Opinion
June 5, 1957.
Shereff Brothers, New York City, for plaintiff, Harry Shereff and Louis H. Shereff, New York City, of counsel.
Levisohn, Niner Cohen, New York City, for defendant, Harry Cohen, Edwin Levisohn, Reuben Spencer, New York City, of counsel.
This is an action for alleged infringement of two patents relating to variable focal lens systems for motion picture and television cameras. These patents are No. 2,454,686 (hereinafter called 686), issued November 23, 1948, and No. 2,718,817 (hereinafter called 817), issued September 27, 1955.
Plaintiff, a New York corporation manufacturing and selling varifocal lenses, alleges that defendant, by selling certain lens systems called Pan-Cinor 70 and Pan-Cinor 100, is infringing the above patents owned by plaintiff. The alleged infringing lens systems are manufactured in France by Societe d'Optique et de Mecanique de Haute Precision and are sold in this country by the defendant, a New York corporation. A United States Patent, No. 2,778,272 covering the alleged infringing devices was issued on January 22, 1957.
The defendant's answer, inter alia, denied infringement and alleged the invalidity of both patents. In addition, the answer contained two counterclaims; one, for a declaratory judgment that the patents in suit are invalid and not infringed, and two, for damages based upon a claim of unfair competition.
A varifocal lens is an optical device composed of lenses which are movable in relation to each other and are operable to vary the size of the image of an object without substantial displacement or defocusing of the image from a predetermined focal plane. In the fields of motion pictures and television, a varifocal lens makes it possible for the cameraman to produce a "zoom" effect, i.e., a continuous enlargement of the image without moving the camera from a fixed position, thereby obviating the use of a dolly or traveling camera support.
The basic principle of varifocal lenses is the physical law that if the distance between two optical elements is changed the combined power and equivalent focal length of these two elements necessarily also changes — producing magnification or demagnification of the object. The lenses, however, not only change their combined focal length, but they also change the position of their combined focal plane in relation to any fixed point of reference. To make such a lens combination usable, therefore, it is necessary to compensate in some way or another for this movement of the focal or image plane.
Prior to the plaintiff's first device, purportedly made in accordance with patent 686, the varifocal lenses utilized in the motion picture field were of the so-called mechanically compensated type. Compensation was achieved in those devices by coordinated or interrelated movement of various cams, gears or levers which displaced the lens elements with respect to each other and also moved the system as a whole to keep the image plane stationary and coinciding with the film plane of the motion picture camera. Mechanical compensation permitted exact focusing throughout the range of the "zoom" operation. On the other hand, due to the need for highly precise and complicated construction, these devices were extremely costly and the mechanical linkages tended to wear out easily, throwing the image out of focus. For these reasons, the mechanically compensated devices did not enjoy a great deal of commercial success although some are still on the market.
The devices involved in the present suit effectuate compensation by purely optical means. They avoid the constructional difficulties by sacrificing, to some extent, the superior optical qualities of the mechanically compensated devices. In the case of devices purportedly made according to plaintiff's patents, the image is in exact focus three times during the zooming operation. In the defendant's devices, exact focus is attained four times during the range of operation. The deviations throughout the remainder of the zoom, however, are within accepted optical tolerances in each case, although concededly superior in the alleged infringing devices.
With regard to plaintiff's first patent application (686), the only prior art cited to the Patent Office were mechanically compensated varifocal lenses for motion picture camera use. The plaintiff successfully contended there that its device was a sufficiently novel and meritorious advance over the mechanically compensated prior art. At trial, plaintiff adequately distinguished its means for effecting compensation optically from the mechanical means, and the defendant did not seriously contest that the Patent Office erred in deciding that this was a valid distinction.
The defendant's position with regard to the validity of 686 rests basically on the following grounds: that the applicable prior art covered more than varifocal lenses for motion picture cameras; that optically compensated varifocal lenses for other uses had been known to the art before 686 but were not considered by the Patent Office; that 686 does not represent an inventive advance over that prior art; and that even if the device allegedly constructed according to patent 686 would represent an inventive advance, the patent itself is nevertheless invalid for failure to sufficiently disclose the invention, for describing it in solely functional terms and for making its claims excessively broad.
Specifically, defendant relies upon two U.S. patents: No. 2,078,586 issued to R. Richter on April 27, 1937 for a Device for Illuminating Microscopic Objects; and No. 2,209,532 issued to K. Michel on July 30, 1940 for a Microscope.
In the Richter patent, the inventor did not intend to make a zoom lens for use in taking motion pictures but rather to create a device which would provide for continuous variation in the size of the image of a source of light, to permit a continuous changing of the numerical aperture of the microscope. He was faced with the same difficulty encountered in the other variable focal instruments in that as the image would be magnified it would tend to leave the desired focal plane. To correct this problem, he employed optical, rather than mechanical compensation.
The patent indicates that Richter considered whether optical compensation would prove useful for other purposes such as the construction of varifocal telescopes. Recognizing that optical compensation would not be exact throughout the zoom range and believing that such "exact images are desired" in telescopes, he rejected the possibility.
The idea of utilizing optical compensation in other varifocal devices, however, was not rejected by Richter's co-worker, Michel. In the Michel patent for a microscope which would permit near simultaneous photography and subjective examination of the microscopic image, the inventor disclosed that "A special improvement can be arrived at * * * by using as a projection system a pancratic" [i.e. varifocal] "optical system which permits a continuous change in the magnification of the image in the image plane of the imaging receiving device without any change in the acuity of the image."
