Ever since Intel launched its ‘Intel Inside’ campaign in 1991, its brand identity has become inseparable not only from PCs that contained its processors but also from PCs and ICT at large. Much has changed in the last 30 years, however, and PCs and ICT themselves have become inseparable parts of today’s world. So at its Vision 2025 event in Las Vegas, Intel introduced its new brand identity that connects Intel Inside to the roles that machines running Intel CPUs, as well as their owners, play in the world around us. 

“That’s the power of Intel Inside” is the centerpiece of Intel’s new brand identity. It brings back a familiar 'Intel Inside' theme from the 1990s while highlighting the role that the company, its partners, and its customers play in the modern world. By reintroducing the phrase with a new purpose, Intel emphasizes its vital role in the industry. 

Brett Hannath, Intel’s chief marketing officer, explained that the message reflects the company’s belief that its products and technologies can unlock the potential of every employee, customer, consumer, community, and partner. Ultimately, the company wants to emphasize how Intel ties together personal impact and global reach using one of the industry’s most iconic brands, with a legacy that spans over 30 years: Intel Inside. 

The original ‘Intel Inside’ campaign was launched in 1991 and marked a shift in tech marketing. While Apple, Microsoft, and various PC makers spoke directly to end-users, their suppliers mostly targeted system designers and remained in the shadows. So, with its Intel Inside campaign, Intel spoke directly to consumers, making the processor a key selling point of a PC. With advertising beyond business media, a widely recognized sticker, and a catchy five-note jingle, the campaign helped build Intel into a household name. 

This co-branding strategy encouraged PC makers to display the Intel Inside logo on their products, which made these systems more ‘familiar’ to average users and thus gave consumers confidence. As the PC market exploded in the 1990s, the Intel Inside campaign played a major role in differentiating Intel from competitors like AMD, Cyrix, IBM, and many others, highlighting Intel’s then-dominance inside PCs, which continues today, even though some believe that AMD now has better brand recognition than Intel. 

One of the most important things about Intel Inside is that it evolved with the company and its use. In the 1990s, Intel Inside promoted Pentium and Celeron CPUs as processors that largely defined user experience back then. Things changed in the early 2000s when Intel introduced its Centrino platform that bundled an Intel CPU, chipset, and wireless network adapter, which were meant to work together, improving user experience and boosting Intel’s balance sheet (while driving some smaller rivals out of business). Eventually, Intel shifted its focus toward performance tiers and user experience with its Core series (i3, i5, i7), thus transitioning Intel Inside to a more supporting role. 

While the Intel Inside slogan and logo still appear widely, Intel’s wide brand ecosystem focused on user experience has largely blurred the importance of the company’s own brand. With the new slogan — That’s the power of Intel Inside — the company is again focusing on promoting its own brand while tying it to user experience and the ubiquity of ICT today. 

Intel has discontinued its RS1 Laminar stock cooler, which was previously the lowest-end model among the company’s three stock coolers.

Introduced in 2021 alongside 12th Gen Alder Lake CPUs, the RS1 uses an all-aluminum design with a plastic shroud that’s just 47mm tall and weighs around 260 grams. It’s essentially an upgraded version of Intel’s old stock cooler, with a more optimally designed heatsink and a larger, five-blade fan, and is rated to cool 65 watt chips.

Intel also has two other stock coolers, the RH1 and the RM1. The RM1 is almost identical to the RS1, except it has a copper slug at the bottom for better heat transfer, as well as an LED light strip that circles the fan. Like the now-discontinued RS1, the RM1 is also rated for 65 watts. The RH1 meanwhile is a wholly different design that’s taller, uses more copper, and a more premium metal shroud.

The RS1 only ever shipped with two dual-core CPUs: the Pentium Gold G7400 and the Celeron G6900. Although Intel’s weakest stock cooler isn’t particularly robust, it doesn’t struggle to cool either the G7400 or the G6900 since both have a TDP of just 46 watts.

That’s probably one of the main reasons why it got discontinued. It’s hard to justify making a stock cooler just for two CPUs, let alone just for dual-core Pentiums and Celerons, which aren’t exactly popular in recent times.

Intel says that the RS1 won’t have a direct successor, but will just be replaced by the RM1. Simplifying its stock cooler offerings could have also been another motivation for Intel to ditch its all-aluminum cooler since the RS1 and RM1 are already nearly identical. The RM1 is already used for the vast majority of locked Intel CPUs, everything from the Core i3s and the Core i7s in the 12th, 13th, and 14th generations.

The RM1 is also used for the Intel Processor 300, a rebranded Pentium G7400. The Processor 300 came out just around a year ago, and the fact that it used the RM1 may have been a sign that the RS1 wasn’t long for this world.

Do you miss the days when Microsoft PDAs roamed the Earth, or the more recent Windows Phone era? Then Japan’s Gloture might be able to tempt you with its new NanoPC (h/t PC Watch). This new product is a highly compact PC with a full-cover 5.5-inch touchscreen on one side, built in battery, and Windows 11 Pro.

Compared to smartphones in 2023, the NanoPC doesn’t look very ergonomic with large bezels, sharp-looking edges, and a thick build. On the positive side, you get Windows 11 in your pocket, and this device bristles with connectivity options and ports including Gigabit Ethernet.

With our description above, and the various images shared in this article, some of you might be thinking ‘wow’, but we are sure another sizable contingent will have responded with a ‘why?’ So, let’s look into the specs:

One of the standout features of the Gloture NanoPC is how much connectivity it packs in. In the promotional imagery for its NanoPC, Gloture shows this device used as a pocket desktop with two extra large screens attached. Users get a decent array of (mostly) full-sized industry standard ports in a world where smartphone makers are starting to be tease devices with fewer and fewer ports.

We feel that the Gloture NanoPC, in this iteration, isn’t looking to tread on the toes of phone makers, due to some major design decisions. Firstly, this new pocketable PC device doesn’t have a SIM card slot (or eSIM). Moreover, it has a built in mic, but no camera of any kind. Lastly, even on such a small device with a 10 W Celeron, and a small 5.5-inch screen, we don’t think 2,500 mAh is meant to provide more than a couple of hours portability away from a wall socket.

Update - Aug. 9, 2023 - 4:54 p.m. ET - An earlier version of this article stated that the  chip's reported specs included a 3.6GHz base frequency; this has been updated to reflect the alleged specs actually mentioned a 3.9GHz base frequency.

Just when we thought dual-cores were dead, Intel is rumored to be bringing back the "obsolete" core configuration for another generation. A Tweet by leaker chi11eddog alleges Intel is developing a new 14th Gen entry-level dual-core chip that will feature one of Intel's three hybrid CPU architectures found in its Best CPUs, Alder Lake, Raptor Lake, or Raptor Lake Refresh.

According to chi11eddog, this new chip will be the spiritual successor of the Pentium Gold G7400. Thanks to Intel's removal of the Celeron and Pentium sub-brands altogether, the new chip will be dubbed the "Intel 300," similar to other entry-level chips (like the "Intel Processor N200").

The chip's reported specifications include a 3.9GHz base frequency (no turbo clock), a 6MB L3 cache capacity, and a core configuration consisting of two P-cores with four threads. Power consumption is rated at 46W. The Twitter leaker did not reveal any specifications for the iGPU, but for obvious reasons, we expect this CPU to sport one of Intel's lower-end configurations — like its UHD 730 Graphics unit.

Compared to Intel's other entry-level CPUs like the Core i3 series, this "Intel 300" will be the cheapest and least powerful entry in Intel's desktop lineup. The chip features two fewer cores than the Core i3-13100 (Intel's lowest-end i3 right now), and a much slower clock speed thanks to the lack of Turbo Boost technology. As previously stated the Intel 300 will also effectively be the replacement of Intel's previous-generation desktop Celeron and Pentium processors which also sported dual-core configurations.

Because of its dual-core design, we don't expect this chip to be great for anything beyond casual web browsing, office work, and video streaming. Even with the addition of Hyper-Threading, the physical limits of dual-core designs make them sub-optimal for gaming tasks, especially in modern game engines which can take advantage of 6 six CPU cores.

Chi11eddog believes the CPU will be arriving as soon as this quarter, so we shouldn't have to wait long for this new entry-level chip to arrive on store shelves — assuming it exists. With its 14th Gen moniker, the new dual-core will allegedly arrive in conjunction with Intel's 14th Gen Raptor Lake Refresh lineup which is also expected to launch very soon.

Intel's Alder Lake-N family of processors aimed at entry-level PCs has its own hierarchy and includes multiple options, including the flagship eight-core Core i3-N305, mid-range quad-core N97, and low-end dual-core N50. The latter belongs to a segment that represents little interest to performance-minded users, which is why such chips never get reviewed. But someone shared test results from the Intel Processor N50 in Geekbench (via @BenchLeaks).

Intel's Processor N50 features two general-purpose cores based on the Gracemont microarchitecture. Operating at 3.40 GHz, the chip comes with UHD-badged Xe-based graphics core with 16 EUs at 750 MHz and supports up to 16GB DDR4, DDR5, and LPDDR5 memory. With a thermal design power of 6W, the CPU can serve various applications, including laptops, thin clients, compact desktops, and many other low-power things. What the SoC is not meant to offer is, of course, high performance.

Indeed, even single-core results of the Intel Processor N50 in Geekbench 6 CPU tests are significantly lower than those of the Processor N97 and Core i3-N305, which is unsurprising given its lower frequency. The multi-core score is predictably over two times lower than the N97 since this is a dual-core CPU. 

As far as OpenCL compute performance is concerned, the Processor N50 comes with a severely cut-down integrated GPU that only has 16 execution units and runs at only 750 MHz. It cannot even compare to iGPUs of the N97 and N305 since they have more EUs that run considerably faster.

Of course, Intel's N50 can run productivity applications just fine, especially if it does not need to compute large XLS files or perform other resource-demanding tasks. Still, this CPU is generally meant to sit in a very simplistic machine that uses cloud resources or processes light workloads.

Geekbench 6 is a synthetic benchmark that does not necessarily represent performance in real-world applications, but when a CPU is times behind even entry-level offerings, it is evident that it is slow.

Intel announced today that it is overhauling its consumer CPU branding for the first time in 15 years, resulting in a simplified naming scheme that looks somewhat similar to AMD’s Ryzen naming scheme. Intel’s new approach focuses on its ‘Core’ branding and splits the chips into ‘Core’ and ‘Core Ultra’ tiers. It also drops the iconic ‘i’ from the i3, i5, i7, and i9 chip identifiers and no longer refers to its processors with a generation moniker, like ‘13th-Generation.’ It also includes new badges, all of which we’ll cover more below. These changes begin with the Meteor Lake processors that come to market later this year but don't apply retroactively to existing chips.

Intel’s announcement comes after the new chip naming scheme emerged a month ago when new ‘Ultra’ badged processors were listed in processor benchmarks on a public database. Intel’s latest rebranding follows other retoolings of its public image, such as changing the iconic Intel logo and jingle, with the latter receiving more than a fair bit of negative feedback. Intel has also recently discarded other Intel hallmarks, like the Pentium and Celeron brands that were once household names. That change also wasn’t well received.

Intel says the Meteor Lake chips are “an inflection point for design, manufacturing, and architecture and delivers significant advances for our company and customers,” and that the changes are the result of customer requests to simplify its brand.

Intel will now split its future chip generations into two tiers: 'Core’ mainstream models and ‘Core Ultra’ premium chips, with the latter indicating a higher tier of performance. Intel has also dropped the generation designator from its brand name, but you can still easily decode that from the product number. For example, the ‘13’ in 13900K indicates the chip is a 13th-Generation model, and that portion of the naming scheme will remain consistent with the coming generation. Intel also has new badges, which you can see in the above album.

Perhaps the biggest outward change is the decision to drop the ‘i’ from the Core i3, i5, i7, and i9 lines. Here's a mockup of how the old branding would have looked for the coming 14900K chips, and a few examples of how it will look under the new official scheme:

• OLD: Intel 14th-Generation Core i9-14900K Processor
• NEW: Intel Core Ultra 9 processor 14900K
• NEW: Intel Core 9 processor 14900K
• AMD: Ryzen 9 7950X

As you can see, Intel also includes the word 'processor' in the official product name. This is technically already present in the current branding scheme even though absolutely no one uses it in common usage -- even Intel doesn't. However, now Intel inexplicably injects the word between the family designator (Core 9, Core Ultra 9, for instance) and the model number, instead of at the end of the name.

We asked Intel's brand team about this, and they said they 'prefer' for the term 'processor' to be used, but that the real focus is on the 'Core' and 'Core Ultra' branding. Given that even Intel itself doesn't use the existing 'processor' designator already, we doubt that the new superfluous 'processor' portion of the branding will catch on. Here's another mockup, but this time of how the branding will look in common usage:

• OLD: Intel 14th-Generation Core i9-14900K
• NEW: Intel Core Ultra 9 14900K
• NEW: Intel Core 9 14900K
• AMD: AMD Ryzen 9 7950X

Make no mistake, this will probably be the most common usage you'll see in reviews and in general reporting. As you can see, without 'processor' injected in the brand string, Intel's new mainstream chip branding is very similar to AMD's Ryzen branding. Yes, the Core Ultra-branded parts will have the extra bit of 'Ultra' to differentiate, but the new scheme feels very Ryzen-esque without the 'i' in Core i9. More on that below.

Intel hasn’t confirmed yet, but we do know from a mountain of external evidence that its next generation of chips will have two swim lanes — processors based on Meteor Lake, a totally new 3D-stacked design with a new microarchitecture, and the Raptor Lake Refresh models, which use the same design and microarchitecture as the currently-shipping processors but have extra tuning to improve performance.

We spoke with Intel’s branding team and asked if the new Core Ultra branding will apply to the newer Meteor Lake models while the refresh models will fall under standard Core branding, and/or if the overclockable K-series processors would be confined to Ultra-branded products only.

Intel didn’t confirm that only Meteor Lake chips would only be marked as Core Ultra but did say that tier ‘represents the latest innovations’ and that they would share more details soon. That means it is at least a rational assumption. However, the representatives did say that overclockability (K-series) isn’t a requirement for a chip to be branded as Core Ultra.

Thoughts

It is certainly a bold yet questionable decision to remove the branding that hearkens back to when Intel had a near-monopoly on the market and was the unquestioned performance leader for over a decade. The perception that Intel is diluting the hard-fought brand recognition built during those years is obviously going to be present, regardless of the market studies that Intel tells us it used to inform the decision.

Intel’s branding change will certainly be confusing for those that have become accustomed to the naming scheme over the last 15 years, but Intel tells us this change is designed to improve brand recognition with mainstream audiences that aren't tech-savvy. In fact, Intel feels that the 'Core' branding is already, well, the core of its brand equity, so it thinks dropping the 'i' sharpens focus on the 'Core' brand.

Today, Intel and AMD are more closely matched than we’ve seen over the last decade. AMD’s Ryzen processors did temporarily catapult the company back into an unquestioned leadership position, but Intel’s 13th-Generation Raptor Lake restored Intel’s overall lead with the best blend of performance and value for the bulk of the desktop PC market. Yes, AMD's Ryzen 7000 still holds the overall gaming lead with its premium lineup of X3D parts, but they carry a premium that relegates them to the highest-end systems, so they aren’t for most users.

That makes the timing of Intel’s rebranding even more confusing. It certainly isn’t uncommon for an underdog to adopt a similar branding scheme as the incumbent, often to leverage the better-known product branding from the dominant player to further their own brand — that’s exactly what AMD did when it launched its Zen-powered processors with a very familiar Intel-like Ryzen 3, 5, 7, and 9 scheme. In fact, AMD even adopted a very similar motherboard chipset branding to Intel, too.

Intel is certainly back on the upswing in the desktop PC market, but the optics of discarding the familiar 'i' portion will be sure to generate plenty of criticism. Though it probably wasn't intentional, some will also insist this looks like AMD's branding scheme for Ryzen. But, as mentioned, AMD first adopted an Intel-like branding scheme, making it hard for Intel to simplify its branding without ending up with a similar-looking approach.

Intel says it isn't sharing all of the details of the new branding scheme yet, and its slides indicate that it is still in the decision-making process on some of the details. That seems a bit dubious, given the proximity of the Meteor Lake launch later this year — surely these decisions have already been made. In either case, we'll learn more about the branding scheme when Meteor Lake arrives later this year. 

MSI brought a ton of motherboards  to Computex, the annual hardware trade show in Taipei. Among its wares are upgraded Z790 Max motherboards and its Project Zero motherboard and case that support connections at the back of the PC.

Most of these aren't coming until at least Q3 2023, but that means we'll see a bunch of faster connectivity and clean build options later this year.

Z790 Max Motherboards and USB4 PD100W Expansion Card

There are four new Z790 Max models, supporting 12 and 13th Gen Intel Core, Pentium, Gold and Celeron processors in LGA 1700 sockets. They are the MEG Z790 Ace Max, MPG Z790 Carbon Max Wifi, MPG Z790 Edge Max Wifi, and MAG Z790 Tomahawk Max Wifi.

The MEG Z790 Max Series adds Wi-Fi 7 and 5 Gb Ethernet LAN compatibility, though the LAN is limited to just two models: the Ace and the Carbon. These boards will be tool free, with M.2 "Frozr" shields using magnetic designs instead of screws. Additionally, MSI is promising an exclusive "DDR5 enhancement technology" it calls Memory Boost. On signage at its Computex booth, it showed up to DDR5-7600 on overlock on the Ace and Carbon, with DDR5-7200 with an OC on the Tomahawk and Edge.

These boards will have Lighting M.2 Gen 5 PCIe slots and USB 3.2 Gen 2 x2, while the Z790 Ace Max will get Thunderbolt 4. The Ace Max, an E-ATX board, will have the most advanced power system with MSI's 24 Duet Rail power system. The MPG boards, — the Carbon Max and Edge Max — appear to be standard ATX boards, with the Carbon packing 19 power phases and the Edge using 16 mirrored power arrangements.  The Tomahawk will have the same arrangement as the Edge.

In most other regards, these four boards are similar to the existing Ace, Carbon, Edge and Tomahawk, but with faster networking, the tool free M.2 covers and faster DDR5 overclocking.

The company didn't announce pricing for the Z790 Max line, but it suggests that they'll launch near the end of Q3 2023.

MSI also has a new expansion card, the USB5 PD100W, which MSI reps at Computex told Deputy Managing editor Paul Alcorn will likely be included with at least one of the new motherboards. The card has two USB 4.0 Type-C ports that support up to 40 Gbps and display support, with up to 100W power charging for one port and 27W for the other. There are also two DisplayPort 1.4 input ports, possibly for capturing from other devices.

The expansion board could come to retail, those spokespeople said, but MSI wouldn't commit to that. In its official listings, MSI doesn't have a price or release date. 

Project Zero

MSI also brought its own take on the rear-connect motherboard, Project Zero. This appears similar to Gigabyte's Project Stealth using a special case that supports connections on the rear, or perhaps Asus' DIY APE standard. It appears MSI is using its own design, though. Reps told Paul Alcorn that MSI changed its designs to avoid patent issues.

The Project Zero board on display at Computex is a take on an AMD B650M board, which MSI partnered with a Ryzen 7 7700X. It was sitting in a take on its own case, the MAG Vampiric Project Zero with a  a 270-degree panoramic glass window.

The motherboard will launch for $219, while the case's price has yet to be determined. We should see these components in Q3.

