Intel’s decision to pull Intel 14A forward by a full year is the clearest sign yet that its foundry recovery is moving from an aspirational roadmap to a capital-backed manufacturing plan. The company now expects risk production for internal 14A products in the second half of 2027 and has committed to a high-volume manufacturing ramp in 2028—placing Intel’s next major process technology on roughly the same production timetable as TSMC’s A14 family.
That is a consequential shift for the semiconductor industry, for Intel Foundry’s credibility, and ultimately for the Windows PC ecosystem. Advanced manufacturing execution determines not only what future CPUs can deliver in performance and battery life, but also whether Intel can regain the economic scale needed to compete in AI servers, client processors, custom silicon, and advanced packaging.
Intel’s latest update arrives alongside a stronger-than-expected second quarter, a higher capital expenditure outlook, improved 18A output, and growing customer engagement around 14A. Taken together, those signals suggest the company is no longer treating 14A as a conditional research project. It is beginning to treat it as the next production platform on which both Intel products and external foundry customers may depend.
Still, a faster roadmap is not the same thing as a guaranteed victory. Intel has made meaningful progress, but the 14A ramp will be one of the most demanding execution tests in the company’s modern history.
Intel has moved several key 14A milestones forward:
The distinction between risk production and high-volume manufacturing matters. Risk production is the stage where a process can be used to fabricate early chips, validate designs, assess yields, and identify defects that remain too costly or too frequent for broad commercial deployment. High-volume manufacturing, by contrast, means the process is expected to run at meaningful scale with yields, costs, and capacity compatible with product launches.
For PC buyers, 2028 may sound distant. In chip development, however, it is uncomfortably close. Complex processors can require years of architecture development, physical design work, IP validation, packaging planning, and software enablement before they reach store shelves. A credible 2028 manufacturing target means Intel and prospective foundry customers must make important design commitments well before then.
What matters is the platform Intel is building around 14A.
Intel has positioned 14A as the successor to Intel 18A, a process family that introduced two major transistor-era technologies:
Intel 14A is expected to build on that foundation with further improvements to transistor performance, density, power efficiency, and manufacturability. The company has previously indicated that the node is designed to provide a meaningful performance-per-watt improvement and a substantial density gain over 18A.
Those projected improvements matter across several categories:
A Process Design Kit, or PDK, is the technical package that allows chip designers to use a manufacturing process. It typically includes design rules, electrical models, libraries, validation data, simulation information, and other tools required to turn a circuit design into a manufacturable chip.
Without a sufficiently mature PDK, an external customer cannot confidently begin a production-bound design.
That is why Intel’s October target deserves close attention. If 14A PDK 0.9 arrives on schedule and demonstrates stable performance models, credible SRAM behavior, usable IP, and practical design rules, it could help convert interest into committed customer designs.
If it slips, the effect could be much larger than a missed date on a presentation slide. External foundry customers plan years ahead. A delay can force them to reconsider capacity reservations, alter product schedules, or stay with an incumbent manufacturing partner.
Intel is right to place such emphasis on the PDK milestone. For a foundry business, a node is not commercially real simply because transistors work in a lab. It becomes commercially relevant when designers can use it with confidence.
That statement should be viewed carefully. It is an internal comparison rather than an independently standardized industry benchmark, and Intel has not publicly disclosed every data point needed to validate it. Yet it remains meaningful because 18A was itself a difficult and strategically important transition.
Defect density measures the number of manufacturing defects across a given area of silicon. Lower defect density generally produces better yield, meaning more functional chips per wafer. For high-end processors with large dies, complex chiplets, dense caches, and expensive packaging, yield is not an abstract engineering metric. It is one of the central determinants of profitability.
Intel’s reference to 256Mb SRAM yield is also important. SRAM is used for caches and other high-speed on-chip memory structures. It is often among the earliest and most revealing indicators of a process node’s maturity because SRAM arrays are dense, repetitive, sensitive to variation, and fundamental to modern CPU and AI accelerator design.
