Page 2 of 4 FirstFirst 1234 LastLast
Results 11 to 20 of 38

Thread: GLX DRI3 GPU Offloading Lands In Mesa

  1. #11
    Join Date
    Sep 2012
    Posts
    369

    Default

    Quote Originally Posted by Adriannho View Post
    Everything ran smoothly at max setting at over 35fps.
    FPS games need 60 fps minimum and approx. 90 fps average to be smooth.

  2. #12
    Join Date
    Sep 2010
    Posts
    491

    Default

    Very good progress!
    If the race problem could be solved, that would be swell!
    And make this a solution without any disadvantages!

  3. #13
    Join Date
    Jul 2013
    Posts
    441

    Default

    Quote Originally Posted by tpruzina View Post
    Your definition of 'smoothness' in not compatible with rest of the world (35fps is barely playable for most games and heavily unsufficient for first person games).
    Quote Originally Posted by JS987 View Post
    FPS games need 60 fps minimum and approx. 90 fps average to be smooth.
    Wow, what rich world full of amazingly powerful computers do YOU guys come from? 35fps IS smooth, and FPS games only need 60+fps if you're playing multiplayer (gotta be twitchy and all that), but if you're playing the storyline 35fps is just fine.
    I WISH I could get 35+fps on any game other than League of Legends at the very lowest graphics settings.

  4. #14
    Join Date
    Mar 2014
    Posts
    32

    Default

    Quote Originally Posted by JS987 View Post
    FPS games need 60 fps minimum and approx. 90 fps average to be smooth.
    By what type of measurement? I used fraps for counting frames...can that be considered as accurate?

    Well, I had absolutely no lags, stuttering and such and the response to commands was instant. From my point of view BF4 (and that btw, is a "next-gen" first-person shooter) was running great at 35 fps average...and this on very high setting....and on a cheap mobile system...

    But what do I know...I'm probably wrong.

  5. #15
    Join Date
    Aug 2012
    Location
    Pennsylvania, United States
    Posts
    1,939

    Default

    Quote Originally Posted by Adriannho View Post
    By what type of measurement? I used fraps for counting frames...can that be considered as accurate?

    Well, I had absolutely no lags, stuttering and such and the response to commands was instant. From my point of view BF4 (and that btw, is a "next-gen" first-person shooter) was running great at 35 fps average...and this on very high setting....and on a cheap mobile system...

    But what do I know...I'm probably wrong.
    35+fps on max settings on a relatively cheap system like that is pretty awesome. You're not wrong in that regard

  6. #16
    Join Date
    Jan 2014
    Location
    Wonderland
    Posts
    111

    Default

    Quote Originally Posted by Adriannho View Post
    By what type of measurement? I used fraps for counting frames...can that be considered as accurate?

    Well, I had absolutely no lags, stuttering and such and the response to commands was instant. From my point of view BF4 (and that btw, is a "next-gen" first-person shooter) was running great at 35 fps average...and this on very high setting....and on a cheap mobile system...

    But what do I know...I'm probably wrong.
    Fps means frames per second and ic does what it says it does.
    Main problem with fps is that at 60fps you can render 59 frames in nanoseconds each and that one 'off' frame that takes nearly second to draw will make perception of 1fps total.
    Yeah its not realistic, but in real games you can have smooth 30fps on console and 'laggish' 60 fps on pc.
    For example, in dota2 on my setup i had 70fps, but longest frames could take 1/12th second to draw, while fastests were in ballpark of 1/100s. Hope this example gives a bit of insight into how deceiving fps can be.

  7. #17
    Join Date
    Sep 2013
    Posts
    250

    Default

    a) There's no logic for GPU offloading needed in the Xserver
    Someone, answer please!

    Does it's mean that usage of DRI_PRIME will no longer able to crash my X server?
    Because some games with newer drivers make my X crash when I run games on AMD GPU and same with Nouveau.

  8. #18
    Join Date
    Mar 2013
    Location
    Porto
    Posts
    275

    Default hm

    Quote Originally Posted by _SXX_ View Post
    Someone, answer please!

    Does it's mean that usage of DRI_PRIME will no longer able to crash my X server?
    Because some games with newer drivers make my X crash when I run games on AMD GPU and same with Nouveau.
    x server don't work well with dual graphics and will never do, the devs never give good support

  9. #19
    Join Date
    Sep 2013
    Posts
    250

    Default

    Quote Originally Posted by rikkinho View Post
    x server don't work well with dual graphics and will never do, the devs never give good support
    Why do you think so?

    I use it at moment with Intel HD4600 (used by X) and AMD HD6950 and it's work well, but Nouveau tend to crash and new graphics stack result in crashes on when AMD GPU used. As far as I understand this news correct then when DRI3 used X shouldn't see the difference between renderng on "default" GPU and one used though PRIME.

  10. #20
    Join Date
    Feb 2008
    Posts
    1,360

    Default

    Quote Originally Posted by _SXX_ View Post
    Why do you think so?

    I use it at moment with Intel HD4600 (used by X) and AMD HD6950 and it's work well, but Nouveau tend to crash and new graphics stack result in crashes on when AMD GPU used.
    Is that works for you, intel uses DRI3 but radeon DRI2 . and nouveau being third using by third nvidia card or what .

    Just asking to have some "picture" of that experience, i don't like those multi gpu system i even avoid buying them .

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •