This is sad to see. A new benchmark video for Windows and Linux using Dota 2 actually shows Windows doing quite a lot better than Linux.
I spoke to the person who did it on twitter, he mentioned both Windows/Linux were done in borderless fullscreen mode. The Linux Nvidia driver was 367.17, while Windows used 368.25. As for AMD Windows used 16.5.3, while Ubuntu used amdgpu-pro.
I said before plenty of times that Vulkan will not be some magical bullet to bring Linux performance up on par with Windows. There can be problems elsewhere contributing to the performance drop on Linux. It could be X11, it could be a compositor issue, it could be lots of things.
I'm certainly no expert in these issues, but seeing Vulkan perform worse than OpenGL at 4K on Linux was quite disappointing to me. At least in lower resolutions Vulkan was performing better than OpenGL on Linux. The major issue is just how much faster Windows is with Vulkan than Linux.
What are your thoughts, any theories? Article taken from GamingOnLinux.com.
YouTube videos require cookies, you must accept their cookies to view. View cookie preferences.
Direct Link
Direct Link
I spoke to the person who did it on twitter, he mentioned both Windows/Linux were done in borderless fullscreen mode. The Linux Nvidia driver was 367.17, while Windows used 368.25. As for AMD Windows used 16.5.3, while Ubuntu used amdgpu-pro.
I said before plenty of times that Vulkan will not be some magical bullet to bring Linux performance up on par with Windows. There can be problems elsewhere contributing to the performance drop on Linux. It could be X11, it could be a compositor issue, it could be lots of things.
I'm certainly no expert in these issues, but seeing Vulkan perform worse than OpenGL at 4K on Linux was quite disappointing to me. At least in lower resolutions Vulkan was performing better than OpenGL on Linux. The major issue is just how much faster Windows is with Vulkan than Linux.
What are your thoughts, any theories? Article taken from GamingOnLinux.com.
0 Likes
Some you may have missed, popular articles from the last month:
The comments on this article are closed.
natewardawg Jun 15, 2016
Quoting: liamdaweI don't see the point, since I have a 980ti, not the 1080 that was tested. Totally different generation.
One issue may be that the 1080 literally just became supported on Linux the day before he posted that video. It may be good to try it on a GPU that's had more time in the oven on Linux. 1080 has been running on Windows since before the 1080 was released.
http://news.softpedia.com/news/nvidia-367-27-video-driver-finally-brings-geforce-gtx-1080-1070-support-to-linux-505199.shtml
http://www.nvidia.com/download/driverResults.aspx/104307/en-us
4 Likes, Who?
Quoting: GuestSo this is strange. i have a GTX 770 and a i7 4770k and i am running the manjaro distro with cinnamon desktop at 1080p with 364.19 drivers. I dint try to replicate the roshan scene but on average i get around 115 fps with vulkan. I usually play with vsync and i never get bellow 60 fps. the strange part here is that the GTX 1080 is more than twice as powerful as my 770, yet the difference isn't all that big as it should considering the difference in the performance of both cards.Vsync limits your framerate to the refresh rate of your monitor. Thus you can only run at 60fps, 30fps, 15fps, 7fps, 3fps, 1fps. If you want to compare performance you will need to turn off vsync and sometimes even gsync.
I'm still calling for this to be tested on a high performance distro rather than Ubuntu.
2 Likes, Who?
Quoting: coruunI wonder, how this relates to Phoronix' results of 107FPS (compared to his 80FPS) for the 4K Vulkan test.I imagine compositor issues (compiz anyone?) mixed with the different demo file.
Is this simply the "difficult" scene or are there some other problems?!
0 Likes
Is borderless screen mode a real fullscreen? Because we all know than on the X server, not fullscreen applications loose many FPS compared to real fullscreen applications
and he also should take a looks at this https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/vulkan/873770-nvidia-vs-amd-opengl-vulkan-benchmarks-with-valve-s-dota-2?p=873829#post873829
Last edited by edo on 15 June 2016 at 1:29 pm UTC
and he also should take a looks at this https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/vulkan/873770-nvidia-vs-amd-opengl-vulkan-benchmarks-with-valve-s-dota-2?p=873829#post873829
Last edited by edo on 15 June 2016 at 1:29 pm UTC
0 Likes
Mblackwell Jun 15, 2016
SteamOS actually uses its own compositor, in order to allow smooth transition between BPM and applications.
Edit:
To the above question about borderless window: No, if an application is not exclusive fullscreen it won't set the flag required to bypass the compositor.
Last edited by Mblackwell on 15 June 2016 at 1:24 pm UTC
Edit:
To the above question about borderless window: No, if an application is not exclusive fullscreen it won't set the flag required to bypass the compositor.
Last edited by Mblackwell on 15 June 2016 at 1:24 pm UTC
3 Likes, Who?
i don't understand this benchmarks... With my 780 i stay always on 120 fps or 110 during heavy teamfight with vulkan api without any problems. So why a gtx 1080 got a minimum of 97 fps? lol. Microsoft already started to pay youtubers for fake benchmark? dafaq
0 Likes
QuoteWhat are your thoughts, any theories?Vulkan won't make any performance difference between Linux and Windows. Only the drivers will.
Vulkan will be good for Linux because it's a cross-platform API and that it's closer to what the game devs wants (something as powerful as DirectX 12)
2 Likes, Who?
Quoting: MblackwellSteamOS actually uses its own compositor, in order to allow smooth transition between BPM and applications.Question is, does this matter if the user isn't using a compositor?
Edit:
To the above question about borderless window: No, if an application is not exclusive fullscreen it won't set the flag required to bypass the compositor.
0 Likes
Don't give a shit about random youtubr posting their benchmark.The source is unreliable.As for Vulkan,Its multi-platform.so,it was never meant to be a silver bullet for Linux.But work as aegis, for shielding the weak form being totally devoured.Its up to Linux community if they can actually turn it into a bullet.
0 Likes
Mountain Man Jun 15, 2016
Quoting: devlandJust once, I'd love to see a test like this that Linux fans don't have to explain away.QuoteWhat are your thoughts, any theories?Vulkan is a new API. So far all performance tests yielded similar results. This was expected and is not news.
Please don't sensationalize the issue. Vulkan will get better in time as will it's implementation by game developers. :)
No matter what happens from here on out, I'm never using Windows again (unless Linux is completely abandoned by developers), but to have it constantly lag behind Windows is frustrating. Of course this isn't a Linux issue exclusively since MacOS performs similarly to Linux in benchmarks, so I think it's a simple fact that Linux/OSX drivers aren't nearly as well optimized as Windows drivers. I just want to see this issue resolved sooner rather than later.
2 Likes, Who?
Patreon
PayPal
See more from me