NuAxi Stream - Real-Time Semiconductor Intelligence

Mon, Oct 26

06:30Impact score 45
SK hynix Halts Solidigm IPO Plans
On Oct. 23, SK hynix announced that, after comprehensively considering capital market conditions and other factors, it has decided to discontinue consideration of pre-IPO fundraising and an initial public offering (IPO) for Solidigm, its indirect subsidiary. The company said recent capital market conditions and share price movements had prompted it to reassess whether an IPO would currently be the most advantageous option for SK hynix and its existing shareholders, and stated that the decision does not reflect any change in Solidigm's growth potential or investment needs, while expanding Solidigm's production capacity and strengthening its competitiveness remain important priorities. The announcement came 79 days after reports of a potential listing first emerged on Aug. 5, and a day after financial regulators explicitly classified the proposed listing as subject to dual-listing regulations. Solidigm is a US company established after SK hynix acquired Intel's NAND business in 2021; in January 2026, SK hynix established AI Company as an intermediate holding company in the US and placed Solidigm under it, and AI Company owns 100% of Solidigm. Reuters and Bloomberg subsequently reported that Solidigm was selecting underwriters and pursuing a Nasdaq listing as early as 2027. Shareholders opposed the proposal, arguing a separate listing would remove value already reflected in SK hynix's share price and require profits to be shared with new shareholders; SK hynix shares fell more than 10% on Aug. 6, the day after the initial reports. On Oct. 19, Financial Supervisory Service governor Lee Chan-jin told the National Assembly's Political Affairs Committee that both the Financial Services Commission (FSC) and the FSS considered the proposed listing to fall under dual-listing regulations; on Oct. 22, FSC chairman Lee Eog-weon said Solidigm was subject to dual-listing guidelines and that SK Square and SK hynix, as parent companies, must fulfill five obligations based on their fiduciary duties (assessing the impact on shareholders, establishing shareholder protection measures, communicating with shareholders, obtaining a board resolution for or against the listing, and making disclosures), meaning South Korea's dual-listing regulations apply even to overseas listings.
NANDSK hynix, Solidigm and 3 totalThe Elec

Sun, Oct 25

21:00Impact score 10
CEO Interview with Lisa Spelman of Cornelis
This is an interview with Lisa Spelman, CEO of Cornelis. She brings more than 20 years of experience across AI, cloud, enterprise infrastructure, data center operations, product strategy, and go-to-market leadership; before joining Cornelis she served as Corporate Vice President at Intel, where she led the Xeon Products and Solutions business within the Data Center & AI Group. At Cornelis she focuses on improving system utilization and reducing networking bottlenecks for AI infrastructure. The interview states: Cornelis is working on the third generation of its high-performance networking platform, with more than 50 OEM and ODM design wins; it is deployed in over 500 datacenters, including new large-scale deployments at Lawrence Livermore National Lab and the Texas Advanced Computing Center; and this month it announced the Active Compute Fabric, which adds programmable RISC-V cores at every hop across every NIC and switch to accelerate collectives, KV cache, and MoE routing, built on open standards (UALink and ESUN for scale-up, Ultra Ethernet for scale-out). The source does not state the cause; this is a background interview with no reported price, capacity, or supply-demand change.
SuperNICsswitcheshigh-performance networking platformCornelisSemiWiki
19:00Impact score 55
AMD CEO Lisa Su proposed bringing back the Ryzen 7 5800X3D, set plan in motion before DDR5 shortages
According to Tom's Hardware, published 25 October 2026, AMD Director of Client Product Management Travis Kirsch told the Japanese publication MyNavi Tech+ that AMD CEO Lisa Su was one of the proponents of bringing back the Ryzen 7 5800X3D processor, which first debuted in April 2022. The relaunch was not a simple re-release: the company had to do extensive re-engineering and update its manufacturing process with TSMC to bring it back. The relaunch of the Ryzen 7 5800X3D coincided with the memory chip shortage, with the world suffering from extremely high DDR5 memory prices (DDR5 finally arrived in the middle of 2026), but Kirsch said 'this was planned before changes in the memory market occurred, not as a countermeasure to DDR5 shortages,' calling it 'the final gift to local users who have used our AM4 for nearly a decade, offering systems that use DDR4 the last change to upgrade their PCs.' Kirsch also said AMD is mostly focused on the premium, high-performance gaming market (AMD closed in on Intel at 48.4% CPU share in the October Steam Survey), while the affordable CPU market is no longer as big as it used to be as laptops have taken over that segment.
Ryzen 7 5800X3D CPUAM4 platformDDR4 memoryAMD, TSMCTom's Hardware

