It's unlikely the CORE One L will have a kit version, as the AC-powered heat bed is not for tinkerers. The 3D printer is still built with easy maintenance in mind and open source parts for lifetime repairability. The printer is also designed to operate completely offline if you choose, without dependence on the cloud for complete security. However, if you enjoy the convenience of mobile apps, the Prusa App offers remote printer operation, full access to Printables.com for model selection, and a lite version of PrusaSlicer for custom slicing on the go.

The CORE One L has a few clever tricks up its sleeve: it can flip the top vent open or close on its own, a flip of a switch makes loading (and running) TPU super easy, and strategically mounted fans control air flow and heat so the door can remain closed at all times. The chamber can hit 60 °C, making engineering-grade filaments as easy to run as PLA. It will also be the first 3D printer compatible with Prusa-backed OpenPrintTag, an open source initiative to create a universal “smart” spool with NFC tags that customers can reuse as they wish.

The CORE One L retails at $1,799 (including import duties and tariffs) and comes with 24/7 customer service and a complete ecosystem. This machine lives up to its reputation and is one of the best 3D printers we’ve tested to date.

Specifications: Prusa CORE One L

Prusa CORE One L: Included in the Box

The Prusa CORE One L comes 98% assembled. It includes a full set of tools and a full 1 kg spool of Galaxy Black Prusament PLA (plus a bag of Gummy Bears). There’s also a USB stick with models to test out and a handbook on 3D printing with your new machine. The PrusaSlicer is available for free download from the website.

Design of the Prusa CORE One L

The CORE One L is at its heart a larger CORE One, with a few critical changes. The most obvious is the size, now sporting a 300 x 300 x 330 build volume, which is about twice the size by volume, if not exactly plate size.

The slim, all-steel exoskeleton now has aluminum walls to cut down on heft, which means the new printer checks in at about the same weight (48 pounds) as the smaller CORE. This keeps it still somewhat manageable to move by a single person and sit on a desk.

It has the same indents on either side to conceal the spool holder and customizable storage space, though now magnets will only stick to the frame. The door and upper parts of the side panels are translucent black polymer. The top panel snaps into place. The top vent can now be opened or closed by the printer itself. It’s still not exactly airtight, with many gaps at the seams, but a few gaps in the frame won’t spoil the system.

The exoskeleton provides a strong, stiff frame for the printer while also minimizing materials and avoiding the cheap plasticky feel we get from other brands. The indented sides reduce the interior space of the printer, making for less space to heat.

The biggest change to the CORE One L is the heat bed, which is now a solid chunk of aluminum, heated with AC power. It provides uniform heating from corner to corner, which is essential with such a large surface. There are two fans underneath the bed to distribute the heat like a convection oven. This eliminates the need for a second heating system, as the bed can now be used more efficiently to warm the entire chamber for technical filaments.

The CORE One L still shares many 3D parts with its Prusa siblings, which are produced at the large Prusa Research print farm in Prague (or in Delaware by Printed Solid if you buy one in America). CORE One 3D parts are made with PC Blend Carbon Fiber, printed on a combo of MK4s and CORE One machines. Prusa also does its own injection molding, which you can see on the LCD case.

Another thoughtful touch is the door hinge, which can be flipped if you rather have it open on the other side. The door can completely open 180 degrees, but Prusa encourages users to go ahead and keep it closed for any material, thanks to their active cooling system. The door has a sensor that will warn you to keep it shut and stop the print when the door is opened. This is a nice safety feature for schools, parents of small children, and pet owners. Thankfully, the sensor can be switched off if you want the freedom to poke your nose or a camera in there.

The CORE One L specs don’t include acceleration rates, as the company is famous for putting quality over speed. The machine does come with a high flow nozzle, input shaping and factory tuned print profiles for over 200 filaments. Still, I opened up PrusaSlicer to see what kind of settings they’re using for speed and acceleration. The top printing acceleration rate of 3000 mm/s² would not impress a Bambu user who easily has twice the speed. However, acceleration isn’t everything. Using their default settings, both the CORE One L and the similarly sized Bambu Lab H2S printed a Maker’s Muse calibration castle in about one hour and 15 minutes using .25mm layer height and default settings. Sure, the Bambu might be faster printing a big, plain box, but how often do you print things that are so boring?

The CORE One L finally has an onboard accelerometer, though it’s completely factory calibrated and ready to print as soon as you plug it in. It also has an NFC antenna to make getting the machine online and connected to the Prusa app and slicer very easy. The app is entirely optional for those who wish to keep their printers offline. The CORE One L has a USB port for manually loading files, which is the printer’s only place for file storage.

Like the CORE One, bed leveling is fully automatic, with no manual adjustments. First layers go down perfectly smooth thanks to a load cell sensor that has replaced the bedslinger’s PINDA probe. The sensor is connected to the nozzle, which allows the CORE One L to physically tap the print surface for a more precise measurement. The sensor is also used to calculate Z Height, and it does it so well that there is no need for human intervention. Of course, there’s still a live Z adjustment available in the menu should you need it. During testing, we did not.

Assembling the Prusa CORE One L

The Prusa CORE One L comes nearly assembled. You just have to unpack it and remove a few screws that held the bed down during shipping. The screen arrives packed inside, attached to the front trim plate, and just needs to be slapped on the printer with magnets.

Our machine came with the optional filtration system, which did need more complex assembly. That involves removing the back panel to access the electronics, plugging the fans to the board and screwing everything back together. It was fairly straight forward, but I did have to hunt down the directions online.

Leveling the Prusa CORE One L

Bed leveling is a no-brainer, as the Prusa CORE One L is fully automatic. There are no manual adjustments and no need to adjust the Z offset. The printer uses a load cell sensor connected to the nozzle, which physically taps the print surface for precise measurement.

The printer trams the bed and calculates the required z height before each print.

Loading Filament on the Prusa CORE One L

Loading filament is very simple. Simply place the spool into side mounted rack and feed the plastic into the reverse Bowden tube until it reaches the hotend. The printer will automatically detect it and pull it in the rest of the way. The screen will then ask what kind of material you’re using to set the temperature. Once it gets hot enough, it will pull the filament into the hotend and ask if the color has changed. If it has, you can say yes and move on. If the answer is no, it will purge more material.

To remove material, go to the Filament menu from the main screen and click unload. It will remember what material was last installed, determine if the filament is still there, and then heat it to the appropriate temperature. Once hot, it will retract the plastic until you can easily remove it.

If you’re running flexible filaments, there’s a switch where the tube enters the chamber that allows you to switch off one of the filament sensors (another is on the tool head). This allows soft TPU to load as easily as PLA and I didn’t have any trouble at all.

Preparing Files / Software for Prusa CORE One L

You’ll need to download PrusaSlicer, which is free and open source. It is developed by Prusa Research and serves as the foundation for many other slicers. It comes with hundreds of profiles for every material and nozzle size that Prusa Research makes.

PrusaSlicer can also directly access Printables.com, which offers a wide selection of free file models for both fun and practical prints. If you’re using the mobile app, files can be sent directly from Printables to the online EasyPrint slicer and then directly to your printer if you choose to have it online.

Printing on the Prusa CORE One L

The CORE One L comes with a full 1 KG roll of Prusament PLA Galaxy Black. If you want more colors and materials like silks and multicolor filaments, you should check out our guide to the best filaments for 3D printing for suggestions.

We used the sample material to print a Benchy to see what the CORE One can do with its fastest default slicer settings. We used Speed Benchy rules for fairness: 2 walls, 3 top and bottom layers, 10% infill, a .25 layer height and .5 layer width. Using these settings, we got a very nice benchy in 23 minutes with a bit of wisps probably due to the humidity in the house.

I printed a few soap dishes I found on Printables for a practical PLA print. These were a little chonky looking despite the .2 layer height, due to the curves. It had 3 walls for strength printed in Prusament PLA Marble Grey, which makes home goods look more posh. It took 5 hours and 54 minutes to print three of them.

I also printed some cute clips, which mix practical with whimsical. These “Vicegriptor” locking clips have surprising bite when printed in PETG. A word of warning – you need to be very careful loosening them up the first time, or you risk breaking the print in place joints. Once they’re limbered up, they work really nice for holding chip bags and the like.

I printed eight at a time on the CORE One L’s roomy bed, taking 3 hours and 13 min, using a .2 layer height and default settings. This was printed in Greengate’s recycled Deepwater Blue, a sparkly filament created for 3D Print Bunny.

For TPU I made another ill-fated dog toy for my Standard Poodle Ripley. The print was excellent, but 3D printed dog toys for strong chewers is still not that great. The filament loaded very easily, thanks to the new switch, and I had no trouble at all pushing it through to the Nextruder.

This printed in 6 hours and 6 minutes, using Inland Rainbow TPU and a .2mm layer height. It lasted about 6 hours after I gave it to the dog too. Before the dog destroyed it, it had very nice smooth and glossy surfaces and very slight blemishes from layer lines.

We ran a lot of high-temperature material from it, like this Fiberon Red PET-GF15 by Polymaker. This is a form of Polyethylene terephthalate, like PETG, but without the G (glycol), which is usually added to make the material behave. Instead, Polymaker added 15% glass fiber to make the PET stiff, dimensionally stable, and thermal resistant. It prints fast and super hot, and this needed 280 °C on the nozzle and 80 °C on the bed. I filled the bed with parts, and it printed very nicely without warping. There were some goobers on the edges that will brush off.

Bottom Line

The CORE One L is a great addition to the Prusa lineup, filling a space between the “medium” MK4/CORE One and the Prusa XL tool changer. It gives you Prusa quality and security in a larger size without having the larger pro price tag. It has a retail price of $1,799 (including import duties and tariffs) and comes with 24/7 customer service and a complete ecosystem.

Its new aluminum bed with even AC-powered heating is excellent, and it has a few perks like the camera, TPU switch, and self-opening top vent that could be easily added to the smaller CORE One. It will also be the first printer to try out OpenSourceTag when it hits the market.

We’re also keeping our fingers crossed for a tool changer mod through a partnership with Bondtech, but details are still vague as of this writing.

If a Prusa CORE One L seems just a bit too spendy for your budget, the smaller Prusa CORE One comes in at $999 for a kit. The delightfully budget-friendly Bambu Lab P1S Combo is currently on sale for $549, and if you need some massive CORE XY size, the Creality K2 Plus is worth checking out at $1199.

So is the €39 Arduino Uno Q just an updated Arduino Uno R4? Oh no, it is something much more. The Q stands for Qualcomm, which recently purchased Arduino and now has its Arm-based Dragonwing SoC on the Arduino Uno Q. Still, the Arm CPU isn’t alone, as the Arduino Uno Q also has an STM32U585 microcontroller. Essentially, an Arm Cortex M33 (the same as the Raspberry Pi Pico 2’s RP2350) that runs the Arduino sketches (your projects) alongside the main CPU.

Who is the Arduino Uno Q for? What can you build with it? Does it perform better than just buying a Raspberry Pi 3 and Pico 2? Let's find out.

• The Arduino Uno Q is the first Arduino board after Qualcomm’s recent acquisition of Arduino.
• The Arduino Uno Q can be used as a single board computer, or it can be used via USB or over a Wi-Fi connection.
• The board integrates a Qualcomm Dragonwing SoC for AI and LLM capabilities, with an STM32 microcontroller for real-time control of GPIO.
• The Arduino Uno Q supports a new IDE, Arduino App Lab, which integrates Python and Arduino’s C language to build projects.

Arduino Uno Q Specifications

Look and Feel of the Arduino Uno Q

At a quick glance, the Arduino Uno Q and the Arduino Uno R4 appear the same, but they are quite different. The first noticeable difference is the omission of a DC power jack. The older Arduino Uno boards had a DC barrel jack and in all my years of tinkering with Arduino, I probably used it less than a dozen times.. If you want / need to supply a power source greater than 5V, the VIN pin can be used with a 7-24V supply.

The USB-C port provides power (5V, 3A max) and data connectivity, allowing you to connect the Arduino Uno Q directly to your PC. It also provides, via a hub, access to DisplayPort / HDMI, USB, and power. Using the Arduino Uno Q with a USB-C dock, HDMI display, keyboard, and mouse takes some getting used to. By which I mean, actually sitting down and writing code directly on the Arduino board. In the past, I’d plug it into a PC, write some Arduino code in the IDE, compile and flash the code to the Arduino, and off it went. But sitting here in the Arduino App Lab, writing the code and uploading it to the STM32 just feels alien.

Interestingly, Arduino recommends using the 4GB model as a single board computer, yet the first model released is the 2GB, which leads to a subpar SBC experience. If you prefer the more traditional Arduino workflow, you can connect via USB or Wi-Fi to a PC running the Arduino App Lab. All of your code runs on the Arduino Uno Q, but development and creation take place on your much more powerful PC. This is my preferred means of using the Arduino Uno Q.

When powering up the Arduino Uno Q, the gorgeous 8 x 13 LED matrix illuminates to show an animated Arduino logo, a nice touch that distracts from the rather slow boot process, 34.6 seconds to be exact. Not horrific; the Raspberry Pi 5 is around 20-25 seconds, depending on whether you use a microSD or an SSD. But it feels like a long time for those of us used to powering up an Arduino and seeing the code run straight away.

Qualcomm Dragonwing

The Arduino Uno Q boots directly into a version of Debian Linux, and it is pretty vanilla. The GNOME interface is pleasant and gets the job done. You could use the Arduino Uno Q as a low-power Linux desktop; it would certainly be a conversation starter. But ultimately, the desktop OS is more for getting the Arduino App Lab running than a full desktop OS. Perhaps this will feel a little different when the 4GB model is released. The paltry 2GB of RAM is just enough to get things working.

The newest addition to an Arduino is born from Qualcomm’s recent purchase of Arduino. The Qualcomm Dragonwing QRB2210 is a quad-core Arm Cortex A53 SoC

The Arduino Uno Q is no Linux PC powerhouse, but it doesn’t have to be. Sitting around the Raspberry Pi 3 / Raspberry Pi Zero 2 W level of computing power, the Qualcomm Dragonwing SoC has enough compute to run Linux, and its two 13MP Image Signal Processors (ISPs) support two cameras, with AI inference models that can run on the CPU and GPU. That makes it a potent package for entry-level AI and IoT projects.

But this is not an Arm desktop PC. If you want that, go for a Raspberry Pi 5. The upside of a lower-power CPU is that when the Arduino Uno Q is used as a desktop PC, it draws around 3.3 W at idle and peaks at 4.5W with all four CPU cores under full load. For 90% of the time under stress, the Arduino Uno Q pulled 4.4W. We have to remember that there is no way to just power the STM32 microcontroller. We need to power up the Qualcomm SoC to access the microcontroller. This means that the ultra-low power afforded by a microcontroller project is lost. If that is a concern, then stick to a microcontroller-based board.

How does the Dragonwing communicate with the real-time STM32 microcontroller, and vice versa? The answer is “Bridge,” specifically Arduino's RPC (Remote Procedure Call) library, which enables sketches written for the STM32 to access Linux services and vice versa. The libraries are written in Python for the Dragonwing and Arduino C for use with the STM32. They are both abstracted enough to be easy to use.

The GPIO

The Arduino Uno Q has the same tried-and-tested GPIO pinout as the older Arduino Uno boards. The pinout reference is printed on the side of the connections, and it is great to see the ~ is still printed, reminding me which pins are PWM-compliant for my robot and Neopixel projects.

The Arduino Uno Q also has two higher-density connectors that are earmarked for high-speed camera/displays, audio, SDMMC, and as an expansion with more GPIO pins. Sadly, I cannot test this aspect of the board because, at present, there are no breakout boards or add-ons that support the interface. But I can see these interfaces being used to dock the Arduino Uno R to another board, enabling extra features like robotics in a simpler form factor.

The QWIIC interface is most interesting. It is the same interface as StemmaQT, MakerPort, and QWIIC, which really does make quick work of connecting up a project. QWIIC is essentially a breakout of the I2C interface, but you don’t strictly have to use it for I2C components, although many add-ons do. You’ll need to purchase compatible QWIIC addons, such as those found in the Arduino Plug and Make kit.

Connect up your sensors, in my case, an accelerometer, then load up your code. I chose to use the accelerometer example, and I quickly saw how the STM32 read the raw data from the sensor, which was processed by the Qualcomm Dragonwing to identify the pattern of movement. Lastly, the output was rendered to a web UI in real time. Powerful learning potential!

Arduino Shield Compatibility

Using the Arduino Uno form factor is a smart move for the Arduino Uno Q, chiefly because it provides access to the extensive Arduino Shield ecosystem. Much like Raspberry Pi HATs, Arduino Shields connect to the Uno form factor GPIO and provide extra functionality for projects. And much like the Raspberry Pi, the Arduino Uno Q introduces a period during which shield/HAT compatibility is in flux, despite the same form factor.

I’ve got Shields for Ethernet, IoT, NFC, GSM phone connectivity, RGB LED matrices, etc. The problem with the shields is that they either require a compatible library or a compatible pinout for the shield. It seems that my Adafruit RGB Matrix does not work with the Arduino Uno Q, my collection of Maplin (a now defunct UK electronics retailer) Arduino shields were also impacted by these conditions. If you rely on a shield for your project, stick with an older, well-supported, and known-to-be-compatible Arduino.

Arduino App Lab

No matter how you choose to use the Arduino Uno Q, you’ll be using the new Arduino App Lab in some form. Initially, I used it directly on the board, connecting a screen, keyboard, mouse, etc., via a USB Type-C hub. It worked, and I was able to write a simple “LED Blink” script, but the experience was slow and left me wondering if something was happening as I waited for something to happen. As I’ve already said, the best way to use the Arduino Uno Q is via the Arduino App Lab on your more powerful PC.

The Arduino App Lab is slick and looks great. But I found myself struggling to understand the concepts and workflows necessary to build a project. I’m not (that much of) an idiot, and I can easily write and flash code to my other Arduinos and clones. But the extra steps for the Arduino Uno Q led me to really sit down and read the documentation. Luckily, Arduino has great documentation, but there were times when I was lost. For example, using the Serial.println() is not supported on the Arduino Uno Q, replaced with Monitor.println(), but I was unable to get this working despite following the instructions.

The Arduino App Lab also introduces “bricks,” packages of Python code that bring functionality via an easy-to-use workflow. Some bricks provide AI-relevant features, such as keyword spotting (used in the Hey Arduino! example), while others link to external APIs for weather and web-based user interfaces. Some bricks run as-is; others require a container to run; all are handled by the Arduino App Lab.

Adding a brick to your project is easy. Just click on Bricks, select what you need, and the brick is part of your project. Follow the API documentation to integrate it into your Python code, and you will soon have a keyword-powered LED, etc. I like this feature, largely because it makes using external resources a trivial process for newcomers.

The Arduino Uno Q workflow is really pushing for using the Qualcomm SoC and the STM32 via the RPC “bridge.” You can just write Arduino code directly in the Arduino App Lab, there is no real reason for using the bridge unless you want the power of the Dragonwing SoC. But by doing that, you are really wasting the potential that this board has. You could just use any Arduino or clone board and save yourself the extra spend.

If the Arduino App Lab isn’t for you, then the Arduino IDE 2.3.6 works with the STM32 side of the Arduino Uno Q. After updating and installing the board, I was able to flash the Blink sketch to the Arduino Uno Q. I also tried to upload the Adafruit NeoPixel strandtest sketch to the Arduino Uno Q, but it didn’t work. I checked the wiring, pins, and configuration, but nothing worked. I transferred the project to an Arduino Uno and flashed the sketch. It worked.

I dug a little deeper, and it seems that the STM32 is not supported by Adafruit’s NeoPixel library, so I had to search for something that was compatible. I couldn’t find anything compatible with the STM32 from the Arduino libraries, so I had to abandon that element of the test.

Overall, the Arduino App Lab is a great start, and as the Arduino Uno Q gains traction in the community, it will see more features and tweaks. The workflow takes a little getting used to, but it does simplify the link between the Python code and the Arduino code, which will help newcomers cut their teeth.

Testing the Included Examples

The Arduino Uno Q includes a set of examples covering a wide range of projects you can build with it. From the simple “blink an LED” to AI-based image identification, via web user interface-controlled LEDs and a voice-controlled LED matrix. Some of these projects rely on just the board, but those that use some form of AI often need extra components connected to the USB Type-C port. So you’ll need to dig out a compatible hub. This is where it gets messy, as we now have a hub, Arduino Uno Q, a microphone, and USB-C power to make a voice-controlled project.

I tested the “Hey Arduino” project and can confirm it worked, but it took ages to get the container running for the example. We’re talking 55 seconds from clicking Run to the project being available. Ok, a minute isn’t too long to wait, but if I were integrating this into a project, I would have to ensure that this and the time it takes the Arduino Uno Q to boot, 34.6 seconds, are both factored in.

The projects themselves are great; they illustrate how to use the Arduino Uno Q, and more importantly, they show how to create a project that uses both the Dragonwing and STM32 in one project.

Why is that important? The Arduino Uno Q is a hybrid device; it is neither a dedicated Linux SBC nor a microcontroller, it is both at the same time. This isn’t the first time that Arduino has created a hybrid board. The Galileo was a joint project with Intel, and it brought an Intel Quark X1000 CPU to a new form factor. It ran a Linux distribution based on Yocto and was compatible with Arduino shields. I’ll be honest, Galileo was an interesting idea, and back in the day, I reviewed a unit, but it was a very different beast from the typical Arduino with an Atmel chip.

There was also the Arduino Yún, a microcontroller-based board featuring the legendary ATmega32u4 and an Atheros AR9331 CPU that supported Linux. More recently, we have the Radxa X4, which features an Intel N100 CPU running any operating system you choose and a built-in Raspberry Pi RP2040, as seen in the Raspberry Pi Pico. Sure, for the Radxa X4, there wasn’t a bridge between the two, but there is a serial connection, and with just a few Python libraries, you could make your own bridge between the Intel and Arm CPUs.

AI Performance

The Dragonwing SoC at the heart of the Arduino Uno Q runs Linux, supports a Python bridge, and handles AI tasks. The thing is, due to the memory and processing restrictions, we can run an LLM, but it is tiny, and performance is slower than other SBCs.

To prove a point, I installed tinyllama:1.1b via ollama, and in a prompt, I asked it “What is an Arduino?” The answer was forthcoming, if a little slow. It took 28 seconds to start formulating an answer, and this led me to test how a series of Raspberry Pis would perform. The Raspberry Pi 500+, a keyboard version of the Raspberry Pi 5, started answering the question in 2 seconds! Even the older Raspberry Pi 4 managed 4 seconds. Sadly, the Raspberry Pi 3B+, a board with similar specifications to the Arduino Uno Q, didn’t have enough RAM to start the LLM.

If you’ve got an old Raspberry Pi 4 laying around, spend $10 on a Raspberry Pi Pico 2, and learn how to connect the two using your own serial bridge. You’ve just built your own Arduino Uno Q clone.

Who is the Arduino Uno Q for and what projects will they create?

Those already invested in the Arduino ecosystem will love the Arduino Uno Q, and I can see many AI / LLM-based projects being created with it. Robotics, sensor monitoring, smart homes, and more can be achieved with the Arduino Uno Q. Merging the AI-centric Qualcomm SoC with the STM32 for the GPIO is a smart move. The electronics for your project run on a real-time microcontroller, while the Arm CPU handles all of the AI, Linux, and Python aspects of the project.

Bottom Line

I love the Arduino. It is the board where I cut my teeth with electronics and microcontrollers. The Arduino Uno Q is an interesting board, but ultimately the spec is a little too old and reproducible with kit that many electronics enthusiasts will already have to hand.

That said, if you are a fan of the Arduino and its form factor, you will love the Arduino Uno Q and the ecosystem it brings. The hardware is good, and the Arduino App Lab is a great starting place to learn how to merge the two sides of the Arduino Uno Q into one project.

