AMD Adrenalin driver 22.5.2 tested: big promises, but no GPU miracle worker

A bar chart showing the average FPS of various games running on both the AMD Radeon 22.4.2 and 22.5.2 graphics drivers.

I’ve been keeping a curious eye on AMD’s Software Preview Driver for May 2022; GPU drivers often promise “optimisations” that are rarely felt in practice, but this one genuinely sounded like a performance enhancing drug for Radeon graphics cards. in a community post, AMD asserted this particular Preview Driver could provide double-digit percentage gains in games running DirectX 11; claims that were backed up by early independent testing, like the creator of benchmarking tool CapFrameX getting an extra 24% out of Crysis Remastered.

this week, AMD Software Adrenaline Edition 22.5.2 released, making that new Radeon driver available to anyone who didn’t fancy switching to preview builds. Armed with an RX 6500 XT (a low-end card that desperately needs a performance kick), I gave the new software a whirl – and found that, for all those big numbers, Radeon owners should probably lower their expectations.

I’m not saying bigger gains won’t be had on other graphics cards, or in games besides the ones I’ve tested. But there’s a pretty wide gap between the reported results of AMD’s internal testing, and my own. For reference, these were recorded using the RX 6500 XT with an Intel Core i5-12600K and 16GB of DDR5 RAM, with all games running at 1080p:


A bar chart showing the average FPS of various games running on both the AMD Radeon 22.4.2 and 22.5.2 graphics drivers.
Across all four of these games, the combined difference was just 2fps.

A bar chart showing the average FPS of various games running on both the AMD Radeon 22.4.2 and 22.5.2 graphics drivers.
Watch Dogs Legion was supposed to be a big winner; instead, it suffered the biggest penalty.

In Watch Dogs Legion, for example, AMD reported performance gains of up to 10%. I in fact lost 10% after switching from the Adrenalin software’s 22.4.2 version to 22.5.2. Granted, I don’t think you’d notice a 3fps difference even around the 30fps mark, but I mean… that’s the wrong way round, right?

Assassin’s Creed Odyssey, meanwhile, can supposedly enjoy gains of up to 28%; Here, the latest driver added a single extra frame per second, which is more like 2%. Most of the other DX11 games I tried either gained or lost frames by similarly miniscule amounts, except for Shadow of the Tomb Raider, which gained about 9% on its Highest preset. And that’s good! But also the sole significant improvement in a pool of eight, and in a game that could run decently well on the RX 6500 XT to begin with.

It’s worth noting the vast gulf between my PC specs and the ones AMD used for their in-house testing, the latter comprising a Radeon RX 6950 XT GPU, Ryzen 7 5800X3D CPU, and 32GB of DDR4 RAM. It could just be a case of more powerful AMD graphics cards extending their leads over the likes of the RX 6500 XT, which doesn’t have much more left to give.

Still, if you own or are considering a Radeon GPU that’s not the brand-new, £1050, top-of-the-range model, don’t count on this driver update propelling it past Nvidia’s best graphics cards. That’s not to say you shouldn’t try it but results, as they say, will vary.



Reference-www.rockpapershotgun.com