FOR PUBLICATION
  UNITED STATES COURT OF APPEALS
       FOR THE NINTH CIRCUIT

JADA TOYS, INC. a California               
corporation,
                                                  No. 05-55627
        Plaintiff-Counterdefendant-
                          Appellee,                 D.C. No.
                  v.                           CV-04-02755-RGK
                                                     (FMO)
MATTEL, INC., a Delaware
                                                  AMENDED
corporation,
                                                   OPINION
       Defendant-Counterclaimant-
                         Appellant.
                                           
         Appeal from the United States District Court
            for the Central District of California
         R. Gary Klausner, District Judge, Presiding

                    Argued and Submitted
             April 12, 2007—Pasadena, California

                     Filed August 2, 2007
                   Amended February 21, 2008

    Before: Harry Pregerson and Johnnie B. Rawlinson,
   Circuit Judges, and Brian E. Sandoval,* District Judge.

                   Opinion by Judge Sandoval




  *The Honorable Brian E. Sandoval, United States District Judge for the
District of Nevada, sitting by designation.

                                 1563
1566               JADA TOYS v. MATTEL


                       COUNSEL

Adam R. Fox, Esq., Los Angeles, California, for plaintiff-
counterdefendant-appellee Jada Toys, Inc.
                      JADA TOYS v. MATTEL                    1567
Jill M. Pietrini, Esq., Los Angeles, California, for defendant-
counterclaimant-appellant Mattel, Inc.


                           OPINION

SANDOVAL, District Judge:

   Defendant-Appellant Mattel, Inc. (“Mattel”) appeals the
grant of summary judgment in favor of Jada Toys, Inc.
(“Jada”) on Mattel’s federal and state trademark infringement
counterclaims. Mattel also challenges the district court’s entry
of summary judgment in favor of Jada as to its dilution and
copyright claims.

   We hold that because the district court erred in its applica-
tion of the relevant infringement test, the district court’s entry
of summary judgment in Jada’s favor as to those claims is
reversed. We also hold that genuine issues of material fact
exist as to Mattel’s copyright and dilution claims and, there-
fore, the district court’s entry of summary judgment as to
those claims in favor of Jada is also reversed.

 I.   FACTUAL AND PROCEDURAL BACKGROUND

   Jada Toys is a California corporation that specializes in the
distribution and sale of miniature diecast toy cars, trucks, and
other vehicles. Generally, these vehicles are scale model repli-
cas of actual vehicles. From 2001 to 2004, Jada produced a
line of toy trucks called HOT RIGZ.

   In 2001, Jada filed an application for a trademark registra-
tion of the term HOT RIGZ with the United States Patent and
Trademark Office (“U.S. PTO”). The trademark was issued
by and registered with the U.S. PTO in 2002, though Jada
used its HOT RIGZ trademark in advertising material and on
its toys and their packaging from 2001 to 2004.
1568                    JADA TOYS v. MATTEL
   Mattel is also a toy company. Among its many lines of toys
is its familiar HOT WHEELS miniature vehicle brand, which
it has been marketing since 1968. The HOT WHEELS vehicle
line includes small scale versions of big rig trucks. Since
1968, Mattel has employed the use of a flame logo (“68
logo”) to identify the HOT WHEELS brand. In 1982 Mattel
developed a complementary version of the 68 logo, this one
incorporating the Mattel seal, to be used in conjunction with
the sale of its product. Mattel owns U.S. federal trademark
registrations for both of these flame logos.

   On April 20, 2004, Jada filed an action against Mattel,
asserting claims for trademark infringement, false designation
of origin, and unfair competition. Jada’s allegations, however,
were not related to its HOT RIGZ mark; rather, Jada claimed
that Mattel’s advertising and sale of its OLD SCHOOL and
NEW SCHOOL lines infringed on Jada’s use of its registered
trademark OLD SKOOL. Mattel asserted various affirmative
defenses and counterclaims. Among the counterclaims were
allegations that Jada’s HOT RIGZ mark infringed on Mattel’s
HOT WHEELS mark.1 Mattel also counterclaimed for copy-
right infringement and dilution. Ultimately, each party sub-
mitted motions for summary judgment.

