553 F2d 740 Digitronics Corp v. New York Racing Association Inc

553 F.2d 740

193 U.S.P.Q. 577

DIGITRONICS CORP., now Amperex Electronic Corp.,
Plaintiff-Appellant-Cross- Appellee,
v.
The NEW YORK RACING ASSOCIATION, INC., et al.,
Defendants-Appellees-Cross- Appellants.

Nos. 338, 339, Dockets 76-7063, 76-7085.

United States Court of Appeals,
Second Circuit.

Argued Jan. 3, 1977.
Decided April 4, 1977.

S. C. Yuter, New York City (Daniel M. Rosen, Wallace E. J. Collins and Yuter & Rosen, New York City, of counsel), for plaintiff-appellant-cross-appellee.

Robert E. Isner, New York City (Charles N. J. Ruggiero, Arthur A. March and Nims, Howes, Collison & Isner, New York City, of counsel), for defendants-appellees-cross-appellants.

Charles W. Sims, Dayton, Ohio, and Francis N. Carten, Stamford, Conn., on brief for Digital Systems Corp., amicus curiae.

Before MOORE, OAKES and TIMBERS, Circuit Judges.

OAKES, Circuit Judge:

view counter
1

This appeal involves a patent issued for a data processing system that, when used in place of the prior art in parimutuel wagering equipment, is said to produce speedier, more accurate and more versatile paraphernalia for the placing of wagers at race tracks by those interested in the ancient and honorable art of improvement of the breed. The United States District Court for the Eastern District of New York, John F. Dooling, Jr., Judge, in a 216-page opinion with 52 formal findings and conclusions, supplemented by explanatory appendices, held the patent invalid for obviousness under 35 U.S.C. § 103.1 He also held that appellees' counterclaim for attorney fees could not be sustained, since the case was not sufficiently "exceptional" to permit the discretionary award of attorney fees under 35 U.S.C. § 285. Part of the extensive opinion has been published at 187 U.S.P.Q. 602 (E.D.N.Y.1975). On both the appeal from the judgment of invalidity and the cross-appeal from the denial of the counterclaim for attorney fees, we affirm the judgment.

I.

2

This action was brought by appellant's predecessor, Digitronics Corp., which sought an injunction and damages from the New York Racing Association, Inc., Automatic Totalisators (U.S.A.) Ltd. (ATUSA), Automatic Totalisators Ltd. (Amtote), and Premier Equipment Proprietary Ltd., for alleged infringement of U. S. Patent No. 3,252,149, issued on May 17, 1966, to Digitronics as assignee of five named inventors. The invention, according to papers filed with the Patent Office, "pertains to data processing systems and more particularly to systems for processing data received from ticket issuing machines," one of the most common of which is "a parimutuel system employed for servicing transactions or wagers made by spectators at sporting events." The patent contains 33 claims, but only claims 20-27 are in issue here. A considerable portion of Judge Dooling's opinion is devoted to developing for the reader a reasonably simplified, albeit unavoidably difficult to follow, primer of knowledge regarding the data processing circuitry and "logic" here involved, including, inter alia, explanations of the meaning and operation of the components used.2 The patent itself frequently refers to R. K. Richards, Arthmetic Operations in Digital Computers (1955), which discloses many of the fundamentals of computer logic applicable in the field of data processing generally.

3

The patent is entitled a patent on a "Data Processing System," and each of the claims in question is one for a "system comprising" a plurality of ticket issuing machines (TIMs) or a TIM followed by a combination of particular means. As Judge Dooling found, the system is primarily an aggregator system, aggregating the number of wagers placed on each entry in two separate aggregators: the central memory register, which aggregates the wagers on each horse in each of the pools, and the TIM aggregator, where the number of wagers at that particular TIM on each horse in each of the pools is similarly (and simultaneously) aggregated. These are updated by a "unit adder" every time an additional wager is registered by any TIM. The computer draws from the central memory register the total wagers in each of the three pools and on each horse in the pool and uses those sums to compute the odds on each entry and to feed to the output display boards the results of the odds changes every 70 seconds. While elaborate circuitry is required to accomplish this, the circuitry itself is unimportant for our purposes; Judge Dooling's finding, that "there is no novel circuitry involved as is clear from the face of the patent and from the trial evidence," is not disputed here.

