HITMAN 3 Runs Well On Linux With Steam Play - Open-Source Radeon Performance Especially Good

[phoronix]

Phoronix: HITMAN 3 Runs Well On Linux With Steam Play - Open-Source Radeon Performance Especially Good

After being an Epic Games exclusive for its first year, HITMAN III launched on Steam last week. While there isn't a native Linux port for HITMAN 3, it does run wonderful under Steam Play with Proton and DXVK for enjoying this Windows game on Linux complete with Vulkan API rendering. Here are some initial benchmarks of HITMAN 3 on Linux with NVIDIA GeForce and AMD Radeon graphics cards.

https://www.phoronix.com/vr.php?view=30876

[Masush5]

FWIW, Hitman 3 uses d3d12 and thus runs on vkd3d-proton and not DXVK. I think it would also be nice if you could include AMDVLK results when testing d3d12 games in the future (another d3d12 game that is already in your testsuite would be Strange Brigade), thanks!

[cytomax55]

Being an AMD fanboy I'd still like to know if the picture quality is the same between the cards or if AMD is simply not rendering something which is what's giving it an edge by the way it's nice to see AMD on top of Nvidia secondary to the drivers...
I wonder how this compares to Windows performance

[programmerjake]

I wonder how this compares to Windows performance

It'd be nice if phoronix could have some linux vs. windows comparisons.

until that happens, you can see some toms hardware results, they used different hardware and slightly different settings, but https://cdn.mos.cms.futurecdn.net/7T...g8-1280-80.png

[aufkrawall]

Being an AMD fanboy I'd still like to know if the picture quality is the same between the cards or if AMD is simply not rendering something which is what's giving it an edge by the way it's nice to see AMD on top of Nvidia secondary to the drivers...

Image quality should be the same. GPU bound performance hit was already lower in Hitman 2 VKD3D-Proton RADV vs. Nvidia where everything rendered correctly.

[pete910]

Being an AMD fanboy I'd still like to know if the picture quality is the same between the cards or if AMD is simply not rendering something which is what's giving it an edge by the way it's nice to see AMD on top of Nvidia secondary to the drivers...
I wonder how this compares to Windows performance

That's normally Nvidias trick

[free2create]

linux Vs Windows ? For a fixed graphics card it would be very interesting to see what the overhead cost is for proton usung HITMAN, eventhough the cost might vary widely from game-to-game

[onlyLinuxLuvUBack]

Thank you for including the rx590 for the wallet challenged people.

[chocolate]

It'd be nice if phoronix could have some linux vs. windows comparisons.

until that happens, you can see some toms hardware results, they used different hardware and slightly different settings, but https://cdn.mos.cms.futurecdn.net/7T...g8-1280-80.png

Interesting. So, Phoronix numbers seem to be quite a bit better both for AMD and Nvidia at 1080p Medium. Main difference is Michael's Ryzen 9 5950X vs. Tom's Hardware's i9-9900K... what else? Sure, operating system and possibly newer/better drivers for Phoronix, but that goes without saying. TH mentions 21.1.1 drivers for AMD on Windows, but doesn't mention Nvidia drivers if my eyesight and my Ctrl+F skills are still okay. Also assuming both numbers are average fps.

	Phoronix	Tom's Hardware
RX 6800 XT	316.36	272.1
RX 6800	280.63	269.2
RTX 3090	251.05	227.6
RTX 3080	234.38	226.0

[birdie]

That's normally Nvidias trick

This was the case almost 20 (!) years ago when NVIDIA sort of cheated on anisotropic filtering. Would be great if people stopped perpetuating this BS.

Many people over the past decade have compared NVIDIA and AMD in terms of image quality and nothing has been found, e.g.