This, is a really important point. LCDs are NOT CRTs. So it isn't that the top left pixel of the screen is getting dimmer and dimmer while the CRT is painting lower pixels, on an LCD it is simply not changing.
What "30Hz" means in LCD terms is that if you go to change a pixel, the fastest you can do that is once every 33 mS but during the whole time it isn't changed, it stays the same intensity.
Basic desktop usage is unbearable at 30fps. Just dragging windows around you have to insert a deliberate pause to make sure the pointer is really where you want it to be because the screen is updating so slowly. Flicker is not the problem, slow screen updates are.
That is completely different from my experience; I think you must have another problem besides the 30 fps frame rate. Perhaps your machine is just having trouble pushing that many pixels? On my machine, animations are just slightly, subtly choppy -- like Android used to be before all the "butter." But there's still very much the sense of instant responsiveness.
I'm pretty sure I mouse as quickly as anybody. It's just that the difference between 17 ms per frame and 33 ms per frame on my system only affects the smoothness of animations. There's no perceptible lag no matter how quickly I move things around.
To add one more data point, I usually set my mouse sensitivity to maximum (people usually complain about how sensitive my mouse is when they borrow my computer), and while 30fps is fine for everything else, it's unbearable to me when mousing.
I just did a test: Over 5 seconds I moved my mouse to the top of my monitor and back down 15 times, which works out to 12000 pixels per second, or 396 pixels per frame - which seems about right for how far the mouse seems to jump per frame. For comparison, Chrome's Close Tab button is 15x15 pixels, and HN's upvote button is 8x7 pixels. Framerate really is the limiting factor for how long it takes me to move my mouse onto the kinds of click targets common in computers.
Another test: it takes around 6 seconds to close 7 tabs with a mouse in Chrome in 30fps, and 4 seconds to close the tabs in Chrome in 60fps.
So I think I've pinpointed the exact problem. There are two parts to moving a cursor to a click target: quickly moving the cursor in the target's general direction, and then slowly moving to its exact position. Both are much more unpleasant in 30fps.
Quickly moving in the target's general direction: it's a lot easier to track a 10x20 pixel cursor when it's jumping 150 pixels per frame than if it's jumping 300 pixels per frame. (These are my own numbers, judging by people's reactions to my mouse sensitivity settings, it's likely yours would be lower.)
Slowly moving to the target's exact position: if the click target is 8x7 pixels like HN's upvote button, I can't move any faster than 2-3 pixels per frame, which is obviously half as fast in 30fps as in 60fps.
Well, if you mouse that fast then your use-case is similar to the aforementioned twitch-based FPS games.
> There are two parts to moving a cursor to a click target: quickly moving the cursor in the target's general direction, and then slowly moving to its exact position
This is a pretty basic usability issue and one of the reasons we put clickable things on the edge of the screen if we can.
Monitors with 40ms of input lag used to be commonplace and regarded as completely usable for non-gaming purposes. The extra input lag on a modern monitor that comes from running at 30Hz instead of 60Hz shouldn't be a huge problem, though depending on how much input lag there is due to other sources it might be at least noticeable.
Yes there is, you just pointed it out. There 16-17 ms of extra lag compared to a typical monitor. When I stop moving my mouse, I have to wait up to 33 before I can see where it actually is. That is easily noticeable. Stick a bunch of people with 30Hz monitors and watch their misclick rate jump.
Can confirm, I am running a UP2414Q @ 30 Hz because my 2013 13" MBPr does not support 4k at 60 Hz and it is pretty bad. I am planning on replacing the laptop soon. Otherwise the screen is great and non-4k 24" screens look terrible to my eyes now.
Dragging windows does not constitute my basic desktop usage. I probably drag or resize a window at most once per day, and I'm not even using a tiling or significantly keyboard-driven window manager, it's just stock OS X.
>Dragging windows does not constitute my basic desktop usage
Fair enough, but it does for most people. The claim I was disagreeing with is that >30fps only matters for action games. It is noticeable for most tasks that use a mouse.
(This is assuming the task at hand doesn't require a high refresh rate of course - a reflex based game like an FPS would be a no-go)