The $648 (review configuration) LattePanda Sigma is a beast of a SBC (single board computer). Packaged deep under a dominating heatsink and fan is a 13th gen Intel I5 CPU, 16GB of DDR5 RAM and an Intel Iris XE GPU (that is good enough for light gaming).

Almost a year ago we reviewed the LattePanda 3 Delta, which was an Intel Celeron N5105 based system and onboard Arduino co-processor. We felt that it was the all-in-one desktop for makers. Providing a full x86 CPU and our choice of operating system, Arduino support, and it could play a decent game from the early 2010s. Fast forward to 2023 and the Sigma’s specifications blow the Delta out of the water. We get more RAM, better CPU and GPU and a plethora of upgrade options that use off the shelf components. Heck, the CPU and GPU upgrades mean we can play some games from 2022!

We put the LattePanda Sigma on the bench and tested how it performed as a maker desktop, and we also did some light gaming.

LattePanda Sigma Versus Delta Specifications

Using LattePanda Sigma

The LattePanda Sigma is powered by a 13th generation Intel Raptor Lake CPU, the i5-1340P to be specific. This CPU has 12 cores in total: four performance cores (max turbo 4.6 GHz) and eight efficiency cores (max turbo 3.4 GHz). There are 16 threads available (via the e-cores). This mix of cores provides plenty of punch when we need it, and a good level of general performance. The more powerful CPU means that the Sigma is larger than the 3 Delta, which features an Intel Celeron N5105. It also means that the Sigma is way larger than a Raspberry Pi 4, approximately 2.5 times the area. 

As the Sigma is an x86 based SBC, we are free to use Windows or a Linux distribution of our choice. Our review unit came with a 500GB WD Black SN770 NVMe SSD with an unregistered version of Windows 11 pre-installed. LattePanda supplies a range of custom operating system images, but you can just install your own choice and tweak it to work.

Boot time for the LattePanda Sigma is fast. At 20.92 seconds the Sigma is a close second to the Raspberry Pi Compute Module 4 with eMMC which clocked in at 20 seconds. Raspberry Pi OS is a lighter OS than Windows 11, so the difference is remarkable given Windows 11’s relative “bloat”. 

A fairer comparison would be Windows 11 on an NVMe drive with the LattePanda 3 Delta. Coming in at 70.13 seconds, the Delta was much slower than the Sigma. This is to be expected given that the Delta has a PCIe 3.0 interface, versus the PCIE 4.0 of the Sigma. We also compared the Raspberry Pi 4 (micro SD) boot time of 30 seconds, and Khadas VIM4’s 36.38 seconds boot for reference.

We noticed that the PCIe 4.0 interface used for the main drive was able to keep up with our WD Black SN850X, with a constant read speed of just under 7GB/s when using Ubuntu 22.04.

We have an AMI BIOS which affords us a great deal of configuration. We can tweak almost every aspect of the Sigma, converting it into a low power desktop, or we can alter the CPU settings to grant us a little more computational power. 

Storage options with the Sigma are condensed when compared to the Delta. Gone is the eMMC. Instead we have three M.2 PCIe slots. The main slot is for PCIe 4.0 NVMe, second is PCIe 3.0. The final slot is M.2 B Key and can be used with a SATA III drive or PCIe 3.0 x1.

All of these options are impressive for what is really a turbo-charged SBC. But whereas we had to manage our expectations with the Delta, the Sigma can be your desktop PC and the brains behind a robotics or machine learning project. The Arduino element is there to control the motors, sensors and other components that make the project move, but the Raptor Lake CPU is there to give your projects the brains.

Gaming is more than possible. The Delta saw a hard limit of mid 2010s era games, but with the Sigma we have an Intel Iris Xe GPU with 80 Execution Units running at up to 1.45 GHz. That is a considerable amount of power for an “SBC” and it means that we can push gaming a little farther. 

We tested Stray, a 2022 indie game which sees the player investigating a mystery in a futuristic world. Oh and you are a cat. To get a constant 60 fps you’ll need to drop the resolution to 720p and everything else to low. If you can survive 30 fps, then 1080p with medium settings is possible, but your framerate will take a dip in congested areas. We noticed a large drop to 8 fps when the game loaded in a new area. 

This prompted us to enter the BIOS and turn off energy saving features. This gave us a little more GPU power and our framerates hovered around 40 fps. As we said, 1080p 30 fps is the sweet spot for Stray. 

We also tested Warhammer 40,000 Boltgun, a game that has just been released. This “boomer shooter” looks like the original Doom, but on modern hardware. It ran well at 1080p, most of the time way above 60fps. There were some dips, but if the fps counter wasn’t there, we would be hard pressed to notice. 

How about something a little older? I went through my Steam library and pulled out the original Call of Duty 4: Modern Warfare and this is where I hit a small issue. Matching my monitor’s resolution of 2560 x 1440 was a breeze, as was using high quality textures. The issue was moving the player to the firing range caused the game to hang, every time! 

The solution seemed to be running the game in safe mode, performing the player orientation and then turning everything back to my preferred settings. I was soon on a container ship in the middle of the ocean with Captain Price and co. If you want the best gaming performance, then you need to think outside of the box. 

The USB-C connectors offer Thunderbolt 4, so with an external enclosure and one of the best GPUs you can unlock plenty of potential. Another means is to use one of the M.2 slots and an Occulink connection with an external GPU. This would provide similar functionality to Thunderbolt 4. YouTuber ETA Prime has connected an Intel ARC A750 to the LattePanda Sigma using Occulink and has seen remarkable results.

Design of LattePanda Sigma

There is no denying that the LattePanda Sigma is a beast of an SBC. It is more like a Mini PC than a true SBC. Dominating the top of the Sigma is a large heatsink and fan. It’s so large, in fact, that ports and pins are cut out of it for easier access. The bulk does not mean that the Sigma is undesirable. All of the ports are located on the long sides, and on the short sides are pinouts for SATA, USB, fans and even a front panel connection. Flipping the Sigma over and we are greeted to a large heatsink.

Removing the four screws and we find the pre-installed 500GB NVMe drive in the PCIe 4.0 slot. We also find the Intel Wi-Fi / Bluetooth card which we included in our review unit. The underside of the Sigma is well designed, with plenty of space for adding drives, and direct contact between the drives and the heatsink.

On the same heatsink are four screw holes, which do not appear to conform to a VESA mount. These screw holes can be used to mount the Sigma to a fixed point, but you’ll need to work out your own fixing solution, perhaps something using the best 3D printers?

Cooling LattePanda Sigma

The Intel i5-1340P Raptor Lake CPU uses Intel 7 lithography and is designed with low power / mobile devices in mind. As the Intel i5-1340P is a much more powerful processor than the N5105 of the Delta, we can expect to see a higher temperature.

We tested the Sigma and Delta using Windows 11, y-cruncher and a PinePower 65W USB C power outlet with realtime power metrics. The Delta hit a max temperature of 80 degrees Celsius. The Sigma, well the performance cores hit 100°C and the efficiency cores hit a high of 88°C. Luckily the included beefy heatsink and fan are whisper quiet, unlike other SBC cooling fans we have tested.

The idle temperature for the Sigma was 25°C compared to the Delta’s 41°C. We tested the idle temps by leaving the board idling for 10 minutes. So the take away from this test is that the Sigma can really heat up when under load.

According to its datasheet, the Intel i5-1340P chip runs between 28 and 64W, but in our tests, we noted that at idle, the Sigma pulled 6W. Under stress, via y-cruncher we pushed the LattePanda Sigma to pull 66W! How does that compare to the Delta? At idle, the delta pulled 4.56W, under stress 19.76W. Again this all boils down to the i5-1340P CPU.

Let's look at the power consumption. Under stress, the LattePanda Sigma consumes 10 times more power than the Raspberry Pi 4 and Khadas VIM4 (60W versus 6W). If you are building a project with a power budget, for example robotics or off-grid data collection, then perhaps the Raspberry Pi or Khadas VIM4 is more applicable.

If your power budget needs to be even lower, perhaps a microcontroller such as the Raspberry Pi Pico W is a better option. That said, the extra power consumption can be factored into your build and if you really need that much processing power in a portable form factor, the Sigma packs a real punch.

LattePanda Sigma GPIO

The GPIO of the LattePanda Sigma is different to that of the Delta. We get less GPIO pins than the Delta, but what we get is pretty much an Arduino Uno in a different form factor. The GPIO is for the onboard Arudino Leonardo compatible coprocessor. It seems that an Arduino compatible co-processor is quite useful for Intel based SBCs. The LattePanda 3 Delta had its own (and a breakout for RS232 and audio) as does Seeed’s Odyssey.

The double row of headers has a pinout printed on its side. Reading the pinout was simple and clear, in fact we prefer it to the Delta which saw its heatsink get in the way.

Controlling the Arduino GPIO is possible using the pre-installed Arduino IDE, an older version of the IDE (1.8.19) which works, but would’ve liked to have seen v2.0 onwards. In fact, we installed the latest Arduino IDE and set the board type to Leonardo, but the IDE would not upload to the board. We attempted to install the board, but alas it didn’t work. Not a massive loss, but the latest Arduino IDE is leaps and bounds over its predecessors. 

Using the pre-installed IDE, we tested the Arduino GPIO with the humble LED, blinking the LED every half second. With that success under our belt, we connected a one meter length of WS2812B NeoPixels and installed the appropriate libraries. Moments later, colors erupted from the chain of RGB LEDs.

This GPIO is a better compromise than the Delta, and for Raspberry Pi alternatives, it is pretty good. Other Raspberry Pi alternatives have attempted to ape (hardware and software) the Raspberry Pi format, and they have varying levels of success. By using a simple pinout form factor, while retaining compatibility with Arduino code and components, the LattePanda Sigma provides the GPIO access that we need. Sure the pinout isn’t Arduino shield compatible, but we can use a few wires to bridge the gap.

LattePanda Sigma Linux Performance

If Windows isn’t your thing, we hear you. Linux is generally a better fit for SBCs and so we installed the latest LTS (Long Term Support) release of Ubuntu 22.04.2 then fully updated the OS. Our installation was contained on a 1TB WD_Black SN850X  NVMe SSD so that our previous Windows installations remained untouched. 

We’d love to say that “Ubuntu just worked”, and while there were no installation issues and it detected our hardware correctly, there was an issue that existed between 22.04 and 23.04. The issue was that the live USB failed to fully boot, complaining of “unable to find a medium containing a live file system.”

A quick Google search revealed that this has been a problem for other computers. The solution? Weirdly it was to remove the live USB during the boot process (just as the Ubuntu logo flashes on the screen) for two seconds, then put it back in. Yes, that worked. I installed the latest Arduino IDE and hit the same Arduino IDE issue as Windows. So I followed the LattePanda 3 Delta guide to configure the LattePanda Leonardo board, but this only works with the older Arduino IDE. Suffice to say I was able to configure the board with Arduino 1.8.19 and complete a series of GPIO tests. 

Ubuntu 22.04.2 is a slightly lighter OS than Windows 11. It felt no different than Windows 11 for its responsiveness,largely down to the powerful Intel CPU and the generous 16GB of DDR5 RAM. We could happily sit down and use Linux on the Sigma for our general work and to write code. We’d leave the gaming side of things to Windows as even after installing the latest drivers, configuring Steam to use Proton and then downloading shaders, both Stray and Warhammer: Boltgun failed to launch.

LattePanda Sigma versus LattePanda 3 Delta versus Raspberry Pi 4

The Raspberry Pi 4 is, like it or not, the benchmark from which other SBCs are measured. Sure it's not a desktop replacement and the hardware package is now five years old (pandemics and global supply chain issues artificially extended the shelf life of the Pi 4), but it is a good all-rounder from which we can measure.

The LattePanda 3 Delta is much more powerful than the Pi 4; we proved that in our review. But the Sigma screams past them both in terms of raw performance, as it should given the 13th gen Intel CPU and 16GB of DDR5. But this performance comes at a price, around $648 for our review configuration. This is just over double the price of the LattePanda 3 Delta and for $648 we can buy eight Raspberry Pi 4 8GB at non-scalper prices.

What about the Jetson Orin Nano? The $499 Jetson Orin Nano is a close contender with an Nvidia GPU featuring 32 Tensor Cores, a six-core Arm A78AE v8.2 64-bit CPU and 8GB of LPDDR5. But we have to reserve our judgment until our review is complete. We had to postpone the review due to software issues at the time of its release.

Bottom Line

If you need lots of computation power and a GPIO then the LattePanda Sigma is a contender for your cash. You will need to justify the spend as $648 is a lot of cash for what is essentially a mini PC with an embedded Arduino. That said, the package is sublime and clearly well thought out.

The $648 price tag is initially hard to swallow but again this isn’t just a replacement for the Raspberry Pi, it is a desktop computer in a slightly larger SBC form factor. You could happily use this machine for day-to-day work, while using less electricity than on a traditional desktop. Or you can embed it into a robotics / autonomous / AI build and harness all of the CPU cores to perform complex operations while on the move.

Arduino compatibility is still the icing on the cake for the LattePanda machines. It is easy to use and works with the many thousands of libraries available in the Arduino ecosystem. Sure, we can’t directly use Arduino shields, but unless you have bought into that system, you aren’t going to miss them.

The LattePanda Sigma is a revolution compared to the Raspberry Pi 4. Compared to the LattePanda 3 Delta, it is an evolution, and one that will entice makers craving more power for their builds. The Sigma is the new maker’s desktop, for makers that need the extra horsepower.

Hackboard today launched the Hackboard 2, a compact and affordable Wintel SBC. Key features of this 4.7 x 3.1 inches (120 x 80 mm) board are its Intel Celeron N4020 dual-core CPU, its good selection of ports and connectivity, the licensed Windows 11 Pro OS, and its $199 price tag.

The Hackboard makers tout it as an ideal maker board for general x86 hobbyist projects, IoT projects, for work from home folk, for edge and embedded solutions, and more. In a press release, Mike Callow, Hackboard's co-founder and CEO said that the Hackboard 2 is also designed "to address the increasing worldwide challenge of the Digital Divide and the needs of education." Those are big asks for a 2C/2T 14nm Celeron-powered SBC.

Hackboard is working to put together some kits built around the Hackboard 2, featuring extras such as a case, a wireless keyboard / touchboard, a webcam, and a portable monitor. There are also optional 4G and 5G cellular modems to expand the connectivity options already on offer.

Regular readers may remember the Hackboard 2 from its crowdfunding stage, way back in Dec 2020 / Jan 2021. However, this SBC is now available from a wide range of resellers, including RS Components, OKdo, Micro Center, Mouser, CrowdSupply, and Vilros.

Perhaps of some importance to the chance of success of the Hackboard 2 is the change in pricing. During the crowdfunding stage, there was an appealing $99 option, which is now gone. Now the cheapest model post-launch model is $175 with Debian Linux, or $199 with a Windows 11 Pro license. With the new price, the above Raspberry Pi 4 comparison might not be as fair as it used to be. And there are full Windows systems with newer hardware available for less money. Having said that, a supply / demand imbalance has affected the Pi in recent months and caused significant price inflation -- when stock is available. So for those looking for a project board who are inclined toward x86, the Hackboard 2 offers up an interesting alternative.

According to two new Product Change Notifications (PCNs), Intel has discontinued the company's Gemini Lake Refresh (GLK-R) processors. The 14nm Pentium and Celeron-branded chips, launched in 2019, were commonly found inside entry-level and ultra-low powered desktops, laptops, mini-PCs, and AIOs.

The original Gemini Lake (GLK) lineup came out in 2017, leveraging Intel's Goldmont Plus microarchitecture. The 14nm low-power chips, which span up to four cores without Hyper-Threading, were the direct replacements for the chipmaker's Apollo Lake processors at the time. However, given Gemini Lake's prevalence in budget-oriented devices, Intel refreshed the lineup in 2019 with moderately improved clock speeds on the 10W and 6W processors.

Intel retired Gemini Lake in 2020, ending a three-year run for the low-power chips. Therefore, Gemini Lake Refresh has had a slightly longer run, but all good things must end. Intel is handling the discontinuance with two different groups. The first group consists of the Celeron N4120, Celeron 4020, and Pentium Silver N5030, while the latter group encompasses the Celeron J4025, Pentium Silver J5040, Celeron N4020C, and Celeron J4125. Intel's clients have until November 24, 2023, and August 25, 2023, to put in their last orders for the first and second retirement groups, respectively. The chipmaker will ship the former by May 24, 2024, and the latter by February 23, 2024.

Intel Gemini Lake Refresh CPUs

Goldmont Plus ultimately has grown long in the tooth and was already been replaced by Tremont in 2020, so it makes sense for Intel to wind down 14nm production and focus on Tremont instead. Tremont powers the 10nm Elkhart Lake and Jasper Lake processors, with the latter being the successors of the Gemini Lake Refresh chips.

Jasper Lake features Tremont cores, which offer a considerable IPC uplift over Goldmont Plus. By Intel's estimates, Tremont has up to 32% higher IPC than Jasper Lake while retaining the same thermal conditions. This means 10W for J-series SKUs that target desktops and 6W for the N-series SKUs that cater to mobile devices. Most of the performance increase is thanks to the Tremont cores since Jasper Lake still maxes out at four cores and only comes with slightly better clock speeds.

With Gemini Lake Refresh almost out of the picture, upcoming really low-budget devices will arrive with Jasper Lake processors as the default. It's excellent news since vendors will have no choice by to use the newer 10nm chips that will offer better performance for budget-conscious consumers.

The budget CPU market continues to be important, at a time when GPU, RAM, and storage prices have shot through the roof. Both AMD and Intel have options if you’re on the hunt for the best budget CPU, but you’ll need to look back a few generations to get the best bang for your buck. If you can stretch your budget higher, our roundup on the best CPUs for gaming will give you entry-level, mainstream, and premium options, while our extensive CPU benchmark hierarchy will show you how processors compare across several generations.

First, we need to define what “budget” is. For this guide, we’re considering CPUs under $200. That’s higher than previous versions of our rankings, but it reflects the reality of CPU prices today. The other consideration is RAM. DDR5 prices are stratospheric, and you’ll likely spend more on a kit of DDR5 than on your CPU if you build a PC from scratch today. We’ve included some DDR4 options, but the unfortunate reality is that DDR5 is now the standard. That adds a significant burden to any cheap CPU, and there isn’t any way around it.

If you’re looking for the go-to best cheap CPU, the Intel Core Ultra 5 250K Plus is what you’re after. It’s right at the high-end of our budget range, and it exclusively supports DDR5, but it’s a CPU that should last any budget builder several generations before an upgrade. It matches AMD’s competitive six-core options in games, and it offers nearly twice the multithreaded performance in applications.

For pure gamers, however, the Core Ultra 5 250K Plus isn’t at the top of the pile. For that crowd, the Ryzen 5 7600X3D is a better choice. It’s now available for purchase online, allowing a much broader audience to pick up the Micro Center-exclusive CPU. At $230 online and $200 in-store, there isn’t another CPU that’s remotely as fast in games.

Both chips require DDR5 memory, but we’ve included some DDR4 options, as well. The Ryzen 5 5600 is a decent all-around CPU if you’re stuck on the AM4 platform, and it works with DDR4. For the most budget-conscious buyers, we recommend the Core i3-14100F. It’s below $100, available on both DDR4 and DDR5 platforms, and even comes with a decent bundled cooler considering the TDP.

