Intel Solves 18A Yield Issues, Production Reaches 30,000 Wafers Per Month

"For the 18A node, the general impression is that it primarily serves Intel's internal products. Future versions, like 18A-P and 18A-PT, aim to attract external customers and are expected to be long-term Intel Foundry products. Intel's 14A node is becoming a critical product for Intel Foundry, with active development targeting a 2027 release. As Intel collaborates with customers on the node design, potential foundry partners are assessing whether the upcoming technology meets their needs. Early customer feedback indicates strong satisfaction with the development progress, with those familiar with the node describing it as genuinely competitive.

Starting in 2026, we expect to see an Intel-NVIDIA partnership formed in mid-September slowly manifest, with the integration of NVIDIA RTX GPUs inside Intel x86 SoCs. It is unclear whether this partnership will involve any foundry services from NVIDIA, but for now, the Intel-NVIDIA deal is focused on product delivery only. When Intel releases the 14A 0.5 PDK, we can expect to hear more feedback from potential foundry customers. All eyes are on the 14A node debut in 2027, alongside the 18A-P and 18A-PT for more specialized applications."

Source: https://www.techpowerup.com/344401/nvidia-tested-intels-18a-node-but-did-not-commit-to-intel-foundry

"Intel Solves 18A Yield Issues, Production Reaches 30,000 Wafers Per Month​

by AleksandarK Today, 04:29 Discuss (21 Comments)
Intel has reportedly resolved all yield issues it experienced with its 18A node over the past few months. According to the sell-side equity research firm BlueFin Research Partners, Intel has fixed these yield issues, making the high-volume production of its latest node sustainable from both a manufacturing yield and economic standpoint. Readers may recall that late last year, Intel confirmed that 18A yields were improving by about 7% per month over several months. This trend continued into the launch of the first 18A product, codenamed "Panther Lake," up until today, when the yield issues have reportedly been resolved. For a mature node, this usually means defect rates of D0=0.1 or D0=0.2, and it is likely that Intel has now achieved the lower end of that spectrum with the node advancements and months of 18A high-volume production.

Regarding manufacturing capacity, the report notes that Intel's 18A is produced at two sites: Fab 52 in Phoenix, Arizona, and another in Hillsboro, Oregon. These fabs currently produce about 30,000 wafers per month with the 18A printing. This capacity is currently sufficient for Intel's internal production, such as the "Panther Lake" processors, but additional capacity will be needed for other internal products."

Excellent news. Hopefully it’s able to at least match performance of TSMC’s 3nm process.

madpistol said:

Excellent news. Hopefully it’s able to at least match performance of TSMC’s 3nm process.

Click to expand...

18A is not a new node, it is the panther lake node, price down/volume for the new intel laptop is what to expect of this higher yield situation, maybe new CPU using it coming out.

panther lake (18A) do a lot of things better than Lunar lake (N3B), but there could be design change itself helping and the packaging has well, that said it seem quite competitive./

LukeTbk said:

18A is not a new node, it is the panther lake node, price down/volume for the new intel laptop is what to expect of this higher yield situation, maybe new CPU using it coming out.

panther lake (18A) do a lot of things better than Lunar lake (N3B), but there could be design change itself helping and the packaging has well, that said it seem quite competitive./

Click to expand...

It's only a year old... So I would still consider it relatively New. Fab 52 opened its doors in July 2025, with the first commercial batches being released in October of that year.

The backside power delivery gives the 18A process advantages over TSMC N3 and N2 in performance per watt (after 40w or so), voltage stability, and clock speeds, but TSMC 3N and 2N do offer better transistor density, which makes them better suited for mass-produced mobile chips.

So TSMC does better for chips that need to maximize processing power, but Intel does better on logic and IO, so using TSMC for the processing core chiplets/tiles, then using Intel and their packaging facilities for the rest of the chip has some very strong benefits.

Hopefully Intel can get some special version of 18A or 14A that focuses on transistor density so they can offer a competing process to the TSMC variants.
But they have their own individual strengths; Intel and their PowerVia can move an absolute crapload of juice through their chips before they catch fire, which is impressive on its own, and their Transistor architecture and designs use a fin-heavy design to handle all that juice, TSMC has optimized for smaller transistors with a much lower power draw, which is ideal for environments where you need to make every watt count.

