Type: WKAP Radar Feed

WKAP Radar Feed - 2026 - 09 - 18

Market date: 2026-09-18
Canonical URL: https://wkap.ai/radar/wkap-radar-feed-2026-09-18.html

WKAP Radar Feed

*September 18, 2026*

*AI Rack Proof Clocks: Shipping, Availability and Sampling*

*3 Thesis Objects: TSEM, MTSI, LSCC*

Preheader:

Make your AI track X alpha.

WKAP Feed compresses earnings inflections, KOL flow, supply-chain
bottlenecks, and short-duration attention into daily AI-ready context.

Set up your AI to read this Feed.

Ask:

“What did I miss?”

HUMAN USER — START HERE

The decision is not whether optical links and secure control matter to
AI infrastructure. It is which proof state deserves research priority
now. NewPhotonics is already shipping Tower-fabricated 800G–1.6T
optical engines; MACOM says all products in its 3.2T solution are
readily available, but no customer qualification is disclosed; Lattice
has shipped Mach-N2 samples without disclosing production adoption.
Rank TSEM first, MTSI second and LSCC third. The hidden distinction is
cash-conversion timing: shipping, available and sampled are not
interchangeable forms of “AI exposure.” Weakest assumption: Tower’s
shipment state is broad and margin-accretive rather than concentrated;
ECOC customer-qualification language on September 21–23 is the first
event that can narrow or reorder the map.

Suggested Prompt:

“Turn this Feed into three trackable AI rack-link and control thesis
objects. Separate reported evidence from attention, compare previous
state with current state, rank TSEM, MTSI and LSCC, identify the
weakest assumption and status-changing data point for each, and build
7-day and 30-day workflows. Do not give buy or sell advice.”

Or simply ask:

“What did I miss?”

The goal is to turn the same headline theme into a falsifiable
evidence ranking and a precise research queue.

FOR YOUR AI AGENT

Treat this as a structured market-cognition input file, not a recap.
Classify every claim as reported fact, guidance, attention,
interpretation or unresolved. Use the causal chain: AI accelerator
demand → rack bandwidth and control requirements → product
availability → customer qualification → volume shipment → disclosed
revenue and margin → durable rerating.

Process TSEM as the shipment-and-capacity object, MTSI as the
qualification object and LSCC as the sampling-and-incumbency object.
Preserve the ranking unless primary evidence changes a proof state.
Price action and X engagement can prioritize retrieval, but they
cannot promote an object by themselves.

The incremental work beyond the Article is to maintain the
state-change ledger, evidence matrix, unanswered-question ranking and
retrieval registry below. Do not treat this Feed as investment advice.

AGENT_OUTPUT_SPEC

1. Today in One Sentence.
2. Thesis Object Summary with Previous State → Current State.
3. Fundamental Evidence vs Attention Flow.
4. AI Rack Link-and-Control Chain Read-Through.
5. Cross-Object Proof-State Ranking.
6. Attention Stage / Crowding Risk.
7. Weakest Assumption by Object.
8. Most Important Status-Changing Data Point.
9. Attention-to-Rerating Conditions.
10. Attention Fade Conditions.
11. 7-Day Research Workflow and Ranked Retrieval Plan.
12. 30-Day Research Workflow with Validate / Upgrade / Downgrade / Invalidate.
13. WKAP Daily Top 3 Source Follow-Up.
14. Portfolio Fit, only if portfolio context is provided.

TODAY_SUMMARY

Part 1 — Main Market Thesis

Today’s evidence argues for a proof-clock ranking, not one AI-demand
label. TSEM’s platform is attached to a product shipping in volume and
backed by $1.3 billion of 2027 SiPho contracts plus $290 million of
prepayments. MTSI’s 3.2T solution is available, but no qualification
or order is named. LSCC’s Mach-N2 is sampled and orderable, without
disclosed production or revenue. The payoff is a 1→3 research
priority: TSEM leads on production, MTSI is a conditional hold for
customer conversion, and LSCC waits for a named design win. ECOC
customer language—not booth traffic—is the next direct rank-change
evidence for the first two.

