[Official] DX9 vs DX11 performance difference

Discussion in 'Technical & Support' started by 4thworld, May 4, 2017.

Thread Status:
Not open for further replies.
  1. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    Any chance to do the same test with dx9?
     
  2. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,854
    Likes Received:
    2,234
    Nope, no VR in DX9 :)
     
  3. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    I know but you can do it with fraps, can't you?
    Did you get rid of all monitors?
    In order to test display mode effect you don't need VR.
     
  4. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,854
    Likes Received:
    2,234
    I'm Benchmarking VR not Monitors, I posted my Results based on what happens with VR (i.e. nothing!) to show the difference.
     
  5. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    In DX9 only one graphics card is available showing only 7 available resolutions
    In DX11 three different cards appear showing 13 available resolutions. Best results (posted ones) selecting first option.
    Is this normal? I don't think so.
    upload_2017-6-6_1-3-18.png
     
  6. BlackGoku

    BlackGoku Registered

    Joined:
    Dec 2, 2016
    Messages:
    116
    Likes Received:
    47
    I think and I hope you will fix it, but at this momment my game is unplayable in DX11 while is +100 fps Ultra in DX9.
     
  7. Marc Collins

    Marc Collins Registered

    Joined:
    Jan 11, 2012
    Messages:
    3,159
    Likes Received:
    162
    Let's guess that you have a 60 Hz monitor with sync on. You aren't actually getting such a steady frame rate. Turn off sync and see.
     
  8. stonec

    stonec Registered

    Joined:
    Jun 19, 2012
    Messages:
    3,399
    Likes Received:
    1,488
    I have exactly the opposite, I have a "secondary" card in DX9 config but not in DX11. I think those duplicate cards come from having many displays connected, one per display.
     
  9. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    Do they also show different available resolutions? For me that is more relevant than the number of cards...
     
  10. Stan

    Stan Registered

    Joined:
    Jan 18, 2016
    Messages:
    166
    Likes Received:
    88
    @Marc Collins Thanks for your answer Marc, as you understood, my monitors are 60Hz (5760*1200), but the fps is only limited by the plr setting MaxFramerate, I don't use any sync (to avoid lags)!
     
  11. Goanna

    Goanna Registered

    Joined:
    Jan 19, 2012
    Messages:
    1,405
    Likes Received:
    953
    Are those resolutions for the same card on both? Curious why one says HD3000 and the other GT555M.
     
  12. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    Yes. That's why I asked when I posted the picture if this is normal. However, it is difficult for me to get an official replay.
     
  13. Goanna

    Goanna Registered

    Joined:
    Jan 19, 2012
    Messages:
    1,405
    Likes Received:
    953
    A quick Google tells me that the hd3000 does not support dx11 so I'm guessing you have both the hd3000 and gt555m, both integrated, one on processor and one on motherboard, which would explain the different supported resolutions.
     
  14. SPASKIS

    SPASKIS Registered

    Joined:
    Sep 7, 2011
    Messages:
    3,155
    Likes Received:
    1,426
    You are right. The problem is that regardless what dX9 says, it is also using GT555M. Older builds showed this info correctly. There is no way that the integrated graphics card for the CPU could run rF2 with the fps I reported with DX9.
     
  15. Marc Collins

    Marc Collins Registered

    Joined:
    Jan 11, 2012
    Messages:
    3,159
    Likes Received:
    162
    So what is the MaxFramerate set at? Another possibility is that you are looking at the new average frame rate indicator instead of the one we actually need (from DX9 and all history of rFactor back to 2005)--live frame rate.

    Or, more likely, that sync is somehow on whether you set it on or not, when using triples.
     
    Last edited: Jun 7, 2017
  16. Stan

    Stan Registered

    Joined:
    Jan 18, 2016
    Messages:
    166
    Likes Received:
    88
    Thanks again for your help @Marc Collins .
    "So what is the MaxFramerate set at?", well... 60, to match the screens.
    Of course I use the "new" ctrl+f, but I do use Fraps also.
    My fps was and still is stable (moving rarely of +/-0.1), just now, with dx11+mutliview, there are stutterings, it's not smooth at all, and as soon as I remove multiview everything run then even faster than with dx9 (15/20%)!

    The problem don't seems to come from the "fake sync" method (MaxFramerate), but rather from the multiview.
     
    Last edited: Jun 7, 2017
  17. Emery

    Emery Registered

    Joined:
    Oct 24, 2010
    Messages:
    3,035
    Likes Received:
    1,654
    Finally got back to a little benchmarking. Looks like we're getting close to parity with triple screens and even solidly in positive territory under light loads!

    settings: Level 4 AA, PPE=off, resolution 5760x1080 multiview, no sync, aggressive threads=off
    system: single GTX 1080, 8 GB VRAM; i7-4770 @3.4 GHz, 32 GB RAM
    view: cockpit camera
    weather: overcast

    Atlanta Motorsports Park, Cobra
    DX9, Avg 88 fps, Min 75 fps, Max 101 fps
    Build 1868941, Avg 87 fps (-1%), Min 75 fps (+0%), Max 100 fps (-1%)

    Palm Beach, Cobra
    DX9, Avg 109 fps, Min 100 fps, Max 122 fps
    Build 1868941, Avg 111 fps (+2%), Min 102 fps (+2%), Max 127 fps (+4%)

    Other thoughts: glad the side panels are no longer flashing! Vehicle labels don't seem to be working in replays... arrows are there, but no names at the top of them.
     
    Last edited: Jun 8, 2017
  18. Marc Collins

    Marc Collins Registered

    Joined:
    Jan 11, 2012
    Messages:
    3,159
    Likes Received:
    162
    So I hope you are no longer surprised that you are getting exactly 60 FPS?

    Set MaxFramerate based on your CPU and GPU performance, not screen refresh rate. Try 100 (or even higher) as a test. You first have to understand what real FPS you can get (hardware versus various quality settings), then fiddle with sync, etc., to eliminate stuttering if you get it.

    Why go higher than your screen refresh rate with FPS? Controller lag. The higher the FPS, the better rF2 will feel, even if your screen(s) limits what you are seeing.
     
  19. Ari Antero

    Ari Antero Registered

    Joined:
    Jul 27, 2012
    Messages:
    1,882
    Likes Received:
    829
    Well, Human brain is wonderful and it learns to handle input lag, but brain has a limit, it don`t handle well bad looking game, tears, stutters, flashing textures etc. I believe that if you have clean OS, good hardware, VIDEO sync on, capped fps and game running steady 60 fps you even notice input lag. I get the best game experience with VIDEO sync on but ofcoure it is up to you which option you feel suits best for you. I don`t claim that have have right I am only charing my my experience about DX11 beta.
    ;)
     
  20. Stan

    Stan Registered

    Joined:
    Jan 18, 2016
    Messages:
    166
    Likes Received:
    88
    @Marc Collins Thanks again Marc, in fact I am not sure anymore that the problem come from Multiview, neither from the fps itself...
    Please, look at this video (which is not mine, but it show the problem very well), have a look at the carousel at the end of the lap, it is drawn very differently in dx9 and in dx11, the stutters are obvious!
     
    TravisRoe666 and Rony1984 like this.
Thread Status:
Not open for further replies.

Share This Page