Levina de Ruijter wrote in post #18416242
My default ISO setting for birds is 800. It's only in Summertime that I can go lower to ISO 400. I also shoot a lot at ISO 1600. Depends on the quality of light and the location on how much noise that will give me but in fairly good conditions ISO 1600 will be very clean. I don't shoot at ISO 3200 unless it's an emergency.
I did notice how the 6DII files are pretty similar to those of the 5D4 at higher ISOs. What concerns me is the wiggle room. In winter I mainly walk the park and it gets dark there and I may have no choice but to underexpose. What I was wondering is this. The 6DII scores for DR are bad at base ISO but are actually better than quite a number of cameras at higher ISO speeds. Now I have only seen people pulling shadows and reducing highlights on files at base ISO and the 6DII does not do so well there, to say the least. However, as the 6DII is relatively better at higher ISO speeds in terms of DR, I was wondering how pulling shadows on files of ISO 800 or ISO 1600 would do. Or does the bad DR at base ISO also influence wiggle room in files with higher ISO speeds?
And thanks, Cary, for responding as I really do need help here.

None of the cameras that you are mentioning here have banding noise of any significance, so the amount of noise as measured by Photons2Photos and DxO are fairly representative of what you can expect. Most of the biggest differences between under-exposing vs using higher ISOs for the same sensor exposure occur at the bottom of the ISO range. The difference between 100 and 200 is the greatest, 200 and 400 a little smaller, 400 and 800 smaller yet, and above that, even smaller and smaller, until you get no noise improvement at all from mathematically pushed ISO settings. Both types of ADC have this pattern, but for on-sensor ADCs, the effect is more diluted.
The absolute read noise of the 6D2 is about 8x as high at base ISO as at the highest ISOs. The absolute read noise of the 5D4 is about 2.67x as high at base ISO, and the absolute read noise of the 6D3 is about 3x that of the 5D4 at base ISO, and they are about the same at the highest ISOs. The higher the ISO, the less of the difference between them. The 6D2 keeps improving in absolute read noise up the ISO scale to 800, slows down a bit for 1600, and is fairly level by 3200. The 5D4, however, is almost level already at 400, and only improves at a very slow rate above that, because it had less room to improve to begin with, with less noise at base ISO.