Announcement

Collapse
No announcement yet.

Ryzen 7 2700 / Ryzen 7 2700X / Core i7 8700K Linux Gaming Performance With RX Vega 64, GTX 1080 Ti

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • boxie
    replied
    Originally posted by humbug View Post
    something is wrong with how people are porting games to Linux. On windows it seems that modern games are more multithreaded which allows Ryzen to keep up.

    Something happens when porting and it becomes more reliant on single threaded performance. When you use 4K which is a more GPU than CPU bottlenecked scenario the difference goes away.

    I dunno maybe it's the same reason every game takes a 20% performance hit when porting, even on Intel CPUs.
    That is an interesting point. Thank you!

    Leave a comment:


  • boxie
    replied
    Originally posted by Jabberwocky View Post

    Sorry for taking it out of context, but stock is not what stock was a year or two ago. With coffee lake and zen+ (depending on what motherboard you have) the CPU will overclock automatically past stock speed as long as you have proper cooling.
    Sure - Stock though in this context is "default behaviour" so, if the 2700X will overclock itself a little, as long as there is thermal overhead and therefore better cooling solutions do better on the auto overclock. AMD also have a better thermal solution compared to Intel though... You could go down that rabbit hole for days!

    I guess a better title would be "As Shipped".

    Leave a comment:


  • Jabberwocky
    replied
    Originally posted by boxie View Post
    oh wow, the Ryzen is getting a spanking.

    Why is the 8700k beating out the 2700X by so much?
    It could be cooling related, maybe SenseMI is not working well, RAM compatibility or Michael did not win the silicon lottery.

    these products are being bencmarked at stock
    Sorry for taking it out of context, but stock is not what stock was a year or two ago. With coffee lake and zen+ (depending on what motherboard you have) the CPU will overclock automatically past stock speed as long as you have proper cooling.



    Like darkbasic asked, I would also like to see test at the same frequency. I would like to know if there's still a big difference compared to Windows results.
    Last edited by Jabberwocky; 22 May 2018, 01:59 PM. Reason: added RAM compatibility as possible cause and fixed typo

    Leave a comment:


  • Michael
    replied
    Originally posted by Dr. Banana View Post
    Those Dota 2 benchmarks look flawed. May I ask you exactly what your testing methodology (for Dota 2) is?
    OpenBenchmarking.org / PTS, all transparent.

    Leave a comment:


  • Jabberwocky
    replied
    Originally posted by Leopard View Post
    TL DR ; go with Nvidia + Intel combo if you don't want to deal with choices.
    With regards to CPU I don't agree with you. My AMD 1800X with high capacity&performance RAM is working wonders for me personally (virtual machines for gaming and work concurrently). For TL;DR advice though, this video summarizes it well and I rather recommend watching it opposed to blindly recommending one above the other.



    Results won't be different i guess.

    Since Hitman uses OpenGL and with OGL single core performance is crucial. So that will go to Intel

    Same with Dirt Rally.

    DoW 3 won't be different for Vega cards either since RADV is very bad for Vega gpu's.

    I learned some valuable things since i started to watch RADV and RadeonSI evolution.

    If you want pure performance on very first day you bought a new gpu ; go with Nvidia and blob driver.

    If you don't mind about getting expected performance after 1.5 or 2 years , stick with AMD gpu's on Mesa. If you don't mind LLVM related bugs too also.

    AMD GpuPro driver is a shitty option since it performs bad than Mesa on OpenGL which most games are using it.
    I agree with you more or less on the GPU side. Your criticism is a bit one-sided, but of course it's just your opinion and you might not have had as much problems with Nvidia blob as other Linux members/kernel developers. Linux gaming as improved drastically over the years, but there still are lots of cons in both open source and proprietary solutions. It is what it is.

    My personal experience: I have been gaming on Linux for more than 15 years and used various Nvidia, AMD, Intel and other graphics cards. The longest time I had been a pure Linux gamer was for ~4 years (without dual-booting to Windows). The problem I faced was that Linux is good for work/privacy/security and Windows is good for gaming/spyware. For the most part I only had one desktop. I obviously desired the best of both worlds.

