This is phenomenal, in the facebook forums in Resolve, color grading and Premiere - this problem comes up daily. I now have a source that I think explains the whole thing and dropping a share to this video is a lot better for my carpal tunnel syndrome typing all this out. Wish I could click the like more times. Great information.
Almost two years later, this still seems to be the best (hybrid) solution for web delivery. So much misinformation or incomplete solutions out there, it's the blind leading the blind. Would be great if Jay has an update on this for 2022/2023 or if Cullen could do either a livestream or video deep dive on this. The 'hackiness' still irks some OCD part of me. At least CSTs making switching deliverables easier. Thanks Le Labo de Jay.
@@StandOnAMountain I think that the transfer function ( second digit in the tagging ) needs to be updated by whoever controls that to include actual gamma. When you render a file in most software at rec709 2.4 it will be tagged 1-2-1 the number 2 is undefined. The think the gamma should be present in the actual video. they can fix this because all tagging for HDR is spot on.
@@JimRobinson-colors what say you to this though? grading in a 2.4 environment on a 2.4 display will look the same as being in a 2.2 environment on a 2.2 display. This means that you should do NO conversion for the internet at all, just tag as 1-1-1 for client review if required. Otherwise you will make things too dark by converting to 2.2! (On the other hand, if you are also converting to 2-2 for client review, you'll end up getting things right but in a backwards way because they can tell you to brighten things up). Another way to think of it: grade on a 2.4 display in a dark (i.e. 2.4) environment. Turn the lights on, thus creating a 2.2 environment - the grade will now look too dark. Switch your monitor to 2.2, brightening it - now it will look correct again. So if you convert to 2.2, it will now look too dark despite changing the environment and the display. You can have your color space transforms convert to 2.2 if you grade on a 2.2 display in a 2.2 environment, but do it at the start of the grade and don't switch later (don't switch even if you have the wrong conversion to start with as your grading decisions would have accounted for this already).
@@joshpickup2581 When services consume their upload, I'm pretty sure they set the gamma they want for the service. Wouldn't surprise me at all if youTube would transcode a gamma 2.4 to 2.2. The lighting environment does change your grading perception and that's where most people make their mistake. With that said a darker grading in rec 2.4 is way easier to control then lighting a grading suite for 2.2. Turning on the lights is a simple explanation. They need to be placed and the right temperature with the now lighted walls being a consideration as well.
Thank you so much for such an in-depth video, which clears the long existed confusion.This is by far the best video for this issue. Merci à vous, c'est génial!
Thank you so much! It's been almost 2 years since the release of this video - has anything changed in that field recently? Any updates in software and web video services you're aware of? At least I noticed that even default export from Davinci 18 is tagged as 1-1-1, not as it used to be.
yes , XDR monitors are a thing now. things are VERY different with XDR, resolve defaults to rec709(scene) as output colorspace which by default sets the tags on export, before it was rec709(gamma 2.4)
The best most in-depth explanation I have seen - thank you to all that contributed With Blackmagic love of manuals wouldn’t it be great if they would make a color management guide which includes recommend workflows and testing?
Thank God for this. I haven't tried this as I'm following the rec709A "hack"...and I'm on PC! I know this hack is for Mac but it seems PC runs into the same problem, unless I'm imagining it. I do notice that there is no color shift using rec709A when YT is the audience. When I use 709-2.4 and grade to that...yes, the video looks washed out (perceptual contrast is lost) on Apple products. Anyone else using the 709a hack on PC?
Just want to say thank you for this video. I've watched it many times over the last year or so as I learn more about color work and today was the first time I was able to fully understand everything along the way. It was extremely satisfying and I can not thank you enough! However I am confused by one thing. I did a test export of a clip of a model with darker skin (as those deep reds I find to be most affected by gamma shifts). Everything was as you said EXCEPT that the 709-a workflow in iOS looked basically identical to what I was seeing in Resolve + QT and in Chrome. I did not measure the displayed values as you have but we can say "that looks the same" with some certainty that there is NOT an evident gamma shift. At least this was the case on an iphone 15 and 12 that I had available. Has something changed in iOS that it now displays 709a tagged files "correctly"? If so that would seem to be major bonus to the 709-a method no? If for no other reason (and I know this might sound cynical) that if I'm grading content for web I know 2 things: 1. I can't control what screen everyone will view on 2. ... but I can be damn near sure that 99% of art directors are reviewing footage on an Apple product of some sort be it in QT or Safari / Chrome on an iPad or iPhone. Please if you have time do let me (us) know if something has changed regarding how iOS devices are displaying 1-1-1 clips tagged with 709-a and thank you again for your informative video!
Hello well time has passed since I made this one. I went to understand a simple fact. We just need to stop exporting rec709 gamma 2.4 It’s useless and source of many pain Just export rec 709 rec709 from resolve and everything will be consistant. Rec 709 gamma 2.4 in dark room will have same perception than rec 709 in bright environment that’s the key
@@LELABODEJAY thank you for taking the time to respond! Just for clarity you would suggest exporting 709/709a with a color space transformation or not even? Just simply grade in 709 2.4 controlled environment and tag as 709a?
@@stephanalessi7720 grade rec 709 gamma 2.4 and export rec 709 rec709 don’t add A Same for everyone So your clients won’t have shift and everything will be much more easier
@@LELABODEJAY I want to believe but the difference between what I see with client in Resolve in the 2.4 environment and what they see when they look at the file on frame / vimeo etc is still too different. Do you have a website or could you send me your email address? I would love to book a 1 hour session with you to try and understand more!
@@LELABODEJAY I don't understand..my exported video after using gamma 2.4 / 2.2 and export in rec709 looks so bright and washed out, so how is it better?? I never had problems before with Rec709-A (it was accurate everywhere except on VLC). I also send my export videos to clients that have a pc and they send me sreenshots when I need to make some modifications, but the screensot looks also the same as my exported video. And when I watch TH-cam on my TV, the colors are also accurate. So I don't understand why suddenly we can't use Rec709-A?
Merci pour ces explications, c'est la première fois que je vois une vidéo aussi détaillée. Malgré tout je suis encore un peu perdu. J'ai un écran LCD sRGB et un écran macbook pro m2. Mon objectif est que le rendu soit cohérent entre Resolve et TH-cam. Je voudrais bosser sur un color management à base de node et et surtout une timeline en Davinci Wide Gamut. Dans ce cas là faut il cocher dans les préférences "Use Mac display color profiles for viewers"? Que dois je mettre dans mes réglages Project Setting/ Color Management? Quelles seraient les réglages CST? (Ce que je faisais à l'origine sur par exemple du sony CST Slog3 -> DWG ---> Etalonnage ---> CST DWG -> Rec709 Gamma 2.4) Que dois je choisir dans les préréglages d'affichage pour mon écran de macbook pro m2 car il propose (Apple XDR display p3, Apple display P3, BT709 BT1886, sRGB etc)? Et pour finir quels seraient les bon tag à mettre à l'export final? En espérant que ma question soit claire
Thank you very much for this work. I'm currently doing a similar experiment but when comparing the "ultimate fix" with the "Rec.709-A Fix" im not seeing an improvement. Did anything change since you uploaded the video or am i missing something? Thanks a lot!
J'arrive 2 ans après la publication de cette vidéo, donc déjà merci pour ce travail énorme ! Maintenant j'ai une question qui va peut-être paraitre un peu bête mais, sachant que de toute manière les clients finaux de nos vidéos (les spectateurs) regarderont notre travail sur 1001 écrans différents, sur 1001 navigateurs différents, sur 1001 players différents, pourquoi s'embêter avec tout ça ? Quand je regarde un bluray sur ma télé et que je regarde ce même bluray sur mon ordi, ou projeté sur un mur, comment font les étalonneurs pour ne pas tomber en dépression à l'idée de toutes ces images qui ne correspondront pas au travail qu'ils ont fait ? 😅
On prend des anti-dépresseurs ! Vaste sujet en effet On étalonne sur un écran de référence car c’est la référence on contrôle sur une tv haut de gamme mais après c’est advienne que pourra
@@LELABODEJAY du coup si sur ma tv "haut de gamme", mon export en rec709-a apparait identique à mon vidéo exportée, je vois pas pourquoi me prendre la tête 😅
@@JS_Leger ta tv s’en fout que tu sois en a gamma 2.4 ou quoi que ce soit. Elle ne lit pas les tag. Elle affichera avec son gamma natif SDR normalement proche 2.4
OMG THANK YOU! merci!!!! (j'ai reconnu l'accent hehe). The last fix is definitely the best! The difference is still there in VLC but not as much as every other methods. Par contre est-ce que je dois faire une conversion 2.4 vers 2.2; ou je peux simplement choisir gamma 2.2 dans les paramètres du projet (output color space : Rec.709 Gamma 2.2)?
I tried the solution for web and it’s got it looking closer to the original grade on my Mac. I think the issue is I use my PC for grading, perhaps I should switch to my Mac?
Merci beaucoup, des explications que je n ai jamais trouvé sur le net....très,très pro.un peu compliqué pour mon niveau, mais commence à éclairer ma lanterne (rec 709) . Bravo. Fred.
Amazing video, thank you. Question, I am using a M1 and have calibrated my display with a calibrite display plus HL - I'm assuming this profile stays in the system settings and isn't used within Resolve to help with color accuracy? I color manage with nodes - so I convert the footage to DWG and Da Vinci Intermediate for grading, then (at timeline level) convert DWG Da Vinci intermediate to rec 709 (Color Space) and rec 709-A (Output Gamma) with a CST. In this case in the delivery page I would leave the tags as same as project right? Am I doing this correctly?
Tanks. It’s getting old should update it Once again rec709A is a trick like my ultimate fix that only work for Apple. This issue only occurred in Mac OS X not on iPhone iPad or in pc world The easiest way is to stop exporting in rec 709 gamma 2.4 from grading as it clearly make hell on earth
Excellent. 1. How do these answers translate to the new MacBook Pro computers? I have an M1 max. It has a BT.1886 profile. 2. How does this work with resolve color management at the project level, instead of the node level? There are a lot of advantages to using project level color management, vs node level management in my experience. I'm still having issues getting my decklink outputs to match my internal viewer and my exports. I can get there with node level management, but not project level management, like I want. Thanks for the AWESOME video man. Subscribed.
Someone correct me if I'm wrong. This response is to test if I understand it correct: 1. Your Mac display profile does not matter, as long as it is the correct profile for your display. Mac OS color management will handle the rest, but to view files correctly in the viewers in resolve you should set: Use mac Display Color Profiles for Viewers. 2. I think you sound set your resolve color managend settings to red709 gamma 2.2 for web export and than still change the tags on export to rec709 and rec709a.
I tested Rec 709 A solution and got a different result from yours, my source on Davinci looks identical to Quicktime, youtube and Vimeo on every browser on Mac, iOS is identical too. I couldn't test on windows and android because I don't have them.
@@ByRaphI created a serial node at the end of my color grade and transformed my timeline color space to rec709 gamma rec 709 A, then at the export setting I changed color space to match (rec 709 gamma rec709 A)
I did this Rec 709A solution and it looks identical in Quicktime and on Vimeo when I upload. However, if I open the same file in VLC, it looks more saturated / higher contrast. I'm wondering if someone viewing on a PC will see what I'm seeing on VLC, and if so, is there any solution to this?
@@skermani Either you don’t use VLC player at all or ask your client which media player they use and export what matches that player. Or you can export one file that gives you best results when uploaded to socials or web and let them know which player the need to use for playback.
Hello, thank you very much for the video. I am working with Arri Alexa Raw files. I am working on a MacBook Pro that is calibrated with a datacolor Spyder X to ITU-R Rec. BT.709. The suggested Gamma is 2,22. So ITU-R Rec. BT.709 should be Rec. 709 BT.1886. The suggested whitepoint in the datacolor app is x 0,313 y 0,329. My Macbook is calibrated to 120 cd/m2. So it is probably not quit BT. 1886 with a suggested 100 cd/m2. I know that my MacBook will never be color accurate as a Flanders monitor. In DaVinci Resolve 17 I am working in ACEScct. Version 1.3. No Input Transform because I am working with Raw and ProRes files. Each clip is individually assigned an IDT. "Apply ACES reference gamut compression" is off. ACES Outputransform is set to Rec. 709. Netflix in one of their color management tutorials said that this Rec. 709 is BT.1886 so I assumed it is Gamma 2.22. th-cam.com/video/u9Rvm5xiuhk/w-d-xo.html "Process node LUTs in..." is set to "ACEScc AP1 timeline space". I have checked the box "Use color space aware grading tools". Graphic white level is set to 100nits. In the "project settings", "Master settings", I have set the Video Monitoring Data levels from "Video" to "Full". Video with depth is set to 10Bit. I did some tests with "Use Mac display color profiles for viewers". Normally it is and was deactivated. When deactivated, the highlights looked bright and colorful as they should. When I exported the video to VLC it looked the same as in DaVinci Resolve, as you said in the video. QuickTime, the NLC Tag was originally HD 1-1-1, it was too bright and washed out. This was before I set up ACES. I used DaVinci Resolve YRGB before. As I understand the NLC tag 1-1-1 is right for Rec.709 BT.1886 Gamma 2.22 for web delivery like Vimeo and TH-cam. When I switched to ACES the ODT is set to Rec. 709 (BT.1886) as mentioned before. Now when I export (Use Mac display color profiles for viewers are still deactivated), and set the color space tag "same as project and set the Gamma Tag to "same as project" my export says when I press command+I that the NLC Tag is 1-2-1 (ITU R BT 709 - UNSPECIFIED - ITU R BT 709). Why is that, that it is now unspecified? In VLC the video looks like in DaVinci Resolve. In Quicktime it looks much better than the 1-1-1 washed out version. But the Image is a tiny bit too dark. The highlights mostly green are not as saturated in DaVinci resolve. The version with the NLC Tag 1-2-1 is washed out in Vimeo and TH-cam on Safari. So the transfer curve is not specified. On Firefox it also does look washed out. Firefox is color managed since recent versions. I have now activated "Use Mac display color profiles for viewers". The image in DaVinci now looks the same as in Quicktime. But it does not change the fact that the transfer curve is not specified. "Automatically tag Rec. 709 Scene clips as Rec. 709-A" ist NOT Activated. I don't want to work in Rec.709-A. My movie will be mostly watched on TH-cam, maybe on Amazon Prime and in cinemas. I am primarily preparing the video for TH-cam. So I want the film to look kind of the same on all mobile devices. So all in all QuickTime should not matter that much. As a test I have changed the color space tag and Gamma Tag. "Color Space Tag": Rec. 709 "Gamma Tag": Gamma 2.2. I assumed because I use Rec. 709 BT.1886 that Gamma 2.2 might be correct. But in Quicktime the file looks way too bright. It is not that extreme as when I used DaVinci resolve YRGB, so ACES might have something to do with it. Because I chose a gamma tag of 2.2 the file says it has an NLC Tag 1-4-1 the transfer function of gamma 2.2 seems correct but not visually. As we can see in your list it is ITU-R BT.470M (Gamma 2.2). But this 1-4-1 is also wrong. If possible I don't want to cheat the NLC Tags and have a work around. Now I have exported it again as "Color Space Tag": Same as Project. Because I want to keep Rec.709. "Gamma Tag": Gamma 2.4. In Quicktime the file looks the same as in DaVinci Resolve as I am not mistaken. The NLC Tag is 1-2-1. I am wondering now. My ACES ODT is Rec. 709. I always assumed that it is BT.1886 with Gamma 2.2. Why does the image looks more or less correct to when I have not chosen a Gamma Tag? The video looks the same as "Gamma Tag": Same as Project. To my understanding this would mean that the ACEScct ODT Rec.709 is not actual Rec. 709 BT.1886 Gamma 2.2 but rather Rec. 709 Gamma 2.4. I have calibrated my Macbook to BT.1886 Gamma 2.2 so when I disable "Use Mac display color profiles for viewers" in DaVinci the Image does not use my calibrated display profile Rec.709 BT.1886 Gamma 2.2 but uses Gamma 2.4. Is this why my image looks better in DaVinci? It is only slightly brighter and the green is more saturated. In DaVinci resolve I can't find the Gamma Tag (NLC 1) ITU R BT709. So that when I export the clip I receive a NLC 1-1-1. Originally before I started using ACEScct I didn't not use any color space tags or gamma tags and received a NLC 1-1-1 but the image was way to bright. (less to no contrast) What should I do? Should I use Rec.709 Gamma 2.4 and recalibrate my Mac to Rec.709 Gamma 2.4 instead of using a calibrated display to 2.2? How to fix the washed out colors when uploading to TH-cam and Vimeo? (NLC 1-2-1)? Is it called NLC or NCLC Tag? Thank you very much.
I have once again exported it as process 422hq "Color space tag: same as project", Gamma Tag: Gamma 2.2 (same as my monitor is calibrated to). When I upload it to TH-cam it looks way too dark and oversaturated. The NLC Tag is 1-4-1. I have changed the color space tag from "same as Project" which should be (Rec. 709) to Color Space Tag "Rec. 709", Gamma tag is still 2.2. I get a NLC Tag 1-4-1. You have talked about Full and Video Range. When I export a Proress 422hq in full Range (Tags Rec.709; Gamma Tag 2.2) it is saturated and looks like the davinci viewer when "Use Mac display color profiles for viewers" is disabled. So in general it is brighter. When I use video range. In Quicktime it is washed out (NLC-1-4-1) but on TH-cam it kind of looks okay (close enough). In full range with no gamma Tag (same as Project) it is still to dark in QuickTime (NLC 1-2-1) and does not look like the davinci viewer. I have exported 31 versions and still haven't found a solution. Is there an option that TH-cam does look the same as in the resolve viewer? Is it right that NLC 1-4-1 is best for TH-cam? Does full Range automatically use Gamma 2.4 (NLC 1-unspecified-1) (Nlc 1-2-1) even if the project gamma is 2.2 Bt.1886 and Tag is set to use project setting? Or is it because aces gamut ODT is Rec. 709 Gamma 2.4 and not Rec. 709 Bt.1886 Gamma 2.2?
Sorry for getting back so late But honestly so much informations so much to read. Sorry I have to see what’s happening. This description is way too long Contact me via email will see if we can arrange a team viewer Lelabodejay@gmail.com
@@worldproductionentertainment i'm having the EXACT problem that you are having and the only "solution" that I could get was to set the output color space of the whole project to rec 709-a, that was getting as close as possible to quick time as possible, as well as on my iphone and youtube. But as the video said, that's only a part of the problem because we are on Apple's ecosystem, so it will look great on their products, but i dont have windows devices to test to see how they would look on comparison. In the end of the day, from what i've searched and tested, was to either find a middle ground, like the suggestion in the video and live with it, or do different exports for the different platforms, which kinda sucks. Also, when opening on different players, like IINA and VLC, the colors are different as well, so i don't think there is a solution for all of this that will work 100% all of the time on all platforms but if it has, i would like to know as well because it's kinda infuriating :( Also, grabbing still from the colors page changes the image a lot, making it brighter and with less contrast
Amazing video, thank you! I'm exporting an .MXF for TV from Resolve with Gamma 2.4. The tag of the exportet file is 1-1-1 for some reason. When I open the file in the MetaData Editor of AMCDX Video Patcher, the Frame MetaData is greyed out and I can't change the tag. Any idea what my problem could be?
@@rusch_meyer it happened to me sometimes with latest version I don’t know why. But remember tag is only for Mac OS your tv don’t care it will be displayed with the gamma of your tv which should be around 2.4
An excellent video! Thanks you for the detailed explanations. One question, I’m using DaVinci Resolve on Windows. If I use node based colour management and use an output CST to Rec.709 2.4 at the timeline level when grading, should I still add a CST after to convert the output from gamma 2.4 to 2.2 and export using gamma tag Rec.709-A or can I simply change the output CST from Rec.709 2.4 to Rec.709 2.2 and use Gamma tag Rec.709-A on “Delivery”? Is there a difference?
Did you solve this? Doing this now and I don't see the reason to add an extra node to covert to 2.4 when the timeline is already 2.4 and you convert to 2.2.
@@JamesWilliams-in3py No, I never received a response. My timeline is in DWG Intermediate, so I use the last node to transform to rec709 2.4 when grading.
I could be wrong, but I think the Rec.709-A is generally the solution for mac users. As for windows I would edit as normal. For web, my project and timeline would be Rec.709 Gamma 2.2 and export with those same settings. The point of using gamma 2.4 is if you know its for broadcast. My take away is be intentional with what you're startup settings are and if you get caught in gamma 2.4 but need to export for web, add a node to timeline and convert to 2.2 (or 709-A for apple devices)
Toujours aussi complet. Est-il prévu de voir le fonctionnement du Color management avec Media composer ? Je sais qu'ils ont amélioré leur Color management. Encore bravo.
@lelabodejay I find that this works great using a Windows Machine - all looks really good - my only problem is playing back on QT on Windows (not tested on a mac yet) - the colours are still washed out. Any possible solution for this? Matches really well when uploaded to youtube though - QT player is the only outlier at the mo. Thanks for all your hard work!
QUESTION: For BlackMagic pocket 6K (braw), which color space would be the best (and will it need a timeline gamma conversion), with best I mean the least problematic result for web purposes across different platforms and devices?
@@LELABODEJAY most of the time all clips are Braw, so would you suggest to leave the camera settings in the color page on timeline instead of selecting clip?
@@LELABODEJAY Thanks for the informative video on the subject. For the web, and to avoid the conversations from 2.4 to 1.9, it would be easier to set a calibrated external monitor to gamma 1.9, and tag Rec.709-A in Resolve Timeline and the Delivery page. This is using, Use Mac Display Color Profiles for Viewer. Both Resolve Viewer and the external monitor would be closer than 2.4 after the conversion. This is what all MacOs player and browsers supporting color management are expecting to see anyway, (1-1-1) = gamma 1.9.
@@LELABODEJAY What about the new Rec709 (Scene) timeline color space in Resolve 17? It allows for 1-1-1 tags, but the rendered files still don't match. And also, in addition to Rec709-A as an output gamma tag, there's now Rec709 and Rec709 (Scene). How do those influence the way we can mix and match gamma tags and CSTs/timeline color spaces to achieve greater consistency?
Hey, im working on a project for a client. On vimeo everything looks washed out My grade in davinci is Blackmagic Wide Gamut, Blackmagic Film Gen 5 to Davinci Wide Gamut, Davinci Intermediate to Rec 709, Cineon Film Log and with a 2383 Film Lut as the last node back to Rec 709 2.4 I export the graded file as DNxHR 444 12bit video levels Rec 709 2.4 and import it in After Effects. After Effects expects DNxHR 444 to be full levels and converts it to video levels. Its the legal vs full issue with DNxHR 444 in adobe software. I did think this was the issue. Would exporting the file at full levels for the import in after effects fix it or not. The grade is finished, so full levels for after effects and web export doesnt need to be. It would be just to fix after effects from washing out the dnxhr 444 file expecting the dnxhr 444 file rendered at video levels to be full range and converting it to video levels. Should i change to Rec 709 2.2 with CTS and Rec 709-A in export settings in davinci before exporting the grade or before rendering the delivery in after effects
Hello Hard to tell from distance Why are you in full Video should be levels If you grade on monitor video it should be levels Tv and web expect levels After effects ignore gamma 2.4 tag It will export 1.1.1 rec 709 May be it’s the first step to watch at If your content will only be on web yes you should try the fix but remember it’s a fix and not accurate Rec 709 gamma 2.4 is for tv in darkness Rec 709 tag 1.1.1 is for monitor on bright environment. Perception should be the same…. Should ! Such a mess Tell me if that helps
@@LELABODEJAY Thank you for the quick answer! I need to say im on windows. I would export as Full Levels DNxHR 444 12bit Rec 709 2.2 Rec 709-A in resolve for after effects. In after effects i would then add the levels effect-computer to video profile to convert full range back to video levels. After effects wouldnt convert video levels to video levels as so the footage wouldnt be washed out and because of the levels effect, the export from after effects for delivery will be in video levels again. What should my timeline colorspace, gamma and my export in after effects be, Rec 709 2.2 for time line and Rec 709-A for export right
@@arceus2015 rec 709 A is only for Mac. No need on windows Why are you going to after effect ? Full level should not be a problem Just export in auto from resolve In windows you don’t have tag problems so if your colors are washed it’s not a tag problem Unless you are in aces or in dolby workflow or with 3D exr elements you should not work in full No need and can cause issues that you are currently experiencing
@@LELABODEJAY In resolve on windows, should i still make a node and convert my timeline with a CTS from Rec 709 2.4 to 2.2 and export tag as rec 709 2.2 for vimeo, instagram or youtube meaning web upload? After Effects has this issue with the DNxHR 444 Codec because it expects dnxhr 444 to be full range, when i put it on Auto in resolve and it auto exports in video levels, when i put it into after effects, the colors will be washed out. In forums they say export in full range for after effects to interper the codec correctly Im rendering in Apple ProRes 4444 XQ in After Effects for the best quality for archive file and Apple ProRes HQ for web.
