SMTC is becoming a high-margin AI connectivity supplier, with accelerating optical adoption and rising content per system creating a path to durable earnings growth.
The growth outlook rests on concentrated, capacity-constrained AI deployments and unusually favorable mix and pricing; execution slippage or normalization could expose elevated estimates and valuation.
Semtech announced an expanded multi-year agreement with Tower Semiconductor, building on their 20-year partnership, and will invest in securing dedicated silicon germanium (SiGe) capacity at Tower.
Semtech's investment in SiGe capacity at Tower Semiconductor is in support of its strong design wins and demand for AI Data Center optical interconnect components, indicating a positive outlook for its products.
TSEM +2.6% after SMTC agreed to fund dedicated SiGe capacity at TSEM for its AI optical-interconnect parts.
Seaport initiated a Buy rating with a $200 price target, believing Semtech has reinvented itself around data-center optical and is well-positioned for the 2027–28 AI ramp.
Seaport sees SMTC as well positioned to benefit as AI optical demand accelerates through 2027–28.
The company reported strong revenue and EPS, exceeding Street expectations, driven by faster 1.6T FiberEdge qualifications and improved mix leverage.
Data Center revenue significantly increased year-over-year and quarter-over-quarter, with 1.6T becoming a major component of the mix.
The existing 800G cycle is strengthening, and 1.6T FiberEdge qualified earlier than anticipated, leading to expectations of significant 1.6T TIA share by fiscal year-end.
The earlier-than-expected 1.6T FiberEdge qualifications were a significant positive, with the company expecting 1.6T FiberEdge and CopperEdge to exceed 50% of Data Center revenue in F3Q.
The 1.6T ramp is not cannibalizing 800G, as industry 800G transceiver expectations have risen, and SMTC's 800G TIA share has increased, creating simultaneous tailwinds.
Adjusted gross and operating margins increased significantly, with F3Q guidance showing further expansion driven by product mix, even before the cellular-module divestiture.
Excluding cellular modules, gross margin is guided to a high level, which management views as a starting point for further improvement as high-margin products become larger contributors.
Operating cash flow and free cash flow were strong, exceeding Street estimates, with CapEx remaining low.
The company is transforming into a focused, high-margin AI connectivity supplier, with its position not limited to one product or architecture.
The cellular-module sale formalizes the strategic shift, removing a low-margin business and concentrating investment on Data Center and LoRa, where the company has strong growth and leadership.
Bookings and backlog extend well into the next fiscal year, indicating strong future demand.
The company's competitive advantage lies in its analog signal-integrity expertise, established module-provider relationships, and expanding content opportunity.
The 800G TIA share increased significantly, and management believes the company can replicate this success with 1.6T, while the 800G market itself has reset higher.
ASPs for 1.6T are significantly higher than 800G, and the potential content per transceiver for 3.2T could be substantially larger, expanding the long-term revenue opportunity.
The earlier FiberEdge qualification shows that the 1.6T thesis is converting into orders faster than expected, with strong Data Center guidance and significant FY28 demand in backlog.
The earnings conversion is stronger than the revenue upside, with F3Q EPS significantly above Street and strong gross and operating margins, indicating the potential for a high-margin business.
The wide upward revisions to out-year EPS across models reflect the margin reset, indicating increased confidence in the company's profitability.
The near-term bear case weakened due to strong evidence against demand uncertainty and margin skepticism, with the 1.6T conversion largely proven.
ZM -7% was also mixed while SMTC +5% the bright spot on a nice DC driven beat and raise. We’ll cover those first, then move onto the usual.
The new estimates capitalize on an unusually favorable combination of accelerated customer qualification, scarcity pricing, elevated backlog, and simultaneous strength in 800G and 1.6T, which may not be sustainable.
Backlog at this stage is not the same as non-cancellable revenue, and customer deployment schedules can change, leading to potential order shifts.
AI infrastructure demand is concentrated among a small number of hyperscalers, and the initial large ACC deployment still heavily relies on a lead customer and a new platform cycle.
Capacity is a two-sided issue; while it validates demand, it can prevent backlog from converting to revenue and may require prepayments and higher CapEx.
Front-end wafer supply could constrain upside in 2H FY28, and qualifying additional facilities introduces yield, timing, and execution risks.
If capacity arrives late, the company could lose share when competitors are expanding their own capacity, while if it arrives into slower demand, it could lead to underutilized commitments.
The high gross margin starting point is established during a period of favorable scarcity pricing and rapid mix shift, and assumes the module sale closes on time, with no return of normal optical price erosion.
The sale price of the cellular-module business was low, reflecting its low profitability, but also means the company is surrendering a meaningful revenue stream.
Outside of AI Data Center and LoRa, High-End Consumer remains down year-over-year, and management did not provide evidence of a broad macro recovery.
The company delivered a very strong earnings print and guide, with Q3 revenue, margins, and EPS all substantially exceeding expectations.
Commentary from the company points to accelerating demand.
The disposal of the held-for-sale business is key for bulls, who favor the faster-growing core portfolio and its potential for upper-50s margins.
