MediaTek: “400G or 448G SerDes IP is really going well. We certainly expect that to be ready sometime in the next year, maybe second half of next year. Very interesting comment.

"2nm optics" sounds exotic, but the 2nm is mostly in the DSP > The real message behind “2nm optical” is less about the photonics itself and more about $MRVL’s DSP/SerDes capability. > Moving coherent/PAM4 DSP into 2nm while still hitting 400G/lane-class performance and aggressive power-per-bit targets is the important part. > It highlights that $MRVL’s moat in AI optical connectivity is not just photonics, but also the DSP behind it.

$MRVL TO SHOWCASE NEW 2NM OPTICAL TECHNOLOGY FOR AI DATA CENTERS Marvell says it will demonstrate several first-of-their-kind 2nm optical interconnect technologies at ECOC 2026 as AI networks move from 1.6T toward 3.2T connectivity. The demos include: • 2nm 400G-per-lane optical PAM4 • 2nm 800G ZR/ZR+ with integrated MACsec • 2nm 1.6T ZR and coherent-lite O-band • 102.4T co-packaged optics platform using 200G-per-lane silicon photonics The focus is higher bandwidth density and lower power per bit as hyperscalers connect increasingly large clusters of GPUs, CPUs, accelerators and memory. Marvell says its CPO platform integrates optics directly with AI accelerators and switch silicon as copper reaches practical distance and power limits in large AI fabrics.

Today's brain food: - chicken quarters from 🇵🇹 local farm with 🥗 brussel sprouts - steak from 🇮🇪 Ireland (it's just too good) In the pan because not frozen (so no airfryer potential) Then finishing off with 400g 🍦 Skyr to hit 180g protein

👀 #MediaTek# confirms its AI ASIC collaboration with #Intel’s# EMIB-T advanced packaging. The chipmaker is now pushing 400G SerDes IP to target next-gen ASIC opportunities, including reportedly #Google’s# TPU v10 and Meta’s custom AI chips.💡More: 🔗

📢 Introducing This Week’s Stockcoin Hong Kong IPO Featured Project 🌐 InnoLight ｜Powering the AI Data Center Era InnoLight provides high-speed optical communication solutions for global cloud computing, data centers, and AI infrastructure. 📌 Key Business Areas: 🔹 High-Speed Optical Modules 400G, 800G, and next-generation optical solutions powering faster AI data transmission. 🔹 AI Data Center Infrastructure Capturing growth from AI computing demand, data center expansion, and rising connectivity needs. 🔹 Next-Generation Optical Technology Developing advanced optical interconnect solutions for future AI networks.

my prediction on the photonics/laser stuff is, there will be more and more SiPho as we add more lanes and need better mux/demux, TFLN is not gonna win for 400G per lane although you hear it a lot, as some folks already made it working on SiPho and the TFLN price and volume won't be even close to beating SiPho. DSP will be dominated by broadcom and then marvell for a long time, but SiPho is just so much easier to design and make, and the Chinese companies will be taping out most of the SiPho chips and making all the way to various engines and modules. Laser moat of the US companies won't last long, it's not that complicated, and more similar to e.g. battery optimization than complex digital circuit design, and Chinese companies already make most of the andriod and lidar vcsel chips, some of them pretty high power and spec. They control most of In and InP production as well. US laser companies will have to and are already building InP fabs in SEA, theres no In or In purification in the US. If I know more chemistry, I would start an ITO recycling company now.

TSMC could create a new “Moore’s Law for CPO,” doubling bandwidth every two years as it accelerates the rollout of co-packaged optics (CPO) to meet the soaring bandwidth requirements of AI data centers, according to a media report. TSMC will focus on three areas to scale CPO: • Per-lane speed: Increase single-channel speeds from 200G today to 400G and beyond • Optical channel density: Expand channel counts from 16 today to 32, then 64 and beyond • WDM (wavelength-division multiplexing): Enable each optical fiber to carry more wavelengths simultaneously The report forecasts that doubling bandwidth every two years could take CPO bandwidth from a 3.2T baseline today to around 410T by 2040 – 128x higher. $TSM

Everything I read out of CIOE this week was a seller talking. $LITE on laser power tiers, $COHR on ultra-high-power ramps, $POET on 6.4T, Sivers on price rises. So I went looking for a buyer, and found one who had already answered the central question in public, in May, and answered it the other way. The sellers' position, from Lumentum management at a broker conference on 9 September: "There's a great concern what happens to the laser when it's so close to the heat source … there's a strong desire to move the laser outside." Two reasons for keeping the light source external. Reliability, and heat. Chen Qin, optical network architect at Alibaba Cloud, speaking at an xPO forum in Shanghai on 28 May, in remarks the CIOE organisers published themselves: deployment of 400G and 800G optical modules, now in the tens of millions, has thoroughly validated laser reliability. The spread of liquid cooling has effectively resolved the thermal bottleneck. NPO and CPO, in his words, now fully possess the objective conditions for putting the laser inside. He then called on the industry to push laser integration forward together. Translation mine. Both of the seller's reasons, marked settled by the customer, fifteen weeks earlier. That is the part worth sitting with, because the external light source story is load-bearing right now. It is the reason indium phosphide capacity is being prepaid years out, the reason a 70 mW part and a 400 mW part are suddenly different businesses, the reason anyone is modelling ELS modules as a separate line at all. His architecture choice is more specific than the headline. Scale-out keeps pluggables, with OSFP iterating. Scale-up moves to near-package or co-packaged, and at 224G per lane and below he says NPO has ample headroom and can fully reuse the existing supply chain. CPO he calls the long-term direction, with its current obstacles named as immaturity and single-vendor lock-in. That last phrase explains something I wrote about earlier this week. If the buyer's stated problem with co-packaged optics is being locked to one vendor, then a group of suppliers writing a common pluggable socket for optical engines is not a standards exercise. It is the answer to a customer objection. And the dates do not line up the way the market has them. Alibaba Cloud lit the first 3.2T NPO system to the OIF standard at the end of 2025, plans small-volume deployment at the end of this year and volume deployment next year. Its 6.4T UPO work expects system bring-up in the second half of this year and beta by year end. Against that, CPO was dated this month by three separate parties to the end of 2027 at the earliest. Near-package is roughly a year ahead, on the customer's own calendar. What I would not do is over-read one buyer. This is a single hyperscaler, in China, with its own silicon and its own supply constraints, and the US hyperscalers sit inside a different vendor ecosystem. It was also May, and plans stated in May are not deliveries. But nobody has publicly walked it back, and the supplier arguments this week were made as though the objection had never been raised. What would change my read: a Western hyperscaler putting a comparable schedule on record, either way. Also worth watching whether VCSEL-based designs gain ground, since he specifically noted they support an internal laser and need no polarisation-maintaining fibre, which removes two of the cost lines the external route carries. Friday's close: $LITE 931.68, $COHR 305.21, $POET 8.01, $AAOI 105.89, $AXTI 66.06.