I’ve been watching a lot of slow mo guys recently and love their videos but I’ve been really interested in the technology of how high speed cameras work. How do they capture such incredible detail at such a high frame rate? Also, why does resolution and exposure matter for the frame rate?
In: Technology
The biggest hurdle is getting enough light. If your shutter is open for half as much time, you need to bring in twice as much light in order to see the same brightness. These cameras going at 10,000 fps need to bring in 160 times as much light as a regular camera, which means you’re going to need a large high quality lens, and very good lighting conditions.
So there’s the basics, which you know. They are high speed cameras that capture footage at frame rates WAY higher than normal. There is the light/exposure part, which is explained in another comment, and lastly we have resolution.
Sooo, it’s the same deal as the first. High speed cameras capture many more frames per second than a regular camera. That means you also have WAY more data being stored per second of footage. There’s an episode where Gavin talks about this, and what it boils down to, is that when you film at ludicrous frame rates like 250k FPS, that means you have over eight thousand times more footage per second. That takes up a ton of space, so to compensate, they lower recording resolution.
You’ll also notice at times the playback they provide is smaller in relation to your screen in accordance with the fact that it’s filmed at a lower resolution.
ALSO, that is why they don’t upload videos at 60FPS. It basically forces them to use twice as much slow footage frames for no reason, to properly pad for the presentation time they want, which results in one of two things
Either they need to film at faster speeds to begin with, which compounds the problems above,
Or they need to lessen the amount of slow footage time they actually have available. It’s stuff they mention on their second channel, and I may be remembering some things incorrectly.
High speed cameras, like the popular Phantom cameras, use digital sensors.
These sensors have a maximum amount of data they can pass to the storage system. It’s typically measured in giga pixels per second.
As the frame rate is increased, this becomes a bottle neck and the choice is to decrease the resolution so there are fewer pixels per frame. Then the frame rate can be increased again.
The storage systems are typically built around buffering the data to Ram and then saving the captured footage to disk.
[Here’s a video of a phantom camera teardown](https://youtu.be/Q8RRRfW-APQ)
Latest Answers