Yeah, when I saw that, I was like "WAITAMINNIT!!!..."
- How did they get the shutter to open and close fast enough?? You would have to do it faster than the speed of the photon packet, which is physically impossible. WTH??
- How did they get a picture of a photon packet traveling through a transparent object when you need those
very same photons striking the image sensor to take the picture?? By the time any photons scattered off the semi-transparent plastic and particulate in the water, the packet would be already gone.
Then I heard the guy talk in the TED video about how it's actually a composite of a bunch of timed shots put together sequentially, and I understood how they did the actual video. So, it's not "real time"
per se.
The answer to the second question is actually twofold and partially self-answering, and how I understand it is, (1) the 'packet' you see is in reality further along than where you see it at that particular frame because it still takes light more time to reach the image sensor than to travel down the length of the bottle, and (2) it's not an image of the
actual packet, but the light from the packet that escaped as it went, refracting off the particulate in the water.
This, BTW is how we are able to see anything at all; we don't
see the thing itself, just the light reflecting off of it.
Without the particulate and the semi-transparency of the bottle itself, you wouldn't see anything. Notice how at the beginning of the video, there is no light beam coming in before it strikes the back end of the bottle, where photons start scattering.
Cool stuff, anyhow
