00201-JRG
MEMORANDUM OPINION AND ORDER
Before the Court is Defendants’ Motion to Exclude the Opinions and Testimony of
Plaintiffs’ Expert Arthur Cobb. (Doc. No. 682.) Plaintiffs have filed a response. (Doc. No. 697),
and Defendants have filed a reply (Doc. No. 709). Upon consideration, Defendants’ Motion
(Doc. No. 682) is GRANTED-IN-PART and DENIED-IN-PART as set forth herein.
On April 25, 2011, Plaintiffs filed suit against AU Optronics Corporation, AU Optronics
Corporation America, Chunghwa Picture Tubes, Ltd., HannStar Display Corporation, Hannspree
North America, Inc., Chi Mei Innolux Corporation, and Chi Mei Optoelectronics USA, Inc.,
alleging infringement of U.S. Patent No. 5,879,958 (“the ’958 Patent”). (Doc. No. 1.) The ’958
Patent is titled “Method of Producing an Electro-Optical Device” and relates to the process of
forming circuitry used in controlling liquid crystal displays (“LCD”). Specifically, the ’958
Patent relates to the process for forming an array of thin film transistors (“TFT”) and pixel
electrodes that are used to control the light emission of an LCD product. Notably, the ’958 Patent
contains 17 embodiments (identified as A-S) providing various manufacturing processes that
reduce the number of photolithographic steps. See ’958 Patent at 4:50–14:18 (describing
processes with four or five lithographic steps as opposed to seven).
Claim 1 is the only issued claim in the ’958 Patent. Claim 1 recites as follows:
1. A method for producing an electro-optical device
in which an electro-optical material is put between
a pair of substrates opposed to each other, at least
a portion of opposing surfaces of the substrates is
insulative, a plurality of source wirings and a
plurality of gate wirings are formed crossing each
other on the surface of one of said pair of
substrates and a transparent pixel electrode and a
thin film transistor are formed at each of the
crossing points between the source wirings and the
gate wirings, wherein the method comprises:
a step G1 of forming a first metal film on
the surface of said one substrate,
a first photolithographic step G2 of
patterning the first metal film to
form a gate electrode and a gate
a step G3 of forming a first insulator film,
a semiconductor film and an ohmic
contact film on the surface of said
one substrate after the first
a second photolithographic step G4 of
patterning the semiconductor active
film and the ohmic contact film to
form a semiconductor portion
above the gate electrode in a state
isolated from other portions,
a step G5 of forming a second metal film
on the surface of said one substrate
after the second photolithographic
a third photolithographic step G6 of
patterning the second metal film
and the ohmic contact film to form
a source electrode, a drain electrode
and a channel portion,
a step G7 of forming a passivation film on
the surface of said one substrate
after the third photolithographic
a fourth photolithographic step G8 of
patterning the passivation film to
form a contact hole reaching the
gate wiring, a contact hole reaching
the drain electrode and a contact
hole for source wiring and gate
wiring connection terminals,
a step G9 of forming a transparent
conductive film on the surface of
said one substrate after the fourth
photolithographic step, and
a fifth photolithographic step G10 of
patterning the transparent
conductive film to form a
transparent pixel electrode.
’958 Patent at 58: 5–47.
On March 6, 2017, Defendants brought the instant Daubert Motion to exclude certain of
Mr. Cobb’s damages opinions. (Doc. No. 682.)
Rule 702 provides that an expert witness may offer opinion testimony if: (a) the expert’s
scientific, technical, or other specialized knowledge will help the trier of fact to understand the
evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c)
the testimony is the product of reliable principles and methods; and (d) the expert has reliably
applied the principles and methods to the facts of the case. Fed.R.Evid. 702.
