Processors - Page 3

NVIDIA will be releasing its Arm-based N1X and N1 processors in 2026, competing directly against CPU giants AMD and Intel, and while the N1X + N1 were meant to be released this year, we can expect a probable launch at Computex 2026.

In a new video from leaker Moore's Law is Dead, who reports from a couple of his sources -- a major OEM and an NVIDIA partner -- N1X and N1 processors will be launched "by the middle of 2026". The first source said: "Just got word (major OEM) that NVIDIA N1X + N1 products will launch by the middle of 2026. Technically, from what I can see in documentation, both products are listed for notebooks... although I was told there could be some form of desktop launch too!"

MLID's second source said: "Yes, I (NVIDIA partner) can confirm that NVIDIA will launch their APUs to consumers by Q2 for sure because Alienware is targeting Q1 for the launch of a 16-inch laptop that utilizes them!"

Apple's new M5 MacBook Pro laptop is severely overheating in new tests, with the new M5-powered MacBook Pro only cooled by a single fan, the M5 chip is hitting 99C under load.

This isn't out of nowhere as Apple's previous-gen M4 MacBook Pro can easily pass 100C+ under loads, but the new M5 MacBook Pro has the same thermal solution as its predecessor. In a new video, YouTuber Max Tech takes a deeper look at both the 14-inch M5 MacBook Pro and the 14-inch M4 MacBook Pro, with some rather surprising results.

While running Cinebench 2024, Max Tech discovered that the thermal solution inside of the new Apple M5-powered MacBook Pro couldn't keep the M5 cool enough. However, the M5 version of the MacBook Pro runs cooler than the M4 version of the MacBook Pro, but we're talking 1-2C which is really nothing at the end of the day.

Intel will have depleted its CPU inventory in the next few months, running out of processors to sell, as demand right now is reportedly higher than anticipated.

Intel CFO David Zinsner spoke with Barron's after its Q3 2025 earnings call, saying that the company is facing a "chip supply" shortage, for both consumer and data center processors. This isn't expected to stop and will continue rolling into Q1 2026, where by then, "chip inventories" will have been depleted -- zilch, nada, gone -- which is a major concern.

Firstly, it means Intel is somehow selling every single processor it is having made (by TSMC) and soon in-house (Intel 18A). Secondly, it means that even with AMD kicking its ass on every front, somehow, Intel is selling its processors so fast, it will run out of inventory in the months ahead. It all feels very strange to me, as there are only a handful of Intel products that I can see consumers willingly buying over an AMD processor (like the Core Ultra 9 285K in some scenarios, even then, the Ryzen 7 9800X3D kicks its ass in gaming).

Apple's next-gen A20 Pro chip is reportedly called "Borneo Ultra" and its A20 non-Pro is called "Borneo", ready to power the next-gen iPhone 18 and foldable iPhone in 2026, fabbed on TSMC's bleeding-edge 2nm process node.

The codenames are coming from Weibo tipster "Mobile phone chip expert" who claims they had dinner with a friend of theirs, who happens to be the IC design boss at Apple. Over dinner, the two discussed the new iPhone and its A-series processor, where we're finding out that the A20 Pro is dubbed "Borneo Ultra" and the A20 dubbed "Borneo".

The base iPhone 18 should be powered by the regular A20 "Borneo", while the higher-end iPhone 18 Pro, iPhone 18 Pro Max, and foldable iPhone, will be powered by the A20 Pro "Borneo Ultra" processor.

Intel will fully release its next-gen Panther Lake CPUs at CES 2026 in just a couple of months' time, with new premium flagship laptop designs shown off in Las Vegas, the first chip Intel will fab and mass produce for consumers on its new in-house Intel 18A process node.

Intel hosted its Q3 2025 earnings call this week with CEO Lip-Bu Tan outlining his plans for the company over the next 18 months or so, with the laptop-ready Panther Lake CPUs dropping at CES 2026 as the new Core Ultra Series 300 family. The first SKUs will be higher-end premium options, but as Intel 18A continues to ramp, Panther Lake will debut in cheaper variants.

