TCampbell
Been spending a lot of time on here!
- Joined
- Mar 31, 2012
- Messages
- 3,614
- Reaction score
- 1,556
- Location
- Dearborn, MI
- Can others edit my Photos
- Photos OK to edit
You're going to need precisely calibrated scientific testing equipment to notice the difference in ISO performance between a 7D, 70D, 60D, or even a T2i/T3i/T4i/T5i. It's not that there isn't a difference... it's just that it's so subtle that you're not going to casually notice the difference. Effectively they will all "seem" to be about the same.
However... the ISO performance of a 5D II, 6D, 5D III, or 1D-X will be extremely obvious.
Part of the reason for this is because is because when it comes down to it... it's the physical size of an individual "photo site" on the sensor that makes the biggest difference and large full-format sensors have physically larger photo-sites. That makes them more efficient collectors of photons.
This is also one of the motivations behind the idea of "binning". For example, 2x2 "binning" means you use a 2x2 matrix (four photo-sites in all) and treat them as if they were just one. Noise elimination is a Poisson distribution/regression. The ability to knock down noise is based on the square root of the number of samples you take. The square root of 4 is 2. Which means if you take 4 samples, you can do twice as good at knocking back the noise.
However... the ISO performance of a 5D II, 6D, 5D III, or 1D-X will be extremely obvious.
Part of the reason for this is because is because when it comes down to it... it's the physical size of an individual "photo site" on the sensor that makes the biggest difference and large full-format sensors have physically larger photo-sites. That makes them more efficient collectors of photons.
This is also one of the motivations behind the idea of "binning". For example, 2x2 "binning" means you use a 2x2 matrix (four photo-sites in all) and treat them as if they were just one. Noise elimination is a Poisson distribution/regression. The ability to knock down noise is based on the square root of the number of samples you take. The square root of 4 is 2. Which means if you take 4 samples, you can do twice as good at knocking back the noise.