US 9404802

I'll research the specific patent US 9,404,802, focusing on its claims (which the provided full text truncates before) and checking for any 2026 CAFC activity.

Let me pull the actual claim set and the abstract, and check for any 2026 CAFC activity.

Let me try to retrieve the granted claims and front-page abstract directly from the patent PDF.

US Patent 9,404,802 B2 — Summary

Scope note: I was asked to search the USPTO database and CAFC 2026 dockets for the specific number 9404802 and to ignore similar numbers. I did that. Two up-front honesty flags, because the rules require them:

1. I could not retrieve the issued claim text of the '802 from the sources available to me. The authoritative full text you supplied is truncated mid‑specification (it ends in the middle of the "inversion / Newton‑Raphson" discussion), and my web searches returned claim sets for sibling applications (11/452,879 → US 7,791,761; 15/155,488 → US 9,894,338), not the '802 (application 14/702,369). I will not present sibling claims as the '802's claims.
2. I ran a CAFC/2026 docket check and found no Federal Circuit activity for US 9,404,802. That is a "no results found" negative from the searches I could run, not a proven universal negative. It is consistent with the patent being expired (see below).

---

Bibliographic data (authoritative — from the patent text you supplied)

Assignment event (recorded): 2016‑05‑09, assignor HOLUB, RICHARD A. → RAH COLOR TECHNOLOGIES LLC, "ASSIGNMENT OF ASSIGNORS INTEREST" (reel/frame not retrieved in the earlier analysis).

---

Abstract

Caveat — I do not have the verbatim front‑page abstract of the '802. The (57) ABSTRACT does not appear in the truncated full text supplied. What I can say authoritatively is the invention's own opening statement, quoted from the supplied text:

"The present invention relates to a system (method and apparatus) for distributing and controlling color reproduction at multiple sites, and particularly to, a system for distributing and controlling the color output of rendering devices, such as color monitors, proofing devices, and presses, at multiple sites or nodes of a network to provide a uniform appearance of color within the output colors attainable at each rendering device. The invention utilizes a color measurement instrument associated with each rendering device for obtaining color calibration data for calibrating the rendering device. The system is controlled by computers at each node and utilizes a data structure, referred to herein as a Virtual Proof, to store and distribute color transformation information in the network. Color image data representing one or more pages or page constituents can be distributed to the nodes separately from the Virtual Proof."

Related‑family abstracts (NOT verbatim '802 — use only as a close paraphrase of the shared disclosure; from US 8,917,394 / US 2016/0261839, members of the same Holub/RAH family):

"In a color imaging system, multiple rendering devices are provided at different nodes along a network. Each rendering device has a color measurement instrument for calibrating the color presented by the rendering device. A rendering device may be a color display in which a member surrounds the outer periphery of the screen… A rendering device may be a printer in which the measuring of color samples on a sheet… is provided by a sensor coupled to a transport mechanism… where the sensor provides light from the sample to a spectrograph. The color measuring instruments provide for non‑contact measurements of color samples… and are self‑calibrating by the use of calibration references in the instrument."

I flag this as paraphrase from sibling members, not a certified '802 abstract.

---

Independent claims — plain‑language overview

Material uncertainty: I could not obtain the granted claim set of US 9,404,802, so I cannot give you claim numbers, claim counts, or the verbatim scope of its independent claims. Any claim‑by‑claim statement here would be fabrication. What follows is a concept‑level overview of the subject matter the independent claims are directed to, derived from the '802's own specification (supplied text) and cross‑checked against how the same family drafted its independent claims. Treat it as a description of the invention's claimed subject‑matter categories, not as quoted claim language or a claim count.

The specification and the family's practice indicate independent claims fall into these categories:

1. A networked color‑reproduction system. A network of nodes, at least some of which each have a rendering device (color monitor, proofing device, or press) plus a color measurement instrument (CMI). The system obtains color calibration data characterizing the output colors (colorants) of each node's rendering device, and produces, responsive to that data, transformation information to convert input color image data into device‑specific output data so that colors appear substantially the same within the output colors attainable by each device. The "Virtual Proof" data structure is central: it stores/distributes the transformation information and is separable from the color image data (shared components vs. local components).

2. A system/method for controlling color reproduction across multiple sites using a computer at a site with memory storing tonal transfer curves for a plurality of color channels, one or more color transformations (converting a first set of coordinates into a second set), and a data structure expressing colors in/out of gamut, plus a network interface communicating that information to another site using a network protocol.

3. A color‑measurement apparatus for a rendering device — the self‑calibrating, non‑contact CMI: e.g., a dual‑beam spectrograph (illumination‑reference beam + sample beam with a dark reference and a known‑wavelength reference) or a linear fiber‑optic array feeding a spectrograph, with calibration references built into the instrument.

4. A display‑mounting apparatus — a member (cowel) surrounding the display screen periphery with a spaced, angled sensor/optics picking up screen light (avoiding specular reflection), used to calibrate the display for soft proofing.

5. A method of maintaining color‑display calibration — adjusting for ambient light with the screen dark; neutral‑balancing the display; measuring per‑channel gamma; and adjusting display output accordingly (stated as especially useful for CRT displays).

6. A printer/transport‑based measuring apparatus — a sensor coupled to a transport mechanism moving the sensor and printed sheet relative to each other, feeding light to a spectrograph.

Confidence: the categories above are well supported by the supplied specification text. The mapping of categories to specific independent claim numbers is NOT verified. Given the family's history (the sibling '761 had claims in the 55–89 range; US 8,917,394 issued with 29 claims; US 8,760,704 with 42 claims), the '802 likely carries a large, multiply‑independent claim set — but I cannot state the count or numbers.

---

CAFC 2026 docket / litigation status

• No AIA trial (IPR/PGR/CBM) on the '802 was found in the earlier-generated sections, and my searches likewise surfaced none. The PTAB activity in this family (Adobe v. RAH, IPR2019‑00627/‑00628/‑00629/‑00646) targets sibling patents, not the '802.
• No 2026 Federal Circuit docket for "9,404,802" was found. Searches for RAH Color Technologies Federal Circuit appeals returned nothing tying to this number.
• Litigation of record (from the patent page's own links): RAH Color Technologies LLC v. Microsoft, E.D. Tex. 4:20‑cv‑00176 (filed 2020‑03‑05; voluntarily dismissed 2020‑03‑11) and E.D. Tex. 2:20‑cv‑00088 (filed 2020‑03‑24); and RAH Color Technologies LLC v. Samsung, N.D. Ill. 1:17‑cv‑05430 (filed 2017‑07‑25). These are 2020‑and‑earlier filings, not 2026 activity.
• Practical point: the patent is recorded as expired (anticipated 2018‑08‑25). An expired patent cannot be asserted prospectively, which is the most likely reason there is no live 2026 appellate docket.

---

Identifier‑disambiguation warning (per the "interpret literally" rule)

The bare string "9404802" returns several entirely unrelated documents, and these must not be conflated with US 9,404,802:

• GB 9404802.2 — a UK catalyst application (Thomson/Sweetman), cited in US 5,817,896, WO 96/23573, and KR 100451027B1 (Pd/ZnO catalyst for CFC hydrogenolysis). This is the single most common false hit.
• SE 9404802 — a Swedish priority number in the PCT/SE97/02214 family (AGA AB).
• FR 9404802, PCT/US1994/004802 (= WO 94/24906, a product‑distribution system) — other unrelated uses.

None of these are the color‑reproduction patent. My summary above is confined to US 9,404,802 B2 only.

---

What I could not confirm (and how to close it)

1. Verbatim granted claims of the '802 and the exact claim count — check USPTO Patent Center / Patent Public Search for application 14/702,369 or the issued PDF's claims. (https://patentcenter.uspto.gov)
2. The '802's verbatim (57) ABSTRACT — on the first page of the issued PDF; my abstract above is the specification's own opening language plus a sibling‑family paraphrase.
3. The negative on any 2026 CAFC docket — confirm via the Federal Circuit's docket search / PACER for "RAH Color Technologies" as appellant/appellee; I found no link to '802 but cannot prove the universal negative.

I'll research litigation involving US 9,404,802 specifically, using litigation-tracking sources. Let me run targeted searches.

