3 ms·

Human perception does not have a sampling rate. This is why our eyes don't produce weird shutter speed effects like the one shown in the video. There is no shut

by fsh 7y ago

Human perception does not have a sampling rate. This is why our eyes don't produce weird shutter speed effects like the one shown in the video. There is no shutter and there are no frames. The same is true for measurements in physics. Modern digital lab equipment may use sampling, but this has nothing to do with the underlying physics. Putting a photographic film behind a double slit "collapses" the wavefunction of the photons going through without any sampling.

ta1234567890 7y ago
You can call "sampling" "interaction" instead, if it's clearer. It still happens. The "shutter speed effect" that your eyes produce are effectively what you perceive as your reality. If you could perceive as fast as a hummingbird or a bee, your reality would look a lot different. Making a photographic film interact with photons is effectively sampling. The film is sampling the photons in the sense that whatever vibrations/waves the film is made of is interacting in a certain way with the vibration/wave of the photon. This process is going on continuously, non-stop every single moment, for everything.
ganzuul 7y ago
That carries the analogy a bit too far. While 'sampling rate' commonly implies a discrete time interval it can also mean an average. Wave function collapse isn't a physical process but instead a commitment to not try to explain what's behind the equations. (And IMO consequently not science. We don't just give up because it's hard.) QFT meanwhile seems to say that samples is what you get when two fields intersect. However, Bell inequality and decoherence further implies that there is no such thing as an abstract 'sample'. The spread of quantum information from an interaction is a subtle physical process.
trhway 7y ago
>Human perception does not have a sampling rate. This is why our eyes don't produce weird shutter speed effects like the one shown in the video. There is no shutter and there are no frames. our visual cortex is sampled at about 200Hz. The pipeline is about 6 stages (similar to the layers in CNN deep net) each taking 5ms, so about 30ms from photons hitting retina to the recognition. To play with eye shutter speed effect - look at a wheel of the car driving along in a near by lane, and now quickly close and open eyelids, may be couple times in a row, and you will catch a glimpse of the non-spinning wheel still-photo like as in those movies.

by fsh 7y ago

ta1234567890 7y ago

You can call "sampling" "interaction" instead, if it's clearer. It still happens. The "shutter speed effect" that your eyes produce are effectively what you perceive as your reality. If you could perceive as fast as a hummingbird or a bee, your reality would look a lot different. Making a photographic film interact with photons is effectively sampling. The film is sampling the photons in the sense that whatever vibrations/waves the film is made of is interacting in a certain way with the vibration/wave of the photon. This process is going on continuously, non-stop every single moment, for everything.

ganzuul 7y ago

That carries the analogy a bit too far. While 'sampling rate' commonly implies a discrete time interval it can also mean an average. Wave function collapse isn't a physical process but instead a commitment to not try to explain what's behind the equations. (And IMO consequently not science. We don't just give up because it's hard.) QFT meanwhile seems to say that samples is what you get when two fields intersect. However, Bell inequality and decoherence further implies that there is no such thing as an abstract 'sample'. The spread of quantum information from an interaction is a subtle physical process.

trhway 7y ago

>Human perception does not have a sampling rate. This is why our eyes don't produce weird shutter speed effects like the one shown in the video. There is no shutter and there are no frames. our visual cortex is sampled at about 200Hz. The pipeline is about 6 stages (similar to the layers in CNN deep net) each taking 5ms, so about 30ms from photons hitting retina to the recognition. To play with eye shutter speed effect - look at a wheel of the car driving along in a near by lane, and now quickly close and open eyelids, may be couple times in a row, and you will catch a glimpse of the non-spinning wheel still-photo like as in those movies.