US 10854683

I'll search for authoritative information on this patent number and its litigation history.

Let me search specifically for the litigation and any 2026 Federal Circuit activity involving this patent.

Let me check the specific Federal Circuit appeals and PTAB proceeding listed for this patent.

US 10,854,683 B2 — Verified Bibliographic Summary

The following is drawn from the patent's own front page (USPTO/Google Patents full text supplied), cross-checked against search results. All identifiers are transcribed literally.

Family / related US continuations: US 9,818,803 (parent), US 11,594,578, US 11,980,077, US 12,324,335, and US 2025/0268051 A1 (pending). Foreign counterparts include EP 2637209 B1, KR 10-1615332 B1, JP 6143043 B2, CN 203260586 U, TW 201338152 A.

Abstract (verbatim): "A pixel arrangement structure of an OLED display is provided. The pixel arrangement structure includes: a first pixel having a center coinciding with a center of a virtual square; a second pixel separated from the first pixel and having a center at a first vertex of the virtual square; and a third pixel separated from the first pixel and the second pixel, and having a center at a second vertex neighboring the first vertex of the virtual square."

Independent Claims — Plain-Language Overview

The patent has 15 claims; the independent claims are 1, 13, 14, and 15. (Claims 2–12 depend from claim 1; claim 2 in turn is the parent of claim 3, and claim 11 of claim 12.)

Claim 1 — the core pixel-layout claim. A pixel arrangement for an OLED display with three kinds of pixels arranged "consecutively" along two crossing lines:

• A first pixel sits at the crossing point.
• A pair of second pixels sits on opposite sides of that first pixel along a first line, and that same line continues on to another first pixel — i.e., the line runs first-pixel / second-pixel / first-pixel / second-pixel / first-pixel, with all their centers on the line.
• A pair of third pixels sits on opposite sides of the same first pixel along a second line, which likewise continues to an additional first pixel, and which crosses the first line at the first pixel.

Additional requirements: (a) the center-to-center distance between the two second pixels is greater than the distance between a second pixel and a neighboring third pixel; (b) the first pixels emit green light, a color different from the second and third pixels; (c) the first pixels are smaller than at least one of the second or third pixels; and (d) each second pixel has a larger area than each third pixel. Plainly: a green-emitter at the center of an X of larger blue/red emitters, with blue bigger than red, and the blue-to-blue spacing larger than the blue-to-red spacing.

Claim 13 — same geometry, narrower wording. Identical structural recitation to claim 1 (lines, consecutive arrangement, crossing at the first pixel, blue-blue distance > blue-red distance, green first pixels, first pixels smaller than something). The differences from claim 1 are only in phrasing: it says the first pixels "are configured to emit green light" without the "different color" qualifier, and recites "smaller size" instead of "smaller." Same final requirement that each second pixel is larger in area than each third pixel.

Claim 14 — the "spacing + convex polygon" variant. Same three-pixel/consecutive-line geometry, but the numeric tests change and new shape language is added:

• The separation between adjacent first pixels (edge-to-edge, stated to be a shortest distance) is greater than the separation between a first pixel and a neighboring second or third pixel.
• First pixels emit a color different from the second and third pixels, and have a different size from at least one of them.
• At least one of the first, second, or third pixels is a convex polygon having five or more sides.
• Both distances are expressly defined as shortest distances between pixels.
• Each second pixel still has a larger area than each third pixel.

Claim 15 — near-duplicate of claim 1. Same elements and same numeric relation (blue-blue distance > blue-red distance; green first pixels; first pixels smaller than at least one of the others; second pixels larger in area than third pixels). The only meaningful difference is drafting: the preamble/body states up front that the first and second lines "pass through respective centers of the first pixel and the second pixels" (respectively the third pixels) before repeating the consecutive-arrangement language.

Notable drafting observation: independent claims 1, 13 and 15 substantially overlap, and all four independents now carry color limitations ("first pixels emit green," and in claims 1/13/15 that blue area > red area), which do not appear in the FIG. 1 description but align with the FIG. 2 embodiment (where "the second pixels 200 that emit blue" are given larger area to address blue's shorter lifetime). The claims thus read on a green-center / blue-red-surround, blue-larger-than-red arrangement.

Litigation and Post-Grant Status (what the searches actually establish)

ITC — Investigation No. 337-TA-1351. "Certain Active Matrix Organic Light-Emitting Diode Display Panels and Modules for Mobile Devices, and Components Thereof." Complaint filed Dec 28, 2022 by Samsung Display; investigation instituted Jan 27 / published Feb 3, 2023. The notice of investigation expressly lists U.S. Patent No. 10,854,683 ("the '683 patent") among the four asserted patents (with the '803, '599 and '593 patents) and identified claims 1, 2, 4–10, and 13 of the '683 patent as asserted. Source: Federal Register 88 FR 7463 (Feb 3, 2023); USITC notice dated Mar 19–25, 2025.

All asserted '683 claims were terminated from 337-TA-1351. The March 2025 Commission notice states that when the final initial determination issued, "only claims 5 and 21 of the '803 patent, claims 5, 10, 17, 40-41, and 47 of the '578 patent, claims 2-3, 13, and 15-16 of the '599 patent, and claim 6 of the '593 patent remain asserted in the investigation as a result of termination of all asserted claims of the '683 patent and certain other asserted claims." So the '683 patent dropped out of the ITC case well before the final ID (Nov 15, 2024, finding no violation on domestic-industry grounds).

District court and appellate dockets (per the Google Patents litigation tab on the '683 record): US case filed in Texas Eastern District Court (case 2:23-cv-00309); US case filed in Virginia Eastern District Court (1:25-cv-00908); a second ITC filing, 337-TA-3661; and appeals to the Court of Appeals for the Federal Circuit (docket nos. 25-1518 and 25-1521) together with a PTAB case, IPR2023-00988.

PTAB IPR2023-00988. This petition (petitioner listed as Unified Patents) is flagged on the '683 record, and I did not obtain a verified final disposition for that specific IPR. Unified Patents' public PTAB litigation-entry URL is https://portal.unifiedpatents.com/ptab/case/IPR2023-00988. I cannot confirm from authoritative sources whether any '683 claims were cancelled or survived; the ITC docket references Board final written decisions only for the '578 and '803 patents.

Global settlement (Nov 2025). Multiple sources (Lexology, IPR Daily, TheElec, BOE investor-relations disclosures) report that in November 2025 Samsung Display and BOE reached a comprehensive settlement of their OLED patent and trade-secret disputes and jointly moved the ITC to terminate the related investigations. The ITC received the joint withdrawal request on Nov 18, 2025 and terminated. Reporting indicates all asserted claims of the '683 patent had already been terminated from the investigation earlier.

CAFC in 2026 — what I could NOT verify. You asked specifically about 2026 Federal Circuit activity. My searches did not surface any 2026 CAFC docket entry, opinion, or oral-argument notice naming US 10,854,683. The only Federal Circuit numbers I could associate with this patent family are 25-1518 and 25-1521, which appear on the patent's Google Patents record without further detail. Given the Nov 2025 Samsung–BOE global settlement, these appeals may well have been dismissed or held in abeyance, but I have no authoritative document confirming that, and I will not speculate. Treat all 2026 appellate status for this patent as unverified.

Explicit Uncertainty

1. No independent confirmation of IPR2023-00988's outcome for the '683 patent.
2. No confirmed 2026 Federal Circuit activity for the '683 patent; the CAFC numbers I found (25-1518, 25-1521) lack verified docket detail in my sources.
3. The Chinese Supreme Court invalidation appeal (Beijing BOE v. CNIPA, with Samsung as third party, heard Sept 16, 2026 per IPR Daily) does not identify the patent number in the public reporting, so I cannot tie it to US 10,854,683 — do not assume it involves this patent.
4. Claim-scope characterizations above are my own paraphrase of the claim text; they are not claim constructions.

Primary sources used: the patent's own text/claims (as supplied); https://patents.google.com/patent/US10854683/en (including its litigation and family tabs); https://patents.justia.com/patent/10854683; Federal Register 88 FR 7463 (Feb 3, 2023) https://regulations.justia.com/regulations/fedreg/2023/02/03/2023-02246.html; USITC notice https://www.usitc.gov/system/files?file=secretary/fed_reg_notices/337/337_1351_notice03192025sgl.pdf; Lexology, "Three Years of Strategic Confrontation Conclude: Implications of the BOE-Samsung OLED Patent Dispute" (Feb 27, 2026).

I'll research litigation involving US 10854683 using litigation-focused sources.

Let me research the ITC investigations and the more recent cases.

Now let me research the PTAB proceeding and Federal Circuit appeals.

