Reply To: A really great read…

[cameraclicker]

Here is a link to a rather bizarre video by Tony Northrup who has done a lot of videos and seems to be offering training:  https://www.youtube.com/watch?v=DtDotqLx6nA

At about 8:14 he is going on about signal to noise ratio and seems to be saying a full frame sensor gathers more light than a micro four thirds sensor; which by itself is true, since the lens’ image circle has to be bigger to cover the bigger sensor, and the full frame sensor is a lot bigger since the diagonal is twice the size, the total light collected is much greater.  The thing is, it’s not how much light the whole sensor gathers, it’s the amount of light a single photosite collects that determines the signal to noise ratio for that photosite, and for a scene reflecting x lumens toward the lens, the same lens, same focal length, same aperture, etc., the amount of light received at some photosite near the middle of the sensor will be the same as for the photosite in the corresponding position on the other sensor.  If, both photosites are the same size and the microprisms on top of the photosites are gathering light from the same sized area and delivering it with equal efficiency.  Photosite sizes vary considerably within a given sensor size, be it full frame, APS-C, micro four thirds, or whatever.  There is only so much space on the chip and the more photosites crammed on, the smaller they have to be.  Improvements in the composition of components and in the software have delivered noise reduction even within a sensor size, so his argument is weird on several levels.

At 17:00 he is comparing to images at about the same size of displayed head, but one is full frame and one is micro four thirds, so he must have moved or changed focal lengths, or cropped a lot!  So, more nonsense.

Around 20:30, he gets into focal lengths and he is trying to apply crop factor to change focal length.  But a 35 mm prime lens is always a 35 mm lens, it doesn’t grow just because you put a small sensor behind it.   He continues on to provide “mathematical proof”.  Very weird!