By your comment, you don't understand how any color management works, but then, very few actually do. I work for/with/teach pro colorists, and have for a decade now ... and am in Resolve daily also. And participate on the BM and LGG forums as well. My system and vieiwng ambient light and surround color/brightness has to meet darn close to broadcast specs, and that is kind of a pain to do, in reality.
And actually, much of color management is way different than most editors think it is.
Another thing to understand, and this is drummed into all colorists: no one ever will see exactly, or necessarily even close to, what you see on your screen. Colorists spend more money on Reference displays and calibration devices and gear than your entire setup including cameras and computers not because anyone ever sees exactly what they do ... that is not physically possible.
You can't even get absolute visual agreement on two "identical" screens, calibrated the same, and fed the same signal from the same system. That's been demonstrated over and over. Manufacturing simply cannot produce that level of identical devices.
All that expense and effort to get a highly calibrated screen image only means a colorist's work will, in relative terms! ... look like other professionally produced media on any screen out there. With the same per-screen differences applied.
Further, NO one grades to the Apple display transform for broadcast/streaming, did you know that? Has that ever been an issue for you watching things on your system? So every pro media you watch was graded different than what you see. That's Reality.
Rec.709 has, ever since "flat screen" displays have been available, specified in the standard that the display shall do a transform roughly equal to gamma 2.4. This was the Bt.1886 addendum to the Rec.709 standards. Posted officially right when flat screens came into existence.
There was never a LUT applied to set the viewing gamma in pro workflows. That display transform is expected to be done in the display device iteself, and is therefore not included in changed data to the file. Or otherwise, any properly specced display system will double-down on darkening the file.
It has always been expected that you graded with the appropriate display transform applied, and that later viewing devices would also use the correct display transform. But then Apple did this ... 1.96 thing. Ouch.
There isn't any actual fix, no matter what you see online. As thanks to Apple, we have two groups of screens, running very different Rec.709 display transforms:
- Macs without Reference modes which use (essentially) a display transform (gamma) of 1.96.
- All other screens, including Macs with Refence modes set to HDTV which apply the correct display transform (gamma) of 2.4.
The latter group includes TVs, any proper broadcast computer system, Android, PCs, and again the Macs with Reference modes.
Some folks argue for using particular NCLC "tags" for the file, but that only helps in some systems, and even then, only under the right circumstances.
So it's pick your poison time! .... and oh that's fun. Not.
1) Set Viewing gamma to 1.96 in Premiere's expanded color management system. Then grade your sequence. And the export will look very similar to within Premiere when viewed in QuickTime Player, Chrome, or Safari. And on other machines in group 1 above.
BUT ... in VLC and Potplayer on your machine, and on all rigs in group 2 above, that file will play too dark and oversaturated.
OR ...
2) Set Viewing gamma to 2.2, unless you are grading in a very dark room as a colorist does!!! ... and your exported file will be light and low in saturation in viewing for machines in group 1 above, but looking 'normal' in all screens in group 2 above.
3) ONLY if you are grading in a very dark room, use a viewing gamma of 2.4 rather than 2.2. As it is the viewing conditions while grading that determines whether to use 2.2 or 2.4.
1 Correct answer
16
Replies
AdChoices