>>13820482Basically(my memory is terrible so bear with me if I get something wrong):
Get a beam splitter that lets 50% of photons go through and 50% be reflected upwards, like in my pic.
The white "sheets" are just normal mirrors to direct the photons to the 2nd beam splitter.
Setup detectors at the ends of the 2nd splitter, A and B.
If there's literally NOTHING between those paths(ignore the bomb for now), the detector at B will never, ever detect a photon there because of destructive interference coming from the bottom path. All photons go to A(by the way, that same thing happens even if you shoot just a single photon, because it interferes with itself, something that was known since the double slit experiment).
Okay great. Now the fun starts.
Get a "bomb" that has a photon detector and place it on the bottom path.
If the detector is broken(bomb is a dud), then nothing changes. Photon is always only detected on A, as the bomb never interacts with the Photon.
Fine for now.
NOW, if the bomb is NOT a dud, there are a few possibilities on what it could happen:
1. The photon takes the bottom path, gets detected by the bomb, and the bomb "explodes". Photon is not detected anywhere(that happens 50% of the time with a live bomb).
2. Photon takes the top path, reaches the 2nd splitter and goes straight through. Photon is detected at A.
3. Photon takes the top path, reaches the 2nd splitter. And gets reflected top. Photon is detected at B.
3 only ever happens if there's a "live" bomb that did not explode. Again, the photon NEVER went through the bottom path, and yet it knows there's a detector there. Interference with itself never happens.
That blew(pun not intended) my mind at the time. That's why stuff like parallel universes make "some" sense. In the normal experiment with no bombs, it's like the photon from our universe was interfered by itself from the photon from another universe that went the opposite path. With the bomb that never happened because it exploded
