top of page
OPTICS · SEMICONDUCTORS · AI INFRASTRUCTURE | stt@simpletechtrend.com
Articles: Blog2
ARTICLES
Articles
Explainers, conference notes, earnings-call reads and market insights on optical communications, silicon photonics, CPO and the AI infrastructure supply chain.
Search


To Understand the 1.6T Optical Module Shortage, First Understand Two Chips: InP Lasers and DSPs
Cisco says optics is the biggest supply constraint. Inside a 1.6T module, two chips set the ceiling: the InP laser and the 3nm DSP. Four figures explain both, and the three questions to ask about any shortage news.
2 days ago16 min read


Understanding NPO: Two Choices That Shape the Module, External vs Internal Laser and Integrated vs Discrete PD
Two NPO modules with the same 3.2T or 6.4T label can carry very different parts. Four figures show how laser placement and PD integration shape NPO's key components and challenges.
5 days ago10 min read


To Understand Slow-and-Wide Optical Interconnects, First Understand SerDes and Die-to-Die
The slow-and-wide optical interconnect race is decided not by the light source, but by whether the chip edge speaks SerDes or Die-to-Die. microLED, VCSEL arrays and DWDM all want to stretch D2D to 50 meters, if XPU vendors change their interfaces.
Nov 1017 min read


The Most-Patented Steps Are Not the Hardest: Inside the 7 Process Steps of a CW DFB Laser and 13 Vendors' Patent Portfolios
A CW DFB laser goes through seven process steps, and power, single-mode yield and reliability each depend on different ones. Mapping 123 patents from 13 vendors shows the most-patented steps, the front-end and DFB+SOA designs, are not the hardest; grating, regrowth and screening are.
Nov 13 min read


The Switch Breaks Before the GPU: Three Counterintuitive Takeaways from OCP's Open Silicon Photonics White Paper
OCP's 294-page open silicon photonics white paper reaches a counterintuitive answer: the switch, not the GPU, outgrows copper first. A 204.8T switch needs 8x the escape bandwidth copper can deliver, and network power in a Kyber-class rack jumps to about 31%. Here is where optics first enters scale-up, by the numbers.
Oct 157 min read


USFP Explained: How NVIDIA and Cisco's New 400G/lane Pluggable Compares with XPO, OSFP-XD and QSFP-DD
At 400G per lane, pluggable optics hit a connector bottleneck, not an optical limit. NVIDIA and Cisco's USFP buys density with faster lanes; Arista's XPO buys it with more lanes and built-in liquid cooling. The pluggable camp has officially split in two.
Oct 1311 min read


Paper Analysis | One Architecture, Five Generations, 3600x Growth: Google Dissects Its Own TPUs — and the Hidden Hero, Optical Circuit Switching
Google reviews five TPU generations: an architecture set in 2017 survived the Transformer takeover and scaled 3600x. For optical readers, the key line: OCS, one of only two designs still unique to TPU, is what keeps a near-10K-chip machine alive, schedulable and deployable in stages. Power has now replaced cost as the new ceiling.
Oct 57 min read


ZR Is the Meat, Coherent Lite Is the Bone: The Long Game in Optical Transceivers for the Next Decade
ZR is the meat, Coherent Lite is the bone. 2026 is the year ZR generates cash; 2028 is when Coherent Lite truly ramps. Three signals could reverse everything: whether Amazon commits, how fast LPO/CPO win sockets, and production-grade demos at OFC 2027. Taiwan's real opportunity lies in upstream components, not the DSP.
Oct 58 min read


VLSI 2026 | Technical Paper Analysis | Intel Packs 16 Wavelengths into One Fiber: Inside an 800Gb/s, 5.7pJ/b Open-Cavity MRR-DWDM Transceiver
At VLSI 2026, Intel showed a fully working silicon photonics microring DWDM transceiver: 16 wavelengths at 50Gb/s each for 800Gb/s per fiber, all with BER<1e-9, at about 5.7pJ/b, with 22nm CMOS stacked directly on the photonic chip in an open-cavity package. We break down this device-to-system CPO reference design figure by figure.
Oct 510 min read


AuthenX: Widening FAU Alignment Tolerance to ±18µm, With Everything Riding on a Single Metalens
AuthenX (7928) uses metalenses to widen FAU alignment tolerance to ±18µm, tackling CPO's hard-to-align, hard-to-repair, hard-to-scale problems. The technology is compelling, but who owns the "patented metalens" and whether it clears certification and yield are open questions to settle before investing.
Oct 52 min read


Technical Analysis | Google's 8th-Gen TPU Dual Architecture: How TPU 8t and TPU 8i Tackle the Agentic AI Compute Bottleneck
Google's eighth-generation TPU splits into two architectures for the first time. STT breaks down how TPU 8t removes the training I/O bottleneck, and how TPU 8i uses large on-chip SRAM and OCS optical circuit switching to cut MoE inference network hops by 56%.
Oct 54 min read
![[CPO Deep Dive 4/6] Fiber-to-Chip: The Least Glamorous Step in CPO That Holds Everyone Back](https://static.wixstatic.com/media/82206f_8308c9afa8cb472cb73ece317e7090f4~mv2.png/v1/fill/w_442,h_250,fp_0.50_0.50,q_35,blur_30,enc_avif,quality_auto/82206f_8308c9afa8cb472cb73ece317e7090f4~mv2.webp)
![[CPO Deep Dive 4/6] Fiber-to-Chip: The Least Glamorous Step in CPO That Holds Everyone Back](https://static.wixstatic.com/media/82206f_8308c9afa8cb472cb73ece317e7090f4~mv2.png/v1/fill/w_276,h_156,fp_0.50_0.50,q_95,enc_avif,quality_auto/82206f_8308c9afa8cb472cb73ece317e7090f4~mv2.webp)
[CPO Deep Dive 4/6] Fiber-to-Chip: The Least Glamorous Step in CPO That Holds Everyone Back
Light leaks a little at every interface, and coupling is where CPO's optical losses concentrate, an invisible ceiling on yield and serviceability. Doubling V-groove density, edge vs. grating coupling, passive vs. active alignment: this unglamorous layer is where volume-production gaps open up.
Oct 54 min read


