Jarred Walton

AMD has launched the RX 7900-series and RX 7600, but nothing in between. When will the RX 7700 and RX 7800 arrive? Because we really want RDNA 3 upgrades in the $400–$600 range.

Swimming with the next generation GPUs

We've tested Diablo IV on a bunch of graphics cards to see how it runs. Here's what hardware and settings you'll need to get the best performance from the game.

These are the 15 most promising products we saw at Computex Taipei.

We tested Lord of the Rings: Gollum on some graphics cards using the launch day release to see how it performs. Maxing out the settings will be problematic if you don't have a monster GPU.

We've compiled a list of all the Radeon RX 7600 models that are available for purchase. The RX 7600 starts at $269, with custom models currently topping out at $339.

Bring on the mainstream RTX 40-series GPUs!

Many games are starting to push well beyond 8GB of VRAM use, particularly at 4K. We discuss what's going on behind the scenes that makes 4K much more demanding on your graphics card.

We tested it on everything from an RTX 4090 to an Intel Arc A380.

Nvidia's new GPU takes on AMD's previous generation king.

The latest patch helps to address excessive VRAM use and seems to run better.

Apparently, there was a glitch in the previous month's survey

Dead Island 2 is here and we've tested it on a bunch of graphics cards to see what sort of settings and performance you can expect from your GPU.

Unreal Engine meets vampires, DLSS 3, and open world gameplay

These are the 25 most innovative products that we've tested or covered in the last year, from multi-screen laptops to Mini-LED gaming laptops and more.

This is why fully ray traced rendering isn't ready for prime time, yet.

Intel's Arc drivers were relatively huge at launch, but in the past two months they've shrunk to less than half their former size. Impressive!

Nvidia announced a new H100 NVL PCI Express solution, which packs dual H100 chips with NVLink into a chunky expansion card. The cards also feature more memory, allowing for larger models.