Prior to this patent it was believed that in order to get an enlarged microphotograph, it would be necessary to move the camera back. (Tr. p. 196). Michel's patent envisioned a device for the taking of still photographs at the various magnifications of the image, without moving the camera. It too, however, did not envision the taking of motion pictures.
This then is the first issue, whether the utilization of an optically compensated varifocal lens system for motion pictures represented an inventive advance over the Richter and Michel patents. The discovery of a new use for a known process "may be patented provided the conditions for patentability are satisfied." See Revisor's Note to 35 U.S.C. § 101. The Patent Act of 1952 sets forth the guiding standard in § 103.
"A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between 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. Patentability shall not be negatived by the manner in which the invention was made."
Until 1952 there was no statutory test of patentability, other than the old requirement that the subject matter be "new and useful." 1 Stat. 318 (1793). Since 1850, however, the courts have held that the subject matter must exhibit "invention" as well. Hotchkiss v. Greenwood, 1850, 11 How. 248, 52 U.S. 248, 13 L.Ed. 683. The definitions of and embellishments on the term "invention" have been plentiful and diverse. The Reviser's Note to section 103 of the Patent Act indicates that Congress adopted a statutory standard in order to bring some measure of stability to this rather confused area. The Note further indicates that the standard enunciated was not intended as a final statement, but was to serve as a basis for the addition of criteria to be worked out at a later time. Until such criteria are set forth by Congress, however, the position taken by this Circuit in Lyon v. Bauseli Lomb Optitical Co., 2 Cir., 1955, 224 F.2d 530, would appear to be the proper approach.
To determine this question, therefore, we must initially inquire who is the "person having ordinary skill in the art to which said subject matter pertains." Plaintiff maintained throughout the trial that the test in this field should not be the ordinary optical worker, but rather, the ordinary skilled lens designer. This position is, of course, correct. See American Stainless Steel Co. v. Ludlum Steel Co., 2 Cir., 1923, 290 F. 103.
The issue is, therefore, whether a skilled lens designer with knowledge of the disclosures of both the Richter and Michel patents would have perceived the utility of adopting those disclosures to the field of motion picture cameras before the inventor, Dr. Back, filed the application for his patent and whether that hypothetical lens designer would have been able to make the necessary adjustments rendering such an adaptation practical.
There is little doubt that ordinary skilled lens designers with knowledge of both of these patents could have recognized the possible application of their teachings to a varifocal lens system for motion picture purposes. Plaintiff's own expert witness, Kaprelian, a lens designer, testified that he studied the Richter patent with this end in view a number of years before the inventor of 686 filed his application for a patent on that device. (Tr. p. 562). At the time, however, he was not aware of the Michel patent and his efforts, the extent of which were not indicated, did not meet with success. (Tr. pp. 277-278).
The question resolves itself into whether or not after having perceived the possible application of these two patents, a lens designer would have been able to take their teachings and construct a varifocal lens for motion picture cameras. If the ordinary skilled lens designer would have been able to do so, then patent 686 would not be valid even though Michel and Richter had not created such a device, and even though they did not realize that their teachings could be used for this purpose.
De Forest Radio Co. v. General Electric Co., 1931, 283 U.S. 664, 51 S.Ct. 563, 75 L.Ed. 1339 ("The question is not whether the prior art had made a practicable * * * [device], but whether it showed how one could be made * *.")
See General Electric Co. v. Jewel Incandescent Lamp Co., 1945, 320 U.S. 242, 66 S.Ct. 81, 90 L.Ed. 43.
In order to determine this question it is necessary to compare the disclosures of the two prior art patents with patent 686.
First, as to the basic functions of the various lenses: The lenses primarily responsible for the zooming effect are three in number. The first of these three serves primarily to magnify or demagnify the image and incidentally to change the location of the plane of the image. In plaintiff's patents 686 and 817, this lens is termed the variator and that term will be used here. The second lens inverts the image presented to it. This lens is termed the erector. The third lens of the zooming group is the so-called compensator which serves to stop the movement of the image and holds it in a focal plane that is substantially in one position.
These three lenses are present in the Richter and Michel patents as well as in plaintiff's two patents and in the other varifocal lenses in the suit. In every one of these devices they serve the same basic function. What is more, their physical setting and operation is almost identical. In the prior art patents, in 686 and in the subsequent devices, the zooming lenses are contained within a single tube structure composed of various inner barrels holding the different lens elements. The erector lens is always in a fixed stationary position. The variator and compensator lenses are always fixed in position relative to each other and both are moveable as a unit with reference to the stationary erector lens.
In addition to these lenses, there are two others which, while they may not participate in the zoom operation directly, are generally necessary to the practical operation of the device. These are a front lens which is focused on a particular subject producing an image which it positions for action by the other lenses and a relay lens which takes the image formed by the compensator lens and directs it to the desired plane. These lenses are fixed in a stationary position relative to the stationary erector lens. In Richter and Michel, there are lenses which perform these functions. In addition, the particular mechanism for moving the variator and compensator lenses relative to the stationary lenses is concededly equivalent in all three devices as is the particular way in which the various lenses are held in place.
