The use case:  Mavic 3 Cine, recording in ProRes D-Log, Davinci Resolve Studio 17.4.3.  I had mixed results trying to use LUTs, so I switched to Resolve's color management (greatly enhanced in version 17 and later). I'm currently testing the following workflow: 1) set project color managementy to Davinci YRGB Color managed, and 2) output to Rec. 709.  3) For each clip in the media pool, I right click and set input color space to "DJI D-Gamut/D-Log."   That can also accomplished by applying a Color Space Transform to one of the nodes (leave at least one node before it for adjustments).  I have also had good results with color correcting with the HDR wheels. Much finer control over exposure and saturation throughout the image than using the RGB wheels only.

Here is my question:  Davinci Resolve 17.4 has only one "input color space" option for DJI drones (DJI D-gamut/D-Log); however, several different  DJI drone models are capable of recording in some flavor of D-Log (for example, DLogM for the Mavic 2 series, and D-Log for the Mavic 3s). Does the one DR input color space option behave the same when grading D-Log footage from the various DJI models?  I also noted that the DJI technical paper on D-Log that is posted on the Mavic 3 documents page is the very same document posted back in 2017, which suggests that all DJI drones since 2017 capable of shooting D-Log will respond similarly to the same color grading settings in DR, and presumably other NLEs. Is this true?

I contacted Blackmagic Design on the above D-Log consistency question and they referred me to DJI for advice--so here I am. DR contains numerous input color space options for Canon and Sony cameras, attributed to subtle color space differences across the respective camera models. I took that as a hint that the same may be true for the various DJI drones. I asked BMD if they planned to add input color space options for the respective DJI drone models, but they did not reply.

Bottom line for those pilots who shoot D-Log and edit/color grade with Davinci Resolve: Resolve color management and HDR wheels may lead to a more natural color correction grade across a wide range of footage than other techniques.  LUTS are "destructive"; Davinci color management is not, and it reduces processing time. Likely similar outcomes with other NLEs. I'd love to hear from anyone with experience and/or expertise on this topic.

Hello there NDI-Orlando. Good day and thank you for sharing these information with us. Again, thank you for your support and have a nice day.

Thanks, DJI Stephen.

Can you or a colleague provide any information on the question I posed:  Several different  DJI drone models are capable of recording in some flavor of D-Log (for example, DLogM for the Mavic 2 series, and D-Log for the Mavic 3s). How does "D-Log" differ (if, indeed, it does) between the various DJI drone models?  To what extent do any differences impact grading in various NLEs?  I noted that the DJI technical paper on D-Log that is posted on the Mavic 3 documents page is the very same document posted back in 2017. Because only one version of this D-Log paper exists, it suggests that log video from all DJI drones since 2017 capable of shooting D-Log is fundamentally similar, if not identical, from a color grading perspective. Is this true?

This is a DJI question, not a Davinci Resolve question. To re-state, in what specific ways do the various drones' log video differ from one another, and where might this be documented?

Thanks.

NDI-Orlando

NDI-Orlando Posted at 1-16 15:38
Thanks, DJI Stephen.

Can you or a colleague provide any information on the question I posed:  Several different  DJI drone models are capable of recording in some flavor of D-Log (for example, DLogM for the Mavic 2 series, and D-Log for the Mavic 3s). How does "D-Log" differ (if, indeed, it does) between the various DJI drone models?  To what extent do any differences impact grading in various NLEs?  I noted that the DJI technical paper on D-Log that is posted on the Mavic 3 documents page is the very same document posted back in 2017. Because only one version of this D-Log paper exists, it suggests that log video from all DJI drones since 2017 capable of shooting D-Log is fundamentally similar, if not identical, from a color grading perspective. Is this true?

I don't have an answer to that for you, but I will concur strongly with using Davinci's color management and d-log color space to begin the grade. LUTS always clip. I am never content with what the initial color space setting gives me for any of the many camera color spaces that Resolve gives me, so I always consider it an approximation. I have definitely seen a significant difference in D-log files between the M2S and my M3. But that would make a lot of sense with the different sensors. I would think that DJI applies the same basic D-log compression to all the drones that can use it, but that the different sensors respond differently and result in files that respond differently to the Resolve color space.

My big disappointment with the M3 is that DJI doesn't give us the ability to shoot D-log on the M3 with any of the tracking systems. Given the cost of this drone and the improved quality of the camera, that requirement for getting the full dynamic range out of the camera should be available on those very helpful flight options. I'm initially finding it too complex than it should be matching shots from the non-D-log footage with the D-log material, and the dynamic range is dramatically reduced. I have been flying DJI since the Phantom 2, and I'm fond of the M3 so far, but I certainly don't think it's living up to its price yet.

Thanks for your reply #MParfit.  Following my earlier post, I experimented with replicating the techniques that worked with M3C D-Log video on DLog-M video from my M2P.  The outcome was not good, which reinforces my hunch that D-Log and DLog-M differ in some undefined, but substantial ways. In any event, I've found methods that basically work (Davinci Resolve's DJI D-Gamut/D-Log color space for M3C video and a Leeming LUT for M2P video, both coupled with use of DR's wheels and curves).  It would be great if Blackmagicdesign would eventually include color space profiles for each drone model, as they have for Sony and Canon camera models.

Am having a kind of similar problem
currently working on a mavic 3 D-Log aerial footage and trying to covert it to Rec709

using the official "LUTS" DJI provided  as a correct conversion from D-Log to Rec709 i like its colors but i don't want to use "LUTS" due to it's destructive nature
so i use the Davinci Resolve's  color space transform with DJI D-Gamut/D-Log to rec 709 but unfortunately it doesn't match the official "LUTs" DJI provided not even close
does this because Davinci Resolve's DJI D-Gamut/D-Log is not compatible with the mavic 3 ?

i hope someone can illuminate us on this matter or maybe a working workaround

Hey man,

I think I’m having a similar issue with d-log.
Looks like my Mavic 3 classic and Mavic 3 pro doesn’t match at all.

djiuser_eqlATrIanKrK Posted at 11-18 16:52
Hey man,

I think I’m having a similar issue with d-log.

If you happen to be using a recent version of Davinci Resolve for color correcting, there is an option on the color page called "shot match to this clip."  To use this feature, you would first color correct the footage from whichever of the drones that initially corrects the best. Then, selecting a clip from the "less accurate" drone you could try to color match it to the first drone's corrected clip. It's not perfect, but IMHO it provides a very good base match, which can be improved by further micro adjustments.  Other editing sofware may also have a similar feature.