• Capcom and Ubisoft game developers weren't informed about Nvidia DLSS 5 being used by their publisher
• One developer stated they found out 'at the same time' as the public
• Former Rockstar game dev condemns DLSS 5, claiming it's a 'complete AI re-render'

Nvidia is under fire for its latest reveal of DLSS 5, and rightly so, particularly after CEO Jensen Huang's response to the ongoing backlash for its generative AI use — and the latest revelation could make the controversy much worse.

As reported by Insider Gaming, Capcom and Ubisoft game developers apparently weren't informed by their publishers regarding their involvement in Nvidia's DLSS 5. This comes after Jensen Huang stated gamers are "completely wrong" about Nvidia's "content-control generative AI".

One of the Ubisoft developers stated, "We found out at the same time as everyone else". Similarly, Capcom developers were reportedly shocked by the publisher's involvement, considering its "anti-AI" stance.

This is so disrespectful to the intentional art direction of devs. If devs wanted to lean in to hyper realism they would. This also drastically changes key aspects of visuals like character features, focal points, lighting and so on. What a terrible invention. Nvidia should shelve this one 😭

— @kortizart.bsky.social (@kortizart.bsky.social.bsky.social) 2026-03-19T12:33:20.060Z

Game developers across multiple studios and publishers have shared the same sentiment (as evident above). Notably, a former Rockstar developer, Mike York (known as York Street Gaming on YouTube), implied that DLSS 5 poses a significant threat to game art, considering the changes it makes to character models.

"This isn't just some lighting, this is like a complete AI re-render," York said. "You're no longer looking at the game [Resident Evil Requiem] anymore. This is scary. This is like an AI filter over every frame."

While Nvidia and Bethesda's Todd Howard have consistently tried to reassure gamers that DLSS 5 use in games will be controlled and tuned by developers, it hasn't done much to allay concerns — and it's not a big surprise, as many are already opposed to generative AI in sectors outside of gaming.

As of its initial reveal, it takes two RTX 5090s to utilize DLSS 5, and that may scale down to one GPU by the time it launches this autumn. However, gamers, developers, and frankly, the majority of consumers online have made it clear that generative AI is not wanted in gaming, and certainly not when it's changing character model details, which is integral to developers' intended art.

For a long while, I've had my eye on an upgrade for my gaming PC, specifically in the GPU department - I'll be honest, I'm obsessed with attaining that perfect powerful system to provide consistent high frame rates.

While using an RTX 3080 Ti (which, make no mistake, is still an absolute powerhouse), at the 3440x1440 resolution on Dell's Alienware AW3423DWF, I found that it couldn't quite push its way through highly demanding AAA titles like Cyberpunk 2077 (especially with ray tracing enabled).

Fortunately, the lovely folks at Nvidia and Asus reached out to provide not only the ROG Swift PG49WCD OLED super ultrawide monitor ($1,199 / £1,399 / around AU$2,699) but also the TUF RTX 4080 Super OC Edition GPU for testing - if you came here for a quick answer on this specific GPU upgrade path, I can tell you without a doubt, it's worth it.

The RTX 4080 Super storms through the 3440x1440 resolution on maximum settings, and does the same with the PG49WCD's 5120x1440 resolution (especially with DLSS 3). This comes after tests in multiple games, most notably - you guessed it - the incredible Resident Evil 4 remake.

While Capcom's 2023 return hit wasn't necessarily a demanding title, I've always wanted to crank graphics settings to the max without worrying about the heavy performance hit I'd see with the 3080 Ti - and the RTX 4080 Super has allowed me to do just that.

What makes the jump from the RTX 3080 Ti to the 4080 Super worthwhile?

It's entirely possible to stick with Nvidia's high-end RTX 3000 series GPU at 3440x1440, 5120x1440, or even 4K, but the problem you might run into is the need to adjust your in-game graphics settings. This is especially the case since so many recent PC titles are poorly optimized (I'll never shut up about this), and the highly beneficial DLSS 3 is only available to RTX 4000 series users - with Team Green's Frame Generation software, even if you're faced with poor performance, it can work wonders by literally making more frames for higher frame rates (despite the slightly increased latency this causes).

The frame rate would wildly fluctuate between around 60 and 100 frames per second in the Resident Evil 4 remake at the 3440x1440 resolution with the 3080 Ti when using the 'High (6GB)' texture quality setting with ray tracing enabled. It didn't help that there's actually no DLSS support for the game, and enabling FSR 2 on an RTX 4000 series GPU just doesn't feel right - especially given how much better DLSS is when it comes to preserving image quality.

Testing with the RTX 4080 Super

Whilst those performance results aren't bad, you could find yourself dealing with constant frame rate dips (also depending on your CPU), which can be incredibly frustrating. I also must mention that the frame rate in the Resident Evil 4 remake specifically can fly all over the place as you enter new locations - most of my testing was done during the iconic village onslaught near the beginning of the game, which is where most of the dips occurred (likely due to the numerous enemies, destructible objects, and corpses).

In came the RTX 4080 Super to my aid! Even without access to DLSS, it outperformed the RTX 3080 Ti comfortably, keeping a consistent frame rate above 99fps at 5120x1440 using the same 'High (6GB)' texture quality and ray tracing enabled. I didn't manage to test the RTX 3080 Ti at the 5120x1440 ultrawide resolution, but the results at 3440x1440 should give you a strong indication of how it may struggle with the PG49WCD monitor's 5120x1440 panel.

Besides the remake, I gave the 4080 Super a run for its money, testing Cyberpunk 2077 at maximum settings using DLSS 3 in 'Quality' mode, and this easily sat above 100fps - setting the 'Overdrive' ray tracing mode on (a ludicrous setting that even the RTX 4090 struggles with) dropped this down to about 30 to 55fps. When compared to the 3080 Ti's 20 to 30fps, the difference is clear to see when both GPUs are put under heavy load in one of the most demanding games available.

Downsides of the RTX 4080 Super

If there's one gripe I have with the RTX 4080 Super itself (and the other high-end RTX 4000 cards, too), it's the power cable adapter that comes with the GPU. Intent on closing your PC's side panel? If you're using the included adapter, forget about it. Not without squashing the power cable to a ridiculous degree - considering previous stories of RTX 4090s burning due to issues of power connectors not fully inserted, I was never going to risk flexing a stiff cable adapter just to secure my case's side panel, ultimately putting my system at risk.

Purchasing a Seasonic 12VHPWR power cable served as a better solution (if of course, you have a compatible Seasonic PSU), providing a far more convenient case setup for me.

Besides this, the RTX 4080 Super impressed me by taking games that would chug along on my previous setup and in many cases literally brute-forcing its way through to fantastic performance while still allowing them to look their best.

How did the Asus PG49WCD OLED fare?

As most OLEDs do, the Asus ROG Swift PG49WCD super ultrawide does an incredible job at providing great deep black levels thanks to the infinite contrast ratio - this paired with a 32:9 aspect ratio at the 5120x1440 display resolution, is the stuff of dreams.

My infatuation with the Resident Evil 4 remake will never die, and while this monitor's resolution pushes the perspective far too close to Leon, a mod known as REFramework from 'Praydog' on GitHub fixes this, adjusting the FOV appropriately.

Like I've said in previous articles, a 32:9 display is fantastic for creating a more immersive experience, and fortunately the PG49WCD (like many other OLED monitors) does a decent job at it. Similar to the move from the RTX 3080 Ti to the RTX 4080 Super, the jump from the Alienware QD-OLED ultrawide to Asus' 49-inch screen was well worth my time.

It single-handedly pushed my immersion up a notch (where I initially hadn't expected it to) - while the best OLED display I've personally used is still the one on the Lenovo Yoga Slim 7x laptop with its True Black HDR 600 VESA certification, this didn't take anything away from my experience with the PG49WCD despite it only being True Black 400.

The one aspect that did leave me with minor annoyances as it does with most OLED monitors, is the OLED care pixel refresh - while this isn't necessarily an issue as it's there to prevent potential burn-in, it becomes irritating using this tool after a while. What doesn't help here either, is the small LED indicator below the monitor, which doesn't get bright enough to tell you whether a pixel refresh is taking place (before realizing refreshes happen automatically, fortunately).

This is quite different with the AW3423DWF QD-OLED, as there is a bright green LED light on the power button to indicate a refresh is occurring - but on a positive note for the PG49WCD, at least the reminder doesn't appear right in the middle of your screen, disrupting gameplay.

Back to the case of immersion, I recently purchased the Meta Quest 3S VR headset, and I can say that this widened perspective comes close enough (with the bonus of an OLED panel). There are small cases of a fisheye effect in some games (specifically Call of Duty: Black Ops 6 for me), but it isn't very noticeable.

Is the 32:9 aspect ratio too much for gaming?

Now, considering the pricing of Asus' PG49WCD monitor ($1,199 / £1,399 / around AU$2,699), it's definitely not an affordable option for most PC gamers - it's worth noting that a large number of games do not have official support for the 32:9 aspect ratio, so I've already had to deal with relying on modders to find a solution where possible. There is also indeed an aspect of having too much screen space, particularly for gaming.

This isn't me criticizing the monitor, but rather letting you know that it isn't an absolute must-buy over a 3440x1440 display - it's certainly worth it don't get me wrong, but if you're not looking to drain your wallet for the monitor and Nvidia's high-end GPU to power through it, you're better off aiming for the likes of the aforementioned AW3423DWF QD-OLED.

You might also like...

• I've never taken Intel's GPU competition seriously, but the Arc B580 has left me no choice - it just surpassed the RTX 4060 and RX 7600 in Vulkan benchmarks
• Intel confirms Arc desktop GPU launch on December 3, while Battlemage B570 leak hints that it could be a dream budget graphics card
• Desperate to see Nvidia’s RTX 5090? A graphics card maker rep just accidentally revealed that next-gen flagship GPU will be unveiled at CES 2025

If you’re in the online competitive shooter scene and own an AMD Radeon graphics card, you might want to turn off the AMD Anti-Lag+ feature exclusive to the Radeon RX 7000-series graphics card, as the results could be devastating. 

According to AMD, Anti-Lag+ is supposed to work by “applying frame alignment within the game code itself, allowing for a better frame syncing which leads to even lower latency and great gaming experiences.” And since AMD often makes some of the best graphics cards, this feature should have worked as expected.

However, it seems that having it enabled for the best AMD graphic cards will net you an automatic ban for several competitive shooters like Counter-Strike 2, Call of Duty, or Apex Legends. This is particularly nasty for CS2 since those bans are much more difficult to reverse, and even the developers warned players to have the feature turned off when playing.

TechRadar reached out to AMD for comment and will update this story once we’ve received a statement.

Why is this happening? 

It seems that this occurs due to the way the feature “detours engine DLL functions”, which resembles a cheat program. Basically, this means that the Anti-Lag+ feature reads the information between the game engine and the display adapter, then goes a step further and replaces/injects functions. It’s essentially like wiretapping a phone call, and then using that info to create things like an aimbot while bypassing the game's executable. 

Not only is it resulting in bans, but it’s also causing significant performance issues in other PC titles like the aforementioned ones as well as Fortnite and Overwatch 2. Luckily, you can disable the feature by ALT + L. You can also verify the hotkey assignment in AMD Radeon Software by clicking the gear icon > Hotkeys > customizing the hotkey for Anti-Lag Plus, according to Engadget.

But as it’s only available in some games through the Adrenaline Edition version 23.30.1 driver, you don’t need to worry about it affecting every game you play.

AMD is struggling 

AMD has been releasing new tools that work with Anti-Lag+, like the ability for its Adrenalin graphics drivers to show you the system lag holding back your PC. It’s meant to compete with Nvidia’s PC Latency measurement. 

However, not only do we have the aforementioned issues with the anti-lag tech, but there’s also the fact that all this tech in general doesn’t even work with the vast majority of the best PC games (only 12 support it so far). This means that Team Red still has a long way to go if it ever hopes to catch up with Nvidia in this regard.

You might also like

• Nvidia's new ultra-low latency mode for GPUs beats AMD's Radeon
• AMD's new drivers boost graphics quality with Radeon RX 5700
• AMD Radeon graphics cards could soon get a much-wanted feature

BenQ Zowie XL2566K: One-minute review

• 360Hz refresh rate
• Incredible pixel response time
• DyAc+ turns off backlight between frames
• Sharpest image for pro FPS players
• Good stand ergonomics
• Not cheap
• Poor for fun and immersive gaming

Sometimes, the old ways are the best and that might be said of BenQ’s Zowie XL2566K gaming monitor. Well, we call it a gaming monitor but it’s really been designed to play one, specific, genre of game: FPS. What’s the big deal? It uses the hideous TN LCD panel technology which blighted the fledgling, consumer LCD monitor industry some 20-years ago. It offers poor colour reproduction, poor contrast, a relatively low Full-HD resolution, horrendous viewing angles and no casual gamer should want to go near it. However, there’s just one more thing... it’s really, really, REALLY fast.

Many potential buyers will be star-struck by the headline ‘360Hz’ refresh rate but there are serious caveats to having a fast display. The ability to operate at a consistent 360 frames per second is only really feasible for casual and competitive games. Cinematic, UHD games, with HDR lighting, require too much processor power to get near that. However, while a fast refresh rate can truly make fast-and-frantic motion in games look buttery smooth, it counts for nothing if the screen’s pixel response time is poor – rapidly displayed objects would still look blurry and horrible.

Indeed, there are several 360Hz screens on the market but most use modern IPS technology. Some even claim to have super-speedy, grey-to-grey (GTG) pixel response times but, in this realm, it’s a claim that’s rendered to meaningless marketing speak. BenQ doesn’t even bother providing a GTG pixel response time.

The quick explanation is that grey-to-grey is almost an entirely arbitrary measure of pixels’ ability to turn from partially on, to fully on, back to partially on again. The industry’s varying definitions of ‘partially on’ (i.e. ‘grey’) are so broad as to be meaningless – especially when measuring a cutting-edge display like this. In this regard, old TN panel technology has more raw speed than IPS.

This preamble is necessary because it’s what the Zowie XL2546K is all about and, even then, we’re not finished. In order for the pixel response time to keep up with super-fast refresh rates, they need to be pre-charged. A cold-start pixel that’s moving from fully off, to fully on and back to fully off, just can’t keep up – it needs to be pre-charged and poised to quickly change and change back again. 

This is what settings like Overdrive adjust – BenQ calls it Advanced Motion Acceleration (AMA). But, if you apply too-little or too-much charge you can get ghosting (smeary ghost images trailing on-screen objects) or inverse ghosting and overshoot (blur that leads on-screen objects and edges that continue to move after an object has stopped), or both. 

Although you need a high frame rate camera to really see it in detail, the Zowie XL2566K’s motion performance in competitive-level FPS games is noticeably sharper than 360Hz IPS screens. For human beings in real-time, the result is having a smidge smoother and sharper imagery where even small-moving objects are rendered that bit clearer. The difference is miniscule, but major world championships have been decided by less. 

Another performance enhancing feature is BenQ’s DyAc+ strobing backlight technology. This turns off the backlight to insert a black frame between each displayed frame. The result is enhanced clarity and motion smoothness. 

The Zowie also features some pro-gamer-friendly ergonomics. There are adjustable site screens on either side of the display to protect your peripheral vision from distractions. The solid, small-footprint stand (which allows an angled keyboard to be placed right up close – just how pro players like it – is highly adjustable and has measured notches for rapid set-up when moving from LAN to LAN. The latter is further facilitated by a carry handle at the top and a PVC carry case (unpadded) to help protect it on its travels.

The bells and whistles are nice to have, but this monitor is defined by its panel speed and almost everything else is irrelevant. If you like your games looking fast, smooth cinematic and immersive, don’t buy this. However, if the BenQ Zowie XL2566K represents a professional tool that will improve your top-level competitiveness, its compromises (and very high price) will be irrelevant.

BenQ Zowie XL2566K: Price and release date

The BenQ Zowie XL2566K is available worldwide for $650 / £599 / AU$1,099. However, in some markets, like the UK, it’s much harder to find than its predecessor, the (£439) BenQ Zowie XL2546K – which is the same but for a 240Hz refresh rate.

These prices are all very high compared to regular 24-inch gaming monitors. However, the premium reflects the Zowie’s skunkworks nature with its high development costs and super-niche market size. Other 360Hz gaming monitors use IPS technology and still cost more, but they tend to be much larger and have more luxury features that suit gaming and multimedia immersion.

• Value score: 4/5

BenQ Zowie XL2566K: Design & features

• Solid, small-footprint, adjustable stand
• Remote control
• Limited connectivity
• Carry handle

The BenQ Zowie XL2566K’s stand has a small-footprint that enables pro gamers to get ridiculously close to the screen. It also has a high degree of adjustability that includes -5o/+23o Tilt, -45o/+45o Swivel and 15.5cm vertical travel – all of which are marked out with notches for rapid set-up. At the top, there’s a headset hook, carry handle, and an unpadded PVC carry cover is included. There are also, removable ‘Shielding Hood’ blinkers at the sides to prevent background distractions.

Connectivity ports are minimal: there are two, 60Hz-limited HDMI 2.0 ports (so console players needn’t apply), DisplayPort 1.4 and a 3.5mm audio jack.

All settings are quickly and intuitively adjusted via a rear-mounted joystick-button and BenQ’s wired S-Remote. They can also be easily saved and activated as pre-sets via BenQ’s 'XL Setting-to-Share' Desktop app which enables teammates (and fans) to import them. 

• Design and features score: 4/5

BenQ Zowie XL2566K: Performance

• Unrivalled speed
• Unrivalled clarity
• Strobing backlight for additional smoothness
• Poor multimedia performance

The superlative image quality afforded by the marriage of the 360Hz refresh rate, the TN-panel’s super-fast pixel-response time and the strobing DyAc+ backlight (which turns off in between frames to insert a black screen to enhance motion smoothness and clarity) is best appreciated when viewed with a very high frame rate camera. Even then, you’ll need a rival, mega-fast gaming monitor for a side-by-side comparison to notice much difference. But, a difference there is!

At real world speeds, the image is faster, sharper and more clear than anything else on the market. It might not be by much, but, and we’ll labour the point, major championships can be decided by less. If you play FPS games, you need the fastest, clearest screen to perform to your potential and this is it. 

While it’s not great for multimedia immersion, the Windows Desktop remains sharp and clear for work and colors retain acceptable vibrancy. Brightness and contrast are mediocre while viewing angles are poor. But, the target market doesn’t want to be blinded by a backlight and they don’t want dark or bright areas hiding enemies. Most users will embrace the ‘Black Equalizer’ setting to destroy contrast, water-down dark areas and ensure there’s nothing hiding in the shadows.

Some settings can be adjusted to suit individual preferences and this includes the AMA (pixel overdrive) setting. Indeed, the XL2566K comes with presets labelled Premium, High and Off, along with a 30-step customizable slider. Premium achieves a fast response time but inverse ghosting is noticeable. High offers a better blend and will be used by many players. However, it’s so easy to adjust the overdrive settings and see instant results (there’s no change-related lag) that tweaking everything is recommended. We found settings between 11-18 represented a sweet spot where response time and inverse ghosting coexisted well, before causing issues.

BenQ’s DyAc+ strobing backlight is also adjustable. Previously, turning the backlight off like this significantly impacted brightness, but here the difference between Off, High and Premium settings is barely perceptible. You can even adjust the brightness while it’s activated, which is unusual. The effect on clarity is extraordinary and it brings even a perfectly overdriven 360Hz TestUfo alien into sharp relief.

It all amounts to the XL2566K being the best monitor for seeing even the smallest, most-rapidly moving objects in relief that’s sharper than on any rival’s screen. The difference may be tiny... but tournaments have been decided by less.

• Performance score: 5/5

Should I buy the BenQ Zowie XL2566K gaming monitor?

Buy it if...

Don't buy if...

Also consider...

As any competitive gamer will know - especially those who value fast reactions and a zero-latency gameplay experience for titles like Valorant, CS:GO, or Apex Legends - having a cutting-edge SSD can virtually eliminate any risk of your SSD slowing you down in the heat of battle.

Getting that competitive edge can be the difference between a victory and a loss, so it’s wise to make sure your hardware is up to the task. If you’ve ever played a game like Rainbow Six Siege, you’ve probably seen people complaining about that one player with slow storage holding up the match as their PC struggles to load between rounds. Don’t be that player.

If you’re a serious PC gamer, chances are that you’ve built your own system at some point - or at least upgraded one you bought previously. It’s the great strength of PC gaming; no need to buy a new console every few years; just swap out some components for next-gen upgrades every now and then, and you’re good to go.

Every PC gamer knows that your GPU is the most important part of your rig, but that doesn’t mean you should overlook the other aspects of your build. Plenty of gamers are out there with RGB-laden systems packed with powerful graphics cards and plenty of RAM, but still rocking outdated storage.

This brings us to the Crucial T700 Gen5 NVMe SSD. The 5th generation of PCIe SSDs is finally here, and it’s turning the dial to 11 - 11.8GB/s, to be exact, since that’s how fast it can write data, with read speeds even faster at 12.4GB/s. If you’re looking for an SSD for your new PC build - or perhaps you’re just upgrading your system - this is the one to pick.

The power of PCIe Gen5

Those are some big numbers when it comes to transfer speeds, and we could wax lyrical about the RDN4K read speeds or the Micron 232-layer TLC NAND that provides the foundation for Crucial T700 SSDs, but we’re not going to do that here.

We’d rather focus on the practical applications, specifically for gamers. When you’re playing PC games - especially online multiplayer games - your SSD is in near-constant use, feeding data from the game’s files stored on it so that you can enjoy whatever you’re playing.

That means the quality and speed of your SSD affect a whole lot of stuff: load times are often cited as the biggest factor, with a faster drive meaning less texture pop-in and less time waiting in front of a loading screen, but a high-speed SSD also helps to cut down latency - helping you ensure your split-second shots in an intense match don’t have any input delay.

Upgrades, people, upgrades

Picking a cutting-edge SSD like the T700 helps to future-proof your build, too, since the PCIe Gen5 storage standard is the new kid on the block right now. You’re not just getting turbo-charged speeds here; you get reliability, longevity, and Micron’s decades of experience in the SSD industry.

If you want to upgrade your storage but can’t afford to buy a whole new motherboard (and potentially a new processor and RAM to accompany it) at the same time as a new SSD, then have no fear: the T700 is backward-compatible with PCIe 3.0 and 4.0 boards, so you can get the best out of your existing motherboard and then upgrade freely when you’re able to.

The bottom line here is that this is the SSD to buy if you want the very best for your gaming PC. The T700 is almost twice as fast as Crucial’s own Gen4 drive, and more than 22 times the speed of a conventional SATA SSD. So if you never want to be the person holding up your team between rounds - get a T700.

Samsung announced that it would begin ramping up production of its 12nm DDR5 RAM, which uses a 12-nanometer (nm) class process node. This mass production of said RAM — coming when the overall market for memory chips has seen a decline — is proof of the manufacturer’s intention to remain an industry leader.

