One of the first things that jumps out is the 5-inch full-color LCD screen, which can show hardware information, Asus animations, and your own customized images. The all-black theme is accented with a silver/chrome-like ROG symbol and “Extreme” branding on the large plate heatsink at the bottom. Along the edge, all except two USB-C front panel connections are hidden under a shroud that runs the length of the board. A plethora of buttons and switches along the bottom also lets you know that it’s not your average motherboard.

The Crosshair Extreme offers three PCIe 5.0 M.2 sockets, with the primary one hidden under an oversized heatsink that ensures cooler-running devices below. The DIMM.2 socket appears next to the DRAM slots, adding two more M.2 (PCIe 4.0) slots for a total of five PCIe drives. Looking for Slim SAS for exotic storage? It’s got that. Fast networking, high-end audio, and robust power delivery are just some of what else is onboard.

Do you want to really push the limits? There are buttons and switches designed to navigate the perils of extreme cooling, or you can use Asus’ AI Software suite to get the most out of your system. And as we’d expect from a flagship board, performance was solid throughout all our tests, whether gaming, rendering, encoding, or basic Office functions.

Below, we’ll examine the board's details, and showcase our benchmark results, which earn the board a spot on our Best Motherboards list. But before we get to all that, here is the extensive list of the board’s specifications from Asus’ website.

Specifications of the Asus Crosshair X870E Extreme

Inside the Box of the Asus ROG Crosshair X870E Extreme

With the Extreme, Asus includes a plethora of accessories designed to enhance connectivity and improve the initial build experience. You get all the basics, from SATA cables to Wi-Fi antennas, thermal pads, thermistors, a VRM fan, and, of course, the Q-DIMM.2 add-in card. There’s plenty to get you going. Below is the long list of all the included extras – including the important RoG bottle opener.

• 1-to-3 ARGB splitter cable
• 1-to-2 ARGB splitter cable
• (2) 1-to-4 fan splitter cable
• (2) ROG weave SATA 6G cable packages
• 3-in-1 Thermistor cables pack
• M.2 pad package for ROG Q-DIMM.2
• (2) thermal pad for ROG Q-DIMM.2
• (2) thermal pad for M.2 22110
• DDR5 Fan holder
• ROG Fan Kit
• ASUS WiFi Q-Antenna
• Q-connector
• ROG logo plate sticker
• ROG screwdriver
• ROG stickers
• ROG VIP card
• M.2 Q-Slide package
• M.2 backplate rubber packages
• ROG Bottle Opener
• USB drive with utilities and drivers
• Quick start guide

Design of the Crosshair X870E Extreme

The Crosshair Extreme’s design exudes premium vibes. It sports an all-black look (matte-black PCB) with silver/chrome reflective accents. It’s relatively conservative for a flagship-class board, but still looks great. Standing out from the other boards is the large 5-inch full-color LCD screen integrated into the VRM heatsink. MSI’s X870E Godlike has a smaller 4-inch display, despite costing more. Overall, the Crosshair Extreme’s stealthy black appearance suits the high-end positioning and is sure to complement most build themes.

On the top left, we get a better look at the 5-inch LCD panel, which can slide to the right for increased rear fan compatibility. You can configure the LCD to display system information, use one of Asus’ included backgrounds, or add your own custom JPG. All of this is configurable in Armoury Crate. The VRM heatsink spans across the top, sharing the load with U-shaped heat pipes connecting the two parts. Above the heatsink are two ProCool II-enabled 8-pin EPS power connectors for the processor.

One of the next items we encounter is the first three (of nine) 4-pin fan/pump headers, one of which is specifically for the included, but optional, VRM fan. Per usual, the board controls PWM- and DC-controlled devices through the headers and the Armoury Crate and Fan Expert 4 software. The CPU, Chassis, Rad, and VRM headers output up to 1A/12W, while the pump headers are both good for 3A/36W each. There should be plenty of headers and power available for your cooling needs.

Next to the fan header is the ROG Probelt, with 12 holes in the motherboard for measuring your system voltage using a multimeter. It covers all kinds of voltages, including VCore/Mem/SOC/Misc, and more. These are cool, but novel additions unless you’re an extreme overclocker, where knowing the actual voltage (without software that can be inaccurate) is key.

Below those two features are four NitroPath-enabled DRAM slots with an easily accessible single locking mechanism located at the top. Asus lists support up to DDR5-8200 for desktop processors (9000 series), but even faster for 8000 series APUs (to 9200 MT/s). If you have PMIC-restricted modules (i.e., no profiles), the AEMP feature should be helpful, as will the flexibility provided by DIMM FIT Pro for increased memory optimization. All that said, if you’re breaking records for memory only, the two-slot boards like the Apex (or ITX boards) are where you’ll probably want to look first. Still, you’ll get plenty of memory speed out of the Extreme, and if you need four stick capacity (up to 256GB), you have the option.

Next is the socket for the included Q-DIMM.2 add-in card, which adds two PCIe 4.0 x4-capable M.2 sockets (up to 110mm). I’m a big fan of this solution for additional M.2 sockets, as its larger heatsinks can help manage thermals better than those sharing a plate-style heatsink with the chipset on the board itself

In the upper right corner are a couple of diagnostic tools in the Q-LEDs and Q-CODE LED. The former is the simpler 4-LED system (CPU, DRAM, BOOT, VGA), while the latter shares a bit more detail in the 2-character display. Hidden under a black shroud with labels on top are the start and Flexkey buttons. Below that, sticking out horizontally, is the first 3-pin ARGB header, followed by two 4-pin headers (W_PUMP1 and CHA_FAN2). Below that is the 24-pin ATX header, which supplies power to the board. Sitting vertically are the two front panel USB-C headers (20 and 10 Gbps, respectively).

Power heads from the 8-pin EPS connectors onto the Asus Digi+ EPU (ASP2205) controller. From there, it reaches the 20 vCore stages (and 2x SOC), utilizing 110A Vishay SiC850A SPS MOSFETs. The MISC gets different Infineon PMC41420 80A MOSFETs. The 2,200A is enough to handle the most extreme conditions. Power delivery won’t get in the way of any overclocking, no matter how hard you push the board.

At the bottom of the board, starting from the left and hidden under heatsinks and shrouds, is the audio solution. Here we spy the flagship Realtek ALC4082 codec, some dedicated audio capacitors, and an ESS ES9219 Quad DAC/Amp. It’s arguably the best integrated solution for this generation and chipset.

In the middle of the board are two full-length PCIe slots, each of which connects directly to the CPU. The top slot, using the updated PCIe SLot Q-Release function, supports speeds of up to PCIe 5.0 x16, while the bottom slot has a maximum speed of PCIe 5.0 x8. There is considerable lane sharing between the PCIe slots and M.2 sockets, though.

When M.2_2 is enabled and M.2_3 is disabled, PCIEX16_1 will run in x8 mode, and PCIEX16_2 will be disabled. When M.2_3 is enabled but M.2_2 is not in use, PCIEX16_1 will operate in x8 mode, and PCIEX16_2 will operate in x4 mode. If both M.2_2/3 are enabled, PCIEX16_1 will run at x8 mode and PCIEX16_2 will be disabled. In other words, if you plan on using the second and third PCIe 5.0 M. 2s, you will sacrifice PCIe slot bandwidth. We would have liked to see the M.2 Q-Release feature on this heatsink as well. As it stands, you have to use tools to access M.2_2/3 (and remove the GPU).

Speaking of M.2 sockets, we find three in the middle of the board: one above the top PCIe slot, under the large heatsink, and the others under the large plate heatsink. The topmost slot, M.2_1, connects directly to the CPU and is the only PCIe 5.0 x4 (128 Gbps) socket that will not affect any other components. It supports devices up to 110mm in size (as does M.2_3). M.2_2 holds 80mm modules. Asus lists RAID 0/1/5/10 support with 9000 series processors.

At the right edge, again hidden under shrouds and connected horizontally, is an 8-pin supplemental board power, a 19-pin USB 3.2 Gen 1 (5 Gbps) header, four SATA ports, another 19-pin USB 3.2 Gen 1 connector, and finally, two more 4-pin fan headers (rad fan).

Along the bottom edge is a range of connectivity and functionality, some of which you don’t usually see. From left to right are the front panel audio BCLK +/- buttons, LN2 Mode jumper, 2-pin thermistor header, and two 3-pin ARGB headers. Next to that are several switches helpful for extreme overclocking, including Pause, Slow Mode, a BIOS Switch, and a PCIe Mode switch. The latter two also have LEDs to display the current mode/BIOS. Next to that is a USB 2.0 header, the SlimSAS connector, the CMOS battery, and another USB 2.0 header. Then there’s the V-Latch switch, Retry and Safe Boot buttons, 4-pin water pump header, and finally, the front panel header.

The rear IO features a substantial amount of connectivity, including eight USB Type-A ports (all supporting 10 Gbps) and four USB Type-C ports (two 40 Gbps and two 10 Gbps), which should be enough for most users (especially those with Type-C devices). Among those are the Marvell 10 GbE and Intel 5 GbE ports. On the left are the Clear CMOS and BIOS Flashback buttons. The HDMI port or the USB4 ports handle video output (requires integrated graphics). On the right are the quick-connect Wi-Fi 7 antenna connections. Finally, the audio stack features a mic-in and line-out (3.5mm) ports, as well as an optical SPDIF output.

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

Firmware

Asus’ BIOS on the X870E Extreme resembles any other ROG board, featuring the familiar black and red ROG theme that is easy to read. Asus starts in Easy Mode, which displays high-level information, including CPU and memory clock speeds, temperatures, fan speeds, and storage information. Advanced Mode has several headers across the top that drop down additional options. The new Q-Dashboard shows all the integrated connectivity. When hardware is connected, there’s a green circle next to it. The BIOS is one of my favorites, as any option you need is readily available, and anything you use frequently isn’t buried deep within menus.

Software

Armoury Crate here also follows the ROG-inspired theme. Several applications exist for various functions, ranging from RGB lighting control and audio to system monitoring and overclocking. It's also worth mentioning the included software. When purchasing this Asus motherboard, you get a one-year AIDA64 license - a helpful application for stress and performance testing. Asus’ Driver Hub (get your updated drivers here!), Dolby Atmos (for audio) and a custom version of Hwinfo for real-time monitoring are also helpful applications. We’ve captured a few screenshots of the applications below.

Test System / Comparison Products

We’ve updated our test system to Windows 11 (23H2) 64-bit OS with all updates applied as of late September 2024 (this includes the Branch Prediction Optimizations for AMD). Hardware-wise, we’ve updated the RAM kits (matching our Intel test system), cooling, storage, and video card. Unless otherwise noted, we use the latest non-beta motherboard BIOS available to the public. Thanks to Asus for providing the RTX 4080 TUF graphics card and Crucial for the 2TB T705 SSDs. The hardware we used is as follows:

Test System Components

Benchmark Settings

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

Benchmark Results

Our standard benchmarks and power tests are performed using the CPU’s stock frequencies (including any default boost/turbo) with all power-saving features enabled. We set optimized defaults in the BIOS and the memory by enabling the XMP profile. For this baseline testing, the Windows power scheme is set to Balanced (default) so the PC idles appropriately.

Synthetic Benchmarks

Synthetics provide a great way to determine how a board runs, as identical settings should produce similar performance results. Turbo boost wattage and advanced memory timings are areas where motherboard makers can still optimize for stability or performance, and these settings can impact some testing.

The ROG Crosshair X870E Extreme performed well on average across our synthetic benchmarks. In fact, it ran slightly faster than average in some, including Cinebench and Blender. That said, most of these results are packed tightly together, so the difference isn’t noticeable unless you’re staring at charts like this.

Timed Applications

Timed applications were the same. Handbrake testing was slightly faster than average, while the Corona and LAME testing was on the slower side of results (literally a tenth-second and one-second difference). Nothing to worry about. This is a performant board out of the box with default settings.

3D Games and 3DMark

Starting with the launch of Zen 5, we’ve updated our game tests. We’re keeping the F1 racing game but have upgraded to F1 24. We also dropped Far Cry 6 in favor of a more popular and visually appealing game. We run both games at 1920x1080 resolution using the Ultra preset (details listed above). Cyberpunk 2077 uses DLSS, while we left F1 24 to native resolution scaling. The goal with these settings is to determine if there are differences in performance at the most commonly used (and CPU/system-bound) resolution with settings most people use or strive for (Ultra). We expect the difference between boards in these tests to be minor, with most falling within the margin of error. We’ve also added a minimum FPS value, which can affect your gameplay and immersion experience.

Gaming results were also solid across UL 3DMark and our two games. Overall, this board performs well across a wide variety of functions.

Overclocking

Over the past few CPU generations, overclocking headroom has been shrinking on both sides of the fence while the out-of-the-box potential has increased. For overclockers, this means there’s less fun to have. For the average consumer, you’re getting the most out of the processor without manual tweaking. Today’s motherboards are more robust than ever, and they easily support power-hungry flagship-class processors, so we know the hardware can handle them. There are multiple ways to extract even more performance from these processors: enabling a canned PBO setting, manually tweaking the PBO settings, or just going for an all-core overclock. Results will vary and depend on the cooling as well. In other words, your mileage may vary. Considering all of the above, we’re not overclocking the CPU. However, we will try out our different memory kits to ensure they meet the specifications.

The Crosshair X870E Extreme was one of the only boards to successfully run our stress test using our DDR5-8000 (Klevv) kit. It was right at the listed limits, and in fact, the kit (there’s only one DDR5-8000 kit on the list from G.Skill) wasn’t listed as supported, but here we are!

I consider the Extreme more of a luxury performance board, whereas the Apex focuses more on performance. However, if performance is what you want, you can achieve it on this board in various ways. Do you prefer manual overclocking? There are plenty of tools in the BIOS (including BCLK overclocking), Turbo V, Core Flex, and the Armory Crate on the software side to keep you busy, even with sub-ambient adventures. If you just want the PC to do most of the work, Asus’s Overclocking technologies, including the Dynamic OC Switcher and AI Overclocking, will get the most out of your processor with little effort from the user.

Power Consumption / VRM Temperatures

We used AIDA64’s System Stability Test with Stress CPU, FPU, Cache, and Memory enabled for power testing, using the peak power consumption value from the processor. The wattage reading is from the wall via a Kill-A-Watt meter to capture the entire PC (minus the monitor). The only variable that changes is the motherboard; all other parts remain the same. Please note that we have transitioned to using only the stock power use/VRM temperature charts, as this section aims to ensure the power delivery can handle flagship-class processors.

The power consumption of the Ryzen 9 9990X is relatively low compared to the 7950X used for the X670/X670E platform. In the past, high-end boards peaked at nearly 300W, but current systems now reach a maximum of 250- 270W during CPU stress tests (gaming with the Nvidia RTX 4080 versus the RTX 3070 is a different matter).

That said, the Extreme peaked at 274W under load (the highest we’ve recorded with this CPU/platform), with the CPU consuming around 160W (around 10W more than we typically see). The idle power consumption was also on the high side, settling at 94W. It’s not the mark of efficiency, but that’s to be expected from a board with everything the platform has to offer available.

VRM temperatures peaked at just over 51 degrees Celsius on our sensors. Interestingly, the internal sensor reading was lower, but only by one degree, at 50 degrees Celsius. Using the included fan, temperatures dropped by a couple of degrees in our test-bed configuration. The fan was audible during the stress test, though you do have some control over it through various means. But it easily handled our test CPU and would also handle the likes of the Ryzen 9 9950X.

Bottom Line

The Crosshair X870E Extreme, priced at just under $1000, stands out as a premium, feature-rich motherboard. It offers robust VRMs, 10 GbE, Wi-Fi 7, and five M.2 sockets (three of which are PCIe 5.0). High-end audio with an integrated DAC/Amp is also included, along with a customizable 5-inch full-color LCD. Beyond its impressive hardware, the motherboard delivered strong performance in testing and offers valuable features, including an AI suite and a comprehensive BIOS.

The ROG Crosshair X870E Extreme is a flagship-class E-ATX motherboard from Asus, priced at $999.99. It offers high-end hardware and a premium look, which helps justify its cost. While Gigabyte's Aorus Extreme AI TOP is a less expensive flagship option at $799.99, MSI's X870E Godlike, whose price has come down to $999.99 (down from its release price of $1,099.99). Despite the various pros and cons of these high-end motherboards, Asus' ROG Crosshair X870E Extreme, when priced below the Godlike as it was a week before publishing, stands out as a solid ‘value’ among current flagship AMD motherboards. At the same price, it’s still worth it, but the value proposition then falls squarely on the $800 Gigabyte.

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

The MAG A850GLS emerges as MSI's attempt to democratize premium power delivery aesthetics without the associated premium pricing—a balancing act that's proven historically challenging in the power supply market. Like a well-tailored suit from a department store, it looks the part from a distance but reveals its compromises upon closer inspection. How does it stack against the best power supplies in the market? This 850W unit represents MSI's move towards the newest ATX 3.1 standard while incorporating distinctive design elements that separate it from the sea of generic black boxes. With its dual 12V-2x6 connectors and geometric chassis styling, the A850GLS targets enthusiasts who demand both form and function but remain mindful of their budget constraints.

Specifications and Design

In the Box

The MSI MAG A850GLS arrives in packaging that transmits the company's design philosophy, a sturdy cardboard box adorned with brushed metal aesthetic themes. The robust construction and foam inserts provide adequate protection, though the presentation feels slightly overwrought for a mainstream product.

MSI keeps the bundle straightforward, including only essential mounting hardware and the requisite AC power cable. There are no cable ties or straps included as a bundle but the cables do have wire combs pre-installed.

The cable selection deserves particular praise, featuring all-black construction with individually sleeved wires that MSI terms "embossed jacket cables." The cables are particularly flexible. This premium treatment extends throughout the cable set, creating a uniform appearance, with pre-installed wire combs on the primary ATX, CPU, and PCIe connectors.

External Appearance

MSI has invested significant resources in the MAG A850GLS's visual presentation, clearly aiming for much more than mere utility. The unit measures a reasonable 150mm in depth which slightly exceeds ATX specifications but should remain compatible with virtually all modern cases except the most space-constrained designs.

The matte black chassis finish provides an excellent backdrop for the unit's geometric patterning, featuring complex embossed shapes and etched lines that create visual interest without appearing ostentatious. The custom fan finger guard deserves special mention, incorporating a distinctive hexagonal shape that adds technical flair while maintaining practical airflow characteristics.

The modular connector layout on the rear panel maintains clean organization, with MSI's full series logo etched discreetly above the connection points. At the front, we can only see the typical AC cable receptacle and on/off switch – there is no switch for disabling the unit’s hybrid mode. The hybrid fan mode operates automatically without user override capability.

Internal Design

Cooling responsibilities fall to a ZETA ZFF132512H 135mm fan equipped with fluid dynamic bearings. FDB fans are a technology that represents the sweet spot between cost and performance for mainstream applications. While lacking the extended lifespan characteristics of hydraulic and/or ball bearing systems, fluid dynamic bearings provide excellent noise characteristics and reasonable durability expectations for typical consumer use patterns. The choice of ZETA as fan supplier raises eyebrows, as this manufacturer maintains virtually no online presence and we practically never find their products in premium PC PSUs.

Channel Well Technology (CWT) serves as the OEM partner, bringing considerable expertise from their extensive work with premium brands over the past couple of decades. CWT's reputation is not unfounded, stemming from their seasoned engineering capabilities and proven track record for performance and reliability, with their platforms frequently appearing in mid-to-high tier PC PSUs from respected manufacturers.

The filtering stage implements a competent design featuring six Y capacitors, two X capacitors, and two filtering inductors. Right after the filtering stage, we find two rectifying bridges on a sizable heatsink. The APFC circuit employs two ST Microelectronics 33N60M2 MOSFETs and a diode placed on a simplistic heatsink near the edge of the board. The passive components consist of one encased inductor and a single Teapo 560μF capacitor.

The primary inversion stage utilizes a half-bridge LLC topology built around two ST Microelectronics 33N60M2 MOSFETs, a mature design choice that prioritizes efficiency over cutting-edge performance characteristics. These active components receive their own dedicated heatsink positioned immediately after the APFC stage, though their size appears somewhat basic for the unit's power output requirements.

At the secondary side of the transformer, we find six IPS 014N04SA MOSFETs positioned on the main PCB that handle primary 12V line generation, with small heatsinks providing indirect cooling. DC-to-DC conversion circuits handle the increasingly less critical 3.3V and 5V rails.

Component selection reveals the unit's mainstream positioning most clearly through capacitor choices. The secondary side features a mixed selection of Teapo and ChengX products - manufacturers with proven reliability records but lacking the prestige and extended lifespan expectations of Japanese alternatives like Nippon Chemi-Con or Rubycon. More concerning are several polymer capacitors bearing no manufacturer markings whatsoever and which we could not identify. This component anonymity is not a common compromise even in mainstream PSUs.

Cold Test Results

Cold Test Results (25°C Ambient)

For the testing of PSUs, we are using high precision electronic loads with a maximum power draw of 2700 Watts, a Rigol DS5042M 40 MHz oscilloscope, an Extech 380803 power analyzer, two high precision UNI-T UT-325 digital thermometers, an Extech HD600 SPL meter, a self-designed hotbox and various other bits and parts.

The MSI MAG A850GLS demonstrates impressive efficiency characteristics that significantly exceed its 80Plus Gold certification requirements. At 115 VAC input, the unit achieves an outstanding 90.5% average nominal load efficiency, while 230 VAC operation improves performance to an exceptional 92.4%. These results earned the unit Cybenetics Platinum recognition. Notably, MSI still chose to market this unit as "Gold" certified, obviously because MAG represents their mainstream-level series and the company did not want internal competition with their MPG series. The efficiency curve peaks at approximately 50% load with consistently excellent performance throughout the entire 10-100% nominal load range. The unit maintains good efficiency at very low loads.

Fan behavior proves exemplary during cold testing, with the fan remaining completely stopped until reaching 50% load, after which point it gradually increases speed with low acoustic output maintained across most of the power range. However, the fan control algorithm exhibits an explosive speed increase at maximum load, probably due to a characteristic that prioritizes reliability under such conditions. The thermal performance of the unit proves excellent considering the relatively simplistic heatsink designs throughout the internal architecture.

Hot Test Results

Hot Test Results (~45°C Ambient)

Elevated temperature testing reveals both the strengths and limitations of the MAG A850GLS's thermal design under stressful operating conditions. High ambient temperatures have a reasonable effect on the electrical performance of the unit, with efficiency dropping to 89.7% at 115 VAC and 91.6% at 230 VAC - reductions of approximately 0.8% that represent just the expected performance degradation for a quality PSU with seemingly zero signs of thermal stress.

The fan control algorithm becomes significantly more aggressive under hot conditions, starting operation at 50% load but ramping speed much faster than during cold operation, reaching maximum RPM when the unit operates at 90% capacity. The unit becomes loud when heavily loaded.

Internal temperature management proves adequate under all tested conditions, with component temperatures rising predictably under sustained high-load operation. The internal temperatures exhibit a noticeable increase when operating at maximum load for prolonged periods, as the fan had already reached its maximum speed at 90% load. The relatively simple heatsink design reaches its thermal limits under extreme conditions, though all temperatures remain well within safe operational levels throughout testing.

PSU Quality and Bottom Line

Power Supply Quality

The MSI MAG A850GLS delivers good electrical performance and power output quality overall that meets modern system requirements without achieving standout excellence. Voltage ripple filtering could be better but proves adequate for a unit of this class, with maximum measurements of 58 mV on the 12V rail, 28 mV on the 5V rail, and 28 mV on the 3.3V rail. While these figures exceed the performance of premium competitors, they remain well within ATX specifications and pose no practical limitations for system operation. Voltage regulation demonstrates exceptional performance across all rails, achieving remarkable 0.8% regulation on the 12V rail, 0.7% on the 5V rail, and 0.8% on the 3.3V rail - performance that proves impressive even on the secondary rails that are often neglected in modern designs.

During our thorough assessment, we evaluate the essential protection features of every power supply unit we review, including Over Current Protection (OCP), Over Voltage Protection (OVP), Over Power Protection (OPP), and Short Circuit Protection (SCP). All protection mechanisms were activated and functioned correctly during testing.

The protection features work well during comprehensive testing, though some calibration decisions are questionable. Over Current Protection appears too lax on the minor rails, even for an ATX 3.1 unit, triggering at 142% for the 3.3V rail and 144% for the 5V rail. However, OCP protection on the 12V rail proves perfect at 120%, while Over Power Protection demonstrates ideal calibration at 122% under hot conditions, showing appropriate engineering focus on the critical primary power delivery circuit.

Bottom Line

The MSI MAG A850GLS represents a thoughtful attempt to bring premium power supply aesthetics and features to mainstream pricing - a challenging proposition that yields mixed results. The partnership with Channel Well Technology has produced a mature, well-engineered platform that demonstrates solid electrical performance and reasonable build quality. The premium cable treatment and distinctive chassis styling provide genuine value for system builders who prioritize aesthetics. The comprehensive 7-year warranty demonstrates MSI's confidence in their engineering decisions and provides reasonable protection for the investment.

The unit's dual 12V-2x6 connector configuration represents forward-thinking design that anticipates future graphics card requirements, though the limited number of traditional 8-pin PCIe connectors may constrain compatibility with current high-end multi-GPU configurations. The excellent efficiency characteristics and solid voltage regulation create a foundation for reliable system operation across a wide range of loading conditions. The unit's ability to maintain tight voltage regulation, combined with adequate ripple suppression and well-calibrated protection circuits on the critical 12V rail, provide the electrical stability necessary for sensitive modern components including high-end CPUs and graphics cards.

However, the component selection reveals the compromises inherent in mainstream positioning. The mid-tier capacitor choices and anonymous polymer components raise questions about long-term reliability expectations, particularly when compared to units utilizing Japanese capacitors throughout. Thermal performance is good but better heatsinks would improve it dramatically and, in turn, acoustics under heavy loads would also improve significantly.

At the current $150 retail pricing, which is the unit’s MSRP price, the MAG A850GLS faces formidable competition that would be extremely difficult to overcome. Products rarely stay at MSRP prices for long however, thus we expect the retail price to drop significantly in the months to come. For builders who value distinctive aesthetics, the unit provides competent power delivery wrapped in an attractive package. While it may not achieve the engineering excellence of premium competitors, it delivers reliable power with style, which is a combination that may prove compelling for budget-conscious enthusiasts willing to accept reasonable compromises if there are modest price cuts.

MORE: Best Power Supplies

MORE: How We Test Power Supplies

MORE: All Power Supply Content

Our review unit came equipped with a 1TB SSD and 32GB of RAM, but those are both pricey upgrade options for what is the entry point into the MacBook Pro family. With an as-tested price of $2,349, it’s an expensive purchase for a MacBook without a “Pro” or “Max” in its CPU name.

Design of the Apple MacBook Pro (M5, 14-inch)

If it ain’t broke, don’t fix it. That seems to be Apple’s mantra regarding the current design of the MacBook Pro. Apple has been using this same chassis for multiple generations, save for some changes to the available colors (like Space Black replacing Space Gray).

The aluminum chassis is still attractive. Our review unit came cloaked in Space Black, which is a fingerprint magnet. The chassis is solid, with none of the flex that you’ll find in systems that use plastic or a combination of plastic and aluminum. The one niggle that I have with the design is the cutout in the keyboard deck that allows your finger to slide in and lift the lid. The corners on either side of the cutout are sharp and can poke your fingers/hand if you’re not careful.

With that said, you get quite a few ports on this laptop, including three Thunderbolt 4 ports, a full-size SDXC, HDMI 2.1, 3.5 mm audio jack and a MagSafe 3 connector. I’ve always been a fan of MagSafe and was happy to see it return when the current chassis design debuted back with the M1 Pro and Max chips.

Many people complain about the notch at the top-center of the display, which houses the Center Stage camera. It’s somewhat of an eyesore, but I’ve become used to it over time. The way it’s positioned, it sits in the space that is usually occupied by the Menu bar for apps, so I don’t feel like I’m being robbed of any additional screen real estate. Do I wish that Apple found a better way to integrate the webcam? Yes. Does the current iteration really encroach on the daily usability of the laptop for me? No.

The other item of note with our MacBook Pro review unit is that it has the nano-texture display option, which costs an additional $150. The screen coating reduces glare and helps to minimize reflections, which can be beneficial outdoors, or in indoor areas with bright light sources.

Our 14-inch MacBook review unit measures 12.31 x 8.71 x 0.61 inches and weighs 3.4 pounds, which is identical to last year’s model with the M4.

Apple MacBook Pro Specifications (M5, 14-inch)

Productivity and Graphics Performance on the Apple MacBook Pro (M5, 14-inch)

Our review unit is equipped with Apple’s newly released M5 processor, which is also found (for now) in the iPad Pro and Vision Pro. When used in the MacBook Pro, the M5 features 10 CPU cores (4 performance cores, 6 efficiency cores),10 GPU cores, and a 16-core NPU.

The new MacBook Pro had no trouble outdistancing the other assembled competition — 14-inch MacBook Pro (M4), Lenovo Yoga 9i 2-in-1 Aura Edition (Core Ultra 7 258V), Dell 16 Premium (Core Ultra 7 255H) — in every performance benchmark.

Apple consistently delivers anywhere from 10- to 20-percent improvements over the prior generation when it comes to CPU performance, and the M5 is no exception. Using the Geekbench 6 synthetic CPU benchmark, our system achieved a single-core score of 4,288, roughly a 10% boost over the preceding MacBook Pro with the M4 chip. The multi-core score saw an uplifting to 17,926 versus 15,114 on the M4 (just over an 18% uplift).

However, the CPU isn’t the only area that Apple decided to address when it comes to hardware upgrades. When Apple announced the M5 MacBook Pro, it boasted a 2x improvement in SSD speeds over the previous generation. In our file transfer test, which copies 25GB of data, we saw 1,917.35 MBps compared to 1,167.29 MBps for the M4 MacBook Pro (also with a 1TB SSD). It wasn’t quite a 2x improvement, but is still a remarkable year-over-year leap.

We also saw a significant improvement in our Handbrake test, where we transcode a 4K video file to 1080p. Last year’s 14-inch MacBook Pro (M4) completed the job in four minutes and 57 seconds. However, the new M5-equipped model finished nearly a minute and a half quicker, taking just three minutes and 31 seconds.

We use Cinebench 2024 to stress test laptops, and the benchmark estimates a single-core clock of 4.6 GHz and a multi-core clock of 3.5 GHz. During our 10-run test, the system started at 1,042 and then settled in the 1,100 to 1,128 range for the duration.

The M5 includes a 10-core GPU, which provides up a 4x improvement in peak GPU compute compared to the M4 thanks to the inclusion of a dedicated Neural Accelerator baked into each core. Overall graphicps performance is said to increase by up to 30 percent over the GPU in the M4.

In the real world, we saw 57 frames per second (FPS) at 1200p and 28 FPS at native resolution in Shadow of the Tomb Raider. For comparison last year’s M4-equipped system recorded 36 FPS and 18 FPS, respectively. In addition, our review unit managed 44 fps at 1200p and 31 fps at native resolution in Cyberpunk 2077 using the “For this Mac” setting.

Display on the Apple MacBook Pro (M5, 14-inch)

Apple’s MacBook Pro family uses an IPS display with Mini LED backlighting. This is a noticeable upgrade from the entry-level MacBook Air, with its more traditional LED-backlit IPS display. The “Liquid Retina” display has a 3,024 x 1,964 resolution and a “ProMotion” refresh rate that maxes out at 120 Hz.

Although this Mini LED can’t quite touch the black level performance of competing OLED panels, it still a great looking display. The nano-texture display option, as configured on our review unit, is a great addition. My personal laptop is an older 14-inch MacBook Pro (M3) with the glossy screen finish. It’s incredible to see the two side by side to highlight the difference in reflection reduction.

I’m not a musical kind of guy by any means, but I went to the theater last year with my wife and thoroughly enjoyed Wicked. Naturally, I’m excited to see the sequel, so I watched the final trailer for Wicked: For Good. Pinks and green are prominent in the trailer and looked wonderful on the display, and the Mini LED panel performed well in the more dimly-lit scenes featuring the Wicked Witch and her flying monkeys.

Among the assembled group, our 14-inch MacBook Pro proved to be the brightest, as measured by our light meter. It pegged 558 nits, just barely edging out last year’s model (556.6 nits). This performance placed it well ahead of the Yoga 9i’s OLED panel (421 nits) and the Dell 16 Premium (367 nits).

Color performance was virtually unchanged compared to M4 MacBook Pro, which shows that Apple is likely using the same panels year-over-year. We measured 81.7 percent coverage of the DCI-P3 color gamut and 115.4 percent of the sRGB color space. However, the Yoga 9i had more saturated colors, thanks to its OLED panel.

Keyboard and Touchpad on the Apple MacBook Pro (M5, 14-inch)

One of the great things about Apple being conservative with redesigns for the MacBook Pro is that the excellent keyboard remains intact. Among laptop keyboards, Apple and Lenovo remain my far and away favorites, and the scissor switches on the MacBook Pro are a delight to use. They’re clicky with just enough travel to satisfy my fingers, and are far superior to problematic butterfly switches that Apple thankfully banished.

Apple employs single-zone white LED backlighting, which turns on automatically in low-light conditions. If you want to manually control the keyboard backlighting, you can do so by navigating to System Settings --> Keyboard.

Using the keyhero.com test, I leisurely strolled to 91 words per minute with 95 percent accuracy, which was in the same ballpark as my performance last year with the 16-inch MacBook Pro (M4 Pro).

Another feature that I like on MacBooks is the large haptic trackpad, which simulates a physical “click.” The benefit of this design over traditional top-hinged trackpads that you find in Windows laptops is that the surface will register a click no matter where you press. The glass surface also is smooth to the touch, allowing my finger to easily glide when navigating through macOS Tahoe.

Audio on the Apple MacBook Pro (M5, 14-inch)

The sound system on our review unit hasn’t changed compared to last year, which is a good thing. The six-speaker setup is one of the best you’ll find on any laptop regardless of price, and includes support for Dolby Digital, Dolby Atmos, Dolby Digital Plus and Apple Spatial Audio.

It’s not often that I can comment on noticeable bass effects in a laptop sound system, but the is most definitely the case with the 14-inch MacBook Pro. I played Rainey Haynes’ “I’m Old Enough to Rock and Roll,” which is a jamming track from one of my favorite 80’s movies: Iron Eagle. The steady drum line, synths and Tina Turner-esque vocals soar on the sound system. Even with the volume dialed up to near max levels, I could barely discern distortion.

Upgradeability of the Apple MacBook Pro (M5, 14-inch)

Upgradeability and MacBooks don't go together. While you can remove four pentalobe screws at the bottom of the chassis to remove the cover, it won't do you any good when it comes other upgrading system components. The M5 SoC is a system-on-a-chip design with our review unit's 32GB included on-die. The 1TB SSD is soldered onto the motherboard, so you're out of luck there as well.

In other words, you'll want to ensure that you have select enough RAM and storage that you'll need for the life of the machine at the time of purchase. Unfortunately, the upgrades are pricey (far more than you would expect in the Windows PC realm), with an upgrade from 16GB to 32GB of RAM costing $400 and boosting the SSD from the base 512GB to 1TB ringing in at $200.

Battery Life on the Apple MacBook Pro (M5, 14-inch)

Battery life has consistently been a strong point for the MacBook Pro in the Apple Silicon era, and the M5 continues that tradition.

Our 14-inch MacBook Pro unit is equipped with a 72.4 Whr battery, enough to deliver 18 hours and 14 minutes of runtime using our battery test, which involves browsing the web, running OpenGL tests, and streaming video at 150 nits of brightness. However, this result was 20 minutes less than last year’s MacBook Pro with the M4.

Despite the 20-minute shortfall compared to last year, the result easily bested the Dell 16 Premium by 2.5x (7:15) despite its larger 99 WHr battery. The Yoga 9i 2-in-1 Aura Edition was also no match, with a runtime of 12 hours and 47 minutes.

Heat on the Apple MacBook Pro (M5, 14-inch)

We measured skin temperatures during our Cinebench 2024 stress test on the 14-inch MacBook Pro. As soon as the test began, the system fans spun into high gear to keep the system temperatures in check.

The hottest point measured was near the left exhaust vent at 108 degrees Fahrenheit, while we saw 104 F between the G and H keys. The touchpad was much cooler at 82.8 F. The system overall felt warm, but it was tolerable to sit on my lap while wearing shorts.

During the Cinebench 2024 stress test, we measured temperatures using the TG Pro app. The overall system temperature, which takes an average of several components (i.e., power supply, battery, SSD, logic board, etc.) was 73 degrees Celsius (163 degrees Fahrenheit). The M5 itself registered 87 degrees C (188 F).

Webcam on the Apple MacBook Pro (M5, 14-inch)

Like last year’s mode, the new M5 uses a 12MP Center Stage camera. Image quality looked impeccable, with good detail in my face (wrinkles, grey hairs in my beard, and everything else), and excellent color representation. I used the camera while sitting on a couch with a brightly lit window behind me, and the picture still looked great without blowing out details or causing halo effects.

The Center Stage feature keeps your face centered in the frame during calls, should you move around in your seat. It’s a neat feature, but I prefer to leave the feature off — sometimes I just want to briefly duck out of frame to sneeze, and not everybody on a call needs to see that.

Software on the Apple MacBook Pro (M5, 14-inch)

The big change with macOS 26 Tahoe is the addition of the “Liquid Glass” makeover, a design language that it shares with iOS 26 and iPadOS 26. This an overabundant use of translucent menus and icons, which reflect displayed content in real time. Liquid Glass has been controversial since it was first announced at WWDC, and Apple tweaked the effect during the beta period to appease the naysayers. Now that Liquid Glass is in its final form, I’m mostly OK with how it looks, mainly because its reach in macOS Tahoe isn’t as broad as it is in iOS or iPadOS.

Besides the fresh coat of paint for the UI, macOS Tahoe brings features like Live Translation for the Messages app and FaceTime, Live Activities for Mac, a refresh Phone app, Call Screening, more intelligent search Spotlight, and a universal clipboard (with history).

Apple Intelligence was a prominent new feature Apple announced last year for macOS Sequoia, but most of the headlining features involving Siri never materialized. Apple took a lot of flak for overpromising and underdelivering, and expect the previously announced Siri features to debut sometime next year. With that said, Apple stayed relatively mum about Apple Intelligence updates for macOS Tahoe.

With that said, macOS continues to be a well-rounded operating system with a slew of Apple-provided apps out of the box. One of my favorites is iPhone Mirroring, which allows you full access to your iPhone’s screen from your macOS Tahoe desktop. It especially comes in handy when I need to perform two-factor authentication verification with my phone, and my phone is in another room or I’m too lazy to dig it out of my pocket.

Apple MacBook Pro (M5, 14-inch) Configurations and Warranty

Apple offers three preconfigured options for the M5 MacBook Pro, all of which feature the M5 chip with a 10-core CPU and a 10-core GPU. The base system features 16GB of unified memory and a 512GB SSD for $1,599. The middle tier includes a 1TB SSD for an additional $200 ($1,799). The top configuration gives you 24GB of unified memory and a 1TB SSD for $1,999.

However, Apple also offers built-to-order configuration that allow you to further tweak the system to your specific needs. Our review unit came with 32GB of unified memory, a 1TB SSD, and the nano-texture display option, which brought the as-tested price to $2,349. That's quite pricey for an entry-level MacBook Pro with Apple's base-level chip.

If you want an M5 MacBook Pro with all the trimmings, opting for 32GB of RAM, a 4TB SSD, the nano-texture display option and the 96W USB-C power adapter balloons the price to $3,369.

Apple's MacBook Pro comes with a one-year warranty. You can opt for Apple's extended warranty, Apple Care Plus, which costs $99.99/year or $9.99/month. The coverage adds unlimited repairs covering accidents, 24/7 priority tech support, and battery replacement.

Bottom Line

Not only did Apple increase CPU performance for the 14-inch MacBook Pro this year with the brawny new M5, but it also delivered a healthy uplift in SSD performance. Despite the performance increases, battery life remains within the margin of error with last year’s model at over 18 hours.

The performance is wrapped in a high-quality chassis with plenty of ports for pro users, although the Space Black finish remains susceptible to fingerprints. Our usual complaints about Apple laptops remain, including the high price of SSD and memory upgrades.

I’ll add one more: the lack of Wi-Fi 7. I find it odd that a laptop starting at $1,599 doesn’t include Wi-Fi 7 as standard equipment, and this is made even more puzzling considering that Apple’s excellent N1 Wi-Fi chip is available already in the iPhone 17/Air/Pro and the new iPad Pro (M5). Both the Dell and Lenovo assembled here include Wi-Fi 7. While we reviewed the Dell 16 Premium and used it as a competitor here, the Dell 14 Premium, which aligns more closely with the 14-inch MacBook Pro in terms of size, includes Wi-Fi 7 standard in its base $1,249 configuration.

Speaking of the Dell 14 Premium, it is a worthy alternative to the 14-inch MacBook Pro. You can spec one out with a Core Ultra 7 255H, 32GB LPDDR5X, a 1TB SSD and a 14.5-inch 3.2K touch OLED for $1,949, undercutting Apple’s offering, as-tested, by $400.

If you already have a 14-inch MacBook Pro with an M4 chip, I would recommend you sit this one out. However, if you’re coming from a much older M1 or M2 version, the performance gains are hard to ignore.

With that said, at its base price of $1,599 with 16GB of RAM and a 512GB SSD, the 14-inch MacBook Pro is a strong productivity workhorse. But once pricing starts cresting the $2,000 mark with hardware upgrades, the value proposition starts to wane.

Last year, we tested Hyte’s THICC Q60 – a “240mm” AIO liquid cooler which features an insane 84mm thickness (with fans installed), which raised the bar for what we consider excellence, providing the best noise-normalized results I tested, when paired with Intel’s i7-13700K.

Today we’ll be looking at Hyte’s latest X50 PC case, which features a unique look with its all-rounded design, top PSU placement, and a variety of colors to choose from: Purple, Pink, Green, Red, Black, or White. We tested the pink variant, but you can see the other color options in our X50 announcement coverage at Computex 2025.

Will Hyte’s X50 make our list of the best PC cases? Let’s take a look at the specifications and features of the case, then we’ll wrap up our review with thermal benchmarks, so you can decide if this is the case for your next build.

Product Specifications

Features of Hyte’s X50 PC Case

▶️ Rounded design

Everything about the X50, from the sides, the glass cover, interior parts, including the PSU cage – even the bottom feet – is rounded. This, combined with the unique colors available, gives the case a softer vibe than the vast majority of case options.

▶️ Top PSU placement

The Hyte X50 is reminiscent of older computers, with its PSU placement on the top rear of the computer case. This means that the PSU’s intake is pulling warmed air from inside the case. That might be a problem if thermals are an issue, but we’ll get to that later.

▶️ Tool-less access

All of the case’s sides can be accessed and secured without the use of tools. Hyte uses rivets to attach the parts, which means they can be removed by simply sliding the part in question. In the photo below, the connecting slots are highlighted in blue.

▶️ Build quality

The build quality of Hyte’s X50 is insane, one of the sturdiest and most well-built products I’ve had the pleasure of testing. There are very few competitors that are built this strong.

The company touts its 4mm thick laminated acoustic glass for noise dampening, which includes a shatter-resistant bonding layer, 1mm thick steel construction, and automotive-grade tooling, which the company says requires 4x tighter tolerances than the industry standard.

The latter is likely at least partially a design necessity since so much of the case is curved, but it’s hard to overemphasize just how solid and premium this case looks and feels.

▶️ Rear view, Storage support, cable management

From the back, we can see the side intake support bracket on the left and the PSU cage at the top. Cable management velcro straps are positioned behind the PSU and along the middle of the unit, and there are five slots for passing cables between the front and rear of the case.

The storage bay sits at the bottom right of the case, and supports two 2.5-inch drives and one 3.5-inch drive.

The case’s design incorporates a rear mesh cover to improve airflow. The lower-priced X50 Air version of this case (which costs $20 less) uses this same type of mesh filter for both sides of the chassis.

▶️ Side view, back-connect support, GPU holder

Looking inside, the thing that stands out to me the most is the exhaust. which uses louvered ventilation blades. Hyte claims this design reduces exhaust impedance while reinforcing chassis rigidity.

▶️ IO panel, Top View

The IO panel is located at the top front of the case. It features a combined audio/mic jack, one USB-C port supporting up to 20 Gbps transfer speeds, and two USB-A ports supporting 5 Gbps. The power button features white LED lighting, and is somewhat unique in that it incorporates a tactile mechanical switch, like a mechanical keybaord.

▶️ Dust filters, Bottom View

The case incorporates an easily removable dust filter at the bottom, and has four rounded feet for support.

▶️Front view

The front of the case showcases a giant, rounded mesh cover.

▶️ Rear view

There’s not much to say about the rear of the case. The main feature that stands out is the gold-colored Hyte logo at the top left corner, and the louvered ventilation blades, which we mentioned earlier.

Thermal tests, test setup, and testing methodology

The measurements I’ve benchmarked this case against focus on the efficiency of the case’s thermal transfer.

Our thermal tests are presented to give you more information about the product’s performance, but aren’t intended as the sole judgment of the chassis. The style, price, features, and noise levels of a case should also be considered, and we all have different preferences. What I might like in a case, you might not, and that’s OK. My goal with these reviews is to give everyone, no matter their preferences, enough information to decide whether or not a product is right for them.

Knowing I was going to be reviewing a pink PC case, I reached out to TeamGroup, who was kind enough to send its T-Force Xtreeem DDR5 7200mhz RAM for use in this review – it is one of the only kits of RAM available in a pastel pink color, which matches the color of the X50 we’re using for today’s review. If you’re interested in this RAM, it can be purchased from Amazon for $215 USD.

There are also a limited number of pink CPU coolers that you can choose to match this case’s aesthetic. I would recommend ID-Cooling’s FX360 INF Pink which we previously reviewed, available for $99 USD on Amazon – or $85 for the 240mm version shown in the picture below. Just remember you’ll have to install an AIO in the front or front side, as the rounded top and PSU placement don’t leave room for cooling there.

With most cases I’ve tested that do not come with pre-installed fans, I’ve used a standardized set of Ocypus fans for testing. However, given the unique aesthetic of the X50, Hyte expects that most buyers will want to pair this case with a color-matching set of FA12 fans. As such, we’ve used these fans for our tests today. The colored variants (blue, pink, red, or green) of these fans cost $49.99 ($12.50/fan), but traditional black or white versions can be obtained for only $39.99 USD ($10/fan).

The pricing of these fans is reasonable, but it would be nice if the company included at least one color-matching fan for exhaust. Especially considering the case supports up to 10 fans.

Maximum thermal performance (fans noise-normalized to 38.9 dBA)

This set of tests is designed to appeal to those who love quietly running PCs, with the combined Hyte FA12 system fans and the CPU fan noise normalized to 38.9 dBA. Consider this a measure of the case’s thermal efficiency when noise levels are set to run quietly. This first set of benchmarks examines CPU performance.

Hyte did well with this CPU-only stress test, with the third-best results I’ve recorded thus far.

The second noise-normalized test I perform is much like the one above, but also fully stresses the RTX 4070 Ti Super GPU I have paired with the system – adding another 295W of heat to the case. The GPU temperatures in this test were the second-best I’ve seen amongst the cases I’ve tested this way, with an average temperature of 63.7 degrees Celsius.

CPU performance in this combined CPU+GPU stress test remained very good, also taking the second-place spot, dropping by only 5W, with 274W dissipated from the CPU cooler!

Maximum Noise Levels

I measured the noise levels of Hyte’s X50 case paired with four FA12 fans (3 intake, 1 exhaust) with my sound meter from 1 meter away, recording a measurement of 44.2 dBA. While this isn’t a low volume level per se, it is on the quieter end in comparison to competing PC cases.

Maximum thermal performance – FA12 fans at full speed

Our noise-normalized results are designed for folks who prefer silence, but what if you don’t mind additional noise and just care about maximum performance? This set of tests covers thermal performance with the Hyte FA12 fans running at full speed.

Starting with a CPU-only stress test, we measured 296W dissipated from the CPU cooler. While that’s a fifth-place result, it’s not far from our best result, which is only 3.2W better.

Adding the heat of a GPU brings the position of Hyte’s X50 closer to the front of the pack in terms of performance, achieving third place, with 288W cooled.

GPU temperatures were about middle of the road – averaging 64.1 degrees C during this test.

Conclusion

Hyte’s X50 sets itself apart with aesthetics – featuring a rounded design a number of striking colors: purple, pink (as tested), green, red, black, or white. The build quality of the X50 is in a class of its own, with higher-quality construction than the vast majority of competitors.

If you’re interested in Hyte’s X50 case, it can be purchased for $159.99 in the United states, or $129.99 for the X50 Air version, which swaps the front glass for a mesh filter. Just note that no fans are included. If you want to add color-matched air movers, Hyte will be happy to sell you a four-pack for $49.99, or $10 less if you opt for white or black fans.

For those looking for a less aggressive case design, or just something more colorful than most builds, the X50 is easy to recommend. It’s not the best performer out there, but our testing didn’t throw up any red flags when it comes to cooling. And if temps are a concern, adding more than the four fans we used for testing (the case supports 10) would no-doubt help on that front.

With a case this design-forward and colorful, some will also lament the lack of RGB here. For that, assuming your components don’t deliver enough lighting, Hyte also sells a selection of magnetic RGB strips, starting at $39.99.

Although the new memory kit retains what Crucial refers to as an origami-inspired heat spreader, the company has made subtle modifications to the design. Instead of a uniform solid white or black color, Crucial has introduced a camouflage pattern with a highly gamer-oriented aesthetic. The aluminum alloy heat spreader remains available in either black or white finishes, although Crucial employs fancier names such as "Stealth Matte Black" and "Snow Fox White," respectively.

Like other memory kits from the Pro Overclocking lineup, there is minimal marketing on the memory modules. The Micron logo remains visible; however, this time, the Crucial logo is presented as a product of CNC machining with a diamond-like finish to enhance light reflection. The Pro Overclocking memory modules have a height of 1.38 inches (35 mm), fitting into the low-profile category.

The memory kit consists of two 16GB memory modules. Each module features a single-rank design with eight integrated circuits (ICs), arranged on a single side. This represents the first occasion that we're seeing these Micron 5QH75 D8HKZ ICs, which, according to the SPD dump, appear to be Micron's new H-die ICs. Given that the memory module is 16GB, each IC has a capacity of 2GB. Meanwhile, the 12=BL 1G1 power management IC (PMIC) indicates to us that it's from Richtek.

As with any overclocked memory kit, the memory modules default to JEDEC's specifications. In this case, it's DDR5-5600 with timings at 46-45-45-90. Both Intel XMP 3.0 and AMD EXPO support are present in Crucial's Pro Overclocking memory modules. The primary profile is configured for DDR5-6400 with timings specified as 32-40-40-103 at 1.35V. Conversely, the secondary profile is designated for DDR5-6000 with timings of 36-38-38-80 and a DRAM voltage of 1.35V. See our PC Memory 101 feature and How to Shop for RAM story for more timings and frequency considerations.

Comparison Hardware

The Intel system features the Core Ultra 9 285K processor and the MSI MEG Z890 Unify-X motherboard with firmware version 7E20v1A60. Conversely, the AMD system utilizes the Ryzen 9 9900X processor and the MSI MPG X870E Carbon WiFi motherboard with firmware version 7E49v1A64. The Corsair iCUE Link Titan 360 RX LCD CPU liquid cooling solution efficiently sustains optimal temperature levels for both the Arrow Lake and Zen 5 processors.

The MSI GeForce RTX 4080 16GB Gaming X Trio addresses the more graphics-intensive workloads, ensuring that our gaming RAM benchmarks do not have a graphics bottleneck. TeamGroup's A440 Lite PCIe 4.0 SSD strikes a balance between performance and storage capacity, offering 2TB of high-speed storage that reaches speeds of up to 7,400 MB/s, ideal for Windows 11 24H2 installations, benchmarking software, and gaming.

Meanwhile, the Corsair RM1000x Shift ATX 3.0 power supply provides a dependable and sufficient power source for our testing systems, directly supplying the GeForce RTX 4080 with a native 16-pin (12VHPWR) power cable. Ultimately, the Streacom BC1 open-air test bench is capable of housing all of our hardware.

Intel Performance

Crucial's latest memory kit demonstrated commendable speed on the Intel platform. It faced no significant issues securing a top position, only trailing behind the Dominator Platinum RGB DDR5-6400 C38 in application performance. Conversely, the Pro Overclocking's gaming performance was comparatively modest and ranked at the lower end of the competitive spectrum.

AMD Performance

Memory kits do not consistently demonstrate identical performance across various platforms; therefore, testing is conducted on both Intel and AMD motherboards. The Pro Overclocking DDR5-6400 C32 memory kit did not perform notably well on the AMD platform and was generally slightly slower than the existing Pro Overclocking DDR5-6400 C38 memory kit.

Overclocking and Latency Tuning

Our best overclocking result on the Pro Overclocking DDR5-6400 C32 was DDR5-6600, maintaining identical memory timings while increasing the DRAM voltage from 1.35V to 1.45V. The result was marginally lower than the DDR5-6400 C38, which was anticipated, given that we were able to preserve the CAS Latency (CL) of 32 clock cycles as opposed to 38 clock cycles on the latter.

Lowest Stable Timings

The factory timings for the DDR5 6400 memory kit were already approaching the stability threshold. Efforts to reduce various memory timings were unsuccessful. Additionally, increasing the voltage to 1.45V, which is 7% above the default setting, did not allow us to lower any of the memory timings.

Bottom Line

The Pro Overclocking DDR5-6400 C32 exhibited a varied array of results. In comparison to the DDR5-6400 C38 variant, Crucial refined the CL timings from 38 to 32 clock cycles; however, this adjustment was accompanied by an increase in tRAS from 84 to 102 clock cycles. The trade-off balances out the performance, which explains why the C32 version is not necessarily consistently faster than the C38 version on every platform. With DDR5 memory, all timings are important, and you should not be fixated solely on the CL timing.

Crucial markets the Pro Overclocking DDR5-6400 C32 memory kit with two distinct pricing options based on the selected color trim. The Stealth Matte Black variant is priced at $174.99, whereas the Snow Fox White variant incurs a $12 premium, resulting in a retail price of $186.99. For comparison, the Pro Overclocking DDR5-6400 C38 memory kit currently sells for $129.98. Regardless of the choice, Crucial provides a limited lifetime warranty on all Pro Overlocking memory kits.

OLED is a relative newcomer to the game, but out of the gate, it’s proven itself to be superior in every way. Okay, they’re more expensive, but can you honestly name one thing that LCD has over the best OLED gaming monitors? If your answer is speed, this review will challenge that. The Asus ROG Strix XG27AQDPG is the first 500 Hz OLED I’ve tested, and it doesn’t sacrifice size or resolution. It’s a 27-inch QHD 2560x1440 panel with HDR10, Quantum Dot wide gamut color, and Adaptive-Sync. Let’s take a look.

Asus ROG Strix XG27AQDPG Specs

LCD has a price advantage but the XG27AQDPG isn’t ridiculously expensive. Think Lexus rather than Rolls Royce. It comes out of the gate for $900 as I write this. For that sum you get everything and the kitchen sink. There are no corners cut here. The 500 Hz is native, not overclocked. There’s plenty of cooling to keep the panel safe with a large heatsink and passive airflow, no fans required. Panel care includes pixel cleaning, an orbiter and multiple protection options to dim persistent taskbars and logos. A user proximity sensor will blank the screen when you leave your desk.

Gaming benefits not only from the high refresh rate but also from Adaptive-Sync and ELMB blur reduction. Asus also includes OLED anti-flicker to prevent artifacts during rapid changes in refresh rate. The moving image is incredibly smooth, with perfect motion resolution and no hint of stutter, tearing, or blur.

The image is stunning, of course, thanks to OLED’s infinite black levels. The XG27AQDPG is a bit brighter than its competitors, with over 500 nits peak available for SDR and HDR. You also get a Uniform Brightness option with its own independent brightness setting and peaks of around 320 nits. This means that there are no changing levels between productivity and gaming.

You also get Quantum Dot color and its associated extra-large gamut. The XG27AQDPG covers over 108% of DCI-P3 and includes an sRGB mode for those who want it. Accuracy comes out of the box, and you get Asus’ usual comprehensive suite of calibration options. These can be accessed in the OSD or from the Windows Desktop using the DisplayWidget Center app. And there’s support for HDR10 with the aforementioned 500-nit peaks and multiple picture modes that are adjustable for brightness and contrast.

Gaming aids include an extensive array of aiming points, sniper mode, timer, stopwatch, frame counter, and display alignment marks. You can line up multiple XG27AQDPGs thanks to the thin bezel around the image. Additionally, this display features USB ports, a headphone jack, and a high-quality stand to complete the premium package.

Assembly and Accessories

The XG27AQDPG arrives in the box version of a plain brown wrapper. The carton is nondescript, but inside, there is a full set of accoutrements. A heavy stamped steel bracket provides a 100mm VESA mount. And a zippered pouch opens to reveal ROG decals, HDMI, USB and DisplayPort cables and two IEC power cords. The stand, base and panel assemble without the need for tools.

Product 360

Asus’ ROG Strix monitors are a bit more simply styled than their Swift counterparts, but are no less capable of high performance. The XG27AQDPG is all business in the front with a thin bezel around a shiny screen. Light rejection is only fair, so one should avoid pointing it towards a sunny window or a harsh light source. A small protrusion at the bottom center displays the ROG logo with an LED backlight. More LEDs shine around the back with a large ROG logo made from a Lite Brite array of lights.

The panel sits on a large component bulge, resulting in a total depth of approximately 2.6 inches. A small round attachment point interfaces with the premium stand that has a 4.3-inch height adjustment along with 45-degree swivel, 5/25-degree tilt, and a 90-degree portrait mode. Movements are solid and firm with no apparent play or wobble. On top of the stand is a quarter-inch threaded socket where you can bolt on a webcam or desk light.

The input complement includes a DisplayPort 1.4 with Display Stream Compression, two HDMI 2.1s, USB 3.2 ports, one up and two downstream, and a 3.5mm headphone jack. Also visible in the last photo above is the OSD joystick, which features an inspirational message: “For Those Who Dare.”

OSD Features

The XG27AQDPG’s OSD has the same layout and design as all the latest ROG monitors, with eight submenus and a complete set of gaming and calibration options. Press the joystick to summon it.

Gaming is up first with video processing at the top of the list. You can toggle between VRR (Adaptive-Sync) and ELMB (Extra Low Motion Blur), which uses black frame insertion to reduce blur. When running over 200 fps, ELMB is completely unnecessary. When you get into the 400 to 500fps range, the moving image becomes almost eerily smooth.

Here too are the GameVisual picture modes, nine in total. The default is Racing, and it doesn’t need calibration. However, you can adjust it with the available single-point white balance, gamma presets, and six-axis saturation sliders. You can also choose the color gamut, sRGB, DCI-P3 or wide gamut, which is the XG27AQDPG’s full color mode that covers over 108% of DCI-P3. Among the picture modes is one called sRGB Cal. It is very accurate, but here’s a hint: stick with Racing mode and simply change the gamut in the color space menu. Then, you retain full control over calibration.

The Image menu has luminance controls for both SDR and HDR. HDR Setting offers four modes plus a switch to make any of them adjustable. You can tweak brightness and contrast and engage uniform brightness, but the color menu remains grayed out. My favorite HDR mode is DisplayHDR 500 True Black because it has the most accurate luminance tracking. All of them use the XG27AQDPG’s large color gamut to full advantage.

Asus provides a huge selection of panel care options. There are three dimming toggles for the full screen or just the perimeter of the image. You can also dim logos that appear in the corners and the status bars that persist in many games. Pixel cleaning can be run periodically to refresh the panel. And you can turn on an orbiter that shifts the pixels from time to time. A proximity sensor can be turned on to dim or darken the screen when you aren’t sitting in front of it.

Aura RGB is the XG27AQDPG’s lighting feature, which can run multiple effects in all colors. You can also sync the LEDs with onscreen movement once a USB connection is made and you install Asus’ DisplayWidget Center app. The joystick directionals are programmable by the user for many possible functions, and you can save settings to two memory slots for later recall.

One important option to visit when setting up your new XG27AQDPG is the Power Setting. By default, it’s set to Power Saving Mode, which limits brightness and grays out many image controls. If you don’t mind drawing around 40 watts at 200 nits brightness, choose Performance Mode.

Asus ROG Strix XG27AQDPG Calibration Settings

The XG27AQDPG doesn’t need calibration in its Racing picture mode (GameVisual), but there is a small improvement available with a few changes to the RGB sliders. Stick with the wide gamut option in the Color Space menu to see full color in both SDR and HDR. If you want the smaller sRGB gamut for SDR, choose it in this menu rather than using the sRGB Cal mode, which has fewer picture options. For HDR signals, I recommend True Black for the best picture. It has nearly spot-on luminance tracking and excellent color accuracy. Below are my SDR settings with brightness values for both settings of the Uniform Brightness toggle.

 Gaming and Hands-on

The XG27AQDPG’s specs already had me predisposed to liking it, but just a few minutes of gameplay made it official. Starting up Doom Eternal, I was immediately enamored with the image, which was extremely colorful and tactile. Though I’ve been spoiled by 4K screens aplenty, there is no sacrifice made when you choose QHD. And you get higher frame rates, which not only ensure smooth motion in all instances, but also the lowest possible input lag. The fps counter hovered between 300 and 450 when driven by a GeForce RTX 4090. And it was working too, as the hot air blowing past my leg could attest. You’ll want as much graphics card as you can afford to get the most from this monitor.

Input lag simply did not exist here as all my movements were instantly translated into action. Quick turns and run-and-gun maneuvers are so easy with a monitor this fast. Slowing down to enjoy the scenery had an equally impressive effect thanks to the superb color. It’s nice to enjoy a super quick monitor that doesn’t sacrifice resolution or color volume like the speedy LCDs I’ve seen.

I always enjoy Asus’ ROG styling and the XG27AQDPG exemplifies the current aesthetic. The screen’s bezel disappears when you play or work, and I appreciated the quality stand with its generous ergonomics. The LED lighting is all in the back except for a ROG logo up front that is always lit, even in standby mode. Isolating the panel from its internal components is an effective way to keep the OLED cool, and I appreciated the large array of care options like pixel orbiting and logo dimming. Asus backs the XG27AQDPG with a three-year warranty, and I don’t expect many will need it.

Workday tasks were dispatched with ease and comfort. The 27-inch flat panel has solid pixel density (109ppi), which is enhanced by that deep OLED contrast. Infinite black levels are a real thing, even when grinding away in a spreadsheet or word processor. It makes text easier to read and small icons clearer.

Graphics tasks in Photoshop were made easier by the XG27AQDPG’s superb color. It’s accurate enough to skip the calibration if you want, but a few tweaks make a small but noticeable difference. Switching between the wide gamut and sRGB is just a few clicks of the OSD joystick away, very handy.

Takeaway: The XG27AQDPG is an ideal monitor for work, games or entertainment. The only thing missing are internal speakers but the 3.5mm headphone jack takes care of that. The picture is stunning and so is the gaming feel. Motion is perfectly smooth with no blur or artifacts of any kind. Control response is as good as it gets, which makes average players like me seem more skilled. This is a premium screen, but you won’t regret buying the best.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Undoubtedly, many of you have skipped to this point, so let’s talk about why we’ve done that. 500 Hz is fast, but does it make a difference? To compare the XG27AQDPG, I’ve rounded up Asus’ PG27AQDP, LG’s 27GX790A, Samsung’s OLED G6, and Philips’ 27M2N8500. Here we go.

Pixel Response and Input Lag

Click here to read up on our pixel response and input lag testing procedures.

The PG27AQDP is nearly as quick on paper as the XG27AQDPG at 480 versus 500 Hz. In practice, there’s no visual or measurable difference between them. OK, if I could measure less than one millisecond, there might be, but I can’t. 2ms is super quick and it means perfect motion resolution. But so does the 4ms posted by the Alienware. If you line it up with the top two, you’ll be hard-pressed to spot a difference. When I say perfect motion resolution, I mean perfect.

However, there might be a difference in input lag. Typically, though not always, higher Hertz means lower lag. This is one of the “not always” examples. The LG and Philips are still the fastest monitors on the planet with just 10ms of total lag. The two Asus screens are next at 15 and 16ms. Yes, 500 Hz seems to be 1ms slower than 480 in this case. And 280 and 360 Hz are just one or two clicks slower. The word slower doesn’t make sense here though because these numbers are incredibly low. Unless you’re Obi-Wan Kenobi, you won’t find a disadvantage in gameplay at 16ms versus 10ms.

Test Takeaway: Yes, the XG27AQDPG is fast and incredibly smooth. But you can get lower input lag from the 360 Hz Philips. And it isn’t visibly smoother than the 280 Hz Alienware. But it doesn’t cost any more than them either and you get the status of owning one of the first 500 Hz OLED gaming monitors. Seriously, the XG27AQDPG is really awesome with incredible smoothness and instant control response. No one will be sad to play on it.

Viewing Angles

I’ve noted slight color shifts with other Quantum Dot OLEDs in the past but the XG27AQDPG seems to avoid that descriptor. At 45 degrees to the sides, there is no change in color, brightness or gamma. This is one of OLED’s big advantages. Even the best IPS screens can’t do what you see above. From the vertical plane, there is a change in gamma and a slight red shift, but no change in brightness. This is excellent performance.

Screen Uniformity

To learn how we measure screen uniformity, click here.

This XG27AQDPG sample recorded one of the lowest uniformity scores in my database of 445 computer monitors. It truly doesn’t get better than 1.55%. There are no visible hotspots, nor is there bleed or glow in any full screen test pattern. I looked at every gray and color brightness level and could see nothing but perfect uniformity from edge to edge. Excellent.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

To read about our monitor tests in-depth, please check out Display Testing Explained: How We Test PC Monitors. We cover brightness and contrast testing on page two.

Uncalibrated – Maximum Backlight Level

The XG27AQDPG is rated for DisplayHDR 500, but it can also achieve this value in SDR mode if you turn Uniform Brightness off and measure a 25% window pattern. Smaller windows would be brighter yet. A full field hits around 320 nits with UB off or on. UB is great for workday tasks, while turning it off makes video and gaming more impactful. I’m glad to see the option because it isn’t there on every OLED. As always, black levels and contrast are immeasurable regardless of the luminance settings.

After Calibration to 200 nits

Calibration makes no difference other than to change the peak brightness value. I set it to 200 nits in both UB on and UB off modes. The XG27AQDPG stores independent settings for both, which is a nice touch. The ANSI test yields the same results as the static with unmeasurable black levels and contrast. This is OLED’s typically excellent contrast performance.

Test Takeaway: The XG27AQDPG is 25% brighter than its competitors, which equates to around 100 nits if Uniform Brightness is turned off. When it’s on, you get 320 nits, which is also about 100 nits more than most OLEDs I’ve reviewed. This gives it an advantage in bright rooms. In fact, it’s bright enough for outdoor environments such as location production tents, AKA video villages. Otherwise, it delivers the same amazing contrast and black levels as other OLEDs.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Like all Asus gaming monitors, the XG27AQDPG ships in its Racing mode, which is very close to the mark. There is no need for calibration, but all the means necessary to do it are provided.

Grayscale and Gamma Tracking

Our grayscale and gamma tests use Calman calibration software from Portrait Displays. We describe our grayscale and gamma tests in detail here.

Default grayscale tracking is a tad red, but the errors don’t crack the 3dE line, the visible threshold, until 100% brightness, and then only just. This is an excellent first run with gamma that’s close to perfect. Calibration tightens it up nicely with all errors now below 1dE, except for 100% which is just over 2dE. Gamma is unchanged.

The XG27AQDPG has an sRGB Cal picture mode, but I preferred to leave it in Racing and change the gamut in the Color Space menu. This gives you the same grayscale tracking as the default and you can calibrate for a small improvement if you wish. Gamma runs a tad light above 60% brightness, but that is a barely noticeable error. This is excellent performance.

Comparisons

OLED monitors remain in a premium category, so they perform well out of the box with just a few exceptions. I recommend calibrating the bottom three screens, but the top three can be enjoyed without tweaking. The XG27AQDPG’s 2.12dE score is very good.

Calibration levels the playing field to where all the monitors become identical in appearance. You won’t be able to discern the LG from the Alienware in a direct comparison, and the XG27AQDPG won’t stand out either. This is excellent performance.

Gamma tracking is a tad better than average with a tight 0.11 range of values and just 0.9% deviation from 2.2. The actual value is 2.18. Accurate gamma tracking is very important and the XG27AQDPG achieves it.

Color Gamut Accuracy

Our color gamut and volume testing use Portrait Displays’ Calman software. For details on our color gamut testing and volume calculations, click here.

Color saturation is what you buy a Quantum Dot monitor for and the XG27AQDPG has plenty. It completely fills the green and blue primaries and exceeds the red targets by a small amount. The progression is linear, which maintains good balance. Red is not overemphasized in practice. The magenta hue error goes away with calibration, taking the error even further below the visible point. This is a seriously colorful and color accurate monitor.

The sRGB mode shows similar tracking with slight red oversaturation and a magenta hue error. If you stick with Racing and change the gamut selection in the Color Space menu, you can tweak this to perfection. Even without adjustment, the average 1.68dE error is low enough to qualify the XG27AQDPG as a professional display.

Comparisons

OLEDs are colorful and accurate, whether you have Quantum Dots or not. The XG27AQDPG sits near the top of this group with a 1.38dE gamut score. It doesn’t get much better than that. There’s plenty of color volume too, though it’s a tad behind the top three. It’d hard to see a 4% difference, but if you look very closely, you might see just a tad more saturation from the Samsung, Alienware and Philips. The XG27AQDPG is well ahead of its non-QD competition, and its nearly 100% coverage of sRGB is ideal.

Test Takeaway: The XG27AQDPG has excellent accuracy and performance with visually perfect color out of the box and a slight improvement available from calibration. It has a huge gamut volume, which makes it great for gaming and entertainment. No one will think a monitor is too colorful. It is well suited for work, play and professional applications in equal measures.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Our HDR benchmarking uses Portrait Displays’ Calman software. To learn about our HDR testing, see our breakdown of how we test PC monitors.

HDR doesn’t get better than a premium OLED like the XG27AQDPG. The tech’s deep blacks and broad contrast create an image that other screens types can’t. This monitor switches modes automatically and offers four presets and some luminance adjustment options that its competitors don’t have.

HDR Brightness and Contrast

The XG27AQDPG is brighter than most other OLEDs at 527 nits peak from a 25% window pattern. Most of the competition are closer to the bottom three screens at 420-450 nits but the LG and Asus at the top are brighter yet. In most content, these differences are minute with just the smallest highlights showing up a little more. But make no mistake, the XG27AQDPG looks spectacular. Remember also that the PG27AQDP and LG 27GX790A have less color volume.

Grayscale, EOTF and Color

There are no color adjustments available in the XG27AQDPG’s four HDR modes, but the red error seen above is barely above the visible threshold. I compared all four presets and found True Black had the best EOTF tracking. This is key to ensuring full detail rendering and the most tactile image. I noted that Gaming HDR had some edge enhancement, which ultimately made the image less sharp.

In the HDR color test you can see general and linear oversaturation in red, magenta and blue. These colors are vivid in all content and there is no clipping of detail. Green is fully covered which sets the XG27AQDPG apart from non-QD monitors that can’t render all of that primary. You can see the same performance in the BT.2020 test where it tops out at 90% red, 75% green and 95% blue. This is excellent performance.

Test Takeaway: The XG27AQDPG is a top example of the HDR format with brightly vivid colors and accurate luminance tracking. Its peaks of over 527 nits are an advantage over much of the competition as well. If you want one of the best available HDR monitors, this one is it.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Finding faults with OLED gaming monitors is a difficult task. They operate in such a narrow performance window that it’s tough to find one that stands apart from its competitors. Even speed increases are hard to quantify because, in practice, anything over 200fps is perfectly smooth with little to no input lag. Is 240 Hz enough, or do you need more?

There will always be users who want more speed, and they are the ones who should take a hard look at the Asus ROG Strix XG27AQDPG. At 500 Hz, it’s the fastest OLED, on paper, currently available. My tests showed that it is indeed quickest in panel response, but not in input lag. That said, there is no perceptible difference between 15ms and 16ms as I noted in the comparison charts. The XG27AQDPG is super quick and super smooth. No one will complain about its gaming performance; it is among the very best.

I was also impressed by its colorful and contrasty image, but that’s true of any OLED. There’s the narrow performance window I’m talking about. The differences in color accuracy between these displays is tiny. The XG27AQDPG has pro-level color like most of its competitors.

Ultimately, the decision to buy it is an easy one. It’s super fast, super colorful, very accurate and gorgeous to look at. Build quality is premium and it’s not too dear at $900. If you want the fastest refresh rate currently available from an OLED gaming monitor, definitely check it out.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

What makes the Hyper 212 3DHP so important? It's in the name of the cooler – the company’s new 3DHP heatpipe technology. Traditional copper heatpipes are typically formed in a “U” shape, and work well enough in most scenarios. But Cooler Master’s 3DHP heatpipes are formed in a trident-like shape, with three ends instead of two.

Will this new cooler make our list of best CPU coolers on the market? Well, let’s put it this way: It features genuine innovation – a rarity in the rather stale air cooling market, and features a nearly silent fan. Let’s go through the features and benchmarks of this cooler, and you can decide for yourself if it is worth sinking $29.99 of your cooling budget into.

Cooler specifications

Packing and included contents

The cooler arrives in a simple brown box, with a purple label showing the design of the product and its specifications. Inside are two smaller boxes: One contains the heatsink and fan, while the other includes the mounting accessories.

Included with the product are the following:

• 1x 120mm fan
• Single tower heatsink with two 3DHP copper heatpipes
• CryoFuze thermal paste
• Mounting for AMD and Intel platforms
• Installation manual

Features of Cooler Master’s Hyper 212 3DHP

▶️ Innovative 3DHP copper heatpipes

This isn’t just another Hyper 212 rebrand – the Hyper 212 3DHP is the first cooler on the market with Cooler Master’s new heatpipe technology. Most traditional heatpipes are formed in a “U” shape, but the 3DHP heatpipes are trident-shaped. As you’ll see in our benchmarks, this groundbreaking and effective design allows for just two of these new heatpipes to outperform four traditional direct-touch copper heatpipes!

Part of the reason these heatpipes perform better is that they more effectively remove heat from the center of the CPU. As an example, Cooler Master has kindly provided this thermal image showing how the hotspot of the CPU is reduced with this new heatpipe design.

▶️ Single tower aluminum heatsink

The Hyper 212 is Cooler Master’s entry-level air-cooling line, and as such, it features a single-tower aluminum heatsink, shown in the photo above.

▶️ Copper CPU plate

The base of the CPU plate is pure copper.

▶️ Full RAM compatibility

As the picture above demonstrates, the Hyper 212 3DHP does not interfere with or overhang DIMM slots in any manner, ensuring compatibility with all sizes of RAM, no matter how tall.

▶️ Low noise 120mm fan

The fan included will appeal to enthusiasts who prefer a PC with the lowest volume levels possible, with our noise measurements reaching just 38.2 dBA. The version reviewed features ARGB support, but there is a solid black version of this cooler without ARGB, if you don’t want lights in your rig.

AM5 and 1851 installation

The installation of the cooler is fairly simple for AMD systems – but it’s somewhat of a pain on Intel systems because the backplate of the cooler must be held in place during the install process.

1. You’ll first need to apply the included backplate if you’re using an Intel CPU. AMD users will remove the default mounting mechanism.
2. Next, you’ll need to install the mounting bars onto the cooler.

3. Apply thermal paste to the CPU. If you’re unsure of the best way to do this, you can follow our thermal paste application guide.
4. Place the heatsink on top of the CPU, and secure it with the included screws.
5. Attach the fan with the included fan clips, and then connect the PWM/ARGB headers to your motherboard – installation is now complete!

Real-world testing configuration –AMD AM5 platform

My results may differ from others because I am looking for results that are comparable to real-world use. This means that I test CPU coolers inside a closed desktop case, which increases cooling difficulty compared to other testing methods.

Many will test CPU coolers outside of a case, on an open test bench. Open benches have lowered ambient temperatures, which in turn makes weak coolers appear stronger than they really are. Some publications have used generic thermal plates to test cooling solutions. I reject both of these methods because they don’t accurately reflect the real-world conditions where a CPU cooler is used.

CPU-only thermal results with AMD Ryzen 9 9950X

Today’s review is going to be a little different than our recent ones, partly because our main testing 9950X3D system is encountering problems at the moment, but also because we wanted to have this review focus on the technological improvement that comes with the 3DHP heatpipes.

We’re using AMD’s Ryzen 9 9950Xm and tested the Cooler Master Hyper 212 3DHP (Single tower, 2x 3DHP heatpipes) against Montech’s NX400 (Single tower, 4x traditional heatpipes) and a prototype dual-tower heatsink with six heatpipes and two fans.

100W thermal results

We’ll start today’s review on the low end, focusing on a 100W workload.

For today’s tests, I’ve compared the Hyper 212 3DHP against Montech’s NX400 and a dual-tower air cooler. Montech’s NX400 is one of the best single-tower air coolers on the market, utilizing four copper heatpipes and a high-power E28 fan. Even with those advantages, the Hyper 212 3DHP outperforms the NX400 by 1.2 degrees Celsius!

It is also worth noting that the Hyper 212 3DHP ran virtually silently in this scenario, with noise levels lower than I could measure, whereas the NX400 was still audible.

200W

For this next test, we’re testing the CPU at its maximum stock power consumption – 200W, twice the thermal load of our previous test.

Here again we see the Hyper 212 3DHP slightly outperforming Montech’s NX400, by 0.8 degrees C.

150W, with fan normalized testing

After completing the first two tests, I started to wonder if the fan on the Hyper 212 3DHP might be holding back the potential of the 3DHP heatsink. After all, it runs very quietly – no more than 38.2 dBA.

To test this theory, I switched out the default fan included by Cooler Master with the newly released high-power E28 fan that comes with Montech’s NX400 – and boy, the results were interesting!

The Hyper 212 3DHP with its default fan continued to outperform the NX400 by a small amount, but when paired with a stronger fan it had significantly higher performance – trailing a dual-tower air cooler by only 0.6 degrees C!

The noise measurements here are intended to show you just how quietly the Hyper 212 3DHP runs with the default fan. In this scenario, it was only 37.3 dBA. In all lower power scenarios, the fan runs quieter than I can measure.

PBO Performance with fan normalized testing

Without power limits enforced, the hottest CPUs on the market – like the AMD Ryzen 9 9950X tested today - will hit their peak temperature (TJ Max) and thermally throttle with even the strongest of air coolers and even most liquid coolers on the market in intensive stress testing.

When the CPU reaches its peak temperature, I’ve measured the CPU package power to determine the maximum wattage cooled to best compare performance. It’s important to note that thermal performance can scale differently depending on the CPU it’s being tested with.

Like our previous test, the results are most interesting with Cooler Master’s 3DHP when it is paired with a high-power fan. In this configuration, it trails a prototype dual-tower cooler by only 13W! This is extremely impressive to see two heatpipes competing with a dual-tower air cooler that has six heatpipes!

If you prefer to use the stock fan, overall cooling performance will only be slightly better than Montech’s NX-400 – but its noise levels will be low, as detailed in the next section.

Maximum noise levels

In terms of noise levels, Cooler Master’s Hyper 212 3DHP is one of the quietest on the market – reaching only 38.2 dBA according to our measurements! This basically makes it the go-to cooler for those who want an entry-level heatsink with low noise levels.

Conclusion

Cooler Master’s Hyper 212 3DHP is an exciting addition to the air cooler market. Its two 3DHP heatpipes are genuinely innovative, improving cooling performance and efficiency. It’s also reasonably priced, at $29.99 in the U.S.

Don’t expect this cooler to surpass the best big air coolers, like those from Thermalright. But keep in mind that’s not what Cooler Master is trying to do here. The company will also release higher-end air coolers with its 3DHP tech, which we look forward to testing soon. If those larger coolers can scale up the 212 3DHP’s capabilities, while keeping costs and fan noise in check, Cooler Master may soon be making a run for a few spots on our Best CPU Coolers list. In the meantime, if you’re looking for a quiet cooler with impressive mainstream performance, the Hyper 212 3DHP is easy to recommend.

The Iskur V2 X sacrifices many of the extras that make the Iskur V2 so special, but its more accessible price point makes it a good fit for those who prioritize value when searching for the best gaming chairs.

Razer Iskur V2 X Specifications

Assembling the Razer Iskur V2

Like the Iskur V2, the Iskur V2 X review unit arrived in a massive black box weighing 41 pounds. With that said, I managed to carry the cumbersome box, intact, up to the second floor of my house and to my office (my back be damned). With the back opened, I found the following parts carefully protected and arranged in the box:

• Fabric gloves for assembly (glued to the inner flap)
• Backrest and seat base (left and right arms pre-attached)
• Metal, 5-arm base
• Casters 5x
• Gas lift, tilt mechanism
• Allen wrench, screws, and plastic covers for left/right side arms

Each primary component for the chair was wrapped in a clear plastic bag to protect the metal finish or fabric during shipping. The assembly instructions are printed on the back of a large piece of green cardstock. The instructions are easy to understand with pictograms to guide you through the building process. I started by affixing the castors to the base, and then banging each into place on the floor (which startled my Mini Schnauzer puppy). The next step was to insert the gas lift cylinder into the base.

I then had to remove four screws from the seat base, using the included Allen wrench tool, attach the tilt mechanism, and then reinsert the previously removed screws. I then slipped the seatback onto the seat base using the metal guides and screwed each side with two screws. The final steps were to attach the plastic covers (to hide the attachment points for the seatback to the seat base). There are also two small round covers to conceal the single screw that secures the larger plastic covers in place.

Razer Iskur V2 X Design

The design of the Iskur V2 X closely resembles that of the Iskur V2. There is significant bolstering on the chair's backrest and seat base, which helps to envelop your posterior and back. My review unit featured a light grey fabric with black contrast stitching (also available in black). It's a more subdued look than the Iskur V2 that I reviewed earlier this year, which was finished in black with green contrast stitching.

The cloth seating surfaces of the chair were pleasing to the touch and comfortable during extended use. One of my niggles with the Iskur V2 was that I found the synthetic leather to be uncomfortable, sticking to my skin in warmer temperatures or when my body temperature increased from prolonged gaming sessions. The cloth on the Iskur V2 breathes better, and I had no issues with the chair sticking to my legs or making my back sweat.

If there's a downside to fabric, it's that it can become a magnet for odors and stains — gaming chairs finished in synthetic leather are easier to clean. With a fabric chair, you'll need to be a little more careful about where you place your greasy hands if you're grabbing for potato chips, Cheetos, or any other kind of food that you may partake in while gaming (or just doing your usual daily workload).

Since this is the cost-optimized version of the Iskur V2, several features have been omitted to deliver a chair that's less than half the Iskur V2's MSRP. The comfy and removable memory foam pillow isn't included (although you can purchase one separately from Razer). You also don't get the two knobs to adjust the back lumbar support. One other change comes with the armrests: the Iskur V2 features 4D armrests, while the Iskur V2X has 2D armrests (adjustable for height and rotatable inward/outward).

One thing that hasn't changed is the ability to adjust the backrest from 90 degrees to a lounge-friendly 152 degrees. While I can recline and use the chair at a 152-degree angle, my anxiety won't allow me to maintain that position for more than 10 or 15 seconds at a time (I have the illogical fear that I'm going to topple backwards and crack my cranium).

Razer Iskur V2 X Comfort and Adjustments

Razer tackles comfort head-on by using a high-density foam for the cushion. That's just a fancy way of saying that the cushion is relatively thick, it molds to your buttocks, and helps to relieve pressure points. During my time testing the chair, my bottom was well-supported, and I didn't experience any fatigue or pain after using it for hours.

Unlike the Iskur V2, the Iskur V2 X doesn't have the adjustable lumbar support for the backrest. On the Iskur V2, you can move the lumbar support that swivels 360 degrees and can move up/down, or in/out. As a result, you might have a hard time finding (or may never find) the optimal "sweet spot" with the chair. It wasn't a problem for me, as my 5-foot-10-inch and 161-pound frame had no trouble adjusting to the backrest, which was set at an angle of about 100 to 110 degrees using the handle on the right side of the seat base. As I mentioned earlier, the chair has a maximum recline of 152 degrees, which is something I was reluctant to test for longer than a few minutes.

You also lose the 4D armrests of the Iskur V2, with the Iskur V2 X featuring 2D armrests instead. You can adjust each armrest independently for height and swivel inwards or outwards. I adjusted the height of the armrests to accommodate my seating position comfortably, but I left them in their original position — pointing straight ahead. (I've never really felt the need to swivel my armrests inward or outward, but to each their own.)

The seat features a height adjustment mechanism, accessible via a handle located on the right side of the seat base. My most comfortable position was to have the seat raised to its maximum setting.

The five castors aren’t PU-coated, like the ones on the Iskur V2, but I still had no trouble effortlessly gliding across the laminate wood flooring in my office.

Bottom Line

The Razer Iskur V2 X is a no-frills gaming chair that I still find to be comfortable to use daily. For gaming to more mundane tasks — such as typing up this review or entering data in Microsoft Excel — I rolled around my office and swiveled from left to right with ease, and even thrust my back into the seat like a maniac when I lost a race in Forza Horizon 5.

You do miss out on several features from the Iskur V2, including the memory foam headrest, 4D armrests, and 6D lumbar support. But if those aren’t features that tickle your fancy for a gaming chair, you can save well over $300 by opting for the Iskur V2 X. The Iskur V2 has an MSRP of $649 versus just $299 for the Iskur V2 X, which makes the latter a prime score for bargain shoppers. However, if you want a bit more adjustability to fit the specific needs of your body, you may also want to consider the Bouiles Master Chair, which offers more adjustability and costs just $70 more than the Iskur V2 X.

The DeathAdder V4 Pro is identical on the outside, but it's been redesigned on the inside: It has a brand-new sensor, new optical switches and scroll wheel, and it's more efficient (it weighs less and offers more battery life). It also comes with a wireless 8K polling dongle in the box — unlike the V3 Pro, which comes with a standard dongle. The 8K dongle is sold separately on the older model. The DeathAdder V4 Pro features Razer's Focus Pro 45K Optical Sensor Gen-2, which has a maximum sensitivity of 45,000 DPI and a maximum speed of 900 IPS, and can handle up to 85 G's of force — it's Razer's latest, fastest, and most accurate sensor yet.

All of this doesn't come cheap, of course, and the DeathAdder V4 Pro will set you back $170, which is on the high end, even for a premium flagship gaming mouse. Like the DeathAdder V3 Pro, it's an excellent mouse — but it doesn't make sense for everyone to upgrade.

Design and Comfort

The design of the DeathAdder V3 Pro was perfect — the ergonomic, contoured right-handed shape fit perfectly in my hand to the point where I forgot I was even using a mouse. The mouse felt so good that I found my hand moving toward it whenever it was on my desk (even while I was trying to test other mice). Razer went all-in on perfecting the DeathAdder V3 Pro's shape, and I don't think there's much (if anything) that could be done to improve it.

Luckily, Razer and I are on the same page, which is why the new DeathAdder V4 Pro is basically identical to its predecessor. The DeathAdder V4 Pro features the exact same size and shape as the V3 Pro: ergonomic and contoured to fit your right hand, with a left-shifted center hump and lightly curved sides to keep the mouse grippable. It has large, separated primary mouse buttons with subtle channeling to help keep your fingers in place, two large, flat thumb buttons on the left side, and a notched, textured rubber scroll wheel.

The DeathAdder V4 Pro is a relatively large mouse overall, measuring 5 inches (128mm) long by 2.67 inches (68mm) wide, and 1.73 inches (44mm) high at its highest point. This is slightly smaller than most general-purpose gaming mice, such as the Logitech G502 X Plus (5.17 x 3.12 x 1.62 inches / 131.32 x 79.25 x 41.15mm) and the Razer Basilisk V3 Pro 35K (5.11 x 2.96 x 1.67 inches / 130 x 75.4 x 42.5mm), but it's still on the larger side in general. It's almost the same size as the Logitech G Pro X Superlight 2 Dex (4.95 x 2.67 x 1.73 inches / 125.8 x 67.7 x 43.9mm), which is Logitech's answer to the DeathAdder lineup.

The DeathAdder V4 Pro may be the same size and shape as its predecessor, but it's not the same mouse. It features a smooth, matte finish (unlike its predecessor's lightly textured finish) and is about 10% lighter: The V4 Pro weighs 56g (57g in white), while the V3 Pro weighs 63g (64g in white). Six grams might not seem like much, but it's a noticeable difference, side-by-side. The mouse also feels a bit sturdier than its predecessor — it doesn't seem to have any of the minor side creaking issues I eventually noticed on the V3 Pro (but I've also only been using it for a month or so).

The bottom of the mouse is expectedly simple, with two large PTFE mouse feet at either end (along with a PTFE strip around the sensor) for smooth gliding, and a power button that doubles as a DPI switch. The mouse has no RGB lighting, but it does have a small, colorful LED indicator light next to the scroll wheel that changes color to indicate your current DPI preset. The mouse comes with five preset DPI steps: 400, 800, 1600, 3200, and 6400.

The DeathAdder V4 Pro also comes with a new wireless dongle — instead of the typical USB-A dongle that plugs directly into your PC, the V4 Pro comes with a small, weighted wireless dongle that connects to your PC via a USB-C to USB-A cable. The dongle measures approximately 1.8 inches (45.6mm) in diameter by 1 inch (25.5mm) high and weighs around 1.6 ounces (45g) and features a grippy rubber base to keep it from moving around on your desk. It has a USB-C port on one side and three small, colorful LED dots on the other, which change color to indicate connection quality, battery level, and polling rate.

In the box, the DeathAdder V4 Pro comes with a 6-foot (1.8m) braided USB-C to USB-A cable, a wireless 8K dongle, and a set of pre-cut grip tape. The USB-C to USB-A cable feels durable and fairly stiff — it's not one of Razer's "SpeedFlex" cables, so it's not an ideal cable for using the mouse in wired mode.

Specs

Performance

The DeathAdder V4 Pro features Razer's Focus Pro 45K optical sensor (gen-2), which has a maximum resolution of 45,000 DPI and a maximum speed of 900 IPS, and can handle up to 85 G's of force. This is Razer's latest, fastest, and most accurate sensor (and this is the first mouse in the brand's lineup to feature this sensor), and it delivers exactly as well as you'd expect, with smooth, accurate, ultra-precise movements on just about any surface. We tested it on a variety of hard, soft, hybrid, and glass mouse pads, as well as on plain surfaces (wood, laminate, polished concrete, and frosted glass), and it had no issues tracking on any of them.

The mouse also features Razer's new gen-4 optical mouse switches and an optical scroll wheel. The switches are rated for 100 million clicks and definitely feel and sound different from the switches in the DeathAdder V3 Pro — they're ever-so-slightly lighter and have a higher, somewhat hollow-sounding click, but they feel great: firm, clicky, and accurate. The notched scroll wheel is also fairly lightweight — it's not as stiff as the DeathAdder V3 Pro's scroll wheel, and it felt even more accurate and precise despite having freer movement.

If you're looking for a simple, ultra-light gaming mouse, the DeathAdder V4 Pro is probably the mouse for you. You don't have to be a competitive player to appreciate how good the mouse feels in hand, or to recognize how precise and accurate its sensor and switches are — and if you are a competitive player, the DeathAdder V4 Pro's all-optical setup and 8K-ready polling dongle are about as cutting-edge as you can get on a gaming mouse right now.

Features and Software

The DeathAdder V4 Pro is a fairly straightforward mouse, but it does have six programmable buttons (eight programmable inputs, including scroll up/down). You can customize the mouse and its dongle using Razer's universal peripheral software, Synapse. You can remap the mouse's buttons and set up a second layer of inputs using Razer's HyperShift, and you can also customize the dongle's LED lighting indicators.

Synapse is also where you can adjust the mouse's preset DPI steps and change the polling rate — the default is 1,000 Hz, but you can permanently move it up to 8,000 Hz or you can turn on the "Smart Polling Rate Switcher," so the mouse will automatically switch to an 8K polling rate when you're running a game in full screen mode.

Synapse also has some calibration features, including a sensitivity matcher that lets you match the DeathAdder V4 Pro's sensitivity to that of your old mouse. You can also set up smart tracking, asymmetric cut-off, dynamic sensitivity, and mouse rotation to customize the mouse to perfectly fit your gaming style.

Wireless Experience and Battery Life

Like its predecessor, the DeathAdder V4 Pro only features two forms of connectivity — low-latency 2.4GHz wireless and wired (USB-C). This is pretty standard for ultra-light stripped-down gaming mice — gamers aren't going to use Bluetooth, anyway, so including it is really just a lifestyle convenience. The polling dongle does keep track of how strong your mouse's connection is, which is a nice feature if you're concerned about how many wireless peripherals you're trying to run simultaneously and/or you're trying to use your mouse from across the room.

Razer rated the DeathAdder V4 Pro's battery life at 120 hours when it launched, but has since issued a firmware update that brings that number up to 150 hours (with a 1,000 Hz polling rate). This is significantly better than the DeathAdder V3 Pro, which topped out at 90 hours, and is higher than most similarly lightweight, stripped-down mice. The Logitech G Pro X Superlight 2 Dex, for example, gets up to 95 hours of battery life. Razer's Naga V2 Pro also gets 150 hours of battery life, but it also weighs more than twice as much at 4.72oz (134g).

Bottom Line

The DeathAdder V4 Pro is everything we loved about the DeathAdder V3 Pro, but better — it has a better sensor, better switches, and a better scroll wheel, and it weighs less and lasts longer on a charge. It also comes with the 8K wireless dongle in the box — though the V3 Pro launched with a $150 price tag, so the dongle is now just baked into the V4 Pro's price. If you're looking for a new ultra-lightweight premium gaming mouse, and you don't need extra buttons, RGB lighting, or Bluetooth connectivity, you won't be disappointed with the DeathAdder V4 Pro.

That said, it's still $170 for what is, ultimately, a pretty simple mouse. While it does improve upon the DeathAdder V3 Pro in many ways, most of those ways are fairly minor when you consider that the V3 Pro is still a fantastic mouse. I'm not saying the V4 Pro isn't worth it, because it is, but if you can find the DeathAdder V3 Pro on sale now that the V4 Pro has launched, you may not need to splurge on Razer’s latest, greatest cursor controller.

The Lumos actually has two lasers: a 10W blue diode for cutting and engraving things like wood, leather, cardboard, and paper; and a 3W infrared laser for engraving metal and some acrylics. The Lumos is a galvo-style laser, which uses a stationary laser mounted in the head and a system of mirrors to move the beam around a circular work surface. This is the secret to both the Lumos’ speed and small size.

The small size limits this machine to smaller projects like coasters, patches, keychains, and jewelry. It has an optional rotary attachment and slider for crafting larger projects, but this requires lifting the lid and switching to goggles for eye protection. The WeCreat website shows happy makers without eye protection with the protective lid up, which I find unsettling.

It uses a built-in HD camera for positioning materials and autofocusing. It needed calibration despite the manual discouraging it. The software had limited material presets, and its project library leans heavily on paid content, with most design files locked behind a premium service. There was also an annoying “nag” to activate a premium membership whenever I booted up the system.

Portability like this comes at a price. The Lumos retails at an eye-watering $2,399.99, but is currently on sale for $1,299.99. Like most lasers of this size, the target audience is entrepreneurs who can get more of a return from a premium tool like the Lumos.

Specifications: WeCreat Lumos

WeCreat Lumos: Included in the Box

The WeCreat Lumos arrived almost completely assembled. The laser unit tucks compactly into its base with the protective cover. Also included was a paper copy of the manual, a 3mm piece of basswood, an exhaust hose and connector, a USB cable, a power brick, a toolkit, positioning pieces, and two working panels (one for cutting and one for engraving). Notably absent is a pair of safety glasses.

We also received the optional rotary, protective cone for handheld use, and slide extension for the review.

Design of the WeCreat Lumos

The WeCreat Lumos has a compact form that fits easily in small spaces, with a carrying handle on top to make it easily portable. It features a built-in safety cover that can shield the workspace completely for smaller items. The laser head is round and rides on a recessed motion system driven by a single z-axis motor. The z-axis is used only for focusing the laser, either by aligning the red and blue dots manually or using the autofocus button.

The Lumos uses small mirrors (galvanometers) to move the laser beam around the work surface with higher speeds than can be achieved by moving the entire tool head. The one drawback to a galvo laser is that the usable work surface is limited to a cone fanning out from the center of the laser.

When cutting, the beam will make a diagonal cut as it moves further from the center point, which is noticeable on larger items or items placed toward the outer edge of the work surface. Most of the time this angle is unimportant, but could cause trouble when fitting together pieces with tight tolerance.

The WeCreat Lumos has a blue 10W diode laser and a red 3W infrared laser. The diode laser really shines while engraving and cutting wood, completely opaque dark acrylic, leather and other organic materials. The laser can cut basswood and opaque acrylic up to 6mm, and it can engrave a painted metal surface like a tumbler. Anything metal is better engraved by the infrared laser. The IR laser cannot cut metal, but engraves it beautifully, with terrific resolution.

Assembling the WeCreat Lumos

Assembling the WeCreat Lumos is mostly taken care of at the factory. The laser is secured to it’s base by a thumb screw, and a wrench is included if you need help.

The exhaust connector is attached to the rear of the unit with four screws, which are already mounted and must be removed. The exhaust hose fits on the connector and is secured with a cable clamp.

After that, remove the lens cap, plug in the laser, and you are ready to go.

Safety Precautions for the WeCreat Lumos

The WeCreat Lumos is a Class 1 laser when the cover is fully in place. When it is up, it becomes a Class 4, requiring laser safe glasses. Unlike many other lasers I’ve reviewed, this laser did not come with safety glasses and the unit does not have an emergency shut off if the lid is raised while printing. The software has no reminder about this, either. As an adult, the safety warnings can sometimes be overdone and are often bypassed. Suffice to say that this laser is suitable for experienced users only, and little hands and pets should be kept away while operating.

Some materials should not be burned with a laser due to their chemical makeup – they could melt, catch fire or produce toxic fumes. Dallas Maker Space has published a list of safe and hazardous materials to use with their laser. The list of no-nos includes plastics, fiberglass and certain foams.

The Wecreat Lumos has an excellent flame sensor that immediately pauses the laser if fire is detected. The machine can be restarted if the issue is small, and the setting can be adjusted if overly sensitive. There is also a tip sensor which will stop the laser if it tilts while running.

Software for the WeCreat Lumos

The WeCreat Lumos uses the WeCreat Makeit! App which can be downloaded from WeCreat’s website. You do need to make an account if you would like to have access to WeCreat’s ecosystem of projects. At the time of this review, there are 24 files listed for beginners, which are free. Currently, all the others require paying by the file, usually $0.99, or paying a premium membership, currently $7.99 monthly or 14 months for $95.99. You do get a free six months of premium membership with the purchase of the laser. If you are not a premium member, there is an annoying nag to sign up, every time you connect to the laser.

The software has a material preview, which does a good job of showing the options for the material and operation you are working with. The material presets are very good though the amount of materials listed is limited. Be prepared to search online for many material settings. Once you have a setting that works, it can be added to the presets.

Engraving / Cutting with the WeCreat Lumos

The review unit came with two pieces of bass plywood. For a first test using the blue laser, I engraved and cut a humming bird from the WeCreat MakeIt’s image library. After putting the material on the ventilated cutting plate and pressing the autofocus button on the machine, I chose to fill engrave the bird twice with different settings and cut it as well. The software accesses the camera by hitting the refresh button, showing exactly where the cutting and engraving will happen. I do recommend that you calibrate the camera, as mine was off quite a bit. The calibration is automatic and does require a piece of unused plywood the size of the work area. The rounded corners show the edge of the laser's cutting ability.

This simple engrave and cut took less than two minutes. The two engravings look identical, though one was done with the machine's preset of 55% power and a speed of 250mm/s. The other was done at 60% power and a speed of 200mm/s. The cut was done at 100% power and a speed of 7mm/s. I used the software’s enhanced cutting feature, which added a second pass of the laser, which was unnecessary in this case. The result was a very nice cut with no charring of the material.

From the free beginner section of the WeCreat MakeIt software, I cut a simple aircraft model. Using the blue laser and the machine preset for 3mm basswood, 100% power, a speed of 7mm/s and 2 passes, the project took 4 minutes and came out very nice, again with little charring of the edges.

I did a more intricate engraving using clip art from Canva and chose fill engraving. The machine default of 100% power and 140mm/s was a little too much for this design. The outside cut at 100% power and 7mm/s speed with 2 passes was perfect. In all, the engrave and cut took a little over 8 minutes.

Going off the beaten path and beyond the material presets listed, I engraved a couple hammers I swiped from my husband’s workshop. One is a 5lbs engineer hammer and the other is a rubber mallet. I’ll let you figure out which is which. I used power of 100% and a speed of 100mm/s. With an engraving time of less than 3 minutes each, both came out very nice, with one needing just a little pre- and post-sanding since the handle had a shellac finish on it. This project also shows the capability of the WeCreate Lumos to not lose focus while engraving on a curve.

For slate, the WeCreat Lumos has settings for both the blue and red laser. I used the blue laser to engrave a favorite college team logo. Engraving the image as a bitmap, I used the material preset of 78% power and DPI of 254. I cranked up the dotting duration to 500 and did 2 passes. In less than 3 minutes, the result was terrific.

While the 10-watt blue diode laser is a respectable cutting machine with limits, the 3-watt infrared laser won’t cut at all. Just for fun, I tried cutting a metal business card at 100% power and 3mm/s. Despite multiple passes, it didn’t make much of a dent. If you just want to engrave metal, then it’s pretty darned good.

For the first use of the infrared laser, I engraved business cards that turned out sharp, clear, and beautiful. The material preset for bitmap engraving of one pass at 100% power, 254 DPI, and 230 dotting duration was perfect, and the laser etched perfectly out to the edges of the work surface. The work time was just over one minute.

Next, I hooked up the optional rotary that was included for this review. It plugs into a port at the rear of the Wecreat Lumos and though it can be secured to the work plate with screws, it was stable enough on its rubber bottom pad that I didn’t bother. This is also the first time the laser cover had to remain up in order to use the laser. Again, I would stress the need for laser safe glasses and the need to keep the laser away from anyone unprotected.

The rotary can be configured for a variety of items from baseball bats to tumblers but the one that intrigued me the most was rings.

The software does a decent job of making the area that will be printed with the camera. The framing function also can help to line up the work. For round objects, you need to measure the perimeter at the engraving area. The included fabric tape measure works well enough. After the perimeter is entered, the 3d preview can show how far around the object the engraving will go and will show where to place your artwork if you are doing the front and back of an item and help avoid any overlap.

I was able to get a really nice engraving on my first try, but I did not realize that the engraving must be mirror imaged in the software to engrave correctly. I used the material preset for 1mm titanium alloy, since there was no preset listed for stainless steel. Using fill engrave at100% power and a speed of 10mm/s and a line density of 199, I engraved one ring with Make All The Things and JRR Tolkien’s One Ring. With a work time of 6 minutes each, they both came out nice and sharp and just look great. The WeCreat Lumos can also engrave the inside of rings by tilting the rotary head up, and it is not bothered at all by engraving items tilted slightly off 90 degrees.

Without even needing to remove the studs used for the rings the outer rubber feet can be used to hold the small tumbler I used next. There is a good preset available for stainless steel tumblers. Using fill engrave at 100% power, a speed of 158mm/s, a line density of 120 it took a little over 7 minutes. Again, the result is excellent.

Bottom Line

The WeCreat Lumos is a tiny power house perfect for crafting on the go. If you want to offer custom engraving at festivals and markets, this is definitely a laser to check out. However, if you prefer a stationary workshop experience, there are other larger lasers that would be more cost-effective.

The Lumos twin galvo lasers let it engrave nearly any surface you can stuff in the chamber. When projects get too large, the machine can be pulled off its mount, wrapped in a protective cone, and used as a handheld laser. I find it troubling that our review unit came with a rotary tool that requires the lid to be lifted, yet it did not come with safety glasses.

WeCreat’s custom software is aimed at beginners and is very easy to use. It is also compatible with Lightburn for advanced users who don’t want to be hounded into purchasing a membership.

The $2,399.99 price tag is high, but typical of portable lasers. I’ve been watching the price while I had it for review, and it seems fairly easy to pick up on sale. Currently, it’s on sale for $1,299.99.

It’s a very good machine, one that we’re pleased to add to the list of best laser cutters.

If your small business needs more room for laser cutting large items, the 40-watt xTool S1 is a great choice, and just a bit more at $1799. If your creativity can not be contained by a laser alone, then the xTool M1 Ultra with its laser, vinyl cutter, and embosser for $1148 can help you make all the things.

The System Power 11 550W positions itself as the entry point to Be Quiet!'s power supply lineup, targeting budget-conscious builders who still expect fundamental quality and quietness. We took a closer look to see if this budget-friendly PSU can compete against the best power supplies in its category. This unit adheres to the modern ATX 3.1 specification and delivers 80 Plus Bronze efficiency certification, promising respectable energy conversion while remaining accessible to mainstream builders. At its retail price of approximately $60 or 50€, the System Power 11 faces fierce competition from established players, making its value proposition dependent on execution quality and the inclusion of forward-looking features like native PCIe 5.1 support.

Specifications and Design

In the Box

The Be Quiet! System Power 11 is supplied in a cardboard box featuring the company's characteristic all-black theme, with the front panel dominated by an image of the power supply itself. Inside, a protective nylon pouch and foam inserts provide adequate protection during shipping.

The included accessories take an absolutely minimal approach, containing only the essential mounting screws and standard AC power cable. Be Quiet! has not included any cable management accessories, zip ties, documentation beyond basic specifications, or other supplementary items.

The power supply features completely hardwired cables throughout, a design decision that has become increasingly uncommon in the modern power supply landscape where modular connectivity has penetrated even budget segments. All cables employ an all-black color scheme with flat, ribbon-like wires and black connectors. Most cables lack any exterior sleeving, though the ATX and PCIe 5.1 connectors receive black nylon sleeving that forms a more traditional circular cable profile.

Of particular interest is the inclusion of a 12V-2x6 connector, demonstrating Be Quiet!'s commitment to forward compatibility despite the unit's budget orientation. However, this connector carries a significant caveat: its power output is limited to 300 watts rather than the 600-watt maximum that the PCIe 5.1 specification theoretically allows. This restriction means the connector will work adequately with potentially with more power-efficient future GPUs, but high-end graphics cards requiring substantial power delivery will be limited by its capabilities. The connector selection also includes only a single Molex connector, which may prove limiting for builders with older peripherals or specialized cooling equipment.

External Appearance

Be Quiet! kept things aesthetically simple and functional with the System Power 11. The chassis adheres precisely to the ATX standard's 140mm length specification, ensuring universal compatibility with any ATX-compliant case without clearance concerns. The unit receives a satin black chassis finish that provides an attractive appearance while remaining neutral enough to complement virtually any build theme. However, this particular coating proves somewhat prone to smudges and fingerprint marks, requiring a bit of care after handling. A decorative logo is subtly embossed on the right side of the unit.

The front panel employs the expected minimalist approach, featuring only the standard AC power receptacle and rocker-style power switch. The completely hardwired design results in a plain rear panel devoid of any modular connectors. The electrical specifications and certifications label covers the left side of the unit. The top surface remains completely flat and free of decorative elements.

The fan guard integrates a typical circular wire guard design, with a large company logo decorating its center and the series branding prominently printed beneath. This approach balances functional characteristics with simplistic, yet tasteful branding that avoids the excessive RGB lighting and aggressive styling.

Internal Design

A Yate Loon D12SM-12 120mm fan equipped with a sleeve bearing system is responsible for the active cooling of this unit. Yate Loon represents a well-established manufacturer in the cooling industry with a proven track record spanning decades, instilling confidence in the component selection despite the budget nature of the unit. However, sleeve bearings represent a cost-conscious choice that prioritizes quiet operation over ultimate longevity. While sleeve bearings operate more quietly than ball bearing alternatives under normal conditions, they prove more susceptible to wear under sustained high-temperature operation and may exhibit reduced lifespan in demanding thermal environments or systems with poor ventilation. The fan carries a maximum rotational speed of 1650 RPM, which proves perfectly adequate for a power supply with this output level and should provide sufficient airflow without becoming excessively loud.

The original equipment manufacturer behind the System Power 11 is Heroichi, also known as HEC or Compucase. This revelation carries significant weight, as HEC represents one of the oldest participants in the power supply manufacturing field, having been founded in 1979. Their longevity in this competitive industry speaks to fundamental competence and manufacturing capability. However, the platform used for the System Power 11 appears to be a subtle refinement of HEC's older Bronze-level design that has been in circulation for well over a decade. This aged architecture brings both advantages and disadvantages: proven reliability and known characteristics balanced against the absence of modern innovations that newer platforms incorporate.

The input filtering stage implements a very basic configuration utilizing just two Y capacitors, two X capacitors, and two filtering inductors. A single rectifying bridge handles AC-to-DC conversion, and interestingly, it lacks a dedicated heatsink of its own, relying instead on convective cooling alone.

The Active Power Factor Correction (APFC) circuitry and primary inversion stage share a common heatsink, a space-saving design approach that proves adequate given the modest power output. The active components of the APFC stage consist of two GPT13N50DG MOSFETs paired with a single diode, while the passive components include one small inductor and a Teapo 330 μF capacitor. The primary inversion stage employs two Infineon 60S180P7 MOSFETs configured in a half-bridge topology, a classic and well-understood arrangement that balances efficiency with simplicity.

The secondary side implementation reveals the fundamental upgrade HEC applied to their decades-old platform design: the inclusion of DC-to-DC conversion circuitry for the minor voltage rails. This represents the essential modernization that distinguishes this iteration from its now-ancient predecessors, where group regulation topology once handled all voltage rails. The DC-to-DC circuits allow the 3.3V and 5V rails to maintain independent regulation regardless of load distribution across different rails, providing more stable voltage delivery for modern computing loads that place increasingly asymmetric demands on different voltage rails. This upgrade demonstrates HEC's effort to keep their aging platform relevant for modern applications, even if the underlying architecture remains largely unchanged from designs that debuted over a decade ago.

Six secondary side MOSFETs handle the 12V rail's synchronous rectification duties, marked as PFR30L60CT. These mount on a sizable heatsink positioned immediately after the main transformer, providing adequate thermal management for the switching components. The secondary side capacitors are supplied entirely by Teapo, all rated for 105 degrees Celsius operation. The capacitor choice here represents a pragmatic decision: Teapo does not command the prestige of Japanese manufacturers like Nippon Chemi-Con or Rubycon, but they have established a proven reliability record in budget and mainstream applications over many years.

Cold Test Results

Cold Test Results (25°C Ambient)

For the testing of PSUs, we are using high precision electronic loads with a maximum power draw of 2700 Watts, a Rigol DS5042M 40 MHz oscilloscope, an Extech 380803 power analyzer, two high precision UNI-T UT-325 digital thermometers, an Extech HD600 SPL meter, a self-designed hotbox and various other bits and parts.

During cold testing, the Be Quiet! System Power 11 demonstrated respectable efficiency performance that comfortably meets its 80 Plus Bronze certification requirements. At 115 VAC input voltage, the unit achieved an average nominal load efficiency of 86.7% across the 10% to 100% load range, while 230 VAC input improved performance to 88.2%. These figures place the unit solidly within Bronze certification parameters, though they trail more efficient designs by several percentage points. The efficiency curve exhibits typical characteristics, peaking at approximately 50% load, where most power supplies achieve their optimal performance due to the inherent characteristics of switching PSU topologies. Low load efficiency proves reasonable for a unit in this market segment, suggesting that users running basic systems will benefit from acceptable power conversion characteristics even during idle or light usage scenarios. The unit does not carry any Cybenetics certification.

The fan behavior during cold testing proved exemplary, maintaining low rotational speeds across most of the load spectrum and becoming audible only when the unit approached heavily loaded conditions exceeding 80% capacity. The thermal performance under normal ambient temperatures proves more than adequate for typical desktop computer environments, with internal temperatures remaining well within reasonable limits during sustained operation.

Hot Test Results

Hot Test Results (~45°C Ambient)

Elevated ambient temperatures reveal the System Power 11's thermal limitations and the age of its underlying platform design. Under hot testing conditions that simulate poor ventilation and/or very warm climate operation, average nominal load efficiency drops precipitously to 85.0% at 115 VAC and 86.5% at 230 VAC. This represents significant performance degradation that suggests thermal stress within the design, with efficiency losses of approximately 1.7% compared to cold testing conditions. This degradation proves more severe than observed in modern platforms utilizing more thermally robust components and advanced topologies.

The fan responds more aggressively under elevated ambient conditions, ramping up rotational speed earlier in the load curve and approaching maximum RPM during sustained high-load operation. However, the relatively modest 1650 RPM maximum speed means the unit never becomes excessively loud even under worst-case thermal conditions. The acoustic profile transitions from virtually silent at low loads to audible but not intrusive at high loads, maintaining Be Quiet!'s acoustic performance standards even when thermally stressed. Internal temperatures remain within safe operational margins.

PSU Quality and Bottom Line

Power Supply Quality

The Be Quiet! System Power 11 delivers commendable electrical performance that exceeds expectations for its budget market positioning and aged platform architecture. Voltage ripple filtering proves quite good for a unit in this price category, with maximum ripple levels of 48 mV on the 12V rail, 30mV on the 5V rail, and 30 mV on the 3.3V rail. These measurements fall comfortably within acceptable industry standards and well below the typical thresholds where ripple begins causing system instability or component stress.

Voltage regulation across all rails proves fair given the aged platform architecture, though it trails the tighter regulation possible by a modest margin. The primary 12V rail maintains 1.7% regulation across the load range, while the secondary 5V and 3.3V rails achieve 2.1% and 2.0% regulation respectively. These figures represent acceptable performance, delivering notably better regulation than the group-regulated designs of the past and are entirely adequate for modern computing applications.

During our thorough assessment, we evaluate the essential protection features of every power supply unit we review, including Over Current Protection (OCP), Over Voltage Protection (OVP), Over Power Protection (OPP), and Short Circuit Protection (SCP). All protection mechanisms were activated and functioned correctly during testing.

The Over Current Protection triggers at appropriate thresholds: 135% for the 3.3V rail, 132% for the 5V rail, and 115% for the 12V rail. These settings prove well-calibrated, providing adequate headroom for transient loads while protecting against sustained overcurrent conditions that could damage components. The OPP activates at 116% under hot test conditions, which represents a reasonable threshold that allows brief power excursions while preventing sustained operation beyond the unit's design capabilities. All protection features function sharply without excessive delay or hunting behavior, suggesting proper tuning of the protection circuitry parameters.

Bottom Line

The Be Quiet! System Power 11 550W occupies a nuanced position in today's competitive power supply landscape, successfully delivering fundamental reliability and modern connectivity while making necessary compromises to achieve its accessible price point. The unit succeeds in its primary mission of providing stable, quiet power delivery with adequate efficiency certification. However, it cannot escape the inherent limitations of its decade-old platform architecture and budget-oriented component selection. HEC's platform engineering proves thoroughly competent if unremarkable, delivering electrical performance that meets industry standards while avoiding the concerning issues that have plagued some truly budget power supply designs. The company's decades of manufacturing experience become evident in the refined nature of this aged platform.

The electrical performance may not impress enthusiasts accustomed to premium units, but it proves entirely adequate for typical desktop computing applications. The assembly quality deserves particular praise, with excellent soldering work, proper component placement, and attention to manufacturing details that suggest HEC takes pride in their work despite the budget nature of this product. Thermal performance under elevated ambient conditions reveals the limitations of both the component selection and platform age, with efficiency degradation and increased noise levels indicating thermal stress within the design. More efficient platforms handle elevated temperatures with less performance degradation. However, the unit maintains stable operation throughout testing without exhibiting dangerous behaviors or protection failures, demonstrating that the design remains fundamentally sound even when pushed beyond its comfort zone. The inclusion of a 12V-2x6 connector demonstrates forward-thinking product planning and provides genuine value for builders planning to use modern graphics cards. However, the 300-watt power limitation is immediately restrictive. This represents a pragmatic compromise given the unit's overall power output and thermal capabilities. The connector will serve adequately for less powerful graphics cards or potentially more efficient future GPU generations, providing reasonable futureproofing for mainstream builds.

The five-year warranty provides adequate peace of mind for budget-conscious builders, though it falls short of the seven-year or even ten-year periods offered by premium alternatives. This warranty duration aligns appropriately with the unit's market positioning and expected use cases, suggesting Be Quiet! has confidence in the platform's reliability without making unrealistic promises about longevity. The current retail price of approximately $ 70 or € 60 represents a reasonable value for a unit with this feature set and performance profile, particularly given the inclusion of the 12V-2x6 connector that many competing Bronze-level units still lack. However, availability proves problematic in the United States market, where the unit faces stock shortages and inflated pricing when available. European buyers enjoy better availability and more favorable pricing, making the System Power 11 a more compelling proposition in those markets.

For budget-conscious builders assembling mainstream systems for office work, casual gaming with mid-range graphics cards, or general computing applications, the System Power 11 provides adequate service with the reassurance of Be Quiet!'s reputation behind it. The unit proves particularly well-suited for typical home and office computers that will not stress the aging platform with extreme loads or thermal conditions. However, enthusiasts planning high-performance builds, overclockers, or users intending to install power-hungry flagship graphics cards would benefit from considering alternatives that offer superior thermal performance, more modern platform architecture, and enhanced long-term reliability prospects. The System Power 11 delivers honest performance for its market segment without pretending to be something it is not, making it a reasonable choice for appropriate applications while acknowledging its clear limitations.

MORE: Best Power Supplies

MORE: How We Test Power Supplies

MORE: All Power Supply Content

Since that time, TP-Link sent us a Deco BE65-Outdoor to review, which is a tri-band Wi-Fi 7 extender, rather than the dual-band Deco BE25-Outdoor. That added band and performance also means that the Deco BE65-Outdoor is more expensive, with an MSRP of $299.99 versus $149.99 for the BE25-Outdoor I had been using.

Design of the TP-Link Deco BE65-Outdoor

The Deco BE65-Outdoor is similar in design to the BE25-Outdoor: It’s a tall, white cylinder with a black base that unscrews to reveal the ports inside for connectivity and power. A rubber gasket prevents moisture and dirt from entering the chassis. With that said, TP-Link says that the Deco BE65-Outdoor is IP65 certified for dust and water resistance.

Inside the unit, you'll find two 2.5 GbE LAN ports that support Power over Ethernet (PoE), and a power port for the included 9-foot cable.

The Deco BE65-Outdoor measures 4.23 × 4.23 × 6.93 inches and weighs 3.48 pounds.

TP-Link Deco BE65-Outdoor Specifications

Setting up the TP-Link Deco BE65-Outdoor

The first step for setting up the Deco BE65-Outdoor was to find a suitable place to mount it outside. This time around, I opted to mount the unit in my backyard on a tree. The tree is roughly 25 feet away from the main router in my Deco BE63 Wi-Fi 7 mesh network. Only one exterior wall separates the primary router from where the Deco BE65-Outdoor was installed.

TP-Link includes a mounting template, four screws, and a plastic mounting bracket, which I installed on the tree at a height of about 6.5 feet. The Deco BE65-Outdoor then just slips securely onto the mounting bracket. Given that this was just a temporary setup for review purposes, I grabbed a 50-foot extension cord from the garage and plugged one end into an exterior wall outlet and the other end into the Deco BE65-Outdoor’s power cord.

TP-Link makes it incredibly simple to add a satellite to your existing network via the Deco app. In case you didn’t know, all of TP-Link’s Deco mesh routers are only configurable through the Deco app, and that remains the same with additional satellites like the Deco BE65-Outdoor and the Deco BE25-Outdoor.

Once I mounted and plugged in the Deco BE65-Outdoor, I went back inside the house and opened the Deco app. When the app opened, I immediately received a prompt saying that the Deco BE65-Outdoor had been found. I was then asked if I wanted to add the device to my existing Deco BE63 mesh network, to which I obliged.

Once the initial network optimization was completed, the Deco BE65-Outdoor opted to connect directly to the primary Deco BE63 router/node upstairs. Initially, the Deco BE65-Outdoor opted for a 5 GHz wireless backhaul to the mothership, rather than the 5 GHz/6 GHz MLO backhaul that the Deco BE63 satellite in my living room uses.

I found this odd, because the Deco BE65-Outdoor and my living room Deco BE63 satellite are both roughly 25 feet away from the main Deco BE63 router – and the living room node has additional walls and floors in its path. Even when I placed the Deco BE65-Outdoor within six feet of the main Deco BE63 node, it would still only connect using the 5 GHz band.

After some trial and error (and an assist from TP-Link), I discovered that the Automated Frequency Coordination (AFC) mode, which is required for outdoor 6 GHz band usage, was not enabled. This was despite my clicking the option during the initial setup to enable AFC. I found the setting buried in that Advanced section of the 6 GHz band in the wireless settings. Attempting to turn it on continued to give me an error message, so I tried rebooting the router several times. On the third reboot, the AFC mode “stuck,” and I was then able to connect reliably on the 6 GHz band.

Finally, I performed a firmware update to get the unit up to par with the rest of my mesh network.

TP-Link Deco BE65-Outdoor Performance

I set up two locations for testing performance: my back porch and my fire pit, which is about 20 feet from my back porch. I decided to do baseline testing to measure performance when connected to my existing Deco BE63 network only. I then performed tests with my laptop, wirelessly connecting to only the Deco BE65-Outdoor node on the mesh network using the wireless backhaul.

I also decided to throw in a third scenario, connecting the Deco BE65-Outdoor to the Deco BE63 living room node using a wired backhaul. In this scenario, I connected a 75-foot CAT6e cable to a free 2.5 GbE port on the Deco BE63 and ran the cable all the way out my back door, across the back porch, and across the yard to the tree-mounted Deco BE65-Outdoor and plugged it into a free 2.5 GbE port. This is the best-case scenario for wireless performance. When an Ethernet backhaul is used, the Deco BE65-Outdoor automatically makes the necessary changes to integrate with the network. I then performed tests with the laptop connecting wirelessly to the Deco BE65-Outdoor node.

For my test system, I used an HP OmniBook X (Qualcomm Snapdragon X Elite) laptop with a Qualcomm FastConnect 6900 Wi-Fi 6E wireless card. I then used my trusty iPerf3 server, which, in this instance, was connected via a 2.5 Gbps Ethernet connection to my Deco BE63 router.

With that said, the numbers speak for themselves regarding performance. The baseline numbers when connecting to the Deco BE63 network yielded 504 Mbps on the porch and a dismal 115 Mbps by the fire pit, using the 5 GHz band. Switching to the 2.4 GHz band, I saw 25 Mbps on the porch and just 6 Mbps at the fire pit.

Shifting to the Deco BE65-Outdoor (wireless backhaul), I saw maximum throughput of 479 Mbps while on the porch, and 343 Mbps at the fire pit (a 3x improvement) on the 5 GHz band. When using the wired backhaul, iPerf3 performance on the porch climbed to 760 Mbps and averaged 601 Mbps near the fire pit.

However, the highest performance was seen using the 6 GHz band, where the OmniBook X nearly hit 900 Mbps on my porch and 570 Mbps at the fire pit when using a wired backhaul.

There were even dramatic improvements on the 2.4 GHz band, reaching 64 Mbps on the porch and 44 Mbps at the fire pit with the wireless backhaul. With the wired backhaul enabled, those numbers climbed to 74 Mbps and 60 Mbps, respectively.

Bottom Line

The TP-Link Deco BE65-Outdoor is a premium wireless satellite that can be used with the company’s Deco mesh systems. Performance across the board on my back porch and in my back yard was dramatically faster when using the Deco BE65-Outdoor versus my “base” Deco BE63 mesh network. I saw upwards of a 10x improvement on the 2.4 GHz band and significant gains on the 5 GHz band. The option to use a wired backhaul further strengthens data throughput.

However, we must then address pricing. The Deco BE65-Outdoor costs $300, which is twice the price of the dual-band Deco BE25-Outdoor. If all you’re hoping to do is expand wireless coverage to your outdoor areas, the Deco BE25-Outdoor should be wholly sufficient for your needs. But if you want a “spare no expense” wireless node that is second to none in performance, the Deco BE65-Outdoor is hard to ignore.

We’ll be taking a close look at the primary upgrade in this digital model, the 6-inch, 1600x720 screen that’s rated to 500 nit of brightness. It can show performance metrics or even be configured as a secondary monitor.

We’ll also run the case through some difficult thermal benchmarks, using Intel’s i9-14900K and an Nvidia RTX 4000 series GPU to test its cooling effectiveness, both at noise-normalized settings and with the fans at maximum speeds.

Will Lian Li Lancool 207 Digital make our list of the best PC cases? You’d better believe it! The thermal performance of this case is among the best on the market, as you’ll see in the benchmarks section. Let’s take a look at the specifications and features of the case, then we’ll wrap up our review with thermal results.

Product Specifications

Features of Lian Li’s Lancool 207 Digital PC case

▶️ GPU length and anti-sag support

While it is a fairly compact mid-sized case, Lian Li’s Lancool 207 supports the biggest RTX 5090 GPUs, with room for graphics cards up to 375mm (14.76 inches )in length. An essential GPU anti-sag bracket is included with the case, outlined below in red.

▶️ PSU up front makes room for cooling at the back

Most PC cases have the power supply installed in the rear of the case, but Lian Li’s Lancool 207 features a front placement for the PSU. This is a trend we’ve seen elsewhere, like in Montech’s HS01 Pro. But here it’s at the bottom, and turned 90 degrees. A bottom mesh cover assures the PSU won’t have an issue dumping its heat outside the chassis.

▶️ Dual fans up front and the bottom of the case, with strong airflow in the back

The case comes with two 140mm fans installed in the front (non-RGB on this model), and two 120mm fans installed on the bottom. Despite lacking – at least in the default configuration – a traditional exhaust fan, airflow through the back of the case (which is also mesh with large holes) is very strong! For those wanting the best cooling possible, an exhaust fan can be installed here. Up to three additional fans, or a 360mm AIO, can be installed at the top of the case.

▶️ Air cooler support

In addition to supporting the largest GPUs on the market, the Lancool 207 Digital also has room for larger air coolers, with towers up to 180mm tall (7.1 inches) supported.

▶️ Bottom view: rubber feet and dust filter

There are four rubber feet on the bottom to ensure airflow to the PSU (there’s also a dust filter here) and prevent it from moving.

▶️ Back side, cable management, and storage support

From the backside of the case, we see a simple but efficient setup. Cables can be managed and tucked in a tidy fashion secured by the three velcro straps on the left side of the case and through the top. Those who have extra cables from their PSUs can stash them above the SSD plates, highlighted in blue. Both 3.5-inch and 2.5-inch drives are supported by the drive plates, shown below.

▶️ ATX, Micro-ATX, and Mini-ATX support

While the Lancool 207 won’t support the largest EATX motherboards, it does support standard-sized ATX boards up to 244mm in width, as well as smaller boards. We’ve installed the Asus Z790-P Prime Wifi, paired with Intel’s i9-14900K CPU for this review’s testing.

▶️ IO panel

The IO panel is located at the front and top of the case. It offers a power button, audio jack, and three USB ports – two USB-A and one USB-C.

▶️ Accessories packaging

Lian Li goes the extra mile in the packaging of case accessories. Many manufacturers include the spare parts in simple plastic bags or a cardboard box. The Lancool 207 comes with a premium box that includes dividers that can be used to customize it as desired.

▶️ Toolless side panels

The side panels of this case can be removed and installed without tools, which some may appreciate for simplicity.

▶️ Six-inch, 500-nit 1720x600 display

The distinguishing feature of this case revision is the inclusion of a 500-nit, 1720x600 display. This can be customized through Lian Li’s L-Connect 3 software, which we’ve covered in-depth in previous reviews. It allows easy customization of the screen to show performance metrics, images, or videos of your choosing. There is a large variety of pre-packaged themes to choose from, I’ve shown a few in the pictures here.

But if you don’t want to use Lian Li’s software or just want to display something else, the screen can also function as a secondary display.

This allows the display to be used as you see fit – I’ve run Furmark at a whopping 450 FPS in the demo above. One thing I like using this feature for is to set up a slideshow of pictures in an album, changing every few minutes to display a different memory.

Most users will need to connect the display to a USB header on their motherboard. However, if you don’t have a spare USB header on your motherboard, Lian Li aslo includes an adapter so that the display can be connected through a traditional USB-A port!

▶️ Rear view

There’s not much to say about the rear of the case. Like most, it supports seven PCI-e expansion slots. The main difference you’ll notice from typical cases is the power plug on the bottom of the case, which connects to the power supply in the front of the unit.

Thermal tests, test setup, and testing methodology

The measurements I’ve benchmarked this case against focus on the efficiency of the case’s thermal transfer.

Our thermal tests are presented to give you more information about the product’s performance, but aren’t intended as the sole judgment of the chassis. The style, price, features, and noise levels of a case should also be considered, and we all have different preferences. What I might like in a case, you might not, and that’s OK. My goal with these reviews is to give everyone, no matter their preferences, enough information to decide whether or not a product is right for them.

Maximum thermal performance – pre-installed fans noise-normalized to 38.9 dBA

This set of tests is designed to appeal to those who love quiet PCs, with both system fans and the CPU fan noise-normalized to 38.9 dBA. Consider this a measure of the case’s thermal efficiency when the fans are set to run quietly.

The first noise-normalized test stresses only the CPU, running Cinebench R23 multi-core with power limits removed. I did a double-take when I saw the results here. The performance shown was better than any other result I’ve had since adopting this method of testing PC cases. This is a testament to the efficiency of the fans and airflow of this PC case.

The second noise-normalized test I perform is much like the above, but also fully stresses the RTX 4070 TI Super GPU I have paired with the system, adding another 295W of heat to the case. The GPU temperatures in this test were the third-best I’ve seen among the cases I’ve tested this way. The result of 64.2 degrees Celsius is only 0.7 degrees C warmer than our best result here!

As a result, the GPU’s fan speeds were also the third-quietest in this scenario, with an average measurement of 1,454 RPM.

Maximum Noise Levels

In terms of maximum volume, Lian Li’s Lancool 207 Digital spins up to 46 dBA with my sound meter from 1 meter away, which is more or less average for PC cases. One thing my charts don’t show, however, is the pitch noise, which isn’t particularly annoying here (at least to me). So I would say that the noise doesn’t “feel” as loud as the dBA measurements would indicate. But your ears may give you a different opinion.

Maximum thermal performance – pre-installed fans at full speed

Our noise-normalized results are designed for folks who prefer silence, but what if you don’t mind additional noise and just care about maximum performance?

Starting with a CPU-only stress test, we measured 296.3W. While not quite as impressive as our noise-normalized results, the Lancool 207 Digital performed well enough here.

Adding the heat of a GPU returns the Lian Li to the front of the pack in terms of performance, showing our best result recorded thus far with, 294.3W cooled (on average) during this test.

Now let’s look at the temperature of the GPU. With the system fans at full speed, the thermal result is our second-best, with an average of 62.5 degrees C.

The thermal efficiency of Lian Li’s fan configuration means that the GPU has to work less to maintain low temperatures, resulting in fan speeds running at an average of 1,307 RPM. Only Montech’s case did better here, and not by much.

Conclusion

Lian Li’s Lancool 207 Digital is an impressive case. Thermal performance is simply excellent, among the best avialable. And its chart-topping noise and thermal results and nice-looking screen, paired with a very reasonable price, makes it my new favorite compact midtower case.

If you’re on a tight budget, the basic version of this case is currently available for only $90 in the U.S. But the Digital version tested here includes a fancy six inch, 1,600x720-resolution display for $105, just $15 more!

Fanttik’s $77 T1 Max wireless soldering iron isn’t as smart as others that I have used, but what it lacks in “smarts” it makes up for in ease of use. With just a dial and a push button, the T1 Max gets to work, but deep inside the soldering iron, there are quality of life improvements to save the 2600mAh battery and the soldering iron tips.

Join me as I put the Fanttik T1 Max through its paces on my new workbench.

Fanttik T1 Max Specifications

Fanttik T1 Max Look and Feel

The design of the Fanttik T1 Max is pleasingly simple for a portable soldering iron. There is no OLED screen or buttons. We just have a dial at the rear of the unit which selects the soldering temperature. Set your temperature, then press and hold the power button to start heating the soldering tip to your desired temperature. The power button is about a third of the way from the hot end of the soldering iron, and the rubberized grip.

The grip is large and somewhat comfortable to use, despite being nearly an inch thick. But because of the soldering iron's balance, with weight at the back, it feels off-center, and I found myself gripping nearer to the power button. This meant that I was gripping an area with little or no friction, and my hand easily slid around. I quickly learnt to move my grip to the correct position, but there were a few times where I shifted my grip.

Another issue that I noticed was with the grip itself. It does provide a small guard from the hot end of the soldering iron, there is no “real” guard that will stop your fingers sliding off. Sure, there is no guard on the TS101, Pinecil, or Fnirsi HS-01 and HS-02 soldering irons that I have tested, but they were smaller and had “grippier” grips, or the grip was further back.

I would’ve preferred to have seen a better guard on the Fanttik T1 Max, but it would ruin the aesthetic. The thickness of the Fanttik T1 Max and uneven weight make it uncomfortable for long soldering sessions. A smaller, lighter soldering iron is much better for those tasks. But for quick projects, the Fanttik T1 Max is fine.

Just in front of the grip is a ring of LED lights that illuminate where you are soldering. This is an interesting idea, and my HOTO electric screwdriver does the same thing, but for both tools, it adds little to the task at hand.

The power button also provides a basic level of indication, alerting you to low battery levels, charging status, heating up, etc.

The Fanttik T1 Max rests in a rather fancy-looking holder. The hot tip points down, and is a couple of centimeters from the base of the holder. The holder can be rotated to bring the iron nearer to your hand, and it can be positioned for left and right-handed persons.

The base itself hides a compartment for a selection of soldering tips. Press the drawer inwards and it pops out with three spare tips (knife, hooked point, and precision point). This is a really nice touch, and it provides somewhere to keep alternative tips, instead of adding them to my overflowing soldering drawer. I did notice that at certain angles, the base was a little wobbly; it only has three points of contact to the desk; four would’ve been a much better choice.

The included sponges, in black and yellow, are housed in a circular dish, and a few squirts of distilled water brought my test sponge to life. They work well enough, but I much prefer brass wool to clean the tip.

Also in the kit was a solder sucker, which looks similar to my Engineer-branded model. I did not use the solder sucker, as it was not pertinent to the review, but it was nice to have.

Soldering with the Fanttik T1 Max

I connected the Fanttik T1 Max soldering iron to my Pinepower desktop station and, using a USB Type-C lead, I charged the unit ready for a morning of soldering. I expected the unit to charge using USB Type C PD, but my unit charged at 5V 1A until it was full.

I selected a small knife-edge soldering tip from the pop-out tray, and I unscrewed the collet and slid the soldering iron tip until there was resistance. I screwed the collet in place and then powered up the soldering iron. A mere 12.14 seconds later, it was up to my working temperature (350 degrees C), and I was ready to go. I really like the auto-sleep function, which drops the tip temperature if the soldering iron’s vibration sensor doesn’t detect movement for 15 seconds. If left for a full 15 minutes, the soldering iron will shut off.

My goal for this test was to solder up an RC2040 Computer Kit from Extreme Electronics. This kit uses a Raspberry Pi Pico to emulate an RC2014 Z80 type system running CP/M-80. I have a thing for old computers and soldering, so the two married nicely.

The kit comprises through-hole and surface-mounted components. Through-hole is no problem for my aging eyes, but surface mount is a challenge for both my eyes and dexterity. Let's see how the Fanttik T1 Max holds up.

Through-hole parts were a breeze. My choice of leaded solder melted easily at 350 degrees C, and I soon had all of the shorter components on the board. Next up, I had to solder in the micro SD card slot and the Raspberry Pi Pico. I powered down the soldering iron and left it to cool. In 2 minutes and 10 seconds, the tip was cool enough for me to remove and replace with the hooked point. Heating the iron back to 350 C, I tinned and prepared the new tip, then started soldering the surface-mount components.

I started with the micro SD card slot and, surprisingly, it went perfectly! No bridges or cold joints, just finely soldered connections to the PCB. The Raspberry Pi Pico was a different story. The Pico has castellated edges for surface-mount soldering, and these are sometimes difficult to solder. This was one of those times. With the hooked tip, it was difficult to get heat into the castellated edge and the PCB pad. I managed it, but after a while, I swapped back to the knife-edge tip to give myself a little more thermal mass, but the tip would occasionally get stuck in the solder. This typically means that the tip has lost temperature and needs a moment to heat up.

This happens with battery-powered soldering irons. The power supply won’t just dump power into the tip until the battery dies. Instead, it regulates the tip temperature, but sometimes what we demand is not what it can constantly provide. For my own peace of mind, I had to check the connections between the castellated edges and the PCB pads. I didn’t want to go to all of this effort for the kit not to work. So I grabbed my trusty UNI-T UT58C multimeter and beeped out (continuity tested) the connections. Some were bad, so I reflowed them and retested. After that, all was good.

I finished off the rest of the connections, and the kit was built and, yes dear reader, it worked!

Next, I wanted to test the Fanttik T1 Max with a larger solder joint. So I dug into my maker box and found some 3mm (AWG 9) thick marine-grade copper wire that I got when a local store closed down. This is the kind of cable used to supply power, so it has multiple strands of pure copper running through it, making it tough to solder. Many makers just use mechanical connections or a really beefy soldering gun. Can the Fanttik T1 Max solder this cable? In a word, “barely.”

There was just enough power at 350 degrees C to get the wire to accept solder, so I bumped the temp up to 450 C and things went better, but it still took a very long time to even tin the copper. Joining two wires together was frustrating, and I never had enough power to make them stick. I retested with the Fanttik T1 Max connected to a USB-C power source, and the result was the same.

For precision soldering on circuit boards etc, the Fanttik T1 Max is a good option, but for beefier solder joints, you’ll need a more powerful soldering iron like iFixit’s Smart Soldering Hub.

Fanttik T1 Max User Interface

As I mentioned earlier, the only visual feedback on the Fanttik T1 Max is the power button LED. It has three colors, red, green, and white, identifying the soldering iron’s current status.

I like this, but sometimes I had to refer to the instructions to understand what was happening. After using the soldering iron for a while, I got a grasp as to what was going on, but it felt awkward.

One thing that didn’t feel awkward was the dial used to control the temperature. This was so simple to use, and luckily it had my preferred working temperature (350 C), but what if I preferred 315C or 360C? Well, I am out of luck as the dial uses 50-degree increments, so I would have to adapt my soldering speed to match the heat of the tip.

Fanttik T1 Max Soldering Tips

The C210 soldering tips are very different to those used on the TS100, Pinecil, and even my faithful Yihua soldering station. But they look similar to those used on the WEP 982 III soldering station and the Fnirsi HS-02 smart soldering iron. I checked, and yes, standard C210 type soldering iron tips can be used. That makes finding alternatives / replacements on Aliexpress or elsewhere very easy (and affordable).

Bottom Line: Who is the Fanttik T1 Max For?

The design of the Fanttik T1 Max makes it look and feel like a lifestyle product, for those who want to solder something but have never picked up a soldering iron before. The performance backs this up, as it only has power for small / light jobs. If you have a big soldering project in mind, then you should buy a more powerful kit.

I enjoyed using the Fanttik T1 Max. It is a great soldering iron for lighter tasks, but its size and uneven weight make it feel a little too “alien” in the hand. That said, the simplicity it affords means that it suits those new to soldering, or for quick and simple soldering projects.

But when Asus teamed up with Xbox on the Asus ROG Xbox Ally X ($999.99 as tested), it didn't just mean new hardware with the AMD Ryzen AI Z2 Extreme chip and comfortable grips; it also meant a full-screen Xbox mode that is easy to navigate with a controller and even hides some of the components from Windows 11 that can slow down gaming. It doesn't have Steam Deck levels of polish, but it's a revelation compared to the standard Windows desktop.

But that Xbox menu hides PC levels of customization and complexity. Me? I loved having my Steam and Epic libraries next to games from my Xbox collection. (Or at least, the ones that support Xbox Play Anywhere.) But to Xbox purists or those who haven't dipped a toe into PC gaming before, I think this might be frustrating, and will raise questions about what an Xbox is and what the future of Xbox looks like.

All in all, the ROG Xbox Ally X feels far more like an Asus gaming handheld than an Xbox one. But outside of tremendously comfortable ergonomics, it's Microsoft's software engineering that makes the Xbox Ally X one of the best PC gaming handhelds around. That is, at least until Microsoft makes the updated operating system available on all Windows 11-based gaming handhelds.

Design of the Asus ROG Xbox Ally X

The Xbox controller is effectively the default controller in PC gaming (other than a mouse and keyboard). So it's honestly a bit surprising it's taken this long for a major PC company to copy that gamepad's comfortable prongs and adapt them to a gaming handheld. It worked for the PlayStation Portal!

Otherwise, though, this Ally looks much like the 2024 Ally X that came before it, with an angular design, plenty of ROG branding, and dedicated buttons for Asus' Armory Crate. Other than an Xbox button, which Windows 11 handhelds have desperately needed for a while now, there is nothing that makes this feel like an Xbox based on Microsoft's current console designs (although there is a repeating ROG XBOX pattern in tiny print on the device). If Asus had released this design without calling it an Xbox, I'd be none the wiser. It has a blacked-out color scheme with hints of red on the buttons. I would've liked to see the Xbox-colored buttons on the vanilla ROG Xbox Ally.

The prongs feel great. This is the most natural a handheld has felt in my hand so far since the Steam Deck. These grips do make the system wider than some other systems, however, which makes it a bit bulky for travel. I was surprised at how good the system felt in my hands. It's wide in a way that makes it look like it would be awkward, but the weight is distributed very evenly, and I was comfortable even for long play sessions. It really does feel like you're using a controller.

Other than that, the general look of the system is almost identical to the ROG Ally X from 2024. You get a very classic Xbox layout here, with ABXY buttons, offset joysticks, and a D-pad sitting on a disk. These are not, however, identical to what you see on an Xbox wireless controller, with all of the Asus ROG styling. I love the rubberized coating around the edges of the joysticks' caps (that reminded me of an Xbox controller), and the buttons are clicky and tactile, if a bit loud. Each joystick is surrounded by a moat of RGB light, which you can customize in Asus's Armoury Crate software. Unlike the Lenovo Legion Go 2, there's no sign of Hall Effect joysticks here.

There are standard bumpers and triggers, though the new Xbox-branded Ally systems use what Xbox calls "impulse" triggers with motors for vibration. Asus notes that these are Hall Effect triggers, so at least they're using that tech somewhere! The bumpers feel a bit cheap for my liking, with clicky plastic, and the trigger vibration, though kind of fun in racing titles like the Forza Horizon series, isn't terribly powerful.

There are just two macro buttons, one on each side of the rear of the device. That's fine for me, but some people may prefer the Legion Go 2's three buttons or the four on the back of the Steam Deck.

The Xbox button is to the right of the left thumb stick. This is as good a place for it as any, and never got in my way. On the left side of the screen are buttons for Armoury Crate and the View button, while the right side features a button to bring you to your library (or, with a long press, call up Copilot) and the standard Menu button.

What there isn't is a touchpad. If you want one, you're better off looking at the Steam Deck. I like touchpads, but their absence here is far from a dealbreaker for me, and I didn't really miss them.

On top of the system, there are two USB Type-C ports: one with USB4 and the other using USB 3.2 Gen 2. The USB4 port is Thunderbolt 4 compatible, and I had no issue plugging it into a Thunderbolt dock to play games on my monitor. There is also a microSD card slot, a 3.5 mm headphone jack, and a fingerprint reader built into the power button. All of the ports are on top.

Like the previous Ally, there are vents in the back of the system between the macro buttons. That brings cool air in to chill the processor, which is then ejected out of the top of the system.

The Xbox Ally X weighs 1.58 pounds, which feels substantially lighter than the Lenovo Legion Go 2’s 2.03 pounds. It's a slight bump over the original Ally X (1.49 pounds), while the Steam Deck OLED is lighter than its Windows-based counterparts (1.41 pounds).

If you include the prongs (and the prongs are permanently attached, so you have to), the Ally X measures 11.42 x 4.76 x 2.00 inches, though it's 1.08 inches deep at its thinnest point. That means that it's technically thicker than the Lenovo Legion Go 2 (11.64 x 5.38 x 1.66 inches), though it doesn't feel as hefty.

The original Ally X measured 11.02 x 4.37 x 1.45 inches, so this iteration is larger all around, while the Steam Deck OLED is still the widest option at 11.73 x 4.6 x 1.97 inches.

The system comes with a small cardboard stand in the box. I'd rather it come with a carrying case, like the Steam Deck and Lenovo Legion Go 2. Instead, you have to pay an extra $70 if you want an official case for this handheld.

Asus ROG Xbox Ally X Specifications and Components

Asus is using the AMD Ryzen AI Z2 Extreme. This chip is almost identical to the regular Ryzen Z2 Extreme we saw in the Legion Go 2, except that it has a 50 TOPS NPU on board, as well as three Zen 5 cores and five Zen 5c cores. The Z2 Extreme has a max boost clock of 5 GHz on the Zen 5 cores, (3.3 GHz on the Zen 5c cores) and a base clock of 2 GHz. That max boost is actually 0.1 GHz slower than the Z1 Extreme.

The Z2 Extreme goes up to 35 watts with its configurable TDP. Here's how Asus uses that with its performance modes, with comparisons to previous models:

Whether you need an NPU in a gaming handheld is debatable. I tried it with Microsoft's Gaming Copilot, but considering it was often wrong and unhelpful about what to do in games, the NPU is more for upcoming features regarding AI-powered upscaling.

There's a 16-core GPU using integrated Radeon RDNA 3.5 graphics. Asus has opted for 24GB of RAM, the same as in the original Ally X but less than the 32GB in the Legion Go 2. The 1TB M.2 SSD is in a standard M.2 2280 form factor. And the 80 WHr battery is identical in size to the cell used in the original Ally X.

Gaming and Graphics Performance on the Asus ROG Xbox Ally X

The Ryzen AI Z2 Extreme has a 16-core integrated GPU with RDNA 3.5 graphics. The CPU and GPU have a shared memory pool, which defaults to 8GB of memory assigned to the GPU in Asus's Armoury Crate SE software. You can customize this, but we ran it at the out of the box settings.

By default, the system runs in a 25W performance mode while unplugged and a 35W turbo mode while plugged in. Again, we stuck to these values in our testing. We ran games at both 720p and 1080p. In our comparisons, the Legion Go 2 and Steam Deck were tested at 1280 x 800 and 1920 x 1200 to fit their 16:10 screens.

Using the Xbox Full Screen Experience should free up some memory and provide performance advantages, as it doesn't load some Windows components. We benchmarked games through the FSE, with the exception of our stress test, which requires additional logging software.

In my spare time with the Xbox Ally X, I played Marvel's Midnight Suns, mostly on battery power. At 720p and medium settings, the game ran between 75 and 111 FPS in missions, though when exploring the Abbey, it was sometimes closer to 60 FPS.

I also tried Alan Wake 2, which I couldn't run on handhelds with the Z1 Extreme. Here, it worked, albeit with some serious caveats. With the screen at 720p and using a 640 x 360 performance render resolution with FSR2 upscaling, the game ran between 32 and 53 FPS is a nightmare version of New York. Down in the subways, away from the open world level, the game hit the 60 FPS cap I had invoked to save the battery. But the game generally looked terrible and blurry from the low resolution.

On the Shadow of the Tomb Raider benchmark (medium settings, DX12), the Xbox Ally X hit 64 FPS at 720p plugged in and 79 FPS unplugged. At 1080, those boosted to 45 and 52 FPS, respectively. Those results are similar to the Legion Go 2, and show a nice but not extraordinary bump over the Asus ROG Ally X.

Cyberpunk 2077's Steam Deck preset showed off the system's power when plugged in, where it hit 63 FPS at 720p and 40 FPS at 1080p. Those both beat the Legion Go 2, though that had the disadvantage of having to push a few more pixels to power its 800p screen.

Red Dead Redemption 2 (favor performance, Vulkan), the Xbox Ally X showed off gains at 720p while unplugged, hitting 66 FPS. It scored almost identically to the Legion Go 2, but showed upticks over the original Ally X all around.

Borderlands 3 (medium, DX11) was the only misstep. At 59 FPS at 720p unplugged and 68 FPS plugged in, it came in behind the Legion Go 2 at 800p. 1080p performance, however, was comparable. In fact, at 720p, the Xbox Ally X had effectively no gains over the original Ally X in this game, though 1080p performance was improved.

Now that we've seen handhelds getting slightly more power, we've upped our stress test. It's still Metro Exodus running 15 times at 720p to simulate about half an hour of gameplay, but we've moved from low to medium settings. The Xbox Ally X achieved an average frame rate of 63.44 FPS on the benchmark.The consistency across the 15 runs suggests the cooling is adequate for the chip.

The Xbox Ally X's three Zen 5 cores ran at an average of 3.24 GHz during the stress test, while the Zen 5c cores hit 2.15 GHz.

Windows 11 and Armoury Crate on the Asus ROG Xbox Ally X: Is it an Xbox?

The Xbox Full Screen Experience is significantly better for handhelds than standard Windows 11.

You boot straight into a version of the Xbox app featuring a home screen with recently played games, though like a standard Xbox, the app also advertises game deals, Game Pass, and has curated lists of games. The most important part is your library (there's even a hardware button for it), which brings everything together from Xbox and launchers like Steam, Epic Games, Battle.net, GOG Galaxy, and Ubisoft Connect.

Wherever you are in the system, you can see a miniaturized version of this experience by hitting the Xbox button, which shows recently played games, launchers, quick settings, your friends list, and most of the features from Xbox Game Bar on PC.

Most importantly, loading directly into the Full Screen Experience means that Windows doesn't even bother loading certain components of the OS that aren't used for gaming, making it lighter, saving RAM, and putting more resources toward games. The handheld even came with Steam preinstalled, booting right into big picture mode. That's a nice welcome mat for PC gamers, though I later did end up in the traditional Steam interface upon further boot ups.

Launchers, however, don't always work well. For instance, if you go into Steam's Big Picture mode, you'll get a great experience with the Xbox controller. But if you end up in Epic Games, as I did after quitting games like Borderlands 3 and Marvel's Midnight Suns, I found myself in an environment where I had to use this PC-style launcher with touch, as there was no mouse control. Even then, when I had the system at 720p, I'd have a gray box around the launcher, making it even harder to hit the touch targets. I had similar issues launching Cyberpunk 2077 from Steam, which forces you to go through CD Projekt Red's launcher. When you pick a game from the Xbox menu, launchers still have to boot and run in the background, which sometimes leaves you with a minute or so of hangtime. You don't have these problems when playing games from the Xbox Store.

Games from the Xbox Store also get other bonuses, like nicer artwork and, if you buy from the Xbox Store, analysis of how well the game will play on your PC.

You can still boot into the desktop (and load all of those extra Windows components upon doing so). If you want to use this system as a standard Windows 11 PC, there's nothing stopping you. It has all of the flaws of other gaming handhelds running Windows, but I'm glad the option is still there. I'm sure there will be at least some people who use the handheld with a keyboard and mouse or just want to tinker.

There are a number of waypoints in and out of the Full Screen Experience, but they feel bolted on to the Xbox app. But the sheer number of ways to get back into it means most people won't get lost. If you do go to full screen Windows 11, it will load the rest of the OS components, likely affecting gaming performance, so you'll want to restart when switching between the Xbox FSE and Windows (the system encourages you to do so).

And there are still aspects of Windows you can't control. Install a game or launcher that Windows runs through User Account Control? UAC doesn't pop up on an Xbox, but it still happens here. Asus also includes some unnecessary software, like MyAsus and GlideX for screen sharing, which feel out of place on a gaming handheld.

One of SteamOS's biggest advantages is a rating system that gives you an idea of how well a game works on the Steam Deck. Microsoft is doing something similar in the Xbox FSE, called the Handheld Compatibility program. Games are labeled as "Handheld Optimized" with a green checkmark, or "Mostly compatible" with a blue icon, suggesting you need to change some settings to get the game to run well. These badges are buried in the Xbox Store, so you won't see equivalents for other launchers. But Microsoft does keep a list of games it and studio partners have tested for handheld compatibility.

Asus's Armoury Crate SE is still here, both as a full application and as a smaller "Command Center" built into the Xbox interface and available through Windows 11. It's important as you switch through performance modes, change wattage to the processor (which can greatly affect battery life), and make other adjustments. (It is nice to have it baked into the Xbox FSE, and accessible easily through the Xbox button). The full Armoury Crate SE software does double up on some features though, including its own unified game library.

All of this leads to one question: Is the new Ally X an Xbox? For almost a year, Microsoft has been pushing a campaign that anything can be an Xbox, from laptops to car entertainment systems to phones – and now, gaming handhelds running Windows.

If you're coming from using an Xbox Series X or Series S, though, this is a fundamentally different experience in many ways. While Xbox Play Anywhere features over 1,000 games and lets you share save states and achievements, it also assumes you bought those games from the Microsoft Store. And plenty of games don't support Play Anywhere, including some of Microsoft's own. (It's nice to see that Doom: The Dark Ages got support after Doom Eternal didn't).

Microsoft needs to get to a point where all Xbox games are playable anywhere with this support to compete with the user experience Valve offers on Steam Deck. It's tough to call something an Xbox when it doesn't play every Xbox game. That will certainly confuse some people. Xbox gamers will also get their choice of using Game Pass or using Remote Play from their console, so there's a lot of flexibility here.

But as a PC handheld, this system also supports Steam, Epic Games, and every other launcher under the sun. All of those launchers, and all of the games within them, have far more settings to tinker with than ever before. Add in the fact that Asus' software gets deep into TDP, and this is a system with far more options than any Xbox owner ever had before. Enthusiasts will love that. But I also think there will be a number of console devotees who wonder how the heck this thing is an Xbox. It sure feels a lot more like an Asus gaming PC.

Microsoft is promising a bunch of upcoming features early next year, but we can only judge the Ally X on what's available today. Promised updates include automatic super resolution to use the AI Z2 Extreme's NPU to upscale games at the system level with no changes needed in game, as well as a game save sync indicator to make sure your save is stored, and improved library management for SD cards, making it easier to store and manage your games from the FSE.

Even without those, though, the Xbox FSE makes the Xbox Ally X the best PC gaming handheld with Windows because it's so much easier to use, despite the fact that Microsoft clearly has work to do to make launchers less janky and make the whole thing more full-featured. None of this is as polished or as easy as Valve's SteamOS, but it's significantly closer than using the regular Windows 11 UI. I can't wait for Microsoft to let every gamer with a handheld get a better experience than they do on the standard Windows 11 desktop.

Display on the Asus ROG Xbox Ally X

Asus is using a 7-inch, 1920 x 1080, IPS touchscreen that runs up 120 Hz on the Xbox Ally X. If that sounds familiar, it's because it's probably the same screen on the previous Ally X. The 16:9 aspect ratio differs from what Valve and Lenovo have done, with taller 16:10 screens that also use OLED technology. In fact, this screen doesn't support HDR.

In general, the screen looks good. In Marvel's Midnight Suns, the Abbey is filled with colorful characters. Agatha's altar includes many blue flames (and, oh, yeah, Agatha Harkness's blue ghost), which created a haunting effect that contrasted against greenery at night. It doesn't, however, look anywhere near as vivid as the Lenovo Legion Go 2 I just tested.

The ROG Xbox Ally X's screen tested almost identically to the ROG Ally X we tested last year. Given they have the same specs, they're almost certainly same panel. The Xbox Ally X's panel covers 110% of the DCI-P3 color gamut and 77.9% of the more challenging DCI-P3 space by volume. That's easily beaten by the OLED screens on the Legion Go 2 and the Steam Deck OLED.

At 515.8 nits of brightness, this panel is more luminous than the Legion Go 2, though the Steam Deck OLED took the crown at 597.2 nits.

Battery Life on the Asus ROG Xbox Ally X

Battery life on a handheld, like its competitors, is largely dependent on your settings and the game you play. I tend to play games in a way to try to preserve battery life as long as possible. If you play low-impact games like Stardew Valley or even Hollow Knight: Silksong, you may experience longer battery life than the effects-heavy games I tested.

One evening when testing the Xbox Ally X, I played some Marvel's Midnight Suns with the system in Performance mode. At medium settings at 720p with the screen locked to 60 Hz and 50% brightness, the battery had dropped to 50% after an hour and 32 minutes.

When I played Alan Wake 2, I kept the same 17W Performance mode, 50% brightness, and 60 Hz refresh rate. At 720p using FSR2 upscaling on a performance-focused 640 x 360 render resolution and low settings preset without ray tracing, the handheld hit 50% in an hour and 35 minutes.

If you extrapolate from these, those suggest about 3 hours on battery, and that's with the limitations I put in place. That seems about standard when compared with the Legion Go 2.

Audio on the Asus ROG Xbox Ally X

The twin speakers on the Xbox Ally X are front facing, playing the sound right toward you. This is the optimal location, similar to what you find on the Steam Deck OLED. Some other handhelds have the speakers facing out of the top, away from you.

I used some of my time with the ROG Xbox Ally X to play a bit of Tetris Effect, which I had through Xbox Play Anywhere. The game sounds best with headphones, but the speakers did a solid job. The "Prayer Circle" level in journey mode features bells and xylophone sounds, which were clear and focused. Some deeper bass sounds, like gongs, were muted at best, however.

In Marvel's Midnight Suns, the speakers swelled with orchestral music during a battle, and kept the dialogue clear even over whirring Hydra helicopter blades. And the game got nice and loud, too, easily filling a whole room.

Heat on the Asus ROG Xbox Ally X

We measured skin temperatures while running our Metro Exodus stress test. Like other handhelds we've tested, the design ensures your hands don't touch any hotspots. The prongs and controls ran from 84 degrees Fahrenheit at the furthest point to 90 F closer to the screen. On the top, the hottest point was at the exhaust vent, which reached 113.7 F. On the rear, the hottest point was near the intake vent, far enough from the two back buttons that you wouldn't place your hands there.

Internally, the CPU package measured an average of 74.7 degrees Celsius, while the integrated graphics reached 70.59 C.

Upgradeability of the Asus ROG Xbox Ally X

Here's one way in which the ROG Xbox Ally X is definitively more like a PC than an Xbox: You can crack it open and upgrade it. The back of the system is held on with eight Phillips-head screws. There are five screws on the back and three on the bottom. Of those, the top middle screw on the rear is captive; the rest all come out.

You'll have to pry carefully from the top and all the way around the sides of the controllers to get the back off. It can be nerve-wracking and time-consuming, as there are a lot of plastic clips to disengage. But go slow — there's a ribbon cable connecting the back to the motherboard to control the M1 and M2 buttons. I would love to see a design in the future with a door that leads directly to the SSD (perhaps similar to what Microsoft has done on some Surface designs) so you don't have to deal with so many screws and tabs. For those without any DIY PC experience, adding storage with the microSD card slot may be the way to go.

The major component you can upgrade is the standard M.2 2280 SSD, which is right at the center when you open the system. The battery cable runs over the SSD, and while you'd probably want to disconnect that before doing work anyway, you'll have to do so to have room to remove the existing drive.

Like the original Ally X, you can see the modular thumbsticks. Since Asus isn't using Hall Effect joysticks, these will probably, eventually, be prone to drift. Asus didn't sell replacement sticks for the initial Ally X, but I'd love to see them do it for this system.

Asus ROG Xbox Ally X Configurations

The Asus ROG Xbox Ally X comes in a single configuration, with a 7-inch, 1080p display, AMD Ryzen AI Z2 Extreme processor, 24GB of RAM, and a 1TB SSD. It costs $999.99, up from $799 on the 2024 Ally X with a Z1 Extreme. That's a big jump for the new processor and slightly new design.

There is another model, the Asus ROG Xbox Ally, which ditches the Xbox Ally X's blackout color scheme for a white controller and colorful buttons. That system, which costs $599.99, uses an AMD Ryzen Z2 A processor, which on paper is similar to what you can find in the Steam Deck OLED. It also has 16GB of RAM and a 512GB SSD.

Bottom Line

The Asus ROG Xbox Ally X makes two things very clear:

1. More handhelds need controller-style prongs.

2. Windows 11 on gaming handhelds has a chance.

The Xbox Ally X has superior ergonomics to every other Windows handheld on the market, and will be more comfortable for some than the Steam Deck, which until now has been the best in this area. It looks a little silly, but controllers have those grips for a reason. Add in some great weight-balancing, and you have a comfortable gaming handheld. Too bad it doesn't come with a case to carry it around in.

The performance of the Ryzen AI Z2 Extreme is strong, but I'm not sure it's such a jump that if you already have an Asus ROG Ally X with Z1 Extreme, you need to upgrade.

At $999.99, this is a pricey handheld. Still, there are more fully featured options out there. If you're willing to skip the Xbox Full Screen Experience (for now), the Lenovo Legion Go 2 adds detachable controllers with Hall Effect joysticks and a gorgeous OLED screen, along with more RAM. But that device is bigger and heavier, and is even more expensive at $1,349.99. The Steam Deck OLED, which is cheaper at $549 for 512GB and $649 for 1TB, also has an OLED screen, though its chip quite as powerful as the Z2 Extreme.

For now, the Xbox Full Screen Experience is exclusive to the Xbox Ally X. All other things being equal, that's the big reason to buy this system. Using it is simply easier, even if Microsoft has plenty of rough edges to clean up.

If you're just now jumping into the gaming handheld market and want something that just plays games and runs Windows, this is the one to get. It's comfortable and is finally starting to get close to the ease of use that Valve offers with SteamOS. When the Xbox Full Screen Experience becomes available more widely, however, the field may blow open.

Erasing the line completely is Samsung’s M9 M90SF Smart Monitor. It features DisplayPort with a refresh rate of 165 Hz and Adaptive-Sync for gaming. On the TV side, it has HDMI, Wi-Fi, and Samsung’s TV Plus Streaming interface. Both sides benefit from a 4K OLED panel with wide gamut color, HDR10, and HDR10+. It truly is a desktop monitor and personal TV in one chassis. Let’s take a look.

Samsung M9 M90SF Smart Monitor Specs

Is the M9 more TV than monitor? Or is it more monitor than TV? That’s what I’m aiming to find out. On the gaming side, it has 165 Hz, Adaptive-Sync, and DisplayPort. TV watchers will appreciate the Wi-Fi with an evolved streaming interface called Samsung TV Plus. It also comes with a remote control befitting a TV, featuring dedicated buttons for Netflix, Prime, and YouTube. And it incorporates a microphone for Google Assistant and Amazon Alexa. The M9 also includes Bluetooth and Apple AirPlay 2. The only thing missing is an antenna input.

Both sides benefit from a gorgeous 4K OLED panel with 138ppi pixel density. The color gamut covers around 97% of DCI-P3, which is a bit less than other QD-OLEDs I’ve reviewed. But there is plenty of light output with over 450 nits peak available for HDR10 and HDR10+ formats. If you’re wondering about the plus, HDR10+ is like HDR10 but with dynamic metadata like Dolby Vision. It requires HDR10+ content, which can be found on some Blu-rays and streamed titles. The M9 also features dynamic tone mapping, which can enhance HDR10 material.

Convenience features abound, starting with a 4K webcam incorporated into the screen bezel. USB ports are included too, with type A and type C accompanying the HDMI 2.1 and DisplayPort 1.4 inputs. A pair of internal speakers delivers five watts of power each.

For those concerned with panel life, the M9 features a heatpipe system to keep temperatures down, along with logo and taskbar detection, which dims those portions of the screen to minimize pixel wear. A refresh routine can also be run periodically. Samsung backs the M9 with a three-year warranty.

This is a premium package at $1,599, but as you’re getting the best of three displays in one, it could be considered a bargain. It’s well-suited for small spaces or users who want a single do-it-all panel in their workspace.

Assembly and Accessories

The M9 is securely packaged in crumbly foam and comes in three parts: base, upright, and panel. You’ll need a Phillips-head screwdriver to secure two captive bolts between the metal base plate and the substantial upright. The stand is monolithic when assembled. The panel snaps in place. Samsung includes a 100mm VESA mount for arms and brackets, but no fasteners. The power supply is an enormous brick with a cord long enough to leave it on the floor. You also get USB-C and HDMI cables.

Product 360

The M9’s styling is the definition of minimalist. There are no textures to be seen except for a small grill visible only in the side view. This vents heat from the heatpipe system and keeps the panel running cool at all times. Heat dissipation is further aided by the panel’s all-metal back cover. Everything else is smoothly finished in silver plastic, except for the thin base, which is a metal plate. The upright is wide and strong with solid ergonomics, 25-degree tilt, 4.6 inches of height, and a 90-degree portrait mode.

The screen has a matte front layer, which effectively keeps ambient light at bay. It’s a nice departure from the glossy screens most OLEDs have. I could not see any downside as the image was sharp, colorful, and free of visible grain or artifacts. The bezel is wider at the top and bottom than at the sides. And you can see the lens for the 4K camera at the top.

The M9’s remote is handy for TV watching with a navigation pad in the middle and dedicated buttons for Netflix, Prime, YouTube, and Samsung TV Plus. The house icon takes you to the streaming interface regardless of which input is currently in use. It has a microphone for use with Amazon Alexa and Google Assistant. The only thing missing is an input selector.

The input panel faces rearwards, making it easy to access and to feed the cables through the hole in the upright. There is one each of HDMI 2.1 and DisplayPort 1.4, along with a USB-C (90 watts charging) and two USB-A downstream ports. The power cord plugs in with a right-angle connector and goes to a large external power supply. Luckily, the cord is long enough to put the brick on the floor.

OSD Features

I’ve commented in the past about Samsung Smart Monitors’ confusing menu systems, which attempt to bridge the gap between TV and computer monitor. The M9 is better than its predecessors but still has some room for improvement.

The panel features a joystick located around the back right side, but it’s far easier to use the remote for menu and streaming navigation. Most of what you’ll need is in the Picture menu, where you’ll find image modes and calibration controls. Expert Settings has a complete set with gamma, two- and twenty-point white balance, color management, video processing, selectable gamuts, and picture enhancements. I noted that the only mode with accurate color was Movie, which meets the BT.709 standard for SDR and DCI-P3 for HDR.

The Game menu features an Adaptive-Sync toggle, aiming points, and a mini-map detector, which is part of the M9’s panel protection suite. Another game-focused menu can be called up from the remote, which I’ll discuss in a moment.

The M9’s audio output includes left/right balance and a multi-band equalizer. Auto Volume compresses the dynamic range for better clarity at low volumes.

Networking is an integral part of the M9’s function, with both Wi-Fi and Bluetooth included. Apple AirPlay 2 is supported along with Bluetooth streaming. These functions are managed with Samsung’s SmartThings app.

Pressing and holding the play/pause button on the remote brings up the game bar, which provides quick access to play aids like the aiming point and sound settings. It also shows signal information and status at the bottom.

Samsung M9 M90SF Calibration Settings

The M9 ships in Eco mode, which is quite dim and will likely only be useful in a darkened room. I measured all the modes and found that Movie came the closest to accurate color standards. It renders BT.709 for SDR and DCI-P3 for HDR. If you want the larger gamut for SDR, choose Native from the color space options. Grayscale tracking is visually perfect, but the indicated BT.1886 gamma is lighter than spec. I used the slider to make it better. My SDR settings are shown below.

For HDR, you can choose from the same picture modes, but a couple of additional settings become active. For best results there, set HDR Tone Mapping to static and Peak Brightness on Medium. This will work well for HDR10 content, while HDR10+ material will automatically be tone-mapped according to its dynamic metadata. If you want brighter midrange tones, set HDR Tone Mapping to Active.

Gaming and Hands-on

I spent a few days putting the M9 through its paces as a gaming monitor, TV, and workday tool. Gaming was the first order of business, and there, I liked it very much. I had noted a little motion blur during testing, which is solely due to its 165 Hz refresh rate. This was harder to spot when mowing through aliens in Doom Eternal. I’m definitely spoiled by speedy screens, especially in this category where there are many 240 Hz models. But the M9’s 165 Hz got the job done with low enough input lag and flawless Adaptive-Sync.

The HDR image was bright and colorful, whether static or moving. I experimented with luminance settings and settled on Static tone mapping and Peak Brightness on Medium. I also set Contrast Enhancer to Medium for a little extra punch in the midtones. This can be used for both HDR and SDR if you wish. I noted that SDR color defaulted to BT.709 while HDR used the M9’s full native color space, which covers around 97% of DCI-P3.

Working in my go-to apps like Photoshop, Excel, and Word, the M9 proved itself capable of getting through productivity tasks with ease and comfort. The picture is superb with no visible pixel structure, deep contrast, and bright color. The screen’s anti-glare layer enhances contrast by successfully filtering out all ambient light. It’s optically pure and more forgiving than the glossy screens found on most OLEDs. You won’t need to be as thoughtful about placement with the M9.

Moving to streaming, I browsed a few titles on Netflix, Disney+, and AppleTV+, which were all available from the main screen that appears when you press the Home button on the remote. The video quality of HDR titles from Netflix, such as Formula One Drive to Survive, was exemplary. I noted truly excellent audio as well. Though the M9 has just two internal speakers, they play a wider range of frequencies than typical monitor speakers and with a wider sound stage. There isn’t a ton of bass, but there is clarity and balance. The dialogue in TV shows like Severance was easy to understand and put just forward enough in the mix that I had no need to adjust the volume more than once.

My only complaint is the menu system. It attempts to adapt itself to each format, showing different options for gaming, desktop, and TV. While it is more integrated than Samsung’s previous efforts, it’s still somewhat confusing. I would prefer to see the same options and picture modes shown in all facets of the M9’s operation. I’d also love to see an input selector button on the remote rather than having to hunt through the OSD to change sources.

Takeaway: Gaming monitor, check. 4K HDR TV, check. Productivity display, check. The M9 does everything well. Even though there are faster gaming monitors, you won’t need that extra speed unless you’re a skilled player. For gamers of average ability, it’s smooth and responsive enough for addictive play. It’s also one of the best desktop TVs I’ve experienced, with a gorgeous picture, perfect integration of streaming apps, and easy control with the included remote. The M9 is a premium purchase, but it is three displays in one, so from that perspective, it’s a bargain.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

To compare the 165 Hz M9’s performance, I’ve rounded up a selection of 32-inch 4K OLEDs. Most of these premium screens run at 240 Hz, like the Asus XG32UCWMG, PA32UCDM, Dough’s Spectrum Black 32, and HP’s Omen 32 OLED. The Dell S3225QC is less focused on gaming and runs at 120 Hz.

Pixel Response and Input Lag

Click here to read up on our pixel response and input lag testing procedures.

Panel response times are in lockstep with refresh rates. The 240 Hz screens all draw a full white field in 4ms and are visibly smoother in operation than the Samsung or Dell monitors. This can be observed in moving test patterns and in content, where the slower panels exhibit a slight motion blur. It is far less than you’d see from an LCD running at 120 or 165 Hz, but it’s more than the 240 Hz displays that are visually perfect.

The M9 is quick but not super-quick. You’ll give up around 6ms of input lag, which for most players won’t be an issue. It didn’t affect my experience, but more skilled players might prefer a faster monitor. It should be noted that the hardware requirements for 240fps in 4K are substantial. I test with a GeForce RTX 4090 and can barely get there. If you have a lesser card, it can’t take advantage of a faster monitor.

Test Takeaway: The Samsung M9 is a solidly capable gaming monitor, but not as quick as other screens in its price class. In fairness, the extra cost goes into smart features and streaming, which typical gaming monitors lack. As an all-around entertainment display, nothing else is as full-featured.

Viewing Angles

The M9 boasts superb off-axis image quality, a hallmark of OLED technology. The side view is visually identical to the head-on photo with no change to color, brightness, or gamma. The top view has a slight red tint with lowered gamma and brightness.

Screen Uniformity

To learn how we measure screen uniformity, click here.

My M9 sample posted one of the better results I’ve seen here. 4.83% is well below the visible threshold of 10%. Historically, OLED has been the most consistent technology for achieving screen uniformity, with no tendencies toward glow or hot spots.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

To read about our monitor tests in-depth, please check out Display Testing Explained: How We Test PC Monitors. We cover brightness and contrast testing on page two.

Uncalibrated – Maximum Backlight Level

The M9 doesn’t include variable brightness for SDR content. That means full white field patterns measure the same as 25% windows. Samsung rates the panel at 250 nits, and I confirmed that with a 260-nit result. There is a Contrast Enhancer option in Expert Settings that tweaks dynamic range, but it doesn’t increase peak output. Black levels and contrast are unmeasurable in any case.

After Calibration to 200 nits

Calibration was a simple matter of adjusting the peak to 200 nits and setting gamma to BT.1886 +1. This did not change black levels or contrast, which were still unmeasurable. The ANSI test yielded the same result.

Test Takeaway: The M9 delivers the same infinite contrast I’ve observed from all OLED displays. Black levels cannot be measured, so the difference is only in peak output. That is also about average for monitors with constant brightness at just over 260 nits. This is the same excellent performance that you will find from any OLED monitor.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

The M9 has only four image modes, but three of them alter color significantly from the BT.709 and DCI-P3 specs. Only Movie offers the right balance of grayscale, gamma and color accuracy.

Grayscale and Gamma Tracking

Our grayscale and gamma tests use Calman calibration software from Portrait Displays. We describe our grayscale and gamma tests in detail here.

Eco mode proved too dark for my tests, so I ran my default measurements in Standard. Grayscale tracking is excellent with no visible errors. The gamma setting is called BT.1886, but my test showed it was closer to 2.1 power. This is a tad light, and the dip at 90% means highlights are a bit muted. You’ll see later that the color gamut is quite a bit off as well.

Switching to the Movie mode, color is much closer to the mark, but gamma is still too light. I calibrated the two-point white balance sliders and only needed to adjust the gains to get all errors below 1dE. The gamma preset worked best on BT.1886 with its slider set to +1. It isn’t ruler straight, but it is much closer to the 2.2 reference than before.

Comparisons

The M9 compares well to its competition with a top finish out of the box and second place after calibration. Adjusting the white balance isn’t necessary, but I recommend changing the gamma to BT.1886 +1 for the best possible luminance tracking. It only misses 2.2 by 0.4%, actual value 2.1, though the range of values is a little larger than the Dough and Dell screens.

Color Gamut Accuracy

Our color gamut and volume testing use Portrait Displays’ Calman software. For details on our color gamut testing and volume calculations, click here.

The color tests show the differences between the M9’s image modes and why I made the choices I did. Standard covers most of DCI-P3, which is a good thing, but the saturation points are off-target. Red and green are undersaturated between 20 and 80% while blue is a little too vivid. The balance is off, and you can see it in content. Color doesn’t look entirely natural.

Movie nails the BT.709 gamut for SDR, coming up just a tad short in red with a slight hue error in magenta. Calibration tightens up the chart and fixes the magenta issue.

If you want the larger gamut in Movie mode, pick Native from the gamut selections. It shows the same behavior I recorded in the Standard mode, with undersaturation in red and green and magenta hue errors. Based on this result, I’d recommend using Auto for the color space setting to keep BT.709 in play for SDR content.

Comparisons

If you keep the M9 set to Auto color space, it runs with the best professional screens with a 1.26dE gamut error. It doesn’t get much better than that. The wide gamut is best used with HDR content as you’ll see on the next page.

In the volume test, the M9 managed 97.25% coverage of DCI-P3, which is a bit below average for Quantum Dot monitors. It is in line with non-QD screens, and it is still a very colorful monitor, just less so than the top three. sRGB coverage is just under 95% which makes it suitable for critical applications.

Test Takeaway: You have to hunt for the M9’s color accuracy, but in Movie mode, it performs well with the only needed change being to the gamma preset. Grayscale and gamut tracking are spot-on out of the box. My only nitpick here is gamut coverage, which is large but not as large as other Quantum Dot displays.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Our HDR benchmarking uses Portrait Displays’ Calman software. To learn about our HDR testing, see our breakdown of how we test PC monitors.

Like any OLED, the M9 delivers superlative HDR. It stands out by supporting HDR10+, which is even more rare among computer monitors than Dolby Vision. There are a few options that bear looking at if you’re seeking maximum performance

HDR Brightness and Contrast

The M9 uses variable brightness for HDR content to peak at over 450 nits when measuring a 25% window pattern. That’s in line with most of its competitors, though Asus and Dough push their panels a bit harder to peak at over 700 nits. The difference can be seen in small highlights but most content is similar in look and feel. To get the most from the M9, set Peak Brightness to Medium and Tone Mapping to Static.

Grayscale, EOTF and Color

The first chart is what I based my recommended HDR settings on. Grayscale tracking is perfect in every case, but the EOTF curve is very different depending on the chosen options. With Peak Brightness on Medium and Tone Mapping set to Static, the M9 hits the marks almost perfectly.

It also does well in the HDR color tests, where it tracks the saturation points correctly and only shows a slight magenta hue error. It’s still a tad shy of full DCI-P3 coverage, but HDR color is vibrant and rich. In the BT.2020 test, it hits all the inner targets properly until running out of color at 85% red, 75% green and 95% blue.

Test Takeaway: The M9 renders excellent HDR with bright highlights, deep shadows and vivid color. It doesn’t quite cover the same gamut volume as other Quantum Dot OLEDs, but it is still very colorful. With the right settings, its quality matches the best gaming monitors available.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

There are situations where a big-screen TV just isn’t practical. But that doesn’t mean you can’t watch one of the tens of thousands of shows or movies available on streaming platforms. We are truly in a golden age of television, no longer bound by the restrictions of broadcast and cable providers. If all you need for TV bliss is a streaming box and a few subscriptions, then it follows that it should be integrated into your display. When you can have only one screen for watching content, gaming, and work, it should be something like Samsung’s M9 M90SF Smart Monitor.

The M9 proves itself to be an excellent gaming monitor thanks to its 165 Hz and Adaptive-Sync. It delivers high-quality streamed movies and TV shows with a solid software interface and an included remote. And it gets through work tasks with ease and comfort. It brings superb image quality with 4K resolution, an effective matte screen layer, accurate color, a wide gamut, and support for both HDR10 and HDR10+ formats. It also delivers better-than-average sound from two speakers with a wide frequency range, a broad soundstage, and balanced dynamics.

The M9 delivers physically, with premium build quality, smooth styling, and convenient ergonomics. USB ports enhance its connectivity options, and Wi-Fi eliminates the need for extra cables for internet access. The price is a bit higher than that of similarly sized competitors, and it doesn’t have a 240 Hz refresh rate. However, it brings a lot to the table that is unique, thanks to its Smart TV side. It’s three displays in one, which makes it something of a bargain. If you’re looking for a do-it-all monitor that meets all your entertainment, gaming, and productivity needs, the Samsung M9 M90SF Smart Monitor is well worth checking out.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Enter the Bambu Lab P2S Combo, a machine that borrows some of the best features from its siblings while remaining an affordable option in four-color printing. It took the color touch screen from the X1 and the swappable nozzle from the A1 – or rather, the H2’s improved versions of the screen and nozzle. It has a better camera, a USB port, an upgraded cooling system, AI camera monitoring to prevent spaghetti monsters, and it comes standard with the new AMS 2 Pro.

Like the P1S, it has everything you need while leaving some of the high end perks on the table. This printer still has the traditional Bambu Lab 256 mm cubed build volume, but has no lasers, no dual nozzles, no fancy air flaps, and definitely no LiDAR. And best of all, it doesn’t need any of that.

At the time of this review, the P2S retails at $549 for the standalone machine and $799 for a combo, only $50 more than its predecessor. It's nearly half the price of the H2S and less than a quarter the cost of a fully loaded H2D.

The P2S maintains Bambu Lab’s exceptional level of speed, quality, and ease of use, making it one of the best 3D printers we’ve seen this year.

Specifications: Bambu Lab P2S

Bambu Lab P2S: Included in the Box

The Bambu Lab P2S comes with everything you need to get your printer set up, as well as a few extra parts. You get tools to maintain the printer, spare nozzle wiping pads, and a .5KG full full-sized (refillable) spool of white PLA.

The printer comes with test models loaded in its memory, including a scraper that you can print and assemble with parts provided in the toolbox. It does NOT come with a USB stick, which you will need to take time-lapse videos. There’s a paper getting-started guide and instructions on how to access the slicer and phone app. The box has a QR code printed on the inside flap that leads to an unpacking video.

Design of the Bambu Lab P2S

The most obvious improvement to the P2S is the touch screen, which is the same as the high end H2 series. The previous machine had a barely functional D-Pad LCD screen that drove its users nuts and made the P1 feel cheap.

The tool head also takes a page from the H2 series and uses the same steel-tipped A1-but-not-an-A1 nozzle. Confused? Me too. The new nozzle looks similar to the removable A1 nozzle and is held on with the same buckle and magnet system. The P2 and H2 nozzles have higher flow rates to match their faster Core XY machines. Bambu says you can use the new nozzle in the A1, but doesn’t recommend putting an A1 nozzle in an H2 (and I assume P2). I wouldn't be surprised if Bambu phases out the old A1 nozzle for more efficient stocking. Either way, I’m happy to see this style in the P2S, as it is way easier to remove for troubleshooting or just swapping sizes.

The P2S also gets a new Active Airflow system, which solves one of my personal pet peeves: leaving the door open while printing PLA. Many new users think the door is there to keep children/cats/dust out of the printer, rather than provide a controlled environment for high-temperature printing. The airflow system feeds cooler air from the room to the auxiliary cooling fan on the sidewall. During testing, the chamber temperature hovered around 33C and I had no issue with nozzle clogs or overhangs.

The printer does not have active heating. But instead, it relies on the bed heater to do all the work. It makes this easier by shutting down internal flaps and only pushing air out of the chamber through a carbon filter. I was able to run ABS without issue.

The P2S also gets a new USB port, which is one of those little things that really means a lot. Having a USB port allows you to easily transfer files without going through the internet, something the older micro SD card made challenging. It's also where the P2S stores timelapse footage.

The camera has also been upgraded to a 1080 high frame rate model, and when combined with enhanced chamber lighting, it delivers much better results. The camera is also part of a new AI computer monitoring system that makes sure you have the same plate installed as the slicer settings and watches for nozzle blobs and spaghetti printing.

Nozzle pressure is measured with an eddy current sensor, and the machine automatically adjusts the flow rates for you. Flow rate is also backed up by the new extruder system, with a PMSM servo delivering 8.5 kg max of extrusion force. That’s about 70% more than the old system. The printer also monitors the system to detect filament grinding or clogs. Combined with a tangle sensor in the filament hub, the P2S is well-equipped to keep printer failures at bay.

Another small, but important improvement is the removal of the dead zone in the forward left corner. The older machines reserved this space for a permanently mounted prong to push the cutter. Now the machine deploys a foldable prong from the side of the frame.

And the much maligned carbon fiber rods for the X gantry are gone, replaced with a more easy to clean steel.

The rest of the printer is largely the same: a 256 mm cubed build volume and a speedy, highly accurate motion system. In case this is your first Core XY, the motion system uses a combination of belts to move the X and Y axes in tandem for smoother, faster printing. The print head stays at the top of the machine while the build plate slowly lowers. When combined with vibration-damping input shaping, the results are exceptionally smooth, high-quality prints.

The printer’s highly sensitive auto bed leveling system makes for a perfect first layer. There is no need to adjust the Z height. After a few minutes of auto calibration, the printer just works.

The P2S Combo comes standard with the new AMS 2 Pro, a four-color Automatic Material System. This new AMS can double as a filament dryer with active venting to maintain low humidity when the lid is kept closed. It still needs desiccant packs to keep moisture at bay because the dryer doesn’t run while it prints.

The AMS 2 Pro is now easier to maintain with the tubes exposed. This makes it a lot easier to fish out scraps of filament that might snap inside the unit.

Like all the other Bambu Lab machines, P2S needs a collection bucket for its inevitable pile of filament poops. There are various community-designed solutions to the problem, or you can just put an empty filament box at the back of the machine. I do wish Bambu would address this issue.

Bambu Lab says the P2S can obtain a top speed of 600mm/s and 20,000 mm/s acceleration, which is still twice the default speed of the slicer. We still have “sport” and “ludicrous” speed modes, which are 124% and 164% faster than normal, but with somewhat sketchy quality results. It’s fine for a quick draft, but I’d still stick with standard speed.

When the P2S has an error, the touch screen can show us the same helpful assistant as the higher-end machines. It will provide a QR code you can scan with your phone, offering troubleshooting advice from the Bambu Wiki, a knowledge base for operating any Bambu printer. The Wiki will help you identify and fix the problem, often with quick video tutorials.

Bambu Lab Automatic Material System (AMS)

The P2S comes standard with Bambu Lab’s new AMS2 Pro system – a four-spool feeder that unlocks the ability to print in four, eight, twelve, and even sixteen colors. Each unit costs $359 and can be placed on top of or next to the printer. It can still use an older AMS if you have extra or just want to save a couple of bucks.

The new screen (love that screen) lets you program filament colors while standing next to the printer, and also lets you change the colors or spool placement on sliced files.

The AMS unit can read RFID ID tags that the company installs on each roll of its own filament for easy identification. These tags tell the printer what type and color of material you have installed, and then match it with presets in Bambu Studio. If you use 3rd party filament, you’ll have to fill that information in by hand.

The AMS has pockets for descant and seals tightly, allowing you to use the AMS as a drybox for temperamental filament like Nylon. The AMS 2 Pro does double duty as a filament dryer, though only while it is not printing.

There’s also a “backup” option for the AMS, which will allow it to automatically switch from one slot to the next instead of tripping a run-out sensor. This could be handy for using up partial spools or prints that will need more than one spool. To use this setting, you’ll need to have identical spools loaded.

Assembling the Bambu Lab P2S

The Bambu Lab P2S comes mostly assembled and only needs to be unpacked. You only need to attach the spool holder if you need it, pop on the touch screen and connect the cords and tubes. The printed guide walks you through each step.

Leveling the Bambu Lab P2S

The Bambu Lab P2S has an excellent auto-leveling system combined with vibration compensation, pressure advance, and automatic belt tensioning. It runs a thorough self-test after unpacking the machine to make sure everything is in working order.

Once you’re set up, the machine will also automatically run a bed leveling check before each print. This takes a few minutes, but you’re free to uncheck the option if you don’t want to take the time.

There is no need to set the Z height or Z offset. In fact, there’s no option or button to do so, because the P2S handles this setting on its own.

Loading Filament on the Bambu Lab P2S

The P2S comes with a single spool rack if you didn’t opt for an AMS, or for running TPU, which can’t go through an AMS. The single spool is mounted on the side and is much easier to deal with than the old rear-mounted P1S spool.

The AMS 2 Pro, which comes with the combo, works the same as the old AMS. Just insert the filament and walk away.

The P2S automatically unloads filament when it’s done printing.

Preparing Files / Software for Bambu Lab P2S

Bambu Studio is the custom slicer for the P2S and its fellow 3D printers. It’s largely based on PrusaSlicer, though the layout is a bit different. If you find the settings overwhelming, it’s generally fine to run with the default presets.

The device tab is where you can send files via LAN or the Cloud, watch time-lapse videos from your camera, and remotely tweak the speed and temperature of your printer.

A link to MakerWorld on the homepage allows you to easily search for models designed to work on Bambu machines. While signed in to MakerWorld (and using the Cloud), you can also send presliced files directly back to your printer.

Printing on the Bambu Lab P2S

Our P2S review unit came with a partially filled spool of while Bambu Lab PLA. This is great to get you started, but you’ll want to check out our guide to the best filaments for 3D printing for more suggestions.

The printer came with several pre-sliced files to print, like this phone stand I printed using Epax PLA Magic, and it turned out pretty nice.

Of course, the first real print had to be a Poop Chute. Otherwise, the P2S was going to spray filament waste all over the floor behind its shelf. Unfortunately, the P2S is a tiny bit larger than the P1P, and as this printer is not public at the time, the chute I printed didn’t quite fit. It could probably be scaled up a bit on the X axis to work.

This chute is very basic and was designed to be super thin, making it a quick print. I further sped up the process by using a .28 layer height. This was printed using Polar Filament Black PLA for four hours and 30 minutes.

More impressive is this Analog Tetris game, which I printed in batches using lots of different filaments. Each plate of blocks took about 2 hours and 40 minutes to print, using standard settings and a .2mm layer height. The blocks use a mix of Bambu Lab PLA and a spool of Flashforge Aurora Red PLA, which has an amazing sparkle.

The game board was a longer print and was designed in two sections with tabs to snap it together. This was printed using Bambu Lab PETG Carbon Fiber, and was Glooped together. I used a .2 layer height, default speeds, and a 15% gyroid infill. The print took 20 hours and came out absolutely perfect.

ABS was super easy to print. This fighting top was printed in PolyLite Galaxy Purple ABS and PolyLite Blue ABS. I wanted it to be very strong, so I used 4 walls and 20% gyroid infill. It took 3 hours and 12 minutes to print. The blue dots were added in Bambu Slicer with the paintbrush. The print is definitely strong, but there was a tiny gap in the layers on one side. It’s less noticeable with the naked eye. The two parts have no trouble fitting together.

This print is a mash-up of Nozzle Spinning Top by Rayjizza and the Spinning Top by KasKap.

TPU ran like a dream, but I did have trouble with the spool holder. You’re supposed to take the lid off and run TPU without the Bowden tube directly into the extruder. I didn’t feel like taking the lid off, so I got out a spare spool holder and ran it through the entry point on the right side of the machine (the yellow part). This avoided about half the Bowden tube, but still used the tubes inside the printer. It worked fine. The eternal spool provided by Bambu was on the wrong side of the printer for this set up to work, but in their defense, filament is meant to go through the tangle sensor located in the buffer.

The purse took 6 hours and 46 minutes to run, using a .8mm nozzle with a .2mm layer height, 4 walls and otherwise standard TPU settings. The pyramid pattern got a little bit wispy, but otherwise the purse looks great and has been very functional as my everyday bag for about a month now. It’s printed in CookieCad Sunset TPU.

Bottom Line

Bambu Lab is continuing to innovate its Core XY printers and stay competitive in a tough market full of copycats. I’m impressed that Bambu was able to keep the price of the P2S so close to the first P1S while adding so many new features.

Retailing at $799 for the Combo, the P2S is a strong choice for anyone wanting to upgrade their 3D printing experience. Its simple setup and ease of use make it an excellent first printer for a newbie or the 10th printer for a farm.

The original Bambu Lab P1S will continue to be sold, and is currently still priced at $749, but I wouldn’t be surprised if that goes on sale soon. If you’re looking for a more affordable entry point into the world of color printing, Bambu Lab’s A1 Mini is still my favorite recommendation for first-timers and is only $399 for a combo unit. And if you want more space with full color, the Bambu H2S for $1,999 is hard to beat.

Design of the Dell 16 Premium (DA16250)

The Dell 16 Premium inherits its design from the XPS 16 (9640). Although the design has matured a bit, it’s still ultramodern, with radical design cues that position it as a cutting-edge system. Standout features include a seamless glass palm rest, which has no visible touchpad border, and a capacitive touch bar replacing the top keyboard row. (See more about these features will be discussed in the keyboard and touchpad section.) I find the design refreshing, as it looks much more futuristic than the average laptop, though it's divisive among Tom's Hardware's reviewers.

Calling the 16 Premium solid doesn't do it justice. Constructed from aluminum and glass, it delivers impressive strength. OLED-equipped models like our review unit also feature a glass display that compliments the premium feel. Build quality is top-notch, with tight panel gaps and a lid that opens effortlessly with one hand. Our model is white, though Dell also offers a darker graphite finish.

Measuring 14.1 x 9.4 x 0.75 inches (WDH), the 16 Premium shares a nearly identical footprint with the Apple MacBook Pro 16-Inch (14.01 x 9.77 x 0.66 inches), and weighs nearly the same, at 4.65 pounds to the Apple’s 4.67 pounds.

Dell’s port selection is minimal, featuring three Thunderbolt 4 ports (two on the left, one on the right), a microSD card slot, and an audio jack. Although USB Type-C has become mainstream, traditional USB-A remains a need for many users, so investing in adapters may be necessary. For wireless connectivity, the 16 Premium boasts an Intel networking card supporting Wi-Fi 7 and Bluetooth 5.4.

Dell 16 Premium (DA16250) Specifications

Productivity Performance on the Dell 16 Premium

Our Dell 16 Premium review unit is a high-end configuration, equipped with Intel’s Core Ultra 7 255H processor, an Nvidia GeForce RTX 5070 GPU, 32GB of RAM, and a 1TB SSD.

The GeForce RTX 5070 is a notable bonus for content creators and media generation. While it's capable of gaming, this laptop is not likely to perform better than an entry-level gaming model, as the GPU has a mere 65 W total graphics power.

In our benchmark comparisons, its primary rival is the Apple MacBook Pro 16-inch, tested in a $3,649 setup with a 14-core M4 Pro CPU, 20-core GPU, and 48GB of RAM. On the Windows side, we included the HP OmniBook Ultra, which was $1,749 as tested featuring an AMD Ryzen AI 9 HX 375 CPU and AMD Radeon 890M integrated graphics. Completing the lineup is Dell’s previous-generation XPS 16 (9640) using older components: a Core Ultra 7 155H and an RTX 4070 GPU.

In Geekbench 6’s synthetic CPU test, the Dell 16 Premium posted a single-core score of 2,784, the second lowest in our group that was only ahead of the XPS 16 (2,373) and trailed the AMD-powered HP OmniBook Ultra (2,847). Its multi-core performance was better, reaching 16,850 points and surpassing the HP’s 14,838. The Apple MacBook Pro 16-inch was in a league of its own, with the M4 Pro producing a score of 3,910 in single-core and a commanding 22,822 in multi-core.

The Dell 16 Premium underperformed in our 25GB file transfer test, averaging just 1,079.96 MBps. In contrast, the HP OmniBook Ultra reached 1,780.23 MBps, while the MacBook Pro 16-inch remained out of reach, delivering 2,237.5 MBps, more than twice the Dell’s throughput.

In our Handbrake test, the Dell 16 Premium completed the 4K-to-1080p transcode in 3 minutes and 50 seconds, decisively ahead of the XPS 16 (4:42) and HP OmniBook Ultra (4:18). Still, it couldn’t touch the MacBook Pro 16-inch, which finished in 2 minutes and 38 seconds.

We stress-tested the 16 Premium with 10 consecutive runs of Cinebench 2024. During the test, scores ranged between 980.7 and 1,011.75 points, indicating stable thermal performance.

Display on the Dell 16 Premium

The base configuration of the 16 Premium features a 16.3-inch non-touch IPS panel with a 1920 x 1200 resolution. Our review unit upgrades to an OLED display, offering a sharper 3840 x 2400 resolution, touch functionality, and a dynamic refresh rate ranging from 48 to 120 Hz.

The picture quality is nothing short of captivating. While watching Planet Earth III, I could discern subtle details in shadowed night scenes and was struck by the HDR brilliance of sunlight piercing through jungle foliage. The 120 Hz refresh rate lends a silky smoothness to everyday interactions, whether dragging windows or scrolling content. Touch responsiveness is excellent, and the hinge provides enough resistance to minimize wobble during use. The only drawback is the glossy glass surface, which tends to attract fingerprints and dust.

These laptops deliver strong color coverage, with the Dell 16 Premium reaching 79.5% of the DCI-P3 gamut and 112.2% of sRGB, putting it within a couple percentage points of its peers. Its peak brightness of 367 nits may seem modest, but that’s typical for OLED panels; by contrast, the MacBook Pro relies on mini-LED technology. In HDR testing, portions of the Dell’s display peaked at 572 nits.

Keyboard and Touchpad on the Dell 16 Premium

Dell made some radical design choices with the 16 Premium around input devices. The most divisive is the replacement of the traditional top-row function keys with a capacitive touch bar. Each “key” responds instantly to touch, though it lacks haptic feedback. By default, the bar displays media and system controls, such as volume, brightness, and the like, and supports Fn-Lock via the Fn + Esc shortcut to make F1 through F12 primary. As someone that relies on keyboard shortcuts whenever possible, such as Alt + F4 to close a window, I found this design counterintuitive, as it required me to look down each time I wanted to hit an F key.

Dell’s zero-lattice keyboard design eliminates spacing between keys, which is an aesthetic rather than a functional choice. The keys deliver decent tactile feedback, with an adequate 1 mm of key travel and a gentle cushion at the bottom of each stroke. I clocked 114 words per minute with 99% accuracy on MonkeyType, which is about as quickly as I can go on a laptop. That said, the arrow key cluster’s half-height up and down keys and full-size left and right keys aren’t intuitive. The power button, located adjacent to the backspace key, doubles as a fingerprint reader for biometric login.

The white keyboard paired with white backlighting is a risky design choice. Under bright lighting, the illuminated typeface can become difficult to make out. Having encountered this issue on previous white XPS models, I was curious to see whether Dell had addressed it. Thankfully, they have. The ambient light sensor now intelligently disables the backlighting in well-lit environments, retaining legibility. Conversely, in low-light settings, the backlighting activates appropriately, ensuring the typeface remains clearly visible.

As for the touchpad, it’s centered below the keyboard as expected, even though it has no visible border. Initially, I had reservations, but the absence of defined edges didn’t hinder usability. The haptic motors deliver a satisfying tactile bump with each press, and right-clicking proved consistently accurate.

Audio on the Dell 16 Premium

For a fairly slim laptop, the 16 Premium delivers first-rate audio. Its quad-speaker setup, consisting of two 3 W woofers and two 2 W tweeters, produces rich, full-bodied sound that matches entry-level Bluetooth speakers. Dubstep tracks from Seven Lions came through with crisp vocals and punchy bass drops, while Steve Winwood’s Valerie showcased the system’s clean instrument separation.

The one wrinkle in this setup: Dolby Atmos equalizers aren’t enabled out of the box. I had to manually download the Dolby Access app from the Windows Store, which felt like an oversight. While the preset modes didn’t dramatically alter the sound (I stuck with Balanced), they did further improve the already excellent clarity.

Upgradeability of the Dell 16 Premium

Accessing the 16 Premium’s internals is straightforward: the bottom panel is secured by eight Torx screws, all conveniently uniform in length. After removing them, the perimeter clips need to be gently disengaged. I used a plastic trim tool for this step, carefully working my way around the edges.

Options for upgrades and repairs are minimal. Only the M.2 storage drive and battery can be replaced; all other components are soldered and non-serviceable.

Battery Life on the Dell 16 Premium

In our battery test, which includes web browsing, video streaming over Wi-Fi, and OpenGL workloads at 150 nits, the Dell 16 Premium lasted just 7 hours and 15 minutes. That’s over three hours less than the previous-generation XPS 16 (10:44) and more than five hours behind the HP OmniBook Ultra (12:52). The MacBook Pro 16-inch remained a superstar, clocking an exceptional 21 hours and 1 minute. Clearly, Dell has room to improve its power efficiency.

Some clue as to what’s going on reside in Dell’s own battery estimates. In their internal testing, OLED-equipped models are rated for just 9 hours of Netflix streaming, compared to 27 hours on the non-OLED variant, a threefold difference. While we couldn’t verify those figures, the takeaway is clear: users prioritizing battery life should avoid the OLED panel.

Heat on the Dell 16 Premium

Thermal management on the 16 Premium is handled by two internal fans, which vent heat through rear exhausts lining the display hinge. The system performs well: during our Cinebench stress loop, surface temperatures peaked at just 78 degrees F on the touchpad, 86 F between the G and H keys, and 93 F on the underside near the exhaust vents. Internally, the Core Ultra 7 255H processor operated at an average temperature of 78 C.

Throughout testing, the chassis remained comfortable to the touch, and the fans operated quietly, blending into ambient household noise.

Webcam on the Dell 16 Premium

Dell’s webcam offers impressive picture quality. Despite only having a 1920 x 1080 video resolution, the picture is sharp, clear, and almost free of noise even in slightly darker than optimal environments. It also offers an IR sensor for facial recognition, allowing password-less logins using Windows Hello. Alas, Dell missed adding a sliding privacy shutter or an e-shutter, which competitors like HP and Lenovo often offer.

Software and Warranty on the Dell 16 Premium

Dell preinstalls a few utilities on the 16 Premium, including the My Dell app, which shows system health, warranty info, and access to some settings like Dolby Vision display modes. Users probably won’t need to interact with this app regularly.

The system also comes with one year of Dell Care Plus, which provides onsite repairs following remote diagnostics. It also includes proactive hardware monitoring via the Support Assist app, which can anticipate component failures and, according to Dell, even initiate automatic part replacements.

Dell also includes a years’ subscription of McAfee Premium anti-virus, not a trial.

A one-year warranty is standard, with extended coverage available. Accidental damage coverage is also available via Dell Premium for $89.

Dell 16 Premium (DA16250) Configurations

Dell’s 16 Premium begins at $1,799, equipped with a Core Ultra 7 255H processor, Arc 140T integrated graphics, 16GB of RAM, and a 512GB SSD. The base configuration features a 1920 x 1200 non-touch display with a variable refresh rate of 30–120 Hz and a peak brightness of 500 nits. Our $2,949 review unit steps up to a 4K OLED touchscreen, discrete Nvidia RTX 5070 graphics, and doubled memory and storage (32GB of RAM and a 1TB SSD).

Customizable models offer a Core Ultra 9 285H CPU, up to 64GB of RAM, and storage options reaching 4TB. Graphics upgrades include in-between options of the RTX 5050 and 5060. The laptop is available in Graphite and Platinum finishes.

Our review unit is priced in line with the Apple MacBook Pro 16-inch featuring the 14-core M4 Pro, which is $2,499 with 24GB of RAM or $2,899 with 48GB. Opting for 1TB of storage pushes the price to $3,499. On the Windows side, Dell faces limited competition. One notable contender we haven’t yet tested is the Lenovo Yoga Pro 9i Aura Edition, listed at $1,899 at the time of writing. It features a Core Ultra 9 285H processor, RTX 5060 GPU, 32GB of RAM, 1TB SSD, and a 3000 × 2000 OLED display.

Bottom Line

Dell’s 16 Premium largely delivers on its high-end ambitions. Its forward-looking design impresses with both quality and style, and it offers excellent entertainment potential with its beautiful OLED screen and quad speakers. Performance is also strong, at least among Windows machines, thanks to an Intel Core H-class CPU and quiet, effective cooling.

However, some design choices detract from usability. The capacitive touch bar, while responsive, lacks the tactile intuitiveness of physical keys, and the absence of USB-A ports limits connectivity. Our most notable complaint is the disappointing battery life of the OLED model.

Despite the Premium rebrand, this laptop still has all of the promise and compromises of the XPS 16. If you want a futuristic rival to the MacBook Pro, it's an option, but you'll have to handle the trade-offs to get the cool design and performance.

Hardware-wise, you get two PCIe 5.0-capable M.2 sockets, a solid audio solution, fast networking including Wi-Fi 7, capable power delivery, and DIY-friendly design - for the most part, not much different than its peers, save for the two fast M.2 slots. The two-DIMM board supports 128GB of RAM, with speed differences largely irrelevant unless you’re trying to break overclocking records. Asus’ features, including AI Advisor, Overclock, Networking II, and AEMP features, help simplify setup and maximize performance. Additionally, it offers DIY-friendly design, including updated PCIe Slot Release functionality and M.2 slide feature.