Re: Blur Buster's G-SYNC 101 Series Discussion
Posted: 24 Jan 2023, 21:04
hey can i use Nvidia control panel to limit FPS or should i use rtss. would there be any negatives if i use the frame limiter thats in Nvidia control panel.
Who you gonna call? The Blur Busters! For Everything Better Than 60Hz™
https://forums.blurbusters.com/
This was answered in the article:
*As of Nvidia driver version 441.87, Nvidia has made an official framerate limiting method available in the NVCP; labeled “Max Frame Rate,” it is a CPU-level FPS limiter, and as such, is comparable to the RTSS framerate limiter in both frametime performance and added delay.
Yes, this is normal. NVidia doesn't consider -3FPS to be enough safety margin and they cap lower than that.Jeroenmfb wrote: ↑18 Mar 2023, 15:38Hi everyone,
I'm having an issue with my monitor and NVIDIA driver. I have G-SYNC enabled + VSYNC (only in nvcp) and the monitor runs at 390Hz. However, when I turn on Reflex + Boost to further lower latency and cap FPS below the monitor refresh rate, it locks my FPS to 351 instead of something close but below 390 FPS.
Has anyone else experienced this or have any suggestions on how to fix it?
As RealNC already menitoned, it's normal for Nvidia to auto-limit the framerate with G-SYNC + V-SYNC + LLM "Ultra" or G-SYNC + V-SYNC + Reflex, and the higher the physical refresh rate, the lower they set the limit.Jeroenmfb wrote: ↑18 Mar 2023, 15:38Hi everyone,
I'm having an issue with my monitor and NVIDIA driver. I have G-SYNC enabled + VSYNC (only in nvcp) and the monitor runs at 390Hz. However, when I turn on Reflex + Boost to further lower latency and cap FPS below the monitor refresh rate, it locks my FPS to 351 instead of something close but below 390 FPS.
Has anyone else experienced this or have any suggestions on how to fix it?
Thanks in advance for your help!
PS: I don't know if this is the correct place to post or if I should have posted this under the XV252QF thread
Haven't tested myself, but I've seen reports that nvidia's limiter is very jittery at high framerates. Maybe that's the reason they use lower caps:jorimt wrote: ↑18 Mar 2023, 19:36Now, while it's true that the higher the physical refresh rate, the lower you should probably set the limit due to frametime tolerances narrowing due to a faster scanout time, Nvidia is being pretty aggressive at 390Hz+, it seems. Why that is, you'd have to take up with them.
Sure, could be a factor, but 1) the auto-limiting behavior occurs both with the LLM "Ultra" combo, which uses the driver-level "Max Frame Rate" limiter, and with the Reflex combo, which uses an engine-level limiter per game, and 2) I found in previous testing that the occasional limiter overshoot will not affect average latency unless it occurs sequentially over a period of several frames. I.E. a frame exceeding the set limit here and there won't be enough to revert G-SYNC + V-SYNC to standalone V-SYNC behavior, etc.RealNC wrote: ↑19 Mar 2023, 09:03Haven't tested myself, but I've seen reports that nvidia's limiter is very jittery at high framerates. Maybe that's the reason they use lower caps:
https://forums.guru3d.com/threads/lock- ... st-6113247
Very interesting!drimzi wrote: ↑01 Jul 2022, 10:42I noticed when Low Latency is set to Ultra, it is capping the frame rate to:Is there any significance to this approach or is it arbitrary?Code: Select all
RefreshRate - RefreshRate^2 / 60^2
I set my 240hz to all of the refresh rates it supported to get the resulting FPS limits, and then created a formula.Chief Blur Buster wrote: ↑21 Mar 2023, 02:37Very interesting!drimzi wrote: ↑01 Jul 2022, 10:42I noticed when Low Latency is set to Ultra, it is capping the frame rate to:Is there any significance to this approach or is it arbitrary?Code: Select all
RefreshRate - RefreshRate^2 / 60^2
How did you find this out?
I'd like to know more.
The formula you provided is used to calculate the minimum frame time required to achieve a certain level of refresh rate stability, and it is not directly related to NVIDIA Ultra-Low Latency (NULL) technology. However, I can explain how this formula relates to gaming and how NVIDIA's technology can help improve the gaming experience.
The formula you provided is derived from the Nyquist-Shannon sampling theorem, which states that to accurately reconstruct a signal (in this case, a video frame) from samples, the sampling rate must be at least twice the highest frequency component of the signal. In the context of gaming, this means that to achieve a stable refresh rate, the frame time (i.e., the time it takes for the GPU to process a frame and send it to the display) must be shorter than the inverse of the refresh rate.
For example, if you have a monitor with a refresh rate of 60 Hz, the minimum frame time required for a stable refresh rate would be 16.67 milliseconds (ms), which is the inverse of 60 Hz. The formula you provided takes this concept further by considering the effect of variations in the frame time over time. It calculates the maximum allowable variation in frame time to achieve a certain level of refresh rate stability.
Now, how does NVIDIA Ultra-Low Latency relate to this? As I mentioned earlier, NULL works by reducing the number of frames that are buffered by the GPU before they are displayed on the screen. By doing so, it can reduce the frame time and input lag, leading to a more responsive and immersive gaming experience.
If the frame time is shorter than the inverse of the refresh rate, the refresh rate will be stable, and the formula you provided would not apply. However, if the frame time varies over time, the formula could be used to determine the maximum allowable variation in frame time to achieve a certain level of refresh rate stability. By reducing input lag and optimizing frame processing, NVIDIA Ultra-Low Latency can help ensure that the frame time stays within the allowable variation range, leading to a more stable and smooth gaming experience.