My Color Grading Workflow - DaVinci Resolve 18 Tutorial 

teal n orange

Agreed! Appealing

More color grading please🙌🏼

Thanks a lot man. Means a lot!

Glad you like them!

Thanks a lot mate!

Thanks! Glad it helped you!

Thanks a lot man!

Thanks a lot Elliot!🙏🏼🙏🏼

Thanks man! Glad this helped you

Glad you enjoyed it

Onto that in the coming weeks/month!

Thanks mate!

De nada!

Thanks a lot!🙏🏼

My pleasure!

Cheers for this mate!

I would over expose a bit perhaps! Worth trying out!

Cheers buddy

Cheers!

Will come mate! Time will tell :-)

Thanks a lot! 1: Yes, generally you want to balance the image in LOG, before converting it into another colorspace. 2: Pocket! I read that the new version can work in ProRes RAW. Not sure tho!

Thanks dad!

Thanks a lot dude. Glad to see you here!

Glad you said the last sentence; if it works it works. Haha. You're right in some way, but I've seen amazing grades done in ways I would never even consider doing it. I believe the end result counts, and the way to achieve it is less important. Of course, a basic knowlegde is important!

@@SjoerdWess Absolutely, it was just worth mentioning, as if you were to be grading a very color-taxing scene, your approach would've significantly limited the possible outcomes as the rec.709 container is times smaller than any arri wide gamut or blackmagic wide gamut. That said, what matters is the end result. Your videos look fabulous and it's obvious you put time into planning all of it and in combination with a great edit - you get a winner! Keep up the good work, brother. All my rambling is to hopefully help you or anyone else that sees it haha.

No tripod. Mainly handheld!

Thanks Scott! It was only me and my AC for camera. And yep, I was responsible for color and camera. There was a small crew for direction/art!

It can definitely!

Hi Hanno! When recording this video, I was convinced that the CST node effect was working destructively, meaning it wouldn't matter wether placing the correction nodes in front or after. As I've gotten loads of feedback, I tried testing the 'original' way, and yet, no difference. Maybe it's me, but I think you can use either way, so: correction (exp, wb, cont, sat, CST - or reversed). Yes, FPE's require CINEON log, Log-C comes very close to Cineon LOG and I personally like it better. It has given me great results, so it's kind of in my workflow. Rules are there to break, or at least see if they matter to you ;-).

Thanks for sharing this video. I am always looking to improve my workflow. I actually do the same and start with a CST to rec709 because the primary wheels work as expected with rec709 as an input. You can correct your footage and then CST back into your favourite flavour of LOG since the CST is non destructive unlike a LUT. It's a really nice workflow because you can get very good result fast. One thing tho @Sjoerd I learned recently is that if you have a proper whitebalance the skintones should be spot on already and it's unnessecary to use qualifier. My mentor told me he only uses qualifier as a last recourse. It's easy to destroy the image or make it look unnatural with qualifiers. In my last 5 projects I graded I used zero qualifiers & his advice really changed the way I approach new grades and above all it improved my grades. Yew

Cheers buddy! Thanks for letting me know. Thanks for this insight mate, appreciate it a lot!

I too have the same doubt. Putting a Rec 709 conversion as the first note make the color space smaller and remaining grading will have less space to work with resulting banding or chroma noise more than working in a wider space like DR intermediate space and then converting it to rec 709 at the last..

Its an interesting debate. I did switch things around myself now. Changing my workflow. But still, i believe after doing loads of tests that it doesn’t matter. But hey, who am I? I care about the end result. All the nitty gritty is too technical to me! Haha

You are so right man. Thanks for letting me know! I will keep this in mind next time I make a video like this. Appreciate it mate!

Cheers dude

Ive never done this tbh. So I cant give out advise on this! Sorry man!

@@SjoerdWess I was under the impression that video footage needs to be color corrected with a color checker, before color grading is applied. Lol now I’m confused!

Ive learned it this way because the node selection could change if you change colors on the same node! High low range is just a way to isolate colors in a specific area.

@@SjoerdWess ahhhh. this explains it alottttt better. Thank you.

Thanks for the feedback, ill make sure to explain such things a but more in depth

@@SjoerdWess you’re fine brotha. I understand that you werent speaking your native language. I appreciate you clearing up the understanding for me as well. Thank you.

CST is non-destructive. So it doesn't matter. The data is kept. My exposure is adjusted before the CTS, which, in my personal opinion, worked best for me. Check out Juan Malera, great colorist going into more detail about this.

@@SjoerdWess I think Diego is right man.. I don't know if you are referring to this video: "Insider Knowledge - An easier way to grade log footage" from Juan Malera. In that video he converts to other color spaces and works in between them, this is different than your case I think. Now your only profiting with your exposure node from the log space and your other nodes not. I could be wrong, but this is what I learned. So just wanted to share the possibility of this workflow.

You’re right to a degree. A CST is different than applying a LUT as a conversion. A cst node works non destructive, so no data is lost after the conversion. It just works differently, but in my opinion that aint a bad thing

@@SjoerdWess Ahh okay, Then I learned something new today, thanks!