4

Appellant characterized below the subject of Claims 20 through 22 as the "totalisator system" invention, Claim 23 as the "nonallowed runner subsystem" invention, Claims 24 and 25 as the "TIM scanning subsystem" invention, and Claims 26 and 27 as the "erroneous data subsystem" invention. Claim 20 consists of the combination of a plurality of conventional TIMs, means of collecting from them and dispatching into the system the betting data on each entry and identifying those data to the original TIM, and means of aggregating the wagering data and sending an acknowledgment signal to the TIM so that it will stamp up and issue a ticket to the bettor with the wagering data on it. The system is composed of a set of commercially available, fairly standardized TIMs, an electronic linkage between them, an aggregator, a unit adder, and an "acknowledgment signal amplifier" between the unit adder and the TIM so that aggregating the wager effects an electromechanical release of the betting ticket. The purpose of the system is to aggregate wagers without losing track of the horse on which and the pool in which the wager was placed or the TIM at which the wager was placed. Judge Dooling found that each component was common in the prior art, existing in various forms, and performed its familiar role in a familiar way, all as basically described in Richards, Arthmetic Operations in Digital Computers, supra.

5

As to Claim 21, the court found that it added nothing to the combination, merely particularizing Claim 20, 187 U.S.P.Q. at 624. Claim 22, the court below found, like Claim 21, furnished particulars of Claim 20, adding the circumstance that the data passes from the "storage address generator" so as also to register simultaneously in the TIM memory. Put another way, the same data was being accumulated in the memory of each TIM as to that machine as well as in the central memory register. Judge Dooling found that Claim 22 really disclosed that the TIM memory and its immediately associated components are simply a conventional memory and unit adder used as a simple aggregator.

6

Claim 23, the "scratch" subsystem, is independent of Claim 20, and prevents betting on a "scratched" or withdrawn horse by using "a conventional equality comparator" to generate a "scratch signal" in the TIM and to prevent the TIM from operating to produce a ticket, "stepping" the "scan" to the next TIM. Put another way, a switch device emits a set of steady scratch signals and a comparison device matches those scratch signals with signals for attempted wagers on scratched entries, resulting in a rejection signal to the TIM. Judge Dooling found this not to go beyond "the obvious teaching of the use of a comparator" as a guard against taking wagers on scratches. As he said, a "comparator" exists to compare signals for identity or differences in a variety of ways, some of which can be gleaned from British Patent 749,836, cited by the Patent Office below.

7

Claim 24 was outlined by the patentee with reference to TIM scanning when no wagering transaction was being conducted at the TIM. Claim 24 and its dependent claim, Claim 25 (involving the converse situation where a transaction was being conducted or sought to be conducted) are the only ones of the claims in issue in which the plurality of TIMs is particularly significant to the claim. The object of the system is to provide a high-speed scanning means for interrogating a plurality of relatively slow operating TIMs. According to the patent itself, even absent a finding by the court below, the device which does the scanning the "scan counter" is a "chain of conventional cascaded binary counters," as discussed in Richards' Arthmetic Operations in Digital Computers, supra. The output of the scan counter is fed to the "scanner," which decodes the data from the scan counter and feeds it to the "interface" of the TIM. When depressed, the keys of the TIMs close switches, and when this is not done the scanner "steps" along to the next TIM or, if a key for a scratched or other wrong entry has been depressed, "the scratch circuitry" previously outlined (Claim 23) "steps" the scan to the next TIM. But when a wager has been processed and error-tested the depressed wager key is unlatched, and the bet ticket is printed and issued to the bettor. Judge Dooling found that the subsystem claim combined conventional components to perform a familiar assignment in a conventional way. Both the binary counter and the scanner themselves being conventional in data processing, the circuitry he found to be "pedestrian," with a very simple goal of furnishing a pulse tracking through a system and on completion to origin ticking a binary counter.

view counter
8

Dependent Claim 25, Judge Dooling found, added nothing, simply covering the alternative that a wager had been made. In that case, as we have said, there is no "step," and the "selected entry transaction" signal is simply processed. The claim covers delaying the stepping of the scan over the transacting TIM to the next TIM until the selected transaction signal is checked out or confirmed and permitted to proceed. Judge Dooling found this to be "detailed and uninspired circuitry doing the routine routinely," "not even approaching the versatility of (a prior patent known as) Schrimpf No. 3029414."