Litigation involving US 10,854,683 ("the '683 patent")

Bottom line: Yes — US 10,854,683 has been asserted in a coordinated set of proceedings by Samsung Display against BOE (and, in the ITC, against a group of US mobile-device repair-parts suppliers). The '683 patent is one of five patents (US 7,414,599; 9,330,593; 9,818,803; 10,854,683; 11,594,578) asserted together in the same disputes. Every case below involves the same patent family (priority date 2012‑03‑06; the '683 patent is a continuation of US 9,818,803).

1. Samsung Display Co., Ltd. v. BOE Technology Co., Ltd. and Mianyang BOE Optoelectronics Technology Co., Ltd.

• Plaintiff: Samsung Display Co., Ltd. (Korea)
• Defendants: BOE Technology Co., Ltd.; Mianyang BOE Optoelectronics Technology Co., Ltd. (later amended to add additional BOE affiliates)
• Jurisdiction: U.S. District Court, Eastern District of Texas (Judge Rodney Gilstrap; Magistrate Judge Roy S. Payne)
• Case No.: 2:23-cv-00309 (full number 2:23-cv-00309-JRG-RSP)
• Filed: June 26, 2023 (amended complaint June 28, 2023)
• Patents asserted: US 7,414,599; US 9,330,593; US 9,818,803; US 10,854,683; US 11,594,578
• Outcome/status: Stayed on July 24, 2023 under 28 U.S.C. § 1659(a) pending the co-pending ITC Investigation No. 337-TA-1351. Voluntarily dismissed without prejudice — notice of voluntary dismissal filed May 30, 2025; order dismissing the case entered June 2/4, 2025 on the docket. No merits determination.
• Sources: https://www.courtlistener.com/docket/67529099/samsung-display-co-ltd-v-boe-technology-co-ltd/ ; https://www.patsnap.com/fr/resources/blog/litigation/samsung-display-v-boe-technology-oled-patent-dismissal-patsnap/

2. Certain Active Matrix Organic Light-Emitting Diode Display Panels and Modules for Mobile Devices, and Components Thereof — ITC Inv. No. 337-TA-1351

• Complainant: Samsung Display Co., Ltd.
• Respondents: 17 US-based mobile-device repair/replacement-parts suppliers, including Apt-Ability LLC d/b/a MobileSentrix; Mobile Defenders, LLC; Injured Gadgets, LLC; Wholesale Gadget Parts, Inc.; Phone LCD Parts LLC and Parts4LCD; Group Vertical, LLC; Electronics Universe, Inc. (Fixez.com/Repairs Universe); LCTech International Inc.; Sourcely Plus, LLC; eTech Parts Plus LLC; Parts4Cells Inc.; Captain Mobile Parts, Inc.; DFW Imports LLC; Mengtor Inc.; Gadgetfix Corp.; plus the Office of Unfair Import Investigations. Mianyang BOE Optoelectronics Technology Co., Ltd. intervened as a respondent on March 22, 2023.
• Jurisdiction: U.S. International Trade Commission (Section 337)
• Investigation No.: 337-TA-1351
• Filed / instituted: Complaint filed December 28, 2022 (supplement January 17, 2023); instituted February 3, 2023 (88 FR 7463–64)
• Patents asserted: '803 patent (claims 1–5, 19–21); '683 patent (claims 1, 2, 4–10, 13); '599 patent (claims 1–18); '593 patent (claims 1–3, 6–8, 14–22); '578 patent added later
• Outcome/status: Two respondents (MobileSentrix and Mobile Defenders) were terminated on consent order; ten respondents were found in default; Injured Gadgets, Wholesale Gadget Parts, and Phone LCD Parts ("BLF Respondents") and BOE remained active. An ALJ initial determination was issued in November 2024 — press accounts characterize it as finding infringement of four Samsung patents but not recommending exclusion/cease-and-desist relief (one Chinese-language account instead reports a finding of no Section 337 violation; I could not resolve that conflict from the sources retrieved). The Commission's final determination proceedings continued into March 2025, per the Federal Register notice of March 25, 2025.
• Sources: https://www.govinfo.gov/content/pkg/FR-2023-02-03/pdf/FR-2023-02-03.pdf ; https://www.usitc.gov/secretary/fed_reg_notices/337/337_1351_notice01082024sgl.pdf ; https://www.govinfo.gov/content/pkg/FR-2025-03-25/pdf/FR-2025-03-25.pdf

3. Samsung Display Co., Ltd. v. [BOE Technology Group Co., Ltd.](/litigations/by-plaintiff/BOE%20Technology%20Group%20Co.%2C%20Ltd.) et al.

• Plaintiff: Samsung Display Co., Ltd.
• Defendants: BOE Technology Group Co., Ltd.; Mianyang BOE Optoelectronics Technology Co., Ltd.; Ordos Yuansheng Optoelectronics Co., Ltd.; Chengdu BOE Optoelectronics Technology Co., Ltd.; Chongqing BOE Optoelectronics Technology Co., Ltd.; Wuhan BOE Optoelectronics Technology Co., Ltd.; Yunnan Invensight Optoelectronics Technology Co., Ltd. (f/k/a BMOT / Kunming BOE Display Technology)
• Jurisdiction: U.S. District Court, Eastern District of Virginia (Judge Rossie D. Alston, Jr.; Magistrate Judge Lindsey R. Vaala)
• Case No.: 1:25-cv-00908 (1:25-cv-00908-RDA-LRV)
• Filed: May 28, 2025
• Patents asserted: US 9,818,803; US 10,854,683; US 11,594,578; US 7,414,599; US 9,330,593
• Outcome/status: Stayed June 25, 2025 pending the final determination in ITC 337-TA-1351; answer deadline set 90 days after the stay is lifted. Dismissed with prejudice by stipulation — stipulation filed November 19, 2025; so ordered December 2, 2025. Case closed.
• Sources: https://www.courtlistener.com/docket/70497017/samsung-display-co-ltd-v-boe-technology-group-co-ltd/

4. Certain ... (Samsung Display / BOE) — ITC Inv. No. 337-TA-3661

• Parties: Samsung Display Co., Ltd. as complainant, with BOE entities as respondents (based on contemporaneous reporting of a second Samsung Display ITC action against BOE filed around April 2025, running parallel to a second E.D. Tex. suit). I could not independently verify the respondent list or the patents asserted in this investigation from the sources retrieved.
• Jurisdiction: U.S. International Trade Commission
• Investigation No.: 337-TA-3661
• Filed: 2025 (exact date not verified)
• Outcome/status: Not verified.
• Sources: Google Patents family-litigation listing for US10854683 (https://patents.google.com/patent/US10854683/en); MLex report on Samsung's second E.D. Tex. suit and parallel ITC action (https://www.law360.com/mlex/articles/[2329838](/patent/2329838)/samsung-files-us-patent-infringement-suit-against-boe-over-oled-screens)

5. Appeals — U.S. Court of Appeals for the Federal Circuit

• Appeal Nos.: 25-1518 and 25-1521
• Filed: 2025. These are the two Federal Circuit appeals identified in the family-litigation data for the '683 patent family. At least one of them arises from the ITC's final determination in Investigation No. 337-TA-1351; an E.D. Va. filing states that "that appeal remains pending, and Mianyang BOE is a party to the appeal."
• Outcome/status: Pending as of the date of the records retrieved; no decision found.
• Sources: https://portal.unifiedpatents.com/litigation/Court%20of%20Appeals%20for%20the%20Federal%20Circuit/case/25-1521 ; https://portal.unifiedpatents.com/litigation/Court%20of%20Appeals%20for%20the%20Federal%20Circuit/case/25-1518 ; https://storage.courtlistener.com/recap/gov.uscourts.vaed.[574882](/patent/574882)/gov.uscourts.vaed.574882.5.0.pdf

6. PTAB proceeding (administrative, not district-court litigation, but directly on the '683 patent)

• Proceeding: IPR2023-00988 (inter partes review)
• Petitioner: Not identified in the source data I retrieved (the family-litigation entry attributes the data set to Unified Patents, not necessarily the petitioner).
• Status: A Final Written Decision is recorded for this proceeding. I could not verify the claims challenged or the outcome.
• Source: https://portal.unifiedpatents.com/ptab/case/IPR2023-00988

Additional context: Chinese-language reporting indicates that BOE, together with TCL CSOT, Tianma, and Visionox as co-petitioners, filed IPR petitions against all five Samsung Display patents asserted in the 1351 Investigation, and that all five were instituted. Those petitions would appear on the PTAB docket alongside IPR2023-00988, but I could not confirm the individual proceeding numbers for the '683 patent.

Cautions and limits of this report

• The '683 patent itself has never been the subject of a standalone infringement suit. It appears only as one of five co-asserted patents in the Samsung Display v. BOE campaigns.
• No case against the '683 patent reached a merits judgment in district court. Both district-court actions were stayed pending the ITC and then resolved by dismissal (without prejudice in E.D. Tex.; with prejudice in E.D. Va.).
• Unverified items: the outcome of ITC 337-TA-1351's final determination; the parties/patents/status of ITC 337-TA-3661; the substance of the two Federal Circuit appeals; and the petitioner/outcome in IPR2023-00988. I am flagging these rather than guessing. Authoritative confirmation would require the ITC EDIS docket, PTAB P-TACTS/PatentsView APIs, and PACER/CM-ECF dockets for the two appeals.
• I did not identify any other litigation (e.g., against Samsung's competitors, retailers, or non-BOE parties) involving US 10,854,683.

I'll research the PTAB proceedings for this patent, starting with the one identified on the patent's own Google Patents page.

Let me dig deeper into IPR2023-00988 and related proceedings.

Let me try more targeted searches on the PTAB case and related IPRs.

I was able to verify one flagged AIA proceeding but could not retrieve the case-level record (institution decision, FWD, or claim-by-claim disposition) for it. I'm reporting the record as it exists rather than filling gaps. Here is the analysis.

Proceedings overview

Total count: 1 flagged AIA proceeding (IPR2023-00988), against a canonical ODP count of zero — the structured "PTAB proceedings on file" block returned no AIA trial proceedings for US10854683, but the patent's own Google Patents litigation record carries a PTAB case entry for IPR2023-00988 with the status label "(Final Written Decision)," sourced from Unified Patents PTAB Data (US10854683B2; Unified Patents case record). Bottom-line defensive posture: the patent is not "hardened," but neither is it dead. It has one PTAB case on record that ran to a Final Written Decision, and — critically and independently — every asserted claim of the '683 patent was terminated out of ITC Investigation No. 337-TA-1351 (ITC notice, 2025-03-25). A defendant today faces a patent that Samsung Display has already stopped pressing in one forum and that has been through a full PTAB trial in another.

Verification caveat, stated plainly: my web searches did not return the IPR2023-00988 docket, institution decision, FWD text, judge panel, petition grounds, or appeal record. The Google Patents entry lists the petitioner field as blank. I therefore cannot and do not assign a petitioner, claims challenged, or a claim-level outcome. Everything below that is not sourced to the ITC notice or the patent's family record is flagged as unverified.

IPR2023-00988 — Petitioner unidentified in source data v. Samsung Display Co., Ltd.

• Type: Inter Partes Review (designation inferred from the "IPR" prefix; the case-number series is the standard IPR series).
• Filed: Not verified. The IPR2023-xxxxx series corresponds to petitions filed in calendar 2023; the numbering is consistent with a filing in roughly the first half of 2023, contemporaneous with the ITC intervention and IPR campaign described below. Treat the exact filing date as unknown.
• Status: Source data records the case as "Final Written Decision" — i.e., trial was instituted and ran to a final decision (the docket would have reached the 35 U.S.C. § 316(a)(11) one-year statutory deadline barring extension). This is not an institution-denied or settled-then-terminated status label; it is an FWD label. The disposition of that FWD is not stated in the source data I retrieved. Note the tension: ODP has no AIA proceeding indexed for this patent, which may reflect an ODP indexing lag, a case-to-patent mapping the ODP dataset does not carry, or a Family-ID mapping issue in the third-party data. Do not treat this as confirmed PTAB activity until the E2E docket is pulled.
• Judge panel: Not available in retrieved sources.
• Petition grounds: Not available. I cannot state which claims were challenged or whether the statutory basis was § 102, § 103, or both.
• Institution decision: Not available — date and panel reasoning unverified.
• Final Written Decision: Issued (per status label), content not retrieved. I will not guess which claims were canceled or sustained. Notably, no claim of US10854683 appears in the Certificate of Correction / cancellation records surfaced in my searches, and the ODP block's silence means I cannot confirm a § 318(b) certificate cancelling claims.
• Settlement / termination: Not confirmed. Separately, Samsung Display and BOE announced a global settlement of their display IP disputes in late 2025 — BOE's investor-relations statement dated 2025-11-26 confirms the parties "已就显示领域知识产权方面的争议达成和解" (reached a settlement on display-field IP disputes) — but I could not verify that this settlement terminated IPR2023-00988 or on what terms.
• Appeal: Not confirmed. The patent's family record lists two Court of Appeals for the Federal Circuit docket entries — 25-1518 and 25-1521 — alongside a 2025 E.D. Va. case (1:25-cv-00908) and a second ITC investigation (337-TA-3661). I cannot confirm these appeals arise from the IPR2023-00988 FWD rather than from the ITC or district court tracks. Anyone relying on this must pull the CAFC dockets (CourtListener).
• Defensive value: Do not build a defense on this proceeding in its current state of knowledge. The status label means a PTAB trial ran to completion — that is a meaningful signal that the claims were contestable and that invalidity art exists — but without the FWD you cannot know whether claims were canceled, sustained, or narrowed. If the FWD canceled claims, § 315(e)(2) estoppel applies to the petitioner and its privies but not to you; you would remain free to raise the same or better art. Until the disposition is confirmed, the single most reliable defensive fact about this patent is the ITC termination of all asserted claims (below).

Strategic summary

Claim status: largely UNTESTED and UNVERIFIED as to the PTAB, but abandoned in the ITC. I cannot report with confidence which claims of US10854683 are canceled, sustained, or untested. US10854683 issued with 15 claims — independent claims 1, 13, 14, and 15, with dependent claims 2–12 hanging from various independents — and the ITC notice does not identify which of those claims SDC originally asserted. What the ITC record does establish is that "all asserted claims of the '683 patent" were terminated from the investigation by Order No. 34 (Oct. 2023), and that by the time the final ID issued on 2024-11-15, only claims of the '803, '578, '599, and '593 patents remained asserted. This is the single most useful fact for a defendant: Samsung Display walked away from this patent in the forum where it had the most leverage. Whether that walk-away was caused by the IPR, by an adverse institution decision, or by litigation economics is not something I can verify.

Estoppel landscape. § 315(e)(2) estoppel attaches only to the petitioner in IPR2023-00988 and its privies/real parties in interest, and only as to grounds raised or that reasonably could have been raised. Because I cannot confirm the petitioner's identity, I cannot tell you who is estopped. For you, as a non-petitioner, estoppel is irrelevant: every § 102/§ 103 ground remains available, including grounds the petition did not raise. Note the strategic asymmetry — if the FWD sustained any claim, the petitioner's loss does not bind you, and you can attack with different art and a different expert. If the FWD canceled claims, those claims are simply gone and any demand letter citing them is a liability for the patent owner.

Pattern signals. Samsung Display is a serial enforcer with a coordinated PTAB/ITC/district-court playbook, not a defensive aggregator target. The enforcement campaign described in the record is Samsung Display → BOE (plus distributors BLF and others) across ITC 337-TA-1351, E.D. Tex. 2:23-cv-00309, and later E.D. Va. 1:25-cv-00908 and ITC 337-TA-3661. On the receiving end, BOE told its investors that it filed IPRs against the patents in the investigation and that five were instituted (BOE investor-relations statement, 2024-11-21: "公司已在美国专利与商标局提出专利无效多方复审程序（IPR），目前5件均已成功立案"). That is a five-patent campaign, and US10854683 is one of the five asserted patents — making it highly likely that IPR2023-00988 is one of BOE's five instituted IPRs, though I am flagging that as inference, not verified fact. There is no defensive aggregator in this chain; the challenger is a direct competitor, which means the invalidity record was built by a party with a commercial, not rent-seeking, incentive — usually a higher-quality record.

Recommended next steps

1. Pull the IPR2023-00988 record before relying on anything above. The Unified Patents case page is a pointer, not evidence: portal.unifiedpatents.com/ptab/case/IPR2023-00988. The authoritative sources are PTAB E2E (https://ptacts.uspto.gov/ptabweb/) and the USPTO Open Data Portal / PTAB API (https://developer.uspto.gov/api-catalog). Download, in order: the Petition (claims and grounds), the Institution Decision (Paper 6–10), the Final Written Decision, and the outcome table at the end of the FWD. Also run a § 318(b) certificate check and a reissue/reissue-application check, since a patent owner facing an adverse FWD sometimes seeks amendment through reissue.

