<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Ynot Photonics — Insights</title><description>How the optical interconnect of the AI data center actually works: the fundamentals, the arithmetic, and what is changing. Every figure sourced.</description><link>https://ynotphotonics.com/</link><language>en-us</language><item><title>A link budget looking for a link.</title><link>https://ynotphotonics.com/insights/a-link-budget-looking-for-a-link/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/a-link-budget-looking-for-a-link/</guid><description>Coherent Lite is dual-polarization 16QAM on a fixed O-band laser, for 2 to 20 km inside a campus. Its estimated power is within the range of OSFP-family 1.6T modules, at about 1.6 to 1.9 times the power of IM-DD. Forecasts span ten times, and nobody placed a volume order. Its market is the set of links where IM-DD runs out of budget, and at 1.6T that set was almost empty.</description><pubDate>Sun, 18 Oct 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>A second source is a new part.</title><link>https://ynotphotonics.com/insights/a-second-source-is-a-new-part/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/a-second-source-is-a-new-part/</guid><description>The first clause of JESD264 says who the silicon photonics qualification standard binds, which events restart a qualification, and what it leaves out. Read together, they tie a qualification to a whole supply configuration, not to a design. A second foundry is therefore a new qualification, with a schedule measured in months.</description><pubDate>Sun, 18 Oct 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>The test menu is public. The results are not.</title><link>https://ynotphotonics.com/insights/the-test-menu-is-public-the-results-are-not/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/the-test-menu-is-public-the-results-are-not/</guid><description>On 1 October 2026 JEDEC published JESD264, the first industry-wide qualification standard for silicon photonics. It tells a foundry what to test. It cannot show which platforms passed, and today almost no foundry publishes that. This essay reads the standard, does the statistics on its sample plans, and follows shipping products back to the fabs that could be asked.</description><pubDate>Sun, 11 Oct 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>Four of these are distances. One is a direction.</title><link>https://ynotphotonics.com/insights/four-of-these-are-distances/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/four-of-these-are-distances/</guid><description>Scale-inside, scale-up, scale-out, scale-across, scale-in, scale-beyond: these six terms appear on one slide as if they were one ladder. Four name a distance that a network must cross. The other two are not distances: one is a direction, and one has no published definition. Sorting them out explains most of what an AI fabric is.</description><pubDate>Wed, 16 Sep 2026 00:00:00 GMT</pubDate><category>Fundamentals</category></item><item><title>The build-out runs on three clocks</title><link>https://ynotphotonics.com/insights/the-build-out-runs-on-three-clocks/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/the-build-out-runs-on-three-clocks/</guid><description>Who is actually fast? Every hyperscaler build now advances on three separate schedules: power, silicon and network. No builder leads all three. This is a scoreboard of who energizes gigawatts, who gets Rubin first, and who still buys 400G while a rival deploys 1.6T.</description><pubDate>Fri, 11 Sep 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>The DSP picks the modulator</title><link>https://ynotphotonics.com/insights/the-dsp-picks-the-modulator/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/the-dsp-picks-the-modulator/</guid><description>The first 400G-per-lane optical DSP arrived at OFC 2026 with two lines in its product bullets. One is an integrated laser driver. The other is proven interoperability with the vendor&apos;s own electro-absorption modulator. Together, those lines make a design decision for the module maker. The reason is the analog bandwidth of a silicon Mach-Zehnder.</description><pubDate>Wed, 02 Sep 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>The wafer under the wafer</title><link>https://ynotphotonics.com/insights/the-wafer-under-the-wafer/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/the-wafer-under-the-wafer/</guid><description>Every silicon photonics chip starts on a specialty substrate, and only one company documents shipping it at 300 mm into a qualified production line. Soitec discloses the revenue but not the wafers. This essay counts them from the outside, removes the one public wafer count that is circular, and finds that capacity is not the constraint.