9

Claims 26 and 27 parallel Claim 23 in that their purpose is to constitute error checks incorporated in the total device. Judge Dooling found both claims to consist of "painstaking articulation of familiar means to perform simple error checks through use of circuitry suggested by the nature of the task to be performed and the nature of the conventionally appropriate components that the tasks themselves pointed out."

10

The findings as to these seven claims formed the basis for Judge Dooling's decision that the patent was obvious. We, of course, may not disturb those findings unless we find them to be "clearly erroneous." Fed.R.Civ.P. 52(a). With regard to obviousness, the statute establishes a legal standard relating to whether

11

the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains.

12

35 U.S.C. § 103. We are to apply this statute by first assessing the scope of the prior art and then determining whether an ordinarily skilled worker conversant with it would think appellant's claims obvious in light of it. See Graham v. John Deere Co., 383 U.S. 1, 12-17, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966); Hotchkiss v. Greenwood, 52 U.S. (11 How.) 248, 267, 13 L.Ed. 683 (1851).

II.

13

Appellant's major argument on appeal is that Judge Dooling erroneously considered the prior art in data processing generally, rather than limiting his consideration to the totalisator business. It purports to find support for this theory in Dann v. Johnston, 425 U.S. 219, 96 S.Ct. 1393, 47 L.Ed.2d 692 (1976), which held invalid for obviousness a patent claim involving an application of data processing in the banking industry. From the discussion in Dann of practices in that industry, appellant deduces that banking, rather than data processing generally, was held to be the relevant prior art. But central to the Dann Court's conclusion was one "Dirks" patent, which made a use of data processing analogous to that of the patent there in suit but was intended for use in any "large business organization," 425 U.S. at 228, 96 S.Ct. 1393, 1398, not just banking. Because of the existence of the "Dirks" patent, "(t)here (was) no need to make the obviousness determination in (Dann) turn solely on the nature of the current use of data processing . . . in the banking industry." Id. As Mr. Justice Marshall concluded:While computer technology is an exploding one, "(i)t is but an evenhanded application to require that those persons granted the benefit of a patent monopoly be charged with an awareness" of that technology.

14

Id. at 229, 96 S.Ct. at 1399, quoting Graham v. John Deere Co., supra, 383 U.S. at 19, 86 S.Ct. 684. Thus the relevant prior art in Dann was held to be that of computer technology generally rather than the art of the banking industry.

15

Since the analogy of this case to Dann is so close, we might not be amiss to decide the question of relevant prior art simply by reference to that case. But in the absence of an explicit statement on this point in Dann, we think it preferable to articulate the rule of law guiding our decision here. In determining the scope of the relevant prior art with which the hypothetical ordinarily skilled person must be presumed to be familiar, we hold simply that the court must look, in light of both the training of the patentee and the elements in the claimed invention which give it its novel quality, at what arts the patentee could reasonably be expected to consult in doing the inventing. See In re Ellis, 476 F.2d 1370 (Cust. & Pat.App.1973) (claimed invention of floor gratings; prior art included shoe scrapers); In re Antle, 444 F.2d 1168, 1171, 58 CCPA 1382, (1971) (claimed invention for mobile produce packing; prior art included produce preserving and plastic film wrapping); Metallurgical International, Inc. v. Kawecki Berylco Industries, Inc., 348 F.Supp. 825, 835 (E.D.Pa.1972) (claimed invention for pneumatically pulverizing material; prior art included sandblasting).

16

Here, as Judge Dooling found, the inventors were trained in data processing, not merely in totalisators. They worked for a company that applied techniques of solid state electronic data processing to any industry which would hire them to do so. The patent was granted as a patent on a data processing system. And the elements of the invention for which novelty is claimed relate to solid state electronic data processing generally, not merely to totalisators. Thus the scope of the prior art in this case, in which the hypothetical reasonable person must be ordinarily skilled, and hence which the inventors here could reasonably be expected to have consulted, encompasses data processing generally and is not restricted to the totalisator business.3

III.