Sat, Oct 24

23:30Impact score 30
RT NEWS: #Intel CTO Pushkar Ranade said on the Dwarkesh Podcast that Intel has already caught up with #TSMC if the measure is #technology alone, while...
Intel CTO Pushkar Ranade said on the Dwarkesh Podcast that, if the measure is technology alone, Intel has already caught up with TSMC, while scale will take longer. He stated that global tools allow roughly 2x more leading-edge output, not 10x. This is a podcast statement by an executive; the source does not state the cause, and no specific price, capacity adjustment plan or other action is mentioned.
technologyleading-edge outputIntel, TSMCTrendForce Official (Market Data)

Fri, Oct 23

23:32Impact score 55
Micron's next-generation 3D NAND to double bit density with over 400 layers, offering extreme performance
At an event this week, Micron announced that its next-generation 3D NAND (G11), which it plans to enter mass production in 2H 2027, is set to offer three significant advantages: doubling storage density (approximately twice the bit density of G9, with 3D TLC going from 21 Gb/mm² to 42 Gb/mm²), increasing I/O performance by up to 50% versus the company's current-generation NAND memory (interface speed from 3,600 MT/s to 5,400 MT/s), and reducing data transfer energy consumption per bit by more than 25%. G11 will adopt wafer-on-wafer bonding, manufacturing the NAND array on a new 3D NAND node and the peripheral circuits and I/O on a logic process before bonding them together, and will transition to more than 400 NAND layers; Micron has not disclosed the exact number of active wordline layers or memory decks. Micron executive Scott DeBoer described G11 as the world's most advanced NAND, ramping in the second half of next year. Micron also expects to disclose more G11 technical details in 2027 and, beyond G11, is developing high-mobility materials and denser cell architectures to continue XYZ scaling; specific architectural details, density targets, or production schedules for post-G11 generations have not been disclosed. These are the company's stated plans and expectations; mass production has not yet occurred.
3D NANDG11 3D NANDMicronTom's HardwareCapacity Change
23:21Impact score 58
Solidigm confirms plans to build a fab in the US, abandons IPO plans
In an early Friday announcement dated 23 October 2026, SK hynix said it had decided to abandon plans for a potential initial public offering (IPO) of Solidigm, its U.S.-based enterprise SSD subsidiary, citing unfavorable stock market conditions and semiconductor share price trends. At the same time, SK hynix confirmed it plans to build a new fab for Solidigm in the U.S. to address growing demand for enterprise SSDs. SK hynix stated that although an IPO could have helped finance a new fab in the U.S., its subsidiary's expansion remains a priority; the company has discontinued its review of both pre-IPO financing and an IPO and will continue to evaluate alternative funding mechanisms. The source states that the planned fab's capacity, investment requirements, and construction schedule have not yet been disclosed. Solidigm specializes in high-capacity data center SSDs based on floating gate 3D QLC NAND memory, and its existing manufacturing capacity is barely sufficient to address its projected needs.
enterprise SSD3D QLC NANDSolidigm, SK hynixTom's HardwareCapacity Change
21:47Impact score 0
Darveen Unveils DPC-1210 21.5-Inch Industrial Panel PC with Intel® Core™ Ultra 5 Processors for Edge AI Applications
Darveen launched the DPC-1210 Industrial Panel PC, powered by Intel Core Ultra 5 125H/U series processors (Meteor Lake), for edge AI applications. The product delivers up to 34 TOPS of total platform AI performance and supports AI performance expansion via an M.2 AI acceleration module such as the DEEPX DX-M1M. It features a 21.5-inch projected capacitive touchscreen with Full HD resolution (1920 × 1080) and 450 nits brightness with an anti-glare surface, a fanless copper heat pipe cooling system, I/O including 2x COM, 4x USB 3.0, 2x LAN, 1x HDMI, and 1x DisplayPort, and support for two 2.5-inch SATA 3.0 HDD/SSD drives with RAID 1 mirroring. Darveen, established in 2007, offers vehicle mount computers, industrial panel PCs, industrial displays, embedded box computers, and rugged tablets.
