ORDER DENYING DEFENDANTS’ MOTION FOR SUMMARY JUDGMENT OF INVALIDITY

I. INTRODUCTION

Defendant’s motion for summary judgment came on for hearing on November 6, 1997. Having carefully considered the papers submitted and oral argument of counsel, the Court hereby DENIES defendant’s motion for summary judgment.

II. BACKGROUND

A. Procedural Background

The instant ease, filed by Robotic Visions Systems, Inc. (“RVSI”) on April 1, 1996, is the most recent of four cases in a series of patent infringement disputes between RVSI and View Engineering, Inc. (“View”), both of whom make automated scanners. This motion arises ft-om a dispute about the patentability of a method for calculating the coplanarity of measured copper pads on the surface of a printed circuit board or a chip substrate, which method takes the form of a software program which may be implemented in a solder paste inspection machine. *1327 RVSI alleges that View is willfully and deliberately infringing its U.S. Patent No. 5, 465,152 (“the ’152 patent”) by making, using, and selling certain integrated circuit testing and inspection equipment. On April 22, 1996, defendant View filed an Answer and Counterclaim for declaratory relief. The case was transferred through the low number process to this Court on April 30, 1996.

On January 22, 1997, this Court granted defendant View’s Motion for Interpretation of the Patent Claims and held that the preamble language “for ball grid array, column grid array, and similar surface mount integrated circuit chips” is not a limitation of the claimed method, but rather is merely an indication of the intended use of the patented method. On June 13,1997, this Court denied defendant’s Motion for Summary Judgment of Non-Infringement.

On September 15, 1997, defendant View filed the instant Motion for Summary Judgment of Invalidity on the grounds that View had anticipated the T52 patent claims, invalidating them under 35 U.S.C. § 102 , and on the grounds of obviousness under 35 U.S.C. § 103 . Plaintiff filed a timely Opposition on September 29, 1997. A timely Reply was filed on October 8, 1997. RVSI lodged a Proposed Surreply on October 15, 1997, which was filed pursuant to this Court’s Order on October 17,1997.

B. The Technology

As set out in this Court’s Order of June 11, 1997, an integrated circuit chip is a component which contains multiple electronic devices built upon a body, or substrate. The integrated circuit chip is connected to other chips and electrical components by exposed conductive “leads,” which may be of different shapes. One such chip is the ball grid array (BGA) chip, on which solder balls are placed on signal pads. A printed circuit board (PCB) is a non-conductive board upon which electronic components, such as integrated circuit chips, are mounted so that the metallic conductors on the board electrically connect the electronic components together to form the desired circuit. (June 11, 1997 Order, p. 3.) Solder paste is the material which is applied to PCBs to allow attachment and electrical connection of chips to PCBs. The proper volume of solder paste is necessary for optimal connection. (RVSI’s Opp. at p. 2, note 3.)

Placed on top of a flat surface like a circuit board, a chip with multiple leads will contact that surface at three points (three points define a plane). However, it is desirable that as many leads as possible (balls, in the case of a BGA chip) contact the PCB (or other component). Thus manufacturers of integrated circuit devices generally aim to ensure that the leads of the devices are all in the same plane, i.e., are coplanar. This coplanarity ensures that when the integrated circuit and the printed circuit board are mated and heated, the solder can melt, or reflow, and form continuous metal links. (’152 Patent, Col. 1, ll. 31-40, View’s Mot., Ex. A.) If the integrated circuit substrate is warped, the leads of the devices will not be coplanar (as the surface they touch is not flat), and the solder reflow will not form the required continuous metal links. (’152 Patent, Col. 1, ll. 31-40; View’s Mot., Ex. A.)

There are technical problems, associated with determining whether chip contacts and pads on PCBs are properly connected. Direct visual inspection is impossible after soldering because the contacts are between the board and the chip. X-ray imaging is possible, but chip removal is difficult. (’152 Patent, Col. 1, ll. 13-15; View’s Mot., Ex. A.)

Plaintiff RVSI and defendant View both manufacture and sell laser scanning machines which perform inspections of a variety of integrated circuit chips (which neither party manufactures). (June 11, 1997 Order, pp. 2-3.) RVSI’s equipment inspects.BGA and other chips. Until 1993, the View 7530 Solder Paste Inspection System (“the 7530”) was sold with the ability to measure solder paste volume and area, but not coplanarity (Rohrer Depo., pp. 25-26.) The versions of the software in use in the machine at that time, versions 3.0 and 3.1, did not measure coplanarity. (Rohrer Depo., pp. 25-26.) View now sells the Model 880, which performs inspection of integrated circuit chips for coplanarity. (RVSI’s Opp., Ex. D.) However, as late as 1995, View application engineers could not configure the 880 to inspect unpopulated (ball-less) BGA substrates *1328 (BSMs). CRVSI’s Opp., Ex. J and K; Baldauf Depo., p. 94-95.)

The ’152 patent technology grew out of RVSI’s development of chip inspection machines. Coplanarity is generally measured optically, with three dimensional (“3D”) vision techniques, as mechanical measurements cannot be made with the necessary speed and accuracy. (’152 Patent, Col. 1, ll. 41-43, View’s Mot., Ex. A.) RVSI found that truly accurate warpage measurement of chip substrates is extremely difficult, as many chip substrates are translucent to the laser light used by 3D vision techniques. (’152 patent, Col. 1, ll. 43-54, View’s Mot., Ex. A.) Conductive lines and surfaces embedded within a translucent substrate may mistakenly be interpreted by the 3D vision system as defining the surface of the substrate. (152 Patent, Col. 1, ll. 50-56, View’s Mot., Ex. A.)

The inventors of the 152 patent overcame this problem by using opaque reference points, called opaque fiducials or index pads, on the surface of the substrate, for coplanarity measurements. These provide a good indication of the warpage of the chip substrate. (June 11,1997 Order, p. 5.)

C. The Patented Invention

RVSI filed its application for the 152 patent on June 3, 1994. “The object of [the invention claimed in the 152 patent] is to provide a practical and reliable means of determining the eoplanarity of [chips and chip substrates].” (152 Patent, Col. 2, ll. 5-7, View’s Mot., Ex. A.) The 152 patent includes twelve claims, of which claims 1, 3, and 12 are presently at issue. (Claim 1 is the only independent claim.)

The claims read as follows:

(These claims are sometimes referred to collectively hereafter as “the ’152 claims.”)

In this Court’s June 11, 1997 Order denying View’s prior motion for summary judgment (of non-infringement) in this case, this Court was called upon to perform claim construction with respect to the language of claims 1 and 12. (June 11, 1997 Order, pp. 10-15.) In that Order, this Court held that the language of “providing,” “correlating,” “disposing,” and “restricting” in claim 1 encompassed or included selecting certain preexisting pads to be used as index pads for measurement purposes (and did not require their manufacture). This Court did not hold that the ’152 claims were limited to the selection of pads. (June 11, 1997 Order, p. 14.) On the contrary: this Court held that under Warner-Jenkinson Co. v. Hilton Davis, 520 U.S. 17 , 117 S.Ct. 1040, 1049 , 137 L.Ed.2d 146 (1997), all steps of a claim are to be given meaning.

*1329 III. ANALYSIS

A. Legal Standards

1. Summary Judgment

Summary judgment shall be granted if the evidence supporting the motion for summary judgment shows that “there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c). A party moving for summary judgment may carry its initial burden by pointing out to the district court that there is an absence of a genuine issue of material fact. Celotex Corp. v. Catrett, 477 U.S. 317, 323 , 106 S.Ct. 2548 , 91 L.Ed.2d 265 (1986). “The plain language of Rule 56(e) mandates the entry of summary judgment, after adequate time for discovery and upon motion, against a party who fails to make a showing sufficient to establish the existence of an element essential to that party’s case, and on which that party will bear the burden of proof at trial.” Celotex, 477 U.S. at 322 .

To avoid summary judgment, the non-movant must set forth specific facts showing that there remains a genuine issue of material fact for trial. Celotex, 477 U.S. at 324 . The non-movant “may not rest upon the mere allegations or denials of the adverse party’s pleading.” A factual dispute is “genuine” if a reasonable jury could return a verdict for the non-moving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 , 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986). The evidence of the non-movant is to be believed, and all justifiable inferences are to be drawn in the nonmovant’s favor. Id. 477 U.S. at 255. If the non-moving party’s evidence is merely color-able or is not significantly probative, then summary judgment may be granted. Id. at 249-250.

2. Summary judgment in the patent invalidity context

Patents are presumed valid. 35 U.S.C. § 282 (1988). The presumption of validity afforded by § 282 places the burdens of going forward and of persuasion on the party asserting invalidity. The party asserting invalidity must establish, by clear and convincing evidence, facts which support the ultimate legal conclusion of invalidity. Checkpoint Systems v. U.S. Intern. Trade Com’n, 54 F.3d 756 (Fed.Cir.1995). When supported by the facts and law, summary judgment is appropriate in patent cases. See, e.g., Barmag Barmer Maschinenfabrik AG v. Murata Mach., Ltd., 731 F.2d 831 (Fed.Cir.1984). For example, a district court can properly grant a motion for summary judgment on patent invalidity when the factual inquiries into obviousness present no genuine issue of material fact. Ryko Mfg. Co. v. Nu-Star, Inc., 950 F.2d 714 (Fed.Cir. 1991).

The production of evidence of prior art not considered by the Patent and Trademark Office may carry more weight as the district court is not called upon to disagree with the PTO or required to defer to it as to the effect of that evidence upon the validity of the patent. American Hoist & Derrick Co. v. Sowa & Sons, 725 F.2d 1350, 1369-60 (Fed.Cir.1984).

B. View’s Anticipation Arguments

View argues that the ’152 patent is invalid because View’s 1990-91 activities anticipate claims 1, 3, and 12, in violation of 35 U.S.C. § 102 . Specifically, View argues that patent T52 was “invented by another who did not abandon, suppress, or conceal it,” in violation of § 102(g); was “known or used by others,” invalidating it under § 102(a); and was “on sale” one year before the patent application, and thus invalid under § 102(b). (View also argues that the ’152 patent is obvious, in violation of 35 U.S.C. § 103 , discussed infra at III.C.)

View’s first three arguments are all variations on the claim that View had anticipated the ’152 patent. “An invention is anticipated if the same device, including all the claim limitations, is shown in a single prior art reference. Every element of the claimed invention must be literally present, arranged as in the claim.” Richardson v. Suzuki Motor Co., Ltd., 868 F.2d 1226 (Fed.Cir.1989).

1. Factual Background

The following facts form the evidentiary basis for all View’s claims under § 102 (that the ’152 invention was anticipated by work done by others). The same facts which *1330 tend to show “invention by another” (invalidating a patent under § 102(g)) may also demonstrate “knowledge] or use by others” under § 102(a) or “public use” under § 102(b). Similarly, an offer for sale before the critical date (by the inventor or another) of the patented device may not only invalidate a patented invention under § 102(b), but may also show non-abandonment by a prior inventor under § 102(g). Facts stated below are relevant and undisputed, except where noted:

In May, 1990, Donald Svetkoff of View (“Svetkoff”) performed an analysis which became the basis of a May 17, 1990 report. (Svetkoff Decl., Ex. A.) The subjects of this report were “(1) Solder Mask/Conduct Height Study (2) Benchmarks for Planar Surface Prediction — Analysis of FlaVQuasiFlat Surfaces — Predict Z Location From Reference Points.” (Svetkoff Decl. ¶ 3, Ex. A, p. 8.) He studied an IBM PCB, and learned and wrote that IBM’s layer of copper signal pads on the PCB was the best target (“benchmark”) for measuring the eoplanarity (“planar surface prediction”) of that device. (Svetkoff Decl. ¶ 3.) Svetkoff selected signal pads as targets, measured their heights, generated a best fit to that height data, and calculated the differences between the signal pads and the best fit. (Svetkoff Decl. ¶ 3.) His best fit calculations employed a line rather than a plane; however, his report does mention the use of a plane. (Svetkoff Decl. ¶ 3, note 1, Ex. A, p. 21.) He also reported on the “potential advantages expected of using solder mask” rather than “special reference points which can pose a severe problem.” (Svetkoff Decl., Ex. A, p. 12.)

On September 18, 1991, Chuck Gamble (“Gamble”), a View sales representative, sent a letter to William Von Voss (“Von Voss”) of IBM. This letter (“the Gamble letter”) includes the statement, “The 7530 is capable of using local fidueials, bare copper pads, or solder mask covered traces for Z referencing.” (Svetkoff Decl., Ex. D, p. 51.)

In September, 1991, Svetkoff performed an analysis of a prototype of the 7530. This analysis was discussed in a September 26, 1991 memorandum, one of whose subjects is “Aii evaluation of coplanarity of the reference pads.” (Svetkoff Decl., Ex. B, p. 36.) An internal heading reads “Coplanarity Deviation of pads.” (Svetkoff Decl., Ex. B, p. 42.) RVSI disputes the accuracy of this descriptive heading, contending that the material presents only data involving “measuring pad heights along a straight line (in both the horizontal and vertical direction) and calculating a line (using an average) from those points.” (RVSI Opp., pp. 5-6.)

In October, 1991, Svetkoff performed an experiment using solder paste inspection software on the 7530, reflected in an October 5, 1991 memorandum. (Svetkoff Decl., Ex. C.) The memorandum’s subject was “Deviation from Coplanarity of Reference Pads (IBM Board) and SPI Inspection Error.” The testing, explicitly borrowed from View’s competitor CyberOptics, included the following steps:

(Svetkoff Decl., Ex. C, p. 45; emphasis in original.) The point of this experiment was to ensure accurate height and volume of solder paste. (Svetkoff Depo., p. 175; RVSI’s Opp., Ex. E.)

On October 10, 1991, Michael Beck (“Beck”) of View sent a letter to Von Voss. This letter (“the Beck letter”) discusses volume measurement employing “a plane defined by the four corner fidueials.” (Svetkoff Decl., Ex. E, p. 54.) It states that “Development is currently progressing to incorporate the modified Z height reference algorithm into the 7530 application software package. This new algorithm will allow reference points to be defined near each pad____ The new algorithm will adjust the measured plane to the Z reference point closest to the measured solder paste deposit minimizing the error induced by board non-coplanarity.” (Svetkoff Decl., Ex. E, p. 54.) It does not explicitly discuss whether the new algorithm *1331 would calculate a differential between measured pad height and a calculated plane.

The solder paste volume algorithms referenced in the Gamble and Beck letters were not available in the 7530 until at least 1992. (Kelley Depo., pp. 63-65, RVSI’s Opp., Ex. B.) Furthermore, the new algorithms did not compute a difference between a measured height of a pad and a calculated threshold plane. (Kelley Depo., p. 59.)

As early as 1991, Shayla Fahey of IBM developed measurement techniques for integrated circuit devices. (Fahey Decl. ¶¶ 6-7, 14.) She measured coplanarity of IC devices, including PCBs and BSMs, by selecting target locations for height measurement, measuring the heights of a device at those target locations, creating a best fit plane from that height data, and calculating the difference in height between each target location and the best fit plane at that location. (Fahey Decl. ¶¶ 8-9.) She sometimes used signal pads as targets, and wrote software for creating the best fit plane and calculating differences between the height of the signal pads and the best fit plane. (Fahey Decl. ¶¶ 11-12.)

2. Invention by another (§ 102(g))

a. Legal Standards

35 U.S.C. § 102 (g) provides that “[a] person shall be entitled to a patent unless— ... (g) before the applicant’s invention thereof the invention was made in this country by another who had not abandoned, suppressed, or concealed it. In determining priority of invention there shall be considered not only the respective dates of conception and reduction to practice of the invention, but also the reasonable diligence of one who was first to conceive and last to reduce to practice, from a time prior to conception by the other.”

A patented invention is previously invented under § 102(g) if the other is the first to conceive and reduce to practice the patented invention. New Idea Farm Equipment Corp. v. Sperry Corp., 916 F.2d 1561, 1566-67 (Fed.Cir.1990). The guiding standard is “whether the public has gained knowledge of the invention which will insure its preservation in the public domain.” Palmer v. Dudzik, 481 F.2d 1377, 1387 (C.C.P.A.1973), cited in Checkpoint, supra at 762.

b. Discussion

View argues that everything purportedly done by RVSI to support the claims of the T52 patent was previously done by Svetkoff in the 1990-91 activities described above; that Svetkoff did not abandon, suppress, or conceal the results; that the information was released freely to IBM; and that the T52 claims are therefore invalid under 35 U.S.C. § 102 (g) because of previous invention by Svetkoff.

RVSI disputes this position, and argues that Svetkoffs activities in 1990-91 do not invalidate the patent because those activities do not include every element of claims 1, 3, and 12 of the ’152 patent, as required for anticipation under Richardson, supra. Specifically RVSI argues that two elements of the ’152 patent were not included in Svetkoffs 1990-91 activities: Svetkoffs use of four corner fiducials on a PCB did not meet the claim requirements for “providing opaque fiducials as index pads where the heights of said index pads are correlated with signal pad heights in a neighborhood about said index pads” (hereafter “opaque fiducial index pads correlated with neighborhood pad heights”), as specified in claim 1. Comer fiducials may or may not be correlated with signal pad heights, and this correlation is part of how the T52 patent works. Second, Svetkoffs use of the heights of four corner fiducials to calculate a plane did not result in a calculation of “the difference in height between each index pad and the preselected surface shape evaluated at the index pad location,” also specified in claim 1. RVSI also argues that the three experiments or tests done by Svetkoff in 1990-91 each differ from the ’152 claims in significant ways, for example, by not including calculation of a plane.

More specifically, RVSI has presented the following admissible evidence of discrepancies between the elements of the T52 invention and the work done by Svetkoff upon which View relies to show anticipation:

(1) Svetkoffs May, 1990 analysis, discussed in the May 17,1990 report

(a) did not include opaque, fiducial index pads correlated with neighborhood pad heights (Svetkoff Decl., Ex. A);'

*1332 (b) favored a technique of reference points covered by solder mask, which is specifically rejected by the ’152 patent (Svetkoff Decl., Ex. A, p. 12; ’152 Patent, Col. 4, 11. 19-30);

(c) rejected “special reference points,” the very technique used in the '152 patent (referred to herein as opaque fiducial index pads correlated with neighborhood pad heights) (Svetkoff Decl., Ex. A, p. 12);

(d) did not include the limitation of index pads that “are identical with the signal pads rather than separately fabricated elements intended solely for use as fiducial pads, which index pads are located at signal pad sites occupied by array elements such as bga balls, columns, or other; and determining coplanarity prior to deposition of grid array elements” (hereafter “identical signal and index pads located at sites occupied by array elements and determining coplanarity prior to deposition of said elements”), as required by claim 12 (Svetkoff Decl., Ex. A); and

(e) did not compute a surface or plane, as required by claim 1, but only a line (Svetkoff Decl. ¶ 3.)

(2) Svetkoffs September, 1991 analysis, discussed in the September 26, 1991 memorandum

(a) did not include opaque fiducial index pads correlated with neighborhood pad heights (Svetkoff Decl., Ex. B);

(b) did not include identical signal and index pads located at sites occupied by array elements and determining coplanarity prior to deposition of said elements (Svetkoff Decl., Ex. B);

(e) although captioned “Coplanarity Deviation of Pads,” the data presented only involve measuring pad heights along a straight line and calculating a line, not a plane, as required by claim 1 (Svetkoff Decl., Ex. B, pp. 42, 44; RVSI Opp., pp. 5-6.)

(d) failed to specify the methodology used, including for example, which pads were used as index pads, and so cannot be compared to the claim language (Svetkoff Decl., ¶ 6); and

(e) despite Svetkoffs testimony that he “collected height data for a representative sample of the 124 signal pads on the PCB, estimated a best fit plane for that height data, and calculated the differences between this height data and this plane” (Svetkoff Deck ¶ 6), the Memorandum of that experiment discusses only deviation between the height of pads and average height only along one row, not a plane (Svetkoff Decl., Ex. B, pp. 42, 44).

(3) Svetkoffs October, 1991 experiment, discussed in the October 5, 1991 memorandum,

(a) did not include opaque fiducial index pads correlated with neighborhood pad heights (Svetkoff Decl., Ex. C; Svetkoff Depo., p. 72);

(b) did not include identical signal and index pads located at sites occupied by array elements and determining eoplanarity prior to depositing of said elements (Svetkoff Decl., Ex. C);

(c) was undertaken to determine height and volume of solder paste, not to determine coplanarity through the use of the four corner fidueials (Svetkoff Depo., p. 175); and

(d) did not include identical signal and index pads, as required by claim 12 (Bilodeau Decl. ¶ 26.)

RVSI also presents admissible evidence which contradicts View’s claims under § 102(g) by showing that whatever work was done by Svetkoff was “abandoned” or “concealed” from the public. Prior invention by another does not invalidate a patent if the prior invention was abandoned or concealed. 35 U.S.C. § 102 (g).

RVSI first points to the absence of any specific evidence presented by View that information about Svetkoffs work was ever communicated to IBM or anyone else, or that IBM ever learned of it from View. Under Checkpoint, supra, the burden is on View, the party challenging the validity of the patent, to present clear and convincing evidence of invalidity. Under Palmer, supra, if the public has not gained knowledge of the previous invention, the patent is not invalid even if previously invented. RVSI has presented admissible evidence that the October, 1991 experimental information was never incorporated into any View inspection machine, at least so far as Svetkoff was aware (Svetkoff Depo., p. 201). RVSI also notes an inconsistency in View’s contention that Svetkoffs September, 1991 Memorandum made the invention known to the public when View itself *1333 has required that the Memorandum be treated as “Confidential/Attorney’s Eyes Only.” (Svetkoff Decl., Ex. B.)

View argues that the Gamble and Beck letters negate abandonment or suppression of Svetkoffs work. However, RVSI presents admissible evidence that

(1) the Gamble letter does not describe Svetkoff s methods, for example, it does not mention coplanarity or warpage calculation (Svetkoff Decl., Ex. D);

(2) the Gamble letter does not mention Svetkoffs testing (Svetkoff Decl., Ex. D; Svetkoff Depo., p. 220);

(3) the Beck letter does not describe the May, 1990 and September, 1991 analyses (Svetkoff Decl., Ex. E);

(4) the Beck letter does not contain all the steps of Svetkoffs testing (Svetkoff Depo., pp. 204-5);

(5) to the extent that the Beck letter communicated Svetkoffs four corner fiducial methodology, it also communicated its abandonment, as the Beck letter reads, “Based on these findingfs] [volume measurement errors of approximately 10%] the SVS reference height algorithm is being revised to better fit the true board surface and result in the most accurate volumetric measurements” (Svetkoff Decl., Ex. E, p. 54)

Abandonment, suppression, or concealment is inferable from View’s failure to use whatever Svetkoff discovered, and failure to mention it in any detail in the Gamble and Beck letters.

RVSI has therefore presented admissible evidence raising genuine issues of material fact both as to whether Svetkoffs work in 1990-91 contained every element of the ’152 claims, as required under Warner-Jenkinson, supra, and Richardson, supra, in order to prove anticipation, and as to whether Svetkoffs work, even if it did anticipate the ’152 invention, was “abandoned, suppressed, or concealed.” Summary judgment on the basis of previous invention under § 102(g) is therefore DENIED.

3. “Known or used by others” (§ 102(a)) and “On sale” bar (§ 102(b))

a. Legal standards

35 U.S.C. § 102 (a) and (b) provide that

Use of the patented invention by one who is neither the inventor nor under any obligation of secrecy to the inventor renders a patented invention “known or used by others” under § 102(a). Lockwood v. American Airlines, Inc., 107 F.3d 1565, 1570 (Fed.Cir.1997). Such use also renders a patented invention “in public use” under § 102(b). Baxter Intern., Inc. v. Cobe Laboratories, Inc., 88 F.3d 1054 (Fed.Cir.1996). A patented invention is “on sale” under § 102(b) when the inventor or another makes an offer to sell the invention with commercial intent. RCA Corp. v. Data General Corp., 887 F.2d 1056, 1059 (Fed.Cir.1989).

b. Discussion

View also argues that Don Svetkoffs 1990-91 work, specifically, the use of signal pads as coplanarity targets, was contemporaneously communicated (and so “known” under § 102(a)) and commercialized (put “on sale” under § 102(b)) when offered to customers, including IBM, by Chuck Gamble and Mike Beck of View’s sales force. The ’152 patent application was filed June 3, 1994, so the “critical date” for purposes of the “on-sale” bar under § 102(b) is June 3, 1993. As the offers to IBM took place before the critical date, View argues that these offers of sale invalidate the patent under both § 102(a) and (b).

RVSI argues that the communications in question did not include all of the elements of the asserted claims, as required under Richardson, supra. As such, they do not eonsti *1334 tute knowledge or use of the invention by others, or the placing of the invention on sale, under § 102(a) and (b). Because the communications were primarily directed to the solder paste measurement capabilities of the 7530, the letters did not describe Svetkoffs activities in sufficient detail. Furthermore, RVSI argues that View’s alleged offers to sell the 7530 could not constitute an offer to sell the patented invention, because at no time prior to the critical date of the ’152 patent could any 7530 machine perform all the steps of the 152 claims.'

RVSI has presented the following admissible evidence raising genuine issues of material fact as to whether the Gamble and Beck letters constitute “knowledge] or use by others” under § 102(a) or a placing of the invention “on sale” under § 102(b):

(1) the solder paste algorithms mentioned by the Gamble and Beck letters did not compute a difference between a measured height of a pad and a calculated threshold plane, as required by Claim 1 (Kelley Depo., pp. 63-65); and

(2) View machines with these algorithms were not available until at least 1992 (Kelley Depo., pp. 59, 63-65).

The second of these items of evidence is relevant, despite 1992 being prior to the critical date for the ’152 patent, because it supports RVSI’s characterization of the statement in the Gamble letter, “The 7530 system is capable of using local fiducials, bare copper pads, or solder mask covered traces for Z referencing,” (Svetkoff Decl., Ex. D, p. 51) as a mere sales pitch, and a sales pitch “for a feature that was not even implemented in the View machine” (RVSI’s Opp., p. 7).

Even assuming, arguendo, that Svetkoffs work had duplicated the 152 invention, RVSI presents admissible evidence raising a genuine issue of material fact as to the public use or knowledge of the patented invention:

(3) the Gamble letter does not describe Svetkoffs methods, for example, it does not mention coplanarity or warpage calculation (Svetkoff Decl., Ex. D);

(4) the Gamble letter does not mention Svetkoffs testing (Svetkoff Decl., Ex. D; Svetkoff Depo., p. 220);

(5) the Beck letter does not describe the May, 1990 and September, 1991 analyses (Svetkoff Decl., Ex. E);

(6) the Beck letter does not contain all the steps of Svetkoffs testing (Svetkoff Depo., pp. 204-5);

(7) to the extent that the Beck letter communicated the Svetkoffs four corner fiducial methodology, it also communicated its abandonment, as the Beck letter reads, “Based on these finding[s] [volume measurement errors of approximately 10%] the SVS reference height algorithm is being revised to better fit the true board surface and result in the most accurate volumetric measurements” (Svetkoff Decl., Ex. E, p. 54); and

(8) the Beck letter does not explain or claim that the 7530 will compute substrate eoplanarity or perform substrate coplanarity inspection (Svetkoff Decl., Ex. E);

From this evidence, taken together, it can be inferred that what Svetkoff learned was neither “known or used by others” under § 102(a) or put “on sale” before the critical date under § 102(b). Therefore, View’s motion for summary judgment of. the invalidity of the ’152 patent under §§ 102(a) and (b) is DENIED.

