Radiant Opto-Electronics, a major backlight module maker, is expanding into new businesses and targeting next-generation optical communications architecture, particularly co-packaged optics (CPO) applications above 6.4T. Chairman and General Manager Yu-Chao Wang said the company plans to provide samples to selected customers in 2027 and could enter mass production as early as 2028. The plan remains at the planning stage; the 2027 sample provision and 2028 mass production are stated plans/targets, not achieved results.
Murata plans to invest about JPY 1 trillion (about EUR 5.7 billion) over five years, from fiscal 2026 to fiscal 2030, to expand its global production capacity for multilayer ceramic capacitors (MLCCs), and expects MLCC sales in fiscal 2030 to be at least double those of fiscal 2025. According to Murata's press release, roughly JPY 350 billion is for buildings and JPY 650 billion for production equipment, directed at new production buildings at existing sites in and outside Japan, new equipment, upgrades to existing lines and some land purchases. Murata says the expansion is aimed at growing demand from data centres in the near term, while also building the production base it will need after 2030, when it expects AI to have spread into edge devices, mobility and robotics. The largest project disclosed so far is in Okayama Prefecture: Okayama Murata Manufacturing in Setouchi will become the company's third MLCC production site in Japan, alongside Fukui Murata Manufacturing and Izumo Murata Manufacturing. Murata will put about JPY 150 billion (about EUR 850 million) into two new buildings there — an MLCC production building with construction starting in April 2027 and completion planned for the end of 2028, and a building for MLCC sheets and raw materials due in 2030 — and aims to employ 1,500–2,000 people in Okayama by then. Izumo Murata will also build a new plant for electronic components, including MLCCs, in Yasugi, Shimane Prefecture, though neither the timing nor the investment has been decided. Outside Japan, the plan includes a new production building at Murata Electronics (Thailand) due for completion in 2030 and new equipment and infrastructure in an unused part of an existing building in the Philippines by 2027, with investment amounts for both to be announced later; Murata will also add equipment and capacity yearly at existing plants in Japan, ASEAN and China without disclosing individual plans. Murata has not said how the remaining roughly JPY 850 billion will be split between its sites.
Shin Seung-cheol, vice president of Samsung Electronics' System LSI Business under its Device Solutions (DS) division, said during the company's third-quarter earnings conference call on Oct. 29 that the Exynos 2700 has improved AI and gaming performance and that development and mass production are proceeding without disruption, with Samsung expecting to further expand market share by continuously strengthening SoC competitiveness. According to Counterpoint Research, Samsung Electronics held a 9% share of the global smartphone AP market by shipment volume in the second quarter of this year, up 2 percentage points from 7% in the previous quarter and up 3 percentage points from a year earlier; the Exynos 2600, used in the Galaxy S26 series, drove the increase in shipments. The Exynos 2700, which recently entered mass production, has increased production volume 10% compared with the previous generation and has been confirmed for use in the new Pro model of the Galaxy S27 series, scheduled for release next year, in addition to the standard and Plus models for South Korea and Europe. Samsung's System LSI Business has also secured a new custom SoC project by establishing a business partnership with a new customer. The source does not state the cause.
Japanese semiconductor package substrate maker Ibiden plans to invest 1.28 trillion won to expand production capacity for high-performance semiconductor package substrates used in AI servers, aiming to meet growing demand for CPU package substrates. Ibiden's board of directors decided on Oct. 29 to expand capacity for high-performance IC package substrates for CPUs at Building No. 6 of its Gama plant in Gifu Prefecture, Japan; the new equipment will begin operations in phases from fiscal 2029. The company also plans to invest in overseas production facilities, but did not disclose the scale of the planned capacity increase. Ibiden expects CPU substrate demand to increase as AI inference workloads grow and agentic AI becomes more widespread. On Oct. 29, Ibiden also announced first-half fiscal 2026 results for the April-September period: revenue totaled 262.546 billion yen, up 34.3% year-on-year, and operating profit reached 60.116 billion yen, up 84.6% year-on-year. The electronics business, which produces semiconductor package substrates, drove the improvement, with segment revenue up 52% year-on-year to 172.54 billion yen and operating profit up 96.7% to 50.028 billion yen, reflecting stronger-than-expected orders for high-value-added products used in AI servers. Ibiden also raised its full-year fiscal 2026 forecast: revenue outlook increased by 2.7% to 565 billion yen from 550 billion yen, and operating profit forecast lifted by 4.7% to 133 billion yen from 127 billion yen, with expectations of continued solid demand for AI and general-purpose server products and growing demand for switch IC products.
