US 6487616

US Patent 6,487,616 — "Controller for data recorder"

Sources consulted: Google Patents (US6487616B1), USPTO-derived gazette text (Indiana VDL, Issue 1264‑4), Justia Patents, Stanford NPE Litigation Database, MediaTek quarterly financial-disclosure notes (ROC years 94–95), E.D. Tex. / C.D. Cal. case records. Google Patents mirrors USPTO bibliographic data; direct PatentCenter/PAIR access was not available to me in this session.

1. Bibliographic data

Related continuations (not to be conflated with the patent itself): US2004/0184374A1 and US2005/0094525A1, both titled "Controller for data recorder" and both claiming the 1999‑11‑22 priority. Sibling patents in the same Sanyo family include US 6,594,213 and US 7,106,675 (per the "Cited By" table).

Representative art cited on the face of the patent includes US 5,436,875 (Sony), US 5,491,677 (Sony), US 5,745,545 (Matsushita), US 5,818,801 (Sanyo) and US 6,317,809 (IBM).

2. Abstract (as issued)

A controller interrupts and restarts writing data from a buffer memory to a medium and prevents buffer underrun errors. The controller includes an address memory for storing a recording medium address or a buffer memory address, which indicate the location of the data when the interruption occurred. A synchronizing circuit sequentially reads data from the recording medium and data from the buffer memory prior to the interruption while synchronizing the written data and the stored data. A restart circuit restarts writing data when the recording medium address or the buffer memory address matches the address stored in the address memory.

3. Plain-language overview of the independent claims

Claim 1 — apparatus (single address memory, "retry" limitation).
A controller that can stop and then resume writing to a disc when data flows out of a buffer memory. Three parts:

1. Address memory — when writing is interrupted, it saves at least one of (a) the address on the recording medium and (b) the address in the buffer memory, i.e., a bookmark of where the interruption happened.
2. Synchronizing circuit — re-reads the tail end of the data already on the medium and the corresponding data still in the buffer, running the two streams together so they line up again (this is the "recording restart mode" of the spec, which goes back m sectors before the interruption to re-establish servo lock and timing).
3. Restart circuit — resumes writing when the re-read address (medium or buffer) matches the saved bookmark. Crucially, if a match shows up before synchronization has actually completed, the restart is suspended and the read-and-synchronize pass is repeated. That repeat loop is the claimed safeguard against an external shock or glitch causing a false/early match.

Claim 4 — apparatus (dual address memories + two detectors + arbiter).
The richer version: a first address memory saves the medium address at interruption and a second address memory saves the buffer address at interruption. The synchronizing circuit re-reads and aligns both streams. A first location detection circuit raises a restart instruction when the medium address read back matches the first memory; a second location detection circuit raises one when the buffer address read back matches the second memory. A determination circuit (the retry determination circuit 44 in FIG. 1) decides whether to actually restart based on both signals — per dependent claims 5–6, restart only when the two signals arrive simultaneously, otherwise the synchronization pass is repeated. Dependent claims 7–9 add sector-unit granularity and the "predetermined number of sectors" back-off, sized to be sufficient for synchronization.

Claim 10 — method counterpart to claim 1. Same steps in method form: store at least one address bookmark; sequentially re-read the previously-written medium data and the pre-interruption buffer data; synchronize them; restart writing on a match; and repeat the read/sync steps if a match occurs before synchronization completes.

Claim 11 — method counterpart to claim 4. Store both the medium address and the buffer address at interruption; re-read and synchronize the two streams; generate a first instruction signal on a medium-address match and a second instruction signal on a buffer-address match; and decide whether to restart based on the two signals.

4. Litigation / docket check

I searched for the exact number 6487616 / 6,487,616 in connection with CAFC 2026 dockets and found no Federal Circuit appeal in 2026 (or any year) involving this patent. Results for similar numbers were excluded, per your instruction.

What does exist on the record is older district-court activity, all now closed:

• Sanyo Electric Co. Ltd v. MediaTek Inc., No. 2:06‑cv‑01225 (C.D. Cal.) — listed in the Stanford NPE Litigation Database as a case involving this patent. The MediaTek financial-statement notes state that Sanyo's counterclaims on US 6,594,213 and US 6,487,616 were transferred from E.D. Tex. to C.D. Cal., and that the parties jointly requested dismissal on 2006‑04‑25; the court formally dismissed the case on 2006‑04‑28.
• MediaTek Inc. v. Sanyo Electric Co. Ltd., No. 6:05‑cv‑00323 (E.D. Tex.) — MediaTek's affirmative suit; the same disclosure notes Sanyo counterclaimed on the '213 and '616 patents there before transfer. This case settled on the eve of trial (reported in a Law360 interview with plaintiff's counsel), and the docket shows status Closed.

Also noted on the Google Patents record is a litigation entry for C.D. Cal. 2:06‑cv‑01225 and E.D. Tex. 6:05‑cv‑00323. The patent's own expiration (2021‑01‑11, adjusted) further makes live 2026 appellate activity improbable.

5. Explicit uncertainties

• I could not independently confirm a 2026 CAFC docket entry for this patent; the absence of results is evidence of absence, but I cannot prove a negative from these searches alone. If you need a certified negative, that requires a direct PACER/CM‑ECF or CAFC docket query by case number, not a web search.
• The "current assignee" field is Google Patents' stated assumption and is not a legal conclusion; the recorded chain is Sanyo → Semiconductor Components Industries, LLC (2011), with a Deutsche Bank security interest (2016) released in 2023.
• I did not retrieve a USPTO PatentCenter/PAIR transaction-history or reexamination record; a check for any ex parte reexam, IPR, or certificate of correction on 6,487,616 would require that direct source.

I'll research litigation involving US patent 6487616, starting with the litigation records referenced in the patent page itself.

I found key sources. Let me dig deeper into each case and the Unified Patents litigation portal.

I have good leads. Let me verify the specifics of both cases and check for any other litigation or appeals.

Based on my research (npe.law.stanford.edu patent page, the Google Patents litigation entries, govinfo docket records for the E.D. Tex. case, and MediaTek's own SEC-equivalent financial-statement disclosures), here is what I found.

Litigation involving US Patent 6,487,616

The '616 patent (Controller for data recorder; assignee Sanyo Electric Co., Ltd.; granted 2002‑11‑26) appears in two federal district court matters, which are actually two stages of the same dispute. It was never asserted in the first Sanyo–MediaTek California suit (CV 05‑2580), which involved different patents.