My worry with the Arduino Uno Q is that it becomes just like the Intel Edison, Galileo, and Arduino Yun. The Intel projects were born of collaboration, whereas the Arduino Uno Q is the result of Qualcomm’s purchase of Arduino. Could that give the Arduino Uno Q more support than previous attempts? Time and popularity will ultimately be the means by which this is measured.

Will this engineering choice result in performance that stands out enough to make our list of the best PC cases? Let’s take a look at the specifications and features of the case, then we’ll get to the design and features before wrapping up our review with thermal and acoustic benchmarks to see how well the case performs.

Product Specifications

Features of HAVN’s BF360 Flow

▶️ Solid black and stone-textured design, massive 180mm fans

The stone-textured front panel covers two of HAVN’s enormous H18 (180mm) fans for air intake.

These fans are much larger than your typical PC parts. To give you a better idea of just how massive they are, here are a couple of photos with HAVN’s H18 placed next to Montech’s extra-thick E28 fan!

▶️ Air scoop for superior GPU cooling performance

The BF360 Flow incorporates an air scoop, which directs the airflow of the bottom fan directly toward the GPU, resulting in some of the best GPU temperatures we’ve seen in any PC case.

▶️ GPU anti-sag support

For lage GPUs, the case includes a highly adjustable anti-sag bracket that mounts on the motherboard tray area and can be slid up and down as well as swiveled out to best support your specific graphics card.

▶️ PWM hub

At the top of the case’s backside is a hardware PWM controller, supporting up to six fans (the maximum supported by this case).

Rear view, Storage support, cable management

Starting at the top left corner, you’ll see the PWM hub. Below it are two drive trays, which can each support two 2.5-inch drives or a single 3.5-inch drive.

If you by chance need more storage, an additional tray (not included) can be installed next to the cable management routing path.

The cable management paths have velcro straps to secure the cables and keep things organized. The cable pass-through paths are lined with a flexible material, and have securing braces (shown above) that can be used to keep cables from moving.

▶️ IO panel

The IO panel is located at the front and top of the case. It features a lighted power button, two USB-A ports, one USB-C port, and an audio jack.

▶️ Dust filters

The BF360 Flow has three dust filters – located in the top (pictured below), front, and bottom (below the PSU).

The dust filter on the bottom of the chassis is independent and easily removed. The other two filters are part of the top panel (shown above) and the stone-textured front panel (shown below), but are also easily removable.

▶️ Rear view

For the most part, if you’ve seen the back of one computer case – you’ve seen most. With this case, you have your standard seven PCI-e slots and exhaust fan, and the PSU slot at the bottom. What stands out in the BF360 Flow is the side panel cutouts or grips, which are used to remove the panels, and a pull tab at up top for removing that panel.

Thermal tests, test setup, and testing methodology

I’ve benchmarked this case primarily to focus on the efficiency of its 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 – 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 noise levels are set to run quietly. This first set of benchmarks examines CPU performance.

With our first noise-normalized test, which stresses only the CPU, the performance was basically “average,” cooling 267.4W compared to an average of 268.8 across the 11 cases we’ve tested this way.

Our second test stresses both the CPU and GPU, to make the case as warm as possible. The added heat of the GPU reduces CPU performance slightly, from 267.4W down to 259.7W.

But let’s not forget: one of the distinguishing features of the BF360 Flow is its air scoop, which directs airflow directly to the GPU.

This engineering design isn’t just for looks – it performs especially well, with the case turning in the second-lowest GPU temperatures we’ve seen in this noise-normalized test. Of course, the large, thick intake fans themselves certainly help thermals all on their own.

Maximum Noise Levels

I measured the noise levels with my sound meter from one meter away from the computer case, with the pre-installed fans running at their maximum RPM speeds, recording a measurement of 46.6 dBA. Personally, I think that’s too loud. But in contrast to other cases on the market, the noise level is about average.

Maximum thermal performance – 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?

Allowing the fans to run at full speed increases CPU performance from 267W to 288W, an increase of about 8%. In contrast to the competition, the BF360 performed a little worse than others, but is only about 10W behind our best result.

Things look better, however, when we add the heat of a GPU to the test. Whereas many cases lose CPU cooling efficiency when the GPU’s heat is added, the impact on the BF360 Flow is almost non-existent. We measured 287.9W, with the HAVN case taking third place in this test.

And when it comes to GPU temperatures, the BF360 Flow shines. With a measurement of only 61.5C, HAVN’s case delivered the best graphics card temperatures we’ve seen in this test yet!

The temperature recorded was 61.8C, which is technically the best we’ve seen, though it is effectively tied (within margin of error) with Montech’s HS02 Pro case here.

One more thing

When HAVN sent along the BF360 Flow for testing, the company also included extra H18 fans. We wanted to see how much (or little) an additional fan might improve performance, and added an H18 exhaust fan above the CPU cooler. For the first test, we set a power limit of 150W for the CPU and also stressed the GPU (consuming 295W).

This test showed a three-degree improvement to CPU temperatures with the addition of an extra fan. GPU temperatures were technically 0.5C higher, but this is a “margin of error” difference, and both results are essentially on par with each other.

For the second test, we removed the power limits from the CPU. The same pattern emerged: slightly better CPU performance, with GPU temperatures essentially the same.

Conclusion

HAVN’s BF360 Flow delivers a highly effective approach to GPU cooling, made possible with its air-scoop intake design and massive 180mm fans. The case consistently produced some of the lowest GPU temperatures we’ve recorded, even outperforming many competitors built for high airflow.

Not everything is perfect with this case, though. The one-year warranty feels underwhelming for a case that’s priced above $150. And the lack of ARGB may limit its appeal to some gamers. But where it counts most, the BF360 Flow excels. Its subdued aesthetic, smart airflow engineering, and chart-topping GPU performance make it a compelling option for users prioritizing cool and consistent graphics card temperatures. If GPU thermals are a top priority, HAVN’s BF360 Flow stands out as one of the strongest options available today.

Design of the Acer Nitro 60

The Nitro 60 sticks to a familiar mid-tower gaming template, built around an all-black chassis with a tempered glass side panel that highlights its internals. Four RGB fans — two front intakes, one on the CPU cooler, and a rear exhaust — provide the expected glow of color. While it certainly looks the part, the design doesn’t differentiate it from the many towers in this price bracket. It’s a welcome demonstration of restraint from Acer, known for bolder and occasionally polarizing designs in its gaming line. This conventional approach provides a sense of normalcy.

The Nitro 60’s stout dimensions of 8.5 x 14.9 x 15.9 inches (WDH) make it smaller than many other mid-towers. Build quality is solid, with clean edges, a blackout interior, and tempered glass that lends a premium feel.

Lighting is divided into two zones: one for the two RGB DIMMs and another for the four fans. You can adjust colors and patterns in the Windows Dynamic Lighting app. Acer includes Gigabyte’s Dynamic Light app, but I wasn’t able to get this to open, preventing me from changing the lighting on the fans or DIMMs. I instead installed Gigabyte’s RGB Fusion, which ultimately enabled independent customization of the fans and DIMMs.

Specifications

Ports and Upgradeability on the Acer Nitro 60

Up front, the Nitro 60 provides two USB 3.2 Type‑A ports, 3.5 mm headphone and microphone jacks, plus power and reset buttons. It’s a decent selection, though the lack of a USB-C port is notable, particularly there is room for another USB port, but the cutout is blocked off. These connections are conveniently placed if the tower sits on a floor, though less so on a desk where cables tend to cascade downward.

Rear I/O includes a USB‑C 3.2 Gen 2x2 port, a USB‑A 3.2 Gen 2 port, four USB 2.0 ports, and three 3.5 mm audio jacks. An old‑school PS/2 port is also present, which feels dated (though it's still fairly common). The motherboard’s HDMI and DisplayPort outputs are disabled since the Core i7‑14700F lacks integrated graphics.

Networking is handled with a 2.5 Gbps Ethernet jack alongside built‑in Wi‑Fi 6 and Bluetooth 5.3. Faster Wi‑Fi 6E or 7 would have been better, but this configuration should suffice for most users. The included wireless antennas must be screwed in.

The side panels detach after you unscrew two thumbscrews for each. The tempered glass panel falls to the side after the screws are removed, so it’s important to support it while doing so.

Inside, the Gigabyte B760M C V2 motherboard and a tall 120 mm CPU air cooler take center stage. It’s notable that Acer used a readily available aftermarket motherboard rather than an OEM or proprietary model. Its expansion is generous, with four DIMM slots, four PCIe slots, and three M.2 slots. Cable management is mostly hidden thanks to the isolated PSU chamber, though the right‑side panel conceals some untidy bundles.

Storage expansion supports one 2.5‑ or 3.5‑inch drive, with SATA cables included. A modular 850 W Thermaltake power supply is a surprising inclusion, as many systems in this price range ship with non‑modular units. Extra cables are provided in the box.

For maintenance, the case features slide‑out dust filters beneath the PSU and on the top panel, simplifying cleaning. The front panel also comes off by tugging the handhold on its bottom edge.

Gaming and Graphics on the Acer Nitro 60

Our Acer Nitro 60 features a Core i7-14700F processor, a GeForce RTX 5070 graphics card, and 32GB of DDR5 RAM. Although the memory is rated for 6,000 MHz, it was only running at 4,800 MHz out of the box, which is how we conducted our testing. I had to access the BIOS to enable XMP1 for it to run at its rated speed.

To test the Nitro 60, I played Call of Duty: Modern Warfare III at 3840 x 2160 at the Extreme preset with DLSS enabled and observed a fluid 90 to 100 fps in most maps, with no stuttering.

For context, we compared the Nitro 60 to several gaming towers. The closest rival is the Asus ROG G700 ($2,029 as tested), which pairs a Core Ultra 7 265KF with the same GeForce RTX 5070. We also included the CyberPowerPC Gamer Supreme ($1,859), built around an AMD Ryzen 7 9800X3D and RTX 5060 Ti, and the iBuyPower Y40 Pro ($1,899) which features a higher-end Ryzen 9 7900X and RTX 5070 Ti.

In Shadow of the Tomb Raider at the game’s Highest detail preset, the Nitro 60 competed with the ROG G700, producing 194 FPS to the latter’s 199 FPS at 1080p, while both produced 68 FPS at 4K. The RTX 5070 Ti-powered iBuyPower led the results, with 228 FPS at 1080p and 89 FPS at 4K, while the CyberPower with its RTX 5060 Ti trailed, with 149 and 46 FPS.

The demanding Cyberpunk 2077 (at the Ray Tracing Ultra preset) saw the desktops in a similar pecking order, with the Nitro 60’s 65 FPS at 1080p just behind the ROG G700’s 70 FPS. Both units also showed similar performance at 4K, with the Nitro’s 21 FPS against the ROG’s 20 FPS. The iBuyPower’s 87 and 29 FPS proved to be the best of the bunch, with the CyberPowerPC trailing with 53 and 15 FPS.

In Far Cry 6 (Ultra), the results were skewed at 1080p, with the CyberPowerPC taking the lead with 138 FPS, the iBuyPower in second place with 125 FPS, and our Nitro 60 narrowly behind it with 121 FPS. The desktops went back to their expected places at 4K, with the Nitro 60’s 81 FPS just ahead of the ROG G700’s 80. It remained behind the iBuyPower (102 FPS) and well ahead of the CyberPowerPC (59 FPS).

In Red Dead Redemption 2 (Medium preset), the Nitro 60’s respective 143 and 54 FPS at 1080p and 4K slightly outperformed the ROG G700’s 120 and 50 FPS, significantly outperformed the CyberPowerPC’s 104 and 37 FPS, and trailed the iBuyPower’s 161 and 66 FPS.

Last, in Black Myth: Wukong (Cinematic preset), the Nitro 60 remained consistent, producing 56 FPS at 1080p and 24 FPS at 4K, which was within one FPS of the ROG G700 at both resolutions. The CyberPowerPC slotted behind, with 41 FPS at 1080p and 17 FPS at 4K, while the iBuyPower’s stronger RTX 5070 Ti propelled it to 66 FPS at 1080p and 32 FPS at 4K.

The Nitro 60 demonstrates strong gaming performance in line with what we expect from an RTX 5070 desktop. Despite using an older Core i7-14700F, its gaming numbers were largely identical to what we saw from the ROG G700 featuring the newer, higher-wattage Core Ultra 7 265KF, particularly at 4K where most games will be GPU-limited. Had Acer enabled the XMP memory settings out of the box, its performance might have been even closer to the ROG G700’s.

To evaluate sustained gaming performance, we run 15 loops of the Metro Exodus stress test at RTX settings. During the approximately 30-minute test, the Nitro 60 produced an average framerate of 121.3 FPS across all runs, with minimal variation between runs suggesting stable thermal performance. The Core i7-14700F maintained an average temperature of 52 degrees Celsius while the GeForce RTX 5070 Ti averaged 61 C. The Nitro 60’s fans were well-behaved throughout testing. While not inaudible, they wouldn't be distracting over typical household background noise.

Productivity Performance on the Acer Nitro 60

Our Nitro 60 features a Core i7-14700F CPU, 32GB of RAM, and a 2TB SSD. While the 65-watt CPU isn’t as formidable as the 125-watt K-series chips, it’s still a formidable performer for almost any task.

In our Geekbench 6 CPU test, the Nitro 60’s Core i7-14700F trailed the single-core results, producing 2,865 points. The iBuyPower did slightly better, with 2,953 from its Ryzen 9 7900X, the ROG G700 produced 3,092 with its Core Ultra 7 265KF, and the CyberPowerPC led with 3,335 courtesy of its Ryzen 7 9800X3D. In multi-core, the Nitro landed mid-pack, its 18,315 points ahead of the iBuyPower (17,226) and just trailing the CyberPowerPC (18,601) but out of range of the ROG G700 (19,057).

Our Handbrake 4K to 1080p video transcoding test saw the Nitro 60 finish dead last, completing the test in 3 minutes and 10 seconds. The others all broke the three-minute mark, with the CyberPowerPC finishing in 2:58, the iBuyPower in 2:29, and the ROG G700 in 2:03. This is the one test where the Nitro 60’s 65-watt CPU is not that adept.

Last, the Nitro 60 produced the best transfer rate of 1,861.6 MBps in our 25 GB file transfer test. The ROG G700 was second (1,816.5) while the CyberPowerPC (1,664.1) and the iBuyPower (1,501.6) picked up the last two spots.

Keyboard and Mouse with the Acer Nitro 60

The Nitro 60 we received did not include any peripherals. This is a mild omission at best, considering bundled peripherals are usually of entry-level quality, which most users will want to replace with a better keyboard and mouse anyway. Some people may even prefer picking out their own accessories.

Software and Warranty

The Nitro 60 ships with relatively little preinstalled software. Acer Care Center is the most practical inclusion, offering system updates and easy access to support resources. Acer also bundles Gigabyte’s Dynamic Light utility for RGB control, though it failed to launch during testing. I installed Gigabyte’s RGB Fusion on my own, which functioned properly and allowed customization of both the RAM modules and fans.

Less welcome is the assortment of bloatware, including a McAfee security trial and several lightweight game demos. Fortunately, these can be quickly uninstalled.

Acer provides a one‑year warranty with the system.

Acer Nitro 60 Configurations

Our review unit, sourced from Micro Center, features an air‑cooled Core i7‑14700F processor, GeForce RTX 5070 graphics, 32GB of RAM, and a 2TB SSD is on sale for $1,599 as of this writing but is $2,099 at normal price..

This configuration is competitively priced with the sale. At Best Buy, desktops with similar RTX 5070 setups—such as models from CyberPowerPC and HP’s Omen 35L—were within $100 of the Nitro 60. It also undercuts the $1,749 Asus ROG G700, which offers comparable specifications but swaps in a Core Ultra 7 265KF.

Entry‑level Nitro 60 models start at $879 at Micro Center, pairing a Core i5‑14400F with an RTX 5060, 16GB of RAM, and a 1TB SSD. A $1,399 tier matches most of our review unit’s specifications but uses an RTX 5060 Ti. For $1,499, buyers can step up to an RTX 5070 paired with a Ryzen 7 7700. The highest‑end configuration, available from Acer’s online store, mirrors our review unit but upgrades to an RTX 5070 Ti for $1,799.

Bottom Line

Acer’s Nitro 60 delivers more than expected for a mid‑tier gaming desktop. With the Core i7‑14700F and GeForce RTX 5070 in our review unit, it proves to be a capable all‑around performer, even stretching into 4K gaming at the right settings. The design has appeal with its tempered glass side panel and the welcome inclusion of a modular power supply, a rarity at this price point.

There are a few shortcomings. Out of the box, the memory runs below its rated speed since XMP settings are not enabled, and no peripherals are included. Networking also tops out at Wi‑Fi 6 rather than newer standards, though this is unlikely to hinder most users.

Competition in this segment is intense. The Asus ROG G700 offers a liquid‑cooled CPU and a more eye-catching design, but at a higher cost. Meanwhile, the Nitro 60 remains price‑competitive against numerous mid‑tower alternatives from Best Buy and Micro Center. Taken together, its performance, design, and value make the Nitro 60 a strong contender for gamers seeking a well‑rounded desktop without overspending.

Creality’s naming system for the K2 lineup is misleading and confusing. The K2 is about the size of a Bambu Lab X1C, but bigger than the first Creality K1, which stuck to the old Ender 3 220 x 220mm build size. The K2 Pro is the size of a K1 Max at 300 x 300 mm, and the K2 Plus is a 350 x 350 mm-sized beast. Confused? So are we.

All that aside, the Creality K2 Pro really shows how far we’ve come since the old Ender 3. Now we expect printers to work out of the box and produce flawless multicolor prints, all without tuning, modding, or going through voluminous test prints with every change of filament. The K2 Pro with its CFS box did just that. It has excellent print quality, even though it produces a lot of purge waste due to its high-flow nozzle. Other than a bit of trouble with TPU, the K2 Pro pleasantly surprised me with a long run of perfect prints and no fails.

The list of features packed into the K2 Pro is impressive. The sturdy die-cast aluminum frame sits on squishy vibration-dampening feet. The machine is fairly quiet with the door closed, and the chamber cooling through the air purifier is good enough in practice to run PLA and PETG without the door open. Chamber heating does take some time to reach its maximum temperature of 60 degrees, but it makes printing ABS and PC (Polycarbonate) a breeze. Dual AI cameras for failure detection and flow compensation, a high flow hotend, step-servos on the X, Y, and extruder, smart auto leveling, which taps only the area to be printed on, and the CFS is the easiest loading filament system on the market.

It retails for $999 as a combo with its four-color CFS and $799 as a standalone. You can also connect up to four CFS units (at $319 each) for 16-color printing. It’s not quite one of the best 3D printers you can buy, but it’s pretty close.

Specifications: Creality K2 Pro

Creality K2 Pro Combo: Included in the Box

The Creality K2 Pro Combo comes with everything you need to get your printer set up. You get tools to maintain the printer, side cutters, a plastic scraper, and a USB stick. There’s also a sample spool of white Creality Hyper PLA to start your printing journey.

You also get a paper copy of the quick installation guide, with QR codes to the Creality Wiki and the Creality Cloud app.

Design of the Creality K2 Pro

The Creality K2 Pro Combo is built like a tank. Its matte-finished die-cast aluminum frame is solid, and the printer by itself tips the scales at over 50 lbs. The glass door and lid with acrylic side panels allow for good viewing of the print area. The interior is well lit, and the print-monitoring camera provides an excellent view of the action, which can be set to run in full screen from the device tab. The primary camera can also create good-quality time-lapse videos, which is nice, while the second camera focuses on the nozzle for flow adjustment and problem detection. Unfortunately, there is no access to the video of the nozzle camera.

Chamber cooling is provided by a rear-mounted exhaust fan with filtration and a chamber fan mounted near the top rear of the printer. When printing lower-temperature filaments, the chamber fan will kick on whenever the chamber reaches 35 °C. The filtration fan runs continuously. There is a large side-mounted fan for part cooling, and a blower fan mounted on the rear of the tool head. The cooling is good enough to run the printer with the door closed for PLA and PETG, which is a personal pet peeve of mine.

The Core XY motion system is driven by step-servo motors on the X and Y axis, running on a linear rail on the X axis and linear rods on the Y axis. Unlike the usual stepper motors, which are designed to move a certain distance on command, step-servo motors provide feedback on how far they actually move. This allows for greater movement precision and alerts the printer to any problems.

The Z axis is driven by a single stepper motor belted to two lead screws, with the bed riding on four linear rods.

The spring steel build plate is single-sided with a textured PEI coating that was nice and sticky from the start. The bed heats up quickly, and the smart-auto leveling only requires the nozzle to tap the area occupied by the print.

The Creality CFS is a well-heeled multicolor system that compares favorably to the Bambu Lab AMS. At this time, it cannot dry filament, but it may be the easiest to load a multicolor system available. The PTFE (Polytetrafluoroethylene) guide tubes make loading effortless, and the filament retracts to within a few inches of the CFS when done printing, making it super easy to swap spools. Creality is still providing extremely short cables to attach the CFS to the printer, which means it can only be set on top of the printer, not next to it on the table. This means powering down and unhooking the cables if you have to take off the lid to get at the tool head.

Thankfully, Creality is still using the same “unicorn” style high flow nozzle for the entire K2 line up. It’s also compatible with the Creality Hi, so hopefully they’ve settled on one size going forward for now.

Like the K2 Plus, the K2 Pro has a single filament spool you can bolt to the side if you didn’t opt for the CFS or have a spool that doesn’t fit inside the box. There is an RFID tag reader on the side of the machine, so it can still read Creality branded filament.

The Core XY motion system can hit acceleration rates of 20,000 mm/s², which is less than the beefy K2 Plus. Rubber feet are included, which help dampen vibrations, though it still jiggles like crazy on some fast prints.

It has the same screen and easy to use menu system as the other core XY printers, with helpful thumbnails supplied by Creality Slicer.

The Creality Filament System (CFS) looks a lot like a Bambu Lab Automatic Material System (AMS), and operates largely the same way. I found it to be much easier to load as there are no stiff buttons to release the feed mechanism tension. It has a similar compartment for desiccant packs to keep filament dry, and a large display on the front of the CFS tells you what the temperature and humidity levels are inside the box. There’s no active heating element in the box, so the temperature fluctuations would only be from sitting on a hot printer.

Assembling the Creality K2 Pro Combo

The Creality K2 Pro arrives fully assembled and only needs to be unboxed and unpacked. There are a few well-marked screws to release the CFS and the bed, some foam to remove and some zip ties to cut. The touch screen slots in without tools after the ribbon cable is attached.

There is a single spool holder that attaches with 2 m3 screws, and the CFS buffer attaches to the rear with two m3 screws. That just leaves the CFS cables and Bowden tube for completion.

Leveling the Creality K2 Pro Combo

The Creality K2 Pro Combo has a smart auto leveling program that uses the nozzle to tap the bed only in the area the current print will be using. Full bed auto leveling is part of the initial calibration and can be refreshed from the calibration options if needed.

Loading Filament on the Creality K2 Pro Combo

Loading and unloading filament using the CFS is super simple. Insert the filament into the PTFE guide tube and the CFS will take it from there. Once the printer is done with the filament it retracts to a point where it is easily removed. Creality filament will be automatically detected and show up on the printer’s screen and in the slicer once you hit the auto mapping option. Other brands of filament will require you to enter the filament type and color from the printer’s touch screen.

Preparing Files / Software for Creality K2 Pro Combo

I downloaded Creality Print 6.3 from Creality.com. It is a custom program based on Orca Slicer and Cura. The slicer is fairly intuitive, but not exactly beginner-friendly. Thankfully, the default settings are pretty good as they are, so you don’t need to poke around too much.

Creality Print connects to Creality Cloud for both downloading models from Creality’s file-sharing site and uploading to your printer. If you don’t like the Cloud, you can avoid it by dropping STL files from your computer into the slicer and transferring them to the printer via LAN or using a USB stick.