17

Once the art in which the ordinarily skilled person must be said to have knowledge is widened to include not only totalisators, but all of data processing, it is manifest that the patent claims here are obvious. They do not perform functions that differ from prior art in the totalisator business; they achieve those functions through means that, while new to the totalisator business, were well established in the wider data processing field; and the improved performance that resulted when solid-state electronic means were applied to perform the totalisator function, far from being unusual, was just what could have been predicted when those means were applied to perform that function.

A. The Totalisator Function

18

Parimutuel betting is, of course, a system of wagering for multi-entrant events, involving the accumulation of wagers in separate pools with betting odds and consequent payoffs being determined by the comparative totals in such pools after deducting therefrom a predetermined percentage based upon the state's and the establishment's statutory shares of the pools. Parimutuel machines replaced bookmakers through the application of technology. ATUSA, an Australian concern, developed a practical mechanical and electromechanical totalisator in 1917, a multiple pool "tote" in 1923, and an automatic odds tote in 1927, installing its first system at Hialeah, Florida, in 1932, a system later sold to the group that formed Amtote. Amtote installed its first totalisator with a public display of betting figures accurate to the dollar in 1933 at Arlington Park in Chicago, and by 1962 it had almost monopolized the American totalisator business, although ATUSA had a few installations in the United States.

19

The totalisators or "totes" automatically permit the issuance of tickets to the bettors at the location of ticket issuing machines; aggregators accumulate the amounts wagered on each race and on each entry in a race for each of the available possible betting combinations (win, place or show, daily double, quinella, etc.); a scanner connects the ticket issuing machines to the aggregators; electrically operable display boards at various locations display entry, odds and payoff information, time of day, time of race, fractional times in the race and the like; checks are made for purposes of accuracy of the displayed information; a calculator determines the probable odds and payoffs; and all components are interconnected so as to provide for information transfer between the components and periodic updating and display of necessary information.

20

As of 1961, the time when the alleged invention here in suit began to take shape, the state of the totalisator art was epitomized by the Amtote Model 7J, in use both at Aqueduct and Roosevelt Raceway. The Model 7J involved the various functions mentioned above: its ticket issuing machines, or TIMs, were of standardized construction and generally manufactured by an independent concern; its aggregators and scanner were electromechanical in operation and the display boards electrically operable; an analog computer was used for calculating probable odds, and a solid state electronic digital price computer for calculating payoffs. The Model 7J included provisions for automatically rejecting wagers on scratched horses and for checking input data for other errors, as well as incorporation of individual TIM devices for recording the bets made on each entry in each race so as to keep a permanent record thereof.

21

The principal prior totalisator art consisted of a series of patents acquired by Amtote on (a) a system including a stepping switch collector or scanner operating only on signal from the TIM (Levy, No. 2,182,875 (1939)); (b) an electromechanical system featuring a "jumping jack" collector or scanner (Levy et al., No. 2,179,698 (1939)); (c) an electromechanical and electronic tote using a vacuum tube non-binary counter and pulse commutators, with stepping "relay" aggregators (Klein, No. 2,557,384 (1951)); (d) an electromechanical tote system with multichannel adding machines (Johnston, No. 2,563,041 (1951)); (e) an analog odds computer for use as a component of a tote system (No. 2,652,977); and (f) the Lange tote system (No. 3,051,384 (Aug. 28, 1962)) and its improvement (No. 3,080,114). The prior art also included a patent for an electromechanical totalisator issued to Handley (No. 2,479,681 (1949)), and its improvement (No. 2,680,561 (1954)), which was also an electromechanical patent involving aggregation of TIM data, means of picking up the wagers or sales or the "handle" of the machines, adding them by classes, performing other calculations on them, and acknowledging or confirming to the ticket issuing machine that the calculation had been made. A switch device permitted the emission of a set of steady scratch signals, and a conventional comparator matched scratch signals with signals for attempted wagers on scratched entries, the signal circuitry effectuating a guard against taking wagers on scratches and other error checks. It is evident from the foregoing description that the patent claims here in suit are meant to perform the same function as this plethora of prior electromechanical data processing art.4B. The Solid State Electronic Means