Paper Analysis | A Symmetrical, Wavelength Agnostic, Bidirectional, Silicon-Photonic Link Proposal and Demonstration | NVIDIA
NVIDIA proposes a symmetrical, wavelength-agnostic bidirectional silicon photonics link that uses polarization multiplexing to break CPO's port-count and cost bottlenecks. Error-free 32 Gbps bidirectional operation makes "colorless" optics a reality and paves the way for scaling AI cluster networks.
Oct 54 min read
CPO at a Turning Point: How Can the Supply Chain Close the Last Mile to Volume? (OFC 2025)
At an OFC 2025 workshop, leaders from NVIDIA, Broadcom, Intel, Marvell and others laid out what it takes to move co-packaged optics from prototype to high-volume manufacturing. Here are their answers to seven core questions.
Oct 53 min read


Breaking Through AI's Memory and Power Walls: How Marvell Photonic Fabric™ Reshapes Scale-Up Architecture | A Deep Dive into Silicon Photonics Resource Pooling
As copper can no longer keep up with the data movement demands of AI clusters, Marvell answers with its Photonic Fabric optical platform. This is not just a speed upgrade, but an architectural shift that breaks through the memory wall and the power bottleneck.
Oct 53 min read
![[CPO Teardown 3/6] The Most Fragile Link in CPO Is the Laser](https://static.wixstatic.com/media/82206f_45e890f73e32485c8274f79ca2ec5361~mv2.png/v1/fill/w_442,h_250,fp_0.50_0.50,q_35,blur_30,enc_avif,quality_auto/82206f_45e890f73e32485c8274f79ca2ec5361~mv2.webp)
![[CPO Teardown 3/6] The Most Fragile Link in CPO Is the Laser](https://static.wixstatic.com/media/82206f_45e890f73e32485c8274f79ca2ec5361~mv2.png/v1/fill/w_276,h_156,fp_0.50_0.50,q_95,enc_avif,quality_auto/82206f_45e890f73e32485c8274f79ca2ec5361~mv2.webp)
[CPO Teardown 3/6] The Most Fragile Link in CPO Is the Laser
Silicon photonics' original sin is that it can't emit light. Lasers hate heat, yet CPO asks them to sit next to a hot chip — making the light source the reliability ceiling, where one failed laser can cost an entire optical engine. External vs integrated sources and four light-source types each bet on a different future.
Oct 53 min read


OFC 2026 - A Paradigm Shift in AI Cluster Architecture: Scale-Out and Scale-Up Optical Interconnects Explained - Broadcom
Copper's physical limits have become the bottleneck to scaling AI. At OFC 2026, Broadcom showed its latest CPO and OCI progress - ultra-low 6 pJ/bit power and exceptional reliability - signaling that optics will dominate both scale-out and scale-up networks.
Oct 53 min read


While We Talk Yield, They Talk Survival: Decoding the "Wolf Pack" Playbook of China's Optical Transceiver Industry
Away from the GPU spotlight, China's optical transceiver makers are fighting a different war — one built on "good enough" chips, domestic substitution, mature process nodes, and ruthless supply-chain cost control. Here's how that mindset works, and where Taiwan fits in.
Oct 53 min read


OFC 2026 - 16-Wavelength 800-Gbps Bidirectional Silicon Photonics Interconnect - Lightmatter
Silicon photonics leader Lightmatter unveiled microring-based 800G/1.6T interconnect technology at OFC 2026. It tackles thermal compensation in AI compute environments and, through 3D-stacked packaging and integration with the TSMC/ASE ecosystem, marks a new milestone for silicon photonics commercialization.
Oct 53 min read
![[CPO Deep Dive 2/6] The Paradigm Shift in Photonic Packaging: Why CPO Is an Advanced Packaging War](https://static.wixstatic.com/media/82206f_a30d6fbdd21248bd828a252a97c481d3~mv2.png/v1/fill/w_442,h_250,fp_0.50_0.50,q_35,blur_30,enc_avif,quality_auto/82206f_a30d6fbdd21248bd828a252a97c481d3~mv2.webp)
![[CPO Deep Dive 2/6] The Paradigm Shift in Photonic Packaging: Why CPO Is an Advanced Packaging War](https://static.wixstatic.com/media/82206f_a30d6fbdd21248bd828a252a97c481d3~mv2.png/v1/fill/w_276,h_156,fp_0.50_0.50,q_95,enc_avif,quality_auto/82206f_a30d6fbdd21248bd828a252a97c481d3~mv2.webp)
[CPO Deep Dive 2/6] The Paradigm Shift in Photonic Packaging: Why CPO Is an Advanced Packaging War
CPO isn't stalled because the components can't be built, but because packaging has switched to an entirely new method. Once the optical engine moves onto the ASIC substrate, photonic packaging must become semiconductor-grade. Whoever standardizes it for foundries and OSATs controls the CPO timeline.
Oct 53 min read
SUBSCRIBE
Want the full supply-chain breakdown?
Subscribe to STT for premium articles: in-depth earnings-call reads, supply-chain maps and weekly industry insights.
bottom of page