1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Nvidia Inspector Frame Rate Limiter different versions explanation?

Discussion in 'Videocards - NVIDIA GeForce Drivers Section' started by 321Boom, Jun 25, 2017.

  1. MaxiJazz

    MaxiJazz Member

    Messages:
    19
    Likes Received:
    0
    GPU:
    Geforce GTX 1070
    CPU Limiters is heavy for CPU?
    Let me explain, i have 2500k@4400 with gtx 1070+G-Sync 144hz display and playing Diablo 3.
    When i'm using RTSS lock to 60 fps - game has 1 frame (or smth) stutter every 5-10 second.
    When i'm using Inspector V2 lock - stutter free, perfect smooth gameplay.
    Without lock my fps 99% of time is 110+, so in teory CPU can it run with RTSS lock 60, but it's stutters.
    Other games is same, but more harder to see stutter (barely visible).

    Windows 10 1603 and 1709 was tested, same results for me.
    This is CPU side problem or what?
     
  2. Agent-A01

    Agent-A01 Ancient Guru

    Messages:
    11,360
    Likes Received:
    896
    GPU:
    1080Ti H20
    RTSS does not use a significant amount of CPU.

    As for stuttering, you can expect stutters with that 60fps lock, especially if gsync is not working properly.

    Do you have Gsync enabled plus driver enabled Vsync?
     
  3. MaxiJazz

    MaxiJazz Member

    Messages:
    19
    Likes Received:
    0
    GPU:
    Geforce GTX 1070
    Yes, G-sync is activated + v-sync nvcp.
    I'm even tried 2500k without OC, so this is not throttling. (Temps is fine)
     
  4. fry178

    fry178 Maha Guru

    Messages:
    1,142
    Likes Received:
    124
    GPU:
    MSI 1080 X@2GHz
    @Haftarun8
    fast sync in Nv panel, capped 2-3fps below refresh + V2 in Nv inspector, ingame vsync off
    seems to work best for that.
     

  5. Martian

    Martian Member

    Messages:
    18
    Likes Received:
    2
    GPU:
    GTX 1080 8GB 12Gbps
    So Forza Horizon 4, kinda different story if you ask me.
    On the full-game you can't use RTSS, but as it turns out the Inspector limiter seems superior in this case ;)
    Yes, I hate and can't stand any tearing - so VSync it is.

    I really don't know if I can trust the in-game benchmark metrics for the "Average Latency".
    It feels like it's telling truth as far as I can say. The results were constant, very small diff. between tests.
    Would be great if someone could prove this with real-life-test measurements (HS camera, button click etc.)

    Setup (SW):
    - RS4 1803 17133.1 (disabledynamictick, nx AlwaysOff, FSE disabled)
    - nvidiaProfileInspector 2.3.0.2 (modified FH4 profile)
    - NV Geforce 430.64 (Display/HDaudio drivers only)
    [​IMG]
    Setup (HW):
    - Asus Prime Z370-A
    - Intel Core i5-8600K
    - Crucial Ballistix Sport LT 16 GB Kit CL16
    - EVGA GTX 1080 FTW 2 Gaming iCX
    - A-DATA SX8200 Pro 1024GB
    - Corsair HX850i
    - LG 29UB67 2560x1080 60Hz
    Forza Horizon 4: Full Game (no RTSS support) + Demo (RTSS support)
    2 screenshots of in-game benchmark result screen for each config run


    1) Inspector only
    - 59 fps, Limiter V2 - Force Off [Enables Limiter V1 (Less Input Lag)]
    - VSync: Force On Standard
    - No VSync, No limiter in-game
    - Average Latency:
    ~46.7 ms (Full game)
    ~45 ms (Demo)
    [​IMG] [​IMG] [​IMG] [​IMG]

    2) Inspector + In-game fps limit
    - No fps limiter, Limiter V2 - Force Off [Enables Limiter V1 (Less Input Lag)]
    - VSync: Force On Standard
    - No VSync, 60 fps limiter in-game
    - Average Latency:
    ~53.65 ms (Full game)
    ~46.35 ms (Demo)
    [​IMG][​IMG][​IMG][​IMG]