1. MediaTek, Inc. v. Sanyo Electric Co., Ltd.; Sanyo North America, Inc.; Sanyo North American Corporation

• Plaintiff: MediaTek, Inc.
• Defendants: Sanyo Electric Co., Ltd.; Sanyo North America, Inc.; Sanyo North American Corporation
• Jurisdiction: U.S. District Court for the Eastern District of Texas (Tyler Division)
• Case No.: 6:05‑cv‑00323 (‑LED; Judge Leonard E. Davis)
• Filed: 2005‑08‑31
• The '616 patent's role: MediaTek filed the affirmative case asserting its own patents (U.S. 5,867,819 and 6,118,486; U.S. 5,751,356 added 2005‑10‑03). Sanyo counterclaimed on 2005‑11‑28, alleging MediaTek's optical‑storage products infringed Sanyo's U.S. 6,594,213 and U.S. 6,487,616.
• Outcome/status: Sanyo asked the court to consolidate MediaTek's three patents and Sanyo's two counterclaim patents into the Central District of California. The court refused, and on 2006‑02‑01 ordered that only Sanyo's two counterclaim patents (including the '616) be severed and transferred to C.D. Cal. The '616 therefore left this case. The remaining MediaTek‑asserted E.D. Tex. case (like the parallel California case) settled — per a first‑hand account by MediaTek's counsel, the parties settled "the night before we were to give opening argument."

2. Sanyo Electric Co., Ltd. v. MediaTek Inc.

• Plaintiff: Sanyo Electric Co., Ltd.
• Defendant: MediaTek Inc.
• Jurisdiction: U.S. District Court for the Central District of California
• Case No.: 2:06‑cv‑01225
• Filed: 2006 (the re‑docketed case created when the '616 and '213 counterclaims were transferred out of E.D. Tex. per the 2006‑02‑01 order)
• Outcome/status: Voluntarily dismissed. Per MediaTek's audited financial statements, at Sanyo's request the parties jointly moved to dismiss on 2006‑04‑25, and the court dismissed the case on 2006‑04‑28. This is the only case listed for the '616 in the Stanford NPE Litigation Database.

Notes and caveats

• Common thread: Both matters are the same Sanyo–MediaTek dispute; the '616 was asserted by Sanyo in each, once as a counterclaim (E.D. Tex.) and once as the transferred case (C.D. Cal.).
• Not the '616: The separately numbered Sanyo Electric Co. v. MediaTek Inc., C.D. Cal. No. CV 05‑2580 RSWL (JTLx) (filed 2005‑04‑07) asserted U.S. 5,499,252 and 5,818,801, with a MediaTek counterclaim on U.S. 6,003,151 — the '616 was not at issue there.
• Related appellate activity (not '616‑specific): In re MediaTek, Fed. Cir. No. 07‑M844 (filed 2007‑02‑07) was an appellate/mandamus matter arising from the E.D. Tex. litigation; I could not confirm that it addressed the '616 itself.
• Post‑2011: The '616 was assigned to Semiconductor Components Industries, LLC (2011, from Sanyo; now an onsemi entity, formerly Fairchild). The patent expired 2021‑01‑11. I found no litigation asserting the '616 by Semiconductor Components Industries/onsemi. I cannot rule out unindexed matters, but no such case surfaced.
• Confidence: The case numbers, parties, courts, and the 2005‑08‑31 / 2005‑11‑28 / 2006‑02‑01 / 2006‑04‑25 / 2006‑04‑28 dates are well grounded in the sources above. I was unable to independently confirm the precise filing date of 2:06‑cv‑01225 beyond "2006," so treat that specific date with lower confidence.

Sources: Google Patents litigation entries for US6487616 (E.D. Tex. 6:05‑cv‑00323; C.D. Cal. 2:06‑cv‑01225); Stanford NPE Litigation Database, patent 6487616; govinfo docket PDF for Case 6:05‑cv‑00323‑LED; MediaTek Inc. Q3/Q4 2006 financial-statement notes (Chinese original describing the Sanyo–MediaTek suits and the transfer/dismissal); Law360/Sheppard Mullin interview on MediaTek v. Sanyo (E.D. Tex.).

⚠️ One disclosure: my final confirmation searches (for docket details of 2:06‑cv‑01225, the exact settlement date, and any later onsemi/SCI assertions) were cut off by a tool‑use limit, so the outcome characterizations for the transferred California case rest on the MediaTek financial statements rather than a docket screenshot.

I'll start from the canonical structured block (no AIA proceedings on file) and then run due-diligence searches to see whether anything the ODP hasn't indexed turns up.

Let me run additional targeted checks — distinguishing this patent from the similarly-numbered Tessera patent, and covering CBM/PGR and any recent filings.

Two more checks: whether any proceeding exists on the sibling/family patents (which sometimes signals IPR interest in the family), and whether a defensive aggregator ever targeted this patent.

Proceedings overview

Total AIA trial proceedings on US 6,487,616: 0. The structured "PTAB proceedings on file" block — the canonical USPTO Open Data Portal list supplied in this prompt — returns no IPR, PGR, or CBM proceeding ever instituted or even filed against this patent, and my independent web searches surfaced nothing the ODP might have missed. There are therefore no "claims invalidated," "claims sustained," "settled," or "institution denied" buckets to fill: the patent is entirely untested at the PTAB. The bottom-line defensive posture is the opposite of the hardened-patent or dead-claims scenarios the template anticipates: every one of claims 1–11 stands as issued, and no petitioner has ever obtained — or even sought — a PTAB invalidity ruling. A defendant cannot point to a prior FWD; the defensive value has to come from the prior art and the prosecution history (including the family's EP/KR/TW counterparts) rather than from any Board outcome.

No AIA proceedings — what was checked

• Structured data (canonical): USPTO ODP API ingest reports no AIA trial proceedings for 6,487,616.
• Web due diligence: searches for 6,487,616/6487616 combined with "IPR," "PTAB," "inter partes review," "institution decision," and "CBM" returned no Board document naming this patent. The only real PTAB record turned up was for a similarly-numbered but unrelated patent, U.S. 6,852,616, in IPR2018-01264, Samsung Electronics v. Tessera Advanced Technologies (Tessera's Patent Owner Preliminary Response arguing discretionary denial under § 314(a)). That is a different patent, different owner, different technology (semiconductor chip-scale packaging, not optical-disc recording control). Do not conflate them — the digit-string similarity (…2,616 vs. …8,616) makes this a live mis-citation trap in any defendant-facing memo.
• Family cross-check: the same searches found no AIA trial on the sibling Sanyo patents in this family (e.g., US 6,594,213, US 7,106,675, US 7,149,165, US 6,992,957), which is consistent with the absence of any PTAB activity on 6,487,616 itself.
• Enforcement history: the only assertion of record is the 2005–2006 Sanyo/MediaTek dispute (E.D. Tex. 6:05-cv-00323, transferred to C.D. Cal. 2:06-cv-01225, dismissed 2006-04-28). It was litigated in the pre-AIA era — MediaTek could not have filed an IPR (available only from 2012-09-16), and no CBM/PGR followed after the AIA took effect. The litigation appears in Google Patents' "litigation" fields but not as any PTAB proceeding.

Strategic summary

Claim status. Because no AIA trial was ever filed, none of claims 1–11 of 6,487,616 is canceled or solely sustained by the Board. All four independent claims — claim 1 (single address memory + repeat-sync safeguard), claim 4 (dual address memories + two location detectors + determination circuit), claim 10 (method counterpart of claim 1), and claim 11 (method counterpart of claim 4) — plus dependents 2–3 and 5–9 remain UNTESTED and presumptively valid. There is no FWD to quote, and I will not invent one.