high-performance semiconductor package substrates for CPUsIbidenThe ElecCapacity Change
Delta Electronics approved plans on October 28 to establish a large manufacturing base in Penang, Malaysia, as the power and thermal management company expands overseas production to support customer demand. The site is expected to ramp up to capacity comparable to that of Delta's Thailand operations over five years, with space reserved for process research and development and a potential future R&D center. The stated purpose is to support customer demand. The event is an approved plan and has not yet been implemented.
According to a TrendForce repost, Samsung's foundry business may finally be turning a corner, but the claims are expectations/plans rather than achieved results: its 2nm projects are expected to more than double YoY in 2026; the second-gen 2nm Exynos 2700 is gearing up for mass production in 4Q; and a related report says Samsung's 8nm-and-below utilization is seen topping 100% in 2027. The source does not state the cause.
2nm8nm-and-belowSamsungTrendForce Official (Market Data)Capacity Change
According to Financial News, Samsung expects demand for its 8nm-and-below processes to exceed capacity, with utilization projected to surpass 100% in 2027; per Yonhap News, the number of 2nm projects secured in 2026 is expected to more than double YoY. According to Yonhap News, Samsung secured additional 2nm projects from major cloud service providers and AI/HPC customers in 3Q26, building on wins in the previous quarter. Per Seoul Economic Daily, Samsung plans to begin mass production of its second-generation 2nm mobile chip, the Exynos 2700, in 4Q26, while ramping up 4nm NPUs for a major customer. Per Central News Agency, Taiwanese ASIC firm Faraday said on October 27 that it had secured a U.S. cloud AI chip project using 2nm gate-all-around (GAA) technology, valued at tens of millions of dollars; Samsung will handle wafer fabrication, with Taiwanese foundry UMC and memory manufacturers also participating. Per Yonhap News, Samsung expects HPC to account for over 50% of its foundry revenue in 2027; as background, TSMC's HPC segment contributed 68% of total revenue in 3Q26. With advanced-node demand outpacing capacity expansion, Samsung is considering investments in new fabs in the U.S. and South Korea; per Green Economic Daily, Samsung plans to continue expanding 8nm, 4nm, and 2nm capacity, and the first Taylor foundry fab in the U.S. remains on schedule, with cleanroom operations underway and equipment installation and wafer runs in progress; per The Guru, the company is reportedly preparing to break ground on its second Taylor fab in November, with mass production targeted for 2030. Samsung expects double-digit foundry revenue growth and further earnings improvement in 4Q26. Per Digital Today citing Kiwoom Securities, the combined operating loss of Samsung's foundry and System LSI businesses is projected to shrink from KRW 6.742 trillion in 2025 to KRW 3.921 trillion in 2026, with quarterly losses estimated to have eased to KRW 777 billion in 3Q26.
8nm-and-below processes2nm4nmSamsung, Faraday Technology and 4 totalTrendForceCapacity Change
ADS-TEC Energy expects to receive a building permit shortly for a planned battery storage facility at the Pulverdingen substation in Baden-Württemberg, Germany. The facility is planned with an output of around 1 GW and a storage capacity of 4 GWh, which would make it one of the largest battery storage systems in Europe. The permitting authority has confirmed the application is complete, according to a press release from ADS-TEC Energy. The permit is a key requirement for concluding financing and partnership agreements, and grid connection is targeted for 2029. ADS-TEC did not give a project cost figure but says facilities of this size typically cost from the high hundreds of millions of euros up to around EUR 1 billion. The batteries will run on ADS-TEC's own battery management system, with in-house developed electronics and software of more than 15 years; German companies will supply the batteries, switchgear, transformers and extra-high-voltage equipment, but none are named. ADS-TEC also plans to sell a majority stake in the project company, preferably to a European energy company, and is in talks with international infrastructure investors, while remaining a long-term partner providing technical services over the plant's lifetime.
battery storage systembattery management system (BMS)batteriesADS-TEC EnergyEvertiqCapacity Change
ESCATEC has completed a new controlled production room at its ESCATEC Mechatronics Ltd (EUK) site in Lutterworth, UK, intended to strengthen the plant's ability to take on manufacturing for medical devices, healthcare and other highly regulated programmes. According to an ESCATEC press release, the 100-square-metre room has real-time temperature and humidity monitoring, ESD-safe flooring and controlled lighting, and is designed for manufacturing where contamination control, traceability, process consistency and product quality are critical. The room is the third investment this year aimed at medical manufacturing at the UK subsidiary ESCATEC Mechatronics: in June the site was certified to ISO 13485, the quality management standard for medical devices, and it has since invested in new SMT equipment for high-mix, high-complexity production. ESCATEC CEO Charles-Alexandre Albin said that, together with the ISO 13485 certification and the recent SMT investment, it strengthens EUK's ability to support medical device, healthcare and other high-reliability manufacturing programmes. ESCATEC is headquartered in Penang, Malaysia, with production in Malaysia, the Czech Republic, the UK, Bulgaria and Switzerland. The company has not disclosed the size of the investment in Lutterworth.
controlled production room for medical manufacturingSMT equipmentESCATEC, ESCATEC Mechatronics Ltd (EUK)EvertiqCapacity Change
Samsung Electronics and SK Hynix are undertaking their largest-ever equipment investments in 2026, with the investment amount stated as KRW110 trillion in the title, but the memory supply increase is not expected to materialize until 2028. By then, new South Korean plants and expanded capacity at China's ChangXin Memory Technologies (CXMT) are expected to come online simultaneously, potentially triggering price competition in commodity DRAM. The source does not state the cause of the delay in the supply increase.
memorycommodity DRAMSamsung Electronics, SK Hynix and 3 totalDigiTimes AsiaCapacity Change
Related manufacturers
Samsung Electronics, SK Hynix, ChangXin Memory Technologies (CXMT)
On its third-quarter earnings call held on October 28, LG Electronics added physical milestones to its two main AI bets; these remain plans and were not stated as implemented. In data center cooling, the company named a US production site and new Korean lines, and listed the Nvidia qualifications its liquid-cooling units have won. In robotics, it outlined an actuator production ramp, a robot-data hub opening this quarter, and a plan to enter the home humanoid market around 2030. The company did not provide a revenue target for either business.
Tesla and SpaceX are building Terafab in Texas, with Musk aiming to establish a facility capable of processing 1 million wafers per month within 10 years. Intel has joined as a technology partner, and TSMC has confirmed discussions on potential cooperation. Procurement officials from Tesla and SpaceX visited Korean suppliers at SEMICON West. Whether Terafab will materialize as planned remains uncertain.
According to an exclusive Reuters report, CXMT is expected to start mass production of its first-ever HBM memory in the HBM3E form factor within the next two months, meaning it will officially start by the end of the year. CXMT began risk production of HBM3E memory in August, but the company has struggled with the yield of HBM3E memory modules, and this situation remains similar today, according to the latest report. The yield challenge stems from the immaturity of the through-silicon via (TSV) technology it employs: front-end manufacturing achieves around 30% yields, while back-end processing adds about 70% on top of the original 30%. Data suggests CXMT is attempting to achieve HBM3E with only 3,000 TSVs per layer and trying to pack 8 layers into the regular 8-high HBM3E configuration (compared with Samsung's roughly 5,000 TSVs per layer on HBM2 and SK hynix's more than 8,000 TSVs per layer on HBM3). Given these challenges, the company is unlikely to transition to a higher-capacity 12-high HBM in the near term as it addresses engineering issues. JEDEC's official data for HBM3E indicates a 1,024-bit width and speeds from 9.2 Gbps per pin to 12.4 Gbps, with most standard configurations targeting 9.6 Gbps per pin, total bandwidth around 1.2 TB/s, and capacities from 24 GB using 8-high stacks to 36 GB using 12-high stacks of 24 Gb layers. The mass production start is an expectation, not a confirmed completed event.