Good — I'm finding specific '802 assertions. Let me dig deeper into the Samsung, Quad/Graphics, and Microsoft cases, and check the Unified Patents portal.

Let me verify the Microsoft Sherman Division case, the Fujifilm and Sony cases, and the Samsung case outcome.

Litigation Involving U.S. Patent No. 9,404,802 ("System for distributing and controlling color reproduction at multiple sites")

Bottom line: Yes — litigation is known. The '802 was asserted in three identified district‑court actions: two against Microsoft (E.D. Tex., 2020 — both voluntarily dismissed within weeks) and one against Samsung Electronics (N.D. Ill., filed 2017 — appears to have resolved by license/settlement). I found no PTAB proceedings and no Federal Circuit activity on the '802, consistent with the earlier‑generated sections and with the patent's recorded expiry.

---

1. Confirmed cases asserting the '802

Case 1 — RAH Color Technologies LLC v. Samsung Electronics Co., Ltd., N.D. Ill. 1:17‑cv‑05430

• Patents‑in‑suit (five): U.S. 8,279,236; 9,404,802; 9,516,288; 8,638,340; 7,830,546. (Complaint, ¶20.)
• '802 claims actually asserted: claims 1, 4, and 8 (Counts V–VII). This is a useful, verifiable data point that partially closes the claim‑identification gap flagged in the earlier sections. The complaint quotes:
  • Claim 4: "The system according to claim 1 wherein said interface comprises a Universal Serial Bus interface and a network interface in which said network interface enables at least partly wireless communication employing one or more network protocols."
  • Claim 8: "The system according to claim 1 further comprising a program executed by one of said one or more processors that transforms said two‑dimensional images into calibrated RGB coordinates for storage in said file."
  • (Claim 1 itself was not quoted in the retrieved excerpt, but claims 1/4/8 as asserted and the Samsung accused‑product narrative — mobile devices with dual image sensors, an RGB ambient‑light sensor, a USB connector, wireless networking, and Exif 2.2/sRGB image files — imply claim 1 recites a camera/system with image sensors, processors, an interface, and a file storing image data.)
• Accused products: Samsung mobile devices (smartphones/tablets with cameras) and displays.
• Status: The action is not reflected as proceeding into 2018–2019. In the brief RAH filed in the parallel Quad/Graphics venue dispute (N.D. Ill. 1:17‑cv‑04931, Memorandum Opinion and Order, Jan. 16, 2018), RAH represented that "it has reached agreements with Sony and Samsung about the licensing of its patents." That representation indicates the Samsung matter was resolved by a license/agreement. I did not retrieve the docket's terminal entry or the exact dismissal date, so treat "settled/licensed" as strongly supported but not docket‑confirmed. (This is consistent with the "average license value under $1 million" characterization RAH gave the MDL court.)
• Sources: Complaint at N.D. Ill. 1:17‑cv‑05430 (available as Exhibit 2228 to IPR2019‑00628, "Complaint‑Samsung"); https://insight.rpxcorp.com/litigation_documents/[12548038](/patent/12548038) ; N.D. Ill. venue opinion (Jan. 16, 2018), https://storage.courtlistener.com/recap/gov.uscourts.ilnd.[341768](/patent/341768)/gov.uscourts.ilnd.341768.70.0.pdf

Case 2 — RAH Color Technologies LLC v. Microsoft Corporation, E.D. Tex. 2:20‑cv‑00088

• Patents‑in‑suit (five): U.S. 7,729,008; 6,995,870; 7,312,897; 9,404,802; 9,516,288.
• Venue basis alleged: Microsoft's Frisco, TX retail store (2601 Preston Rd., Ste. 1176) as a "regular and established place of business."
• Status: Voluntarily dismissed. Docket entries on the Unified Patents portal: NOTICE of Voluntary Dismissal by RAH (Alison Richards) entered 06/01/2020; ORDER OF DISMISSAL signed by Judge Rodney Gilstrap 6/2/2020. Complaint filed 2020‑03‑24, so the case lasted roughly ten weeks.
• Source: https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2:20-cv-00088

Case 3 — RAH Color Technologies LLC v. Microsoft Corporation, E.D. Tex. 4:20‑cv‑00176

• Patents‑in‑suit: five patents, including the '802, the '870, '897, '008 (per the Sherman Division complaint, which recites "36 patents" awarded to Dr. Holub and lists the '870, '897, '008, '802, and '288 as the Patents‑in‑Suit).
• '802 awareness allegations: The complaint recites pre‑suit licensing contacts (2009 NDA with Microsoft; 2015–2016 letters from Global IP Law Group identifying the '870, '897, '008).
• Status: Voluntarily dismissed (2020‑03‑11, per the previously‑generated section — flagged as not independently re‑confirmed in this session). The '802's inclusion in this Sherman Division complaint is directly supported by the complaint text at RPX insight (litigation_documents/13705327, "EASTERN DISTRICT OF TEXAS SHERMAN DIVISION").
• Source: https://insight.rpxcorp.com/litigation_documents/13705327 ; Google Patents litigation link, https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/4%3A20-cv-00176

---

2. Cases in the same RAH campaign where the '802's presence is NOT confirmed (do not attribute)

I checked these because they are commonly lumped with the '802 in third‑party roundups; I could not confirm the '802 in any of them, and in one case the retrieved complaint affirmatively omits it:

• RAH Color Technologies LLC v. Quad/Graphics, Inc., N.D. Ill. 1:17‑cv‑04931 (filed 2017‑06‑30), later transferred to E.D. Wis. 2:18‑cv‑00087. The complaint I retrieved lists the '870, '897, '008, '704, '546 and '357 patents — no '802. A July 2017 trade report ("Retail Patent Litigation Report") lists the '802 and '288 in a combined entry captioned "RAH Color Technologies LLC v. Quad/Graphics, Inc. … (multiple cases)" with both Quad/Graphics and Samsung as defendants, which best explains the '802 by the Samsung case, not the Quad case. Working conclusion: the '802 was not asserted against Quad/Graphics, but I flag that I could not read every amended complaint.
• RAH Color Technologies LLC v. Fujifilm Holdings Corporation et al., N.D. Ill. 1:17‑cv‑02924. Surfaced in the Stanford NPE litigation database as an RAH 2017 N.D. Ill. case; the '802's inclusion is unconfirmed (I could not retrieve its patent list before hitting tool limits).
• RAH/Sony matter. Referenced only indirectly (RAH's statement that it had "reached agreements with Sony and Samsung"). No case number, court, or '802 attribution retrieved. Unconfirmed.
• In re RAH Color Technologies LLC Patent Litigation, MDL No. 2874 / N.D. Cal. 3:18‑md‑02874‑SI. The MDL's patents‑in‑suit (per the Jan. 2019 Joint Preliminary Status Report) were the '399/'870/'897/'251/'008/'761/'444/'357/'704 family — the '802 is not listed among the MDL patents. The Microsoft Texas actions were pitched by RAH as "tag‑along" actions for the MDL, but were dismissed before consolidation.
• Adobe IPRs (IPR2019‑00627/‑00628/‑00629/‑00646). These target the '008, '444, '897 and '761 patents — not the '802. Confirmed again in the petition text ("The 'sole real party‑in‑interest is Petitioner Adobe Inc.'" challenging specific claims of the '897, etc.).
• Pre‑2016 suits (Seiko Epson 2010, Ricoh 2015, Xerox 2017, Heidelberger, Dalim, Kodak, X‑Rite, Adobe). These predate or do not include the '802 (which issued 2016‑08‑02); no '802 assertion found in any of their retrieved complaints.

---

3. Negative findings (stated as "not found," not as proven universal negatives)

• PTAB: No IPR, PGR, or CBM naming US 9,404,802 was found — consistent with the earlier PTAB section. (No § 315(e)(2) estoppel attaches to this patent.)
• Federal Circuit / CAFC 2026: No appeal, and no 2026 docket, tying to the '802 was found. Consistent with the patent's recorded status "Expired – Fee Related," anticipated expiration 2018‑08‑25 — an expired patent cannot be asserted prospectively, which is the most likely reason there is no live appellate activity.
• ITC / Section 337: No investigation involving the '802 was found.
• Later filings (2018–2026): No litigation asserting the '802 after the two 2020 Microsoft suits was found. Every '802 assertion I located is dated 2017 or 2020.