Estoppel landscape. With zero IPRs, there is no § 315(e)(2) estoppel running against anyone on this patent. No petitioner has been estopped, and equally no petitioner has been through the process to lock in a narrowed claim construction or a favorable institution record. For a defendant today, the entire prior-art field is open — including all 16 references cited on the face of the patent (e.g., US 5,436,875 and US 5,491,677 to Sony, US 5,745,545 to Matsushita, US 5,818,801 to Sanyo, and US 6,317,809 to IBM) and the family-cited art (e.g., JP H11-185375, KR 100288402, JP 3163064, JP 3505452). Any IPR would be a clean-slate filing, not a follow-on barred by earlier grounds.

Pattern signals. No repeat petitioner, no serial filings, no defensive aggregator. Note that Unified Patents appears in this record only as the source of the district-court litigation data (the portal.unifiedpatents.com/litigation/... links on the Google Patents page), not as a petitioner — do not read that as Unified having challenged the patent. The patent's adjusted expiration of 2021-01-11 means it is expired, which removes the practical incentive for anyone to mount a new AIA challenge now (an IPR cannot be filed against an expired patent to obtain cancellation of claims that no longer exclude anyone, and there is no injunctive exposure).

Recommended next steps

• No PTAB proceeding exists — say so plainly. There is nothing to link at PTAB E2E or CourtListener because no Board decision or CAFC opinion for 6,487,616 exists. The absence is itself the signal: this is an expired (2021-01-11) patent that was never worth an IPR to anyone, even during its 2005–2006 assertion window.
• If you are a defendant receiving a demand citing 6,487,616: (1) confirm the patent is expired and that the demand is therefore at most a past-damages theory; (2) note that no claim has ever been invalidated, so you cannot short-circuit the demand with an FWD — you must build your own invalidity case from the cited art; and (3) be careful not to cite the Tessera/Samsung 6,852,616 IPR2018-01264 materials as if they concerned this patent.
• If you (or a client) were contemplating an IPR: an AIA petition against an expired patent is generally not a viable strategy, and the statutory 1-year § 315(b) window and the "no live controversy" posture both counsel against it. Reexamination is similarly of limited value where no exclusionary right remains.
• Verification caveat. I could not run a direct PTAB E2E / PatentCenter query in this session, and my last two due-diligence searches were cut off by a tool-step limit. The ODP-derived "no proceedings" result is the authoritative statement here; if you need a certified negative for a filing or opinion, confirm via a direct PTAB E2E search on application 09/718,173 and patent 6,487,616. I found nothing suggesting the ODP list is incomplete, but I cannot prove a negative beyond the sources searched.

I'll research the assignment record for US 6,487,616, starting with the USPTO Assignment Center and cross-referencing the recorded chain.

Let me dig into the specific assignment records and correspondents.

I found the key reel/frame references. Let me search for the full assignment details and correspondents.

I've hit the tool-use limit for this session. Below is what I could reconstruct from the portable records already in hand (the Google Patents legal-events mirror, which reproduces the USPTO recording date and, in the corrective entries, the underlying reel/frame, plus the ON Semiconductor / MediaTek SEC-style filings). I was not able to open assignmentcenter.uspto.gov directly, so the correspondent-of-record fields are a gap I flag explicitly rather than fill in.

Assignment record — US 6,487,616 ("Controller for data recorder")

Sourcing note: The reel/frame anchors below are quoted from the corrective-assignment entries on the patent's own legal-events record, which cite the prior recordings they amend. I could not independently verify them against the Assignment Center's raw register in this session; treat the two reel/frames as high-confidence and any unlisted reel/frame as unknown.

Inventors

Pattern check: Not unusual. A single inventor assigning to his employer on the filing date is the ordinary course; there is no signal of pre‑filing inventor flight.

Original assignee

Sanyo Electric Co., Ltd. (Japan) — a large consumer‑electronics and semiconductor manufacturer. It shipped products embodying the claims: the patent is directed to a CD‑R drive controller, and Sanyo was an active CD‑R/CD‑RW drive and optical‑pickup maker in this era (the spec's "recording restart mode" is the operational behaviour of those drives). Primary line of business: consumer electronics, batteries, and semiconductors. Current status: acquired/absorbed — Sanyo's semiconductor business ("SANYO Semiconductor") was sold to ON Semiconductor via Semiconductor Components Industries, LLC, closing 2011‑01‑01 (Purchase Agreement with Amendment No. 1 dated 2010‑11‑30; 8‑K filed 2011‑01‑06); the remaining Sanyo Electric business was absorbed into Panasonic and the Sanyo brand wound down. Sanyo did not enter bankruptcy.

Assignment timeline

Execution dates are shown where a public instrument pins them; where the recorded date is all that is available, the entry is marked "(execution date not retrieved)". Google Patents' "reassignment" events report recording dates.

• 2000‑11‑21 (executed) / recorded 2000‑11‑21 — Reel not retrieved

  • Conveyance: Assignment of assignors' interest
  • Assignor: Hayashi, Koji
  • Assignee: Sanyo Electric Co., Ltd.
  • Correspondent: not retrieved — this is the field I most wanted and could not obtain; the Assignment Center record will show it.
  • Context: ordinary inventor‑to‑employer assignment, executed on the filing date.

• Executed ~2010‑11/2011‑01 (Sanyo Semiconductor Purchase Agreement; closing 2011‑01‑01) / recorded 2011‑07‑14 — Reel 026594 / Frame 0385

  • Conveyance: Assignment (portfolio transfer; not a standalone patent sale)
  • Assignor: Sanyo Electric Co., Ltd.
  • Assignee: Semiconductor Components Industries, LLC (the ON Semiconductor acquisition vehicle, a Delaware LLC)
  • Correspondent: not retrieved
  • Context: corporate divestiture — the '616 rode along inside the SANYO Semiconductor business sale, not a fire‑sale or an IP‑only carve‑out. (Reel/frame anchor: the 2014 corrective entry amends the recording placed at 026594/0385.)

• (execution date not retrieved) / recorded 2014‑05‑06 — Reel not retrieved (amendatory)

  • Conveyance: Corrective Assignment — "TO CORRECT THE INCORRECT #12/577882 PREVIOUSLY RECORDED ON REEL 026594 FRAME 0385"
  • Assignor: Sanyo Electric Co., Ltd.
  • Assignee: Semiconductor Components Industries, LLC
  • Correspondent: not retrieved
  • Context: administrative cleanup of the 2011 portfolio recording (a wrong application number listed in the schedule).