@@arceus2015 you should export rec 709 rec709 tag won’t matter on windows Test several method to see what’s the less corruptive My video is 2 years old some things has evolved Really hard to tell without clip and testing and seeing
you say 'don't touch the grade', but then you apply CSTs. do you mean you grade and THEN apply the CST, just for the export-for-web? for example your ultimate fix: do you grade on your rec.709 gamma 2.4 timeline and monitor (ignoring your sRGB screen), then apply the CST only for the web export? and you only set your UI monitor to sRGB for doing your testing right? at least on my system, Display P3 looks much closer to the video monitor, and sRGB will confuse my clients with darker shadows and more saturation.
Hello Sorry for late reply Yes you do your grade on reference monitor Then apply cst for web delivery I don’t really trust my ui For your information if your monitor is Apple P3, it’s a P3 gamut combined with srgb gamma.
Hi I also found that 709A. works well in all applications. if you have a broadcast project like a tv show, would you still use 709A or change the tag to 2.4?
Thank you a lot for this video, really clarified some things! Quick question: What would happen if you export as follows: Rec709/Gamma 2.2 and then tag it as 1.4.1 (Rec709/Gamma 2.2/Rec709)? Would that tag get misread / re-tagged by some of those apps as well?
It will be ignored as everyone ignore tags except Mac color managed applications The only way to avoid this shit is a lot more simplier that what I have explained in this video Stop tagging rec 709 gamma 2.4 So we will never have shifts Color sync can be so tricky and leads to error The ultimate fix is a trick Like every trick it generates problems. I should redo a video about it This one is old
@@LELABODEJAY Thanks for the reply. Would be awesome to get an update video... even if it is just a short guide without elaborate explanation. I've found your video when I re-searched a color space problem I had. I also posted on a bunch of message boards about it and got referred to your video at least on 3 separate instances.... Your explanation is by far the best and most thorough one I found.
Salut Jay, merci encore pour la vidéo! J'aurais besoin de ton aide si tu as 2 min: Je suis sur Resolve Windows et mon moniteur est un Flanders CM250 calibré. Si j'exporte un QT 1-2-1 de Resolve et que je le visionne ensuite sur mon Mac, c'est beaucoup plus sombre que sur mon Flanders. Est-ce signe que ma calibration n'est pas bonne? Par contre, si j'utilise une node CST Gamma 2.2 et que j'exporte en 1-1-1 (avec le Rec709a), c'est plus proche. Salutations du Québec!
What you say about converting to 2.2 for the internet is incorrect. Ironically you get things right early on when you say that grading in a 2.4 environment on a 2.4 display will look the same as being in a 2.2 environment on a 2.2 display. This means that you should do NO conversion for the internet at all, just tag as 1-1-1 for client review if required. Otherwise you will make things too dark by converting to 2.2! (On the other hand, if you are also converting to 2-2 for client review, you'll end up getting things right but in a backwards way because they can tell you to brighten things up). Another way to think of it: grade on a 2.4 display in a dark (i.e. 2.4) environment. Turn the lights on, thus creating a 2.2 environment - the grade will now look too dark. Switch your monitor to 2.2, brightening it - now it will look correct again. So if you convert to 2.2, it will now look too dark despite changing the environment and the display. You can have your color space transforms convert to 2.2 if you grade on a 2.2 display in a 2.2 environment, but do it at the start of the grade and don't switch later (don't switch even if you have the wrong conversion to start with as your grading decisions would have accounted for this already).
Does this mean when sending files for these 3 web deliveries you have to change the timeline output color space and adjust the footage to look like your final grade across all three output tags? Pls let me know if am correct
Hi just export in rec709 for web and do the Cst to convert gamma 2.4 to 2.2 Don’t touch the grade Remember that you have gamma 2.4 for monitor in dark environment and gamma 2.2 was set for monitor in office in daylight You can’t have one file thad works everywhere 🤷♀️
@@LELABODEJAY you mean I should do a cst conversion to my final grade at the end? Is that correct? My only screen is a macbook with p3 display I don’t mind having different files for all deliveries.
@@LELABODEJAY One last thing that confuses me when using srgb monitor is the workflow because if my monitor is srgb and DaVinci source timeline is rec 709 2.4 and with my understanding the use mac display option is supposed to be turned off here. So how do I make sure that what am seeing on resolve on the srgb monitor will be the same on the web, do I still need to do the Cst to convert gamma 2.4 to 2.2 or how?
Thanks a lot for this explanation, CST output gamma 2.2 and export advanced setting output gamma to Rec.709-A worked to produce the video with correct colors when viewed on Apple Studio Display with P3 gamut
@@LELABODEJAY Thanks for your reply. I think there must be a Davinci version difference. I see it in a very different way. In my project ALL clips already have a color node (I can't identify which type) because the "settings" tab next to "library" is greyed out. Also, if I add an explicit "color space transform" there's no "gamma 2.2" or "gamma 2.4" options on the Input Gamma and Output Gamma menus. My version of davincy (version 18) looks very different than your video from 2 years ago: I suspect many things have changed. Also, how do you tell that a color space transform should apply to the entire timeline, rather than just one single clip?
@@lsfornells yes lots of things have changed. You can apply effect to all timeline on the top of the window on the right you have two dots click on first you apply to clip on second you apply to the timeline
Blackmagic should release a Mac viewer. TH-cam should respect ICC tags. Job done.
I love how you appreciate everyone's help, regardless you spent most of the effort creating this. Kudos!
This is phenomenal, in the facebook forums in Resolve, color grading and Premiere - this problem comes up daily. I now have a source that I think explains the whole thing and dropping a share to this video is a lot better for my carpal tunnel syndrome typing all this out. Wish I could click the like more times. Great information.
Thanks. Feel free to share 😊
Almost two years later, this still seems to be the best (hybrid) solution for web delivery. So much misinformation or incomplete solutions out there, it's the blind leading the blind. Would be great if Jay has an update on this for 2022/2023 or if Cullen could do either a livestream or video deep dive on this. The 'hackiness' still irks some OCD part of me. At least CSTs making switching deliverables easier. Thanks Le Labo de Jay.
@@StandOnAMountain I think that the transfer function ( second digit in the tagging ) needs to be updated by whoever controls that to include actual gamma. When you render a file in most software at rec709 2.4 it will be tagged 1-2-1 the number 2 is undefined. The think the gamma should be present in the actual video. they can fix this because all tagging for HDR is spot on.
@@JimRobinson-colors what say you to this though? grading in a 2.4 environment on a 2.4 display will look the same as being in a 2.2 environment on a 2.2 display. This means that you should do NO conversion for the internet at all, just tag as 1-1-1 for client review if required. Otherwise you will make things too dark by converting to 2.2! (On the other hand, if you are also converting to 2-2 for client review, you'll end up getting things right but in a backwards way because they can tell you to brighten things up). Another way to think of it: grade on a 2.4 display in a dark (i.e. 2.4) environment. Turn the lights on, thus creating a 2.2 environment - the grade will now look too dark. Switch your monitor to 2.2, brightening it - now it will look correct again. So if you convert to 2.2, it will now look too dark despite changing the environment and the display. You can have your color space transforms convert to 2.2 if you grade on a 2.2 display in a 2.2 environment, but do it at the start of the grade and don't switch later (don't switch even if you have the wrong conversion to start with as your grading decisions would have accounted for this already).
@@joshpickup2581 When services consume their upload, I'm pretty sure they set the gamma they want for the service. Wouldn't surprise me at all if youTube would transcode a gamma 2.4 to 2.2.
The lighting environment does change your grading perception and that's where most people make their mistake.
With that said a darker grading in rec 2.4 is way easier to control then lighting a grading suite for 2.2. Turning on the lights is a simple explanation. They need to be placed and the right temperature with the now lighted walls being a consideration as well.