Limits on these negatives: they rest on web‑indexed sources (Unified Patents portal, RPX Insight, Docket Alarm, CourtListener, Stanford NPE database, Justia) rather than a direct PACER/PTAB E2E docket pull, and my tool budget was exhausted before I could run the remaining confirmation searches. I could not verify every month between 2016 and today, and I could not pull the Samsung and Microsoft terminal docket entries directly. To close these:

• PTAB E2E search by patent number 9,404,802: https://ptacts.uspto.gov/
• PACER / CourtListener for N.D. Ill. 1:17‑cv‑05430 (Samsung) terminal entry and E.D. Tex. 4:20‑cv‑00176 (Microsoft, Sherman) dismissal date.
• Unified Patents portal caselist: https://portal.unifiedpatents.com/litigation/caselist
• RPX Insight: https://insight.rpxcorp.com/

---

4. Practical read

The '802's litigation footprint is narrow and short‑lived: one 2017 device case (Samsung, resolved by license) and two 2020 Microsoft suits dismissed on the courthouse steps. It was never an MDL patent, was never IPR'd, and was never appealed. Combined with the 2018‑08‑25 anticipated expiration, any residual exposure is past‑damages only and capped by the six‑year § 286 lookback — and the two Microsoft complaints were dismissed within ~2–10 weeks of filing with no reported adjudication on the merits. The single most useful new fact for the earlier sections' open items is the verified assertion of '802 claims 1, 4, and 8 in the Samsung complaint, with the verbatim text of claims 4 and 8 quoted above.

I'll research PTAB proceedings on US 9404802, starting with the structured data's baseline and verifying via web search.

Now let me search specifically for any PTAB proceeding naming the '802 patent itself.

Let me verify whether any petition has ever targeted the '802 patent itself, and check the litigation context.

PTAB Proceedings on US 9,404,802 ("System for distributing and controlling color reproduction at multiple sites")

Bottom line up front: According to the structured USPTO Open Data Portal block provided with this task — the canonical source — there are zero AIA trial proceedings on US 9,404,802. My independent web searching corroborates that: I could not find a single IPR, PGR, or CBM petition naming the '802 patent. Every PTAB decision in this family that I could locate is on sibling Holub patents sharing the same specification (the '008, '444, '897, and '761 patents), not on the '802.

---

Proceedings overview

Total: 0 AIA trial proceedings on US 9,404,802 (0 active / 0 with claims invalidated / 0 with claims sustained / 0 settled / 0 institution denials). Because no petitioner ever challenged the '802 at the Board, there is no IPR estoppel, no claim cancellation, and no PTAB claim construction attaching to this patent — a defendant faces a completely untested patent whose validity record at the PTAB is empty, which is materially different from the "hardened by two survived IPRs" posture the prompt asks about.

Two things about the '802's posture matter more than the empty PTAB docket:

1. The patent is expired. The Google Patents record (fetched 2026-09-27) lists an anticipated expiration of 2018-08-25 and legal status "Expired – Fee Related." An expired patent cannot be infringed prospectively; any assertion of the '802 today is necessarily a claim for past damages, capped by 35 U.S.C. § 286's six-year lookback (i.e., roughly 2020-forward from a 2026 filing, potentially capturing part of the pre-2018-08-25 window only on a timely-filed case).
2. The '802 was asserted in litigation, not invalidated at the PTAB. It appears as a Patent-in-Suit in RAH Color Technologies' campaign (see "Related matters" below), which is why a defendant researching it will find MDL and E.D. Tex. activity but no PTAB paper trail. I found no evidence of a petition, so I will not invent one.

---

(No proceedings to list for US 9,404,802)

There is nothing to summarize at the proceeding level: no petition number, no petitioner, no institution decision, no FWD, no appeal. I want to be explicit about the limits of this statement — the ODP block is authoritative as of its ingest date, and my web searches did not surface an older or un-indexed petition. I could not independently confirm the negative for every month between the '802's issuance (2016-08-02) and today, so I flag it as "no proceedings on file," not a proven universal negative. The fastest way to close that gap is PTAB E2E / PTAB Center search by patent number: https://ptacts.uspto.gov/

---

Related (non-'802) PTAB proceedings — context only, NOT on this patent

These are the Adobe IPRs on sibling Holub patents sharing the '802's specification. Do not cite any of these as invalidating US 9,404,802. They are useful only because the art, the expert testimony, and the credibility findings are transferable to a future '802 challenge.

IPR2019-00646 — Adobe Inc. v. RAH Color Technologies, LLC (US 7,791,761 B2)

• Type: Inter Partes Review
• Filed: 2019-02-01 (petition)
• Status: Final Written Decision — some challenged claims unpatentable (Paper 119, dated 2020-10-08)
• Judge panel: Eleni Mantis Mercader, Frances L. Ippolito, John A. Hudalla
• Petition grounds: Claims 7–11, 13, and 15 of the '761 patent; § 102/§ 103 over ICC Profile Format v3.0 (Ex. 1013) and v3.01 (Ex. 1014), among others
• Institution decision: Instituted on all challenged claims and all grounds, 2019-09-04 (Paper 34). Panel: "we are persuaded that Petitioner has demonstrated a reasonable likelihood of prevailing in establishing unpatentability of at least one claim."
• Final Written Decision: Petitioner proved by a preponderance that claims 7, 8, 10, 11, and 13 of the '761 patent are unpatentable. Claim 15 was not held unpatentable — the N.D. Cal. MDL court later referenced "the '761 Patent Final Written Decision with regard to the finding about claim 15" (Case 3:18-cv-03277-SI, Doc. 120, 2021-01-22), indicating claim 15 survived.
• Settlement / termination: None found.
• Appeal: No Federal Circuit appeal confirmed in my sources — treat as unverified.
• Defensive value (sibling only): Shows Adobe successfully attacked a claims set drawn from the same specification using the ICC Profile Format as § 102 prior art, with the patentee's own admissions about the ICC spec doing much of the work. The claim 15 survival is the cautionary data point.

IPR2019-00629 — Adobe Inc. v. RAH Color Technologies, LLC (US 7,312,897 B2)

• Type: Inter Partes Review
• Filed: 2019-02-01
• Status: Final Written Decision — claims unpatentable
• Judge panel: Same consolidated panel (Mercader, Ippolito, Hudalla)
• Petition grounds: § 103 combinations built on Sugiura (US 6,177,934) and Beretta, with ICC v3.0/v3.01
• Final Written Decision: "Petitioner has demonstrated by a preponderance of the evidence that claims 32, 33, and 37 of the '897 patent are unpatentable."
• Appeal / settlement: None confirmed.
• Defensive value (sibling only): Another Holub independent claim killed on the ICC-spec + Sugiura/Beretta theory.

IPR2019-00627 — Adobe Inc. v. RAH Color Technologies, LLC (US 7,729,008 B2)

• Type: Inter Partes Review
• Filed: 2019-02-01
• Status: Final Written Decision — patent owner prevailed
• Judge panel: Docket lists Frances Ippolito, John Hudalla, Nathan Engels, J. John Lee, Eleni Mantis Mercader
• Petition grounds: Claims 28–33 and 36–38; § 103 over ICC v3.0/v3.01, Beretta, and the UCR User Guide
• Final Written Decision (public version, ~2020-09): "we determine that Petitioner has not demonstrated by a preponderance of the evidence that claims 28–33 and 36–38 of the '008 patent are unpatentable."
• Notable procedural history: Consolidated oral hearing held 2020-05-28 across -627, -628, -629, -646; extensive motions practice (two motions to exclude, a motion to strike, a motion for supplemental information); the Board rejected RAH's real-party-in-interest theory, finding Quad/Graphics, Dalim, Xerox, Heidelberger, and EFI were not RPIs.
• Defensive value (sibling only): Demonstrates the Board did not rubber-stamp Adobe's gamut-descriptor theories — claim-tier-specific outcomes matter.