Tower Semiconductor's capacity expansion is considered 'hyper bullish' for Semtech, which manufactures the amplifier chips that fill that capacity.
The company delivered a solid beat and raise with numbers and guide above expectations, implying significant acceleration.
The call confirmed the multi-legged product story and the DC 2H FY27 acceleration commentary.
Management called out upside tied to data center, optical, and ACC momentum as driving the upside.
Data center revenue hit $71.6M, +39% y/y, and management guided FQ2 data center to roughly $97M, +85% y/y / +35% q/q.
Management walked away from the prior “50% y/y floor” framing and instead said growth should accelerate in 2H FY27, supported by backlog, bookings, 1.6T FiberEdge, CopperEdge, and HieFo optical components.
Demand for 800G FiberEdge/TIA remains “exceptionally strong,” with wins across established and emerging module suppliers, including some sole-source sockets.
The FQ2 guide appears driven meaningfully by continued 800G strength, before the larger 1.6T contribution really builds in the back half.
Management highlighted significant 1.6T optical design wins with major module makers using latest-generation DSPs, with shipments launching in FQ2 and module ramps supported by backlog in 2H.
Early 1.6T volume is still FRO, while 1.6T LRO/LPO is under evaluation, creating additional upside if linearized optics adoption broadens.
Management emphasized it began preparing capacity ~18 months ago, can support drop-in demand, and is working with foundry/OSAT/test partners to double or triple current capacity.
HieFo gain-chip demand is running about 3x supply, with capacity targeted to rise 3-4x by year-end and another 3-4x by the end of next year.
The company started shipping 1.6T CopperEdge ICs to cable partners for a U.S. hyperscaler, and management said ACC and on-board linear equalizer engagements should convert into design wins.
FQ1 gross margin was 53.0% vs Street 52.5%, and FQ2 guide is 54.0% vs Street 52.6%.
Management expects the 1.6T data center portfolio to be accretive to semiconductor products and Signal Integrity GM as it scales.
The company is rapidly transitioning into a scarce AI data-center connectivity compounder.
The data-center ramp is not dependent on a single product cycle, with multiple strong product lines contributing.
This quarter made the portfolio breadth much more tangible, as management tied the FQ2 guide and 2H acceleration to bookings, backlog, and specific product ramps.
Visibility improved enough to move numbers materially higher, with data center growth potentially exceeding 100%+ exiting the year.
The capacity story supports the bull thesis, as management can support drop-in demand and is adding capacity, reducing the risk that demand exists but cannot convert to revenue.
We’ll hit up SMTC +10% and ZS -25% first then get to the usual, where a lot of good stuff awaits today…
The stock has already discounted a lot of the data-center upside, and the burden of proof has risen sharply.
The new guide needs to be followed by sustained backlog conversion, capacity execution, hyperscaler adoption, and margin expansion.
800G strength could normalize as 1.6T ramps, pricing/ASP pressure could emerge, and ACC/LPO adoption could be slower or more concentrated than bulls expect.
A key concern is customer concentration around early deployments, especially the lead U.S. hyperscaler ACC ramp, and whether additional hyperscalers move fast enough to support the multiple.
Execution complexity is a concern, as the company is simultaneously ramping multiple product lines.
HieFo demand is above supply, but capacity, yield, and margins have to scale.
R&D is stepping up, which reduces near-term operating leverage.
Outside data center, high-end consumer missed Street, IoT Systems remains lower growth and is still awaiting divestiture, and broad macro, IT spending, tariffs, and geopolitics remain risks to the non-AI parts of the portfolio.
Vik's Newsletter argues that increasing AI rack power density is creating a new interconnect category where Active Copper Cable (ACC) becomes the optimal solution between passive copper and optics.
The note highlights the company currently leading in ACC through Meta Catalina wins and efficient power usage.
ACC/Interconnects/SMTC/MRVL: Vik’s Newsletter Says Rising Rack Power Density Is Creating a New Active Copper Opportunity
The company delivered record quarterly results and expanded design wins across its data center and LoRa businesses, indicating a strong start to fiscal year 2027.
Data center growth is expected to accelerate throughout the year due to revenues from FiberEdge and CopperEdge 1.6T layering onto a strong growth base.
The company has strong conviction in its growth trajectory due to the strength of its bookings and backlog, a broad and well-positioned product portfolio, and targeted R&D investments aligned with customer roadmaps.
The company's Q1 results exceeded expectations with higher revenue, EPS, EBITDA, and gross margin compared to estimates.
The Q2 guidance for revenue, EPS, and EBITDA was significantly above street expectations.
Northland downgraded the company on valuation, despite acknowledging improving fundamental setup.
Research through 2026-10-13 · 2026-05-26 to 2026-10-13 · 7 source reports.
| Range | Prices through |
|---|---|
| 1W | |
| 1M | |
| 3M | |
| 6M | |
| 1Y | |
| 2Y | |
| 3Y |
Research through 2026-11-03. Artifact validity does not establish factual accuracy or research freshness.