Announcement

Collapse
No announcement yet.

RadeonSI/RADV Mesa 17.3 + AMDGPU DC vs. NVIDIA 387.12 Linux Gaming Performance

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

  • Strunkenbold
    replied
    Originally posted by pal666 View Post
    acutally, in dxmd vega 21% faster than fury
    if you will play only select games and will not use any new hardware features( like double vram)
    I think you are missing an important point, Michael is just testing a Fury non X. The performance delta should be even higher between Fury and Vega 64, approx 30-35%. The driver is either very well optimized for the Fury cards or there is something very wrong with Vega. Thats why Im asking if there is still room for improvement left for them.
    On those benchmark figures:
    DeusEX: MD LQ: Fury non X reaches 84% of Vega performance, the Vega 64 is 18% faster
    DeusEX: MD HQ: Fury non X reaches 82% of Vega performance, the Vega 64 is 21% faster
    MadMax Scene 1: Fury non X reaches 93% of Vega performance, the Vega 64 is 8% faster
    MadMax Scene 2: Fury non X reaches 96% of Vega performance, the Vega 64 is 4% faster
    MadMax Scene 3: Fury non X reaches 91% of Vega performance, the Vega 64 is 10% faster
    MadMax Scene 4: Fury non X reaches 92% of Vega performance, the Vega 64 is 9% faster

    If Michael would have tested a Fury X, results would be on par. Thats strange IMO.

    Leave a comment:


  • oleyska
    replied
    Originally posted by starshipeleven View Post
    While with a 4k on 43-ish inches you get 4 times the space (4k is 4 times the resolution of fullHD and I fully expect to be able to use it), which is my main point. I'm not buying a 4k screen to strain my eyes and still not use even half of that high res.

    And contrary to popular opinion, I've seen enough 4k displays and TVs too.
    you're not straining your eyes with higher dpi, you're making it easier on the eyes.
    Adjust scaling if it's too small.
    One adjusts scaling and not resolution.

    Leave a comment:


  • starshipeleven
    replied
    Originally posted by zanny View Post
    There is a subtle fuzziness in text at 95-105 DPI and 2-3' view distances that goes away completely upping it to ~160 DPI.
    If your eyes are at less than 20-ish cm from your screen (at around FullHD resolutions), you're using it wrong.

    Ergonomics seems to be an unknown concept.

    Leave a comment:


  • starshipeleven
    replied
    Originally posted by tomtomme View Post
    Spoken like someone who has no 4k screen. Bravo.
    If you had you would know that you only need to scale up a bit because of the higher dpi. Everything is so much sharper, so it can bei smaller. You may Not double your space but you get like 1.5 Times the space.
    While with a 4k on 43-ish inches you get 4 times the space (4k is 4 times the resolution of fullHD and I fully expect to be able to use it), which is my main point. I'm not buying a 4k screen to strain my eyes and still not use even half of that high res.

    And contrary to popular opinion, I've seen enough 4k displays and TVs too.

    Leave a comment:


  • leipero
    replied
    Originally posted by FireBurn View Post
    Can you also try with the xf86-video-amdgpu driver rather than modesetting, you're running 1.19.3 which was released back in March
    In my experience xf86-video-x is better for perception, but modesetting shows better FPS most of the time, regardless of that I still prefer xf86 for xorg.

    Leave a comment:


  • tomtomme
    replied
    Originally posted by pal666 View Post
    of static picture surely. how about modern games 4k vs 2.5k ?
    also staggering. 4k is a no brainer. you just have to see it. I did not believe it myself until I started 4k gaming a year ago. Even 2d indy-platformers look so much sharper / better. and the price difference to a fullHD monitor is not even very high nowadays. Only competitive gamers have to pay high prices to also get 120+fps

    Leave a comment:


  • oleyska
    replied
    Originally posted by pal666 View Post
    of static picture surely. how about modern games 4k vs 2.5k ?
    I was talking about games mainly... I was so surprised cause I thought of it more as a gimmick but I stand corrected.
    I bought a 4K monitor for desktop use not for gaming but I immediately saw the massive difference in a lot of games, Wolfenstein New order, Witcher2 is the ones I definitely can confirm a big difference on.

    Leave a comment:


  • pal666
    replied
    Originally posted by Strunkenbold View Post
    In MadMax and Deus:Ex, the performance delta is a joke.
    acutally, in dxmd vega 21% faster than fury
    Originally posted by Strunkenbold View Post
    Right now it makes no sense to buy a new Vega when you can get used Furys for 250€ less and have little to no performance loss.
    if you will play only select games and will not use any new hardware features( like double vram)

    Leave a comment:


  • pal666
    replied
    Originally posted by oleyska View Post
    I use a 28" screen and the quality compared to 1440P on a 28" is absolutely staggering.
    of static picture surely. how about modern games 4k vs 2.5k ?

    Leave a comment:


  • MrCooper
    replied
    Michael will you ever stop perpetuating the myth that the triangle test is "purely limited by video memory bandwidth"? Just for one thing, Vega doesn't have more memory bandwidth than Fiji.

    Leave a comment:

Working...
X