</description><pubDate>Wed, 19 Aug 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>How many optical modules does AI actually need?</title><link>https://ynotphotonics.com/insights/how-many-optical-modules-does-ai-actually-need/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/how-many-optical-modules-does-ai-actually-need/</guid><description>Published 2026 forecasts for 800G module shipments are between 33.5 million and 63 million, a factor of two for the same product and year. This essay does not select one. It counts the market in three independent ways, then follows the winning technology down to wafers and substrates. The 2025 results are now available, a market model is now in a regulated filing, and one widely cited wafer count is circular.</description><pubDate>Wed, 05 Aug 2026 00:00:00 GMT</pubDate><category>How to calculate</category></item><item><title>Trust the FEC, not the thermometer</title><link>https://ynotphotonics.com/insights/trust-the-fec-not-the-thermometer/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/trust-the-fec-not-the-thermometer/</guid><description>An optical module measures its laser temperature, bias current, output power and modulator bias point, and reports all of it to the network. Almost none of it is accurate enough to act on. The measuring chip stopped being a dedicated part years ago. The number that operators actually use comes from the FEC.</description><pubDate>Wed, 22 Jul 2026 00:00:00 GMT</pubDate><category>Fundamentals</category></item><item><title>What a gigawatt of AI data center actually costs</title><link>https://ynotphotonics.com/insights/what-a-gigawatt-of-ai-data-center-costs/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/what-a-gigawatt-of-ai-data-center-costs/</guid><description>Four respected research houses priced the same one-gigawatt build between $35 billion and $60 billion. An average is tempting. It would remove exactly the information that the spread contains.</description><pubDate>Wed, 08 Jul 2026 00:00:00 GMT</pubDate><category>How to calculate</category></item><item><title>CapEx, OpEx and the depreciation argument</title><link>https://ynotphotonics.com/insights/capex-opex-depreciation-payback/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/capex-opex-depreciation-payback/</guid><description>An AI cluster with a tenant can show a 98 percent margin while its owner reports a multi-billion-dollar loss. Both statements can be true. The most important number in AI economics is not a measurement. Management declares it.</description><pubDate>Wed, 17 Jun 2026 00:00:00 GMT</pubDate><category>How to calculate</category></item><item><title>A thousand gigahertz is not a product</title><link>https://ynotphotonics.com/insights/a-thousand-gigahertz-is-not-a-product/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/a-thousand-gigahertz-is-not-a-product/</guid><description>In one postdeadline session at OFC 2026, five different modulator materials reached 400 gigabits on a single lane. So bandwidth no longer decides the 3.2T generation. The rest of the datasheet does. A census of nine routes shows that the leaderboard everybody quotes is almost the reverse of the one that matters.</description><pubDate>Wed, 27 May 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>Who makes what, and what you can actually second-source</title><link>https://ynotphotonics.com/insights/who-makes-what-multi-sourcing-map/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/who-makes-what-multi-sourcing-map/</guid><description>When an AI build is a year late, the review rarely finds a GPU problem. It finds a transformer. A supply chain can be mapped in dollars or in lead times, and almost everyone plans with the dollar map.</description><pubDate>Wed, 13 May 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>Reading the XPO MSA, the pluggable fights back</title><link>https://ynotphotonics.com/insights/xpo-msa-the-pluggable-fights-back/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/xpo-msa-the-pluggable-fights-back/</guid><description>Many analysts expected co-packaged optics to replace the pluggable module. Then Arista and its partners announced a 12.8-terabit, liquid-cooled, 400-watt pluggable that puts a whole switch generation on one rack unit of faceplate. The arithmetic shows why the pluggable stays.</description><pubDate>Tue, 28 Apr 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>Reading the OpenCPX MSA, the socket that makes co-packaging serviceable</title><link>https://ynotphotonics.com/insights/opencpx-msa-the-socket-that-makes-cpo-serviceable/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/opencpx-msa-the-socket-that-makes-cpo-serviceable/</guid><description>The least visible document of OFC 2026 may have the largest effect. OpenCPX standardizes a connector. That connector decides whether an operations team accepts co-packaged optics for production, or keeps it as a demonstration.