2. If the FWD canceled claims, use it affirmatively and quote the disposition. There is no public FWD text in what I retrieved, so I will not quote one. Once pulled, the operative language will be of the form "claims X–Y have been shown to be unpatentable" / "claims Z have not been shown to be unpatentable," followed by the ORDER paragraph. If a demand letter cites a canceled claim, the response is that the claim no longer exists — it cannot be infringed and cannot be asserted.

3. Lean on the ITC termination as an independent defensive fact. The ITC's own notice states that the remaining asserted claims at final ID were only those of the '803, '578, '599, and '593 patents "as a result of termination of all asserted claims of the '683 patent." The final ID of 2024-11-15 then found no violation of section 337, because SDC failed the economic prong of the domestic-industry requirement. Two independent fact findings in a single Commission proceeding support the position that this patent is not carrying the enforcement load.

4. Model the appeal risk. CAFC dockets 25-1518 and 25-1521 are live on this patent's family record as of the 2026-09-26 fetch. Confirm (a) whether either is an appeal of the IPR2023-00988 FWD, and (b) whether the 2025 Samsung–BOE global settlement moots them. An appeal of an adverse FWD changes the estoppel and claim-status picture materially, and a settlement-driven dismissal-with-prejudice would lock in a PTAB loss.

5. Note the one hard date you can rely on. The '683 patent carries an anticipated expiration of 2032-09-13 per the patent record (i.e., 20 years from the 2012-09-13 filing of parent application 13/614,197). It is a live patent with roughly six years of term remaining, so this is a patent worth fully clearing rather than waiting out.

I'll research the assignment record for this patent before writing the analysis.

Let me dig further for the specific recorded assignment details.

Let me try to pin down the specific reel/frame and correspondent for this patent's assignment.

Inventors

Employer note (flagged as inference, not a recorded fact). The application traces to Korean priority application 10-2012-0022967, filed 2012-03-06. At that date the Yongin-based display affiliate was still Samsung Mobile Display Co., Ltd.; it was renamed Samsung Display Co., Ltd. in July 2012, before the 2012-09-13 filing of the US parent (13/614,197). By the time the continuation at issue here (15/811,599) was filed on 2017-11-13, the employer of record was unambiguously Samsung Display Co., Ltd. The inventor is listed on the KR family member KR 10-1615332 B1, whose assignee is 삼성디스플레이 주식회사 (Samsung Display Co., Ltd.), which corroborates employment. I could not independently verify Lee's employment contract or job title.

Unusual-pattern check. The task asks me to look for inventors exiting within 12 months of filing (a classic precursor to a portfolio fire-sale). No such signal here: there is a single inventor; he does not appear as an assignor to any entity other than Samsung Display in the records I could reach; and no inventor-side assignment to a third party exists on this chain. Not present.

Original assignee

Samsung Display Co., Ltd. — 1, Samsung-ro, Giheung-gu, Yongin-si, Gyeonggi-do 17113, Republic of Korea (as recorded on the face of the patent).

• Primary line of business: design and manufacture of flat-panel displays — OLED and, historically, LCD panels — for mobile devices, IT products and televisions. This is a component/panel supplier, not a consumer-facing brand.
• Did it ship a product embodying the claims? Yes, on the evidence in the file. The asserted claims (1, 2, 4–10, 13) read on a green-center / blue-and-red-surround RGB pixel layout with blue area > red area and blue-to-blue center spacing > blue-to-red spacing — i.e., a mass-production OLED sub-pixel arrangement. Samsung Display commercialized exactly this class of layout (the "Diamond Pixel" / PenTile-derived arrangement) in its AMOLED mobile panels and, per the ITC pleadings summarized in the prior section, accused BOE of importing panels that infringe it. A company does not obtain a general exclusion order over a layout it has never practiced. Product-practicing: yes.
• Current status: Operating. No bankruptcy, dissolution, or divestiture appears anywhere on this chain. It remains the current assignee and the named complainant in the related ITC and district-court actions.

Assignment timeline

Critical limitation on sourcing, stated up front. I was unable to retrieve the reel/frame numbers or the recorded correspondent for this patent. The Google Patents legal-events record supplied to me — which mirrors the USPTO assignment index but does not print reel/frame or correspondent fields — shows exactly one post-filing reassignment entry. I could not reach assignmentcenter.uspto.gov or assignment.uspto.gov record detail through the tools available in this session, and my searches did not surface a PDF or notice-of-recordation copy bearing a reel/frame for 15/811,599. I will not invent a reel/frame number. Everything below is what the record actually supports.

• 2018-01-25 (date on record) / recorded 2018-01-25 — Reel not retrievable / Frame not retrievable
  • Conveyance: Assignment — recorded with the brief "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)."
  • Assignor: LEE, SANG-SHIN (sole named inventor)
  • Assignee: SAMSUNG DISPLAY CO., LTD.
  • Correspondent: Not obtainable from my sources. ⚠️ Context only, not a verification for this record: Samsung Display's US assignments of this era were routinely filed through Knobbe Martens Olson & Bear, LLP, Irvine, California (e.g., Samsung Display assignment reel 033308/0763, submitted by Paul C. Steinhardt, for a different Samsung Display application). I have no document tying Knobbe Martens to the 15/811,599 record, so treat it as an unverified pattern observation, not a finding.
  • Context: Internal — inventor-to-employer conveyance. The sole inventor assigned his right, title and interest in the continuation application to his employer, the original assignee. This is the ordinary first link of a corporate chain of title, not a transfer to an acquirer. It is dated roughly five years after the 2012 priority filing and approximately five years before the first infringement complaint (Dec 2022), so it fails the "pre-litigation transfer" test by a wide margin.

That is the complete chain. No other recorded assignments appear. Specifically, the record shows no security agreement, no release, no merger, no change-of-name, no license recordation, and no transfer out of Samsung Display. The chain terminates where it started: at the operating company.

Verification links

• Assignment Center (search by patent number 10854683): https://assignmentcenter.uspto.gov/
• Legacy assignment search mirror: https://assignment.uspto.gov/patent/index.html
• Google Patents legal-events/mirror record: https://patents.google.com/patent/[US10854683](/patent/US10854683)/en

Contradiction flag against the previously generated section

The earlier bibliographic section lists the related-family continuations (US 11,594,578, US 11,980,077, US 12,324,335, and pending US 2025/0268051 A1). Important for an ownership analysis: every one of those continuations carried the same "Priority to" entry with the assignee unchanged. There is no divestiture, no continuation-to-NPE spinout, and no orphaned child in the family. If a shell transfer had occurred, that is precisely where it would have surfaced — 2023–2025 continuation filings are a common vehicle for moving claims out of an operating parent. It did not happen here. This is consistent with, and reinforces, the earlier section's characterization of the family as a single operating-company portfolio.

Timeline diagram

timeline
    title Ownership of US 10854683
    2012 : Korean priority filed by Samsung affiliate
         : US parent filed
         : Samsung Mobile Display renamed to Samsung Display
    2017 : Continuation application filed
    2018 : Sole inventor assigns to Samsung Display
    2020 : Patent US 10854683 granted
    2022 : ITC complaint filed against BOE
    2023 : ITC investigation instituted
    2025 : Global settlement with BOE
         : Asserted claims already dropped

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
The patent has never left Samsung Display. There is no "IP / Patents / Licensing / Holdings / Ventures" transferee anywhere in the chain, no single-purpose Delaware or Texas LLC, and no registered-agent address of record. The only recorded conveyance (2018-01-25) runs from an individual natural person to his corporate employer, which is the opposite of a shell-entity transfer pattern.

2. Known asserter in the chain — NOT PRESENT.
Neither the assignor nor the assignee matches any entity on the referenced NPE lists — Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg vehicle. Samsung Display is a FORTUNE Global 500-class operating manufacturer, and it appears in RPX/Unified data as a defendant-side and plaintiff-side operating company, not as a high-frequency NPE plaintiff.

3. Repeat correspondent across the chain — NOT APPLICABLE / UNCLEAR.
This signal is structurally defeated by the chain itself: there is only one recorded assignment. A "repeat correspondent" requires two or more links. Even if the correspondent were confirmed as Knobbe Martens, a single appearance is expressly not a finding under the stated methodology, and Knobbe Martens does substantial operating-company prosecution work (including Samsung Display's own post-issue assignments — see the 2014 recording at reel 033308/0763). Marked unclear only because the correspondent field could not be retrieved, not because any adverse evidence exists.

4. Cascading transfers — NOT PRESENT.
Zero. There is one transfer across the entire life of the patent, in 2018. No chain of LLCs, no shared correspondence addresses, no common principals, and no sub-24-month succession. The patent sat with one owner for its full life.

