I'm having difficulty understanding what is happening to the radial gradient between keyframes. I am attempting to animate the grad to follow the Milky Way throughout the timelapse, but the shape of the gradient changes. Quite dramatically in one instance. See the dramatized version of my edit here:

https://youtu.be/3f6fwb7C7yc

I followed the proper workflow and made sure to only use the existing gradients. I always used the scripts in Lightrooom to sync everything. And I have the latest Adobe DNG converter software.

It appears there's something strange about how the animation of the grad is created. I've tried rotating the points used for the narrow/wide ends of the grad to see if that helped. But I got the same results.

Not sure if it helps, but for reference, I started by editing the keyframe in the middle of the sequence. Then moved the edits to the keyframes forward and backward from there. Always using the sync keyframes script. I never changed the shape of the grad from my original edit. I only moved the center point, and rotated it. But you can clearly see that the shape of the grad changes throughout the sequence. Sometimes getting thinner, and then getting circular before going back to long and thin.

Attached are some screenshots showing what I was attempting to do by looking at the keyframes in LR. And a screenshot showing the Mask 5Angle plotted against the visual preview. This seems to be showing a dramatic change in the Mask 5Angle curve about the same time as the dramatic change in shape of the grad. I tried doing a linear transition for this column, but still had strange results. Not sure if this means something, or is even helpful. But it does seem to show some sort of correlation.

Can anyone explain why the animation changes the shape of the grad? And so dramatically?

At some point, when the circular gradient's angle passes 360°, Lightroom "Flips" it. To avoid this, rotate the starting gradient 90° left, then skew it to match the former shape and bring this gradient to the next keyframes. This should avoid that behavior.

This post was last modified: 2022-06-02, 21:19 by cwc474.

Thanks for the reply Gunther

I actually had already tried doing that. That's what I meant when I said "I've tried rotating the points used for the narrow/wide ends of the grad to see if that helped." in my initial post.

But just to make sure, I tried doing it again. First with a 90° rotation. And again with a 180° rotation. I get the same results every time. It makes intuitive sense that this would fix the problem, but it doesn't seem to work.

If it does have something to do with the 'Mask 5Angle'. I'm curios, what is this actually measuring? it seems a bit redundant since there are already data points for the 'corners' of the mask. Mask 5 Top, Bottom, Left, and Right in LRT. I would think the 'angle' of the gradient could be derived from these data points. Perhaps I'm missing something.

In any case, rotating the gradient before beginning the edit, and then starting to skew it to follow the Milky Way doesn't fix the behavior I was seeing.

Do you have any other suggestions?

This post was last modified: 2022-06-03, 12:21 by Gunther.

Maybe I was not clear. Try to exchange the X and Y axis. Make the wide axis short and and the short axis wide. See if that helps.
It can be a bit tricky - Lightroom was not designed to do this kind of stuff that LRTimelapse does :-)

Check out the angle column in the table maybe that cam help you to identify where th overflow occurs and why.