Part of the Khronos Group
OpenGL.org

The Industry's Foundation for High Performance Graphics

from games to virtual reality, mobile phones to supercomputers

Search:

Type: Posts; User: Aleksandar

Page 1 of 20 1 2 3 4

Search: Search took 0.02 seconds.

  1. Here are some new results with the same datasets...

    Here are some new results with the same datasets on GTX470 and GTX850M:





    What can we conclude from this results?

    1. Shader optimization is going beyond shader boundaries in monolithic...
  2. Yes, it could be interesting to see what's...

    Yes, it could be interesting to see what's happening, but I'll rather try something that malexander proposed...


    Very useful tip! Thanks!

    I haven't had time today to carry out all intended...
  3. That is not a belief. A HW tessellation is not...

    That is not a belief. A HW tessellation is not possible in OpenGL 3.3. You need to buy new hardware if you want to play with it.
  4. What actually does GL_RASTERIZER_DISCARD

    Hi All,

    The title of this thread is a little bit cryptic, but I'll try to explain it immediately.

    According to all sources, glEnable(GL_RASTERIZER_DISCARD) should disable rasterization and...
  5. Replies
    4
    Views
    154

    On AMD it is called "AMD Hybrid Graphics". I have...

    On AMD it is called "AMD Hybrid Graphics". I have no idea how to choose GPU using that technology since I don't have an AMD card. Try to google. Each technology/vendor has its own way for choosing...
  6. Replies
    4
    Views
    154

    For the Optimus platform, just add: extern "C"...

    For the Optimus platform, just add:

    extern "C" { _declspec(dllexport) DWORD NvOptimusEnablement = 0x00000001; }
    to activate NV graphics card.
  7. Replies
    2
    Views
    235

    Actually, GTX470 has 640MB L2 cache, since L2...

    Actually, GTX470 has 640MB L2 cache, since L2 cache is associated with the memory controllers on Fermi architecture.
    Fermi also has a 12kB of texture cache per SM. On Maxwell there is no separate...
  8. Replies
    2
    Views
    235

    NVIDIA Maxwell texturing performance

    Hi All,

    Recently I have noticed really unusual behavior which leads me to a conclusion something substantially changed in texture handling in recent NVIDIA GPUs. (No, I don't mean "bindless...
  9. Since it is for illustration purpose only, you...

    Since it is for illustration purpose only, you could draw CPU and GPU time diagrams separately (of course, preserve a sense of latency by moving GPU time-line a little bit to the right). On the other...
  10. Oh, I have forgotten to post a code, in a case...

    Oh, I have forgotten to post a code, in a case someone want to reproduce the test.
    Here it is:


    double GLRenderer::CPU2GPUCorrelation()
    {
    unsigned int id;
    GLuint64 startTimeGPU;...
  11. In spite of being in a rush, I implemented what...

    In spite of being in a rush, I implemented what you proposed and ... Results are useless!
    I read the difference between CPU and GPU time stamps at the beginning of the application (it was...
  12. Yes. That's the way to see a variation in...

    Yes. That's the way to see a variation in reading. If difference between CPU and GPU time varies a lot, it is very hard to estimate the values.


    Your assumption is wrong. There is no such delay....
  13. Do you mean "chapter 34. The OpenGL Timer Query",...

    Do you mean "chapter 34. The OpenGL Timer Query", written by Chris Lux?
    It is a nice article, but it doesn't address CPU-GPU correlation, something Frank want to do.
    If it is some other chapter,...
  14. Yes, timer_query is a part of the core...

    Yes, timer_query is a part of the core specification since ver. 3.3, but I like to use Registry instead of searching through the core specification.


    Probably it is a quite correct statement, bur...
  15. We are not reading the same spec probably. :) ...

    We are not reading the same spec probably. :)


    What you could eventually do is to measure time between reaching the server (not quite clear what does it mean, i.e. which command-queue has been...
  16. Alfonse is not active on the forum for a long...

    Alfonse is not active on the forum for a long time. It's a pity. The forum has lost much of its liveness with his retreat.

    Back to the question. Although I'm not an expert, let's try to find an...
  17. Replies
    9
    Views
    1,001

    Can you provide some proof of this? A screen shot...

    Can you provide some proof of this? A screen shot or something else.
    Because, with that settings the polygons are scattered all over the scene much before 30km. Take a look at the link....
  18. Replies
    9
    Views
    1,001

    That is not an extremely huge terrain. Even if...

    That is not an extremely huge terrain. Even if the viewer stands on the edge of that block, the whole terrain should be visible without artifacts with 24-bit depth buffer by setting the front...
  19. Replies
    5
    Views
    667

    Of course! texture2D() samples texture at (r,a),...

    Of course! texture2D() samples texture at (r,a), where r is actually x-coordinate and a is y-coordinate in Cartesian coordinate system, starting from the lower left corner of the texture.
    But (r,a)...
  20. Replies
    1
    Views
    342

    The reason is in the fact that 4.0.0.14142...

    The reason is in the fact that 4.0.0.14142 version (and probably earlier versions also) of PerfKit cannot create a context on Optimus platforms. :(
    It is a known issue. I'm crossing fingers and...
  21. Replies
    5
    Views
    667

    And I don't understand what you don't understand....

    And I don't understand what you don't understand.
    There is a full shader code with description of all uniform variables used in the code.
    There is no hidden internal work. GL_TEXTURE_WRAP_ is...
  22. Replies
    1
    Views
    342

    NV PerfKit Context Creation Issue

    I have recently downloaded the latest NV PerfKit (4.0.0.14142) and tried it on my new laptop with GTX850M (Win 8.1 64-bit).
    Unfortunately, I have to report a problem in the initialization process....
  23. Replies
    10
    Views
    1,272

    I'm sorry if I'm rude, but I think you are...

    I'm sorry if I'm rude, but I think you are wasting your time.

    The post-transform cache size vary from GPU to GPU, from generation to generation. There is no purpose to make something that works...
  24. Very nice! Thank you NV for the instant support! ...

    Very nice! Thank you NV for the instant support!
    Maybe it would be nice to publish notebook drivers too.
  25. Replies
    30
    Views
    7,005

    When I talked about the mem cache, I didn't mean...

    When I talked about the mem cache, I didn't mean graphics card memory. The cache is in the main memory. So, use regular way to cut proper rectangle from it and send to the texture.
    You won't have...
Results 1 to 25 of 500
Page 1 of 20 1 2 3 4