IPR2019-00628 — Adobe Inc. v. RAH Color Technologies, LLC (US 8,416,444 B2)

• Type: Inter Partes Review
• Filed: 2019-02-01
• Status: I could not confirm the final disposition from the sources I retrieved. Consolidated with -627/-629/-646 for hearing. Do not assume an outcome for this one.

Common context for all four (not '802):

• Petitioner counsel: Perkins Coie LLP (James F. Valentine; Wing H. Liang). Patent owner counsel: Global IP Law Group LLC (Alison A. Richards; Irwin Park; David Berten).
• Art relied on: ICC Profile Format v3.0 / v3.01, Stewart, Beretta, Scott-Taggart, Sugiura (US 6,177,934), and the UCR User Guide.
• Parallel district court: RAH Color Technologies LLC v. Adobe Inc., N.D. Cal. 3:18-cv-03277-SI, consolidated in MDL No. 2874; the MDL was stayed 2019-09-10 pending the IPRs.

---

Strategic summary

Cancelled vs. sustained vs. untested — as to US 9,404,802 specifically: every claim is UNTESTED. No claim of the '802 has been cancelled by the PTAB, and no claim has been confirmed by the PTAB either. Contrast that with its siblings: the '897 lost claims 32, 33, 37; the '761 lost claims 7, 8, 10, 11, 13 while claim 15 survived; the '008 survived the challenge outright. Those outcomes tell you nothing adjudicative about the '802's own claim set — which, as a continuation in the same family, I could not fully resolve from the material provided, so I am not going to state which '802 claims were asserted in the E.D. Tex. campaign or which mirror the invalidated siblings. Verify the '802 claim set directly before relying on any theory.

Estoppel landscape: there is essentially no IPR-driven estoppel to worry about. 35 U.S.C. § 315(e)(2) estoppel attaches only to grounds a petitioner "raised or reasonably could have raised" in an IPR on that patent. Because the '802 was never the subject of an IPR, no § 315(e)(2) estoppel reaches the '802 at all — for Adobe, for the MDL defendants, or for anyone. Separately, and usefully: the Board's finding in the -627/-629/-646 FWDs that Quad, Dalim, Xerox, Heidelberger, and EFI were not Adobe's RPIs cuts against treating those defendants as Adobe's privies, so Adobe's IPR work product and grounds are not estoppel-protected against them on the sibling patents. Two practical constraints remain for any new IPR on the '802: (i) § 315(b)'s one-year bar runs from service of a complaint alleging infringement of the '802, so the 2020-filed E.D. Tex. actions likely place those served defendants beyond the window; and (ii) the '802's expired status means the Board may apply Phillips claim construction rather than the Phillips-like approach — an IPR on an expired patent is permissible but of diminishing practical value. The real lever for a defendant today is § 102/§ 103 in district court, not the PTAB.

Pattern signals. The same petitioner (Adobe) filed a coordinated four-petition campaign — a textbook "one-big-defendant-files-for-the-group" pattern — but it targeted four other Holub patents, not the '802. There is no evidence of a defensive aggregator (e.g., Unified Patents) in this chain. RAH/Holub is a one-man licensing operation (Dr. Holub is the sole member of RAH Color Technologies per his own declaration, Ex. 2121 in IPR2019-00646), and the patentee litigated the IPRs hard (motions to exclude, motions to strike, RPI discovery fights, a joint-defense-agreement discovery motion) — but I found no confirmation of any Federal Circuit appeal from any of the four FWDs, so I will not assert one. The family-wide outcome is mixed: one patent fully survived, two were partially or fully invalidated, one is unconfirmed.

---

Recommended next steps

1. Do not treat this as "hardened." The absence of PTAB activity on the '802 is a genuine gap, not a validation. It most likely reflects that defendants in the 2018–2020 MDL concentrated fire on the higher-value siblings, and that the '802 expired (2018-08-25) before an IPR on it made economic sense. Confirm current status on Google Patents / USPTO Patent Center before relying on expiration.
2. Verify the docket on PTAB E2E by patent number to close the negative: https://ptacts.uspto.gov/ (search "9,404,802"). Also check the litigation links in the structured block — E.D. Tex. 4:20-cv-00176, E.D. Tex. 2:20-cv-00088 (RAH Color Technologies LLC v. Microsoft Corporation, filed 2020-03-24), and N.D. Ill. 1:17-cv-05430 — to see which '802 claims were actually asserted and when each defendant was served (that date drives the § 315(b) window).
3. If you are a served defendant with § 315(b) time remaining, an IPR is available on an untested patent, and the Adobe record is directly reusable ammunition: the ICC Profile Format v3.0 (Ex. 1013) and v3.01 (Ex. 1014) were held to be § 102 prior art against claims from this exact specification, in part because Dr. Holub's own submissions to the PTO and his litigation conduct bound RAH to that admission (see the -646 briefing). The panel's reasoning on authenticating and applying the ICC spec — and RAH's expert/credibility defeats — is portable. Budget for § 325(d) discretion if the same art was before the examiner.
4. If the patent is expired and your exposure is past damages only, weigh the cost of an IPR against a district-court invalidity case, and note that the six-year § 286 lookback plus the 2018-08-25 expiration sharply caps the damages period. For an expired, never-IPR'd patent, a strong § 102/§ 103 summary-judgment or Markman strategy may beat a PTAB filing on cost-effectiveness.
5. Confirm the unverified items before drafting anything: (a) the '628 ('444) FWD outcome; (b) whether any of the four Adobe FWDs was appealed to the Federal Circuit; (c) whether a later, un-indexed petition ever named the '802. I could not confirm (a)–(c) from the sources available to me, and I would rather flag them than guess.

Sources for the sibling proceedings cited above:

• IPR2019-00646 FWD (claims 7, 8, 10, 11, 13 unpatentable), Paper 119, 2020-10-08: https://www.docketalarm.com/cases/PTAB/IPR2019-00646/
• IPR2019-00627 FWD (claims 28–33, 36–38 not shown unpatentable), public version: https://bannerwitcoff.com/wp-content/uploads/2020/09/PTAB-IPR2019-00627-124.pdf
• IPR2019-00629 FWD (claims 32, 33, 37 unpatentable): https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1523317](/patent/1523317)
• N.D. Cal. MDL order referencing the '761 FWD and claim 15, Case 3:18-cv-03277-SI Doc. 120, 2021-01-22: https://storage.courtlistener.com/recap/gov.uscourts.cand.[327437](/patent/327437)/gov.uscourts.cand.327437.120.0.pdf
• US 9,404,802 patent record: https://patents.google.com/patent/US9404802/en

I'll research the assignment chain for US 9404802 using search, then cross-check litigation and entity records.

Good foundation. Now let me pin down the recorded assignments (reel/frame) and any later transfers.

Ownership Reconstruction — US 9,404,802 B2

"System for distributing and controlling color reproduction at multiple sites"
App. 14/702,369 · filed 2015-05-01 · issued 2016-08-02 · priority 1997-08-25 · Applicant/Assignee of record: RAH Color Technologies LLC (Alexandria, VA)

Data-quality caveat up front. I was able to confirm that a post-filing assignment exists (Holub → RAH Color Technologies LLC, recorded 2016-05-09, per Google Patents legal events), but I was not able to retrieve the reel/frame numbers or the correspondent of record in this session — the Assignment Center is a JavaScript application that my search tooling could not render, and no indexed mirror surfaced the reel/frame for this specific patent. I will not invent reel/frame numbers. Where the field is required but unknown, it is marked [NOT RETRIEVED]. Verify at the USPTO Assignment Center search page: https://assignmentcenter.uspto.gov/ (search patent 9404802 or app. 14/702,369) or https://assignment.uspto.gov/patent/index.html.

---

Inventors

• Holub's employment history is on the record in his own litigation filings and 2020 IPR deposition (Adobe v. RAH Color Technologies, IPR2019-00627/-00628/-00629/-00646): Eikonix Corp. → Eastman Kodak (1983–1990), Agfa/Compugraphics (1990–1993), SuperMac Technologies (1993–1994). He left SuperMac in 1994, before the 1996/1997 priority filings, and worked out of rented space at RIT Research Corp. in Rochester, NY. So at the critical date he was a true independent inventor — not an employee-inventor whose rights would have flowed to an operating company.
• He did not incorporate until Imagicolor Corporation (1998), and that entity "eventually dissolved" without shipping a product (RAH complaints, e.g. N.D. Ill. 1:17-cv-05430 ¶¶19–20; E.D. Pa. complaint ¶¶14–15).
• Unusual pattern — none of the classic fire-sale kind. There are no co-inventors to depart, no employer co-assignee to abandon the portfolio, and no 12-month inventor exodus. The relevant anomaly is the inverse: a single inventor personally owning a 30–35 patent portfolio and later assigning it into his own single-member LLC — a self-monetization structure, not a distress structure.

---

Original assignee

RAH Color Technologies LLC (listed on Google Patents as "RAH Color Tech LLC"; the front page of sibling family members, e.g. US 8,279,236, reads "Assignee: Rah Color Technologies LLC, Rochester, NY (US)").

• Entity type: Virginia limited liability company, single-member, owned and managed by Dr. Holub personally. Earlier address of record Rochester, NY; address used in litigation: 7012 Colgate Drive, Alexandria, Virginia 22307 — a residential address, not a registered-agent commercial mail drop.
• Primary line of business: patent licensing and enforcement only. Its own complaints state it was "formed in 2005 as a vehicle for an on-going licensing program for companies whose products depend on Dr. Holub's innovations" and that "RAH Color Technologies is owned by Dr. Richard A. Holub, who is a named inventor of all its patent assets."
• Did it ship a product embodying the claims? No. The only commercialization attempt was Imagicolor Corporation (1998), which was dissolved after investment failed. RAH itself has no manufacturing, no product revenue, and no distribution.
• Revenue model as stated in its own MDL filing: ~27 companies licensed over ten years, average license value "less than $1 million" (In re RAH Color Technologies LLC Patent Litig., 18-md-02874-SI, Joint Preliminary Status Report).
• Enforcement record (not product-competitor suits): Seiko Epson (W.D.N.Y. 10-cv-06710, 2010); Ricoh, Xerox, Adobe, Quad/Graphics, R.R. Donnelley, Heidelberg, Kodak, X-Rite, EFI (declaratory judgment); Microsoft (E.D. Tex. 4:20-cv-00176 and 2:20-cv-00088, March 2020, both voluntarily dismissed); MDL 18-md-02874 (N.D. Cal. → E.D. Va.).
• Current status of the patent: Google Patents lists "Expired - Fee Related," anticipated expiration 2018-08-25 (20-year term from the 1997-08-25 priority plus PTA). US 9,404,802 is therefore no longer enforceable, and Google's "Current Assignee" field still shows RAH Color Technologies LLC — i.e., it was never sold on. The assignee appears to remain an active licensing entity, but this particular asset is dead.

---

Assignment timeline

Recorded assignments found for US 9,404,802 / app. 14/702,369: one post-filing record, plus one family-level predecessor record that is reported but not verified against this application number. Reel/frame and correspondent are missing — see caveat above.

• Executed date not shown in retrieved source / recorded 2016-05-09 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment (Google Patents legal event title: "ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: HOLUB, RICHARD A. (sole assignor)
  • Assignee: RAH COLOR TECHNOLOGIES LLC
  • Correspondent: [NOT RETRIEVED] — cannot assess recurrence. (Litigation-side counsel of record worth cross-checking against the recording correspondent: Irwin I. Park, Park Vaughan Fleming & Dowler LLP, who states in IPR2019-00628 that he has handled RAH's licensing/litigation for ~6 years; Alison Aubry Richards, who filed RAH's Rule 7.1 corporate disclosure statements in the 2020 E.D. Tex. cases; David Berten (IL Bar #6200898), who appears as RAH's counsel in the MDL; and Global IP Law Group, identified in the Kodak complaint as RAH's licensing counsel.)
  • Context: Confirmatory/clean-up assignment of this specific continuation application from the sole inventor into his own pre-existing licensing LLC, recorded ~3 months before the patent issued (2016-08-02). Not a third-party acquisition.
  • Source: https://patents.google.com/patent/[US9404802B2](/patent/US9404802B2)/en → Legal Events ("2016-05-09 Assigned to RAH COLOR TECHNOLOGIES LLC … Assignors: HOLUB, RICHARD A.").

• Reported, not verified against this application — circa 2005–2006 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment (family-level, presumed)
  • Assignor: Richard A. Holub (individually)
  • Assignee: RAH Color Technologies LLC
  • Basis: RAH was formed in 2005; family members issuing in 2006–2010 (e.g. US 7,075,643; US 6,995,870; US 7,312,897; docketed to "Rah Color Technologies Llc") carry RAH as assignee, while 1996–2004 family members (US 6,043,909; 6,157,735; 6,459,425; 6,750,992) issued in Holub's name alone. That step-change implies a Holub → RAH assignment in the 2005–2006 window covering then-pending continuations. I could not confirm a recording lodged against 14/702,369, which was parented-in later; that is why the 2016 record above exists at all.
  • Context: Internal reorganization — inventor assigns portfolio to his own newly formed licensing vehicle.
  • Flag: If the Assignment Center shows only the 2016-05-09 record against this application number, treat the 2005–2006 entry as family background rather than a link in this patent's chain.

Not found / no record: no security agreement, no merger, no change-of-name, no release, no correction, and no transfer to any third-party acquirer or aggregator on US 9,404,802.

---

Timeline diagram

timeline
    title Ownership of US 9404802
    1997 : Family priority date
    1998 : Inventor forms Imagicolor Corporation
    2005 : RAH Color Technologies LLC formed
    2015 : Continuation application filed
    2016 : Holub assigns rights to RAH Color Technologies
         : Patent US 9404802 issues
    2017 : Portfolio suits filed in Illinois
    2018 : Anticipated term expiration
         : Portfolio consolidated in multidistrict litigation
    2020 : Texas suits filed against Microsoft

---

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT (in substance, inventor-controlled form).
   The patent was never held by an operating company; it went from an individual inventor into his own licensing-only LLC. The substantive tells are documented, not inferred from naming: RAH admits in its own complaints that it exists "as a vehicle for an on-going licensing program"; it is a single-member LLC whose sole member is Holub; it has no products in commerce; and its office of record is a residential address (7012 Colgate Drive, Alexandria, VA). Corresponding record: the 2016-05-09 assignment, Holub → RAH Color Technologies LLC (reel/frame [NOT RETRIEVED]). Distinguishing note: this is not the classic purchased-shell pattern (no "IP Holdings" name bought in from a bankrupt operating company, no third-party acquisition). It is the inventor's own enforcement vehicle.

2. Known asserter in the chain — PRESENT via the Unified/RPX clause, NOT via the named roster.
   RAH Color Technologies does not appear on the Acacia / Marathon / IV / IPNav / Wi-LAN / Conversant / Vringo / Pendrell / Innovatio / MPHJ / Lumen View / Round Rock / DGC / Spangenberg lists. It is independently surfaced as a high-frequency filer by the sources the task points to: the Google Patents page itself carries Unified Patents litigation-portal entries for E.D. Tex. 4:20-cv-00176 and 2:20-cv-00088, and RPX Empower maintains dedicated RAH litigation pages (https://litigation.rpxcorp.com/litigation/txedce-[196195](/patent/196195)-rah-color-technologies-v-microsoft). Cumulative docket: suits against Seiko Epson, Ricoh, Xerox, Adobe, EFI, Heidelberg, Kodak, X-Rite, Quad/Graphics, R.R. Donnelley and Microsoft across ~2010–2020.

3. Repeat correspondent across the chain — UNCLEAR (cannot be scored).
   There is only one assignee-to-assignee link, so recurrence is not demonstrable. The recording correspondent for the 2016-05-09 entry is [NOT RETRIEVED] — this is the single highest-value field still missing, and the one that would most cleanly confirm or kill the "same lawyer runs all the paper" hypothesis. Do not treat the litigation-counsel names above (Berten, Park, Richards, Global IP Law Group) as assignment correspondents without checking the recorded document.

4. Cascading transfers — NOT PRESENT.
   One recorded transfer, into an entity that has held the asset continuously since 2016 (Google's "Current Assignee" still reads RAH Color Technologies LLC). No chained LLCs, no <24-month sequence.

