[Official] DX9 vs DX11 performance difference

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

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

    stonec Member

    Joined:
    Jun 19, 2012
    Messages:
    3,286
    Likes Received:
    1,402
    I wonder why the performance varies so much depending on the car I drive. I measured the FPS on starting grid at Silverstone with 15 cars, and these are the results I got.

    Radical: 44 FPS
    USF2000: 62 FPS
    Civic: 57 FPS
    FISI: 72 FPS

    Especially the Radical is striking, almost twice less FPS than with the FISI. Is it a matter of tuning LODs or are the shaders that different? Surely Radical shouldn't be that much lower than the USF, both have less complexity in the model than the Civic for instance. I think in DX11 optimizations we speak about a few % of gains, whereas with a car like Radical we lose 20%, so maybe the optimization needs to be more focused on content.
     
  2. demerzel

    demerzel Registered

    Joined:
    Aug 30, 2015
    Messages:
    197
    Likes Received:
    127
    Because the quality of the content in rF2 is not consistent.
     
    DaVeX and vegaguy5555 like this.
  3. Emery

    Emery Registered

    Joined:
    Oct 24, 2010
    Messages:
    2,805
    Likes Received:
    1,398
    There are two components to fps performance on any given track: the car you drive (cockpit view) and the cars you race against. Some cars have an fps-intensive cockpit (Cobra, Radical). Others, like the GT2 Corvette, are very light in the cockpit, but more fps-intensive if they're an opponent.

    I haven't measured the Radical, but it could be that cockpit fps are heavy and it's fps-light as an opponent. Or it could be fps-heavy on both sides.
     
  4. not_him_59

    not_him_59 Registered

    Joined:
    Jan 20, 2017
    Messages:
    60
    Likes Received:
    31
    I want to report something to the devs that I am very very happy about. Last night at REVS following our tues night club race we set up and tried the new version of Sao Paolo with the new Nissan GT500, at night. With a total of 20 cars running on my tripple screens, I was maintaining a (better) min of 90 FPS, the motion was silky smooth and the lighting (of DX11) is just gobsmackingly beautiful and so convincing. :D With the FFB now working properly for me I'm very happy to be giving lots of attention to this sim again. Thankyou devs, great job.
     
    Obbzy_XR_ and vegaguy5555 like this.
  5. Risto Kappet

    Risto Kappet Registered

    Joined:
    Feb 14, 2012
    Messages:
    207
    Likes Received:
    338
    Untitled2.png

    Unofficial mods: GPVWC 2017 Superleague formula car, Nürburgring GP track.

    Specs: Athlon X4 870K 3,9ghz, AMD R9 270 2GB, DDR3 RAM, Windows 7, 1920x1080 borderless mode.
    All settings at lowest and with many identical player.json tweaks on both DX versions.

    Summary: DX9 runs far smoother and averages higher. You can remove ~5 fps more from DX11 if trackmap was running. Black line is with agressive threading off, gray line with agressive threading on. No reason to move to DX11 so far but on the first releases the performance was far worse, so keep up the work. DX11 desperately needs all the UI and trackmap mods to pick up users and increase the userbase I imagine.
     
  6. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,732
    Likes Received:
    2,095
    Please do not test with unofficial mods, anything can be wrong with them from not begging DX11 ready or other bugs that could skew the results, test with official mods only is best.
     
    Ari Antero likes this.
  7. Gijs van Elderen

    Gijs van Elderen Member

    Joined:
    May 1, 2014
    Messages:
    1,396
    Likes Received:
    470
    DX11 ready means textures adjusted to fit in the HDR environment. That applies in DX9 too. There are no DX11 only shaders yet. And i think the mods doesn't make use of custom shaders.

    I don't see the problem with his test. :)

    If it performs bad, good or buggy in DX9, it will do the same in DX11. Adjusting some textures: I don't think it will have an impact on performance.
     
    The Iron Wolf likes this.
  8. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,732
    Likes Received:
    2,095
    yeah fair point but I still don't think testing with unofficial mods is the right way to go personally, if the car mod isn't up to scratch it may show odd effects in DX11 (as has been seen) which could affect performance.
     
    Ari Antero likes this.
  9. suryaprihadi

    suryaprihadi Registered

    Joined:
    Apr 16, 2017
    Messages:
    87
    Likes Received:
    32
    1 question to S397
    Today's DX11 supposedly can use multi GPU , that is good news.
    I am doing test now, not complete yet....