A process that demonstrates healthy SRAM behavior early can give chip architects more flexibility. It can support larger caches, denser compute tiles, and potentially more competitive designs without requiring excessive area or power compromises.
However, early silicon results do not eliminate later risks. Manufacturing progress must hold as the company transitions from test structures to large product dies, from small pilot lots to volume output, and from internal products to the more varied design requirements of external customers.
The fact that 14A may be ahead of 18A at this point is encouraging. The real achievement will be proving that the advantage survives the move into production.
Intel 18A is now being ramped for major internal products, including systems based on Panther Lake and Wildcat Lake, while the enhanced 18A-P process has entered risk production. This matters because 18A is effectively Intel Foundry’s first broad proof point for its next-generation manufacturing model.
A company cannot credibly sell future process technology solely on projections. It has to demonstrate that its current node can:
Those figures point to a familiar semiconductor reality: early-node manufacturing can be expensive, but costs can decline rapidly as yield improves, defects fall, process flows stabilize, and factory utilization rises.
For Windows PC users, this cost curve will shape how quickly leading-edge manufacturing becomes available in mainstream systems rather than only premium laptops and flagship desktops. Better yields and lower wafer costs give Intel more room to use newer manufacturing technology across higher-volume product tiers.
That compatibility is strategically valuable. A customer that has already invested in an 18A design may be able to move to 18A-P with less redevelopment than a full process migration would require. Intel can therefore offer customers an intermediate upgrade path while it prepares 14A for a later generation of products.
This improves the company’s foundry story in two ways:
For years, Intel’s manufacturing narrative was defined by delays, missed transitions, and the perception that TSMC had established an unassailable lead in leading-edge logic. Timing parity at the 2028 generation does not prove technical parity. It also does not prove that Intel will match TSMC’s yields, customer ecosystem, capacity, pricing, or packaging breadth.
But it changes the conversation.
Intel is no longer presenting 14A as a distant catch-up effort scheduled after the competition has moved on. It is presenting 14A as a contemporaneous process platform aimed at the same broad era of AI, high-performance computing, premium client devices, and custom silicon.
A useful comparison must consider:
Intel’s advantage is different. It owns major CPU product businesses, has extensive U.S. and European manufacturing assets, operates advanced packaging technologies, and can potentially combine chip design, foundry services, and system-level engineering for customers that want a more integrated partner.
The contest will not be settled by a single 2028 production date. It will be settled by which company can deliver usable technology at scale with predictable yields, capacity, economics, and support.
A recognizable customer matters for several reasons:
The stronger indicator will be whether Intel can convert broad engagement into a pipeline of multiple customers across different segments. A healthy foundry cannot depend on one high-profile AI customer. It needs sustained demand from a mix of compute, networking, automotive, communications, industrial, and consumer silicon programs.
The company says much of the increased investment is aimed at tooling, with spending directed toward Intel 3, 18A, 18A-P, and the broader manufacturing network needed to support future growth. It is also accelerating tool purchase orders, clean-room build-outs, and sourcing for constrained materials such as substrates and memory-related components.
That is important because leading-edge capacity cannot be created overnight. A high-volume 2028 ramp requires decisions now about equipment, facilities, supplier relationships, workforce development, and customer allocation.
Intel Foundry still reported an operating loss in the quarter despite improved revenue and better manufacturing performance. That is not surprising for a foundry organization undergoing a major investment cycle, but it is a reminder that volume, cost discipline, and external customer adoption must all improve together.
The core risk is straightforward: Intel could build capacity faster than it secures profitable demand. Management has emphasized that it intends to be more disciplined about investment and less willing to make large bets without customer signals. The 14A acceleration suggests those signals are improving, but the company has not yet disclosed enough binding customer detail to remove uncertainty.
If Intel executes well, 14A could enable future generations of Core Ultra, desktop CPUs, mobile processors, AI accelerators, and potentially custom chips with better energy efficiency and higher transistor budgets. For Windows PCs, that can translate into more than benchmark gains.
