Emulating Film Halation in DaVinci Resolve

Is it just me or there is no video?

We’re not sure why it disappeared. But it’s back now.

This couldn’t be more timely! I’ve been trying to emulate helation for a while now and I landed on sort of a mixture of these two methods. But I had to tweak it on almost every shot due to a luminance shift that affected the shots and sent highlights into clipping territory. Thanks so much for sharing both of these methods, I’m excited to try them!

Thanks! yea thats part of why i like to put it at the beginning of the pipeline, since I’m not adjusting luminance in any significant way – it shouldn’t blow out highlights – it just acts like the halation was there on the original source.

I was just messing around with it on a super 16 emulation I’ve been working on for a while. I started the look from scratch, focusing on tweaking the halation to get it looking right on the raw clip, made that a compound node, and then re-built the whole look with a denoise (optional to use but in the first node), the halation on second node, and then everything else after that in my fixed node tree. It’s really responding well, so cool!

Wow, really knocking it out of the park to start off the year! Awesome insight, @tao-ml-d98c1545b7619bd99b817cb3169cdfde:disqus . I’ve been using a method similar to the one you said Dan uses but I LOVE the concept of creating a matte with edge detect. Super flexible and may even look better than the Log-Lin + CST method. Thanks again for the great insight and inspiration to experiment !

Great Insight! And, as usual, the technical aspects and versatility of your signal path is mind expanding.

Thanks for this Insight. I might have missed it but did not see a mention from where the linear halation technique originated. As far as I know this technique was first very clearly explained on the LGG forums, by Matthias Tomasi. He also goes deeper on the logic behind this technique than this insight. I don’t mind seeing it repeated here in this video it’s a great technique! But again, no mention of Matthias feels a little wrong since this is paid material.

Thanks Wietse for the heads up. I wasn’t aware of a LGG post by Matthias. I’m certainly happy to give credit where it’s due and didn’t mean any disrespect – I’ve added a link to the article to the LGG thread. In talking to Dan it seems like he had a power grade of that technique, but didn’t remember its origin.

As you can see in the insight, I prefer the method that I show building from scratch. Both methods produce great results – but I feel it’s more accurate to the physics of halation to isolate based not only on brightness, but also on edge-detection, to limit the effect to hard transitions of light, not all bright areas.

Hey Patrick, I think the video has disappeared again..?

And – it’s back again. Thanks for the heads-up.

Hi Jou,

I really enjoyed this insight.

I believe I’m the one who gave the powergrade to Dan (indirectly). I developed that powergrade in Resolve based on what Matthias described on the LGG forum, as his approach was designed in Nucoda. I believe the powergrade Dan is using to be the one I shared on a group you’re also in Joey. I also shared it at the time on that LGG thread.

I really like your method, I downloaded the powergrade, it gives excellent results, however I still prefer the one I developed based on Matthias’ technique, I think there’s power in it’s simplicity and the fact that it’s converts the log to linear means you can use it with any log curve from any camera, whereas with keying there’s more guess work involved.

Also I have to disagree with your argument that you cannot customise it: by changing the location of the 100% control point of the custom curve toll on the lower node you can control the threshold of the effect and the intensity is controlled by the amount of blur you dial

The added contrast you mention can also be avoided solved by using a parallel node instead of a layer node and pulling the top right point of the custom curve considerably to the left on the top node (essentially recreating the add blend mode). I started doing the halation emulation this way since blend modes do not work in an ACES workflow, which I use often.

Thanks! And thanks for making the original powergrade as well!

All great points. I don’t want to say that my method is the only, or even the best method – but it’s definitely given good results for me. I actually haven’t tried mine in ACES- I’ll be curious to see how it holds up there.

And yea I should have been more specific – both methods are dfinitely customizable, but the customization I was more referring to was to do with the shape of the halation – ie how much you blur the mask, and the different edge detection settings that are available to effect its width, spread, and softness.

Mind blown! Will need to watch this a few more times and try it. Thanks!

brilliant