Search:

Type: Posts; User: przemoli; Keyword(s):

Page 1 of 5 1 2 3 4

Search: Search took 0.04 seconds.

  1. @Michael Are those comments about dx/webgl...

    @Michael

    Are those comments about dx/webgl about gpu profiler or about bugfix to that bug?
  2. "has landed" usually mean that code for that...

    "has landed" usually mean that code for that feature was freshly merged to code repositories for aurora channel. Beta and up get stuff "enabled".
  3. Michael Another test round with some tests...

    Michael

    Another test round with some tests that score very similarly across different GPUs:

    World of Padman

    Can You add CPU usage % (to clarify if that is GPU constrain or CPU)
  4. Those numbers mean CPU is now the bottleneck? ...

    Those numbers mean CPU is now the bottleneck?

    So what about CPU usage stats added to the next such test, Michael?
  5. Replies
    121
    Views
    12,251

    LMAO about that "easy use of C++". Yeah. Right....

    LMAO about that "easy use of C++". Yeah. Right.
    Inner class overloaded functions anyone?

    LMAO about that "speed of C++". Yeah. Right.
    That's why game engine devs steer away from:
    * Boost
    *...
  6. Which mean that we get MORE and MORE solutions...

    Which mean that we get MORE and MORE solutions using ABI we want DEAD & replaced by new model developed by Nvidia?

    Hope that approach is adapted for all APIs exposed by Mesa.
  7. @Michael Another thing to test in next Win vs...

    @Michael

    Another thing to test in next Win vs Lin comparison :D
  8. Michael can You check AI settings next time You...

    Michael can You check AI settings next time You benchmark Lin vs Win?

    This way we could see if that same behavior on both platforms, or if Linux version is stripped.
  9. What there is to "see, analyse, adapt" is FPS? ...

    What there is to "see, analyse, adapt" is FPS?

    Its binary decision "is there head or nor?", after that its reflex of shooting :) (and hearing "head shoooooot!" ;) )

    Not that I'm saying that 120...
  10. Correct. But humans need FEEDBACK. This...

    Correct.

    But humans need FEEDBACK.

    This require 25 FPS minimum.
    And there are people who claim that they can SEE difference up to 60 FPS.

    Lack of feedback => more discomfort.
  11. OpenGL Next can focus solely on CPU side of...

    OpenGL Next can focus solely on CPU side of things*.

    AZDO, DSA, glfush/context work etc. are good enough for GPU... For graphics.

    Easier integration with (big) compute would be nice though.
    ...
  12. Replies
    14
    Views
    1,959

    No. C&P is illegal. You can create Your own...

    No.

    C&P is illegal.

    You can create Your own derivate but only if You stick to LGPL. What You can do is to put that code in .dll/.so, so only that part need to be LGPL'ed

    RTFM (RTFL! ;) )
  13. Disagree. Game devs if using special...

    Disagree.

    Game devs if using special techniques could actually give lower latency to users.

    And human reflex can go as low as 6ms -> 160 fps...

    Yes 120FPS with adaptive refresh rate, with...
  14. Replies
    16
    Views
    2,015

    LOL Witcher 2 have BAD gfx? Man did You...

    LOL Witcher 2 have BAD gfx?

    Man did You SEE the game?
    Anyway Witcher 3 is just around the corner (Q1 2015)...
  15. Good news about AMD/Intel floss strategies is...

    Good news about AMD/Intel floss strategies is that there is abundance of 3rd party or independent devs hacking.

    So if Crossfire become reality for any of AMD floss drivers there should be enough...
  16. "atime" ... breaks POSIX. And if only few apps...

    "atime" ... breaks POSIX.

    And if only few apps use POSIX for atime... then maybe there is reason for that...

    Like abnormal performance penalty, where good alternatives do exist?


    "Wayland...
  17. 260x vs 270x Anyone have those cards here? ...

    260x vs 270x

    Anyone have those cards here?

    How are non-performance things playing out for them? Fans? Video encode/decode? Issues with DEs?

    Currently 270x (with 2GB) is twice as expensive as...
  18. But then they need to rewire their whole infra. ...

    But then they need to rewire their whole infra.

    And I mean WHOLE infra.

    Hardly justifiable.
  19. In theory there is nothing in Mesa/Gallium/LLVM...

    In theory there is nothing in Mesa/Gallium/LLVM that would make such "profiles" impossible.

    Especially hand written shaders SHOULD be reusable between Catalyst and radeon drivers.

    Driver quirks...
  20. Steam Survey also underreport OpenGL version......

    Steam Survey also underreport OpenGL version...
    As it only count Compatibility Profile (OGL 3.0 instead of 3.3 on my 5730M with r600g)
Results 1 to 20 of 100
Page 1 of 5 1 2 3 4