TLDR: Just change the export settings in Davinci Resolve to Color Space Tag: REC 709 and Gamma Tag: REC 709-A - worked like a charm! Thanks :)
Thank you so much for such an in-depth video, which clears the long existed confusion.This is by far the best video for this issue. Merci à vous, c'est génial!
Whaou !! Quel boulot de récap ! Merci mille fois pour ce tour complet !
DAMN! This is huge! Thanks ♥️ 👏🏽👏🏽👏🏽👏🏽 we should share this video and show it to rest of the world!
Feel free to do it 😊
This is amazing.. best video i've seen this year so far!! Thank you so much!
Jay, you did a great job. Really important for deliveries in my opinion. Bravo
Thanks 🙏
Thank you so much! It's been almost 2 years since the release of this video - has anything changed in that field recently? Any updates in software and web video services you're aware of? At least I noticed that even default export from Davinci 18 is tagged as 1-1-1, not as it used to be.
yes , XDR monitors are a thing now. things are VERY different with XDR, resolve defaults to rec709(scene) as output colorspace which by default sets the tags on export, before it was rec709(gamma 2.4)
@@FinnJaeger1337what does this mean for the end user?
Excellent, finally some one that explains everything (a lot of information though), thank you Jay.
The best most in-depth explanation I have seen - thank you to all that contributed With Blackmagic love of manuals wouldn’t it be great if they would make a color management guide which includes recommend workflows and testing?
I love you! Thanks from the bottom of my heart for bringing light into this mess. Things are a lot clearer now.
Thanks 🙏
Thanks a million for your time preparing such a delightful 'answer' to this dreadful subject . Appreciated !!
and for Finn +1 Karma point
Very technical but VERY VERY useful. It explain why we struggle with colors. Thanks. :))
Thank you 🙏
I have been looking for these answers for a few years , Thank you
Thank God for this. I haven't tried this as I'm following the rec709A "hack"...and I'm on PC! I know this hack is for Mac but it seems PC runs into the same problem, unless I'm imagining it. I do notice that there is no color shift using rec709A when YT is the audience. When I use 709-2.4 and grade to that...yes, the video looks washed out (perceptual contrast is lost) on Apple products. Anyone else using the 709a hack on PC?
Great video. Thank you! Finn, thank you as well.
This is such a great explanation. Thank you so much!
Thank you. This is such a complicated topic and I *think I understand now. Mainly I shoot & edit for web delivery, so I will study again from 24:18 :)
Just want to say thank you for this video. I've watched it many times over the last year or so as I learn more about color work and today was the first time I was able to fully understand everything along the way. It was extremely satisfying and I can not thank you enough!
However I am confused by one thing. I did a test export of a clip of a model with darker skin (as those deep reds I find to be most affected by gamma shifts). Everything was as you said EXCEPT that the 709-a workflow in iOS looked basically identical to what I was seeing in Resolve + QT and in Chrome. I did not measure the displayed values as you have but we can say "that looks the same" with some certainty that there is NOT an evident gamma shift. At least this was the case on an iphone 15 and 12 that I had available.
Has something changed in iOS that it now displays 709a tagged files "correctly"? If so that would seem to be major bonus to the 709-a method no? If for no other reason (and I know this might sound cynical) that if I'm grading content for web I know 2 things:
1. I can't control what screen everyone will view on
2. ... but I can be damn near sure that 99% of art directors are reviewing footage on an Apple product of some sort be it in QT or Safari / Chrome on an iPad or iPhone.
Please if you have time do let me (us) know if something has changed regarding how iOS devices are displaying 1-1-1 clips tagged with 709-a and thank you again for your informative video!
Hello well time has passed since I made this one. I went to understand a simple fact. We just need to stop exporting rec709 gamma 2.4
It’s useless and source of many pain
Just export rec 709 rec709 from resolve and everything will be consistant. Rec 709 gamma 2.4 in dark room will have same perception than rec 709 in bright environment that’s the key
@@LELABODEJAY thank you for taking the time to respond! Just for clarity you would suggest exporting 709/709a with a color space transformation or not even? Just simply grade in 709 2.4 controlled environment and tag as 709a?
@@stephanalessi7720 grade rec 709 gamma 2.4 and export rec 709 rec709 don’t add A
Same for everyone
So your clients won’t have shift and everything will be much more easier
@@LELABODEJAY I want to believe but the difference between what I see with client in Resolve in the 2.4 environment and what they see when they look at the file on frame / vimeo etc is still too different. Do you have a website or could you send me your email address? I would love to book a 1 hour session with you to try and understand more!
@@LELABODEJAY I don't understand..my exported video after using gamma 2.4 / 2.2 and export in rec709 looks so bright and washed out, so how is it better?? I never had problems before with Rec709-A (it was accurate everywhere except on VLC). I also send my export videos to clients that have a pc and they send me sreenshots when I need to make some modifications, but the screensot looks also the same as my exported video. And when I watch TH-cam on my TV, the colors are also accurate. So I don't understand why suddenly we can't use Rec709-A?
Excellent lesson on this silly colour issue. Thank you, very well made.
Thanks sir
The Ultimate Solution helped so much thank you!
Thanks for posting! Awesome information
This is exactly what I've been looking for. Thank you, thank you!
Thanks 😊
Merci pour ces explications, c'est la première fois que je vois une vidéo aussi détaillée.
Malgré tout je suis encore un peu perdu.
J'ai un écran LCD sRGB et un écran macbook pro m2.
Mon objectif est que le rendu soit cohérent entre Resolve et TH-cam.
Je voudrais bosser sur un color management à base de node et et surtout une timeline en Davinci Wide Gamut.
Dans ce cas là faut il cocher dans les préférences "Use Mac display color profiles for viewers"?
Que dois je mettre dans mes réglages Project Setting/ Color Management?
Quelles seraient les réglages CST? (Ce que je faisais à l'origine sur par exemple du sony CST Slog3 -> DWG ---> Etalonnage ---> CST DWG -> Rec709 Gamma 2.4)
Que dois je choisir dans les préréglages d'affichage pour mon écran de macbook pro m2 car il propose (Apple XDR display p3, Apple display P3, BT709 BT1886, sRGB etc)?
Et pour finir quels seraient les bon tag à mettre à l'export final?
En espérant que ma question soit claire
Thank you very much for this work. I'm currently doing a similar experiment but when comparing the "ultimate fix" with the "Rec.709-A Fix" im not seeing an improvement. Did anything change since you uploaded the video or am i missing something? Thanks a lot!
From the comments and update for the video would be highly welcomed can’t wait to watch it
WOW Thanks for your great video - finally a professional answer!
J'arrive 2 ans après la publication de cette vidéo, donc déjà merci pour ce travail énorme !
Maintenant j'ai une question qui va peut-être paraitre un peu bête mais, sachant que de toute manière les clients finaux de nos vidéos (les spectateurs) regarderont notre travail sur 1001 écrans différents, sur 1001 navigateurs différents, sur 1001 players différents, pourquoi s'embêter avec tout ça ?
Quand je regarde un bluray sur ma télé et que je regarde ce même bluray sur mon ordi, ou projeté sur un mur, comment font les étalonneurs pour ne pas tomber en dépression à l'idée de toutes ces images qui ne correspondront pas au travail qu'ils ont fait ? 😅
On prend des anti-dépresseurs !
Vaste sujet en effet
On étalonne sur un écran de référence car c’est la référence on contrôle sur une tv haut de gamme mais après c’est advienne que pourra
@@LELABODEJAY du coup si sur ma tv "haut de gamme", mon export en rec709-a apparait identique à mon vidéo exportée, je vois pas pourquoi me prendre la tête 😅
@@JS_Leger ta tv s’en fout que tu sois en a gamma 2.4 ou quoi que ce soit. Elle ne lit pas les tag. Elle affichera avec son gamma natif SDR normalement proche 2.4
Very comprehensive, congratulations
Hi, great video! Quick question: what should I have my monitor calibrated to for grading under the ultimate fix? Cheers!
@@cnihon5894 hello if it’s a liked to computer and not video card then 2.2 I guess
Thank you so much, God bless🙏🏼❤️♾️
OMG THANK YOU! merci!!!! (j'ai reconnu l'accent hehe). The last fix is definitely the best! The difference is still there in VLC but not as much as every other methods. Par contre est-ce que je dois faire une conversion 2.4 vers 2.2; ou je peux simplement choisir gamma 2.2 dans les paramètres du projet (output color space : Rec.709 Gamma 2.2)?