I might be wrong as well hey. Just my ‘logical’ view on it haha

I personally never grade in LOG. I've done this, but results were way better for me when grading in a rec709 basis. Also, as mentioned in someone elses comment, I used the same nodes for skintones to isolate particular colors in the frames where no skintones were present. Shouldve mentioned this :-p. Using Filmconvert is a bit different than using traditional FPE's. But how I usually work, is from a Arri LOG-C profile, which comes close enough to CINEON and gives amazing results. I know things are done differently than they 'should'. That's the whole thing, there is no right or wrong in any creative artform. The output matters, right? Thanks for checking the video dude! Appreciate it.

@@SjoerdWess Well, there's no law to prevent you from doing the way you want of course but there is a reason why colorists work in log space and why fpe's are created for cineon log. Guys like Kelly or Darren Mostyn can explain why this is so far better than I can. And once you know the technical boundaries of the tools you're using there's plenty of room to be creative.

I can totally get what you're saying. I will do research on this matter. Everyone always learns and gets better. So do I. Thanks for the headsup dude!

Ah for real! Thats interesting. Would you mind sharing your opinion? Im curious

@@SjoerdWess Well for Starters, the Film Stock from Dehancer is just way more plentiful than FC. You have a lot more to choose from. you have more options for Film Print Emulation over FC. But that is not where it ends. You have better options for push pull EV, you can change the color heads as if you were printing actual film stock. Film Grain is a little more customizable. but the option for defringing, film breath, gateweave film bloom(completely different from bloom plug in from DR). The halation effect adds a more authentic touch. And its nice that it has false color when you are exposing in the plug in. Another thing that annoyed me about CF is I would have to add the vignette in another node. not anymore. Dehancer has that. Also, if you are big time. you can actually output this stuff out to film negatives to be printed authentically. There is a reason why dehancer is worth 400 and FC is work like a little over 100. its just superior. if you ever have the extra cash, go get dehancer.

Destructive to an image or not, if it looks good my goal is achieved. I dont follow the books or do what everyone elses does and thats something I try to bring across. There is a ‘perfect’ way of doing something for everything but greatness is achieved once someone dares to break free. See legendary cinematographers, directors etc. Every node you make is destructing the image in some way, and thats totally cool.

@@SjoerdWess Breaking free sure but every node is destructing in some kind of way? No. I mean that’s saying every node is purposefully causing the image to become worse and there is no fix. I suppose there is no wrong or right.

Thanks man

Thanks Lina! Glad it was helpful! And double thanks for that compliment, ha!🙏🏼

I havent had any troubles tbh!

Always. I build my entire grade these days before CST. Building the grade in LOG grants for more flexibility.

@@SjoerdWess Please have a chat with some Color Scientist or at least an exp colorist to understand the difference. There is a reason for everyone using IDT and ODTs in their usual places.

@JustPassingByBaby oh I did. With BMD engineers, colorists, all of them.

@@SjoerdWess And they told you to put the IDT at the back?

Thanks for the feedback

Well, I generally used it since a lot of the tones where present and this way I could control them regardless of skintones or not.. but I respect you.

@@SjoerdWess Dont you get problems with artifacts of the matte when keying the skin like this?

Nope, you need to play with the blur, black/white radius etc to ease that off! Works wonders

This is a legit reply. Teaching this as a technique, to new users is especially steering them in the wrong direction. Almost all modern cameras actually record skin tones accurately. What does happen is that the temperature or ND filters or lighting controls the image and it appears that the skin is misrepresented. But the skin tone is in most cases, is the subject of the frame, and is the most recognizable to anyone as being "off". So in the case of lighting or white balance or whatever is making the skin off, it is usually present in the whole frame, but is not as apparent as it is on the skin. If the balance leaned toward green- the face would show it right away, but not in the background of cooler color hues. Deal with the balance and get the skin correct - there is no need to qualify skin as a goto procedure. Of course the better the photography the easier this is to do. Using a skin qualification means that every frame of your video needs to be scrutinized. Which of course is possible but in every project I have ever worked on, there is no budget to do this and the end result is no better, and in a lot of cases creates more work and problems. The qualifier will react to different trim changes and in the case where someone steps in or out of the light - there will be chatter. This chatter to the end user will look like noise. So I agree with Allen here - Walter Volpatto is one of the world's best Colorist and if you go take his masterclass with TAC, there is no time where he qualifies skin as a procedural step in grading or color correction in his teaching. So why would a beginner, learn this before learning to color correct in the most logical way? Pick the subject and grade for the subject. If it is a person - skin first. If it is a mountain - grade the mountain. The end viewer should be looking at the subject and it is the most important asset to be correct and everything else is secondary. The tutorial here is literally the opposite.

Thanks for the comment, Jim. Appreciate you taking the time to write all this! The title says ‘my workflow’. I get that people get upset because its not ‘by the books’. But this worked for me for this job, and so it might do for others. Its here to get people going and go from there. The end result is what matters and if thats acceptable, than thats good. Dont take all this too serious. Its free advice, and there is enough in there for people to learn at least something.