5. Pre-litigation transfer — NOT PRESENT.
The controlling date comparison: assignment 2018-01-25; earliest infringement complaint naming this patent 2022-12-28 (ITC complaint). That is ~59 months, roughly ten times the 6-month window the methodology treats as suspicious. The assignment was executed for the routine purpose of perfecting title in a continuation filing, not to set venue or manufacture standing.

6. Bankruptcy fire-sale — NOT PRESENT.
Samsung Display has not filed Chapter 7 or 11. No trustee's sale, no §363 sale order, no assignment of record arising from insolvency proceedings. Samsung Display was profitable throughout the relevant period; the display unit's economics are irrelevant here because no insolvency proceeding exists.

7. Privateering — NOT PRESENT.
This is the signal most worth addressing head-on, because Samsung Display is a serial enforcer and it does assert in the ITC. But privateering is defined as an operating company transferring to an NPE that then asserts on the operating company's behalf. Here the opposite is true: Samsung Display asserts in its own name. It is the named complainant in ITC Inv. No. 337-TA-1351 (88 FR 7463), the named plaintiff in the Texas action, and it sought a general exclusion order itself. No assertion intermediary is interposed between Samsung Display and the accused products. Not present.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. Note the interesting inversion: Unified Patents is on the petitioner side here (IPR2023-00988 is flagged as a Unified Patents filing on the patent's record), i.e., the defensive aggregator is attacking this patent on behalf of the industry, not acquiring it. That is a pro-assertion posture indicator, not an anti-NPE one.

Verdict

Operating-company assertion

Justification. There is exactly one recorded assignment on the entire chain — the sole inventor Sang-Shin Lee conveying his interest to his employer Samsung Display Co., Ltd., dated 2018-01-25 — and the assignee has never transferred the patent to any party, shell or otherwise. The assignee is a product-practicing panel manufacturer whose mobile AMOLED panels embody the green-center/blue-larger-than-red pixel layout recited in claims 1, 13, and 15, and it asserted the patent in its own name against a direct competitor (BOE) in ITC Inv. No. 337-TA-1351 and in district court, seeking a general exclusion order — the classic posture of an operating company protecting a manufacturing technology against imports, not a licensing-only monetization play. No NPE signal is present, and the only "unusual" feature of the record — an inventor-to-employer assignment recorded in 2018 for an application with 2012 priority — is explained by the intervening continuation filing, not by a title laundering scheme.

Caveat on completeness (stated plainly, not as a hedge): the reel/frame and correspondent fields for the 2018-01-25 record could not be retrieved with the tools available in this session, and one PTAB proceeding (IPR2023-00988) has an unverified outcome per the prior section. Neither gap could plausibly change the verdict — a missing reel/frame cannot conjure an assignee that does not appear in the ownership chain, and an adverse IPR outcome would affect validity, not ownership. But a reader who needs the reel/frame for a title opinion should pull it directly from the Assignment Center: https://assignmentcenter.uspto.gov/ (search patent number 10854683). Do not accept any reel/frame number for this patent that is not sourced from that page or a notice of recordation.

I'll search the USPTO/patent databases for the specific patent and its cited references.

Let me research the specific prior-art references and the IPR grounds.

Verification of the target patent (USPTO record)

I searched for the specific number 10,854,683. The authoritative document returned is the granted patent's own full text (PDF/HTML) — I am not relying on any similarly-numbered document:

• US 10,854,683 B2, "Pixel arrangement structure for organic light emitting display device," Appl. No. 15/811,599, filed Nov 13, 2017, granted Dec 1, 2020, inventor Sang-Shin Lee, assignee Samsung Display Co., Ltd. Pre-grant pub. US 2018/0069057 A1 (Mar 8, 2018). Priority KR 10-2012-0022967 (Mar 6, 2012); continuation of US 13/614,197 (now US 9,818,803 B2). Source: https://patentimages.storage.googleapis.com/8f/07/0e/22093dfc98de53/US10854683.pdf and https://patents.google.com/patent/US10854683/en

Explicit non-matches I discarded (this is the trap the instruction warns about): search returned several different patents that happen to end in "683" — e.g., US 8,338,683 (a musical-instrument tuner, § 101/Alice litigation), an IPR involving a ScentAir/Prolitec liquid-diffusion "'683 patent," and a router "'683 patent" (Decasper). None of these is US 10,854,683, and none is cited by it. I have not merged them into the analysis.

Cross-reference to the earlier sections (one correction)

The previously generated "Explicit Uncertainty" item #1 said the outcome of IPR2023-00988 was unverified. That is now resolved. I retrieved the PTAB Final Written Decision text for IPR2023-00988 (U.S. Pat. 10,854,683 B2):

• The Board held that claims 1, 4–10, 13, and 15 are unpatentable (obviousness, § 103), and that Petitioner did not prove claim 2 unpatentable.
• Real parties-in-interest identified by the Petitioner: Mianyang BOE Optoelectronics Technology Co., Ltd. and [Visionox Technology, Inc.](/litigations/by-plaintiff/Visionox%20Technology%2C%20Inc.) (and subsidiaries); Patent Owner is Samsung Display. So the "'683 IPR" was a BOE/Visionox challenge — the "Unified Patents" string on the Google Patents litigation tab is a data-source label, not the petitioner.
• Grounds pivoted on the Phan reference, discussed below.
  Source: PTAB FWD text, IPR2023-00988, doc. 44 (via ai-lab.exparte.com / PTAB).

This is important context for the prior-art question: the claims that were cancelled in the IPR are the same claims asserted in the ITC (1, 2, 4–10, 13). Claim 2 was the only challenged claim that survived.

Preliminary: what the prior art actually has to meet

The asserted/cancelled claim set imposes (all in claim 1, repeated in 13 and 15):

1. Pairs of second and third pixels on two crossing lines that each continue to neighboring first pixels (consecutive first/second/first/second/first and first/third/first/third/first);
2. second-pixel-to-second-pixel center distance > second-to-third center distance;
3. first pixels emit green, a different color;
4. first pixels smaller than second or third;
5. each second pixel larger in area than each third pixel (blue > red).

Only a single reference that discloses every one of these is a § 102 anticipation reference. In practice the cited art splits into two groups: (a) references that teach the geometry (X/crossing arrangement, hexagonal/octagonal sub-pixels, spacing to improve FMM deposition), and (b) references that teach the color/area trade-off (blue = largest area because blue has the shortest lifetime). The '683 claims sit at the intersection, which is why the record was fought on § 103, not § 102. I flag § 102 vs. § 103 for each reference below.

Part A — The most relevant prior art

A1. Phan (the actual IPR ground) — most probative reference

This is the single most consequential prior art for the '683 patent, because it is the reference that produced the actual cancellation of claims 1, 4–10, 13, and 15.

A2. The black-and-white "X with crossing lines + hexagonal/octagonal sub-pixels" references

These teach the geometry half of the claims (requirement 1, and parts of 2/4).

A3. The "color/area trade-off" references (aperture ratio / lifetime)

A4. Clairvoyante / sub-pixel-rendering lineage (background for "pentile" and pixel geometry)

A5. Imaging-sensor / CCD references (peripheral — cited for polygonal or shifted pixel geometry, not for OLED color rules)

These appear on the face of the '683 patent and in the family but teach image-sensor color-filter arrays rather than OLED arrangements. They are not plausible § 102 references for claim 1's OLED/color/area limitations; at most they are § 103 auxiliary art for the "polygonal pixel with vertices/centers" geometry language:

• US 4,642,619 A (Togashi, Citizen Watch) — Feb 10, 1987 — non-light-emitting liquid crystal color display device.
• US 6,882,364 B1 (Inuiya et al., Fuji Photo Film) — Apr 19, 2005 — solid-state imaging apparatus; honeycomb→video signal transform.
• US 7,110,031 B2 (Kondo et al., Fuji Photo Film) — Sep 19, 2006 — state image pickup apparatus with pixel shift layout.
• US 7,710,484 B2 (Oda, Fujifilm) — May 4, 2010 — CCD image-sensor driving/imaging method.
• US 7,982,786 B2 (Nishida et al., Fujifilm) — Jul 19, 2011 — solid-state imaging device for color imaging.
• US 2002/0140833 A1 (Hirai) — Oct 3, 2002 — signal processing / imaging.
• US 2008/0273793 A1 (Oishi) — Nov 6, 2008 — signal processing, noise reduction.
• US 2008/0308819 A1 (Louwsma, TPO Displays) — Dec 18, 2008 — light-emitting diode arrays.
• US 2009/0302331 A1 (Smith et al.) — Dec 3, 2009 — (pixel/imaging-related).
• US 2013/0002911 A1 (Miyashita) — Jan 3, 2013 — imaging device / image processing.
• US 2015/0192834 A1 (Sharp) — Jul 9, 2015 — liquid crystal display device. Post-priority; peripheral.

A6. Remaining face-of-patent references (grouped, lower relevance)

Display/driving and miscellaneous U.S. publications: US 2004/0183764 A1 (LG Electronics, Sep 23, 2004); US 2005/0162079 A1 (Sakamoto, Jul 28, 2005); US 2006/0152531 A1 (Lin et al., Jul 13, 2006); US 8,026,669 B2 (Kawasaki, Canon, Sep 27, 2011); US 8,350,468 B2 (Ko et al., Jan 1, 2013); US 2010/0117528 A1 (Fukuda, Sony, May 13, 2010); US 2010/0171440 A1 (Mitsubishi Chemical, Jul 8, 2010); US 2012/0319564 A1 (Emagin) — note: "US 2012/0319564 A1" appears in the family citation list as Emagin "Color organic light-emitting diode display device" (filed Jun 20, 2011; pub. Dec 20, 2012).

Foreign patent documents cited: JP H03-36239, JP H10-39791, JP 2000-235891, JP 2003-203770 (Toshiba), JP 2004-179028 (Optrex), JP 2004-192813 (Toshiba Matsushita), JP 2005-005227 (Hitachi Displays), JP 2006-018195 (NHK), JP 2006-309182 (Semiconductor Energy Lab), JP 2007-156126 (Epson Imaging), JP 2009-230096 (Mitsubishi Electric), JP 2010-003880 (Sony), JP 2010-153173 (SEL), JP 2011-076760 (Toppan), JP 2011-198761 (Samsung Mobile Display); WO 03/053068 (Clairvoyante), WO 2004/073356 (Fujitsu), WO 2005/067352 (Samsung Electronics); KR 10-0742370 (Samsung SDI), KR 10-0807524 (LG.Philips), KR 10-2004-0096706 (LG.Philips), KR 10-2009-0017910 (LG Display), KR 10-2009-0049515 (Samsung), KR 10-2011-0023996 (LG Display), KR 10-2011-0106565 (Samsung Mobile Display); CN 101009304 (AU Optronics); GB 2437110 (Cambridge Display Tech). None of these, on the face of the record, discloses the complete combination; they read as general background/support, i.e., § 103-only.

Date caution applying to the whole list: Because the '683 claims are entitled to a Mar 6, 2012 priority date, any reference whose publication occurred after that date (e.g., US 2012/0313844, US 2012/0287605, US 2013/0037827, US 2013/0057521, US 2013/0065891, US 2013/0127689, US 9,431,469, US 9,837,476, US 2015/0192834) is prior art only if an earlier effective date cannot be relied on (e.g., via its own § 102(e)/§ 102(a)(2) filing date, or if the '683's priority chain were broken). I have not independently established the effective date of every such reference; treat the "date-sensitive" tags as flags, not conclusions.

Part B — Answer to "which claim(s) each reference potentially anticipates under § 102"

Straightforward, defensible bottom line:

• No single reference of record cleanly anticipates claim 1 (or independent claims 13, 14, 15). Anticipation under § 102 requires one reference disclosing all limitations, including the color rule ("first pixels emit green, different color"), the area rule ("second pixel area > third pixel area"), and the spacing rule ("second–second distance > second–third distance"). The cited art splits geometry references from color/area references.
• The closest single-reference candidates to a § 102 case are the area/color references (US 2011/0291550 A1 and US 6,867,549 / US 2004/0108818 for the blue-largest/red-smaller rule; US 2011/0012820 / US 8,354,789 for polygonal sub-pixels) — but each lacks the crossing-line X geometry of claim 1, so even these are properly § 103 references, not § 102.
• The § 102(e)/§ 102(a)(2) candidates worth checking are the pre-priority-filed publications US 2012/0049726 A1 (filed Aug 27, 2010) and US 2012/0056531 A1 (filed Sep 8, 2010), because their filing dates predate Mar 6, 2012 even though they published in Mar 2012.
• The reference with actual, adjudicated impact is Phan (US 7,091,986 / 7,215,347), which in IPR2023-00988 rendered claims 1, 4–10, 13, and 15 unpatentable — on § 103, not § 102. Claim 2 survived the challenge.
• I see no reference that would support a § 102 rejection of surviving claim 2, consistent with the Board's finding.

Explicit uncertainty (updates to the earlier section)

1. Resolved: IPR2023-00988's FWD cancelled claims 1, 4–10, 13, 15 (obviousness) and upheld claim 2; petitioner real parties-in-interest are BOE/Visionox entities, not Unified Patents (the Google Patents label is a data source). Source: PTAB IPR2023-00988 FWD (doc. 44).
2. Not verified: the precise claim-by-claim mapping of the Phan ground beyond the Board's summary conclusion, and the identity of the secondary reference(s) combined with Phan — I retrieved the FWD discussion of Phan but not the full petition claim charts. Do not treat my claim mapping for Phan as the Board's exact reasoning.
3. Not verified: the detailed technical content of ~40 of the foreign/peripheral references above (JPs, KRs, imaging-sensor patents). My descriptions are based on their titles, assignees, and the role they play in the citation record, not on a full read of each.
4. Date flags in Part A are flags, not determinations of prior-art status.
5. Everything above about claim scope is a preliminary reading, not a claim construction.

Primary sources: US 10,854,683 B2 full text/claims (patentimages.googleapis / Google Patents); the patent's own "References Cited" and family/citation tabs (https://patents.google.com/patent/US10854683/en); PTAB FWD IPR2023-00988 doc. 44; US 2011/0291550 A1 and US 2011/0012820 A1 full texts (Google Patents / FreePatentsOnline); US 8,330,352 B2 and US 2013/0057521 A1 (Google Patents / FreePatentsOnline).

I'll ground this § 103 analysis by verifying the key prior-art references and the IPR outcome before writing.

Strong results. Let me verify the appeal status and the Phan-based ground details, since those govern whether the cancelled claims are actually cancelled.

Two more targeted checks — the exact ground structure of the winning Phan challenge and the secondary references the Board rejected.

Obviousness Analysis — US 10,854,683 B2 under 35 U.S.C. § 103

Building on the Prior Art section already generated. I do not repeat the reference inventory there; I use it. Two of that section's flagged uncertainties are now resolved from primary PTAB text, and one statement in it needs correction — flagged in §1.

1. What changed since the Prior Art section (corrections first)

Three verified facts from the PTAB Final Written Decision in IPR2023-00988 materially sharpen the § 103 picture:

(a) The winning ground was Phan standing alone, not a Phan combination. The FWD's conclusion paragraph states: "Petitioner has shown, by a preponderance of the evidence, that the subject matter of claims 1, 4-10, 13, and 15 of the '683 patent would have been obvious over Phan." The secondary reference Yamada appears only in grounds the Petitioner lost ("Phan and Yamada" for claim 2; "Matthies and Yamada"; "Murai and Yamada"). ⚠️ This corrects the Prior Art section's statement that "Phan is a combination ground (it was not a clean single-reference § 102 anticipation)." It was a single-reference § 103 ground. Source: IPR2023-00988 FWD, doc. 44.

(b) The petitioner is identified. Real parties-in-interest: Mianyang BOE Optoelectronics Technology Co., Ltd. and [Visionox Technology, Inc.](/litigations/by-plaintiff/Visionox%20Technology%2C%20Inc.) and their subsidiaries. Patent Owner: Samsung Display. The "Unified Patents" string on Google Patents is a data-source label, as the prior section suspected. Same source.

(c) The FWD is on appeal, so the claims are not yet cancelled. A 2025/2026 PTAB petition involving a related family patent states that "IPR2023-00987, IPR2023-00988, and IPR2023-01075 … are on appeal to the Federal Circuit, Nos. 25-1517, 25-1518, 25-1519, 25-1520, 25-1521." No § 318(b) certificate can issue while the appeal is live. Source: PTAB petition 1558494.

Consequence for the § 103 question: the target is a claim set that has already been held obvious by the Board on one reference, with the appeal pending. Obviousness is therefore not a hypothetical for this patent — it is an adjudicated (if non-final) finding. The analysis below separates (i) the ground that worked, (ii) grounds that are redundant with it, and (iii) grounds that failed and which a defendant should not assume are available.

2. Governing law and the level of ordinary skill

Effective filing date for prior-art qualification: 6 March 2012. The Board expressly so held: "we apply the March 6, 2012, date for qualifying the asserted references as prior art" (FWD § I.B). References publishing after that date are prior art only on a § 102(a)(2)/§ 102(e) basis (their own earlier filing), which is why several items in the prior-art table carry "date-sensitive" flags.

