US 6857716

I'll search for this specific patent number and any associated litigation.

Let me check for litigation and verify the USPTO record.

Let me do a couple more targeted checks for any 2026 appellate activity and the patent's legal status.

US Patent 6,857,716 — Analysis

Sourcing note

The authoritative record used here is the full granted text fetched from Google Patents (US6857716B1, https://patents.google.com/patent/[US6857716](/patent/US6857716)/en). I ran searches against general web sources (which index USPTO data and the patent text) and for Federal Circuit / 2026 appellate activity. Below I flag where I could not confirm something.

One important ID caveat (literal reading, no auto-correction): Searching "6857716" surfaces two unrelated documents. The one you asked about is US 6,857,716 B1 (Seiko Epson, printer control). There is also a Japanese patent JP 6857716 B2 — a Panasonic Intellectual Property Corporation video-coding (encoder/decoder) patent, filed 2018, registered 2021. These are separate national numbering series and separate inventions; JP6857716 is not the Japanese counterpart of US6857716. The actual Japanese family member of the US patent is JP4352529B2.

Bibliographic data (US 6,857,716 B1)

Assignee/inventor confirmation: a patent-leaderboard listing for Seiko Epson inventor Toshinori Nagahashi confirms 6857716, issued 2005-02-22. Justia cross-references also list "6857716 | February 22, 2005 | Nagahashi." Consistent with the Google Patents record.

Abstract (as granted)

The invention provides high-speed, high-quality printing when the vertical and horizontal resolutions of print data differ from those a printer can process. A table is provided in which the positional relationship, in each printing pass, of each of a plurality of vertically aligned printing-head pins to a plurality of raster lines in a predetermined range is examined, so that the pins to be actuated in each pass are determined from the pin-to-raster relationship. This is the "raster-line/pin-relationship table." A printer driver receives print data from a print-data forming unit, consults the table, and actuates the pins for each printing pass accordingly.

Plain-language overview of the independent claims

The patent has 12 claims; the independent claims are 1, 6, 11, and 12.

Claim 1 — Method (pins). A method of controlling a printer when two conditions hold: (a) the head's vertical transfer step is not the reciprocal of an integral vertical resolution of the print data, and (b) the pin pitch is neither an integral multiple of the vertical resolution nor the reciprocal of an integral resolution. The method: examine, for each printing pass, where each of several vertically arranged pins falls relative to the raster lines to be printed in a given range (using the vertical resolution, pin pitch, vertical transfer amount, print-starting raster line position, and the number of horizontal passes needed); build a "raster-line/pin-relationship table" that says which pins to fire for which raster lines in each pass; determine the number of passes and the print-starting raster line; consult the table accordingly; and print by firing the pins per the table.

Claim 6 — Device (pins). The apparatus counterpart: a raster-line/pin-relationship table (built for the same two above conditions/parameters); a print-data forming unit; a printer driver that determines the print-starting raster line and pass count, consults the table, and outputs pin-driving signals per pass; a printing-head transferring unit that positions the head based on the driver's signal; and a data transmitting unit that sends print data and printing information from the driver.

Claim 11 — Method (marking elements). Substantively the same method as claim 1, but the pins are generalized to "marking elements," with a "raster-line/marking element-relationship table."

Claim 12 — Device (marking elements). The apparatus counterpart to claim 11, generalizing the "pin" terminology of claim 6 to "marking elements."

Dependent claims 2–5 and 7–10 add the refinements that (i) multiple relationship tables are prepared according to the varying print-starting raster line, and (ii) the correct one is selected based on the determined starting line and then consulted to fire the pins/marking elements.

CAFC 2026 docket check — result and uncertainty

I searched for Federal Circuit / 2026 appellate activity tied to patent number 6857716 and found no docket entry, opinion, or appeal associated with this patent. Searches for "6857716" plus "Federal Circuit 2026" returned only unrelated 2026 Federal Circuit materials (e.g., a Gibson Dunn Federal Circuit update covering cases such as AGI/Farmers Edge, Ollnova v. ecobee, Ironburg v. Valve, and the Newman v. Moore Supreme Court docket) — none involving the '716 patent.

Explicit uncertainty: I could not access a primary CAFC/PTAB docket database directly and cannot rule out a case that is not well indexed by general web search. Given the patent expired (fee-related) with an anticipated expiration of 2020-10-03, active infringement litigation or a 2026 appeal asserting it is unlikely, but I am not able to state that as a certainty. I also found no record of a 2026 (or any) Federal Circuit appeal for this patent in the sources reachable here.

Context on Epson litigation found (not this patent): The Epson enforcement actions my searches surfaced concern ink-cartridge patents (e.g., U.S. 6,502,917; 6,550,902; 6,955,422; 7,008,053; 7,011,397; 8,366,233; etc.) in ITC investigations 337-TA-565 and 337-TA-946 and related Oregon district-court cases. US 6,857,716 does not appear in those actions; it is a printer-control/raster-pin method patent, not an ink-cartridge patent.

Items I could not independently verify

• Whether the "current assignee" ever changed from Seiko Epson (Google Patents lists Seiko Epson; no reassignment away from Epson was found, but I did not inspect the full USPTO assignment record).
• Any maintenance-fee lapse detail beyond the "Expired – Fee Related" status and 2020-10-03 anticipated expiration date.
• The complete, verbatim claim 1 ending: the text available here renders claim 1's final clause ("the pitch of the pins of the printing head, the amount of the unit of vertical transfer…") as a trailing fragment of the examining step rather than a cleanly separated element; I have summarized the substance rather than reproducing a possibly garbled claim boundary.

I'll search for litigation involving US patent 6857716 across multiple sources.

Initial searches returned mostly unrelated results. Let me search litigation-specific databases.

Let me search more targeted litigation databases and the patent's specific number in court contexts.

Findings: Litigation involving US Patent 6,857,716

No litigation involving US Patent 6,857,716 was found.

I searched litigation-oriented sources (including queries aimed at the Unified Patents litigation caselist, CourtListener/RECAP dockets, PACER-derived aggregators, RPX Insight, and general patent-litigation coverage), and none returned any case in which U.S. Patent No. 6,857,716 was asserted, challenged, or otherwise placed at issue.

What the searches did surface (and why it is not litigation for this patent)

1. The patent itself, but no enforcement record. Searches for the number resolve to the patent's own documents — Print‑controlling method and print‑controlling device for printer, issued Feb. 22, 2005 to inventor Toshinori Nagahashi, originally assigned to Seiko Epson Corp., filed Oct. 3, 2000, priority Oct. 4, 1999 (JP 28324799). Google Patents lists its legal status as Expired – Fee Related, with an anticipated expiration of 2020‑10‑03. No litigation "Litigation" or docket field appears on its page; only prior-art citations and citing documents (prosecution citations, not court cases).

2. Seiko Epson / Epson enforcement campaigns are unrelated patents. Numerous Epson suits and ITC investigations (e.g., ink-cartridge matters such as ITC 337‑TA‑565, 337‑TA‑946, and related Oregon district court actions like Civil No. 06‑477‑BR) involve different Epson patents (e.g., the '422, '749, '116, '917, '902, '053 patents). None of the retrieved materials list 6,857,716 among the asserted patents.

3. The "Cited By" / "Citations" lists are not litigation. The Flexiworld Technologies and other documents that reference 6,857,716 (e.g., US 7,905,401; US 7,805,720) do so as prosecution prior-art citations, not as infringement or validity challenges in a court or the PTAB.

Important disambiguation (I did NOT auto-correct or merge these)

• US 6,841,716 — appears in a Noven Pharmaceuticals v. Mylan Technologies infringement case (D. Del., filed 2015‑04‑23). This is a different patent number and unrelated to the Epson patent. I am not treating it as the patent you asked about.
• JP 6857716 B2 — a Japanese patent (Panasonic, video coding) that matched the raw number string. This is a different patent, different jurisdiction, different owner, and not US 6,857,716.

Confidence and caveats

• Confidence that no asserted litigation exists: Moderate-to-high, based on the absence of any docket, complaint, or PTAB record tying this patent number to a case in the sources reachable in this session.
• Caveats: I could not directly query the paywalled/JS-heavy litigation portals (Unified Patents caselist, Docket Navigator, Lex Machina) for a certified "zero results" record; my conclusions rest on indexed search results. It is also possible that this patent was asserted in a case that is not well indexed, or never asserted at all.

Recommendation for a definitive negative: run the patent number directly in Docket Navigator, Lex Machina, Unified Patents' Patent/Litigation search, and a PACER docket search for the patent number. If those return nothing, that is a strong basis to state the patent has no known litigation history — which is consistent with the patent's age (expired 2020) and its narrow subject matter (dot-matrix/impact-printer raster-line-to-pin mapping, a technology largely superseded before this patent's term ran).

If you can share access to a specific docket database, I can help interpret the results.

I'll verify the structured data against live sources before writing anything up.

Two searches so far returned no PTAB hits on this patent (and several false positives on unrelated "685"-numbered patents). Let me run a few more targeted checks before concluding.

Proceedings overview

Zero AIA trial proceedings are on file for US 6,857,716. The USPTO Open Data Portal (ODP) structured record returns no IPR, PGR, or CBM petitions as of the most recent ingest, and my independent web checks (Google Patents' proceeding listings, PTAB E2E/P-TACTS petition text, and general searches for the patent number) surfaced no AIA trial, no final written decision, no institution denial, and no Federal Circuit appeal arising from a PTAB trial — only unrelated false positives (e.g., a Panasonic JP 6857716 B2 and the Trinity Info Media v. Covalent "’685 patent," neither of which is this patent). The defensive read is therefore not "hardened by surviving IPRs" nor "claims canceled" — it is "untested at the PTAB." There is no estoppel on the books, no FWD to cite, and no claim-level validity adjudication from the Board to lean on. Any invalidity case you build is a first-generation, clean-slate challenge.

Proceedings

No AIA trial proceeding identified — no docket number to report

Because the canonical structured block returns an empty set, there is no proceeding number to list, and per the operating rules I will not invent one. For completeness, here is what each diligence check returned:

• Type: N/A (no IPR / PGR / CBM found)
• Filed: N/A
• Status: N/A — no PTAB activity on file
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A — no claim of US 6,857,716 has ever been canceled, confirmed, or otherwise adjudicated by the Board
• Settlement / termination: N/A
• Appeal: N/A — no PTAB-originating CAFC appeal located
• Defensive value: The absence of any PTAB track record means the patent's claims are completely untested before the Board. You get no free claim cancellations, but you also inherit no adverse precedent and no § 315(e) estoppel constraining your invalidity theories. You would be the first petitioner, which carries both first-mover cost and first-mover freedom.

Strategic summary

Claim status: all claims UNTESTED. US 6,857,716 issued 2005-02-22 with 12 claims: claims 1–5 (method for controlling a printer using raster-line/pin-relationship tables), claims 6–10 (corresponding print-controlling device), and claims 11–12 (method and device reciting "marking elements" instead of "pins" — an apparent claim-drafting variant covering non-impact marking). No claim has been canceled by the Board, and no claim has been confirmed by the Board. There is nothing in the PTAB record that narrows your exposure and nothing that locks in a claim construction. Note that the patent reached its 20-year, anticipated expiration on 2020-10-03 (per the ODP legal-status record), and the Google Patents status is "Expired – Fee Related." That matters more than any PTAB analysis: infringement liability for a now-expired patent is limited to pre-expiration damages, and prospective injunctive theories are off the table. Any current demand built on ongoing infringement should be scrutinized on that basis first.

Estoppel landscape: wide open. § 315(e)(2) estoppel only attaches to a party that was a petitioner, RPI, or privy in an IPR that reached a final written decision. Here, no IPR exists, so no estoppel operates against anyone. Every prior-art ground — § 102 anticipation, § 103 obviousness, plus § 112 written-description/enablement attacks — remains fully available in district court and at the Board. If you file an IPR, the standard § 315(b) one-year bar runs from service of the complaint on you, so calendar that clock immediately upon being served.

Pattern signals: none. There is no recurring petitioner, no defensive aggregator (no Unified Patents, RPX, or similar) in the chain, and no serial-challenger behavior. The patent's "Cited By" list in Google Patents is dominated by Flexiworld Technologies wireless-printing continuations, not by invalidity-driven activity. On the patent-owner side, the assignee of record is Seiko Epson Corporation (a large operating company, not an NPE), and the patent never appears to have been sold into an assertion vehicle. The absence of PTAB filings is itself informative: a 2000-priority printer-control patent that expired in 2020 and was never IPR'd suggests it was never a serious litigation lever — which is consistent with a low-value, expired asset rather than a well-asserted one.

Recommended next steps

• No PTAB workstream is available to you by default. There is nothing to distinguish on institution, no FWD to mine for claim constructions, and no cancellation order to cite. Do not represent to a court or adversary that any claim of this patent has been held invalid — it has not.
• Lead with expiration, not invalidity. The patent's anticipated expiration date is 2020-10-03 and its status is expired. Confirm the maintenance-fee/expiration record on the USPTO Patent Center for US 09/677,578 and the corresponding Global Dossier entry before responding to any demand. If a demand letter asserts ongoing infringement of an expired claim, that is a threshold defect worth quoting back at the sender.
• If you nevertheless need an invalidity posture, you have an unimpeded field: § 102/§ 103 on the raster-line/pin-relationship-table teachings, and § 112 on claims 11–12's "marking element" broadening, are all unencumbered by estoppel. File any IPR petition before the § 315(b) one-year anniversary of service on you; the Board's trial deadline is then a statutory 12 months from institution (extendable up to six months for good cause under § 316(a)(11)).
• Verify my finding independently before relying on it. I could not locate any PTAB proceeding, but the authoritative source for this determination is the USPTO PTAB E2E / P-TACTS system and the ODP structured record. Search by patent number 6,857,716 there directly; if a recently-filed petition exists but is not yet indexed, it would appear there first, and the structured block's "no proceedings" result should be treated as current-as-of-ingest rather than eternally true.

One caveat on the source text: the full-text claim 1 and claim 6/11/12 reproductions in the provided patent HTML show OCR/reflow artifacts — the limitation enumerating "the pitch of the pins … the amount of the unit of vertical transfer … the position of the print-starting raster line …" dangles after the "printing" step rather than sitting in the examining clause. If claim language becomes outcome-determinative, read it against the USPTO's official claims page rather than the Google Patents scrape.

I'll research the assignment record for this patent carefully, starting with the USPTO Assignment Center and cross-referencing sources.

Let me dig into the specific assignment records and cross-reference sources.

Let me try to locate the specific reel/frame and check NPE/asserter databases.

Ownership & Assignment Record — US Patent 6,857,716

Sourcing note: The USPTO Assignment Center (https://assignmentcenter.uspto.gov/) and its legacy mirror are JavaScript/POST-driven and did not return a machine-readable record to my searches; I could not load a reel/frame page directly. What I can confirm comes from the Google Patents "Reassignment" event table on the authoritative record (https://patents.google.com/patent/[US6857716](/patent/US6857716)/en), the Justia cross-reference, and the PatentLeaderboard inventor record. I did not obtain the reel/frame number or the correspondent of record. I flag every place where a value is unverified rather than guess. (This builds on the earlier-session note that raw string "6857716" also matches JP 6857716 B2, a Panasonic video-coding patent — a different series and invention, not to be merged here.)

Inventors

Pattern notes:

• No co-inventors on this patent — the record shows "—" for co-inventors. All four independent claims and all dependent claims trace to a single inventor.
• No evidence of inventor departure or fire-sale precursor. I found no June-2000-to-2001 departure, resignation, or inventor-to-third-party conveyance. Nagahashi is also named on later Epson filings (e.g., JP 2006-127465, "Printing system, printing device, client device…", Seiko Epson applicant, 2006), which indicates he remained with Epson well past this filing — the opposite of the "all inventors gone within 12 months" pattern.
• Employer-of-record address: Epson inventor addresses of this vintage are typically listed c/o Seiko Epson Corporation, 3-5 Owa 3-chome, Suwa-shi, Nagano 392-8502 (the printer-division address) — not directly confirmed on this document, so treat as an inference from sibling Epson filings, not a verified fact for the '716.

Original assignee

Seiko Epson Corporation (Shinjuku-ku, Tokyo, Japan) — the entity named on the issued patent and the only assignee ever recorded.

• Line of business: Consumer and commercial imaging/hardware — inkjet and impact (dot-matrix) printers, projectors, watches, semiconductor and electronic devices. The '716 is squarely a printer-control patent (pin/raster firing logic for impact print heads), i.e., Epson's core business, not a licensing arm.
• Product embodying the claims: Yes — the specification is written around an impact printer ("the printer 1 is a common impact printer") whose driver consults raster-line/pin-relationship tables to map 120×120 dpi binary data onto a 120×72 dpi mechanism. Epson shipped dot-matrix/impact printers in this era, so the claimed subject matter maps to a shipping product line.
• Current status: Operating. Seiko Epson Corporation remains a publicly listed operating company (Tokyo Stock Exchange). I found no bankruptcy, merger-away, dissolution, or divestiture of this patent. The patent itself is Expired – Fee Related, anticipated expiration 2020-10-03.

Assignment timeline

I could confirm exactly one recorded event — the original inventor-to-employer assignment. The Google Patents reassignment table shows a single 2001-01-17 entry; there is no post-issuance transfer anywhere in the reachable record.

• Executed [date not retrieved] / recorded 2001-01-17 — Reel NNNNNN/NNNN (not retrieved)
  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Toshinori Nagahashi (sole inventor)
  • Assignee: Seiko Epson Corporation
  • Correspondent: Not retrieved — I could not obtain the attorney/firm or authorized user of record from the assignment cover sheet. (Note: the cover sheets I retrieved in this session for other Epson patents of the 2005 era name "G. Gregory Schivley"/"Bryant E. Wade" as submitting attorneys and "Epson R&D" as authorized user; those belong to different patents and I will not attribute them to the '716.)
  • Context: Original inventor→employer assignment (work-for-hire confirmation), executed in the ordinary course of prosecution — not a fire-sale, reorg, securitization, or transfer-to-asserter. It is the only conveyance on the property.

No further assignments are recorded. The patent never left Seiko Epson: no transfer to an "IP/Holdings/Ventures" entity, no security interest, no license record, no change of name. Per the assignment-search limitation above, I state plainly: the only ownership record I could verify is the original assignment to Seiko Epson; I found no subsequent assignment.

Timeline diagram

timeline
    title Ownership of US 6857716
    1999 : JP priority filed by Nagahashi
    2000 : US application filed
    2001 : Assignment recorded to Seiko Epson
    2005 : Patent issued
    2020 : Patent expired fee related

NPE / troll-pattern signals

Verdict

Insufficient data — (as the rubric defines it: only the original assignment is on record).

By the strict category definitions, the only verifiable event is the inventor→employer assignment recorded 2001-01-17 to Seiko Epson Corporation; there is no post-issuance chain to analyze, so no NPE signal can be affirmatively scored. Substantively this is the cleanest possible non-NPE outcome: the patent stayed with the original operating-company assignee (Seiko Epson), issued 2005-02-22, and expired fee-related on 2020-10-03 with zero recorded transfers — consistent with a captive corporate patent that was never monetized or asserted. I flag that I could not retrieve the reel/frame or correspondent, so a definitive negative rests on index-based evidence; to certify, run the patent number directly in USPTO Assignment Center (https://assignmentcenter.uspto.gov/), PEDS/ODP, and PEDS-adjacent docket tools.

Verification links:

• USPTO Assignment Center (search by patent number 6857716): https://assignmentcenter.uspto.gov/
• Google Patents legal events / reassignment table: https://patents.google.com/patent/US6857716/en
• Legacy assignment search mirror: https://assignment.uspto.gov/patent/index.html

I'll research the prior-art citations for US 6,857,716 and verify each reference against primary sources.

Let me verify inventor details and check the Japanese family member and any additional references.

Prior-Art Analysis for US 6,857,716 B1

Scope note on method. I re-verified the target through its USPTO-derived record (Google Patents, https://patents.google.com/patent/US6857716/en, and the Justia mirror https://patents.justia.com/patent/6857716). As noted in the earlier sections, the identifier 6857716 resolves to two unrelated documents; this analysis concerns only US 6,857,716 B1 (Seiko Epson, "Print-controlling method and print-controlling device for printer," filed 2000-10-03, issued 2005-02-22, priority 1999-10-04 via JP 28324799). I did not treat JP 6857716 B2 (Panasonic video coding) or US 6,841,716 (Noven v. Mylan) as this patent.

Important limiting fact about the prior-art universe. The face of US 6,857,716 lists exactly three patent citations, all examiner-cited, and no non-patent literature. There is no "Other References" field on the record. The "Cited By" documents (Flexiworld Technologies et al.) are later documents that cite the '716 patent — they are not prior art against it and are excluded. The Japanese family member JP4352529B2 and the priority application JP28324799 are the same invention, not prior art.

The three references are examined below.

Reference 1 — US 5,959,646 A (Lexmark)

Brief description. A thermal ink-jet printer method in which a shingle/interlace level (100%, 50%, 25%, etc.) is assigned to each raster independently, per color plane. Bit masks ("SHINGLE 100_1", "SHINGLE 50_1/2", "SHINGLE 25_1…25_4") define which pixels of a raster are available in each printhead scan; the printer computes the "most limiting interlace level" to set the paper-advance distance, and re-aligns rasters with different orifices across successive scans. It is fundamentally a multi-pass / interlace-management and media-advance disclosure.

Relevance to US 6,857,716. This is the closest reference on the multi-pass framework — it uses the vocabulary of "rasters," "raster lines," "passes," and per-pass bit allocations that the '716 patent also uses. It arguably touches claim 1's element "a number of passes in the horizontal direction required for printing the predetermined printing range."

Section 102 assessment — no anticipation. US 5,959,646 is a qualifying §102(a)/(b) reference (its printed publication date of 1999-09-28 precedes the '716 priority date of 1999-10-04). However, anticipation requires every claimed element in a single reference. US 5,959,646 does not disclose:

• the two threshold conditions of claim 1 — i.e., that the vertical transfer unit is not the reciprocal of an integral vertical resolution of the print data, and that the pin pitch is neither an integral multiple nor the reciprocal of an integral resolution;
• any "raster-line/pin-relationship table" mapping vertically arranged pins to raster lines based on pin position in each pass; and
• the determination of a print-starting raster line and selection among multiple pre-computed tables keyed to it (claims 2–5).

It is at best a §103 secondary reference for the multi-pass/interlace context — not a §102 anticipatory reference against any of claims 1–12.

Reference 2 — US 6,336,701 B1 (Hewlett-Packard) (best §103 candidate)

Brief description. An ink-jet method for placing drops from multiple printheads per colorant in which the media is advanced by a "microstep" distance d = (m·S) + S/n (S = nozzle spacing; n = integer > 1), the printhead's new positional relationship to the pixel matrix is determined after each advance, and nozzles are "selected in closest proximity to target pixels for dotting each targeted pixel." Pen-to-pen spacing is expressly not required to be an integer multiple of nozzle spacing, and the target resolution in the media-advance axis exceeds the nozzle packing density.

Relevance to US 6,857,716. This is the most technically relevant of the three. It independently discloses two ideas that map onto the '716 core:

1. a sub-nozzle-spacing (non-integral) media/head advance — conceptually parallel to the '716 condition that the vertical transfer unit is not the reciprocal of an integral vertical resolution; and
2. choosing the nozzle/pin nearest the target raster row ("selecting nozzles in closest proximity to target pixels"), which is exactly the '716 selection principle ("pins on the same positions as … the raster lines … or pins nearest to the positions").

Section 102 assessment — no anticipation (but real §103 exposure).

• Date qualification (§102(e)): Filed 1999-12-22 — after the '716 Japanese priority date (1999-10-04) but before the '716 U.S. filing date (2000-10-03). Because of the Hilmer doctrine, an applicant's foreign priority date is generally unavailable to antedate a §102(e) reference; the '716 applicant would need an actual pre-1999-12-22 invention date. This reference therefore appears to be properly citable under §102(e) as a U.S. patent granted on an application filed before the applicant's invention. (I flag this rather than assert it as final, since it depends on facts not in the public record.)
• Element coverage: Even granting its prior-art status, US 6,336,701 does not disclose the pre-computed raster-line/pin-relationship table that is the heart of claims 1, 6, 11, and 12, nor the selection of that table based on a print-starting raster line (claims 2–5, 7–10). Its context is nozzle-misalignment correction across multiple same-color printheads, not resolution-mismatch resolution by table lookup.

Conclusion: Not a §102 anticipatory reference for any claim. It is the strongest §103 primary reference for the "microstep / nearest-element selection" limitations, and would likely be combined with US 5,959,646 (multi-pass interlace).

Reference 3 — US 2003/0151640 A1 (Marra) — not §102 prior art

Brief description. An ink-jet method that selectively underfeeds the print medium (advances slightly less than the swath/pass height) during single-pass printing but not during multiple-pass (shingling) printing, to mask indexing errors without introducing shingling-mode artifacts. It discusses subdivided nozzle groups and four-pass shingling as background.

Relevance and §102 assessment — it is NOT prior art against US 6,857,716. Both of its critical dates post-date the '716 patent:

• '716 priority: 1999-10-04; '716 U.S. filing: 2000-10-03
• Marra publication: 2002-02-11 (filed) / 2003-08-14 (published) — about two years after the '716 U.S. filing.

Under §102(a), (b), and (e), a reference must predate the applicant's invention/filing date. Because this publication was filed and published entirely after the '716 filing date, it cannot anticipate any claim 1–12 (and, absent an earlier provisional not shown on the record, has no qualifying §102 date at all). Its appearance in the citation list is best read as a citation to the general state of multi-pass inkjet printing or as an anomalous/later-added citation; I am not auto-correcting or re-characterizing it, only reporting that on its stated dates it is not statutory prior art. This is the one item in the citation list that a validity analysis should treat as non-prior-art background.

Summary table — §102 exposure by claim

Overall conclusion: None of the three citations anticipates any of claims 1–12 under 35 U.S.C. §102. Each fails to disclose at least (i) the two negative conditions recited in every independent claim (non-integral vertical transfer unit; pin pitch neither an integral multiple nor the reciprocal of an integral resolution), and (ii) the pre-computed raster-line/pin-relationship table with selection among multiple tables based on the print-starting raster line. The realistic invalidity theory is a §103 combination led by US 6,336,701 (microstepping + closest-element selection) in view of US 5,959,646 (multi-pass interlace), not §102 anticipation.

Caveats and confidence

• Confidence in the citation list: High — the three references match across Google Patents and the Justia mirror, and Justia independently names the same three (Bates et al.; Hickman; Marra et al.).
• Date anomaly (flagged, not corrected): The Marra publication (filed 2002-02-11, published 2003-08-14) post-dates the '716 filing date, so it is not §102 prior art. I report this literally rather than "fixing" it; if a validity opinion is needed, confirm whether Marra claims an earlier provisional that the database record omits.
• §102(e)/Hilmer nuance: I could not complete a live USPTO Patent Center/file-wrapper pull (tool-call limit reached), so the §102(e) status of US 6,336,701 and any unreported references are based on the public bibliographic record and the Hilmer doctrine, not a certified file history. For a definitive date-of-invention and citation analysis, pull the '716 file wrapper from USPTO Patent Center and check for any cited NPL or remarks distinguishing these references.
• Missing data: I did not verify the complete prosecution history (e.g., examiner's reasons for citing Marra, or any interview/amendment). Those would clarify the intended role of each reference.

I'll verify the three cited references' dates, content, and applicability before running the § 103 analysis.

Now let me verify priority/filing dates that affect whether these references can even qualify as prior art, particularly US6336701B1 and the Marra publication.

Obviousness Analysis — US 6,857,716 B1 under 35 U.S.C. § 103

Builds on the earlier sections. Priority date 1999-10-04 (JP 28324799); US filing 2000-10-03; granted 2005-02-22. Pre-AIA § 103 governs.

1. The threshold problem: only one of the three "cited" references is cleanly prior art

Before combining anything, each reference must be available as prior art as of the operative date (here, at least the 2000-10-03 US filing date for § 102(b)/(e) purposes, and the 1999-10-04 priority date for § 102(a)/(e) purposes if the JP priority is perfected).

Consequence: the only unambiguously available reference on this record is Lexmark '646. HP '701 is the reference that actually carries the dispositive teaching, and its availability turns on whether the Japanese priority document supports the claims. Marra is legally unusable here (it appears in the citation list, but its dates are fatal). Everything below is therefore presented as contingent on '701 being reachable, with the '701-antendated scenario analyzed separately.

(Caveat carried forward from the earlier section: claim 1's closing parameter list — "the pitch of the pins …, the amount of the unit of vertical transfer …, the position of the print-starting raster line, and a number of passes in the horizontal direction …" — renders as a trailing fragment. I read it as belonging to the "examining" step, as the specification confirms.)

2. Level of ordinary skill

A POSITA here is a printer-control engineer with ~2–3 years' experience in raster-based printing (inkjet interlacing/shingling, dot-matrix head control) and familiarity with the mathematics of mapping a print head's discrete marking elements onto a target raster grid under sub-line media/head indexing. This is the level against which both references are read; both are ordinary engineering disclosures in the same art.

3. Independent claim 1 — element mapping

Claim 1 requires, in the non-commensurate case (transfer step ≠ 1/integer-resolution; pin pitch ≠ integer multiple or 1/integer multiple of resolution):

Gap on the face of the art: the precomputed lookup table indexed by print-starting raster line is not expressly disclosed; '701's selection is characterized as an algorithmic/real-time "logical selection," and '646's table is a per-raster interlace-level/mask table rather than a table keyed to the starting line. This is the narrowest point of novelty.

Why that gap does not save the claim (KSR analysis):

• The '716's own specification states the pin/raster relationship repeats at a period of five raster lines (L0→L5→L10…), so there are only a finite, small number of distinct start-line "phases." Reducing a phase-dependent calculation to a set of precomputed tables is a recognized, predictable programming expedient ("common sense" optimization under KSR), not an inventive contribution.
• The table's contents follow arithmetically from the parameters already recited and disclosed by '701 (resolution, spacing, step, pass count). A tabulation of a known algorithm that produces a known mapping is obvious where the underlying method is obvious.
• "Pins" vs. "nozzles" is a substitution of one known marking element for another (impact pins ↔ inkjet orifices) to obtain the same predictable result — and the applicant's own claims 11/12 concede this by generalizing to "marking elements."

Motivation to combine (if using '646 + '701): both are multi-pass, raster-by-raster print-control disclosures in the same art, addressing the same problem — which marking elements fire on which raster lines in which pass when the head's position relative to the raster grid shifts between passes. '701 supplies the non-commensurate, sub-pitch advance and the nearest-element selection; '646 supplies the table/mask-per-raster implementation of pass assignment. A POSITA implementing '701's microstepping for a fixed print-mode/resolution would naturally precompute the per-start-line element/raster assignment rather than recompute it each pass, with a reasonable expectation of success (both references report improved resolution/quality).

Confidence that claim 1 is obvious over '701 alone (or '701+'646): Moderate-high on the merits, subject to the § 102(e) availability problem in § 1.

4. Independent claim 6 (device) and claims 11–12

• Claim 6 is the apparatus mirror of claim 1: table + print-data forming unit + printer driver (determines start line and pass count, consults table, outputs pin-driving signals) + head-transferring unit + data-transmitting unit. '701 is expressly a computer memory / computer-readable-code claim set with print-mode selection, transport control and firing-algorithm routines — i.e., the same driver/controller architecture — and '646's host-transmits-raster-information model supplies the forming-unit→driver→head data path. The '716's own FIG. 11 blocks (printer driver 22, pin selector 24, head transferring unit 25, data transmitting unit 26) are conventional controller modules; the only substantive element is again the table.
• Claims 11–12 merely rename "pins" as "marking elements." This is the applicant's own express generalization and removes any argument that the pin/nozzle distinction is inventive. These claims are obvious to the same or greater degree as claims 1/6 ('701 is an inkjet, nozzle-based disclosure).

5. Dependent claims 2–5 and 7–10

These add: (i) multiple tables prepared according to the varying print-starting raster line; (ii) selection of the correct table from the determined start line; (iii) consulting the selected table to fire the elements.

This is a design choice / finite-set enumeration: '716 acknowledges the pattern repeats every five lines, so at most five tables are needed. Preparing a table for each possible phase and indexing by the detected start position is the textbook routine implementation of claim 1's concept. Under KSR, "when a work is available in one field of endeavor, design incentives and other market forces can prompt variations of it," and a POSITA would have foreseen the benefit (avoiding per-pass recomputation). These claims are obvious if claim 1 is, and are not independently patentable.

6. Scenario if HP '701 is antedated (the real defensive position)

If the '716's JP priority (Hilmer) defeats '701's § 102(e) date, the surviving cited art is Lexmark '646 alone. On that record the obviousness case is weak:

• '646 is a color-inkjet shingling patent aimed at banding/bleed artifacts. Its advance is "a distance corresponding to a most limiting interlace level" — it does not teach or suggest a sub-pitch, non-commensurate head/media step, nor nearest-element selection to achieve a vertical resolution finer than the element pitch.
• The two claim-1 preconditions are therefore not taught by '646, and the nearest-pin mapping that is the heart of the claims is absent.
• Marra '640/'685 cannot fill the gap — it is not prior art by date (2002 filing), and its advance in multipass mode is "approximately equal to the multiple pass height," i.e., the commensurate case, which cuts against, not toward, the claim.

So the outcome is binary and driven entirely by the date question: avail '701 → claims 1/6/11/12 look obvious; exclude '701 → the cited record does not establish a prima facie § 103 case. A practitioner would want to look beyond the three cited documents — e.g., the Hickman interlacing family that '701 itself incorporates (US 4,965,593; US 5,583,550 "Ink drop placement for improved imaging"; and '701's cited US 5,825,378, US 5,663,624) — plus contemporaneous dot-matrix/impact interlace art, to rebuild the rejection on references whose dates are not in doubt.

7. Secondary considerations / rebuttal levers

• No objective indicia on this record. No evidence of unexpected results, commercial success with nexus, long-felt need, or industry praise tied to the claimed table structure. The patent expired (fee-related; anticipated 2020-10-03) and, per the earlier section, shows no litigation or PTAB history, so no adjudicated secondary considerations exist.
• Applicant's strongest argument is not the table but the specificity of the two preconditions (transfer step not a reciprocal of an integral resolution; pin pitch neither an integer multiple nor a reciprocal of an integral resolution). Applicant would argue '701's microstep is a compensation for pen-to-pen misalignment and '646's masks are artifact-reduction, neither being directed to the resolution-mismatch up-conversion problem. Rebuttal: '701 expressly claims "printing resolution … greater than nozzle packing density" and "closest proximity" element selection — the same functional result the '716 achieves for 120×120 data on a 120×72 machine.
• Avoid hindsight: the examiner must cite a reference teaching the non-commensurate step + nearest-element selection ('701's claim 7 and its d = m·S + S/n advance), not merely the general idea of tables.

8. Bottom line

Confidence: moderate-high as to claims 11/12 and the table/multiple-table claims; moderate as to claims 1/6. The analysis is gated by two date issues I flag rather than resolve: (1) whether HP '701's 1999-12-22 filing survives the '716's 1999-10-04 JP priority under Hilmer (if it does not, the cited record fails), and (2) that Marra US 2003/0151640 / US 6,612,685 is not prior art at all and must be excluded from any rejection. No litigation or PTAB decisions construing these claims were found, so no authoritative claim construction or obviousness holding exists to rely on.