The Rules also “assign to the trial judge the task of ensuring that an expert’s testimony
both rests on a reliable foundation and is relevant to the task at hand.” Daubert v. Merrell Dow
Pharms. Inc., 509 U.S. 579, 594, 597 (1993). “The relevance prong [of Daubert] requires the
proponent [of the expert testimony] to demonstrate that the expert’s ‘reasoning or methodology
can be properly applied to the facts in issue.’” Johnson v. Arkema, Inc., 685 F.3d 452, 459 (5th
Cir. 2012) (quoting Curtis v. M & S Petroleum, Inc., 174 F.3d 661, 668 (5th Cir. 1999)). “The
reliability prong [of Daubert] mandates that expert opinion ‘be grounded in the methods and
procedures of science and . . . be more than unsupported speculation or subjective belief.’”
Johnson, 685 F.3d at 459 (quoting Curtis, 174 F.3d at 668).
In assessing the “reliability” of an expert’s opinion, the trial court may consider a list of
factors including: “whether a theory or technique . . . can be (and has been) tested,” “whether the
theory or technique has been subjected to peer review and publication,” “the known or potential
rate of error,” “the existence and maintenance of standards,” and “general acceptance” of a
theory in the “relevant scientific community.” Daubert, 509 U.S. at 593–94; see also Kumho
Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 150 (1999) (“Daubert makes clear that the factors it
mentions do not constitute a ‘definitive checklist or test.’”); U.S. v. Valencia, 600 F.3d 389, 424
(5th Cir. 2010). “The proponent need not prove to the judge that the expert’s testimony is
correct, but she must prove by a preponderance of the evidence that the testimony is reliable.”
Johnson, 685 F.3d at 459 (quoting Moore v. Ashland Chem., Inc., 151 F.3d 269, 276 (5th Cir.
1998) (en banc)). At base, “the question of whether the expert is credible or the opinion is
correct is generally a question for the fact finder, not the court.” Summit 6, LLC v. Samsung
Elecs. Co., Ltd., 802 F.3d 1283, 1296 (Fed. Cir. 2015).
Defendants move to exclude Mr. Cobb’s damages opinions on 5 bases: (1) Mr. Cobb’s
opinions violate the Entire Market Value Rule (“EMVR”); (2) Mr. Cobb does not assess the
economic footprint of Claim 1; (3) Mr. Cobb attempts to correct his over-inclusive royalty base
by lowering his rate; (4) Mr. Cobb calculates damages on “indirect sales” not linked to an act of
infringement; and (5) Mr. Cobb uses an incorrect hypothetical negotiation date. (Doc. No. 682.)
Defendants argue that Mr. Cobb admits the smallest saleable unit is the LCD panel but
improperly uses LCD modules as his base. (Doc. No. 682, at 9, citing Doc. No. 682-4, 222:7–
11.) Defendants contend that therefore in order to use the LCD modules as his base, Mr. Cobb
must meet the requirements of the EMVR—that is that the process covered by claim 1 is the
basis for consumer demand. Id. Defendants contend Mr. Cobb fails to do so and instead only
opines that the technology in claim 1 reduces manufacturing costs for TFT arrays and the cost
reductions are the basis for consumer demand. Id. at 10. Therefore, Defendants argue that Mr.
Cobb should not be able to base his opinions on the billions of dollars in accused sales because it
will only skew the damage horizon for the jury. Id. at 11.
Eidos argues that Mr. Cobb’s royalty base “comports with consistent industry practice to
include module and panel sales in LCD patent license agreements.” (Doc. No. 697, at 15.) Eidos
argues that numerous LCD patent licenses confirm the standard approach is to pay a royalty on
all LCD products sole under the license, including modules and panels. Id. at 16. Eidos argues
that even if EMVR applies, there is a question for the jury because claim 1 creates price
reductions that drive the sales of accused products. Id. at 17–18.
In determining a reasonable royalty for a multi-component product, it is generally
required that “royalties be based not on the entire product, but instead on the ‘smallest salable
patent-practicing unit.’” LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51, 67 (Fed.
Cir. 2012). The entire market value rule is an exception to this rule, and “allows a patentee to
assess damages based on the entire market value of the accused product only where the patented
feature creates the ‘basis for customer demand’ or ‘substantially create[s] the value of the
component parts.’” Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292, 1318 (Fed. Cir. 2011)
(quoting Lucent Techs. V. Gateway, Inc., 580 F.3d 1301, 1336 (Fed. Cir. 2009); Rite–Hite Corp.
v. Kelley Co., 56 F.3d 1538, 1549–50 (Fed. Cir. 1995)).
