20-Way Linux Graphics Card Comparison For Total War: Three Kingdoms

[phoronix]

Phoronix: 20-Way Linux Graphics Card Comparison For Total War: Three Kingdoms

Total War: Three Kingdoms is the newest Linux game port from Feral Interactive and saw a same-day release back in May. While back then it was said there weren't benchmarking capabilities for this game, there now is a test profile. For those wondering how Three Kingdoms performs on Linux, here is a twenty way graphics card comparison using the newest AMD Radeon and NVIDIA drivers.

http://www.phoronix.com/vr.php?view=28196

[tildearrow]

Typo:

Or for achieiving 60 FPS

[skeevy420]

Michael
Could the FPS bar graphs be tweaked so the average reported is the only thing in the bar itself with the min/max numbers in black right below the bar? Bottom of page one and on, most of the information below the Radeon VII and 5700 XT becomes gibberish with all the overlaying text due to how it compresses it to try to fit it all within the confines of the average of the bar used. That would probably require shifting the bars to the top of the separating hash mark to fit the text below, but it would make the FPS graphs a lot more readable.

Anyways, thanks for letting me know my 580 sucks

[cybertraveler]

Strange that the RX 5700 often had an equal or higher minimum FPS than the RX 5700 XT. On many benchmarks it came out higher.

[deppman]

Hey Michael, what's the hold up on the super cards? Are Nvidia going to provide you with samples? I'm curious of the performance delta. The gtx2060 super looks especially impressive.

[disobeyedtoast]

I'd be very interested to see a CPU comparison on this game considering the Total War series is usually CPU bound. Speaking of which, I wonder when that Ryzen 3000 gaming benchmark is going to be.

[Michael]

Hey Michael, what's the hold up on the super cards? Are Nvidia going to provide you with samples? I'm curious of the performance delta. The gtx2060 super looks especially impressive.

NVIDIA isn't responding at all to my inquiries about SUPER, so probably no tests unless I end up buying one in which case would only be the RTX 2060 SUPER.

[Michael]

I'd be very interested to see a CPU comparison on this game considering the Total War series is usually CPU bound. Speaking of which, I wonder when that Ryzen 3000 gaming benchmark is going to be.

Thursday should be for the 9900K vs. 3900X.

[Michael]

Michael
Could the FPS bar graphs be tweaked so the average reported is the only thing in the bar itself with the min/max numbers in black right below the bar? Bottom of page one and on, most of the information below the Radeon VII and 5700 XT becomes gibberish with all the overlaying text due to how it compresses it to try to fit it all within the confines of the average of the bar used. That would probably require shifting the bars to the top of the separating hash mark to fit the text below, but it would make the FPS graphs a lot more readable.

Anyways, thanks for letting me know my 580 sucks

Patches certainly welcome for any graphing improvements.
The particular code in question is right: https://github.com/phoronix-test-sui..._bars.php#L184

[Venemo]

Is this a Vulkan or an OpenGL game? In other words, did the benchmark use radv or radeonsi?