Rounding out our list if the Ryzen 5 5600GT, which is an aging APU that we wouldn’t recommend under normal circumstances. AMD’s newer Ryzen 5 8600G is faster, more readily available, and not much more expensive. However, if you don’t have the money to spare on a discrete GPU, you probably don’t have the money for a kit of speedy DDR5, and the Ryzen 5 5600GT is the DDR4 APU we’d recommend right now.

We’ll provide more details on all our picks for the best budget processors below. For now, here’s the shortlist of our picks.

Prime Day Exceptional CPU deals

Here is a standout budget CPU deal from the Prime Day event, which is currently taking place. See our best overall picks below.

Best Budget CPUs in 2026 at a Glance

Best Budget CPU ($150-$200):

1. Intel Core Ultra 5 250K Plus

Best Budget CPU ($100-$150):

2. AMD Ryzen 5 7600X3D

Best Alternative Budget CPU ($100-$150):

3. AMD Ryzen 5 5600

Best Entry-Level Budget CPU (<$100):

4. Intel Core i3-14100F

Best Budget APU:

5. AMD Ryzen 5 5600GT

Best Budget CPUs 2026

Best Budget CPU ($150-$200)

Although Intel’s new Core Ultra 5 250K Plus carries the Core Ultra 5 name, it’s priced more like a budget chip. It launched at $200, and you can find it now for around $220, and there isn’t another CPU that comes close at that price. The Core Ultra 5 250K Plus is faster than AMD’s Ryzen 5 7600X and 9600X in games, and nearly twice as fast as the latter in multithreaded application performance. There are valid concerns about upgradability given that we don’t expect new CPU releases on the LGA 1851 socket. However, if you want to get the best balance of performance for $200 today, the Core Ultra 5 250K Plus is a clear choice.

That’s partially because it’s built like a higher-class chip. It comes with 18 cores (and threads), split across six Lion Cove P-cores and12 Skymont E-cores. It clocks up to 5.3 GHz, nearly matching the Core Ultra 7 265K, and it comes with a healthy 60 MB of combined L2/L3 cache. Beyond standard Arrow Lake, the Core Ultra 5 250K Plus also officially supported DDR5 speeds up to 7200 MT/s. The chip itself is unlocked for overclocking on Z890 motherboards, which also support PCIe 5.0.

The main rub with the Core Ultra 5 250K Plus is memory. It exclusively uses DDR5, unlike previous-gen Raptor Lake Refresh chips, which drives up the platform cost significantly. You can easily drop double (or more) on DDR5 as the cost of the Core Ultra 5 250K Plus right now. It’s a good option if you already have a kit of DDR5, or if you need to bite the bullet and upgrade regardless.

Although we’re recommending the Core Ultra 5 250K Plus here, the Core Ultra 5 250KF Plus is available, as well. It’s about $10 cheaper and identical to the base model, short of the integrated graphics, which are cut from the KF version. For memory, keep our best RAM combo deals page bookmarked, which we update daily. We constantly see bundles bringing together a CPU, motherboard, and memory with sizable discounts, and it’s probably the most sensible way to get to building a new PC right now.

READ: Intel Core Ultra 5 250K Plus review

Best Budget CPU ($100-$150)

If your sole focus is gaming, the Ryzen 5 7600X3D is your best option right now. It’s not as fast as AMD’s eight-core Ryzen 7 7800X3D and 9800X3D, but it’s surprisingly close considering the price. Based on our testing, it’s just 4.5% slower than the Ryzen 7 7800X3D, giving you most of the gaming performance for close to $150 less.

The Ryzen 5 7600X3D is available for $230, but if you have a local Micro Center, you can pick it up in-store for $200 (or even less bundled with a motherboard). Currently, the Ryzen 5 7600X3D is the most inexpensive X3D CPU on the market. The Ryzen 5 5600X3D has been discontinued, and the Ryzen 5 5500X3D is only available in Latin America, if you can find it in stock at all.

You can slot the Ryzen 5 7600X3D into existing AM5 motherboards, which AMD recently confirmed it will support until at least 2029. That gives you quite a long runway for future upgrades. AM5 motherboards exclusively use DDR5, with officially supported speeds going up to 5600 MT/s (though, we’ve found that 6000 MT/s is the sweet spot for most AMD chips). Given that the Ryzen 5 7600X3D has a large L3 cache, memory speed is less important; there’s only a marginal performance difference between overclocked memory and JEDEC standards on X3D chips.

READ: AMD Ryzen 5 7600X3D review

Best Alternative Budget CPU ($100-$150)

The $107 AMD Ryzen 5 5600 delivers a solid blend of performance in both gaming and productivity applications, bringing a new level of value to the Zen 3 lineup. If you're fine sticking with a previous-gen AM4 motherboard, the Ryzen 5 5600 makes a great budget build. The primary trade-off for the AM4 platform is that you're limited to DDR4, and you don't have access to PCIe 5.0, but this isn't a problem for most users at this price point. However, the Core i5-12400F is a better deal if you're looking for DDR5 and PCIe 5.0 support.

The Ryzen 5 5600 also makes an absolutely unbeatable budget chip if you're updating a first-gen Ryzen system. The 5600 unseats the Ryzen 5 5600X, a long-time favorite. The 5600X is only a mostly imperceptible ~1% faster in gaming and multi-threaded PC work than the non-X model, but provides a 4% advantage in single-threaded work.

You can find the six-core 12-thread Ryzen 5 5600 at $107, a deep discount brought on by competitive pressure from Intel. In fact, our testing shows that the Ryzen 5 5600 generally matches the gaming performance of its more expensive sibling, the ~$200 Ryzen 7 5800X. That makes the 5600 an incredibly well-rounded chip that can handle gaming well, from competitive-class performance with high refresh rate monitors to multi-tasking gaming workloads like streaming, while also serving up more than enough performance for day-to-day productivity apps. As with all AMD CPUs for gaming, you can fully overclock the CPU.

The Ryzen 5 5600 has a 3.7 GHz base and 4.6 GHz boost clock. The chip also has a 65W TDP rating, so it runs cool and quiet. Existing AMD owners with a 500-series motherboard will be happy, as the 5600X drops right into existing 500-, 400-, and 300-series motherboards. If you need a new motherboard to support the chip, AMD's AM4 motherboards are plentiful and relatively affordable, with the B-series lineup offering the best overall value for this class of chip.

Read More: AMD Ryzen 5 5600 Review

Best Entry-Level Budget CPU (<$100)

Recently, Tom’s Hardware underwent an experiment to find the best CPU under $100, and the Core i3-14100F from Intel is what came out on top. It’s still available for sale at major retailers (though the price has slipped slightly higher to $115), and it’s faster than AMD’s Ryzen 5 5500 and the Core i3-12100F in both games and applications, which are the two CPUs competing around this price that are still available for sale.

The Core i3-14100F is a straightforward CPU. It uses the same silicon as the older Core i3-12100F and Core i3-13100F, but it’s clocked higher and has a larger power budget available. Otherwise, it comes packed with four Golden Cove performance cores and eight threads, and no efficiency cores. As with other LGA 1700 CPUs, it supports DDR4 (up to 3,200 MT/s) and DDR5 (up to 4,800 MT/s) with a compatible motherboard.

At $100, the Core i3-14100F is one of the cheapest modern CPUs you can buy, but it also saves you a lot of money on an overall platform. The option to use DDR4 memory significantly reduces the cost of building a PC, and the Core i3-14100F comes bundled with Intel’s Laminar RM1 cooler, which is enough to keep the Core i3-14100F in check with its 58W TDP.

Read More: $100 CPU Shootout: Comparing the Ryzen 5 5500, Core i3-14100F, and Core i3-12100F to find the top DDR4 CPU

Best Budget APU

The Ryzen 5 5600GT steps into the arena as the value champ for APUs, which are chips with strong enough integrated graphics that they don't require a discrete GPU for light gaming — just be sure you're willing to accept lowered quality settings. The Ryzen 5 5600GT is a refreshed version of the Ryzen 5 5600G, which was our previous pick in this slot. They’re the same CPU minus a slight bump in clock speed, but whereas the Ryzen 5 5600G is out of stock at most retailers, you can actually buy the Ryzen 5 5600GT.

At $150, the Ryzen 5 5600GT gives you 96% of the gaming performance on integrated graphics than its more expensive sibling, the $208 Ryzen 7 5700G, but for 28% less cash. Our testing shows that its level of performance makes it the best value APU on the market. As long as you're willing to sacrifice fidelity and resolution and keep your expectations in check, the Ryzen 5 5600GT's Vega graphics have surprisingly good performance in gaming.

The 5600GT's Vega graphics served up comparatively great 1280x720 gaming across numerous titles in our tests, but options become more restricted at 1080p. Of course, you can get away with 1080p gaming, but you'll need to severely limit the fidelity settings with most titles.

With eight cores and 16 threads that operate at a 3.6 GHz base and boost up to 4.6 GHz, the Ryzen 5 5600GT also offers solid performance for its price point in standard desktop PC applications. The chip also comes with a bundled Wraith Stealth cooler, sweetening the value prop, and drops into existing 500-series and some 400-series motherboards, though support on the latter will vary by vendor.

Read More: AMD Ryzen 5 5600G Review

• MORE: AMD vs Intel
• MORE: How to Overclock a CPU
• MORE: How to check CPU Temperature

The LattePanda Sigma, launched today by the LattePanda team from $579, improves upon the power of the LattePanda 3 Delta with a Raptor Lake-based CPU, 16GB of LPDDR5, and an Intel Iris Xe GPU. All of this makes it a much more powerful machine than the LattePanda 3 Delta and the Raspberry Pi 4. 

The LattePanda 3 Delta was already an impressive board, much more powerful than a Raspberry Pi 4. However, the specs for the Sigma blow the Delta firmly out of the water.

The choice of a 13th-Gen Intel Raptor Lake Core i5-1340P (12 core, 16 threads) is a big upgrade from the quad-core Intel Celeron N5105 found in the LattePanda 3 Delta. The GPU also sees an upgrade, now sporting an Intel Iris Xe GPU with 80 execution units and up to 1.45 GHZ, compared to the Delta's now rather pedestrian Intel UHD GPU running between 450 and 800 MHz.

The Sigma's RAM upgrade is also welcome. While the 8GB of LPDDR4 RAM afforded by the Delta was enough for most tasks, 16GB of LPDDR5 running at 6400 MHz means we could get a big performance boost for all projects. 

What the two share, though, is an onboard Arduino Leonardo-compatible co-processor, which is located this time along the short edge of the board. In our review of the LattePanda 3 Delta, we loved the easy GPIO access, and using the Arduino IDE and Python firmware, we easily created projects. The Sigma seems to share the same pin arrangement, and while it isn't a true Arduino Uno pinout, it is functional.

The LattePanda Sigma is pitched as a "Hackable Single Board Sever with Mighty Power," but that doesn't restrict it to just sitting in an office. The onboard Arduino Leonardo provides a GPIO interface accessible via a USB-to-serial interface. With some Arduino and Python code and a USB camera, building a powerful machine learning/computer vision project would be simple. Sure, the Sigma is larger than a Raspberry Pi 4, but it could be easily integrated into a robotics project. The CPU / GPU combo is potent and could easily be used as a low-power desktop machine with four 4K displays running simultaneously. To keep it cool, we see a combined heatsink and fan assembly dominating the board's top, taking up more space than the LattePanda 3 Delta's cooling solution.

The dual Thunderbolt 4 ports are an enticing proposition. We potentially benefit from fast data transfer speeds and potentially external GPUs. Gaming is possible, on the Intel Iris Xe GPU, heck we managed a competent session of Call of Duty: Modern Warfare on the LattePanda 3 Delta, so it should be a cakewalk for Sigma.

The LattePanda Sigma is now available direct from DFRobot, starting from $579 for a base unit, $648 with a 500GB SSD and Wi-Fi 6E module.

Earlier this year it transpired that in 2023, Intel intends to cease using its Celeron and Pentium brands for entry-level notebook CPUs. Instead, Intel will use the 'Intel Processor' branding. With 2023 being mere weeks away, it is about time for one of the first Intel Processor-badged CPUs to allegedly get tested. For now take the test results with a pinch of salt until they are verified.

The first Intel Processor product to hit Primate Labs' Geekbench 5 database appears to be the Intel N95 (via @Benchleaks), which seems to be one of the entry-level Alder Lake-N CPUs featuring four Atom-class energy-efficient cores based on the Gracement microarchitecture. The processor features a 1.70 GHz base clock, a 2.80 GHz boost clock, 2MB of L2 cache, and 6MB of L3 cache, based on the entry in the Geekbench database. This is not the first time when an Alder Lake-N CPU gets benchmarked, but this is the first time when we see a quad-core Alder Lake-N. 

 Since the Intel N95 is an entry-level product with only four efficient cores and rather low frequencies, it was certainly not designed to offer breakthrough performance. Therefore, it is not surprising that it ranked substantially slower than Intel's eight-core Core i3-N305 'Alder Lake-N' processor. Yet, it stacks up very well against Intel's existing Celeron 7305 CPU with one high-performance and four energy-efficient cores that operate at 1.10 GHz.

While we yet have to determine how exactly Intel intends to position its N95 processor, we also included scores obtained on AMD's entry-level Athlon Gold 7220U CPU with two Zen 2 cores at 2.40 GHz ~ 3.70 GHz and with a 4MB of L3 cache for comparison. Evidently, the Athlon Gold 7220U shows better GeekBench 5 scores than Intel's N95, though we should bear in mind that AMD's lowest-end next-generation Athlon will be the dual-core Athlon Silver 7120U (which yet has to get benchmarked). 

Even though Geekbench 5 is not exactly the best way to estimate performance of a CPU in real-life applications, it should be noted that it can still be used to find out how processors stack up against each other. 

Meanwhile, keep in mind that for now we are dealing with pre-production Alder Lake-N hardware and it might be too early to draw any conclusions about performance of these CPUs.

AAEON, a Taiwanese maker of industrial computers and an associate company of Asus, has revealed the UP Xtreme i12 developer board, a curious mix of an Alder Lake processor with a 40-pin GPIO connector rather similar to that seen on Raspberry Pi boards, and is allegedly compatible with Raspberry Pi HATs.

At least, that’s the claim. Compatibility will be a mixture of hardware and software support, and there's a list of tested HATs on GitHub, including some from Adafruit and PiFace. The tested HATs all appear to be I2C and SPI based devices. You can add screens, accelerometers, stepper motor control, and Grove sensors through those spiky pins. It is not impossible to use another HAT with this or other SBCs, but you will need to do a lot of work port libraries and code to make it work.

Elsewhere, the board is a more conventional Intel-based design, with a choice of 12th-gen P-series processors all the way from Celeron to i7 with Intel UHD or Iris Xe graphics. Up to 32GB of LPDDR5 RAM can be specified, and there's a 2.5Gb Ethernet port, another gigabit port, two PCIe 4.0 M.2 sockets plus a SATA 3 port, USB 4, 3 and 2, full-size DisplayPort and HDMI connectors (with extra displays pushed from USB-C and Embedded DisplayPort, for a total of four 4K/60 screens), headphone socket, and front-panel connectors for putting it in a case. 

Optional modules add Wi-Fi 6, Bluetooth 5.2, and 5G connectivity. Power comes via a DC input, and it can accept feeds from 12V to 36V. The board comes with active cooling, and while a passively cooled case, the UP Xtreme i12 Edge, is also listed, along with bundles including power supplies and cables, it’s not available yet. The board runs Windows or Ubuntu, with Yocto support coming. As this is an x86 based board you are free to choose your own operating system.

The board is not yet showing on the AAEON e-shop, but the previous version, the UP Xtreme i11, starts from $299.

The first benchmark results of Intel's yet unannounced eight-core Core i3-N300-series processors have been posted online. The eight-core processors appear to be slower than Intel's quad-core Core i3-12100 and six-core Core i3-1210U in Bapco's CrossMark as well as Primate Labs's Geekbench 5, which may suggest that we could be dealing with Intel's anticipated Alder Lake-N CPUs that only feature Atom-class energy-efficient cores. As ever with leaks, treat the news with a pinch of salt until it can be verified.

Intel's Core i3-N300-Series

Intel yet has to formally introduce its Core i3-N300-series processors for notebooks. It seems as though their results in Bapco's CrossMark and Primate Labs's Geekbench 5 benchmark suites have already been published by someone either accidentally or by design. Based on entries in Bapco's database (via @Benchleaks), Intel's Core i3-N300 and Core i3-N305 processors feature eight physical cores and eight logical cores (i.e., they do not support simultaneous multithreading) as well as Intel's built-in graphics. The Core i3-N305 has a base frequency of 1.80 GHz and a maximum frequency of 3.78 GHz, according to the Geekbench 5 database. 

The Core i3-N300-powered system was equipped with 8GB of memory as well as an SCY 512GB NVMe SSD, whereas the Core i3-N305-based machine came with 16GB of DDR4-3200 memory as well as an SCY 512GB NVMe drive. In both cases systems reported about one memory channel, which is consistent with rumors about Alder Lake-N having only one memory channel.

CrossMark Performance

Benchmark results of Intel's eight Core i3-N300 and Core i3-N305 processors do not look overly impressive when compared to Intel's quad-core Core i3-12100 (4P cores, eight threads) and six-core Core i3-1210U (2P + 4E cores, eight threads). Intel's currently available parts are better for Productivity and Creativity workloads emulated by CrossMark as well as offer better responsiveness. It is noteworthy that the Core i3-N300 is significantly behind the Core i3-N305 despite similar number of cores. Perhaps, the former runs at tangibly lower clocks than the latter, or maybe a misconfiguration on the pre-production system used to run the tests.

Geekbench 5 Performance

Performance of Intel's eight-core Core i3-N305 processor in Geekbench 5 is not too impressive either as it again falls behind Intel's six-core Core i3-1210U as well as quad-core Core i3-12100. The Core i3-N305 run at around 3.78 GHz most of the time (its maximum supported clock speed), so it was clearly not constrained by frequency or thermals. Nonetheless, keep in mind that we are dealing with pre-production hardware, so take these results with a grain of salt anyway.

Since Intel has not formally announced either Core i3-N300-series or its codenamed Alder Lake-N processors, we cannot say that the former CPUs are based on the ADL-N silicon with 100% probability. While Intel could have rolled-out two eight-core Core i3 parts based on some outdated silicon and microarchitecture and decided to use all-new model numbers for them, another explanation could be that Core i3-N300-series parts are indeed Alder Lake-N CPUs that feature only energy-efficient cores that are significantly slower when compared to Intel's Golden Cove cores. If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck. 

Furthermore, considering that Intel's mobile processors currently sold under Celeron and Pentium brands will be marketed under Intel Processor N100 and N200 monikers in 2023, it might be logical for Intel to offer N300-series processors under its Core i3 brand to address users who understand the value of its Core trademark. In the end, N100, N200, and N300 model numbers will be somewhat more consistent than today's hoard of Pentium Gold, Pentium Silver, Celeron Gold, and Celeron N.

Intel's New Mobile CPU Lineup

Intel formally introduced its 13th Generation Core family of CPUs earlier this week. High-end models in the family will use all-new silicon with up to eight high-performance Raptor Cove and 16 Gracemont energy-efficient cores. By contrast, most midrange Core i5 and all entry-level Core i3 processors will continue to rely on Alder Lake 6P silicon with up to six high-performance cores. 