Here, Eidos does not dispute that its royalty base includes LCD modules that are beyond
the smallest saleable patent-practicing unit. While the Court agrees there is no firm rule that
requires the royalty base to be comprised of the smallest saleable patent-practicing unit, Mr.
Cobb’s royalty base necessarily implicates the EMVR where he ultimately opines on a royalty
based on the entire market value of the LCD module. Notably, Eidos has not shown that claim 1
of the ’958 Patent is the basis for consumer demand of the LCD modules. Eidos only contends
that the benefits of claim 1 resulted in more competitive prices and price reductions that drove
the sale of the accused products. The fact that the claimed process reduced costs—even if true—
does not show it was the basis for consumer demand of the products themselves. Put simply,
Eidos provides no evidence of the requisite consumer demand.
Moreover, Eidos is incorrect in suggesting that this question could be presented to the
jury because there is no evidence on which the jury could reach a conclusion that claim 1 drove
the consumer demand for LCD modules. Rather, posing the sales of the entire product to the jury
where there is no evidence of consumer demand based on the patented feature would only work
to improperly inflate damages for the jury. See LaserDynamics, 694 F.3d at 68 (“Admission of
such overall revenues, which have no demonstrated correlation to the value of the patented
feature alone, only serve to make a patentee’s proffered damages amount appear modest by
comparison, and to artificially inflate the jury’s damages calculation beyond that which is
‘adequate to compensate for the infringement.’”) (quoting 35 U.S.C. § 284).
Similarly, Eidos’s argument that it is the typical practice to license LCD patents this way
is unavailing. While comparable licenses can serve to inform what might have occurred during a
hypothetical negotiation, they cannot expand a royalty beyond the economic demand for the
Here, the parties agree that the smallest saleable unit is the LCD panel. Mr. Cobb will not
be permitted to include the LCD modules in his royalty base. Moreover, as will be discussed
further herein, Mr. Cobb must further apportion the value of the patented process even where
using the LCD panel as the royalty base. Accordingly, Defendants’ Motion is GRANTED as to
Mr. Cobb’s use of LCD modules in his royalty base.
Defendants argue that Mr. Cobb admits that claim 1 relates solely to the manufacture of
the TFT array and that he must therefore apportion his base down to the sales of TFT arrays or
use the TFT array as his base. (Doc. No. 682, at 11.) Defendants contend that Mr. Cobb did no
such apportionment. Id. at 12. Instead, Defendants argue that Mr. Cobb opines that the “[c]ost
reduction allowed by the technologies of the ‘958 Patent create substantial economic value by
allowing the LCD manufacturer to compete on the basis of price in the highly competitive
market.” Id. at 13, citing Ex. A at 76. Defendants contend that even if this is true, it says nothing
about the value of claim 1 over methods described in the prior art. Id.
Eidos argues that Mr. Cobb properly ties his analysis to the “economic footprint” of
claim 1 as his primary consideration. (Doc. No. 697, at 12.) Eidos cites to the following
deposition testimony as evidence of Mr. Cobb’s apportionment:
A. So what I – I’ve looked at is the ’958 patent. I understand from my reading
that it reduces the number of steps, improving yield, as compared to prior
method by production methods in prior art.
Q. I'm asking you about –
A. And then I understand through discussions with -- with Dr. Smith that there
are significant cost savings, significant yield savings. So I have one claim in
the patent. I have the -- the patent reducing the number of steps, increasing
yield and reducing costs. So what I've done is made considerations of such
reductions, that such reductions exist, and then have conducted an analysis of
a licensing of that patent, with the understanding that the patent has a single
claim and that the patent improves yield and improves cost performance.
(Doc. No. 697, at 13, citing Doc. No. 697-10, at 369:1–18.)