• Executed 2016‑04‑15 (Guarantee and Collateral Agreement date) / recorded 2016‑04‑15 — Reel 038620 / Frame 0087

  • Conveyance: Security Interest (grant of security interest in patent collateral)
  • Assignor: Semiconductor Components Industries, LLC
  • Assignee: Deutsche Bank AG, New York Branch
  • Correspondent: not retrieved (the collateral-agent work products are New York‑law governed, per the Guarantee and Collateral Agreement filed with the SEC)
  • Context: securitization — blanket lien over the patent estate to secure a credit facility; no change in beneficial ownership.

• recorded 2016‑08‑25 — amends Reel 038620 / Frame 0087

  • Conveyance: Corrective Security Interest — "TO CORRECT THE INCORRECT PATENT NUMBER 5859768 AND TO RECITE COLLATERAL AGENT ROLE OF RECEIVING PARTY IN THE SECURITY INTEREST"
  • Assignor: Semiconductor Components Industries, LLC
  • Assignee: Deutsche Bank AG New York Branch, as Collateral Agent
  • Correspondent: not retrieved
  • Context: administrative cleanup (typo'd patent number; corrects the capacity in which Deutsche Bank takes).

• recorded 2023‑06‑22 — releases Reel 038620 / Frame 0087

  • Conveyance: Release of Security Interest ("RELEASE OF SECURITY INTEREST IN PATENTS RECORDED AT REEL 038620, FRAME 0087")
  • Assignor: Deutsche Bank AG New York Branch, as Collateral Agent
  • Assignee: Fairchild Semiconductor Corporation and Semiconductor Components Industries, LLC
  • Correspondent: not retrieved
  • Context: release of collateral on termination of the secured facility (note the patent had already expired 2021‑01‑11 by this date).

Net ownership outcome: the chain is Sanyo → SCI/onsemi, with Deutsche Bank appearing only as collateral agent (2016) and releasing in 2023. There is no IP‑holding LLC, no licensing vehicle, and no NPE anywhere in the recorded chain.

Timeline diagram

timeline
    title Ownership of US 6487616
    1999 : Priority JP 11 331417 filed
    2000 : Filed and assigned to Sanyo Electric
    2002 : Patent issued 26 Nov
    2005 : Sanyo asserts patent against MediaTek
    2006 : Assertion transferred then dismissed
    2011 : Sanyo Semiconductor sold to onsemi
         : Assigned to SCI LLC reel 026594 frame 0385
    2014 : Corrective assignment recorded
    2016 : Security interest to Deutsche Bank reel 038620 frame 0087
    2023 : Security interest released

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. The only operating→operating transfer is Sanyo → Semiconductor Components Industries, LLC at Reel 026594/0385 (recorded 2011‑07‑14). SCI LLC is the Delaware acquisition subsidiary of ON Semiconductor (now onsemi) and ships semiconductor products; its principal business is manufacturing, not licensing. No "IP / Holdings / Ventures / Licensing" vehicle appears.

2. Known asserter in the chain — NOT PRESENT. No assignee matches Acacia, Marathon, IV, IPNav, Wi‑LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, or an Erich Spangenberg entity. The only "asserter" framing on this patent comes from the Stanford NPE database, which classifies the 2005–06 plaintiff Sanyo as a "Product company" (category 8), not an NPE — consistent with the operating‑company picture here.

3. Repeat correspondent across the chain — UNCLEAR (data gap). I could not retrieve the correspondent of record for any of the six recordings, because I could not reach the Assignment Center register in this session. A single firm handling both the 2011 SCI recording and the 2016 Deutsche Bank security recording would be expected and benign (it is the same corporate patent department's outside counsel), so even a repeat correspondent here would not be an NPE tell. No finding either way.

4. Cascading transfers (<24 months through chained LLCs) — NOT PRESENT. Substantive ownership changed once in 26 years (Sanyo → SCI, 2011). The 2014 corrective and 2016 corrective are amendatory recordings over existing ones; the 2016 security interest is a lien, not a transfer; the 2023 release unwinds it.

5. Pre-litigation transfer — NOT PRESENT (inverse). The assertion preceded the transfer by ~6 years: Sanyo counterclaimed on the '616 in E.D. Tex. 6:05‑cv‑00323 (counterclaim 2005‑11‑28), which was severed/transferred to C.D. Cal. 2:06‑cv‑01225 (2006‑02‑01) and dismissed (2006‑04‑28) — all before the 2011 SCI assignment. The transfer was not arranged to enable this patent's assertion.

6. Bankruptcy fire‑sale — NOT PRESENT. Sanyo never filed Chapter 7/11. The 2011 divestiture was a going‑concern sale of the SANYO Semiconductor business to ON Semiconductor, not a distressed IP auction.

7. Privateering — NOT PRESENT. No evidence (SEC filings, Patent Progress/EFF coverage, or the recorded chain) that Sanyo or SCI transferred the '616 to an NPE to assert on the transferor's behalf.

8. Defensive aggregator — NOT PRESENT. The chain terminates at Semiconductor Components Industries, LLC / Fairchild Semiconductor Corporation (operating onsemi entities), not at RPX, AST, LOT, Unified, or OIN. The patent has not been neutralized by a defensive aggregator; it simply expired (2021‑01‑11) in operating‑company hands.

Recurring-principal check (ancillary): the same Sanyo→MediaTek dispute shows up in the ON Semiconductor and MediaTek filings (MediaTek Q3/Q4 2006 financial‑statement notes describe the transfer/dismissal of the '616 counterclaim). No common principal links any assignee to an NPE.

Verdict

Operating-company assertion.

The record shows a single, clean Sanyo→Semiconductor Components Industries, LLC transfer at Reel 026594 / Frame 0385 (recorded 2011‑07‑14) inside the SANYO Semiconductor divestiture, with the only other substantive filings being a Deutsche Bank security interest at Reel 038620 / Frame 0087 (2016‑04‑15) and its 2023‑06‑22 release — i.e., liens, not ownership changes. The one assertion episode (Sanyo v. MediaTek, E.D. Tex. 6:05‑cv‑00323 → C.D. Cal. 2:06‑cv‑01225, 2005–06) was brought by an operating competitor against an actual competitor, and Stanford's database classifies Sanyo as a "Product company," not an NPE. The chain terminates at operating onsemi/Fairchild entities, not at a defensive aggregator, so the patent was neither privatized nor neutralized — it simply ran to its 2021 expiration.

Two caveats I want to be explicit about: (a) every reel/frame anchor above is derived from the patent's own corrective-assignment entries, not from a fresh Assignment Center query — verify at https://assignmentcenter.uspto.gov/ (patent‑number search) before relying on them; and (b) the correspondent-of-record field — the single most useful tell for this task — is genuinely missing from my reconstruction, not "not applicable." If you can supply screenshots of the six recording entries, I can complete signal #3 (repeat correspondent) properly.

I'll verify the primary citations and enrich descriptions before mapping them to claims. Let me search for the key references.

Let me pull details on the remaining significant references.