    3) In-game only
    - No fps limiter, No fps limiter mode
    - VSync: 3D Application setting
    - VSync ON, 60 fps limiter in-game
    - Average Latency:
    ~66.45 ms (Full game)
    ~67.95 ms (Demo)
    [​IMG][​IMG][​IMG][​IMG]

    4.1) RTSS + In-game (Demo only!)
    - No fps limiter, No fps limiter mode
    - VSync: 3D Application setting
    - VSync ON, 60 fps limiter in-game
    - RTSS limit (Limit=59989, LimitDenominator=1000)
    - Average Latency:
    ~54.55 ms (Demo)
    [​IMG][​IMG]

    4.2) RTSS (Demo only!)
    - No fps limiter, No fps limiter mode
    - VSync: Force On Standard
    - No VSync, No limiter in-game
    - RTSS limit (Limit=59989, LimitDenominator=1000)
    - Average Latency:
    ~54.95 ms (Demo)
    [​IMG][​IMG]
     
  6. Prabin Sharma

    Prabin Sharma New Member

    Messages:
    1
    Likes Received:
    0
    GPU:
    MSI GTX 980 TI 6GB
    I heard lot of guys talking about latency on Nvidia Profile Inspector (NPI) and Riva Tuna Statistic Server (RTSS). Here's what i have to say, RTSS is not GPU-level limiting, it is same like in-game limiting that's why it does not produce latency because the GPU is still sending those maximum frames even if you have 60fps limiter set on RTSS. Limiting on RTSS and limiting on In-Game is the same thing. RTSS does not have input lag because it is not limiting the frames send by GPU, it is only limiting the frames of your monitor. RTSS is useless if you want to fix screen tearing. But what NPI does is it limits the GPU from sending more frames than specified value. i.e, if you put 60 fps max on NPI, the GPU is limited to produce only 60 frames that's why it generates little bit of input lag. The input lag created by Vsync is FAR more worse then created by NPI. Also NPI completely solves the screen tearing issue making your game look much smoother than before. If you have 60hz monitor and you get above 100fps on most of the games, your monitor will produce a bad screen tearing/lag and in this case even if your frames are at 100fps in your 60hz monitor your game wont feel smooth as 60 fps should feel like. This problem can be solved by NPI by setting fps limit to 60 will little bit of input lag (just little bit you will get used to it) and your can experience your game in true 60fps. But this is not the case in RTSS, limiting your frames in RTSS will make you feel like game got less laggy and more smooth BUT with addition to unbelievably BAD screen tearing which screws your gaming experience. So, for all those folks who have less hz monitor but a GPU powerful enough to push higher fps than your monitor refresh rate then use NPI and set your refresh rate matching to your monitor's HZ and BOOM you experience zero tearing with true 60fps smoothness with just a tiny bit (does not even matter for me) input latency but a LOT less latency than what V-sync produces. That's it folks i think i made everything clear.
    P.S Lastly i want to say this, if you want none of this to be issue (the input lag, the tearing) NOTHING at all, then buy a monitor which has higher refresh rate than your GPU can push. For example if your GPU can push 120 fps max buy a 120hz/144hz/240hz monitor. If your monitor HZ is always higher than the FPS your GPU can produce at MAX, this will never be an issue. A guy who has used 240hz monitor for his lifetime has no idea what screen tearing or input lag is. THATS IT IM DONE
     
    Last edited: Jul 24, 2019
  7. Chastity

    Chastity Ancient Guru

    Messages:
    1,946
    Likes Received:
    479
    GPU:
    Nitro 390/GTX1070M
  8. André Moraes

    André Moraes New Member

    Messages:
    1
    Likes Received:
    0
    GPU:
    1080ti
    I think i'm getting even lower latency with NVIDIA inspector Vsync On and In-game or RTSS limiter to 59,96fps for 60hz Monitor.
     

Share This Page