Radeon RX 560 Linux OpenGL/Vulkan Benchmarks

[phoronix]

Phoronix: Radeon RX 560 Linux OpenGL/Vulkan Benchmarks

Last week the "Polaris Evolved" Radeon RX 560 graphics card launched. I picked up a Sapphire Radeon RX 560 for Linux testing and have those results to share today for OpenGL and Vulkan workloads under Linux using the 4.12 development kernel and Mesa 17.2-dev compared to a range of Radeon and GeForce graphics cards.

http://www.phoronix.com/vr.php?view=24686

[satai]

Nice card but I am missing a variant with at least DP outputs :-/

[nuetzel]

Michael
Please redo with 'ARB_bindless_texture-for-RadeonSI'.
With this RX480 is _faster_ than GTX 1060 on DOW3 and others, now.

You can grep the whole series from Mesa patchwork.

https://patchwork.freedesktop.org/series/24711/

https://patchwork.freedesktop.org/api/1.0/series/24711/revisions/1/mbox/

[quoted from Samuel's cover letter]
=== Performance results for DOW3 ===

DOW3 exposes two bindless texture modes:
- mode 1: all bindless (ie. no bound samplers)
- mode 2: bound/bindless (ie. only bindless when the limit is reached)

CPU: Intel(R) Core(TM) i5-4460 CPU @ 3.20GHz
NVIDIA blob: 381.22

== GTX 1060 ==

LOW:
- mode 1: 89 FPS
- mode 2: 51 FPS

MEDIUM:
- mode 1: 49 FPS
- mode 2: 28 FPS

HIGH:
- mode 1: 32 FPS
- mode 2: 19 FPS

The GTX 1060 performs very well with the all bindless mode (default), while
the bound/bindless mode is not good at all.

== RX480 ==

LOW:
- mode 1: 67 FPS (-32%)
- mode 2: 75 FPS (+32%)

MEDIUM:
- mode 1: 38 FPS (-28%)
- mode 2: 44 FPS (+57%)

HIGH:
- mode 1: 26 FPS (-23%)
- mode 2: 29 FPS (+52%)

The RX 480 performs very well with the bound/bindless mode (default), while
the all bindless mode still has to be improved.

The most important bottleneck with the all bindless mode is the number of
buffers that have to be added for every command stream. The overhead in the
winsys and in the kernel (amdgpu_cs_ioctl) becomes important in this situation.
This mode is still clearly CPU bound and should be improved (see the "Future
work" section).

Btw, without any optimisations, it was around 35FPS in low (mode 1).

=== Performance results for other Feral titles ===

I didn't record any numbers because these games have been initially
developed/tested against the NVIDIA blob which it's unaffected by a VERY huge
number of resident handles. While the AMD stack is really slow in this
situation. Though, as I said, all Feral games that use bindless work fine, we
just need to improve perf on both sides.

[boxie]

Michael
Please redo with 'ARB_bindless_texture-for-RadeonSI'.
With this RX480 is _faster_ than GTX 1060 on DOW3 and others, now.

How about we let the work land first! there is more yet to be done. Be patient young padowan!

[nuetzel]

How about we let the work land first! there is more yet to be done. Be patient young padowan!

What? - It _is_ running here since Friday for testing (the latest version).
Looks all good.
Tested-by: is under way.
Have a look in .mailmap...;-)

[davidbepo]

rx 460 still perfroms badly, at least rx 560 is a nice updat,e but both cards should be performing better especially in xonotic and csgo

[hugo8621]

Didn't think the 560 would be that obvious faster than the 460. Supposedly the new 5XX series has a significantly lower idle power consumption, particularly for multi monitor setups. Would you mind verifying that, the 560 against the 460?

[dungeon]

That Michael's RX 460 behave weird since release Not sure if that is just Michael or driver problem for that one in general.

[Minute]

What is the lowest Kernel and Mesa that support the RX 560?
I suppose it is the 4.7 Kernel?

Thanks if someone can help me.

[humbug]

Didn't think the 560 would be that obvious faster than the 460.

Cause rx 560 has more CUs in addition to higher clocks. It's full fat Polaris 11. Not a rebrand.