Prior-Art Analysis — US 6,487,616 B1 ("Controller for data recorder")

0. Record confirmation (literal, no auto-correction)

The number you gave resolves to exactly one record, and I did not return anything for neighboring numbers (e.g., 6,487,615 / 6,487,617 are not discussed):

Statutory frame. Because the application was filed 2000-11-21, pre-AIA 35 U.S.C. §102 (subsections (a), (b), (e), and (g)) governs. Critical dates for the analysis below:

• Effective invention/priority date: 1999-11-22 (JP 11-331417); second JP priority 2000-10-20.
• §102(b) one-year bar relative to US filing: ~1999-11-21.
• Anything published before 1999-11-22 is solidly pre-priority art; documents published between 1999-11-22 and 2000-11-21 are §102(a) art at best unless they are US patents/US applications with an earlier effective filing date (§102(e)), or JWIPO/PCT designating the US (§102(e) via pre-AIA 102(e)(2)).

Sources / method note. I worked from the authoritative full text supplied to me (Google Patents mirror of the USPTO front page, https://patents.google.com/patent/US6487616/en) plus Google Patents / EPO publication-server / WIPO pages for the individual references. I could not reach USPTO PatentCenter/PAIR in this session, so the "References Cited" list below is taken verbatim from the printed patent's front page as mirrored by Google Patents. No reexamination, IPR, or certificate-of-correction record was checked (that requires a direct USPTO source).

1. What the claims actually require (the yardstick for §102)

Before mapping references, the four independent claims reduce to these elements:

E4 is the architectural heart of the disclosure (spec: "the recording of data is not restarted and data reproduction in the recording restart mode is repeated… Thus the effect of an external disturbance is avoided"). This matters because, as shown below, no cited reference discloses E4, which is why the claims issued over the art.

2. Table A — The 16 references cited on the face of US 6,487,616

Dates are given as printed (first date = filing/priority as listed; second = publication/issue). Descriptions are drawn from the references' own text where I could retrieve it.

(The printed list is captioned "Citations (16)"; the table above transcribes all rows as they appear, including the duplicate-seeming entries.)

3. Table B — The 10 "Family Cites Families" references

These appear on the record as citations that appear in other members of the same patent family (the JP/EP/TW siblings), i.e., art considered during the family prosecutions rather than necessarily the US '616. They are still candidate prior art against the '616 claims and several are directly on point.

(All rows are transcribed from the "Family Cites Families (10)" table on the record.)

4. What is not prior art (avoid the classic error)

The 22 documents in the "Cited By (22)" table are later documents that cite the '616 — i.e., forward citations. They are not §102 art against the '616. Note in particular that several of them are Sanyo siblings with the same 1999-11-22 priority (US 6,594,213 "Data recorder"; US 7,106,675 "Data recorder"; US 7,167,427 "Data recorder"; and the two Hayashi publications US 2001/0006496 and US 2001/0006499), and the continuations US 2004/0184374 A1 and US 2005/0094525 A1. These are same-family, same-day filings — not citable prior art against the '616 (they share its priority).

Also not prior art: JP 3653459 B2 ("Control device," granted from JP 2000-321249) — that is the JP family member of the '616 itself.

5. Claim-by-claim §102 mapping (element analysis)

5.1 US 5,436,875 (Sony) — the closest reference

The retrieved text states: "the data previously stored in the memory is read until the optical pick up returns to the address where the track jump occurred, even when the track jump occurs by vibration, etc., so that the optical disc device can effectively avoid the discontinuity of data" and, for recording, "if the trouble such as track jump occurs, the optical disc device performs re-write processing… in order that the discontinuity of data does not occur."

→ Potentially anticipates a controller of the claim-1 genus and would be the primary §102/§103 reference against claim 1 and claim 10, but for the E4 retry clause, which is absent. It cannot, on this record, fully anticipate claims 1/10 (all-elements rule), and it does not disclose the dual address memories + two location-detection circuits + determination circuit of claims 4/11 (& 2–3, 5–6). Best target: claim 7 (sector-unit address correspondence) and claims 8–9 (reading back a predetermined number of sectors), if its seek-back distance is expressed in sectors.

5.2 US 5,491,677 (Sony)

→ Strong §103 art against the buffer-amount-based interruption concepts implicit in claims 1/4/10/11; not an anticipation of any independent claim.

5.3 US 6,317,809 (IBM)

→ §102(e)/§102(b). Good art for the "buffer underrun determination circuit" element and for the background/problem; teaches away from the claimed solution, so it is a poor anticipation reference for claims 1/4/10/11.

5.4 US 5,745,545 (Matsushita)

Discloses the buffer between input and recorder/reproducer and a controller that computes buffer occupancy to avoid overflow. → §102(b). Useful against the buffer-occupancy/buffer-determination elements; discloses no address bookmark, no re-read-and-synchronize, no match-triggered restart. Not an anticipation of the independent claims.

5.5 JP H11-185375 (Toshiba) — likely the most on-point §102(b) foreign publication

Published 1999-07-09, ~4.5 months before the '616 priority. Titled "CD-R writing control system, computer system, and information recording/reproducing device." On its face this is the closest pre-priority document to a "stop the CD-R write and resume it" control scheme. I could not retrieve the full Japanese text in this session, so I flag it as the highest-priority reference to obtain and read in full. If it discloses storing a disc address and/or buffer address at the interruption and resuming on an address match, it becomes the leading §102(b) candidate against claims 1/4/10/11; if it lacks E4, it is a leading §103 reference.

5.6 Shock/vibration cluster: US 5,818,801; US 5,910,935; US 6,115,337; US 5,333,138

All four address interruption of an optical-drive operation by mechanical disturbance and the recovery of continuity. They map well onto the problem statement and onto the '616's embodiment (4) (displacement of the optical head), and to E2 (re-synchronization). None discloses E1's dual address bookmarks or E3/E4's match-triggered restart-and-retry. → §102(b) (5,818,801; 5,910,935; 5,333,138) and §102(e) (6,115,337 — 1994 effective filing). Useful as §103 secondary references, not anticipatory.

5.7 General optical-recording art (A3–A5, A7–A10, B1–B2, B4, B6)

US 5,212,678, US 5,289,440, US 5,220,545, JP H02-223066, JP H03-228266, JP H10-049990, JP H10-063433, JP S61-139984, JP S63-61423, US 6,438,083, JP H09-035270 — background as to constant-speed recording, jump-back, and continuous recording. Individually they antedate the '616 but reach none of E1–E4 as a complete set.

5.8 Anomalies / unverified

• JP 3228266 B2 — listed as a pallet-transfer device; inconsistent with this art and with its dates. Treated as a source-record anomaly (see Table A, row A17).
• KR 100288402 B1 and JP 3163064 B2 / JP 3505452 B2 — publication dates fall after the '616 priority date; they are §102(e)-relevant only via a US counterpart with an earlier effective filing date. Not verified.

6. Ranking — most relevant prior art

Tier 1 — closest to the independent claims (best §102/§103 bases):

1. US 5,436,875 (Sony; §102(b)) — buffer memory + interrupt on trouble + replay-until-the-pickup-returns-to-the-address. Hits E1–E3; misses E4.
2. US 5,491,677 (Sony; §102(b)) — buffer fill level drives intermittent recording; read/write pointer tracking; servo-gain adjustment during the interruption.

