- January 13, 2019 at 7:16 pm
when grading in DaVinci resolve, the values on the scopes are displayed on the scale from 0 – 1023, meaning that colours are previewed as 10bit. Does DaVinci always previews colours this way?
If I for example import a clip that has 12 or 16 bit depth and the values on the scopes are still between 0 and 1023, does it mean that it downgrades footage to 10bit? I find this highly unlikely, so I presume that this 10bit values must only be for preview, while DaVinci treats colours in greater bit depth. Is there a specific value for the bit depth that DaVinci always uses?
Does DaVinci perhaps always use the bit depth of the clip with highest bit depth that is present on the timeline to grade all the clips on the timeline?
- January 13, 2019 at 7:54 pm
Processing is at 32 bit, unless using performance mode, when it is less (I think 16 bit) for playback only. All rendering processing for delivery is 32 bit unless I’m mistaken.
- January 13, 2019 at 10:44 pm
Thank you for your reply! Can anyone confirm for sure?
- February 3, 2019 at 12:32 pm
yes, resolve works with 32bits float internally (as does fcpX by the way) but monitoring (waveforms etc) are limited to 10bit because people usually only need it for broadcast? 12bits is the maximum and used in cinemas only. as already mentioned, that does not mean the information is not there! in avid you are stuck with 8bit monitoring for example. does not mean correct 10bit export is not possible.
- February 4, 2019 at 12:43 am
I thought that 4K required 64 bit processing partly because of the bigger colour spectrum that it can resolve. I think that information would be in the HTML 2.1 specification. Perhaps this is all irrelevant but I know it is discussed.
- February 4, 2019 at 12:49 am
I apologise for the error I should have said HDMI
- February 4, 2019 at 1:23 am
[Terence Christopher] “I thought that 4K required 64 bit processing..”
I think you might be confused by Operating System bit depth, file wrap formats like Quicktime and the actual bit depth process of digital video. Resolve is 32 bit floating point which has a greater dynamic range than 64 bit fixed. It has more than enough for a typical 10 bit 4.2.2 4k digital video signal
- February 4, 2019 at 4:50 am
Michael thanks for that correction. I did not know these issues. I have a lot to learn!!
- February 4, 2019 at 6:38 am
I had a head start with my father who designed microwave communications gear including broadcast TV links, plus working in broadcast telecine 1976, then dabbling in digital audio from 1984, through to digital video when it first appeared. Color space, bit depth and technical specs have been bread & butter for many years and it does take some time to get your head around what it all means.
If it’s any consolation, I still have to bone up on the changing landscape of high dynamic range variances and compatibility in all the different color spaces. Never stop learning
Log in to reply.