Intel Core Ultra 5 125H/U series processors (Meteor Lake)M.2 AI acceleration module (DEEPX DX-M1M)21.5-inch projected capacitive touchscreenDarveen, IntelEE Journal
12:00Impact score 50
FCC grants Intel a waiver to foreign robot ban to prevent chip disruptions
The US Federal Communications Commission (FCC) granted Intel a waiver from its ban on foreign-made mobile robots, which will allow Intel to update its fleet of autonomous guided vehicles (AGVs). Intel had earlier petitioned the agency for the exemption, saying that not granting it would threaten chip production in the country.
chipsIntelDigiTimes AsiaPolicy/Regulation
11:58Impact score 45
Analysis: Nvidia's Vera CPU takes on Intel and AMD with a platform play
Nvidia has won a CPU supply order from AI startup Modal, challenging Intel and AMD in the server CPU market as it expands beyond GPUs into general-purpose computing. Modal's annual revenue is below US$1 billion, and neither the order's value nor its deployment scale has been disclosed. The deal underscores a broader shift in AI infrastructure competition toward full-stack system integration. The source does not state the cause.
CPUGPUNvidia, Intel and 4 totalDigiTimes AsiaPartnership/M&A
05:41Impact score 60
Intel Fab 62 Moves to Next Phase, Begins Equipment Installation
Intel has begun installing equipment at Fab 62 within its Ocotillo campus in Chandler, Arizona. According to Intel officials, construction and equipment installation are currently underway, with production projected to begin by the end of 2028. Naga Chandrasekaran, Executive Vice President of Intel Foundry and Chief Technology and Operations Officer, stated: 'Today, I am pleased to share that construction and tool installation at Fab 62 are now fully underway, and we expect operations to begin by the end of 2028.' Reaching this stage comes more than a year after Intel had Fab 52 at full production capacity. Fab 62 could support the 18A, 18A-P, and 14A process nodes, however the company has not confirmed which ones; the next-door Fab 52 already manufactures chips on the 18A node, currently producing Panther Lake processors and Clearwater Forest server CPUs using the 18A process. It can be presumed that the new facility could either expand existing capacity or support the next-generation 14A process, but this remains unconfirmed by the company. Intel also reported progress in Ohio, where the first major structural truss has been installed on the first fab in New Albany.
18A18A-P14AIntelTechPowerUpCapacity Change
01:54Impact score 20
Xiaomi Redmi Book 14 2027 With Intel Wildcat Lake CPU and 2.8K Display Confirmed
Xiaomi has officially confirmed details of the 2027 update to its Redmi Book laptop series via its Redmi account on Weibo, announcing an October 27 release date for the Redmi Book 14 and likely the 16 alongside it. Xiaomi confirms that the Redmi Book 14 and 16 will use Intel's Wildcat Lake Core Series 3 processors, with the fine print on marketing images confirming the Core 5 320 for the Redmi Book 14 (previous leaks had tipped the Core 5 320 for the Redmi Book 16). The Redmi Book 14 2027 is confirmed to ship with a 72 Wh battery, which Xiaomi claims will deliver over 20 hours of office use and a claimed 1.26 days of local video playback; it will use LPDDR5X-7467 RAM, with capacities unclear aside from confirmation of a 16 GB version; the display is confirmed as 2.8K, though the panel tech is not mentioned. International launch timing and pricing remain unclear; the outgoing Redmi Book 14 starts at 6,499 CNY with 16 GB of RAM and 512 GB of storage. The source does not state the cause (no price movement or supply event involved).
Intel Wildcat Lake Core Series 3 processorsCore 5 3202.8K displayXiaomi, IntelTechPowerUp

