@wolf480pl I was just reading about Lorentz contraction as a solution to the Michelson–Morley null result, and I'm back to not understanding it.

Here's my logic, lets imagine the arms are 10 units long and the half-silvered mirror is at x,y of 0,0.

At rest: Light on the X axis bounces from 0,0 to 10,0, to 0,0, doing a total of 20 units of motion. Light on the Y axis bounces from 0,0 to 0,10 to 0,0, also doing 20 units of motion.

At speed (X travel of 2 units per light-bounce-time), light on the X axis bounces from 0,0 to 11,0 to 2,0, doing a total of 20 units of motion. Light on the Y axis bounces from 0,0 to 1,10 to 2,0 doing 20.1 units (if I'm not mistaken, the hypotenuse is 10.05 so the sum is 20.1).

You had me convinced earlier that because the X beam starts at 0,0 and ends at 2,0, it has to have taken extra time to move those 2 units, but considering the mirrors instantaneously moving from points 0->1->2, the numbers seem to indicate that the opposite is true.

But now the problem is that there's no amount of X contraction that can fix this, when the X travel is 20 and the Y travel is 20.1.

If light travels at 5 units per second normally, then at rest it takes 2 seconds out and 2 seconds back which is 4 seconds.

If the speed is 1 unit per second then light will only do 4 units per second on the outbound but does 6 units per second on the return. 10/4+10/6 is 4.16.

Okay I see the problem, it takes more than half of the time for the light to get to the mirror, so it's not 0,0 to 11,0 to 2,0, it's more like 0,0 to 11.3,0 to 2,0 (rough estimate).

@cjd I was about to point out it's not 11.0

do I need to read the rest of your post or is it all clear now?

I got it, thanks. This experiment is quite subtle and in fact McCulloch misrepresented it in his 2014 book - not sure what to make of that yet...

@cjd
> quite subtle

idk, seems like a typical mistake one makes at a high school physics contest :P