I tried the solution for web and it’s got it looking closer to the original grade on my Mac. I think the issue is I use my PC for grading, perhaps I should switch to my Mac?
Thank you for this video, it is amazing!
@@Mr_G thanks 🙏
Merci beaucoup, des explications que je n ai jamais trouvé sur le net....très,très pro.un peu compliqué pour mon niveau, mais commence à éclairer ma lanterne (rec 709) . Bravo. Fred.
Amazing video, thank you. Question, I am using a M1 and have calibrated my display with a calibrite display plus HL - I'm assuming this profile stays in the system settings and isn't used within Resolve to help with color accuracy?
I color manage with nodes - so I convert the footage to DWG and Da Vinci Intermediate for grading, then (at timeline level) convert DWG Da Vinci intermediate to rec 709 (Color Space) and rec 709-A (Output Gamma) with a CST. In this case in the delivery page I would leave the tags as same as project right? Am I doing this correctly?
Tanks. It’s getting old should update it
Once again rec709A is a trick like my ultimate fix that only work for Apple.
This issue only occurred in Mac OS X not on iPhone iPad or in pc world
The easiest way is to stop exporting in rec 709 gamma 2.4 from grading as it clearly make hell on earth
What’s the equivalent of rec 709 a for premiere?
Excellent. 1. How do these answers translate to the new MacBook Pro computers? I have an M1 max. It has a BT.1886 profile. 2. How does this work with resolve color management at the project level, instead of the node level? There are a lot of advantages to using project level color management, vs node level management in my experience. I'm still having issues getting my decklink outputs to match my internal viewer and my exports. I can get there with node level management, but not project level management, like I want. Thanks for the AWESOME video man. Subscribed.
Someone correct me if I'm wrong. This response is to test if I understand it correct:
1. Your Mac display profile does not matter, as long as it is the correct profile for your display. Mac OS color management will handle the rest, but to view files correctly in the viewers in resolve you should set: Use mac Display Color Profiles for Viewers. 2. I think you sound set your resolve color managend settings to red709 gamma 2.2 for web export and than still change the tags on export to rec709 and rec709a.
The absurdity of this clusterfuck makes me think of browser "support" of CSS in 00s.
Thank you for this!
Extremely good video. Thank you!
I tested Rec 709 A solution and got a different result from yours, my source on Davinci looks identical to Quicktime, youtube and Vimeo on every browser on Mac, iOS is identical too. I couldn't test on windows and android because I don't have them.
So what did you use as settings to get this result?
@@ByRaphI created a serial node at the end of my color grade and transformed my timeline color space to rec709 gamma rec 709 A, then at the export setting I changed color space to match (rec 709 gamma rec709 A)
@@sondp Same
I did this Rec 709A solution and it looks identical in Quicktime and on Vimeo when I upload. However, if I open the same file in VLC, it looks more saturated / higher contrast. I'm wondering if someone viewing on a PC will see what I'm seeing on VLC, and if so, is there any solution to this?
@@skermani Either you don’t use VLC player at all or ask your client which media player they use and export what matches that player. Or you can export one file that gives you best results when uploaded to socials or web and let them know which player the need to use for playback.
Hello, thank you very much for the video. I am working with Arri Alexa Raw files. I am working on a MacBook Pro that is calibrated with a datacolor Spyder X to ITU-R Rec. BT.709. The suggested Gamma is 2,22. So ITU-R Rec. BT.709 should be Rec. 709 BT.1886. The suggested whitepoint in the datacolor app is x 0,313 y 0,329. My Macbook is calibrated to 120 cd/m2. So it is probably not quit BT. 1886 with a suggested 100 cd/m2. I know that my MacBook will never be color accurate as a Flanders monitor.
In DaVinci Resolve 17 I am working in ACEScct. Version 1.3. No Input Transform because I am working with Raw and ProRes files. Each clip is individually assigned an IDT. "Apply ACES reference gamut compression" is off. ACES Outputransform is set to Rec. 709. Netflix in one of their color management tutorials said that this Rec. 709 is BT.1886 so I assumed it is Gamma 2.22. th-cam.com/video/u9Rvm5xiuhk/w-d-xo.html
"Process node LUTs in..." is set to "ACEScc AP1 timeline space". I have checked the box "Use color space aware grading tools". Graphic white level is set to 100nits. In the "project settings", "Master settings", I have set the Video Monitoring Data levels from "Video" to "Full". Video with depth is set to 10Bit.
I did some tests with "Use Mac display color profiles for viewers". Normally it is and was deactivated. When deactivated, the highlights looked bright and colorful as they should. When I exported the video to VLC it looked the same as in DaVinci Resolve, as you said in the video. QuickTime, the NLC Tag was originally HD 1-1-1, it was too bright and washed out.
This was before I set up ACES. I used DaVinci Resolve YRGB before. As I understand the NLC tag 1-1-1 is right for Rec.709 BT.1886 Gamma 2.22 for web delivery like Vimeo and TH-cam. When I switched to ACES the ODT is set to Rec. 709 (BT.1886) as mentioned before. Now when I export (Use Mac display color profiles for viewers are still deactivated), and set the color space tag "same as project and set the Gamma Tag to "same as project" my export says when I press command+I that the NLC Tag is 1-2-1 (ITU R BT 709 - UNSPECIFIED - ITU R BT 709). Why is that, that it is now unspecified? In VLC the video looks like in DaVinci Resolve. In Quicktime it looks much better than the 1-1-1 washed out version. But the Image is a tiny bit too dark. The highlights mostly green are not as saturated in DaVinci resolve. The version with the NLC Tag 1-2-1 is washed out in Vimeo and TH-cam on Safari. So the transfer curve is not specified. On Firefox it also does look washed out. Firefox is color managed since recent versions.
I have now activated "Use Mac display color profiles for viewers". The image in DaVinci now looks the same as in Quicktime. But it does not change the fact that the transfer curve is not specified. "Automatically tag Rec. 709 Scene clips as Rec. 709-A" ist NOT Activated. I don't want to work in Rec.709-A. My movie will be mostly watched on TH-cam, maybe on Amazon Prime and in cinemas. I am primarily preparing the video for TH-cam. So I want the film to look kind of the same on all mobile devices. So all in all QuickTime should not matter that much. As a test I have changed the color space tag and Gamma Tag. "Color Space Tag": Rec. 709 "Gamma Tag": Gamma 2.2. I assumed because I use Rec. 709 BT.1886 that Gamma 2.2 might be correct. But in Quicktime the file looks way too bright. It is not that extreme as when I used DaVinci resolve YRGB, so ACES might have something to do with it. Because I chose a gamma tag of 2.2 the file says it has an NLC Tag 1-4-1 the transfer function of gamma 2.2 seems correct but not visually. As we can see in your list it is ITU-R BT.470M (Gamma 2.2). But this 1-4-1 is also wrong. If possible I don't want to cheat the NLC Tags and have a work around. Now I have exported it again as "Color Space Tag": Same as Project. Because I want to keep Rec.709. "Gamma Tag": Gamma 2.4. In Quicktime the file looks the same as in DaVinci Resolve as I am not mistaken. The NLC Tag is 1-2-1. I am wondering now. My ACES ODT is Rec. 709. I always assumed that it is BT.1886 with Gamma 2.2. Why does the image looks more or less correct to when I have not chosen a Gamma Tag? The video looks the same as "Gamma Tag": Same as Project. To my understanding this would mean that the ACEScct ODT Rec.709 is not actual Rec. 709 BT.1886 Gamma 2.2 but rather Rec. 709 Gamma 2.4. I have calibrated my Macbook to BT.1886 Gamma 2.2 so when I disable "Use Mac display color profiles for viewers" in DaVinci the Image does not use my calibrated display profile Rec.709 BT.1886 Gamma 2.2 but uses Gamma 2.4. Is this why my image looks better in DaVinci? It is only slightly brighter and the green is more saturated.
In DaVinci resolve I can't find the Gamma Tag (NLC 1) ITU R BT709. So that when I export the clip I receive a NLC 1-1-1. Originally before I started using ACEScct I didn't not use any color space tags or gamma tags and received a NLC 1-1-1 but the image was way to bright. (less to no contrast)
What should I do? Should I use Rec.709 Gamma 2.4 and recalibrate my Mac to Rec.709 Gamma 2.4 instead of using a calibrated display to 2.2? How to fix the washed out colors when uploading to TH-cam and Vimeo? (NLC 1-2-1)?