5. Pre-litigation transfer — NOT PRESENT.
   The 2016-05-09 assignment predates the first portfolio suits by ~1 year (N.D. Ill. 1:17-cv-05430, 2017) and the MDL (2018) and Texas filings (2020) by 2–4 years, and its assignee is the assignor's own pre-existing company — it was not created to set venue or manufacture standing. (Separate optics point, not a scored signal: RAH, a Virginia LLC, filed against Microsoft in E.D. Tex. in March 2020 and voluntarily dismissed both cases within ~90 days.)

6. Bankruptcy fire-sale — NOT PRESENT.
   No Chapter 7/11 sale of this asset. Imagicolor Corporation's failure was a wind-down, not a §363 sale, and Eastman Kodak appears in this story only as a defendant to RAH's infringement claims (N.D. Ill. 6:19-cv-06112), never as a seller of these patents.

7. Privateering — NOT PRESENT.
   Privateering requires an operating company to hand patents to an NPE that asserts on its behalf. Here the operating company never existed — Holub self-enforces, and he is the patent's author, not a corporate sponsor's proxy.

8. Defensive aggregator (anti-NPE) — NOT PRESENT.
   The chain does not terminate at RPX, AST, LOT, Unified Patents or OIN. Inverse conclusion: the patent has not been neutralized by aggregation; it died by term expiration (anticipated 2018-08-25, status "Expired - Fee Related" per Google Patents) while still in the asserter's hands.

---

Verdict

NPE — high confidence.

Two strong, independently documented signals drive this: (i) the assignee of record, RAH Color Technologies LLC, is a non-practicing, licensing-only single-member LLC that says so in its own pleadings, holds the patent by virtue of the 2016-05-09 assignment from its sole member Richard A. Holub (reel/frame [NOT RETRIEVED]), has no products, and operated from a residential Alexandria, VA address; and (ii) it is a serial plaintiff tracked by Unified Patents and RPX (E.D. Tex. 4:20-cv-00176 / 2:20-cv-00088; MDL 18-md-02874), asserting a low-value-per-license portfolio. Note the mitigating nuance for accuracy: this is inventor-owned, self-monetized enforcement of the inventor's own patents, not a third-party shell cascade or a bankruptcy fire-sale — signals 4, 5, 6, 7 and 8 are all not present, and the asset is now expired and unenforceable.

Verify the two open fields here: https://assignmentcenter.uspto.gov/ (search 9404802 / app. 14/702,369) — confirm the 2016 assignment's reel/frame, execution date, and correspondent attorney, and whether any earlier 2005–2006 Holub → RAH recording was lodged against this application number.

I'll research US Patent 9404802 and its cited prior art. Let me start with searches.

Let me gather the specific citation data from the patent's front page.

Let me look at the Justia and FreePatentsOnline pages which typically list "References Cited" distinctly.

Important caveat up front (read this first)

I must be straight with you about two limits on this task, because the rules you set require me not to fabricate:

1. I searched public patent databases via web search (Google Patents, FreePatentsOnline, Justia, PTAB/PTACTS filings) — not the USPTO PatentCenter search interface directly. The USPTO's PatentCenter and Global Dossier links for this patent are listed at the bottom.
2. I was not able to retrieve a clean, complete image of the front-page "(56) References Cited — U.S. Patent Documents" list for US 9,404,802. What the searches returned is a combined citation table on FreePatentsOnline that appears to mix examiner-cited prior art with forward citations ("Referenced By") and the patent's own family/continuation members. I therefore cannot certify for every entry whether it is § 102 prior art, a § 103 secondary reference, a family member, or a forward citation. Where I am unsure I say so.

I will not invent citations, dates, or claim mappings to fill that gap.

---

1. The patent, interpreted literally

Identifier-disambiguation warning (per your literal-interpretation rule). The numeric string "9404802" surfaces in several entirely unrelated documents and must not be conflated with US 9,404,802:

• GB 9404802.2 — a UK catalyst application (Thomson), cited in US 5,817,896 / US 5,899,678 and KR 100451027B1.
• FR 9404802 — a French application (search report dated Jan. 6, 1995).
• SE 9404802 — a Swedish priority number appearing in a PCT/SE97/02214 family.
• PCT/US1994/004802 = WO 1994024906 ("Pinch-actuated product distribution system"). Note the Google Patents family field literally prints "US9404802W" as an internal ID here — this is not the color patent.

None of these are US 9,404,802. I flag them because search engines returned them for the bare query.

---

2. Prior art expressly discussed in the '802 specification (authoritative — from the full text you supplied)

These are the references the patent itself identifies in its Background/"Prior Art" discussion. They are the most defensible "relevant prior art" list because they come straight from the specification (US 9,404,802 column text):

U.S. patents discussed:

• US 2,790,844 (Neugebauer) — system extending the Hardy–Wurzburg machine; representation of color imagery in a colorimetric (device-independent) coordinate system; soft proof by projecting onto the target paper stock. Relevant to the "soft proof / device-independent color" concepts.
• US 4,500,919 (Schreiber) — system to match a hard-copy proof to the monitor image.
• US 5,272,518 (Vincent) — portable spectral colorimeter for system-wide calibration (linearly variable spectral filter + linear sensor array). The specification notes Vincent "mentions applicability to insuring consistent color across a network, but does not discuss how distributed calibration would be implemented."
• US 5,107,332 (Chan) — closed-loop control of digital printers (desktop/inkjet; flatbed scanner in the loop).
• US 5,185,673 (Sobol) — same family of closed-loop desktop printer calibration systems.
• US 5,319,437 (Van Aken et al.) — spectrophotometer cited as an example of a standard-observer meter (SOM).

Non-patent literature discussed: Hardy & Wurzburg (J. Opt. Soc. Amer. 38:30); Neugebauer (TAGA 1956); Pobboravsky (TAGA 1962); Schwartz et al. (TAGA 1985); Gordon et al. (TAGA 1987); Boll (SPIE Vol. 2170, 1994); Kipphan et al. (on-press spectral color bars); Sodergård et al. (TAGA 1993/1995 and "A general teleproofing system," TAGA 1991); Garini (SpectraCube application note); Éouzan et al. (projection video display control); Bonino et al. (harmonizing color devices); ICC Profile Format v3.01 (May 8, 1995); Metric Color Tag (MCT) Spec, Rev 1.1d, 1993, Electronics for Imaging; CIE Publication 15.2, 1986; Holub, Kearsley & Pearson, J. Imaging Technology 14(2), 1988 (Parts I & II).

---

3. U.S. patent-document citation table retrieved (classification uncertain — see caveat)

The FreePatentsOnline citation table for 9,404,802 (https://www.freepatentsonline.com/9404802.html) returned the following. I present it verbatim as retrieved, without asserting that each is examiner-cited § 102 art. Entries marked (H) are Holub/RAH family members and are not third-party prior art:

Analytical read of this table:

• The Holub entries (8,416,444 → 6,044,909 / 6,157,735 / 6,750,992 / 6,995,870 etc.) are continuation/priority-family members, not third-party prior art. They are relevant to priority/§ 101–112 continuity, not to § 102 anticipation of a third party.
• The genuinely third-party references with apparent § 102/§ 103 relevance cluster around three themes: (a) distributed/multi-engine print & RIP routing (Barry, Zuber — 6,657,741; 6,633,396; 6,606,165; 6,636,326; 6,271,937; 6,219,155); (b) device-specific color profiles / colorimetric display characterization / interpolation (Schwartz 6,075,888; Rozzi 6,232,954; Kim 6,320,668; Walker 6,373,580; Kumada; Kuwata; Hidaka); and (c) double-beam spectrometry / incremental printer calibration (Barshad 6,204,919; Harrington 6,178,007).

I did not recover the older (pre-1997) § 102 references — e.g., the Vincent, Chan, Sobol, Van Aken patents discussed in the text — as explicit entries in the retrieved "(56)" block, though the specification cites them.