Meanwhile since Intel's 7 (aka 10nm Enhanced SuperFin) production node has now been used for well over a year, expect regular continuous process improvements (CPI) to kick in and improve yields and overclockability of Alder Lake 6P-based products. As a result, it might be more preferable to buy 13th Generation Core i5 and Core i3 CPUs than stick to last year's 12th Generation Core i5/i3 parts. 

Formally, Intel's Core i3-N300-series processors are not parts of the company's 13th Generation Core processors family, even though they seem to use Gracemont cores and Intel 7 process technology. In fact, the introduction of Core i3 N300-series looks like a wise move by Intel as it clearly distinguishes between its hybrid processors featuring high-performance and energy-efficient cores as well as energy-efficient system-on-chips based solely on low-power Gracemont cores.

The main advantage of Intel's Alder Lake-N processors (and therefore Core i3-N300 parts) is of course their energy efficiency, which promises to enable a very long battery life for laptops featuring these parts. The only question is whether PC makers will actually use relatively expensive displays, solid-state storage, and memory with lowered power consumption to build long-lasting notebooks based on Core i3-N300-series CPUs. The prototype machines used to run Bapco's CrossMark tests were equipped with a cheap 1920 x 1080 display, DDR4 memory, and an SCY SSD, so we are probably not talking about prototypes of high-end or even midrange laptops here. 

What remains to be seen is how Intel will divide its N300, N200, and N100-series processors for entry-level low-power laptops. Intel's Core i3-N300-series CPUs featuring eight cores and running at 1.80 GHz ~ 3.80 GHz, so we wonder whether N200 and N100-series SoCs will just feature lower clocks, or some will also have fewer cores.

According to a report by Coelacanth's Dream, a couple of new Alder Lake-N processor names have been discovered in an Intel graphics bootlog. These chips are known as the N100 and N200. Traditionally, the 'N' badging would signify a Celeron or Pentium product; however, there is no such terminology to be found this time. That leads the outlet to believe these chips will be the first to incorporate Intel's new "Intel Processor" brand name.

A few days ago, Intel announced that it would retire the 20-year-old Pentium and Celeron naming scheme. In its place, Intel will use a very generic "Intel Processor" term. Starting in 2023, Intel Processor will represent all of Intel's budget-friendly CPUs, which are cheaper and less powerful than the Core i3 lineup.

The most notable hint at the name change is the drastically different model numbers found with the N100 and N200. Previous chips have used four digits plus the N to signify a full model name - such as the Pentium N6000. But here we see a reduction of the thousandth digit to hundreds only, signifying these chips could use the 'Intel processor' name instead.

The name change would be appropriate for the new CPUs, as these chips are believed to be a part of Intel's new budget-friendly Alder Lake-N architecture — built to replace current Pentium and Celeron Jasper Lake CPUs.

Alder Lake-N will be built on the same Gracemount cores found in Alder Lakes E-cores, with a maximum configuration of eight cores on the flagship model. As a result, these chips won't be particularly fast compared to today's chips. However, they will be very efficient and incredibly compact yet still retain performance akin to Intel's Skylake parts. That's a perfect combination for Chromebooks and entry-level Windows laptops.

According to the bootlog data, the N100 and N200 are quad-core parts with no hyperthreading. Naturally, we should expect a flagship eight-core eight-thread part to be in the works to complete the lineup. The release date for Alder Lake-N is expected to be in 2023.

The shortage of Raspberry Pis continues. Over a year has passed since Eben Upton came on The Pi Cast to talk about Raspberry Pi and the supply chain woes, and it is still a challenge to pick-up everyone’s favorite single board computer (SBC).

Over the last decade there have been many Raspberry Pi alternatives. OrangePi, Asus Tinkerboard and Khadas VIM4 to name but a few. These alternatives often have high spec hardware, but come up short with software support.

Another group of alternatives are boards that can still be classified as SBCs but feature Intel CPUs at their core. The latest to cross our bench is the $279 LattePanda 3 Delta, a rather large and powerful SBC powered by an 11th Gen Intel Jasper Lake CPU. Because it is Intel based, we have an expansive choice of potential operating systems. LattePanda provides both Windows 10 and 11, but if Linux is more your thing, you are well catered for.

So if this board is Intel based, how do we access the GPIO? Just as with Seeed’s Odyssey, there is a built-in ATMEGA32U4 microcontroller. This chip is perhaps best known as the brains of many Arduino boards, such as the Arduino Leonardo. No matter the OS, we can access the Arduino and use it as an interface with the LattePanda 3 Delta.

Is the LattePanda 3 Delta a Raspberry Pi killer? Is it worth the extra money? To determine this, and to learn more, we need to put the LattePanda 3 Delta on the bench.

LattePanda 3 Delta Specifications

Using LattePanda 3 Delta

Essentially the LattePanda 3 Delta is an Intel x86 Jasper Lake based PC in a large single board computer form factor. This means that we are free to use Windows (LattePanda offers Windows 10 and 11 downloads via its site) or many other Linux distributions. We tested the board with Windows 10, downloaded directly from LattePanda. The installation process is entirely automated. We just plug in the USB stick, power on the board and the install script copies Windows 10 to the onboard 64GB eMMC. Note that LattePanda doesn’t supply a Windows registration key, unless you pay for the $339 model. The install script can also install Windows 10/11 to an alternative drive. We installed Windows 11 to a spare 256GB SATA drive of similar performance to the onboard eMMC. Also, we installed Ubuntu 22.04.1 to a Toshiba 256GB NVMe drive.

Boot times were longer than those of the Khadas VIM4 and the Raspberry Pi 4. The slowest was our Windows 10 install on the onboard eMMC which clocked in at 53.64 seconds. Second was Windows 11 installed to a SATA3 drive with a boot time of 47.14 seconds. First place goes to Ubuntu 22.04.1, installed to our NVMe drive which came in at 42.99 seconds. These boot times aren’t stellar, and when compared to the Raspberry Pi 4 (micro SD) boot time of 30 seconds, and Khadas VIM4’s 36.38 seconds boot, they seem glacial. But consider this, 10 seconds of the LattePanda’s boot time is a prompt to enter the BIOS.

Speaking of the BIOS, we have an unlocked American Megatrends BIOS which affords us a great deal of configuration. We can tweak our CPU and GPU settings, along with disabling the onboard eMMC storage. This was particularly helpful for our tests, as we didn’t want to create a dual-boot on the limited 64GB eMMC. The BIOS can also be used to set the boot device, be it USB, SATA3 or NVMe.Our Ubuntu installation was much snappier thanks to the improved bandwidth of NVMe.

Whether we were using the internal storage or SATA, Windows worked rather well. But lets manage expectations a little; this is a low power CPU in an SBC configuration. You will be able to perform most admin tasks, some light video editing and photo work. Think of the LattePanda 3 Delta as a powerful platform for robotics and machine learning. Projects that traditionally require more processing power to perform a task. The Arduino element is there to control the motors, sensors and other components that make the project move.

Our Ubuntu installation was much snappier thanks to the improved bandwidth of NVMe.

Gaming is possible, but best to treat the catalogue from the early 2010s as the limit. We tested the original Call of Duty: Modern Warfare at a native 2560 x 1440 and it was just too much for the board. Dropping the resolution down to 1080p provided a better result, but 720P returned the best results.If you need a little more speed, invest in an NVMe drive. Design of LattePanda 3 Delta

The LattePanda 3 Delta has an almost sci-fi look and feel. The large heatsink and fan bearing the LattePanda logo bear a resemblance to the bridge of a starship. The black PCB is clear and easy to read, largely down to a nice silkscreen print and a matte finish.

The board is obviously larger than a Raspberry Pi 4, and unlike the Pi it favors the main ports at either end of the board. Port placements are good. There is plenty of space between the ports, unlike the Khadas VIM4 which has a tight fit between the USB-C and HDMI port. 

The top layer of the PCB is dominated by the heatsink and fan, with a perimeter of GPIO pins around it. But the underside is where we find the most interesting features of this board. On this underside we spot an NVMe M.2 slot that supports PCIe 3.0 and it is compatible with 2280 format drives. Another M.2 slot B provides expansion via SATA SSD (2242, 2260 and 2280), or with 4/5G modules. The board also has a combined micro SD and SIM card slot, necessary should you wish to use 4/5G modules.

Cooling LattePanda 3 Delta

The Intel Celeron N5105 Jasper Lake CPU uses 10nm lithography and is designed with low power devices in mind. The TDP may only be 10W, but this CPU requires active cooling to prevent thermal throttling. Luckily the included fan is whisper quiet. At 80% we could barely hear it. We tested using Windows 10, leaving the board idling for 10 minutes.

At idle, the board pulled 15.2V at 100 milliamps (1.52W) and reached a maximum temperature of 34 degrees Celsius. Under stress, via y-cruncher we pushed the LattePanda 3 Delta to pull 15.2V at 1.2A, 18.24W with a peak temperature of 77 degrees Celsius. All the while, the cooling fan was quiet.

Let's look at the power consumption. Under stress, the LattePanda 3 Delta consumes three times more power than the Raspberry Pi 4 and Khadas VIM4. This isn’t grotesque but if you are building a project with a power budget, for example robotics or off-grid data collection, then perhaps the Raspberry Pi or Khadas VIM4 is more applicable. If your power budget needs to be even lower, perhaps a microcontroller such as the Raspberry Pi Pico W is a better option.

LattePanda 3 Delta GPIO

The GPIO of the LattePanda 3 Delta is two-fold. We have breakouts for USB, RS232 (serial), I2C and Audio. Via these breakouts we can directly access hardware at the OS level. The other GPIO is for the onboard Arudino Leonardo. In a similar manner to Seeed’s Odyssey, LattePanda 3 Delta features its own onboard Arduino microcontroller.

It does not share the typical Arduino Uno form factor, instead it has a double row of headers with the pinout printed to the side of the header. The heatsink does get in the way at times, and it made it a little tricky to identify the pins that we wished to use.

Controlling the Arduino GPIO is possible using the Arduino IDE, the version of Windows 10 provided via the LattePanda site has this pre-installed and ready to go. We tested the Arduino GPIO with the humble LED, blinking the LED every half second. With that success under our belt we connected a one meter length of “neon-like” NeoPixels and installed the appropriate libraries. Moments later, our desk was bathed with the glow of 90 RGB LEDs.

Buoyed by our success we pushed forward and connected an I2C LCD to the GPIO, and this is where we hit an issue. Our I2C LCD requires power (5V and GND) and connects to the I2C bus, except it doesn’t work with the labelled headers. The I2C header is reserved for OS level access, not the Arduino. For the Arduino to see the LCD display we had to determine the correct I2C pins via a handy pinout, and then scan the I2C bus for the address of our screen. With that out of the way it was a simple matter of installing the LiquidCrystal I2C library and modifying the code to show a custom message. 

If the Arduino IDE isn’t your area of expertise, you can use Python to control the GPIO. You will need to briefly use the Arduino IDE to flash Firmata, a firmware that we can use with Python. Installing the PyFirmata library via Python 3’s package manager we then created a short script to flash the onboard LED and then an external LED connected via jumper wires. The Python approach worked well, albeit much slower than the Arduino workflow. It took our Python code a few seconds before it was executed. This is the result of PyFirmata interpreting our Python code into something the ATMEGA32U4 can understand.

This GPIO is the best compromise we have seen on a Raspberry Pi alternative. Others have attempted to emulate (hardware and software) the Raspberry Pi format, and they have varying levels of success. By eschewing this form factor, while retaining compatibility with Arduino code and components, the LattePanda 3 Delta provides the GPIO access that we take for granted. Sure we can’t directly connect an Arduino shield, but we can use a few wires to bridge the gap.

LattePanda 3 Delta Networking Performance

With a Gigabit Ethernet connection, LattePanda 3 Delta has plenty of network bandwidth. Using iperf to test a connection from the LattePanda to our desktop machine over a Gigabit network connection we saw 947 Mbits/sec.

The same test using the onboard AX201D2W Wi-Fi card saw 162 Mbits/sec via our 5-GHz Wi-Fi access point. We were unable to test Wi-Fi 6. Based on our tests the LattePanda 3 Delta could be pressed into service as a DIY network storage device.

Windows 11 on the Latte Panda 3 Delta

With support for TPM 2.0 backed into the board, the LattePanda 3 Delta is fully compliant with Windows 11 and dare we say it runs rather well. Installed to a spare SATA3 drive, close in specification to the eMMC (the write speeds were much faster, but read was similar), Windows 11 booted slightly faster than Windows 10 (SATA3 performance).

Overall performance felt the same as Windows 10 and we even managed to undertake a short Blender test where we rendered a simple scene. Windows 11 felt decent. it won’t beat a higher spec machine, but it would work as your low power daily driver. The Arduino IDE worked with the onboard Arduino Leonardo and we had no issues creating projects and having fun.

If you purchase a LattePanda 3 Delta without a Windows license, then perhaps now is the time to make the jump to Windows 11?

LattePanda 3 Delta Linux Performance

If Windows isn’t your thing, we hear you. Linux is generally a better fit for SBCs and so we installed the latest version of Ubuntu 22.04.1 and fully updated the OS. Our installation was contained on a 256GB NVMe SSD so that our previous Windows installations remained untouched. 

Ubuntu just worked, there were no installation issues and it detected our hardware correctly, including the Arduino Leonardo. After specifying the board and the device port we were able to recreate our LCD display project with no issues.

Ubuntu is a slightly lighter OS than Windows. Sure we can tweak Windows to remove some of the bloat, but on the whole, Linux is generally lighter. Ubuntu on the LattePanda 3 Delta is a joy. We get the full desktop Linux experience, and access to the GPIO.

Linux graphics performance, specifically YouTube playback is a bit of a mixed bag. We noted that Ubuntu dropped 212 from a total of 2846 frames when playing back 4k30 video. The same test on Windows 10 dropped on 4 frames. At 1440 there were no dropped frames under Ubuntu and only 6 from a total of 2373 under Windows. So when it comes to graphics performance, Windows is the most consistent experience. That said, how often will you be consuming 1440P and 4K media?

On the whole, Ubuntu felt great. As a long term Linux user, I felt at home with Ubuntu on the LattePanda 3 Delta. The ease at which we could access the onboard Arduino, and the familiarity of the OS provided us with a frictionless onboarding process. The graphics issue is something to consider, but it really is a non-entity if you are just watching 1080P media and general PC use.

LattePanda 3 Delta versus Raspberry Pi 4

It is fair to say that the LattePanda 3 Delta is much more powerful than the Raspberry Pi 4. The quad-core 2 GHz Intel CPU has more raw horsepower than the Arm CPU in the Pi 4, while staying relatively conservative with power consumption. If raw power is your consideration, then the LattePanda 3 Delta is the winner. It has plenty of power for machine learning and AI projects and the Arduino provides plenty of GPIO for our projects. 

The cons for the LattePanda 3 Delta are its size, approximately twice the size of the Raspberry Pi 4 and the price. The $279 price tag comes in even higher than the $239 of the Khadas VIM4 and much higher than the RRP of the $95 Raspberry Pi 4 8GB. But with the price of Pi being driven by supply and demand, and demand is still outstripping supply, RRP is a unicorn that we have yet to catch.

Bottom Line

If you are looking for a Raspberry Pi replacement, something small and powerful then the LattePanda 3 Delta isn’t a done deal. Larger in size, higher in price and with three times the power consumption of a Pi 4, LattePanda 3 Delta is not a direct competitor to the Raspberry Pi. Instead the LattePanda 3 Delta is more a low-power computer that inhabits the SBC form factor. It brings the CPU power necessary for advanced autonomous robots, projects with machine vision and artificial intelligence. The computational power is where the larger size (for the active cooling) and power consumption come into play.

The cost, $279, is initially hard to swallow but again this isn’t just a replacement for the Raspberry P;, it is a desktop computer in an SBC form factor. You could happily use this machine for day to day work, while using less electricity than on a traditional desktop. 

The Arduino compatibility is icing on the cake. It is easy to use and works with the many thousands of libraries available in the Arduino ecosystem. Sure, we can’t directly use Arduino shields, but unless you have bought into that system, you aren’t going to miss them.

If we need an all-in-one for our maker workstation, and we crave more power than the Pi can provide, then the LattePanda 3 Delta is the obvious choice.

Intel will decommission its legendary Celeron and Pentium brands used for basic notebook CPUs. Instead, starting in 2023, they will be referred to as the rather humble "Intel Processor" name. The move will allow the company to sharpen its focus on premium Core, Evo, and vPro brands and sell more premium CPUs. 

For now, Intel will continue to use Celeron and Pentium brands for desktop and embedded applications. When asked how this branding change would affect desktops, Intel told us, "Desktop has no new products in this segment slated for Q1 '23."

Intel's notebook product stack today includes a variety of Core-branded processors for high-performance, mid-range, and entry-level laptops, as well as Pentium and Celeron-badged CPUs for inexpensive essential systems. Starting from 2023 with notebooks, Intel will unite its Pentium and Celeron product families under the Intel Processor umbrella and will continue to address respective market segments. The new brand leaves unchanged Intel's existing processors (they will sell under their own names) and the company's product roadmap. 

"The new Intel Processor branding will simplify our offerings so users can focus on choosing the right processor for their needs," said Josh Newman, Intel vice president and interim general manager of Mobile Client Platforms.

While the merge may clear up some confusion between various Pentium and Celeron-branded laptop offerings, it might also create some new puzzlements. 

Intel's mobile Pentium Gold products are based on the company's designs featuring high-performance or high-performance and energy-efficient cores, therefore offering an experience comparable to that of fully-fledged Core processors. For example, the company currently offers Pentium Gold-badged Alder Lake CPUs with one high-performance Golden Cove and four energy-efficient Gracemont cores (8505). 

Intel's mobile Celeron processors are also based on the company's designs with performance (6305) or performance and efficient cores (7305). However, since they sit below mobile Pentiums, they have lower clocks, have smaller caches, or lack certain features. Yet, these are still quite capable chips for basic workloads. 

Merging mobile Pentium Gold and Celeron product families into one Intel Processor lineup makes sense on the condition that model numbers accurately represent their performance and capabilities. 

There are other Celeron and Pentium-branded processors for laptops too. Intel's mobile Celeron N and Pentium Silver processors are based on energy-efficient Atom-class cores and barely offer performance that is even slightly close to that of mobile Pentium Gold and 'vanilla' Celeron CPUs with high-performance cores. Intel did not update these product families this year as its Alder Lake-N design featuring only energy-efficient Gracemont cores is not yet here. But Intel is certainly prepping these system-on-chips for launch, and they might address market segments of Celeron N and Pentium Silver. 

We do not know whether or not Intel plans to add Alder Lake-N into the Intel Processor lineup for mobile PCs. But if it does, pouring in designs based solely on energy-efficient Atom-class cores into the lineup with quite different chips will create a lot of confusion.  

Single-board computers (SBC) with Intel processors onboard are outnumbered by Arm-based designs, but the latest release from Aaeon, whose boards we’ve covered before, ups the ante with Alder Lake, DDR5, and up to four independent displays. Alder Lake CPUs of course rank among the best CPUs, though we're dealing with slightly different models in this case.

Known as the GENE-ADP6, it’s aimed at machine vision and edge computing applications, as well as digital signage. However, at just 5.75in x 4in (for comparison, a Raspberry Pi 4 measures 3.35in x 2.25in) it would also make a pretty good small-form-factor PC if you can find a suitable case. Mini-ITX for reference uses a 6.7 x 6.7 inch form factor.