Is it called NLC or NCLC Tag?
Thank you very much.
I have once again exported it as process 422hq "Color space tag: same as project", Gamma Tag: Gamma 2.2 (same as my monitor is calibrated to). When I upload it to TH-cam it looks way too dark and oversaturated. The NLC Tag is 1-4-1. I have changed the color space tag from "same as Project" which should be (Rec. 709) to Color Space Tag "Rec. 709", Gamma tag is still 2.2. I get a NLC Tag 1-4-1. You have talked about Full and Video Range. When I export a Proress 422hq in full Range (Tags Rec.709; Gamma Tag 2.2) it is saturated and looks like the davinci viewer when "Use Mac display color profiles for viewers" is disabled. So in general it is brighter. When I use video range. In Quicktime it is washed out (NLC-1-4-1) but on TH-cam it kind of looks okay (close enough). In full range with no gamma Tag (same as Project) it is still to dark in QuickTime (NLC 1-2-1) and does not look like the davinci viewer. I have exported 31 versions and still haven't found a solution. Is there an option that TH-cam does look the same as in the resolve viewer? Is it right that NLC 1-4-1 is best for TH-cam? Does full Range automatically use Gamma 2.4 (NLC 1-unspecified-1) (Nlc 1-2-1) even if the project gamma is 2.2 Bt.1886 and Tag is set to use project setting? Or is it because aces gamut ODT is Rec. 709 Gamma 2.4 and not Rec. 709 Bt.1886 Gamma 2.2?
Sorry for getting back so late
But honestly so much informations so much to read. Sorry I have to see what’s happening. This description is way too long
Contact me via email will see if we can arrange a team viewer
Lelabodejay@gmail.com
@@worldproductionentertainment i'm having the EXACT problem that you are having and the only "solution" that I could get was to set the output color space of the whole project to rec 709-a, that was getting as close as possible to quick time as possible, as well as on my iphone and youtube. But as the video said, that's only a part of the problem because we are on Apple's ecosystem, so it will look great on their products, but i dont have windows devices to test to see how they would look on comparison.
In the end of the day, from what i've searched and tested, was to either find a middle ground, like the suggestion in the video and live with it, or do different exports for the different platforms, which kinda sucks.
Also, when opening on different players, like IINA and VLC, the colors are different as well, so i don't think there is a solution for all of this that will work 100% all of the time on all platforms but if it has, i would like to know as well because it's kinda infuriating :(
Also, grabbing still from the colors page changes the image a lot, making it brighter and with less contrast
Amazing video, thank you! I'm exporting an .MXF for TV from Resolve with Gamma 2.4. The tag of the exportet file is 1-1-1 for some reason. When I open the file in the MetaData Editor of AMCDX Video Patcher, the Frame MetaData is greyed out and I can't change the tag. Any idea what my problem could be?
@@rusch_meyer it happened to me sometimes with latest version I don’t know why. But remember tag is only for Mac OS your tv don’t care it will be displayed with the gamma of your tv which should be around 2.4
An excellent video! Thanks you for the detailed explanations. One question, I’m using DaVinci Resolve on Windows. If I use node based colour management and use an output CST to Rec.709 2.4 at the timeline level when grading, should I still add a CST after to convert the output from gamma 2.4 to 2.2 and export using gamma tag Rec.709-A or can I simply change the output CST from Rec.709 2.4 to Rec.709 2.2 and use Gamma tag Rec.709-A on “Delivery”? Is there a difference?
Did you solve this? Doing this now and I don't see the reason to add an extra node to covert to 2.4 when the timeline is already 2.4 and you convert to 2.2.
@@JamesWilliams-in3py No, I never received a response.
My timeline is in DWG Intermediate, so I use the last node to transform to rec709 2.4 when grading.
I could be wrong, but I think the Rec.709-A is generally the solution for mac users. As for windows I would edit as normal. For web, my project and timeline would be Rec.709 Gamma 2.2 and export with those same settings. The point of using gamma 2.4 is if you know its for broadcast.
My take away is be intentional with what you're startup settings are and if you get caught in gamma 2.4 but need to export for web, add a node to timeline and convert to 2.2 (or 709-A for apple devices)
Toujours aussi complet. Est-il prévu de voir le fonctionnement du Color management avec Media composer ? Je sais qu'ils ont amélioré leur Color management. Encore bravo.
hello desole je ne bosse pas sur media composer
the amcdx video patcher doesn't allow me to edit... :[
@@WalkingHeadPro sometimes it happens I don’t know why
Don’t work with mp4 and sometimes doesn’t work on ProRes 🤷🏻♂️
what's the name of your color picker?
Thank you for this amazing video. Truly helped me so much. Do you know if anything has gotten better? Or are we still tagging Rec-709A for now?
Any tips on exporting stills? My video files look great now, but my stills are super washed out.
Just change the tagg in Apple preview app as I show in video
@lelabodejay I find that this works great using a Windows Machine - all looks really good - my only problem is playing back on QT on Windows (not tested on a mac yet) - the colours are still washed out. Any possible solution for this? Matches really well when uploaded to youtube though - QT player is the only outlier at the mo. Thanks for all your hard work!
QUESTION: For BlackMagic pocket 6K (braw), which color space would be the best (and will it need a timeline gamma conversion), with best I mean the least problematic result for web purposes across different platforms and devices?
Hello. Grade with rec709 gamma 2.4 on reference monitor. Then do a cst to rec709 gamma 2.2 and tag rec709 rec709
That’s the least problematic
@@LELABODEJAY most of the time all clips are Braw, so would you suggest to leave the camera settings in the color page on timeline instead of selecting clip?
@@adeeboberoi5362 you can decode in rec 709 or leave as log and grade it
@@LELABODEJAY Thanks for the informative video on the subject.
For the web, and to avoid the conversations from 2.4 to 1.9, it would be easier to set a calibrated external monitor to gamma 1.9, and tag Rec.709-A in Resolve Timeline and the Delivery page. This is using, Use Mac Display Color Profiles for Viewer. Both Resolve Viewer and the external monitor would be closer than 2.4 after the conversion.
This is what all MacOs player and browsers supporting color management are expecting to see anyway, (1-1-1) = gamma 1.9.
@@LELABODEJAY What about the new Rec709 (Scene) timeline color space in Resolve 17? It allows for 1-1-1 tags, but the rendered files still don't match. And also, in addition to Rec709-A as an output gamma tag, there's now Rec709 and Rec709 (Scene). How do those influence the way we can mix and match gamma tags and CSTs/timeline color spaces to achieve greater consistency?
Hey, im working on a project for a client.
On vimeo everything looks washed out
My grade in davinci is
Blackmagic Wide Gamut, Blackmagic Film Gen 5 to
Davinci Wide Gamut, Davinci Intermediate to
Rec 709, Cineon Film Log and
with a 2383 Film Lut as the last node back to Rec 709 2.4
I export the graded file as DNxHR 444 12bit video levels Rec 709 2.4 and import it in After Effects. After Effects expects DNxHR 444 to be full levels and converts it to video levels. Its the legal vs full issue with DNxHR 444 in adobe software. I did think this was the issue. Would exporting the file at full levels for the import in after effects fix it or not.
The grade is finished, so full levels for after effects and web export doesnt need to be. It would be just to fix after effects from washing out the dnxhr 444 file expecting the dnxhr 444 file rendered at video levels to be full range and converting it to video levels.
Should i change to Rec 709 2.2 with CTS and Rec 709-A in export settings in davinci before exporting the grade or before rendering the delivery in after effects
Hello
Hard to tell from distance
Why are you in full
Video should be levels
If you grade on monitor video it should be levels
Tv and web expect levels
After effects ignore gamma 2.4 tag
It will export 1.1.1 rec 709
May be it’s the first step to watch at
If your content will only be on web yes you should try the fix but remember it’s a fix and not accurate
Rec 709 gamma 2.4 is for tv in darkness
Rec 709 tag 1.1.1 is for monitor on bright environment. Perception should be the same….
Should !
Such a mess
Tell me if that helps
@@LELABODEJAY Thank you for the quick answer!
I need to say im on windows.
I would export as Full Levels DNxHR 444 12bit Rec 709 2.2 Rec 709-A in resolve for after effects.
