I disagree. There’s absolutely diminishing returns above 120 or so, and at 120hz you’d do as well to remove input latency rather than output latency. On a non full screen app you’re pretty much guaranteed to have a full frame from the OS/compositor, and more if your application has an event loop that is in any way abstracted from the OS.
Any meaningful amount of work is either going to be done “immediately” (see pretty much every spreadsheet in existence except for that one on the network drive) or it will take a perceptible amount of time. Giving progress in 4ms increments instead of 8ms isn’t really going to affect the perception of the task.
There are definitely real world tools that take time - pumping petrol/gas isn’t an instant action, neither is cooking.
Also - the device that I would consider quietist in my life has a sub 1hz refresh rate, and an 800MHz cpu with a 3 week battery life.
I agree that input latency matters a LOT but disagree about 120hz being some sort of sweet spot.
On the input side, yeah even a few 10’s of ms is very noticeable when typing or doing something quick.
But also I use 240hz monitors for everything, and when I had a config bug on my Mac that degraded my connection to 180hz, my computer felt slow and laggy, and I only realized when it drove me crazy enough to check the settings again. I’m sure the same would happen if I was on a 480hz monitor and it got degraded to the 240hz that I’m used to now.
And I agree there are diminishing returns, but there are still very perceptible returns.