Thompsonrh wrote:I'm not a video expert but am learning a lot about the most recent technologies and how as a professional photographer you have to think differently. The importance of understanding the subject scene, in terms of the amount of shadow and highlights in the scene and setting ISO accordingly (somewhat the same as Ansel Adams Zone system). See Page 50 in the manual. If you landed on ISO 640 as part of the best setup for the scene then the question to ask, is this a mainly dark scene or mainly bright scene? If it is mainly dark, then a better setting might be ISO 1250 because your dynamic range in the shadows goes from 7 stops (ISO 640) to 8.5 stops (ISO 1250) on the 6K model. If the scene is mainly bright, then ISO 640 might be better although ISO 1000 would give more dynamic range in the bright areas than 640, but ISO 1250 would be a bad choice in this case of more bright tones. This is all because of the dual ISO capability. Seems to me its like another dimension in the setup process. I appreciate that dynamic range and digital noise are not the same thing but with the better dynamic range, there's less curve adjustments needed in the deep shadows. Having said all this, I do agree with the original Post, I am surprised at the amount of noise in my 4K at native ISO 3200. Although exposed well and adjustments made for input and output color space, I definitely need to apply NR.
Applying a film analogy to ISO as its implemented on BMD cameras, as you're doing above, is only going to get you into trouble. It doesn't help that most of the posts above don't distinguish between braw and Prores, or offer a word about how the shots were actually exposed.
And the infamous Facebook group hysteria, promoted by John Hess and others, that shooting at ISO 1000 magically offers more "highlight headroom" simply ignores the fact that you can only get to 1000 by upping the digital gain (noise!), and stopping down the lens by the same amount you're boosting the signal. Big surprise! You can just easily shoot the scene at a lower ISO, meter for iso 1000, and raise levels in post, and get the same or better result. If you ask me, that blasted chart is the source of endless and pointless disputation over ISO.