In a reasonable royalty analysis, damages must be tied “to the claimed invention’s
footprint in the marketplace.” ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 869 (Fed. Cir.
2010). Thus, “where multi-component products are involved, the governing rule is that the
ultimate combination of royalty base and royalty rate must reflect the value attributable to the
infringing features of the product, and no more.” Ericsson, Inc. v. D-Link Sys., Inc., 773 F.3d
1201, 1226 (Fed. Cir. 2014) (citing VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1326 (Fed.
Cir. 2014)). When the accused products have both patented and non-patented features, the
damages analysis requires an apportionment analysis in order to determine the value added by
such features. Id.
Here, the parties’ ultimate dispute is over the novelty of the claimed process—i.e. the
incremental benefit of the claimed invention added over the processes already known in the art.
This dispute was similarly evidenced in Plaintiffs’ Daubert motion challenging the opinions of
Defendants’ damages expert, Mr. Jackson. (Doc. No. 678.) There, Eidos argues that Mr. Jackson
improperly focuses his apportionment analysis on step G8 of claim 1 as the incremental benefit
and value of the claimed technology. (Doc. No. 678, at 10.) Eidos contends that the “economic
footprint” of the claimed process is broader than the G8 step or the TFT arrays. (Doc. No. 697, at
6–7.) Eidos therefore contends that it need not apportion from the LCD panels. See Doc. No.
697, at 14 (“Because LCD panels are closely related to the invention, LCD panels should be
included in the royalty base even if apportionment is required, which is disputed.”) As this Court
has explained in ruling on Eidos’s Daubert motion (Doc. No. 678), the dispute over the novelty
of the claimed invention is a question for the fact finder. However, the question ofwhether an
expert’s opinion comports with the settled principles of apportionment, is for the Court to decide
in exercising its gatekeeping authority. VirnetX, 767 at 1328 (explaining that a district court must
exercise “its gatekeeping authority to ensure that only theories comporting with settled principles
of apportionment were allowed to reach the jury”).
Claim 1 of the ’958 Patent is ultimately directed at forming a TFT array. ’958 Patent at
58: 5–47. As Eidos even concedes, “[t]he technical experts agree that the method in Claim 1
covers the entire TFT (or ‘active’) array substrate.” (Doc. No. 697, at 14, citing Doc. No. 697-12
at 493; Doc. No. 697-14, at 28:15–18; Doc. No. 697-15, at 161:08–18.) While the Court does not
disagree that the ’958 Patent is generally related to the fabrication of a LCD panel, what is
ultimately claimed in claim 1 is a method that creates the TFT array that is used to control the
light emission of the LCD panel. ’958 Patent at 58: 5–47. Here, what is claimed is indisputably
not an LCD panel, and the LCD panel indisputably has features not attributable to the claimed
invention. Therefore, the Court finds that apportionment must be done to value the claimed
invention. While Eidos may provide its evidence that the claimed process is integral to the
development of the LCD panels, Mr. Cobb nonetheless must separate the value of the patented
features from the unpatented features of the LCD panels. CISRO v. Cisco Sys., Inc., 809 F.3d
1295, 1301 (Fed. Cir. 2015) (“to be admissible, all expert damages opinions must separate the
value of the allegedly infringing features from the value of all other features.”) Here, there is no
evidence that Mr. Cobb undertook this analysis in rendering his opinions on a reasonable royalty
in this case. Accordingly, Defendants’ Motion as to Mr. Cobb’s failure to apportion is
c. Rate Adjustment Depending on Base
Defendants contend that Mr. Cobb sets forth three royalty base unit scenarios—Module,
Panel, and TFT Arrays—and proposes royalty rates specific to each royalty base unit. Doc. No.
682, at 14, citing Doc. No. 682-1, at 3, 97. Defendants contend that as the royalty base goes
down from modules to panels and arrays, Mr. Cobb’s alleged royalty rate increases. For
example, Defendants contend that as the dollar amount of his base decreases from modules to
panels to arrays, Mr. Cobb simply increases his alleged running royalty rate from 0.75% to 1% to
2%, without explanation. Id. Defendants argue that Mr. Cobb’s analysis of the royalty rate is
further undermined by his inability to point to a single comparable license on which he relied in
reaching his conclusion. Id. Eidos argues that Mr. Cobb adjusted his rates based on GeorgiaPacific factors and did not improperly adjust his rates to maintain the same measure of damages
for each royalty base. (Doc. No. 697, at 11–12.)
The moving rates for Mr. Cobb’s range of bases is entirely unclear. Mr. Cobb does not
explain, for example, how the different bases would have impacted the hypothetical negotiation
such that the corresponding rates would be increased. Regardless, Mr. Cobb’s base including
LCD modules will not be permitted for the reasons explained above. Similarly, Mr. Cobb has
been directed to conduct an apportionment analysis on his LCD panel base. To the extent Mr.
Cobb wishes to maintain his alternative bases and rates between the LCD panels and TFT arrays
in supplementing his report, Mr. Cobb must explain why the royalty rate doubles when the base
is comprised of LCD panels instead of TFT arrays. Accordingly, Defendants’ Motion as to Mr.
Cobb’s rate adjustment depending on base is GRANTED.
d. Indirect Sales
Defendants argue that Mr. Cobb’s statistics on indirect sales are not tied to actual
purported acts of infringement and therefore must be excluded as unreliable. (Doc. No. 682, at
16.) Eidos argues that Mr. Cobb’s royalty calculations for indirect U.S. sales are directly tied to
acts of infringement based on a reliable estimate of the number of accused products that have
been induced to be imported and sold in the U.S. through Defendants’ customers. (Doc. No. 697,
Defendants rely on Power Integrations to exclude Mr. Cobb’s opinions on indirect sales.
711 F.3d 1348 (Fed. Cir. 2013). In Power Integrations, the Federal Circuit found that the
damages expert’s testimony was both based on an unreliable data source and employed
unreliable methodology. Id. at 1373. Specifically, the evidence of worldwide sales came from a
source that was unclear and assumed to have come only from the Internet. Id. In addition, while
the technology accused was found in phone chargers, the damages expert included worldwide
sales of mobile phones, assuming that each phone was shipped with a charger. Id. The Federal
Circuit found that this leap was speculative as well as his assumption that each of the chargers
included an infringing power circuit. Id. at 1374.
Here, Mr. Cobb relies on general statistics of the number of TFT LCD products sold
within the United States. (Doc. No. 697-8, at 9–11.) Mr. Cobb also relies on data from form 10Ks for Dell, Apple, and HP—customers of Defendants—to show the net revenue for sales of
notebook computers and desktop monitors. Id. at 10.
As this Court stated in ruling on Defendants’ motions for summary judgment, Mr. Cobb
is permitted to testify on circumstantial evidence of indirect infringement through the use of
general statistics and specific data from each Defendant and their respective customers.
However, Mr. Cobb may not use this data to support damages for Eidos’s claims of indirect
infringement. If permitted to do so, Mr. Cobb’s methodology would suffer from many of the
same speculative leaps described in Power Integrations. For example, to rely on this data to
establish damages, one would have to assume that each of the LCD products sold within the
United States was made using the accused processes. Even if that speculative leap could be made
from the data, which it cannot, there is no way to determine what percentage is attributable to
any one Defendants’ customers or accused products. For example, according to Mr. Cobb’s
report, the data comes from “six conspiring manufacturers”: AU Optronics Corporation,
Chunghwa Picture Tubes, Chi Mei Optoelectronics Corp., HannStar Display Corp., Samsung
Electronics Corp. and LG Display Co. Id. at 9. The remaining Defendants constitute only 3 of 6
of the listed companies. Moreover, even among the Defendants one would have to speculate as
to what percentages each Defendant is responsible for selling through various U.S. customers.
Worse yet, even if that speculative leap could be made, there is no way to distinguish which of
these sales are the result of Defendants’ direct sales or indirect sales.