Potential benefits include:
Intel also appears to be positioning advanced packaging as an equally important part of that future. Chiplets and heterogeneous integration are becoming central to performance scaling. Rather than relying on a single monolithic die, future processors can combine compute tiles, graphics tiles, I/O dies, cache, accelerators, and memory-adjacent components in one package.
That strategy could be especially valuable in PCs, where Intel must balance performance leadership against battery life, price, and device size.
The company is not merely saying that 14A exists in development. It is saying that the PDK is progressing, defect density and transistor performance are encouraging, external customer engagement is increasing, internal demand is growing, and a high-volume 2028 ramp is now important enough to fund.
That is a markedly stronger position than treating 14A as a node that might proceed only if a major customer appeared.
The next milestones are equally clear. Intel needs to deliver PDK 0.9 on time, demonstrate that 18A continues to improve as volume grows, show that 18A-P is commercially useful, disclose more evidence of 14A customer commitments, and maintain financial discipline as capital spending rises.
If those pieces fall into place, Intel 14A could become the process node that turns Intel Foundry from a strategic promise into a credible second source for the world’s most advanced chips. If they do not, the 2028 target will be remembered as another ambitious schedule that ran ahead of the economics and ecosystem required to support it.
For now, Intel has earned a more serious hearing. Its 14A roadmap is no longer simply about catching up; it is becoming a test of whether the company can compete at the leading edge on the same calendar as the industry’s most established foundry leader.
The report also associates 14A with Intel’s planned use of High-NA EUV lithography. That equipment could reduce reliance on some complex multi-patterning steps, but its production value will depend on throughput, maintenance reliability, resist performance, and wafer economics during the 2028 ramp.
That is a consequential shift for the semiconductor industry, for Intel Foundry’s credibility, and ultimately for the Windows PC ecosystem. Advanced manufacturing execution determines not only what future CPUs can deliver in performance and battery life, but also whether Intel can regain the economic scale needed to compete in AI servers, client processors, custom silicon, and advanced packaging.
Intel’s latest update arrives alongside a stronger-than-expected second quarter, a higher capital expenditure outlook, improved 18A output, and growing customer engagement around 14A. Taken together, those signals suggest the company is no longer treating 14A as a conditional research project. It is beginning to treat it as the next production platform on which both Intel products and external foundry customers may depend.
Still, a faster roadmap is not the same thing as a guaranteed victory. Intel has made meaningful progress, but the 14A ramp will be one of the most demanding execution tests in the company’s modern history.
Overview: What Intel’s New 14A Timeline Actually Means
Intel has moved several key 14A milestones forward:- 14A PDK 0.5 is complete, giving customers and internal teams an early foundation for test chips and design evaluation.
- 14A PDK 0.9 remains targeted for October, a much more important milestone for prospective customers preparing real products.
- Risk production for internal products is planned for the second half of 2027.
- High-volume manufacturing is now targeted for 2028.
The distinction between risk production and high-volume manufacturing matters. Risk production is the stage where a process can be used to fabricate early chips, validate designs, assess yields, and identify defects that remain too costly or too frequent for broad commercial deployment. High-volume manufacturing, by contrast, means the process is expected to run at meaningful scale with yields, costs, and capacity compatible with product launches.
For PC buyers, 2028 may sound distant. In chip development, however, it is uncomfortably close. Complex processors can require years of architecture development, physical design work, IP validation, packaging planning, and software enablement before they reach store shelves. A credible 2028 manufacturing target means Intel and prospective foundry customers must make important design commitments well before then.
Intel 14A Is More Than a Smaller Number
The “14A” name refers to Intel’s angstrom-era process naming system. It should not be interpreted as a literal measurement of transistor features. Modern node names are branding labels that indicate a new generation of manufacturing technology rather than a directly comparable physical dimension.What matters is the platform Intel is building around 14A.
Intel has positioned 14A as the successor to Intel 18A, a process family that introduced two major transistor-era technologies:
- RibbonFET, Intel’s gate-all-around transistor implementation
- PowerVia, its backside power delivery technology
Intel 14A is expected to build on that foundation with further improvements to transistor performance, density, power efficiency, and manufacturability. The company has previously indicated that the node is designed to provide a meaningful performance-per-watt improvement and a substantial density gain over 18A.