Intel CEO Lip-Bu Tan explained: "In addition, and just as important, the group will spearhead the build out of a new basic and design service business to deliver purpose-built silicon to a broad range of external customers. This will not only extend the reach of our core x86 IP, but also leverage our design strength to deliver an array of solutions from general purpose to fixed function computing. In client, we are on track to launch our first Panther Lake SKU by year end, followed by additional SKUs in the first half of next year".

Intel will be launching its next-generation Nova Lake CPUs in 2026, its next-gen processor family for the desktop, with up to 52 cores, its upgraded Xe3 GPU, and fabbed on Intel's new Intel 18A process node.

During the Q3 2025 earnings call, Intel CEO Lip-Bu Tan talked about his plans for moving the company forward in its client, server, and foundry-side businesses. The CEO reaffirmed some of the things we already know like its new Panther Lake CPUs fabbed on its new Intel 18A node launching at CES 2026 with a full unveiling, unlike the tease and small details from its recent Tech Tour 2025 event.

However, he did say that Intel is reaffirming the high-end processor market is an intense battlefield, but there are things in the pipeline for that. First, the company will be launching its refreshed Arrow Lake CPUs under the Core Ultra 200S Plus processors on its current LGA 1851 socket, but after that comes Nova Lake-S on the new LGA 1954 socket -- meaning new motherboards and CPUs -- in the second half of 2026, with a new architecture, software changes, and more.

NVIDIA CEO Jensen Huang is currently celebrating the first Blackwell wafer to be produced at TSMC's Arizona facility, and while that is a substantial achievement, the Arizona facility doesn't have all the necessary tools to complete the product, meaning those newly produced chips need to be shipped elsewhere.

If your first guess for the destination of the chips you are correct. Yes, after producing the Blackwell wafer on American soil, as per the plan of US President Donald Trump, it will still need the magic touch of TSMC Taiwan, as the Blackwell wafer will remain in its most rudimentary state until it's shipped back to Taiwan where it will undergo an assembly process that turns the wafer into the usable chips tech companies, and PC gamers around the world are after.

Here's how it works. NVIDIA's Blackwell family of data center chips features two reticle-sized compute dies and eight stacks of HBM3e memory. To connect all these components together, TSMC implements a stitching process that it calls Chip-on-Wafer-on-Substrate, or CoWoS. Essentially, this process stitches all of the components together to form the Blackwell datacenter chip, and currently, the only facilities capable of carrying out the CoWoS process are located in Taiwan.

Whispers are again circulating that AMD has new Ryzen 3D V-Cache processors on the horizon for its Zen 5 family, one of which will have the cache on both of its two chiplets.

That's the top-end CPU, of course, which according to the new leak from Chi11eddog on X (flagged by VideoCardz), will be the AMD Ryzen 9 9950X3D2.

The '2' tacked onto the end of the X3D indicates that this is the first (theoretical) 3D V-Cache offering with such a dual-cache configuration, carrying a total of 192MB of L3 cache (in two 96MB pools).

Intel CEO Lip-Bu Tan has headed to the Middle East to sit down with Saudi officials to reportedly ink a new partnership deal.

Reports from Arab News indicate that Lip-Bu Tan has met with the Saudi Minister of Communications and Information Technology, Abdullah Al-Swaha, to discuss new areas of which Intel and Saudi Arabia can benefit from developing Intel's semiconductor business, along with advanced computing technology. Additionally, the discussion involved developing Intel's artificial intelligence infrastructure.

Lip-Bu Tan has been on a tear when it comes to deals as the new Intel CEO recently inked a partnership with NVIDIA, SoftBank, and the Trump Administration, and now it seems a new deal is in the works with Saudi Arabia.

Intel's 18A process or node, created in-house at its US-based Fab 52 semiconductor facility or Intel Foundry in Arizona, is set to be used in the company's upcoming Panther Lake and Clearwater Forest chips. The production-ready technology and Foundry are open for business, and Intel is looking to partner with big tech to manufacture new chips and technologies using the 18A or 18A-P process.