The interface has a LOT going on, but once you find all the buttons, it’s as good as Bambu or OrcaSlicer. The “device tab” is a simplified Klipper screen that shows your printer’s file storage, temperatures, motion system, and camera. To access the full Fluidd screen (Klipper UI), type your machine’s IP address followed by :4408.

Creality Slicer now allows you to tweak your purge volumes by color as well as the overall “multiplier,” which is a welcome change. You’ll need to use caution with tuning light colored filaments, as they still require a hefty purge amount due to the size of the high-flow nozzle.

Printing on the Creality K2 Pro Combo

The Creality K2 Pro Combo comes with a partial spool of white Hyper PLA. If you want more colors and materials, you should check out our guide to the best filaments for 3D printing for recommendations.

For the first print, I ran the printer’s presliced four-color Benchy, which printed in 35 minutes using Creality’s Hyper PLA. The Benchy looks fantastic, as you’d expect from a presliced file.

I decided to slice something myself in the same colors. The stock purge volumes and the super long melt zone in the unicorn hotend produce a ton of purge waste, but the results on this four-color Flexi catfish are really, really nice. This file is from FilliFuu on Makerworld. It showed excellent color transition with zero bleed-through going from dark to light colors. A plate of four catfish took 13 and a half hours to print using a .16 layer height and average print speed of 180 mm/s. The result is a very nice print with a nice matte finish using four colors of Creality Hyper PLA.

For PETG, I printed something a bit more involved: a kilo’s worth of Gridfinity storage for hardware. It took three build plates and 28 hours to print. The print came out very nice, with just a little stringing, using a .2mm layer height and average print speed of 170 mm/s using Prusa Jet Black PETG.

The K2 Pro Combo can print TPU (allegedly, but I have yet to see proof of it. I tried multiple TPU filaments, from hard TPU for AMS from Bambu Lab, to my favorite rainbow TPUs from Inland. The hard TPU print did finish, but the result was an under-extruded mess. Bumping up the flow in the slicer made no difference, nor did running from the external spool holder or directly to the tool head.

(An internet search found that others also have difficulty with TPU on Creality K2 Pro, and some found success, but only if you’re willing to replace a spring inside the toolhead.)

Thankfully, with the ability to run 300 °C on the nozzle, 110 °C on the bed, and 60 °C in the chamber, the Creality K2 Pro Combo can print the majority of the more technical filaments out there. It prints ABS well, but my favorite, though slightly persnickety, technical filament is Polycarbonate (PC). The CFS cannot dry filament, but it can prevent further moisture absorption if you keep it shut with fresh desiccant packets in place. It did an excellent job printing dry Polymaker Poly Max PC with no warping and excellent bed adhesion.

One of many prints I did in PC is a handy little Utility Blade Scraper. Using a .2 mm layer height and an average speed of 100 mm/s, the scraper took a little under 30 minutes to print, and it is both a pretty and functional print.

Bottom Line

The Creality K2 Pro Combo is a send-it-and-forget-it printer for anything other than TPU. It has Core XY speed and produces excellent quality prints. It’s easy to load, levels perfectly, and features dual AI cameras. Retailing for $999, the K2 Pro Combo is not cheap, but it makes up for it with its extra 300 mm³ of “helmet class” build volume.

Ultimately, the K2 Pro Combo lands in a sweet spot: larger and more refined than entry-level machines, but easier to live with than the heavyweight K2 Plus. It’s not a revolution, but it’s Creality’s most polished and capable Core XY to date.

Competition among Core XY printers is getting fierce. If you want more bang for your buck, there’s always the smaller 250 x 250 x 250 mm Anycubic Kobra S1 Combo on sale for $459, or the Bambu Lab P1S Combo, which has been marked down to $549 on the anticipation of the P2S launch in the US. If you want an even bigger color Core XY, check out the Bambu Lab H2S with a 340 x 320 x 340 mm print volume for $1,499.

That said, we're here to answer the question: Is the best 4K webcam actually two 4K webcams?

The Piko+ features a 1/2.55-inch Sony sensor with an aperture of f/1.8, and records video at 4K / 30fps or 1080p / 60fps. It has a 73-degree field of view, 1.5x digital zoom, and features two lenses — one for imaging and one to help out the webcam's AI with focusing and metering. It also has three built-in noise-cancelling microphones and works with Emeet Studio, which lets you link and switch between multiple cameras (complete with multi-device audio mixing). Perhaps the most enticing thing about the Piko+ is its price — its MSRP is listed at $95.99, which is already pretty good for a 4K webcam, and it's currently $16 off at Emeet, bringing it down to just $79.99.

Design

The Piko+ is smaller than most webcams — but, thanks to its design, it’s also higher-profile. This small, oval-shaped webcam measures 2.2 inches (56mm) high by 1.69 inches (43mm) wide, and is 0.87 inches (22mm) deep. On the front of the webcam, you'll see two lenses under a round glass cover — similar to what you see on the back of most smartphones these days. The Piko+ has two lenses: a primary lens (used for imaging) and an AI lens, which is used to help with focus, face detection, and metering. The Piko+ comes in three colorways — black, white, and mint green (our review unit came in black). There's a small indicator light, right under the dual lenses, that glows blue when the camera is streaming.

The webcam has a glossy finish that's pretty fingerprint-prone — not that you'll be touching your webcam too often. However, it was noticeable enough that I did start to wish the company had sent over a lighter color, where fingerprints would be less noticeable.

The back of the webcam features a USB-C port, and the bottom has a 1/4-inch threaded mounting point for attaching the webcam to a standard monitor mount or tripod.

The Piko doesn't have a built-in privacy shutter, but it does come with this cute, rubbery panda-wearing-sunglasses lens cover thing, that attaches to the Piko's lens magnetically. The lens cover is cute and it works pretty well — the magnet isn't very strong, so you can easily remove it from the webcam without moving the camera itself — but it is a completely separate piece. I don't love this, because it means it's easy to lose. I don't know about you, but I definitely misplace things often enough that I know it's only a matter of time before this goes missing. It's also not terribly useful for anything other than sitting on your desk — the moment you toss the Piko+ into a bag or, I don't know, lightly brush it against anything, the privacy cover falls off. Webcam privacy covers aren't necessarily designed to protect the lens, but most of them do — this little panda is cute, but gimmicky.

My review unit came with a lightweight aluminum tripod in lieu of a traditional monitor mount. I don't mind a webcam tripod, but everything about this tripod just feels cheap. It has three legs that measure 3.76 inches (95.5mm) and extend 2.44 inches ( 62mm) for a total length of 6.16 inches (156.4mm). With the legs fully extended and spread, the tripod lifts the webcam about 7 inches (177.8mm) off the surface of your desk. The tripod also tilts and swivels, so it shouldn't be too hard to keep yourself in-frame.

I like the idea of a webcam tripod, but not as a primary mount — it's hard to find a place on your desk for something like this, even if your desk is clutter-free. This tripod is also very flimsy and easy to knock over (I know this because I kept knocking it over). Anyway, I'm not the only one who doesn't particularly care for webcam tripods, and the Piko+ now ships with a clamp-style monitor mount instead of a tripod.

The thing about the Piko+ is that, at 2.2 inches (56mm) high, it's taller than most webcams. This might not be a problem for most people, but I have four monitors — two horizontally stacked in the middle, with a vertical 27-inch on each side — and the Piko+ adds just enough extra height to interfere with my top screen. I realize this is due entirely to my unique monitor setup, but I'm sure I'm not the only one who prefers a low-profile webcam.

Speaking of which, the Piko+ comes with just a handful of accessories: a 5-foot (1.5m) USB-C to USB-C cable with a USB-C to USB-A adapter, a small desktop tripod (now a clamp-style monitor mount), and a magnetic, panda-shaped lens cover.

SPECS

Well-lit Performance

Our well-lit webcam scenario is very well-lit — more well-lit than the average well-lit lighting scenario. Most webcams perform nicely in a well-lit setting, and this is usually where 4K webcams really shine — especially when it comes to all that extra detail. The Piko+'s performance was just okay in this scenario. The webcam did a decent job with color balancing, but the default settings looked a little overexposed. Also, getting good picture quality was tricky — the autofocus worked well enough for video, but couldn't keep me in focus long enough to get a good still photo.

The Piko+'s well-lit image wasn't nearly as crisply detailed as the images from other 4K webcams, such as the Logitech MX Brio, Elgato Facecam Pro, and Razer Kiyo Pro Ultra, but this seems to be partly due to the dual lenses attempting to autofocus and create artificial depth of focus. Too much light meant too much to try to focus on, and the algorithm ended up kind of just putting everything in focus — but it also uses the AI lens to apply a pseudo depth of focus effect, and it ended up also applying this effect (mildly) to the whole picture.

This also explains why the Piko+'s image looks flatter and less dynamic than the pictures from other 4K webcams. In videos, however, the AI lens did a pretty good job of keeping a moving subject in focus, though the depth-of-field effect was still a little touchy.

(I should point out that the default Windows Camera app doesn't support 4K resolution, so unless a 4K webcam comes with its own app for taking photos, it can be tricky to get good still pictures of the webcam's image quality.)

Low-Light Performance

Our low-light scenario is basically just darkness — the only source of light in the room is one 27-inch monitor, set to 70% brightness, with a dark wallpaper and the Google homepage open in a browser window on half the screen. The Piko+ performed adequately in this lighting scenario — it tried to light the scene evenly, but it was still pretty dark overall (not to mention very grainy/noisy). It's pretty clear that I'm sitting in a dark room, and the Piko+ doesn't even try to make up for that.

Overexposed Performance

Our overexposed lighting scenario might seem unusual, but it's actually more common than you might expect — most people put lights everywhere except behind their monitor/laptop screen (screens are backlit for a reason). So, while you won't necessarily end up in a scenario manufactured to be overexposed, like ours — which consists of a 27-inch monitor with its brightness set to 70% and a screen that's split between a dark wallpaper and relatively white webpage (Google home page) and a bunch of overhead and background lighting — the average home or work office lighting scenario leans toward overexposed.

The Piko+ performed relatively well in our overexposed lighting scenario. Like just about every webcam, the Piko+ struggled to adjust the light balance between the subject and the background. The subject is fairly dark, and the background is still overexposed, but it's more balanced than the Logitech MX Brio — which lights the subject well but blows out the background. The Razer Kiyo Pro Ultra lights the background nicely, at the expense of the subject.

Features and Software

The Piko+ works with Emeet's combination peripheral / streaming software, Emeet Studio. You'll find the standard webcam settings in Emeet Studio — image adjustments and auto/manual exposure, white balance, and focus. The software also has some streamer-oriented features, including the ability to set up multiple linked cameras and switch between them (e.g., for switching between angles/scenes/close-ups). You will, of course, need more than one webcam to set up multiple cameras, which is why Emeet also sells a three-camera MultiCam Bundle (currently $209.99 — $69.99 each).

The Piko+ has three built-in noise-canceling microphones, which sound better than most webcam microphones — but that's not saying much. I was impressed with how well the Piko+'s mics picked up my voice: I was still pretty loud and clear even when I was a couple of feet away and facing a different direction while speaking. But don't be fooled into thinking this means the Piko+'s microphones are good — they're still webcam mics, after all. Voices are loud and clear, but sound a bit thin and hollow, a pretty common issue with webcam mics. I wouldn't recommend anyone actually use the mics in the Piko+ for anything real. But if you happen to be in an emergency that can only be solved with a webcam microphone, the mics in the Piko+ are better than most.

The Piko+ also has some other features, such as whiteboard mode — which autofocuses on the text or diagrams on the "whiteboard" behind you, and AI-powered autofocusing and light metering.

Bottom Line

The Emeet Piko+ packs a decent amount of power in a small package. It features two 4K lenses and uses AI to improve images (or, well, at least.... try to). I'm glad to hear the company scrapped the wobbly tripod for an actual monitor mount, and the sale is tempting me a little.

While we definitely got 4K video quality from the Piko+, there's more to image quality than just resolution — from the lens and sensor size to firmware updates and software workarounds/adjustments. The Piko+'s AI lens did a decent job of keeping a moving subject in focus, but it was a bit too finicky if you weren't moving all over the place.

So don't run out and buy this webcam just because it's relatively budget-priced — it comes down to what you're going to be using it for. If you'll be presenting or lecturing or otherwise moving around, the Emeet Piko+ is a pretty good budget alternative to fancier AI-enhanced action webcams, such as the Insta360 Link. But if you'll mainly be sitting in front of your PC and streaming, and you're looking for impressive 4K image quality, I still recommend the Razer Kiyo Pro Ultra.

This product slots in under the tri-band Archer GE550, GE650, and GE800. The Archer GE400 features a wealth of ports, along with a host of software features that cater to gamers looking to get that extra edge during fierce online competitions. While all of this sounds good on paper, we found that the Archer GE400 had spotty performance on our wireless tests, which made us question its value proposition against the cheaper Archer BE3600, which is one of the best Wi-Fi routers.

Design of the TP-Link Archer GE400 Gaming Router

For a wireless router, the Archer GE400 has a bold design with sharp lines and six movable antennas to improve wireless transmission. There are triangular banks of ventilation slats on the top of the router, and a mesh ventilation network on the bottom. However, the eye-catching feature of the Archer GE400 is a single, vertical LED strip mounted on the top of the router. It’s configurable using either the web GUI or TP-Link’s Tether smartphone app for its routers. You can turn off the LED strip altogether or choose from several lighting patterns.

TP-Link includes a USB 3.0 port on the side of the router for connecting storage devices, but the rest of the ports are on the back. Here (from left to right), you’ll find a button to turn off the LED strip, a reset button, a 2.5 GbE WAN port, a 2.5 GbE gaming port, three GbE LAN ports, the power port, and the on/off button.

There are two mounting points on the bottom of the router for attaching it to the wall.

TP-Link Archer GE400 Gaming Router Specifications

Setting up the TP-Link Archer GE400

The Archer GE400 can be initialized using either a web GUI (in a desktop browser) or the TP-Link Tether app. To speed things along, I opted to use the Tether app due to its ease of use. Once I opened the app, I tapped the “+” symbol in the top-right corner and chose to add a standard router. The setup program then asked me to scan the QR code on the bottom of the router with my smartphone’s camera. Next, my phone asked me to join the temporary wireless network to set up the Archer GE400.

From there, things were pretty simple: I was tasked with creating an admin password to log in to the router and choosing an SSID. By default, the Archer GE400 enabled Smart Connect, which combines the 2.4 GHz and 5 GHz bands under a single SSID (I chose Archer_GE400). More on that later.

Finally, I was given the option to turn on automatic firmware updates for the router. The entire process took less than 5 minutes, with the required steps clearly outlined and explained for novices.

TP-Link Archer GE400 Gaming Router Software

Although I used the Tether app to set up the Archer GE400, I switched to the web GUI for the rest of my testing. The interface's overall aesthetic is pleasing, dominated by black and red. This is a stark departure from the white and blue design theme that you’ll find on non-gaming Deco and Archer routers from TP-Link.

There are six tabs across the top of the GUI: Network Map, Game Center, Internet, Wireless, HomeShield, and Advanced. Network Map outlines your current connection type, internet IP address, and current router uptime. You’ll also see an overview of the currently enabled SSIDs, CPU/memory loads, and a visual representation of which physical WAN/LAN ports are active. Finally, a Connect Clients feature shows you which devices are currently connected to the router and how they are connected (wired, 2.4 GHz, 5 GHz).

The Game Center tab is what separates the Archer GE400 from TP-Link's other, more basic dual-band Wi-Fi 7 routers like the Archer BE3600. The primary Game Panel shows you current network traffic, CPU/memory load, connected USB devices, and allows you to configure the RGB strip on the top of the router.

Delving deeper into the submenus, there are options for Game Application Boost (which lets you specify the total bandwidth you want to dedicate to gaming devices), Game Server Acceleration via WTFast, and traffic priority per LAN port. Other features include Game Port Forwarding, network traffic statistics for the games you play, and Game Detector, which shows you ping times in real-time for popular game servers.

The Internet tab is basic, as it only shows your current connection type and allows you to clone a MAC address for the router.

The Wireless tab lets you control all the available SSIDs on the Archer GE400. By default, the Archer GE400 enabled Smart Connect, which combines the 2.4 GHz and 5 GHz bands into a single SSID and routes devices to the optimal band. I prefer to keep a separate SSID for each band, so I disabled Smart Connect and created two SSIDs: Archer_GE400 for the 2.4 GHz band and Archer_GE400_5G for the 5 GHz band. You can also create an MLO network, allowing you to simultaneously send and receive data on the 2.4 GHz and 5 GHz bands. Finally, TP-Link will enable you to create separate Guest and IoT networks — you can configure either to allow only 2.4 GHz traffic, only 5 GHz traffic, or both.

The HomeShield tab includes a Network Check function that scans your network for security issues. Parental Controls are also available which you can specify per device.

Finally, the Advanced tab is where you’ll find more granular control of features that gaming enthusiasts might want to tweak. Many of the functions from the previous five main tabs are also duplicated here, but you’ll also find other additions like SpeedTest, an option to create an EasyMesh network with another supported router, VPN integration, firmware updates, backup/restore of router settings, and a router reboot schedule (among many other features).

TP-Link Archer GE400 Gaming Router Performance

I tested the Archer GE400 using our tried and true, purpose-built testing client rig, which includes an MSI Pro B650M-A Wi-Fi motherboard, AMD Ryzen 5 7600 processor, 32GB of DDR5 memory, a 1TB PCIe 4.0 SSD, an MSI Herald-BE Wi-Fi 7 adapter (Qualcomm NCM865), and Windows 11 Home. Windows 11 has all the latest software updates installed, while the Herald-BE is running the latest drivers available from MSI’s support website. The server is a Windows 11 Home machine with a 10 GbE network card connected to the 2.5 GbE Game LAN port on the Archer GE400 router.

As always, benchmark results may vary widely depending on the layout of your home, apartment, or office. Home construction, wall thickness, the choice of materials in the flooring, and even the number of walls between the router and clients can affect performance. So, while the results that I present are consistent, given that the testing locations and methodology that I use in my home don’t change, they might not align with what you might see in your specific home or office environment.

Our iPerf3 throughput tests are conducted at six-foot and 25-foot distances, with and without traffic on the network. In the congested traffic tests, we add six wireless clients streaming 4K YouTube videos across multiple wireless bands.

One word sums up the performance I was able to extract from the Archer GE400: disappointing. When it comes to baseline wireless performance, the Archer GE400 should offer comparable performance to the dual-band Archer BE3600. However, our results showed the Archer GE400 lagging behind its cheaper sibling in nearly every metric.

The Archer GE400 brought up the rear of the pack in our 5GHz iPerf3 test with no additional network traffic. The router achieved 951 Mbps at close range (6 feet), which put it nearly 200 Mbps behind the Archer BE3600 and even further behind the Asus RT-BE86U (1,473 Mbps) and the RT-BE88U (1,256 Mbps). It wasn't much better at 25 feet, though the Archer GE400 managed 552 Mbps, compared to just 382 Mbps for the sub-$100 Archer BE3600.

Things looked slightly better for the Archer GE400 with the 2.4 GHz test, where it hit 135 Mbps at 6 feet, putting it in second place behind the Archer BE3600 (214 Mbps). However, when stretching out to 25 feet, the Archer GE400 lagged all competitors at 36 Mbps.

In our iPerf3 test, with congested traffic, the Archer GE400 secured third place with 882 Mbps at 6 feet and even came in first place (478 Mbps) at 25 feet. It was another mixed bag with congested traffic on the 2.4 GHz band -- the Archer GE400 snagged second place with 102 Mbps, while the Archer BE3600 led the field at 205 Mbps. However, the Archer GE400's fickleness reared its ugly head again at 25 feet, where it finished last with 31 Mbps.

Bottom Line

The TP-Link Archer GE400 features an angular design and RGB lighting meant to appeal to gamers, and includes a 2.5 GbE WAN and an additional dedicated 2.5 GbE gaming port. TP-Link also provides a comprehensive suite of gaming-related enhancements that can be configured via the Tether smartphone app or the web GUI.

However, those pluses clash with its real-world wireless performance, which was a mixed bag. Short-range performance on the 5 GHz band lagged its rivals, although the tables turned at longer distances. Conversely, the Archer GE400 showed relatively strong performance at short range on the 2.4 GHz band but faltered at long range.

The biggest knock against the Archer GE400, however, is pricing. The router has an MSRP of $219.99, but is currently available at Amazon for $169.99 after a $50 instant discount. For comparison, the Archer BE3600, which offers faster and more consistent wireless performance, is available for just $89.99. If you can look past the fancy RGB lighting and don’t need the gaming-specific tweaks, it’s hard to justify the $80 price differential.

And despite some recent troubles in Europe, the company is still prodicing new products, like the Big Shuriken 4 we’re looking at today. Despite the name, this cooler has a petite profile designed to be compatible with SFF Mini-ITX builds. The diminutive design packs six heatpipes and features the new Kaze Flex II 120AH fan to move heat away from the processor below.

Will this low-profile cooler limbo its way onto our best CPU cooler list? We’ll cover the features and benchmark results, then you can decide for yourself if it is worth purchasing.

Cooler specifications

Packing and included contents

Both Scythe’s Big Shuriken 4 and the smaller Shuriken 3 arrive in small blue and white packages, just large enough to protect and house the cooler.

The CPU cooler and fan are protected with molded foam, and the accessories are housed in a small brown box.

Included with the product are the following:

• CPU heatsink
• Kaze Flex 120 II Slim AH PWM fan
• Small tube of thermal paste
• Mounting for AMD and Intel platforms
• Installation manual

Features of Scythe Big Shuriken 4

▶️ Slim 67mm height for SFF builds

This cooler is designed for the smallest Mini-ITX SFF systems on the market, with a height of just 67mm.

▶️ Six copper heatpipes

To transfer heat from the CPU to the heatsink’s fins, the Big Shuriken 4 incorporates six standard-sized copper heatpipes.

▶️ Full RAM compatibility

The cooler does not interfere with or overhang DIMM slots. All sizes of RAM, regardless of height, are compatible with this cooler.

▶️ Kaze Flex 120 II Slim AH PWM

Scythe includes its newest 15mm (height) fan, the Kaze Flex 120 II Slim AH, with the Big Shuriken 4.

AM5 installation

1. AMD users will first have to remove the default mounting mechanism and then place the grey standoffs around the CPU.

2. Take the mounting brackets and place them on top of the standoffs, using the included screws to secure the brackets.

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. Press the heatsink against the mounting bars and use a screwdriver to secure it through the holes on top of the heatsink (in red below). You’ll want a thin screwdriver for this purpose, as larger ones may not fit.

5. The last step is to attach the fan’s PWM cord to the corresponding motherboard header.

Real-world testing configuration – AMD AM5 platform

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

That being said, I am using my standardized testing setup for this review, which features a full-size case and motherboard. As with every review, your mileage may vary, as every case and its airflow are different. However, you may notice larger differences here, as I’m not testing in a Mini-ITX case.

Today’s test bench has yielded more limited results than normal, as our 9950X3D system has suffered an unfortunate failure. As such, we’ve begun testing coolers on a system with AMD’s Ryzen 9 9950X.

There are some differences in how thermal events impact the 9950X and 9950X3D CPUs. While the heat output of the CCDs in AMD’s 9950X3D is relatively balanced, the 9950X I’m using has one CCD that runs much hotter than the other, with a difference of over 10 degrees Celsius in some scenarios.

One of the purposes of today’s review is to establish how well, or poorly, a smaller SFF CPU coolers handle AMD’s Ryzen 9 9950X processor. While this review will focus on the Big Shuriken 4, we’ve tested and will also go over the results of the smaller Scythe Shuriken 3 CPU cooler.

Maximum noise levels

In terms of maximum volume, both Scythe’s Big Shuriken 4 and Shuriken 3 had relatively low noise levels, measuring 39.6 and 40.3 dBA, respectively.

PBO Performance

Enabling PBO allows AMD’s Ryzen 9 9950X to stretch its legs and guzzle power, but those legs won’t stretch very far with SFF cooling. Despite the limitations of a small heatsink, the Big Shuriken 4 is still able to handle 90% of this CPU’s default power limit in Cinebench R23.

The smaller Shuriken 3 doesn’t fare as well, dropping to 134W. But that’s kind of impressive, considering that cooler is about the size of a pack of cigarettes.

150W

For the next thermal test, I’ve set the power limit to 150W. This is about the highest power limit I would recommend for users considering the Big Shuriken 4 to keep their CPU under TJ Max, as the average temperature was measured at 79.5 degrees Celsius.

The smaller Shuriken 3 isn’t able to handle this workload, so it has been omitted from the results in this chart.

150W Noise measurements

Now, as the 150W thermal measurements demonstrated above are similar to what users might encounter in gaming scenarios, I’ve measured the noise levels of coolers in this benchmark to see how loud they get in everyday situations.

Scythe’s Big Shuriken 4 measured 39.3 dBA, a moderately low volume level.

100W thermal results

Our next test is the least-demanding benchmark featured in this review, with a workload that consumes only 100W. With this lower power limit in place, both of Scythe’s latest SFF coolers can handle the CPU with relative ease. Scythe’s Big Shuriken 4 reached a temperature of 62.7 degrees Celsius, only 5-6 degrees behind traditional single tower air coolers.

Noise-normalized testing

While the preceding results used the default fan curve of my MSI x870E Carbon motherboard, I understand many readers prefer data where cooler noise levels are equalized. For this review, two noise-normalized tests were conducted. The first test was run with Precision Boost Overdrive (PBO) active to allow for the highest possible power consumption.

Looking at this first round of results, what stands out to me is how little performance is lost when the CPU’s fans are set to 39.6 dBA and the system fans are reduced from 60% to 25% PWM.

The Big Shuriken 4 cooled 178.7W with fans set normally, but only lost 1.2W of cooling power – practically a margin of error difference when noise levels were lowered and standardized.

The other noise-normalized test we have for this review also stresses the CPU to its maximum, as above, but also adds the heat of a GPU to the mix – our MSI RTX 4070 Ti Super consumes ~295W.

Conclusion

Scythe’s Big Shuriken 4 is a worthwhile choice for builders targeting SFF systems. Its slim 67mm profile, combined with low-noise operation, delivers cooling performance superior to that of smaller SFF alternatives. It’s a compelling balance of price, performance, and size for any system where space is at a premium.

Many of the latest ultra-wide screens are OLEDs, and they live in a premium price category north of $1,000. If you’re willing to go with VA technology, Samsung has a $900 proposition: the Odyssey G7 S40FG75. This 40-inch, 21:9, 1000R curved panel has 5120x2160 resolution, 180 Hz, Adaptive-Sync HDR, and a wide color gamut. Let’s take a look.

Samsung Odyssey G7 S40FG75 Specs

A 1000R curve means that a circle of Odyssey G7s is just two meters across. You can sit close to this one thanks to its 140ppi pixel density, well above QHD resolution. Of course, moving that many pixels takes time, so the maximum refresh rate is 180 Hz. That’s plenty quick, and though you can find higher rates in the 16:9 aspect category, the Odyssey G7 is faster than most of the ultra-wides I’ve tested. And most of them are 34 inches diagonal with lower resolution. The 40-inch genre is a small group.

The panel uses Vertical Alignment (VA) tech, providing the highest contrast of any LCD variant. It doesn’t quite have the viewing angles of IPS, but it’s far better than old-school TN panels. The Odyssey G7 features an edge LED backlight with eight dimming zones, providing a nice contrast boost in HDR mode. The native ratio is around 2,500:1, but I measured over 9,000:1 for HDR content, with a peak output of around 550 nits. The color is there too, with over 88% coverage of DCI-P3. I also discovered excellent out-of-the-box accuracy, with no need for calibration.

Gaming aids are a bit sparse; you get aiming points and a black level adjuster, but no sniper mode, frame counter, or timers. The overdrive is well implemented, with precise motion-blur removal, but I found it unavailable when Adaptive-Sync is enabled. You must choose one or the other. However, if you can maintain 180 fps, you can turn off AS without penalty.

There are some nice extras here, like LED lighting with color and effect control, and USB ports that face backwards for easy hookup of peripherals. Inputs include DisplayPort 1.4 and HDMI 2.1. and you get a headphone jack. There are no internal speakers. An intuitive nav pad sits under the center of the panel for easy access to the comprehensive menu system.

The Odyssey G7 is a solid gaming package that promises an immersive experience for less than the price of a premium OLED. It also has everything needed for high performance gaming and a bright and colorful picture. And if an OLED is out of h, this makes a great alternative.

Assembly and Accessories

The Odyssey G7 comes in a coffin-like box with its contents protected by crumbly foam. The curvaceous panel is well secured in a large, molded block. The chance of damage is small. The solid metal base and stout upright snap together, then the panel hooks on for a toolless assembly. A large external power supply comes with right-angle plugs, and you get USB, DisplayPort and HDMI cables. A trim ring covers up the panel’s attachment point and is part of the lighting effect.

Product 360

The Odyssey G7 makes a bold statement with its extreme 1000R curve and three-foot width. You will need some space to accommodate that 40-inch screen. The base is relatively small but super solid and weighty as it’s made from a thick aluminum plate. The upright snaps on and offers a 4.7-inch height adjustment, 5/20 degrees tilt and 20 degrees swivel in either direction. Movements are firm and precise, the way a premium monitor should be.

The only branding in front is a super tiny “Samsung” screened onto an equally tiny bulge at the bottom center. This is where you’ll find the nav pad OSD control. It isn’t a joystick, but it works like one with four directional keys and a center selector. They’re rubberized and very easy to operate. They control all monitor functions.

The back is mostly smooth except for a ribbed band across the bottom. You can see that the inputs face back instead of down. Though it’s a little harder to hide cables, it’s a lot easier to make and change connections. You can manage the wiring with a small rubber band on the upright. In the photo, you can see the LED lighting that rings the fulcrum in different color and effects that are found in the OSD.

The top and side photos show just how extreme the Odyssey G7 is. It’s super curvy, but the image is rendered with minimal distortion. The 40-inch size and 1000R shape strike a good balance between immersion and clarity. The picture is super sharp with no visible pixel structure or grain from the anti-glare layer. 140ppi is nearly as dense as an Ultra HD monitor.

The inputs include two HDMI 2.1 and one DisplayPort 1.4. You also get USB 3.2 with one upstream and two downstream ports. A 3.5mm jack accommodates headphones or powered speakers. There are none inside.

OSD Features

The Odyssey G7’s OSD has a cool dashboardy look with signal info at the top and a logically laid out menu tree. Press the center button to summon it.

In the Game menu there is a three-level overdrive, Adaptive-Sync toggle, shadow detail enhancer, and a selection of aiming points. I noted that you can’t have Adaptive-Sync and overdrive at the same time, which is unusual. However, if you can keep the action humming at 180 fps, that isn’t crucial. There is no backlight strobe feature either.

The LEDs are called Infinity Core, and they come with multiple color and effect options. All the lighting is in a ring around the back that encircles the stand’s attachment point.

In the picture menu, you get 10 image modes along with all the basic and advanced controls for calibration, including color temp, gamma, and a color gamut selector. You also get local dimming in three levels. On high, you’ll see around 9,000:1 in SDR and HDR. It’s an edge zone proposition and though Samsung doesn’t specify the number, I counted eight zones when viewing a moving test pattern.

For HDR signals, you retain most of the picture adjustments and you can engage dynamic tone mapping, which is very effective. This is a rare feature that should be included on every HDR monitor. It goes beyond the static tone mapping found in HDR10 content and ensures that the display shows the full dynamic range of the material. It significantly improves the HDR experience.

A 21:9 monitor is perfect for PBP and PIP, so you can view two video sources at once. The Odyssey G7 has options for window position and aspect ratio, as well as audio source and sizing.

Samsung Odyssey G7 S40FG75 Calibration Settings

The Odyssey G7 can be enjoyed without calibration in its Eco mode. Despite the name, it does not limit brightness, and it’s fully adjustable. I tweaked the RGB sliders, which are very precise, and adjusted gamma, to get a tiny gain in quality. The main improvement was slightly more color saturation, which is certainly a good thing. My settings are shown below. For HDR content, I recommend turning on the Dynamic Tone Mapping and setting Local Dimming to High.

Gaming and Hands-on

I have played games on every type of monitor, and no format has the in-your-face realism of an extremely curved ultra-wide. The wraparound effect is truly a thing, super effective. Not only was my peripheral vision completely covered, but the enemies were larger and more invasive of my personal space, virtually that is. Even 3D displays don’t impart that feeling. And comparing the Odyssey G7 to 34-inch screens is almost no comparison. Six inches doesn’t sound like much, but when you add in the 1000R curve, it is almost night and day.

The Odyssey G7 won’t give you the perfect motion resolution of an OLED, but its overdrive was excellent enough that I preferred it over Adaptive-Sync. You’ll need a fast video card to accomplish this since tear-free operation only comes when the frame rate is locked at maximum. A GeForce RTX 4090 provided enough juice to keep the action at 180fps. If you go slower, AS might be the better choice. This said, Samsung should make it possible to use overdrive and AS at the same time.

HDR was superb in every respect. Though you won’t find as much saturation as a QD-OLED here, the Odyssey G7 still makes a very colorful impression with accuracy that is plain to see. It looked great whether I calibrated or not, but I saw just a tiny bit more saturation after tweaking the RGB sliders. Check out my settings above if you want the best possible picture.

Functionally, the Odyssey G7 was a pleasure to use. The OSD is intuitive and easy to navigate using the pad-style buttons underneath. I also appreciated the rear-facing inputs. This is a small thing, but it really does make wiring easier. And it’s something Samsung almost always does, and almost everybody else doesn’t. The only other brand that commonly features rear-facing inputs is LG. I appreciated the inclusion of USB ports. A type B means you can make the Odyssey G7 a hub for two other peripheral devices. However, there’s no KVM functionality. I briefly missed the internal speakers (there aren’t any) until I plugged in a pair of headphones and remembered that even cheap cans sound better than nearly all monitor speakers.

The Odyssey G7 is very useful day to day thanks to its PBP capability. The 21:9 aspect ratio means you can have two rectangular windows showing two sources, which is great for a PC and a gaming console, for instance. Working on documents is best done in the center third of the screen. Placing windows on the sides shows a little distortion, though it was never a distraction.

Takeaway: The Odyssey G7 is a super cool monitor with excellent gaming performance and a vivid image. For well below the cost of an OLED, you get about 90% of the experience. Its tight curve creates an immersive effect that can’t be equaled by a less curved or flat screen. And its 40-inch size puts the game right in your lap, and, in your face. It’s one of the more addictive screens I’ve had the pleasure of playing and working on.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

It was hard to find five other ultra-wide monitors that weren’t OLEDs to compare with the Odyssey G7. They are Titan Army’s C34A1R, Gigabyte’s G34WQCP and GS34WQC, Xiaomi’s G34WQi, and HP’s Omen 34c.

Pixel Response and Input Lag

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

180 Hz doesn’t seem to be quicker than 165 Hz in the response test. 6ms is plenty fast, and it will deliver motion that is almost free of blur. Add in the overdrive and you’ll get a smooth experience from any of these screens. The Odyssey G7 has excellent video processing, but it cannot use Adaptive-Sync and overdrive simultaneously.

In the input lag test, the Odyssey G7 is up with the top panels at 23ms total. This is respectable for any monitor running below 200 Hz. You can spend more money for more Hertz and get a marginal gain, but in practice, the difference between 23 and 20ms is nearly imperceptible. To mitigate the loss of Adaptive-Sync when the overdrive is engaged, a faster system is key, so you can keep the frame rate maxed.

Test Takeaway: Aside from the inability to run overdrive and Adaptive-Sync together, the Odyssey G7 delivers excellent video processing and very quick response. The OD effectively eliminates motion blur, and the play experience is very smooth. Input lag is lower than most of the ultra-wides I’ve tested, making this Samsung one of the fastest in its class.

Viewing Angles

The Odyssey G7 is one of the better VA panels I’ve photographed for the viewing angle test. You can see a slight red shift at 45 degrees to the side, but light output stays consistent while gamma gets a tad lighter. From the top, gamma is much lighter, and the red shift is more obvious. This is excellent performance.

Screen Uniformity

To learn how we measure screen uniformity, click here.

I noted slightly visible hotspots in the corners of my Odyssey G7 sample. They didn’t affect actual content, and they weren’t visible in test patterns above zero percent brightness. I’ve noted a greater range of results in this test among ultra-wide curved panels so the Odyssey G7’s is typical.

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

In SDR mode, with local dimming turned off, I measured a peak of 341 nits. That’s plenty bright for such a large panel, but if you need even more output, the top four screens have a little extra. For indoor use, you’ll never need to max the brightness slider unless you wear sunglasses while playing. In the black level and contrast test, the Odyssey G7 is solid for a VA screen but has a little less dynamic range than the others. However, it’s still visibly better than any IPS monitor.

After Calibration to 200 nits

The Odyssey G7 maintains consistent contrast throughout its brightness range with a solid 2,516.9:1 score after calibration. Don’t let the last-place finish concern you; it’s still a dramatic screen with more depth than any IPS display can boast. In the ANSI test, I observed consistent results with a ratio of 2,300.9:1. This indicates Samsung has made sound component selections and is maintaining excellent quality control.

Test Takeaway: The Odyssey G7 shows average brightness and contrast performance for its category. It doesn’t quite reach the 3,000:1 average I’ve observed for VA panels, but it comes close at around 2,600:1, more than IPS by almost 3x. It’s the closest you’ll get to an OLED for under $1,000.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

The Odyssey G7 comes out of the box set to Eco mode, which, despite the name, does not limit brightness or adjustability. In fact, it’s spot-on for color and gamma and doesn’t need calibrating. Samsung includes a factory calibration report in the monitor’s firmware, which you can view from the OSD.

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.

The default grayscale chart is about as good as it gets. Grayscale errors are all well below 2dE, with the bottom end under 1dE. That’s pro-level performance. Gamma runs a tad light, but tracks straight, which means there is no clipping of detail anywhere in the brightness range.

I tweaked the RGB sliders to see if I could make any impact and saw a tiny gain in grayscale accuracy. Going for one click darker gamma brings all but the 90% step up to the 2.2 reference. That point is still too light, though the error will be hard to spot in content.

The sRGB test is similar to the out-of-box result with no visible grayscale errors and slightly light gamma. You can access that gamut using the Color Space option in the Picture menu while remaining in the Eco picture mode.

Comparisons

The Odyssey G7 wins the out-of-box contest with a 1.06dE average grayscale error. That’s better than some monitors can boast after calibration. I have no complaint there. When all the screens are adjusted, the Odyssey G7 hangs onto third place in a group of monitors with no visible grayscale errors. That puts them on equal footing.

The gamma results are also very close between the six displays. The Odyssey G7 has a small 0.15 range of values and an even lower 0.91% deviation from the 2.2 reference. The actual value is 2.18, and all but the 90% step is right on 2.2. This is excellent performance.

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 first two gamut charts above are nearly identical. Though I adjusted the Odyssey G7’s grayscale tracking, it made almost no difference to color accuracy. In fact, the post-calibration result has a slightly higher DeltaE. But if you look closely, you can see that with adjustment, color is a tad more saturated. It’s a subtly visible difference that is worthy of your time. The sRGB gamut is even closer to the mark with all points on or near their targets. The Odyssey G7 is suitable for color-critical applications.

Comparisons

Though the Odyssey G7 comes last in this group, a 2.23dE color error is nothing to complain about. This is solid performance among an excellent group of monitors. In the volume test, the Odyssey G7 is also right in the thick of it with 88.41% coverage of DCI-P3 and a near-ideal 97.02% value for sRGB. None of these screens has Quantum Dot layers, and visually, the differences are slight. They are very colorfu,l but if you want the ultimate in saturation, a QD-OLED is the way to go.

Test Takeaway: The Odyssey G7 delivers pro-level accuracy right out of the box. That puts it near the top of its class. Remember that I’m comparing 32-inch screens here because the 40-inch Samsung is pretty much in a class by itself. It delivers a lot for the money in the color accuracy department.

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.

The Odyssey G7 accepts both HDR10 and HDR10+ signals. The latter is significant because it includes dynamic tone mapping metadata, much like Dolby Vision. You’ll need specifically encoded content to take advantage of this standard, which is found in many Samsung displays. For HDR10, you get a dynamic tone mapping option that works very well.

HDR Brightness and Contrast

Samsung rates the Odyssey G7 for HDR600, but I only measured 548 nits. Is this a failure? No, VESA’s criteria state that a display must exceed 600 nits from an 8% window pattern. I use 25% because that’s the smallest window available from my pattern generator. The Odyssey G7 would easily top 600 nits with a smaller window. And it’s the brightest monitor of the bunch, thanks to its edge-zone dimming backlight. That puts it on top in the black level and contrast tests, with a final ratio of 9,011.2:1, excellent performance. This gives the HDR image a lot of depth and makes it clearly superior to SDR.

Grayscale, EOTF and Color

Measuring the Odyssey G7’s HDR color showed me some of the best results I’ve seen to date. Grayscale tracking is visually perfect and rides well below the 3dE threshold. I’ve tested only a tiny handful of monitors that are this close to the reference. Luminance tracking is also nearly perfect, with just slight darkness observed at zero and 20% brightness. These are tiny errors that you won’t spot in content. The Odyssey G7 clearly renders every detail throughout the brightness range and transitions to tone mapping at 65% which is correct for the measured white and black levels.

In the color tests, it doesn’t fully cover DCI-P3, but it looks very colorful thanks to a bit of oversaturation in the low- and mid-tone areas. Hue tracking is on point, which keeps color looking natural and vivid with no loss of detail. In the BT.2020 test, the Odyssey G7 runs out of color at 82% red, 65% green, and 95% blue. This is typical coverage for the ultra-wide monitors I’ve tested.

Test Takeaway: The Odyssey G7 has some of the most accurate HDR color I’ve seen to date. It also has plenty of dynamic range, triple that of its SDR mode. When compared to other ultra-wides, its HDR rendering is superior in every way, and made even better when Dynamic Tone Mapping is active. This is something you won’t find elsewhere. And you also won’t find many monitors that support HDR10+. The Odyssey G7 stands out from other LCD HDR monitors in general.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

With so many shapes and sizes of gaming monitors available, users can easily find one that fits their priorities. For maximum versatility, you can’t go wrong with a 27-inch 16:9 flat screen. But when immersive gameplay is at the top of the wish list, nothing is better than a jumbo ultra-wide with an extreme curve.

The Samsung Odyssey G7 S40FG75 is extreme in every good way. A 1000R curve and a 40-inch screen put the gaming environment right in your face, making everything seem bigger, closer, and more real. It’s not an OLED, but it doesn’t cost like one either. And it delivers excellent contrast from a VA panel with solid color volume that covers over 88% of DCI-P3. With edge zone backlight dimming, it takes contrast up and over 9,000:1, putting it ahead of most LCDs. Color accuracy out of the box means no calibration is necessary. And that’s true for HDR as well. The Odyssey G7 aced my HDR color tests most impressively.

If I were to recommend the ultimate gaming monitor, it would be a jumbo 1000R 21:9 curved QD-OLED. But that prospect comes with a $1,800 price tag if it’s the LG 45GX950 I recently reviewed. The Samsung Odyssey G7 S40FG75 is $900 at this writing, half the money for 90% of the experience. That’s something well worth checking out.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

The X870E Tomahawk MAX Wifi PZ ($319.99) is a well-rounded solution in the mid-range PC space. In addition to its back-mount features, you get ample storage options, fast networking, 40 Gbps USB 4, robust power delivery, EZ DIY functionality, and the new(ish) Click BIOS X. As far as looks go, the PZ/back connect element really cleans up the appearance and does a great job of hiding a vast majority of wires, especially when using PZ graphics cards. The white/silver theme looks great, especially in the available MSI PZ chassis.

For under $320, MSI’s board comes with four M.2 sockets (two PCIe 5.0), 5 GbE and Wi-Fi 7 networking, plenty of USB ports (including two 40 Gbps USB4 ports) on the rear IO, and three full-length PCIe slots for a graphics card and expansion. MSI’s EZ DIY features, such as the M.2 Shield Frozr II and EZ M.2 Clip II, make installing your M.2 drive effortless, eliminating the need for fiddling with tiny screws. It also includes the EZ PCIE Release, a simple physical button release mechanism that helps remove any PCIe card with a single tap. The Tomahawk MAX Wifi PZ has all the specifications you’d expect around the $300 mark, and is an improvement, if only by looks, over the non-MAX X870E Tomahawk Wi-Fi we covered in mid-2025.

Performance on this board was average, with some instances being faster and others slower. As we usually see, the results aren’t significantly different in most tests, so if something is slower or faster than average, chances are you won’t be able to notice unless you’re actually looking at benchmarking results. Whether you’re gaming or working, this board won’t hold you back.Below, we’ll examine the board's details and determine whether it deserves a spot on our Best Motherboards list. But, before we get to the test results, discuss the details, and showcase the board in a PZ ecosystem, we have included the specifications from MSI below.

Specifications of the MSI X870E Tomahawk MAX Wifi PZ

MSI’s Project Zero - Pano 100R PZ and RTX 5070 Ti Ventus 3x PZ

Over the past couple of years, we’ve seen this idea of rear-connecting ecosystems gain momentum, and MSI wasn’t going to miss that boat. So far, the company has released several PZ motherboards for both AMD (4x) and Intel-based (3x) systems, as well as what appears to be five PZ chassis of varying sizes to accommodate rear-facing connectivity. Price points, at least for the motherboards, can be slightly higher than comparable non-PZ offerings, but if you want to hide all your wires, the meager difference is worth it.

MSI sent the E-ATX MAG Pano 100R PZ chassis ($159.99) to showcase the clean look of the back-connect motherboard. We reviewed the wrap-around case earlier this year and found it spacious, with ample room for expansion, and it exhibited excellent thermal performance with the included RGB fans. Additionally, it supports a 4-slot vertical GPU (although it does not include a riser cable). When your dream build doesn’t include white, you can also find this case in black for the same price.

Rounding out the PZ Ecosystem is the RTX 5070 Ti Ventus 3x PZ ($859.99) video card. This black-and-silver GPU appears identical to its non-PZ counterpart, except it features a more back-connect-friendly location for the 12VHPWR connector. Instead of at the top, it’s offset and pointing towards the PCIe slot. It’s notably more PZ-friendly and looks way better than the top or rear locations we’re more familiar with, back-connect or not. The Ventus 3x PZ also includes a magnetic plate to conceal the cable path, which is visible when the graphics card is mounted horizontally on the motherboard.

Inside the Box of the MSI X870E Tomahawk MAX Wifi PZ

Inside the box are a few accessories to help you get started. The only thing that stands out is that you get a USB drive for drivers, which is something you don’t typically see in mid-range class boards. Below is a list of everything that’s included.

• Quick installation guide/EU Reg. Notice
• Cable stickers
• USB drive (drivers)
• (2) SATA cables
• EZ M.2 Clip II remover
• (3) EZ M.2 Clip II mounts
• M.2 plate screws
• EZ Wi-Fi antenna
• 1 to 3 EZ Conn-cable (V2)
• EZ Front panel cable

Design of the Tomahawk MAX Wifi PZ

The X870E Tomahawk MAX Wifi PZ sits on a silver-and-white 8-layer server-grade PCB, with large aluminum heatsinks covering the VRM and plate-style heatsinks cooling the M.2 sockets and chipset. On the heatsinks, there is branding with the MAG symbol, and the Tomahawk/Arsenal Gaming branding appears on the rest of the heatsinks. RGB lovers will need to add their own via the headers, as none are built-in.

Obviously, the most significant aesthetic change on the PZ is the lack of front connectors on the front. Shrouds and heatsinks now cover the right edge where the connectors should be, with only the essential sockets and slots visible on the front.

We won't cover the front of the board in too much detail, as the essential connectivity is located on the back. But that doesn’t mean there isn’t anything important on the visible side. Outside of the extended Frozr Guard heatsinks on the power delivery, the next notable feature is the four reinforced DRAM slots. MSI lists speeds up to DDR5-8400+ MT/s, with a capacity up to 256GB. AMD’s sweetspot is still somewhere in the DDR5-6000 to 6400 MT/s range, and this, like every board, covers it easily.

In the upper-right corner are the EZ Debug LEDs, which light up during the POST process. If there’s a problem, one of the LEDs remains lit, providing some insight into the source of the boot issues. Just below that is the 2-character Debug LED, which gives more detailed insight into POST issues (and displays CPU temperatures by default when in Windows).

Flipping the board around to the rear, in the upper right corner (as we’re looking at it), are the two 8-pin EPS connectors (one required) that power the processor. To the left are the first three (of eight) 4-pin fan headers. In this case, CPU_Fan1, PUMP_Fan1, and SYS_Fan1. Each header supports PWM and DC-controlled devices. Output varies depending on the header, with the system fan headers rated at 1A/12W, the CPU fan at 2A/24W, and the pump, which is capable of outputting 3A/36W. Fan Control is handled through the MSI Center software and the User Scenario applet.

Down the left edge (we’re still on the back of the board) is our first (of four) 3-pin ARGB headers. If you want RGB lighting through the motherboard, you’ll have to buy and attach your own. Next is a system fan header, the 24-pin ATX power for the board, the EZ Conn header (connects ARGB strips, fans, USB 2.0 devices, and even the company’s MPG EZ120 ARGB fans on a single cable), and two SATA ports.

Power delivery on the Tomahawk MAX Wifi PZ is the same as that of the front-facing Tomahawk Wifi, consisting of 17 phases, 14 of which are dedicated to Vcore. Power is supplied by the 8-pin EPS connector(s) to a Monolithic Power Systems (MPS) MPS2412 controller. From there, it moves to the 80A MPS MPS2433 SPS MOSFETs. The 1,120A is more than enough for the flagship Ryzen 9 9950X/X3D. While the specs here aren’t the most robust we’ve seen, they do not hold you back, even if you decide to overclock.

Making our way back to the front of the board on the bottom half, and starting on the left, under the shroud is the audio solution comprised of the Realtek ALC4080 codec, a few dedicated audio capacitors, and like that isolates the audio signal to reduce EMI from the board. This is the latest gen codec from Realtek and is fitting for the price.

In the middle of the board are three full-length PCIe slots and five M.2 sockets. Starting with PCIe, the top slot (PCI_E1) connects directly to the CPU, runs at up to PCIe 5.0 x16, and serves as your primary graphics slot. It’s also reinforced and uses MSI’s EZ PCIE Release, an easy-to-access button that locks and unlocks the PCIe card with a single tap. The other full-length slots connect through the chipset, running at PCIe 3.0 x1 (PCI_E2) and PCIe 4.0 x4 (PCI_E3), which should provide sufficient bandwidth for most expansion cards.

Mixed in among the PCIe slots are four M.2 sockets. The Tomahawk MAX PZ offers two PCIe 5.0 (128 Gbps) sockets (M.2_1/2), and the bottom two (M.2_3/4) connect through the chipset and run up to PCie 4.0 x4 (64 Gbps). M.2_1 supports up to 110mm modules, while the rest support up to 80mm length. The board supports RAID 0/1/5/10 modes if you need extra speed or redundancy. Do note the second PCIe 5.0 M.2 socket (M2_2) shares bandwidth with USB 4. By default, when M2_2 is in use, it runs at PCIe 5.0 x2 (64 Gbps). You can get the full x4 width by adjusting it in the BIOS, but that disables the USB 4 ports on the rear IO.

Flipping the board around to the back one more time, we’ll start with the left edge and work our way down. The first item we run into is the front panel 3.2 Gen 2x2 (20 Gbps) header, followed by the other two vertically mounted SATA ports. Finally, below that is the 19-pin USB 3.2 Gen 1 (5 Gbps) header.

Along the bottom edge (we’re still on the back!), there is a slew of headers and connectors. From left to right, you have the following:

• Front Panel
• (2) 4-pin system fan headers
• 3-pin ARGB header
• 19-pin USB 3.2 Gen1 connector
• 2-pin CMOS reset/battery jumper
• 4-pin system fan header
• (2) USB 2.0 headers
• (2) 2-pin BCLK frequency jumpers
• 8-pin supplemental PCIe power
• 4-pin system fan header
• 4-pin RGB, 3-pin ARGB
• Front panel audio

The rear IO on the Tomahawk Wifi PZ is the same on the non-PZ board. Starting on the left side of the rear IO are two tiny buttons: one for clearing the CMOS and the other for BIOS Flashback. Next to that is the HDMI port to use with integrated graphics. There are three USB Type-C ports, two of which support 40 Gbps (the third is 10 Gbps). You also get two 10 Gbps (red), four 5 Gbps (blue), and four USB 2.0 (black) Type-A ports, for a total of nine. Above one of the red USB ports is the Realtek 5 Gbps Ethernet. Next are the two Wi-Fi 7 connections and the audio stack, with two 3.5mm (mic in/line out) and SPDIF.

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

Firmware

With X870, MSI has updated its BIOS, renaming it “Click X,” and improved the aesthetic and functionality in the process. The company changed the format with the headings on the left side, details in the middle, and system status information on the right. Like most other board partners, it features an informational EZ Mode, which allows for limited functionality, including Game Boost, PBO, and more.

Overall, I like the new layout, and maneuvering around is intuitive after a short time, though it does take some getting used to. The dark, fading yellow/green background (that matches the board’s aesthetic) and white labeling are easy to read, and most of what you need is at your fingertips.

Software

MSI Center is a single utility that offers a wide range of functionality. From hardware monitoring to RGB control with Mystic Light, the software features numerous applets and serves as a one-stop shop for downloading additional utilities, including overclocking and fan control. Those looking to use the Gamebar feature, Super Charger, or any other utilities MSI offers will find them all in MSI Center.

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. The hardware we used is as follows:

TEST SYSTEM COMPONENTS

• CPU - AMD Ryzen 9 9900X
• Cooling - Arctic Liquid Freezer II 420
• Storage - Crucial 2TB T705 M.2 PCIe 5.0 NVMe SSD
• RAM - Kingston Fury Beast DDR5-6000 CL36 (KF560C36BBEAK2-32)
• RAM - Teamgroup T-Froce Delta DDR5-7200 CL34 (FF3D518G7200HC34ABK)
• RAM - Klevv Cras XR5 RGB DDR5-8000 (KD5AGUA80-80R380S)
• GPU - Asus TUF RTX 4080 16G
• PSU - EVGA Supernova 850W P6
• Windows 11 64-bit (24H2)
• NVIDIA Driver 561.09

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 offer a valuable method for evaluating a board's performance, as identical settings are expected to yield similar results. Turbo boost wattage and advanced memory timings are areas where motherboard manufacturers can still optimize for stability or performance, though, and these settings can impact specific testing scenarios.

Performance in our synthetics tests was right around average for the Tomahawk. The Procyon tests were all over the map, displaying generally slower performance in the Office apps and video editing, but the fastest result was in photo editing. Still, there’s nothing to be concerned about so far.

Timed Applications

In the timed applications, the system's results were consistent with those of the other tests. Nothing out of the ordinary here, either.

3D Games and 3DMark

Starting with the launch of Zen 5, we’ve updated our game tests. We’re keeping the EA’s F1 racing game and have upgraded to the most current version, F1 24. We also dropped Far Cry 6 in favor of an even more popular and good-looking game in Cyberpunk 2077. 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 differences. We’ve also added a minimum FPS value, which can affect your gameplay and immersion experience.

Gaming results were also solid, particularly well in Cyberpunk 2077 testing, leading the pack in minimums and one frame per second shy of the average. F1 24 also had a decent showing, with good minimums and average FPS compared to the rest.

Overclocking

Over the past few CPU generations, overclocking headroom has been shrinking on both sides of the fence while the out-of-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 from the BIOS, 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 the above, we will not be overclocking the CPU. However, we will try out all our different memory kits to ensure they meet the specifications.

Our memory adventures were similar to those of other boards of this ilk. In this case, the Klevv DDR5-8000 kit wouldn’t boot (not on the QVL), but the TeamGroup DDR5-7200 kit worked without issue. AMD’s sweetspot is still in that DDR5-6000 to 6400 MT/s range, so pushing to higher speeds isn’t terribly beneficial, especially when looking at price-to-performance increases.

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 MSI X870E Tomahawk MAX WiFi PZ peaked at 260W and idled at 85W. Surprisingly, it uses more power than most boards we’ve tested, but isn't out of line to make a real complaint.

The Tomahawk MAX’s VRM temperatures were well within specification. The 80A MOSFETs and large heatsinks easily kept things in check. Our Ryzen 9 9900X peaks at approximately 46 degrees Celsius on our sensors (48 degrees on the integrated one), and the CPU averaged 150W during the test. You can easily run a Ryzen 9 9950X/X3D and even overclock it with this board. Your cooling will hold you back before the board does.

Bottom Line

MSI’s Project Zero ecosystem provides a comprehensive solution for users seeking a back-connect system, encompassing a range of compatible cases and motherboards across various sizes and price points. Currently, the company only offers one video card, but we hope that will be expanded to other models for improved cable management. MSI’s PZ hardware is also compatible with competitors' BTF/Stealth back-connect hardware, offering a range of options that work with MSI PZ motherboards (and vice versa). One of the main drawbacks of moving to a back-connection-supported system is that you’ll need to purchase a new case or get creative with a Dremel to modify your current one to accommodate the motherboards.

The X870E Tomahawk MAX Wifi PZ motherboard’s feature list is as comprehensive as the non-PZ Tomahawk model. Priced at just under $320, it includes capable VRMs, USB 4 support, ample storage options, high-speed networking, multiple DIY/EZ features to simplify the building process, and, of course, a visually appealing, connector-free aesthetic. If you’re in the market for a back-connect system, MSI’s X870E Tomahawk MAX Wifi PZ is a great base to build your back-connect system around, but it fell just short of making our best motherboard list.

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

A 27-inch 4K monitor delivers a tremendous pixel density of 163 ppi, which means the image is incredibly sharp with no visible dot structure. It’s great for close-in viewing, whether you’re gaming or working on your next Photoshop masterpiece.

There’s great value in this genre as well, and Titan Army is one of the leaders there. Its latest offering, the P2712V, delivers a 27-inch IPS screen with 4K at 160 Hz, Full HD at 320 Hz, Adaptive-Sync, HDR10, and wide gamut color. More importantly, it’s just $325, so let’s take a look.

Titan Army P2712V Specs

The P2712V starts with a 27-inch IPS panel sporting an edge backlight. Cost is kept down by leaving out a full-array LED and local dimming; neither is included. But brightness is solid at around 350 nits for HDR and SDR. There is no dynamic contrast in HDR mode, so the ratio remains around 1,000:1 across all content. You also get wide gamut color with a measured 93% coverage of DCI-P3. Accuracy is so good out of the box that I couldn’t improve the image with calibration.

Not too much has been sacrificed on the altar of value. There are aiming points, static and dynamic, sniper modes, night vision, frame counters, timers, alignment marks and solid video processing with a well-implemented overdrive. LED lighting accents the back of the panel with colorful vertical bands. And the stand is fully adjustable with solid ergonomics.

The dual mode switch takes a couple of button presses to execute, but once set up, you can quickly change between 4K at 160 Hz and Full HD at 320 Hz. You lose pixel density, of course, but at 27 inches, the P2712V has a sharper FHD picture than a 32-inch screen.

There are plenty of inputs for all the latest systems, PCs, and consoles. Two HDMI 2.1 and two DisplayPort 1.4 operate at full bandwidth, 3840x2160 at 160 Hz with Adaptive-Sync and HDR. You also get a headphone jack, but there are no USB ports.

Titan Army has always impressed me with its value. The P2712V’s bang-for-the-buck quotient is very high with little left out for $325 at this writing. Let’s check out some test results.

Assembly and Accessories

The P2712V’s plain brown wrapper is so subtle that you have to look at the stick-on label to know which model is inside. Once the crumbly foam is parted, you’ll find the base, upright, and panel securely packed. A small external power supply delivers the AC, and you get a heavy-gauge DisplayPort cable.

Product 360

My apologies for the stock photos above; the P2712V doesn’t actually make characters climb out of the screen. The picture looks awesome, though, thanks to the packing of 8.3 million pixels into a 27-inch IPS panel. The anti-glare layer cuts reflections nicely without spoiling that razor-sharp image. A thin bezel extends past the picture by just a few millimeters, creating a flush and frameless look. Only a Titan Army logo appears at the bottom, which reminds me a bit of a mouth harp. You know, one of those thingies that twangs when you pluck it. At the lower right is a row of OSD control keys and a backlit power button.

The LED lights trim two sides of the P2712V’s backside, making a nice contrast to all the textures going on. The perimeter looks like carbon fiber weave, and the center section has a brushed finish. The base also has a carbon fiber look. The lighting can be controlled in the OSD with multiple color and effect options.

The upright features 5/20 degrees tilt and a 90-degree portrait mode. You also get a solid four-inch height adjustment. There is no swivel. Movements are firm with confident positioning, but there is a little wobble where the panel attaches to the stand.

Up and under, you’ll find four video inputs, two HDMI 2.1 and two DisplayPort 1.4. All operate at full resolution and 160 Hz with HDR and Adaptive-Sync. You also get a 3.5mm headphone jack with volume control in the OSD. There are no USB ports or internal speakers.

OSD Features

The P2712V’s OSD is the same one I’ve seen in all Titan Army monitors. It offers many options for picture adjustment, gaming aids, and video processing, but it is a bit clunky to operate. After a bit of back-and-forth, I figured it out.

The OSD appears when you press any key and has a purposeful look suited for gaming. Resolution and refresh rate are shown at the top of all screens. Starting with the TI menu, it has aiming points, night vision, and two sniper modes. It also has a dual-mode switch that changes between 160 and 320 Hz.

The P2712V has 12 picture modes, though it could be said there are 24, since each has a default and a custom memory. To keep confusion at bay, stick with Standard mode in its default configuration. It doesn’t need calibration, and in fact, tweaking provides no benefit. If you need sRGB, it’s on the second screen of mode choices. It too has default and custom modes and can be calibrated if you wish.

In Picture Settings, you can pick from multiple gamma and color temp presets or adjustable RGB sliders. I found them quite coarse in practice, so for me, they made no improvement. If you get lost in the settings, you can reset just this menu back to factory defaults.

Some of the TI options are repeated in the Game+ menu, and you also get aspect ratios, HDR modes (Auto, Game and Movie) and picture enhancements. That’s where you’ll find the overdrive, which works well on all its levels except Extreme, which caused ghosting on my sample. Here also is the frame counter, more crosshairs, a timer and stopwatch, and sniper mode. This menu too can be reset independent of the others, very handy. Game Illumination refers to the LED in the back, which can be configured for different colors and effects.

Titan Army P2712V Calibration Settings

My initial measurements of the P2712V were excellent. It doesn’t need calibrating in the Standard mode. It also works well in sRGB mode where you get accuracy suitable for color-critical tasks. The native gamut covers 93% of DCI-P3, so it will be oversaturated for SDR content but not in a way that anyone will object to. I tried calibrating the RGB sliders, but found they weren’t fine enough to make any improvement. Below are the brightness settings for commonly used peak levels. HDR signals get three of their own modes, Auto, Movie and Game. Testing showed that Movie was the best choice.

Gaming and Hands-on

To prepare the P2712V for dual-refresh operation, I started in 4K and set the rate in the Nvidia Control Panel to 160 Hz. There were no issues running at full honk for hours of gaming. The P2712V was completely stable. To set up 320 Hz, first activate the dual-mode option in the OSD. On means 320 Hz at FHD resolution. Then return to the Nvidia Control Panel and set the refresh rate to 320 Hz. In both modes, use level 3 for the overdrive. It removes nearly all motion blur and doesn’t cause ghosting. Extreme will add artifacts, so it’s best avoided. There is no backlight strobe feature here, but it isn’t necessary if you run 4K faster than 140 Hz. My setup uses a GeForce RTX 4090, so I ran steadily at 160fps.

Games like Doom Eternal proved to be very responsive; I could not detect any input lag. The picture is super sharp, whether moving or stationary. Fast camera pans don’t cause the smearing that calls out a lesser display. Titan Army’s video processing is very well done.

To switch refresh modes, I had to return to the Windows desktop. Activating dual-mode in a game usually ends in a visit to Task Manager to close the crashed program. This is true of all the dual-refresh monitors I’ve reviewed. At 320 Hz/FHD, the P2712V is very smooth with no visible motion blur. You won’t be wishing for backlight strobing. The overdrive should still be set to level 3. The loss of pixel density is certainly obvious, but not a major negative. It will be attractive to users with less powerful PCs that can’t maintain 160fps at 4K. When it comes to frenetic gameplay, smoothness beats sheer resolution.

The HDR image was colorful and bright, but not more dynamic than SDR. This is solely due to the P2712V’s lack of dynamic contrast. A field dimming option would be welcome here, but in fairness, it is far from the only screen with this deficiency. It supports HDR but doesn’t embrace it.

For everyday tasks, the P2712V is a great choice. Pixel density like this is super nice for photo and graphics work. There’s no visible pixel structure at close viewing distances; I’m talking about 18 to 24 inches away from the screen. Color accuracy is also good enough for critical tasks, especially if you need sRGB. Titan Army nails that mode.

My only complaint about the day-to-day operation of the P2712V is its OSD. The menu has some redundancies that are confusing. And it’s challenging to navigate with just buttons. I’m spoiled by joysticks, but I think they should come on every monitor, regardless of price.

Takeaway: The P2712V punches above its weight class in gaming performance with premium video processing, responsive play and refresh rate flexibility. It can serve equally well with high-end and mid-grade systems. HDR imagery is color accurate and well saturated, but lacks the depth of monitors with dynamic contrast, which is absent here. I’m not a fan of the OSD, but in fairness, you won’t have to use it much since calibration isn’t necessary. Overall, the P2712V gives you a lot to be happy about at a low price.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

To compare the P2712V’s performance, I rounded up speedy 4K screens, mostly 27-inch and one 32-incher, the BenQ PD3226G. Two of them are also dual-refresh: Acer’s XB273K V5 and Alienware’s AW2725QF. Fixed at 160 Hz are Gigabyte’s GS27U and Asus’ XG27UCS.

Pixel Response and Input Lag

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

At 160 Hz, the P2712V is 1ms slower to draw a full white field than its most direct competition. But that is made up for by its overdrive, which is extremely precise. Visually, it’s enough to eliminate the 1ms deficit. The P2712V is very smooth with only the tiniest bit of motion blur at 160 fps. At 320 Hz, there’s no blur to speak of. And that is also true of the Acer and Alienware.

In the lag test, the P2712V is a tad slower than the rest, but it is still within the realm of very fast. Hardcore competitors might prefer the Acer or Alienware for ultimate speed, but most of us won’t be able to perceive a difference between them and the Titan Army. I found it very responsive in all scenarios.

Test Takeaway: If you want the lowest possible input lag and panel response, you’ll have to buy an OLED. But for $325, it’s hard to beat the P2712V. It’s more than quick enough for all but those with very high skill levels. And you can ramp up the fun with the dual-mode switch. The overdrive works very well at 160 and 320 Hz and delivers solid blur reduction without ghosting on its first three levels. The P2712V’s performance compares well with that of more expensive displays.

Viewing Angles

The P2712V’s viewing angles are slightly better than those of the average IPS monitor. At 45 degrees to the sides, there’s a slight red tint which impacts content less than the more common green shift. There is little change in brightness or gamma, so detail and clarity remain high. From the top, you can see a blue tint and a roughly 30% reduction in brightness with light gamma.

Screen Uniformity

To learn how we measure screen uniformity, click here.

My P2712V sample showed excellent screen uniformity with just a 6.01% deviation from the center zone. This exceeds expectations for a $325 4K display. The competition shows solid performance too, but also costs more.

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

Titan Army rates the P2712V at 350 nits, but my sample couldn’t quite get there. It’s close, though, at 334 nits, more than enough output for any indoor space. If you need more than that, the competition promises over 400 nits, which is very bright.

Black levels are solid for an IPS panel, resulting in a decent contrast ratio of 1,026.8:1. That’s an average result and good enough for second place here. The P2712V would certainly benefit from a Mini LED backlight, but then it would cost more.

After Calibration to 200 nits

I didn’t calibrate my P2712V, just lowered brightness to 200 nits. That maintained a consistent contrast of 1,020.7:1, putting it in the top spot. In practice, there are no significant differences among the monitors in the group.

The P2712V also wins the ANSI test with a consistent 936:1 result. It is value-priced but still maintains good QC with well-selected components and solid build quality.

Test Takeaway: The P2712V provides solidly average brightness and contrast when compared to other 4K IPS gaming monitors. It’s average among IPS screens but slightly better than its most direct competitors. It maintains good black levels and a consistent contrast ratio in both static and intra-image tests.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

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.

The P2712V gets off to a great start with nearly perfect grayscale tracking and gamma that tracks with little deviation from the reference. All gray errors are below 2dE, and the gamma average is 2.16, with the only error being a slight dip at 90%. It doesn’t get much better than this for out-of-the-box performance, especially at this price point. I attempted a calibration but could not improve upon what you see above.

In the sRGB test, grayscale tracking is even better with a 0.78dE average. That’s pro-level performance. Gamma is right on the reference line except for 10% brightness, which renders a bit light. This makes shadow detail clearer but also less dark than it should be. It’s a minor issue at best. I have no complaints.

Comparisons

At default settings, the P2712V beats all comers, except the XG27UCS, which shows exceptional grayscale tracking. 1.21dE is an excellent score. Once the other screens are calibrated, they draw ahead, but visually, there’s no difference between the top and bottom monitors. This is excellent performance.

The P2712V’s gamma test result is very good, but the others are a bit better. The light number I measured at 90% brightness is the main culprit. It contributes to a 0.13 range of values and a 1.82% deviation from the 2.2 reference. The actual value is 2.16. Again, I have no complaint.

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 P2712V’s gamut results are equally impressive. Magenta and blue show a tiny bit of undersaturation, while the other colors are nearly perfect. Green is a tad under as well, but there is more to that primary than I typically see from other non-Quantum Dot displays. Hue values are also spot-on.

The sRGB gamut chart looks like it came from an expensive professional display. Aside from a tiny undersaturation at 90% red, it is pretty much perfect. A 0.80dE average error is something you rarely find in a $325 monitor.

Comparisons

None of the monitors have significant color errors; they are all exemplary. The P2712V sits in fifth with 1.79dE, but that is still well below the visible threshold of 3dE. Though I didn’t show the sRGB values, the P2712V has a lower error in that gamut than the other screens.

Value displays rarely pass 90% coverage of DCI-P3, but the P2712V manages a solid result of 92.98%. It comes up the tiniest bit short in green, blue and magenta, but I’m definitely picking nits there. It is very colorful, and you won’t see any deficiencies in practice. This is excellent performance.

Test Takeaway: The P2712V is a very colorful monitor, especially when its low price is considered. It’s also more accurate than most value screens and manages to outperform more expensive monitors. You don’t need to calibrate it and in fact, the available adjustments provide no benefit. It also has a very accurate sRGB mode, which is something rarely found in any budget screen. For $325, it’s hard to beat.

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.