The invention, if any, in patent 686 must be sought elsewhere than in the specific construction of the lens holding members, or the mechanism by which the movable lens assemblies are moved relative to the fixed lens assemblies. Patent 686 emphasized that "The requirement is simply that the variator and the compensator move as an integral unit relative to the front lens, erector and relay lens assemblies which themselves are fixed as a single unit." (Column 5, lines 51-55). This too, however, was not a novel disclosure, since it was basic to the varifocal structure in the prior optically compensated devices.
The Richter patent stated "The additional optical system is conveniently a system, consisting of two displaceable members which have a constant distance apart and of a stationary member disposed between these two displaceable members. When the two displaceable members of the additional optical system are convergent and when the said stationary member is divergent, the focal lengths may be so chosen that the light source image produced permanently lies in one and the same plane, irrespective of the relative positions of the members of this optical system, which means that this system is approximately pancratic." Column 2, lines 20-32. Pancratic means variable focal. The Richter patent goes on in some detail discussing the operation of the varifocal system, with extensive diagrams, charts and specific lens data.
Plaintiff contends basically that the degree of compensation in Richter is not sufficient for purposes of cinematography and that developing a method by which greater compensation may be obtained constituted invention. More specifically, the plaintiff demonstrated that using the calculations for the various lenses given by Mr. Richter, the image would be in exact focus only twice during the zoom range and the disparity during the remainder of the operation would be too great for motion picture purposes. The real question is whether a competent lens designer could have solved this compensation problem. At this point, the expert witnesses were in complete disagreement. Plaintiff's expert testified that "* * * if you took the disclosures and the ideas [in Richter and Michel] and created a zoom lens around that — that is, use the disclosures with the knowledge of a good lens designer — as far as I know you will still end up with having only two compensation points with this large deviation in between." (Tr. p. 563).
The defendant's expert testified, on the other hand, that it was within the general knowledge of the art at the time to be able to construct a device from the disclosures in Richter which would have four exact compensation points and a relatively slight deviation during the remainder of the zoom operation. He arrived at this conclusion by the following reasoning: the diagrams and lens data in Richter relating to the 3 lens zooming group indicated that he intended to construct a symmetrical system — i.e. the compensator and variator lenses were both positive and of equal power, etc. A symmetrical system may be represented graphically by a "W" shaped curve. To construct such a curve, Dr. Rosin took a ray parallel to the axis in the middle lens, traced the ray to one side and chose that as the starting point. There was a slight discrepancy between this point and the figures indicated by Richter, but Dr. Rosin assumed that Richter made a slight miscalculation in working out the arithmetic. From that point he charted a graph which did resemble a "W" shaped curve. By a slight variation of the focal plane on this graph, he demonstrated that the image would be in exact focus four times during a zoom operation and would be well within points of acceptable optical tolerance during the remainder of the operation. (Defendant's Exhibits J and U). As to the movement of the focal plane to the desired position, he testified that this was well within the knowledge of the art "since the beginning of photography." (Tr. p. 474).
Plaintiff's expert conceded that Richter showed at least a "nominally symmetrical system", but he contended that when the figures presented by Richter are traced out they do not show an exact symmetrical system. (Tr. p. 582). He relied on Richter's calculations rather than on the other indicia in his patent indicating a symmetrical system. He agreed, however, that if Richter intended to create a symmetrical system, the method by which Dr. Rosin calculated his graph would have been a correct one. (Tr. p. 583). With regard to the movement of the focal plane in Exhibits J and U, Kaprelian did not directly dispute that this could be done. His testimony with regard to shifting compensation planes was directed at Plaintiff's Exhibit 27 which was a graph charting image displacement in the Richter 3 lens zoom group according to the figures given by Richter. Kaprelian testified that he could presently transform that graph into the graph charted by Dr. Rosin, but he testified that he would not have been able to do so before the issuance of patent 686. His reasons were twofold: one, that before the issuance of 686 and the publication of an article by Dr. Back in 1952 (Plaintiff's Exhibit 31), no one knew how to make such transformations and secondly, apparently there was sufficient knowledge in the art at that time to make the transformation, but there was insufficient disclosure in Richter of the functions of the various lenses to enable one to make this transformation. (Tr. pp. 571-572). Kaprelian did not testify that the transformation made in Exhibits J and U would have been beyond the knowledge of the art at the time. And indeed, the inventor of 686, Dr. Back, admitted that it was customary in the art to make adjustments in order to obtain the best focal sharpness. (Tr. pp. 593-594).
There was sufficient lens data in the Richter patent, however, to enable a lens designer to reconstruct a model of that device to see just how each lens operated.
The question is not really what Richter's intent was with regard to a symmetrical or asymmetrical system, but whether the competent lens designer would have taken the disclosures in Richter's patent and would have checked his figures by drawing a graph based upon a symmetrical system as indicated in the diagrams and in Richter's making the variator and compensator lenses of equal power or whether, if he relied upon Richter's calculations, he would have been able to make the necessary transformation from that graph.
If one or the other possibility would have been obvious to the skilled lens designer concerned with solving the problem, then there is no invention in 686. While I am mindful of the dangers of hindsight, I am led to the inescapable conclusion that it would have been.
Of course, after a designer had solved this problem of compensation range, he would still be faced with what one expert in the field described as the "well-nigh insuperable" (Plaintiff's Exhibit 19, p. 136) difficulties involved in actually designing the lens system in such a manner as to correct all the problems of chromatic and spherical aberration, astigmatism, adequate light transmission, etc. He would have to determine the proper thickness of the lenses and their spacing, etc. Solving these problems may well take many years of careful experimentation.