Tier 2 — same problem, different solution (underrun-specific):
3. JP H11-185375 (Toshiba; §102(b), published 1999-07-09) — CD-R write control for under-run; read this one in the original.
4. US 6,317,809 (IBM; §102(e), EP counterpart §102(b)) — under-run detection + padding sectors; teaches away from stop/restart.
5. US 5,745,545 (Matsushita; §102(b)) — buffer occupancy calculation to prevent data loss.

Tier 3 — interruption/resume under mechanical disturbance (embodiment (4) support):
6. US 6,115,337 / US 5,910,935 (Deutsche Thomson-Brandt; §102(e)/§102(b)) — vibration-resistant playback with improved synchronization.
7. US 5,818,801 (Sanyo; §102(b)) — shockproof reproduction.
8. US 5,333,138 (Ministor Peripherals; §102(b)) — preventing write-data corruption from mechanical shock.
9. US 6,438,083 (Philips; §102(e)) — continuous-stream recording into track gaps.

Tier 4 — background/field art (address & continuous recording): US 5,220,545 (jump-back), US 5,212,678 and US 5,289,440 (Philips), JP H03-228266, JP H10-049990, JP H10-063433, JP H02-223066, JP S61-139984, JP S63-61423, JP H09-035270, US 6,055,216, JP H04-225433 (cancelable synchronization — see B3).

Closed-out (not prior art): all 22 "Cited By" documents (including the Sanyo siblings US 6,594,213, US 7,106,675, US 7,167,427, and the Hayashi US 2001/0006496, US 2001/0006499, US 2004/0184374, US 2005/0094525), and JP 3653459 B2 (the '616's own JP family member).

7. Bottom line for validity

• Anticipation (§102) of the independent claims 1, 4, 10, 11 by any single cited reference: unlikely on this record. No cited document discloses E4 — the requirement that restart be suspended and the read/synchronize pass repeated when a match occurs before synchronization completes — and claim 4's specific trio of (two address memories + two location-detection circuits + determination circuit) is likewise not shown in any one reference.
• The strongest pre-priority art is US 5,436,875 and US 5,491,677 (both Sony), with JP H11-185375 (Toshiba) as the sleeper. Together they disclose: buffering to bridge data-rate mismatch, interruption of recording when the buffer runs low, retention of the disc position where the interruption/trouble occurred, and re-playing buffered data until the head returns to that position. A §103 combination of US 5,436,875 + US 5,491,677 (optionally + US 6,317,809 for the under-run detector, or + US 5,818,801/US 6,115,337 for the disturbance-triggered interruption) would be the natural attack on claims 1–3 and 10.
• The narrow dependent claims are the more vulnerable ones. Claim 7 (sector-unit address correspondence) and claims 8–9 (read back a "predetermined number of sectors" sufficient to complete synchronization) are the most likely to be met by the Sony references once their seek-back granularity is confirmed in the original figures.
• The commercially significant real-world prior art is BURN-Proof (Sanyo, mid-2000) — but note this postdates the '616 priority (1999-11-22) and is the patentee's own commercial embodiment of this very family, so it is not §102 art here.

8. Explicit uncertainties

1. I could not retrieve full texts for US 5,212,678, US 5,220,545, US 5,289,440, US 5,910,935, US 6,115,337, US 6,438,083, US 5,333,138, or for the Japanese-language JP H02-223066, JP H03-228266, JP H10-049990, JP H10-063433, JP H09-035270, JP H11-185375, or KR 100288402 B1; descriptions for those rest on titles/bibliographic data, and the §102 mapping for them is consequently provisional.
2. The §102 subsection and critical date for a given reference depends on its effective date, which for US documents means the earliest US filing (§102(e)) and for foreign documents means the actual publication date of the printed publication. I flagged every case where the listed date creates a timing problem (KR 100288402; JP 3163064; JP 3505452).
3. JP 3228266 B2 is internally inconsistent with this technology and citation list; I have flagged it rather than force a mapping.
4. I did not verify any reexamination/IPR/certificate-of-correction history, nor did I confirm whether US counterparts exist for KR 100288402, JP 3163064, or JP 3505452 (which would change their §102 status from ineligible to §102(e) art).
5. The whole analysis uses the Google Patents mirror of the USPTO front page; no direct PatentCenter/PAIR transaction-history pull was possible in this session.

I'll ground this obviousness analysis in the actual prior-art references listed on the patent page. Let me pull details on the highest-value references first.

Now the shock-detection, Samsung underrun, and Sanyo Japanese-language references — these are the likely primary references.

I have enough grounded material to build the analysis. One note before the memo: my verification searches were cut off partway (the KR, JP, IBM and NEC queries returned no results before the tool limit), so those references are characterized from their face-of-record titles/metadata and flagged accordingly.

Obviousness Analysis — US 6,487,616 B1 under 35 U.S.C. § 103

Scope note: The '616 application was filed 2000‑11‑21, so pre‑AIA § 102/§ 103 govern (Graham factors + KSR). Because the patent expired 2021‑01‑11, this analysis is retrospective/defensive — relevant to past damages and to the validity of the family, not to any live injunction.

1. The invention, stripped to its gist

The '616 solves buffer underrun in a CD‑R drive by a "write‑interrupt‑then‑resume‑at‑the‑same‑sector" scheme (spec's "recording restart mode"). The technical content is in three moves, all recited in claim 1:

Claim 4 is the same invention with two bookmarks and two comparators feeding an arbiter ("determination circuit"); claims 5–6 add simultaneous‑match and re‑sync‑on‑mismatch; claims 7–9 add sector granularity and the m‑sector back‑off sized to servo+sync settling time (m≈10–30 at 4×–8×). Claims 10–11 are method counterparts.

PHOSITA: an engineer with a B.S. in EE/CS (or equivalent) and ~2–3 years in optical‑drive controller/firmware design, familiar with the CD‑R Orange Book, CLV/ATIP addressing, EFM/CIRC encoding, and DRAM buffer management. This is the art level used by the examiner in this family's EP counterpart.

Effective date: 1999‑11‑22 (JP 11‑331417). Note a wrinkle carried over from the earlier section: the family also lists JP 2000‑321249 (filed 2000‑10‑20, granted as JP3653459B2, "Control device"). Subject matter first disclosed only in that later document would be dated 2000‑10‑20, which would open a ~11‑month additional window of prior art. None of the combinations below depends on that distinction, since every reference relied on predates both dates.