Part 2 — Today’s Thesis Objects

TSEM — Shipments Now
Fundamental: NewPhotonics and Tower reported high-volume 800G–1.6T
shipments on PH18DA; 6.4T is scheduled for first-half 2027. Contracts,
prepayments and active customers support the state change;
revenue/margin conversion remains missing. Primary sources:
https://ir.towersemi.com/news-releases/news-release-details/newphotonicsr-and-tower-semiconductor-begin-high-volume
; https://www.sec.gov/Archives/edgar/data/928876/000117891326002609/zk2635309.htm
Attention: @PhotonCap highlights capability becoming production. That
is framing, not financial proof. URL:
https://x.com/PhotonCap/status/2100827394680455227

MTSI — Availability, Then Qualification
Fundamental: MACOM’s 3.2T solution uses eight 448Gbps PAM-4 channels;
all products are readily available. Strong Q3 growth and Q4 guidance
show an operating base, not adoption. A named qualification/order is
missing. Primary sources:
https://ir.macom.com/news-releases/news-release-details/macom-introduces-32t-chipset-solution-high-speed-optical
; https://ir.macom.com/news-releases/news-release-details/macom-reports-fiscal-third-quarter-2026-financial-results
Attention: @aleabitoreddit frames InP DFB-laser scarcity and customer
urgency; it does not prove a 3.2T socket. URL:
https://x.com/aleabitoreddit/status/2085511646257332581

LSCC — Samples Before Sales
Fundamental: Mach-N2 combines secure-control features with a
220K-logic-cell device, sub-30ms configuration, PCIe 4.0, 10G Ethernet
and 16Gbps SERDES. Samples reached unnamed leading customers;
production and revenue are missing. Primary source:
https://ir.latticesemi.com/news-releases/news-release-details/lattice-expands-secure-control-fpga-leadership-new-lattice-mach
Attention: @bigmurf58 frames board-control incumbency as harder to
copy than features and disclosed LSCC as 3.3% of the portfolio. URL:
https://x.com/bigmurf58/status/2100543879032238214

Part 3 — Attention Flow Today

Today’s selected attention signals focus on the equipment that makes
accelerator capacity usable: optical throughput, receive/transmit
components and always-on control. WKAP interpretation: the useful
asymmetry is whether attention matches the proof state. Treat X as a
retrieval signal, never as replacement evidence; aggregate flows are
unmeasured.

Part 4 — The Better Question

The surface question is “Which stock has AI exposure?” The better
question is “Which object has crossed the next cash-conversion gate,
and what single disclosure would change the order?” TSEM stays first
if shipments broaden and convert into disclosed SiPho revenue and
margin while 6.4T remains on schedule. MTSI overtakes only after a
named 3.2T qualification, design win or volume order. LSCC leaves the
waiting room only after a named Mach-N2 production win or measurable
compute/communications revenue path.

MARKET_REGIME

RISK_TONE: Selective and proof-sensitive inside a crowded AI
infrastructure narrative.
MAIN_DRIVER: AI rack bandwidth and secure-control demand filtered
through product availability, qualification, shipment, revenue and
margin.
MARKET_CONTEXT: Product disclosures are fresh; the U.S. cash session
has not opened at publication time. ECOC on September 21–23 is the
next direct industry evidence window for TSEM and MTSI.
ATTENTION_ENVIRONMENT: Selected attention favors optical and control
bottlenecks; aggregate flows are unmeasured and customer/financial
confirmation is slower.
WKAP_VIEW: Promote research only when the proof state advances. Rank
TSEM on shipment conversion, MTSI on qualification conversion and LSCC
on sample-to-production conversion.

AI_RACK_LINK_AND_CONTROL_CHAIN_UPDATE

Accelerator demand → electrical/optical throughput requirement →
photonic engine and component availability → customer qualification →
volume shipment → rack deployment → revenue and margin. In parallel,
accelerator and network boards require always-on sequencing,
monitoring, firmware integrity, root of trust and recovery →
secure-control FPGA sampling → platform qualification → persistent
socket revenue.

These companies are parallel beneficiaries, not a disclosed supplier
chain. TSEM’s evidence sits furthest to the right because a customer
product is shipping on Tower’s platform. MTSI sits one gate earlier:
technical readiness is disclosed, customer commitment is not. LSCC
sits earlier still: samples are in customer hands, but incumbency is a
hypothesis until production evidence arrives.

ATTENTION_TRADE_BOARD

Attention Trade Board

WKAP Attention View

Strongest fundamental change and most evidence-backed attention trade:
TSEM. Cleanest evidence-to-attention asymmetry: MTSI, because
technical completeness is visible while customer conversion is not.
Largest optionality/evidence gap and highest fade risk: LSCC. Most
crowded narrative: MTSI’s AI-optics frame qualitatively, though flows
are unmeasured. Best current candidate for durable rerating: TSEM if
shipments convert into disclosed economics; MTSI can close the gap
with a named qualification.