One of these new customers, according to a report over at SemiAccurate (via TechPowerUp), is Microsoft. The company is apparently planning to use Intel's 18A process for its next-gen Maia 3 accelerator for AI, codenamed "Griffin." And with that, depending on the yields and success of its latest AI hardware, Microsoft could also tap future Intel process technology like 18A-PT and 14A for future Maia accelerators.

Microsoft expressing interest in leveraging Intel technology for its AI accelerators isn't new. Microsoft CEO Satya Nadella released a statement about Microsoft's plans to use 18A for an undisclosed chip when Intel announced its Foundry roadmap in early 2024.

Intel's Core Ultra X9 388H processor, rumored to be the flagship for the incoming Panther Lake mobile range, could have some seriously powerful integrated graphics going by a new leak.

As Wccftech reports (via VideoCardz), this leak surfaced at Laptop Review Club, and if you haven't heard of that, we wouldn't blame you - but you probably have heard of Golden Pig Upgrade, a regular hardware leaker who works with the site.

Intel has already told us that the 12-core Arc Xe3 integrated GPU (in the Core Ultra X9 388H) is set to be up to 50% quicker than Lunar Lake, and the leak shows the iGPU hit around 6300 in 3DMark Time Spy - which is some 30% faster than Lunar Lake (add your own seasoning with this, of course).

Apple has just announced its new M5 processor, which has some big upgrades across the board for new MacBook Pro laptops, a new Vision Pro headset, and new M5 iPad Pro.

The new Apple M5 processor features up to 10 CPU cores and 10 GPU cores, which might not sound like a lot, but they are much faster and more efficient than the previous-gen M4 chip. Apple's new M5 processor and its 10-core CPU configuration splits into 6 performance and 4 efficiency cores, just like the M4, but each of the cores has up to 15% more multi-threaded performance than the M4.

Apple's new M5 silicon also provides a bigger 30% upgrade in graphics performance, alongside a Neural Accelerator inside of each CPU and GPU core, Apple's new M5 processor is up to 4x faster in compute capabilities over the M4. Not bad for a single generation.

Intel will be launching its new Panther Lake laptop processors as the Core Ultra 300 series, but with tweaked naming with the flagship reportedly called the Core Ultra X9 388H, with more members of the new SKUs revealed in HWMonitor. Check it out:

Intel formally announced its new Panther Lake CPUs with some details revealed at its recent Tech Tour 2025 event, but now CPU ID has added in support for the new Panther Lake CPUs and their naming convention being confirmed, and the SKUs to be launched.

HWMonitor v1.60 includes support for the Core Ultra X 3X8H, Core Ultra 3X5H, and Core Ultra 3X0U processors, with the new Core Ultra X series having four members:

Apple is expected to announce its next-gen M5 processor this week along with other product announcements, with news that the new M5 chip will be fabbed on TSMC's 3nm "N3E" process node.

In a new report from the Commercial Times, we have new details on Apple's next-gen silicon, where we can expect the next-next-gen M6 processors to be unveiled in 2026 and they could be Apple's first foray into the bleeding-edge 2nm process node for its next-next-gen MacBook laptops.

Apple's upcoming M5 processor will compete with other laptop processors including Qualcomm's newly-released Snapdragon processors and MediaTek's new chips. However, its competitors' chips are fabbed on the higher-end N3P process node, compared to the M5 reportedly fabbed on N3E.

Intel and AMD have partnered to introduce new improvements to the x86 instruction set, with the two competing companies coming together in a joint initiative to strengthen the future of x86 computing.

The joint initiative was formed in October 2024, and today marks a one-year anniversary, with Intel and AMD outlining a bunch of new standardizing features for x86. The goal of the initiative is to enhance compatibility, predictability, and consistency across x86 processor-based productions, which includes everything from supercomputers to handheld gaming devices.

The advisory group has outlined key technical milestones to include as features in upcoming x86 processors, which have been outlined below. One of the announced features is FRED, or Flexible Return and Event Delivery, which AMD and Intel intend to reduce latencies and increase software reliability with a "modern interrupt model." Additionally, AVX10 is the next-generation vector and general-purpose instruction set extension, designed to boost throughput and enable portability across client, workstation, and server CPUs.