Those projected improvements matter across several categories:
- Windows laptops, where power efficiency can translate into better battery life, thinner systems, and lower fan noise.
- Desktop PCs, where more efficient transistors can support higher performance within realistic cooling and power limits.
- Data centers, where every improvement in performance per watt affects operating costs, rack density, and infrastructure planning.
- AI infrastructure, where power delivery, memory bandwidth, packaging, and compute density increasingly matter as much as raw transistor counts.
- Custom ASICs, which may require different balances of frequency, power, area, and integration than a general-purpose CPU.
Why PDK 0.9 Is the Crucial Near-Term Test
Intel’s 14A PDK 0.5 completion is a significant milestone, but PDK 0.9 is likely to be the more consequential checkpoint for external customers.A Process Design Kit, or PDK, is the technical package that allows chip designers to use a manufacturing process. It typically includes design rules, electrical models, libraries, validation data, simulation information, and other tools required to turn a circuit design into a manufacturable chip.
Without a sufficiently mature PDK, an external customer cannot confidently begin a production-bound design.
PDK 0.5: Early Access and Validation
The 0.5 PDK stage is useful for:- Evaluating the process architecture
- Running early design experiments
- Building test structures
- Assessing yield behavior
- Measuring SRAM and transistor characteristics
- Estimating the likely benefits of moving from another process node
PDK 0.9: The Customer Commitment Milestone
A 0.9 PDK is closer to the point where customers can begin more serious implementation work. It is still not the final production kit, but it is far enough along to reveal whether the process has the design maturity, IP readiness, and manufacturability needed for an actual product program.That is why Intel’s October target deserves close attention. If 14A PDK 0.9 arrives on schedule and demonstrates stable performance models, credible SRAM behavior, usable IP, and practical design rules, it could help convert interest into committed customer designs.
If it slips, the effect could be much larger than a missed date on a presentation slide. External foundry customers plan years ahead. A delay can force them to reconsider capacity reservations, alter product schedules, or stay with an incumbent manufacturing partner.
Intel is right to place such emphasis on the PDK milestone. For a foundry business, a node is not commercially real simply because transistors work in a lab. It becomes commercially relevant when designers can use it with confidence.
Intel Says 14A Is Tracking Ahead of 18A Development
One of the most encouraging elements of Intel’s update is management’s assertion that 14A is progressing ahead of 18A at comparable development stages in defect density and transistor performance.That statement should be viewed carefully. It is an internal comparison rather than an independently standardized industry benchmark, and Intel has not publicly disclosed every data point needed to validate it. Yet it remains meaningful because 18A was itself a difficult and strategically important transition.
Defect density measures the number of manufacturing defects across a given area of silicon. Lower defect density generally produces better yield, meaning more functional chips per wafer. For high-end processors with large dies, complex chiplets, dense caches, and expensive packaging, yield is not an abstract engineering metric. It is one of the central determinants of profitability.
Intel’s reference to 256Mb SRAM yield is also important. SRAM is used for caches and other high-speed on-chip memory structures. It is often among the earliest and most revealing indicators of a process node’s maturity because SRAM arrays are dense, repetitive, sensitive to variation, and fundamental to modern CPU and AI accelerator design.
A process that demonstrates healthy SRAM behavior early can give chip architects more flexibility. It can support larger caches, denser compute tiles, and potentially more competitive designs without requiring excessive area or power compromises.
However, early silicon results do not eliminate later risks. Manufacturing progress must hold as the company transitions from test structures to large product dies, from small pilot lots to volume output, and from internal products to the more varied design requirements of external customers.
The fact that 14A may be ahead of 18A at this point is encouraging. The real achievement will be proving that the advantage survives the move into production.