In after effects i would then add the levels effect-computer to video profile to convert full range back to video levels.
After effects wouldnt convert video levels to video levels as so the footage wouldnt be washed out and because of the levels effect, the export from after effects for delivery will be in video levels again.
What should my timeline colorspace, gamma and my export in after effects be, Rec 709 2.2 for time line and Rec 709-A for export right
@@arceus2015 rec 709 A is only for Mac. No need on windows
Why are you going to after effect ?
Full level should not be a problem
Just export in auto from resolve
In windows you don’t have tag problems so if your colors are washed it’s not a tag problem
Unless you are in aces or in dolby workflow or with 3D exr elements you should not work in full
No need and can cause issues that you are currently experiencing
@@LELABODEJAY In resolve on windows, should i still make a node and convert my timeline with a CTS from Rec 709 2.4 to 2.2 and export tag as rec 709 2.2 for vimeo, instagram or youtube meaning web upload?
After Effects has this issue with the DNxHR 444 Codec because it expects dnxhr 444 to be full range, when i put it on Auto in resolve and it auto exports in video levels, when i put it into after effects, the colors will be washed out. In forums they say export in full range for after effects to interper the codec correctly
Im rendering in Apple ProRes 4444 XQ in After Effects for the best quality for archive file and Apple ProRes HQ for web.
@@arceus2015 you should export rec 709 rec709 tag won’t matter on windows
Test several method to see what’s the less corruptive
My video is 2 years old some things has evolved
Really hard to tell without clip and testing and seeing
this video solved everthing haha, thank you
Thanks
Glad if that helped you
So for TV delivery, the correct tag is 1.2.1?
Tag only concern max viewing
Just try to not modify what you received from grading and what your client has reviewed
Tv will ignore tagg anyway
you say 'don't touch the grade', but then you apply CSTs.
do you mean you grade and THEN apply the CST, just for the export-for-web?
for example your ultimate fix: do you grade on your rec.709 gamma 2.4 timeline and monitor (ignoring your sRGB screen), then apply the CST only for the web export?
and you only set your UI monitor to sRGB for doing your testing right? at least on my system, Display P3 looks much closer to the video monitor, and sRGB will confuse my clients with darker shadows and more saturation.
Hello
Sorry for late reply
Yes you do your grade on reference monitor
Then apply cst for web delivery
I don’t really trust my ui
For your information if your monitor is Apple P3, it’s a P3 gamut combined with srgb gamma.
Hi I also found that 709A. works well in all applications. if you have a broadcast project like a tv show, would you still use 709A or change the tag to 2.4?
Great & interesting video, Good job !
🙏
Thank you a lot for this video, really clarified some things! Quick question: What would happen if you export as follows: Rec709/Gamma 2.2 and then tag it as 1.4.1 (Rec709/Gamma 2.2/Rec709)? Would that tag get misread / re-tagged by some of those apps as well?
It will be ignored as everyone ignore tags except Mac color managed applications
The only way to avoid this shit is a lot more simplier that what I have explained in this video
Stop tagging rec 709 gamma 2.4
So we will never have shifts
Color sync can be so tricky and leads to error
The ultimate fix is a trick
Like every trick it generates problems.
I should redo a video about it
This one is old
@@LELABODEJAY Thanks for the reply. Would be awesome to get an update video... even if it is just a short guide without elaborate explanation. I've found your video when I re-searched a color space problem I had. I also posted on a bunch of message boards about it and got referred to your video at least on 3 separate instances.... Your explanation is by far the best and most thorough one I found.
@@LELABODEJAY That would be highly appreciated! Subscribe
thanks for your video.
Salut Jay, merci encore pour la vidéo! J'aurais besoin de ton aide si tu as 2 min: Je suis sur Resolve Windows et mon moniteur est un Flanders CM250 calibré. Si j'exporte un QT 1-2-1 de Resolve et que je le visionne ensuite sur mon Mac, c'est beaucoup plus sombre que sur mon Flanders. Est-ce signe que ma calibration n'est pas bonne? Par contre, si j'utilise une node CST Gamma 2.2 et que j'exporte en 1-1-1 (avec le Rec709a), c'est plus proche. Salutations du Québec!
Hello Mp sur Facebook envoyé
What you say about converting to 2.2 for the internet is incorrect. Ironically you get things right early on when you say that grading in a 2.4 environment on a 2.4 display will look the same as being in a 2.2 environment on a 2.2 display. This means that you should do NO conversion for the internet at all, just tag as 1-1-1 for client review if required. Otherwise you will make things too dark by converting to 2.2! (On the other hand, if you are also converting to 2-2 for client review, you'll end up getting things right but in a backwards way because they can tell you to brighten things up). Another way to think of it: grade on a 2.4 display in a dark (i.e. 2.4) environment. Turn the lights on, thus creating a 2.2 environment - the grade will now look too dark. Switch your monitor to 2.2, brightening it - now it will look correct again. So if you convert to 2.2, it will now look too dark despite changing the environment and the display. You can have your color space transforms convert to 2.2 if you grade on a 2.2 display in a 2.2 environment, but do it at the start of the grade and don't switch later (don't switch even if you have the wrong conversion to start with as your grading decisions would have accounted for this already).
....u sure?
"Turn the lights on, thus creating a 2.2 environment - " What do you mean? The actual lights in the room?
MERCI !
Are there any updates in software or Mac OS that have fixed this nonsense yet? Would love an update from Apple or Resolve.
Does this mean when sending files for these 3 web deliveries you have to change the timeline output color space and adjust the footage to look like your final grade across all three output tags? Pls let me know if am correct
Hi just export in rec709 for web and do the Cst to convert gamma 2.4 to 2.2
Don’t touch the grade
Remember that you have gamma 2.4 for monitor in dark environment and gamma 2.2 was set for monitor in office in daylight
You can’t have one file thad works everywhere 🤷♀️
@@LELABODEJAY you mean I should do a cst conversion to my final grade at the end? Is that correct?
My only screen is a macbook with p3 display
I don’t mind having different files for all deliveries.
@@sondp you should not trust your MacBook for delivery
Get a video card and a monitor for grading. It will avoid any critical issues
@@LELABODEJAY One last thing that confuses me when using srgb monitor is the workflow because if my
monitor is srgb and DaVinci source timeline is rec 709 2.4 and with my understanding the use mac display option is supposed to be turned off here. So how do I make sure that what am seeing on resolve on the srgb monitor will be the same on the web, do I still need to do the Cst to convert gamma 2.4 to 2.2 or how?
so great!! thanks
so helpful
thanks!
You re welcome 🤗
Merci beaucoup!
It's 2024 but still...thank you!
THANK U!
You re welcome
Merci bg
Thanks a lot for this explanation, CST output gamma 2.2 and export advanced setting output gamma to Rec.709-A worked to produce the video with correct colors when viewed on Apple Studio Display with P3 gamut
What did you select in your project settings at "Output color space"?
Hardcore
Great
Brilliant video, thank you. One question: Do you have that girl's number?
🤣 that’s the best question I ever had on that video
Unfortunately no 😉
🤣@@LELABODEJAY
"go to resolve and do a color transform". Ok, it may be obvious to you but I did't find that menu anywhere. You lost me after that
It’s showed in the video go to color page add serial node and add color space transform
@@LELABODEJAY Thanks for your reply. I think there must be a Davinci version difference. I see it in a very different way. In my project ALL clips already have a color node (I can't identify which type) because the "settings" tab next to "library" is greyed out. Also, if I add an explicit "color space transform" there's no "gamma 2.2" or "gamma 2.4" options on the Input Gamma and Output Gamma menus. My version of davincy (version 18) looks very different than your video from 2 years ago: I suspect many things have changed. Also, how do you tell that a color space transform should apply to the entire timeline, rather than just one single clip?
@@lsfornells yes lots of things have changed. You can apply effect to all timeline on the top of the window on the right you have two dots click on first you apply to clip on second you apply to the timeline
@@LELABODEJAY Ok, thanks for the tip. Merci beaucoup
Absolutely mind blowing that we can ONLY grade for either Apple Devices or Non-Apple devices. Apple, GET THIS FIXED ALREADY.
You did not understand the video if you think this is apple's fault.
@@mpeg2000 Is that so? Why don't you go ahead and enlighten me then.
@@issawrapppp what do you want me to tell you when he explained it perfectly fine in the video?!? 🤷♂️