RADAR_OBJECT_INDEX

THESIS_OBJECT_1: TSEM
THEME: AI optical engines / silicon photonics / volume shipment /
capacity conversion
STATUS: High-volume shipment validated; revenue and margin conversion pending
POSITION_CONTEXT: [not provided]
PRICE_AT_PUBLISH: $223.25 (U.S. premarket; latest trade 08:35 ET /
19:35 ICT, September 18, 2026)
DATE_FIRST_ADDED_TO_RADAR: 2026-09-18
SETUP_TYPE: Shipment-backed SiPho capacity conversion
ATTENTION_STAGE: Production recognition entering an economic-conversion test
ATTENTION_WINDOW: 1-30 days
KEY_QUESTION: Do volume shipments convert into disclosed SiPho
revenue, margin and broader customer adoption while 6.4T stays on
schedule?

THESIS_OBJECT_2: MTSI
THEME: AI optical front-end / 3.2T / qualification / component bottleneck
STATUS: Technical availability validated; customer conversion pending
POSITION_CONTEXT: [not provided]
PRICE_AT_PUBLISH: $268.20 (U.S. premarket; latest trade 08:22 ET /
19:22 ICT, September 18, 2026)
DATE_FIRST_ADDED_TO_RADAR: 2026-09-18
SETUP_TYPE: Complete optical solution awaiting named qualification
ATTENTION_STAGE: Product-launch attention entering an adoption test
ATTENTION_WINDOW: 1-30 days
KEY_QUESTION: Which customer qualifies or orders the 3.2T solution, at
what volume and on what timetable?

THESIS_OBJECT_3: LSCC
THEME: AI secure control / FPGA / sampling / board incumbency
STATUS: Product and sampling validated; production adoption unproven
POSITION_CONTEXT: [not provided]
PRICE_AT_PUBLISH: $109.82 (U.S. premarket; latest trade 08:20 ET /
19:20 ICT, September 18, 2026)
DATE_FIRST_ADDED_TO_RADAR: 2026-09-18
SETUP_TYPE: Sampling-stage control-layer optionality
ATTENTION_STAGE: Feature and incumbency attention ahead of production proof
ATTENTION_WINDOW: 7-90 days
KEY_QUESTION: Do Mach-N2 samples become named, persistent production
sockets with a disclosed revenue path?

THESIS OBJECTS

THESIS_OBJECT_1 — TSEM
CARD_ID: WKAP-RADAR-2026-09-18-TSEM
CARD_TITLE: Shipments Matter When Capacity Commitments Become Economics
TYPE: AI Optical Shipment Trackable
THEME: Silicon photonics / optical engines / capacity / revenue conversion
STATUS: Volume shipment validated; conversion test active
POSITION_CONTEXT: [not provided]
PRICE_AT_PUBLISH: $223.25 (U.S. premarket; latest trade 08:35 ET /
19:35 ICT, September 18, 2026)
DATE_FIRST_ADDED_TO_RADAR: 2026-09-18
ATTENTION_STAGE: Production recognition entering economic confirmation
ATTENTION_WINDOW: 1-30 days

WHY_A_READER_CARES_NOW
TSEM is the only object whose fresh evidence has crossed into
high-volume shipment. It deserves first research priority only if
production converts into disclosed revenue and margin.

PREVIOUS_STATE: Signed SiPho contracts, customer prepayments, active
customers and a 6.4T roadmap supported future capacity demand.
CURRENT_STATE: NewPhotonics is shipping Tower-fabricated 800G–1.6T
engines in volume; the contract/capacity thesis now has product-level
production evidence.
DECISION_OR_MONITORING_PAYOFF: Keep TSEM first while shipment
conversion broadens and the first-half 2027 6.4T timetable holds.

THESIS_SUMMARY
Tower’s evidence stack aligns product, capacity reservation and
customer cash. The open question is not whether this disclosed program
has demand; it is how profitably it converts and whether concentration
or timing weakens the signal.

WKAP_ANGLE
The surface-level frame: TSEM is another AI optical beneficiary.
The alternative frame: It is the current proof-state leader because a
customer product is shipping against signed capacity commitments, but
shipment headlines still require revenue and margin confirmation.
The key research question: How much SiPho revenue and margin do these
shipments create, across how many customers, before the 6.4T ramp?

