The second test is slightly luck based. 1) Sometimes the sheep that starts running is offscreen. Time starts running when the sheep appears in the screen, but the running sound starts as soon as the sheep starts running (so you have an advantage if you react by sound instead of images). 2) Sheep that appear in the upper part are much easier to catch than the ones that appear in the bottom.
Also, it seems if you manage to catch a sheep before it is onscreen you get a 0.0 sec reaction time. (I averaged 200.2 in the first test and around 240 in the second one)
I get very slow times on the second one, because there is constant distraction from the sheeps moving and making sound. I can't even get a single sheep under 0.200 seconds.
I guess that's the difference between playing with or without storyboard/background?
I'm guessing there's some kind of "object recognition" thing at the first three sheeps which then results in faster response time at the last two sheeps.
I doubt anyone is ever going to flawlessly be able to read AR10 + DT without the use of follow points. It'll probably need some memory work but not nearly as much as flashlight.
Also, this thing, it measures how fast you click given a visual cue. In OSU, you need different types of reaction to different types of visual cues, so...
if I let my cursor just idle and hit on the green as soon as possible I get 250ish consistently if I constantly move my cursor in circles on it and then click on green I get 215ish consistently
better reaction time when more active
...and still worse than almost everybody in this thread.
Something wrong with the sheep test though, the results are a bit skewed to lower values, and i'm always seeing 101, 167, 217... never other numbers. On the human benchmark one though, I average about 190 after 10 tries.