Samsung stated that the new chips’ power consumption, when compared to the previous generation, has been reduced by 23% meaning that server and data center operators are able to reduce their energy consumption and carbon footprint. The new chip also has a maximum speed of 7.2Gbps, giving it a processing speed of 60GB per second. And though these chips are mainly targeted for data centers, AI, and more advanced computing applications, hardcore gamers are sure to benefit from the increased specs.

This follows the news of Samsung’s development of the 14nm based 16GB DDR5 RAM that was announced in November 2021 and was validated for use on Qualcomm Technologies’ Snapdragon mobile platforms in March 2022. 

ZDNet reported that the new RAM “has been verified for compatibility with AMD in December [2022] already” and that Samsung is currently working with more global IT companies.

This could be a game-changer for gamers 

Samsung announcing mass production of the 12nm DDR5 RAM is great news for plenty of industries, including gaming. Not only because it’ll make the tech more readily available but for cheaper too. One of the ways this is supported is that the new chip has 20% more wafer productivity, meaning that more chips can be produced out of a single wafer due to the smaller size compared to the previous generation.

Back in 2022, it was already reported that pricing per GB had been decreasing from 2021. According to statistics from ComputerBase, the price of memory in 2022 was €5 per gigabyte (around $5 / £4 / AU$7.50), a massive decrease from the €15 per gigabyte (around $16 / £13 / AU$22) average at the end of 2021.

Now with this Samsung news, we’ll not only be seeing more of the advanced 12nm-based RAM but an overall decrease in pricing for DDR5 RAM due to a much higher supply. And since the tech giant has confirmed that its new memory is compatible with AMD products, expect to see the new RAM in both its graphics cards and processors.

Of course, compatibility is another issue, as DDR5, in general, requires an AMD AM5 or Intel LGA 1700 socket motherboard to support it. While AMD’s Zen 4 chips already require it, the problem is a lack of backward compatibility with DDR4, but one that isn't as big a deal as it was last year. That will just have to be smoothed out by time and advancement, as more companies adapt to the DDR5 format.

Two-minute review

Asus ROG Rapture GT-AXE16000: Two-minute review

Asus is in a mad race with itself to produce the fastest, most overpowered, feature-laden wireless router. It must already hold at least nine of the top 10 positions for super-premium gaming routers, yet the new ROG Rapture GT-AXE16000 is the latest standard bearer. 

It’s big, brash, spiky and my wife hated it. The RGB top (and its seven preset patterns) illuminates half the house at night and there’s no timer to turn it off. Setup is simple: just scan a QR code on the device's bottom with Asus’ phone app, set the network name and password, and voila. 

The phone and browser apps are well designed. They’re responsive, show connection statuses, alert you to issues (and suggest solutions), provide extensive QoS options, include server-grade apps, enable mesh network creation (from other Asus routers) and offer a plethora of premium consumer-router level features. There are extensive, easy-to-setup parental controls that bind children and devices together and set either online or offline schedules for weekdays and weekends. There’s also (adjustable) content blocking with presets for different age groups and, unlike some rivals, you don’t need to pay a premium for the feature. Nor must you pay for the built-in security, which comes via Asus’ Trend Micro-powered AiProtection feature. 

In addition to four Gigabit Ethernet ports (two can be aggregated) and USB 2.0 and USB 3.0 ports, there’s a 2.5GbE WAN port (which can be aggregated with one 1GbE LAN port) and two (count ’em) 10GbE ports, one of which can also be used for WAN. Inside there’s a 2GHz quad-core CPU that keeps everything ticking under load.

The GT-AXE16000 also supports quad-band Wi-Fi with a crazy 16,000Mbps theoretical throughput. That’s spread across a 2.4GHz band, two 5GHz bands and the new 6GHz band. Despite the power and the simple design, we did have issues testing it. Neither our new Wi-Fi 6E sporting Dell XPS 15 nor our new Asus Zenbook recognized the 6GHz band, whether it was bound together with other bands or separated. Only our Asus Zenfone did! That’s despite driver updates and location hacks (not every global region allows the same 6GHz channels to be used). 

Still, it flew in my tests. Up close I downloaded large movie files from our NAS at 825Mbps. Two rooms away, in my weatherboard cottage, it only dropped to 520Mbps, which rivals the best two-node, mega mesh systems like the Asus ZenWiFi Pro XT12. At the bottom of the garden (25m away) it managed 101.5Mbps. While the latter is a little slow compared to rivals, it’s worth remembering that once the ‘new’ Wi-Fi 6E connection technology settles down, this router has a great deal more headroom to go faster.

Other special gaming features include shortest-route-to-game-server optimization, game packet priority (even when a VPN is running) and a Mobile Game Mode that ensures phone-based games get top priority. 

Ultimately, it’s the highest-spec consumer router available at the time of testing (March 2023) – but that also makes it extreme overkill for most users. Still, we have to admit, nothing else carries the aspirational cachet of this monster from Asus.

Asus ROG Rapture GT-AXE16000 review: price and availability

The Asus ROG Rapture GT-AXE16000 router is available now in most major global markets and is priced at $700 / £580 / AU$1,299. 

While it’s incredibly expensive, it does arguably offer the performance, features and new technology to justify the price tag... as long as you actually need it.

For most buyers, Synology’s dramatically cheaper consumer and SMB routers – the WRX560 and the RT6600ax (respectively) – represent a much better choice in terms of value for money. With the latter, you can even buy multiple units to build an all-powerful mesh network and still save a fortune vs the GT-AXE16000.

If you’re desperate for a ROG gaming router, we'd recommend checking out this model's slightly older sibling, the Asus ROG Rapture GT-AX6000. It’s still overkill for most buyers, but it shares many of the firmware features of the GT-AXE16000 and costs half the price. All you’re missing out on is the next-gen Wi-Fi 6E connectivity.

Asus ROG Rapture GT-AXE16000 review: key specs

• Wireless connectivity: Wi-Fi 6E (IEEE 802.11ax), quad-band (2.4GHz, 2x 5GHz, 6GHz)
• Processor: Quad-core @ 2GHz
• Memory: 256MB NAND, 2GB RAM
• Ports: 2.5Gb WAN/LAN, 4x Gigabit Ethernet LAN; 2x 10GbE; USB-A 3.2 Gen 1, USB-A 2.0
• Dimensions (HxWxD): 353 x 353 x 193 mm

Asus ROG Rapture GT-AXE16000 review: design and features

• Imposing design
• Best connectivity options on the market
• Dripping with features

You can’t get away from the fact that the Asus ROG Rapture GT-AXE16000 is physically huge. You’ll need some serious desk space to house it. But, once positioned and turned on, you can bask in the infinity-mirrored lid and RGB goodness that sits within it. 

Setup is as simple as we’d expect nowadays; you just scan a QR code with the phone app and step through a short process to choose a network name and password. 

Once activated, you can access the extensive settings via the phone app or web browser. The phone app is excellent and well-polished. It displays what’s on your network, how everything is performing and gives access, priority and bandwidth control over everything on it. It offers insights into security issues and service improvements. It also provides excellent Parental Controls which assign comprehensive, timed access settings for different users and/or devices. There are even content-filtering options (these are always fallible) with different (customizable) presets for different age groups.

You can set up most other features via the app – and there are many of them – but several are better served by going through the web browser.

All the stalwart consumer-configurables are there, and everything is rendered responsive and rapid thanks to the speedy 2GHz quad-core processor. 

There's frankly too many key features to delve into in the space of a review, but a non-exhaustive list of others includes: AiMesh which allows you to network multiple Asus routers together for additional coverage; configurable Edge protection from Trend Micro which protects everything on your network, blocks malicious sites and isolates infected devices; extensive VPN controls; extensive traffic analysis and logging tools; RGB controls; wireless channel combining and separation; guest network controls; extensive WAN and LAN routing controls; extensive USB device options including 4G WAN, print server, backup, BitTorrent and general server options; cloud syncing; voice control via Alexa and Google Assistant and extensive firewall settings.

Where the ROG Rapture routers really stand apart is with the game optimization settings. You can prioritize gaming traffic in many ways. The first set of speed-ups operates across three levels. Level One prioritizes traffic to particular devices. Level Two uses QoS to prioritize gaming specific traffic. It can also automatically recognize other ROG products on your network (motherboards, laptops, desktops etc) and automatically prioritize traffic to and from them. Level Three integrates the third-party WTFast VPN service to help reduce pings.

The Game Radar feature measures the speed of different games’ different servers so you can quickly see what has the healthiest ping. Meanwhile, Mobile Game Mode on the phone app prioritizes all traffic to your handheld device.

Beyond all the software smarts, it’s in wireless connectivity that GT-AXE16000 really shines. It supports quad-band Wi-Fi with a 16,000Mbps theoretical throughput. That’s spread across a 2.4GHz band, two 5GHz bands and the newer 6GHz band. My only problem was that I struggled to use the latter, as hardly any laptops recognized the 6GHz network – even though they had compatible Wi-Fi 6E adapters and were set to the same region. Only my Asus ROG phone recognized it. Still, I'd expect this to become less of an issue as the new technology settles. Right now, it either provides a speed boost or headroom for the future.

The physical network ports will tempt any router enthusiast on their own. There are four Gigabit Ethernet ports, two of which can be aggregated. There are USB 2.0 and USB 3.2 Gen 1 ports which can make use of Asus’ extensive and advanced USB features. The WAN port supports 2.5GbE connections and can be aggregated with one 1GbE LAN port. There are also two 10GbE ports and one of them can be used as a second WAN.

You could run a competitive LAN off the Asus ROG Rapture GT-AXE16000 router.

• Design and feature score: 5/5

Asus ROG Rapture GT-AXE16000 review: performance

• Extremely fast
• Headroom for growth
• Not the best long distance

With great throughput comes great processing requirements, so Asus includes a 2GHz quad-core CPU, 256MB of NAND memory and 2GB of DDR4 RAM. It’s bigger and more powerful than some cheap laptops.

I tested it with a series of updated Wi-Fi tests. These saw me download large video files to my upgraded, Wi-Fi 6E-sporting Dell XPS 15 laptop from a Synology NAS that was physically connected to the router. The latter was positioned in the middle of my single-storey weatherboard house.

I combined all four bands into one SSID and let the router pick what was best. However, it quickly became clear that the 6GHz band was not being used – even when separated into its own discrete network. I tried with several Wi-Fi 6E laptops, but none of them would see it, no matter how I tweaked the drivers and location settings. I also wasn’t in a position to test the 10GbE connections either, so straight off the bat the Asus ROG Rapture GT-AXE16000 is so fast that I need to update my updated tests before I've even used them.

Nonetheless, with last-gen 5GHz connectivity tech, it scored 825Mbps for transfer speeds up close, which is up there with the fastest we’ve seen. Two rooms away, it managed 519.3Mbps which rivals competitor ‘mega mesh’ systems that have a second node in the room that’s connected by a dedicated, wireless, backhaul channel.

Finally, 25 meters away, at the bottom of my garden, it managed 101.5Mbps. That’s a little slow compared to the best performers and mesh systems, but certainly better than your average router. It’s also worth remembering that other cheap Asus routers can be used to extend the GT-AXE16000's network using the company's AiMesh technology. But you don’t really want to be considering buying an upgrade after investing in a beast like this.

• Performance score: 5/5

Should I buy the Asus ROG Rapture GT-AXE16000?

Buy it if...

Don't buy it if...

Also consider

• [First reviewed March 2022]

Well, it’s confirmed: Asus really wasn’t joking when it said it was making a handheld gaming PC on April Fools Day. The Asus ROG Ally is an AMD-powered handheld that looks set to take on Valve’s Steam Deck - with some seriously impressive specs.

Asus claims that the AMD APU powering the ROG Ally is the fastest chip yet, and packs support for connecting to Asus’ own XG Mobile external GPU - letting you hook up an RTX 4090 for potentially unbelievable performance.

The APU is a custom-designed 4nm Zen 4 chip with an RDNA 3 iGPU, which is practically guaranteed to deliver superior performance to the Zen 2+/RDNA 2 APU found in the Steam Deck. This will likely make the XG Mobile a bit redundant; why you’d need to connect a crazy-powerful $1,999 GPU, I can’t fathom. The ROG Ally will use the same ‘ROG Intelligent Cooling’ found in Asus’s Republic of Gamers laptops.

The rest of the specs look set to blow the Steam Deck away, too: the Ally shares the same 7-inch screen size, but it packs a higher 1080p resolution over the Deck’s 1280 x 800 and doubles the refresh rate at 120Hz. Asus also claims the maximum brightness is 25% better than the Steam Deck’s. We also know (from the Canadian Asus website, which strangely is the only region with any info on the Ally) that it’ll run Windows 11.

We don’t know anything about how the battery life will measure up yet, but we do know that the ROG Ally will be a bit smaller and lighter than the Steam Deck, weighing in at 1.3 pounds (about 0.59kg). It doesn’t have the Deck’s fancy touchpad controls, instead opting for a layout that almost perfectly matches the Nintendo Switch.

The Steam Deck is great, but it’s time for something new

I’m massively happy to see this. Why? Because we’ve seen a whole bunch of Steam Deck imitators so far - like the solid AyaNeo 2 - but this is the first product I’ve looked at and thought “that’s not an imitator - that’s a competitor”.

In fact, the ROG Ally could be more than just that: it could be the Steam Deck killer. Just off the specs that have been revealed so far (via a pair of videos from two YouTubers who were given exclusive early access), it should absolutely wreck the Deck in raw performance and graphical fidelity.

I’m amped - for two reasons. First of all, because I’m (hopefully) gonna be reviewing this bad boy, and I just can’t wait to get my hands on it. But secondly, because this will hopefully set a new bar for the performance of gaming handhelds - the Steam Deck is now more than a year old, while the hardware powering the Nintendo Switch is even older.

It’s time for new hardware! And AMD is proving that we don’t need expensive discrete graphics cards: APUs could really be the future of gaming, and this is the proving ground for that technology. In short, I’m hoping that Valve and Nintendo are taking notes.

As we recently reported, Nvidia’s budget desktop GPUs are expected to land in the next couple of months, with the RTX 4060 rumored to be launching in May – and the RTX 4060 Ti and 4050 supposedly dropping around the same May-June window.

What we didn’t know at the time was that freshly leaked specs for the RTX 4060 and 4060 Ti had also surfaced, with VideoCardz reporting on that – as well as further corroborating the suggested May release date.

Those leaked specs have me pretty worried, though. My colleague John Loeffler already raised concerns about the RTX 4060 back in February, and I’m here to double down on those claims: the most recent rumors have all but confirmed that both the 4060 and its big brother the 4060 Ti will only have 8GB of GDDR6 (not the GDDR6X found in the currently available RTX 4000 cards) with a 128-bit bus. That’s… not good.

Compared to the RTX 3060 Ti, that’s the same amount of video memory with literally half the bus width, and both of these cards have fewer shader cores too. Sure, it’s built on newer, faster architecture, so the baseline performance will be better, but the VRAM situation is concerning.

Analysis: 8GB is rapidly becoming insufficient for triple-A PC games

See, while 8GB of on-card video memory is fine in most instances, we’re seeing more and more games start to really guzzle that VRAM. As our good friends over at PCGamer pointed out, the recent PC port of The Last of Us: Part 1 glugs down more than 8GB of memory – and that’s just at 1080p High. Crank up the resolution or graphical quality and you’re going to be looking at much worse.

Sure, TLOU’s PC port has been notoriously poorly optimized, but the reality here is that it’s a phenomenally popular game that you’d struggle to run well at 1080p without 12GB of VRAM. Plenty of recent games can chew up more than 8GB, especially at 1440p and 4K.

Although we don’t have pricing for Nvidia’s next-gen budget GPUs yet, I’d expect the RTX 4060 Ti to land somewhere in the $300-500 range – a price comparable to AMD’s RX 6700 XT, which notably does have 12GB of VRAM. I’m basing that price tag assumption on the RTX 3060 Ti’s list price, and the newly-leaked pricing for the RTX 4070, incidentally.

This could be a make-or-break moment for Nvidia

The RTX 4060 (and, to a lesser extent, the 4060 Ti) are the GPUs I’ve personally been waiting for. Yes, I know the RTX 4090’s performance is earth-shatteringly fast, but it’s the RTX 4070 Ti that sits at the top of our best graphics cards list – and that’s because it’s better value for money.

The fact is that most gamers (and plenty of professional content creators) simply can’t afford to drop over $1,000 on a new GPU. I’m more interested in the best cheap graphics cards - I used a trusty old Nvidia GTX 1060 for years, and it served me well. If we take a look at the Steam Hardware Survey, we can see that Nvidia’s xx60 cards account for four of the top five most-used GPUs among PC gamers.

It’s clear that these more affordable graphics cards are the ones gamers actually want (or at least, can reasonably afford to buy), so there’s a lot riding on the RTX 4060. I’m seriously worried that it’ll underperform – and Nvidia really needs at least one of its budget RTX 4000 lineup to offer excellent bang for your buck.

I hope this latest round of rumors is incorrect; or at least that Nvidia can make that 8GB of GDDR6 stretch farther than expected. The RTX 4060 needs to follow the 3060 and 2060 as the most-used desktop card of this new generation. Prove the naysayers wrong, Nvidia!

Update: An earlier version of this article didn't make it clear that the 5800 XT doesn't actually exist. This has been clarified.

Last week the PC specifications for The Last of Us (or TLOU if you’re cool) PC port had been released by Naughty Dog and Sony. If those minimum requirements gave you a fright, then there's some good news, as they have been updated and are not quite as demanding. Interestingly, the Iron Galaxy logo has also been slapped onto the new specification sheet which suggests it is handling the port.

The Last of Us is a game dear to many peoples' hearts, and as well as being a critically-acclaimed game, a HBO TV show based on the game has been extremely well received as well, so hype for the PC port is higher than ever.

As well as the addition of the Iron Galaxy logo, we can see a few minor changes to the recommended requirements for gamers looking to play the game soon. 

The good news is the updated specifications are actually lower than before, which means more PC gamers are will be able to play the game without having to shell out for equipment upgrades. For example, in the graphic below you can see that if your performance goal is 60 fps at 1080p, rather than the AMD Radeon RX 5800 XT (8GB), players with a seemingly less powerful AMD Radeon RX 5700 XT (8GB) will be able to meet that specification.

As a lot of people have pointed out, despite there being an AMD Radeon RX 5700 XT (8GB) and even an AMD Radeon RX 6800 XT, AMD never ended up releasing a AMD Radeon RX 5800 XT. The reference to the GPU in the initial system requirements appears to have been an unfortunate typo.

We've got high hopes that the game will run smoothly and be just as much fun for PC gamers as it was for console players since Iron Galaxy did a pretty good job with the Uncharted: Legacy of Thieves Collection port. Not a sensational port by any means, but definitely solid enough that the gaming experience wasn't hindered.

Great news for everyone!

As we gear up for the PC version of The Last of Us to release on March 28, we’re pretty pleased to see the adjustments that have been made to allow gamers with older equipment to still enjoy the game at a reasonable capacity. While the reference to the non-existent AMD Radeon RX 5800 XT was an error, we're glad that it seems like it was meant to say the more affordable RX 5700 XT, rather than the more expensive AMD Radeon RX 6800 XT. When it comes to PC specs, a stray number can change the context quite drastically.

With the popularity of the show and the subsequent resurgence of the fandom, loads of newcomers to TLOU will be pleased with the improved specification list. 

Hopefully, this is a trend we will see more in the future and have those of us stuck on older, more familiar equipment kept in mind when porting games over from console to PC. There’s no drastic change between hardware specifications in this case, but enough to hopefully warrant a positive reaction from the community. 

In news that will probably shock exactly nobody with their finger on the pulse of the PC hardware industry, the average price of a new graphics card has effectively doubled since 2020.

Thanks to some analysis of sales figures from German retailer Mindfactory conducted by Digital Trends, we’ve seen that prices for current-generation Nvidia and AMD GPUs have risen since launch and remained high, with frequent stock issues. Amusingly, the only exception is the sub-par RTX 4080 - which Nvidia is allegedly desperate to sell off.

While the stock situation for last-gen RTX 3000 and RX 6000 graphics cards is largely fine, prices aren’t dropping much - barring some reductions by AMD last year, most GPUs are holding steady when it comes to cost.

Taking a look at the pricing data, the average sale price across all AMD and Nvidia GPUs in February 2020 was €361 (around $385/£318/AU$562), while three years later in February 2023, the average has risen to a whopping €713 (around $760/£627/AU$1,111) - almost double the 2020 figures.

Who is to blame?

Of course, it’s worth noting that these sales stats are only from a single retailer in Germany and as such don’t necessarily paint the full picture of GPU sales, but any PC builder will happily (or unhappily) confirm that graphics card prices have just been going bonkers over the past few years.

There are naturally a ton of reasons behind this steep increase in GPU pricing, so I can’t really pin the blame on any one company or faction. The COVID-19 pandemic and global chip shortage led to a bevy of production and development issues for GPU manufacturers, while general inflation will have also seen companies raising prices in an effort to offset rising costs.

The cryptocurrency boom was also a huge contributor, with vast numbers of cards bought up for use in massive crypto mining rigs. This had a particularly significant impact on the previous generation of GPUs from AMD and Nvidia through to 2022 when the crypto market crash saw miners desperately selling off their used cards in an effort to recoup their losses - much to the amusement of our own John Loeffler.

If there is one single company that bears a significant chunk of responsibility here, I regretfully say that it has to be Nvidia. I mean, the RTX 4090 might be the best graphics card ever made, but it costs $1,499 / £1,679 / AU$2,959! You’ll be extremely lucky to find one for that price right now, too - a quick scan through Newegg.com showed that the cheapest one available is selling for a hefty $1,769.99.

Nvidia has been insistent that ‘Moore’s Law is dead’, a common claim among computing experts these days. Moore’s Law, for those not in the know, is the statement that the number of transistors that can be put into an integrated circuit doubles about every two years. Following the launch of RTX 4000 series, Nvidia CEO Jensen Huang declared that falling generational GPU prices are ‘a thing of the past’, which prompted Intel and AMD to declare themselves the guardians of Moore’s Law in the GPU industry.

Things are going to get worse

Depressingly, this is probably a portent of doom for PC gaming. Nvidia seems determined to push its GPUs for enterprise purposes these days - anyone buying an RTX GPU for gaming won’t care about Team Green’s omniverse nonsense - and targeting professional industries rather than gaming consumers leads to inevitable price hikes, especially in a space where Nvidia has very little competition.

The thing is, it looks like gamers aren’t even really the ones buying these high-end new GPUs. A look at the Steam Hardware Survey’s data on GPU usage shows that the most-used graphics cards are the GTX 1650, GTX 1060, and the laptop variant of the RTX 3060 - all older budget offerings from Nvidia. RTX 4090s account for just 0.24% of gamers using Steam.

Does Nvidia learn from this? Leo DiCaprio said it best in Wolf of Wall Street: "Absolutely fucking not." Even the RTX 4070 Ti, currently the cheapest offering in the 4000-series lineup, costs $799 / £799 / AU$1,479 - that’s $200 more than the launch price of the previous-gen RTX 3070 Ti, which we actually criticized in our review at the time for being too expensive.