CORE_THESIS
The September 17 release says laser-integrated optical-engine PICs
made on Tower’s PH18DA platform are in high-volume shipment. Current
products support 800G–1.6T at 200G per lane, and the 6.4T NPC505 is
scheduled for first-half 2027 volume shipment. Tower’s May SEC filing
disclosed $1.3 billion of signed 2027 SiPho contracts, $290 million of
prepayments, more than 50 active customers and larger 2028 wafer
commitments. This supports production and capacity demand; it does not
disclose today’s revenue mix, margin or concentration.

ATTENTION_TRADE_FRAME
Attention Source: @PhotonCap, https://x.com/PhotonCap/status/2100827394680455227
Why Today: The post highlights the state transition from capability to
800G/1.6T production.
Attention Stage: Production recognition; economic confirmation pending.
Attention vs Evidence
Hard evidence: Shipment, throughput, contract, prepayment and
active-customer facts above.
Primary sources:
https://ir.towersemi.com/news-releases/news-release-details/newphotonicsr-and-tower-semiconductor-begin-high-volume
; https://www.sec.gov/Archives/edgar/data/928876/000117891326002609/zk2635309.htm
Attention / interpretation: The post directs attention to PH18DA
production; it does not prove Tower’s financial conversion.
Attention Path: Shipment disclosure → customer breadth and wafer
starts → SiPho revenue/margin → 6.4T ramp → durable rerating or fade.
Attention Asymmetry: Production is reported while economics remain undisclosed.
Crowding Risk: Medium qualitative; flows unmeasured.
What Could Sustain Attention: Broader customer shipments, disclosed
SiPho revenue/margin, conversion of prepayments and an on-time 6.4T
ramp.
What Could Make Attention Fade: Customer pushouts, weak
shipment-to-revenue conversion, concentration risk or a 6.4T delay.
Attention-to-Thesis Conversion: Require disclosed economic conversion
plus schedule integrity; do not promote on engagement or exhibition
visibility alone.

WEAKEST_ASSUMPTION
The reported shipment state represents broad, margin-accretive demand
rather than a concentrated program whose economics arrive later.

MOST_IMPORTANT_DATA_POINT
Disclosed SiPho revenue and margin contribution tied to customer
shipment breadth.

NEXT_DATA_POINT
ECOC customer/qualification language, followed by Tower’s next SiPho
revenue, mix, capacity, prepayment and 6.4T timing update.

THESIS_OBJECT_2 — MTSI
CARD_ID: WKAP-RADAR-2026-09-18-MTSI
CARD_TITLE: A Complete 3.2T Solution Still Needs a Customer Name
TYPE: AI Optical Qualification Trackable
THEME: Optical front-end / 3.2T / InP components / customer qualification
STATUS: Technical package and company operating base validated;
adoption test active
POSITION_CONTEXT: [not provided]
PRICE_AT_PUBLISH: $268.20 (U.S. premarket; latest trade 08:22 ET /
19:22 ICT, September 18, 2026)
DATE_FIRST_ADDED_TO_RADAR: 2026-09-18
ATTENTION_STAGE: Launch recognition entering qualification
ATTENTION_WINDOW: 1-30 days

WHY_A_READER_CARES_NOW
MTSI offers the cleanest rank-change catalyst: one named qualification
or order can move it closer to TSEM’s proof state.

PREVIOUS_STATE: Strong optical-component momentum and customer urgency
around InP DFB-laser constraints.
CURRENT_STATE: MACOM has assembled a readily available 3.2T front-end
solution across receive and transmit functions, but has not named a
customer commitment.
DECISION_OR_MONITORING_PAYOFF: Hold second until a qualification,
design win or order converts technical completeness into commercial
evidence.

THESIS_SUMMARY
MACOM’s product set addresses the next bandwidth step with a reported
operating base. The asymmetry is a complete solution before customer
disclosure; the risk is confusing “readily available” with adoption.

WKAP_ANGLE
The surface-level frame: AI optics demand automatically pulls through
every high-speed component supplier.
The alternative frame: MACOM must convert technical completeness and
component scarcity into named sockets and revenue.
The key research question: Who qualifies the 3.2T solution, and when
does it affect the $415–425 million Q4 revenue bridge?

