Mix and match Log3G10 LUTs in Resolve (Primers, philmColor, IPP2, etc)

[Marc Wielage]

2 years ago, I switched from video to ACES (ACEScct) and don't look back (with camera footage). On IPP2 based Red colour science I found that using as IDT Alexa LogC gives me nicer colour to start with. In everything else I totally follow Mark's elaborations on LUTs.

I haven't embraced ACES as much as some do, but I know that some swear by it, and some studios (particularly Netflix) insist on it. In a situation where we have to deliver a show in both Rec709 and for HDR, that would probably be a preferable way to go. I feel like the LogC IDT is a different experience, but it's not always the best experience with Red for me.

But at the same time: I always try to tell other colorists, "you know, if it works for your project, you like the pictures, the client likes the pictures, and the check clears, then there is no problem." There's probably nine different ways to do the work effectively (and maybe another 100 ways to fail). If one of those methods uses a LUT or an IDT or whatever you have out there, and it works, god bless you. ACES is a bit of a moving target, and when ACES 2.0 comes out later on this year, we could well have a whole new set of options (and potential challenges). I mainly care about productivity and how quickly I can come up with a look the client likes; I'm a little thrown by how the knobs respond in ACES, and there are other methods, like using Color Space Transforms within the node tree in Resolve, which use elements of ACES but not the entire signal path. So there are a lot of choices out there that don't make it an easy decision.

Note also that there are big shows out there using EXR (including many of this year's blockbusters), particularly those heavy on VFX, and you basically have to throw traditional LUTs out the window in that kind of environment. Once the EXR is normalized for a given color space, then there are things that can be done in color science to bend the image in a predictable way. For me, PowerGrades are a better idea than LUTs, and they help save a little time and instantly give the client some quick looks to give them an idea on range. There are equivalents for Lustre and Baselight and all the other systems out there, but where LUTs are helpful is just to get a quick look on the set to know, "yes, this is going to work for post." For that, I'm all for a temp LUT just to give the filmmakers confidence that the DP knows what they're doing.

 

[Hrvoje Simic]

So IPP2 see the skin as slightly greenish in that light, with that lens and white balance.

You are not seeing only IPP2 here but:

-Sensor properties
-Base CS properties
-Image transformation properties

IPP2 always does the same, regardless of the light, lens and white balance.

Primers for RED work on IPP2 CS basis.

What if she had a too warm tone to start with or to blue? you clamp all skin color towards that color?

NO clamping.

What is correct is kept correct, what isn't and can be corrected - gets corrected.

You are attempting to conceptualize this through a concept of a grade.
Again > This is NOT a grade.

Grade is shot-dependent. Light, lens, scene bias, ND tint, white balance.

This is signal re-structuring and normalization.
It is not image oriented. It is signal oriented. Images are a result. This is shot-independent.


With other scenarios, other optimizations emerge.



In this shot 4 are visible. There are many others.

Helium, Erich Roland

 

[rand thompson]

Hrvoje,


No Ill will was meant from my post and I do apologize for getting your name wrong, I used a Tablet which can autocorrect spellings of words sometimes and I fail to catch it. My post was meant to show the benefits of your Primers not to troll you.

 

[Hrvoje Simic]

When the image signal is stabilized by Primers...all kinds of cool stuff happens.



Gemini, Luigi

And oh yeah....happens with other cameras too.