Viewing topic Warhammer 40.000 Dawn of War II Benchmark

Title: Warhammer 40.000 Dawn of War II Benchmark

Vortex_Acherontic 10 Apr 2024

Link

While I am at it I also ran the Dawn of War II Benchmark using the native Linux version made by Feral Interactive against the Windows version using Proton Experimental.

Preset: Ultra
Resolution: 2560x1440

Best out of two:

Linux Native:
Avg. FPS: 172.47
Max FPS: 349.65
Min FPS: 95.23

External Media: You need to be logged in to view this.

Proton Experimental:

Version: experimental-9.0-20240328c
DXVK: v2.3.1-3-g855b2746

Avg. FPS: 182.6
Max FPS: 362.63
Min FPS: 105

External Media: You need to be logged in to view this.

Conclusion: Even though this is one of the older Feral Interactive ports (Linux release: September 29, 2016) using OpenGL the performance does not differentiate from the Windows version using DXVK that much. Well at least not as anticipated by me. I mean we talk about 5% differences in avg. FPS here which is not really noteworthy tbh.
I suspect these results to be fairly close to each other might be related to the Windows version written to use DirectX 9.0c. As both APIs, OpenGL and DirectX 9, are rather similar in the way they worked internally.

Also, in addition to a recent discussion below the DiRT Rally Benchmarks, it surprises me that these fairly old Feral Interactive Port worked out-of-the-box for me with no tweaking. Espcially since DiRT Rally was release only a year later (2017) for Linux and should make use of similar porting tech as Dawn of War II. But yet the native DiRT refuses to work for me.

Again, this seems to be a very well made Linux Port by Feral Interactive. Good Job! I really wish you guys will make Linux Ports some day in the future again.
However the native version was still a bit flunky with fullscreen (in a way it did not fullscreened) but that might be related to XWayland or maybe an old version of SDL. Dunno though.

Edit: Dawn of War III Linux native does not work for me. What ever Feral Changed in their porting from 2016 to 2017 they shouldn't.

System:

OS: openSUSE Aeon (Kernel 6.8.4-rc1)
CPU: AMD Ryzen 7 7800X3D
GPU: nVidia GeForece RTX 3080 (Driver: 550.67)
RAM: 16 GB DDR-5 + ZRam
DE: Gnome 46 (Wayland)

Last edited by Vortex_Acherontic on 10 Apr 2024 at 12:24 pm UTC

1 Likes, Who?

PublicNuisance 14 Apr 2024

Link

Here are my results.

Game From: Steam
Proton 8.0-5
Game Settings: Highest, AA On
Resolution: 2560x1440

Min FPS: 37.81
AVG FPS: 120.30
Max FPS: 260.00

Game From: Steam
Native OpenGL
Game Settings: Highest, AA On
Resolution: 2560x1440

Min FPS: 28.62
AVG FPS: 86.21
Max FPS: 175.82

My System:

Intel i5-12600K | 32GB DDR4-3200 CL16 | XFX RX 590 8GB Fat Boy | Mesa 24.0.2-manjaro1.1 | Western Digital Black SN850x 2TB | MSI Pro Z690-A DDR4 | Dasharo 1.1.1 | Manjaro | Mate 1.26.1 | Kernel 6.8.5-1-MANJARO| MSI G2730QPF 2560*1440 @ 165hz

1 Likes, Who?

Vortex_Acherontic 14 Apr 2024

Link

Proton:

Quoting: PublicNuisanceMin FPS: 37.81
AVG FPS: 120.30
Max FPS: 260.00

Native:

Quoting: PublicNuisanceMin FPS: 28.62
AVG FPS: 86.21
Max FPS: 175.82

I find it quite interesting that there is a ~28% difference in FPS between the native version vs the Windows version in this case. Then I noticed you had AA enabled which I didn't. Therefore I re-run the benchmark with AA enabled.

Res: 2560x1440
Preset: Ultra
AA: Enabled

Linux native:

Avg: 176.77
Max: 352.64
Min: 98.7

Proton (8.0-5):

Avg: 187.24
Max: 372.62
Min: 104.16

Which still leads to a ~5.9% difference in FPS between native and Proton but not like 29%. Which seems odd to me. Any ideas how this could happen? Especially since you run a fairly up-to-date Mesa version.

System used:

OS: openSUSE Aeon (Kernel 6.8.4-rc1)
CPU: AMD Ryzen 7 7800X3D
GPU: nVidia GeForece RTX 3080 (Driver: 550.67)
RAM: 16 GB DDR-5 + ZRam
DE: Gnome 46 (Wayland)

1 Likes, Who?

PublicNuisance 14 Apr 2024

Link

Quoting: Vortex_AcheronticProton:

I find it quite interesting that there is a ~28% difference in FPS between the native version vs the Windows version in this case. Then I noticed you had AA enabled which I didn't. Therefore I re-run the benchmark with AA enabled.

Res: 2560x1440
Preset: Ultra
AA: Enabled

Linux native:

Avg: 176.77
Max: 352.64
Min: 98.7

Proton (8.0-5):

Avg: 187.24
Max: 372.62
Min: 104.16

Which still leads to a ~5.9% difference in FPS between native and Proton but not like 29%. Which seems odd to me. Any ideas how this could happen? Especially since you run a fairly up-to-date Mesa version.

System used:

OS: openSUSE Aeon (Kernel 6.8.4-rc1)
CPU: AMD Ryzen 7 7800X3D
GPU: nVidia GeForece RTX 3080 (Driver: 550.67)
RAM: 16 GB DDR-5 + ZRam
DE: Gnome 46 (Wayland)

Hard to say. The differences in software betwween us is Wayland vs X11, Gnome vs Mate, Proton 8.0-5 vs Experimental, and Manjaro vs OpenSuse, none of which should really make a huge difference. In my experience AMD GPUs can have some pretty poor results with OpenGL even when I was on Windows so maybe it is the reason. I have a GTX 780 Ti on a Mint system I will try to test in the next few days to see what kind of difference it has in native vs Proton.

1 Likes, Who?

While you're here, please consider supporting GamingOnLinux on:

Reward Tiers: Patreon. Plain Donations:

PayPal.

This ensures all of our main content remains totally free for everyone! Patreon supporters can also remove all adverts and sponsors! Supporting us helps bring good, fresh content. Without your continued support, we simply could not continue!

You can find even more ways to support us on this dedicated page any time. If you already are, thank you!