DYK, Circuit Judge.

Plaintiff-Appellant International Seaway Trading Corporation (“Seaway”) filed suit against Walgreens Corporation (“Wal-greens”) and Touchsport Footwear USA, Inc. (“Touchsport”) claiming infringement of Seaway’s patents, U.S. Design Patents Nos. D529,263 (“the '263 patent”), D545,-032 (“the '032 patent”), and D545,033 (“the '033 patent”). The district court granted summary judgment for defendants, finding that the claims of the asserted patents were invalid under 35 U.S.C. § 102 as anticipated by a patent assigned to Crocs, Inc. (“Crocs”), U.S. Design Patent No. D517,789 (“the Crocs '789 patent”). Int’l Seaway Trading Corp. v. Walgreens Corp., 599 F.Supp.2d 1307, 1319 (S.D.Fla.2009).

On appeal, Seaway contends that the district court erred by basing its invalidity determination solely on the ordinary observer test and by failing to apply the point of novelty test. We agree with the district court that the point of novelty test should not be utilized for anticipation, and that only the ordinary observer test applies. However, while we conclude that the exterior appearance of the patented designs would be substantially similar to the prior art in the eyes of an ordinary observer, we conclude that the district court erred in failing to compare the insoles of the patents-in-suit with the prior art from the perspective of the ordinary observer. Accordingly, we affirm-in-part, vacate-in-part, and remand.

BACKGROUND

Plaintiff-Appellant Seaway is an Ohio corporation that acts as a buyer’s agent and/or importer of footwear to mass merchandise retailers, as well as to footwear, apparel, and sporting goods stores. Seaway also creates its own shoe and boot designs and pursues design patents for them. Defendantr-Appellee Walgreens is an Illinois corporation with retail drug stores across the country that sell footwear, among other products. Defendant-Appellee Touchsport is a California corporation that, like Seaway, serves as a buyer’s agent and/or importer of footwear to retailers, including Walgreens.

*1236Seaway’s '263, '032, and '033 patents (collectively “the patents-in-suit”) claim designs for casual, lightweight footwear, which are typically referred to as “clogs.” The '263 patent application was filed on February 18, 2005, and issued on October 3, 2006. The '032 and '033 patents were filed as continuations-in-part of the '263 patent in February 2006, and both issued on June 26, 2007. It is undisputed that the '032 and '033 patents are “substantially the same as the '263 patent design” except that the heel strap is in a forward position overlying a portion of the clog upper in the '032 patent and that the heel strap is not part of the claimed design for the '033 patent.

A single Patent Office Examiner examined and allowed each of the three patents-in-suit. During prosecution of the '263 patent application, the examiner considered and found the '263 patent design patentable over: (a) four pages from the website of Crocs that depicted various models of Crocs clogs, including the Beach model clog; (b) five pages of photographs of the Crocs Beach model clog; and (c) a December 2002 archival version of the Crocs website, depicting various views of a Crocs clog. In the examination of the '263 patent application, the examiner did not have the benefit of the Crocs '789 patent, which depicted the Crocs Beach model clog, even though the '789 patent issued before the examination was concluded. For the '032 and '033 patent applications, the examiner considered and found the '032 and '033 designs patentable over the references considered during the '263 prosecution as well as the '789 patent.

On February 15, 2008, Seaway filed a complaint against Walgreens and Touchs-port in the United States District Court for the Southern District of Florida, alleging that they were infringing its '263, '032 and '033 patents. Seaway asserted that Touchsport had imported and continued to import shoes that infringed the Seaway patents and that Walgreens had sold and continued to sell the allegedly infringing shoes. Walgreens and Touchsport filed a motion for summary judgment on June 24, 2008, contending that Seaway’s patents were invalid as anticipated under 35 U.S.C. § 102(a) & (e) by the Crocs Beach and Cayman model clogs and/or the Crocs '789 patent, or as obvious under 35 U.S.C. § 103 in view of the Crocs Beach and Cayman model clogs and/or the Crocs '789 patent. Figures depicting the design in the '789 patent, and the design in the '263 patent, which is representative of the designs in the '032 and '033 patents, are set forth below.

Figure 1 in the '789 Patent

*1237 Figure 2 in the '263 Patent

On January 22, 2009, the district court granted summary judgment of anticipation, finding that the three Seaway patents were anticipated by the Crocs '789 patent. Int’l Seaway Trading Corp., 599 F.Supp.2d at 1317. The district court held that the ordinary observer test was the sole test of design patent invalidity under 35 U.S.C. § 102. Id. In granting summary judgment, the district court considered and compared the exterior portions of the designs of the patents-in-suit, but not the insoles of the designs, to the prior art. Id. at 1318-19. The district court concluded that comparison of the insoles was not required, holding “that the law requires a court to consider only those portions of the product that are visible during normal use, regardless of whether those portions are visible during the point of sale.” Id. at 1315. The court then explained its rationale:

When a shoe is in use, it’s [sic] insole is, obviously, hidden by the user’s foot. The sole of the shoe, however is sometimes visible while a person is sitting or walking. As such, this Court will not consider any aspects of the insoles of the shoes, but will consider the sole of the shoes as those are visible during use.

Id. The district court did not determine whether the patents-in-suit were invalid as obvious.

Seaway timely appealed, and we have jurisdiction under 28 U.S.C. § 1295(a)(1). Because this appeal is from a grant of summary judgment, we view the record in the light most favorable to the non-moving party (here Seaway). OddzOn Prods., Inc. v. Just Toys, Inc., 122 F.3d 1396, 1401 (Fed.Cir.1997). Anticipation is a question of fact. Akzo N.V. v. U.S. Int’l Trade Comm’n, 808 F.2d 1471, 1479 (Fed.Cir.1986).

DISCUSSION

I

Seaway first contends that the district court should have applied the point of novelty test in addition to the ordinary observer test during its anticipation analysis. Our decision in Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed.Cir.2008) (en banc), changed the test for infringement. In doing so, we held “that the ‘point of novelty’ test should no longer be used in the analysis of a claim of design patent infringement” and that “the ‘ordinary observer’ test should be the sole test for determining whether a design patent has been infringed.” Egyptian Goddess, 543 F.3d at 678. The issue remains whether Egyptian Goddess also requires a similar change in the test for invalidity. We reserved this question in Egyptian Goddess and more recently in Titan Tire Corp. v. Case New Holland, Inc., 566 F.3d 1372, 1384 (Fed.Cir.2009). After careful consideration, we have concluded that the district court was correct in concluding that Egyptian Goddess necessarily re*1238quires a change in the standard for anticipation.

Section 171 of Title 35 provides the criteria for obtaining a design patent. It provides that: “Whoever invents any new, original and ornamental design for an article of manufacture may obtain a patent therefor, subject to the conditions and requirements of this title.” 35 U.S.C. § 171. There are two differences in wording between the requirements for a design patent under § 171 and for a utility patent under 35 U.S.C. § 101. Section 171 excludes the word “useful” (to distinguish design patents from utility patents) and adds the word “original.” The originality requirement in § 171 dates back to 1842 when Congress enacted the first design patent law.1 The purpose of incorporating an originality requirement is unclear; it likely was designed to incorporate the copyright concept of originality — requiring that the work be original with the author, although this concept did not find its way into the language of the Copyright Act until 1909. See 1-2 Melville B. Nimmer & David Nimmer, Nimmer On Copyright § 2.01 (2005) (subject matter of copyright). In any event, the courts have not construed the word “original” as requiring that design patents be treated differently than utility patents. Section 171 requires that the “conditions and requirements of this title” be applied to design patents, thus requiring application of the provisions of sections 102 (anticipation) and 103 (invalidity). Our cases have recognized that in the past we have applied a dual test for anticipation identical to the then-applicable test for infringement, namely the ordinary observer and point of novelty tests. See Bernhardt, L.L.C. v. Collezione Europa USA, Inc., 386 F.3d 1371, 1383 (Fed.Cir.2004).2 But the application of those tests in the context of infringement and anticipation was necessarily different.

In the case of infringement, in applying the ordinary observer test, we compared the patented design with the accused design. Contessa, 282 F.3d at 1377. In the case of anticipation, we compared the patented design with the alleged anticipatory reference. Door-Master Corp. v. Yorktowne, Inc., 256 F.3d 1308, 1313 (Fed.Cir.2001).

In applying the point of novelty test in the case of infringement, we looked at whether the accused design appropriated the points of novelty of the patented design. Litton Sys., Inc. v. Whirlpool Corp., 728 F.2d 1423, 1444 (Fed.Cir.1984) (“For a design patent to be infringed ... no matter how similar two items look, ‘the accused device must appropriate the novelty in the patented device which distinguishes it from the prior art.’ ” (internal citations omitted)). The points of novelty for the patented design were determined by comparing the patented design to the prior art designs. Bernhardt, 386 F.3d at 1382. In the case of anticipation, we compared the patented design with the alleged anticipatory reference to see if it appropriated the points of novelty of the prior art reference. The points of novelty of the prior art reference were determined by looking to earlier prior art to determine the points of novelty in the anticipatory reference. The ordinary observer and point of novelty tests were applied in much the same manner for *1239obviousness as for anticipation,3 except that in the ease of obviousness the features of the prior art could be combined to create a single anticipatory reference or an earlier single reference could be modified based on the knowledge of a skilled artisan. See, e.g., Durling v. Spectrum Furniture Co., 101 F.3d 100, 103 (Fed.Cir.1996).

While our cases have utilized the point of novelty test for infringement and anticipation, as we pointed out in Egyptian Goddess, 543 F.3d at 672, this test was not mandated by Smith v. Whitman Saddle Co., 148 U.S. 674, 13 S.Ct. 768, 37 L.Ed. 606 (1893), or precedent from other courts. In Whitman, the Supreme Court utilized only the ordinary observer test for determining infringement and invalidity, as did at least one later circuit case following Whitman (Bevin Bros. Mfg. Co. v. Starr Bros. Bell Co., 114 F. 362, 362 (C.C.D.Conn.1902)).

Moreover, it has been well established for over a century that the same test must be used for both infringement and anticipation.4 This general rule derives from the Supreme Court’s proclamation 120 years ago in the context of utility patents: “[t]hat which infringes, if later, would anticipate, if earlier.” Peters v. Active Mfg. Co., 129 U.S. 530, 537, 9 S.Ct. 389, 32 L.Ed. 738 (1889). The same rule applies for design patents. See Bernhardt, 386 F.3d at 1378 (explaining that the test for determining anticipation of a design patent is the same as the test for infringement); Door-Master, 256 F.3d at 1312 (stating that the test for infringement is the same as the test for anticipation in the design patent context); Litton, 728 F.2d at 1440.

In Egyptian Goddess, we abandoned the point of novelty test for design patent infringement and held that the ordinary observer test should serve as the sole test for design patent infringement. 543 F.3d at 678. The ordinary observer test originated in 1871 when the Supreme Court held

that if, in the eye of an ordinary observer, giving such attention as a purchaser usually gives, two designs are substantially the same, if the resemblance is such as to deceive such an observer, inducing him to purchase one supposing it to be the other, the first one patented is infringed by the other.

Gorham Mfg. Co. v. White, 14 Wall. 511, 81 U.S. 511, 528, 20 L.Ed. 731 (1871). In Egyptian Goddess, we also refined the ordinary observer test by characterizing the ordinary observer as being “deemed to view the differences between the patented design and the accused product in the context of the prior art.” 543 F.3d at 676. We explained:

When the differences between the claimed and accused design are viewed in light of the prior art, the attention of the hypothetical ordinary observer will be drawn to those aspects of the claimed *1240design that differ from the prior art. And when the claimed design is close to the prior art designs, small differences between the accused design and the claimed design are likely to be important to the eye of the hypothetical ordinary observer.

Id. We further determined that the point of novelty test, as a second and free-standing requirement for proof of design patent infringement, was inconsistent with the ordinary observer test laid down in Gorham and was not mandated by Supreme Court cases or other precedent. Id. at 672.

In light of Supreme Court precedent and our precedent holding that the same tests must be applied to infringement and anticipation, and our holding in Egyptian Goddess that the ordinary observer test is the sole test for infringement, we now conclude that the ordinary observer test must logically be the sole test for anticipation as well. In doing so, we will prevent an inconsistency from developing between the infringement and anticipation analyses, and we will continue our well-established practice of maintaining identical tests for infringement and anticipation.