Governing framework: Graham v. John Deere (scope and content of the prior art; differences; PHOSITA level; objective indicia), with KSR Int'l v. Teleflex supplying the rationales and the "predictable results" and "design incentive" tests. Because the Board applied KSR to reject Samsung's argument that "Phan does not teach about OLED displays," the KSR framework is not academic here — it is the ratio decidendi.

PHOSITA (my formulation — not the Board's, and not a construction): a person with a bachelor's degree in electrical engineering, applied physics, or materials science, plus approximately 3–5 years' experience designing active-matrix OLED display architectures, including sub-pixel layout, fine-metal-mask (FMM) deposition, color-balance/lifetime budgeting, and display-driver integration. This is the standard the art supports: the asserted claims are arrangement claims, and every motivation discussed below turns on display-engineering trade-offs (aperture ratio vs. FMM shadowing, blue lifetime vs. white balance) that a display engineer of ordinary skill would have known as of 2012.

3. The § 103 checklist the claims impose

Not restating the prior section's claim overview; compressing the independent claims into the elements a combination must supply:

The crucial structural insight for § 103: the "hard" limitations are [1-e] (blue area > red area) and, for claim 14, the longer-first-pixel-pitch rule. Neither is a geometry-of-record problem; both are allocation problems — i.e., "which color gets the bigger aperture, and how do you space the greens." Those are precisely the questions the prior art answers explicitly and with stated reasons.

4. The ground that actually worked: Phan alone

4.1 The mapping the Board accepted

Phan's "square quad pixels" (implementation of Figure 11b's dot structure in Figure 12's repeating grid) were mapped:

The Board's language on the spacing element: "We are persuaded that the dot/pixel structure of Phan's Figure 11b implemented in the repeating grid pattern of Phan's Figure 12 results in the arrangement recited in this limitation with a 'first distance' between blue dots … being greater than a 'second distance' between a blue dot … and a red dot."

4.2 Why a PHOSITA would have arrived there (the motivation, as the Board found it)

1. Phan is in the same field. Both Phan and the '683 are directed to how light-emitting elements are spatially arranged to control perceived resolution and luminance; Phan's specification addresses arrangements of emissive dots with inter-element spacing — structurally the same problem as sub-pixel layout for an OLED panel.
2. The Board invoked KSR to bridge display technology. Samsung argued Phan was not an OLED reference. The Board rejected that, holding it would have been obvious to a PHOSITA to implement Phan's arrangements in an OLED. Under KSR, a reference in an analogous display art is reasonably pertinent to the problem of sub-pixel arrangement.
3. The '683 specification supplies the admission of the problem. Its own Background states: "When reducing a gap between the neighboring pixels to obtain a high aperture ratio of the pixel, deposition reliability may be deteriorated. On the other hand, when increasing the gap between the pixels to improve the deposition reliability, the aperture ratio of the pixel may be deteriorated." That is an express, applicant-authored statement of the design trade-off — textbook KSR "design incentive" evidence, and it is in the patent being attacked.
4. The '683 specification also supplies the blue-lifetime premise. The FIG. 2 passage states "the second pixels 200 that emit blue have the shortest life span among the first pixels 100, the second pixels 200, and the third pixels 300," and that enlarging them "suppress[es] the deterioration of the life span." An applicant's own recitation of a known problem, coupled with a known solution in the art, is a strong obviousness posture.

4.3 Where the Board's Phan ground is exposed on appeal (Samsung's likely attack)

Flagging this because it matters to whether a defendant can rely on the FWD:

• Measurement basis for [1-b]. The Board's ratio analysis (2x vs. 1.4x) is center-to-center. But claim 8 — which depends from claim 1 — separately recites that "the first distance is measured between respective centers of the second pixels, and the second distance is measured between respective centers…." If claim 1 already required center-to-center measurement, claim 8 would be redundant. A patent owner can therefore argue claim 1's "distance" is edge-to-edge (gap), in which case Phan's equal-sized dots with equal gaps could yield equal first and second distances — defeating [1-b]. The flip side: the FWD records that Samsung "does not dispute this limitation apart from its arguments discussed above with respect to the preamble," which reads as a trial-level concession that will be hard to revive. I cannot resolve the construction from the text I have.
• "Pixels" vs. "dots." The preamble dispute (whether Phan's dots are "pixels" in the claim sense) was rejected, but the Board's reasoning there is the appellate fulcrum.
• Claim 2 survived. The Board held claim 2 was not shown obvious over Phan + Yamada. That is a real, adjudicated limit on this art family.

5. Redundant and alternative § 103 combinations

These are combinations a defendant could plead independently of the FWD (and are not barred by § 315(e)(2) as to a non-petitioner). All are untested on this claim set unless noted.

A. Phan + Kim '550 — the belt-and-braces version of the winning ground

Kim '550 = US 2011/0291550 A1 (Kim, Gun-Shik; published 1 Dec. 2011; filed 26 May 2010 — pre-priority, so § 102(e)-qualified and available for § 103). Teaching, verbatim: "In general, the blue sub-pixel B in the organic light emitting display device has the shortest life property. Therefore, the third sub-pixel 26 that has the largest area may be set up as the blue sub-pixel B, so that the life property can be improved." Source: FPO y2011/0291550.

What it cures: the two Phan limitations Samsung will litigate hardest — [1-c] (green first pixels) and [1-e] (blue area > red area). Kim '550 supplies an express reason to give blue the largest area, in an OLED, to fix lifetime.

Motivation: both references are OLED sub-pixel-arrangement art; Phan supplies the geometry, Kim '550 supplies the area-allocation rule with the reason stated in the reference itself. Result: the lifetime of a fixed layout is improved without changing resolution — a predictable, mechanically-applied improvement. This is KSR rationale (3) ("known technique to improve a similar device in the same way") plus (1) ("predictable results").

B. Credelle + Kim '550 — the combination already used against this family

A PTAB petition/appeal record in this patent family recites the reasoning verbatim: "Credelle clearly teaches that a shortest distance (a gap) between the first pixel (the blue pixel 102) and the second pixel (the green pixel 106) as well as the shortest distance … between the first pixel … and the third pixel (the red pixel 104) is a same length … Credelle also teaches that a shortest distance … between the first pixels (two blue pixels 102) is a second length that is longer than the first length," and pairs it with Kim: "it would have been obvious … to have used different sub-pixel sizes as taught by Kim in a subpixel grouping of Credelle because the different sub-pixel sizes can correct the white balance and the color coordinate. … Kim teaches that the second pixel (the green pixel 106) has a larger area than … the third pixel (the red pixel 104)." Source: PTAB petition 1558493.

Motivation: white-balance and color-coordinate correction. That is a concrete, articulable engineering reason — the best kind of § 103 motivation.

⚠️ The vulnerability, and it is a real one: Credelle designates the blue sub-pixel as "first," the green as "second," the red as "third." Claim 1 requires the green emitter to be the central/"first" pixel, smaller than the others, with blue as the second pixel and larger than red. Reaching claim 1 from Credelle therefore requires re-designating/re-coloring the arrangement — precisely the objection recorded in the same record: "Combining Kim with Credelle requires arbitrary designation of 'first,' 'second,' and 'third' pixels … which amounts to impermissible hindsight." Whether that objection prevails depends on whether the re-coloring produces a new structural relationship or merely relabels; on these claims the relationship (pitch and area ratios keyed to the blue/red pair) is arguably unchanged, but I would not treat Credelle + Kim as a clean § 103 win, and note the Board in IPR2023-00988 did not rely on it.

C. Claim-2/3/11/12-specific combination: Phan + Kim '820/'789 + Cok/Chao

Claim 2 requires polygonal first/second/third pixels with second and third larger in area than the first. The Prior Art section's US 2011/0012820 A1 / US 8,354,789 B2 (Kim, Gun-Shik) supplies hexagonal sub-pixel geometry expressly for "aperture ratio, efficiency, life, and image sticking" in a Pentile-matrix variant — a direct teaching that polygonal (hexagonal) sub-pixels are used for the very purposes the '683 recites. Add Kim '550 for blue-largest and Cok (US 6,867,549 / US 2004/0108818) or Chao (US 2008/0001525 A1) for repeated-unit-cell arrangements with differing emitter areas.

Motivation: geometric substitution of one known sub-pixel footprint for another, to satisfy an already-known need (aperture ratio without sacrificing FMM reliability). Claim 3 (octagonal second/third) is then a routine design choice over Kim '789's hexagons — the '683 specification itself describes both hexagons and octagons as interchangeable options (claim 3 vs. the FIG. 3/FIG. 5 descriptions), which weakens any argument that octagon-vs-hexagon is inventive. Claims 11 and 12 (non-quadrilateral; ≥5 interior angles) are met squarely by Kim '789's hexagons.