Unimicron Technology is reshaping its global production footprint and steering resources toward Taiwan's higher-margin AI supply chain. The company's board has approved an additional NT$50.9 billion (US$1.57 billion) in long-lead equipment orders and plans to sell its Japanese subsidiary to PCB maker Meiko Electronics. According to the source, the cause is surging AI and high-performance computing (HPC) chip demand, which drives the need for high-layer-count, large-size IC substrates. Both the equipment orders and the planned sale are board-approved plans; the source does not state that implementation has been completed.
AI substrateshigh-layer-count, large-size IC substratesUnimicron Technology, Meiko ElectronicsDigiTimes AsiaCapacity Change
Samsung Foundry is scheduled to present 8LPU+, the second generation of its 8LPU platform, at IEDM 2026 on December 14. Nintendo Patents Watch has already pegged it as a possible fit for a Switch 2 Lite, but this is speculation for now: Nintendo has not announced a Switch 2 Lite, and Samsung's material does not mention Nintendo or NVIDIA. The link exists because the Switch 2's NVIDIA T239 SoC is built on Samsung's 8 nm node, so a refined version of that process is an obvious candidate for a cheaper handheld-only model. Samsung's IEDM abstract lists 12% higher performance, 15% lower power, and an 8% area shrink over 8LPU, along with roughly 10% smaller I/O area and lower standby leakage. It is a DUV-only node, and Samsung pitches it at automotive, robotics and edge AI chips. Korea's Electronic Times reports the platform should arrive before year's end, and quoted an industry official who named existing 8 nm customers as likely candidates. These are process-level figures, so they do not directly translate into a future Switch 2 Lite that runs 12% faster or lasts 15% longer per charge, and it is unknown how closely NVIDIA's custom 8N variant maps onto 8LPU+.
Samsung 8LPU+Samsung 8LPUSamsung 8 nm nodeSamsung Foundry, Nintendo and 3 totalTechPowerUpCapacity Change
On Oct. 27, Doosan announced it signed a memorandum of understanding (MOU) with Jeonbuk State and Gimje City to establish dedicated copper-clad laminate (CCL) production lines for optical modules used in artificial intelligence (AI) data centers. The project is the first detailed domestic expansion plan under the company's investment program of approximately 970 billion won announced in September. Doosan will invest 399.2 billion won by 2028 to build the plant on a site adjacent to its flexible copper-clad laminate (FCCL) factory in the Gimje Jipyeongseon General Industrial Complex. The new facility will cover approximately 33,000 square meters, create about 200 jobs, and the production lines are scheduled to begin phased operations in the second half of 2028. The expansion is aimed at meeting growing demand for optical module CCL used in AI data centers; Doosan currently supplies CCL for high-speed optical modules capable of transmitting data at 800 gigabits per second (Gbps). No price movement is stated; this is a planned capacity expansion, not yet an operational capacity change.