The 18A Ramp Is the Foundation for Everything That Follows
Intel’s 14A optimism would carry much less weight if 18A were struggling. Instead, the company says 18A output rose sharply during the second quarter, with factories exceeding internal volume goals across Intel 7, Intel 3, and Intel 18A.Intel 18A is now being ramped for major internal products, including systems based on Panther Lake and Wildcat Lake, while the enhanced 18A-P process has entered risk production. This matters because 18A is effectively Intel Foundry’s first broad proof point for its next-generation manufacturing model.
A company cannot credibly sell future process technology solely on projections. It has to demonstrate that its current node can:
- Reach yield targets.
- Improve cycle times.
- reduce wafer costs.
- support real commercial products.
- deliver enough volume to serve internal and external demand.
- provide reliable packaging and supply-chain capacity.
Those figures point to a familiar semiconductor reality: early-node manufacturing can be expensive, but costs can decline rapidly as yield improves, defects fall, process flows stabilize, and factory utilization rises.
For Windows PC users, this cost curve will shape how quickly leading-edge manufacturing becomes available in mainstream systems rather than only premium laptops and flagship desktops. Better yields and lower wafer costs give Intel more room to use newer manufacturing technology across higher-volume product tiers.
Why 18A-P Matters
Intel 18A-P is not simply a replacement for 18A. It is a performance-enhanced variant intended to retain design compatibility while offering better performance and power characteristics.That compatibility is strategically valuable. A customer that has already invested in an 18A design may be able to move to 18A-P with less redevelopment than a full process migration would require. Intel can therefore offer customers an intermediate upgrade path while it prepares 14A for a later generation of products.
This improves the company’s foundry story in two ways:
- It provides a more continuous technology cadence.
- It reduces the all-or-nothing nature of a customer’s manufacturing decision.
Matching TSMC’s A14 Schedule Is Symbolically Important
TSMC plans volume production of its own A14 process technology in 2028. Intel’s revised 14A schedule now lines up with that timeframe, which is a meaningful psychological and commercial milestone.For years, Intel’s manufacturing narrative was defined by delays, missed transitions, and the perception that TSMC had established an unassailable lead in leading-edge logic. Timing parity at the 2028 generation does not prove technical parity. It also does not prove that Intel will match TSMC’s yields, customer ecosystem, capacity, pricing, or packaging breadth.
But it changes the conversation.
Intel is no longer presenting 14A as a distant catch-up effort scheduled after the competition has moved on. It is presenting 14A as a contemporaneous process platform aimed at the same broad era of AI, high-performance computing, premium client devices, and custom silicon.
Schedule Parity Does Not Equal Process Parity
The industry should resist simplistic comparisons between Intel 14A and TSMC A14. Node names cannot be used as direct measurements of density, power efficiency, transistor performance, or manufacturing maturity.A useful comparison must consider:
- Logic density
- SRAM density and yield
- Frequency capability
- Leakage characteristics
- Power delivery design
- High-NA EUV deployment
- Design-rule complexity
- Availability of IP and EDA tools
- Packaging options
- Geographic capacity
- Pricing and long-term supply commitments
Intel’s advantage is different. It owns major CPU product businesses, has extensive U.S. and European manufacturing assets, operates advanced packaging technologies, and can potentially combine chip design, foundry services, and system-level engineering for customers that want a more integrated partner.
The contest will not be settled by a single 2028 production date. It will be settled by which company can deliver usable technology at scale with predictable yields, capacity, economics, and support.
Tesla Could Be an Important Early Signal—But It Is Not the Entire Foundry Story
Tesla has publicly indicated an intention to use Intel’s 14A process for chips associated with its proposed Terafab AI effort in Austin. If that relationship translates into a substantial manufacturing program, it would represent a significant validation for Intel Foundry.A recognizable customer matters for several reasons:
- It demonstrates that an external company is willing to consider Intel for leading-edge manufacturing.
- It helps justify capital investment in future node capacity.
- It creates an opportunity to validate Intel’s design, wafer, packaging, and supply-chain services under real commercial requirements.
- It could attract other customers that want an alternative source of advanced manufacturing capacity.