22

Appellant's contribution to the totalisator business involved simply the upgrading of a well-defined, existing data processing system by converting some components of the tote system from electromechanical to solid state electronic data processing. Specifically, Digitronics' totalisator, the First Tote, differed from prior totalisators in using electronic solid state aggregators for accumulating wagers and total amounts, an electronic scanner for sequentially connecting the ticket issuing machines to the aggregators, digital solid state computing devices for calculating probable odds and payoffs, and an electronic solid state memory for recording the bets made on each TIM.

23

While these devices were new to totalisators, they were not new to data processing. The court below found the testimony of appellees' expert witness, Highleyman, persuasive in this regard. Judge Dooling also mentioned six United States patents on these subjects; certain pages from a British patent on Univac; three publications (Functional Description of the EDVAC, The System Organization of MOBIDIC, and Project Mercury Real-Time Computational and Data Flow System); and the standard text, Richards, Digital Computer Components and Circuits (1957). These together describe all of the solid state electronic means used by appellant. 187 U.S.P.Q. at 620-22. Further, Judge Dooling quoted the statement of Digitronics' chief executive officer to the effect that "there is virtually nothing in our device which has not already been incorporated in other machines which we have installed for other purposes in such companies as . . .," naming a railway and three large industrial concerns. The judge specifically found that Digitronics had previously supplied, for example, a medical access and analysis system for the Schering Drug Company "which involved the electronic scanning of a plurality of lever operable data input stations, the collection and storage of data obtained therefrom on magnetic tape units and the return of selectively generated signals to the data input stations." Thus we have here the routine seeking of business by a qualified data processing "system" house, followed by the routine application of then-current state of the art data processing technology to the updating of an existing data processing system of known and defined functional characteristics.

C. The Predictable Result

24

If the application of obvious means to an obvious function has " synergistic" results that would not be obvious to one of ordinary skill in the art, a patent still may issue. Sakraida v. Ag Pro, Inc., 425 U.S. 273, 282, 96 S.Ct. 1532, 47 L.Ed.2d 784 (1976); Anderson's-Black Rock, Inc. v. Pavement Salvage Co., 396 U.S. 57, 61, 90 S.Ct. 305, 24 L.Ed.2d 258 (1969); Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., 340 U.S. 147, 152, 71 S.Ct. 127, 95 L.Ed. 162 (1950); Roanwell Corp. v. Plantronics, Inc.,429 U.S. 1004, 97 S.Ct. 538, 50 L.Ed.2d 617 (1976) (White, J., joined by Brennan, J., dissenting from denial of certiorari). Here, however, the benefits that accrued from the replacement of electromechanical by solid state electronic means increased accuracy, speed, compactness, flexibility, reliability, and economy were, as Judge Dooling found, nothing more than the recognized advantages of electronic upgrading of a data processing system. Once the prior art is expanded to include all of data processing, appellant's only claim of "synergistic" results is based on its "multi-function TIM number signal generation means" (Claim 20b). But the court specifically found that this claim was nothing more than semantics, because in electronic data processing any system is energized from a common source and ordered in real time by a common step signal. Thus, once two different memories are used, here one central and one at each TIM, not to have simultaneous storage in the central memory register and in the TIM memory would require "a special ingenuity of indirection." 187 U.S.P.Q. at 632. Based as this finding is on expert testimony and voluminous prior art references, we cannot hold it clearly erroneous.

25

Mr. Leonard of Digitronics, Judge Dooling found, had "no doubts or reservations" as to Digitronics' ability to provide an electronic data processing system to perform the totalisator function. There was good reason for his confidence. As Judge Dooling put it:

26

The demands of a racetrack parimutuel system on data processing resources are modest: the arithmetic calculations required are rudimentary; the range of input data is narrow, and the data are simple and easily translated into binary terms; no elaborated long-term memory is requisite; programming is simple and direct, and much of it is reducible to permanent wiring of system components; pre-existing electromechanical circuitry readily supplies patterns which are in considerable part directly adaptable to solid state electronic data processing; the service demands on data processing systems of a racetrack parimutuel system present no identifiable difference of technical significance from those of other multiple input situations.

