[CapacitorSet]

Silly question: isn't this phenomenon bound by the Nyquist limit, which means it would take at the very least 4k fps to reconstruct music (considering that a soprano can get no higher than 2 kHz)? That sounds like an impossibly high framerate.

[coldtea]

4K fps? Impossibly high? Not at all. There are consumer compact cameras and even smartphone sensors that can hit 1K fps [1] -- and those are not even specialist high speed cameras.

Specialist cameras hit 4k fps at second gear, and can go as high as million fps (and even trillion fps [2])!

Besides, this is about reproducing music frequencies, whereas the most probable target for this technology is eavesdropping, not as a replacement for Neumann U87 -- and for that you need way less range.

[1] https://www.theverge.com/circuitbreaker/2017/2/7/14532610/so...

[2] https://newatlas.com/fastest-camera-44-trillion-frames-per-s...

[dorgo]

wait, 4.4 trillion fps with a resolution of 450 x 450 ? Isn't it petabytes of data per second?

[leoedin]

They address that in the paper. They make use of the rolling shutter common in CMOS imaging chips to effectively increase the sampling rate to well above the frame rate.

[Monory]

They show in the last demo usage of this technique with rolling shutter. Considering image height for most cameras is now in thousands of pixels, this somewhat overcomes Nyquist limit problem