    My personal solution to my desktop gaming problem: Do not mix open source and proprietary drivers. The past 6 years Mesa did not allow me to play the games that my friends played and I usually ended up dual booting, which is another problem of itself. I also tried using two desktops, one of Linux and other Windows. It was great for a while, but not practical in the long run nor something I would recommend for others. A few months ago I tried IOMMU setup and this really is the best of both worlds that I desired. I now have Mesa+Radeon for Linux, my host OS. Windows+blob for my gaming VM. I currently have a Radeon rx 560 (Linux) and rx 480 (Windows VM) but if you need better gaming performance you can just slap a geforce 1080 ti and pass that to windows. I personally have not had the need for it and bought my rx 480 before I did my IOMMU setup. Now I am able to game on Linux and Windows and I don't need to have multiple desktops, ridiculously priced KVM switches. or software that is supposed share my keyboard/mouse over the network without problems. It's the way to go if you have the knowledge and want to game on both platforms using a single desktop. It makes backups much easier too, Windows just a image on my Linux filesystem.

    Linux gaming with laptops: I try to block out any thoughts regarding this, it's my dark place.

    If engineers was able to "streamline" the process so that the average person was able to run Linux and Windows concurrently and switch between the two dynamically, then IMHO Linux would have much more usage and support in the personal computer industry. You need a good transitioning mechanism to move from one technology to another... anyway I have gone completely off topic!

    PS: Thanks for the tests Michael I found the selection of hardware more valuable than the average gaming benchmarks.
    Last edited by Jabberwocky; 21 May 2018, 08:33 PM. Reason: typo

    Leave a comment:


  • abott
    replied
    Wonder what is making the Ryzen chips so abysmal.

    Leave a comment:


  • Leopard
    replied
    Originally posted by Jahimself View Post
    Was waiting for this one. I was expecting r5 2600 to see if there is a difference with windows (2600 is faster than 2700 on windows for gaming)
    We have a lot of triple A title now on linux. It would be interesting to bench heavy title such as Hitman and Dawn of War 3 and also to include Dirt Rally.
    Results won't be different i guess.

    Since Hitman uses OpenGL and with OGL single core performance is crucial. So that will go to Intel

    Same with Dirt Rally.

    DoW 3 won't be different for Vega cards either since RADV is very bad for Vega gpu's.

    I learned some valuable things since i started to watch RADV and RadeonSI evolution.

    If you want pure performance on very first day you bought a new gpu ; go with Nvidia and blob driver.

    If you don't mind about getting expected performance after 1.5 or 2 years , stick with AMD gpu's on Mesa. If you don't mind LLVM related bugs too also.

    AMD GpuPro driver is a shitty option since it performs bad than Mesa on OpenGL which most games are using it.

    TL DR ; go with Nvidia + Intel combo if you don't want to deal with choices.

    Leave a comment:


  • vegabook
    replied
    Funny though how good ol' RX480/580s completely spank Nvidia when it comes to "raw" compute such as mining. It takes a 1080ti get faster compute than an RX580.... and the Vega64 just sails right past even that.

    Leave a comment:


  • pete910
    replied
    CPU difference aside, I took that as a reasonably good showing for the rx64 on the mesa driver

    Leave a comment:


  • BNieuwenhuizen
    replied
    https://www.tomsguide.com/us/nvidia-...view-4241.html
    Originally posted by Azpegath View Post
    Wow, what is going on at 4K for the Vegas? marek, any idea? Michael, could you test that with the official AMD Vulkan driver as well?
    So I think for some of the vulkan benches it indeed plays a role that radv is a bit worse for vega.

    However also remember that it is being compared to a 1080 Ti, not a 1080, and from other benchmarks I can see the Ti tends to push another 20-30% compared to the 1080 on windows: https://www.tomsguide.com/us/nvidia-...view-4241.html

    Generally I'd expect the vega 64 to keep up with the 1080 but not the 1080 Ti.

    Leave a comment:

Working...
X