We note as well that the problems inherent in the point of novelty test in the infringement context also exist in the anticipation context. The test is just as difficult to apply in the context of anticipation as in the context of infringement, encouraging the focus on minor differences between the allegedly anticipatory reference and the patented design. So too, applying the point of novelty test in the context of anticipation, as in the context of infringement, creates the need to canvass the entire prior art to identify the points of novelty. In addition, eliminating the point of novelty test for anticipation “has the advantage of avoiding the debate over the extent to which a combination of old design features can serve as a point of novelty under the point of novelty test.” Id. at 677. Just as the problems deriving from the point of novelty test exist in both the infringement and anticipation contexts, the benefits of applying the refined ordinary observer test are identical in both.

Seaway’s arguments to preserve the point of novelty test for invalidity are unconvincing. Seaway argues that adopting the ordinary observer test for anticipation will blur the distinction between the tests for obviousness under § 103 and for anticipation under § 102, resulting in jury confusion. According to Seaway, the test for invalidity due to obviousness is whether a designer of ordinary skill in the art would have found the patented design, as a whole, obvious in light of the prior art. Litton, 728 F.2d at 1443. The test for invalidity due to anticipation, on the other hand, requires the jury to consider the perspective of the ordinary consumer. There is in fact no potential for confusion. For design patents, the role of one skilled in the art in the obviousness context lies only in determining whether to combine earlier references to arrive at a single piece of art for comparison with the potential design or to modify a single prior art reference.5 Once that piece of prior art has been constructed, obviousness, like anticipation, requires application of the ordinary observer test, not the view of one skilled in the art. And, as noted by Seaway, “[b]oth the ordinary observer test, whether applied for infringement or invalidity, and the obviousness test, applied for invalidity under Section 103, focus on the *1241 overall designs.” Appellant’s Br. 28 (citing OddzOn, 122 F.3d at 1405, and Litton, 728 F.2d at 1443). Under these circumstances, we see no potential for jury confusion.

In summary, the district court did not err in concluding that the ordinary observer test is the sole test for design patent invalidity under § 102.

II

A

Seaway’s second contention is that, even if the ordinary observer test is found to be the sole and proper test for anticipation under § 102, the district court misapplied the ordinary observer test by failing to compare the entirety of the patented designs, including the clogs’ insoles, with the Crocs '789 patent. The district court, relying on Contessa, held “that the law requires a court to consider only those portions of the product that are visible during normal use, regardless of whether those portions are visible during the point of sale.” Int’l Seaway Trading Corp., 599 F.Supp.2d at 1315. The court did “not consider any aspects of the insoles of the shoes” because the insoles are “hidden by the user’s foot.” Id. We conclude that the district court erred, and we vacate and remand for a determination of whether the differences between the insole patterns in the patents-in-suit and in the prior Crocs art bar a finding of anticipation or obviousness.

In Contessa, we considered the issue of infringement with regard to a shrimp serving tray. The district court held that “any reasonable fact finder would conclude that the competing designs are substantially similar despite the minor differences in tray structure.” Contessa, 282 F.3d at 1377. The district court did not consider the undersides of the trays because they were not visible at the time of sale. Id. at 1377-78. On appeal, we stated: “Our precedent makes clear that all of the ornamental features illustrated in the figures must be considered in evaluating design patent infringement.” Id. at 1378. We found that the district court in Contessa erred by limiting its infringement inquiry to those features visible at the time of sale, rather than to those features visible at any time in the “normal use” lifetime of the accused product. Id. at 1379. We explained that “normal use” in the design patent context extends from the completion of manufacture or assembly until the ultimate destruction, loss, or disappearance of the article. Id. at 1379-80 (citing In re Webb, 916 F.2d 1553, 1557-58 (Fed.Cir.1990)). The same test necessarily applies to anticipation.

The district court here misconstrued Contessa as requiring that the normal use of a clog be limited to the time when it is worn. Contessa did not exclude the point of sale from the normal use of a product. Rather, it emphasized that normal use should not be limited to only one phase or portion of the normal use lifetime of an accused product. Id. at 1380 (citing KeyStone Retaining Wall Sys., Inc. v. Westrock, Inc., 997 F.2d 1444, 1450 (Fed.Cir.1993)). The sale of a clog occurs after it has been manufactured and before it is ultimately destroyed. Thus, the point of sale for a clog clearly occurs during its normal use lifetime. At the point of sale, the insole is visible to potential purchasers when the clog is displayed on a shelf or rack and when the clog is picked up for examination. Similarly, removing a clog from a wearer’s foot also occurs after manufacture and before destruction of the clog, so it also falls squarely within the clog’s normal use lifetime. The wearer may remove the clog temporarily to stretch out his or her toes, leave the clogs on the *1242beach to go for a swim, or engage in countless other activities that would leave the insole exposed.

Walgreens and Touchsport acknowledge that the district court misinterpreted Contessa but argue that it was harmless error. They assert that it was a harmless error because insoles have an insignificant effect on the overall visual appearance of the clogs. They claim there is a “universal truth that consumers buy shoes primarily for their exterior appearance. The insole therefore contributes little to the overall appearance of the shoe to an ordinary observer with knowledge of the prior art.” Appellees’ Br. 51. We reject this argument. The burden is on an accused infringer to show by clear and convincing evidence facts supporting the conclusion that the patent is invalid. Mentor H/S, Inc. v. Med. Device Alliance, Inc., 244 F.3d 1365, 1375 (Fed.Cir.2001). Hence, Walgreens and Touchsport had the burden to establish by clear and convincing evidence that consumers do not consider the insoles of shoes to be significant. The appellees failed to present any evidence in support of this argument.

Walgreens and Touchsport also argue that the district court’s error was harmless because the asserted differences between the insoles of the patents-in-suit and the prior art “were at most slight variations of design elements already present in the Crocs prior art.” Appellees’ Br. 51. We disagree. The insole pattern for the patents-in-suit is distinctly different than the Crocs insole pattern.

Figure 6 in the '789 Patent

Figure 6 in the '268 Patent

The Crocs '789 patent, as shown above m the figure on the left, contains a long, U-shaped dimpling pattern on the insole. In *1243contrast, the patents-in-suit, as demonstrated above in the figure on the right, have a dimpling pattern that includes multiple short rows of dimples. Because we cannot say that these differences are insignificant as a matter of law, a genuine issue of material fact exists as to whether the designs would be viewed as substantially similar in the eyes of the ordinary observer armed with the knowledge of the prior art.

Beyond the insole features of its patented designs, Seaway argues that a genuine issue of material fact exists as to whether the exterior features of its designs preclude a finding of anticipation. It claims that four exterior features differ from the prior art to the degree necessary to preclude summary judgment: (1) the number and arrangement of the circular openings on the upper of the clog; (2) the number and position of the rectangular cut-outs in the lower portion of the upper of the clog; (3) the shape of the toe portion of the clog; and (4) the raised pattern of the outsole of the clog. These features are identical in all three of Seaway’s patents-in-suit. With regard to these alleged dissimilarities, the district court stated:

Slight variations on the number and position of the circular holes on the top of the shoe, the rectangular holes on the toe of the shoe as well as the design of different shaped rectangles on the sole of the shoe would not convince a reasonable jury, or an ordinary observer with knowledge of the prior art, that the limitations were not inherently disclosed in the '789 patent. This conclusion does not change merely because plaintiff slightly changed the arrangement of the textured portions on the top and around the bottom portion of the sides of the shoe.

Int’l Seaway Trading Corp., 599 F.Supp.2d at 1318. We agree with the district court that these minor variations in the shoe are insufficient to preclude a finding of anticipation because they do not change the overall visual impression of the shoe. Although the ordinary observer test requires consideration of the design as a whole, Egyptian Goddess, 543 F.3d at 675; Contessa, 282 F.3d at 1378; Braun, Inc. v. Dynamics Corp. of Am., 975 F.2d 815, 820 (Fed.Cir.1992), this does not prevent the district court on summary judgment from determining that individual features of the design are insignificant from the point of view of the ordinary observer and should not be considered as part of the overall comparison. The mandated overall comparison is a comparison taking into account significant differences between the two designs, not minor or trivial differences that necessarily exist between any two designs that are not exact copies of one another. Just as “minor differences between a patented design and an accused article’s design cannot, and shall not, prevent a finding of infringement,” Litton, 728 F.2d at 1444, so too minor differences cannot prevent a finding of anticipation.

B

As an alternative basis for affirming the judgment, Walgreens and Touehs-port argue that the patents-in-suit are invalid as obvious under § 103. The district court did not determine whether obviousness was a ground for finding invalidity. Int’l Seaway Trading Corp., 599 F.Supp.2d at 1313 (“Since this Court holds that the patent is anticipated under § 102 it will not determine whether the patent was also obvious under § 103.”). Nonetheless, Walgreens and Touchsport assert that the issue can be considered on appeal because they raised both anticipation and obviousness arguments below.

Obviousness, like anticipation, requires courts to consider the perspective of *1244the ordinary observer. Therefore, for the same reasons that the district court’s failure to compare the insoles of the patented designs to the prior art designs precludes a finding of anticipation, it also precludes a finding of obviousness.

We thus conclude that, while the district court correctly determined that there was no genuine issue of material fact that the exterior features of the patented designs were substantially similar to the Crocs '789 patent, the court erred by failing to consider the insoles of the patents-in-suit in its invalidity analysis.

CONCLUSION

In summary, we conclude that the district court correctly held that the ordinary observer test is the sole test of invalidity. The district court erred, however, in failing to compare the insole patterns in Seaway’s patented designs to the prior art as part of an overall comparison of the designs. We vacate and remand for further proceedings on the limited issue of whether the differences in the insole patterns between the prior (Crocs) art and the patented designs bar a finding of anticipation or obviousness.

AFFIRMED-IN-PART, VACATED-IN-PART, and REMANDED

COSTS

No costs.

CLEVENGER, Circuit Judge,

dissenting in part.

I part company with the court on only one issue, which is the scope of the district court’s further action on remand.

The majority concludes that the district court was correct as a matter of law in holding that the ordinary observer must find anticipation when comparing the four different exterior design elements of the patented designs with the Crocs design patent. However, with regard to the comparable insole designs, the majority ascertains a sufficient difference to preclude anticipation as a matter of law. The majority thus concludes that the law permits dissection of a design as a whole into its component pieces. With laser-like focus, the fact finder is permitted to decide that the changes on the top of the clogs are trivial enough to sustain anticipation. The same holds for changed design elements on the lower portions of the clogs, the shape of the toe portions and the raised pattern of the outsoles. Because the majority is not satisfied that the same can be said of the changed insole design, which the district court erroneously refused to assess, the insole design is carved out of the overall design and independently remanded for further proceedings.

As recognized by the majority, the ordinary observer test requires assessment of the designs as a whole. See Egyptian Goddess v. Swisa, Inc., 543 F.3d 665, 675 (Fed.Cir.2008) (en banc); Contessa Food Prods., Inc. v. Conagra, Inc., 282 F.3d 1370, 1378 (Fed.Cir.2002); OddzOn Prods., Inc. v. Just Toys, Inc., 122 F.3d 1396, 1405 (Fed.Cir.1997). I agree that the differences in the inner sole designs are to be assessed as part of the anticipation inquiry. But the differences in the inner sole designs must be appreciated in conjunction with all of the design differences. This is so especially with regard to the differing number and arrangement of the circular openings on the upper of the clogs.

*1245

When the two designs are observed from above, as shown in the above figures, the distinctions in the different number and location of the circular openings on the upper of the clogs are apparent, in addition to the different insole designs. And when the differing insoles are sorted in the mind of the ordinary observer along with the four external differences of the clogs, the ordinary observer surely reaches a different conclusion about the designs as a whole than when the ordinary observer only looks at the differences in the insoles.

The effect of the majority bifurcation of the insole design differences from the exterior design differences, and the piecemeal adjudication of the exterior design differences, is to treat the patents on remand as without any exterior design. The fact finder will only assess anticipation on the basis of design differences on the insoles. I think this violates the rule for anticipation that the designs have to be compared as a whole. The effect of the summation of all the design differences is what counts, not the comparison of differences one by one, isolated from each other. Such an approach invites the problems we sought to eliminate by rejecting the “point of novelty” test. As the court stated in Egyptian Goddess, when there are several different alleged points of novelty, “[t]he attention of the court may therefore be focused on whether the accused design has appropriated a single specified feature of the claimed design, rather than on the proper inquiry, i.e., whether the accused design has appropriated the claimed design as a whole.” 543 F.3d at 667. Remanding for adjudication of anticipation solely on the insole inappropriately focuses the fact finder on a single specified feature of the claimed design.