The stronger indicator will be whether Intel can convert broad engagement into a pipeline of multiple customers across different segments. A healthy foundry cannot depend on one high-profile AI customer. It needs sustained demand from a mix of compute, networking, automotive, communications, industrial, and consumer silicon programs.
Capital Spending Shows Intel Is Beginning to Commit
Intel’s decision to lift its 2026 capital expenditure outlook to more than $20 billion is one of the most tangible elements of the announcement. Process roadmaps are inexpensive to present. Fabs, lithography systems, clean rooms, metrology tools, packaging lines, and supply contracts are not.The company says much of the increased investment is aimed at tooling, with spending directed toward Intel 3, 18A, 18A-P, and the broader manufacturing network needed to support future growth. It is also accelerating tool purchase orders, clean-room build-outs, and sourcing for constrained materials such as substrates and memory-related components.
That is important because leading-edge capacity cannot be created overnight. A high-volume 2028 ramp requires decisions now about equipment, facilities, supplier relationships, workforce development, and customer allocation.
The CapEx Risk Remains Real
Greater investment is a sign of confidence, but it also raises financial risk. Semiconductor fabs require enormous upfront spending, while returns arrive only if customer demand, yields, and pricing hold up after the equipment is installed.Intel Foundry still reported an operating loss in the quarter despite improved revenue and better manufacturing performance. That is not surprising for a foundry organization undergoing a major investment cycle, but it is a reminder that volume, cost discipline, and external customer adoption must all improve together.
The core risk is straightforward: Intel could build capacity faster than it secures profitable demand. Management has emphasized that it intends to be more disciplined about investment and less willing to make large bets without customer signals. The 14A acceleration suggests those signals are improving, but the company has not yet disclosed enough binding customer detail to remove uncertainty.
What This Means for Windows PCs and AI PCs
Intel’s 14A timeline will not affect Windows laptops this year or next. Its impact will emerge later, through the processors and platform designs built on the node.If Intel executes well, 14A could enable future generations of Core Ultra, desktop CPUs, mobile processors, AI accelerators, and potentially custom chips with better energy efficiency and higher transistor budgets. For Windows PCs, that can translate into more than benchmark gains.
Potential benefits include:
- Longer battery life in thin-and-light laptops
- Higher sustained performance in compact systems
- More capable on-device AI workloads
- Larger CPU and GPU caches
- Better integrated graphics performance within constrained power envelopes
- More efficient NPUs and media engines
- Reduced cooling requirements for some form factors
- Greater room for premium features without proportionally increasing power draw
Intel also appears to be positioning advanced packaging as an equally important part of that future. Chiplets and heterogeneous integration are becoming central to performance scaling. Rather than relying on a single monolithic die, future processors can combine compute tiles, graphics tiles, I/O dies, cache, accelerators, and memory-adjacent components in one package.
That strategy could be especially valuable in PCs, where Intel must balance performance leadership against battery life, price, and device size.
The Biggest Risks Intel Still Has to Navigate
Intel’s roadmap acceleration is positive, but the road from promising data to high-volume production remains hazardous.Yield Risk
Early yield indicators are useful, but large commercial dies are much harder to manufacture than test structures. A process can look strong in SRAM data or small test chips and still encounter problems when complex products begin ramping.Customer-Ecosystem Risk
TSMC’s ecosystem is deeply entrenched. Intel must prove that its PDKs, IP availability, electronic design automation flows, customer support, and production logistics are competitive enough to persuade designers to move.Capacity Risk
A 2028 high-volume ramp requires enough fabs, tools, trained employees, substrates, packaging capacity, and supply-chain resilience. The industry is already experiencing constraints in multiple areas tied to AI infrastructure.Economic Risk
Leading-edge manufacturing is expensive, and Intel must improve the foundry business’s profitability while continuing to invest heavily. Better yields are necessary, but not sufficient.Product Risk
Intel’s internal products need to execute as well. External customers will judge the company’s manufacturing claims partly by how Panther Lake, Wildcat Lake, Xeon platforms, and future Intel-designed silicon perform in the market.Competitive Risk
TSMC will not stand still, and neither will Samsung, Rapidus, or the broader ecosystem of specialized foundries and packaging providers. Intel does not need to win every customer to succeed, but it must offer a compelling reason for customers to diversify their manufacturing supply.A More Credible Roadmap, Not Yet a Completed Turnaround
Intel’s accelerated 14A plan is a major development because it combines three things that had too often been disconnected in the company’s recent history: a technology roadmap, early manufacturing evidence, and capital commitments.The company is not merely saying that 14A exists in development. It is saying that the PDK is progressing, defect density and transistor performance are encouraging, external customer engagement is increasing, internal demand is growing, and a high-volume 2028 ramp is now important enough to fund.