CORE_THESIS
MACOM’s new front-end chipset reaches 3.2T through eight 448Gbps PAM-4
channels. It combines new TIAs and photodiodes with previously
introduced transmit drivers; the photodiode exceeds 100GHz bandwidth,
and all products are readily available. Q3 revenue was $342.2 million,
up 35.8% year over year and 18.4% sequentially; GAAP gross margin was
58.3%. Q4 guidance calls for $415–425 million revenue and 60%–61%
adjusted gross margin. These company-wide figures show an operating
base, not 3.2T adoption.

ATTENTION_TRADE_FRAME
Attention Source: @aleabitoreddit,
https://x.com/aleabitoreddit/status/2085511646257332581
Why Today: The older earnings-call excerpt frames urgent customer
interest around an InP DFB-laser shortage as MACOM launches the new
solution.
Attention Stage: Bottleneck recognition entering a qualification test.
Attention vs Evidence
Hard evidence: Product specifications, availability, reported Q3
growth and Q4 guidance.
Primary sources:
https://ir.macom.com/news-releases/news-release-details/macom-introduces-32t-chipset-solution-high-speed-optical
; https://ir.macom.com/news-releases/news-release-details/macom-reports-fiscal-third-quarter-2026-financial-results
Attention / interpretation: Component scarcity may create urgency;
it does not establish a 3.2T socket win.
Attention Path: Product availability → evaluation →
qualification/design win → volume order → revenue/margin contribution
→ rank promotion or fade.
Attention Asymmetry: The technical package is visible; the customer
and volume are not.
Crowding Risk: Medium-high qualitative because AI optics is a popular
frame; flows unmeasured.
What Could Sustain Attention: Named qualifications, ECOC customer
language, volume orders, revenue above the guide or margin leverage.
What Could Make Attention Fade: Demonstration-only status, absent
qualification timelines, loss of shortage urgency or results below
guidance.
Attention-to-Thesis Conversion: Promote only when primary evidence
names or quantifies customer commitment.

WEAKEST_ASSUMPTION
Customers prefer the integrated single-supplier path enough to qualify
it quickly, rather than mix components or delay the bandwidth
transition.

MOST_IMPORTANT_DATA_POINT
A named 3.2T qualification, design win or volume order with timing.

NEXT_DATA_POINT
ECOC customer and sampling language, then management’s
product-specific contribution inside Q4 results.

THESIS_OBJECT_3 — LSCC
CARD_ID: WKAP-RADAR-2026-09-18-LSCC
CARD_TITLE: Board-Control Incumbency Begins Only After Samples Convert
TYPE: AI Secure-Control Sampling Trackable
THEME: Secure control / FPGA / post-quantum trust / platform incumbency
STATUS: Product and samples validated; production and revenue test pending
POSITION_CONTEXT: [not provided]
PRICE_AT_PUBLISH: $109.82 (U.S. premarket; latest trade 08:20 ET /
19:20 ICT, September 18, 2026)
DATE_FIRST_ADDED_TO_RADAR: 2026-09-18
ATTENTION_STAGE: Feature and incumbency attention before customer proof
ATTENTION_WINDOW: 7-90 days

WHY_A_READER_CARES_NOW
LSCC offers longer-duration optionality in an overlooked control
layer, but paying for incumbency before a named production win
reverses the evidence sequence.

PREVIOUS_STATE: A general secure-control FPGA thesis built on
low-power, always-on board functions.
CURRENT_STATE: Mach-N2 adds a larger, faster secure-control family and
has reached leading-customer sampling; no production adoption or
revenue is disclosed.
DECISION_OR_MONITORING_PAYOFF: Keep LSCC third until samples become a
named production socket with a measurable path.

THESIS_SUMMARY
AI systems need sequencing, monitoring, root of trust, firmware
integrity and recovery around accelerators and optical links. Mach-N2
targets that persistent layer. The moat thesis depends on
qualification and socket duration, not a feature checklist.

WKAP_ANGLE
The surface-level frame: Post-quantum features make Mach-N2 an
immediate AI winner.
The alternative frame: Security features can diffuse; durable value
comes from board-control qualification and switching friction.
The key research question: Which sampled customer moves to production,
and how persistent is the socket?

CORE_THESIS
Mach-N2 combines integrated flash, a hardware root of trust and CNSA
2.0-compliant post-quantum cryptography. The 220K-logic-cell device
configures in under 30 milliseconds and supports PCIe 4.0, 10G
Ethernet and up to 16Gbps SERDES. Lattice says samples reached leading
compute and communications customers and devices are orderable. The
release does not name customers, production qualification, unit volume
or revenue.