It looks like 2026 flagship smartphones will be even more expensive than the already expensive smartphones of today, as Qualcomm's new Snapdragon 8 Elite Gen 5 processor costs more to make than previous gen Snapdragon chips.

In a new post on X from Abhishek Yadav, we have some price estimates for each of Qualcomm's flagship Snapdragon smartphone processors, ranging from the first Snapdragon 8 Gen 1 through to the just-announced Snapdragon 8 Elite Gen 5... and boy, things are getting more and more expensive with each generation of mobile chip released.

Qualcomm is reportedly charging between $240 and $280 per Snapdragon 8 Elite Gen 5 processor, and this doesn't include the royalties that the companies have to pay to Qualcomm to use their chips... so it is a little more than that on top. Qualcomm started out charging $120-$130 for its Snapdragon 8 Gen 1 a few years ago, but the price has slowly been rising and now it's closer to $300 per chip (once again, without OEMs paying royalties).

Intel's new Fab 52 semiconductor facility at its Ocotillo campus in Chandler, Arizona, produces the most advanced logic chips on American soil, and is part of the $100 billion that Intel is investing to expand its domestic semiconductor operations.

The company has advanced R&D and production in Oregon, joined by high-volume fabrication in Arizona, as well as packaging operations in New Mexico, providing Intel a unique position to support key national priorities and provide strategic capacity for Intel Foundry customers.

The company says that its Fab 52 facility builds on Intel's vast 56 years of US-based R&D and manufacturing advancement, and marks a major milestone as the company builds a trusted leading-edge US foundry for the AI era.

Intel will be launching its next-generation Panther Lake "Core Ultra Series 3" processors at CES 2026, available inside of new laptops in Q1 2026.

The new Panther Lake CPUs will feature new technologies including new P-Cores, E-Cores, Xe3 GPU cores, and NPU/IPU. Intel is combining all of this technology into a multi-tile solution that will not just be for laptops, but new AI PCs, and other products.

One of the more exciting things about Panther Lake is that the CPU tile is being fabbed in-house on Intel's new 18A process node, while the Xe3 GPU will be fabbed at TSMC on its N3E process node.

Apple's new M5 series processors are going to be an interesting change from the current-gen M4, with the M5 Pro and M5 Max chips to have a new chip design that separates the CPU and GPU blocks.

This is particularly useful for Apple as it can offer a chip with less CPU cores but maxed-out GPU cores, or vice versa, maxed-out CPU cores and less GPU cores. Apple looks to be using TSMC's new SoIC-MH advanced packaging, as well as fabbing its new M5 series processors on TSMC's new 3nm "N3P" process node, just like its new A19 and A19 Pro chips.

The regular M5 chip won't be using the new SoIC-MH advanced packaging, but the higher-end M5 Pro and M5 Max processors would be using it.

AMD has announced a new line of Zen5 CPUs, the Ryzen Embedded 9000 Series processors, designed for "industrial PCs, automation systems, and machine vision applications." AMD notes that its embedded lineup is designed to meet the "performance, energy efficiency, and long-term reliability" requirements of industrial computing applications. The new Ryzen Embedded 9000 Series ships with "seven years of product availability and reliability."

As for the lineup, it mirrors the specs and configuration of the desktop Ryzen 9000 Series, including X3D models that utilize the company's 3D V-Cache technology. The full lineup, which can be found below, includes the entry-level 6-core AMD Ryzen Embedded 9600X processor, featuring a maximum frequency of 5.4 GHz and a TDP rating of 65W, and the flagship 16-core AMD Ryzen Embedded 9950X3D processor, boasting a maximum frequency of 5.7 GHz and a TDP rating of 170W.

Like the desktop series, these Embedded variants are all built using the company's latest Zen 5 architecture, with support for PCIe Gen5, DDR5 memory, and are also based on the AM5 socket. It's unclear what separates these new CPUs from the non-Embedded counterparts, other than AMD confirming that they're designed for long-term use and reliability for at least seven years.