In this trademark infringement case, Eclipse Associates Limited (“EAL”), a computer company, appeals the district court’s order in favor of Data General Corporation (“Data General”) enjoining EAL from using “ECLIPSE” as its trademark and its business name in the United States. The issues are (1) whether the district judge erred in finding there was a likelihood of confusion in EAL’s use of ECLIPSE; (2) whether the lack of evidence of actual confusion should be determinative; and (3) whether it was error to exclude evidence of use of the mark by unrelated third parties. We affirm the district court.

I. FACTS AND PROCEDURAL HISTORY

A. Data General’s Use of the Mark Eclipse

Data General is a manufacturer and distributor of a full range of computer products. Data General adopted the ECLIPSE mark in 1974 to identify its digital computers and hardware. In 1975 the mark was adopted for its software and computer publications. Since that time, Data General has continuously used the mark ECLIPSE in the United States in connection with the sale, distribution and advertising of these goods and has adopted other ECLIPSE marks (e.g., ECLIPSE, ECLIPSE MV/2000, ECLIPSE MV/9000) to identify specific computer products. Data General is the owner of the trademarks ECLIPSE and ECLIPSE with additional terms for computer products as described in Data General’s U.S. Registration Nos. 1,017,909, 1,107,836, 1,227,384, 1,188,463, 1,227,383 and 1,213,102.

Since 1974, Data General has sold over 15,000 ECLIPSE computer products in the United States. Data General sells to a broad market, including the architectural, engineering and construction (“AEC”) market and the digital mapping market. Sales revenues for these markets have totaled more than $3 billion. In fiscal year 1987, Data General’s sales were about $500 million for ECLIPSE computer products.

B. EAL’s Use of the Mark Eclipse

In 1978, Eclipse Computer Services was formed in the United Kingdom. It developed and sold sophisticated software systems, primarily to the AEC market. Eclipse Computer Services adopted the mark ECLIPSE to identify its application software and services. In 1983, Eclipse Computer Services was incorporated in the United Kingdom as Eclipse Associates Limited (“EAL”).

EAL has used the marks ECLIPSE and ECLIPSE ASSOCIATES in connection with its international sales since approximately 1978. In 1984, EAL entered the United States market and has used ECLIPSE and ECLIPSE ASSOCIATES to identify its software in the AEC market. EAL does not use ECLIPSE or any form thereof to identify itself as a source of computer hardware or operating system software.

C. Course of Prior Proceedings and Dispositions

This case commenced when EAL sought a declaratory judgment, under 28 U.S.C. §§ 2201 & 2202, that its use of the word ECLIPSE did not violate any statutory or common law trademark rights of Data General. Furthermore, EAL sought to cancel or otherwise modify Data General’s U.S. Trademark Registration No. 1,107,836 .

Data General counterclaimed for trademark infringement under 15 U.S.C. § 1125 (a), unfair competition, trademark dilution under several state statutes, and common law trademark infringement. The district court found that EAL infringed on Data General’s mark and enjoined EAL from using the mark ECLIPSE.

EAL moved to suspend the injunction and to amend the judgment to permit it to use “ECLIPSE” as a trademark and “Eclipse Associates Limited” as a business name with a disclaimer of any association with Data General. The trial court denied this motion and refused to revise the injunction against the use of the ECLIPSE logo. EAL’s motion to suspend the trial court’s injunction pending appeal was also denied. EAL filed a timely Notice of Appeal.

II. LIKELIHOOD OF CONFUSION

In Levi Strauss & Co. v. Blue Bell, Inc., 778 F.2d 1352, 1355 (9th Cir.1985), an en banc panel of our court was convened to resolve the appropriate standard of review for a district court’s finding of likelihood of confusion. The court held that the “clearly erroneous standard” applied.

The en banc panel listed several reasons for this conclusion. First, the issue of likelihood of confusion is a mixed question of law and fact, which is predominately factual in nature. Levi Strauss, 778 F.2d at 1355 . (“[Wjhere a mixed question of law and fact is predominately one of fact, we review the trial court’s determination under the clearly erroneous standard.” (citing United States v. McConney, 728 F.2d 1195, 1203-04 (9th Cir.) (en banc), cert. denied, 469 U.S. 824 , 105 S.Ct. 101 , 83 L.Ed.2d 46 (1984)). Further, we reasoned that de novo review of likelihood of confusion issues would “demand a significant diversion of appellate court resources to a task which more properly belongs to the district court judge.” Levi Strauss, 778 F.2d at 1355 (citing Miss Universe, Inc. v. Flesher, 605 F.2d 1130, 1133 (9th Cir.1979)). Finally, we noted, “the limited precedential value of likelihood of confusion decisions, each of which stands upon its own facts, reduces the need for de novo review.” Levi Strauss, 778 F.2d at 1356 . See also McConney, 728 F.2d at 1201 .