27

187 U.S.P.Q. at 643. This finding seems compelled by the evidence. Having found that the function, means, and results of appellant's claims are fully anticipated by the prior art, we hold that they would have been obvious to one ordinarily skilled in data processing at the time the invention at issue was made.

D. Secondary Considerations

28

Perhaps foreseeing our agreement with Judge Dooling's conclusion that appellant's claims are obvious in light of the prior art, appellant argues strenuously that the true test for obviousness is not the court's evaluation of the patentee's inventiveness, but rather such objective, often called "secondary," considerations as long-felt need, immediate copying, and commercial success. Despite the persuasive advocacy of Judge Learned Hand, Reiner v. I. Leon Co., 285 F.2d 501 (2d Cir. 1960); Lyon v. Bausch & Lomb Optical Co., 224 F.2d 530 (2d Cir. 1955), this view does not represent current law. Any theory that "secondary" considerations must be given weight before a determination of obviousness can be made was laid to rest in Sakraida v. Ag Pro, Inc., supra, 425 U.S. at 282-83, 96 S.Ct. at 1537, where Mr. Justice Brennan concluded for a unanimous Court:

29

Though doubtless a matter of great convenience, producing a desired result in a cheaper and faster way, and enjoying commercial success, (the invention) "did not produce a 'new or different function' . . . within the test of validity of combination patents." Anderson's-Black Rock v. Pavement Co., supra, at 60, (90 S.Ct. 305, 308). These desirable benefits "without invention will not make patentability." Great A & P Tea Co. v. Supermarket Corp., 340 U.S., at 153, 71 S.Ct. 127, 130.

30

Accord, Timely Products Corp. v. Arron, 523 F.2d 288, 295 (2d Cir. 1975); Julie Research Laboratories, Inc. v. Guildline Instruments, Inc., 501 F.2d 1131, 1135 (2d Cir. 1974); Formal Fashions, Inc. v. Braiman Bows, Inc., 369 F.2d 536, 539 (2d Cir. 1966). See also Maclaren v. B-I-W Group, Inc., 535 F.2d 1367, 1376 (2d Cir.) (citing Julie Research and Formal Fashions for proposition that objective criteria "are of distinctly secondary importance"), cert. denied, 429 U.S. 1001, 97 S.Ct. 531, 50 L.Ed.2d 612 (1976). Only in a close case, in which application of the subjective criteria of nonobviousness in 35 U.S.C. § 103 does not produce a firm conclusion, can these objective or secondary considerations be used to " 'tip the scales in favor of patentability' ". Roanwell Corp. v. Plantronics, Inc., supra, 429 U.S. at 1008, 97 S.Ct. at 541 (White, J., joined by Brennan, J., dissenting from denial of certiorari),5 quoting Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 321 U.S. 275, 279, 64 S.Ct. 593, 88 L.Ed. 721 (1944). Because we hold that the claims made here are clearly obvious, we need not examine secondary considerations. Compare U. S. Philips Corp. v. National Micronetics Inc., 550 F.2d 716 (2d Cir. 1977) (secondary considerations important because claim not clearly obvious).

31

Moreover, these secondary considerations are markedly less accurate as guides to nonobviousness in this case than might be the situation in another case. This is due to the domination of the totalisator business by one large company, Amtote. Thus any "long-felt need" for the improvement to which the Digitronics patent related, converting electromechanically operated totalisators into electronically operated ones, and its immediate use and commercial success do not necessarily mean that no one else could have made a similar improvement. On the contrary, it very likely was simply a reflection of the fact that the dominant company had no incentive to make the change because its former system, exemplified by the Model 7J, was good enough to dominate the market, and no one else was willing to make the needed investment further to upgrade its operation, in view of that domination. Once the improvement was introduced into the market, the dominant company, Amtote, was quick to introduce electronic tote to maintain its dominant market position.

IV.