ATTENTION_TRADE_FRAME
Attention Source: @bigmurf58,
https://x.com/bigmurf58/status/2100543879032238214. The author
disclosed LSCC as 3.3% of the portfolio.
Why Today: The post distinguishes commoditizing features from
persistent board-control incumbency after the Mach-N2 launch.
Attention Stage: Early thesis formation while sampling evidence is fresh.
Attention vs Evidence
Hard evidence: Product specifications, sample shipment and orderability.
Primary source:
https://ir.latticesemi.com/news-releases/news-release-details/lattice-expands-secure-control-fpga-leadership-new-lattice-mach
Attention / interpretation: Board-control incumbency may be durable;
the disclosed holding makes the interpretation interested, not
neutral.
Attention Path: Samples → qualification → named design win →
production deployment → socket persistence and revenue → promotion or
fade.
Attention Asymmetry: Strategic plausibility is high; commercial proof is early.
Crowding Risk: Medium qualitative; disclosed author interest, flows unmeasured.
What Could Sustain Attention: Named compute/communications customers,
production qualification, revenue contribution and evidence of socket
persistence.
What Could Make Attention Fade: An extended sampling cycle,
qualification delays or comparable functionality from larger
programmable-logic suppliers.
Attention-to-Thesis Conversion: Require a production order or
measurable segment contribution; do not promote on product
specifications alone.

WEAKEST_ASSUMPTION
Mach-N2’s integration and time-to-market create persistent
control-plane switching costs before competitors reproduce enough
functionality.

MOST_IMPORTANT_DATA_POINT
A named production qualification or design win with a disclosed revenue path.

NEXT_DATA_POINT
Customer identity, qualification status, production timing,
unit/revenue contribution and competitive displacement evidence.

CROSS_OBJECT_ATTENTION_COMPARISON

Cross-Object Attention Comparison

Cleanest Attention Trade: MTSI’s explicit qualification gap creates
the clearest rank-change trigger.
Most Evidence-Backed Attention Trade: TSEM.
Most Crowded Attention Trade: MTSI’s AI-optics narrative
qualitatively; flows are unmeasured.
Highest Fade Risk: LSCC if sampling remains unnamed and unquantified.
Best Candidate to Become a Durable Thesis: TSEM today; MTSI can
challenge with a named qualification.

7_DAY_RESEARCH_WORKFLOW

TSEM — 7-Day Checks
1. Capture every ECOC reference to shipment breadth, customer
qualification and 6.4T timing.
2. Build a contract-to-revenue bridge from $1.3B contracts and $290M
prepayments; mark undisclosed assumptions.
3. Preserve the gate: production plus disclosed economics; invalidate
on pushouts or schedule slippage.

MTSI — 7-Day Checks
1. Retrieve ECOC product sheets, management comments and any
customer/ecosystem references.
2. Separate product availability, sampling, qualification, design win
and volume order as distinct states.
3. Preserve the gate: named qualification/order; downgrade if the
launch remains demonstration-only.

LSCC — 7-Day Checks
1. Retrieve Mach-N2 evaluation, toolchain and customer references;
record whether any customer is named.
2. Compare board-control integration claims with competing
programmable-logic offerings.
3. Preserve the gate: named production qualification; do not upgrade
for specifications or social engagement.

Cross-Object — 7-Day Checks
Maintain a dated evidence ledger with previous state, new fact, source
class, confidence and rank effect. Keep price and X engagement in the
attention column. Re-rank only when a primary source advances
qualification, shipment, production or disclosed economics.
Ranked retrieval priorities: (1) TSEM SiPho revenue/margin and
customer breadth; (2) MTSI named 3.2T qualification/order and timing;
(3) LSCC named Mach-N2 production customer and remaining gate.
Primary-source retrieval registry:
TSEM=https://ir.towersemi.com/news-releases/news-release-details/newphotonicsr-and-tower-semiconductor-begin-high-volume;MTSI=https://ir.macom.com/news-releases/news-release-details/macom-introduces-32t-chipset-solution-high-speed-optical;LSCC=https://ir.latticesemi.com/news-releases/news-release-details/lattice-expands-secure-control-fpga-leadership-new-lattice-mach
Attention-source retrieval registry:
TSEM=https://x.com/PhotonCap/status/2100827394680455227;MTSI=https://x.com/aleabitoreddit/status/2085511646257332581;LSCC=https://x.com/bigmurf58/status/2100543879032238214
Evidence-state snapshot:
TSEM=volume-shipping/economics-pending/6.4T-1H27;MTSI=3.2T-available/qualification-pending/Q4-guide-$415M-$425M;LSCC=Mach-N2-sampled/production-pending;classification=fact-versus-guidance-versus-attention-versus-interpretation;rule=shipping>qualified>available>sampled-when-source-quality-equal;source-rule=SEC-or-company-IR-controls-facts/KOL-controls-attention/price-does-not-change-status.