Thu, Oct 22

23:00Impact score 58
Evertiq Expo Warsaw: Top 10 chip companies took two-thirds of a record market
According to Omdia data presented at Evertiq Expo Warsaw on October 22, 2026, the ten largest semiconductor suppliers grew revenue by more than 30% in 2025 (30.4%, to USD 562.1 billion), almost three times as fast as the rest of the industry, and now account for nearly 68% (67.6%) of a record market worth USD 831 billion that grew 23.3%. Nvidia's revenue rose by USD 52.9 billion to USD 150.3 billion (up 54.3%), its share of total semiconductor revenue rose from about 14.5% to just over 18% (18.1%), its lead over second-placed Samsung Electronics nearly tripled from USD 22.3 billion to USD 64.5 billion, and it accounted for roughly a third of the industry's growth. Samsung grew 14.2% to USD 85.8 billion, its share fell from 11.1% to 10.3%, and its lead over SK hynix shrank from USD 27.8 billion to USD 18.5 billion. SK hynix grew 42.3% to USD 67.2 billion and moved into third place; Micron, the fastest grower in the top ten at 56.3% (up USD 16.4 billion to USD 45.6 billion), climbed from seventh to fifth past Broadcom and Qualcomm; Broadcom grew 28.0% to USD 39.5 billion but stayed sixth; Qualcomm grew 11.6% to USD 38.9 billion yet dropped from fifth to seventh with its share falling from 5.2% to 4.7%; AMD grew 33.4% to USD 32.7 billion and stayed ninth; Apple's USD 34.4 billion is a value Omdia assigns to its in-house semiconductor activity, not external chip sales; Intel was the only top-ten company whose revenue fell, declining 3.7% to USD 48.5 billion and dropping from third to fourth, with its share falling from 7.5% to 5.8%. Between 2023 and 2025 total semiconductor revenue grew about 53%, the top ten grew about 90%, all other suppliers roughly 8% combined, and Nvidia plus the three major memory makers (Samsung, SK hynix, Micron) went from 24% of industry revenue in 2023 to 42% in 2025. According to Omdia, the DRAM market nearly tripled in two years, from just over USD 50 billion in 2023 to more than USD 150 billion in 2025, growing more than 50% in 2025 alone; Lino Jeng, Senior Principal Analyst at Omdia, said AI servers require not only more HBM but also significantly more system memory, particularly DDR5; Ewelina Bednarz of Evertiq explained that as manufacturers shift capacity toward HBM and high-capacity server DRAM, supply of other memory types tightens and prices rise beyond HBM as well, and noted the figures measure revenue, not volume, with price and product mix explaining much of the growth. Omdia's September update shows industry revenue reached USD 752 billion in the first half of 2026, about 90% of the full-year 2025 total; second-quarter revenue alone passed USD 425 billion, up more than 31% from the first quarter; memory made up more than half of the market in the second quarter, and DRAM, NAND and NOR all set quarterly revenue records; excluding memory, revenue rose just over 10% from the previous quarter against a historical second-quarter average of just over 3%. The SIA reported that global semiconductor sales passed USD 1 trillion year to date through August 2026, the first time the industry reached that level, with August sales of USD 159.7 billion, up 144.3% year on year and the 18th consecutive month of growth; SIA expects full-year 2026 sales to exceed USD 1.6 trillion; SIA figures are based on World Semiconductor Trade Statistics data and use a different method from Omdia's, so the two sets of numbers are not directly comparable.
semiconductor marketmemoryDRAMNvidia, Samsung Electronics and 10 totalEvertiq
21:00Impact score 60
#Intel’s U.S. fab expansion is moving ahead! Arizona’s Fab 62 has begun equipment installation, targeting late-2028 production for potential 18A or ...
According to a TrendForce post dated October 22, 2026, Intel's U.S. fab expansion is moving ahead: Arizona's Fab 62 has begun equipment installation, targeting late-2028 production for potential 18A or 14A capacity, while Ohio One's first fab begins taking structural shape. This is a planned expansion; the process (18A or 14A) remains potential/undetermined, and the source does not state the cause.