---

4. Anticipation analysis under 35 U.S.C. § 102 — hedged

Two structural problems with a definitive answer:

1. I do not have the issued claim set of US 9,404,802. The materials retrieved give the abstract, the specification, classifications, and a sibling patent's claim count (US 8,917,394 = 29 claims), but not the '802 claims themselves. Anticipation must be analyzed claim-by-claim (every element present in a single reference). Without the '802 claim text I cannot responsibly say "Reference X anticipates claim Y."
2. A granted patent's issued claims were, by definition, not anticipated by the art of record — the examiner allowed them over that art. So the correct frame is: which cited references could have supported a § 102 rejection, and which were overcome (often by amendment, as the PTAB petitions on the sibling '444 patent show). Any present-day § 102 argument would be made de novo in litigation, as in the E.D. Tex./N.D. Ill. cases noted on the front page.

General (non-claim-specific) mapping of the third-party art to claim categories of the '802 family:

On the specification-cited art, the strongest § 102 candidates are the ones closest to the claimed subject matter:

• Vincent US 5,272,518 — spectral colorimeter + express mention of network-wide consistent color. The '802 explicitly distinguishes it because Vincent "does not discuss how distributed calibration would be implemented" and has no self-calibration/verification. That distinction is exactly the kind of gap that defeats anticipation and shifts the argument to § 103.
• Chan US 5,107,332 / Sobol US 5,185,673 — closed-loop printer calibration; distinguished as desktop-only, scanner-in-loop, not imaging the image area.

Legal-standard note: Because the '802 claims priority to 1997-08-25, whether pre-AIA § 102(a)/(b)/(e) or AIA § 102(a)(1)/(a)(2) governs depends on whether the issued claims are actually entitled to that 1997 priority date and on the pre/post-March-16-2013 status of each reference. A reference published after the 1997 priority date but before the 2015 filing is only prior art if the claims lose the 1997 priority (a very live question in the litigation, given the RCE/claim-churn history visible in the PTAB petitions on the sibling '444 patent: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1523316](/patent/1523316) and .../1523296).

---

5. What I could not confirm, and how to confirm it

• The definitive "(56) References Cited" and "(10)/(45)" front-page data for US 9,404,802 — verify at:
  • USPTO PatentCenter: https://patentcenter.uspto.gov (application 14/702,369)
  • USPTO Global Dossier / Espacenet family view (linked from the Google Patents page)
  • The issued PDF: the "References Cited" block is on the patent's first page
• The full claim set of US 9,404,802 — needed for any real claim-by-claim § 102 mapping. Not in the material I retrieved.
• Which of the FreePatentsOnline table entries are examiner-cited vs. forward citations — the table's ordering (newest-first, dominated by Holub family members) strongly suggests it is (or is mixed with) the "Referenced By" list.

Bottom line: The most reliable "relevant prior art" for US 9,404,802 that I can substantiate is the set the patent itself discusses in its Background (Neugebauer US 2,790,844; Schreiber US 4,500,919; Vincent US 5,272,518; Chan US 5,107,332; Sobol US 5,185,673; Van Aken US 5,319,437, plus the TAGA/SPIE/ICC literature), and the third-party U.S. patents in the retrieved citation table (notably Schwartz US 6,075,888; Rozzi US 6,232,954; Walker US 6,373,580; Barshad US 6,204,919; Harrington US 6,178,007; the Zuber and Barry print-engine families). I cannot, on the evidence retrieved, tie any specific reference to a specific claim of the '802 under § 102 without the issued claim text, and I will not guess.

Sources: https://patents.google.com/patent/US9404802/en · https://www.freepatentsonline.com/9404802.html · https://patents.justia.com/patent/9404802 · https://ptacts.uspto.gov/ptacts/public-informations/petitions/1523316 · https://ptacts.uspto.gov/ptacts/public-informations/petitions/1523296

I'll verify the key prior-art references cited in this patent's specification before building the §103 analysis.

Obviousness Analysis — U.S. 9,404,802 B2 (35 U.S.C. § 103)

Analyst note / scope: Consistent with the earlier section, I could not retrieve the verbatim granted claims of US 9,404,802. The analysis below is therefore organized by claim concept (each a candidate independent/dependent claim grouping), derived from the specification's own "Summary" and "Definitions" text and from the family disclosure. Claim numbers are not asserted. Where a reference detail is drawn only from the patent's own description rather than independently verified, I say so.

---

A. Legal framework

Because the patent is expired, § 103 invalidity would matter only for past‑damages/retrospective purposes — but the analysis is unchanged.

---

B. The prior art of record (from the specification's own background/trailing citations)

---

C. Claim concepts and the § 103 mapping

Concept 1 — Networked system: multiple nodes, each with a rendering device + color measurement instrument; transformation info distributed separately from image data ("Virtual Proof," shared/local components)

Primary: Vincent '518. It already discloses (i) a colorimeter monitoring/controlling "one or more computer peripheral devices controlled by a computer" (cl. 15), (ii) a network connection (element 67) and a remote printer accessed over a network by a printer server, (iii) transmission of spectral/calibration data over the network to recalibrate the remote device, and (iv) color mapping means "transforming image data from a first color system ... to a second color system" (cl. 21) — i.e., transform files separate from the image.

Secondary A: ICC Profile Format v3.01 / MCT 1993 — supplies the standardized file carrying color transforms between devices, i.e., the "shared component."

Secondary B: Holub/Kearsley/Pearson 1988 — supplies per‑device input and output calibration transforms to a common colorimetric standard (device‑independent intermediary), which is what the shared component must contain.

Motivation (KSR‑sufficient):

• Vincent itself names the problem ("data sent along the network channel would be converted to a standard color data format, preferably based on ... CIE Lab") — an express motivation to standardize the transform file.
• Transmitting the compact transform rather than the bulky page image is the ordinary engineering response to known bandwidth constraints; the ICC spec's own "embedding profiles with image data" practice was criticized for "large data transfers over a network whenever profiles are updated," which is a design incentive to separate the two.
• All references are in the same field of endeavor (color reproduction and color measurement) → analogous art.

Assessment: Concept 1 is likely obvious over Vincent + ICC/MCT + Holub/Kearsley/Pearson. The only genuine differentiation — the two‑tier "prototype vs. production node" taxonomy and the explicit shared/local split of a single data structure — is largely naming and arrangement of known functions, absent evidence that the split yields an unexpected result.

---

Concept 2 — Per‑node calibration: rendering calibration forms, measuring them with the CMI, building a forward model, inverting it to a rendering/SEP transform, with gamut descriptor/operator data

• Chan '332: test pattern → measure → fit transforms → update LUT, all closed‑loop.
• Sobol '673: automated calibration image printing, automatic location, automatic coefficient calculation with little user intervention → directly meets the '802 object of "self‑ or auto‑calibrating, minimize user involvement."
• Holub/Kearsley/Pearson 1988 (I & II), Neugebauer 1956, Gordon et al. 1987: forward colorant→color models and inverse color→colorant transforms; gamut operators mapping input gamut into output gamut.
• Kipphan '721: spectral measurement + CIELAB comparison + ΔE‑based correction of the marking process.
• Johnson & Wichern: PCA to trim polynomial terms (the '802 specification expressly adopts this).

Motivation: Chaotic to run a printer calibration without a measurement device; Vincent '518 supplies the colorimeter and Van Aken '437 supplies a reference‑beam spectrophotometer for accuracy. Reducing operator skill is stated explicitly in the '802 background as a motivation for adding instrumentation — a market/design driver under KSR. Combining Chan's LUT‑update loop with Kipphan's spectral measurement (instead of Chan's scanner) is the predictable substitution of a more accurate, known element (spectral colorimetry) for a known less accurate one — the '802 specification itself concedes Chan/Sobol's scanner‑as‑sensor is limited and that Kipphan's spectral approach is superior.

Assessment: Likely obvious. The forward model (polynomial), its inversion (Newton/least‑squares/interpolation), and the gamut descriptor are taught or suggested collectively by Chan, Holub/Kearsley/Pearson, Gordon et al., and Neugebauer. The specific 17‑point (16‑interval) hypercube grid is a routine design choice, not a patentable distinction.