The plaintiff, however, may not be heard to contend that the solution of these difficulties requires invention for patent 686 gives no indication as to how they were solved in the 686 device.
This brings us to defendant's alternative ground for its claim that patent 686 is invalid — insufficient disclosure.
The 1952 Patent Act like its predecessors requires that "The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention." 35 U.S.C. § 112. (Emphasis added.)
Patent 686, as stated above, makes no reference to the method by which the inventor was able to correct the aberrations which appear in any lens device but which are particularly acute in a device which contains lens elements which are movable in relation to each other. The patent not only does not indicate how these matters were to be corrected but it does not contain the most elementary data with regard to the optical construction of the 686 device which would enable a skilled designer to reconstruct that device without excessive difficulty and experimentation.
The inventor's position is that the methods for correcting such optical problems were well known to the art and beyond the scope of his invention. The publications introduced by the plaintiff to demonstrate that the device was well received by experts in the optical field, indicate, however, that these experts were particularly impressed with the inventor's ability to master just those difficulties.
See e.g. Plaintiff's Ex. 19 at p. 136. Kingslake, Chief Optical Designer for Eastman-Kodak and formerly Dean of the Optical University in Rochester, wrote in "Lenses for Photography" in 1951, as follows:
"* * * the extreme difficulty of both designing and manufacturing a sufficiently well-corrected zoom lens has made them almost non-existent until quite recently. Even today, the best-known zoom lens, the Zoomar, for 16 mm. photography embodies 22 separate lens elements, while for 35 mm. cameras it contains even more. It is hard enough to design a fixed lens with adequate definition for most purposes, and the difficulties of designing a system in which one part of the lens may be moved relative to another part to change the focal length, are well-nigh insuperable."
In Plaintiff's Exhibit 18, at p. 22, another expert stated as follows:
"It is claimed that each lens element has to be corrected for chromatic aberration, spherical aberration and astigmatism, and that coma and distortion are corrected by the `concerted interplay' of the different lenses. There is no mention in the published account of the lens" [nor in the patent] " of the manner in which these aberrations remain corrected on varying the image size. Moreover, all the lenses of components 1, 2, 3 and 4 are of positive power, and consequently must have a large positive Petzval sum. Components 5 and 6 appear to be conventional triplet lenses, so that the Petzval sum of these two components would certainly not be less than zero. For this reason it would seem impossible to obtain really good definition over the whole field of view."
This would lead one to suspect that one of the principal reasons for the delay in developing a zoom lens for motion picture use was the anticipated designing difficulties rather than the theoretical problem of achieving optical compensation.
See Plaintiff's Exhibit 22 at page 64, wherein another expert states:
"Since each lens element is a separate piece of glass, this means 56 surfaces; and when it is considered that each one reflects and loses up to 10 per cent of the light that falls on it, one wonders how any light reaches the plate at all. By coating the lenses with magnesium fluoride, however, this reflection can be cut down to 1½ per cent at each surface. Only since the development of antireflection coating has such a lens design been made possible."
Still another expert stated in reference to the 35 mm. Zoomar lens, in plaintiff's exhibit 20, at pages 145-146, as follows:
"This lens, developed by Dr. Frank G. Back, has twenty-eight elements operating inside the barrel. At once the tremendous value of the lens-coating process so recently developed will be apparent. Without this, the loss of light in passing through all these lenses, would be tremendous, and it would not be half as effective. As it is, the f value is a maximum of f.5.6 with a minimum of f.22. It is thirty inches long and weighs eleven pounds."
The development of anti-reflection coating, not known at the time of the Richter and Michel patents, may be another factor in accounting for their not having developed a varifocal lens for motion picture use.
Patent 686, however, unlike most patents for optical devices, fails to set forth a table indicating the radii and thickness of the various lenses, their refractive index and spacing. These elements are critical, not only to the general designing problem, but to the particular zoom operation of the lens.
Compare, for example, the 817 patent, the Reymond patent, and the Richter patent.
In a pre-trial deposition read into the record, the inventor, Dr. Back, indicated that the power and position of the various lenses were critical factors. He admitted that the patent did not disclose the distance that the front lens should be spaced from the other stationary lenses. (Tr. pp. 548-549). While the powers of certain lenses or lens groups are given, their thickness is not. The inventor conceded that the thickness of a lens could not be determined from its power but that it would be necessary to have the radii, glass coefficients and other factors to determine thickness. He admitted that the thickness of the lenses was an important factor in producing the zoom effect with regard to every lens except the front lens. (Tr. pp. 550-552).
When plaintiff's expert witness, Kaprelian, was asked to explain just what patent 686 disclosed, he answered as follows:
"In patent 686, Dr. Back discloses a method for changing the size of the image on the film without having any non-linear motion required. He does it by moving two groups of based lenses with a third lens between them, and other lens groups to the outside. By properly balancing these powers and choosing his spacings, he is able to achieve this result." (Tr. p. 215).
As far as the movement of the two groups of based lenses with a third, this was clearly taught in the prior art patents of Richter and Michel. And as far as the proper balancing of powers and spacings, it is simply not disclosed in the 686 patent. Thus, even apart from the total failure to disclose the radii, thickness and coefficients of the various lenses, the feeble attempt to set forth powers and spacing which is made, is woefully inadequate.
