Brad Hurley wrote:I'd dispense with the CSTs and use Resolve Color Management, which is designed to help you match footage from different cameras.
To be strictly clear, it isn't designed to match cameras. Rather to generalize display rendering and manage color in a slightly more automated fashion. You can achieve the exact same results with using CSTs in a manual setup.
@Roberto, The main aspect to matching cameras is to use the same display rendering, which you automatically do if you use RCM, ACES, or manual setups. The basic principle of 'modern' managed grading setups is shoot in log or raw -> convert to a log grading space of choice -> render to display with a single transform. From what you've explained you're using Arri's 'classic' rendering on the OG BMPCC and DaVinci on the iPhone.
For both camera's you'd want for example to convert to DaVinci Wide Gamut / DaVinci Intermediate, then grade, then a CST from DWG/I to Rec.709. Or convert both to AWG3/LogC3 and then use ARRI's rendering instead. (You could even grade in DWG/I and convert to AWG3/LogC3 at the end to then apply the ARRI LUT but I'd focus on simplicity first.)
When using manual management the most efficient way is to use groups where the pre-clip group node tree would be just the CST to convert to the grading space. The group post-clip node tree where you convert to Rec.709. Doing it on the timeline is not practical unless grading your footage is the only thing you do. Otherwise you apply it to graphics/text and other elements that shouldn't be part of it.
Color Managed does the same tranformations but 'hidden' with input conversion before the entire node tree and out all the way after timeline node tree. So it frees up your group node trees for other purposes which may be preferential. But you won't be able to do any operations before input conversion or after display conversion which in some cases you may actually want.
Try both and see which you prefer. When using manual management make sure you still set your timeline in the project settings to DWG/I if that's the grading space you chose. This will make sure that conversions are done as intended and HDR wheels work as they should without manually setting them to the correct space.
Once everything is solidly set up managment wise, all that's left to deal with is plain sensor, lens, encoding quality differences that you have to deal with. That only comes with practice and doing it a lot. And some things will never match fully ofcourse. The main goal should be to make them feel like they belong in the same world, not match 1:1.