At the heart of the board is 12th-gen Intel silicon, ranging from the Celeron 7305E (five cores, five threads, boost up to 4.2GHz) to the mighty i7-1270PE (12 cores, 16 threads, boost up to 4.5GHz). This is backed by dual-channel DDR5-4800, with capacity for up to 64GB. There's a UEFI BIOS and a TPM2 chip, Iris Xe graphics, an M.2 2280 slot for a PCIe 4.0 x4 SSD, and a SATA port too. Additional M.2 modules can be used to add Wi-Fi and 5G connectivity.

Around the back, we find a pair of Ethernet sockets, one 2.5Gb, the other 1Gb, alongside three USB 3 Type-A ports and a single Type-C. USB 2 is available via a pin header on the board itself, while both HDMI 2.1 and DisplayPort 1.4a are also present. Additional screens can be pushed from the LVDS and Embedded DisplayPort interfaces, allowing it to hit the four-screen target.

Power comes via AT or ATX power supplies, with the board able to accept anything from 9 to 36 volts. Software support comes in the form of Windows and Ubuntu 22.04. The board is for sale on the Aaeon online store, but it's currently sold out. Also, at the time of writing, it's showing an eye-watering price of $9,999/ That's either a mistake or some sort of bulk pricing. You can get notifications of stock levels or price drops, or make enquiries about the board at the product page.

Lately, we're seeing a slew of small systems with with big processors seem, so we were weren't too surprised to see the Aaeon EPIC-TGH7 SBC reported in the pages of CNX Software. This four-inch (165 x 115mm) board has a list of compatible processors that begins with the Intel Celeron 6600HE, but which makes its way through the i-numbers to the mighty Xeon W 11865MRE. 

None of these processors would make our list of the best CPUs for gaming. However, the config with the Xeon W 11865MRE promises workstation class power on a motherboard that's significantly smaller than a mini ITX motherboard (170 x 170 mm), the smallest mainstream size.

The TGH7 appears to come with the Tiger Lake H CPU+GPU combo soldered to the board; only the i3 and i7 models are listed on Aaeon’s site, and they’re both out of stock at the time of writing. But theoretically you could get one with a Xeon W installed, and run ECC memory with it too. This appears to be the only benefit of the workstation chip, which in every other way is identical to the i7 11850HE.

These two hot chips (their temperature rating tops out at 100°C or 212°F) both run to eight cores and 16 threads, with a top boost speed of 4.7 GHz, two memory channels, and identical integrated graphics. They both draw the same 45W, and don't appear to come with any cooling solutions, though holes for a CPU fan are visible on the board.

Elsewhere on the board, you’ll find a lot of ports crammed into the small space. There are two DisplayPorts and an HDMI port, four USB 3.2 Type-A sockets, a pair of Ethernet ports - one gigabit, the other 2.5Gb - a pair of SATA ports, a PCIe 4.0 x8 slot supplying a maximum of 25W, and two M.2 slots, one of which can take NVMe storage, while the other is for optional modules such as Wi-Fi and Bluetooth, or a cellular modem. 

Drivers for Windows 10 are available, raising hopes it will work under Windows 11 as well, while a single ‘peripheral’ driver exists for Ubuntu 20.04.2. 

Assuming availability picks up a bit, you should be able to get the boards direct from Aaeon’s eshop, with pages online for the i3 ($812) and i7 ($1,167) boards. 

LattePanda, a Chinese electronics manufacturer that has released two generations of single-board computers with Intel CPUs, has taken the covers off its latest product, the LattePanda 3 Delta, which is now on sale following some teasing from the brand’s Twitter account and a Kickstarter campaign last year. 

It’s no shock to discover that the board uses an Intel CPU, and the new board follows its predecessors by also sporting an Arduino Leonardo coprocessor. This unusual addition is an ATmega32u4 (a RISC-based microcontroller) with built-in USB, and can appear to the connected computer as a keyboard and mouse, as well as a virtual COM port. 

Slightly larger than a Raspberry Pi at 125 x 78 x 16mm (4.9″ x 3.1″ x 0.6″), the 3 Delta can run desktop versions of both Linux and Windows, the Celeron N5105 (four Jasper Lake cores, four threads, boosting to 2.9GHz and drawing 10W) is backed by 8GB of RAM and 64GB of eMMC. You can slot an M.2 NVME SSD into one of its slots, while the other is compatible with SATA drives or mobile data modems. There's Wi-Fi 6 and Bluetooth 5.2 onboard, an HDMI, gigabit Ethernet, three USB 3.2 Type-A ports — one of which is a 10Gbps Gen 2 — and a Type-C port for power (there’s a separate 12v input), data, and DisplayPort. There's also a 3.5mm audio/mic jack.

GPIO pins run down both sides of the board, with Arduino and BIOS pins, including 3.3v and 5v power output on one side, and RS232, USB 2, I2C, audio and system control pins on the other. The CPU requires active cooling, so there's a fan header at one end, and with the shroud-like blower cooler attached the board looks rather like a small graphics card, or the inside of an old Mac Mini.

With the ability to run Windows (we’d be more inclined to put a lightweight Linux distro such as Mint’s Xfce Edition on it, but we’re weird like that) the possibilities of what to do with the board are endless — especially when you take the Arduino and the compatibility with 4K touchscreens into account. The need for active cooling, however, means it may not be entirely silent. 

Originally meant to enable secure execution in an isolated environment, Intel's Software Guard Extensions (SGX) memory encryption technology could do more harm than good. It turns out, processors featuring Intel's Sunny Cove microarchitecture may expose data located in the memory-mapped registers of the local Advanced Programmable Interrupt Controller (APIC), reports The Register. 

The registers are reportedly not initialized cleanly and therefore reading them exposes stale date of recent sample data transferred between the L2 and last-level cache, including SGX enclave data, from the super queue. Researchers call the vulnerability ÆPIC Leak (aka CWE-665: Improper Initialization) and claim that the bug has hardware origins. 

Intel claims that the affected processors include all chips based on the Sunny Cove/Cypress Cove microarchitectures, which covers 10th Generation Core 'Tiger Lake' and 'Rocket Lake', 3rd Generation Xeon Scalable 'Ice Lake-SP', and Xeon D-1700/2700 products. In addition, Atom, Celeron, and Pentium system-on-chips featuring the Gemini Lake microarchitecture are vulnerable to the same kind of attack. 

Meanwhile, to access data from APIC registers, perpetrators need to have admin or root privileges. Which makes the use of this weakness slightly harder to exploit (but not impossible). In virtualized environments hypervisors do not allow virtual machines access to APIC registers. 

Intel admits the problems with its SGX technology and has issued a set of recommendations on how to avoid potential problems with the vulnerability. Meanwhile, the researchers who discovered the bug late last year offer their own fix for the problem. 

Interestingly, some of the investigators who exposed the ÆPIC Leak bug also recently identified the first side-channel attack on scheduler queues. The vulnerability affects all of AMD's existing Ryzen processors featuring Zen 1/2/3 microarchitectures. To exploit the weakness and get access to data processed by the same CPU core, perpetrators need to run malicious code on that CPU core first, which is not particularly easy.  

"An attacker running on the same host and CPU core as you, could spy on which types of instructions you are executing due to the split-scheduler design on AMD CPUs," explained Gruss. "Apple's M1 (probably also M2) follows the same design but is not affected yet as they haven't introduced SMT in their CPUs yet." 

AMD reportedly confirmed the problem — currently called AMD-SB-1039: Execution Unit Scheduler Contention Side-Channel vulnerability on AMD Processors — and said that the company considers it a 'medium severity' threat.

Taiwanese tiny-PC company Shuttle has announced via its EU site a new model in its XPC Barebone line, capable of accepting Intel Xeon W processors, sporting four Ethernet ports, and the first Shuttle PC compatible with error-correcting RAM.

The SW580R8 isn't going to win any prizes for its name, but it's innovative in other ways. Extremely compact, as is the Shuttle way, the case comes with just the motherboard and PSU included - it's up to the customer to provide their own CPU, RAM sticks and graphics. It's small, but not as small as some we've seen.

The included chipset is Intel's W580, a product aimed at workstations and capable of housing Intel's 10th and 11th gen Xeon W processors, which in their Rocket Lake incarnation have six or eight hyper-threading cores, turbo speeds up to 5.3GHz, and draw up to 125W. Of course, you don't have to accept this much processing power; however: anything down to a Celeron G5905 can also be fitted.

The Xeon W chips all come with an Intel UHD P750 integrated GPU, but the case can accommodate a PCIe 4.0 x16 dual-slot AIB. There are four RAM slots, for a maximum of 128GB of ECC RAM, and a plethora of ports, including one HDMI 2.0b, two DisplayPort 1.4, four 10Gbit USB 3.2 plus a 5 Gbit USB 3.2 Type C. There are three USB 3.2 and four USB 2.0 Type-A ports too, audio around the back, and four Ethernet sockets (two of which are 2.5Gbps and the others plain old gigabit). One supports vPro and AMT, Intel's Active Management Technology, which can provide remote management even when the PC is switched off.

Inside, you can fit six drives by exploiting every port, and the case comes with a cooling system designed to keep it all cool. Accessories include wireless upgrade kits, an installation frame for hard drives, and an uprated 850W power supply should the built-in 500W unit prove insufficient. This is a small black cube with a lot of power inside, measuring just 33.2 x 21.5 x 19.0 cm (13 x 8.5 x 7.5 inches).

The recommended price is 483 Euros, which converts to $495 thanks to the current closeness of the two currencies. The case should be available from specialist retailers imminently.

Intel has revamped how it will deliver graphics driver updates to modern and legacy processors. The chipmaker has decided to move its 6th to 10th Generation processors to a legacy support model.

Under the new model, Intel will only provide critical fixes and security vulnerabilities for processors from the Skylake to Comet Lake family. That means that the aforementioned processors will no longer receive Day 0 game support updates. The change applies to all the SKUs in the lineup, including Core, Atom, Celeron, and Xeon chips. Instead, Intel will deploy the software updates quarterly or when it needs to address critical issues or security vulnerabilities.

Intel's 11th Generation Rocket Lake and newer processors will be the only chips to continue to enjoy Day 0 game support. Intel expects to launch regular updates through a standard monthly cadence.

The Intel Graphics Driver will now pack two drivers in the same package. It'll contain driver files for Intel's 10th Generation processors and older and driver files for 11th Generation chips and newer. Installation remains the same: you download and run the executable. The executable automatically picks the adequate driver for your system, so you don't have to ponder which one to install.

It makes sense why Intel would drop support for Skylake since the 14nm chips came out seven years ago. It seems odd, however, that Intel would forsake Comet Lake too since the processors are relatively new and only debuted two years ago. Although Intel's integrated graphics solution has improved tremendously over the years, we don't expect anyone to use an Intel iGPU for serious gaming. According to the latest Steam Hardware Survey, less than 2% of Steam users are gaming on Intel's UHD Graphics. Intel's new graphics driver model shouldn't affect most consumers since the chipmaker will continue to offer security updates. It's just that the old iGPUs won't have access to new games.

List of Affected Intel Processors

• 10th Generation Intel® Core® processors with Intel® Iris® Plus graphics (Codename Ice Lake)
• 10th Generation Intel® Core® processors with Intel® UHD Graphics (Codename Comet Lake)
• 9th Generation Intel® Core® processors, related Pentium®/Celeron® processors, and Intel® Xeon® processors, with Intel® UHD Graphics 630 (Codename Coffee Lake-R)
• 8th Generation Intel® Core® processors, related Pentium®/ Celeron® processors, and Intel® Xeon® processors, with Intel® Iris® Plus Graphics 655 and Intel® UHD Graphics 610, 620, 630, P630 (Codename Kaby Lake-R, Coffee Lake)
• Intel Pentium® and Celeron® processor family (Codename Gemini Lake)
• 7th Generation Intel® Core® processors, related Pentium®/Celeron® processors, and Intel® Xeon® processors, with Intel® Iris® Plus Graphics 640, 650 and Intel® HD Graphics 610, 615, 620, 630, P630 (Codename Kaby Lake)
• 6th Generation Intel® Core®, Intel® Core® M, and related Pentium® processors with Intel Iris® Graphics 540, Intel® Iris® Graphics 550, Intel® Iris® Pro Graphics 580, and Intel® HD Graphics 510, 515, 520, 530 (Codename Skylake)
• Intel® Pentium® Processor family and Intel® Celeron® Processor family (Codename Jasper Lake),
• Intel® Core® Processor with Intel® Hybrid Technology (Codename Lakefield)
• Intel® Atom®, Pentium® and Celeron® processor family (Codename Elkhart Lake)

Once upon a time, the notion of a pocket PC was just that - Windows XP’s touchscreen extensions packaged to fit in your Levi’s, ports and all. Then iPhones happened and everything became a black rectangle, so it’s good to see a proper pocket PC crowdfunding on Indiegogo: the Higole PC from rugged tablet company Gole out of Shenzhen, China.

Don’t go expecting a pants pocket powerhouse, however. The Higole uses the Celeron J4125 (Gemini Lake 14nm, four cores, four threads, 2.7GHz boost, 10W) from 2020, which is probably not going to trouble our hierarchy of the best CPUs for gaming, but which should acquit itself well if running a web browser or note-taking app. Oddly, the Celeron N4000 (Gemini Lake, 14nm, two cores, two threads, 2.6GHz boost, 6W) from 2017 is also available as a slightly cheaper option, the discontinued chip doing its best in the face of the competition.

Whichever Celeron you choose, it’s backed by 8GB of LPDDR4X RAM and integrated Intel UHD 600 graphics, allowing the PC to output 4K to a couple of extra screens. A variety of ports are scattered around the edge of the fanless case, including a pair of full-size HDMI 2.0s, mic and speaker, four USB 3 Type-As, one Type-C for power, Ethernet, audio jack, Micro SD, and a power button. Wi-Fi and Bluetooth depend on the CPU you choose, with the N4000 getting 5.0 for both, and the J4125 getting Wi-Fi 6 and Bluetooth 5.2. There's a 128GB SSD fitted, an internal 5,000mAh battery good for about four hours, and it comes with Windows 11.

The whole thing is squeezed into a case about the size of two smartphones pressed back to back: 5.6in tall, 3.6in wide, and 0.7 of an inch thick. The touch-enabled IPS display is 5.5in 1280x800px, and while the creators claim it will be the lightest PC in the world ever, we can’t see a weight figure quoted.

Still, one to watch, perhaps, especially as early bird offers are going for around $200 for the N4000 and $230 for the J4125 version. The crowdfunding campaign is already over 200% funded, with 42 days left, and the aim of shipping products in September. Remember that crowdfunding a project is not a guarantee of receiving a finished product. Backing a crowdfunded project is akin to an investment; you believe in the project and want it to succeed. You are not purchasing a retail product.

The Raspberry Pi is rather hard to get a hold of right now (we have a few tips on where to buy a Raspberry Pi and a list of best Raspberry Pi deals if you have your heart set on one) but you can still find alternatives on the market for example the Khadas VIM 4.

The latest board in the line up of UP products, is the UP 4000 board. The UP 4000 offers greater options than the previous UP boards with more memory, more IO support, and 12V power input. The processor choices are firmly based on Intel, rather than Arm. The choice of processors ranges from an Intel Atom x7-E3950 which can reach up to 2.0 GHz, the Intel Celeron N3350 reaching as high as 2.4 GHz, Apollo Lake Intel Pentium N4200 which caps out at 2.5 GHz.

UP 4000 CPU Options

There are six varieties of the UP 4000 available to purchase. As of writing, preorders are currently being accepted with a 10% discount offered until July 15th. The UP 4000 boards are planned to be shipped in August of 2022.

UP 4000 General Specifications

The UP 4000 board has a small form factor of 3.37 x 2.22 inch (85.6 x 56.5mm) and uses the same pin definition as a Raspberry Pi, but this does not guarantee compatibility with HATs. The UP 4000 has onboard TPM 2.0 which adds an extra layer of hardware-based security necessary for Windows 11. The team are also planning to launch a carrier board for the UP 4000 in late 2022. The carrier board brings M.2 slots for AI, 5G and Wi-Fi expansion.

If you want to take a closer look at the new UP 4000 boards check out the UP 4000 product page and explore the full UP 4000 specifications. Preorders are available for purchase on the official UP shop website.

MSI will soon release its highly anticipated MSI MAG B660M Mortar Max motherboard for Intel Alder Lake processors, according to Twitter leaker Chi11eddog. The special sauce this motherboard centers on is a Renesas RC26008 external clock generator, which will allow BCLK overclocking for the first time on a shipping B660 motherboard. Moreover, if this leak is correct, the Renesas chip will bring PCIe 5.0 support to this Micro ATX DDR4 motherboard.

We saw the MSI MAG B660M Mortar Max in May when the same Twitterer shared some purported images of this motherboard and a close-up of the external clock generator sitting proudly on the PCB.

The introduction of this motherboard could be critical to the PC DIY community. If you read the stories about BCLK overclocking on Alder Lake non-K processors from early this year, you will know why. In brief, some of Intel's 'locked' processors like the Core i5-12400, Core i5-12490F, and even the Core i3 -12100 and Celeron G6900 are very impressive overclockers if your motherboard lets you have your wicked ways. However, most of these fantastic overclocking feats rely on not just any Z690 motherboard but premium models like the ASRock Z690 Aqua or the MSI MEG Z690I Unify, which have external clock generators.

So, Chi11eddog has spilled the beans that the MSI MAG B660M Mortar Max motherboard will arrive in July, but he also kindly leaked some authentic-looking images, one of which appears to show the back of the motherboard box with full specs.

In addition to the specs listed above, MSI boasts features such as its extended heatsink design, the M.2 Frozr Shield heatsinks, its memory boost technology, and core boost digital power design.

Last but not least, and in keeping with our headline, MSI's packaging highlights the use of the OC Engine with clock generator, allowing for BCLK overclocking and "releasing the maximum power of the CPU."

MSI MAG B660M Mortar Max Specifications

To make the MAG B660M Mortar Max a success, MSI mustn't charge too much of a premium for this 'Max' addition to its product portfolio. The standard MSI MAG B660M Mortar WIFI DDR4 was reviewed in January, winning some acclaim, and currently sells for $159.99 on Newegg.

If the Max edition price is too high, PC DIYers and enthusiasts might easily resist the Mortar Max's charms and miss out on a lot of potential Alder Lake non-K overclocking fun. In other words, an inexpensive motherboard is the best match for the affordable CPUs that would gain the most from being unleashed by the Mortar Max. But, of course, those with deep pockets can already buy one of the Intel Alder Lake K processors and their pick of a wide range of Z690 motherboards.

The new Zotac ZBOX PI336 Pico (brought to our attention by Liliputing) looks to be a portable and decently equipped pocketable device, for anyone in the market for a portable media player. It has dual 4K outputs and comes with Windows 11 Pro installed, making it a good digital signage option, but it can also serve other uses.

The presence of a Celeron chip likely won't make anyone jump for joy, but the N6211 chip in this machine acquits itself quite nicely, taking its two cores to a 3GHz burst while drawing just 6.5W of power. It’s a 2021 model built on the 10nm Elkhart Lake architecture, and it comes with 10th Gen Intel UHD graphics.