The P2712V supports HDR10 signals with an automatic switch. You get three specific modes, Auto, Movie and Game. There is no dynamic contrast, so you won’t get much drama, but it delivers accuracy on par with most of the other HDR screens I’ve tested.

HDR Brightness and Contrast

Bright highlights help boost HDR quality and the P2712V is a bit below average in that regard. Though it tops out slightly higher than its SDR value, it still falls short of the other screens, which can all exceed 400 nits. It’s obvious in the black level and contrast charts which displays have dynamic contrast. The P2712V doesn’t have it, so its HDR contrast ratio is the same as its SDR, just over 1,000:1. This is often one of the things sacrificed for lower cost, which is not unlike some of its competitors.

Grayscale, EOTF and Color

I measured all three HDR modes and noted subtle differences between them. None had ideal grayscale tracking, but Movie had the least visible errors thanks to a little extra red. Auto and Game are decidedly green in tone. Movie also has the best EOTF tracking with near-perfect adherence to the reference line and a soft transition to tone mapping at 65%. Shadow detail is clear but a little light at levels between zero and 10%.

In the gamut tests, the P2712V is generally oversaturated in the primaries’ inner points. This gives HDR images a little more punch than their SDR counterparts. The only visible hue error is magenta, which is a tad too red in tone. In the BT.2020 test, color runs out at 88% red, 65% green and 90% blue. This is typical HDR coverage for a budget display.

Test Takeaway: The P2712V doesn’t have a precisely accurate HDR mode, but its Movie option has very good luminance tracking, which goes a long way towards maximizing detail and clarity. Color is well saturated and reasonably accurate as well. My only gripe is the lack of dynamic contrast which is a trait shared by other value-oriented monitors. The HDR picture is more colorful than the SDR, but with no additional depth or texture.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

A 4K monitor might be an elusive purchase for some budgets, but more choices appear all the time. Titan Army caters to the value-conscious market with its products and consistently delivers a lot of bang for the buck. It’s also challenging to pair 4K with mid-level gaming systems because to push 8.3 million pixels at 100 fps, you need to spend a bit of money on a good video card. Dual-mode displays provide a nice alternative: you can switch to FHD for gameplay and high frame rates, then back to 4K for productivity and video.

The P2712V isn’t the first dual-mode monitor I’ve seen, but it is one of the least expensive. At $325, it provides everything needed for high-performance gaming and then some. The only things left out here are USB ports and internal speakers. And dynamic contrast for HDR. That’s my only real complaint. It provides the saturated color needed for good HDR but not a backlight dimming option. That would take the P2712V to another level, especially at its current price point.

On the upside, SDR color accuracy is superb. Calibration is completely unnecessary, and I found that I could not improve upon the factory settings. Titan Army has done a superb job there. And SDR contrast is solid with more than enough brightness and decent black levels.

The P2712V’s biggest draw is its flexibility. You can have a high pixel density monitor for productivity and video, and a super-fast screen for gaming that runs at 320 Hz. And you get accurate color too, and it’s all just $325. I can remember when 4K enterprise screens cost 10 times that.

If you are longing for high pixel density and high frame rates but don’t have a high bank balance, the Titan Army P2712V is well worth checking out.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Fuelling the Gamestation Go’s potential success are several special features. Probably the most alluring features of this 7-inch handheld are its pleasing array of retro arcade--appropriate controls, and My Arcade’s inclusion of a whopping 200+ classic games from Atari, Jaleco, Namco, and others built-in.

The combination of a bumper selection of classic console and arcade titles, alongside those unique for its form factor controls, should make this product a winner. However, the price, compared to portable retro-gaming alternatives out there, and a failure to play to its strengths, means we are marking it down quite strongly.

Design of the Atari Gamestation Go

The Atari Gamestation Go takes its cues from the most popular PC handheld designs right now. In other words, it looks a lot like a Steam Deck or ROG Ally – from a distance. Once you take a little more time to absorb what you are seeing, though, some interesting differences will catch your eye.

Scanning beyond the now familiar 7-inch landscape form factor, several things become obvious. To the right, there is a numeric keypad. That’s different. But to the right are some of the most important physical controller design ‘innovations’ – in our experience. From top to bottom, you will see an integrated paddle/dial/spinner; below that, there’s a more familiar D-pad, and below that a trackball. Control options like these can really impact the retro-gaming experience – and there are several famous Atari console and arcade games that used controls like paddles and trackballs to differentiate the gaming experience.

Unpacking and getting hands-on

Atari and My Arcade have produced some excellent packaging for this device. The box looks glossy, bright, and eye catching, highlighting the console design and its classic arcade contents. This positive impression grows when you flip up the faux-top to reveal a gatefold gaming content reveal. The games that My Arcade must expect to exert the strongest pull are shown in a grid on the upper half of the gatefold; below you will see the full 200+ games listed.

Inside the box, we see everything is very well padded for protection in transit. Moreover, the makers haven’t skimped on accessories. There’s an HDMI cable, USB-C charging cable, and power brick, all provided. A couple of bits of paperwork are also to be found in the box. One asks users to connect to Wi-Fi to keep the system up to date, and the other advertises the upcoming GamePad and Arcade Stick controller peripherals, made especially for Gamestation users. These will surely plug directly into the two USB-C ports on the back of the device.

In my hands, the Gamestation Go feels plastic but very solid. Twisting it, there is a little give, but nothing dramatic, and there was no easily induced creaking heard, or threat of panels buckling observed.

Atari Gamestation Go Specifications

Teardown attempt – not entirely successful

Atari and My Arcade don't disclose the components used inside the Gamestation Go, so I was more than curious to open up this handheld up to take a look at the PCB.

First, the good news: Accessing the battery was a cinch. Two T6 screws needed to be removed from the back panel, and then the battery cover, sat between the legs of the integrated kickstand, could slide down and reveal the 7,200mAh 3.7W battery pack.

It was simple enough to unplug the battery, as the two adhesive strips fixing it in place just needed some gentle persuasion. This battery is attached to the PCB via a small 3-wire removable connector.

Now onto the bad news. I could find no more screws and couldn’t pry apart the console by levering any of its seams with my tools. Stumped, I went along to my local smart device repair emporium, and they thought the screen would need to be heated up, and the PCB would be uncovered by pulling this up and away, like one might open up a tablet. However, they couldn’t guarantee the device would still work after such an operation, so I decided not to sacrifice this handheld.

We have seen no other teardown online, but Atari enthusiasts buzzing around forums generally think that the innards of the new Gamestation Go are closely related to the non-portable Gamestation Pro, released two years prior. This non-portable TV-companion console was torn down here.

If the assumption that these consoles have a very similar set of inner components is correct, the new Gamestation Go will be Rockchip RK3032-powered. This is a very modest SoC, with dual Arm Cortex-A7 cores, supported by Mali-400MP graphics and up to 512MB of DDR3 RAM.

For software support, an RK3032 retro gaming device maker will usually look to Linux SpectralElec 3.0 and EmuELEC v9.0 builds. It will also struggle with any games newer than the 16-bit era, which tallies with my experience of this system (as we'll discuss later).

Gaming experience

From sliding the red switch to the on-position, to reaching the game UI of the Gamestation Go, takes an unwelcome full 30 seconds. Once there, you will be presented with a colorful interface where the D-Pad is used to scroll horizontally between ‘systems.’ After a system has been selected, there are usually subcategories to flick through vertically, followed by a cascading list of actual games.

It can take a while to get to a game you like, and (perhaps obviously) the Atari family contains a long list of consoles and platforms, each with rather long game lists inside. Thankfully, you can favorite/bookmark titles, and there is a recent list, too, but these shortcut lists aren’t global. Thus, you have favorite lists for Atari games, for PIKO, and Jaleco titles. Moreover, smaller main categories like Atari Recharged, Balls of Steel (pinball), and Pac-Man titles are offered up in a horizontally scrolling format, like the home screen.

As you scroll through the list of games in your chosen category, the console UI shows a picture of the cartridge or cabinet, plus an animation showing gameplay, and beneath that, there’s a more-info section.

In your chosen game, it is quick and easy to head to the settings menu to adjust brightness, screen aspect ratio, load and save states, and view (but not adjust) the controls available. Moreover, the Home button works quickly for getting straight back to the game selection UI and picking something else to enjoy.

The games library

I was keen to dive into some of the classics I remember from the arcades of yesteryear. Some of these titles ate through quite a few coins in my time. Thus, it was exciting to see Asteroids, Berzerk, Breakout, Millipede, Missile Command, Space Duel, Breakout, and Tempest among the roster of titles.

Quite a few of my favorite old arcade titles are ‘vector’ style, and I enjoyed playing them. However, bear in mind that many arcade games get hard and frantic quite quickly. Remember, these titles were designed to hook people in and extract cash as quickly and efficiently as possible – not for lengthy gaming sessions.

Still on the topic of the arcade games on this machine, it was interesting to check each one, and the SmartGlow lighting choices for controls that had been pre-ordained by My Arcade.

For example, Asteroids used the D-pad for ship rotation and thrust, and one of the ABXY buttons for fire. I also enjoyed using the rotational paddle dial in Super Breakout and Tempest. Some fun trackball games that got me playing were Millipede and Missile Command.

It's also important to note the Gamestation Go ahs many duplicate games. Sometimes I wondered why there were so many duplicates, especially in the Atari library. For example, there are versions of Centipede for the Atari 2600, 5200, 7800, and arcade. Then titles like Asteroids and Breakout also have modern Recharged Editions, as well as various old console and arcade versions.

Controllers don’t live up to their potential

While control method authenticity is something of a star attraction for Gamestation Go gamers, there are some clear issues with it, in my experience.

One of the first things you may notice when gaming on this handheld is that you can’t change the controller configuration. The pre-defined SmartGlow indicated controller configuration is what you have to live with. If a game is configured to use the D-pad for movement by Atari / My Arcade, then trying the paddle or trackball will be futile. There’s no way to change this in any of the firmwares I had during testing (v 1.0 and v1.1).

Not only did the SmartGlow highlighted pre-selected controller choices and adjustment possibilities disappoint, the physical interfaces aren’t the best. Specifically, I found the inertia of the paddle, dial, spinner (or whatever you prefer to call it) was too strong. So it was often cold-treacle slow to move in response to on-screen action. I didn’t enjoy the trackball either. I’d instead call it the track-marble, as it is so small. This control method is really overly sensitive, with no way to tone it down. The D-Pad, ABXY buttons, and shoulder buttons all felt good enough, though.

Games performance and compatibility

With the pre-installed games, I didn’t notice any poor performance or symptoms of the hardware not being up to the job. I’d say the most processor-intensive (newest) titles available in the pre-installed list all worked without a hitch, without slowdowns, glitching, or audio pops or crackling. Thus, this handheld, with its underlying hardware, is fit for purpose.

Micro SD card fun

Though not documented by Atari or My Arcade on the official product pages, Gamestation Go users can expand their playable games library using the system’s microSD card slot. This feature is another one that is shared by the prior Gamestation Pro console.

To get an SD card to be recognized by the system, you must format it using exFAT. ROMs for particular systems need to use pre-set extensions. It also makes sense to keep ROMs for different systems in different sensibly named folders, for your hygiene and sanity.

To access these files, you must turn on the system with the SD card already inserted. Just before the UI settles down and is ready for user interaction, you then select between the 'Local Storage' or 'Micro SD card' options (see above gallery). It is easy to press the Home button and get back to the basic folders and files SD card list. If you press Home twice in a row, it will take you to the built-in games. From here, I didn't find a way to get back to the SD card contents, except via a system restart.

I successfully tested and enjoyed some games from the Sega Genesis (AKA Mega Drive), Super Nintendo (SNES), Atari Lynx, and NeoGeo Pocket Color (NGPC) systems. All the titles I tried from these systems played nicely, albeit with just the D-Pad and ABXY buttons usable during play. Perhaps some SNES games pushed the system a little too far, as I heard a bit of audio crackling in some of these classic titles.

Moving beyond the 16-bit era, I had no success with original PlayStation, Nintendo 64, or PlayStation Portable titles, whatsoever. While the latter two platforms may have been limited by the lack of analog joystick controls provided by the Gamestation Go, the PS1 should have been fine in this respect.

I could have tried more titles, and I’ve read of other testers having success with them, but I gave up when all my picks just black-screened and returned me to the UI. I’d only very limited hopes that the Rockchip CPU could have coped with the newer systems, anyway.

Bottom line

My Arcade and Atari have crafted a handheld retro gaming console graced by a bevy of unique retro-futuristic controls. It is clever to differentiate this device by having a paddle and trackball built-in, and this will surely exert a gravitational pull on retro gamers in its orbit. While the Gamestation Go largely works as intended, its appeal is dulled by the lack of user-configuration options, rendering quirks in classic vector games, and a price point that feels misaligned with its level of hardware and software polish. We remain hopeful that further firmware updates can improve matters, but canny customers don't buy on a promise.

Another thorn in the Gamestation Go's sides is that nearly everyone owns a smartphone, with a screen approaching the 7-inch diagonal of this handheld. Many more will have access to a portable in the form of a tablet, laptop, or handheld - often with even larger, more immersive, screens. Add a nice game controller to the powerful mobile device you already own, and you can probably enjoy playing emulated consoles from eras much more recent than the Gamestation Go can hope to support.

These flexible alternative platforms also have legitimate access to many of the same Atari games via Google Play, the Apple App store, and the various PC/Steam games libraries.

Providing yet more competition to the Atari Gamestation Go are consoles like the Nintendo Switch. Users of this popular hybrid console can load up carts / downloads like Atari Mania, and several Atari Recharged collections.

Last but not least, there are dozens of gaming handhelds on Amazon, AliExpress and other marketplaces that are designed primarily for playing retro console and arcade ROMs. These platforms may have fewer legitimately licensed game collection options, but that won't bother some. While I haven't seen any rivals with the trackball or paddle that the Atari offers, they often come with analog joysticks - and enough processing power to make later generations of console and arcade games playable. Moreover, there are lots of options around $50, if a 4-inch screen is enough for your retro gaming fix.

Among these alternatives, to be fair, is the My Arcade Atari Pocket Player Pro at just $39.99 on Amazon at 11% off right now. It has a small screen at just 2.75-inches, and relatively limited controls. However, it still has a decent 100+ licensed games built in and similar cool styling to the newer and more substantial Gamestation Go. That alternative Atari-branded product might just be enough for an enjoyable blast from the past. We'd also expect the larger new console to have some attractive price cuts in due course, rewarding those who are patient.

During my testing, the Gamestation Go’s firmware received an update from version 1.0 to 1.1. It was good to see some improvements, but I feel there’s still plenty more to do in future revisions.

If you're still in for the nostalgia, you can grab the Atari Gamestation Go from Amazon.com for $179.99 today.

The good and bad news is that this means you know what you’re going to get: an affordable, passable drive with usually wide availability but unknown hardware. The first part is good. As for the controller and flash, well, you know that you can never be sure, as the NV series uses whatever is available. That’s bad. How much does this matter for M.2 2230? Well, given the drive’s specifications, which set the minimum, this drive is plenty fast for any device that can take it. The controllers that pass muster and can fit on M.2 2230 are largely decent and will usually be paired with adequate flash – our sample has good hardware even with its QLC flash. This usually means a power-efficient drive, which is critically important for this form factor. The highest capacities also need newer or future flash to work, although we wouldn’t be surprised if some lingering QLC stock could end up on the drive, and that’s less ideal at the most common capacities.

We offer a hedge by relying on the old adage: you get what you pay for. The Kingston NV 2230 is inexpensive and fundamentally sound, but not high-end. If you’re looking for the fastest or best, this isn’t the drive for you. If you want something that will offer an excellent experience at a lower cost, then it should certainly be on your list. It’s likely equal to or better than many no-name brands, and the wide capacity range on offer lets you dial in exactly what you need. It can always be extended in length to fit M.2 2242 or 2280 slots, too. Our sample has excellent power efficiency, so we see no problem with that from a cooling perspective, keeping in mind that usually M.2 2280 versions of drives have more surface area for heat dissipation. To wrap it up simply: Kingston has once again delivered on a simple budget premise, this time in a smaller form factor.

Kingston NV3 2230 Specifications

The Kingston NV3 2230, unlike many M.2 2230 SSDs, is available in a wide range of capacities: 500GB, 1TB, 2TB, and eventually 4TB. Current pricing is at $50.23, $79.99, and $139.69. This pricing is competitive as one would expect from a Kingston NV3 drive. The drive is capable of hitting up to 6,000 / 5,000 MB/s for sequential reads and writes, with no specifications given for random read and write IOPS. We would expect a ceiling around 1,000K IOPS with this hardware.

These specifications are effectively identical to the full-length NV3, but there are changes worth noting. The first is that while the TBW is the same, the 2230 version of the drive has a five-year rather than a three-year warranty. This means the drive writes per day (DWPD) is actually lower on this drive, although still higher than the Crucial P310 2230. Write endurance shouldn’t be an issue with a drive of this type, but be aware of this distinction. Another difference is weight, with the 2280 NV3 being much heavier, as one would expect. The final difference is the flash: BiCS6 QLC on the 2280 and BiCS8 QLC on the 2230 NV3, a change we will dig into in more depth later.

Kingston NV3 2230 Software and Accessories

Kingston’s primary software offering is the Kingston SSD Manager, which is your typical SSD toolbox application. It’s designed to help you monitor disk health, verify properties of the drive, and upgrade the drive’s firmware. Kingston also has its own OEM version of Acronis True Image, which is used to back up, clone, and restore data. These two applications cover almost everything you’d need for a drive, although of course they are designed for Windows.

Kingston NV3 2230: A Closer Look

The Kingston NV3 is a single-sided drive at 2TB, which is pretty cool for an M.2 drive. We’ve seen that before, yes, but it still surprises you when you have such a tiny drive in your hands. Drives in this form factor benefit from being single-sided for compatibility. The Steam Deck, for one, is best with a single-sided drive for fitment, given motherboard components and the EMI shield. Other systems that can take longer drives, from 2242 to 2280, can meet this requirement at higher capacities more easily, as there is space on the board for between two and four NAND flash packages versus just one for 2230.

That brings one thing into question: Kingston has suggested a 4TB SKU is on the way for this drive, can it be single-sided? Currently, consumer flash is stacked at up to 16 dies per package, which, with 1Tb dies, means 2TB at the most.

32-stack technology is out there, but it's not reasonable for this type of device. Even 16-die packages can be challenging at higher speeds, as we saw with some 1TB TLC-based 2230 drives from Phison that had trouble in rare edge cases. The NV3 would have to either be double-sided or use 2Tb dies at 4TB. The latter is very possible – we may even see 2Tb dies at lower capacities – and it will be interesting to see what Kingston does.

Here we can see the controller, the Silicon Motion SM2268XT2, and the single NAND flash package. The XT2 is very similar to the original XT, the main difference being that the XT2 can handle up to 3,600 MT/s flash, while the XT is designed for 3,200 MT/s. There aren’t too many drives with either controller variant, the Kioxia BG6 probably being the most well-known one with the XT. The subtle change is similar to Phison’s “upgrade” from the E27T to the E29T – with flash support going from ONFi/Toggle 5.0 to 5.1 – with the latter being found on Micron’s 2600, a drive we may review in the near future. In practice, there’s no huge benefit to increasing the I/O rate as a four-channel PCIe 4.0 SSD controller can saturate the interface with just 2,400 MT/s, but enabling a higher rate can improve performance and/or efficiency in some cases.

The BiCS8 QLC flash is an upgrade over the regular NV3's BiCS6 QLC. There are improvements across the board as the newer architecture has better power efficiency and, in general, some of the best latency results we’ve seen. Specifically, the SanDisk WD Blue SN5100 is using this flash to very good effect, offering a TLC-like experience. It’s perfect for an M.2 2230 drive. One thing we can point out is that the 1TB and 2TB SKUs are slightly heavier than the 500GB SKU, which could mean heavier flash packages or more circuitry. It could also imply the use of 2Tb dies, now or in the future. However, it’s not uncommon for flash packages to maintain size and weight regardless of functional die count because fillers and dummy dies are used for uniformity.

None of that matters for the end user, but when looking at the weaker performance specifications for this drive and remembering that Kingston’s NV series of drives uses variable hardware, we must caution that the hardware might not remain what we’re reviewing today. Any new hardware must meet existing specifications, so even minor discrepancies can be a tell. Given the relatively low TBW and even lower DWPD than the 2280 variant, it’s likely this drive will only have QLC or lower-end TLC flash. The future 4TB SKU, if it comes, will be more standardized given the rarity. We think that, considering all these factors, the 500GB SKU should mostly be avoided unless it’s your only choice.

MORE: Best SSDs

MORE: Best External SSDs

MORE: Best SSD for the Steam Deck

Comparison Products

The Kingston NV3 2230 has its work cut out for it: it’s late to the M.2 2230 game with significant competition already in place. There are multiple tiers of competitors which we will briefly outline here so you can get a good feel for what drive is right for you. Before we get into that, we remind the reader that the Steam Deck is only PCIe 3.0, so the impact will be different. Power efficiency and latency results do transfer over at lower interface speeds, but conversely, less emphasis should be placed on sequential results.

The first tier is QLC-based drives with higher-end PCIe 4.0 performance. Right now, the main competitor there is the Crucial P310 2230. We scored that drive highly because it offers excellent performance and power efficiency, backed by reliable hardware. The NV3 2230 is rated for lower speeds, but our sample is using decent hardware. The issue here is that the specifications allow Kingston to use worse hardware down the line, as was the case with the full-length NV3.

The next tier, if you can call it that, is high-end 4.0 drives using TLC flash instead. This includes the Corsair MP600 Mini E27T variant and the PNY CS2342. These drives are going to be faster than the 2230 NV3 and P310 with higher TBW. However, these drives also cost more and will bump up against capacity limits with pricing more readily, especially if the NV3 gets 2Tb flash dies later on for 4TB. Generally, extra performance and write endurance aren’t necessary for where M.2 2230 drives are going to go, so that’s a secondary concern.

As some devices like the Steam Deck are PCIe 3.0, slower TLC-based drives like the WD Black SN770M can get you high performance with older hardware. This is also true of QLC drives such as the Sabrent Rocket Q4 and TeamGroup MP44S. At one time, all three of these drives – and similar ones from other brands – were pioneers of the form factor. Before they came out, options were limited to OEM drives and much older hardware. However, these drives can easily get the job done on a Steam Deck, though a 4.0 host will do better with one of the newer, faster drives listed above.

Trace Testing — 3DMark Storage Benchmark

Built for gamers, 3DMark’s Storage Benchmark focuses on real-world gaming performance. Each round in this benchmark stresses storage based on gaming activities including loading games, saving progress, installing game files, and recording gameplay video streams. Future gaming benchmarks will be DirectStorage-inclusive and we also include notes about which drives may be future-proofed.

The number one use of an M.2 2230 drive today is probably gaming, thanks to the soaring popularity of portable, small-form-factor systems. Certainly, if you’re getting a drive for the Steam Deck, your primary interest is gaming. There are a few characteristics worth analyzing for this, the first and most important of which is game load time. This correlates with 3DMark performance to some degree, with the latency results giving the most understandable metric – nobody wants to wait for a game to load, especially in multiplayer.

What we see here is that drive bandwidth is the number one factor for score. The Rocket Q4 and MP44S are QLC-based, sure, but also limited to around 5 GB/s for reads. Yet the QLC-based P310 tops the list, and the NV3, also with QLC flash but at a lower 6 GB/s, is in the middle. This requires some explanation because QLC flash is inherently slower than TLC, but this is often not reflected in benchmarks or even real-world feel, as is the case here.