C. View’s Obviousness Argument

1. Legal standards

35 U.S.C. § 103 provides that “[a] patent may not be obtained ... 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 the subject matter pertains.”

“It is black letter law that the ultimate question of obviousness is a question of law.” Richardson-Vicks, Inc. v. Upjohn Co., 122 F.3d 1476 (Fed.Cir.1997), citing Graham v. John Deere Co., 383 U.S. 1 , 86 S.Ct. 684 , 15 L.Ed.2d 545 (1966). “At the same time, it is well understood that there are factual issues underlying the ultimate obviousness decision. These so-called Graham factors include (1) the scope and content of the prior art; (2) the differences between the *1335 claimed invention and the prior art; (3) the level of ordinary skill in the art; and (4) certain secondary considerations.” Richardson-Vicks, supra at 1479 . The question of what constitutes ordinary level of skill in the pertinent art, for purposes of determining whether an invention is obvious and whether a patent is valid, is one of fact. Graham, supra 383 U.S. at 17 ; AMP, Inc. v. Fujitsu Microelectronics, Inc., 853 F.Supp. 808 (M.D.Pa.1994). The failure of others to arrive at the patented invention is a “secondary consideration” tending to show nonobviousness. Panduit Corp. v. Dennison Mfg. Co., 810 F.2d 1561 (Fed.Cir.1987).

2. Discussion

a. Prior art

RVSI does not dispute that measurement of BGA ball top eoplanarity, employing measurement of the distance from the tip of the ball to the reference surface or seating plane, was known in the prior art. (’152 Patent, Col. 3, ll. 4-29.) It is also undisputed that the following elements of the ’152 claim were known in the prior art: selecting ball tops on BGAs as targets and determining the best fit plane (Bilodeau Depo., pp. 24-25); collecting ball height data on BGA devices and calculating the coplanarity of the balls (Jacovino Depo., p. 54); selecting QFP leads as targets, measuring their heights, generating a best fit plane to the lead height data, and measuring the deviation from each lead tip to the best fit plane (Bilodeau Depo., pp. 19-20.) The Patent Office granted the ’152 patent in light of prior art concerning BGA ball top eoplanarity and flattening of ball tops to facilitate optical inspection. (RVSI’s Opp., Ex. F.)

RVSI has presented admissible evidence that methods for measuring the eoplanarity of ball tips on BGA devices was known and considered by the Patent Office through disclosure in the ’152 patent specification. (’152 patent, col. 3, ll. 4-46.)

b. Differences from prior art

View relies in part on the testimony of Shayla Fahey to support its claim of the obviousness of the T52 invention. Fahey has testified that the use of signal pads, rather than ball tops is “the only arguable difference between the admitted prior art and claims 1, 3, and 12 [of the ’152 patent].” (Fahey Decl. ¶ 18b.) On this basis, together with her own level of skill in the art, she' concluded that “claims 1, 3 and 12 would have been obvious to one of ordinary skill in the art in 1991, and certainly in 1993.” (Fa-hey Decl. ¶¶ 17,19.)

RVSI argues that despite Fahey’s statement that the ’152 invention was obvious, her own work did not include all the elements of the ’152 invention, and therefore her work does not demonstrate that the innovations of the ’152 patent Were obvious to her. RVSI also argues that Fahey is not a person “of ordinary skill in the art,” so that even if the ’152 patent were obvious to her, it would not be invalid under § 103 (see c., infra).

RVSI has presented admissible evidence raising a genuine issue as to whether the ’152 invention was obvious to Fahey:

(1) Fahey did not provide “opaque fiducial index pads correlated with neighborhood pad heights,” or “disposing any index pads in a pre-arranged pattern over a substrate,” as required by Claim 1 (Bilodeau Decl. If 34); and

(2) “none of the documentation provided by Ms. Fahey indicates that she used only signal pads in a determination of the eoplanarity of substrates” (Bilodeau Decl. ¶ 34);

c. Ordinary skill in the art

The determination of whether a patent is invalid under § 103 crucially relies on what would be obvious to “a person having ordinary skill in the art.” 35 U.S.C. § 103 . The level of ordinary skill in the pertinent art is a question of fact under Graham and AMP, supra. In this case, the parties are in dispute over the factual question of what constitutes ordinary skill in the art. RVSI adopts the level of ordinary skill specified by View’s witnesses, Shayla Fahey and Svetkoff, together with RVSI’s witness, Steven Bilodeau, president of the Electronics Division of RVSI. RVSI has presented the following admissible evidence of ordinary skill in the art:

(1) one who “had knowledge of mathematics, a person probably with a bachelor’s of science degree, not necessarily in engineering, *1336 but in some sort of technology, computer science, perhaps” (Fahey Depo., pp. 57-58);

(2) “applications engineers that worked for View” (Fahey Depo., pp. 57-58);

(3) “[e]ngineers that, including application engineers, that work in the machine vision industry in measurement algorithm development” (Svetkoff Depo., p. 117);

(4) “a person who would implement the inventions on a regular basis” (Bilodeau Decl. ¶30); and

(5) “applications engineers]... [who] generally have at most a college education or the equivalent through their experience in the device inspection field” (Bilodeau Decl. ¶ 30.)