That is a markedly stronger position than treating 14A as a node that might proceed only if a major customer appeared.
The next milestones are equally clear. Intel needs to deliver PDK 0.9 on time, demonstrate that 18A continues to improve as volume grows, show that 18A-P is commercially useful, disclose more evidence of 14A customer commitments, and maintain financial discipline as capital spending rises.
If those pieces fall into place, Intel 14A could become the process node that turns Intel Foundry from a strategic promise into a credible second source for the world’s most advanced chips. If they do not, the 2028 target will be remembered as another ambitious schedule that ran ahead of the economics and ecosystem required to support it.
For now, Intel has earned a more serious hearing. Its 14A roadmap is no longer simply about catching up; it is becoming a test of whether the company can compete at the leading edge on the same calendar as the industry’s most established foundry leader.
Update: Additional details (July 24, 2026)
Wccftech reports that Intel 14A is expected to introduce RibbonFET 2, a second-generation gate-all-around transistor design, alongside PowerDirect, a direct-contact evolution of Intel’s backside power-delivery approach. Intel 18A introduced RibbonFET and PowerVia; the new report characterizes 14A as extending both technology paths rather than merely shrinking the preceding node.The report also associates 14A with Intel’s planned use of High-NA EUV lithography. That equipment could reduce reliance on some complex multi-patterning steps, but its production value will depend on throughput, maintenance reliability, resist performance, and wafer economics during the 2028 ramp.
References
- Primary source: TrendForce
Published: 2026-07-24T10:07:27+00:00
[News] Intel Pulls Forward 14A Timeline by One Year, Targeting 2028 Mass Production Near TSMC’s Roadmap
As Intel delivered stronger-than-expected second-quarter results, reportedly posting its fastest quarterly revenue growth since 2011 and hiked its 202...www.trendforce.com - Related coverage: techspot.com
Intel says 14A is on track for 2028 risk production, 10A and 7A now in development | TechSpot
Speaking at J.P. Morgan' 54th annual Global Technology, Media and Communications Conference in Boston, Tan revealed that development of Intel's 14A process node is progressing as planned,...www.techspot.com - Related coverage: design-reuse.com
Intel Accelerates 1.4nm Foundry Push, Targets Mass Production in 2029
Design And Reuse - Catalog of IP Cores and Silicon on Chip solutions for IoT, Automotive, Security, RISC-V, AI, ... and Asic Design Platforms and Resourceswww.design-reuse.com
- Related coverage: anysilicon.com
Intel Begins Work on 10A and 7A Process Nodes as 14A Roadmap Moves Toward Key PDK Release - AnySilicon
Intel has started early development of its future 10A and 7A process nodes while keeping 14A on track for a key PDK milestone. Here is what it means for advanced semiconductor manufacturing.
anysilicon.com
- Related coverage: tomshardware.com
Tesla hires 17-year Intel veteran responsible for billion-dollar fab startups — Gary Jiang likely chosen to oversee fab efforts for Terafab's licensing of 14A | Tom's Hardware
Musk's Terafab now has advanced tooling talent.www.tomshardware.com - Related coverage: pcgamer.com
TSMC reportedly plots ultra-advanced sub-1 nm chips with 'trial' production starting in 2029 [Updated] | PC Gamer
Meanwhile, TSMC's A14 node is supposedly go for 2028.www.pcgamer.com
Last edited:
