NVIDIA 387.12 Vulkan vs. OpenGL Performance Across Multiple CPUs

[phoronix]

Phoronix: NVIDIA 387.12 Vulkan vs. OpenGL Performance Across Multiple CPUs

Published earlier this week was the Intel Core i3, i5, i7 With NVIDIA vs. AMD Radeon For Linux Gaming results that are quite interesting while in this article is looking at the OpenGL vs. Vulkan Linux gaming performance using NVIDIA's first-rate binary driver while also doing this graphics API/renderer comparison across the Intel Coffeelake processors from low-end to high-end.

http://www.phoronix.com/vr.php?view=25329

[devius]

So, basically a Core i3-8100 is enough for all of the tested games. Is there any linux game where the more powerful CPUs provide increased performance?

[Michael]

So, basically a Core i3-8100 is enough for all of the tested games. Is there any linux game where the more powerful CPUs provide increased performance?

See the other linked article for more CPU/GPU game tests from a greater variety.

[johanb]

> The Core i3 8100 average CPU usage dropped from 37% to 25% when using Vulkan, the Core i5 8100 went from 24% to 17%, and the Core i7 8700K went from 12% to 8%.

Intel released both an i3 and an i5 named 8100?
I knew intels marketing was bad, but not this bad.

Seriously though, please fix.

[microcode]

Wouldn't the CPU utilization only really make sense for limited framerate operation?

For example, with the Serious Sam BFE comparison, the framerates are so much higher, but the CPU usage is only a little bit higher: this means that it's using considerably less CPU time to deliver each frame. V-sync tests for games that reliably exceed 60fps could be interesting, especially if combined with full system power usage monitoring.

[pinguinpc]

Michael

Good article but lack of 1080p tests again (4K dont care for now)

[Guest]

I can't wait to see such advantages on AMD hardware with Open Source drivers.

[bosjc]

So, basically a Core i3-8100 is enough for all of the tested games. Is there any linux game where the more powerful CPUs provide increased performance?

At 4K, yes - everything is GPU limited from these tests (although they show Vulkan has a bit more headroom there). As GPUs advance, the bottleneck should move back to the CPU, like what we see with 720p/1080p. Wish we had Ryzen in there, too.

[oleyska]

At 4K, yes - everything is GPU limited from these tests (although they show Vulkan has a bit more headroom there). As GPUs advance, the bottleneck should move back to the CPU, like what we see with 720p/1080p. Wish we had Ryzen in there, too.

Ryzen Yes, would be awesome.
720P to make cpu bottleneck visible is just a stupid kinda thing to sell a cpu on.
There is hardly anyone playing with a modern system at less than 1080p.
We can complain about cpu game performance when competitive games cannot run 144hz\fps and or cpu delivers 60 fps instead of 80 in story games.

So, basically a Core i3-8100 is enough for all of the tested games. Is there any linux game where the more powerful CPUs provide increased performance?

Yup, civ6 and Factorio (Singlecore).


And finally my own comment:
Add in some downright crap into the mix, which is usually where we saw the benefits of Vulkan\dx12 and not with modern overpowered cpu's.
I am thankful that we can call an I3 overpowered for games in 2017 which is the first time ever!

[humbug]

Good stuff from Nvidia and the game devs. Vulkan faster across the board.

also something for AMD to aspire to. In fact they should expect even bigger gains...