The majority has forged a new rule for design patent anticipation, if not for infringement as well. The new rule is that the “design as a whole” rule

does not prevent the district court on summary judgment from determining that individual features of the design are insignificant from the point of view of *1246the ordinary observer and should not be considered as part of the overall comparison. The mandated overall comparison is a comparison taking into account significant differences between the two designs, not minor or trivial differences that necessarily exist between any two designs that are not exact copies of one another. Just as “minor differences between a patented design and an accused article’s design cannot, and shall not, prevent a finding of infringement,” Litton, 728 F.2d at 1444, so too minor differences cannot prevent a finding of anticipation.

Maj. Op. at 1243.

The majority’s reliance on Litton Systems, Inc. v. Whirlpool Corp., 728 F.2d 1423 (Fed.Cir.1984), to support its new rule is misplaced. The quotation from the Litton opinion is in the context of that court’s faithful application of the “design as a whole” rule. Minor differences between one design taken as a whole and another design likewise appreciated (the “mandated overall comparison”), of course, cannot fool the ordinary observer. So the court in Litton, after earlier quoting verbatim from Gorham Co. v. White, was correct in stating “that minor differences between a patented design and an accused article’s design cannot, and shall not, prevent a finding of infringement.” 728 F.2d at 1444 (emphasis added). The majority would have us believe that the Litton decision authorizes dissection of designs into component parts for purposes of partial summary judgments of anticipation or infringement. That is not correct and runs counter to precedent.1

The district court should be directed on remand to evaluate the differences in the designs as a whole. Partial judgments of anticipation on segments of a design prohibit assessment of designs as a whole, in violation of long-standing law, starting with Gorham.

ORDER

DALY, Chief Judge.

After review and over objection, the ruling of the Magistrate is hereby ADOPTED and RATIFIED in its entirety.

After review and absent objection, the ruling of the Magistrate is hereby ADOPTED, APPROVED and RATIFIED. The Court finds that the additional affidavits submitted by plaintiffs do not provide a basis for reversal or modification of the ruling granting defendant’s motion for partial summary judgment adopted, approved and ratified by the Court today.

RECOMMENDED RULING GRANTING DEFENDANT’S MOTION FOR PARTIAL SUMMARY JUDGMENT

July 24, 1985

JOAN GLAZER MARGOLIS, United States Magistrate.

On February 13, 1984, plaintiffs Black & Decker, Inc. and Black & Decker (U.S.), Inc. (“Black & Decker” or “plaintiffs”) filed this patent and trademark infringement suit against defendant North American Philips Corporation (“NAPC,” “Norelco” or “defendant”) in the United States District Court for the District of Utah. Upon NAPC’s motion for change of venue, the Utah court ordered the action transfer*189red to this district, in which NAPC has its executive offices and manufacturing plants.

Black & Decker’s Second Amended Complaint, filed February 5, 1985, alleges three causes of action: Count I is for infringement of Design Pat. No. 257,661 (’661 patent), issued on December 28, 1980, and infringement of mechanical Patent No. 4,209,875 (’875 patent), issued July 1, 1980; Count II raises a claim for unfair competition under state law pendent to the claim for patent infringement (Count I) and trademark infringement (Count III), in that plaintiffs allege that defendant’s portable rechargeable vacuum cleaner, NORELCO CLEAN UP MACHINE, substantially embodies the appearance of Black & Decker’s DUSTBUSTER vacuum cleaner, that defendant has usurped the good will and exploited the advertising and promotion of plaintiffs’ DUSTBUSTER vacuum cleaner, and that the defendant has copied plaintiffs’ trade dress and intends to pass off the Norelco vacuum cleaners as those of plaintiffs; Count III claims that defendant’s activities constitute “false description or representation and false designation as to origin and are likely to cause confusion” and deceive buyers as to the origin and quality of defendant’s machines.

In its Answer, filed February 27, 1985, NAPC denies having infringed Black & Decker’s patents and has asserted counterclaims which seek to invalidate the ’661 and ’875 patents, inter alia, for lack of patentable invention over the prior art, and to invalidate the trademark covering the design of the DUSTBUSTER portable vacuum cleaner.

Defendant has moved for partial summary judgment on Count I, in that the design of its NORELCO CLEAN UP MACHINE, sold under the Norelco trademark, does not infringe the plaintiffs’ design patent ’661, and on Counts II and III for trademark infringement and unfair competition. This motion for summary judgment does not attack plaintiffs’ action for infringement of the ’875 mechanical patent.

For the reasons stated herein, defendant’s motion for partial summary judgment is granted.

I. STANDARDS FOR GRANTING SUMMARY JUDGMENT

Rule 56 of the Federal Rules of Civil Procedure requires that summary judgment be entered whenever the evidence presented indicates that no genuine issue of fact exists and that the moving party is entitled to the judgment as a matter of law. Summary judgment is “without question intended to be effectuated in patent litigation as in any other type of suit and in accordance with the same standards.” Union Carbide Corp. v. American Can Co., 724 F.2d 1567, 1571 (Fed.Cir.1984). Although the court must decide the issue in favor of the non-moving party if it has any doubts about the existence of material facts; summary judgment is proper when no “significant probative evidence [exists] tending to support the complaint.” Peterson Mfg. Co. v. Central Purchasing, Inc., 740 F.2d 1541, 1551 (Fed.Cir.1984). See also Scovill, Inc. v. Dynamics Corp. of America, Civil No. B-81-149, slip op. at 2-4 (D.Conn. Mar. 18, 1985).

The Court, mindful of the heavy burden borne by defendant in this motion, nevertheless rules that no evidence has been presented to show that a genuine issue of fact exists and that defendant is entitled as a matter of law to partial summary judgment. The Court finds that the defendant has infringed neither the plaintiffs’ design patent ’661 nor the DUSTBUSTER trademark.

II. ’661 PATENT . INFRINGEMENT CLAIM

A. The Gorham “Eye” Test

The accepted test of infringement of a patent design is the “eye” test, articulated in Gorham Co. v. White, 81 U.S. (14 Wall) 511, 528, 20 L.Ed. 731 (1871):

[I]f in the eye of an ordinary observer, giving such attention as a purchaser usually gives, two designs are substan*190tially the same, if the resemblance is such as to deceive such an observer, inducing him to purchase one supposing it to be the other, the first one patented is infringed by the other.

The Court has applied the test to the two products in issue, plaintiffs’ DUSTBUSTER vacuum cleaner (Defendant’s Exhibit 10), covered by the ’661 patent, and defendant’s accused structures, the NORELCO CLEAN UP MACHINE (Defendant's Exhibit 14). Based upon this physical inspection, the Court concludes that the overall appearance of the plaintiffs’ DUSTBUSTER and defendant’s CLEAN UP MACHINE are substantially different, so that there is no patent infringement.

There are many physical differences between the two products, the most obvious of which is the shape of their respective handles. The Norelco machine has a large closed-loop design, shaped like a parallelogram, whereas the handle of the Black & Decker machine appears as an extension of the dustbowl and extends rearward. Other obvious distinctions between the two machines include the air vents and the on/off buttons. The vents in the Norelco design are plain lateral slats located on the vertical sides of the housing in front of the handle; the air vents of the Black & Decker patented design extend around the sides and top of the housing. The on/off button of the Norelco machine operates like a trigger and is located inside the loop of the handle, whereas the on/off button of the patented design is located on top of the handle member.

In addition to the differences in the handles, the vents, and the on/off buttons of the machines, the contours of the two products are distinctive. The bottom contours are different because of the different handle designs. The top line profile of the Norelco machine is a slight curved diagonal. The top line profile of the DUSTBUSTER profile resembles an open inverted V.

The plaintiff argues that summary judgment is improper in light of Park v. Milton Industries Inc., 223 U.S.P.Q. 719 (N.D.Ill.1983), in which the court denied a similar motion for summary judgment. The facts in Park, however, are easily distinguished from the present case. In Park the plaintiff claimed that the defendant infringed the patent design of a “blow-gun” product. The defendant moved for summary judgment, arguing that three changes had been made in its product which distinguished it from the plaintiff’s product. The court found that because the changes were minor, it could not rule as a matter of law that there was no resemblance between the products so as to eliminate any genuine issue of material fact. In contrast, the design differences between the Black & Decker DUSTBUSTER and the NORELCO CLEAN UP MACHINE are substantial. Both machines consist basically of two separate parts, the wedged shaped dustbowl and the power handle. While the wedged shape bowls of the two machines are similar (See Part II. B. below), the handles are so distinct that an ordinary buyer would not purchase the NAPC product believing it to be the DUSTBUSTER. The Black & Decker handle extends rearward from the housing, while the handle of the CLEAN UP MACHINE forms a parallelogram. In addition, the other differences noted above combined with this distinctive arrangement leave no doubt in the eye of the ordinary observer that the products are not the same.

Plaintiffs also argue that comparison of the individual elements of the respective machines is impermissible. While the Gorham “eye” test anticipates that the court will make a judgment based on an overall impression of the products in issue and not compare individual components of a product the court cannot avoid a consideration of the constituent elements of the machines, if only to use these elements as a point of reference for making a general comparison. Moreover, when the elemental differences are numerous and substantial, their combined effect causes the product in overall appearance to be distinct and different. The Court notes the differences between the constituent parts of the opposing designs not as determinative by them*191selves, but as part of the process of evaluating the overall appearances of the product. The Court would “have been deficient if it had not so specified the underlying reasons for its appraisal.” Tappan Co. v. General Motors Corp. and Halle Bros. Co. Inc., 380 F.2d 888, 891 (6th Cir.1967). See also Unette Corp. v. Unit Pack Co., Inc., Civil No. 84-2166, slip op. at 6-7 (D.N.J. Mar. 8, 1985).

Thus, the Court concludes that the overall appearance of the Black & Decker DUSTBUSTER vacuum cleaner is quite distinct from the overall appearance of the NORELCO CLEAN UP MACHINE, based on comparison of the two machines and their constituent elements. Such a comparison is necessary in applying the “eye” test of Gorham v. White, supra.1

B. The “Point of Novelty” Test

The different appearances of the Black & Decker DUSTBUSTER and the NORELCO CLEAN UP MACHINE, identified above, are by themselves sufficient evidence for the Court to rule that the two products are not so substantially similar as to deceive the ordinary observer. The Court, however, has noted not only the differences between the two machines, but also the similarities. In examining those similarities, the “point of novelty” test has been utilized as such test supplements the Gorham “eye” test, “the two requirements [of similarity and novelty] being conjunctive.” Shellar v. Durham Industries, Inc., 745 F.2d 621, 628 (Fed.Cir.1984). Therefore, in addition to considering the many dissimilar-, ities between the two products in issue, the Court has also focused on the similar features shared by both products. As the court stated in Litton Systems Inc. v. Whirlpool Corp., 728 F.2d 1423, 1444 (Fed.Cir.1984) (“Litton”)

Even though the Court compares two items through the eyes of the ordinary observer, it must, nevertheless, to find infringement, attribute their similarity to the novelty which distinguishes the patented device from the prior art.

The Court finds that the wedged shaped dustbowl of the Black & Decker DUSTBUSTER is not a novel feature, as the wedged shaped bowl in a vacuum cleaner design has been described in the British Licentia Patents. (Defendant’s Exhibit 8).

The diagrams in the Licentia Patents, particularly figures 2 and 3, show distinctly a side view of a machine that has a wedge shape. The description of the machine (Defendant’s Exhibit 8, p. 1, Col. 1, L. 44-46 and.p. 2, L. 70-81) clearly depicts a “wedge form.” The primary distinction between the wedge shape of the Licentia vacuum cleaner and the vacuum cleaners in the present case is that the Licentia machine has “sides rounded out” (Defendant’s Exhibit 8, p. 2, L.70-81) while the Black & Decker DUSTBUSTER and the NORELCO CLEAN UP MACHINE are rectangular. This distinction, however, does not diminish the importance of the fact that the concept and the diagram of the wedged shaped casing are clearly described in the prior art. The wedge shaped feature is therefore not novel.

