Core i3 vs. Core i5 Performance Impact On OpenGL/Vulkan Linux Gaming

[phoronix]

Phoronix: Core i3 vs. Core i5 Performance Impact On OpenGL/Vulkan Linux Gaming

For a while now there have been some requests to post GPU benchmarks from some modern low-end and higher-end CPUs while testing different graphics cards, particularly to see the impact of the Vulkan API. With all the recent Kabylake testing, I've run some open-source AMD graphics tests using a Core i3 7100 and Core i5 7600K for those that may be weighing CPU options for a Linux gaming system upgrade.

http://www.phoronix.com/vr.php?view=24063

[chuckula]

So using a real-world comparison with these GPUs, there's basically zero practical difference between the two CPUs. Sure the 7600K can add a bunch of FPS to a game that was already running at over 200 FPS on the i3 anyway, but there is literally zero real-world impact to that "victory".

Maybe if you were using more powerful GPUs, like a GTX-1070 or 1080, and used more intensive games we could get a worthwhile comparison between these chips. These benchmarks, however, are pretty meaningless.

[L_A_G]

No Nvidia tests? On the Windows side it's well known that Nvidia's drivers scale much better with additional physical threads and you'd think that with their proprietary drivers sharing much of the code base this would also happen on Linux. AMD on the other hand is known to scale pretty badly, but it's an almost completely different code base so you can't really draw any Linux conclusions based on it.

So let's hope there's a follow-up to this using a GPU from Nvidia and their proprietary drivers.

[ResponseWriter]

Thanks for the benchmarks, Michael.

I guess the conclusion is that CPU makes little difference to GPU performance for a lot of games, at least where modern Intel chips are concerned. Hopefully this is also true for AMD with (Ry)zen.

Do you have a recent comparison of Intel CPUs with/without Iris (Pro/Plus) as well? Looking at laptops/NUCs so comparisons with the low-power 'U' chips would be helpful, too.

[gamerk2]

So using a real-world comparison with these GPUs, there's basically zero practical difference between the two CPUs. Sure the 7600K can add a bunch of FPS to a game that was already running at over 200 FPS on the i3 anyway, but there is literally zero real-world impact to that "victory".

Not quite true. While the i3 can often keep up in terms of FPS, it fares significantly worse in terms of latency. That's where those extra cores in the i5 make a huge difference.

[schmidtbag]

Maybe if you were using more powerful GPUs, like a GTX-1070 or 1080, and used more intensive games we could get a worthwhile comparison between these chips. These benchmarks, however, are pretty meaningless.

You are so biased against AMD that you don't even realize how contradictory or simply wrong you are:
First of all, the Fury is plenty powerful enough. It's not much worse than the 1070, so no, getting a more powerful GPU is not really going to make that big of a difference.
The 1080 would achieve frame rates even higher than 200FPS, so why is that of interest to you?
AMD drivers are known to have more CPU overhead. I'd say Nvidia would actually be a worse choice for CPU benchmarks.
You whine about there being no practical difference at 200FPS yet you somehow think the tests will show more useful results when testing more GPU intensive games (which reduces CPU load) and again, on drivers that are also less CPU intensive. Sure, the test may prove the i3 makes more sense (economically) but it is otherwise very boring and doesn't prove much.

You want meaningful benchmarks, and yet you completely lose sight that this is testing CPUs, not GPUs. I'm not saying Michael shouldn't test Nvidia, in fact I'm very interested in those results. But the difference is you somehow think your idea for benchmarks will add meaning to a CPU test. Even on Windows, some crappy FX series CPU can keep up with a modern 6-core i7 when benchmarking modern titles with a GTX 1080.

[Xicronic]

Michael, do Kaby Lake chips significantly improve Radeon gaming performance relative to older generations like Sandy and Ivy Bridge?

[chuckula]

You are so biased against AMD that you don't even realize how contradictory or simply wrong you are:
First of all, the Fury is plenty powerful enough.

What planet did you come from your koolaid drinking shill?
Have you bothered to look at ANY of the benchmarks that have been posted on here for the last year or so? AMD has major performance problems under Linux. MAJOR.

That Furry that you are so hot for* has the same theoretical computer power as a GTX-1070 but loses by a large margin in practically every benchmark that's ever been run here.
It's not even close.

Hell, not even bothering to leave the confines of this article, that Furry is literally losing to the far less powerful Rx 480, and the Rx 480 barely manages to beat the far less powerful Rx 460. The support is absolute crap and if Nvidia had reasults similar to this, the douchetard shills like you would be wetting yourselves crying about how Nvidia has "gimped" its older cards. But when AMD does worse than that.... acknowledging facts that a four year old could plainly read from a chart is being a "fanboy".

As for the Rx 480, it's a power hog that's supposed to be 40 - 50% faster than the GTX-1060 on paper. Doesn't work out that well in the real-world though.

* And by "hot", I'm talking about the almost irresponsible power consumption levels for a rather pedestrian level of performance.

[Michael]

No Nvidia tests? On the Windows side it's well known that Nvidia's drivers scale much better with additional physical threads and you'd think that with their proprietary drivers sharing much of the code base this would also happen on Linux. AMD on the other hand is known to scale pretty badly, but it's an almost completely different code base so you can't really draw any Linux conclusions based on it.

So let's hope there's a follow-up to this using a GPU from Nvidia and their proprietary drivers.

Yes of course follow-up comparison will have NVIDIA results, these were just some results for those curious about Vulkan and RadeonSI/RADV in particular. Doing more tests when my 7700K arrives and I also have Celeron/Pentium Kaby coming for more entertainment.

[nomadewolf]

Just my 2 cents:
On intel processores from i3 (including) up, what really counts are the cores.
Bottom line: if you're unsure about two processors with the same number of cores and same technology, of course, just get the cheapest.