Within games, the data(not screen)refresh rate has two predominant things that I can think of. Data Updates, and Cohesive Immersion.
With Data Updates, the only thing that is relevant to your mind, is what has changed in a scène and what has stayed the same. This is context sensitive.
Example 1- If you are in a flight simulation, and you are on the runway just looking at a side display map- it doesn’t matter to you if the screen is showing this static information at 30 or 60fps.
Example 2- In the same flight sim if you are on an bombing run in the sky, you would want to watch the cross hairs to drop the bomb effectively, but still be aware of your surrounding in case of enemy aircraft. You want the information updates to match the display device. If the scene stays relatively static you devote more attention to the cross hairs, but the moment an enemy appears in the distance you want that to be relayed on screen, so that you can divert your attention to the new threat if necessary.
Data updates from the machine to the screen and then to you eye can never exceed the refresh of the display device. Internally the machine may be changing system states at far greater than what your display can handle- Sucks for you. But unless it is output on screen or through other less relevant to this discussion, audio or rumble cues, you can’t react directly to it. We can only process the information that we are given and use indirect reactions –interpolation- to try to compensate for perceived holes in that information.
One key piece that I think gets overlooked is that the game world e.g. game state changes are occurring at incredible speeds. And because of this we can falsely assume that we play first person shooters, better at 120fps on a display that refreshes at 60hz. When in reality, it’s that we are playing better because the frames have more applicable frame variation, in relation to the game state.
This is long so I will continue with Cohesive Immersion in my next post.