-
FCP X – steady as she goes.
-
[Bill Davis] “That process could care less whether the original data represents RAW, Rec. 709 or SD footage from a security camera.
A LUT is a LUT is a LUT.”That is just so incorrect and untrue. And yes, the process absolutely cares what the format is. No offense meant, but it seems like maybe you aren’t clear on what raw is. So with my apologies, here’s lengthy explanation.
Camera raw data (not capitalized, BTW), like .cr2 or .r3d is monochrome, raw Bayer-pattern (usually) sensor data, that cannot be seen without decoding. A raw decoder module – not a LUT – is used to do that. Like Adobe Camera Raw, Redcine-X or a RED plug-in. These do not use a LUT to generate viewable video.
Most cameras do not expose the raw values to post production. Many do encode viewable video in log gamma space. This means that a logarithmic curve is applied internally as part of the raw-to-video encoding, that brings the shadow and highlight detail within range without clipping. However, that video is viewable in Rec709 color space, but with a flat appearance.
There are two types of LUTs – and they often use the same file format – technical LUTs and aesthetic LUTs (“looks”). Technical LUTs are the mathematical inverse of the log curve used when the camera image is encoded to log space. This “expands” the flat image back to something else. Usually that’s Rec709 color space, but not always.
These LUTs are mathematically correct, but don’t necessarily give you the desired look, especially if you want more brightness, contrast, saturation, etc. So you’ll often grade on top of the LUT’s conversion. IOW, the LUT gets you 75% of the way. In addition, you can generally get the same look without a LUT, though it might not be mathematically accurate.
The other types of LUTs are subjective, creative, custom “looks” that aren’t mathematically correct. Film stock emulation LUTs are one example. These are usually someone’s creative interpretation of what a certain type of Kodak or Fuji negative or print looks like when transferred to video (its own subjective process).
In my experience most of the technical LUTs yield less than desirable results. ARRI’s LogC-to-Rec709 LUT when applied to Alexa and Amira cameras is usually OK. Sony SLog3 LUTs are usually OK with Sony cameras, when they are properly set-up. After that, it’s the wild west.
So your original premise, that simply adding camera LUTs automatically gets most editors the color grading they need, doesn’t work very well in actual practice.
– Oliver
Oliver Peters – oliverpeters.com
-
[Oliver Peters]”So your original premise, that simply adding camera LUTs automatically gets most editors the color grading they need, doesn’t work very well in actual practice.”
Indeed using LUTs in practice has taught me to insert a LUT on the second node in Resolve followed by a third node. this allows me to take a log or film look image and do highlight/lowlight recovery on the first node, before a LUT is applied. In my experience chucking a LUT on log images always needs pre and post LUT tweaks. And that is before we start working on a look. This is basic grading for a doco which thankfully more and more is being shot with these log or flattened gamma presets.
Raw is a whole different level with complex tweaking of many parameters like ISO, color balance, sharpening etc, all before we start adding correction nodes. Depending on the raw source, LUTs may not be applicable at all.
-
[Michael Gissing] “Indeed using LUTs in practice has taught me to insert a LUT on the second node in Resolve followed by a third node. this allows me to take a log or film look image and do highlight/lowlight recovery on the first node, before a LUT is applied. “
Depending on what you’re doing in the first node, you probably only need 2 nodes. First node is preLUT + LUT, second node is post LUT. See the order of operations here:
https://vanhurkman.com/wordpress/?attachment_id=3331
Only HSL curves, Channel Mix, Soft Clip, Defocus, and Node Sizing are applied after a LUT, on a node. So if you’re just adjusting exposure/contrast/color balance you can do it on the same node as the LUT and it will be identical to your 3-node structure.
—————-
Michael Hancock
Editor -
[Michael Hancock] “Depending on what you’re doing in the first node, you probably only need 2 nodes.”
I sometimes do just the 2 nodes but I often like to disable nodes to look at cumulative grade work and it helps to separate. I’ve grown used to adding nodes to separate for that reason and also to unpick or modify more easily. Having started with Color, I found you had to be precise about order when directors or DPs wanted an aspect changed. Resolve lets me jump in with new nodes anywhere in the chain.
It’s less about actual signal flow than control of aspects to bypass
-
[Oliver Peters] “So there are a number of companies out there with full editing consoles – just not the cheap ones.”
Tangent has been working with Autodesk on the high-end Tsunami concept panel, initially for Lustre. Here’s Andy Knox from Tangent, interviewed by Jonny Elwyn:
“The Tsunami Concept is a versatile platform, allowing us to produce both generic and bespoke panels… Internally the panels are truly modular which allows us to rapidly alter the layout or even introduce completely different panels to meet with the specific demands of our software partners.”
Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive | RenderBreak [blog] | Profile [LinkedIn]
Reply to this Discussion! Login or Sign Up