</description><pubDate>Tue, 21 Apr 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>Reading the OCI MSA, where scale-up optics chose NRZ</title><link>https://ynotphotonics.com/insights/oci-msa-nrz-and-the-scale-up-fabric/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/oci-msa-nrz-and-the-scale-up-fabric/</guid><description>Six of the most powerful companies in computing formed a consortium to define optical scale-up links. Their first technical decision was to replace PAM4 with NRZ, the simplest modulation there is. That choice explains the whole specification, and a short calculation predicts it.</description><pubDate>Tue, 14 Apr 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>The placement ladder, or what LPO, LRO, NPO, CPO and OIO actually mean</title><link>https://ynotphotonics.com/insights/placement-ladder-pluggable-lpo-lro-npo-cpo-oio/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/placement-ladder-pluggable-lpo-lro-npo-cpo-oio/</guid><description>Pluggable, LPO, LRO, NPO, CPO and OIO are six answers to one question: how close to the silicon should the optics sit? Each step closer saves energy per bit and makes a repair harder.</description><pubDate>Tue, 07 Apr 2026 00:00:00 GMT</pubDate><category>Fundamentals</category></item><item><title>Silicon has no Pockels effect, and everything about modulator materials follows from that</title><link>https://ynotphotonics.com/insights/modulator-materials-lithium-niobate-and-the-alternatives/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/modulator-materials-lithium-niobate-and-the-alternatives/</guid><description>Silicon, the most manufacturable material in history, has a crystal symmetry that gives it no linear electro-optic effect. So silicon modulates light by injecting or removing carriers, with side effects. The materials that avoid this show why the best physics rarely wins, and why it does not need to.</description><pubDate>Thu, 19 Feb 2026 00:00:00 GMT</pubDate><category>Fundamentals</category></item><item><title>Copper is out of bandwidth. Optics is out of SNR. That asymmetry is the roadmap.</title><link>https://ynotphotonics.com/insights/copper-out-of-bandwidth-optics-out-of-snr/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/copper-out-of-bandwidth-optics-out-of-snr/</guid><description>The SerDes world seriously debates PAM6 and PAM8. The optics world will not use either, and adds lanes instead. Both are right, and the reason explains most of the interconnect roadmap.</description><pubDate>Thu, 29 Jan 2026 00:00:00 GMT</pubDate><category>Fundamentals</category></item><item><title>Retrofits, mining sites, and whether a new build is obsolete on day one</title><link>https://ynotphotonics.com/insights/retrofits-and-the-obsolescence-question/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/retrofits-and-the-obsolescence-question/</guid><description>Experienced buyers pay nearly new-build prices to convert bitcoin mines that cost a tenth as much to build. They know what they are doing. They buy the only part of a data center that a factory cannot make.</description><pubDate>Thu, 15 Jan 2026 00:00:00 GMT</pubDate><category>Industry watch</category></item><item><title>Inside an optical module, where the watts and the dollars go</title><link>https://ynotphotonics.com/insights/inside-an-optical-module-bom-power-supply-chain/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/inside-an-optical-module-bom-power-supply-chain/</guid><description>An 800G module that fits in your hand has a bill of materials over a thousand dollars and a power budget over ten watts. One chip inside it takes a third of the first and half of the second. That chip explains how the module economy works.</description><pubDate>Fri, 19 Dec 2025 00:00:00 GMT</pubDate><category>Fundamentals</category></item><item><title>Why AI capacity is counted in gigawatts</title><link>https://ynotphotonics.com/insights/why-ai-capacity-is-measured-in-gigawatts/</link><guid isPermaLink="true">https://ynotphotonics.com/insights/why-ai-capacity-is-measured-in-gigawatts/</guid><description>The industry measures computing buildings in units of electricity, and nobody finds this strange now. Three physical limits explain how that happened. The third one is why optics now sits in the critical path.</description><pubDate>Mon, 08 Dec 2025 00:00:00 GMT</pubDate><category>Fundamentals</category></item></channel></rss>