Higher queued frames Higher Output lag?

[xiaomantou]

I am fighting with someone.I only found one testhttps://i.imgur.com/4lc78kN.png,so I drew some pic and I found the same result.
Just to make sure:

Without any sync,
Does it increase output lag when I add the mumbers of maximum pre-rendered frames(queued frames)?
How many buffers in total?

[RealNC]

With vsync OFF, the max pre-rendered frames setting only has an effect on lag if the GPU is maxed out and becomes the bottleneck, and if the game itself doesn't use some method to limit pre-rendering itself.

This isn't very difficult to test yourself. Use a game that has high GPU load and supports 4K. You can use nvidia DSR to set the game to 4K. Once you do that, you should be able to get the GPU to completely max out and become the bottleneck. It should then be easy to tell the difference between the different max pre-rendered frames settings. The lower the FPS, the more apparent the difference becomes.

The game I usually use for this test is Witcher 3. It easily maxes out my GPU at 4K with high graphics settings.

In previous nvidia driver versions it was much more easier to tell, because you could set MPRF to 4, and a setting of 4 vs 1 had a huge input lag difference. I could also set MPRF to 8 using Profile Inspector. The input lag in that case was beyond huge. About 1/3 of a second of input lag when moving the mouse.

[xiaomantou]

With vsync OFF, the max pre-rendered frames setting only has an effect on lag if the GPU is maxed out and becomes the bottleneck, and if the game itself doesn't use some method to limit pre-rendering itself.

This isn't very difficult to test yourself. Use a game that has high GPU load and supports 4K. You can use nvidia DSR to set the game to 4K. Once you do that, you should be able to get the GPU to completely max out and become the bottleneck. It should then be easy to tell the difference between the different max pre-rendered frames settings. The lower the FPS, the more apparent the difference becomes.

The game I usually use for this test is Witcher 3. It easily maxes out my GPU at 4K with high graphics settings.

In previous nvidia driver versions it was much more easier to tell, because you could set MPRF to 4, and a setting of 4 vs 1 had a huge input lag difference. I could also set MPRF to 8 using Profile Inspector. The input lag in that case was beyond huge. About 1/3 of a second of input lag when moving the mouse.

I set MRPF to 8 and max csgo video settings and I got 99 gpu usage and 40~99(most around 75) usage of cpu,but I could not feel the '300ms'.I can't understand why MRPF effect on lag (vram full?) with vsync off,could you please explain it?

[RealNC]

I set MRPF to 8 and max csgo video settings and I got 99 gpu usage and 40~99(most around 75) usage of cpu,but I could not feel the '300ms'.I can't understand why MRPF effect on lag (vram full?) with vsync off,could you please explain it?

https://forums.guru3d.com/threads/anti- ... st-5714400