better way to apply LUT (pro-colorist approach)
Вставка
- Опубліковано 4 сер 2022
- This week we will guide you on how to apply a LUT appropriately using DaVinci Resolve so that you don’t lose your artistic expression for technical accuracy. If you like this video, make sure to subscribe! Review our tutorials and share how we can improve in the comments down below!
► Notice! Colorist Factory is now Colorist Foundry.
Checkout our math-true emulations:
• coloristfoundry
Film Emulation Plugin (2024)
• coloristfoundry.com/filmverse...
Film Emulation LUT Kit (Camera & other NLEs) (2024)
• coloristfoundry.com/filmcurve...
Instagram (Important)
• / coloristfoundry
What is Colorist Foundry?
• A constant exploration of what's beyond cinematic look made us form Colorist Foundry. UA-cam is our way to simplify our learnings for filmmakers and storytellers.
This is a great video
what I want to understand is do we need to change our settings to tetrahedral before we add out lut or do we need to invert?
You don’t have to invert the image. That’s only to make sure that what we are trying to show is clearly visible after UA-cam compression.
Apart from that, you’ll have to change the interpolation method before applying LUTs in your project.
Another great video! Super important when creating/using LUT's.
I’m learning a lot from this channel 🙏🏽
I love this kind of videos, thanks so much
espectacular! gracias
thanks alot for this
Thanks!
Great video
I have only one question - why isn't the tetrahedral the default?
Exactly my q
probably performance but honestly the difference cant be that drastic
Probably because it's kind of subjective. If you apply a neutral LUT that is meant to produce the same output as its input, tetrahedral interpolation will instead shift colors.
So it's perceptually better, but not exactly mathematically correct (or desired).
Awesome video! Can you make another version for Premiere?
Watching this without audio and seeing all the crazy color stuff happening with no context made me think this was a shitpost.
But I learned things when I turned on the audio!
hey hey hey the intro is blaaast. Can you tell me where to find it
Also thanks for the video, it was really infomative
I bought all the packs and power grades ... game-changing ... Can you please tell us where can we get the music track used in the video? Thanks
So what's the catch? Why is tetrahedral interpolation not the default or even exclusive method? What's the downside of using tetrahedral vs trilinear interpolation?
This.
what if you want to bring down the percent down?
sorry for newbie asking, but do you really need slog video for applying luts/color grading best? or i can apply luts in my raw smartphone videos too? i wonder the results
So is there any reason it would be better to choose trilinear for a grade?
trilinear uses less cpu power. for slower computers
@@user-ll9ln1uu4b thanks
i prefer LUT's that slightly change the atmosphere and put a slight artistic change to the environment, but too much change to natural colors, i don't like very much. maybe i would want to be a filmmaker to show what i mean, the perfect splash of artficial color grading, that I don't find excessive.
so never use trilinear? :P
False information here in the Video about LUTs. A cube size of 17 doesn't mean "less information". The transitions between ONE color transform to another es even BETTER or smoother than using a cube size of 33 of 65! You also say that it is "A Bigger challange for the software" which is wrong. It is easyier for the software to work with a smaller cube size. And it is not "less accurate". It is just different. I agree that Tetrahedral is better. But if you got COLOR banding, you should not use the LUT at all. And it is better to "stress test" a LUT correctly.
Allen do you have a minute to tell me some more info about what you mentioned on the differences between 17 to 65 cubes? Which is better to use if you need better accuracy and resolution and why? Many thanks in advance!
@@vfxd8372 It is just a different size. A cube can have a size of 3, 10, 30, 100. YOu can google it and find your answer. But one size is NOT better than another size. People saying bigger cube size is more accurate but that is not true. A cube size 17 is also accurate. And if you have 65 the "LUT designer" must be a professional. Because it is easier to make mistakes. A 17 cube can be "accurate", too. "Better" and "more accurate" are the wrong terms.
@@allenpayne9182 Thank you for your time, Allen. I understand what you're saying. We're talking about a different approach when choosing one of them. I'll google it and take a better look at the usage of each one of them. 🙏
The video doesn't say that a 17-Point cube is better than a 33-Point cube. I think you are working on too many wrong assumptions at this point. Here’s the excerpt from the script where we speak about 17-point cube LUT: “Why 17-point? It’s the least information LUT that Davinci can create. This means that when we apply this LUT back it’ll be a bigger challenge for the software to re-create our adjustments accurately using just the 17-Point LUT.”
I think you are missing the entire point. It's an experiment designed to showcase how different interpolations work on a 2D image. It's very successful I must say, given that our target audience is the people who are creators with no intentions of becoming colorists. I say this because it’s not easy to showcase the effects of even a “bad” LUT on a 2D image. On a 3D cube, it's simple. But on a 2D image, it’s a matter of trial and error. That forces you to design a pipeline with certain assumptions and certain extreme starting points. And if anyone wants to take a deep dive into it, our blog links more such resources to go in-depth color science.
It's not a good practice to come into a room full of newbies and try to be a know-it-all. You have to acknowledge the learning curve of someone who is a complete outsider to all these concepts. You cannot walk in here and write a careless comment with false assumptions and half-baked knowledge.
@@ColoristFoundry
You say:
"t’ll be a bigger challenge for the software to re-create our adjustments accurately using just the 17-Point LUT"
This statement is wrong. Whyt d do you mean with "bigger challenge". If the LUT creator chooses 17 as size it is totaly fine.
By wording it the way you say it in the video it seems that the "creator of the LUT" designed it for 33 or 65? But if you design it that way you should export it as 33 or 65.
For the software itself is HARDER to interpolate a larger cube size. So "less challenging" for the software. Not "bigger"
Sorry, but when it comes to LUTs I don't like simplifications. That is why people writing comments like: "what is better". But one size is not better or worse than other sizes. And "bigger challenge" statement is wrong.