At this year’s IBC I had the opportunity to discuss my favorite version 19 features with Simon Hall from the DaVinci Blackmagic Design team, which began beta testing on April 12 and was released in its polished version in late August. As I am an advanced beginner editor without a lot of time to learn fusion, I am most interested in features that will save me time while delivering professional-level effects, and DaVinci’s neural engine in version 18 seemed designed for users like me. It’s been confirmed in version 19.
Tasks that required a steep learning curve in the past often also required multiple hours to tweak or extend through a long clip, but now the developers at Blackmagic Design have focused on increasing the number of tools that are more intuitive and require less time.
dialogue leveler
The dialogue leveler does this while giving even more control for beginners than could be found in the slightly more complex method using the compressor. In fact, the dialogue leveler is fundamentally a compressor with algorithms managing some of the decisions you’d make if you used a compressor.
Audio ducking
Audio ducking no longer has to be manually keyframed, and the interface allows just enough sliders to control the effect without it getting confusing. I was able to get the desired timing of the level of my background music to go quiet and then return within two adjustments and about a minute altogether. This is a huge leap forward for editors of videos with constant background music that has to fill in long gaps between dialogue or editors of shorts who want to repeat the same exact timing for intro music fade and return, just to name two examples.
magic mask
Discussing edge detection with Simon made me realize I knew nothing about the magic mask, a fundamental feature with which many other features are used, such as object removal, defocus background, and edge detection. The AI neural engine manages all these features, so removing a distracting object in the frame, “fixing” a slightly out-of-focus subject by defocusing the background and adding sharpening to the subject with a magic mask, applying different color temperatures to different parts of the image, all become tasks that can be accomplished in seconds instead of hours, particularly for editors who are not yet advanced.
Music remix
Music remix is a fun and incredibly useful feature, and I think it is a harbinger of things to come in the realm of music track manipulation. Now we can increase or decrease the intensity of various “parts” of a song, such as different instruments and voices, by moving just a few sliders, rather than spending hours learning a complicated workflow in Fairlight that I might forget by the time I need to use it again.
Object removal and patch replacer
Object removal and patch replacer can be compared to popular features found in Adobe Lightroom and Photoshop for fixing something in an image. Object removal applies a mask to identify and remove the undesired object, and then AI is used to pull information from the immediate surrounding area to fill in the new empty space. The patch replacer does this by actually copying a portion of what it sees in another part of the image, and this gives more manual control.
depth map
Perhaps the coolest feature is the depth map, where again, AI is used to detect what areas of the frame have the sharpest focus and what areas have the least. This information is represented somewhat like a false color effect on an external monitor, but the added benefit here is that Da Vinci can apply effects to the frame and modulate how much is applied based on the amount of focus, resulting in a precise application of the effect that would have been painstaking to perform by hand. This presents an alternative to using the magic mask so that together editors now have more tools to quickly analyze what’s in the frame and take the next step.
multi-user collaboration
One more feature that’s been added in version 19 that will save users time, though it’s not necessarily related to AI advancements is in multi-user collaboration, when you need to share a project with a single user. Let’s say you want to hire a colorist for a few hours to go over your clips. Now you can rent and assign single licenses for only as long as you need them. Although I knew this was where I’ll be using the cloud first, working with another editor, a colorist, or an audio specialist, I was convinced the cloud system was too large scale to be applicable to me. Now I’m thinking of who I want to collaborate with first.
The free and paid version of DaVinci Resolve can be found on the official website, and most of the features discussed here can be accessed for free. Let us know if using DaVinci’s AI features has sped up your workflow, enabled you to do more, or both.
Check out the full LensVid coverage of IBC 2024 on this link—special thanks to our sponsors for IBC 2024 – SmallRig and Hollyland.
You can support LensVid by shopping with our affiliate partners
Affiliates: Amazon, B&H, Adorama and E-bay.
Why should you trust us?