To illustrate, the front lens is composed of two lens elements. The patent does not set forth their separate powers but indicates that together their power is plus 12-¾ diopters. The distance that the front lens is spaced from the stationary erector or relay assemblies is not disclosed at all. The variator lens assembly is composed of four separate lenses in groups of two. The sole reference in the patent to the construction of these lenses is that there are that many and they are spaced the way they are "to avoid aberration" (no indication of what type of aberration) and that each group of two is of plus 36-¼ diopters and that the groups are spaced 22-¾ mm. apart. The erector assembly consists of eight lenses, cemented into four lens groups spaced in groups of two. The sole statement in the patent is that the entire assembly has a power of plus 32-¾ diopters "and comprising four aplantic doublets to correct aberrations." The compensator assembly consists of two lens elements cemented together. The patent states merely that together they have power of plus 14 diopters. The relay lens consists of two groups of 3 lenses each. The sole information about these lenses given in the patent is that they are anastigmats and each group has a power of plus 23-½ diopters.
With regard to the crucial element of the fixed distance between the variator and compensator lenses, there is no disclosure. Instead, the patent presents a relational formula which is admittedly insoluable mathematically, for determining the characteristics and spacing of the compensator lens. The patent states: "It is impossible to find a value for L, L', and f where the above mentioned equation holds true for every position of the barrel but by judiciously choosing these values, this condition can be exactly fulfilled for three points while the deviations of the other points are so small that they fall within accepted optical tolerances." (Column 4, lines 55-61).
Even the plaintiff did not try to make much of this formula as disclosing the invention apparently because the alleged infringing devices were constructed under a different formula providing for four points of exact compensation. The 686 formula may be translated into an S curve, whereas the defendant's formula would resemble a W curve. To cover the defendant's devices, the plaintiffs had to give their claims the broadest possible reading, bringing them inevitably within the prior art of Richter and Michel.
Even in setting out the functions of the various lens groups the disclosure was not "full". Thus, throughout the trial the plaintiff stressed repeatedly that the relay lens serves to further diminish deviations from the focal plane which the compensator is unable to catch and that it will change the size of the image presented to it to varying degrees. (Tr. p. 187). There is no indication of this function anywhere in the patent. In the mechanically compensated prior art patents for varifocal motion picture lenses, however, this function for the relay lens is disclosed. See e.g. Capstaff. (Tr. pp. 504-505).
Despite these multiple failings, plaintiff's expert witness testified that although he never tried to construct a varifocal device from the disclosures in patent 686, he believed that a competent lens designer would "very probably" be able to do it. (Tr. p. 561).
It is interesting to note how that hypothetical figure, the ordinary skilled lens designer, shifts from a rather dull, unimaginative fellow when the question is whether the device is an advance over prior art, to a rather brilliant, industrious and imaginative person when the question is whether there was sufficient disclosure and whether devices seemingly totally different are mere equivalents.
Even if a competent lens designer could construct such a device from the disclosure in 686, and I find this difficult to believe, plaintiff's evidence indicates that it would certainly take him no less than six years. And it is extremely doubtful that the 686 patent would serve any more of a function than to encourage the ordinary lens designer during this period with the thought that it could be done.
Plaintiff contends that the Pan-Cinor 60 varifocal device, constructed according to the Cuvillier patent (Defendant's Exhibit A) as well as Dr. Back's own patent 817 would both be infringements of patent 686 as mere optical equivalents. The Cuvillier patent introduced a negative erector lens which formed virtual rather than real images and permitted the construction of a much smaller and simpler device. The plaintiff contended that the ordinary lens designer realized that negative lenses may be substituted for positive lenses, and that a negative lens would create a virtual image. (Tr. p. 102). This was Dr. Back's primary change in patent 817. Nevertheless, Dr. Back testified that although he knew that a negative lens could be substituted and would have this desirable effect when he filed for his patent in 686, and although he started work at that time to design a verifocal system with a negative lens, and despite his intimate knowledge of the designing problems to be met in constructing a varifocal lens for motion picture purposes, nevertheless it took him 6 years to go through the necessary mathematical calculations and optical experiments in order to design such a lens. (Tr. p. 177; Plaintiff's Reply Brief p. 10).
With regard to the disclosure problem, it is interesting to note what Dr. Back had to say about 686 in his application and patent for 817.
"It is the object of the present invention to overcome these disadvantages inherent in the device set forth and described in U.S. Patent 2,454,686 * * *. This arrangement [686] requires a rather complex optical construction for the lens necessitating a considerable number of lens elements. In order to keep the overall length of the system in reasonable proportions the lens groups between the images had to have considerable power and had to be split up into several components. The great number of elements thus required caused a substantial decrease in light transmission even when anti-reflection coatings were applied to each of the glass air surfaces. All these lens groups had to have positive powers and therefore caused a strong Petzval curve. Field flattening had to be achieved by introducing either negative astigmatism or distortion." Patent 817, column 1, lines 36-60.
Unfortunately, these factors were not mentioned in patent 686. While it is true that some of these factors relate principally to the achievement of a sharper image rather than to the reduction of image displacement, one of the central grounds upon which the plaintiff seeks to distinguish Richter and Michel is that these factors would have to be adjusted in order to convert the disclosures in those patents into a workable varifocal system for motion picture purposes.
Plaintiff's position, however, is that by describing the functions of the various lenses a competent lens designer would immediately know just what type of lens is required to perform that function and the problems peculiar to the designing of an overall varifocal system and their solutions were well within the knowledge of the art although it would take time to work out the various formulas.