This is my analytical proposal, not an adjudicated ground. Note the Board's contrary finding: claim 2 was not proven obvious over Phan + Yamada. A defendant pleading Phan + Kim '820/'789 must therefore explain why that combination succeeds where Phan + Yamada failed — the best answer is that Kim '820/'789 is OLED-pixel-arrangement art (unlike the more remote Yamada), making the combination far more predictable.

D. The combinations that failed — do not assume they are available

Analytical takeaway: Matthies + Yamada and Murai + Yamada are worse combinations than they look on paper. Both depend on a resizing step imported from Yamada onto a structure whose dots are not arranged as the claims require; the Board found the resizing wouldn't have been done. By contrast, Phan already embodies the crossing-line/quad arrangement, so no structural modification was needed — which is exactly why Phan won alone and the two-reference grounds lost. The pattern is instructive: for this patent, more references = weaker ground.

E. Claim 14 — the spacing-test variant, and its best combination

Claim 14 replaces [1-b] with "first-pixel-to-first-pixel separation > first-pixel-to-neighbor separation" — which is the FIG. 1 L2 > L1 relationship the specification describes as producing "improved deposition reliability when using a fine metal mask." It additionally requires a convex polygon with ≥5 sides.

Best combination: Phan + Kim '820/'789 (+ Phan's own grid spacing). Phan's grid pattern contains green dots between and around blue/red dots; where the greens are the "first pixels," the green-to-green separation along a row relative to green-to-blue separation is a function of the grid pitch and is measurable from Figures 11b/12. Kim '820/'789 supplies the ≥5-sided convex polygon (hexagon) as a known FMM/aperture-ratio expedient, and the '683's own Background supplies the FMM-reliability rationale for the spacing rule.

Caveat: claim 14's numeric inequality is the one element I have not seen mapped against any reference in the record I retrieved. It is plausible but not verified.

F. Kodama-style "white balance" secondary references (auxiliary only)

For claim 5's color assignment (blue = second, red = third) the record's Koo (US 2008/0001527 A1, "improved white balance") and Kim '549 (US 2011/0291549 A1, largest sub-pixel = blue) both supply reasons to weight blue and red differently. § 103 auxiliary art at best; neither adds the geometry.

6. Claim-by-claim § 103 disposition

Note on the claims the Board did not reach: claims 3, 11, 12, and 14 were not in the challenged set (the IPR covered 1, 2, 4–10, 13, 15). They are therefore born-again claims if the appeal reverses: if the Federal Circuit vacates the Phan ground, they are also untested, and they carry the shape/geometry limitations that are the easiest to invalidate over Kim '820/'789 but which no tribunal has yet addressed.

7. The motivation-to-combine case, summarized (mapped to KSR rationales)

The single strongest motivation sentence in the whole record is the applicant's own: reducing gaps improves aperture ratio but degrades FMM deposition reliability, and increasing gaps does the reverse. Once that trade-off is conceded, and once the art teaches both the crossing-line lattice and "blue gets the biggest area because blue dies first," the claimed arrangement is the predictable output of ordinary design work.

8. Objective indicia (what Samsung has and does not have)

• Commercial success / product practice: Samsung Display commercialized Diamond/PenTile-derived AMOLED panels; the Prior Art and Assignment sections both treat the '683 claims as reading on a mass-production layout. But the nexus problem is significant — the asserted claims cover one arrangement among many panel features, and third-party (BOE) designs were accused as infringing, which is evidence the patented feature is not the driver of any particular commercial success.
• Copying: Samsung's ITC/EDTX pleadings allege BOE imported infringing panels. KSR "copying" requires evidence of copying the patented feature, and an infringement allegation is not copying evidence. Weak.
• Long-felt need / failure of others: The '683 Background states the aperture-ratio/FMM-reliability tension as an open problem, which can support long-felt need — but the same passage is an admission that the problem and its trade-off were known, which cuts the other way under § 103.
• Unexpected results: Not evident in the specification, which asserts improvement ("aperture ratio … may be improved") without data. Kim '550 shows the blue-largest allocation was already the expected result, defeating unexpectedness for [1-e].
• Skepticism of experts: The Board's rejection of Matthies + Yamada and Murai + Yamada shows a PHOSITA-skepticism argument can win against a weak combination — but it did not defeat Phan.

9. Confidence, verification, and explicit uncertainty

9.1 Verified from primary text

9.2 Contradictions to flag explicitly

1. ⚠️ Docket-number collision on "25-1521." One document reports "in appeal no. 25-1521, the Federal Circuit affirmed the Board's Final Written Decisions … in proceeding no. IPR2022-00711 … claims 1-20 of U.S. Patent No. 10,860,506 are unpatentable" — a Netlist/Samsung Electronics matter, unrelated to this patent. Another reports 25-1521 as part of the Samsung Display v. Mianyang BOE consolidation. Both cannot be the same appeal. Do not map any individual CAFC docket number to IPR2023-00988 without pulling PACER. What is verified is that IPR2023-00987/00988/01075 are collectively on appeal within the 25-1517–25-1521 block.
2. ⚠️ Date inconsistency in the working materials. The task header sets the current date as April 26, 2026, while the patent fetch recorded on the prior sections is 2026-09-26, and one prior section cites a September 16, 2026 hearing and a February 27, 2026 article. Events dated after April 26, 2026 should be treated as reported-but-not-yet-occurred relative to this analysis. This does not affect the § 103 conclusions.
3. ⚠️ New correction to the earlier sections: the prior Prior Art section described Phan as a "combination ground." Per the FWD's conclusion language, the winning ground reads as Phan alone; Yamada was confined to the losing grounds. The prior section's §1(a) correction above supersedes this.

9.3 Not verified — do not rely

• Full text of the FWD's limitation-by-limitation analysis for [1-c] (green), [1-d] (first pixels smaller) and [1-e] (blue area > red area). I have the Board's conclusion of obviousness-over-Phan and its explicit reasoning for the preamble, the "separated from" element, and [1-b]. I do not have the Board's stated rationale for how Phan alone met [1-c]/[1-d]/[1-e]. My statement that Phan's "square quad pixels" supply those elements is an inference from the conclusion, not a quotation.
• The exact composition of Phan's ground and the identity of the Phan reference version used (US 7,091,986 vs. US 7,215,347). Both are Gia Chuong Phan references with a 1997-09-13 priority, and both appear in the family citation record; the FWD abbreviations I retrieved reference "Phan" generically.
• Claim 14's numeric spacing inequality has not been mapped to any reference in any record I retrieved.
• A "'616 patent" is identified in a 2025/2026 PTAB petition as related to US 9,818,803, US 10,854,683, and US 11,594,578, and is itself under challenge with family-member unpatentability findings cited against it. I cannot map "'616" to any specific patent number in the family list supplied in the earlier sections (which shows US 11,594,578, US 11,980,077, US 12,324,335, and US 2025/0268051 A1). It may be a newly issued continuation of the '803 parent, but I will not guess.
• Whether the Nov 2025 Samsung Display–BOE settlement terminates the '683 IPR appeal. BOE's investor filing confirms a display-IP settlement, and Mianyang BOE withdrew from CAFC 26-1092 on Nov 19, 2025 — but Tianma, Visionox, and Wuhan CSOT remain appellants in 26-1092 and are briefing in 2026, and the 25-1517 block is still reported as on appeal. The two facts are in tension and I cannot reconcile them.
• All claim-scope statements above are my paraphrase, not claim constructions.

9.4 Bottom line for a § 103 challenge

• If you want the highest-probability ground, it is the one that already won: Phan alone (or Phan as the primary reference), against claims 1, 4–10, 13, and 15. It is adjudicated, articulated, and resting on the applicant's own admissions about the aperture-ratio/FMM trade-off and blue's short lifetime.
• For claim 2 (and 3, 11, 12, 14), the winning ground must change — add Kim '820/'789 for polygonal/hexagonal geometry and Kim '550 for the blue-largest allocation. The Board's failure to invalidate claim 2 over Phan + Yamada does not pre-judge that different combination.
• Avoid the loser combinations (Matthies + Yamada; Murai + Yamada). The Board's reasoning there — the extra reference's teaching wouldn't have been applied to the primary reference — will be recited back at any defendant who pleads them.
• The record's most under-appreciated opening is the [1-b] measurement problem (§4.3): claim 8's express center-to-center recitation is a foothold for reading claim 1's distances as gap measurements, which would break the Phan mapping. Samsung will very likely press exactly this on appeal.