18A14AIntelTrendForce Official (Market Data)Capacity Change
11:30Impact score 58
RT NEWS: [Breaking] Google Reportedly Sticks With Intel EMIB-T Despite 35% Substrate Yields as HumuFish Schedule Tightens Uncertainty is growing over ...
According to industry sources (reportedly, unconfirmed), Google is reportedly sticking with Intel EMIB-T despite substrate yields of only 35%, as the HumuFish variation's schedule tightens. Industry sources point to the HumuFish variation as a key area of concern, with reported yield problems. Uncertainty is growing over potential delays to Google's TPU v9; the source does not state that any delay has occurred and gives no specific timeline, with the reported yield problems serving as the stated cause of concern.
TPU v9HumuFishEMIB-T substrateGoogle, IntelTrendForce Official (Market Data)
11:22Impact score 60
Intel begins Arizona Fab 62 equipment installation, advances Ohio project as AI CPU demand surges
Intel is accelerating its long-delayed US semiconductor manufacturing projects: equipment installation is underway at its Fab 62 facility in Arizona, and construction is progressing at its Ohio site. The source states these developments come as stronger demand for AI data center processors supports the chipmaker's efforts to expand capacity and strengthen its foundry business. The source does not specify expansion volume or production start dates.
AI data center processorsIntelDigiTimes AsiaCapacity Change
09:05Impact score 62
[News] Tesla Reportedly to Invest KRW 1.1T to Secure Samsung Electro-Mechanics Substrate Capacity; NVIDIA Also Commits
According to Maeil Business Newspaper, citing industry sources, Samsung Electro-Mechanics is reportedly set to receive a combined KRW 3.7 trillion in investment from Tesla and NVIDIA to expand advanced chip substrate production; the investment is a reported commitment. The company has signed a long-term flip-chip ball grid array (FC-BGA) substrate supply agreement with Tesla and is expanding dedicated production lines at its Sejong facility. The substrates are expected to be used in AI chips for Tesla's autonomous electric vehicles and Optimus humanoid robots under a five-year agreement running from 2029 through 2033. Tesla will contribute approximately KRW 1.1 trillion toward the expansion in exchange for guaranteed substrate supplies; the funding forms part of Samsung Electro-Mechanics' KRW 4.27 trillion Sejong investment plan disclosed in September. Construction of the new FC-BGA substrate production facilities at the Sejong plant is expected to be completed by the end of 2027, with equipment installation and trial operations scheduled for the first half of 2028, mass production in the second half, and deliveries to Tesla expected to begin in 2029. According to TechNews, NVIDIA is also expanding global substrate capacity through supplier partnerships and aims to increase its total IC substrate capacity to 2.5 times its current level over the next three years: in Japan, NVIDIA has committed billions of dollars in capacity reservation prepayments to Ibiden, whose Ono plant aims to triple production capacity, with additional output expected to begin ramping in the first quarter of 2028; in South Korea, Samsung Electro-Mechanics is expanding its Sejong and Busan facilities, with NVIDIA expected to account for a significant share of its annual package substrate capacity by 2028 — while NVIDIA did not disclose an investment amount, Maeil Business Newspaper reports the commitment is believed to be worth approximately KRW 2.6 trillion; NVIDIA is also supporting Taiwan's Unimicron in expanding its KF substrate capacity through a capacity reservation agreement, with additional output expected from the second quarter of 2027 and full capacity targeted by mid-2028. Separately, Samsung Electro-Mechanics disclosed a KRW 360.5 billion contract on October 19 to supply memory semiconductor package substrates to an undisclosed global customer throughout 2027, with the customer not identified citing business confidentiality, according to Chosun Biz. The broader expansion comes as high-end FC-BGA production remains concentrated among a handful of suppliers, including Samsung Electro-Mechanics, Japan's Ibiden, and Taiwan's Unimicron; industry estimates cited by Maeil Business Newspaper put Ibiden's market share at more than 70%, with Samsung Electro-Mechanics working to narrow the gap.
FC-BGA substratememory semiconductor package substrateKF substrateSamsung Electro-Mechanics, Tesla and 5 totalTrendForceCapacity Change
03:41Impact score 96
[News] PSMC Q3 Gross Margin Hits 48% on Memory Price Gains; Eyes 10–20% Mature-Node Price Hikes in 2027
PSMC's third-quarter revenue rose 23% sequentially and 80% annually to NT$21.3 billion, while net profit surged 98% and 338%, respectively, to NT$6.5 billion, with gross margin climbing 20 percentage points to 48%, driven by rising memory prices and tight foundry capacity. According to Liberty Times, PSMC President Martin Chu expects fourth-quarter revenue to grow by double digits sequentially as ASPs continue to rise; in the third quarter, completed memory wafers had ASPs more than 30% below those entering production during the same period, meaning recent price increases have yet to be fully reflected in revenue, with higher-priced wafers expected to contribute more in the fourth quarter. According to TechNews, global DRAM supply remains tight as major manufacturers shift capacity toward HBM and gradually exit specialty DRAM, keeping PSMC's specialty DRAM foundry prices elevated; PSMC is advancing its 1p DRAM collaboration with Micron, with a PDK due by the end of 2026, equipment arriving in 1Q27, and trial production starting in 2Q27. PSMC notes tight MLC NAND supply is driving demand for high-capacity SLC NAND, with the company shifting its SLC foundry platform to 24nm; AI servers require three to five times as much NOR Flash per system as conventional servers, and PSMC is the only NOR Flash foundry outside China. According to PSMC's earnings presentation, wafer input ASPs for its 8-inch and 12-inch logic foundry services rose 8–10% in the third quarter; with capacity constraints persisting, the company plans further price hikes of 10–20% in 2027 and is negotiating long-term agreements (LTAs) to secure pricing and volumes. PSMC has begun shipping 55nm BCD products with customers seeking three-year contracts; fully utilized 2026 foundry capacity is squeezing smartphone PMIC production, prompting major handset makers to seek 2027 capacity; 80% of 2027 MOSFET and IGBT capacity is already covered by LTAs; GaN products have entered small-volume wafer production pending final customer validation. IPD order visibility extends into 2028 with mass production underway on 8-inch and 12-inch wafers; WoW hybrid bonding small-scale trial production is underway with mass production targeted for 1H27. The Micron PWF project remains on schedule, with pilot line construction expected to finish by the end of 2026 and mass production planned for 4Q27.
memoryspecialty DRAMNAND FlashPowerchip Semiconductor Manufacturing Corporation (PSMC), MicronTrendForcePrice Movement
03:35Impact score 60
Intel installs production tools at Fab 62 in Arizona, provides updates on Ohio chip fab
On 21 October 2026 (Wednesday), Intel announced that it had started installing production tools at its Fab 62 in Arizona, with the fab on track for mass production by late 2028; however, the company has not disclosed when it intends to begin high-volume production, the planned wafer processing capacity, the capital expenditures required to complete the facility, and it has not officially assigned Intel 14A to Fab 62 (Fab 62 remains unassigned, while Fab 52 has been designated for Intel 18A). Per Intel's manufacturing roadmap disclosed earlier this year, Fab 62 will be able to make chips on 18A, 18A-P, and 14A nodes. Intel also confirmed that it has integrated High-NA EUV lithography into high-volume manufacturing (HVM), used to produce select tiles for Panther Lake CPUs at its D1X fab in Oregon; Intel has not explicitly confirmed whether High-NA EUV scanners are currently used to produce the same circuits as Low-NA EUV machines. In addition, construction workers recently installed the first major structural truss at the first fab of the Ohio One complex in New Albany, Ohio, but Intel did not provide an updated construction schedule, disclose when the first fab is expected to begin operations, or reveal which process technology it will use or its estimated capacity. The source does not state the cause (this is a capacity expansion progress update, not a price or disruption event).