The CPU supports up to eight PCIe lanes, but they’re not getting much work here as there's neither M.2 nor PCIe slots for cards. You get 128GB of eMMC storage along with 4GB of RAM (a shame, as the chipset supports up to 32GB), and some USB 3.1 Gen 2 ports, including one that’s Type-C. Additional connectivity is taken care of via Wi-Fi 6E, Gigabit Ethernet, and Bluetooth 5.2. There's a Micro SD slot for extra storage.

Video out is where the tiny PC excels, with HDMI 2.0 and DisplayPort 1.4 outputs capable of driving two 4K screens. That makes it useful for a tiny media PC, or to drive retail or other digital display applications. It all squeezes into a case with a volume of 0.18L (11 cubic inches) and dimensions of 115 x 76 x 20.7mm (4.3in x 3in x 0.8in). We've seen larger and thicker wallets!

It also manages to be passively cooled, with the case acting as a heat sink in the same way as other tiny PCs we've seen recently. There's a 5V DC input at the side of the case, so a separate power brick is required and helps to avoid adding heat to the PC interior.

There's no confirmed release date yet nor pricing, but it’s nice to see a mini PC using an Intel chip and running Windows instead of some Arm/Linux combination. Zotac’s previous small-form-factor model, the ZBOX Magnus, arrived earlier this year, but it  was much more powerful and also much larger than this diminutive video-pusher.

A device packing an AMD Mero APU has emerged online. According to a sys-info screenshot shared by Rogame on Twitter, the device is the Magic Leap Demophon, rumored to be an AR headset that is currently in development.

The Tweeted screenshot above offers quite a lot of information to decode and take in. For example, AMD's Mero APU appears to be an octa-core CPU partnered with a Mero GPU. But what is a Mero GPU? Thankfully, we have some previous Mero tech leaks and spills to look back on and provide clues.

In 2020, PC hardware hound Komachi shared a screenshot of a document that tabulated functions and device IDs of AMD APUs. Specifically, it mentioned AMD Renoir, Van Gogh, and Mero. Komachi later clarified that Mero and Van Gogh were "the same." However, some changes or tweaks may exist to distinguish between AMD Van Gogh and AMD Mero APUs.

AMD's Van Gogh APU famously powers the Steam Deck, but due to being customized for Valve, it doesn't carry any AMD codenames, just a line saying 'Powered by AMD.' With the Steam Deck launched, reviewed, and in the hands of many a tech enthusiast and gamer, we now have a good understanding of the Van Gogh APU. In March this year, when the Steam Deck was released, we featured it in an in-depth review, and you can check out the complete device and APU specs there.

In brief, the Steam Deck's Van Gogh APU offers the following processing technologies; a Zen 2 CPU cluster with 4C/8T running between 2.4 and 3.5 GHz and an RDNA2 GPU with eight compute units running at between 1.00 and 1.6 GHz.

Checking back on the purported Magic Leap Demophon device specs, we see it is quite a different proposition to the Steam Deck. The contrast in purpose and function means that Magic Leap may have got AMD to implement a few APU design tweaks and rebalances to suit it better than a PC gaming handheld.

Magic Leap's Android OS AR device is likely going to be for the central part projecting 3D rendered elements into your field of vision, popping up some widgets and menus, as well as simultaneously processing a wide range of sensor data. Makers of wearables are very reluctant to pack in large and heavy battery packs. Due to device comfort and portability, we expect this mysterious Mero APU to be less power-hungry than the version in the Steam Deck.

Things will get very interesting in the mobile space thanks to AMD moving its newer gen CPU and GPU cores into low-power APUs. As well as the Magic Leap Demophon, we'd like to see Van Gogh-a-like APUs inaccessibly priced laptops and 2-in-1s that currently rely on weedy Celeron or Pentium chips.

The latest version of the CPU-Z hardware diagnostic and benchmarking software suite adds support for multiple yet-to-be-released processors and graphics processors, including those from AMD, Intel, and even the first discrete GPU from Chinese CPU developer Zhaoxin.  

Support by CPU-Z shows that certain CPUs and GPUs are indeed incoming; their respective vendors have supplied developers of the program with the information required to enable the detection of appropriate hardware. Typically, support by utilities like CPU-Z and HWInfo is an indicator that the development of certain upcoming products is near the finish line. 

Among the pieces of hardware now supported by CPU-Z version 2.01 for Windows are the following parts:

• AMD Rembrandt and Raphael APUs (RDNA 2).
• AMD Mendocino APU (Zen 2 + RDNA 2).
• Preliminary support for Intel Arc 3/5/7 (DG2).
• Preliminary support for Intel Raptor Lake (13th gen Core).
• Zhaoxin's Glenfly Arise-GT10C0 GPU.

AMD's Ryzen 6000-series 'Rembrandt' APUs set to use Zen 3+ cores are designed primarily for notebooks and are expected to hit the market shortly. By contrast, AMD's Ryzen 7000-series 'Raphael' are the company's all-new processors featuring the Zen 4 microarchitecture, are aimed at desktops, and are scheduled to arrive later this year. AMD's Mendocino is a somewhat less well-known name, but this one is a Zen 2-based mobile APU with an RDNA 2-powered integrated GPU targeted at gamers. Interestingly, readers with a good memory might remember Intel using the Mendocino codename for its 2nd Generation Celeron processors in 1998 – 1999. 

Intel's Arc Alchemist graphics processors are due to be released in high volumes in Q2 – Q3 for both laptops and desktops, so it is about time for Intel to add its support to diagnostic software. Meanwhile, Raptor Lake is expected to be released in Q4, yet Intel's partners are already testing this CPU, so they need CPU-Z. 

Unfortunately, we do not know much about Zhaoxin's GlenFly Aris-GT10C0 graphics chip. However, we understand that this part is a DirectX 11.1 or DirectX 12-capable GPU made using TSMC's 28nm fabrication process and featuring a 70W thermal design power. Perhaps, Zhaoxin is starting to ship this chip to interested parties and needs to add its support to CPU-Z. 

In addition, the latest version of CPU-Z supports numerous already existing or shortly incoming processors:

• Intel Core i9-12900T, Core i5-12600T (35W).
• Intel Atom x6427FE, x6425RE, x6425E, x6414RE, x6413E, x6212RE, x6211E, x6200FE (Elkhart Lake, FCBGA1493).
• Intel Pentium J6425, N6415 (Elkhart Lake, FCBGA1493).
• Intel Celeron J6413, N6211 (Elkhart Lake, FCBGA1493).
• AMD Ryzen 7 5700X, Ryzen 5 5600/5500.
• AMD Ryzen 3 5300GE, Ryzen 3 PRO 5350GE, Ryzen 5 PRO 5650GE, Ryzen 7 PRO 5750GE (Cezanne).
• AMD Ryzen 9 6980HX, 6900HX, Ryzen 7 6800H, Ryzen 5 6600H (45W).
• AMD Ryzen 9 6980HS, 6900HS, Ryzen 7 6800HS, Ryzen 5 6600HS (35W).
• AMD Ryzen 7 6800U, Ryzen 5 6600U (15-28W).
• AMD Ryzen 7 5825U, Ryzen 5 5625U, Ryzen 3 5425U (15W).
• AMD Radeon RX 6850M XT GPU (Navi 22).
• AMD RX 6800S, RX 6700S, RX 6650M, RX 6650M XT GPUs (Navi 23).
• NVIDIA GeForce RTX 3090 Ti GPU (GA102-350, 450W).
• Improved information accuracy when core isolation is enabled.
• Improved validation process for high clock submissions (>6GHz).

Biostar has announced the new iMiner 660MX8D2 with eight Radeon RX 6600 graphics cards for cryptocurrency miners. And no, it's not an April Fool's Day joke.

Believe it or not, Biostar is pretty active in the cryptocurrency mining scene. The vendor offers everything from mining-oriented motherboards to complete, plug-and-play mining machines. Biostar previously employed AMD's Polaris-based graphics cards, such as the Radeon RX 580 or RX 570. This time, the manufacturer opted for an RDNA 2 worker for the iMiner 660MX8D2, more specifically, the Radeon RX 6600M.

The iMiner 660MX8D2 utilizes eight Radeon RX 6600M graphics cards, the mobile variants of the Radeon RX 6600. However, the Radeon RX 6600 and RX 6600M use the same Navi 23 silicon with an identical shader count, with the only significant difference being the clock speeds. Therefore, Biostar probably just repurposed the mobile dies into desktop graphics. However, it's a common practice, and we've previously seen it with Nvidia's mobile GeForce RTX 3060 graphics card.

Biostar's Radeon RX 6600M 8GB features a cooler with a dual-heat pipe and dual-fan design. The graphics card has a single 8-pin PCIe power connector and offers three DisplayPort outputs and an HDMI port, which aren't necessary for a mining graphics card. It's a shame because these are legit gaming graphics cards that could have gone into the hands of gamers.

The eight Radeon RX 6600M graphics cards reside on Biostar's custom TB360-BTC D+ motherboard with the B360 chipset that supports old 8th and 9th Generation Intel processors. Biostar outfits the iMiner 660MX8D with a feeble Celeron G4900 dual-core chip to minimize power consumption and a 120GB SSD for storage. Unfortunately, there's only a single SO-DIMM memory slot for a DDR4-2666 stick up to 16GB. However, the important thing is that the motherboard provides eight PCIe x16 expansion slots (seven PCIe 2.0 x16 at x1, one PCIe 3.0 x16), so no riser cards are required. In addition, Biostar includes a 2000W 220V power supply to feed the eight RDNA 2 graphics cards.

Biostar advertises the iMiner 660MX8D2 as a hassle-free mining system; even beginners can get it up and working. It's a plug and mine product. The machine boasts an ETH hash rate of 248 MH/s, around 7% higher performance than Biostar's prior iMiner A588x8D2 miner with eight Radeon RX 580. The miner supports all the popular cryptocurrencies, including Ethereum, Ethereum Classic, Monero, BTG, and Zcash - to name a few.

The manufacturer didn't share the price tag for the iMiner 660MX8D2, so potential customers will have to send an inquiry to Biostar about pricing.

Prime Computer of Switzerland has launched a new laptop called the PrimeBook Circular, heralding the new laptop as being modular and "100% climate neutral." The PrimeBook appears to have some quality build choices; its chassis, screen, input devices, and so on appear to be premium in nature. This is important as these aren't upgradeable parts. However, the processor (Intel Tiger Lake), RAM, and Wi-Fi solution all live in an easily replaceable module that's about the size of a 2.5 inch SSD.

You may feel there is something familiar about the PrimeBook Circular, and that is because it appears to be a Prime Computer-branded version of the Intel NUC P14E whitebook design. The specs of the host laptop match, as does the replaceable modularity. Though it isn't mentioned in the press release or official product pages, the video below includes a short section that confirms the PrimeBook relies on Intel NUC 11 Compute Element modules. Moreover, the three PrimeBook SKUs align with three of the available NUC 11 Compute Element modules you can see in the Intel Ark.

Prime Computer brings some of its green and sustainable ethos to this laptop design, making it more than just a rebrand. For example, Prime Computer will offset the average 408kg of CO2 used when manufacturing this device to make the product "carbon neutral." It is also setting up a program to repurchase and repurpose the modules which you no longer want/need, as and when you upgrade.

Before going on to look at the module specs you can choose, let us have a look at the rest of the Prime specs which, if you buy in, you will have to live with for multiple processor generations.

Starting from the outside, the laptop body is made of CNCed anodized aluminum. Inside this sturdy frame are a 13.9 inch IPS touch screen with 3000 x 2000 pixels with 100% sRGB coverage and a max brightness of 400 nits. There's an HD webcam in the bezel that supports IR for Windows Hello.

On the other side of the clamshell, you will find a backlit keyboard (various layouts are available for your language/region) underlined by a large glass surface touchpad, and there's a fingerprint reader too.

A decent set of ports is available on this thin (34 x 23 x 1.6cm) and light (1.5kg) design. It has HDMI 2.0b, Mini DisplayPort 1.4a, Thunderbolt 4/USB 4 via Type-C, 2x USB 3.2 Gen2 Type-A, Gigabit Ethernet, and a 3.5mm stereo jack. There are built-in speakers and a mic too, naturally. Users will be able to fast charge the laptop, or rather its 77Whr battery, via a 65W USB-C power supply.

PrimeBook Circular Compute Element Specs

The above RAM amounts specified above are fixed, as the module has soldered RAM and no spare slot. Prime Computer's configurator also lets you choose the keyboard, storage (single M.2 slot) of none, 250GB, 500GB, or 1TB, as well as OS of none, Ubuntu, Windows 10 or 11. Going through the configurator and submitting your choices will precipitate an email, presumably with availability and pricing info. We don't have that information right now.

On the surface, the modules don't seem entirely green or sustainable, especially as they have non-upgradable RAM. However, Prime Computer's buy-back and resell idea helps gloss over that concern. Hopefully, the supporting structure/components will last long enough to stay serviceable long enough to last multiple Compute Element generations, and Intel will keep up with the production of these standard-sized compute blocks. Unfortunately, laptop components seem to age at different rates, though this sounds like a decent platform to settle with for a few years.

The PrimeBook Circular provided some echoes of the Framework laptop project. However, Framework attacks the customization and e-waste reduction initiative from a different angle with its easy reparability and custom (USB-C) slot-in I/O modules. Our laptops editor reviewed the Framework design last summer, if you want to read more about this rival offering. More recently, we reported on the Marketplace for upgrades it launched and started up its Expansion Card Developers Program.

AMD looks set to release a salvo of new processors to try and recapture some of the gains made by Intel's extensive set of Alder Lake desktop CPU releases this year. Prolific PC hardware leaker momomo_us has spotted some European retail pricing for the new and as yet officially unannounced AMD CPUs. We must take this with a pinch of salt, but are happy to see the reliable source has included both pre-and post-VAT prices for us to ponder over this time.

As well as the glamorous and highly anticipated new Ryzen 7 5800X3D, AMD is fleshing out its Ryzen 5000 range with some inbetweeners like the Ryzen 7 5700X and a decent selection of processors to compete directly against Intel's surge in sales in the lower-end Core i5 and i3 performance (plus Pentium and Celeron) categories. This thrust will be represented by the likes of the Ryzen 5 5600, 5500, 4500, and Ryzen 3 4100.

We reported on most of these processors being confirmed by a retailer in the Philippines only a couple of days ago, but now we have the actual pricing. The table below contains the Ryzen lineup, with new processors shown in italics. In addition, we have original Euro pricing alongside US dollar equivalents for your reference.

In addition to the above, the leaker's SKU pricing data suggests we will get a new Ryzen 5 4500 CPU at $146 / €132 and a Ryzen 3 4100 at $115 / €104. Due to being in the Ryzen 4000 family, we expect some of them to be based upon the Zen 2 CPU architecture or a modernized/modified form of it.

Data also indicates a trio of new Ryzen 5000 APUs (codenamed Cezanne, mixing Zen 3 CPU and Vega GPU) will be arriving. Oddly, these are unnamed in the inventory and price data, but one will be a 4C/8T part, with the other two offering 6C/12T configurations.

Referencing back to our earlier report on these processors again, rumors suggest that some of the new lower-end Ryzen chips fleshing out the series have been created from Cezanne APUs which didn't make the grade due to iGPU issues.

The logic behind the new Ryzen 4000 series (Zen 2, Renoir-X) CPUs isn't as clear, but we must consider the extent and depth of Intel's Alder Lake SKUs and AMD's desire to answer as best as it can. It would be hard to deny that AMD has neglected lower- and mid-range price points for quite some time.

With purported pricing beginning to leak, it shouldn't be long until we see AMD officially launch and release the above processors. Launch press releases and presentations will give us a better idea of how some of the more unexpected releases outlined above fit into its plans, and who might be expected to purchase these CPUs and APUs.

CPU-Z has been officially updated to version 2.00, and with it comes support for a plethora of new CPUs and one new GPU. Most notably, AMD Ryzen 7 5800X3D and Intel Core i9-12900KS support is here. These two new CPUs haven't been released yet, but promise to push the performance envelope of their respected architectures as high as possible.

The full details of version 2.00 can be found below. Other additions include support for several of Intel's P series and U series mobile Alder Lake processors, as well as the Intel Pentium 8500, 8505, and Celeron 7300 and 7305. 

Other additions include Intel's new mid-range and entry-level core i3 and i5 desktop processors such as the 12500, 12400, 12300, and 12100.

On the AMD side, preliminary support for AMD's Ryzen 6000 series mobile APUs has been added, along with the newly released Radeon RX 6500 XT desktop GPU (which we must say is late to the party).

• Intel Core i9 12900KS
• Intel Core i7-1280P/1270P/1260P, Core i5-1250P/1240P, Core i3-1220P (28W)
• Intel Core i7-1265U/1255U, Core i5 1245U/1235U, Core i3 1215U (15W)
• Intel Core i7-1260U/1250U, Core i5 1240U/1230U, Core i3 1210U (9W)
• Intel Pentium 8505, Celeron 7305 (15W)
• Intel Pentium 8500, Celeron 7300 (9W)
• AMD Ryzen 7 5800X3D
• AMD Radeon RX 6500 XT
• Preliminary support of AMD Ryzen 6000 "Rembrandt" APUs

5800X3D & 12900KS

The Ryzen 7 5800X3D could be one of the last Zen 3 desktop CPUs to launch before Zen 4 arrives later in 2022. The 5800X3D improves on the already superb Zen 3 architecture by adding AMD's brand new 3D V-Cache technology to the chip, giving the speedy octa-core CPU an additional 32MB of L3 cache for a total of 96MB.

In a recent AMD press event, CEO Lisa Su gave a demonstration of its Ryzen 7 5800X3D running several games in comparison to a Ryzen 9 5900X, and more interestingly, Intel's current flagship Core i9-12900K.

According to AMD, the Ryzen 7 5800X3D offers a 1.2 to 1.4x average performance jump over its bigger brother, the Ryzen 9 5900X. When compared to the 12900K, the results were tied on three of the games, with the rest yielding a 1.1x to 1.2x performance improvement in favor of the 5800X3D.

Overall, AMD measured an average 15% gain in performance over its current Zen 3 CPUs. That's a rather impressive gain considering the 5800X3D has a 400MHz clock speed deficit compared to the stock 5800X.

However, pricing and availability remain unknown. Due to the additional material required to make 3D V-Cache a reality, we wouldn't be surprised to see the 5800X3D cost just as much as a 5900X. Plus, with Zen 4 just around the corner, the 5800X3D could be one of AMD's shortest-lived CPUs to date, with barely a couple of months in the limelight.

The Core i9-12900KS will be Intel's new flagship for the Alder Lake generation, with the S marking the chip as a special edition product. The only thing changing with the 12900KS is higher clock speeds with a supposed 3.4GHz base clock and 5.2GHz all-core boost clock on the P cores. That is a 200MHz improvement compared to the 12900K's base clock, and it can only do 5.2GHz on a single core.

According to recent reports we've covered, the performance of Intel's new special edition chip isn't that far off from its current 12900K twin, with 4% additional single-core performance and an 11% boost multi-core performance in a recent Geekbench 5 score.

Other limited edition or special edition parts such as the Core i7-8086K and Core i9-9900KS featured slightly higher clock speeds to earn their namesake, but the difference in performance was so slight compared to their vanilla parts that they barely mattered at all.