EAL agrees that the district court’s determination of a likelihood of confusion is reviewed under the clearly erroneous standard. However, EAL contends that certain specific factors must be applied, as a matter of law, in making that determination. Thus, it is EAL’s position that we must review de novo the question of what factors are to be applied in making the determination. It is our legal conclusion that there is no set formulation of factors that must be applied as a matter of law.

EAL argues that since AMF Inc. v. Sleekcraft Boats, 599 F.2d 341 (9th Cir.1979), the Ninth Circuit has used two distinct “likelihood of confusion” tests: one involving inexpensive impulse purchase-type goods (a 5-factor test), the other involving expensive goods purchased with a great deal of care by sophisticated consumers (an 8-factor test). EAL contends that the district court erred in not applying the 8-factor test, since EAL’s computer products are sophisticated goods.

The Ninth Circuit has made no firm distinction between the 5-factor and 8-factor tests for likelihood of confusion in trademark infringement cases. In some cases the first test is used, in others the second is used. The difference does not rest, as EAL contends, on the distinction between inexpensive goods with a mass base of consumers and expensive goods that appeal to sophisticated consumers. Both tests have been applied to both situations. See, e.g., Alpha Indus., Inc. v. Alpha Steel Tube & Shapes, Inc., 616 F.2d 440, 445 (9th Cir.1980) (applying the 5-factor test to steel tubing, “where the purchasers are ... knowledgable [and] where the items are expensive (implying greater buyer care)”); Levi Strauss, 778 F.2d at 1360 (applying the 8-factor test to casual shirts, inexpensive goods which appeal to a mass market).

In fact, although the tests are described in slightly different terms, the content of the tests appears to have been applied interchangeably. For example, in Alpha Indus., the court used the 5-factor analysis, but it also discussed the degree of care factor, not formally present in the 5-factor test, and it cited Sleekcraft, the case that established the 8-factor test.

The Ninth Circuit enumerated likelihood of confusion tests as helpful guidelines to the district court. These tests were not meant to be requirements or hoops that a district court need jump through to make the determination. Our court has never articulated specific factors that a district court must recite and apply, instead, we have identified a non-exclusive series of factors that are helpful in making the ultimate factual determination. The Second Circuit has followed a similar approach. See Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 872 (2d Cir.1986) (specific factors are designed to help in the ultimate determination of a likelihood of confusion, not to be applied as a rigid formula).

The district court’s primary task, in determining infringement of registered trademarks, is to make the factual determination whether the public is likely to be deceived or confused by the similarity of the marks as to source, relationship or sponsorship. See Smith v. Chanel, Inc., 402 F.2d 562, 563 (9th Cir.1968). Regardless of what specific factors are discussed by the district court, we review the district court’s ultimate factual finding under the clearly erroneous standard.

In the instant case, the district court stated that it considered five elements in determining that there was a likelihood of confusion, but did not specifically mention consumer sophistication. While the district court discussed only five factors, there is no reason to believe that other factors were not implicitly considered as well. See Alpha Indus., 616 F.2d at 444 (the court used the 5-factor test, but considered customer sophistication). The district court received extensive evidence from both parties on the matter of customer sophistication. In fact, customer sophistication was the crux of EAL’s argument. Undoubtedly, this evidence was considered by the district court, even if it was not listed separately as a relevant factor.

The district court had substantial evidence from which to determine that customers, regardless of their sophistication, were likely to be confused by two computer companies using the mark ECLIPSE: the ECLIPSE mark was strong; both companies used the same mark on computer software; both companies sold to the AEC market; and EAL had prior knowledge of Data General’s use of the mark. We find no basis to rule that the district court's finding, that there was a likelihood of confusion, was clearly erroneous.

III. LACK OF EVIDENCE OF ACTUAL CONFUSION

EAL’s second argument is that the district court erred in holding that lack of evidence of actual confusion was not a significant factor in determining likelihood of confusion.

Appellant mischaracterizes the district court’s opinion. The district court held that “although the presence of actual confusion is strong evidence of the likelihood of confusion, it is not incumbent upon defendant herein to show actual confusion because of the difficulty of gathering such evidence.”

