[maximumPC] richard huddy interview

[antihelten]

It doesnt make sense to compare G-Sync to V-Sync off. V-Sync is the fastest method to send frames to the monitor. But the drawbacks are tearing and stuttering.

With Adaptive-Sync there will be always "lag" over V-Sync off because the display needs to be triggered by the GPU.

The reason why I'm comparing to vsync off, is exactly because it's the fastest method, it's the baseline so to speak.

Also the latency I'm talking about is in addition to the latency caused by the G-SYNC module pooling the monitor. But again all of it added together should still be fairly insignificant.

With that being said I still don't have any idea what's going on with the results blur buster got from CS:GO, as the magnitude of the latency they saw there really doesn't match up with either double or triple buffering, so there must be some other reason for it.

 

[Flapdrol1337]

The higher latency of the game without fps limiter is probably because of the default (3) pre rendered frames setting. With the fps limiter the output is faster than the frame creation and the buffer is not allowed to build up.

In the nvidia control panel I use "display - no scaling", and 1 pre rendered frame in the global profile. Only problem I've had with it so far was 3dmark 11 refusing to start.

Of course this is only important if you're gpu limited, if you're not (like with an fps limiter) the gpu is waiting for the cpu and the buffer doesn't build up. Waiting for vsync forces a gpu limit, on 60 hz it's very much unplayable with 3 pre rendered frames, it's also why some people use an 58 fps limiter with 60hz vsync.

As far as I know amd also uses 3 as default, but you can change it with radeonpro (or in games cfg files). Since tech review sites don't do latency testing, but only fps and smoothness there's no real incentive for amd and nvidia to work on latency unfortunately.

 

[antihelten]

The higher latency of the game without fps limiter is probably because of the default (3) pre rendered frames setting. With the fps limiter the output is faster than the frame creation and the buffer is not allowed to build up.

In the nvidia control panel I use "display - no scaling", and 1 pre rendered frame in the global profile. Only problem I've had with it so far was 3dmark 11 refusing to start.

Of course this is only important if you're gpu limited, if you're not (like with an fps limiter) the gpu is waiting for the cpu and the buffer doesn't build up. Waiting for vsync forces a gpu limit, on 60 hz it's very much unplayable with 3 pre rendered frames, it's also why some people use an 58 fps limiter with 60hz vsync.

As far as I know amd also uses 3 as default, but you can change it with radeonpro (or in games cfg files). Since tech review sites don't do latency testing, but only fps and smoothness there's no real incentive for amd and nvidia to work on latency unfortunately.

That might very well be the case, although it still seems a bit weird that 143 fps cap and the 120 fps cap aren't closer in latency then, but as blur busters notes, the 143 fps cap might be colliding with the refresh rate somehow, due to the way the G-SYNC module works.