AMD has been following suit in worrying fashion, with its new flagship Radeon RX 7900 XTX costing $999 / £1049 / AU$1,499 - though that’s still a cool $500 less than Nvidia’s leading card. Price cuts on previous-gen cards are a good sign (though it’s not enough) and AMD likely has some solid budget offerings in store for the RX 7000 line, but gamers are still predominantly using Nvidia hardware - AMD doesn’t even make it into the top ten in the Steam Survey.

So we’re probably looking down the barrel of another generational price hike once RTX 5000 and RDNA 4 GPUs arrive, which will presumably be sometime in 2024 judging by Nvidia and AMD’s usual release cadence. That sucks - maybe not for creative professionals, but definitely for gamers.

Does it actually make me as miserable as the experience of playing Duke Nukem Forever? Perhaps not quite as much; at least Jensen Huang doesn’t feel the need to make dick jokes while telling me I’m going to pay more than a grand for a single PC component. But it’s a bad sign for what’s to come - I never thought I’d say this, but I’m actually pinning my hopes on Intel Arc to save me from GPU hell in 2024.

AMD executives David Wang and Rick Bergman have confirmed that we’ll be seeing a lot more AI in the next generation of graphics cards from the tech giant, which will be built on AMD’s RDNA 4 architecture.

In a recent interview with the Japanese gaming website 4gamer, the AMD execs detailed some of what we can expect from RDNA 4. Naturally, front and center was confirmation that we’ll be seeing the second iteration of Team Red’s AI Accelerator cores (similar to Nvidia’s Tensor cores), which were first introduced in the current-gen RDNA 3 GPUs - such as the excellent Radeon RX 7900 XTX, currently the best AMD graphics card on the market.

Nvidia’s tech is still lightyears ahead of AMD when it comes to AI processes - just look at the RTX 4090 - but these second-gen AI cores should offer a serious step up. Beyond the Accelerator cores, the pair also discussed some other nifty new features, most importantly a new self-contained GPU pipeline that allows for rendering and texture processes to be generated exclusively on the GPU without needing to communicate with the CPU.

This has massive potential to boost the processing speed of RDNA 4 GPUs, since it won’t need to rely on the CPU and system RAM to carry out some of its workloads, effectively cutting out two potential system bottlenecks. According to Wang and Bergman, we can expect a massive 2.2x performance boost over the current RDNA 3 cards.

Analysis: AMD is now fully aboard the AI bandwagon

Perhaps even more interestingly, Wang was keen to discuss the implementation of AI within the gaming space. In the 4gamer interview, he espoused that AI hardware shouldn’t just be used purely for improving the graphical aspect of games, positing that inferences cores on the GPU could actually be used to improve the gameplay experience itself.

This has some potentially frightening implications - imagine playing a shooter where the enemies are actually powered by a deep-learning AI housed in your PC, which genuinely reacts, learns, and adapts to your actions. It does sound just the teensiest bit dystopian to imagine an in-game boss battle against ChatGPT.

Still, on a slightly less terrifying level, it’s interesting to consider the uses AI cores could have for things like NPC pathfinding in games. Every gamer has, at some point, complained about game characters walking too slowly or getting stuck on a doorframe due to wonky pathfinding - Wang believes AI cores have the potential to solve a lot of immersion-breaking issues such as this.

It’s somewhat amusing to see AMD commit so strongly to AI now, since Team Red basically mocked Nvidia last year for relying too heavily on AI tech in its RTX GPUs and DLSS. In its defense, AMD did prove that upscaling tech such as DLSS can be achieved without the use of AI, as seen in the Radeon FidelityFX Super Resolution (FSR) software. It sounds like AMD’s higher-ups have become AI converts, but they have different - and intriguing - ideas about how it should be employed compared to Nvidia’s plans. We can’t blame them for wanting to go in a different direction; Nvidia has been in a lot of hot water recently.

While we don’t know exactly when RDNA 4 will land, AMD’s 2-year cadence when it comes to GPU launches means that we can most likely expect them sometime in late 2024. Bergman assured 4gamer that the new cards will be released in the near future, though ‘near’ might be a stretch here.

Microsoft’s DirectStorage tech has had its first outing on PC courtesy of Forspoken, and according to some new testing, the feature boosts not just loading times considerably (for those with the right hardware), but also in-game frame rates in certain scenarios.

Neowin flagged up a YouTube video in which Compusemble tested Forspoken, observing that: “DirectStorage has a pretty significant effect on frame rate, frame time consistency, and GPU utilization at 1080p, whereas it has virtually no effect at higher resolutions.”

What seems to be happening here, Compusemble elaborates, is that DirectStorage only kicks in like this to bolster frame rates when the game is CPU-limited, meaning the processor is struggling with its workload, whereas the graphics card is ticking along just fine.

That’s the case when playing Forspoken at 1080p (Full HD resolution) for Compusemble, but when the resolution is cranked up to 1440p, the YouTuber becomes GPU-limited instead – and the frame rate boost vanishes into thin air.

The testing was performed with a Sabrent Rocket 4 Plus-G SSD (2TB) in a gaming PC with an RTX 3080 Ti graphics card plus Ryzen 7 7700X CPU. Remember, to get the full benefit of DirectStorage you need an NVMe SSD, and furthermore running Windows 11 gives better results than Windows 10 (the test PC used Windows 11).

Analysis: New GPU decompression tech at work? Not the case, in fact…

We knew that DirectStorage would do more than just speed up loading times, and that it’d make a difference in big open world environments, supercharging the in-game loading of assets on-the-fly. But this is the first tantalizing evidence that the tech really can boost frame rates on PC – at least in certain scenarios, namely where the GPU isn’t breaking a sweat, but the CPU is getting hammered.

So, this could make you think that this is clear evidence that the GPU decompression tech in DirectStorage 1.1 – the latest version that Forspoken uses – is coming into play here. This allows the GPU to handle decompression of game assets (which are compressed for size reasons), and the graphics card can do this far more efficiently than the CPU – helping to take a load off the latter when it’s struggling, and thus improving frame rates in CPU-limited scenarios.

That’s the obvious theory, but as Digital Foundry separately points out, even though Forspoken does utilize DirectStorage 1.1, apparently GPU decompression isn’t actually active here, as there is “no GPU compute usage spike when the game is doing a dedicated load” (when obviously there should be if the card was busy with decompression work).

All in all, this is a bit of an odd one, and as Digital Foundry (and others) have observed, there are quite a few issues and glitches with Forspoken on PC. (Not exactly uncommon with games which are subject to multiple delays, and indeed console ports, it has to be said).

The conclusions we can draw from a single game are, of course, limited, but we can say that if these are the results without GPU decompression being made use of, it’s exciting to imagine what other games that actually employ this tech might do for frame rates (at least away from high resolutions). Although that excitement is somewhat tempered when we think about what the next game to use DirectStorage on PC might be – which, well, isn’t yet clear.

We simply don’t know anything else about what games will be supported on the Windows platform after Forspoken, which is somewhat ominous in terms of suggesting the wait for a second title – with any luck featuring an even better implementation of DS 1.1 – could be a long one. But next year, maybe, DirectStorage could become a compelling reason for gamers to upgrade to Windows 11 if they haven’t already made the leap from Windows 10.

AMD’s RX 6000 graphics cards are looking like compelling purchases right now, with prices falling steadily as seen in a new report for this month, while Nvidia GPUs are stubbornly refusing to drop.

This is going by TechSpot’s GPU price update for January 2023 (the site produces these reports monthly), which shows a whole load of data for graphics card sales, including retail pricing (in the US), and extra nuances such as the best value boards (in terms of raw ‘cost per frame’, meaning price/performance ratio).

The good news is that for last-gen products on shelves, some of AMD’s GPUs have seen some sizeable drops in price. Indeed, the top five RDNA 2 models have dropped in price, with the Radeon RX 6800 XT and 6950 XT leading the charge with 14% and 13% falls in price compared to last month.

The 6900 XT is also down a solid amount with a 7% reduction compared to December, and slightly more modest falls – but still worthwhile given that this is month-to-month – were witnessed with the RX 6800 and 6750 XT, dropping 4% and 3% respectively.

Mid-range offerings like the RX 6650 XT and 6600 XT also fell by 4% and 3% respectively, and the 6500 XT was down by 6%, but elsewhere, prices mainly stuck at the same levels. There was an outlier in the RX 6600 going up by 12%, mind.

The story was very different with Nvidia’s last-gen RTX 3000 GPUs, where the only card to actually fall in price was the GeForce RTX 3060 which was down a healthy 6% – and to be fair, that’s a pretty useful product to see a decent drop.

While the RTX 3070 stayed the same price, the other Ampere GPUs still available at US retailers rose – by a slight 2% for the RTX 3070 Ti, but slightly bigger increases of 4% and 5% were in evidence with the RTX 3050 and RTX 3060 Ti.

What about RTX 3080 and 3090 variants? Well, there’s no stock of these boards left, aside from scattered RTX 3080 10GB models (but TechSpot didn’t have any price to compare with in this case, as these cards were out of stock in the past few months).

As for new products, the RTX 4070 Ti came to market this month, and as TechSpot noted, the latest Lovelace graphics card remained in stock and sold for its MSRP (recommended price), an unusual situation – and one that might indicate a somewhat weak demand. TechSpot does clarify that the RTX 4070 Ti is supposedly selling better than the RTX 4080 – something we’ve seen clear evidence of elsewhere – while also pointing out that this isn’t a particularly major achievement, as that wasn’t (and still isn’t) a particularly popular GPU.

Still, in TechSpot’s price/performance (cost per frame) table based on MSRP pricing, the 4070 Ti sits in second place overall, hot on the heels of the leader: AMD’s RX 7900 XTX. Those are the top performers ahead of the RTX 3070 in third place, and the RX 6800 XT and RX 6800 following that in fourth and fifth positions.

Behind those cards lies the RX 7900 XT in sixth place, but you have to go further down the table to find the RTX 4080 and RTX 4090 in ninth and tenth spots respectively. (Remember, this is based on recommended pricing, not actual current price tags).

Analysis: Some seriously tempting options from Team Red

Both RX 6900 XT and 6950 XT graphics cards look like great buys right now, having come down a large amount over the past nine months or so. Particularly the latter which hit a lowest price of $700 (around £570 / AU$1,000) in January, which represents over a third knocked off the MSRP.

The RX 6800 XT also came down heavily, although due to a price spike in December, it has been a bit cheaper previously. All of these are compelling options for those looking for a powerful graphics card, but not wanting to pay more for the new generation of RDNA 3 or Lovelace.

In the mid-range area for AMD’s RX 6000 series, declines are more modest as noted, but there are still some really tempting options – like the RX 6650 XT at an asking price of $275 (around £220 / AU$395).

It’s disappointing that Nvidia graphics cards aren’t seeing the same kind of falls, but not exactly surprising. When GPU prices first started dropping (back from truly inflated levels of double the MSRP, or even more than that), it’s been a theme that Nvidia’s products have been more stubbornly resistant to reductions. Although in fairness, that’s partly a reflection of their relative popularity, of course.

The brighter spot of news with Team Green is that at least the RTX 3060 and 3060 Ti – key mid-range models – are now almost at MSRP in the US (they’re just $10 over in both cases). Maybe they’ll fall further, but we’re not banking on it, as Nvidia’s Lovelace mid-range does not appear to be anywhere near on the horizon. Indeed, the RTX 4060 isn’t rumored to be turning up until the middle of 2023.

That said, the high-end RTX 3000 stock has all sold out now, so perhaps it won’t be too long before excess RTX 3060 inventory is cleared, too; and if so, maybe the Lovelace successor will turn up a bit sooner than is widely expected.

AMD’s Ryzen 7000X3D processors are set to launch in February, the company told us at CES 2023, but now we have a purported release date for the supercharged chips that’ll be of great interest to gamers – and it’s February 14.

OC3D spotted that this release date was shared via the official web page for the Ryzen 7 7800X3D on the AMD site, but looking at the page now, that information has been removed (not before it was screen grabbed).

So, we can assume this was an accidental leak by a less-than-careful AMD employee, though we do need to be a bit careful about assuming the screenshot is genuine, as ever. All we have is an image here, and as we know, pics can be manipulated.

However, OC3D seemingly took the screen grab itself, as the image source is credited to AMD and not a third-party, so it seems unlikely to be a fabrication.

Come February 14, then, we’ll see not just the Ryzen 7 7800X3D, but also the 7900X3D and 7950X3D. All these chips will use 3D V-Cache to pep up gaming performance considerably, as we saw with the 5800X3D of the Zen 3 family.

Analysis: Say it with flowers, or chocolates… or 3D V-Cache?

So, the launch date will be Valentine’s Day, but will we love AMD’s pricing? That remains to be seen. On the one hand, late last year we witnessed price cuts on the still fresh vanilla Ryzen 7000 CPUs, after sales weren’t great to begin with. However, that was partly a reflection of the total cost of upgrade, taking into account the necessary AM5 motherboard (with no wallet-friendly options) and need for DDR5 RAM (unlike Raptor Lake, you can’t use DDR4, and DDR5 is still fairly pricey).

Now more affordable AM5 motherboards are around (and even cheaper ones shouldn’t be far off), and gamers will likely be keen to get a piece of the Zen 4 3D V-Cache action, we can foresee AMD pushing a bit harder with the pricing here. Time will tell, but at any rate, the Ryzen 9 spins on X3D are obviously going to be expensive, anyway.

Aside from the potential hole they might blow in your wallet, the other niggling worry about the higher-tier Ryzen 9 X3D models is how Windows might deal with their new design. In a nutshell, these CPUs have two CCDs (chiplets), but only one has the 3D V-Cache on top – the other runs with a faster boost speed instead. Now, some games will benefit from more cache, and some will find the higher clocks more of a boon, so any given game needs to be marshaled to favor the appropriate CCD that’s best for it.

Could that introduce issues and performance teething problems? Microsoft and AMD are working closely to ensure this doesn’t happen in Windows gaming, but of course, we’ll only know when we get these processors in for review. Not long now, and we can’t wait to see how they’ll perform, and how much of a threat the new X3D CPUs will be to Intel’s powerful Raptor Lake line-up.

What kind of 2022 did Microsoft experience? The company made some big moves during the course of this year, including a gargantuan near-$70 billion acquisition – mired in controversy – and it also brought the first major feature update to Windows 11. Let’s take a deep dive into how Microsoft fared across the software, hardware, and gaming fronts over the past year.

Windows 11’s big update (and bugged updates)

Windows 11 is still relatively new, having arrived late in 2021, and the biggest thing to happen to the OS this year was the debut of its first major update. (Remember that there’ll only be one feature update per year now, rather than twice yearly as was the case in the past). The Windows 11 2022 Update (informally known as 22H2) pitched up in September, applying a whole lot of polish, alongside some major and useful changes to the interface. We were also treated to a new system of ‘Moments’, fancy Microsoft jargon for smaller updates more agilely applied as needed outside of the big yearly feature upgrade.

Some of the notable improvements piped through included the introduction of tabs for File Explorer, allowing for multiple folders to be opened in one File Explorer window, in the same way there are multiple tabs in a browser. That represented the delivery of a long-promised feature from Microsoft. The taskbar also received a lot of attention, with a taskbar overflow panel being brought in, and after lots of clamoring from a good many Windows 11 users, drag-and-drop functionality was finally returned to the taskbar.

Indeed, it seems that something of a theme for the year was Microsoft listening more to feedback and user needs, as another much complained about aspect of the taskbar – the removal of the option to ungroup apps on the bar, a piece of functionality that’s in Windows 10 – looks set to be remedied. Or at least that facility has been present in testing with Windows 11, and should be here soon, fingers crossed, for those who don’t want multiple instances of the same app combined on the taskbar.

The feeling that Windows 11 is a work in progress still remained, with Microsoft needing to do a good deal of honing and revamping to knock various bits of the interface into better shape (and clear up cluttered legacy corners of the UI) – plus frustrating bugs were still a problem.

Throughout 2022, while there weren’t any show-stoppers in the way of Windows 11 bugs (as has been seen with Windows 10 in the past), we’ve reported on far too many glitches, and some nasty enough ones. Like printers having much of their functionality stripped away, for example, as well as gremlins that dragged down performance levels in various ways, and gamers came under fire from some seriously annoying bugs, too. Flaws in File Explorer seemed particularly rife, as well, which is a concern given that this is the core pillar of the UI (the very files and folders you work with every day).

Perhaps the real acid test we can use to measure the success of Windows 11 in 2022 is looking at how many people made the move to the OS (or bought a new PC with it installed, of course). According to Statcounter’s figures (as of November 2022, the latest available at the time of writing), 16% of all Windows users are on Windows 11. That’s not great for an operating system which has been out for over a year now, especially considering Windows 7 holds almost a 10% share of the market. A completely outdated OS is not far behind, in other words.

Furthermore, a year into its reign as the cutting-edge operating system for Microsoft, Windows 10 had secured almost 30% adoption, nearly twice as much as Windows 11 has managed. Although that said, Windows 11 is hampered somewhat because some people can’t upgrade, as their PC doesn’t meet the more stringent requirements (particularly on the security front), so Microsoft will have expected migration to be slower when they baked those elements into the design of the OS.

Overall, Windows 11 made reasonable headway in applying important interface changes, for sure, but we’d like to see more progress still, and especially a better effort in combating bugs, with an overhaul long overdue for Microsoft’s QA processes in our opinion.

Surface so-so

Just like any other year, another bunch of refreshed Surface devices emerged.

In June we got our first glimpse of the Surface Laptop Go 2, which wasn’t hugely different to the original laptop, but made some useful improvements, and remains a great choice for a compact notebook.

Then in October, at a big Microsoft event we witnessed the revelation of the Surface Pro 9 and Surface Laptop 5, alongside the Surface Studio 2+.

This wasn’t the best crop of Surface hardware for Microsoft, sadly. For starters, the Surface Laptop 5 failed to wow us, turning out to be spectacularly middle-of-the-road. Not terrible, but hampered by performance issues, and definitely not worth the hefty asking price pinned to the laptop.

The Surface Pro 9 turned out to be a better device, for sure, taking some laudable steps forward from the Pro 8. It also made an interesting move in having two flavors: a 5G model with a Qualcomm SQ3 (ARM) CPU, and a version with an Intel processor, the former being a great option for some (businesspeople on the move, for instance), albeit with ARM-related gremlins in the works. The Pro 9 remained pricey, but an undeniably slick device overall, and a good upgrade from its predecessor in the performance stakes.

The Surface Studio 2+ received just a minor upgrade, as the 2+ (rather than a 3) denoted, but Microsoft implemented some solid changes, albeit nothing particularly exciting. Which about sums up what happened on the Surface front overall: nothing bad, by any means – certainly not with the Surface Pro 9 – but nothing great, either.

Game for an acquisition

In January, Microsoft pounced to buy gaming giant Activision Blizzard in a huge acquisition for close to $70 billion. Yes, that Activision – the one with franchises like Call of Duty, Diablo, Overwatch, World of Warcraft and many other big-name games in its stable.

The magnitude of this deal can’t be underestimated, meaning that in the future, Xbox Game Pass owners could get the next Call of Duty blockbuster for nothing as part of that subscription (whereas PlayStation fans will have to stump up cash, and we’re all very familiar with the eye-watering cost of games these days).

Indeed, the ripples from this deal are still ongoing now, because at the time of writing this piece, the Federal Trade Commission (FTC) has filed a lawsuit to block Microsoft’s acquisition on the grounds that it could suppress rivals to the Xbox console.

Naturally, Microsoft has given us a whole raft of assurances on how it won’t throw its weight around in anti-competitive fashion after snapping up Activision, but there are plenty of doubters, and the FTC points to Microsoft’s past history and the ZeniMax (Bethesda) acquisition. While in that case, promises were also made about not withholding games from rival consoles, the FTC observes that post-acquisition, titles like Starfield, Redfall, and Elder Scrolls 6 were made Xbox exclusives.

Microsoft’s Activision buyout could still come a cropper at the final hurdles, then, potentially.

Ads nauseum – but with a glimmer of something positive?

While you might expect to see ads in a web browser, you don’t in your operating system. Because, you know, you paid good money for that OS, in the case of Windows anyway (or good money for the machine it came on, and that was part of the bill).

But ads taking up real-estate in Windows menus or panels again became a worry in 2022, as Microsoft tested – and note, these were just experiments that didn’t come to the release version of Windows 11 – adverts for OneDrive in the user session (shutdown) menu. And even more galling, the company briefly flirted with ads in File Explorer too, that central pillar of the Windows interface. That those waters were even tested is a concern, really.

All this could make you think that adverts are going to be a thorn in the side of Windows 11 users in the future – especially given Microsoft’s past form, this is not a new thing, of course – but instead, another possibility emerged late in the year.

Namely the notion that Microsoft could be angling towards low-cost PCs which are sold on subscription, rely on cloud services, and are supported by ads (to make them cheaper – so there’d be a benefit in this case). This was just a rumor, picked up (in November) via a Microsoft job listing, but earlier in July of this year, Microsoft CEO Satya Nadella talked about incorporating ads in its products.

There were clearly heard groans at this prospect, but as we observed, in a cost-of-living crisis, where not everyone might be able to afford a new PC, this could be a good thing to make Windows computers more accessible to those struggling with money.

Head in the clouds

Speaking of cloud services tied in with those ad-supported cheap PCs, this is an area where Microsoft went from strength to strength during 2022. Cloud revenue was strong right from the start of the year, and then in October, Microsoft’s Q1 2023 fiscal results showed that while the Windows division faltered – and was down 15%, in fact, partly due to the slump in PC sales this year – cloud revenue was buoyed by 24% compared to the same quarter in 2021.

Microsoft made $25.7 billion in just one quarter – over half its total revenue for that period – if you had any doubt about where the future lies for swelling the software giant’s coffers.

Concluding thoughts

With the PC slump, it wasn’t the best year for Microsoft – but cloud revenue was a clear highlight, and going by those various experiments with advertising in Windows 11, the company may have future plans to stoke profits in a different way, with low-cost ad-supported systems.

2022 was a pretty run-of-the-mill year for Surface hardware, albeit with some solid upgrades for some devices. And Windows 11 was a similar case of some reasonable progress being made, in the main, albeit with thorny asides in the form of bugs. However, a highlight with Windows 11 was seeing Microsoft listen more to user feedback, although arguably, some of the introductions made based on popular demand should have been addressed more swiftly.

That mammoth Activision Blizzard acquisition very much remains something to watch, as it’ll have potentially huge ramifications for the gaming industry next year if it goes through; but it feels like there’s still a distinct possibility of it coming off the rails.

Intel’s Arc graphics card driver has improved considerably over the course of a month according to a new report, with frame rates showing some impressive gains to the tune of almost 20% in the best-case scenario.

This comes from PC World, which used an Intel Arc A770 and measured frame rates in a small selection of games with the latest graphics driver, comparing that to the results obtained a month ago when the flagship GPU was released and the site first reviewed the card.

The benchmark rig contained the A770 GPU plus a Ryzen 5900X processor, and 32GB of system RAM (Resizable BAR was enabled, which helps boost frame rates). It’s the same hardware configuration (and OS version) as the original rig the review was performed with, as you’d expect.