As such, these statistics cannot be relied upon to expand or infer the damages for indirect
sales. Rather, Mr. Cobb must provide damages opinions for Eidos’s claims of indirect
infringement only where there is specific evidence of importation or sales by a specific customer
such that Mr. Cobb can apply reliable principles and methods that are closely tied to the facts of
this case. That is, where, at a minimum, Eidos’s evidence provides the identity of the third-party
infringer and the accused products sold in the United States.
Accordingly, Defendants’ Motion is GRANTED in so much as Mr. Cobb employs
general statistics to estimate damages for indirect infringement.
e. Hypothetical Negotiation Date
Lastly, Defendants argue that Mr. Cobb assumes that a hypothetical negotiation date for
Innolux would be April 2002, but that Innolux began using the alleged 5-mask process in 1999.
(Doc. No. 682, at 17, citing Doc. No. 682-1, at 82.) Because Mr. Cobb does not use a
hypothetical negotiation date of 1999, Defendants argue his opinions are unreliable and should
be excluded. Id. Eidos contends that its hypothetical negotiation date is based on evidence of first
infringement because Innolux produced no sales information prior to 2002. (Doc. No. 697, at
“In general, the date of the hypothetical negotiation is the date that the infringement
began.” LaserDynamics, 694 F.3d at 75; see also Applied Med. Res. Corp. v. U.S. Surgical
Corp., 435 F.3d 1356, 1363–64 (Fed.Cir.2006) (“[T]he hypothetical negotiation relates to the
date of first infringement.”); State Indus., Inc. v. Mor–Flo Indus., Inc., 883 F.2d 1573, 1580
(Fed.Cir.1989) (“The determination of a reasonable royalty ... [is based] on what a willing
licensor and licensee would bargain for at hypothetical negotiations on the date infringement
Here, Innolux claims that the hypothetical negotiation date is 1999 because Innolux
“began using the alleged 5-mask process in 1999” and “produced the first large TGT-LCD
display panel to be made from Taiwan-developed technology in July 1999.” (Doc. No. 682, at
13.) But here, as Innolux points out in its own briefing, “the date of first infringement begins
when both the patent has issued and accused products have been sold.” (Doc. No. 682, at 17)
(emphasis added) (citing Wang Labs., Inc. v. Toshiba Corp., 993 F.2d 858, 869 (Fed. Cir. 1993).
The Court agrees that the hypothetical negotiation date should be the “start of infringement, i.e.,
when both a patent had issued and accused products were sold.” Wang Labs, 993 F.2d at 869.
Notably, Innolux only argues that it began using the alleged 5-mask process in 1999 and
produced the first large TGT-LCD display in July 1999. But Innolux fails to provide any
evidence that it sold an accused product in 1999. Importantly, because the process was carried
out abroad, merely using that process, as Innolux contends it did in 1999, would not constitute
infringement of the ’958 Patent. See Power Integrations, 711 F.3d at 1371 (“It is axiomatic that
U.S. patent law does not operate extraterritorially to prohibit infringement abroad.”). Similarly
producing a prototype abroad fails to establish an act of infringement. Id. Therefore, the first act
of infringement in this instance must be the first importation or sale within the United States.
Here, the evidence in the record indicates that such infringing sales began in 2002.
Accordingly, Mr. Cobb did not err in using a hypothetical negotiation date of April 2002.
The Court DENIES Defendants’ Motion as to Innolux’s hypothetical negotiation date.
Consistent with the rulings set forth herein, Defendants’ Motion (Doc. No. 682) is
GRANTED-IN-PART and DENIED-IN-PART. Within 14 days of the issuance of this Order,
Mr. Cobb shall file a supplemental report consistent with the findings herein. Defendants may
seek an additional deposition, not to exceed 3 hours, within 14 days of any supplementation by
Mr. Cobb. Defendants’ damages expert may also file a rebuttal report to Mr. Cobb’s
supplemental report within 14 days of being served, not to exceed 5 pages.
Within seven (7) days of the issuance of this Order, the parties shall file a notice with the
Court as to whether this Order can be unsealed, or request appropriate redaction.
So ORDERED and SIGNED this 29th day of March, 2017.
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?