According to Nikkei, citing industry sources, rising component costs are reportedly prompting leading PC brands including HP, ASUS, and Acer to move part of their production from Thailand and Vietnam back to China, while Dell is also reportedly considering a similar move (reported move, not confirmed as implemented). The stated cause is that surging AI demand drives shortages of memory and storage chips, raising PC bills of materials: per TrendForce, CPU, DRAM, and SSD costs accounted for approximately 68% of a mainstream notebook's BOM in the third quarter of 2026, up from 45% in the first quarter of 2025. The development partially reverses the 'China+1' diversification strategy pursued amid U.S.-China tensions and trade restrictions. TrendForce estimates the share of global notebook production outside China will decline from approximately 24% in 2025 to 21% in 2026, potentially falling below 20% in 2027, and expects global notebook shipments to decline by a low-single-digit percentage in 2027, potentially widening to a high-single-digit rate if elevated DRAM and CPU prices are passed on to consumers. Separately, Quanta reportedly suspended HP notebook production for about a week in September amid a dispute over rising component costs, resuming after further negotiations; servers accounted for over 80% of Quanta's revenue in the second quarter of 2026, and supply chain sources suggest it has gained greater leverage in notebook pricing negotiations. Chinese ODMs (Luxshare, Huaqin Technology, Longcheer Technology) are reportedly offering lower upfront costs, including waiving engineering and tooling fees, and benefit from access to domestically produced memory chips.
memory chipsstorage chipsDRAMHP, ASUS and 11 totalTrendForceCapacity Change
On October 21, during GTC Berlin, Nvidia announced several multi-billion-dollar deals with six major supply chain partners — Ibiden, Unimicron, Samsung Electro-Mechanics, SPIL, Amkor and KYEC — across substrates, advanced packaging, semiconductor testing and optical engines, to expand production capacity and reserve capacity for its AI accelerators; expanded supply is to be available from 2027 through 2028, with commitments and production roadmaps stretching through 2028. For substrates, Nvidia plans to secure 2.5x dedicated substrate capacity over the next three years: it earlier signed a deal to provide a multibillion-dollar capacity reservation advance to support Ibiden's Ono Plant in Japan (targeting 3x the plant's current capacity); the Samsung Electro-Mechanics deal will support facility upgrades and new plants in Busan and Sejong, with Nvidia projected to account for a significant portion of its output by 2028; a capacity reservation agreement with Unimicron supports its substrate expansion in Taiwan, expected to start providing capacity by Q2 2027 and reach full capacity by mid-2028. For co-packaged optics, Nvidia signed with SPIL to expand optical engine production capacity 15x (1,400%) by mid-2027, with plans to double again by the end of 2028. With Amkor, Nvidia targets a more-than-300% increase in interposer wafer capacity by the end of 2028 through advanced packaging expansion in South Korea; Amkor confirmed the deal is worth $1.3 billion, bringing Nvidia's total investment across Amkor's Arizona and South Korea facilities to $3 billion. For testing, on October 13 Nvidia signed a deal to pay KYEC $1.1 billion to reserve testing capacity at KYEC's new $1.4 billion U.S. facility (high-volume production slated to begin in April 2027; Singapore high-volume production expected to begin February 2027), with testing capacity expected to double from calendar 2026 to 2027; KYEC currently handles nearly all of Nvidia's chip testing. Per its August 2026 disclosures, Nvidia had accumulated $279 billion in outstanding supply and manufacturing-capacity commitments by July 2026. All expansion figures are announced plans and targets, not yet realized.
substratesadvanced packaginginterposer wafersNvidia, Ibiden and 7 totalTom's HardwareCapacity Change
Fujifilm Electronic Materials is planning to expand its advanced semiconductor materials capacity in South Korea, specifically increasing extreme ultraviolet (EUV) photoresist capacity in Pyeongtaek and expanding chemical mechanical polishing (CMP) slurry production in Cheonan. The investments will deepen local support for customers including Samsung Electronics, which the source attributes to growing AI and advanced-process demand. This expansion is in a planned status; the source gives no specific investment amount or timeline.
According to ZDNet Korea, reported on October 26, Samsung Electronics and SK Hynix are targeting the second quarter of 2027 to begin volume shipments of eight-layer HBM4 memory to Nvidia, as the AI chipmaker reassesses its memory strategy for next-generation Rubin Ultra accelerators. This is a plan/target and actual shipments have not yet occurred.
8-layer HBM4 memorySamsung Electronics, SK Hynix and 3 totalDigiTimes AsiaCapacity Change