Fujifilm (4901.JP) expects CMP slurry use to intensify in advanced semiconductor manufacturing, directly benefiting suppliers of abrasive grains (FUSO Chemical, 4368.JP) and polishing pads (Fujibo Holdings, 3104.JP). Fujifilm leads the copper interconnect CMP slurry market and is the second-largest slurry supplier globally. It works closely with FUSO to develop CMP slurries using ultra-high-purity colloidal silica for polishing wiring layers on silicon wafers. Fujifilm sees several structural drivers behind accelerating slurry demand: (1) increasing layer count in logic and memory chips, (2) proliferation of GAA transistors with complex geometry, (3) adoption of backside power delivery, and (4) transition to more complex bonding processes (hybrid bonding) in chip packaging. The supply can hardly meet this demand, giving the key producers considerable pricing power. I think they will finally use it. This year, FUSO is increasing its silica prices by 5-10% for the first time (to my knowledge). FUSO Chemical is the second-largest position in my portfolio. I see about 70% upside to ¥5,300 per share. I think investors underestimate the silica demand multipliers, operating leverage price hikes, and the land-grab opportunity in China.

$6135.T In a March interview, Makino's sales executive said: “We’ve heard about Starlink-related components, so that might be a positive development for our company as well.” I checked, and he is right - SpaceX likely uses Makino's machines to make Starlink satellites. SpaceX manufactures Starlink satellites at the dedicated facility in Redmond, Washington. GPT 5.5 helped me unearth local county permit records, which show new foundations for relocated machinery and related fire-alarm alterations in early 2025 under the project name "SPACEX SE03 - MAKINO EXPANSION AND HITL". In addition, SpaceX is looking for experienced machinists for the Redmond facility to operate multi-axis CNC machines, including 5-axis mills and 6-axis lathes. This convinced me that Elon is also a fan of Makino's gigantic "mother machines".

After digging around, I finally settled on my favorite Japanese machine tool producer: Makino $6135.T From giga-casting molds and satellite structures to gas turbine cooling holes and vascular stents, Makino’s tools excel in the toughest machining jobs. The failed Nidec acquisition in 2025 and the blocked PE buyout in April revealed the company’s strategic value. The Japanese government blocked MBK Partners’ acquisition of Makino on national security grounds, arguing that its machines “are widely utilized by manufacturers of defense equipment in Japan.” I found several examples of Makino machines inside defense manufacturing ecosystems across Japan, Europe, and the US. They are used to make flight-critical structures and propulsion components for F-35 and Typhoon fighter jets, CH-47 and OH-1 helicopter engines, and gas turbines of Japanese destroyers. Beyond defense, management is also targeting aerospace, energy, semiconductor, and medical markets, where customers need high-precision machining of difficult materials and are willing to pay up for productivity, accuracy, and process know-how. Key applications are aircraft engine parts, gas turbines, semiconductor equipment, and medical devices. Makino’s opportunity in the emerging space economy deserves a separate post. The US satellite producer Astranis uses Makino MAG3.EX to machine full satellite panels in-house (video linked below). I’m also fairly confident that Makino machines are integrated into the supply chains of SpaceX and Blue Origin. After the acquisition fiasco, management is under pressure to unlock Makino’s true potential. The “Corporate Value Enhancement” plan announced in May focuses on larger and more complex machines, pricing discipline, shorter lead times, higher-margin engineering and aftermarket services, dividends, and buybacks. The plan is good, but I think it undersells the opportunity. I assume the company will achieve at least 7% revenue CAGR and reach 15% operating margin by 2030. That gets me to ¥21,500 per share or ~50% upside. At ~13 P/E, Makino looks too cheap for its strategic importance, technical depth, and end-market optionality. I took an initial position this week and will be watching for the buyback announcement and continued order build-up.

$6135.T In a March interview, Makino's sales executive said: “We’ve heard about Starlink-related components, so that might be a positive development for our company as well.” I checked, and he is right - SpaceX likely uses Makino's machines to make Starlink satellites. SpaceX manufactures Starlink satellites at the dedicated facility in Redmond, Washington. GPT 5.5 helped me unearth local county permit records, which show new foundations for relocated machinery and related fire-alarm alterations in early 2025 under the project name "SPACEX SE03 - MAKINO EXPANSION AND HITL". In addition, SpaceX is looking for experienced machinists for the Redmond facility to operate multi-axis CNC machines, including 5-axis mills and 6-axis lathes. This convinced me that Elon is also a fan of Makino's gigantic "mother machines".

The Makino MAG3.EX (top right image) is a beast. 130 kW spindle. 33,000 rpm. Can remove 75 pounds of aluminum per minute. 5 micron tolerance (!). 1,600 sq ft footprint. And it lets us make a part in a single day that would take a *year* to get from an outside shop.

After digging around, I finally settled on my favorite Japanese machine tool producer: Makino $6135.T From giga-casting molds and satellite structures to gas turbine cooling holes and vascular stents, Makino’s tools excel in the toughest machining jobs. The failed Nidec acquisition in 2025 and the blocked PE buyout in April revealed the company’s strategic value. The Japanese government blocked MBK Partners’ acquisition of Makino on national security grounds, arguing that its machines “are widely utilized by manufacturers of defense equipment in Japan.” I found several examples of Makino machines inside defense manufacturing ecosystems across Japan, Europe, and the US. They are used to make flight-critical structures and propulsion components for F-35 and Typhoon fighter jets, CH-47 and OH-1 helicopter engines, and gas turbines of Japanese destroyers. Beyond defense, management is also targeting aerospace, energy, semiconductor, and medical markets, where customers need high-precision machining of difficult materials and are willing to pay up for productivity, accuracy, and process know-how. Key applications are aircraft engine parts, gas turbines, semiconductor equipment, and medical devices. Makino’s opportunity in the emerging space economy deserves a separate post. The US satellite producer Astranis uses Makino MAG3.EX to machine full satellite panels in-house (video linked below). I’m also fairly confident that Makino machines are integrated into the supply chains of SpaceX and Blue Origin. After the acquisition fiasco, management is under pressure to unlock Makino’s true potential. The “Corporate Value Enhancement” plan announced in May focuses on larger and more complex machines, pricing discipline, shorter lead times, higher-margin engineering and aftermarket services, dividends, and buybacks. The plan is good, but I think it undersells the opportunity. I assume the company will achieve at least 7% revenue CAGR and reach 15% operating margin by 2030. That gets me to ¥21,500 per share or ~50% upside. At ~13 P/E, Makino looks too cheap for its strategic importance, technical depth, and end-market optionality. I took an initial position this week and will be watching for the buyback announcement and continued order build-up.

Using proceeds from the Santec $6777.T sale, I built up a sizable position in GLTechno $255A.T GLTechno has two core businesses: (1) analytical instruments and (2) semiconductor equipment parts. (1) The analytical instruments business, GL Sciences, is a steadily growing cash cow, selling chromatographs and related consumables to chemical producers, government institutions, pharma/biotech firms, and food manufacturers. (2) The semiconductor business, Techno Quartz, is the main earnings driver and thesis anchor. Techno Quartz manufactures high-precision quartz glass and crystalline silicon components for semiconductor manufacturing equipment primarily used in etching and deposition processes. Applied Materials $AMAT has historically been the largest client, followed by Tokyo Electron $8035.T and Kokusai Electric $6525.T Importantly, repeat purchases of replacement parts account for ~70% of total sales. The case is straightforward: overwhelming compute demand -> rising foundry utilization rates and expanding production capacity -> accelerated depreciation of installed equipment and increasing orders for new tools -> greater demand for Techno Quartz parts. This demand is already evident in the company’s backlog: semiconductor orders increased 42% QoQ in the last quarter, after a 27% increase in the previous one. The key clients of Techno Quartz are all guiding for 20-30% growth in new equipment sales and 15% increase in service revenues this year, with demand visibility extending beyond 2027. Just in time for this, Techno Quartz's production capacity is set to expand 30% by January 2027. At 14 P/E, I think GLTechno is too cheap. I see about 60% upside to ¥9,500 per share in the next 18 months.

Increased my position in FUSO Chemical (4368.JP) by 25% today. For those unfamiliar with the company, FUSO leads the market for ultra-high-purity silica particles used in chemical slurries to polish silicon wafers. After updating the valuation model, I see about 80% upside to ¥16,000 per share. I don’t think investors fully recognize structural changes in FUSO’s electronic materials business and its impact on consolidated figures. I expect this segment to reach 50% operating margins over the next two years, as the recently built silica reactors expand output and management raises prices amid tight supply. The new medium-term plan in May should provide more legs to this story. I’m surprised FUSO hasn’t been discussed much here, despite its vital role in advanced-node polishing processes. Anyway, the numbers will ultimately do the talking.

Sold 60% of Santec $6777.T position today, locking a 700% gain. I invested 30% of my portfolio in Santec during 2023-2024 at ¥4,015, before optics exploded. This proven compounder was stupid cheap under 12 P/E. At ~50, the setup is very different. I think the current valuation fully discounts industry demand for transceivers and fiber cable testing. The upside should come from emerging CPO applications. If Santec successfully integrates its optical instruments into high-volume SiPh/CPO manufacturing, it will likely compound for years. I believe management has a fair shot, but it's not worth risking 50% of my capital.

Very strong execution by Santec $6777.T: sales rose 53% and operating income jumped 74% YoY in Q4. This was driven entirely by the optical instrument business and demand for fiber cable testing in North America. I hope to see some clues about CPO in the management briefing.

Santec Holdings ($6777.JP). Cheap, highly profitable, and rapidly growing producer of optical instruments and components that dominates several niche optoelectronic markets. The company has also successfully expanded into medical and semiconductor devices in the past 10 years.