Taxboy wrote in post #19403673
I'm still deliberating between the R6 and R7. Based on my reading would appreciate if people could confirm or correct my thinking please
The R6 deals with higher ISO noise better than the R7 producing cleaner RAW files.
That's not how I would put things. The R6 has no real noise advantage other than the brute force of a larger sensor area, which makes it capable of less noise at the same exposure and ISO, but that does not automatically become possible, depending on shooting conditions and glass. There is a false belief prevalent out there that total final image light comes from your sensor size. It does not; total light that you keep in your composition if cropping comes from your lens' pupil and its distance from the subject, for any subject size in any given ambient light. The bottom line is that you must have shallower DOF to get a noise advantage from FF, above base ISO. Only at base ISO, when you have so much light that you can use an unnecessary shutter speed, does FF get more light without shallower DOF. If you want and can get shallower DOF with FF, like 85/1.2 vs 50/1.2 on APS-C, it is win-win, but in many situations in which people shoot small/distant subjects and are always focal-length-limited, that this shallower DOF never materializes, and FF noise superiority is a miscalculation of scale.
However it has a full frame sensor compared to the R7's APS C. If you add a 1.4x TC to the R6 this will give a broadly similar field of view to the R7; The TC will require an additional stop of exposure to be applied but the R6's cleaner files permit the use of a higher ISO with minimal impact.
The FOV gets closer in size, but the R7 is still giving 2x as many pixels-on-subject. If you put a 2x on the R6 for the same pixels-on-subject, then the R6 is at 4x the ISO, where 100% pixel views, which are now safe for comparing the sensors doing the same pixels-on-subject "job", should show slightly less noise with the R7. So, capturing the subject at the same number of pixels with the same main glass gives zero noise benefit to the R6.
The R6 produces a smaller file size so does not allow a great deal of cropping in post if required to maintain a reasonable size file. The R7 produces a bigger file size allowing cropping but noise can be cleaned up in post up to about 6400 ISO before some impact is seen on image quality
All input appreciated
I think it is pretty clear that there is an empirical fact that R7 captures are never really noisier than R6 captures DOING THE SAME THING with the same amount of light, but many people will interpret the higher resolution of the R7 as a "flaw" because they can see individual pixel noise impulses at 2x the resolution that you can see anything at all coming from the R6 pixels, with the same ISO and exposure. You have a choice to use those finer details at magnification, in which case you need to see noise or use a lot of reduction, or you can just not sharpen them and use them for better sampling quality.
Not everyone is looking at 200% crops from the R6 and comparing them to not-heavily-sharpened 100% crops from the R7, which is what it would take to even begin to be equitable for focal-length-limited photography, so many people are seeing "more noise" where it does not actually exist. And no matter how many times I or anyone else explains this to you, you too may feel that the R7 is noisier with the same light, even though it isn't. Take someone with smooth skin that makes other people jealous, and put it under a microscope, and it will show all kinds of "imperfections", but you don't have to use 400x power if 287x will do what you need.