High-NA EUV lithographyIntel 18AIntel 18A-PIntelTom's HardwareCapacity Change
03:34Impact score 20
Leak: Intel Core Ultra 9 4995HX Leads Nova Lake Laptop Lineup
Leaker Jaykihn on X says Intel's Core Ultra 9 4995HX looks set to be the fastest Nova Lake chip for laptops, using an 8+16+4+2 layout: 8 P-cores, 16 E-cores, 4 LPE cores, and 2 Xe3 GPU cores, totaling 28 CPU cores—four more than Intel's current Arrow Lake-HX chips, with the extra cores coming from the LPE cluster Arrow Lake-HX lacked. The layout matches the previously reported 8P+16E+4LP ceiling, and the name fits the Core Ultra 4000 branding said to be extending to HX parts. The source explicitly notes Intel hasn't confirmed any of this, and clocks, cache, and power figures are still unknown. The same post lists the Core Ultra 9 4970H as 4+8+4+12, i.e., 16 CPU cores and a 12-core Xe3P iGPU—the same core count as Panther Lake's flagship Core Ultra X9 388H, so any gains should come from the new Coyote Cove P-cores, Arctic Wolf E-cores, and Xe3P graphics rather than extra cores; the 4995HX's tiny iGPU is less of a concern since HX chips almost always pair with a discrete GPU. Jaykihn also put the final 8+16 HX compute tile, built on TSMC N2P, at 14.8 x 6.6 mm, roughly 98 mm². Current reports point to the first Nova Lake-H and HX laptops showing up around CES 2027, though that timing should be treated as tentative.
Nova Lake-H/HX laptop CPUsXe3 GPU coresXe3P iGPUIntel, TSMCTechPowerUp
02:30Impact score 60
[Breaking] Google Reportedly Sticks With Intel EMIB-T Despite 35% Substrate Yields as HumuFish Schedule Tightens
Google reportedly remains committed to Intel's EMIB-T for HumuFish despite poor substrate yields. According to industry sources, EMIB-T substrate yields were around 35% in 3Q26, well short of the 50% target set for year-end; the reported causes are thermal expansion differences between ABF substrates and embedded silicon bridges that make precise alignment difficult during high-temperature processing, and NCF (Non-Conductive Film) applied beneath the bridges' TSVs trapping air bubbles during bonding, potentially weakening adhesion. Despite market rumors of a HumuFish delay and the possibility of a switch to TSMC's CoWoS-L, supply chain checks indicate HumuFish remains '100%' committed to EMIB-T. Intel has adjusted its substrate electroplating process, and suppliers including Ibiden, Unimicron, and Shinko continue working to improve yields. Google's TPU v9 lineup is confirmed to include three variants: Broadcom-designed WhaleFish (expected to adopt TSMC's CoWoS-L, with mass production slated for 2H27), MediaTek-designed HumuFish (set for 1H28 mass production, relying exclusively on Intel's EMIB-T), and TriggerFish (also planned to use EMIB-T, with production targeted for 2H28 or later). Per TechNews citing analyst Ming-Chi Kuo, TriggerFish is positioned as an upgraded TPU v9 with stronger inference capabilities and features three key upgrades over HumuFish: 2–3 times the SRAM capacity, an additional simulation die, and a shift to HBM4E from HumuFish's HBM4. On delays, the tape-out of Google-designed Compute Die A, built on TSMC's N2 process, has been pushed back to late October while other dies remain on schedule, and the delay is expected to push back trial production of the complete package; Commercial Times further notes the MediaTek-involved Google TPU v9 project (HumuFish) could see its tape-out pushed back to January 2027 due to design revisions, potentially delaying shipments until 1Q28. With HumuFish not slated for mass production until 1H28, the supply chain still has time to address EMIB-T's remaining issues, but a tape-out slip to January 2027 could compress that cushion.
Intel EMIB-T substratesABF substratesNCF (Non-Conductive Film)Google, Intel and 8 totalTrendForce

Analog	$0.68	−5.9%
Connector	$0.75	+0.2%
CPU	$386.74	−21.3%
DDR4	$88.40	−0.2%
DDR52026-W13	$14.89	−11.2%
GPU	$478.20	+8.2%
MCU	$0.68	−61.7%Unusual
MLCC	$0.01	+54.0%Unusual
MOSFET	$0.12	−82.8%Unusual
PMIC2026-W13	$0.44	−12.5%
SSD/HDD	$288.43	n/a

NuAxi Stream - Real-Time Semiconductor Intelligence

Mon, Oct 26

SK hynix Halts Solidigm IPO Plans

Sun, Oct 25

CEO Interview with Lisa Spelman of Cornelis

AMD CEO Lisa Su proposed bringing back the Ryzen 7 5800X3D, set plan in motion before DDR5 shortages

Sat, Oct 24

RT NEWS: #Intel CTO Pushkar Ranade said on the Dwarkesh Podcast that Intel has already caught up with #TSMC if the measure is #technology alone, while...

Fri, Oct 23

Micron's next-generation 3D NAND to double bit density with over 400 layers, offering extreme performance

Solidigm confirms plans to build a fab in the US, abandons IPO plans

Darveen Unveils DPC-1210 21.5-Inch Industrial Panel PC with Intel® Core™ Ultra 5 Processors for Edge AI Applications

FCC grants Intel a waiver to foreign robot ban to prevent chip disruptions

Analysis: Nvidia's Vera CPU takes on Intel and AMD with a platform play

Intel Fab 62 Moves to Next Phase, Begins Equipment Installation

Xiaomi Redmi Book 14 2027 With Intel Wildcat Lake CPU and 2.8K Display Confirmed

Thu, Oct 22

Evertiq Expo Warsaw: Top 10 chip companies took two-thirds of a record market

#Intel’s U.S. fab expansion is moving ahead! Arizona’s Fab 62 has begun equipment installation, targeting late-2028 production for potential 18A or ...

RT NEWS: [Breaking] Google Reportedly Sticks With Intel EMIB-T Despite 35% Substrate Yields as HumuFish Schedule Tightens Uncertainty is growing over ...

Intel begins Arizona Fab 62 equipment installation, advances Ohio project as AI CPU demand surges

[News] Tesla Reportedly to Invest KRW 1.1T to Secure Samsung Electro-Mechanics Substrate Capacity; NVIDIA Also Commits

[News] PSMC Q3 Gross Margin Hits 48% on Memory Price Gains; Eyes 10–20% Mature-Node Price Hikes in 2027

Intel installs production tools at Fab 62 in Arizona, provides updates on Ohio chip fab

Leak: Intel Core Ultra 9 4995HX Leads Nova Lake Laptop Lineup

[Breaking] Google Reportedly Sticks With Intel EMIB-T Despite 35% Substrate Yields as HumuFish Schedule Tightens

Chip Teardown Lab2 opened

Component price indexNPI, vs. prior period

Trending components