  On March 15, 2005, the district court issued its ruling
regarding the parties’ motions for summary judgment. As to
Jada’s claims relating to Mattel’s OLD SCHOOL and NEW
SCHOOL marks, the court granted summary judgment in
Mattel’s favor. And as to Mattel’s counterclaims for infringe-
ment, copyright, and dilution, relating to Jada’s use of its
HOT RIGZ logo, the court granted summary judgment in
Jada’s favor. In granting summary judgment as to Mattel’s
  1
    Mattel alleged four separate infringement claims: (1) trademark
infringement under 15 U.S.C. § 1114; (2) false designation of origin under
15 U.S.C. § 1125(a); (3) common law unfair competition and unfair com-
petition under California Business and Professions Code section 17200;
and (4) common law passing off.
                     JADA TOYS v. MATTEL                    1569
infringement claims, the district court relied on the dissimilar-
ity of the marks alone to determine that no likelihood of con-
fusion existed. Mattel timely appealed the grant of summary
judgment as to its counterclaims.

              II.   STANDARD OF REVIEW

   The review of a grant of summary judgment as to an
infringement claim is de novo. Surfvivor Media, Inc. v. Survi-
vor Prods., 

, 630 (9th Cir. 2005). As such, we
“must determine whether, ‘viewing the evidence in the light
most favorable to the nonmoving party, . . . there are any gen-
uine issues of material fact, and whether the district court cor-
rectly applied the relevant substantive law.’ ” Entrepreneur
Media, Inc. v. Smith, 

, 1140 (9th Cir. 2002)
(quoting Wendt v. Host Int’l, Inc., 

, 809-10 (9th
Cir. 1997)). “Because of the intensely factual nature of trade-
mark disputes, summary judgment is generally disfavored in
the trademark arena.” 

Starship Servs.,
Ltd. v. Epix, Inc., 

, 1109 (9th Cir. 1999)). The
Court may affirm a judgment on any ground fairly supported
by the record. Narell v. Freeman, 

, 910 (9th Cir.
1989) (citing DeNardo v. Murphy, 

, 1347 (9th
Cir. 1986), cert. denied, 

(1986)).

                      III.   ANALYSIS

A.   MATTEL’S TRADEMARK CLAIMS

   All of Mattel’s infringement claims are subject to the same
test. See Century 21 Real Estate Corp. v. Sandlin, 

, 1178 (9th Cir. 1988) (citing Rodeo Collection, Ltd. v.
W. Seventh, 

, 1217 (9th Cir. 1987)); see also
Vuitton Et Fils S.A. v. J. Young Enter., 

, 777 (9th
Cir. 1981) (noting that if a product is marketed in such a way
so as to cause a likelihood of confusion, the defendant might
be guilty of “palming off”). Therefore, the critical determina-
tion is “ ‘whether an alleged trademark infringer’s use of a
1570                 JADA TOYS v. MATTEL
mark creates a likelihood that the consuming public will be
confused as to who makes what product.’ ” Brother Records,
Inc. v. Jardine, 

, 908 (9th Cir. 2003) (quoting
Thane Int’l Inc. v. Trek Bicycle Corp., 

, 901 (9th
Cir. 2002)).

   [1] We employ an eight factor test (“Sleekcraft factors”) in
determining the likelihood of confusion. Brookfield
Commc’ns, Inc. v. W. Coast Entm’t Corp., 

,
1053-54 (9th Cir. 1999). These factors include: (1) strength of
the mark; (2) proximity of the goods; (3) similarity of the
marks; (4) evidence of actual confusion; (5) marketing chan-
nels used; (6) type of goods and the degree of care likely to
be exercised by the purchaser; (7) defendant’s intent in select-
ing the mark; and (8) likelihood of expansion of the product
lines. AMF Inc. v. Sleekcraft Boats, 

, 348-49 (9th
Cir. 1979) (citations omitted).

  [2] We have previously noted that the test for likelihood of
confusion is “pliant,” and that “[s]ome factors are much more
important than others, . . . .” 

.
Jada seizes on such language as an indication that the district
court’s consideration of the dissimilarity of the marks alone
was sufficient to support its conclusion. We disagree.

   [3] Though the language we have used warns against “ex-
cessive rigidity” in applying the test, see 

never
countenanced a likelihood of confusion determination based
on a consideration of dissimilarity alone or, indeed, on the
consideration of any single factor. Instead, we have regularly
applied all the relevant factors, noting that a final likelihood
of confusion determination may rest on those factors that are
of the most relative importance in any particular case. See,
e.g., 

(applying all eight factors
and noting that “the test is a fluid one and the plaintiff need
not satisfy every factor, provided that strong showings are
made with respect to some of them.”). Thus, while a likeli-
hood of confusion determination may ultimately rest on a sub-
                     JADA TOYS v. MATTEL                   1571
set of factors, evidence of relatively important factors must be
considered as part of that set. In the context of two subjec-
tively dissimilar marks, evidence of actual confusion and evi-
dence defining the context in which the goods are sold are
particularly relevant.

   To hold otherwise would allow the possibility that persua-
sive evidence of a particular factor may be considered at the
expense of relevant evidence of others. This problem is partic-
ularly acute where, as here, a court relies on the dissimilarity
of the marks to conclude that no likelihood of confusion
exists. In such a case, the potential for a judge to elevate his
or her own subjective impressions of the relative dissimilarity
of the marks over evidence of, for example, actual confusion,
is great. And where the subjective impressions of a particular
judge are weighed at the expense of other relevant evidence,
the value of the multi-factor approach sanctioned by this
Court is undermined.

   [4] Today’s holding is consistent with our prior decisions.
For while Jada accurately cites language from the case law of
this Circuit suggesting that dissimilarity alone may be a suffi-
cient basis upon which to judge the likelihood of confusion,

; Sleeper Lounge Co. v. Bell Mfg
Co., 

, 723 (9th Cir. 1958), we conclude that the
language employed in those cases constitutes dicta and, there-
fore, we are not bound by it. See Inlandboatmens Union of
Pac. v. Dutra Group, 

, 1081 (9th Cir. 2002)
(holding that a panel is not bound by dicta in prior decisions).

   In Brookfield, for example, we stated that “[w]here the two
marks are entirely dissimilar, there is no likelihood of confu-
sion. ‘Pepsi’ does not infringe Coca-Cola’s ‘Coke.’ Nothing
further need be said.” 

. In Brook-
field, however, we did not conclude that the marks in question
were so dissimilar that no likelihood of confusion could exist.

To the contrary, we found that the marks were
quite similar. 

is no indication that in
1572                 JADA TOYS v. MATTEL
Brookfield we ignored evidence of other important factors in
determining that dissimilarity could be determinative. See 

And, beyond that, there was no evidence of actual
confusion presented in the case. 

Thus, in Brook-
field we were not confronted with a situation where, as here,
strong evidence of other important factors existed to counter
the conclusion that dissimilarity alone could be dispositive.
Consequently, in Brookfield we were not afforded the oppor-
tunity to consider fully the ramifications of our comment.

   The same is true of our decision in Sleeper Lounge. There,
we noted that “[similarity of the marks] is the one essential
feature, without which the others have no probative value.”
Sleeper 

. But as was the case in
Brookfield, in Sleeper Lounge we expressly noted that no evi-
dence of actual confusion had been presented. 

Sleeper Lounge we were no more able to consider fully the
impact of our statement than we were in Brookfield.

   [5] While the Brookfield and Sleeper Lounge decisions
offer important insight into the persuasiveness of dissimilar-
ity, we must, nonetheless, conclude that though it may be true
that very dissimilar marks will rarely present a significant
likelihood of confusion, dissimilarity alone does not obviate
the need to inquire into evidence of other important factors.
Only upon such an inquiry may a court ensure that its judg-
ment as to the likelihood of confusion is fully informed and
without error. Therefore, because the district court considered
only the dissimilarity of the marks in question, we reverse its
grant of summary judgment in favor of Jada.

B.     MATTEL’S DILUTION CLAIM

   [6] Mattel brought dilution claims under both federal and
California state law. The analysis under each is the same.
Panavision Int’l, L.P. v. Toeppen, 

, 1324 (9th
Cir. 1998). In order to prove a violation, a plaintiff must show
that (1) the mark is famous and distinctive; (2) the defendant
                         JADA TOYS v. MATTEL                           1573
is making use of the mark in commerce; (3) the defendant’s
use began after the mark became famous; and (4) the defen-
dant’s use of the mark is likely to cause dilution by blurring
or dilution by tarnishment.2 15 U.S.C. § 1125(c)(1); California
Business and Professions Code section 14330.

   [7] Neither federal law nor California state law requires a
showing of competition or likelihood of confusion to succeed
on a dilution claim. Nissan Motor Co. v. Nissan Computer
Corp., 

, 1011 (9th Cir. 2004) (citing 15 U.S.C.
§ 1127).; Cal. Bus. & Prof. Code § 14330. However, for a
plaintiff to establish that the mark is being used in commerce
(as per step two),“the mark used by the alleged diluter must
be identical, or nearly identical, to the protected mark.” Thane

(noting that this circuit’s description of
dilution by blurring and by tarnishment requires a defendant
to use the plaintiff’s actual mark) (citation omitted). In order
  2
    We note that in this case the district court applied the prior version of
the Federal Trademark Dilution Act (“FTDA”), 15 U.S.C. § 1125(c)
(2000), which required a showing of actual dilution. The actual dilution
requirement was a product of the Supreme Court’s decision in Moseley v.
V Secret Catalogue, Inc., 

, 433 (2003), where the Court held
that the federal dilution statute required a showing of actual dilution. How-
ever, since that time the FTDA has been amended so as to require only a
likelihood of dilution to succeed. Trademark Dilution Revision Act of
2006 (“TDRA”), Pub. L. No. 109-312 § 2(1), 120 Stat. 1730. In this case,
we choose to apply the standard currently in operation so as to adhere to
our prior precedent established in Nissan Motor Co. v. Nissan Computer
Corp., 

, 1009-10 (9th Cir. 2004), in which we held that
application of the FTDA to an alleged diluting mark that was in use before
the statute’s passage was not retroactive because the FTDA authorizes
only prospective relief.
   We are aware that in Horphag Research Ltd. v. Garcia, 

,
1035 (9th Cir. 2007), we applied the FTDA retroactively, thereby creating
an unintentional intra-circuit conflict with Nissan. In Horphag, however,
neither party mentioned the TDRA in its briefs, nor moved for a petition
for panel rehearing or to vacate the mandate in light of the TDRA’s pas-
sage. Moreover, the plaintiff in Horphag prevailed under the more strin-
gent version of the federal dilution statute. Accordingly, recalling the
Horphag mandate at this point would serve no purpose.
1574                     JADA TOYS v. MATTEL
to be nearly identical, two marks “must be ‘similar enough
that a significant segment of the target group of customers
sees the two marks as essentially the same.’ ” 

(quot-
ing Playboy Enters. Inc. v. Welles, 

, 806 n.41
(9th Cir. 2002)).

   Here, the district court granted summary judgment in favor
of Jada on Mattel’s dilution claims because the marks at issue
were not identical or nearly identical. Mattel claims that
because it has shown a likelihood of confusion, that finding
was inappropriate.

   Jada claims, on the other hand, that the district court’s
determination that the marks were not identical is supported
by the record and, therefore, summary judgment was appro-
priate. Jada also contends that Mattel has not made a showing
of actual dilution sufficient to support its claim under federal
law.3 And as to Mattel’s state law claim, Jada argues that Mat-
tel is barred from such a claim because Jada owns a valid fed-
eral registration mark, which acts as a complete bar to a
dilution claim under state law. 15 U.S.C. § 1125(c)(6).4

   [8] Turning first to the district court’s determination that
the marks in question were not sufficiently identical, we con-
clude that a reasonable trier of fact could find that the HOT
WHEELS and HOT RIGZ marks are nearly identical. They
both contain the word “hot,” they both are accompanied by a
flame, and they use similar colors; they also convey a similar
meaning and connotation vis a vis their use of the modifier
   3
     This argument is largely moot in light of the TDRA’s requirement that
a showing of only a likelihood of dilution is necessary to succeed on a
dilution claim. See 15 U.S.C. § 1125(c)(1).
   4
     Jada’s claim that its HOT RIGZ registration acts as a bar to Mattel’s
state law claim is misplaced because Mattel’s eighth counterclaim is for
cancellation of the HOT RIGZ registration under 15 U.S.C. § 1064. As
such, if Mattel can succeed in obtaining that cancellation, the bar to state
unfair competition actions found in 15 U.S.C. § 1125(c)(6) will be inappli-
cable.
                     JADA TOYS v. MATTEL                    1575
“hot.” Cf. Thane 

(holding that a reason-
able trier of fact could conclude that the marks “OrbiTrek”
and “TREK” were nearly identical).

   In determining the degree of fame a mark retains, 15 U.S.C.
§ 1125(c)(2)(A) provides a non-exclusive list of four factors
that a court may consider. These factors include:

    (i) The duration, extent, and geographic reach of
    advertising and publicity of the mark, whether adver-
    tised or publicized by the owner or third parties.

    (ii) The amount, volume, and geographic extent of
    sales of goods or services offered under the mark.

    (iii)   The extent of actual recognition of the mark.

    (iv) Whether the mark was registered under the
    Act of March 3, 1881, or the Act of February 20,
    1905, or on the principal register. 

a reasonable trier of fact could conclude that the
HOT WHEELS mark is famous: it has been in use for over
thirty-seven years; 350 million dollars have been expended in
advertising the mark; three billion HOT WHEELS units have
been sold since the inception of the mark; and HOT WHEELS
are sold in all fifty states and throughout the world.

   [10] A reasonable trier of fact could also conclude that Jada
began using its mark after HOT WHEELS became famous.
The HOT WHEELS mark has been in use since 1968, but it
was only in 2001 that Jada began to employ the use of HOT
RIGZ. Thus, the HOT WHEELS mark had been acquiring
recognition for over thirty years before HOT RIGZ entered
the market.

   [11] The federal dilution statute also requires a showing of
likelihood of dilution either by blurring or by tarnishment. 15
1576                 JADA TOYS v. MATTEL
U.S.C. § 1125(c)(1). Where, as here, a plaintiff’s claim is
based on a dilution by blurring theory, the question is whether
the “association arising from the similarity between a mark or
trade name and a famous mark . . . impairs the distinctiveness
of the famous mark.” 

A court may con-
sider all relevant factors in making this determination, includ-
ing the six identified by the statute:

    (i) The degree of similarity between the mark or
    trade name and the famous mark.

    (ii) The degree of inherent or acquired distinctive-
    ness of the famous mark.

    (iii) The extent to which the owner of the famous
    mark is engaging in substantially exclusive use of
    the mark.

    (iv)   The degree of recognition of the famous mark.

    (v) Whether the user of the mark or trade name
    intended to create an association with the famous
    mark.

    (vi) Any actual association between the mark or
    trade name and the famous mark.

   [12] As has been stated, in this case a reasonable trier of
fact could conclude that the marks at issue are quite similar.
Moreover, there is significant evidence of actual association
between the alleged diluting mark and the famous mark. As
evidence of such, Mattel has submitted two surveys. In the
first, respondents were exposed to the HOT RIGZ name and
asked who they believed “puts out or makes” a toy vehicle
with that name. Twenty-eight percent of respondents thought
that the toy vehicle put out under that name was either made
by Mattel or by the same company that produced HOT
                         JADA TOYS v. MATTEL                           1577
WHEELS, or that whatever company did produce it required
permission from Mattel to sell the product.

   In a separate survey, respondents were shown a HOT RIGZ
package and asked who they thought put out that product. Of
the respondents, 7% believed it was either made by Mattel or
by the same company that produced HOT WHEELS, or that
whatever company did produce it required permission from
Mattel to sell the product.