    But my question is: is there any DX9 update changelog data, like how DX11 change log is stated here :
    https://forum.studio-397.com/index.php?threads/official-dx11-beta-updates.55357/

    I am asking because the previous DX11 update 27/06/2017 Build ID : 1915820 ,
    the DX9 was 1906069. Today with DX11 6/07/2017 Build ID : 1941092 , I am seeing DX9 now
    becomes # 1929188.

    Just curious :)

    Thanks Mod.
     
  10. MacaRonny

    MacaRonny Registered

    Joined:
    Jan 12, 2014
    Messages:
    8
    Likes Received:
    1
    The new build seems to have multi gpu support “ Added support for mulit-gpus and multi-monitor selection, available via drop downs in Launcher config.” Can someone tell me where I can find this launcher config drop downs?
     
  11. Rui Santos

    Rui Santos Registered

    Joined:
    Jan 8, 2012
    Messages:
    823
    Likes Received:
    902
    After a long period without using my PC, fired up and updated rF2 and performance is not good...

    Matsuzaka Circuit, 19 AI USF2000 cars and average 70FPS

    In Video Settings should i keep AUTO or should i choose my GPU?

    DX9 performance was way better... don't know what's going wrong with the development of this sim...
     
    Last edited: Jul 7, 2017
  12. MaD_King

    MaD_King Registered

    Joined:
    Oct 5, 2010
    Messages:
    1,830
    Likes Received:
    611
    To be able to compare with DX9, you need to disable the Post Effects and check, by default the DX11 version has the Post Effect enable I think and so impact the FPS.
     
  13. Rui Santos

    Rui Santos Registered

    Joined:
    Jan 8, 2012
    Messages:
    823
    Likes Received:
    902
    Post effects are those on video settings? I have on Ultra, my GPU should handle it i think!
     
  14. demerzel

    demerzel Registered

    Joined:
    Aug 30, 2015
    Messages:
    197
    Likes Received:
    127
    Drop down to medium. That's the official recommendation. Ultra is for screenshots.
     
    Rui Santos likes this.
  15. Drathuu

    Drathuu Registered

    Joined:
    Oct 14, 2010
    Messages:
    350
    Likes Received:
    98
    upload_2017-7-7_21-25-1.png

    upload_2017-7-7_21-25-26.png
     
    MacaRonny and Rui Santos like this.
  16. Drathuu

    Drathuu Registered

    Joined:
    Oct 14, 2010
    Messages:
    350
    Likes Received:
    98
  17. Rui Santos

    Rui Santos Registered

    Joined:
    Jan 8, 2012
    Messages:
    823
    Likes Received:
    902
    So i should change it to MEDIUM right? The image quality loss is high? Will test it later, thank you very much!
     
  18. MaD_King

    MaD_King Registered

    Joined:
    Oct 5, 2010
    Messages:
    1,830
    Likes Received:
    611
    If you just want to compare the FPS difference bewteen DX9 & DX11, I think you need to set Post Effects : None
    If you want to get better FPS than your previous test, yes put to Post Effects : Medium
     
    Rui Santos likes this.
  19. Rui Santos

    Rui Santos Registered

    Joined:
    Jan 8, 2012
    Messages:
    823
    Likes Received:
    902
    So i guess DX11 it isn't very well optimized yet, usually DX11 offers much better graphics with better performance, we have other examples i won't mention here, so we'll have to wait for future updates...

    Thanks!
     
  20. Drathuu

    Drathuu Registered

    Joined:
    Oct 14, 2010
    Messages:
    350
    Likes Received:
    98
    VERY BASIC - it is more technical than this and im not over the intracies to the level needed to code directly - but

    You are rendering polygons.. you do math, convert 3d points to a 2d image, you do some more maths to work out where light is to determine where / how triangle is shaded. you shade triangle with texture map bump map spec map.. Dx9 to DX11.. almost the same.. (Same math same process). (So should take identical time and hence identical frames)

    However DX11 then offers advantages with tesselation (RF doesnt use it).. Post processing effects and the ability to handle more complex shaders

    So you should find (Post processing off) Dx9 - dx11 very similar frames in all instances. You will find with PP on any level higher than off you will be a little worse off..

    The equation changes if you start using some custom tailored shaders for DX11.. then you may see performance gains.. however none of the current content uses them. (as it wouldnt work on people using dx9 version).

    Put simply, if they abandon DX9, Tailor it only for DX11 and everyone reworks content with new more complex shaders, you could get better performance improvements - but you will allienate any DX9 users or lose them altogether.

    So best case dx9 to dx11 with no effects = framerate the same. Custom content built with custom dx11 shaders (And will only work on dx11) will be faster - Of which there is none.

     
    Last edited: Jul 7, 2017
    Rui Santos likes this.
Thread Status:
Not open for further replies.

Share This Page