The findings were that some games enjoyed hefty boosts, the biggest beneficiary of the updated Arc graphics driver being Total War: Troy which improved from 91 frames per second (fps) to 108 fps at 1080p resolution (ultra details). That’s an increase of 19%, no less, and at 1440p the game improved by 9%.

Horizon Zero Dawn witnessed an improvement of 10% (from 81 fps to 89 fps, at 1080p ultra), and Cyberpunk 2077 framerates were hiked by 5% at both 1080p and 1440p (ultra details).

Elsewhere, other games such as Assassin’s Creed Valhalla, Borderlands 3, Call of Duty: Modern Warfare 2, and Shadow of the Tomb Raider showed slight increases to the tune of 1-2 fps.

Analysis: Arcing upwards as time goes on…

With some of these games, then, a couple of extra frames per second is not something you’re going to notice in real-world gameplay, but it’s still an indication that things are at least moving in the right direction for Intel (something we’ve seen highlighted elsewhere, too).

And the large boosts for a few titles here are certainly some welcome news to focus on for Intel, and those hoping for a serious third competitor in the desktop GPU arena (to challenge Nvidia and AMD). Boosts of 10% and 19% in particular, over just the course of a month’s work on the graphics driver, are actually very impressive.

One game went backward, by the way, with Watch Dog Legions actually dropping by 5% at 1080p (and dropping 1 fps at 1440p too, but that’s not much of an impact). We’ll chalk this one down to an anomaly here, and hopefully now the problem has been highlighted it’ll be cured in the next version of the Arc graphics driver.

While this is admittedly a pretty limited run of tests from PC World, if Intel can keep up the general pace of improvement suggested here, or at least something like it, then the performance of Arc GPUs is going to go from strength to strength, month by month – at least in these early days when there are still a lot of kinks in the driver code.

Perhaps by mid-2023, Arc GPUs will be considerably more competitive as a result, and hopefully the likes of visual glitches and artifacts will also be ironed out more and more, as that’s the other task on the smoothing-out front for Intel, as well as achieving higher frame rates. Certainly, the kind of gains visible here in just a month is a positive sign that Intel really is changing things up a gear.

Remember that as well as updates to the Arc driver, there will also be patches applied to games in which developers improve performance on their side. So, measuring improvements in the way PC World has done here doesn’t guarantee they come from the driver, though in the space of a month, it’s likely that the gains mostly will. But at any rate, wherever those extra fps are coming from, the overall gaming experience is getting better, and that’s a good thing for Intel - and anyone with an Arc graphics card.

Via PC Gamer

Valve’s iconic game Portal is getting reimagined this December courtesy of Nvidia, and we're absolutely buzzing. The original Portal is widely regarded as one of the most iconic puzzle games of all time, and Nvidia will be bringing it to users complete with advanced graphics features like full ray tracing and Nvidia DLSS 3 capabilities to give the game a fresh look and feel.

Portal with RTX will be made available as a free official DLC (downloadable content) for existing owners of Portal on December 8, just in time for the game’s 15th anniversary. It’s also perfect timing for those of us that like to wind down during the holidays and play video games. 

The game's new look is achieved using graphics techniques that are heavily reliant on Nvidia DLSS tech, with an emphasis on the use of next-generation DLSS 3. Users of RTX 40-series graphics cards should see a 2.8x increase in performance according to Nvidia, while 30-series users will be able to enjoy the reimaging at a slightly lower - but still appreciable - level with DLSS 2. 

You can check if your system is compatible with Portal with RTX by consulting the system specifications here. 

Nvidia will be hosting a global Portal with RTX launch party on December 6, which the entire Portal community is invited to join via Twitch, Youtube, and Steam for behind-the-scenes access, new gameplay, and - arguably the most exciting part - giveaways! Virtual attendees will have the chance to win not just Portal with RTX game codes but GeForce RTX 4090 and GeForce RTX 4080 graphics cards!

Although the RTX 4080 might've fallen on hard times due to its high asking price and confusing nomenclature, it's still a powerful (if overpriced) GPU, and the RTX 4090 is obviously the big draw here. In fact, it might just be the only way to get your hands on Nvidia's newest flagship card right now, since stock is still virtually non-existent.

We’re deeply sad - heartbroken, even - to report that it looks like Nvidia is sunsetting the RTX 2060, long considered one of the best budget graphics cards ever made.

Recent reports seen on Chinese tech website MyDrivers appear to indicate that Nvidia has stopped supplying its vendors in Asia with RTX 2060 chips (as well as the upgraded RTX 2060 Super and RTX 2060 12GB), as well as requesting that its production and sales partners limit supply of the cards by the end of November.

It might be more than three years old now, but we’re looking rather wistfully at the departure of the 2060 family. These graphics cards were incredible: sensibly priced with sound performance, introducing many budget games to the magic of Nvidia’s DLSS tech as well as ray-traced graphics.

We weren’t the only ones who appreciated this budget GPU. Even today, the RTX 2060 is the second most popular GPU among Steam users, as per the most recent Steam Hardware Survey. The number one spot is held by the GTX 1060, another stellar cheap graphics card.

Why is Nvidia putting the RTX 2060 to sleep now?

It’s difficult to speculate regarding the reasoning behind Nvidia’s decision to end the RTX 2060’s life, but it’s most likely a move to push out the RTX 3000-series stock we know the GPU giant still has knocking about.

The xx60 and xx50 versions of Nvidia GeForce GPUs are typically much more popular than Team Green’s crazy-expensive flagship cards - like the recently launched RTX 4090. Looking at the survey results, the top ten most-used cards are all xx60 or xx50 budget variants, with the exception of the excellent midrange RTX 3070 in seventh place.

This really shouldn’t be surprising to anyone who hasn’t been living in a deep-sea cavern for the past few years. The cost-of-living crisis is squeezing our wallets, deepening the gulf between the wealthy and poor, and the simple fact is that right now most gamers can’t afford to drop $1,000+ on a GPU.

The RTX 4080 is proving to be a particularly pertinent example of this. In the UK, where skyrocketing energy bills and food prices are wreaking havoc on the finances of people all across the country, we’ve seen distinctly more stock availability than in the US. The average PC gamer doesn’t have hundreds of pounds to spend, so we’re all left sitting with our trusty RTX 2060s.

A new dawn for budget GPUs - we hope

Nvidia’s planned wind-down of the RTX 2060 will clear the path for (hopefully) less expensive Lovelace GPUs like the RTX 4060 and the newly-rumored RTX 4050. Let’s face it: these are the cards people actually want, provided they’re not so absurdly priced as the currently-available RTX 4000 cards.

It’s mildly depressing to see Nvidia and AMD engaged in an apparent race to the top when it comes to GPU pricing. AMD has long provided solid budget alternatives to Nvidia’s cards, but its recent unveiling of its new Radeon GPUs left us wanting more - or, more accurately, wanting fewer digits on the price tags. The RTX 2060 first launched at a very respectable $329 (£300/AU$599); today, you can pick one up for around $250.

If you’re still rocking an old GPU that desperately needs upgrading, have no fear: the RTX 2060 will likely remain on shelves for months to come while Nvidia shakes its remaining stock out of its shirtsleeves.

In fact, this news means there’s a very good chance that we’ll see some serious discounts on older RTX GPUs this Black Friday, so we’d heartily recommend you keep an eye out for the best graphics card deals with our handy guide.

Now, if you'll excuse us, we're going to go dig out our old office RTX 2060 and cradle it gently while listening to R.E.M.'s 'Everybody Hurts'.

Asus TUF Gaming AX5400: One-minute review

Asus has a separate brand, called Republic Of Gamers (RoG), that it uses for its high-end gaming gear, including its Rapture range of high-performance gaming routers. However, Asus also sells a couple of more affordable routers under its own name, using the TUF Gaming label. 

The Asus TUF Gaming AX5400 is one of the more expensive TUF models, but it still comes in at under $200, and its dual-band Wi-Fi 6 provides a top speed of 5.4Gbps, which should be more than fast enough for gaming on most domestic broadband services. And, for really serious gamers, the TUF router also provides ‘link aggregation’ features that can handle super-fast fibre broadband connections too. 

It also supports Asus’ AiMesh technology, which allows it to work as a mesh router in conjunction with other Asus routers, so you can further upgrade your home or office wi-fi in the future if you need to.

Asus TUF Gaming AX5400: Price and availability

• How much does it cost? $199 (£170, AU$369)
• Where is it available? Available now
• Where can you get it? Available in the US, the UK, and Australia

The TUF AX5400 is competitively priced for a high-speed Wi-Fi 6 router, costing $199 (£170, AU$369). That compares well with similar Wi-Fi 6 routers such as the Linksys Hydra Pro 6 ($249 / £169), whilst also offering link aggregation and other gaming features. And, for more casual gaming action, there’s a model called the Asus TUF Gaming AX3000, which offers 3.0Gbps speed for $189 (£127, AU$214).

• Value: 5 / 5

Asus TUF Gaming AX5400: Design

• Easy-to-use app
• Good parental controls
• Browser interface for advanced users

It may not have the over-the-top lighting of Asus’ Rapture routers, but the Asus TUF Gaming AX5400 certainly looks like a gaming router, with a chunky, angular design and an array of six separate antennae that helps to provide a fast, reliable wi-fi signal.

Like most conventional routers, the Asus TUF Gaming AX5400 includes one Gigabit Ethernet WAN port for connecting to your existing broadband modem or router, along with four additional Gigabit Ethernet LAN ports to provide a lag-free wired connection for PCs or gaming consoles. It’s also possible to combine - or ‘aggregate’ - the WAN port with one of the LAN ports in order to handle broadband speeds as high as 2Gbps. 

One of the Ethernet LAN ports is also assigned as the ‘gaming port’ and takes priority over the other Ethernet ports to ensure that your gaming rig always gets the highest possible speed. You can also aggregate the gaming port with a second LAN port to provide 2Gbps speed over Ethernet if required. There’s also a USB 3.2 port provided as well, so that you can connect a USB storage device and share it with other users on your network.

Getting started is very straightforward, as the Asus Router app allows you to simply scan a QR code in order to connect to the TUF router’s wi-fi network. Somewhat oddly, this new network is not initially protected by a password - something we’ve seen with other Asus routers in the past - so you need to quickly assign your own password to the network in order to keep it secure.

The Router app works well, though, giving you the option of creating two separate networks on the 2.4GHz and 5.0GHz frequency bands, or combining them into a single network. There’s a handy network overview that allows you to monitor the data traffic and all the devices connected to your network. The QoS option - ‘quality of service’ - allows you to prioritise tasks such as gaming or streaming video in order to provide top performance for your gaming sessions. 

It also provides good parental controls, allowing you to create schedules for your children’s Internet use, and to select filters for adult content and other unsuitable material. There’s also a web browser interface available for advanced users who may want more detailed control over the network settings.

• Design: 5 / 5

Asus TUF Gaming AX5400: Performance

• Good performance on 2.4GHz and 5.0GHz bands
• High-speed wired connections for gaming

The Asus TUF Gaming AX5400 router performs well too, easily handling Steam downloads and gaming sessions on our 100Mbps office broadband service. Our normal broadband router can manage full speed when connecting to devices in the same room, achieving 100Mbps with the Ookla speed test, and 12.5MB/s for Steam downloads. 

As we expected, the Asus TUF Gaming AX5400 router simply matches those speeds when connecting to nearby devices, but we also have an office in the back of the building that our old wi-fi router struggles to reach, and forcing me to use PowerLine adaptors to provide a wired connection for my office iMac. 

However, the Asus TUF Gaming AX5400 router wins hands-down here, comfortably maintaining 100Mbps in the Ookla test and 12.5Mbps on both the 2.4GHz and 5.0GHz bands. That will make it a great - and affordable - option for gamers who need to boost their home Wi-Fi.

• Performance: 5 / 5

Should I buy the Asus TUF Gaming AX5400?

Buy it if...

Don't buy it if...

Also consider

Asus TUF Gaming AX5400: Report card

• First reviewed October 2022

How we test

We pride ourselves on our independence and our rigorous review-testing process, offering up long-term attention to the products we review and making sure our reviews are updated and maintained - regardless of when a device was released, if you can still buy it, it's on our radar.

Read more about how we test

Alright, listen. RTX Remix is a genuinely cool bit of tech wizardry, and it undeniably has the potential to revitalize the modding scene for a ton of old games. I’m already trying to work out how I can possibly afford an RTX 4090; my plans may or may not involve selling a kidney.

As a long-time Elder Scrolls fan, I won’t lie: seeing Morrowind lovingly re-rendered with RTX Remix did excite me a little. I’m already imagining what a Remixed Fallout: New Vegas would look like, and it’s making me less depressed about the current state of the Fallout franchise. I can hear Marty Robbins singing ‘Big Iron’ already.

But there’s something about the whole thing that sets off the tiny ‘something’s wrong here’ alarm in the back of my head. About halfway through my second viewing of the Portal RTX trailer, it hit me: sure, this looks better, but it doesn’t look how it’s supposed to look.

I’m making a note here: huge success

Games are universally bound by the technological hardware capabilities of their time, and a big part of game design involves figuring out how to create a game that looks good within the confines of those capabilities. Today, many triple-A games are edging closer and closer to genuine realism; a few short years ago, that simply wasn’t possible.

So, developers were generally given the choice of two paths: heavy stylization, such as using pixel art or cel-shading, or trying to create as realistic a world as possible with the tools available. The former method tends to age better (provided that the original style was, you know, good), while the latter is a bit more nebulous.

Take TES: Morrowind, for example. While Bethesda clearly wasn’t going for hyper-realism in its game about elves and giant clockwork golems, the game does try to create at least a reasonable facsimile of a live-action fantasy world that could run on 2002 hardware. In a post-Fellowship of the Ring world, it made sense. The problem is that today, Morrowind looks like shit.

Ray-traced role-playing

That might sound cruel, but anyone who has played Morrowind will likely agree; while the game’s story and worldbuilding have stood the test of time, its graphics just… haven’t. The textures are flat, the overworld’s color palette is frequently bland, and the character faces are frankly haunting.

It’s in games like Morrowind where I do agree that RTX Remix can shine. There’s already a gigantic modding effort to recreate the game in the Skyrim engine - titled ‘Skywind’ - and it looks great. Modders are a determined and phenomenally creative lot, as evidenced by huge projects like Skywind and Fallout: London. Giving them a tool as powerful as Remix makes me excited for what modders could create in the future.

But Portal with RTX? No, I’m not excited about that one. For starters, it’s an official project - this might seem like an odd thing to pick at, but if it’s made in-house by a bunch of paid developers, it’s not a mod. It’s an expansion. Portal’s shiny new reflective surfaces and reactive lighting weren’t produced out of a love for the game and a desire to expand it, the way most mods are. It was created to sell overpriced graphics cards.

That’s a cynical view, sure, and one that falters a bit when you consider that Portal’s newly Remixed environments do actually look great. The thing is, Portal already looked great. It’s one of the rare games that aim for realism and manages to make it timeless, through a combination of smart design choices and excellent art direction.

The simple tiled walls, the harsh lighting, the polished white bodies of the sentry turrets, and the portal gun itself - this all comes together to produce a still-engaging rendition of an inescapable, sterile laboratory complex, feeding effectively into the game’s minimalist but still intriguing plot.

Portal 2, another stone-cold masterpiece, took things further with the benefits of more powerful technology; less mechanical animations for characters like the hyper-animated Wheatley, more detailed environments as the Aperture Science Labs fell into disrepair and overgrowth, and wacky physics effects for gravity-warping paint. But the original Portal still holds up, using the capabilities of 2007 hardware to excellent effect.

We do what we must because we can

My fear here is that we’re going to see a wave of Remixed classics with little regard for the original style of the games in question. Some of these - like Morrowind - will be good, a way to graphically update old games to better realize the vision the developers originally had but couldn’t create due to hardware limitations.

Some - like Portal - will be unnecessary redesigns of games that already looked good, created merely because they could be rather than because they were needed. I’ll be very interested to see Portal’s player counts on Steam once the RTX 4000 cards arrive; I’d put money on very few people actually going back to replay it with RTX.

Another big problem with Remixing games like this is that it throws any intentional design decisions related to lighting straight out of the window. Environmental lighting has long been a vital part of building a game’s visuals, after all.

Light can be used to subtly illuminate an object or path, telling the player where to go or what to do without the need for a glowing, immersion-breaking HUD icon. Lighting draws the eye, putting a character (sometimes literally) in the spotlight during an important moment. It’s an oft-unmentioned element of game design that RTX Remix threatens to override.

The light fantastic

Take Portal, for example. Valve is no stranger to using lighting to direct the player; the exit door (or another item of importance) in many of the test rooms is deliberately positioned so that the room’s light sources fall upon it. I reinstalled the game to reacquaint myself with it before writing this article, and noticed that this design choice is visible in multiple screenshots just on the Steam store page.

This sort of thing is more common than most realize in games; when footage of The Last Of Us Part 1 remake was revealed, there was some discussion on Twitter about how the remake’s more ‘realistic’ lighting actually damaged the original game’s careful art direction and by extension the way the player was guided through the world.

The tweet above is about someone complaining about a door-kicking sound effect, so feel free to ignore that part. But take a look at the lighting in the scene when the door is opened; in the original PS4 remaster, bright orange sunlight hits the wall opposite an adjoining alley, so you immediately know there’s a route leading that way, even though you can’t actually see the other alley yet.

In the gritty realism of the remake, this lighting cue is essentially abandoned; the alley is sparsely lit, and the addition of a little bloom when Joel kicks open the door - again, an effect added to promote realism as you move from a dark environment to a brighter one - further muddies the scene.

This door-bullying moment is part of a chase sequence, so the player knowing where to go is key. If you lose sight of your quarry, lighting is one of the factors used to keep you on track. It’s worth noting that the Part 1 remake doesn’t appear to use ray-traced lighting, but this is a good example of how important lighting can be as a game mechanic.

Remix and match

This is what I’m worried about seeing. Sure, the original Portal wasn’t a game that struggled massively with player direction, thanks to its self-contained puzzle rooms and generally clean aesthetics. But there are plenty of games in which a developer will have painstakingly placed fixed light sources for gameplay purposes, only for RTX Remix to come along and ruin everything.

My last point - and one that I’m aware might come across as a bit petty - is that Portal RTX is really just a waste of time. Seeing dynamic shadows or slightly more realistic water textures isn’t going to bring back the same feeling of mystery and intrigue Portal held when I first played it on my dad’s struggling desktop PC all those years ago.

Yes, I get the need to showcase Remix’s capabilities for the big RTX 4090 reveal livestream, but was doing it to a whole game really necessary? The Morrowind clip was more than sufficient to generate excitement, especially since a lot of newer Elder Scrolls fans will have bounced off the older games due to their janky graphics.

In short: RTX Remix is an exciting bit of tech for modders, but we really don’t need soulless corporate remakes of already-great games that have been fed through the ray tracing grinder to remove any sense of deliberate art direction. Not every game has to look real. Hell, I play games to escape from real life, not be reminded of it.

Put cool RTX tech in the hands of modders by all means, but this is my plea to big developers: your games don’t need the Remix treatment, and you don’t need to devote time to remakes when the original titles are still out there for people to play. Put those funky ray-traced reflections in a new game instead. Oh, and Nvidia? I’ve watched your GeForce Beyond presentation three times now, and I still don’t really know why Remix has to be part of an ‘Omniverse’, or what an ‘Omniverse’ actually is.

Intel has put together an incentive bundle to persuade folks to buy prebuilt PCs with an Arc Alchemist graphics card plus Alder Lake CPU, consisting of a bunch of games and more serious software.

The all-Intel machines come with $190 (around £165, AU$280) worth of PC games (and in-game content in one case) plus a choice of three out of five featured apps that, if you select the most expensive products, adds up to $284 (around £250, AU$420). All of this is going by Intel’s working out of the average retail price.

This offer or ‘Intel Software Advantage Program’ was spotted by @momomo_us on Twitter (via VideoCardz), and pertains to specific PCs with Arc A5 or A7 graphics cards alongside Alder Lake Core i5, i7 or i9 processors (both desktops and laptops are included).

The game bundle consists of Call of Duty: Modern Warfare II, Ghostbusters: Spirits Unleashed, Gotham Knights and in-game content for Vampire: The Masquerade – Bloodhunt.

On the software front, you can select three of the following five: PowerDirector 365, D5 Render, Magix Video Pro X14, Topaz Gigapixel AI, or XSplit Premium Suite.

As mentioned, with the most expensive apps of the latter bunch picked, you’re looking at a bundle worth $474 (around £410, AU$705), or with the cheapest software selected, $380 (around £330, AU$565). All of these products are delivered as digital downloads.

The offer is active now, or rather, when the A5 and A7 graphics cards emerge – we’re expecting the higher-end A7 models, namely the A770 and A750, to arrive very soon, Intel recently informed us. Of course, the games haven’t yet been released either, with the three titles on offer not coming out until next month.

The promotion itself lasts until the end of 2022, by which time you must claim your code to grab your download.

Analysis: Give it away, now…

Whatever software choices you make here, this is a pretty chunky giveaway, with some nifty freebies available. And as you might expect, the games are picked due to their features with Arc GPUs, boasting support for XeSS allowing for frame rate acceleration via Intel’s DLSS rival – and we already know that Team Blue has implemented ray tracing for Gotham Knights. (Intel has recently been stoking the hype around XeSS, too, and it sounds pretty impressive on the face of it).

Clearly, Intel is serious about making its higher-end Arc Alchemist graphics cards more attractive given that they will not be able to compete even remotely with AMD and Nvidia’s top dogs – and that’s before the next-gen GPUs from Team Red and Green arrive, which won’t be too long now.

Intel has indicated that while Arc performance can be boosted somewhat by drivers as time rolls on, it won’t be to any major extent, and so the main route that the chip giant will have for competing with the established rivals in the graphics card world is to do so on pricing. And evidently part of that, beyond dropping MSRPs, will be value bundles like this – though we have to remember this isn’t a pure GPU bundle, but one that comes with a whole PC (an entirely more expensive proposition, of course).

The trouble with the bundle scene is that right now, AMD and Nvidia are offering some quite attractive packages with standalone graphics cards, so Intel does have its work cut out to compete. (For example, Nvidia’s recent ‘face your demons’ bundle sports Doom Eternal – plus two expansions for the shooter – and Ghostwire: Tokyo).

In some respects, the problem for Intel on the value front is that it’s coming into an environment where Nvidia in particular needs to shift current-gen RTX 3000 GPUs before the next-gen launch, and price cutting action is expected to move up a gear as soon as this month, really. Still, whatever the case, we’re not going to argue with big giveaways or hefty price drops regardless of where they’re coming from – the more, the merrier, quite simply.

At Lenovo’s Tech Life | IFA event, several monitors were showcased with the highlights being the upcoming gaming monitor Legion Y32p-30 and the curved monitor ThinkVision T34w-30.

The Legion Y32p-30 is an impressive 31.5-inch monitor with an equally impressive 3840 x 2160 resolution. Its response time is an extremely fast 0.2ms with a refresh rate of 144 Hz. Those are already great specs for gamers but it has something to offer creatives as well, with a 99% sRGB for its coverage ratio and a 112% sRGB for its color gamut coverage.

