Page 1 of 1

Question about capping frame

PostPosted: 07 May 2019, 01:51
by del1ks
Hey,

So for me it's clear when using g-sync why you should cap your frame lower to your refresh rate. I do understand doing so with RivaTuner could add a little of input lag and it's better to use the frame cap in the game if there is one.

My question is, if for example I'm using g-sync with my 240Hz monitor and I use RivaTuner to cap at 237, if some games runs between 100 and 160 fps for example, so never reach really the 237 limit, the cap is just useless but doesn't bother or is it somehow still adding a little bit of input lag ?

Thanks !

Re: Question about capping frame

PostPosted: 07 May 2019, 03:38
by RealNC
del1ks wrote:if some games runs between 100 and 160 fps for example, so never reach really the 237 limit, the cap is just useless but doesn't bother or is it somehow still adding a little bit of input lag ?

It doesn't add input lag to begin with. In many games, it lowers input lag. So when the cap is not reached, in some games you will get higher input lag.

An in-game limiter can lower input lag more than RTSS can.

Re: Question about capping frame

PostPosted: 07 May 2019, 10:38
by Chief Blur Buster
del1ks wrote:So for me it's clear when using g-sync why you should cap your frame lower to your refresh rate. I do understand doing so with RivaTuner could add a little of input lag and it's better to use the frame cap in the game if there is one.

Correct. The rule of thumb is capping with the game is always lower lag than capping with an external framerate capper.

del1ks wrote:My question is, if for example I'm using g-sync with my 240Hz monitor and I use RivaTuner to cap at 237, if some games runs between 100 and 160 fps for example, so never reach really the 237 limit, the cap is just useless but doesn't bother or is it somehow still adding a little bit of input lag ?

Software implementation dependant.

In a well-designed frame rate capping utility, nothing happens if the framerate is always lower than the cap.

That said, once the cap is hit, there can be added lag, but you're avoiding a worse lag.

The thing is that when a variable refresh rate monitor in its default mode (VSYNC ON), automatically switches to VSYNC ON when framerates max out at refresh rate. And, we know VSYNC ON adds input lag. The capping lag is typically less than the VSYNC ON lag. So, capping 237fps on a 240Hz GSYNC monitor -- in order to avoid getting the dreaded VSYNC ON lag.

Blur Busters was the world's first website to discover the lag-increase effect of framerates reaching refreshrate on a GSYNC monitor, in past GSYNC articles. So we pioneered the advice specific to GSYNC/FreeSync to use a frame rate capping utility. That way, your input lag is much more consistent without the sudden lag increase/decrease (inconsistency) as your framerate fluctuates up and down, hitting max Hz repeatedly and falling below max Hz repeatedly, with all the repeated continual switching between GSYNC and VSYNC ON.

Thus, the lagfeel of GSYNC/FreeSync is much better with the "cap a few frames per second below" especially with an in-game framerate cap. Your aiming feels better.

The capping lag is capping-dependant, some well-designed games's built-in framerate caps have no noticeable lag, while external cappers (e.g. RTSS) may have slight amount of lag (but less than simply letting VSYNC ON happen everytime framerates max-out on a GSYNC monitor).

Another workaround to avoid this, is the GSYNC+VSYNC OFF mode. Then you don't need a cap. However, the feel of GSYNC is different from the feel of VSYNC OFF, and not all gamers like the transitions between those modes whenever framerate go above/below the refresh rate.

Jorim's GSYNC 101 here on Blur Busters is the best detailed info about all this on the Internet.