Lakados said:

Hopefully Intel can get some special version of 18A or 14A that focuses on transistor density so they can offer a competing process to the TSMC variants.
But they have their own individual strengths; Intel and their PowerVia can move an absolute crapload of juice through their chips before they catch fire, which is impressive on its own, and their Transistor architecture and designs use a fin-heavy design to handle all that juice, TSMC has optimized for smaller transistors with a much lower power draw, which is ideal for environments where you need to make every watt count.

Click to expand...

To add to that, this is why you see nVidia using TSMC "4N" which is their own special TSMC 4nm(ish) node rather than the smaller ones. They needed something optimized for MOAR POWER not max density or low heat or any of that. They'd rather have a larger process that can handle more power and scale up than the smallest and most efficient and you can't have all things these days because of the limits of physics.

While some people love to slobber Apple's knob for their ARM chips and the N2 node they are made on, they are not the kind of thing that would scale well to high power units and high performance per core. That's fine, particularly for a phone, there size, thermals, and power usage is king. But there's a reason to want other processes, particularly for gamers. We want single core performance to absolutely rip, because even though games are multi-threaded, there will always be that one thread that everything else depends on and there's only so much you can spin things off.

Sycraft said:

To add to that, this is why you see nVidia using TSMC "4N" which is their own special TSMC 4nm(ish) node rather than the smaller ones. They needed something optimized for MOAR POWER not max density or low heat or any of that. They'd rather have a larger process that can handle more power and scale up than the smallest and most efficient and you can't have all things these days because of the limits of physics.

While some people love to slobber Apple's knob for their ARM chips and the N2 node they are made on, they are not the kind of thing that would scale well to high power units and high performance per core. That's fine, particularly for a phone, there size, thermals, and power usage is king. But there's a reason to want other processes, particularly for gamers. We want single core performance to absolutely rip, because even though games are multi-threaded, there will always be that one thread that everything else depends on and there's only so much you can spin things off.

Click to expand...

“Intel 18A Silicon Goes to Space with "Starfire" Processors​

by AleksandarK Today, 05:38 Discuss (0 Comments)
Intel's 18A node is officially space-grade, according to documentation on Intel's website. A document that initially went unnoticed has been published, detailing a new "Starfire" generation of processors built on the 18A node and designed for space operations. Intel has designated two SKUs for orbital computing, each equipped with four P-Cores and four LPE-Cores, resulting in an eight-core SoC design. One is optimized for low-power operations with clocks adjusted for efficiency, while the other is designed for performance. The Low Power edition tunes its four P-Cores to 1.0 GHz, while the four LPE-Cores run at 850 MHz. For the Performance version, Intel has increased the frequency of P-Cores to 3.1 GHz, with LPE-Cores reaching up to 2.1 GHz.

Both SKUs feature a GPU tile built on the Intel 3 node, which includes 4 Xe cores with 64 EUs. On the Low Power SKU, these Xe cores operate between 800 MHz and 1.0 GHz, while the Performance SKU boosts this to 2.0 GHz. The total package for the Low Power version remains at 10 W TDP, while the Performance SKU is capped at 35 W TDP. Additionally, Intel includes an NPU on both versions, with a compute capability of 45 TOPS on the Low Power version and up to 75 TOPS on the Performance SKU using INT8 precision. Beyond specifications, "Starfire" can operate within a T-Junction range from -55°C to 125°C. This range, alongside TID, SEL, and SEE radiation hardening certifications, introduces the first 18A silicon to space, ideal for orbital computation. Intel targets this quarter, Q3, for samples.“

Sycraft said:

While some people love to slobber Apple's knob for their ARM chips and the N2 node they are made on, they are not the kind of thing that would scale well to high power units and high performance per core. That's fine, particularly for a phone, there size, thermals, and power usage is king. But there's a reason to want other processes, particularly for gamers. We want single core performance to absolutely rip, because even though games are multi-threaded, there will always be that one thread that everything else depends on and there's only so much you can spin things off.

Click to expand...

Apple has yet to release an N2 node chips and I really do not think single core performance is the reason you want an other processor over an M5 pro/A19 pro (which are not on N2 Node, but the quite old N3 family)