The Court rejects plaintiffs’ argument that defendant’s motion for summary judgment rests on ah “illegally impermissible” dissection of plaintiff’s design patent. *192(Plaintiffs’ Brief in Opposition to Motion for Summary Judgment, filed April 1, 1985, at 16). Defendant identifies those aspects of the configurations of the two machines that are similar and those that are distinctive and then asserts that the similar feature is not a novel one. Defendant’s line of reasoning is identical to that in the Litton decision, supra, where the court isolated the points of novelty in the plaintiff’s microwave oven, identifying as novel features the oven door with a three stripe door frame, the absence of a handle on the door, and a latch release lever on the door frame. Litton, supra, 728 F.2d at 1446. The court then distinguished the Litton and Whirlpool products by pointing out that the Whirlpool product lacked these novel features. See also Unette Corp. v. Unit Pack Co., Inc., supra. In the instant case, defendant has not infringed the plaintiffs’ design patent ’661 because the wedge shaped bowl, the only point of similarity between the two products in issue, is not a novel feature of that patent.

Finally, the Court, in ruling that the defendant may use the wedge shaped bowl as part of its configuration of the NORELCO CLEAN UP MACHINE, makes no assertion as to the validity or invalidity of the plaintiffs’ design Patent ’661 — the Black & Decker DUSTBUSTER, as an overall configuration, may be a novel design; the patent may be valid, even if it used individual features that are in the public domain, such as the wedge-shaped bowl. Lancaster Colony Corp. v. Alden, 506 F.2d 1197 (2d Cir.1974).

III. TRADEMARK INFRINGEMENT CLAIM

In order to show that NAPC has infringed its trademark, Black & Decker, under Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), must prove: (1) that the trade dress or product configuration of the two competing products are confusingly similar; (2) that the appropriated features of the trade dress or product configuration are primarily non-functional; and (3) that the product or trade dress has acquired secondary meaning. Litton, supra, 728 F.2d at 1445. Cf. Warner Bros. v. American Broadcasting Co., 720 F.2d 231, 246 (2d Cir.1983) (“Warner Bros.”).

Defendant raises three defenses with respect to the first two requirements. First, NAPC argues that the configuration of the NORELCO CLEAN UP MACHINE is substantially different from Black & Decker’s DUSTBUSTER, is prominently labeled, and therefore is not confusingly similar. Second, it asserts a right to use the configuration because the Black & Decker design has been disclosed in two previous mechanical patents, so that designs are functional in nature. Finally, defendant argues that the features of the Black & Decker design are either functional or have not been used by NAPC. Defendant’s first defense is meritorious.

A. Absence of Likelihood of Confusion

Among the factors to be considered when determining the likelihood of confusion are: “the degree of similarity between the marks in appearance and suggestion; the similarity of the products; ... the degree of care likely to be exercised by the consumers; the strength of the complainant’s mark; [and] actual confusion.” B & L Sales Associates v. H. Daroff & Sons, Inc., 421 F.2d 352, 354 (2d Cir.1970), citing Miss Universe, Inc. v. Patricilli, 408 F.2d 506, 508 (2d Cir.1969).

While the likelihood of confusion is “fairly frequently an issue of fact on which reasonable minds may differ, ... the issue is amenable to summary judgment in appropriate cases.” Warner Bros., supra, 720 F.2d at 246. Three factors are appropriate here: (1) the noted differences in the physical appearances of the product; (2) the clear labeling by NAPC of the NORELCO CLEAN UP MACHINE; and (3) the expense of the machines.

As previously noted, the Black & Decker DUSTBUSTER and the NORELCO CLEAN UP MACHINE have a similar appearance in that each machine has a wedge shaped bowl. This fact is significant in order to evaluate the likelihood of confusion that exists by virtue of this similar *193appearance. In its evaluation, the Court has judged the likelihood of confusion based not on its own subjective assessment of the products, but on “whether the ordinary pruduent customer in the marketplace would likely be confused.” 2 McCarthy on Trademarks and Unfair Competition § 23.-20 (1st ed. 1984). The similarity in the wedge shaped bowl by itself is not enough to cause the likelihood of confusion. The two products, when their general features are compared, are so dissimilar that they leave a general overall impression in the minds of prudent buyers that the two products do not come from the same source. Again, as noted in Part II. A., supra, three major differences between the Black & Decker DUSTBUSTER machine and the NORELCO CLEAN UP MACHINE distinguish the two products such that there exists no likelihood of confusion. First, the Norelco machine has a power handle with a closed looped design resembling a parallelogram, whereas the Black & Decker DUSTBUSTER handle is elongated and extends rearward from the dust bowl. Second, the air vents in the Norelco design are plain lateral slits located forward of the handle; in contrast, the Black & Decker vents extend around the sides of the housing. Third, the Norelco handle houses an on/off button like a trigger, whereas the on/off button of the Black & Decker DUSTBUSTER is located conspicuously on top of the handle.

In addition to these general overall differences between the two machines, the NORELCO CLEAN UP MACHINE is so clearly and prominently labeled that no likelihood of confusion exists between it and the Black & Decker DUSTBUSTER. The conspicuous labeling is found both on the box of the NORELCO CLEAN UP MACHINE and on the machine itself. “The most common and effective means of apprising the intended purchasers of the source of goods is a prominent disclosure ... of the manufacturer’s trade name [and when] that is done, there is no basis for the charge of unfair competition.” Litton, supra, 728 F.2d at 1446. Moreover, the defendant’s brand, Norelco, is a well-known consumer brand and trademark and the conspicuous use of this brand on products is further evidence that consumers are not likely to be confused as to the source of those products.

While labeling by itself may not in every case be sufficient to overcome the likelihood of confusion, other factors, such as the prominence of the label, the type of product involved, the cost of the product and the level of sophistication of the buyers may support the strength of the label in avoiding confusion. “As in many areas of the law, there can be no bright line.” Id. at 1443. One easily identifiable factor to consider is cost. High fidelity speakers, microwave ovens, and electric ranges are sufficiently expensive that the clearly labeled product of one manufacturer will not be confusingly similar to a like product of another manufacturer. “The ordinary consumers of such goods would exercise much more discriminating care in the purchase of such items than would the purchaser of an inexpensive routinely purchased product, such as a desk-style checkbook.” Litton, supra, 728 F.2d at 1446. Other examples of such expensive items include carpets, mattresses, furnaces, air conditioning units, sets of silverware, cameras, and snowmobiles. McCarthy, supra, at § 23.-20. The portable vacuum cleaners here cost between $20 and $40, which is substantially less than the prices of the items listed above, but nonetheless is a price range sufficiently high that an average consumer would examine carefully the product before purchase. Such an examination would, at a minimum, cause the ordinary prudent buyer to read the manufacturer’s clearly labeled.name on the product. Moreover, such an examination is likely because the product is not the kind of household item that one purchases frequently, but rather is an item that the purchaser expects will last for a lengthy duration and therefore would require care in its purchase.

B. Plaintiffs’ Survey Information

The standard for determining the likelihood of confusion is whether “an appreciable number of ordinary prudent purchasers [will] be misled or indeed simply confused *194as to the source of the goods in question.” Mushroom Makers, Inc. v. R.G. Barry Corp., 580 F.2d 44, 47 (2d Cir.1978), cert. denied, 439 U.S. 1116, 99 S.Ct. 1022, 59 L.Ed.2d 75 (1979). Plaintiffs’ evidence fails to meet this standard. Plaintiffs’ evidence consists of survey evidence collected by plaintiffs and of affidavits attesting to the secondary meaning of the Black & Decker configuration and trade dress.

Plaintiffs conducted a survey in order to demonstrate that the two products in issue are confusingly similar. The survey was conducted by showing to qualified respondents a newspaper advertisement with a variety of household products, including the DUSTBUSTER vacuum. The ad was then removed from view and respondents were asked to recall any products advertised. Those who recalled the DUSTBUSTER vacuum cleaner were later shown several household products on display, including a NORELCO CLEAN UP MACHINE. At this point the respondents were asked to identify the products in the display which had been depicted in the newspaper advertisement. ' According to plaintiffs, 44% of those respondents incorrectly identified the NORELCO CLEAN UP MACHINE as the Black & Decker DUSTBUSTER that had been shown in the advertisement.

While survey evidence is one of the “evidentiary routes to prove the likelihood of confusion”, McCarthy, supra, at § 23.20, the survey is not necessarily determinative of the issue of the likelihood of confusion. Cf. Warner Bros., supra, 720 F.2d at 245 (“cautioning against ‘the dangerous precedent of allowing trial by the court to be replaced by public opinion poll,’ ” citing Ideal Corp. v. Kenner Products Division of General Mills Fun Group, Inc., 443 F.Supp. 291, 304 (S.D.N.Y.1977)). The survey must be relevant “as to whether an ordinary shopper might be confused into buying [one product] when intending to buy [the other].” Litton, supra, 728 F.2d at 1447. Plaintiffs’ survey lacks relevancy, primarily because it lacks adequate control for the possibility of ambiguity and confusion in the questions to the respondents. The questions that were posed do not adequately distinguish between products and brand names. Since the NORELCO CLEAN UP MACHINE was the only hand held vacuum cleaner among the various household items on display, the only firm conclusion that can be drawn from the study is that the Black & Decker DUSTBUSTER and the NORELCO CLEAN UP MACHINE share several features in common. See Litton, supra, 728 F.2d at 1447.

The affidavits that plaintiffs presented to prove secondary meaning also do not present a material issue of fact relevant to the likelihood of confusion. The affidavits instead summarize the scope and cost of Black & Decker’s marketing efforts in regard to the DUSTBUSTER machine. These affidavits are relevant to the issue of secondary meaning. Plaintiffs attempt to link this evidence with the great potential for the likelihood of confusion, but plaintiffs’ logic is flawed — secondary meaning and the likelihood of confusion, though related, are separate legal issues. Secondary meaning is a prerequisite for confusion, for without secondary meaning confusion in the mind of the consumer does not exist. Beyond providing evidence for the circumstances under which the likelihood of confusion may exist, evidence of marketing strategy does not relate to the likelihood of confusion. To show confusion, other factors must be considered, such as evidence of actual confusion or an “inference arising from a judicial comparison of the conflicting marks themselves.” McCarthy, supra, at § 23.20.

C. Conclusions

Four factors were considered in deciding that the Black & Decker DUSTBUSTER and the NORELCO CLEAN UP MACHINE are not confusingly similar — (1) the Court’s own inferences drawn from an overall examination of the products, including the differently shaped handles, air vents, and location of on/off buttons; (2) the prominent display of the Norelco trademark on the NAPC product; (3) the relationship between the price of the products and the level of sophistication of the probable buyers; and (4) a review of the plaintiffs’ survey evidence and the affidavits *195attesting to secondary meaning. Based on these factors, the Court concludes that no likelihood of confusion of the products exists. This conclusion is determinative of the issue of trademark infringement and “[i]t is, therefore, not necessary for this Court to inquire further into questions of functionality or secondary meaning.” Scovill, Inc. v. Dynamics Corp. of America, supra, slip op. at 17.

IV. PENDENT STATE CLAIMS

The second count of Black & Decker’s complaint alleges claims under state law — that defendant’s vacuum cleaners substantially embody the appearance of the Black & Decker DUSTBUSTER vacuum machine; that defendant has usurped the good will and exploited the advertising and promotion of plaintiffs’ DUSTBUSTER, and that defendant has copied plaintiffs’ design in order to pass off the Norelco Clean Up machine as that of the plaintiffs.

The Supreme Court has held that federal law preempts any state unfair competition claim for copying designs that have no patent. Sears, Roebuck & Co. v. Stiffel Co., 376 U.S. 225, 84 S.Ct. 784, 11 L.Ed.2d 661 (1964). More recently, the Litton. court applied this preemption doctrine to cases in which a product has been found not to infringe the plaintiffs’ design. Litton, supra, 728 F.2d at 1448. Claims of trademark infringement and “passing off” do not, however, protect rights equivalent to the rights protected by patent law and are therefore not preempted. Warner Bros., supra, 720 F.2d at 247.

For the same reasons as stated in Part III, supra, NAPC’s motion for summary judgment is granted with respect to the Second Count.