With Intel planning to launch the Core i9-12900KS soon, it'll be the first time that a special edition part has been released in two generations (the last being the Core i9-9900KS). Supposedly, Intel had plans to introduce a Core i9-10900KS Comet Lake CPU to the market during the Rocket Lake generation, but those plans never came to fruition.

Intel has already released dozens of 12th Generation Core 'Alder Lake' processors for desktops, laptops, and ultra-thin laptops. But there are other categories of PCs that are yet to be addressed by the new family: ultra-low-power and ultra-low-cost computers. Intel is prepping its Alder-Lake N system-on-chips that only feature small energy-efficient Gracemont cores for these PCs, but these SoCs may actually be pretty serious performers. 

Intel's Alder Lake-N processor apparently packs up to eight cores based on the energy-efficient Gracemont microarchitecture that come in two quad-core clusters with 2MB L2 cache per cluster and a shared L3 cache, according to a partial boot log of Alder Lake-N (ADL-N RVP) uploaded by Intel to the Linux repository in the Sound Open Firmware (SOF) Project (and discovered by Coelacanth's Dream). 

Gracemont is a 5-wide out-of-order architecture with enhanced branch predictor with pattern recognition, new integer and floating point execution units, increased number of execution ports, and even L3 cache support. Eight Gracemont cores promise to offer quite formidable performance even when compared to previous generation 'big' cores like Skylake. That said, Intel's forthcoming low-power/low-cost PC platform (at least in its eight-core incarnation) may offer performance on par with premium PCs released some three or four years ago.

The SoC also has an integrated GPU with 32 execution units based on Intel's Xe-LP architecture, media, and display engines. Intel's Xe-LP has up-to-date media encode/decode capabilities (with AV1, H.264, H.265, Dolby Vision, and 8Kp60 support), so expect Alder Lake-N systems to offer premium multimedia playback features.  Just what the doctor ordered for home theater PCs. 

It remains to be seen how Intel plans to market its Alder Lake-N processors, but it is reasonable to expect this design to power next-generation Pentium Silver and Celeron CPUs for desktops, thin client, NAS, embedded systems, and low-cost notebooks. Eventually, Intel might offer some specialized Atom-branded processors based on Gracemont cores, though they will not necessarily be based on the Alder Lake-N design.

MSI has finally released its highly anticipated MEG Z690 Godlike motherboard. It claims that this premium board deserves the title of "One Board to Rule Them All," with regard to Intel's Alder Lake platform. At this end of the market, with price seemingly no object, MSI had better deliver. The retail price of the MSI MEG Z690 Godlike is $2,099 and, as per our headline, it is claimed to be available now. However, you have to go through a registration process if you are interested.

Intel launched its unlocked 12th Gen Intel Core Processors in late October 2021, and MSI had some timely Z690 chipset motherboards available to carry the initial wave of enthusiasts. MSI then started to tease the sequel to last year's MSI MEG Z590 Godlike. We reported on multiple MSI MEG Godlike teasing campaigns through November and December, before the official launch of the product in mid-December at $2,099 (remember this is a motherboard bundle with MSI MEG CoreLiquid S360 AiO CPU cooler and 32GB Kingston Fury Beast DDR5 RAM kit).

In case you missed our previous coverage of the MSI MEG Z690 Godlike specs and features, the highlight of this E-ATX design (other than the shock price and limited edition sales strategy) might be the M-vision touch screen. This 3.5-inch IPS real-time monitoring screen resides near the DDR5 slots by default, but can be removed and used as a USB-C tethered remote control if preferred.

Other key specs [PDF] of the MEG Z690 Godlike motherboard are:

• 4 x Memory slots supporting DDR5-6666 RAM
• 1 x Lightning Gen 5 128Gbps M.2 slot
• 5 x M.2 PCIe slots (4 x Gen 4, 1 x Gen3)
• 2 x Thunderbolt 4 ports
• 2 x Lightning USB 20G ports
• 8 x USB 3.2 Gen 2 ports
• Dual LAN (Aquantia AQC113CS 10G LAN, Intel I225-V 2.5G LAN)
• Intel Wi-Fi 6E module, Bluetooth 5.2
• Direct 20+2 Phases with 105A Smart Power Stage power design
• Dual 8-pin power connectors with Core Boost technology
• 8 Layer Server Grade with 2oz Copper PCB

Please note that the MEG Z690 Godlike is a limited edition product from MSI. The PCs, accessories and components maker is offering a "prioritized qualification to those who purchased a previous generation of Godlike motherboard or an MSI RTX 3080 or 3090 series graphics card." In other words, it seeks to reward brand loyalty and previous Godlike owners with this product. You must reside in one of the following regions: PRC, JP, KR, AU, TW, AE, TH, VN, USA, DE, UK, ES, FR, RU, DK, FI, SE, or NO.

Now the MSI MEG Z690 Godlike is finally "available" those all-important reviews are going to emerge, so you can really see if you get what you pay for, and this latest generation Godlike lives up to its pedigree. We should also find out whether it can OC your Celeron G6900 to unheard of MHz.

The release of Intel's Alder Lake processors and accompanying new motherboards seems to have inspired a significant wave of new benchmark world records. Of course, many of the achievements were stemmed from CPU-based tests, but over recent days some important 3DMark graphics benchmark world records have also fallen. Both the Fire Strike and Fire Strike Extreme world records have been claimed by overclocker Biso Biso wielding an AMD Radeon RX 6900 XT. Meanwhile, the 3DMark2001 SE world record has been claimed by overclocker Rauf using an Nvidia GeForce GTX 980 Ti.

3DMark Fire Strike / Extreme World Records

A 3DMark Fire Strike score of 62,389 was achieved by Korean OC expert Biso Biso on Thursday. This record is notable for a couple of reasons. First, it was done using the AMD Radeon RX 6900 XT pushed to a 3,147 MHz core clock. Secondly, it unseated a Quad-SLI Nvidia GeForce GTX 1080 Ti-powered record by Kingpin (61,206) that had stood for almost two years.

On Sunday, Biso Biso achieved another world record pole position, this time in Fire Strike Extreme 1x GPU, and validated by HWBot. Checking through the details of the feat, you can see it confirmed that the Korean overclocker used liquid nitrogen (LN2) cooling on both the Intel Core i9 12900K and XFX Radeon RX 6900 XT used in the benchmarks.

This Fire Strike Extreme record of 39,515 wasn't a big leap over the previous holders, and the top 10 all appear to use the same GPU, so this test looks like a sweet spot for AMD's top-end consumer GPU.

3DMark2001 SE World Record

Yes, people still compete fiercely over achieving the highest score in 3DMark2001 SE, which was released in, you guessed it, 2002. With such aging software, it might not be surprising that only older GPU architectures can run, using older OSes, to complete the benchmark successfully. The new world record, recorded on Sunday by Overclocker Rauf, stands at a 253,975 marks, as validated by HWBot.

Rauf used an Intel Core i9 12900K with eight P-cores active, on an Asus ROG Maximus Z690 Apex motherboard, with DDR5 RAM, to grab this world record. The GPU behind his success was an Asus GeForce GTX 980 Ti Matrix Platinum Edition.

Interestingly, the CPU was cooled using LN2 and running at 7,125 MHz, but the GPU was air cooled. In the photo you can see that the GPU is covered in paper towel to prevent condensation from the CPU pot affecting or damaging it.

We have covered some other interesting benchmark and overclocking world records recently. Certain new Intel Alder Lake motherboards have a feature where users can unlock BCLK, which has inspired a flood of new records using chips you can't usually overclock, such as value-orientated Core i5, i3, Pentium and even Celeron CPUs.

We're sure many readers have had their imaginations fired up by the revelation that non-K Alder Lake processors can be overclocked using certain motherboards, especially given that this wasn't possible on previous-gen processors. We reached out to Intel for an official statement about the matter. In brief, Intel tells us that overclocking non-K CPUs is not covered by the warranty and may cause damage to the processor and associated PC components. The statement implies that this new overclockability isn't sanctioned by Intel, but the company hasn't stated that directly.

Last weekend, we reported on overclocker Der8auer's first video, in which he shared his discovery of BCLK overclocking options for locked Alder Lake-S CPUs. The story developed further on Monday when he tested some lower-end ADL-S CPUs like the Celeron G6900 and Core i3-12100. The overclocking feats achievable with these humble parts were quite breathtaking. For example, the Celeron achieved a 57% OC (hitting 5,338 MHz), with standard off-the-shelf cooling.

Moving forward a day or two we noticed that, as predicted, the non-K ADL-S CPUs were kicking up a storm on HWBot, grabbing all sorts of world records. This was particularly apparent in the quad-core charts, as Intel hasn't produced an unlocked quad-core since its 7th Gen Core processors. In the same article, we first noted that a trio of Intel B660 chipset motherboards also appear to have unlock BCLK functionality in their BIOS: the Asus Strix B660G, Asus B660F and the ASRock B660 Steel Legend. These were still pretty high-end, high priced B660 samples, but thankfully much more accessible than the Asus ROG Maximus series where this non-K ADL-S overclocking was uncovered.

Earlier today we saw der8auer mention that "one of the big manufacturers is working on a B660 board with DDR4 for Non-K OC," which could bring the entry ticket price to ADL-S overclocking fun down much further.

Intel's Statement on Non-K Processor Overclocking

In recent hours, Intel commented to Tom's Hardware about non-K processor overclocking:

"Intel’s 12th Gen non-K processors were not designed for overclocking. Intel does not warranty the operation of processors beyond their specifications. Altering clock frequency or voltage may damage or reduce the useful life of the processor and other system components, and may reduce system stability and performance."

We had been worried that enthusiasts overclocking Intel's non-K processors would cause great consternation at Intel, with the expectation that this avenue to lower-priced OC antics will be closed pretty quickly. However, while Intel's statement appears to firmly disapprove of this overclocking activity, on the surface at least, it isn't suggesting it will be shut down ASAP.

"Errors" Will Be Fixed

It's rumored that the unlocked BCLK functionality for non-K CPUs was born of an error in Intel's microcode. Considering Intel's history of shutting down similar loopholes, it's likely that this non-K overclocking fun will be removed in any newer BIOS updates to boards where it is currently available.

On Sunday, we reported on the discovery that base clock (BCLK) overclocking is now available on non-K Intel Alder Lake-S CPUs. The only prerequisite seemed to be that you would need a high-end Z690 motherboard with a BCLK unlock feature in the BIOS. Today, OC expert der8auer, aka Roman Hartung, continued his investigations to the lower ranks of the ADL-S line-up and discovered that the lowly Intel Celeron G6900 could run at 5,338 MHz – a 57% overclock.

Celeron G6900 hits 5,338 MHz

Regarding the Alder Lake Intel Celeron G6900, in the embedded video you can see der8auer install this CPU in the same Asus ROG Maximus Z690 Apex motherboard as used in his previous video, and in a few moments push its two non-Hyperthreaded cores to beyond 5 GHz. The Celeron required a significantly higher BCLK push to get it beyond 5 GHz, as it started from such a low clock speed. Interestingly, der8auer noted he made the first use of the hardware buttons on the Apex motherboard for nudging BCLK in real-time, in order to achieve a BCLK of 157 MHz, and a resulting 5,338 MHz with the x34 multiplier.

In benchmarks and tests the overclocked Intel Alder Lake Celeron G6900 still didn't really shine, however. Remember, this is still a 2C/2T processor in 2022, and though this is a great increase for the Celeron it isn't enough to push past other Celeron limitations.

If you want to read and see more information about the Intel Celeron G6900, on Saturday we reported on a YouTuber's gaming benchmarks on a system based around this processor. Spoiler alert – this ADL-S 2C/2T chip was not very good in AAA games from the last half decade or so.

Core i3 -12100 hits 5,400 MHz across all four cores

Another processor featured in the video today was the Intel Core i3 -12100, with a much more useful configuration of 4C/8T. Only a few years ago Intel Core i7 chips rocked 4C/8T, and der8auer sees much more potential for HWBot world record attempts with this sub-family of the ADL-S range. The overclocker was impressed by the ease of achieving this all-cores 5,400 MHz overclock with the i3, which is about 26% faster than the standard boost clock.

In Cinebench R20 multi-threaded tests, the overclocked Intel Core i3 -12100 was able to closely challenge the likes of the AMD Ryzen 5 5600X. Intel's Core i3 is approximately half the price of the Ryzen.

With the above tests der8auer had moved from an AiO CPU cooler to a more capable liquid loop using a Hydro X Series XC7 Pro block from Corsair. Moreover he wanted to point out that he had also upgraded the thermal paste to Thermal Grizzly Kryonaut Extreme.

Once overclockers get a feel for the full ADL-S range we should expect a slew of new world records on HWBot, when we will see which processors respond best to extreme cooling using LH2, Helium and so on.

Summing up his video, der8auer was obviously impressed and said that he hadn't seen overclocks of this scale since the Intel 775 era.

If you are interested in choosing the right gaming CPU for your budget in 2022, please check out our Best CPU for Gaming in 2022 guide. 

Intel's NUC 11 Essential Kits, previously codenamed "Atlas Canyon", have now officially launched. These are low-power NUCs for entry level computing and not normally on the radar of PC enthusiasts, but the new 10nm Tremont based Jasper Lake SoCs inside can deliver a significant CPU and GPU performance uplift vs previous generations. Rough comparisons from Intel, to whet your appetite, indicate that Jasper Lake CPU cores are around 33% faster than Gemini Lake, and the GPU is even more convincing delivering up to a 78% uplift.

We first reported on Atlas Canyon NUCs last May, when a product data sheet was unearthed and shared by Fanless Tech. At the time the source tipped a release of Q1 2022, and this has turned out to be accurate.

Pondering over the new Intel NUC 11 Essential Kits, you can see there are three CPU choices available to would-be buyers of the five newly launched SKUs. The choice is basically between barebones, full systems and bare boards with the Intel Celeron Processor N4505 with 2C/2T at up to 2.90 GHz, Intel Celeron Processor N5105 with 4C/4T at up to 2.90 GHz, or Intel Pentium Silver N6005 Processor with 4C/4T at up to 3.30 GHz. These processors are all 10W parts with Intel UHD graphics. WhIle the Celerons support up to 16GB of RAM, the Pentium is capable of supporting 32GB of DDR4-2933 via two SO-DIMM slots. Unsurprisingly, the Pentium has the best GPU too, with 32 Execution Units helping it stand out from the Celerons with 16 or 24EUs. Storage is via one M.2 2280 SSD, plus an optional 64GB eMMC.

Considering the fuller system specs, one of these Intel NUC 11 Essential Kits measures 135 x 115 x 36mm like the classic compact low-power NUCs before it. 

Ports/connectivity features include 1x DP1.4 and 1x HDMI 2.0b video outputs for dual displays, six USB ports (2 x front and 2 x rear USB 3.2; 2 x USB 2.0), 1 x 3.5mm stereo out jack, 1 x 3.5mm microphone jack, 10/100/1000 LAN, Intel Wireless-AC 9462 + Bluetooth 5.0. A barrel power jack takes input from the 65W 19V DC power brick. These systems are designed to support Windows 10, 11, and Linux distros.

Previous gen entry level NUCs started at around $200 for a bare bones offering with the cheapest CPU option, hopefully this will also be the starting price for the Atlas Canyon NUCs. We don't have an availability date to share but these low-power NUCs should start to show up at retailers and etailers soon.

Earlier in the week, we reported how the new Celeron G6900 from Intel's 12th Generation Alder Lake family matched the Core i9-10900K (Comet Lake) in single-threaded performance. In addition, the Geekbench benchmark results highlighted the punch of the new Golden Cove cores. However, it is much better to see and test processor performance in the real world, and Steve at Random Gaming in HD (RGIHD) has obliged with a nine-game test suite.

To recap the Intel Celeron G6900's specifications, it has two Golden Cove cores and no Hyper-Threading. It runs at 3.4 GHz, with 4MB Smart Cache and 2.5 MB of L2 cache. The processor supports up to DDR4-3200 or DDR5-4800. Intel's UHD Graphics 710 iGPU has up to a 1.3 GHz boost and 16 EUs. This Celeron has a 46W TDP.

As he was testing the base model in the Alder Lake lineup, which cost £55 (MSRP $42 in the US), Steve saw it to match it with the cheapest Intel LGA1700 motherboard he could find, the Gigabyte H610M S2H DDR4 for which he paid £80 (about $110). The RAM quantity wasn’t skimped upon for a budget build, with 16 GB (8 x 2 DIMMs) of DDR4-3000.

He didn’t say anything about the storage, but the GPU in the system for game testing was the Nvidia T1000, a creator-focused card with very similar specifications to the GTX 1650. In a previous video, RGIHD said he picked up this T1000 as it had better availability/pricing than the gamer/consumer-focused GTX 1650. So it might be the same for you, depending on your region.

The system, as mentioned earlier, was tasked with a decent range of popular games, new and old, using sensible quality settings.

The results showed The Witcher 3 at 1080p high with an average of 43 FPS, marred by lengthy intermittent freezes, which couldn’t accurately qualify as stutters – these pauses were far too long (1% low 0 FPS)

In GTA San Andreas at 1080p high, the system achieved an average of 58 FPS, and the glitches and slowdowns were much more bearable (1% low 15 FPS).

For Cyberpunk 2077 playing at 1080p low in the Badlands, we saw an average of 47 FPS 1% low of 11 FPS). However, this game has a bigger problem, as it will not load save games when this lowly 2C/2T processor is your CPU. It is hard to complain about this “bug” as the Celeron is two cores short of the minimum specced CPU.

The YouTuber ran CS:GO at 1080p low settings on the Dust II map. The average in-game performance was 120 FPS (1% low 42 FPS), so performance was much more respectable.

Steve went on to test Fortnite, Forza Horizon 5 (hangs at loading screen), GTA 5, Far Cry 6 (so slow it is unplayable), and Red Dead Redemption 2 ran at an average 34 FPS using 1080p “console settings.”

A common observation throughout the testing was that the Celeron held this system back in gaming. If you flick through the video, you will often see CPU utilization at or near 100% while the GPU isn't really under duress (unless it is in Cyberpunk 2077). The system is not balanced for gaming.

To conclude his exciting insight into the G6900, Steve indicated that there wasn't much point in going for this latest-gen Celeron over any previous-gen entry-level Intel processor that is still available due to the higher motherboard costs at this time. However, on an optimistic note, RGIHD intends to look at the new Pentium Gold G7400, which Steve will be testing with the iGPU as well as an appropriate discrete GPU. The Pentium has several upgrades over the Celeron, including its 2C/4T configuration, a faster clock speed of 3.7 GHz, and double the Smart Cache.

Intel's Celeron and Pentium chips don't always get the love they deserve. However, the Alder Lake Celeron SKUs, specifically the Celeron G6900, could pack a serious punch if the latest Geekbench 5 results (via Benchleaks) are accurate.

The Celeron G6900, which has a $42 MSRP, rocks two Golden Cove cores and lacks Hyper-Threading. In addition, the dual-core processor doesn't even have access to Intel's boost technologies and sticks to a 3.5 GHz base clock. In order words, the Celeron G6900 is at the bottom of the Alder Lake barrel, but the chip does carry Intel's latest Golden Cove cores, so you shouldn't underestimate it.

According to the Geekbench 5 submissions, the Celeron G6900 resided on an ASRock Z690M Phantom Gaming 4 motherboard paired with 16GB of DDR4 memory. Using an ASRock motherboard has its perks, such as access to the company's Base Frequency Boost (BFB) technology that increases the processor's power limit to enable base clock overclocking. The result varies from processor to processor, but the Celeron G6900 benefitted from a 1 GHz increase, allowing it to run at 4.4 GHz during the benchmarks.