View in its motion instead defines Fahey, Svetkoff, and Bilodeau as themselves “reflecting] the level of ordinary skill in the art.” (View’s Mot., p. 21.) Not only does View reject the testimony of its own witnesses in favor of a much higher level of skill: “advanced degrees, and all have considerable experience” (View’s Mot., p. 21), it does so without evidentiary support.

Fahey states in her Declaration that “The level of skill in the art is high.” (Fahey Decl. ¶ 18c.) It seems likelier that she is referring to her own level of skill and that of the inventors of the ’152 patent, whom she mentions in ¶ 18c, than describing an “ordinary” level of skill. When specifically asked about the “ordinary” level of skill, she testified as set out above as (1) and (2).

Where only the non-movant has presented admissible evidence on a material factual issue, summary judgment on that issue in favor of movant must be DENIED.

d. Failure of others

Under Panduit Corp., supra, the failure of others to combine every element of the patented invention is a secondary consideration tending to support nonobviousness. RVSI has presented admissible evidence that others failed to solve the problem addressed by the 152 invention:

(1) the work Fahey did resembled initial work done by RVSI which yielded “unsatisfactory results” (Bilodeau Decl. ¶ 34); and

(2) Fahey did not avoid using the substrate as a height reference, the very problem the 152 invention was designed to avoid (Bilodeau Decl. ¶ 34; Fahey Depo., p. 110);

(3) even four years after View claims it practiced the steps of the 152 patent, View application engineers were unable to configure the Model 880 to inspect the eoplanarity of BSMs (Baldauf Depo., pp. 95-96; RVSI’s Opp., Ex. J and K).

More should be said about the third item of evidence. View application engineer Russ Baldauf has given contradictory testimony respecting the obviousness of the 152 patent. In Baldauf s October 8, 1997 Declaration, he testified that “it was obvious to me that I could determine the eoplanarity of the BSM pads in the same manner that I determined the eoplanarity of the BGA ball tops.” (Baldauf Decl. ¶ 6.) However, in a Memorandum dated May 17, 1995, Baldauf reported, “I have no knowledge of how to set up a BSM on the 880 system.” (RVSI’s Opp., Ex. J, p. 86.) In deposition testimony given March 5, 1997, he confirmed this; in answer to the question, “Is it fair to say that as of May 17, 1995 ... you had no knowledge of how to set up the inspection of a BSM device on an 880 machine?,” Baldauf answered, “That is correct.” (Baldauf Depo., p. 95.) In answer to the question, “Do you think that no other applications engineer [at View] had that knowledge as of May 17, 1995?,” Baldauf again answered, “That is correct.” (Baldauf Depo., p. 96.) In a Memorandum dated August 21, 1995, Baldauf reported that he was unable to perform successful BSM inspections on an 880 machine, and that “resident expert Mr. Warren Lin” was equally unsuccessful. (RVSI Opp., Ex. K.) In other words, by his Declaration Baldauf has testified that something he was unable to do, did not know how to do, and believed no one else knew how to do, was nevertheless “obvious.” The August 21, 1995 Memorandum describes his and Lin’s failure; his Declaration simply contains a conclusory statement of obviousness. Baldaufs earlier testimony, together with other admissible evidence presented by View, is sufficient to raise a genuine issue of material fact as to the nonobviousness of the ’152 invention as suggested by the secondary consideration of failure by others.

*1337 e. Conclusion as to obviousness

RVSI has presented admissible evidence tending to show that a genuine issue of material fact remains as to the nonobviousness of the 152 patent; specifically, that the patent was granted by the Patent Office in light of the prior art View claims invalidates it; that the 152 invention was different from that prior art; that the level of ordinary skill in the art is low enough that obviousness to someone like Fahey does not invalidate the patent; and that the failure of others to arrive at the patented invention supports its nonobviousness. Therefore, View’s motion for summary judgment of the invalidity of the 152 patent on grounds of obviousness is DENIED.

IV. Conclusion

RVSI has presented admissible evidence raising genuine issues of material fact as to the invention of the 152 patent by another, the abandonment or suppression by any pri- or inventor, the use or knowledge of the 152 patent by others, the placing of the invention on sale, and its obviousness to one of ordinary skill in the art. Therefore, View’s motion for summary judgment of invalidity on grounds of anticipation under § 102 and obviousness under § 103 is hereby DENIED. IT IS SO ORDERED.