   [13] Not only do these surveys indicate that consumers
associate one mark with the other; they suggest, too, that Mat-
tel’s HOT WHEELS mark does not adequately identify its
product because Jada is able to convey, through the use of its
HOT RIGZ mark, the impression that Mattel either produces
or allows the production of HOT RIGZ. See 

(holding that such evidence is sufficient to
meet the heightened standard of actual dilution). Thus, a rea-
sonable trier of fact could conclude that this evidence was suf-
ficient to establish the existence of a likelihood of dilution.5
The district court’s entry of summary judgment is, therefore,
reversed.

C.    MATTEL’S COPYRIGHT CLAIMS

   In order to succeed in a copyright infringement claim, “a
plaintiff must show that he or she owns the copyright and that
defendant copied protected elements of the work.” Cavalier v.
Random House, Inc., 

, 822 (9th Cir. 2002) (cit-
ing Shaw v. Lindheim, 

, 1356 (9th Cir. 1990)).
The protected elements of a work are those elements which
are “original,” which is to say they are the product of “inde-
pendent creation, not novelty.” Roth Greeting Cards v. United
Card Co., 

, 1109 (9th Cir. 1970) (citations omit-
  5
   The evidence of likelihood of dilution here is also enough to support
such a finding with respect to the California dilution statute. See Cal. Bus.
& Prof. Code § 14330 (requiring a showing of only likelihood of dilution).
1578                     JADA TOYS v. MATTEL
ted).6 Copying can be proved by evidence indicating that the
infringer had access to the copyrighted work and that the pro-
tected portions of the works are substantially similar. 

   This Court has traditionally employed a two-part test to
determine whether two works are substantially similar: an
extrinsic test and an intrinsic test. Sid & Marty Krofft Televi-
sion Prod., Inc. v. McDonald’s Corp., 

, 1164
(9th Cir. 1977). The specifics of the extrinsic test have
evolved during the years following Marty Krofft. As we
explained in Apple Computer, Inc. v. Microsoft Corp., 

, 1442 (9th Cir. 1994):

      As originally adopted in [Krofft], the extrinsic prong
      was a test for similarity of ideas based on external
      criteria; analytic dissection and expert testimony
      could be used, if helpful. The intrinsic prong was a
      test for similarity of expression from the standpoint
      of the ordinary reasonable observer, with no expert
      assistance. 

has evolved, however, the extrin-
      sic test now objectively considers whether there are
      substantial similarities in both ideas and expression,
      whereas the intrinsic test continues to measure
      expression subjectively. (citations omitted).

   In this case, the district court concluded that because the
logos at issue were not similar, the extrinsic test was not met.
It also found that since the logos were not similar, no reason-
able person could believe that they conveyed a similar expres-
sion and, therefore, the intrinsic test had not been satisfied.
  6
    Jada does not dispute that Mattel retains copyrights for its HOT
WHEELS flame logos. Jada also only tangentially disputes that some ele-
ments of the flame logos are not protectable. However, the combination
of the words used, the stylization of the flame graphic, and the colors cho-
sen, suggest that elements of the flame logo are protectable for the pur-
poses of a copyright infringement action. Cf. Roth Greeting 

.
                      JADA TOYS v. MATTEL                     1579
   [14] We hold, however, that the district court did not con-
duct an objective test as to both ideas and expression as
required under the extrinsic test outlined in Apple Computer.
In addition, as to the intrinsic test, Mattel has raised a material
issue of fact. A trier of fact could conclude that the logos are
similar “from the standpoint of the ordinary reasonable
observer.” 

district court’s entry of summary
judgment in favor of Jada as to Mattel’s copyright claim is
reversed.

                     IV.   CONCLUSION

   The district court’s reliance on a single Sleekcraft factor —
dissimilarity of the marks — in its likelihood of confusion
determination was in error; therefore, its entry of summary
judgment in Jada’s favor as to Mattel’s trademark claims is
reversed. Those claims are hereby remanded for consideration
consistent with this opinion. The district court must analyze
Mattel’s copyright claim under the Apple Computer frame-
work. In addition, Mattel raised a genuine issue of material
fact as to its dilution and copyright claims; thus, the district
court’s entry of summary judgment in favor of Jada as to
those claims is also reversed.

  REVERSED AND REMANDED FOR FURTHER PRO-
CEEDINGS CONSISTENT WITH THIS OPINION.