The Celeron G6900 scored 1,391 points and 1,408 points on the single-core tests. For comparison, the Core i9-10900K (Comet Lake) averages 1,393 issues in the same test. Therefore, plebeian Celeron G6900 is on the same level as a Core i9-10900K for single-core performance. We're looking at a Golden Cove core at 4.4 GHz matching the performance of a 5.3 GHz Skylake-derived Comet Lake core. Logically, the latter wipes the floor with the Celeron G6900, given its much higher core count.

Compared to AMD chips, the Celeron G6900 was only 5% slower than AMD's Ryzen 5 5600G (Zen 3 at 4.4 GHz) APU in single-core performance, according to Geekbench 5's average processor scores. However, the Celeron G6900 was slightly faster than the tuned-up Ryzen 7 3800XT (Zen 2 at 4.7 GHz).

The Celeron G6900 is part of Intel's recently announced non-K Alder Lake SKUs that address the needs of budget consumers. When in stock, the dual-core Alder Lake chip retails for $59.99.

Qnap has introduced the industry's first 4-bay NAS with two Thunderbolt 4 ports and 10GbE connectivity. The TS-464T4 NAS can house four 20TB 3.5-inch hard drives (80TB in JBOD mode), uses Intel's quad-core Jasper Lake system-on-chip, has two M.2 slots with a PCIe interface for SSDs as well as 2.5GbE and 10GbE connectivity. Perhaps the most critical improvement over TB3-supporting NAS is that this one promises to improve Thunderbolt bandwidth significantly. 

Modern 4-bay NAS devices with RAID 5 and SSD caching support tend to offer relatively high performance with premium drives. Still, when you need to load a large amount of data on them, their network interface becomes a performance bottleneck. Therefore, Thunderbolt 3/4-enabled NAS comes into play with their 40 Gbps interface. The problem is that some TB3-supporting NAS (including Qnap's previous-generation TS-453BT3) utilize outdated SoCs that can provide a TB3 controller with only about 1.6 GB/s of usable bandwidth (PCIe 2.0 offers 5 GT/s per lane and uses an 8b/10b encoding with a 20% overhead). It's far from saturating TB3's peak usable bandwidth of 25.92 GT/s (3.24 GB/s). 

Qnap's TS-464T4 brings numerous significant improvements over its predecessor in pretty much every aspect possible, including enabling higher TB3/TB4 bandwidth thanks to its eight PCIe 3.0 lanes.  

The new unit employs Intel's quad-core Celeron N5105/N5095 'Jasper Lake' SoC that has four Tremont cores at 2.0 – 2.9 GHz with 4MB of L3 and supports 16GB of dual-channel DDR4-2933. Other attributes include a Gen11 GPU, HEVC, H.264, VP9 encoding/decoding for 4Kp60 videos, eight PCIe Gen3 lanes, 2.5GbE MAC, and 10/15W TDP. As a result, it features considerably higher general-purpose performance than Apollo Lake (Goldmont) and Gemini Lake (Goldmont Plus) SoCs used for Qnap's more or less recent 4-bay NAS. That extra performance, along with support for 16GB of DDR4-2933 memory, will likely make performance-demanding workloads like rebuilding a RAID array faster. 

While eight PCIe 3.0 lanes is a significant improvement over six PCIe 2.0 lanes on the TS-453BT3, there are still some caveats. Assuming that Qnap uses four PCIe 3.0 lanes for its TB4 controller (Intel's JHL8440/Goshen Ridge, we presume), there are only four lanes left for the 10GbE controller and two M.2 slots for caching/Qtier SSDs. Probably, one PCIe 3.0 lane (8 GT/s or 0.985 GB/s with a 128b/130b encoding) is enough for the 10GbE controller, and we can speculate that one M.2 slot uses a PCIe 3.0 x2 interface, while another uses a PCIe 3.0 x1 interface. Even TB4 provides 25.92 GT/s, but the actual bandwidth available will be limited to 16 GT/s to the end-user if this drive becomes a medium for writing/reading. 

Like other modern 4-bay NAS from Qnap, this one comes with an HDMI 2.0 output and may be equipped with a remote to work as an HTPC. In addition to two Thunderbolt 3 ports on the front, it also has a USB 3.2 Gen 2 Type-A port near them. 

Unlike some other Qnap NAS, the TS-464T4 no longer has a PCIe slot for expansion cards. It's a logical compromise as there are no spare PCIe lanes left, but this is a good thing since original Qnap accessories (10GbE adapter or M.2 to PCIe card converter) are pretty expensive. Besides, nobody guarantees that third-party hardware will work with Qnap's QTS operating system. 

Speaking of QTS, the manufacturer will ship the TS-464T4 with the latest QTS 5 version that improves performance, enhances security. It also supports Coral Edge TPU that can do image recognition and even attempt to predict hardware failures using cloud-based AI. 

Unfortunately, QNAP has neither announced its official MSRP for the TS-464T4 nor disclosed its launch window.

One of AMD's yet unreleased Athlon Gold 4000G series (Renoir) APUs, the Athlon Gold Pro 4150GE, has appeared at online retailer AliExpress. As Komachi_Ensaka tweeted, the new chip is available for $118.29 -- discounted from $168.99. Unfortunately, the listing shows only one CPU remains in stock, so it wouldn't be surprising to see this listing disappear once someone picks it up.

The seller notes that the CPUs are used and not brand new products. The chips also come from various locations, including China, Malaysia, Vietnam, Costa Rica, and more. If we had to guess from this information, the seller has probably acquired engineering samples of the product.

The seller also released a CPU-Z snapshot, showcasing critical specifications of the CPU. The CPU appears to have a base clock of 3.3 GHz and 4MB of L3 cache, along with four cores and four threads.

This listing is super interesting because AMD has not officially launched any new Athlon Gold CPUs with the 4000G series branding just yet. There have also been no rumors about a new Athlon refresh from AMD, making this new listing very intriguing to see.

The Athlon Gold 3000G lineup currently consists of the quad-core (4 threaded) Athlon Gold 3150 GE, 3150G, and dual-core 3050 GE. In addition, Pro equivalents, such as the Athlon Gold Pro 3150G, Pro 3150GE, and Pro 3125GE, are also available. However, these parts are outdated as they utilize Zen+ cores and launched over a year ago. So it makes sense that AMD would be refreshing the Athlon series under the 4000G series branding to remain competitive in the entry-level segment. In addition, Intel has prepared new Alder Lake-based Pentiums and Celerons that may launch relatively soon.

Since this new Athlon Gold Pro 4150GE uses the 4000G series branding, we suspect that this chip features AMD's more contemporary and much faster Zen 2 cores. Unfortunately, we have no idea or hints of when AMD will announce the Athlon Gold 4000G chips. The only other indication of AMD re-using its 4000G series branding in new CPUs is the rumors revolving around Renior-X, including refreshed versions of the Ryzen 3 3100 and Ryzen 3 3300X under the 4000G series lineup to fend off the forthcoming Alder Lake's Core i3 SKUs. These new Athlon chips could be under that Renoir-X umbrella as well, but it's only a wild guess at this point.

Aaeon has launched one of the industry's first credit-card-sized platforms based on Intel's 11th Gen 'Tiger Lake-UP3' chips. Despite its tiny dimensions, the new NanoCOM-TGU single-board computer (SBC) offers everything that Intel's Tiger Lake can. That includes four high-performance cores and rather decent Iris Xe graphics, which is plenty of horsepower for the embedded computing target market. 

Aaeon's NanoCOM-TGU COM Express Type 10 module comes with up to Intel's Core i7-1185G7E or down to Intel's Celeron 6305E to address both entry-level and premium applications.

The processor is mated with up to 16GB of LPDDR4x-4266 memory and up to a 256GB soldered-down NVMe SSD with a PCIe x1 interface. Depending on the exact SKU, the SBC will come with either a CPU cooler, a heat spreader, or just the board.

• Core i7-1185G7E: 4C/8T, 1.80/4.40GHz, 12MB cache, 96 EU, 15W
• Celeron 6305E: 2C/2T, 1.80GHz, 4MB cache, 48 EU, 15W

The NanoCOM-TGU measures only 3.31×2.17 inches (84×55 mm), yet its expansion and connectivity department is quite decent as it includes (via an internal header, of course) a 2.5 GbE port (Intel I225-LM), two display outputs (one embedded DisplayPort, one DDI), two SATA ports, PCIe expandability, one USB 3.2 Gen 2, one USB 2.0, I2Cm LPC, SMBUS, UART, and high-definition audio support. 

Being aimed at embedded applications like aerospace, industrial, gaming, medical, military, and IoT, the NanoCOM-TGU is equipped with things like a watchdog timer and fTPM.

Aaeon's NanoCOM-TGU single-board computer is not expected to be available at mainstream retail outlets, so don't expect it to compete against the best motherboards for PCs. However, the tiny SBC could theoretically be used to build ultra-compact form-factor (UCFF) PCs, but it would be rather expensive.

Chinese news outlet NetEase snagged one of Intel's new refreshed stock coolers and tested it with a Core i5-12400. The new RM1 is one of three new stock coolers for the Alder Lake generation and will address the mid-range 12th-Gen Core i3, i5, and i7 CPUs, and it proved to be pretty capable when paired with the Core i5-12400.

A few months ago, we saw reports of Intel revamping its stock cooler lineup specifically for Alder Lake. The revamped lineup also includes the RS1 and RH1. The RS1 will come with new Celeron and Pentium processors and have a similar form factor to Intel's previous cooler designed for its 10th Gen processors and older, so it lacks a copper core. The RH1 will be Intel's flagship cooler, so to speak, and it will only ship with 12th Gen Core i9 chips and has a much larger profile than the RM1. The RH1 will also be the most aesthetically pleasing cooler of the three, with RGB lighting effects on the fan.

Allegedly, these coolers will be bundled with Intel's 65W parts and not Intel's high-wattage K SKUs.

The RM1 cooler features an all-black plastic shroud encasing a blacked-out metal heatsink in the middle and surrounded by a blue ring at the top. At its core, a copper slug directly contacts the CPU to enhance the cooler's capabilities over traditional metal designs. However, the copper slug does look noticeably larger than Intel's previous coolers featuring the same copper cores. Presumably, Intel did this to ensure Alder Lake's larger LGA 1700 form factor could be cooled correctly.

In testing with a Core i5-12400, the cooler kept the CPU at a steady 73C with 20C ambient temperatures while running a full AIDA64 stress test, resulting in around 80W of power consumption. According to the outlet, the only downside was the fan's rather high 3100 RPM, which produced a lot of wind noise.

Nonetheless, this is an impressive result for Intel's new stock cooler. Previous versions of Intel's stock cooler, even with copper slugs, could barely keep 65-80W parts from approaching near 100C, or at best, kept the locked chips at 90-95C under ideal conditions at max load.

If these results are indeed true, then the new RM1 cooler is a huge upgrade over Intel's previous cooler design, and you might not have to ditch it immediately in favor of an aftermarket heatsink. In fact, with over 20C of headroom, there would technically be some overclocking headroom if Intel ever unlocked its 65W parts.

Since all three of these coolers will be designed for Intel's 65W locked SKUs, expect these coolers to arrive when Intel officially releases 65W versions of its Alder Lake desktop CPUs next year.

Intel's 12th Gen Alder Lake processors have barreled into the list of best CPUs for gaming, but more SKUs are on the way. In addition, Best Buy (via momomo_us) has revealed the specifications and pricing for the upcoming non-K models that target consumers with stricter budgets.

According to rumors making the rounds in hardware circles, Intel will reportedly announce Alder Lake's non-K chips at CES 2022. Considering the time frame of Best Buy's listings, it's doubtful that they're placeholders. The non-K series may launch at CES 2022. Nonetheless, we recommend you approach the information with caution.

For buyers who don't need the power or unlocked multiplier of the Core i9-12900K ($599), the Core i9-12900 and Core i9-12900F are legit alternatives at $529.99 and $509.99, respectively. You still get all the 8P + 8E configuration benefits, albeit at lower clock speeds.

The Core i7-12700K ($419) is one of the more balanced options in Intel's Alder Lake product stack. Nonetheless, the more budget-friendly and power-efficient options include the Core i7-12700 and Core i7-12700F, which may retail for $359.99 and $329.99, respectively.

Intel Alder Lake Non-K Specifications And Pricing

Currently, we consider the Core i5-12600K ($299) as the best overall processor for gaming. But if your budget doesn't allow it, the Core i5-12600, Core i5-12500, and Core i5-12400 are good 65W options as well. However, there will be a performance gap between the models above and the K-series chip.

The Core i5-12600 could be available for $239.99, whereas the Core i5-12500 may retail for $219.99. Like last generation's Core i5-11400, the Core i5-12400 will probably be the best budget gaming Alder Lake chip. At $209.99, the Core i5-12400 could potentially offer the best bang for your buck. Or, if you don't require integrated graphics, the Core i5-12400F might sell for as low as $179.99. The Core i3-12100, Core i3-12100F, Pentium Gold G7400, and Celeron G6900 will also have their place in the market with price tags below the $140 mark.

The pricing for Alder Lake non-K processors certainly looks very attractive, but their success will depend partially on the price of the upcoming H670, B660, and H610 motherboards as well. In addition, it'll be interesting to see how AMD reacts to the new Alder Lake chips. The chipmaker's Ryzen 5000 (Zen 3) had crushed it before Alder Lake arrived and stole the show, but a small price cut could put Ryzen back in the picture.

The single-board computer (SBC) market is exploding with options for tinkerers looking to power their pet projects and enterprise customers that need pint-sized machines for industrial and robotics applications. There are numerous competitors in this field, including Aaeon, which just announced its Up Square 6000 board (via CNX-Software).

Instead of a power-sipping Arm-based design, the Up Squared 6000 goes with x86-64-based Intel Elkhart Lake SoCs. Aaeon gives customers four processor options to choose from, including the Celeron N6210 (dual-core, 1.2/2.6GHz), Pentium J6426 (quad-core, 1.8/3GHz), Atom x6413E (quad-core, 1.5/3GHz) and Atom x6425RE (quad-core, 1.9GHz). Base boards come with 2GB of onboard LPDDR4 memory and 32GB of eMMC storage, while mid-range configurations double memory to 4GB. The flagship SKU with an Atom x6425RE SoC comes with 8GB of RAM and 64GB of eMMC storage.

When it comes to interfaces, the Up Squared 6000 features HDMI 2.0b, DisplayPort 1.2, two USB 3.1 (Type-A), USB 3.2 (Type-C), M.2 slot (NVMe supported), SATA, 40-pin GPIO connector and a 100-pin board connector. In addition, AAEON has outfitted the board with GbE and 2.5GbE LAN ports for wired networking duties. Wi-Fi, Bluetooth and 5G connectivity are optional via the M.2 slot. The Wi-Fi/Bluetooth kit is currently available to purchase for $29.99.

Aaeon also sells the Up Squared 6000 Edge (which includes a fanless chassis) and the Up Squared 6000 Edge Computing Kit (which features a carrier board and comes preinstalled with Ubuntu 20.04 LTS). The carrier board beefs up the system with the following additional ports: one SATA, one SATA with power connector, one microUSB, one Mini PCIe slot (via USB 2.0), one RS232/422/485, one 40-pin GP I/O and two GbE ports. 

According to Aaeon, the Up Squared 6000 will bow in February 2022, priced from $219 with the Celeron/2GB/32GB configuration. The Up Squared 6000 Edge will arrive the same month with a starting price of $289. The Up Squared 6000 Edge Computing Kit doesn’t currently have a price and will be available later in 2022.

Two new Intel processors will arrive on the market very soon according to Twitter user and known hardware leaker @momomo_us. There's no indication whether these two unreleased chips hail from the chipmaker's latest Alder Lake family or just another Comet Lake refresh. However, the specifications insinuate the former.

The last Pentium Gold or Celeron processor was from Intel's 10th Generation Comet Lake lineup. With Rocket Lake, the chipmaker didn't launch any Core i3 and below SKUs, but preferred to rewarm its Comet Lake chips with a small clock speed bumps. It would seem that Intel may finally give the entry-level processors a much needed renovation.

It seems that the Pentium Gold G7400 and Celeron G6900 are the direct replacements for the Pentium Gold 6400 and Celeron G5900, respectively. As such, these processor should retain the same core configurations as their predecessors. The Pentium Gold G7400 and Celeron G5900 will likely feature a dual-core setup with the first arriving with Hyper-Threading and the latter without.

Core i3 models and some Core i5 variants don't leverage Alder Lake's hybrid microarchitecture, meaning they only sport the Golden Cove (Performance) cores. The Pentium Gold G7400 and Celeron G5900 will follow suit.

Pentium Gold G7400, Celeron G6900 Specifications

*Specifications are unconfirmed.

The clock speeds and L3 cache imply that the Pentium Gold G7400 and Celeron G6900 shouldn't be a Comet Lake refresh. The Pentium Gold G7400 has a 300 MHz lower base clock than the existing Pentium Gold G6400, while the Celeron G6900 preserves the same 3.4 GHz base clock as the Celeron G5900. If the unannounced processors were a refresh, they should be sporting improve clock speeds. 

Another aspect that doesn't add up is amount of L3 cache. The Pentium Gold G6400 and Celeron G5900 have a 4MB and 2MB L3 cache, respectively. The Pentium Gold G7400 and Celeron G6900, on the other hand, utilizes a different cache configuration. Assuming that the Canadian retailer's specifications are spot on, the Pentium Gold G7400 and Celeron G6900 are equipped with 6MB and 4MB of L3 cache, respectively, a considerable increase over the Comet Lake models.

According to the retailer listings, the part numbers for the Pentium Gold G7400 and Celeron G5900 start with "BX80715,", the same string that Intel utilizes for its other Alder Lake parts. The chipmaker uses "BX80708" for Rocket Lake and "BX80701" for Comet Lake.

Intel intelligently launch Alder Lake K-series processors first, while saving the less expensive SKUs for a later stage in time. The chipmaker has confirmed that the non-K chips should arrive early 2022, therefore, the Pentium Gold G6400 and Celeron G5900 are probably patch of the second batch.

A vulnerability has been revealed in Intel’s Goldmont and Goldmont Plus low-power architectures that could potentially reveal low-level security keys, according to security firm Positive Technologies (via by The Register). 

The chips in question are Apollo Lake and Gemini Lake (plus Refresh) Atom, Celeron, and Pentium products. They’re all low-power chips used in embedded systems, mobile devices, and cheap laptops. The Atom E3900 is also found in over 30 cars, including the Tesla Model 3 (if you believe a guy on Twitter).

Positive Technologies responsibly disclosed the flaw to Intel (which has put out an advisory) before going public, and it has been assigned the reference CVE-2021-0146. It requires physical access to the computer and sees the chip tricked into entering a test debugging mode that has excessively high privileges, from which root encryption keys can be extracted. “The bug can also be exploited in targeted attacks across the supply chain,” said Positive’s Mark Ermolov in a statement. “For example, an employee of an Intel processor-based device supplier could, in theory, extract the Intel CSME firmware key and deploy spyware that security software would not detect.”

A UEFI BIOS update can plug the security hole, and owners of affected systems are advised to look out for an update from their device’s manufacturer.