V. CONCLUSION

Defendant’s motion for partial summary judgment is granted with respect to Count I regarding the design patent ‘661 and with respect to Counts II and III.

See 28 U.S.C. Section 636(b); F.R.Civ.P. 72; Rule 2 of the Local Rules for United States Magistrates, United States District Court of Connecticut.

RECOMMENDED RULING ON PLAINTIFF’S MOTION FOR RECONSIDERATION

Nov. 26, 1985

On July 24, 1985, this Magistrate filed a Recommended Ruling Granting Defendant’s Motion for Partial Summary Judgment (“Recommended Ruling”), in this patent and trademark infringement suit. With respect to the patent infringement claim, the court relied upon the “eye” test articulated in Gorham Co. v. White, 81 U.S. (14 Wall) 511, 528, 20 L.Ed. 731 (1871), under which the court is to look toward “the eye of an ordinary observer.” Recommended Ruling at 4-9. In addition, the court applied the “point of novelty” test, which supplements the Gorham “eye” test. Id. at 9-12. With respect to the trademark infringement claim, the court again looked to the ordinary consumer, to ascertain that there was a likelihood of confusion between the plaintiff’s and defendant’s products. Id. at 12-18. The court considered at some length plaintiff’s survey evidence. Id. at 18-21. As the Second Circuit has stated, the standard for determining the likelihood of confusion is whether “an appreciable number of ordinary prudent purchasers [will] be misled or indeed simply confused as to the source of the goods in question.” Mushroom Makers, Inc. v. R.G. Barry Corp., 580 F.2d 44, 47 (2d Cir.1978), cert. denied, 439 U.S. 1116, 99 S.Ct. 1022, 59 L.Ed.2d 75 (1979) (emphasis added).

Plaintiff filed an objection to the Recommended Ruling on October 28, 1985. On November 18, 1985, it filed an affidavit, dated November 8, 1985, by a nineteen year-old student at a hair design school, who resides in Eldridge, Iowa. The affiant stated that on or about December 3, 1984, she purchased a Christmas present for her mother, which she believed was plaintiff’s DUSTBUSTER; she later learned, after having applied for a rebate, that she had purchased defendant’s NORELCO CLEAN UP MACHINE 2. This affidavit was referred to the Magistrate on November 19, 1985, to be treated as a motion for reconsideration.

*196This single affidavit does not demonstrate that “an appreciable number of ordinary prudent purchasers” will be misled or confused as to the source of the goods in question. The affidavit only indicates that a single purchaser in Iowa was confused.

Accordingly, plaintiffs motion for reconsideration is denied.

See 28 U.S.C. Section 636(b); F.R.Civ.P. 72; Rule 2 of the Local Rules for United States Magistrates, United States District Court for the District of Connecticut.

CUMMINGS, Circuit Judge.

Defendant, Russ Berrie & Co., Inc. (Ber-rie), appeals from a judgment entered on a jury verdict of $4.3 million in favor of plaintiff, Georgia Lee Miller Roulo (Roulo), for infringement of her copyright and trade-dress rights in a greeting card line known as “Feeling Sensitive” (FS) cards. Roulo’s FS cards are single-faced greeting cards with engraved sentimental messages appropriate for holidays and other emotional occasions. She first introduced her FS cards in July 1977 and promoted them herself until October of 1977, when she began negotiations with Berrie for a distribution agreement. The parties entered into a two-year contract under which Berrie would exclusively manufacture, distribute and sell FS cards, remitting 10% of the sales receipts to Roulo as compensation for the use of her cards, while Roulo retained the ownership of her copyright and trade dress.

The FS cards were sold by Berrie pursuant to the agreement from approximately April 1978 to April 1980. Roulo indicated in late 1979 that she did not intend to renew the contract, prompting Berrie to begin development of a comparable greeting card line to be known as “Touching You” (TY) to substitute for the lost FS product, which it introduced in July 1980 at the Chicago Gift Show. Roulo observed the TY line at the Chicago Gift Show where she was also promoting another line of greeting cards. This suit for infringement of her FS line under the Lanham (15 U.S.C. *935§ 1051, et seq.) and Copyright Acts (17 U.S.C. § 101, et seq.) was filed in April 1982.

Roulo’s action was tried to a jury in a bifurcated trial of liability and damages. The jury returned a verdict in her favor under both the Lanham and Copyright Acts, awarding Roulo $4.3 million in damages based on Berrie’s profits from the TY line. Berrie has launched an exhaustive attack on the jury’s verdict and the district judge’s rulings on pre-trial and post-trial motions. Roulo has cross-appealed from the district judge’s denial of attorney’s fees.

Since Berrie’s challenges on appeal inextricably involve the visual appearance of the two greeting card lines, and indeed a display of each was presented to the jury for a side-by-side comparison, a detailed description of each is necessary and black and white reproductions of the cards are appended to this opinion. Roulo’s FS cards are beige, single-face (no fold) cards containing sentimental verses and frequently using ellipses written in Roulo’s handwriting with brown ink. An example of a verse reads: “I want to shout and tell the world how much I love you ... but instead I’ll just ... whisper.” Flanking the message on the left and right borders are a series of four stripes, two silver foil stripes enveloping one brown and one colored stripe in the middle. The FS cards are displayed in a four-sided freestanding rotating rack, each side containing eight cards of the same colored stripe displayed vertically. At the top of the rack is a removable header bearing a sketch of Roulo’s likeness, her name, and the words “ ‘Feeling Sensitive’... A lost and found department for those with feelings ... in search of the words.” The backs of the FS cards indicate that the verses are taken from Roulo’s book “I’ll See You Sometime ... Between Now and Soon.” Each card was priced at eighty cents.

The TY cards are designed for occasions similar to the FS line, employing like, although not identical, sentimental verses. The TY cards are identically sized and priced, single-face cards on cream paper with cursive written messages in brown ink. The cards combine two stripes of col- or on the left side and one colored stripe on the right side, none of which are in foil. A colored foil butterfly is superimposed on the left stripes. The TY cards were also displayed in a four-sided, 32-card rack with each side displaying a single color scheme.

I.

TRADE-DRESS INFRINGEMENT

“Trade dress” refers to the total image of a product, including features such as “size, shape, color or color combinations, texture, graphics, or even particular sales techniques.” John H. Harland Co. v. Clarke Checks, Inc., 711 F.2d 966, 980 (11th Cir.1983). An infringement of the trade dress is proven if: (1) the plaintiff’s trade dress is inherently distinctive or has acquired secondary meaning, (2) the plaintiffs trade dress is primarily non-functional, and (3) the defendant’s trade dress is confusingly similar, engendering a likelihood of confusion in the marketplace. Am-Brit, Inc. v. Kraft, Inc., 805 F.2d 974, 978 (11th Cir.1986), certiorari denied, 481 U.S. 1041, 107 S.Ct. 1983, 95 L.Ed.2d 822. As an affirmative defense, the alleged infringer may demonstrate that the trade dress has been abandoned. Prima facie evidence of abandonment is established by nonuse of the trade dress for two years, shifting the burden of production to the trade-dress owner to explain the nonuse or establish the existence of an intent to resume use. Exxon Corp. v. Humble Exploration Co., Inc., 695 F.2d 96, 99 (5th Cir.1983). Berrie challenges the trade-dress verdict, arguing that Roulo failed to demonstrate that her trade dress was distinctive or had acquired secondary meaning, failed to establish a likelihood of confusion between the two greeting card lines, and failed to rebut the presumption of abandonment of the greeting cards arising from their nonuse for two years.

The purpose of trademark and trade-dress protection is to enable a business to identify itself efficiently as the source of a given product through the *936adoption of a mark which may be in the form of a slogan, symbol, ornamental design or other visual insignia. A mark or trade dress which is fanciful, arbitrary or otherwise distinctive is given protection more readily than a generic or descriptive trademark or trade dress that is functional since appropriation of generic words, marks or dress would prevent producers from accurately describing or denoting the quality or content of their goods. Blau Plumbing, Inc. v. S.O.S. Fix-It, Inc., 781 F.2d 604, 609 (7th Cir.1986). Hence common words necessary to describe the goods in question such as “greeting card” or a trade-dress element such as an envelope for a greeting card are not capable of appropriation, at least not in the absence of extraordinary evidence indicating secondary meaning, i.e., that the common features have come to denote a single producer in the minds of the consuming public. It is therefore easier to secure protection for a fanciful mark or trade dress such as “Twinkie” than a more suggestive or descriptive mark such as “M-TV”. When a trademark or trade dress is inherently distinctive or fanciful, it is unnecessary to make the further showing that the mark or dress has become associated with a single producer. Blau Plumbing, 781 F.2d at 608; AmBrit, 805 F.2d at 979. If the mark is not distinctive such that a showing of secondary meaning is required, it is not necessary that the public be aware of the identity of the producer, but simply that the public associate the mark or dress with a single source. Processed Plastic Co. v. Warner Communications, Inc., 675 F.2d 852, 856 (7th Cir.1982).

Here Berrie asserts that Roulo has failed to demonstrate that the elements of her trade dress are distinctive or have acquired secondary meaning. Specifically, Berrie contends that Roulo’s cards incorporated several common features such as stripes, dots, handwriting and other common design elements which are indigenous to all greeting cards and do not represent a trade-dress element capable of protection. To establish her claim of distinctiveness, Roulo introduced testimony of Sally Hopkins, a Hallmark cards curator, who testified that she was unable to find any cards in their collection from this century containing the same combination of elements as used in the FS line. In the alternative, Roulo introduced as evidence of secondary meaning the fact that her name and likeness appeared on the header to the rack displaying the cards, evidence of the success of her greeting cards as revealed by letters written to Berrie inquiring about Roulo’s book of verse, an article in Greetings Magazine featuring her card line, and testimony by Berrie that her cards were as unusual as those of Boynton and Davis, other famous card designers. Roulo was also given 10% of sales, the highest percentage share of the sales receipts ever agreed to by Berrie for any card designer.

In order to undermine Roulo’s evidence, Berrie simply asserts that Roulo failed to offer any evidence to demonstrate that her cards had acquired secondary meaning. Although Roulo did not offer any empirical evidence that consumers identified her trade dress with the Roulo name, there was certainly adequate evidence to allow the jury to conclude that Roulo’s FS cards were distinctive. The adequacy of this evidence is augmented by Berrie’s admission that the TY line was modeled after Roulo’s FS cards once it was clear that Roulo would not renew her contract with Berrie. The fact that Berrie gave Roulo one of the highest commissions for her cards is also strong evidence that her line was unique. The fact that her cards incorporated common, indistinct elements such as lines and handwriting does not refute the fact that Roulo’s combination of these elements was sufficiently unique to warrant trade-dress protection. Trade dress encompasses the overall appearance of a product, including its size, color or color combinations, texture, graphics, packaging or other visual features. Hartford House, Ltd. v. Hallmark Cards, Inc., 846 F.2d 1268, 1271 (10th Cir.1988), certiorari denied, — U.S. -, 109 S.Ct. 260, 102 L.Ed.2d 248. Roulo did not offer as much evidence on the secondary meaning of her cards as on their distinctiveness, no doubt since it is easier to show the cards *937are unique than to poll greeting card consumers on whether they identify the FS cards with a single source. However, secondary meaning is not necessary where the trade dress is distinctive.

Once a trade dress is found to warrant protection, the plaintiff must establish that defendant’s trade dress is sufficiently similar to create the likelihood of confusion between the two greeting card lines. In assessing the likelihood of marketplace confusion, the factors to be considered include the similarity of the trade dresses, the products to which the trade dresses are attached, the area and manner of concurrent use, the degree of care likely to be exercised by consumers, the strength of the plaintiffs trade dress, and the actual confusion and intent on the part of the alleged infringer to pass off the infringer’s goods as those of the plaintiff. Intema-tional Kennel Club v. Mighty Star, Inc., 846 F.2d 1079, 1087 (7th Cir.1988). Berrie contends that the sole evidence of a side-by-side comparison at trial was insufficient to establish the existence of marketplace confusion. In addition, Berrie argues that its use of its own name on the header of the display rack demonstrates a lack of intent to infringe and militates against a finding of likelihood of confusion.