While there may be an element of truth in this contention, where does it leave plaintiff's 686 patent in relation to the prior art patents Richter and Michel? If there was room for invention after those patents, and there well might have been, it could not be satisfied by a patent which did little more than state what role the lenses in the Richter system performed and merely announced to the world that the designing problems could be worked out by diligent labor. Surely, it would be a novel theory to grant patent monopolies to one whose "invention" consisted of stating what function is performed by the elements in a prior patent. This is true, particularly where the prior Richter patent does give sufficient lens data to enable a lens designer to reconstruct the device and see for himself.
Patent 686, while informing the lens designer that "by judiciously choosing" his values he may be able to keep deviations within accepted optical tolerances, does not even inform him what those acceptable tolerances are or what relation they have to the possible range of magnification. The patent gives no indication of the extent to which the compensator actually does compensate for image shift.
Compare discussion by the inventor in Plaintiff's Exhibit 31.
"The original Zoomar lens was laid out in such a way that the end position of the variator barrel always placed the first intermediate real image into the principal plane of a variator component thus eliminating its influence on the displacement error at the end position. Power and spacing of the compensator were chosen to achieve minimum displacement. The compensator therefor was the only element determining the image travel compensation. This way, the best compensation achievable was still approximately half a per cent of its maximum focal length and with a range of 3 to 1, one and a half per cent of the minimum focal length." (p. 1198)
Dr. Back conceded that in 686, despite compensation, the image shift was "still too large to allow a design of a zoom system with a range larger than 4 to 1." Plaintiff's Exhibit 31 at p. 1196. The 686 device actually provided for a range of 3 to 1, whereas the Richter device was designed to provide a range of 9 to 1. Kaprelian, plaintiff's expert witness, admitted that "if you took the system and changed the zoom range from 9 to 1, to say, 3 to 1, you should be able to reduce the excursions from the film plane between compensation points." (Tr. p. 584).
The principal questions facing the art after the Richter and Michel patents were the following: Just how effective would optical compensation be for motion picture purposes? For what range of magnification? And just how could the "well-nigh insurmountable" designing problems be surmounted? The patent answered none of these questions. Certainly, in a field such as optical design, where experimentation involves such lengthy, arduous and costly work, including the construction and testing of innumerable models to see whether they might satisfy the essential requirements, failure to make full disclosure is totally inexcusable.
The requirement of "full" disclosure is no mere technicality, but is basic to the underlying policies of the patent law. A principal purpose of the patent law is to make sure "that the subject of the patent will be dedicated ultimately to the public." General Electric Co. v. Wabash Appliance Corp., 1938. 304 U.S. 364, 369, 58 S.Ct. 899, 902, 82 L.Ed. 1402.
"Specifications of the patent, including the description and the claims, constitute a contract between the public and the patentee, under which the public, through the government, agrees that, in consideration of the inventor's disclosure of his teaching and his grant of the right to use the same after his monopoly expires, he shall be protected in his exclusive use during the life of the patent." National Carbon Co. v. Western Shade Cloth Co., 7 Cir., 1938, 93 F.2d 94, 96.
"In return for the monopoly that the government grants to the inventor for a term of years, he is to disclose enough about his device to bring it within reach of the public when his patent expires. A disclosure that would enable a man of exceptional training and experience in the art, but not a person of ordinary skill in it, to make the device might not serve this important public purpose. Persons wishing to produce the device when the patent expires might not be able to find such a man; in fact, there might be no such man. The statute therefore requires the applicant for a patent to describe his invention, and how to make and use it, so that `any person skilled in the art' can make and use it." International Standard Electric Corp. v. Marzall, 1950, 87 U.S.App.D.C. 198, 184 F.2d 592, 593.
In the instant situation, the disclosure is so scanty that not only would the ordinary lens designer find it an almost impossible task to reconstruct the device, but even the most exceptional lens designer would encounter grave difficulties.
Section 112 seeks not only to enable the public to reconstruct the device at the expiration of the patent, but also "to guard against unreasonable advantages to the patentee and disadvantages to others arising from uncertainty as to their rights. The inventor must `inform the public during the life of the patent of the limits of the monopoly asserted, so that it may be known which features may be safely used or manufactured without a license and which may not.'" General Electric Co. v. Wabash Appliance Corp., 1938, 304 U.S. 364, 369, 58 S.Ct. 899, 902, 82 L.Ed. 1402.
By phrasing the specifications and claims in the broadest possible language, the inventor sought to extend the range of equivalents to every conceivable method of compensating optically for image shift in varifocal devices. But, "Certainly, if we are to be consistent with Rev.Stat. § 4888, a patentee cannot obtain greater coverage by failing to describe his invention than by describing it as the statute commands." Halliburton Oil Well Cementing Co. v. Walker, 1946, 329 U.S. 1, 13, 67 S.Ct. 6, 12, 91 L.Ed. 3.
By phrasing its claims and specifications in the broadest possible terms, the 686 patent effectively posed a threat to every conceivable optically compensated varifocal device. Thus, plaintiff contended throughout the trial that not only the defendant's device, but their own 817 as well, would infringe 686. The following chart points up some of the differences between these devices:
Alleged Infringing 686 817 Devices ------------ -------------------- -------------------- -------------------- Front Movable for focusing Composed of separate Movable for focusing Lens purposes lenses purposes only; no variable movable in relation only; no variable power; but to each other power. of high positive to vary their power. combined power from a negative power to a very low positive power. May have zero power. Variator Of positive power. Of positive power. Of negative Varies size In order power. of image by itself. for this lens to Necessary magnify the image Variator produces to have field lens age must have magnification as element in erector lens in without variator assembly. place. stationary third Image lens. never formed to Compensator the right of 2d has greater power Variator and variator lens. than variator. compensator are Variator has of equal power. greater power Magnification than compensator. occurs when Magnification Magnification variator moves occurs when occurs when toward the front variator is variator is lens. moved away moved away from the front from front lens. lens. Erector Of positive power. Of negative Of positive power. Erects the power. Erects Inverts the image formed the image. image formed by the first 2 by the first 2 lenses. The image lenses. is real. Compensator Of positive power. Of positive Of negative As zoom power. Image power. Changes group is moved formed by this real, inverted from front lens, lens is real and image to an erect real image inverted. Lens and virtual formed by the changes the size image. This occurs erector is of the image. throughout changed to a virtual zoom operation. image by the compensator. Relay Forms a real inverted Does not invert Forms a real inverted image. the image. image.In addition, plaintiff contended that the Pan-Cinor 60 device, Defendant's Exhibit A, would also infringe 686, although that device does not have a front lens. The variations in the position and formation of images, real and virtual, in these devices, are substantial and extremely diverse. The only thread running through each of these lens devices is that they all contain at least 4 lenses designed to accomplish the varifocal effect optically. They each contain a lens which somehow magnifies the image, one which then somehow inverts that image, one which somehow compensates for image shift and one which somehow takes the image formed by the compensator and transfers it to the desired focal plane. In addition, the so-called variator and the compensator lens assemblies move as a unit in relation to the stationary lenses.
If the claim of 686 may be read this broadly, then clearly those claims can have no validity against the prior art patents of Michel and Richter which also contain an optically compensated varifocal lens system, with a lens which magnifies the image, one which inverts that image, one which compensates for image shift and one which transfers that image to the desired focal plane. In both Richter and Michel, the variator and compensator lens assemblies move as a unit in relation to the stationary lenses.
In addition, the prior art patents also fully disclose mechanisms for holding and moving these lenses which are concededly equivalent to those shown in 686.
Even assuming that 686 is invention and that the invention is sufficiently disclosed, its claims could not extend beyond the equivalent of that invention. In this case, those equivalents could not conceivably extend beyond an S compensation curve system. At most, that is all that could possibly be considered novel in the patent.
The defendant's device is not based upon an S curve compensation system, but rather upon a W curve symmetrical compensation system permitting four points of exact focus and a substantial decrease in deviations throughout the rest of the operational range. This is more than a mere improvement, it is a distinct theory beyond the range of equivalents to which 686 would be entitled even were it disclosed invention. In fact, the defendant's device is much closer to the prior art patents than it is to patent 686. The plaintiff cannot preclude all other optically compensated varifocal lens systems for motion picture use, merely because Dr. Back was the first to think of utilizing varifocal lenses for this purpose. As Judge Learned Hand said:
The Patent Office accepted this as a validly distinguishing feature when they granted a patent for defendant's devices to the assignor of defendant's manufacturer (Defendant's Exhibit H; Column 1, lines 41-49).
"A patentee may indeed monopolize all the uses of his invention, but we must first learn what that invention is. It is not the idea in vacuo; the consideration for the grant is a chart which others can follow, not an intimation they must work out for themselves. If they embody the idea by quite different means, they need pay him no tribute; he gets his reward for marking the path, not for pointing to the goal * * *." Grubman Engineering Mfg. Co. v. Goldberger, 2 Cir., 1931, 47 F.2d 151, 153.
I find that the defendant has overwhelmingly satisfied its burden of proving that patent 686 is invalid. Furthermore, plaintiff has failed to prove that the patent is infringed by defendant's devices.
We now come to the second patent in suit, Back 817. Plaintiff contends that the defendant's devices infringe claims one and three of this patent. Defendant alleges here, as before, that the patent is invalid and that alternatively, it is not infringed. As noted earlier, patent 817 was granted as an inventive advance over Back 686. Here again, however, the most pertinent prior art was not considered by the Patent Office. Particularly, the Michel and Richter patents and the Cuvillier patent, U.S. Patent No. 2,566,485, issued on September 4, 1951, over a year before the filing date of patent 817.
Plaintiff claims two principal advances in this patent.
"817 is an improvement on 686. It adopts and uses the basic attachments of 686 in that it has the identical three-barrel, five-lens attachment. The only differences lie in that in 817 a negative erector lens is introduced, instead of a positive lens, and a change is made in the front lens. Instead of moving the entire front lens for focusing, the front lens is now composed of two elements and spaced apart and variable, so that they can be moved with respect to each other for focusing, but after such focusing remains fixed in position. The effect of this change was that instead of using real images as were used in 686, virtual images were used in 817. * * * The effect of the use of virtual images was to enable the plaintiff to make a shorter and better lens. However, the functions, purposes and operation of each lens group still remain the same. The change was merely an improvement." Plaintiff's Summation Brief, pp. 14-15.