2. What the prior art actually discloses

The face‑of‑patent citations are unusually on‑point. Verified from primary text:

US 5,436,875 (Sony, 1995‑07‑25) — Recording/reproducing apparatus having buffer memory for preventing discontinuity… (https://patents.google.com/patent/[US5436875A](/patent/US5436875A)/en; https://www.freepatentsonline.com/[5436875](/patent/5436875).html). Teaches storing reproduced data in a memory at a higher rate than it is read out; on a track jump caused by vibration, the pickup is moved back to the address where the track jump occurred, the data is re‑read and re‑stored, and "the data previously stored in the memory is read until the optical pick up returns to the address where the track jump occurred," so discontinuity is avoided. In recording, on trouble such as a track jump the device "performs re‑write processing, and effectively avoids… discontinuity of data." This is limitations A, B and (part of) C, for both the "medium address as bookmark" and the "buffer address as bookmark" concepts, in an optical disc recorder.

US 5,333,138 (Ministor Peripherals, 1994‑07‑26) — …preventing data corruption in disk drives from mechanical shock during write operations (https://patents.google.com/patent/[US5333138A](/patent/US5333138A)/RED_FLAGS_Oct.2007_.pdf; EP counterpart EP 0 560 600 A3). Teaches: sense a shock above threshold; "storing information identifying the data being written at the onset of the sensed shock"; interrupt the write current; reposition the data head to the original track; and rewrite the data which was interrupted; and "the controller maintains the original data in the data buffer until it is written successfully… the drive will quit transferring new data into the data buffer until the data already in the data buffer is successfully written." That is limitation A (both bookmarks — the medium block and the buffer block) plus the interrupt/resume‑at‑the‑same‑point idea, triggered by external mechanical disturbance rather than buffer depletion.

US 5,491,677 (Sony, 1996‑02‑13) — Recording and/or reproducing apparatus and method of controlling same (https://patents.google.com/patent/US5491677). Teaches temporarily interrupting recording based on the amount of data stored in the memory, and autonomously adjusting servo gain while recording is interrupted — i.e., the trigger for interruption is buffer fullness and the interval is used to re‑establish servo conditions. This is the functional equivalent of the '616's buffer‑underrun determination circuit 20 and m‑sector servo re‑lock period.

US 5,745,545 (Matsushita, 1998‑04‑28) — Digital recording and reproducing apparatus and method which prevents or manages a data loss (https://patents.google.com/patent/[US5745445A](/patent/US5745445A)/en). Teaches a buffer absorbing the difference between input and write transfer rates, and computing the buffer occupancy at each access‑mode change (a running address/pointer bookkeeping of the buffer) to keep the buffer within capacity. This supplies the buffer‑address tracking and the predictive "will underrun occur" determination.

US 6,437,883 (Philips / Koninklijke Philips) — Apparatus for recording a continuous information stream in available gaps… (https://patents.google.com/patent/[US6438083](/patent/US6438083)). Teaches an input buffer decoupling the incoming stream from the write head; detecting when the scan nears the end of a recordable gap; interrupting the information supply; searching for the next recordable portion; and resuming supply from the input buffer so the stream stays continuous. Caveat: the US grant is 2002‑08‑20, after the '616 priority, so it is usable only via its earlier‑published family member (the EP publication I retrieved, EP 0 543 446 A1, published 1993 — I matched it to the family by its shared background citation to EP‑A‑0 275 972; I could not open the EP bibliographic face to confirm the link with certainty).

US 6,317,809 (IBM, granted 2001‑11‑13) — Optical disk under‑run protection using formatted padding sectors. Usable only under pre‑AIA § 102(e) (US filing 1997‑02‑19). Establishes that optical‑disk under‑run protection was a known, named problem with known solutions — padding/link blocks — motivating the search for better ones.

US 5,910,935 and US 6,115,337 (Deutsche Thomson‑Brandt, 1999‑06‑08 / 2000‑09‑05) — Vibration‑resistant playback device having improved synchronization / Vibration‑resistant playback device. These are the two references the examiner actually cited on the '616 face (marked "*"), and they address precisely the "re‑synchronize the read channel after a disturbance" half of the problem.

US 5,220,545 (NEC, 1993‑06‑15) — Disk controller including format control unit instructing directly jump back operation. A disk controller that commands a jump‑back to a prior address — i.e., a controller that decides to return the head to an earlier location.

KR 100288402 B1 (Samsung, priority 1997‑12‑29) — Method for confronting buffer underrun of optical disc drive. ⚠️ Not verified beyond the title/metadata (search cut off). By title it is the single closest reference on the problem statement; if its disclosure includes any address‑bookmark‑and‑resume teaching it becomes a strong primary reference.

JP 3 503 452 B2 / JP 3 163 064 B2 (Sanyo, "Disk recording device") — ⚠️ same‑assignee art, not verified. Note a date problem: JP 3 505 452 B2 is listed with a 1999‑11‑05 date — 17 days before the '616 priority — and a 1999 filing would not have laid open before 1999‑11‑22, so it is probably not prior art; JP 3 163 064 B2 (1998‑07‑22) likely laid open ~2000‑02, also after. These should be treated as § 102/§ 103 art only after confirming actual publication dates.

Plainly irrelevant or background: JP 3228266 B2 (NKK — pallet transfer direction changing device), JP H04‑225433 (Fujitsu — semaphore synchronization with canceling function), JP S61‑139984 / S63‑61423 (early optical disk devices), US 5,289,440 and US 5,212,678 (Philips scanning/bit‑rate).

3. Element‑by‑element mapping for claim 1

4. The combinations, with motivation

Combination I (primary): Sony '875 + Ministor '138, optionally + Sony '677 and Matsushita '545.
Sony '875 supplies the optical‑disc, buffer‑memory, bookmark‑and‑resume architecture (limitations A–C, broadly). Ministor '138 supplies the explicit "store the identity of the data being written at the moment of interruption → reposition to the original track → rewrite" sequence, i.e., the address‑bookmark concept stated in the exact terms of claim 1, plus the discipline of holding the buffer contents until the write succeeds.
Motivation: both are in the same field (data‑storage write interruption/recovery) and address the same failure mode — an unexpected interruption mid‑sector producing an unrecoverable discontinuity. A PHOSITA facing the CD‑R problem in 1999 would treat the hard‑disk art's shock‑interrupt/rewrite routine as a directly applicable template, because a CD‑R is write‑once: unlike a CD‑RW or MO disc, a bad sector cannot be overwritten, so the incentive to stop before the discontinuity and resume at it is even stronger than on a magnetic disk. Sony '677 and Matsushita '545 add the trigger (buffer occupancy) so the same hardware covers both interruption causes — buffer depletion and mechanical shock — which the '616 itself contemplates (spec, alternative (4): "the present invention may be applied when data writing is interrupted due to the displacement of the optical head 4… a mechanism for determining the displacement… may be used in lieu of the buffer underrun determination circuit 20").

Combination II: Sony '875 + Sony '677 + Matsushita '545 + Samsung KR '402.
If KR '402 discloses any address‑bookmark/resume method, it is art of the same character as the '616 and combines with the Sony/Matsushita buffer‑control teachings on the same rationale (same problem, same field, predictable combination under KSR). ⚠️ This combination is contingent on verifying KR '402's disclosure.

Combination III: Philips US 6,437,883 (via EP 0 543 446 A1) + Ministor '138 + Sony '677.
The Philips disclosure supplies buffer‑decoupled interrupt/resume of a continuous stream at a new location; Ministor supplies storing the identity of the interrupted data and returning to the original position; Sony '677 supplies the buffer‑fullness trigger. Motivation: the Philips gap‑recording scheme minimizes wasted capacity but leaves a discontinuity at each interruption; combining it with a return‑to‑the‑same‑address recovery (Ministor) yields the '616's continuous‑sector result — an obvious improvement with a known benefit ("sectors of data are recorded without any interruptions when restarting recording," '616 spec).

Combination IV (for claims 4–9 and 11): the above + NEC US 5,220,545 + DT Brandt US 5,910,935/6,115,337.
Claim 4's additions map as: NEC '545 → a controller that commands a jump‑back (the "first location detection circuit" function in controller form); DT Brandt '935/'337 → re‑synchronization of the read channel after a vibration disturbance (the "synchronizing circuit"); Ministor/'875 → the two address bookmarks; Matsushita '545 → buffer pointer bookkeeping (the "second location detection circuit" input). The "determination circuit" is then a logic arbiter over already‑available signals — the classic KSR "arrangement of old elements yielding no more than predictable results."

5. Where the claims are strongest (and how an examiner/petitioner would still attack)

The only limitation with no clean anticipatory or explicit obviousness teaching in the cited art is the claim 1 tail:

"…the restart circuit suspends the restarting and the synchronizing circuit repeats the reading and synchronizing of data when one of the address of the data read from the recording medium and the address of the data read from the buffer memory matches the address stored in the address memory before the synchronizing circuit completes the data synchronization."

and its claim 4–6 cousins: two bookmarks, two detectors, and restart only if their outputs are simultaneous, else re‑sync (retry determination circuit 44).

Petitioner's best case (§ 103):

1. Known problem, known fix, predictable combination. US 6,317,809 proves optical under‑run was a named problem; the '616 spec itself concedes the alternatives (packet writing, link blocks). Given Sony/Ministor's return‑to‑interruption‑point recovery, guarding the restart with a verification step is the routine way an engineer avoids writing garbage onto write‑once media — a "predictable variation" (KSR).
2. Second bookmark is a design convenience, not an invention. Ministor already requires both the medium block identity and the buffer contents to be retained; storing the buffer address/comparator output is the natural way to implement the same check in an SDRAM ring buffer (spec: "the buffer memory 13 is a ring buffer"). Adding a second comparator is KSR's "duplication of parts."
3. Simultaneity check is a data‑integrity double‑check. A PHOSITA who knows that servo/decoder lock can be lost by shock (DT Brandt '935) and that the two streams can slip is motivated to require agreement of two independent indicators before committing an irreversible write.
4. m‑sector back‑off (claims 8–9) is optimization. Setting the re‑read look‑back to cover servo settle + sync time (m≈10–30) is the kind of parameter tuning that is prima facie obvious absent evidence the range is critical (In re Aller); note, however, that the spec recites an effect ("increase m as the recording speed increases"), which the patentee would argue is a recognized criticality.

Patentee's best case (nonobviousness):

• None of the cited art discloses two independent bookmarks compared against each other, and none discloses aborting an in‑progress restart when the indicators disagree. Sony '875, Ministor '138, Philips '883 and NEC '545 each restart on a single criterion; they contain no teaching that a single criterion is insufficient.
• The claimed retry loop is a specific remedy for a specific undetected fault — the spec says: "if an external disturbance occurs… the elements 2‑22 of the CD‑R drive 1 may function erroneously such that the two restart signals are not synchronously activated." Solving an intermittent, previously‑unappreciated failure mode is not "predictable variation."
• No teaching‑away exists in the relevant art. The '616 spec disparages packet writing and link blocks ("the link blocks decrease the recording capacity"), but that criticism is directed at padding/partition schemes (IBM '809, Philips '883), not at same‑point resume art (Sony '875, Ministor '138). So the patentee cannot buy nonobviousness by arguing teaching‑away from the secondary references; conversely, a petitioner cannot use the spec's disparagement of packet writing as a motivation for the Sony/Ministor route.
• Secondary considerations (Graham). The '616 is a foundational member of Sanyo's buffer‑underrun family (US 6,594,213, US 7,106,675, US 7,167,427) and drew a 22‑document forward‑citation cluster plus a 2005–2006 Sanyo–MediaTek assertion (E.D. Tex. 6:05‑cv‑00323, transferred as C.D. Cal. 2:06‑cv‑01225 and voluntarily dismissed 2006‑04‑28 — see the earlier litigation section). That is evidence of industry recognition and of copying/design‑around pressure. But it is not probative weight: neither case produced a validity adjudication, and no secondary‑considerations record (nexus, commercial success figures) exists in the material available.

Net assessment: Claims 1 and 10 are the mostly likely to fall to a Sony '875 + Ministor '138 + Sony '677 combination; claims 4, 5, 6, 11 are the most defensible, because "two bookmarks, two detectors, restart only on simultaneity, else retry" is not expressly taught by any cited reference and rests on an articulated failure mode. Claims 2, 3, 7, 8 sit in between (2–3 are near‑conventional circuit duplication; 7–8 are routine sector‑granularity choices). Claim 9 is the weakest nonobviousness candidate of the dependent set.

6. Explicit uncertainties

1. Verification was truncated. My searches on KR 100288402 B1, JP 3 505 452 B2, JP 3 163 064 B2, JP H09‑035270 A, JP H11‑185375 A and US 6,317,809 (IBM) returned no content before the tool limit; those references are characterized from the face‑of‑patent metadata only. Combination II is contingent.
2. I did not retrieve the '616's file history (USPTO PatentCenter/PAIR). The examiner cited DT Brandt US 5,910,935 and IBM US 6,317,809 (the two asterisked "* cited by examiner" entries), which strongly suggests the vibration‑resync and under‑run arts were considered and the claims allowed over them — but whether the allowance turned on the retry/simultaneity limitation, or on some other feature, I cannot confirm without the office actions.
3. Prior‑art dates need per‑reference checking. Several "Citations" entries carry US grant dates after the 1999‑11‑22 priority (US 6,317,809 → 2001; US 6,437,883 → 2002; US 6,055,216 → 2000; US 6,115,337 → 2000). These qualify, if at all, only under pre‑AIA § 102(e) (earlier US filing) or through earlier‑published foreign family members. I flagged the Philips family link (EP 0 543 446 A1) as inferred from the shared background citation to EP‑A‑0 275 972 rather than confirmed from the EP face page.
4. No contradiction found between this analysis and the previously generated sections. One refinement, not a contradiction: the earlier summary lists both JP 11‑331417 (1999‑11‑22) and JP 2000‑321249 (2000‑10‑20) as priority documents; the second one post‑dates the first by 11 months and is titled "Control device" (granted JP3653459B2), so the '616's effective date could be later for subject matter appearing only in that filing — a point worth resolving from the file history if the '616 is ever re‑examined.