Although side-by-side comparison is only one factor to be considered in evaluating the likelihood of confusion, the similarity in this instance is striking. The cards in each line are identically sized with handwritten messages containing ellipses in brown ink flanked by colored stripes imprinted on beige or cream-colored textured paper. The two lines of cards are displayed in identical racks (the TY rack is brass-toned and the FS rack is silver-toned) in four vertical rows of eight cards bearing the same color but varying messages for a total of thirty-two cards.

In an attempt to accentuate the dissimilarities between the two card lines, Berrie points out that the TY line uses a foil butterfly and no foil stripes and the header on the display rack bears the name “Touching You” and identifies Berrie as the source of the cards. A side-by-side comparison, however, reveals that these differences are easily overlooked, especially given that the marketing channels for greeting cards do not promote a careful analysis of the product prior to purchase as indicated by Berrie’s testimony that the cards were impulse purchase items. Berrie argues that the side-by-side comparison should not be given an inordinate amount of weight. However, where the trade dress relates to the same products which are sold side by side in retail channels, a direct visual comparison is appropriate, AmBrit, supra (“Klondike” and “Polar B’ar” brand ice cream bars), even though such a comparison may be inappropriate where a similar trade dress or mark is used on dissimilar products. James Burrough Ltd. v. Sign of the Beefeater, Inc., 540 F.2d 266 (7th Cir.1976) (“Beefeater” used on restaurant sign and gin).

To further add to the confusion in the marketplace, Berrie’s TY cards were sold through the same retailers as Roulo’s FS line had been under her contract with Ber-rie. Had Roulo desired to market her FS cards through these retailers, the FS cards would have been displayed along with other greeting cards somewhere near the TY cards. The substantial visual similarity between the cards cannot be underestimated given that these cards are impulse purchase items likely to be marketed near each other at a price less than one dollar, discouraging much thoughtful research before purchase.

There was also sufficient evidence from which the jury could have discerned an intent to imitate Roulo’s FS cards. Given the myriad of artistic expressions which could have been chosen for Berrie’s greeting cards, the near identity of their appearance is itself strong evidence of copying. In addition, Berrie testified that once Roulo indicated that she would not renew her contract, Berrie began work on a card line to replace the FS cards to fit into the FS display with verses “a la Feeling Sensitive,” the name of Roulo’s line.

The strength of a particular trade dress relates in part to its uniqueness. As evidence of the strength of the FS trade *938dress, in addition to the other testimony, Bill Dawson, Berrie’s Vice President of Marketing, testified that the FS line of cards was the “finest sensitive verse greeting card line ever released.” Thus, even absent the difficult-to-acquire evidence of actual confusion, there was adequate evidence for the jury, over and above the side-by-side comparison, to determine that the public was likely to confuse the two card lines.

Of slightly more merit is Berrie’s assertion that Roulo abandoned her trade dress. After Roulo’s contract with Berrie ended in April of 1980, she did not attempt to market the FS cards again but instead sought to publish a book of verse in the spring of 1980 which proved unsuccessful. Upon the termination of her contract with Berrie, Roulo was entitled, but declined, to receive the lists of distributors who had carried the FS cards and to receive any unsold inventory of the cards at Berrie’s cost. At the Chicago Gift Show in the summer of 1980, Roulo had brought a new line of emotional verse cards to market when she discovered Berrie’s TY cards. Later in January of 1981, Roulo released this new line of emotional verse greeting cards and marketed them until 1983.

A prima facie case of abandonment was established by Roulo’s failure to use the FS trade dress for more than two years. 15 U.S.C. § 1127. However, this presumption may be rebutted by evidence explaining the nonuse or demonstrating the lack of an intent not to resume use. P.A.B. Produits Et Appareils de Beaute v. Satinine Societa in Nome Collettivo di S.A. e. M. Usellini, 570 F.2d 328, 334 (C.C.P.A.1978); Star-Kist Foods, Inc. v. P.J. Rhodes & Co., 769 F.2d 1393, 1396 (9th Cir.1985); Chandon Champagne Corp. v. San Marino Wine Corp., 335 F.2d 531, 535 (2d Cir.1964). Some courts require the owner of the trade dress not used for over two years to demonstrate the intent to resume use. AmBrit, 805 F.2d at 994; Exxon Corp. v. Humble Exploration Co., 695 F.2d 96, 102-103 (5th Cir.1983); Silverman v. CBS, Inc., 870 F.2d 40, 46 (2d Cir.1989). Other courts require the owner to prove the presumably lesser showing of the absence of an intent to abandon. Sterling Brewers, Inc. v. Schenley Industries, Inc., 441 F.2d 675 (C.C.P.A.1971); Saratoga Vichy Spring Co. v. Lehman, 625 F.2d 1037, 1044 (2d Cir.1980). We think the proper approach is to require evidence to demonstrate the intent not to resume use. This reading comports with the language^of the statute and the seeming intent of Congress to prevent hoarding of trademarks where the owner has no intent to abandon, although no intent to resume use of the trademark within the reasonably foreseeable future. Intent to abandon connotes a permanent intent not to resume use which would be virtually impossible to prove. Our interpretation is consistent with the legislative history indicating that Congress substituted the present phrase “intent not to resume use” for the phrase “intent to abandon”. See Silverman, 870 F.2d at 46. As noted in Exxon, this distinction is pivotal where the trademark or trade-dress owner is attempting to “hoard” the mark or dress with no intent to resume meaningful commercial use. Exxon, 695 F.2d at 102-103.

However, in contradiction to defendant’s contention that the plaintiff must also bear the burden of persuasion on this affirmative defense, the owner of the trademark need only produce evidence to rebut the presumption while the ultimate burden of persuasion rests on the defendant. Silverman, 870 F.2d at 47 (“A proprietor who temporarily suspends use of a mark can rebut the presumption of abandonment by showing reasonable grounds for the suspension and plans to resume use in the reasonably forseeable future when the conditions requiring suspension abate.”). Therefore it was not erroneous to refuse to give Berrie’s proposed instruction No. 37 which would have placed the burden of proof on Roulo. The trial judge properly instructed the jury that the defendant must prove by a preponderance of the evidence that plaintiff abandoned the FS trade dress, that abandonment is found when use has been discontinued with intent not to resume use, and that such intent may be presumed from nonuse for two *939consecutive years. Finally, the judge instructed: “The purely subjective intention in the-owner’s mind to reengage in a former enterprise at some further time is not, standing alone, sufficient to avoid abandonment.” (Tr. 982).

As evidence of her intent to resume use within the reasonably foreseeable future, Roulo testified that she would have marketed the FS cards if the TY cards had not been developed. She cites her presence at the 1980 Chicago Gift Show as evidence of this intent. Berrie weakly contends that Roulo’s failure to begin marketing her FS cards immediately or to purchase the remaining inventory indicates an intent not to resume use. Given that the evidence is meager on either side of the issue, the jury's special verdict that Roulo did not abandon her trade dress is not against the weight of the evidence. It was within the jury’s prerogative to credit plaintiff’s testimony against abandonment.

II.

COPYRIGHT INFRINGEMENT

To demonstrate an infringement of Roulo’s copyright, Roulo must establish: (1) that she owns the copyright in question; (2) that Berrie had access to the copyrighted work; and (3) that there was substantial similarity between the two works. Atari, Inc. v. North American Philips Consumer Electronics Corp., 672 F.2d 607, 614 (7th Cir.1982), certiorari denied, 459 U.S. 880, 103 S.Ct. 176, 74 L.Ed.2d 145. Defendant concedes the existence of the second element of the test but challenges the sufficiency of the evidence on the scope of Roulo’s copyright and the jury’s finding of substantial similarity between the two card lines. Roulo owns a registered copyright in the FS cards themselves as well as in a display header bearing her likeness, although she used a different, but similar, header than the one registered. Berrie claims that the material subject to copyright protection did not include the display rack, the pattern in which the cards were displayed, the header actually used and the price. Consequently, Berrie claims the judge erred in refusing to instruct the jury to render separate damage awards and in commenting that the trade dress and material protected by copyright were the same.

In Atari, this Court examined the idea/expression dichotomy in the context of video games to determine which elements of the Pac-Man video game were expression, subject to protection, rather than ideas which are in the public domain. In so doing, this Court stated that dissection of the subject matter into copyrighted and unprotected elements is generally rejected in favor of examining the “total concept and feel” of the copyrighted work. Atari, 672 F.2d at 614. The Court cautioned that while such dissection was not appropriate, the substantial similarity inquiry must “take into account that the copyright laws preclude appropriation of only those elements of the work that are protected by the copyright.” Such dissection is inappropriate since the substantial similarity inquiry is conducted from the perspective of the “ordinary observer” who “unless he set out to detect the disparities, would be disposed to overlook them, and regard their aesthetic appeal as the same.” Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487, 489 (2d Cir.1960).

As discussed above, the two card lines are strikingly similar in appearance. Further, the jury’s finding of substantial similarity is buttressed by the evidence that Berrie patterned the TY cards after Roulo’s cards. Berrie insists that the jury was allowed to consider elements of similarity unprotected by copyright laws, such as the size of the cards, the color of the paper, ink, and border designs, the general concept of stripes, the ellipses and the single-side format. While it is true that these elements are not individually capable of protection, just as individual words do not deserve copyright protection, it is the unique combination of these common elements which form the copyrighted material. The district court correctly instructed the jury by adopting the defendant’s proposed instruction that the common elements noted above are not copyrightable, adding that “[t]he combination of these ele*940ments in an arrangement or layout is entitled to copyright protection, however, if you find that arrangement or layout to be original and unique.” (Instruction No. 3). Defendant fails to acknowledge that its TY cards not only used the same unprotectable elements as the FS cards, but also used them in a format nearly identical to that used by the FS cards. It cannot be disputed that Berrie could have produced a non-infringing card with colored stripes, but Berrie used similar stripes flanking the verse on both the left and right side from top to bottom just as the FS cards did. The color of paper is unprotected, yet the TY cards are printed on a nearly identical beige paper. The use of cursive writing is also a common element as well as the brown ink used, yet TY cards, out of the entire range of colors and handwritings available, chose brown ink and a similar cursive writing. The minute differences which Berrie painstakingly incorporated to prevent a mirror image copy of FS cards are insignificant in light of the nearly identical composition and appearance used in both card lines. Berrie’s contention that there is no support in law for the proposition that a compilation of common elements is subject to copyright protection is ridiculous. “[Cjards, considered as a whole, represent a tangible expression of an idea_ Greeting cards are protected under 17 U.S.C. § 5(a) or (k) as a book (citations omitted). They are the embodiment of humor, praise, regret or some other message in a pictorial and literary arrangement.” Roth Greeting Cards v. United Card Co., 429 F.2d 1106, 1110 (9th Cir.1970). Of course Roulo could not copyright the use of beige paper for greeting cards. Her copyright is in the artistic combination of all the common elements in the previously described design. There was ample evidence for the jury to find substantial similarity and the jury was properly instructed as to which elements were subject to copyright protection.

Berrie also contends that the district judge erroneously excluded an expert witness, a copyright lawyer, from testifying on the scope of protection to be afforded Roulo’s copyright. Presumably the expert would have testified that the only material subject to copyright was the verse of the FS cards — which was not copied — and not the design of the cards as a whole. As previously noted, greeting cards have been held to be proper subjects of copyright as literary and pictorial compositions. Roth, 429 F.2d at 1110. The court correctly instructed the jury on which elements were subject to copyright protection without the assistance of the expert. There was no error in excluding this expert testimony.

III.

DAMAGES

A. Merger of Trade-Dress and Copyright Claims

Berrie’s barrage of challenges continues in its contention that the district judge committed reversible error by equating the subject matter protected by Roulo’s copyright with the material cognizable under Roulo's trade dress infringement claim under the Lanham Act, although the jury was instructed individually and rendered separate verdicts as to liability under each Act. In so doing, Berrie asserts that features of Roulo’s FS cards that were unprotected under copyright law but cognizable in a trade-dress infringement claim were considered by the jury in determining whether an infringement under either claim occurred. Specifically, Judge Kocoras explained: “Under the unique facts of this ease, the protected copyright material is the same thing as the trade dress.” Although the jury rendered separate verdicts in favor of the plaintiff on the trade dress and copyright liability claims, the jury assessed a single damage award. The district judge committed no error in instructing the jury to render a single verdict on damages since both the Lanham Act and the Copyright Act provide for the same measure of damages under like circumstances. Under each Act, the plaintiff is required to present proof of the infringer’s gross revenues, with the burden on the infringer to establish any deductions for expenses or to demonstrate that the profits introduced by the plaintiff were either “at*941tributable to factors other than the copyrighted work,” 17 U.S.C. § 504, or not attributable to infringing use under the Lanham Act. Maltina Corp. v. Cawy Bottling Co., 613 F.2d 582, 586 (5th Cir.1980); 15 U.S.C. § 1117.

Berrie also contends that the district court should have directed a verdict against the recovery of profits under the Lanham Act because Roulo failed to demonstrate any actual damages, confusion, competition between the parties or wilfulness on the' part of Berrie. The Lanham Act specifically provides for the awarding of profits in the discretion of the judge subject only to principles of equity. As stated by this Court, “The trial court’s primary function is to make violations of the Lanham Act unprofitable to the infringing party.” Otis Clapp & Son, Inc. v. Filmare Vitamin Co., 754 F.2d 738, 744 (7th Cir.1985). Other than general equitable considerations, there is no express requirement that the parties be in direct competition or that the infringer wilfully infringe the trade dress to justify an award of profits. See, e.g., Baker v. Simmons Co., 325 F.2d 580 (1st Cir.1963), certiorari denied, 382 U.S. 820, 86 S.Ct. 49, 15 L.Ed.2d 67. Profits are awarded under different rationales including unjust enrichment, deterrence, and compensation. Maltina, 613 F.2d at 584-585, Maier Brewing Co. v. Fleischmann Distilling Corp., 390 F.2d 117, 123-124 (9th Cir.1968), certiorari denied, 391 U.S. 966, 88 S.Ct. 2037, 20 L.Ed.2d 879. The case of Schutt Manufacturing Company v. Riddell, Inc., 673 F.2d 202, 206 (7th Cir.1982), cited by defendant addresses an award of plaintiffs damages based on plaintiffs actual losses rather than on defendant’s profits. Where plaintiff seeks an award of damages, plaintiff must show that defendant’s infringement caused those losses. Here, however, an award of profits was appropriate under either a deterrence or unjust enrichment theory even if plaintiff’s actual sustained losses may have been less. Given the evidence of intentional imitation and the substantial similarity between the two card lines, the district judge’s decision to instruct the jury that an award of profits would be appropriate was not an abuse of discretion.

In order to sustain her burden to demonstrate the amount of profits secured through sales of the infringing greeting cards, Roulo introduced expert testimony of John Moran, a Certified Public Accountant, who testified that Berrie’s sales receipts of TY cards were $5.9 million and that the direct cost of producing the cards was $945,000. In order to meet its burden of proving all deductions from profits, defendant’s witnesses, Steven Willis and A. Curtis Cooke, testified that incremental profits were only $2.9 million and that the amount of profits properly attributable to the protected elements after deductions for costs was $38,601. The jury was free to reject this calculation based on cross-examination testimony that documents evidencing the alleged operating expenses for certain years were lost in a fire and based on Berrie’s general inability to substantiate his deductions. In addition, Berrie’s witness, Willis, improperly deducted certain administrative expenses without demonstrating they were variable costs. Fixed costs are not deducted from the profit calculation.

Berrie also argues that the jury was not free to ignore its evidence apportioning profits between the infringing and non-infringing elements of the cards as required under 17 U.S.C. § 504(b). Berrie’s evidence indicated that the infringing elements of the TY cards contributed to only $38,601 in profits using Berrie’s gross profit figure, or $65,645 using Roulo’s gross profit figure. An infringer is entitled to an apportionment when “the evidence is sufficient to provide a fair basis of division so as to give the copyright proprietor all the profits that can be deemed to have resulted from the use of what belonged to him.” Sheldon v. Metro-Goldwyn Pictures Corp., 309 U.S. 390, 402, 60 S.Ct. 681, 685, 84 L.Ed. 825. The burden of proving an apportionment is on the defendant. Frank Music Corp. v. Metro-Goldwyn-Mayer, Inc., 772 F.2d 505, 518 (9th Cir.1985). The jury’s verdict of $4.3 million rather than *942the entire $5 million proposed by the plaintiff is evidence that the jury either apportioned the award or accepted some of Ber-rie’s other deductions. Given that Berrie appropriated the “total concept and feel” of Roulo’s cards, rather than a few distinct protected elements, the jury’s failure to accept defendant’s suggested apportionment is not against the weight of the evidence.

B. Laches

[30] As a final matter, Berrie contends that Roulo’s recovery of profits should be barred by the equitable doctrine of laches due to her 21-month delay in filing this action after learning of the TY card line. Berrie asserts that it has established that: (1) Roulo had knowledge of the use of her trade dress or copyrighted material; (2) Roulo inexcusably and unreasonably delayed in filing this action; and (3) Berrie relied to its detriment and was prejudiced by Roulo’s delay. The Money Store v. Harriscorp Finance, Inc., 689 F.2d 666, 674 (7th Cir.1982). A two year delay in filing an action following knowledge of the infringement has rarely been held sufficient to constitute laches. Piper Aircraft Corp. v. Wag-Aero, Inc., 741 F.2d 925, 933 (7th Cir.1984). Roulo explained that her delay in filing was due to her inquiry into the facts of the case to determine the merit of her claim against Berrie. Berrie has not cited any conduct to indicate that it actually relied to its detriment on Roulo’s failure to bring suit sooner except for the actual marketing of the TY cards. Since Berrie began its promotion of the TY cards prior to any knowledge of Roulo or her purported acquiescence, Berrie cannot assert that its distribution of the TY line was made in actual reliance on Roulo’s conduct. The jury’s verdict was not against the weight of this evidence.

C. Attorney’s Fees

Roulo claims that the district judge erroneously denied her attorney’s fees under both the Lanham and Copyright Acts. The Lanham Act provides that attorney’s fees may be awarded “in exceptional cases.” This language apparently was intended to allow the recovery of fees “in infringement cases where the acts of infringement can be characterized as ‘malicious,’ ‘fraudulent,’ ‘deliberate,’ or ‘willful.’” S.Rep. No. 1400, 93d Cong., 2d Sess. 2, reprinted in 1974 U.S.Code Cong. & Admin.News 7132, 7133. This Circuit has interpreted this section to require a finding of wilful infringement. Hairline Creations, Inc. v. Refalas, 664 F.2d 652, 657-658 (7th Cir.1981). We agree with the district court that this was not an exceptional case warranting an award of attorney’s fees. Berrie attempted to create a line of greeting cards similar to the FS line but made conscious efforts to create elements of dissimilarity, although the dissimilarities were not sufficient in this case.

The Copyright Act also allows the award of attorney’s fees to the prevailing party in the discretion of the court. 17 U.S.C. § 505. Attorney’s fees under this section have been awarded for the purposes of encouraging the assertion of color-able copyright claims, deterring infringement, and making the plaintiff whole. McCulloch v. Albert E. Price, Inc., 823 F.2d 316, 321 (9th Cir.1987).

The district judge awarded costs to Rou-lo but denied attorney’s fees under either Act on the basis that Berrie’s conduct was not “flagrant” and that plaintiff received a substantial and considerable award of profits, such that an award of attorney’s fees would be an unwarranted windfall to the plaintiff. This Court has determined that a finding of wilful copyright infringement will support an award of attorney’s fees, although wilfulness is not necessarily a prerequisite. International Korwin Corp. v. Kowalczyk, 855 F.2d 375 (7th Cir.1988), and cases cited therein. The Copyright Act is silent on the circumstances that warrant a fees award and the circuit courts are not in agreement on this issue. Some courts require a finding of bad faith or frivolousness. Jartech, Inc. v. Clancy, 666 F.2d 403 (9th Cir.1982), certiorari denied, 459 U.S. 826, 879, 103 S.Ct. 58, 175, 74 L.Ed.2d 62, 143; Del Madera Properties v. Rhodes and Gardner, Inc., 820 F.2d 973 *943(9th Cir. 1987). Others award fees routinely to the prevailing party, especially to the plaintiff to encourage the assertion of colorable claims. Diamond v. AmLaw Publishing Co., 745 F.2d 142, 148 (2d Cir.1984); The Micromanipulator Co. v. Bough, 779 F.2d 255, 259 (5th Cir.1985).

The district judge declined to award fees due to Berrie’s lack of wilfulness and in light of the magnitude of the award of profits. Based on these considerations he did not abuse his discretion under any of the above rationales. Plaintiff secured an award of profits significantly greater than she would have received if she had renewed her contract with Berrie and perhaps more than she would have made if she had independently marketed the FS cards following the expiration of her contract with Berrie. Attorney’s fees were not necessary to compensate the plaintiff and the defendant was sufficiently deterred by the substantial award of profits. Awarding fees to the plaintiff under these facts would serve only to sanction the defendant which is inappropriate where the infringement was not wilful.

The decision of the district court is affirmed, and the cross-appeal is dismissed.

APPENDIX

*944

RADER, Circuit Judge.

Anthony California, Inc. (Anthony) designed, and sells, the Sonoran and Hercules collections of bedroom furniture. James Chang is Anthony’s president and principal shareholder. Anthony’s competitor, Amini Innovation Corp. (Amini), sells the LaFrancaise and Paradisio collections of bedroom furniture. . Amini sued Anthony for infringement of its copyrights and design patent. The United States District Court for the Central District of California granted summary ■ judgment of non-infringement of the copyrights and design patent. , Because the record discloses a lingering genuine issue of material fact, this court reverses and remands for further proceedings as appropriate.

I.

Amini holds U.S. Copyright Registrations for, as the registrations state, “carved ornamental woodwork” in the bed and dresser with mirror in the LaFran-caise line, and the bed, dresser with mirror, armoire, and night stand in the Parad-isio line. Amini also owns U.S. Patent No. D475,218 (June 3, 2003) (the ’218 patent) for the design of its Paradisio bed frame. These products were first sold in 2000 and 2001. After Anthony’s products went on sale, in August 2003, Amini informed James Chang of its belief that Anthony was infringing Amini’s copyrights and patents. Amini further demanded that Anthony stop any further sale, of its accused products. Anthony did not agree. Amini sued Anthony for six counts of. copyright infringement and one count of design patent infringement on December 1, 2003. After discovery, which included expert testimony, the trial court granted Anthony’s motion for summary judgment of non-infringement of the copyrights and the design patent, and denied Amini’s motion for partial summary judgment of infringement *1368of its copyright claims. Amini Innovation Corp. v. Anthony California, Inc., CV OS-8749 SJO, CV 04-1192 SJO, CV-04-1316 SJO (C.D.Cal., Dec. 6, 2004) (Order). This appeal followed.

II.

This court has jurisdiction over this appeal because it includes claims for patent, as well as copyright, infringement. 28 U.S.C. §§ 1292, 1295, 1338 (2005). This court applies copyright law as interpreted by the regional circuits, in this case the for the Ninth Circuit. Atari Games v. Nintendo of Am., 897 F.2d 1572, 1575 (Fed. Cir.1990). The United States Court of Appeals for the Ninth Circuit reviews interpretations of the copyright act without deference. See Ets-Hokin v. Skyy Spirits, Inc., 225 F.3d 1068, 1073 (9th Cir. 2000). The protection of a particular design under the copyright law raises a mixed question of law and fact that the Ninth Circuit also reviews without deference. Id. Where reasonable minds could differ on the issue of substantial similarity, however, summary judgment is improper. Cavalier v. Random House, Inc., 297 F.3d 815, 822 (9th Cir.2002).

In patent infringement suits involving the grant of a motion for summary judgment, the law enunciated in Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986) governs. This court reviews those determinations without deference. Golan v. Pingel Enter., Inc., 310 F.3d 1360, 1367 (Fed.Cir. 2002). An appellate court does not review a district court’s “findings” of fact underlying summary judgment under the clearly erroneous standard of Rule 52. Avia Group Intern., Inc. v. L.A. Gear California, Inc., 853 F.2d 1557, 1561 (Fed.Cir. 1988). To overturn a summary judgment, the non-movant need only show that one or more of the facts on which the district court relied was “genuinely in dispute,” as that phrase is interpreted in Anderson, and was material to the judgment. Avia, 853 F.2d at 1561-62 (citing Anderson, 477 U.S. at 248, 106 S.Ct. 2505).

III.

Copyright Infringement

To prevail on a copyright infringement claim in the Ninth Circuit, a plaintiff must establish ownership and unauthorized copying of protected expression. Data East USA Inc. v. Epyx, Inc., 862 F.2d 204, 206 (9th Cir.1988). Copying requires evidence that a defendant literally copied the designs or, alternatively, that a defendant had access to the protected designs before creating the accused designs with an additional showing of “substantial similarity not only of the general ideas but of the expression of those ideas as well.” Shaw v. Lindheim, 919 F.2d 1353, 1356 (9th Cir.1990). The parties do not disagree that Amini owns the disputed copyrights and that they are valid. The parties also do not dispute the catalogue of standard ornamental woodwork features which, according to the trial court, make up the protected “carved ornamental woodwork.” These features include a lion’s paw, ball, reeds, leaf-and-flower motifs, foliate scrolls, C— and S-shaped scrolls, a serpentine decoration, a seashell motif, laurel wreaths, an iron-canopy rail, beads, and moldings. The accused furniture also has these features. The parties dispute, however, Mr. Chang’s access to the protected work as well as the degree of similarity between the protected designs and the accused designs.

Access

The import of a defendant’s access to the disputed designs is governed, in the Ninth Circuit, by an inverse-ratio rule that allows “a lesser showing of substantial similarity if there is a strong showing of access.” Three Boys Music Corp. v. Bol *1369 ton, 212 F.3d 477, 486 (9th Cir.2000). Access may be inferred from circumstantial evidence, and may be shown by at least a “reasonable,” beyond a “bare,” possibility that the defendant viewed the protected work. Id. at 482. A wide dissemination of the protected works leads to a reasonable possibility of access. Id.

The record does not conclusively resolve the question of the access of Mr. Chang, Anthony’s designer, to the protected work before the creation of the Anthony products. Specifically, the record includes a disputed translation from Chinese of Mr. Chang’s deposition, as well as evidence that Amini displayed its designs at furniture trade shows which Mr. Chang attended. The record does not evince an analysis of this access evidence. Instead, the trial court apparently based its finding of access on Mr. Chang’s deposition testimony that neither he nor Anthony made any effort to determine if their designs violated intellectual property rights. Mr. Chang’s deposition did not give any other hints about Anthony’s access to the protected designs. On the basis of this evidence, without more, a strong showing of substantial similarity is required to prove infringement in this case, i.e., one approaching “striking” similarity. Id. at 485.

Similarity

In assessing copyright infringement, a court must first distinguish protectable expression from unprotectable design elements. Johnson Controls, Inc. v. Phoenix Control Sys., Inc., 886 F.2d 1173, 1175 (9th Cir.1989). Copyright protects only expression of ideas, not ideas themselves. See Rachel v. Banana Republic, Inc., 831 F.2d 1503, 1507 (9th Cir.1987). In this case, the trial court construed the protectable expression in the copyright registrations to be the “carved ornamental woodwork,” not the furniture designs as a whole. In other words, the trial court correctly applied the Ninth Circuit standard that a purely utilitarian article — such as bedroom furniture — receives no protection. On the other hand, “if the shape of a utilitarian article incorporates features, such as artistic sculpture, carving, or pictorial representation, which can be identified separately and are capable of existence independently as a work of art, such features will be eligible for registration.” Fa-brica Inc. v. El Dorado Corp., 697 F.2d 890, 893 (9th Cir.1983). Thus, the trial court correctly accorded protection to the expressive ornamental carvings on the furniture, but not the furniture pieces as a whole.

After identifying the protectable features, the trial court turned to the assessment of similarity.. The Ninth Circuit uses a two-part analysis — an “extrinsic test” and an “intrinsic test” — to determine whether two works are substantially similar. Id. The “extrinsic test” is an objective comparison of specific expressive elements. Cavalier, 297 F.3d at 822 (citing Shaw, 919 F.2d at 1356). Thus, the objective-extrinsic test determines whether two works “share a similarity of ideas and expression as measured by external, objective criteria.” Swirsky v. Carey, 376 F.3d 841, 845 (9th Cir.2004). The objective-extrinsic test requires “analytical dissection of a work and expert testimony.” Three Boys, 212 F.3d at 485. It assesses specific criteria including “the type of artwork involved, the materials used, the subject matter, and the setting for the subject.” Shaw, 919 F.2d at 1356.

The “intrinsic test” is a subjective comparison that focuses on whether the ordinary reasonable audience would find the works substantially similar in the “the total concept and feel of the works.” Cavalier v. Random House, Inc., 297 F.3d 815, 822 (9th Cir.2002). Once the extrinsic test is satisfied, the factfinder applies the in-*1370trinsie test. Three Boys, 212 F.3d at 485. Therefore, K[f]or the purposes of summary judgment, only the extrinsic test is important because the subjective question whether works are intrinsically similar must be left to the jury.” Swirsky, 376 F.3d at 845; accord Shaw, 919 F.2d at 1355-56 (observing that summary judgment is not highly favored on questions of substantial similarity in copyright cases).

In this case, the trial court erred in expanding its application of the “extrinsic” part of the infringement test to encompass an examination of “the total concept and feel of the works.” Thus, the trial court mistakenly expanded the extrinsic elements of the infringement test to encompass the intrinsic elements as well. The trial court’s error is significant because on summary judgment the Ninth Circuit places the extrinsic part of the infringement test within the court’s purview, but disfavors application of the subjective-intrinsic part.

In applying the extrinsic test, the district court opined that the ornamental features, themselves not original, also did not exhibit an original “selection and arrangement.” Whether a feature is original falls reasonably within the district court’s role of assessing the extrinsic prong of infringement on summary judgment, and we see no error in this assessment. However, the trial court then proceeded to examine, on summary judgment, the fact intensive question of the total concept and feel of the carvings in the furniture. The court’s conclusion appears to be based primarily on its own visual inspection. Thus, the court itself performed the intrinsic test.

In addition, to reach its conclusion on summary judgment, the trial court must have deduced that no “reasonable jury” could reach a different decision. See Freedman Seating Co. v. Am. Seating Co., 420 F.3d 1350, 1356 (Fed.Cir.2005). This court finds that a reasonable jury could conclude that at least some of the accused designs satisfy the heightened showing of substantial similarity thus far required in this case. This disagreement weighs in favor of reversing the trial court’s grant of summary judgment. See Twentieth Century-Fox Film Corp. v. MCA, Inc., 715 F.2d 1327, 1329-30 (9th Cir.1983) (reversing a grant of summary judgment because reasonable minds could differ on the question of substantial similarity). For instance, the protected design uses a unique lion’s paw with five toes; the accused design incorporates a near-copy of this whimsical device, including the anatomically incorrect presentation of the toes. In some cases, the legs rising from the lion’s paws in both products are themselves decorated with similar reed and scroll motifs. Headboards in both the protected and accused designs have a virtually identical serpentine shape, are decorated with moldings of identical shape and placement, and flourish similar bedposts with scrollwork embellishing spherical bases. Vertical decorations in some of the protected and accused products both use a motif of bundled reeds. Furthermore, it is well settled that a jury may even find a combination of unprotectible elements to be protectible under the extrinsic test because “ ‘the over-all impact and effect indicate substantial appropriation.’” Three Boys, 212 F.3d at 485. In sum, this court perceives that a consideration of these facts by a jury could reveal a variety of opinions on the question of infringement. This court cannot agree that no reasonable jury could find infringement on this record.

IV

Design Patent Infringement

“A design patent protects the non-functional aspects of an ornamental design as seen as a whole and as shown in the patent.” KeyStone Retaining Wall *1371 Sys., Inc. v. Westrock, Inc., 997 F.2d 1444, 1450 (Fed.Cir.1993). An aspect is functional “if it is essential to the use or purpose of the article or if it affects the cost or quality of the article.” Inwood Labs., Inc. v. Ives Labs., Inc., 456 U.S. 844, 851, 102 S.Ct. 2182, 72 L.Ed.2d 606 (1982). It is the drawings in the patent, not just one feature of the claimed design, that define the patented design. KeyStone, 997 F.2d at 1450. An ordinary observer test governs design patent infringement: “[If] in the eye of an ordinary observer, giving such attention as a purchaser usually gives, two designs are substantially the same, if the resemblance is such as to deceive such an observer, inducing him to purchase one supposing it to be the other, the first one patented is infringed by the other.” Gorham Co. v. White, 14 Wall. 511, 81 U.S. 511, 528, 20 L.Ed. 731 (1871). In addition, the accused design must appropriate the novel ornamental features of the patented design that distinguish it from the prior art. Oakley, Inc. v. Int’l Tropic-Cal, Inc., 923 F.2d 167, 169 (Fed. Cir.1991).

Claim Construction

The ’218 patent claims “[t]he ornamental design for a bed frame, as shown and described.” Under the KeyStone standard, the claim extends to the overall design of the bed frame. 997 F.2d at 1450. If a design includes both functional and ornamental features, infringement occurs if an ordinary person “would be deceived by reason of the common features in the claimed and accused designs which are ornamental.” Read Corp. v. Portec, Inc., 970 F.2d 816, 825 (Fed.Cir.1992). It is noteworthy, however, that the deception that arises is a result of similarities in the overall design, not of similarities in ornamental features considered in isolation. See, e.g., KeyStone, 997 F.2d at 1450; see also In re Lapworth, 59 C.C.P.A. 738, 740, 451 F.2d 1094 (1971) (“It is distinctiveness in overall appearance of an object, when compared with the prior art, rather than minute details or small variations in configuration ... that constitutes the test of design patentability.”) Here, the trial court perhaps focused too narrowly on the isolated ornamental features, namely “the ornamental wood carvings shown in the application.” The trial court specifically noted that “[e]ach of the patent’s seven drawing figures show[s] the designs of certain portions of the bed frame.” While literally true, that statement overlooks that the claimed drawing on the face of the ’218 patent and three of the drawings in the patent itself are views of the bed as a whole, two are views of the entire headboard, and two illustrate entire leg-post assemblies. These drawings incorporate almost all of the design features cata-logued by the trial court. These patent drawings show the complete bed frame with many ornamental features. These drawings affirm that it is the overall “bed frame” that is patented — not just the details of its ornamentation.

Infringement

The trial court decided that no reasonable jury could find infringement of Ami-ni’s design patent, pointing out only that “[t]he claimed design has four hollow metal orb and bed posts which are absent in the accused bed frame,” and that plaintiff “failed to proffer evidence showing that the resemblance of the feet ... will induce a purchaser to buy the accused product supposing it to be the [patented] bed frame.”

At the outset, this court perceives again that conclusions about reasonable jurors are difficult to make on an issue of this factual dimension. See Bernhardt, L.L.C. v. Collezione Europa USA, Inc., 386 F.3d 1371, 1383 (Fed.Cir.2004) (“Both the ordinary observer and point of novelty *1372tests are factual inquiries that are undertaken by the fact finder during the infringement stage of proceedings, after the claim has been construed by the court.”). More important, however, this court perceives a procedural error. Specifically, as noted above, the trial court mistakenly analyzed each element separately instead of analyzing the design as a whole from the perspective of an ordinary observer. The trial court is correct to factor out the functional aspects of various design elements, but that discounting of functional elements must not convert the overall infringement test to an element-by-element comparison. Thus, the trial court erred in its application of the “overall similarity” test on summary judgment. This court cannot agree on this record that “no reasonable jury could find that the designs of these two pieces of furniture are substantially similar.”

Novelty

With respect to the points of novelty test, Amini has not yet introduced sufficient evidence to establish with particularity its points of novelty. On remand, in addition to the evidence on record, Amini “must introduce ... at a minimum ... [the] prosecution history [for the ’218 patent], and the relevant prior art references cited in the prosecution history; and must present, in some form, its contentions as to points of novelty.” Id. at 1384. Moreover, Amini may find it advantageous “to present additional evidence, such as expert testimony, to assist the fact finder in understanding its contentions pertaining to the points of novelty.” Id.

Conclusion

Because the trial court mistakenly applied both the extrinsic and intrinsic tests for substantial similarity in copyright infringement, and an element-by-element test for design patent infringement, this court reverses the grants of summary judgment of non-infringement and remands for further proceedings as appropriate.

COSTS

Each party shall bear its own costs.

REVERSE and REMAND