Seeing it in action truly reflected its high quality, with a crystal clear picture that’s sure to make any video game or creative project pop in color. The brightness helps along as well, with a peak of 400 cd / m2.

It also features an excellent port selection for a monitor: two HDMI 2.1 ports, one audio jack, one Display Port 1.4, one USB Type-C port, two USB Type-A, and one USB Type-B. These features and specs could easily put it in the running as one of the best gaming monitors to be released this year.

The Legion Y32p-30 gaming monitor is expected to launch in December 2022 for $749.99 (around £650 / AU$1,100).

There’s also the ThinkVision T34w-30, which is Lenovo’s 34-inch curved monitor option. It boasts a 3440 x 1440 resolution (21:9) at a refresh rate of 60 Hz, which is perfect for more casual use as a business PC monitor or a screen for watching movies. Its 99% sRGB coverage ratio lets it double as a monitor for creative projects.

The curved aspect really adds to the monitor, allowing for a more complete view of the screen, but not so curved that it becomes overwhelming or claustrophobic. And the size helps, as it has great screen real estate without forcing you to crane your neck or lean back in uncomfortable positions to view the whole screen.

Its port selection is also quite good as it includes an ethernet port, an HDMI 2.0, one DP 1.4, one USB Type-C port, four USB Type-A ports, and an audio jack.

The ThinkVision T34w-30 is expected to launch in January 2023 for $849 (around £740 / AU$1,250).

IFA 2022 is Europe's biggest tech show, and TechRadar is in Berlin to bring you all the breaking news and announcements, plus our hands-on first impressions of the new TVs, wearables, audio devices and other gadgets on show.

AMD’s Ryzen 7000 processors which use 3D V-cache to pep up things for gaming will be revealed at CES 2023, a new rumor claims.

This comes from Greymon55 on Twitter (via Wccftech), who made the assertion in a somewhat cryptic tweet which just listed ‘V95, V9, V8’, referring to the following respective models which will supposedly be the first to arrive with V-cache: the 7950X3D, 7900X3D and 7800X3D.

Naturally, we must be skeptical about anything like this from the rumor mill, but it’s not the first time we’ve heard that AMD is planning to get some 3D V-cache toting Ryzen 7000 silicon out there relatively early in 2023.

Note that Greymon clarifies that CES will just be an initial unveiling of the purported trio of X3D models, with the CPUs not actually going on sale until February or March (or that’s the “likely” scenario, the leaker believes).

Analysis: A quick follow-up to keep Intel on the ropes?

As you doubtless saw, AMD unveiled its Ryzen 7000 processors earlier this week, with the initial models to be revealed consisting of the Ryzen 9 7950X, Ryzen 9 7900X, Ryzen 7 7700X, and Ryzen 5 7600X, just as was rumored. These mentioned 3D V-cache models will be the follow-up batch, in theory, and AMD did say that it’s already working on these chips.

One notable move with the new Zen 4 CPUs is pushing out a 7700X, rather than an 7800X – remember that the Ryzen 5000 range launched with a 5800X (the 5700X didn’t come until much later, earlier this year in fact). So, if AMD is indeed planning a 7800X3D to come out soon, the omission of the 7800X makes sense in that respect.

Furthermore, regarding the release date as in hitting the shelves, we can see the 7800X3D being first out of the door, as the most affordable and mainstream model, to really tempt gamers like the 5800X3D has done – so perhaps that will arrive in February, and the other models could follow in March.

It’s a bit of a surprise to see that AMD could be planning a pair of high-end 3D V-cache chips, the 7950X3D and 7900X3D, as these will be pricier and more niche propositions. Previous chatter from the grapevine has talked about an X3D flagship, but just that and the 7800X3D, which we feel sounds more likely than two top-end offerings, as that feels like overkill.

Particularly as AMD will want to focus on getting a decent amount of stock of these Zen 4 V-cache CPUs out there right from the get-go, because as we’ve been informed via the rumor mill, this is one of the key points with the initial Ryzen 7000 launch (with those first processors coming out on September 27).

If AMD can get plenty of X3D processor stock readied for a February or March launch, that could well keep it ahead of the game in terms of the battle with Intel. Right now, with those initial Ryzen 7000 CPUs coming at the end of September, they’re set to have a head-start of a few weeks on Intel’s next-gen processors (Raptor Lake) if the rumors are right – with plentiful stock to make a killing on sales, in theory.

Raptor Lake might potentially shift the direction of the limelight subsequently – a lot, of course, will depend on relative performance versus Zen 4, and crucially pricing – but if that does happen, AMD could have a pretty quick reply on hand with these V-cache models.

All of which is to say that it’s starting to look like AMD could have the upper hand in the battle of the next-gen CPUs, although that prospect might just drive Intel to be more competitive with Raptor Lake – rather than hiking prices, as is rumored – which would be a big win for consumers, of course.

Spider-Man Remastered has swung onto PC after a few years as a PlayStation exclusive, and it's one of the best PC ports we've played.

Not only does Spider-Man Remastered work great on the Steam Deck handheld console, but it comes with a huge amount of graphical settings and features, such as ray tracing, DLSS upscaling, uncapped frame rates and ultrawide monitor support.

This is another PlayStation PC port from Sony that puts other ports, such as Elden Ring, to shame.

So, naturally, I was excited to see how Spider-Man Remastered runs at 8K on our beastly test PC, especially as I had finally got my hands on the Nvidia GeForce RTX 3090 Ti graphics card, currently the most powerful gaming GPU in the world.

However, before I was able to play the game, disaster struck: I broke the PC.

A special kind of broken

I've been tinkering with PCs since the early 1990s, and in that time, I've seemingly perfected the art of breaking PCs. Among friends and family (and now on the internet), I'm the first person people go to when they have a problem with their PC or laptop.

These amateur PC destroyers usually have problems I can identify and fix pretty quickly. At worst, they need a new hard drive or a Windows reinstall to fix things. Pathetic.

When I have a problem with my PC, I really have a problem. And so, that came to pass as I replaced the RTX 3090 with the RTX 3090 Ti. Everything seemed fine... except the PC wouldn't boot, USB ports weren't working, and the RAM's RGB lights didn't fire up.

After ensuring everything was properly seated I tried again – still no luck. However, a white light on the motherboard next to the letters VGA gave me a clue – it was to do with the video card. Makes sense, I thought, as I'd just installed a new one. So I took it out, put it back in, and... had the same issue.

So, I put the older GPU back in, worried that it may be the RTX 3090 Ti's fault. But it wasn't. The PC still wouldn't boot, and the light next to VGA stayed on. Taking out the GPU and using it in another slot didn't help, and worryingly, having no GPU in there at all and using the onboard graphics also didn't help.

I was stuck. I tried swapping out the RAM (luckily I have plenty lying about, being a big ol' PC geek), which was tricky thanks to the huge CPU cooler in the case. I also tried to reflash the BIOS, the last refuge of a scoundrel, but no luck. The PC was dead, Jim.

Scouring the internet, it seemed like other people had similar issues with the same motherboard I was using, but any fixes they offered didn't help.

It looked like I would have to do some major surgery on the PC.

Time for a brain transplant

It looked like the issue was either to do with the motherboard or the CPU (seeing as using integrated graphics of the processor didn't fix things). This meant I'd have to replace those. While I could have done some more troubleshooting to determine which one was causing the issue, time was of the essence, as I really wanted to test out Spider-Man Remastered at 8K as close to its launch as possible.

While having to replace the motherboard and CPU was certainly frustrating – and is one of the more time-consuming upgrades you can make – I decided to put a positive spin on it. I used it as a chance to make some big upgrades to the 8K rig and future-proof it for upcoming games.

So, I swapped out the 8-core Intel Core i9-9900K for a 16-core Intel Core i9-12900K chip to make the PC more modern.

A new CPU means a new motherboard, which is just as well, as the mobo was one of the components I suspected of being faulty. So, out went the Asus ROG STRIX Z390-E GAMING, and in its place (literally) went the MSI MPG Z690 Carbon WiFi.

One of the main selling points of this motherboard is that it comes with a large selection of M.2 SSD ports, so it was easy to move all of the existing storage from the old motherboard into the new one. I also added another 1TB WD_Black SN770 SSD for good measure.

I wasn't done upgrading, however. The 12th-gen Intel processor supports the new, faster DDR5 RAM. So I took the opportunity to switch to 32GB of Corsair Dominator Platinum RGB DDR5 5200MHz RAM, replacing the 32GB of G.Skill Trident Z RGB DDR4 3200MHz memory.

Oh, I also needed a new CPU cooler, and while the large air cooler I did have was impressively quiet, it was so big that it made tinkering difficult. So, I used the Corsair iCUE H150i RGB ELITE Liquid CPU Cooler instead.

As with many all-in-one liquid coolers, the cooling plate that rests on the CPU is connected via tubes to fans and a radiator in another part of the chassis (I installed it at the front due to the case layout, with the fans pulling in cool air from outside). This gives you a lot more space on the motherboard itself.

Finally, as this ended up being a major upgrade operation, I took the opportunity to replace the Corsair RM850x 80 PLUS Gold 850W PSU with the Corsair HX1500i 80 PLUS Platinum 1500W.

While this made the upgrade more complex (despite both PSUs being modular and by Corsair, you should always change the cables when switching power supply units), it's a forward-thinking upgrade. The RTX 3090 Ti is a demanding enough beast that the 850W PSU may have struggled but, if rumors about the upcoming RTX 4090 are true, the next generation of graphics cards could be incredibly demanding when it comes to power, especially at the high-end. So, while I was elbow-deep in components and cables, I thought it wise to upgrade to a hefty 1500W model.

Because I was keeping everything else, including the case, the upgrade process was quicker than building a PC from scratch.

Once completed, I pressed the power button with bated breath and... success! The PC booted.

Even better, Windows 11 booted straight away, and after a little bit of automatic configuring, I was back on my desktop, with everything where I left it. I didn't even need to reactivate Windows 11. It was impressively smooth, and I have to give Microsoft and Windows 11 credit here. Previously, swapping out the CPU and motherboard could cause all kinds of nonsense with the operating system and would sometimes make the OS consider the upgraded PC a whole new device, requiring you to purchase another activation key.

Not this time, though. So, with a sigh of relief, I fired up Spider-Man Remastered and got started.

Spectacular performance

Usually, when I try 8K games I set the graphics levels to their highest setting, see how poorly the game performs, then begin tweaking settings, such as lowering graphics quality or adding upscaling tech like Nvidia DLSS to get a playable frame rate.

That's what I set out to do with Spider-Man Remastered, with graphics set to 'Very High' and ray tracing effects enabled. Jumping into the game and swinging around New York, I was blown away by how good it felt, despite the intensive settings and resolution of 7,680 × 4,320.

At this level, average frame rates were an impressive 32fps, just above the 30fps minimum I consider playable – and which the original Spider-Man game ran at on the PlayStation 4.

While framerates did dip to a minimum of 26.4fps, it also hit highs of 39.3fps. For a game as visually stunning as Spider-Man Remastered, with a large and busy open world, that's remarkable. It's a testament to the power of the Nvidia RTX 3090 Ti, as well as the hard work of developer Insomniac, plus Nixxes Software, which handled the PC port.

And while the original game is four years old – and made for hardware that launched in 2013 (finding out the PS4 is nearly a decade old has made me feel old, thanks time!) – the remaster looks fantastic. The increased details and realistic ray-traced reflections really make the game's recreation of New York feel alive.

Switching off ray tracing gave the game a bit of a bump to 36fps on average and a minimum of 30fps. This was surprising as ray tracing usually makes a big impact on frame rates – sometimes as much as halving them when turned on.

The slight increase in performance wasn't worth losing the ray tracing effects, so I turned them back on, but this time, I also turned on DLSS using its 'Balanced' setting.

DLSS is a clever method of using AI and machine learning to upscale games running at a lower resolution so that they appear to run at a higher resolution.

It can help you use intensive graphical settings while getting a performance boost. When done well, it means image quality isn't much different from running the game at the higher resolution.

In the past, this has made previously unplayable games playable at 8K. The balanced setting, as the name suggests, offers a mix of graphical fidelity and performance, and I saw an instant improvement. Average frame rates were now at 51.4fps and hitting a maximum of 60fps, which is the ideal frame rate for 8K.

The game felt smoother to play and still looked very good, with little noticeable graphical difference while swinging through New York.

Switching to DLSS Performance mode, where performance takes precedence over image quality, there was another decent boost to 57.5fps on average. While some textures were a bit softer, the game still looked great, and those extra frames made it feel sublime. After all, this is a game that focuses on fast and fluid movement and action.

Switching to DLSS Ultra Performance got the game up to 59.7fps, basically at the golden 60fps mark. Graphical fidelity did take another knock. This time there was a perceptible haze around Spider-Man when he moved, but nothing too distracting, and it was undeniably impressive that I was playing the game at essentially 60fps.

After swinging around New York for a bit, I took on some muggers during a randomly generated crime, and frame rates during the fight remained at a solid 59.9fps. Because the fighting in this game relies on quick combos and strict timing, a dependable frame rate is essential, and that's exactly what I got.

Is 8K gaming more realistic?

When I started out with 8K gaming, back with an Nvidia RTX Titan, the idea of playing games at 7,680 × 4,320 resolution was, frankly, far-fetched. Not only were 8K displays hard to come by, but the performance impact of playing games at 8K, even on what was then the most powerful – and expensive – graphics card in the world, just wasn't worth it.

Since then, we've seen more 8K TVs come out, more powerful GPUs with the RTX 3090 and RTX 3090 Ti, and games that are better optimized for ultra-high resolutions. Technology like DLSS and FidelityFX Super Resolution – both of which Spider-Man Remastered supports – have made gaming at 8K even more achievable, as long as you have the expensive hardware that can handle it.

The fact that Spider-Man Remastered runs at 30fps at 8K with graphics settings at their maximum level and ray tracing on, is incredible. While 4K remains the sweet spot between looks and performance (the exact same settings at 4K got us 70fps on average), it shows that gaming at 8K is no longer a distant dream. Thanks to DLSS pushing the average frame rate up to 60fps, the future looks bright for 8K gaming.

I should once again give a shout-out to Sony, Insomniac, and Nixxes Software (especially the last one) for a brilliant PC port that includes a huge range of settings. Because of the excellent performance straight out the box, I didn't need to look at other upscaling options, such as AMD's FidelityFX Super Resolution, or use the Dynamic Scaling option, which lets you set a preferred frame rate, and the game will scale the resolution accordingly to make sure it hits that frame rate.

While I didn't make use of it, this could be another handy tool for people with lower-powered hardware who want more consistent performance.

This is great to see. While 8K gaming is great for testing the limits of hardware and software, using the technology to make modern games more accessible to people who can't afford powerful hardware is where this really counts.

Considering that gaming routers come at a premium, figuring out whether the investment is worth it or is just a bit of marketing can be tough without a little bit of help. After all, you don’t want to spend that extra cash on something with RGB lighting or an aggressive aesthetic when any old router will meet your network needs.

Of course, the best gaming routers do offer more than just a particular aesthetic and some gamer-centric buzzwords. They cater to certain types of users that need blazing fast speeds and rock-solid performance and it’s not because they’ll be streaming the latest Netflix blockbuster and checking emails.

That said, investing in a router made for gaming is a little less exciting and often harder to justify than, say, upgrading your GPU. But, depending on what you need out of your router, it might be worth it. In that vein, we’ll consider all sorts of options including wireless routers and mesh routers to see if you should upgrade to one.

We’ll dig into whether getting one of these routers will help you get the most out of that gaming PC or gaming laptop. Not all gamers will want to make the jump, but for some, it will be worth it. We’ll factor in the unique features these routers come with as well as the type of gamer to help you decide.

What makes a gaming router?

• Performance and features set gaming routers apart
• They come with robust bandwidth solutions

Getting a gaming router will not all of a sudden make you a better player. But, it could make a difference in how seamless your sessions are by minimizing network-related issues or limitations. Experiencing lag, screen tearing, or other interruptions because multiple devices are vying for bandwidth all play a part in whether you need one of these routers.

To start, gaming routers offer the kind of fast data transfer speeds that you find in more robust routers. And, the current generation routers such as the Asus ROG Rapture GT-AX6000 are more likely to come with the latest Wi-Fi 6 standard, allowing even faster speeds for computers and laptops that also support the standard.

Having enough bandwidth is always an issue, and one of the main things that gaming routers try to tackle is managing that bandwidth so that gaming devices keep a steady and low ping (or latency). While just about every Wi-Fi router younger than Tik-Tok comes with at least a 2.4GHz and 5GHz band to organize device traffic, gaming routers sometimes offer tri-band connectivity for even more capacity to handle for multiple devices. The Asus ROG Rapture GT-AXE11000, for example, includes a blazing 6GHz frequency band for a faster, more continuous signal that’s less prone to interruptions from other devices on the network.

Even without that next level performance and bandwidth, a gaming router’s feature-set is still unique, not to mention useful. It’s able to help you prioritize your gaming devices during your gaming sessions, find servers that are geographically closer or are experiencing less traffic, and have a plethora of ports for dedicated connections.

Gaming router features explained

• Gaming router can more easily prioritize gaming network traffic
• They generally come with more ports

The included feature set on a gaming router separates it from the pack as much as its performance and bandwidth do. Though most routers come with physical ports for the devices that need it, models aimed at gamers come with more ports as well as more specialized ones dedicated to gaming consoles and PCs. USB ports are also typically on hand to plug in printers or external hard drives you want available to the whole network.

Whether or not you’re physically plugged in, these routers have a number of ways to minimize the amount of latency you experience while gaming. QoS (quality of service), a feature available typically through software that interfaces with the router, does just that by putting whatever you’re playing at the top of the list of devices and tasks vying for bandwidth. Many gaming routers also come with the ability to find the shortest path to the server to which you’re connected or even force a connection to the closest server to minimize latency (sometimes called geofiltering or geolocating).

You’ll also find MU-MIMO (multi-user, multiple-input, multiple-output) support on most gaming routers, a feature that allows multiple devices almost instantaneous access to the network. This way, someone watching Netflix will still get a solid signal regardless of how poorly you’re doing in Fortnite at the moment.

That’s just the tip of the iceberg when it comes to features. Security features, the ability to test and monitor ping and bandwidth, and even sometimes RGB lighting are part of the package.

Strengths and weaknesses of gaming routers

• Gaming routers are adept at prioritizing network tasks
• Mesh systems and range extenders are still needed for dead spots

What gaming routers do best, in essence, is provide a number of ways to prioritize and protect the connection handling your current gaming session. And, they do it better than other routers, thanks to that combination of performance and features. These are basically really good routers that happen to have some gamer-focused specialization.

That said, while some newer models such as the Asus TUF Gaming AX5400 (which still requires you to buy additional units) offer mesh-like features, many keep to the standard single-unit setup. That means that homes or offices with dead spots or range issues will still have those same problems when using a gaming router. If you insist on a gaming router and still have these coverage-related problems, you might need to add on a Wi-Fi extender.

Also, keep in mind that these are fantastic for organizing in-network traffic but can only do so much with the signal coming in. To take full advantage of one of these routers, you’ll want to make sure that the modem you’re using and the speeds that your provider offers aren’t also creating bottlenecks.

Should you get a gaming router?

• Online competitive gamers will appreciate what these routers have to offer
• Those that play mostly offline games should save their money

Not every gamer should invest in a gaming router. Their specialization means that they’re pricier than equivalent non-gaming options (not counting mesh routers, which are also heavy on the pocketbook).

Also, if you prefer single-player adventures, you’re paying quite a bit extra for some slightly improved download times. If your online gaming isn’t that intense, you probably won’t experience that much of a difference either.

Gamers that play anything online that’s remotely competitive are the ones who will benefit the most from a gaming router and see the biggest difference. After all, every part of a system needs to have the lowest latency possible and that includes that network connection. While a powerful non-gaming router may still work well, once another device starts doing something that requires a large amount of data, you need a network solution that can organize large downloads, streams, or whatever else is hogging the frequency band around your game instead of sharing it equally. And, nothing can do that quite like a gaming router.

As we get closer to the anticipated launch of Nvidia Lovelace later this year, more spec and benchmark leaks are starting to pop up online, and boy howdy, the flagship Lovelace card could be an absolute monster of a GPU.

The latest news comes courtesy of reliable Twitter hardware leaker kopite7kimi, who says that Nvidia is currently testing a variant of its AD102 GPU with an astonishing 18,176 CUDA cores, or nearly double the number of cores of the Nvidia RTX 3090 Ti, currently the best graphics card on the consumer market in terms of raw performance.

Kopite7kimi calls the card "the beast", and that is entirely appropriate if the leaked specs are anywhere close to what the actual card has at launch. In addition to the 18,176 CUDA cores, the AD-102-450-A1 GPU will also support 48GB of GDDR6X VRAM and could be rated for about 800W of TGP, which is in line with a lot of what we've heard about Lovelace's power-hungry design.

There are some other things that we could deduce from these specs. Assuming that there isn't some further revolutionary architectural change in the stream multiprocessors (SM), each SM should have 128 CUDA cores, which would mean that the card would have 142 SMs. If each SM only has a single ray tracing core, as is the case with Nvidia Ampere, that would also mean 142 ray tracing cores.

If Lovelace follows Ampere's lead and includes four tensor cores per SM, that would mean 568 tensor cores capable of accelerating the advanced FP32 operations required for machine learning and ray tracing features of the card. All in all, this is about a 70% increase over the RTX 3090 Ti, but the actual performance gains could be even greater than that.  

We can't say for certain whether this is an engineering sample of an RTX 4090 Ti or some other higher end variant of the next-gen BFGPU. A couple of months back, we got word of a Titan-class AD102 GPU with higher core counts and TGP, but this would likely be a workstation graphics card, assuming it was ever actually released. 

The expectation is that this new design is a consumer card straddling the border between a card found in the very best gaming PCs and professional workstations.

Analysis: Holy CUDA, this is going to be an absolutely massive graphics card

The RTX 3090 Ti is huge. It weighs several pounds, is over a foot long, and putting one in an upright tower case requires additional physical support so that the torque on the PCIe connector doesn't damage the board. It simply won't fit in a lot of cases on the market.

Let's say for argument's sake that this is an RTX 4090 Ti. Just figuring out how you're going to fit this kind of card in even a full tower case is a serious design challenge, because if this card is sucking up as much as 800W of energy, it's going to require an enormous cooling solution to keep it running safely.

What's more, the RTX 3090 Ti is already a $2,000+ graphics card. How much more expensive is this card going to be if this is the hardware it's packing? We need to take any leak with a grain of salt, if not a heaping spoonful, but everything is pointing towards a massive increase in the number of cores, VRAM, and power draw for the flagship Lovelace cards. 

We can't know for sure until there's some kind of official announcement, but if it's anything close to what these specs indicate, our concerns about the rising costs of graphics cards look downright quaint.

Intel’s Raptor Lake flagship CPU has seen some further leaked benchmarks, and unlike the previous spillage, this is a juicier affair because they’re actual gaming tests.