32

Appellees counterclaimed for attorney fees under 35 U.S.C. § 285, which permits a discretionary award in "exceptional" cases, as where there is fraud in obtaining allowance of the patent by the Patent Office or other significant misconduct. See Timely Products Corp. v. Arron, supra, 523 F.2d at 305; Kramer v. Duralite Co., 514 F.2d 1076 (2d Cir.) (per curiam), cert. denied, 423 U.S. 927, 96 S.Ct. 275, 46 L.Ed.2d 255 (1975); Kahn v. Dynamics Corp. of America, 508 F.2d 939, 945 (2d Cir. 1974) ("unclean hands" render case exceptional), cert. denied, 421 U.S. 930, 95 S.Ct. 1657, 44 L.Ed.2d 88 (1975). Appellees argue that Digitronics at least had unclean hands, since in the Patent Office proceeding it knew of but did not refer to the existence of the demonstrator prototype of the First Tote, see note 4 supra, or to its demonstration and attempted commercial exploitation.

33

But the court found as fact that there was no "conscious" participation by patentees and their draftsman, who was himself not told about the prototype and who drafted the application hurriedly (between January and March 28, 1963) in view of an intended sale or license of the patent. At another point the judge stated: "(I)t cannot be said that the case was pursued in bad faith, or was so wholly devoid of substance that plaintiff could not fairly be supposed to be proceeding in good faith." Although the court alluded rather darkly to the lack of "any satisfactory and reasonable explanation" for the disappearance of the prototype drawings and to the fact that they were "uniquely missing," reflecting "a discontinuity in the files," the court went on to conclude as a matter of law that "(i)ssuance of the patent was not procured through fraud or concealment or culpable nondisclosure," and hence that the case was not "exceptional" within 35 U.S.C. § 285. The weight of the evidence supports this conclusion.

34

Judgment affirmed.

1

Judge Dooling also found the patent invalid under 35 U.S.C. § 102 for lack of novelty and held that certain data processing systems of the appellees did not infringe the patent. In view of our agreement with the judge that the patent is obvious, we need not reach these issues

2

The judge, for example, explained that one such component, a "memory" core, contains a very large number of tiny "toroids", each of which has the capacity to represent a one or a zero (a binary number system) and in turn represents part of the numerically encoded information that is in "storage" or is being brought out of "storage." He referred to the "address" of the information in storage in the "memory" and discussed how each toroid is individually accessible to wire-conveyed impulses that impress upon it a clockwise or counterclockwise magnetic field and give it thereby a significance of one or zero. To read out the meaning, a current is applied that alters the polarity of magnetization if the toroid is in one state but not if it is in the other state, thus extracting the "bit" of information from the toroid. The judge explicated also how sections of the memory may store instructions, duly coded in binary numerical form, to govern the steps in the functioning of the device, so that the machine literally operates upon itself. And he explained how each toroid is wired with two "impulse" wires and one "inhibit" wire, the last to prevent the loss of information stored upon "reading" the "bit."

3

Our holding on this point disposes of appellant's subsidiary claim that the court below erred in not allowing it to examine the witness Fosse on the level of ordinary skill in the totalisator art. Since the applicable art was far broader than the totalisator business, the court was well within its discretion in barring the requested examination

4

Judge Dooling also held that a prototype that Digitronics constructed as a testing and advertising device, before producing the fully operable "First Tote," its all-electronic totalisator, had to be considered part of the prior art. Whether this is so depends on whether the prototype was "commercially marketable" under Timely Products Corp. v. Arron, 523 F.2d 288, 302 (2d Cir. 1975). We need not decide this question, because we conclude that here the prior art apart from the prototype indicates performance of the same functions

5

In Roanwell, the Supreme Court was considering certiorari in a Second Circuit case. Plantronics, Inc. v. Roanwell Corp., 403 F.Supp. 138 (S.D.N.Y.1975), aff'd, 535 F.2d 1397 (2d Cir. 1976) (per curiam). The panel decision there is not inconsistent with the above analysis; the court said: "In view of the ample evidence of obviousness, plaintiff's arguments concerning secondary factors are not persuasive." 535 F.2d at 1398. It was the district court opinion, which relied on secondary considerations, with which Justices White and Brennan took issue in their dissent from denial of certiorari