Core i7 8700K vs. Ryzen 7 1800X For NVIDIA/Radeon Linux Gaming

[VikingGe]

What? Have you looked at a Linux game and found it only uses one process? I know OpenGL is supposedly unable to make use of multiple command queues (or something like that), but that doesn't mean the rest of the application can't be multithreaded.

It does however mean that the main rendering thread is more likely the main bottleneck, especially since a lot of overhead seems to be added anyway when porting from D3D11 to OpenGL.

Wine kind of takes that to the extreme, The Witcher 3 is easily four times as fast on Windows as it is on wine on my machine, and while it can easily reach near 100% utilization on my 6-core CPU on Windows, it never goes above 70% on Wine thanks to one single thread holding it back.

[gurv]

Sigh, another case of Ryzen inter-ccx latency problem (unless Michael ran the 1800x with low speed DDR4 which could also be a problem. The review states that 3200 ram was used but that doesn't mean it was running at that speed).
This is seriously harming AMD's brand under Linux

It's a pity since all it generally takes to solve is a simple taskset -c 0-7 %command% in Steam.
I'm especially disappointed by Feral : Dawn of War 3 was released well after Ryzen.
They should have known by then that scheduling is really important with Ryzen.

The 8700k has a 4.3ghz all core frequency and a 7 to 10 % IPC advantage.
So it should be around 4.3/3.7 * 1.1 ~= 30% faster in games (because no game use more than 6 core / 12 threads)
It could reach 40% in extreme cases but not more.
That's why these benchmarks with a ludicrous 70% lead for the 8700k are not representative of the performance that Ryzen is capable of.
Unfortunately for AMD though, they are representative of the out of the box experience

[monte84]

Because there's little point in e.g. benchmarking them all at 1080p high quality or so when each game/engine behaves differently... In a game like Xonotic where it can often run several hundred frames per second, it makes more sense stressing it more at 4K than e.g. Deus Ex at 4K where it would be running at a couple frames per second.

But like the RX 580 review, which is a video card review, Tomb Raider was benchmarked on normal settings, why not a higher quality or higher resolution? Xonotic, i get it, but in this review, what about Deus Ex? it was benchmarked at 1440, which even at 1080 is going to be pretty demanding, but Tomb Raider was 1080 and Normal quality. This is what I am complaining about, there is no consistency. At least give some explanation as to what settings are used and why. Im just trying to let you know, as a consumer, what I would like to see. I would have no problems grabbing premium when some changes get implimented. I know its a lot of work, I have made a few benchmark videos for youtube ad plan on going a bit more in depth with it when I finish classes. Thanks for the response and effort!

[monte84]

Sigh, another case of Ryzen inter-ccx latency problem (unless Michael ran the 1800x with low speed DDR4 which could also be a problem. The review states that 3200 ram was used but that doesn't mean it was running at that speed).
This is seriously harming AMD's brand under Linux

It's a pity since all it generally takes to solve is a simple taskset -c 0-7 %command% in Steam.
I'm especially disappointed by Feral : Dawn of War 3 was released well after Ryzen.
They should have known by then that scheduling is really important with Ryzen.

The 8700k has a 4.3ghz all core frequency and a 7 to 10 % IPC advantage.
So it should be around 4.3/3.7 * 1.1 ~= 30% faster in games (because no game use more than 6 core / 12 threads)
It could reach 40% in extreme cases but not more.
That's why these benchmarks with a ludicrous 70% lead for the 8700k are not representative of the performance that Ryzen is capable of.
Unfortunately for AMD though, they are representative of the out of the box experience

I have tested my 1700 with a 4+0 config (so no inter ccx) and it does nothing for helping the performance on some of the games I have tested, like Dirt Rally, War Thunder, and Shadow of Mordor. Also tried disabling SMT, to not much avail. Now, my friend who also has a 1700 noticed performance improvement on his nvidia card, but less so now. Feral needs to do some optimizations i belive.

[Jabberwocky]

It does however mean that the main rendering thread is more likely the main bottleneck, especially since a lot of overhead seems to be added anyway when porting from D3D11 to OpenGL.

Wine kind of takes that to the extreme, The Witcher 3 is easily four times as fast on Windows as it is on wine on my machine, and while it can easily reach near 100% utilization on my 6-core CPU on Windows, it never goes above 70% on Wine thanks to one single thread holding it back.

It's crazy to think that you can gain significant performance when running Windows in a virtual machine. I'm interested in Ryzen Native-Linux vs Ryzen vfio-Windows-guest.

[xpander]

What? Have you looked at a Linux game and found it only uses one process? I know OpenGL is supposedly unable to make use of multiple command queues (or something like that), but that doesn't mean the rest of the application can't be multithreaded.

one process yes. most games just push rendering to another thread, so it makes them use 2 threads, ofc there are games that take advantage of more, but 90% of opengl games wont use more than 3 threads, thats what i have seen at least

[pheldens]

https://www.cpubenchmark.net/singleThread.html many games rely on single thread performance, intel is just 30% faster in that area, ask Arma III users.

[M@GOid]

But like the RX 580 review, which is a video card review, Tomb Raider was benchmarked on normal settings, why not a higher quality or higher resolution? Xonotic, i get it, but in this review, what about Deus Ex? it was benchmarked at 1440, which even at 1080 is going to be pretty demanding, but Tomb Raider was 1080 and Normal quality. This is what I am complaining about, there is no consistency. At least give some explanation as to what settings are used and why. Im just trying to let you know, as a consumer, what I would like to see. I would have no problems grabbing premium when some changes get implimented. I know its a lot of work, I have made a few benchmark videos for youtube ad plan on going a bit more in depth with it when I finish classes. Thanks for the response and effort!

I had told Michael to push up Tomb Rider a couple times. On my machine (i7 3770k + RX470) I can run ~ 60 fps on the benchmark, on Ultimate, with the hair thingy on. I'm sure the much more powerful systems Michael has can deliver close to 100 fps, on a high end machine.

[pinguinpc]

Michael

Thanks for add 1080p tests

See results intel is 1080p* game king

*Without forget 1080p is more desktop used resolution, in many cases at 144hz

Hopefully amd can improve for ryzen 12nm

https://www.cpubenchmark.net/singleThread.html many games rely on single thread performance, intel is just 30% faster in that area, ask Arma III users.

Yeah for disgrace amd single thread is too low compared with intel**, especially with coffelake*** higher frecuencies

**Sadly ryzen oc have similar single thread of haswell at 4.2 - 4.4ghz but haswell 4.2 and upper stay avalaible since 2014

***Various models can stay around 5.0ghz but with delid have better chances

This situation appears in dolphin benchmark and others for courtesy of anandtech

Dolphin Benchmark:

Many emulators are often bound by single thread CPU performance, and general reports tended to suggest that Haswell provided a significant boost to emulator performance.

This benchmark runs a Wii program that ray traces a complex 3D scene inside the Dolphin Wii emulator.

erformance on this benchmark is a good proxy of the speed of Dolphin CPU emulation, which is an intensive single core task using most aspects of a CPU.

Results are given in minutes, where the Wii itself scores 17.53 minutes.

[Jumbotron]

Both the AMD Ryzen 5 1500X and the 1600 are a very good value according to at least this one benchmark. The Dolphin Emulator is a very nice piece of work.