Try the experiment a few times. It only works if you happen to look right after it changed to the number you see. Eg it changes to :57, 0.1s later you start loo

Try the experiment a few times. It only works if you happen to look right after it changed to the number you see. Eg it changes to :57, 0.1s later you start looking at it, 0.9s later it changes to :58, 1s later it changes to :59. But to you the delay between :57 and :58 will seem noticeably longer than :58 and :59. If you happen to look at :57 when it's already been on :57 for, say, 0.5s, then it'll change to :58 0.5s later. It might seem like 0.7s to you but you wouldn't have any way to know it was supposed to be 0.5s.