Radeon RX 560/570/580 vs. GeForce GTX 1060/1650/1660 Linux Gaming Performance

[phoronix]

Phoronix: Radeon RX 560/570/580 vs. GeForce GTX 1060/1650/1660 Linux Gaming Performance

If you are looking to soon upgrade your graphics card for Linux gaming -- especially with the increasing number of titles running well under Steam Play -- but only have a budget of around $200 USD for the graphics card, this comparison is for you. In this article we're looking at the AMD Radeon RX 560 / RX 570 / RX 580 against the NVIDIA GeForce GTX 1060 / GTX 1650 / GTX 1660 graphics cards. Not only are we looking at the OpenGL/Vulkan Linux gaming performance both for native titles and Steam Play but also the GPU power consumption and performance-per-dollar metrics to help guide your next budget GPU purchasing decision.

http://www.phoronix.com/vr.php?view=27817

[davidbepo]

nice testing, the only thing i dont like is seeing the gtx 1060 above the rx 580 in the harmonic mean, it should be the reverse

[gukin]

Being an owner of an RX 570 this test reaffirms something I already suspected: I'm a cheap SOB.

That aside, thank you Michael for doing a comparison of the lower-end cards, it gets a little dull looking at the amazing but nearly useless performance of $500+ cards. Some of us get a kick out of biggest bang for the buck.

[nadro]

nice testing, the only thing i dont like is seeing the gtx 1060 above the rx 580 in the harmonic mean, it should be the reverse

RX580 lost a lot to GTX1060 in Strange Brigade benchmark. Without this result I think that RX580 should be higher than GTX1060. I hope that fix to RADV for this Vsync issue will be available soon. At now I use AMDVLK driver for Strange Brigade.

[Grinness]

RX580 lost a lot to GTX1060 in Strange Brigade benchmark. Without this result I think that RX580 should be higher than GTX1060. I hope that fix to RADV for this Vsync issue will be available soon. At now I use AMDVLK driver for Strange Brigade.

To be fair the harmonic mean should not consider performance on Strange Brigade -- it essentially runs at 60 fps locked.
It is a known bug (already highlighted in some old articles), it makes sense to do check tests to verify if resolved, but completely miss-leading if included in performance comparisons.

[loganj]

those radeon cards sure use a lot of power.

[tildearrow]

Typo:

the RADV Radeon performance was coming in a bit short wit hthe RX 580 losing to the GTX 1060

[skeevy420]

I know that I've been very happy with my 4GB RX 580. I can play most games at 1080p60fps using slightly tweaked ultra settings. Slightly tweaked means lowering various shadow options and disabling SSAO.

[Girolamo_Cavazzoni]

In Germany RX 590 and GTX 1660 are roughly equally priced and therefore should be matched against each other... I don't know the US price tag of the 590 though.

[humbug]

in the Dota 2 and hitman 2 Vulkan benchmarks each GPU's min, avg and Max fps are virtually the same. Basically each GPU is maintaining a fixed framerate throughout the test.

This is highly unusual behaviour in a graphical / gaming bench.