---

Concept 3 — Self‑/auto‑calibrating CMI with built‑in references (dual‑beam spectrograph; fiber array with light‑source reference, dark reference, and a known‑wavelength reference; non‑contact fiber‑optic pickup)

This is the strongest candidate for nonobviousness on a single‑reference basis, because the specification expressly says Vincent has "no provision for self‑checking of calibration" and Kipphan's colorimeter "is not capable of calibration self‑check." But the combination undercuts it:

• Van Aken '437 (dual‑beam spectrophotometer) supplies the reference beam concept (light source reflected from a known reflector compared to the sample beam) — precisely the "second fiber optic transmits light from the light source to the first input of the spectrograph."
• Vincent '518 supplies the dark reference (dark‑current compensation by blocking light) and the known‑wavelength reference — its filter‑alignment module transmits an Hg 546.1 nm line to identify which photosensor corresponds to which wavelength.
• Van Aken + Vincent also supply the fiber/array + spectrograph architecture.
• Sobol '673 supplies the minimal‑user‑intervention automation (the sheet‑insertion‑and‑measure workflow described in the '802 summary).

Motivation: The '802 background itself identifies the problem to be solved ("instrument‑to‑instrument variation precludes calibration of devices at different sites to a tolerance that will support confident, remote proofing") and names the failings of manual black traps/white plaques. Where the problem is expressly recognized in the prior art and the solution is the combination of known calibration techniques (dual beam + dark reference + known‑line reference), the combination is obvious under KSR absent unexpected results.

Assessment: Likely obvious as a combination of Van Aken + Vincent + Sobol. The "one‑dimensional array of fiber optics" ordering (reference / dark / wavelength standard / sample) is a routine arrangement of known reference elements.

---

Concept 4 — Display‑mounted sensor assembly (cowel surrounding the display periphery; sensor spaced at an angle; optics focusing to a sensor)

• Eouzan '481: camera retakes the projected image, i.e., a fixed/associated image‑capture device monitoring the display.
• Vincent '518, FIGS. 6–7 and cl. 3/23–24: a probe placed contiguous to the monitor screen, movable from patch to patch; measurement of gun transfer functions, gain, offset, gamma, and the explicit note that monitor calibrators are "electronically tethered to a host computer."
• Vincent FIG. 1B/1C/1D discloses a hood 26 "to shield the window ... from external stray light" — the functional equivalent of the '802 cowel. The '802's added refinement — angling the line of sight so the sensor does not see specularly reflected light off the faceplate — is a known optics technique (the specification notes the standard 45°/2°–10° incidence/reflection geometry, which appears in Vincent's text).

Motivation: Shielding ambient light and avoiding specular pickup are art‑recognized necessities for accurate display measurement; a POSITA mounting a Vincent probe to a CRT would predictably add a light‑shielding surround.

Assessment: Likely obvious over Vincent + Eouzan. The cowel geometry is a predictable mechanical design choice.

---

Concept 5 — Method of maintaining display calibration (adjust dark/ambient light level → neutral‑balance → measure per‑channel gamma → adjust display)

Every step is in Vincent '518:

• per‑channel gamma via measurement of each gun's transfer function (FIGS. 5; cl. 23–24);
• offset/gain = the dark level (the monitor‑calibrator discussion describes turning each gun on "over a sequence of successively greater brightnesses to determine the gamma, offset and gain");
• neutral balance = measurement of "the monitor's full intensity white produced by a fixed combination of the R, G and B colors";
• ambient handling via spectral compensation and comparison to a reference spectrum.

Bonino adds the step of adjusting the V/L functions of all devices to a common achievable maximum (harmonization). Eouzan '481 adds black‑level detection.

Assessment: Likely obvious — this is essentially Vincent's monitor‑calibration disclosure with method‑step claim formatting. Under pre‑AIA § 103, converting an apparatus disclosure into a method of using it is normally not patentable subject matter unless an unexpected result is shown.

---

Concept 6 — Imaging colorimeter ("imagical") for the press: image‑area colorimetry, spectral interpretation, variable focus/anti‑aliasing, illuminant substitution

• Sodergard et al. (1993/1995): frame grabbing from the image area of a moving web with a CCD camera.
• Garini/SpectraCube: complete visible spectrum at each pixel → supplies the spectral capability whose absence the '802 criticizes in Sodergard.
• Kipphan '721 / US 6,119,594‑type disclosure: monitoring both the control strip and the printed image area.
• Greivenkamp (anti‑aliasing) and Shnitser/Hoyt (tunable filters) supply the ancillary features as cited background.

Motivation: The '802 specification expressly states Sodergard's drawback is that "color control lacks the necessary precision" — that is a stated, recognized problem with a known solution (spectral imaging). Substituting a spectral sensor for a 3‑channel CCD to cure a known precision deficiency is the classic KSR "known technique to improve similar device in the same way" rationale.

Assessment: Likely obvious.

---

D. The obviousness bottom line (concept‑by‑concept)

---

E. Anticipated counterarguments and their weaknesses

1. "Teaching away." The '802 background criticizes Chan/Sobol (scanner‑based) and Sodergard (imprecise). But this is criticism of specific features, not a teaching away from the references' disclosures (automated calibration forms, LUT updating, image‑area frame grabbing). In re Fulton / KSR: disparaging a reference's accuracy does not teach away from adopting its other teachings, and here the '802 avoids the disfavored features by substituting Kipphan's/Van Aken's spectral measurement.
2. "Vincent lacks self‑check." True as to Vincent alone, but it is cured by Van Aken's reference‑beam and Vincent's own dark‑current and Hg‑line alignment structures — a combination of known elements.
3. "The Virtual Proof is a new data structure." A new file format whose only asserted advantages are transmission economy and re‑usability is generally obvious where the constituent transforms and the network are known (Vincent) and standardization is expressly suggested (ICC/MCT).
4. "Unexpected results / synergy." Nothing in the specification's data suggests results beyond the predictable sum of the parts (uniform appearance within each device's attainable gamut). This would need to be established with evidence.
5. Objective indicia. The record so far discloses licensing/asserted‑infringement activity (E.D. Tex. 4:20‑cv‑00176; 2:20‑cv‑00088 RAH v. Microsoft; N.D. Ill. 1:17‑cv‑05430) but no evidence yet of long‑felt need, failure of others, or industry praise tied to the '802's specific claims. Litigation alone is not a secondary consideration.
6. Prosecution/PTAB context (context, not proof). Earlier‑generated section noted the RAH Color Technologies v. Adobe, Nos. 21‑1067/-1068 Federal Circuit appeals (2021) involving IPRs against related family patents. Flag: I have not verified which family patents/claims were held unpatentable, and that is not admissible proof of the '802's invalidity; but if the family's IPR outcomes included § 103 holdings on the network/calibration concepts, they would be strong corroboration of the analysis above.

---

F. Confidence, caveats, and what would change the conclusion

• Caveat (material): this analysis is concept‑level, not claim‑level, because the granted claims were not retrieved. A single, tightly drafted limitation could change the outcome — e.g., a specific, non‑predictable numeric relationship among the shared/local file components; a specific reference‑fiber ordering with a claimed calibration algorithm producing a demonstrated accuracy improvement; or a specific cowel geometry with proven specular‑rejection performance.
• Caveat: Van Aken '437, Bonino, Garini, Sodergard, Neugebauer, and Gordon et al. were taken from the patent's own description (only Holub/Kearsley/Pearson 1988 was independently confirmed in search). Their content should be verified before use in a formal opinion.
• Recommended verification steps: (1) pull the granted claims via USPTO Patent Public Search/PatentCenter for US 9,404,802; (2) confirm the AIA/pre‑AIA determination against the benefit chain; (3) confirm the exact bibliographic content of Van Aken '437, Bonino, Garini, and Sodergard; (4) if a formal § 103 opinion is needed, run a claim chart against Vincent '518 + Chan '332 + Sobol '673 for the calibration concepts and Vincent '518 + ICC v3.01 for the networked‑transform concept, since those are the highest‑confidence grounds.

Net assessment: on the concepts disclosed and on the prior art the applicant itself placed of record, US 9,404,802 is highly vulnerable to a § 103 challenge, with the strongest grounds being (i) Vincent '518 as the base reference for both the networked color‑reproduction system and the display‑calibration method, and (ii) Chan '332/Sobol '673/Kipphan '721/Holub et al. 1988 for the print‑side calibration, modeling, and inversion features — all combined for the reasons of art‑recognized problem, predictable result, and design incentive identified above.