The initial question is whether these changes constitute invention. Firstly, as to the substitution of a negative erector lens, there is clearly no invention. The erector lens in both Richter and Michel was negative. In both of those cases, virtual images were utilized. The inventor, Dr. Back, and plaintiff's expert witness, Kaprelian, testified repeatedly throughout the trial that it was well known in the art that negative lenses could be substituted for positive ones, and that they would produce virtual rather than real images. (Tr. pp. 102, 247).
It was equally well known that the utilization of virtual images would permit a shorter optical device reducing the light transmission problem and allowing for lenses of lower power.
Moreover, the defendants were selling an optically compensated varifocal device with a negative erector lens for use in conjunction with motion picture cameras, a year before Dr. Back filed his 817 application. Defendant's Exhibit A. This device, called Pan-Cinor 60, made in accordance with the Cuvillier patent, resembles 817 in all respects (including virtually identical ratios of focal lengths — see Defendant's Exhibit P and Tr. pp. 386-390) except for two. The first difference is that Pan-Cinor 60 is an afocal camera attachment rather than a complete camera system. For this reason, plaintiff contends that Pan-Cinor 60 and Cuvillier may not be considered anticipatory prior art. The distinction between afocal and complete objective systems, however, is not at all material in the art. Dr. Back agreed that "the difference between a complete zoom lens and an afocal attachment of the zoom lens is that in the relay part of the afocal attachment you omit parts that correspond to a camera objective whereas in the complete zoom lens you include the parts which would normally be supplied by the camera objective lens." (Tr. p. 192).
In fact, in the 686 patent, the inventor clearly explains how that device, which was a complete objective, may be converted into an afocal attachment by merely removing half of the relay group. Certainly, adding an additional relay lens does not constitute an inventive advance over Cuvillier.
"If only half a relay system with an equivalent focal length equal to the focal length of a standard motion picture objective is used the varifocal lens becomes an afocal system (both focal points at infinity). Such a system can be placed in front of any standard camera lens thereby converting the latter into a varifocal lens." (Column 5, lines 6-13). Dr. Back also indicated that he considered Pan-Cinor 60 an infringement of 686 although the Pan-Cinor device was an afocal attachment.
The second element distinguishing 817 from Pan-Cinor 60 is that the latter does not contain a front lens, whereas the former contains a front lens of variable power. In Pan-Cinor 60, the variator lens serves the same function as a front lens. It is variable for focusing purposes and then becomes fixed during the zoom operation. The addition of a front lens that is variable for focusing purposes, was certainly not new in the art, since the 686 device contained such a front lens. The only new element in 817 is that there is a gap between the two elements comprising the front lens assembly which may be varied to increase or decrease the combined power of that lens from a relatively slight negative power to a relatively slight positive power. The front lens in 686 on the other hand is not of variable power. The patent, however, does not disclose at any point why this makes any difference. The only function attributed to the front lens in the patent is that it is variable for focusing. Furthermore, in the alleged infringing devices, the two front lens elements are cemented together as they are in 686. Since the plaintiff contends that the defendant's front lens is equivalent to the front lens in 817, the front lens in 686 must necessarily be equivalent and anticipate the front lens in 817.
During trial, Dr. Back testified that the "front lens has power also to reduce the zoom deviation because that has something to do with it." (Tr. p. 130). This is not indicated in the patent.
Plaintiff's witnesses testified that Pan-Cinor 60 gave inferior picture quality because it did not have a front lens. All that 817 does is add a front lens to the disclosures in Cuvillier and Pan-Cinor 60. Dr. Back conceded that 817, unlike 686, would work without a front lens, but he stated that the picture quality would be inferior without that lens. (Tr. p. 198).
Even if the variable gap front lens in 817 is somehow effective in enabling the achievement of a sharper image, it certainly cannot rise to the dignity of a patentable advance. Nor can Pan-Cinor 60 be validly distinguished on this basis.
The one additional distinction between Pan-Cinor 60 and 817 is the mechanism by which movement of the variator and compensator lenses is affected. Plaintiff concedes, however, that the different methods are equivalent. (Tr. pp. 147 and 256).
Even if the 817 patent could somehow be considered invention, its range of equivalents would be extremely small. They would certainly not cover the defendant's devices which, as stated earlier, operate on a different theory of optical compensation permitting four points of exact focus and a smaller curve of deviation between those points.
In addition, if there is invention in having the variable gap front lens, that invention is not infringed by the defendant's devices because they do not have front lenses of variable power.
Patent 817 is invalid because of anticipation by Richter, Michel, 686, Cuvillier and Pan-Cinor 60. Even if it were valid, however, it would not be infringed by the accused devices.
We come lastly to defendant's counterclaim based on their contention that it was unfair competition for the plaintiff to send certain notices to the trade relating to the bringing of the instant action. There seems to be no doubt that this court has jurisdiction over this matter. Cutting Room Appliances Corp. v. Empire Cutting Machine Co., 2 Cir., 1951, 186 F.2d 997. However, I find the claim itself without merit. Beyond a doubt, the notices were sent in good faith. As was clearly enunciated in Kaplan v. Helenhart Novelty Corp., 2 Cir., 1950, 182 F.2d 311, 314,
"* * * it is not an actionable wrong for one in good faith to make plain to whomsoever he will that it is his purpose to insist upon what he believes to be his legal rights, even though he may misconceive what those rights are. * * * And this has long been recognized in respect to patent rights."
Settle decree promptly in accordance with this decision. The exceptional circumstances necessary for the allowance of reasonable attorney's fees ( 35 U.S.C. § 285) are not present in this case.