The game benchmarks come from the same source on Chinese video sharing website Bilibili who provided results for the likes of CineBench and Geekbench recently, as flagged up by @harukaze5719 on Twitter.

Again, this is an engineering sample of the Core i9-13900K processor which is compared to its predecessor the 12900K, but we have to bear firmly in mind that it’s a pre-release chip, and also be very cautious about any leaked benchmarking, anyway.

Bearing those caveats in mind, across a selection of games including some popular titles like Red Dead Redemption 2, Horizon Zero Dawn and PUBG, the 13900K turned out to be about 5% faster for average frame rates at 1080p resolution, around 7% quicker at 1440p, and 3% better at 4K resolution.

The rough upshot of this is around a 5% increase for gaming performance compared to the 12900K, which feels rather disappointing on the face of it. Plus power usage is considerably cranked up for this 13900K sample, too, at least for the provided peak consumption figures which see the Raptor Lake chip hit 150W to 160W, when the 12900K sits at more like 120W to 140W (the figures vary quite widely depending on the game being played, as you’d expect).

Analysis: Reasons to be cautious – and there are plenty of them

This isn’t the best look for the incoming 13900K, then, but we definitely shouldn’t jump to any conclusions here. Realistically, we’d expect more of an increase for a new generation of CPUs than this, and we’re likely to get exactly that for a number of reasons.

Firstly, as we’ve already mentioned, this is a (purported) pre-release sample, and the finished flagship processor will be faster anyway. For starters, in these tests, the 13900K is boosting to 5.5GHz, and the rumored boost speed for the flagship is more like 5.8GHz or thereabouts. And the Raptor Lake platform is obviously not fully realized yet either, with for example early BIOS versions for Raptor Lake motherboards that are still not ready for launch – these too are works in progress.

The power consumption figures also seem a bit off to us, likely pointing to some wonkiness around the engineering sample being run through its paces here. Some of the individual benchmark results themselves look a little odd, too, with some seriously hefty hikes in performance for minimum frame rates here and there.

All of which is to say we wouldn’t be running around panicking about Raptor Lake being a damp squib just yet. Remember that the mentioned earlier benchmarking looked pretty promising, particularly around multi-core performance, with the chip still in a pre-release incarnation at that.

Intel made good progress with 12th-gen Alder Lake processors, snaring many an accolade particularly in terms of gaming versus Ryzen, so even just a reasonable step forward with this Raptor Lake refresh will still mean AMD needs to pull something out of the bag with Ryzen 7000.

Fortunately for Team Red, all rumors point to the new Zen 4 chips being something special, so it’s likely that the next-gen CPU battle will be a close fought race; certainly in terms of performance. This new 13th-gen leak does also hint at AMD being more likely to get a firmer lead in terms of the power efficiency crown, mind, and that could be more important than usual given the state of power bills which are currently getting out of control for many folks, and set to worsen in the future.

Via Tom’s Hardware

AMD’s FSR 2.0 (FidelityFX Super Resolution) is still relatively fresh to the gaming scene, and hardly in any games at all yet – but modders have taken the situation into their own hands in some cases, when it comes to unofficially pushing support for the frame rate boosting tech.

We’ve already seen a mod for Cyberpunk 2077 from the amusingly-named modder PotatoOfDoom, which essentially leverages DLSS support to bring in FSR 2.0, and that workaround is now also live in mods for Red Dead Redemption 2 and Dying Light 2. The latter were worked on by other modders on Nexus Mods, but still based on PotatoOfDoom’s original mod.

As PC Gamer reports, other gamers are now trying to employ this technique with the likes of Metro Exodus: Enhanced Edition and Guardians of the Galaxy, and doubtless further DLSS-supporting titles to follow.

So how good are the results from these sort-of-thrown-together mod solutions? Well, they’re not ideal as you might anticipate, although a performance boost is certainly delivered, even if the upscaled image is noticeably softer compared to running in the native target resolution (and there are some minor issues with artifacts as well).

With full FSR 2.0 support, the difference in quality (to native resolution) is barely noticeable and closely rivals DLSS 2.0 in terms of upscaling chops.

Of course, we wouldn’t expect modders to be able to achieve the same results, and it sounds like what they have done here works well enough, and could certainly act as a stopgap while waiting for developers to get on the case with bringing in FSR 2.0 support.

Analysis: Patience is a virtue, but not when it comes to frame rates…

This is a measure of just how keen gamers are to get FSR 2.0 support for the titles they enjoy, while at the same time, it certainly highlights a degree of impatience about getting wider support for the Nvidia DLSS rival. So far, only three games support FSR 2.0 officially (Deathloop, God of War, and Farming Simulator 22), but more are promised to be inbound soon.

Now, there are some big hitters on the incoming list – like Hitman 3, Microsoft Flight Simulator, and EVE Online – but also a lot of relative gaming minnows in all honesty. So it’s not too surprising that gamers with modding chops are taking matters into their own hands to at least see the kind of results which are possible with workarounds.

It’s going to take time to build up a library of games that officially support FSR 2.0, of course – just look at how long it took Nvidia to push forward with DLSS support. Team Green, of course, has a big head-start in that respect, seeing as DLSS has been around for coming on four years now (believe it or not).

There are key advantages AMD has with FSR 2.0 being open source and freely available, with promises already having been made about how easy it is to implement for games that already support DLSS 2.0. In fact, AMD claims that for those titles, FSR 2.0 could be bolted on in just a matter of a few days (best-case scenario, but still, you get the general idea – it can happen very fast given the will). And these modding efforts appear to back up that assertion, what with them coming together pretty speedily.

Other fast routes to FSR 2.0 adoption include devs with games built on Unreal Engine (UE 4 or 5) that can use AMD’s plug-in, and so before too long, we’re hoping to see more support coming through quite quickly – in much faster fashion than the timeline witnessed with DLSS, for sure. That doesn’t mean Team Red hasn’t still got some considerable ground to make up, but it should be able to cover that turf with quite a turn of pace, if these early signs and promises are anything to go by.

AMD will also be pushing its other big advantages over DLSS, namely a wider coverage for graphics cards with FSR, including Nvidia models – as opposed to DLSS which requires an RTX GPU – and also the maintenance of FSR 1.0 as an admittedly lower quality solution than FSR 2.0, but one which allows for lesser-spec GPUs to still benefit from the frame rate boosting tech.

Back in April, AMD launched the Ryzen 7 5800X3D - a modified version of the solid 5800X CPU that came equipped with the manufacturer’s impressive new ‘3D Vertical Cache’ technology, also referred to as 3D V-cache. Now, fresh leaks are suggesting that AMD might be planning on bringing this technology to new CPUs in the Zen 3 line.

The V-cache chiplet design, in layman’s terms, allowed AMD to ‘stack’ cache memory on the processor die, creating additional L3 cache for boosted performance in gaming and server processes. The 5800X3D is a gaming powerhouse, but at $449 it’s out of reach for many looking to build their own gaming PC.

According to fresh info from well-known leaker Greymon55, though, AMD has plans to expand the X3D lineup before its next-generation Zen 4 processors arrive. Zen 4 (which will likely launch under the Ryzen 7000 brand) is expected to integrate 3D V-cache on all its chips, while the 5800X3D is currently the only Zen 3 chip to do so.

Although Greymon states that more information will come next month, speculation is already circulating that the new chips will be the 5600X3D and the 5900X3D (or even a uber-powerful 5950X3D), cornering the high end and low end of the market simultaneously. Since the regular 5600X is already a $199 beast for gaming, an updated version with X3D tech built-in could be an incredible option for PC gamers.

Opinion: AMD remains the best choice for gamers, and this could reinforce that

Yes, Intel’s current-gen Alder Lake CPUs are doing great, with its smart performance/efficiency core design proving to be highly effective. And yes, the Intel Core i9-12900K is essentially the most powerful consumer CPU you can buy right now. But AMD has positioned itself as the choice for gamers, with more affordable processors and technology aimed at improving performance in games.

Obviously, leaks like this should be taken with a hearty pinch of salt - although Greymon has been reliable in the past, nothing is set in stone until AMD confirms it. However, a possible Ryzen 5 5600X3D represents a boon for PC gamers building on a budget.

If you’re looking for the best processor for a gaming PC build (or need an upgrade to your existing Ryzen rig), then consider holding off until next month when new details on these X3D chips should surface. Greymon also suggests that new ‘low-end’ processors using the current AM4 motherboard socket could be arriving in the near future too, so budget gamers could have a lot to look forward to.

Sony is rumored to be readying a couple of gaming monitors aimed at the PS5, and a new leak has detailed some purported specs and pricing for the 27-inch screens, as well as a claim that these ‘InZone’ products will be launched today. We know for sure that something’s coming in a press event later today which as the tagline ‘Find Your New Zone’.

Chatter about the new InZone gaming brand from Sony sprang up around a week ago, with the word being spread that the initial products could consist of some gaming headsets, and the two aforementioned monitors which will be marketed as ‘perfect for PlayStation 5’ – but equally good for PC gaming purposes, as you might imagine.

The fresh spillage on Twitter reveals the supposed Sony InZone M3 and M9 monitors, along with the specs they might run with. 

Both are 27-inch models, with the M3 apparently being a 1080p screen with a 240Hz refresh rate, 1ms response time, and 400 nits brightness, along with 99% coverage of the sRGB color gamut.

The M9 ups the ante to 4K resolution with 600 nits of brightness, and 95% coverage of the DCI-P3 color gamut, offering a 144Hz refresh rate (with 4K / 120fps support), plus both will sport a pair of HDMI 2.1 ports with VRR (variable refresh rates) aiming to make them among the best PS5 monitors for gamers.

Those console gamers will also benefit from some PS5-targeted features (hence that ‘perfect for PS5’ labeling). These reportedly include Auto HDR Tone Mapping (to better show detailing in very dark or light areas), and Auto Genre Picture Mode to automatically optimize the screen’s settings based on the content being displayed.

As for the price in Europe, rumor has it that we’re looking at €800 (around $840, £690) for the M3, and a predictably beefier outlay of €1,300 (around $1,370, £1,120) for the M9 monitor.

Analysis: How might Sony fare in the monitor world?

As mentioned, these 27-inch monitors could be great screens for PC gamers, not just PS5 addicts, and indeed we can see that in the higher levels the refresh rates have been pushed to here (240Hz for the 1080p panel, and 144Hz with the 4K screen). Sony is also including an automatic KVM switch feature, which allows you to switch seamlessly between two different PCs at will (using the same monitor, mouse, and keyboard with either computer).

We’ll certainly be interested to see how monitors from Sony will stack up to the competition from existing and well-established gaming monitor brands. Obviously Sony is no stranger to quality panels when it comes to TVs, but monitors are something of a different kettle of pixels.

Regarding the extra features pitched at PS5 owners, we’ve seen these on some Bravia TVs already. Auto HDR Tone Mapping is a nifty boon, for sure, although all the automatic picture adjustment function does is simply switch between gaming and movie modes on Sony’s supporting TVs. (Might it be expanded to separate display modes for different game genres, perhaps?)

The good news is that we should get a glimpse of these monitors in action very shortly, and confirmation on the spec and pricing – assuming this rumor doesn’t go wonky at the last minute, of course.

Via Notebookcheck.net

AMD’s FSR 2.0 (FidelityFX Super Resolution) frame rate boosting feature that uses temporal upscaling is coming to Hitman 3 and a bunch of other games in the not too distant future.

Team Red just published a blog post revealing five new games that’ll support FSR 2.0, which include Hitman 3 as mentioned, as well as Abyss World, Rescue Party: Live, Super People, and The Callisto Protocol.

AMD also reminded us of 11 other games which are going to get FSR 2.0 in the future, and that includes Microsoft Flight Simulator, Eve Online, and Forspoken (when it comes out later in the year – also this will be the first game to support DirectStorage, notably).

The full list of 16 games due to support FSR 2.0 is as follows:

• Abyss World
• Asterigos
• Delysium
• EVE Online
• Forspoken
• Grounded
• Hitman 3
• Microsoft Flight Simulator
• NiShuiHan
• Overprime
• Perfect World Remake
• Rescue Party: Live
• Super People
• Swordsman Remake
• The Callisto Protocol
• Unknown 9: Awakening

FSR 2.0 uses temporal upscaling, which makes it a major step up from FSR 1.0 that used spatial upscaling. The difference is that with temporal, FSR is drawing data to work its magic not just from the current frame in-game, but previous frames as well – just as Nvidia DLSS does – and this makes for a better upscaled image in terms of quality.

Analysis: Impressive progress, and FSR 1.0 games are still coming

With those 16 games inbound, and three titles already boasting the upscaling tech, that makes 19 games in total which have brought, or are bringing, FSR 2.0 on-board.

In case you missed it, the launch game with FSR 2.0 was Deathloop, and since then, we have witnessed the tech come to God of War and Farming Simulator 22. AMD also provided some new benchmarks for the latter games, showing for example that God of War at 4K ‘ultra’ settings on an RX 6950 XT hit 66 frames per second (fps) by default. That increased considerably to 96 fps with FSR 2.0 in performance mode, and 83 fps with FSR 2.0 in quality mode (for the best upscaled image quality, which looks very close to native 4K).

It's good to see FSR 2.0 making commendable progress in raw numbers, even if it’s fair to say that a sizeable chunk of the games getting on-board are more obscure affairs, there are still some definite big names in the mix.

What’s interesting to note is that support is still being announced for FSR 1.0 with some games, with AMD observes that this is because the developers in question began working on incorporating the tech some time ago. Those games getting FSR 1.0 include Arma Reforger, Assassin’s Creed Valhalla, Dolmen, Hitman 3, iRacing, Sniper Elite 5, The Elder Scrolls Online, and V Rising.

So yes, Hitman 3 is getting both FSR 1.0 and 2.0 support, giving folks the choice, with FSR 1.0 being less of a strain on GPUs – for inferior results, of course – but that means those with a lesser graphics card which is below the hardware recommendations for FSR 2.0 still have an option to fall back on for frame rate boosting.

Via Tom’s Hardware

The latest roadmap for AMD’s products has been shown, revealing a whole host of planned hardware releases ranging from Zen 4 CPUs launching later this year to new APUs for datacenter use. Perhaps most exciting, though, is the news of new hardware slated for launch further down the line: RDNA 4 graphics cards and Zen 5 processors, both due for release in 2024.

RDNA 4 will (shockingly enough) follow the new RDNA 3 GPU architecture scheduled to launch later this year. That means we’ve just been given our very first glimpse of Radeon RX 8000-series gaming GPUs, codenamed ‘Navi 4x’.

We don’t have any concrete numbers on the RX 8000 graphics cards yet, but the flagship will likely be the RX 8900 XT and is presumed to employ a dual-node architecture using 3nm and 5nm processes. It appears that the Navi 4x family will exclusively use RDNA 4 architecture, while rumors suggest that Navi 3x (the RX 7000 series coming this year) may use RDNA 3 for the higher-end GPUs and a refreshed RDNA 2 for the cheaper ones.

Details on the Zen 5 CPUs are even thinner on the ground, beyond the 2024 release date and whatever we can extrapolate from what we know about the Zen 4 microarchitecture. AMD is promising a serious 35% performance increase from Zen 3 to Zen 4, so we can reasonably expect a significant boost in Zen 5.

Interestingly, AMD also mentions that Zen 5 includes ‘optimizations for AI and machine learning’, but doesn’t elaborate on what that means. With AMD’s FSR upscaling tech notably not using dedicated AI cores (unlike Nvidia’s DLSS and Intel’s XeSS), could we expect to see AMD diving into AI-powered hardware with these new CPUs? In any case, we’re excited to see more.

Analysis: Will RDNA 4 be worth waiting for?

We still don’t even have a concrete release date for AMD’s Radeon RX 7000 GPUs yes, which will be going up against both Nvidia’s Lovelace graphics cards and Intel’s Arc desktop GPUs later this year (assuming Intel manages to stick to their ever-slipping release plans). So it might feel silly to get pumped up about RX 8000 this soon.

But today’s news has given more details on the 7000-series cards, and the reveal that some of these lower-end Navi 3x GPUs will be running on the current RDNA 2 architecture (with a single GPU node) might prove to be a blow to anyone looking for a more affordable graphics card.

We’ll have to wait for actual performance figures, but this decision could lead to a significant rift in gaming performance between the higher-end RDNA 3 GPUs and the cheaper refreshed RDNA 2 models. If that’s the case, anyone looking to upgrade for gaming at 1080p or potentially even 1440p may want to wait until RX 8000 drops and leap straight from RDNA 2 to RDNA 4.

From WCCFTech

After a less than convincing start, cloud gaming has grown exponentially in the last few years. Services such as Nvidia GeForce Now, Google Stadia, Amazon Luna, PS Now, and Xbox Cloud Gaming can be seen as viable alternatives – depending on your internet speed – to traditional, local-based hardware. 

What’s more, cloud gaming enables devices, such as Chromebooks or smartphones, to run graphically intensive games that wouldn’t otherwise be possible. If you subscribe to the highest tier of Nvidia GeForce Now, you can get the performance of an RTX 3080 – a GPU that costs thousands of dollars and continues to be hard to find due to demand and the global semiconductor crisis – on a device that costs less than a few hundred. That means you can enjoy games at resolutions of up to 4K, and frame rates up to 120fps, should your device support it.

TechRadar Gaming spoke to Nvidia Geforce Now’s vice president and general manager Phil Eisler about the future of the company’s cloud streaming service, the challenges that remain, and whether or not cloud gaming can replace consoles and gaming PCs once and for all.

Invisible barriers 

Although cloud gaming can be considered to be more accessible than ever, challenges remain. Eisler outlined the main hurdles that Nvidia GeForce Now has had to overcome and is still tackling today.

“Initially, the challenge was just getting it to work reliably and to build a service that can scale to millions of members,” says Eisler. “We have a great engineering team that created an architecture that scaled to 30 global data centers and supports millions of gaming sessions from a catalog of over 1,000 games. 

“Next, we had challenges updating those games around the world. To alleviate that, we created an automated patching system that detects game patches and distributes them around the world in under an hour.

“And finally, there’s the challenge of providing a consistent smooth streaming experience with low latency to our members around the globe. The move from 30fps to 60fps and now to 120fps rendering and streaming solved the latency issue,” Eisler explains. “Nvidia engineering continues to make improvements to our streaming protocol to manage hitches in network connections. And, network connections keep getting better with more Fiber and DOCSIS internet service links, and less ADSL. The increased performance of home-based routers has helped, as well, from 2.4GHz to 5.0GHz and WiFi-6.”

In terms of the biggest bottleneck for cloud gaming right now, Eisler admits that latency is perhaps the most troublesome problem, but technological solutions and network improvements are on the way.

“I’m excited about future network improvements like Low Latency Low Loss Scalable Throughput (L4S) that we are working on with 5G network and Low Latency DOCSIS (LLD) network provider,” says Eisler. “These technologies will reduce cloud gaming latency even further, and make our level of service more consistent.

“The original GRID beta in 2013 used Kepler GPUs at 720p and 30fps. In 2015, we launched GeForce Now using Pascal GPUs at 1080p and 60fps,” Eisler continues. “And most recently, we introduced the Ampere generation SuperPOD (RTX 3080) that streams up to 1440p and 120fps. Nvidia will always continue to lead the way in cloud gaming technology and performance.”

A stop-gap for the GPU shortage? 

With the GPU shortage continuing to drag on and cards often selling for far above their recommended retail price, Nvidia GeForce Now has ultimately benefited according to Eisler. However, he stressed that Nvidia isn’t ignoring supply issues, despite the service’s increase in popularity.

“There has been an uptick in subscribers," says Eisler. "Nvidia continues to work hard to improve the supply situation. In the meantime, playing on GeForce Now can be a great solution for gamers waiting for a new card to arrive.”

However, Nvidia GeForce Now isn’t a replacement for PS5 or a high-end gaming PC.

“We are not there yet, but cloud gaming is a great alternative for those who can’t afford a new console or gaming PC or just want to play their games on other devices,” Eisler admits. “Chromebooks and phones are now the fastest-growing client types on GeForce Now, and this segment is quickly catching up to integrated graphics laptops and Macs.

“A GeForce gaming PC is always going to be the best machine for gamers by offering the lowest possible latency and highest resolution. So far, GeForce Now has been extending the GeForce gaming experience to new machines and gamers: 80% of our sessions are hosted on machines that could not play the game locally.

“Cloud gaming is an opportunity to expand the gaming market by instantly making one billion underpowered PCs and other devices, game-ready. It’s an opportunity to introduce hundreds of millions of people to PC gaming.”

Competition in the cloud 

When it comes to the novelty of GeForce Now, Eisler says that even though there are other competitors in the cloud gaming space like Xbox Cloud Gaming and Luna, it’s the gamers that Nvidia is listening to and learning from. 

“We spend more time learning from gamers. PC gamers have told us that they want to be able to play free-to-play games, or the games that they already own on Steam, Epic, Uplay, and Origin game stores,” says Eisler. “And they want to play on all of their devices using GeForce rigs in the cloud. This simple formula has connected with gamers and set GeForce Now apart from other cloud gaming services.”

Eisler also says that Nvidia has plans to bring Nvidia GeForce Now to more devices, and didn't rule out the Nintendo Switch if there’s demand. 

“Yes, based on member requests, we started supporting smart TVs beginning with the latest LG TV models in December 2021. At CES in January, we announced that we’re collaborating with Samsung to bring GeForce Now to its Smart TVs by adding the service to the Samsung Gaming Hub. We continue to evaluate other devices that our members are interested in.”

GeForce Now could also come to VR headsets in the future, according to Eisler. 

“Nvidia already has CloudXR technology that enterprises use today to stream VR from the cloud,” says Eisler. “Now that the latest consumer VR headsets have become untethered and GeForce Now has the ability to render and stream at 120fps, I believe that we have the building blocks in place where cloud VR will make sense in the future.”

Getting games approved for GeForce Now

Like Xbox Cloud Gaming, not every game is available on GeForce Now. Eisler shared some insight into the process of getting games onto the service.

"GeForce Now is a great platform for publishers to distribute their games because we use standard PC builds that don’t require any extra work by publishers," Eisler explains. "It’s the only platform that doesn’t charge publishers any revenue share. 

"Game publishers have access to a rapidly growing market of cloud gamers they couldn’t reach previously who can now play AAA games on older integrated laptops, macOS computers, Chromebooks, and more devices. We are also opening up new regions with more access to GeForce PCs via the cloud such as Turkey and Brazil. 

“The result is millions of games being purchased by gamers – from the stores publishers offer their games on – to play on GeForce Now. That continues to entice more publishers to offer their games on GeForce Now. Our member response has been terrific, with many encouraging publishers to list their games. Over time, we’ve grown a massive catalog of playable games and we expect to see even more great games on GeForce Now in the future.”

But can we expect any exclusive games to hit Nvidia GeForce Now in the future? Not according to Eisler.

"Our goal is to be an open platform that supports all PC games, and therefore, none of our publishing deals are exclusive," says Eisler. "Today, many popular Free-to-Play games are only available in the cloud on GeForce Now, like Fortnite, League of Legends and Apex Legends. However, this is because we focus on bringing the best PC games to the cloud, and not because of any exclusive deals."

