Broadcast Legalization Part 5 – Hard Clipping in DaVinci Resolve

Patrick, I finally found time to watch your first grade-along training and this insight make me think about LUTs and the best place to apply them.

Instead of applying the LUT in the last node of every shot of the timeline at the clip level, maybe a better strategy is to put the LUT on the first node at the track level, just before the hard clip node. Of course, this suppose that all the clips are from the same camera.

Do you see any inconvenient in that?

Steve,

I’ve addressed this notion about where I like to place my LUT in it’s own Insight here in the Library:

http://mixinglight.com/portfolio/working-with-luts-a-3-node-approach/

I’d love to hear your thoughts after watching that 🙂 It’s a great topic for discussion.

While I understand why you put your LUT at the end of the pipeline (like Truelight at layer 99 on Baselight), I fail to understand why you would want to grade after applying your LUT. Won’t you loose the 32-bit processing of Resolve after the LUT and then limiting yourself? That why I though it might be a good idea to put it on the track level, before applying hard clipping, grain or addition overall effects & looks.

DaVinci doesn’t clip after a LUT. The LUT may clip, which means you need to manage the image feeding the LUT… but as I explain in those tutorials – controls act differently before and after a LUT and we can use that different behavior to our creative advantage.

That’s why I intended by “loosing the 32-bit process”: the LUT will clip everything that out of range and you won’t get them back after it.

So you are suggesting to put a third node after the LUT for creative purposes. I tend to avoid LUT in general where I can as I like to build everything myself but it might be a good solution for fast creative looks. I will experiment a little bit with them and see what I can get from them. (^_^)

Thanks for your insight!

I use LUTs as often as I discard them. If they work, great. If not, cool. I definitely encourage to watch my series on LUTs here on ML. One of the things I discuss… not using them at all…

And yes, I almost always work after the LUT. If nothing else to precisely place my black point. But also the controls are linear after the LUT… which I sometimes find useful.

As I suggest in my LUT series, I consider them a creative tool masquerading as science 🙂

Hey Steve –

I use the 3 Node approach as well with the LUT in the middle – but its important to note you don’t HAVE to do that. In Resolve a LUT applied on a node is the last thing processed in that node – so if you wanted to you could make a grade pre LUT by simply making a correction on the node that has the LUT. Then for grades post LUT just add a new node.

In other words you could simplify by have 2 nodes instead.

However like you I’m more DIY these days.

Hi Patrick great insight as always!

I use scopebox connected via a mini Recorder and I have a question. I have no issues on the Waveform or on the Vectorscope but I often have excursions on the RGB Parade, I’ve never had any job get back to me due to QC failure but I’m wondering if those excursions mean anything:

I turned peak on and scrubbed through the timeline of a project I’m working on (which in this case is for the web distribution) and this is what I get: No issues on the Waveform or the Vectorscope but if I look at my RGB Parade it has all those peaks above and below 0 and 100%. I have the parade transform set to Studio RGB.

https://uploads.disquscdn.com/