Search:

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

Page 1 of 2 1 2

Search: Search took 0.22 seconds.

  1. Odd...Can you fill a bug on github ? ...

    Odd...Can you fill a bug on github ?
    https://github.com/supertuxkart/stk-code/issues
    With your gpu, mesa version and log from stk ?

    By the way did you build Mesa with floating point texture ?
  2. Sorry my fault, I removed an "#extension ... :...

    Sorry my fault, I removed an "#extension ... : enable" because catalyst complained about it (it's enabled by default) but I didn't check on Mesa. It's reverted in master.
  3. I wasn't involved in the process of writing the...

    I wasn't involved in the process of writing the initial announcement otherwise I would have asked to remove the note about proprietary driver.

    Mesa drivers run STK fine here with Hawaii last time...
  4. You may need this patch for Mesa :...

    You may need this patch for Mesa :
    http://lists.freedesktop.org/archives/mesa-dev/2014-September/067645.html
  5. Please note that STK uses some OpenGL features...

    Please note that STK uses some OpenGL features when available to increase perf like Compute shader or buffer storage + indirect addressing. There is even an azdo implementation that can be enabled...
  6. I can confirm that it works on Fedora 20 with...

    I can confirm that it works on Fedora 20 with kernel-drm-3.17-wip + glamor 0.6 + patched git ddx, although not reliably.
    I can see gdm login screen, but I soon as I enter my session, Xorg hangs (but...
  7. Note that hawaii works correctly with clover and...

    Note that hawaii works correctly with clover and with radeonfb (ie the tty are "accelerated" and Plymouth works) so it's possible to run Xorg without 3d acceleration (if you use Windows to play game)...
  8. Replies
    12
    Views
    5,054

    I'm a bit surprised by these results to be honest...

    I'm a bit surprised by these results to be honest ; some years ago with pre Fermi card it was necessary to design gpgpu kernel around coalesced memory access, and it was said that developper was...
  9. Replies
    98
    Views
    18,992

    I wanted to answer something like this, but you...

    I wanted to answer something like this, but you did it better than I would ever have.
    By the way, why people bring the achievement of Sarah Sharp in the debate ? It looks like most people in this...
  10. Replies
    135
    Views
    27,551

    Did someone test mir with a non Intel igp ? Last...

    Did someone test mir with a non Intel igp ?
    Last time I tried Ubuntu and mir (a couple of weeks ago) radeon didn't work at all. There was the cursor of course, but the rest of the screen was...
  11. Replies
    59
    Views
    12,110

    OpenCL support on r600 hw is making progress,...

    OpenCL support on r600 hw is making progress, even if the gallium compute page does not reflect it.
    Constant Address and local address are supported ; I'm not sure about private address, I think we...
  12. I think opencl will be more usefull at the ap...

    I think opencl will be more usefull at the ap level than at the os level.
    Some app already uses opencl : afaik gimp already has an opencl backend for gegl, one of blender lighting backend uses it...
  13. Replies
    8
    Views
    2,271

    For end users it does not matter that much,...

    For end users it does not matter that much, however for dev it means that R600 backend is now built and tested automaticaly on a daily basis, which helps spot regressions much faster.
    For instance...
  14. How is the fillrate computed ? If the measured...

    How is the fillrate computed ? If the measured time includes shader compilation, overhead of shader optimizations for the second shader may outperform the benefit of tiling.
  15. Replies
    33
    Views
    6,138

    A good benchmark is not always a slow benchmark,...

    A good benchmark is not always a slow benchmark, some very old game engine are likely to be slow if they are used to render scene too complex wrt the target they were designed for.
    Some custom map...
  16. Do you use Wine to run HL2 Lost Coast ? (HL2 is...

    Do you use Wine to run HL2 Lost Coast ? (HL2 is ported to Linux, but I'm not sure for Lost Coast)
  17. Replies
    23
    Views
    5,786

    Gains are more noticeable in Unigine Heaven and...

    Gains are more noticeable in Unigine Heaven and Lightmarks. (Xonotic/Doom/Warsow aren't really shader limited).
  18. I'm a bit surprised with Unigine results, in the...

    I'm a bit surprised with Unigine results, in the Heaven demo there is a significant performance improvement due to better shader, with llvm or sb backend.
    Lightmark 2008.2 is also a benchmark where...
  19. Replies
    14
    Views
    5,364

    It seems related to...

    It seems related to https://bugs.freedesktop.org/show_bug.cgi?id=63632 . Can you provide some info on your config on buzilla ?
  20. The performance aren't that bad for a proof of...

    The performance aren't that bad for a proof of concept, IIRC Tom get 98Mh/s which is 60% of fglrx throughput. Some OpenCL intrinsic are not optimized at all, for instance we fetch thread block size...
Results 1 to 20 of 32
Page 1 of 2 1 2