The impact of 5G

One of the biggest boons to cloud gaming is the rollout of 5G. 5G can offer 10 times the speed of 4G and support more simultaneous connections. It's something that will take cloud gaming out of the home and become accessible almost anywhere. 

"I’m very excited about the prospects of 5G and GeForce Now. 5G network service provides excellent bandwidth and low latency, which we have been testing with our Korean partner LGU+ for two years. In Seoul, you can even play GeForce Now on the subway, and it works great!

"In America, we have a partnership with AT&T to bundle 6-months of GeForce Now with 5G phones that really shows the benefits of a 5G phone.

"We’ve also been working to improve the native touch-to-cloud experience on GeForce Now. Earlier this year we started a beta test with Epic to add native touch controls to Fortnite on our GeForce Now cloud version. The results are fantastic with a very intuitive way to enjoy cloud games on your 5G mobile phone with touchscreen inputs. We plan to rinse and repeat with more games."

Unlimited possibilities

So looking into the future, what are the biggest improvements we can expect to see in cloud gaming in the coming years?

“More Devices. More Games. More consistent quality of service. More links to launch cloud gaming from anywhere,” says Eisler. “Games that scale better from 6” to 60” screens. Cloud-native games that only run in the cloud and are designed to leverage the full power of the SuperPODs that we’re building. The future of cloud gaming is only limited by our imagination.”

• Cloud gaming: which service is right for you?

Dolmen, a sci-fi action RPG set to emerge in just a few days on May 20, won’t be launching with Intel’s XeSS tech on-board – that’s Team Blue’s DLSS rival – with the developer having clarified this.

Originally, the developer of Dolmen was questioned by Wccftech in an interview regarding whether the RPG would have XeSS support at launch, and (as flagged by Tom’s Hardware) said: “I think I can tell you that everything, including the partnership with Intel, will be released in the Day One patch [for Dolmen].”

Wccftech subsequently updated its interview to say that Koch Media PR had been in touch to advise that in fact XeSS support won’t come until the summer. Also, as VideoCardz picked up, the developer just tweeted to confirm a summer release date for XeSS in Dolmen.

Analysis: Did we really think XeSS was imminent, when desktop Arc wasn’t?

This is a bit of an odd one, really, in that we wouldn’t have expected Intel to be pushing out XeSS as soon as this week. Remember, there are no Arc Alchemist desktop GPUs out there, or even nearing launch – these graphics cards won’t come out until probably late June, and that’ll be only in China to begin with (we may not see discrete Arc desktop GPUs until later in August in Europe and the US).

On top of that, even Alchemist laptop GPUs aren’t present outside of Korea yet, so what would be the point of launching XeSS to just a tiny user base – potentially frustrating the folks waiting for Intel’s Arc graphics cards even more (“we’ve released something else – that you can’t benefit from”).

Dolmen – which was a game Intel used to showcase its frame rate booster back at the Arc laptop GPU launch – should get XeSS in maybe July or August, then, and it would certainly make sense to hold off the launch of this tech until Arc cards are more widely available across the globe (and desktop models, too). The game does also support Nvidia DLSS and AMD’s FSR (just v1.0 in the latter case, though FSR 2.0 support is apparently coming).

The (effective) delay of Arc desktop graphics cards beyond Q2 (at least outside of Asia) is, of course, already bad news for Intel, with AMD and Nvidia potentially having next-gen GPUs ready for September – and Nvidia might have some kind of a launch planned as early as July, going by a fresh rumor. And with FSR 2.0 now having emerged – looking like a big step forward for Team Red – well before XeSS hits the streets, that’s just another area in which Intel is now falling behind in its bid to make the best impression when Alchemist desktop debuts.

In short, it’s starting to seem like things are really not going to plan for Intel, and that puts us in a frame of mind whereby further delays would not be surprising.

• Who’s top dog right now in the GPU world: AMD or Nvidia?

It took almost an entire year from their initial release, but the LHR (low hashrate) versions of Nvidia's RTX 30 series graphics cards have finally been completely unlocked by a mining software called Nicehash, restoring each card's respective mining capabilities.

Many of the best graphics cards were nearly impossible to find available to buy over the last two years because of the global chip shortage, a broader supply chain crisis at ports around the world and demand for consumer tech putting even more pressure on the availability of semiconductors needed by AMD and Nvidia for their products.

Part of this demand also included cryptominers who were buying up the available stock in bulk of popular GPUs during the height of the recent crypto currency boom. And while there are mixed opinions about how this affected overall availability for gamers and building hobbyists, there was certainly no love lost between the two groups which resulted in Team Green putting measures in place to make its consumer graphics cards less desirable to those hoping to use them to mine currencies such as Ethereum.

Almost the entire RTX 3000-series of cards was re-released with hash rate limiters (not including the GeForce RTX 3090 as it was deemed too expensive to attract the mining market) which limited their ability to mine currencies like ETH, but several of these were quickly defeated, with some software able to unlock cards to 70-80% efficiency in a matter of weeks.

As reported by WCCFTech, NiceHash has now announced that it can utilize 100% of available mining speeds using these previously restricted Ampere-series cards, stating on its website that:

"We are very excited to tell you that NiceHash QuickMiner (Excavator) is the first mining software to FULLY (100%) UNLOCK LHR cards! Now you can earn more profits than any other mining software on the market if you are using LHR graphics cards with NiceHash QuickMiner".

We don't have an LHR model GPU to verify this claim ourselves, so take the statement with a pinch of salt, but its also worth noting that GPU stock levels and pricing is currently sitting at the best numbers we've seen for many months, so it's unlikely you'll be fighting off swarms of competition if cards are viable for mining again.

Analysis: Don't worry about it – no, seriously

This news is unlikely to cause another shortage of graphics cards or even drive up the price of the stock currently available, and that's for a few different reasons. The main piece of good news for gamers is that Ethereum, the world's second most popular cryptocurrency after Bitcoin, started moving to proof-of-stake at the end of 2021.

Let me explain. Proof-of-work is a system where computers compete against each other to be the first to solve complex puzzles, which left it open to miners using warehouses full of consumer graphics cards to solve those puzzles, earning some ETH in the process. Proof-of-stake instead uses validators to find a block based on the number of tokens they hold, removing the need for those 'puzzles' to be solved.

This switch from a proof-of-work to a proof-of-stake model has removed a lot of the demand of the graphics cards, so where Ethereum was previously the largest cryptocurrency that is primarily mined using graphics cards (Bitcoin has mostly been mined on ASICs over the last few years), it will soon be inefficient to do so.

Sure, there are other currencies available that still use a Proof-of-work system, but these don't have the same grasp on the market that ETH did last year, so unless another crypto starts to rocket upwards in value, it'll take some time to see a similar situation unfold.

In reality, this shouldn't impact consumers who are trying to buy a new graphics card in any way, though it does mean that LHR versions of GPUs no longer serve the same purpose. With any luck, Nvidia continues to develop its hashrate limiting technology just in case we see a similar scenario unfold when it releases its Lovelace series of graphics cards to act as a deterrent.

• If you don't want to build a PC, buy a gaming laptop instead for gaming on the go

PC components could soon see a much better situation in terms of supply meeting demand according to a new report – but a few areas could still remain troublesome to an extent, including graphics cards.

As spotted by Tom’s Hardware, Counterpoint Research recently published stats on global PC shipments (which fell 4.3% year-on-year in Q1 2022), complete with an observation that the component shortage is very much set to ease in the second half of this year.

The Counterpoint report states: “Since late 2021, demand-supply gaps have been narrowing, signaling an approaching end to supply tightness across the broader ecosystem. Among all PCs and laptops, the supply gap for the most important components such as power management ICs, Wi-Fi and I/O interface IC has narrowed.”

In fact, as we head towards H2 2022, just four areas of the wider PC component market will remain short on supply, and even those will only be around 10% adrift of meeting demand going by Counterpoint’s estimates (whereas in 2021, supply was falling short of demand by up to 20% or even 30% in some areas, causing big problems as we’ve seen most obviously with graphics cards last year).

Sadly, one of those troublesome areas that could remain a thorny prospect which the report highlights is GPUs, with supply still not predicted to meet demand later in 2022. Other problem areas will be Power Management Integrated Circuit for LCDs (monitors), as well as some motherboard components and Wi-Fi chips.

Analysis: GPU situation still looks promising enough to us

The major area of concern for the PC ecosystem here, then, is that graphics cards are not flagged for a full recovery this year. But then, a full return to normal is one thing, and a perhaps marginal (sub-10%) supply deficit – maybe within certain categories or price brackets of GPU – should still allow pricing to very much normalize, by and large, and take the wind out of the sails (and sales) of scalpers.

We’d also caution against reading too much into this guestimate of GPU availability anyway, as the report is a very generalized prediction here – a broad overview of the upcoming state of the entire component market – and we’ve already seen quite a few signals elsewhere that the graphics card market is making something of a recovery already, well before H2 of 2022 kicks in. Those can’t be ignored.

Also, it’s the second half of the year which both AMD and Nvidia have openly stated that we should see much better GPU stock levels coming through, plus we also shouldn’t forget that Intel is coming to the discrete graphics card market as well with its Arc desktop offerings (within the next couple of months in theory). That in itself will cause an uptick in stock and should stoke competitivity with pricing, too (though the extent of the impact on price tags could depend on exactly what tack Intel takes, and how much production Team Blue can muster).

In our books, things look optimistic enough for GPUs to soon be mostly out of the woods we’ve been stuck in for far too long now – not that there aren’t potential dangers ahead for all PC components when we look at the situation in China of late, with more supply chain disruption being caused by Covid lockdowns.

Equally, there are other forces to consider on the demand side of the equation in terms of inflation and possible drops in spending, with weakness in consumer demand already observed by Counterpoint in those falling PC shipment levels, and for example Chromebooks losing sales momentum. So, to some extent, the supply and demand seesaw will also be righting itself with a lightening of the latter load…

• You might want to wait for next-gen GPUs like Nvidia’s RTX 4000 range which could come sooner than you think

Shadow, the cloud gaming platform which does things rather differently – as it isn’t just a straightforward streaming service, but rather gives you a full gaming PC in the cloud – has announced a major relaunch which is coming next month.

The company – which was acquired by OVHcloud founder Octave Klaba back in May 2021 – revealed that a ‘Shadow Spotlight’ keynote will reveal its relaunch plans and “strategic vision for the coming years”. What’s more, it’ll “deliver on promises made to its community”, the firm wrote in a teaser press release for the event which will happen on May 18.

There’ll be a bunch of new announcements about the cloud gaming service, and plenty to get excited about, apparently. Indeed, we’re told that: “2022 will be a massive year for Shadow and we cannot wait to update our community with everything we have for them!”

The launch event kicks off at 8am Pacific Time in the US (4pm UK time) on May 18 as mentioned, on Shadow’s YouTube and Twitch channels (plus Twitter).

Analysis: What’s lurking in the shadows?

The strength of Shadow is, of course, the fact that you get what’s essentially an entire remote PC, not just a streaming service packaged for specific games or products. In other words, you can use that virtual Windows 10 PC to do whatever you want (even upgrade it to Windows 11, as we’ve seen, though Microsoft’s latest OS still isn’t officially supported on a Shadow PC installation).

Exactly what changes are coming, we can’t be sure – maybe Windows 11 support for one thing – as the info in this initial teaser is pretty thin on the ground. Although the press bumph does mention creatives and business users, so that could point to different tiers of subscription aimed at those folks, perhaps.

What’s also welcome is a mention in the press material of making the service more “accessible and affordable”, which can only be a good thing, assuming, naturally, that this isn’t just general marketing blather.

Better stability is another vague promise, too, and we can guess at some pepped-up specs to boot, as the current GPU in the Shadow PC, a GTX 1080 equivalent, doesn’t exactly feel cutting-edge for contemporary gaming (remember, Nvidia now offers an RTX 3080 tier with its GeForce Now streaming service).

Whatever the relaunch brings, we’ll find out soon enough, and we’re certainly keen to see how Shadow reshapes its cloud PC offering to better take on rival gaming services like the aforementioned GeForce Now or indeed Google Stadia.

• Don’t know which cloud gaming service to pick? Here’s all you need to know

AMD’s Radeon RX 6750 XT is one of the rumored refreshed graphics cards the company is supposedly readying for the near future, and we’ve just spotted a leaked benchmark featuring the GPU, which is disappointing on the face of it – but with some clear caveats.

This spillage comes from Komachi, a well-known hardware leaker on Twitter, who spotted the RX 6750 XT being tested in GFXBench, as reported by VideoCardz.

The precise benchmark – and the only test run here, sadly – is ‘Aztec Ruins High Tier’ in which the RX 6750 XT purportedly hits 366.5 frames per second (fps), which is barely any faster than the existing 6700 XT that this GPU supersedes.

In fact, the 6700 XT is shown for comparison running at 362.2 fps, so the 6750 XT is only around 1.2% quicker in this particular test – not a huge difference by any means. Apply salt as ever at this point.

Analysis: Plenty of caution is required here

As always with an individual leak – aside from the given skepticism around how genuine it might be – we need to remember that this is a very limited picture of the potential abilities of a pre-release GPU. In this case, it’s a more obscure benchmark as well – not one of the commonly seen metrics used to gauge graphics card performance – and of course the 6750 XT is still in testing, anyway, so this won’t represent its full performance level.

The real-world gaming performance of the finished product will likely see the refreshed GPU distance itself to a greater extent from the 6700 XT, but then again, this leak suggests that maybe the difference won’t be all that much (mind you, we wouldn’t expect it to be a big gulf anyhow – this is just a refresh, after all).

Going by the grapevine, we’ve heard that the 6750 XT – and the other 6x50 XT refreshes, with three expected in total – will likely employ faster VRAM (juiced up to 18Gbps) compared to the 6700 XT, and AMD is supposedly looking to up clock speeds as well. Maybe any clock speed gains might be minimal, though, if this benchmark is any reliable hint of what’s to come (remember that the leak regarding considerably faster clocks was about the flagship replacement, the 6950 XT, so perhaps further down the range we won’t see such a jump).

All of this is still speculation, naturally, so we just have to sit and wait at this point, but we shouldn’t have to be patient for long to see the true specs of these rumored models. The theory is that the refreshed GPUs will land around mid-May, and a recent regulatory filing for the 6750 XT lends further weight to the idea that these products are coming sooner rather than later.

As indeed does the simple fact that AMD needs to get them out of the door pretty sharpish, because RDNA 3 graphics cards are coming later in 2022, so Team Red will need to leave enough distance between these refreshes and the next-generation GPUs, otherwise people are simply going to wait for the latter to arrive (RDNA 3 will be a huge step up, if the rumor mill is right).

Another noteworthy possibility is that with the arrival of the 6750 XT, we could see the 6700 XT getting discounted – and the same might be true of all these refreshes, of course. Indeed, prices are already falling and coming back to normal in the GPU world as it is, so this could be another downward pricing pressure for those looking to pick up an affordable AMD card; fingers crossed.

• With RTX 3060 prices falling fast, a reasonable 1080p gaming PC is finally possible

There are more options than ever before with the best 4K curved monitors on the market, and we've reviewed many different models which have been hand-picked below. Our list looks at budget, mid-range, and premium offerings from established manufacturers such as Dell, Samsung, LG, AOC, and others to give you an encompassing picture of what's available and worth your money in 2024. 

What's exciting about the best 4K curved monitors in 2024 is the fact that they have become substantially cheaper and more available as the 2160p resolution has become standardized. Previously, you were paying a high price in comparison to 1080p and 1440p alternatives, but now things are far more level, even when weighing up with ultrawide and dedicated gaming panels. For more offerings, I recommend checking out our full list of the best gaming monitors. 

A curved form factor can be a better fit for desk work and gaming over a more standard straight model. It also means enhanced viewing angles when sat up close, particularly when paired with an aggressive curve. For reference, the lower the number the more intense the curvature. Traditionally, this ranges anywhere from 1000R (a deep curve) to 4000R which is more subtle. 

Then there's the size of 4K curved displays to consider. You'll find a range of sizes between 32 inches and 40 inches available in this buying guide to give you ample range which can benefit from having a curvature. For even more options, I recommend checking out the best 4K monitors and the best monitors overall to see what else is out there. 

The Quick List

Best overall 4K curved montior

The Dell S3221QS’s sleek and classy design, just the type of look that can fit anywhere, is only surpassed by the beautiful visuals it produces on its 3840 x 2160 60Hz panel. The image quality this 4K curved monitor delivers is sharp and vibrant, made better by its 300-nit brightness and 3000:1 contrast ratio. 

The color gamuts – 99% sRGB and 90% DCI-P3 color – aren’t wide enough for content creators, but that doesn’t stop it from producing colors vivid enough for media consumption. 

And, of course, that 1800R curvature offers deep immersion and a wider field of view. Its sub-$400/£400 price tag rounds it all out nicely, offering an attractive option for those looking for a display that sees them through most tasks, including narrative-driven games.

Read our full Dell S3221QS review

Deepest curvature on a 4K monitor

Curved monitors have varying curve ratings that typically range from 1800R to 4000R. But, then there are those special ones like the Samsung UR59C, whose 1500R curvature on a 32-inch panel means you’ll get a higher level of immersion without feeling engulfed by the screen. Its higher contrast ratio of 2,500:1 and color coverage are decent, which means you’re getting great dynamic range and vivid colors. 

Plus, features like Game Mode and Picture-by-Picture mean you can switch from productivity to gaming in a snap – as long as you don’t need a higher refresh rate for your games. But it’s worth keeping in mind that you’re paying for the curvature here, as features like swivel and height adjustability are non-existent and USB ports are missing.

Best ultrawide 4K curved monitor

There are two things to bear in mind before jumping on the Dell UltraSharp U4021QW. One, it’s not exactly 4K. It’s specifically a 5K2K resolution panel, but this is as close as you’re going to get to an ultrawide 4K. Two, this productivity display is more expensive than many kitted-out gaming monitors out there. 

However, if you can afford it, it might just be well worth it with its plethora of ports that include Thunderbolt 3 and RJ45 ports and massive screen real estate that allows for the ultimate multitasking. And that 2500R curvature is just enough to offer a deeper immersion on an ultrawide display. Last but hardly least, it comes with 100% sRGB, 100% Rec. 709 and 98% DCI-P3, which makes it a great 4K curved monitor for content creators.

Read our full Dell UltraSharp U4021QW review

Best curved 4K monitor for gaming

Who says you can’t have high-resolution gaming resolution while satisfying your need for speed too? While most 4K panels are still at 60Hz, a newer class of 4K 144Hz models is on the rise so that gamers can have the best of both worlds. 

The LG UltraGear 38GN950 might not be exactly 4K, sacrificing some pixel space for that ultrawide look at 3840 x 1600, but it boasts superior picture quality, great features for gaming and elevated specs like 450-nit brightness and VESA DisplayHDR 600. Better than other 4K 144Hz monitors, it comes with a 2300R curvature and an ultrawide design that delivers deeper immersion and more screen real estate.

Read our full LG UltraGear 38GN950 review

Best budget 4K monitor

Also the least expensive option here, which means it’s the best value as long as you don’t need the extra frills, the AOC CU32V3 is best for general use if you’re looking for a 4K curved monitor for your home. 

This 32-inch panel comes with that deep 1500R curve as well as a thin-bezel design. Of course, this being an AOC monitor, it touts an imaginative modern design as well, for better or worse. It’s a bit stripped down on features. Even its ports are limited. But most users don’t need those, and if you’re such a user, you’ll get a lot out of its sub-$400 price.

Best 4K curved monitors - FAQs

How I tested the best 4K curved monitors

Here at TechRadar, we endeavour to review as many products as possible before recommending them here in our best lists. That's the case with the best 4K curved monitors as well. We've gone hands-on with many of the models featured here; testing their performance with media playback and gaming where possible. 

You can find out more about how we test, review and rate products here. The cliff notes version is this: if a company makes claims about refresh rates, peak brightness, and the port selection then we will verify to ensure it's accurate. We're frequently testing and scoring more screens, so we'll be expanding on this list as more offerings hit the scene. 

AMD’s RX 6950 XT, the rumored refresh of its flagship GPU, might come with a seriously weighty price tag if a fresh leak is anything near the right ballpark.

There are some caveats here, as you might expect – and more than normal around a pricing leak, we’ll come back to that – but on the face of it, the RX 6950 XT has seen pre-release product listings at a couple of Australian retailers, Megabuy and ITnest (as PC Gamer spotted).

The damage that the 6950 XT will supposedly do to your wallet is a hefty AU$3,240 (around $2,400, £1,800) at Megabuy – ouch indeed.

Now, note that the product listing actually says the graphics card is a ‘Gigabyte Radeon RX 6900 XT Gaming OC’, which is the existing flagship, but the product code clearly states ‘GV-R695XTGAMING-OC-16GD’, referring to a 6950 XT model.

It’s likely that someone has made a mistake typing in the full name – or used a template and not changed that bit – although of course this could indicate the entire listing is a mistake. We can’t rule that out, anyway.

However, note that there is an existing listing for the Gigabyte 6900 XT Gaming OC, product code ‘GV-R69XTGAMING OC-16GD’ – so R69, not R695 – and this is priced much lower at AU$1,949 (around $1,450, £1,100).

Furthermore, as VideoCardz picked up, Gigabyte has recently registered the GV-R695XTGAMING OC-16GD product code with the EEC (Eurasian Economic Commission), which is another tick in the box for its likely authenticity – though none of this represents anything like a cast-iron guarantee, of course.

Analysis: Performance expectations will be high if there’s anything to this leak

If true, this price tag would represent a massive leap compared to the cost of the 6900 XT – a hike of 66% on the current flagship going by the pricing observed above at Megabuy. Now obviously, we’ve got to be careful around any leak like this, as the price given could merely be a placeholder – in other words, a guess.

However, if it’s true, or close to the mark, you can be sure that the performance increase which is delivered by the 6950 XT over the current flagship will obviously be far more modest than the level of hike seen with the pricing.

Although if AMD is setting the price at around this kind of level, we should expect a good extra dollop of frame rates from this GPU refresh (remember that some really fast boost speeds are expected from the 6950 XT going by the grapevine).

It’s also worth bearing in mind that the Megabuy price for Gigabyte’s RTX 3090 Ti cards is only a few percent higher than this purported 6950 XT price tag, so the expectation would certainly be that it competes with Nvidia’s new flagship. If this pricing is correct, of course…

We shall see, and interestingly the order delivery date is pegged at 4+ weeks, which coincides roughly with the latest rumored launch date of mid-May for the RX 6950 XT (and the other two refreshes supposedly due, namely the 6750 XT and 6650 XT).

The evidence is really mounting up regarding this trio of RDNA 2 refreshes, and we’ve even been treated to pics of them – with the graphics cards being smart-looking black editions.

It seems more and more likely that this will be the next step AMD takes with its GPU launch schedule, before debuting a whole new RDNA 3 range later in 2022 (and if the 6950 XT price is high, it’ll surely be even more tempting to be patient and wait for the next-gen flagship for those who want a powerhouse card).

• Rumor has it that AMD’s next-gen GPUs could get the better of Nvidia in one major way

AMD’s Radeon RX 6750 XT is inbound according to an official filing that’s been spotted for the rumored graphics card.

Tom’s Hardware reports that a PowerColor RX 6750 XT Devil D6 graphics card has been filed with the South Korean National Radio Research Agency (at the end of March), as flagged up on Twitter by well-known leaker @harukaze5719.

According to the description, the GPU comes equipped with 12GB of GDDR6 VRAM, although no other details are imparted (though they aren’t normally with this kind of leak – you only get the product name and codes).

The appearance of such a filing usually indicates that the launch of a product isn’t too far away, but then these kinds of things can be done speculatively by manufacturers, plus we must treat any leak with a healthy dollop of caution anyway.

What makes this seem more likely is that we have heard a fair few rumors around AMD refreshing its GPUs with 6x50 models, and the 6750 XT has been speculated about before. So there is something of a weight of evidence which has now built up around this purported card, and the others which are expected to arrive – namely the 6950 XT and 6650 XT. Indeed, we’ve recently seen leaked pics of all three models (although they could be fakes, it doesn’t look like that’s the case).

Analysis: An imminent launch for trio of RDNA 2 refreshes?

So what else do we already know about these rumored refreshed RDNA 2 graphics cards? The latest gossip contends that they will be black editions – meaning that they’ll follow the design of the RX 6800 XT Midnight Black model, a smart-looking card – and they could arrive in mid-May.

Originally the grapevine believed the 6950 XT, 6750 XT and 6650 XT were due to land in April, but that launch timeframe has supposedly been pushed back (apply sufficient condiments with all this, naturally). Hopefully this fresh filing is an indication that things are on track for an imminent launch – it certainly makes sense that AMD needs to get these GPUs out quickly if they are coming, because RDNA 3 isn’t that far off (with next-gen models debuting later in 2022).

Of course, even now we’re wondering exactly what the point of these refreshes is, seeing as many folks will want to wait for the next-gen (AMD or Nvidia) at this stage of the game – especially as they’re not likely to be a hugely compelling upgrade.

We don’t know much about the spec bump that these graphics cards could offer over and above the respective 6900 XT, 6700 XT and 6600 XT, but the theory is that they could use faster VRAM (running at 18Gbps), and it’d make sense that clock speeds might be further boosted.

Indeed, there’s been talk of the RX 6950 XT boosting to 2.5GHz out of the box, or maybe even faster, with power usage driven proportionally higher. AMD no doubt will be looking at Nvidia’s new RTX 3090 Ti and the performance uptick that offers (albeit with a seriously heavy power draw, although the supercharged Ampere flagship can still beat the RX 6900 XT even when its power consumption is ramped down in a big way, interestingly).

• Bargain hunters note that AMD may soon launch its cheapest ever RDNA 2 GPU

AMD’s next-gen RX 7700 XT, which will be a mid-range RDNA 3 offering, is rumored to be equally as powerful as the current 6900 XT (RDNA 2) flagship – or even better performing in some respects.

This is the latest from Moore’s Law is Dead (MLID), a YouTube leaker who often chimes in on the subject of GPUs and CPUs, but as ever with anything from the rumor mill, arm yourself with a boat-load of skepticism.

So, the theory as outlined by MLID is that the mid-range RDNA 3 graphics card, presumably set to be called the RX 7700 XT (the follow-on to the 6700, of course), will match the performance of the 6900 XT in rasterization, meaning traditional rendering as opposed to ray tracing. Indeed, performance could be slightly faster in 1080p (and a little slower in 4K resolution), but likely the same ballpark for 1440p which is what the 7700 XT will be pitched at.

The further assertion is that when it comes to ray tracing, the RX 7700 XT – which is built around the Navi 33 GPU – will outdo the 6900 XT, although MLID is a bit more vague about how much in the way of gains we can expect here.

MLID reminds us that the 7700 XT is set to be a 6nm monolithc GPU (meaning it’s a single chip, rather than chiplets – the latter is supposedly the route AMD will take with the higher-end RDNA 3 GPUs, meaning Navi 31 and 32).

The 7700 XT will see speedier clocks than Team Red’s current products, and better power-efficiency – indeed that’ll be a hallmark of the entire RDNA 3 family, as you might guess. The GPU will supposedly run with 8GB of GDDR6 video memory, with a 128-bit memory bus, so cut down in other words (the 6700 XT is 192-bit), but the faster VRAM will make up for that, MLID reckons.

Analysis: Mid-range – or even lower mid-range – killer GPU for AMD?

MLID notes that the design decisions purportedly made with this GPU, including limiting to 8GB of video RAM, show that this is clearly a mid-range, or even lower mid-range, graphics card.

The leaker estimates that the price could be as low as $400, maybe $500 in the US, but MLID makes it clear enough that this graphics card could well represent quite the bargain if it delivers the rumored performance which is essentially equivalent to the 6900 XT (and maybe surpassing the existing flagship for ray tracing, and for 1080p gaming).

The icing on top for would-be next-gen AMD buyers is power-efficiency, with the 7700 XT delivering these kind of frame rates on a power budget of around 200W (MLID claims power usage will be about 180W to 230W, so it could even be under 200W).

Compare that to the talk of Nvidia’s Lovelace (presumably RTX 4000) graphics cards being seriously power-hungry – like the RTX 4080 maybe chugging 450W for example – and things look bad for Team Green in that respect. Particularly if the top-end next-gen Nvidia GPUs are going to force some gamers to look at a PSU upgrade (not just an added expense, but a serious additional complication for the less tech-savvy out there).

All that said, while there’s a lot of big talk here about how RDNA 3 is shaping up to deliver on both performance and efficiency fronts, we’d do well to remember these are just rumors. Time will tell if they pan out, with both AMD and Nvidia set to spring their fresh GPUs on the market later in 2022 (Nvidia may beat Team Red to the punch for launch timeframe, though, going by the grapevine).

• Worried about needing a beefier PSU? Here are all the best options for an upgrade

Via Wccftech

Graphics card prices are likely to come “crashing down” from their inflated price tags in the future, according to a new report from an analyst.

Jon Peddie, who heads up Jon Peddie Research (JPR), shared his latest thoughts on the graphics market, and in particular desktop cards for PCs, in an article for Graphic Speak (spotted by Tom’s Hardware).

The report includes an interesting analysis of the average selling price of AIBs (meaning add-in boards, a fancy term for PC graphics cards), complete with a graph which shows pricing has been rising gradually since 2014, but really shot up during 2020 and 2021.

In the US from 2014 through to 2019, we saw the desktop GPU go from an average price tag of about $270 to around $440, but in 2020 the average selling price leapt to over $650, and further to about $780 in 2021. Which is, of course, ridiculous, but we all know the reasons why, and Peddie goes on to underline them.

Peddie observes: “PC gaming and mining AIBs have increased in price by at least 2x, maybe 3x, over PC notebook GPUs. So, supply shortage has to be ruled out as the reason for the surge in PC AIB pricing. That leaves miners, speculators, and gougers. This is no joke.”

He continues: “Who is benefitting? The channel organizations like Amazon, NewEgg, BestBuy, and others, while the speculators sell AIBs on eBay at prices 2x to 3x more than the manufacturer’s suggested retail price (MSRP).”

The article concludes optimistically though, with the assertion that “there’s a good probability that these inflated prices will come crashing down as gamers just say no”, leaving the price gougers with graphics cards they can’t sell (or rather they’ll be forced to sell them at a loss – what a shame that’d be).

Analysis: Are GPU prices really going to crash?

We’d broadly agree with Jon Peddie here, and certainly as we’ve already said, pricing has clearly got ridiculous, particularly around mid-2021 when certain Nvidia graphics cards were costing triple their MSRP in some cases.

Where to place the blame is equally clear, with the pressures of crypto and Ethereum mining causing demand, taking cards out of the hands of gamers, while the pandemic-related supply problems worsened everything – and then caused the third factor, namely scalpers looking to buy that scarce inventory to resell for big profits.

As to whether GPU pricing will ‘crash’ as Peddie hints that there’s a good chance of happening – well, we’re not so sure on that. True, there are a lot of signs of late that availability of graphics cards is strengthening, albeit with prices drifting down and normalizing slowly rather than in any hurry.

But the theory has always been that the second half of 2022 will see more of a recovery from the component shortage for both AMD and Nvidia, with new Arc cards coming to market from Intel as well to bolster supply levels further. It’s a promising sounding mix of positive steps forward, for sure, and so there’s certainly reason to hope that the GPU market will see prices dropping at more of a speed akin to a stone than a feather.

On balance, it certainly seems like pulling the trigger on a GPU purchase this month would be rash, and the best bet is to wait and keep an eye on those price tags which are only heading in one direction right now (with the other eye on potential market developments that could muddy any recovery, like further lockdowns in China for example).

• Check out our thoughts on whether 2022 will be a PC gaming apocalypse or utopia

PC graphics card sales were up strongly last year according to a new piece of research, and the AIB or add-in board (meaning desktop graphics card) market grew to a value of $51.8 billion (around £40 billion, AU$68 billion) in 2021.

This comes from Graphic Speak (a publication of analyst firm Jon Peddie Research or JPR), and the report observes that in Q4 2021, shipments numbers of GPUs were up 29.5% compared to the same quarter in 2020, which is quite a boom (they were also up 3% on the previous quarter). In total, $13.5 billion (around £10 billion, AU$18 billion) worth of graphics cards were shipped in the final quarter of last year.

AMD actually gained more ground than Nvidia in Q4 2021, interestingly, with Team Red’s graphics card shipments up 35.7% year-on-year, whereas Nvidia didn’t make quite the same level of progress, albeit still managing very robust growth of 27.7%.

Team Green still remains by far the dominant player for discrete desktop GPUs, of course, holding 77.2% of the market.

Analysis: AMD turning the graphics card tide?

Nvidia generally holds around an 80% desktop graphics card market share as per JPR’s regular figures, although in recent times it has reached as high as 83%. That said, we’ve seen it drop below 80%, too, but 77% is about the leanest market share figure we have witnessed in a while – and it’ll be interesting to see if AMD can further leverage any more sales momentum here.

Team Red does, of course, have a trio of refreshed RDNA 2 GPUs rumored to be close to arrival – along with fresh cheaper RX 6000 series models, too – and that could propel AMD further forwards, maybe even attacking to drive Nvidia below the 75% mark. A 25% share is as good as we’ve seen AMD do in recent times, going by JPR’s stats (that was back at the start of 2020), and just maybe that could be bettered in the near future.

The report also notes that PC gaming is getting more popular, and is partly responsible for the overall growth in GPUs, stating that: “With the help of the rise of esports and the growing popularity of PC gaming, AMD and Nvidia have been reporting record game segment revenues over the past quarters.”

Of course, discrete desktop graphics cards are not the sole preserve of PC gamers, as they are also used in professional settings (workstations or servers), and let’s not forget crypto-mining (even though we might want to). The latter was certainly a strong driver of GPU sales last year, and part of the reason why while more cards may have been shipped, it certainly wasn’t easy to buy the current-gen GPU you wanted to slot into your PC.

Crypto sales are, however, now on the wane, or that’s certainly the prevailing sentiment, and coupled with slowly recovering supply levels, we are finally seeing graphics card prices normalize somewhat, with that trend expected to continue. Don’t forget that Intel’s entry to the graphics card market will soon make big waves, too, in terms of supply, and hopefully pricing with its Arc cards (coming to laptops first, but desktops as well at some point in Q2).

• Here’s why we think Intel’s Arc laptop GPUs are exactly what gaming notebooks need

Via Tom’s Hardware

AMD’s incoming refreshed RDNA 2 GPUs, of which there are rumored to be three new models led by a fresh flagship RX 6950 XT, have been pushed back slightly to May, rather than launching in April as the rumor mill initially theorized.

According to a new report from VideoCardz, its sources claim that the Radeon RX 6950 XT, 6750 XT, and 6650 XT are due to land on May 10 (rather than April 20, the previous launch date which was floated via the grapevine).

The tech site provides a leaked pic of the three new RDNA 2 graphics cards, as well, noting it’s from a trusted source, Disclosuzen, and this highlights another interesting nugget of info – that they will all be black edition models.

The aforementioned previous leak suggested that the RX 6950 XT alone could get the black paint job, looking just like the RX 6800 XT Midnight Black model, but in fact, all three of these GPUs will copy that exact same design (or at least that certainly looks to be the case from the image provided).

A final piece of this round of rumor spillage is that the existing RX 6400 budget graphics card, which is currently OEM-only, is set to be released as a standalone model on May 10 also.

This means that you’ll be able to buy it off the shelf (whereas at the moment, it’s currently only present in prebuilt PCs). VideoCardz didn’t mention anything about the purported RX 6300 which we heard about only yesterday.

Analysis: Let’s face it, a delay would be no surprise

Of course, all of this is just speculation, so we must remain skeptical about this trio of incoming GPUs. Assuming the previous rumor of an April launch was correct at the time it was circulated, the cards have now theoretically been delayed, and if that is true, it’d be no big surprise.

Delays happen all the time with hardware launches, and especially these days given the ongoing component shortages still affecting the tech industry (albeit we are hearing more positive noises of late that supply problems are easing, slowly but surely, and if nothing else GPU pricing is drifting downwards and normalizing somewhat).

What’s undeniably true, though, is that the publishing of an actual picture – assuming it’s not faked – of the purported 6950 XT, 6750 XT and 6650 XT does lend more weight to their rumored existence, naturally. And it’s an interesting move that the entire refreshed range could be black editions.

Perhaps this is AMD looking to make these GPUs seem a touch more attractive in light of the fact that by the time they’re on shelves – in May supposedly, or potentially even later, as we can’t rule that out – it won’t be that long until RDNA 3 graphics cards are due to emerge (they’re scheduled to come out before the end of 2022, as you may recall).

• Good reasons to buy an AMD GPU include Team Red’s plans for turbocharging its products

AMD is purportedly going to launch a new even cheaper budget RDNA 2 graphics card, at least if the latest rumor proves to be on the money.

The theory is that Team Red is preparing an RX 6300 GPU to slot in below the current wallet-friendly RX 6400, as per speculation shared by Komachi_Ensaka, a hardware leaker on Twitter (whose tweet was spotted by Wccftech).

The tweet mentions that the RX 6400 is ‘no longer exclusively for OEMs’, meaning that it isn’t just going to be present in pre-built PCs, and will be available to buy as a standalone product on shelves. Apparently the RX 6300 is incoming as an ‘OEM exclusive’ – meaning this GPU will only be sold to PC manufacturers for them to include in new machines.

In short, the RX 6300 is theoretically in the pipeline, but you won’t be able to buy one as a standalone model to put in your existing PC; it’ll only be out in new computers.

Analysis: Another wallet-friendly GPU would be very welcome

Assuming this is true – large pinches of salt to hand, as ever – bear in mind that in the future, you may still be able to pick up a resold RX 6300 on an auction site, say. And also it’ll mean budget PCs can soon become even more wallet-friendly when OEMs get the option to use the 6300 and shave a bit more off the cost of making a cheap system.

There is also the chance that if the RX 6300 appears as an OEM-only graphics card to begin with, it might eventually get produced and sold as a standalone product. As noted above, this is what’s rumored to be happening with the RX 6400, which was introduced as an OEM-only GPU back in January, but could be sold on its own very soon – supposedly in April – if the rumor mill is right (note that this is not confirmed).

We’ll just have to see, but fingers crossed that the RX 6300 is in the cards, as another useful option for the budget end of the GPU market certainly wouldn’t go amiss. We’d imagine we are more likely to see such a card sooner rather than later, as next-gen RDNA 3 products aren’t all that far off, after all (they’re set to land later in 2022).

• Want to build cheap PC? One that doesn’t suck? We’ve got some thoughts on that

Asus has stated that it’ll soon be reducing the prices of its graphics cards across the board, and by a considerable amount as well.

This further good news for GPU buyers – which comes on top of a lot of recent positive noises regarding price tags – comes from PC Gamer, who spoke to the card manufacturer directly.

Asus noted that it has “been dropping pricing across all SKUs,” meaning all its different models of graphics cards, and that the firm is “reducing MSRP aggressively” right now.

The use of the term 'aggressively' certainly bodes well for some chunky price cuts on Asus products, and doubtless soon enough, from other card manufacturers who face a similar situation to Asus, and the need to remain competitive on pricing, of course.

Analysis: GPU market is slowly but surely reverting to normal

As we’ve seen, the amount of GPU stock on shelves is now ticking up meaningfully, and so the first manufacturer to make a move dropping pricing by a hefty amount could hoover up a lot of sales in a short period of time. This would appear to be the intention here from Asus.

A supply chain expert, Dr Thomas Goldsby (of the Haslam College of Business, University of Tennessee), told PC Gamer: “A supplier out there will get nervous that they are sitting on considerable supply for an item that has a limited shelf life (with newer cards entering the market) and will make the move to clear out that excess inventory.

“Peer suppliers will then have to follow that lead. And back to equilibrium we fall.”

As stock levels pick up, card makers will indeed want to ensure they’re selling enough through before next-gen models from AMD and Nvidia hove into view later in 2022, and crucially, before what’s likely to be another major turning point in the GPU arena – Intel’s entry to the market.

Intel Arc Alchemist graphics cards are going to be launched this week (tomorrow, in fact), although initially they’ll be laptop GPUs only, but desktop models will follow in Q2 (possibly landing in May or June, from what we’ve heard on the grapevine). Rumor also has it that Intel is set to produce major quantities of these new Alchemist cards, which could help to considerably ease supply issues that are already being remedied as we’ve seen.

Recently, Nvidia said that it was passing on cost savings to third-party graphics card makers, and that’s part of a bigger picture where GPU price reductions are occurring more often across different models, and starting to come down much closer to normal levels (recommended pricing, in other words).

On top of all this, we’ve just heard from a major UK retailer, Box, that by the end of April, or start of May, GPUs should start on a path where they “return to a more attractive price” – and with all these predictions stacking up, on top of what we can take to be concrete news of reductions from Asus, maybe it is time to finally be optimistic about being able to find and buy a decent graphics card without paying (perhaps well) over the odds.

• Check out all the best gaming PCs

The most popular graphics card out there (at least going by Steam’s hardware survey stats) falls below the recommended requirements for AMD’s incoming next-gen frame rate boosting feature.

Team Red recently revealed FSR 2.0 (the successor to the original FidelityFX Super Resolution tech), and made it clear that version 2.0 will not be usable (in theory – more on that later) by those with an Nvidia GTX 1060 graphics card (whereas v1.0 was compatible).

The new baseline where support for Nvidia cards begins is the GTX 16 Series (Turing) and GTX 1070 from the previous Pascal generation. On the AMD side, the base GPUs are the RX 6500 XT and RX 590.

This is a recommendation, mind, so not a hard-and-fast rule, but nonetheless the GTX 1060 – and RX 580 for that matter – can’t clear the bar with FSR 2.0 running at 1080p.

Higher resolutions make even greater demands on your graphics hardware, of course. We’re talking an RX 6600 or RX 5600 (or RX Vega) or better, or an RTX 2060 and GTX 1080 (or RTX 3060 for current-gen) when it comes to 1440p.

4K calls for an RX 6700 XT or RX 5700 and above, and an RTX 2070 or RTX 3070 for those with Team Green.

FSR 2.0 is expected to be released at some point in Q2.

Analysis: The temporal price… but all is not lost for GTX 1060 owners

First off, why the greater demands here? Well, FSR 2.0 moves to use temporal upscaling, rather than spatial upscaling as seen in FSR 1.0, the difference being that the former uses data drawn from past frames (not just the current frame) to improve overall quality. Unsurprisingly, upping the quality to better compete with Nvidia’s rival DLSS tech is more taxing for the GPU, and there’s no way around that.

Still, these are only recommendations from AMD, and it could be the case that in certain PCs, maybe with faster CPUs or other components working alongside the popular GTX 1060, the results could be palatable (at least in some games).

In short, this isn’t a black-and-white supported-or-unsupported situation; there’ll be shades of grey, and some owners with slightly underpowered GPUs compared to AMD’s suggestions might get away with it at 1080p.

Team Red itself noted: “Depending on your specific system specifications, the system requirements of individual games that support FSR 2.0, and your target resolution, you may be still able to have a good upscaling experience on lower-performing or older GPUs [than the recommended ones].”

Further tweaks to FSR 2.0 could improve the situation going forward, as well, as our sister site PC Gamer points out, while also observing that it seems like the new tech depends considerably on memory bandwidth to work well. (The RTX 3050 is notably not recommended, whereas the GTX 1070 gets a shout – and the latter offers more bandwidth, even though technically the 3050 is a slightly faster card).

Finally, it’s worth remembering that AMD also now has Radeon Super Resolution (RSR) on the table as a more basic frame rate booster, and one that applies across a great deal of games, not just those coded to support it (FSR support must be baked-in by game developers). The catch with RSR is that it’s directly in the Radeon driver, so while it’s good for AMD GPUs, Nvidia card owners can’t benefit from it.

• Check out all the best gaming PCs

AMD could unleash a new RDNA 2 budget GPU, an RX 6400 card which might arrive as soon as April, according to the latest speculation.

You may recall that the RX 6400 graphics card has already been launched by Team Red at the start of 2022, but as an OEM-only GPU, meaning that AMD is just supplying them to system builders, and you can only get one in a prebuilt PC (or resold, perhaps). However, that’s set to change next month, at least if this VideoCardz report is correct.

The tech site has spotted a filing for a Gigabyte RX 6400 Eagle graphics card made in South Korea (with the National Radio Research Agency).

The theory therefore is that new standalone RX 6400 GPUs could be coming alongside AMD’s other big graphics card launches that the rumor mill has indicated are happening in April. Indeed, we have a rumored launch date of April 20 for the RX 6950 XT, 6750 XT and 6650 XT, which are supposedly pepped-up refreshed versions of the 6900 XT and other respective GPUs currently out there from Team Red.

Analysis: Growing evidence for a release outside the OEM market

While these kinds of regulatory filings being flagged up can indicate the upcoming launch of a product, and indeed have done exactly that in the past, we can’t take this South Korean sighting as any kind of cast-iron guarantee.

However, this done line up with a previous rumor that as well as the stopgap refreshes for existing RDNA 2 models mentioned above, AMD intends to move the RX 6400 from being just OEM to an actual product that’s on shelves available for consumers to purchase. So, in short, there’s growing evidence that this is Team Red’s plan.

Whether the spec of this would-be RX 6400 spin might be identical to the OEM variant of the graphics card, we obviously don’t know, but the grapevine has previously theorized that it may be slightly different – though any changes would be minor. Indeed, past rumors have suggested the GPU may even be renamed (perhaps as the RX 6500), even though it would remain much the same, but with this sighting, it looks like the name will remain as is. Take all of this with a good deal of skepticism, naturally.

The RX 6400 is a low-profile card (and furthermore needs no power hook-up from the PSU), and it’d bring pricing down to lower than ever levels for an RDNA 2 model – though very much temper expectations when it comes to performance, of course. Pricing would, as ever, be key, and AMD could easily make this a robust addition to the wallet-friendly end of the GPU spectrum.

• Check out all the best gaming PCs