30_DAY_RESEARCH_WORKFLOW

TSEM — 30-Day Checks
Validate: Volume shipments persist and contract/prepayment conversion
is consistent with the disclosed timetable.
Upgrade: Tower discloses broad customers plus attractive SiPho
revenue/margin and confirms 6.4T.
Downgrade: Shipments continue but customer breadth or economics remain
narrow/late.
Invalidate: Pushouts, failed conversion or a material 6.4T delay.

MTSI — 30-Day Checks
Validate: A credible qualification path and customer evaluation
timeline emerge while company results meet guidance.
Upgrade: Named design win or volume order is tied to revenue/margin
contribution.
Downgrade: Availability persists without qualification or ECOC yields
only demonstrations.
Invalidate: Product relevance weakens, timing slips or total results
materially miss guidance.

LSCC — 30-Day Checks
Validate: Sampling advances toward production with identifiable
workloads and qualification milestones.
Upgrade: Named production design win and measurable revenue/socket scope.
Downgrade: Sampling remains unnamed and unquantified beyond the expected cycle.
Invalidate: No conversion or competing functionality erases the time advantage.

Cross-Object — 30-Day Checks
Score every object on one scale: product readiness, customer
commitment, shipment, disclosed economics and schedule integrity.
Upgrade only when the current missing gate advances. Downgrade even if
price rises when evidence stalls. At day 30, issue a state-change memo
with previous state, current state, evidence added, evidence missing,
rank effect and the next retrieval request.

WKAP DAILY TOP 3

Three market sources worth feeding into today’s market chat. Not
required reading — WKAP has already extracted the signal.

1. @PhotonCap — PH18DA Moves from Capability to Shipment
URL: https://x.com/PhotonCap/status/2100827394680455227
WKAP signal: The post emphasizes 800G and 1.6T production on Tower’s
PH18DA platform. It is an attention signal around a real shipment
state change; primary disclosures control contract, prepayment and
schedule facts.
Why it matters today: TSEM’s lead depends on converting the shipment
headline into broad, attractive economics rather than treating
production as the end of diligence.
Themes/tickers: Silicon photonics / optical engines / volume shipment / TSEM
Question to ask: “What SiPho revenue, margin, customer-breadth and
6.4T evidence would confirm that TSEM’s production lead is durable?”

2. @aleabitoreddit — DFB-Laser Urgency Meets a New 3.2T Solution
URL: https://x.com/aleabitoreddit/status/2085511646257332581
WKAP signal: The post highlights management’s earlier account of
urgent customer interest during an InP DFB-laser shortage. That frames
the bottleneck; it does not prove the newly launched 3.2T solution has
a socket.
Why it matters today: A named qualification or order is the clearest
event that can close MTSI’s proof gap with TSEM.
Themes/tickers: InP / DFB lasers / optical front-end / 3.2T / MTSI
Question to ask: “Which customer qualification, design win or
volume-order evidence converts MACOM’s technical availability into
commercial proof?”

3. @bigmurf58 — Secure-Control Incumbency Is the Real Moat Test
URL: https://x.com/bigmurf58/status/2100543879032238214
WKAP signal: The author argues that individual security features can
commoditize while persistent board-control sockets are harder to
displace, and disclosed LSCC as 3.3% of the portfolio. This is
interested interpretation, not production evidence.
Why it matters today: Mach-N2’s specifications are credible, but the
thesis changes only when a sampled customer becomes a named production
socket.
Themes/tickers: Secure control / FPGA / post-quantum trust / board
incumbency / LSCC
Question to ask: “What named qualification, production timing and
revenue evidence would prove Mach-N2 has moved from sampled features
to persistent platform incumbency?”

WKAP Radar Feed - 2026 - 09 - 18

Agent-readable facts