The first thing to keep in mind is that it’s commonly believed that games load with random, 4KB I/O at low queue depth. It’s true that low queue-depth transfers are the most common for gaming and consumer workloads, and that 4KB random read performance does correlate with storage responsiveness to a significant degree. However, games often rely on sequential reads, which incur higher I/O. This reduces any potential gap between TLC and QLC flash, not to mention it also eliminates the need for DRAM on a drive. In the future, with DirectStorage, this may change as it can require larger, random I/O with very high QD. We’re also not talking about niche scenarios, such as a mod-heavy Minecraft build or something along those lines.

The second thing to know is that QLC flash is designed with optimizations to improve read performance. More performance could be realized with DirectStorage, and some manufacturers even have software or driver-based optimizations – Solidigm’s Synergy 2.0 comes to mind. However, we’re talking about hardware or architectural optimization, with perhaps some assistance from firmware. Most of the I/O or access latency is at the level of the media, that is, the NAND flash. Improvements in other areas, such as the drive having DRAM and pathway optimization, or even just running the drive over CPU rather than PCH lanes, have a much smaller effect. Therefore, it pays to keep QLC’s plane count down with algorithms designed to boost everyday reads. There are a number of ways to do this, with the essential purpose being to reduce latency. This can be achieved by superior mixed I/O or program-interrupted read performance algorithms, by leveraging knowledge/feedback on expected voltage levels for superior start states, by leveraging data locality, and more.

What this means is that, as small as the difference is for game load times between one NVMe drive and another, it’s even smaller than you would expect given the flash type. QLC drives are perfectly suitable for read-heavy workloads, which include gaming. The difference between the 58µs Rocket Q4 and 39µs P310 is going to be really small, too. Where you’re going to see a real difference is how the drive handles edge cases with multiple I/O simultaneously, longer/larger updates, games with a million small files, a fuller drive, and so forth. In these cases, it’s worth jumping up to the NV3 from earlier QLC-based 2230 drives, since it’s faster and better prepared for those scenarios, but going with a TLC-based drive like the MP600 Mini (E27T) is even better.

Trace Testing — PCMark 10 Storage Benchmark

PCMark 10 is a trace-based benchmark that uses a wide-ranging set of real-world traces from popular applications and everyday tasks to measure the performance of storage devices. The results are particularly useful when analyzing drives for their use as primary/boot storage devices and in work environments.

People do engage in productivity tasks with M.2 2230 drives, too, thanks to the form factor being used in the Microsoft Surface and other lightweight, portable computers. Some of them, including Lenovo’s Thinkpad, have sometimes opted for 2242 instead. 2242 potentially allows for higher capacity with double the NAND flash package potential. You can also extend 2230 drives to fit in 2242 slots at lower capacities. Most of these devices don’t allow for thicker, double-sided drives, which is why most 2230 and 2242 drives are single-sided. The relatively small space allowed for these drives can make cooling difficult, especially with the Deck’s M.2 SSD EMI shield, so that’s something to keep in mind when buying.

PCMark 10 is also bandwidth-bound, which means newer drives will perform better even if the hardware isn’t necessarily faster in many real-world tasks. Still, to add on to the examples we gave in the previous section, higher speeds can correspond to superior responsiveness. One reason for this is that higher-bandwidth drives will have flash rated at a higher I/O rate, which, aside from implying the flash is newer and therefore may have additional optimizations, means that the delay or latency is reduced. Again, this is tiny compared to the average latency of the flash. Having more bandwidth available also means some tasks can be completed more quickly, which in rare cases could reduce bottlenecks but more generally can improve power efficiency.

In this case, the QLC-based NV3 has no problem keeping up with the fast, TLC-based Black SN770M. This shows the power of BiCS8 QLC flash and the SM2268XT2 controller. The argument could be made that none of this really matters for Microsoft Word or YouTube in Chrome, but with pricing being an important component for drive selection, it’s worth looking at the experience you get from the drive. We all know someone who keeps more tabs open than is sane – and that someone might be you – and if your computer is not CPU or RAM-bound, your storage can be a bottleneck in terms of responsiveness. This is even more true if you have many apps and/or games open, are doing content creation simultaneously, Discord is lagging your GPU acceleration, and so on.

Why would you get a less responsive drive for nearly the same price? Why shun QLC flash – which is used on the P310, the dominant winner in this test – when it can actually be faster in everyday use than many TLC-based drives? Know your workload before you buy.

Steam Deck Testing — Gaming, KDiskMark, and Temperature

The Steam Deck is not the only portable gaming system in town, but it was the first and most popular to take M.2 2230 SSDs. While some systems have moved on to fit 2280 length drives, 2230 remains popular for many systems and such drives will work fine in longer slots with the proper standoff or extender. The Deck operates in PCIe 3.0 mode for its SSD which limits maximum bandwidth but that has less of an impact on responsiveness/latency and the Deck is still useful for gauging drive temperature and power efficiency.

Our current testing for the Deck involves analyzing game load times for some popular games. This is probably the most important metric for gamers, but the difference between one SSD and another can be small. We also use KDiskMark, a CrystalDiskMark-like substitute that uses the flexible I/O (FIO) tester instead of diskspd for its underlying benchmarks. We also check the drive’s maximum temperature during this test.

The tests in this section are run under the stock Arch-based SteamOS Linux platform but our other tests are conducted as per our normal reviews, using Windows. Many portable gaming systems today use or can use Windows with multi-boot also as an option. This testing section is instead designed to give an idea of Linux performance, which does involve the use of Proton.

As we’ve stated many times, the difference in game load times from one NVMe drive to another is usually minimal. There will always be some outliers, and, additionally, it can still be informative to look at smaller differences as the games being tested have distinct loading patterns. Let’s expand on this a little bit, as gaming is a primary target workload for M.2 2230 SSDs.

We’ve said before that BiCS8 flash can have advantages in latency, particularly 4KB random read latency, yet we see the NV3 is dead last in Elden Ring. Elden Ring’s level loading is, in fact, almost all sequential, with much of the I/O being larger than a typical 16KB page. This means it favors drives with more bandwidth, like the P310 and CS2342. The P310 uses newer flash than the CS2342, which might give it an edge. Larger I/O can transfer faster, and the same amount of data takes fewer transfers in comparison to smaller I/O, which can make analysis counterintuitive to the uninitiated, especially as SSDs are designed to get maximum performance with parallelization.

That’s more difficult to achieve with a low queue depth. We test both random and sequential performance at QD1, which we’ve stated tends to reflect real-world transfers pretty well. This also applies to games where your average transfer is effectively QD1. Some multiplayer games might have smaller, more random transfers at higher queue depths, though we don’t suspect many people are competitive gaming on a portable gaming system. More realistically, we’d be looking at edge cases – a very full drive updating a game that’s immediately played – where a TLC-based drive like the CS2342 is probably a safer bet. Or maybe you do have a majorly modded game, in which case we would again lean towards TLC flash. For everything else, the NV3 is plenty fast.

We also test in KDiskMark to give CDM-like results. This also shows how the Steam Deck in particular handles I/O. Let’s be honest, it’s not a super-powerful system, and therefore it often looks last-gen on this test. In fac,t you could definitely get by with a Gen3 SSD, but we would recommend against that. Gen4 drives have newer hardware that’s often much more efficient, which is easier on the system. The only exception might be the OEM BC711 from SK hyni,x which has the same hardware as the Gold P31 with embedded DRAM, something rare in M.2 2230 drives. It can also be found in M.2 2242 inside the Tube T31 and Beetle X31 for those willing to shuck. Otherwise, we reiterate that the Gen 4 NV3 is more than enough.

Transfer Rates — DiskBench

We use the DiskBench storage benchmarking tool to test file transfer performance with a custom, 50GB dataset. We write 31,227 files of various types, such as pictures, PDFs, and videos to the test drive, then make a copy of that data to a new folder, and follow up with a reading test of a newly-written 6.5GB zip file. This is a real world type workload that fits into the cache of most drives.

File transfers are probably the most common storage workload that you’ll do, at least the most common one that will push the drive to some extent. Our DiskBench methodology reflects a real-world transfer. Usually, this will fit inside the drive’s pSLC cache, meaning you don’t get an idea of how the drive performs in its native mode – see our Write Saturation section for a more detailed breakdown. The fact is, most of everything you do with an SSD should fit inside the cache. Your operating system is also using system memory or RAM to cache data. Caching is useful in both cases because it allows smaller writes to be combined in a way that improves performance and reduces drive wear.

This is another case where drives with higher bandwidth will score better. This is, in part, due to the fact that bandwidth can improve even in cases where you’re not getting anywhere near the theoretical maximum. That maximum is designed for large I/O with lots of queue depth, but a faster bus or I/O rate can trickle down to improve performance even for regular I/O. For that reason, a drive like the NV3 can outperform the Rocket Q4 and MP44S in this test, and the P310 can outshine everything. The P310 uses the same controller as the MP600 Mini (E27T) and QLC flash rather than TLC, yet it scores higher. One reason for this is that the MP600 Mini has a very small pSLC cache, and another is that the P310’s QLC flash is of a higher generation. The former is more impactful, but don’t be misled by this result. The MP600 Mini will be faster in many cases.

Synthetic Testing — ATTO / CrystalDiskMark

ATTO and CrystalDiskMark (CDM) are free and easy-to-use storage benchmarking tools that SSD vendors commonly use to assign performance specifications to their products. Both of these tools give us insight into how each device handles different file sizes and at different queue depths for both sequential and random workloads.

It should be immediately apparent that the NV3 has some problems in ATTO. The read chart indicates that the drive’s performance is well below the competition’s. This is almost certainly due to the controller. Our NV3 (2280) review had the same controller but with BiCS6 QLC flash, which is inferior to the BiCS8 on the 2230 version. That drive has the same issue, and for that matter, we see some similarities with the Adata Legend 960 and Legend 960 Max. Those drives are using a different SMI controller, the SM2264, so this seems to be an SMI problem. No such issue exists for the SM2508 so we’re guessing we can narrow this down to a Gen 4 problem. We do, in fact, see reduced sequential read performance from the drive in CDM, too, so this is a weakness of the hardware. It’s not make-or-break. Sequential write performance is less important, but the NV3 performs as expected with QLC flash and the rated speed.

Random write latency is less good. Luckily, this isn’t a huge deal. Random writes are designed to be coalesced or combined through caching, which can happen in system memory or SSD volatile memory, but even if it hits non-volatile medium, it’s pSLC rather than QLC. These writes will be combined in the cache before being written to the native QLC flash. Arguably, you will sometimes notice performance gaps here from, say, large Windows or game updates, but this should be hidden by caching unless the drive is very full. The NV3’s read latency is excellent, matching the P310's, which also uses QLC flash, and within reach of some of the best 2230 drives. We have no complaints about that.

Sustained Write Performance and Cache Recovery

Official write specifications are only part of the performance picture. Most SSDs implement a write cache, which is a fast area of pseudo-SLC (single-bit) programmed flash that absorbs incoming data. Sustained write speeds can suffer tremendously once the workload spills outside of the cache and into the "native" TLC (three-bit) or QLC (four-bit) flash. Performance can suffer even more if the drive is forced to fold, which is the process of migrating data out of the cache in order to free up space for further incoming data.

We use Iometer to hammer the SSD with sequential writes for 15 minutes to measure both the size of the write cache and performance after the cache is saturated. We also monitor cache recovery via multiple idle rounds. This process shows the performance of the drive in various states as well as the steady state write performance.

The 2TB NV3 (223) writes in the pSLC cache mode at 5.77 GB/s for over 92 seconds. This is a massive cache, using all of the flash available in the single-bit mode. As the drive is converted from 4-bit QLC flash, that means the cache is wholly one-fourth the maximum capacity. This capacity can include flash not exposed to the user or OS, and given the normal binary-to-decimal translation, this means that the 533GB cache seen here is possible.

The drive is then forced to wait for data to be moved from the cache to the native flash, which is laborious and slow – this drive manages just under 300 MB/s when looking at the steady state average. While this is poor by today’s standards, it’s actually not that bad for QLC flash. It’s fast enough for any portable device that relies on gigabit Ethernet or a wireless connection and is plenty fast to match memory cards, too. External storage at 10 Gbps or higher will be bottlenecked by the USB connection. It’s true that this is simplifying things, but it does mean that most 2230-accepting devices won’t have a problem with this potential slowdown, but faster ones coming down the pipeline may be better served by faster storage solutions.

Power Consumption and Temperature

We use the Quarch HD Programmable Power Module to gain a deeper understanding of power characteristics. Idle power consumption is an important aspect to consider, especially if you're looking for a laptop upgrade as even the best ultrabooks can have mediocre stock storage. Desktops may be more performance-oriented with less support for power-saving features, so we show the worst-case.

Some SSDs can consume watts of power at idle while better-suited ones sip just milliwatts. Average workload power consumption and max consumption are two other aspects of power consumption but performance-per-watt, or efficiency, is more important. A drive might consume more power during any given workload, but accomplishing a task faster allows the drive to drop into an idle state more quickly, ultimately saving energy.

For temperature recording we currently poll the drive’s primary composite sensor during testing with a ~22°C ambient. Our testing is rigorous enough to heat the drive to a realistic ceiling temperature.

Power efficiency is a shining point for the NV3. It’s the most efficient drive in our lineup and by a significant amount at that. Considering that we have the fastest QLC-based and the two fastest TLC-based 2230 SSDs on the list, the NV3 is in a good place as a compromise between performance and power efficiency. We do point out that this is in our Windows testing and that the max or peak power draw isn’t anything special.

That said, this drive will be very efficient, reducing power consumption on your battery-powered devices. Is this amount significant in daily use? No, but over the device's lifetime, this could add up when considering battery wear. Further, the NV3 runs cooler as a resul,t which reduces the likelihood of throttling – especially as devices age and get clogged up – and mitigates general wear. QLC flash is perfect for this kind of SSD, as the performance you need is there, and you can get the best cost per GB in a form factor that normally makes higher capacities challenging. If the drive runs cooler, as we would expect from BiCS8 QLC flash in particular, that’s just an added bonus.

Test Bench and Testing Notes

We use an Alder Lake platform with most background applications such as indexing, Windows updates, and anti-virus disabled in the OS to reduce run-to-run variability. Each SSD is prefilled to 50% capacity and tested as a secondary device. Unless noted, we use active cooling for all SSDs.

Kingston NV3 2230 Bottom Line

This one was difficult to grade. We definitely think the NV3 has some advantages over older M.2 2230 SSDs, but it just doesn’t hit all the marks like the Crucial P310 does. The bright spots: excellent power efficiency, good all-around performance, and pretty good pricing as well. The last one is probably the most important since this is a QLC flash drive. With QLC, you’re going for capacity as well as price, and in terms of GB per $, the NV3 is unsurpassed at 2TB. However, the P310 is only $11 more at the time of review, and it’s a more complete package. If you want more performance, you jump up to the TLC-based WD Black SN770M. This is an oversimplification, but right now that’s the general approach for 2TB.

If we drop to 1TB, and to be clear, we don’t think you should get a QLC-based drive at any capacity lower than that, the NV3 is still the best value. For $10 more right now, you can get the Sabrent Rocket 4 2230 if you want TLC flash. The P310 is also in that range and is a better drive on paper than the NV3, so you have three good options there. We think all the drives listed so far, with the SN770M as an exception, are power-efficient and run cool enough on any 2230 host. Battery savings are not a real factor, as the difference will be small, but in terms of longevity, we do think getting a more efficient drive can be worthwhile. We don’t think the use of QLC flash is an issue here, and therefore, the NV3 is a good choice.

We should also touch on use cases not involving the Deck, which is limited to PCIe 3.0 speeds, which makes the 2230 drive price more relevant. There are portable PCIe 4.0 systems, but 2230 is also useful for small-form-factor systems, HTPCs, mini PCs, and the like. Power efficiency is usually king in these use cases, making the NV3 2230 incredibly effective given its low price. This would be even more the case if the 4TB model ever comes to light. You can always use the drive in a regular slot later with an M.2 extender, if needed. It’s not a bad investment.

It’s hard to dislike the NV3 because it uses excellent flash and a reasonable controller. However, we have to lower its score further due to history. Kingston’s NV series of drives is notorious for hardware swaps, and that’s one reason the specifications are so modest. This risk keeps the drive from being a champion in our minds, but we emphasize that the NV3 2230 is about price before all else. The fact that our sample is also super efficient, with the best QLC flash available right now, is just icing on the cake. If that fits your priorities, then this drive is probably the best one on the market right now. If, on the other hand, you’re looking for TLC flash, maximum performance, or high endurance, then it’s not the best option. It’s perfect for the Deck, but for faster and future hardware, the NV3 2230 will eventually be left behind.

MORE: Best SSDs

MORE: Best External SSDs

MORE: Best SSD for the Steam Deck

For the money, the X870I Edge Ti Evo Wifi (man, that's a mouthful!), gives you everything you’d expect from the X870 platform, including two USB 4 (40 Gbps) Type-C ports, 5 GbE and Wi-Fi 7, a quality audio solution, and of course, the three M.2 sockets (one PCIe 5.0). Additionally, it features a DIY-friendly design with EZ features on the main M.2 socket and Wi-Fi antenna, as well as an EZ Conn header for connecting to MSI EZ series fans and MPG P-Series liquid cooling. The silver and white aesthetic, along with its array of mirror-finish design elements, is sure to look great inside one of the best mini-ITX cases.

Overall, performance during testing was above average, with strong results in gaming and office activities. Surprisingly, it didn’t perform well in the Procyon video editing test, but it was the top result in photo editing. Please keep in mind that most results are very close, and the difference between worst and first is negligible at times. Below, we’ll examine the board's details and determine whether it deserves a spot on our list of the Best Motherboards. But before we share test results and discuss details, here are the specifications from MSI’s website.

Specifications of the MSI X870I Edge Ti Evo Wifi

Inside the Box of the MSI MPG X870I Edge Ti Evo Wifi

• Installation guides
• Cable stickers
• EZ Wi-Fi Antenna
• 5-in-1 XPander card
• 5-in-1 Xpander USB 5 Gbps cable
• (2) SATA cables
• USB Drive
• EZ Front panel cable
• 1 to 3 EZ Conn-cable (v2)

Inside the box of the Mini-ITX board are several accessories to help you get started. You get the typical things, like Wi-Fi antennas, SATA cables, and guides, among a few other goodies. We’ve listed everything that comes in the package below.

Design of the Edge Ti Evo Wifi

TThe X870I Edge Ti Evo Wifi uses a white, 12-layer PCB and MSI’s large Frozr design heatsinks to cool the VRMs underneath. On top of the VRM heatsink is a parallel line pattern with venting for the ridiculously tiny fan hiding below. You’ll find MSI branding in the form of the dragon, the MPG (MSI Performance Gaming) logo, and the model name emblazoned on the heatsinks. Everything outside the heatsinks is white, except for one connector (a gray fan header). It’s a clean look and will look great in a white-themed build, but where are the Project Zero versions of Mini-ITX boards to really clean things up?

Starting with the top half of the board, we get a better look at the reflective finish (and my knuckles) and the VRM venting on the VRM heatsink. Above that is the single 8-pin EPS connector, which powers the processor. The square 4-pin next to it isn’t listed on the documentation (maintenance access, perhaps?), but the 2-pin header immediately to the right is for chassis intrusion. Next, we find three 4-pin fan headers (CPU_FAN1, PUMP_SYS1, and SYS_FAN1). Each header supports PWM and DC-controlled devices. Output varies across each at 2A (CPU), 3A (Pump), 1A (Sys), but is plenty to support an AIO and case fans. You can control the fans through the BIOS, MSI Center’s User Scenario, or Cooling Wizard.

The EZ Conn-Header V2 sits between the fan headers and the DRAM slots. You can connect MSI’s compatible fans and AIOs, or use the included 1-to-3 EZ Conn-Cable (V2), which splits into 3-pin ARGB, 4-pin fan, and USB 2.0 ports.

Next are the two reinforced DRAM slots, which feature locking mechanisms at the top (where there’s room). MSI lists support up to wicked-fast DDR5-10000, but that’s only with the APUs/Ryzen 8000 series processors. 9000-series processors are still rated to DDR5-8400, but that’s still fast for the platform. Our DDR5-8000 kit didn’t work, but it’s not listed in the memory QVL either, and there were no miracles. Finally, in the upper-right corner is the 24-pin ATX connector, which powers the board.

Power delivery to our little board consists of 11 phases, with eight dedicated to Vcore. Power heads from the Renesas RAA229139 8-phase (duet rail config) controller and on to the 110A SPS MOSFETs. The 880A isn’t a lot, but it should handle flagship-class processors, even the Ryzen 9 9950X3D. The actively cooled VRMs ran hot during stress testing on our Ryzen 9 9900X (the fan was audible but not bothersome), but were within specification even in the worst circumstances. If you plan to push these high-end processors for extended periods, you’ll want good airflow to keep temperatures down.

Focusing on the ‘bottom’ half, we’ll start with the 5-in-1 Xpander card protruding from the motherboard like a misshapen SO-DIMM. The add-in-card (AIC) provides a total of one Gen 4 M.2 slot, one USB 3.2 Gen 2x2 (20 Gbps) port (with 27W PD), two USB 3.2 Gen 2 (5 Gbps) ports, two SATA ports, and a set of front panel headers. It’s a lot of connectivity, but if you use it extensively, the cables stick up and look unsightly. There’s not much you can do about it with this board, though.

At the bottom of the board, we’ll work our way top-down from the M.2 Shield Frozr II heatsink, which pops off with the push of a button. It’s cooling the primary (PCIe 5.0 x4) M.2 socket. On the same raised PCB is a USB 2.0 header and a 3-pin ARGB header. Underneath that is a silver heatsink that cools the PROM 21 chip. The sole PCIe slot (PCIe 5.0 x16) is along the bottom edge. To the left is the audio solution, based on the Realtek ALC4080 codec, and the front panel audio connector. Finally, to the right of the PCIe socket is the main front-panel header.

If you’re looking at the rear IO area and it feels a bit sparse, you’re not alone. The pre-mounted IO plate features a white background with dark, easy-to-read letters. You get three total Type-C ports (two 40 Gbps, one 10 Gbps), but only five USB 3.2 Gen 2 (10 Gbps) Type-A ports. An HDMI output for integrated video is located on the left, with the Clear CMOS and BIOS Flashback buttons positioned in the middle. On the right is the quick-connect Wi-Fi 7 (5.8 Gbps) and the audio stack (mic-in, line-out plus SPDIF out). Five USB-A ports may be a concern for some users, but the front panel or a hub can be used as an alternative if needed.

MORE: Best Motherboard Deals

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

Firmware

With X870, MSI has updated its BIOS, renaming it “Click X,” and improved the aesthetic and functionality in the process. The company changed the format with the headings on the left side, details in the middle, and system status information on the right. Like most other board partners, it features an informational EZ Mode, which allows for limited functionality, including Game Boost, PBO, and more. Overall, I like the new layout, and maneuvering around is intuitive after a short time, though it does take some getting used to. The black background with purple/blue highlights and gray labels is easy to read, and most of what you need is readily available at your fingertips.

Software

MSI Center is a single utility that offers a wide range of functionality. From hardware monitoring to RGB control with Mystic Light, the software features numerous applets and serves as a one-stop shop for downloading additional utilities, including overclocking and fan control. Those looking to use the Gamebar feature, Super Charger, or any other utilities MSI offers will find them all in MSI Center.

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 publicly available non-beta motherboard BIOS. The hardware we used is as follows:

Test System Components

Benchmark Settings

MORE: Best Motherboard Deals

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

Benchmark Results and Final Analysis

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. The Windows power scheme is set to Balanced (default) for this baseline testing, so the PC idles appropriately.

Synthetic Benchmarks

Synthetics offer a valuable method for evaluating a board's performance, as identical settings are expected to yield similar results. Turbo boost wattage and advanced memory timings are areas where motherboard manufacturers can still optimize for stability or performance, though, and these settings can impact specific testing scenarios.

Through our synthetic benchmarks, the Edge Ti Evo performed well overall, with an average or above-average result across most tests. The Procyon video editing was, for some reason, our slowest result, but it was the second fastest in photo editing. There’s nothing to worry about so far.

Timed Applications

Our timed results were also good, with our board performing very well in Handbrake (producing the fastest and second-fastest results) and averaging in the Corona and LAME testing. All good in these tests, too!

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 an even more popular and good-looking game in Cyberpunk 2077. 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 performance was some of the best we’ve seen in the actual games. In Cyberpunk, it had the highest minimums and the second-fastest average frame rate, at 107 and 141, respectively. F1 24 was similar, with solid minimum and average frame rates. Overall, the MSI board performs very well across a wide variety of tasks.

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 the above, we will not be overclocking the CPU. However, we will try out our different memory kits to ensure they meet the specifications.

Our memory testing results sang a familiar tune, where our Klevv DDR5-8000 kit booted to Windows and ran the AIDA64 memory bandwidth test, but couldn’t pass the stress test. As many know, that’s well past the price-to-performance sweet spot anyway, and our kit isn’t on MSI’s Memory QVL list, so we couldn’t expect it to work. The TeamGroup DDR5-7200 kit worked as expected, as did our Kingston DDR5-6000 kit, which we use for baseline testing (it has been finicky on the latest ASRock boards).

Power Consumption / VRM Temperatures

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

Power usage during our testing was also within normal limits. Idle sat at 75W, which is on the efficient side, with a peak of 248W. This places it towards the efficient side of our results, using less power than most other boards.

VRM temperatures ran hot during our stress test. Our external sensors peaked at almost 76 degrees Celsius, while the internal sensor crept up to a very hot 87 degrees Celsius. If you plan to run a chip like our Ryzen 9 9900X (or any higher-powered chip) in stress-test-like situations, ensure you have adequate airflow in your chassis, as excessive heat can cause the chip to throttle.

Bottom Line

MSI’s MPG X870I Edge Ti Evo Wi-Fi was late to the Mini-ITX party, but does arrive on the scene as one of the most well-equipped models, specifically in storage, thanks to the inclusion of a 5-in-1 Xpander card that adds another M.2 socket, among multiple other ports and headers. In addition to being the only Mini-ITX board on this platform with three M.2 sockets, you get a clean, premium white aesthetic, a solid audio solution, and fast networking, rounding out this comprehensive offering.

The elephant in the room is the price. At $449.99, it’s tied for the most expensive, along with Asus’ ROG Strix X870-I Gaming Wifi. Between these two, the Asus offers an audio solution with an ESS-based DAC, more USB ports on the rear IO, and more robust power delivery, but it has slower Ethernet (2.5 GbE). It looks good, but if you’re in the market for a white board, Asus’s option is out. Another premium white Mini-ITX option in the X870 chipset space is Gigabyte’s X870-I Aorus Pro Ice, which costs $284.99. This board offers comparable audio and networking, two M.2 slots, and a white design aesthetic similar to the MSI—though I prefer the MSI's look. While ASRock's B850I Lightning Wifi is a competitive sub-$200 board, it is more limited in connectivity due to the chipset, and is not technically a direct competitor, as it’s B850-based.

In all, the X870I Edge Ti Evo is a feature-packed Mini-ITX motherboard that offers users the best of the platform in a small form factor. And while some features make the $449.99 price tag more palatable, many may not need the three M.2 sockets and can get by with something that costs significantly less. At the same price, I prefer the MSI over the Asus, but I wish these were both under $400, as they offer little more than the sub-$300 Gigabyte Mini-ITX board. The Edge Ti Evo didn’t make our top motherboard list, but if you’re in the market for a premium Mini-ITX motherboard, MSI’s X870I Edge Ti Evo should be on your consideration list.

MORE: Best Motherboard Deals

MORE: Best Motherboards

MORE: How To Choose A Motherboard

MORE: All Motherboard Content

Simulators are dominated by the best gaming monitors with ultra-wide and mega-wide screens in either 21:9 or 32:9 aspect ratios. The mantra is, “the wider the better,” but height is essential too. Enter the 45-inch 21:9 category, where you get tremendous screen area and, if the curve is extreme enough, the true feeling of being in a cockpit.

LG’s Ultra Gear 45GX950A is ideally suited for driving and flying sims. It delivers a high pixel density of 125ppi with a 5120x2160 (5K2K) array on an OLED panel with an 800R curve. It also sports a 165 Hz refresh rate with a dual-mode switch to 330 Hz at 2560x1080, plus Adaptive-Sync, DisplayHDR 400, and wide-gamut color. Let’s take a look.

LG Ultra Gear 45GX950A Specs

If you’re building a driving or flying simulator, a monitor like the 45GX950A is a natural choice. You could expand your view with multiple screens, but then there are those annoying dividing lines that you can’t unsee. A 32:9 screen brings the width, but current 49-inch models aren’t very tall. They impart a view that’s like looking out of a racing helmet, which is fine for driving but not so much for flying.

45 inches at 21:9 means 18 inches vertical, more than a 32-inch 16:9 monitor. To bring the sides into the user’s peripheral vision, the 45GX950A has a very tight 800R curve radius. That means a circle of them would only be 1.6 meters across. It’s so tight that you can hear your voice bounce off the sides of the screen — it’s a unique experience.

Onto the important stuff: The 45GX950A is a very high-resolution panel with a resolution of 5120x2160 pixels. That equates to 125 pixels per inch, denser than a 27-inch QHD monitor but not quite as dense as a 32-inch Ultra HD screen. It’s more total pixels — over 11 million — so fast frame rates will require a stout video card.

However, you won’t need to worry about motion blur, thanks to the 45GX950A’s OLED panel. It doesn’t have a Quantum Dot layer, but the color gamut covers almost 98% of DCI-P3, so the vivid saturation is there. You also get high brightness with 1,300 claimed nits in HDR mode for a 1.5% window. I measured 673 nits from a 25% window so that tracks. There are two levels of variable brightness as well, so SDR peaks can top 450 nits.

The native refresh rate is 165 Hz, and you get a 330 Hz mode at 2560x1920 resolution. It’s easily accessed with a switch in the OSD. The 45GX950A is certified G-Sync Compatible and for FreeSync Premium Pro as well. There’s plenty of bandwidth available through the DisplayPort 2.1 input, and you get two HDMI 2.1 ports, along with a USB-C port that supports DisplayPort and delivers 90 watts of power. Two USB downstream ports handle peripherals, and you get a nice pair of internal speakers. The headphone jack is a four-pole design that supports surround sound and DTS Headphone:X.

There is little that the 45GX950A doesn’t have, and all that goodness will cost you $1,600 at this writing. For that sum, you’re getting something relatively unique that brings a new level of immersion and realism to your favorite flying and driving simulators.

Assembly and Accessories

The 45GX950A arrives in a huge box with plenty of crumbly foam inside. It’s needed to support that extremely curved panel, and my sample showed up unscathed. The stand and base are substantial and assemble without tools. Cables include DisplayPort, HDMI, and USB-C. You also get an enormous external power brick that thankfully has a cord long enough that you can put it on the floor.

Product 360

The 45GX950A is impressive from any angle, not only in size but in shape. An 800R curve is tighter than anything you’ve seen before, unless you read my review of the AOC AG456UCZD, which is the same size but has lower resolution. When you sit two or three feet from the screen’s center, the sides aren’t that far from your ears. It’s about as immersive as you can get short of VR goggles.

The stand has a relatively small footprint thanks to its heavy cast metal base. The upright is wide and features a 4.7-inch height range, plus 10/15 degrees tilt and 10 degrees swivel. I appreciated that it set the screen high enough to keep it perfectly vertical with my eyepoint centered. This is especially important for a monitor as curved as the 45GX950A. Any tilt ruins the immersive effect. The height adjustment is firm, nicely dampened, and wobble-free.

The back is perfectly smooth and features a ring of LEDs controllable from the OSD and an LG logo. The cable hole in the upright is trimmed in purple. Underneath the center is a single joystick that controls all monitor functions. The headphone jack is also under the front edge, making it easy to access. The remaining ports are up and under and include two HDMI 2.1 ports, a DisplayPort 2.1 port, and a USB-C port with DisplayPort function and 90 watts of power. You also get two USB downstream ports. A snap-on cover helps to hide the cabling.

OSD Features

The 45GX950A’s OSD appears when you press the joystick. It’s game-focused with a slick look that includes color and fade effects, with important info always displayed at the top.

Game Adjust is the first menu up, and it starts with the 45GX950A’s picture modes. There are 13 in total, including sRGB, HDR emulation, and two calibration memories that work with LG’s Calibration Studio app. Personalized Picture is another memory that connects with the LG Switch app.

Dual-Mode is the refresh rate/resolution toggle, offering 5120x2160 at 165 Hz and 2560x1080 at 330 Hz. You can also change the aspect to 16:9 or 21:9/39” to accommodate games that distort the image at full-screen settings. The switch was seamless every time; I didn’t need to set it up in the Nvidia Control Panel beforehand. LG and I recommend that you make the switch before starting up a game. Additional gaming aids include aiming points and a frame rate counter.

In the Picture Adjust menu, you’ll find the Peak Brightness setting, which is LG’s term for variable brightness. It has two levels, both of which render the image without clipping. On High, you’ll get 457 nits for SDR and 673 nits for HDR from a 25% window pattern. That’s very bright indeed. This menu also has four gamma presets and five color temps. Manual changes the temp by Kelvin value, and Custom uses the RGB sliders to help tweak grayscale tracking. You also get hue and saturation sliders for all six colors. If you get lost during calibration, you can reset just this menu.

Sound comes from two internal speakers that play with 10 watts of power. Or you can use the headphone output, which supports DTS Headphone:X. This requires compatible headphones with a four-pole connector, but once plugged in, the surround-sound experience is incredible. If you typically play with headphones, I recommend going for DTS Headphone:X models. It takes audio to another level.

The 45GX950A supports PIP and PBP, so you can view two video signals at once. The side-by-side view is especially useful if you have a PC and console hooked up. In the General menu are OLED care options that include pixel orbiting, a screen saver and image cleaning, which runs automatically at regular intervals. And the two of the joystick directions are programmable for different quick access functions.

LG Ultra Gear 45GX950A Calibration Settings

The 45GX950A can be enjoyed without calibration in its default Gamer 1 mode. When I tweaked the RGB sliders, I was able to improve grayscale tracking slightly, but gamma was negatively affected. I wound up leaving the settings at their defaults for the best picture. Color is spot-on in either case. I noted that the sRGB mode allowed calibration, which is something I rarely see. I’ve provided my brightness settings below. They are valid when Peak Brightness is turned off. If you use the high or low setting, you might need to lower them a tad to compensate.

In HDR mode, you get six of the 13 picture modes, Gamer 1 is still the best. You can also adjust brightness, and the Peak Brightness setting is available too, which is a nice touch. Most monitors have no brightness options in HDR mode.

Gaming and Hands-on

I rarely have a new gameplay experience, given the over 450 monitors I’ve reviewed in the past 10 years. But the 45GX950A provided just that. You’ve undoubtedly heard the cliché, “it brings the action closer.” Well, it does, truly. I had to readjust my perceptions when playing familiar maps in Doom Eternal and Call of Duty WWII. Getting close to enemies makes them uncomfortably close until you adapt.

Movement and control are blindingly quick. I had no trouble with lag or hesitation of any kind. 165 Hz means there’s a tiny bit of motion blur, but it never distracted during gameplay. I played for a bit at 330 Hz and found it was smoother, but the loss of resolution was significant. While I gained clarity by removing all traces of blur, it was lost in the lowered pixel density. I noted that it was no problem to maintain 165 fps using a GeForce RTX 4090. This same setup typically runs around 220 fps with the best 4K gaming monitors.

The HDR image was simply stunning. The 45GX950A is a bit brighter than other ultra-wide OLED monitors I’ve reviewed. That made up for its lack of Quantum Dot color. It was rich and vibrant with highlights that popped and deep shadow detail. It also had a textural quality that made me want to reach out and touch objects on the screen.

For daily use, the extreme curve took some getting used to. The 45GX950A is different from anything else except the AOC AG456UCZD which I reviewed last year. And the LG has a higher resolution than that screen, making it well-suited for graphics and photo work. There is some image distortion from the extreme curve, but it should be easy for users to adjust. It’s important to keep the screen vertical so your eyepoint is equal to all parts of the screen. The 45GX950A’s stand is high enough for this purpose.

My physical interactions with the 45GX950A were all positive. The menu is well laid out and easy to navigate using the joystick. I didn’t have to calibrate, so there’s little to no setup involved. The only choice to make is whether to use Peak Brightness or not. This is LG’s version of variable brightness, and it has low and high settings. I left it off for work tasks, set it on low for video, and high for HDR content. That extra flexibility was a nice touch that not all HDR monitors have.

Takeaway: Despite its extreme shape and size, the 45GX950A is a superb gaming monitor in all respects. It might be a little too in-your-face for shooters, but I can’t imagine a better screen for racing or flying sims. The dual refresh feature isn’t terribly useful because it lowers pixel density too much. But it was more than smooth and responsive enough at 165 Hz to be addictive. This is truly a monitor to crave. It takes gaming to a new level of immersion and realism.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

The 45GX950A has no direct competitor at this time. The closest screen I’ve reviewed is AOC’s AG45456UCZD, which is the same size and shape but runs at 3440x1440 pixels and 240 Hz. There’s also Alienware’s AW3425DW, AOC’s PD49 and AG346UCD, and Philips’ 49M2C8900.

Pixel Response and Input Lag

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

The 45GX950A moves a lot of pixels around, so 165 Hz is quite fast for a screen with over 11 million dots to manage. The 6 ms draw time is fast enough to prevent almost all motion blur. Remember that OLEDs are smoother at a given refresh rate than LCDs and don’t use overdrive or strobing. That said, there is a slight blur below 200 Hz. There is no blur whatsoever at 330 Hz, where the draw time drops to 3ms.

The input lag score is respectable but not life-changing. 26 ms is fast enough for most players, and it’s quick enough for driving or flying sims. It might be noticed by the best gamers in a twitchy shooter, though. Switching to 330 Hz lowers lag to a competition-worthy 19 ms. The only downside is that pixel density drops to 63 ppi.

Test Takeaway: The 45GX950A is more than smooth and quick enough for flight and race simulation applications. Motion blur is minimal at 165 Hz and non-existent at 330 Hz. Low pixel density at 2560x1080 might be an issue for some, but at full res, the picture is glorious at 125ppi. Video processing is as good as it gets. The only consideration here is the high pixel count. You’ll need a stout video card to drive it at full resolution and refresh.

Viewing Angles

A screen this curved needs to be an OLED for the best possible experience. My sample looked consistent from every angle. In the photos, you can see that there is no change in color, light output, or gamma. Even the top view is decent with just a slight drop in gamma and a barely visible green tint. This is excellent performance.

Screen Uniformity

To learn how we measure screen uniformity, click here.

I had an interesting experience when testing the 45GX950A’s screen uniformity. I ran the sample for four hours, then measured a 10% field pattern. I could clearly see bands of dark and light, and my initial result was over 22%. I ran the screen clean routine in the OLED Care menu overnight and tested again the next day. The banding was completely erased, and the result dropped to a more expected 6.42%. So, if you’ve bought one of these and are dreading a call to LG support, run the screen clean before you pick up the phone. It works as advertised.

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 45GX950A is very bright, especially given its large size. It draws a bit more wattage than its competitors to achieve this, but it’s worth it for that 457 nits for SDR content. That’s the maximum value with the Peak Brightness option set to high. If you turn PB (variable brightness) off, the peak is around 285 nits. Like all OLEDs, black levels and contrast cannot be measured.

After Calibration to 200 nits

Calibration changes nothing about black levels or contrast; they are still unmeasurable. For SDR content, I preferred Peak Brightness off, which is equivalent to constant brightness. When watching videos, PB Low improves the image a bit with brighter highlights. PB High is the go-to for HDR content.

Test Takeaway: The 45GX950A delivers typical OLED SDR brightness and contrast, with output over 450 nits and perfect black levels. It includes variable brightness, which not all OLED monitors have, and you get two settings — a point in its favor.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

The 45GX950A comes set to its Gamer 1 picture mode, which is factory-calibrated to a 6500K white point, BT.1886 gamma, and DCI-P3 color. I noted that the adjustment produced some effects that did not improve picture quality.

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.

In the default chart, only 100% brightness shows a barely visible red error. You won’t see this in actual content. Gamma rides the BT.1886 reference line with slight variations, but nothing to cause complaint. When I adjusted the RGB sliders, gamma became very skewed, resulting in reduced image quality. Shadow and highlight detail lost focus and became clipped making the whole picture flat and drab. Clearly, there is some interaction between the RGB sliders and gamma, which I could not work around. No biggie since the picture was fine to start with.

The sRGB mode has fair grayscale tracking with a slightly warm tone. Gamma is skewed to be too dark at the low end and too bright up top. This reduces both highlight and shadow detail. There is no way to correct this in the OSD.

Comparisons

The 45GX950A fares well against its competitors with a respectable 2.63dE out-of-box grayscale score. It can be enjoyed without calibration. Since I could not improve this value, it falls to last place after the other screens are adjusted. It still has almost no visible errors, so I have no complaint.

In the gamma test, I measured the 45GX950A against BT.1886 as LG intended. The range of values is fairly tight at 0.19, and the deviation from 2.4 is 2.5% with an actual value of 2.34. This is very good performance. However, it would be nice to have a 2.2 option.

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.

In the default color test, the 45GX950A shows no visible errors and a very low average of 1.61dE. Few monitors are this good out of the box. Again, this is to a BT.1886 gamma reference. The sRGB mode is slightly oversaturated in red, magenta and blue, while cyan, green and yellow are on target. This is mainly due to the skewed gamma, which throws off color balance. This can’t be corrected in the OSD, but it is possible to create custom calibrations with a meter and LG’s Calibration Studio app.

Comparisons

The 45GX950A takes second place in the color comparison despite running at its default settings. The other screens have all been calibrated, which means LG has done a good job with its factory setup. This is excellent performance.

In the volume test, you can see that the 45GX950A doesn’t have a Quantum Dot layer, which puts it on par with the AG456UCZD and behind the top four screens, which are more saturated. 97.43% of DCI-P3 is still very colorful, though. I noted some oversaturation in the sRGB mode, but that final volume calculation is an ideal 98.34%.

Test Takeaway: The 45GX950A has superb out-of-the-box accuracy with no need for calibration. Adjustments to the RGB sliders caused some gamma issues, which should be addressed in a future firmware update. The gamma reference here is BT.1886, which is fine, but a 2.2 choice would be welcome. Color volume is very good at nearly 100%, but not quite as high as the Quantum Dot screens I’ve tested.

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.

The 45GX950A switches to HDR mode automatically when fed an HDR10 signal. It leaves six of its picture modes accessible and includes adjustable brightness and the Peak Brightness options too.

HDR Brightness and Contrast

I measured a 25% window pattern with Peak Brightness on high and came up with 673 nits. LG’s claim of 1,300 nits from a 1.5% window is not in doubt. This is a seriously bright monitor and that is amplified by its size. If you play in a dark environment, you might be reaching for the brightness slider. Few monitors allow adjustments in HDR mode, but the 45GX950A does. I recommend Peak Brightness high for all HDR content, but if you find that too bright, you can dial it down to low or off. Black levels and contrast are immeasurable. This is excellent performance.

Grayscale, EOTF and Color

The 45GX950A shows a slightly warm grayscale with most errors around 4dE or less. This is a forgivable issue because red errors are less visible than green or blue ones. The EOTF tracks a little dark from zero to 30% then rides above the line to the tone map transition at 70% brightness. I was able to see all the fine detail in actual content. I noted that lower settings of Peak Brightness dropped the trace further below the reference, making some shadow detail hard to see.

HDR color is generally oversaturated in all colors by 5-15%. This is typical of the HDR monitors I’ve tested. There is plenty of punch in all HDR content and I never had difficulty seeing fine detail. This is solid performance. In the BT.2020 chart, the 45GX950A runs out of color at 85% red, 70% green and 95% blue.

Test Takeaway: The 45GX950A has an HDR image on par with other non-Quantum Dot OLEDs. Color is accurate with natural hues and clear detail. It fully covers DCI-P3 but has a little less BT.2020 volume than a QD-OLED. It is a bit brighter than its competitors and offers adjustments for peak level and variable brightness. Most HDR monitors don’t include those options.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

Gaming monitors should be considered tools for a task. While you can play any game on a 27-inch flat screen, greater immersion and realism can be achieved with ultra-wide, curved and jumbo screens. When you can combine those three attributes into a single display, you have something special. The LG Ultra Gear 45GX950A does this expertly.

At 45 inches and a 21:9 aspect ratio, it’s taller than a 32-inch screen. The 800R curve brings the sides completely into one’s field of view, where you are literally in a 3D window looking out into the game environment. All of this is enhanced by 5120x2160 resolution, 125ppi, and a stunning OLED panel that’s brighter than other ultra-wide displays. It also has flexibility with two levels of variable brightness and adjustable levels in HDR mode. The 45GX950A can adapt to any room lighting scenario. It also manages near-perfect color out of the box and almost 100% coverage of DCI-P3. I’d have loved to see a Quantum Dot layer here, but that’s a minor gripe. I also noted that calibration caused a skewed gamma, but it did not affect my experience.

The 45GX950A is a truly unique gaming monitor, at least for now. It immerses the player like nothing else and works well as an everyday screen. The price isn’t low, but considering it has no direct competition, its value is hard to quantify. If you want realism on par with VR goggles and the ultimate tool for a driving or flying simulator, the LG Ultra Gear 45GX950A is worth considering.

MORE: Best Gaming Monitors

MORE: How We Test PC Monitors

MORE: How to Buy a PC Monitor

But, of course, even office-style gaming chairs are pretty pricey. So it's not surprising that a chair that claims to be on par with high-end task chairs like the Herman Miller x Logitech G Embody, but is priced at around half that, is making waves.

The LiberNovo Omni is not your typical chair. Not only is it (according to the company) the "world's first dynamic ergonomic chair," it's also a Kickstarter phenomenon. LiberNovo launched the Omni on Kickstarter in June 2025 with a $50,000 goal; the project has since raised over $9.5 million. It seems everyone is clamoring for a sub-$1,000 ergonomic office-style gaming chair (shocking, I know). The Kickstarter campaign is over, but the chair is now shipping — and you can still pick one up for around $800 (MSRP is listed at $1,099, but we suspect it will be permanently discounted) in black or light gray.

Assembly

The Omni comes in a large box and requires assembly. It's more complicated to assemble than the average gaming chair, thanks to the dynamic frame and motorized lumbar support, but you should be able to do it by yourself. LiberNovo estimates the Omni will take around 15 - 20 minutes to assemble, and says the process is "comparable to the 15 - 20 minutes required for the Herman Miller Aeron." However, I have a Herman Miller Aeron, and it came fully-assembled, so I'm not sure what this means.

The Omni wasn't too difficult to assemble, but it did take me longer than 15 - 20 minutes. None of the parts were heavy or unwieldy, but this chair had more steps than the average gaming chair. For example, I needed to secure the armrests to the seat so they would move dynamically with the chair — a step most gaming chairs don't require.

The Omni comes with a large, illustrated step-by-step instruction sheet. It's pretty straightforward — some of the illustrations were a little confusing, but any issues were easily cleared up with trial and error. In addition to the assembly guide, the Omni also comes with a separate user manual, as well as an initial setup manual that guides you through adjusting and calibrating the chair for the first time.

The chair came packed neatly in a box that was manageable in size. It came with an accessories box that included a hex screwdriver, hardware, and smaller parts and accessories — everything you need to assemble the chair.

The Omni was trickier to put together than most gaming chairs, but everything is well-engineered, and the parts slot together easily. The chair does feature motorized lumbar support, so you'll need to connect the seat base to the backrest with a short, built-in USB-C cable, but that's all the wiring there is.

Specs