In Sleekcraft, this court held: “the failure to prove instances of actual confusion is not dispositive.” 599 F.2d at 353 . In Park ‘N Fly, Inc. v. Dollar Park and Fly, Inc., 782 F.2d 1508, 1509 (9th Cir.1986), we reiterated: “Though the court found no evidence of actual confusion, this is merely one factor to be considered ... and it is not determinative.”

The district court was well aware that proof of actual confusion is only one factor to be weighed in determining likelihood of confusion and specifically mentioned this in its list of factors to consider. Although no actual confusion was proven by Data General, the district court was not clearly erro neous in determining that “likelihood of confusion” existed.

IV. EXCLUSION OF TESTIMONY

Appellant also argued that the district court erred in excluding the testimony of various third-party users of the mark ECLIPSE. EAL attempted to introduce into evidence five deposition transcripts describing the use of the mark ECLIPSE for goods unrelated to the computer field. Evidence of other unrelated potential infringers is irrelevant to claims of trademark infringement and unfair competition under federal law. Lanham Trade Mark Act, §§ 32, 43(a), as amended, 15 U.S.C. §§ 1114 , 1125(a). See also Century 21 Real Estate Corp. v. Sandlin, 846 F.2d 1175, 1181 (9th Cir.1988); Sleekcraft, 599 F.2d at 348 ; National Lead Co. v. Wolfe, 223 F.2d 195, 203-05 (9th Cir.), cert. denied, 350 U.S. 883 , 76 S.Ct. 135 , 100 L.Ed. 778 (1955). The fact that a floor cleaning product, commercial laundry folding equipment and industrial process heating equipment bear the name ECLIPSE is irrelevant to the case at bar and was correctly excluded.

V. CONCLUSION

The district court’s finding, that a likelihood of confusion existed, was not clearly erroneous. The district court properly recognized that lack of evidence of actual confusion, though relevant, is not essential to prove likelihood of confusion. Finally, the court did not err by excluding evidence of unrelated third-party users of the mark ECLIPSE. We affirm the district court.

AFFIRMED.

. A panel of our court inadvertently later misstated the standard of review in Miss World (UK) Ltd. v. Mrs. America Pageants, Inc., 856 F.2d 1445, 1448 (9th Cir.1988), overlooking the resolution of that issue by the en banc panel in Levi Strauss. It is apparent that this was simply an oversight. . In the instant case, the district court used the following 5-factor test to determine if likelihood of confusion existed: 1. strength of the mark; 2. similarity of the mark; 3. class of goods and marketing channels; 4. evidence of actual confusion; and 5. intent of second user. J.B. Williams Co., Inc. v. Le Conte Cosmetics Inc., 523 F.2d 187, 191 (9th Cir.1975), cert. denied, 424 U.S. 913 , 96 S.Ct. 1110 , 47 L.Ed.2d 317 (1976). . The Sleekcraft 8-factor test lists the following factors as relevant in determining likelihood of confusion: 1. strength of the mark; 2. proximity of the goods; 3. similarity of the marks; 4. evidence of actual confusion; 5. marketing channels used; 6. type of goods and the degree of care likely to be exercised by the purchaser; 7. defendant’s intent in selecting the mark; and 8.likelihood of expansion of the product lines. Sleekcraft, 599 F.2d at 348-49 . .The Ninth Circuit test for likelihood of confusion has been described as an eight factor test in Sleekcraft, a six factor test see J.B. Williams Co. v. Le Conte Cosmetics, Inc., 523 F.2d 187, 191 (9th Cir.1975), cert. denied, 424 U.S. 913 , 96 S.Ct. 1110 , 47 L.Ed.2d 317 (1976), and Century 21 Real Estate Corp. v. Sandlin, 846 F.2d 1175, 1179 (9th Cir.1988), and a five factor test, see Rodeo Collection, Ltd. v. West Seventh, 812 F.2d 1215, 1217 (9th Cir.1987). Accuride Intern, Inc. v. Accuride Corp., 871 F.2d 1531 , 1533-34 n. 1 (9th Cir.1989). There is no indication that the cases applying five or six factors were intended to undermine Sleekcraft, and eliminate consideration of the two or three Sleekcraft factors excluded in those cases. J.B. Williams explicitly states that the six factors in this list are not meant to be exclusive. 523 F.2d at 191 . . Data General contends that customer sophistication is implicitly considered in the 5-factor test. This is a strong argument, especially since J.B. Williams, the case that cites the 5-factor test, considered the degree of care that customers used when purchasing items, even though it did not list degree-of-care (a customer sophistication) in its list of relevant factors. See 523 F.2d at 193 .