Hi Gunther,

Sometimes (1 out of 10 sequences or so), after I have generated visual previews and then want to deflicker, LRT tells me I need to save before being able to deflicker. I save, and LRT generates the initial visual previews all over again. Only then can I deflicker.

A bit annoying for sequences of 375 images, but maddening for sequences of 1500 images.

Is there something I'm doing wrong (note: I'm doing the exact same workflow every time), and is there a way I can avoid this, or is it an actual bug?

Windows 10 Pro Ver 1809
LRT 5.3.3
LR Classic 9.1
Camera Raw 12.1

Many thanks in advance for any help.

Chas

LRTimelapse will only tell you that it needs to save, if something changed (background of image number is red) LRT has some intelligent algorithms to fix inconsitant metadata, which might lead to a "changed" sequence.
Please check, what is making the images appear as changed. To do so, if the image number appears with red background, right click on the image, then "Image Properties" (last menu item). Then on the third tab you will be able to see, what's changed.
Please post it here.

Hi Gunther, sorry it's taken me so long to test and get back to you.

I think I have an idea of what's going on: it's not a bug, but user-error (of course :-)

Once in a while I shoot a non-holy-grail sequence (all same exposure settings, unchanging throughout the sequence), but still want to use the visual deflicker capabilities of the Visual Workflow. So, I edit one photo in LR, copy those edits to all other raws, load those edits into LRT, skip "Auto transition", and start the Visual Previews.

What I think may be happening with my "LRT needs to save before you can deflicker" issue is that I've started the Visual Previews before every XMP has been read by LRT (because there are a lot, and I'm eager to start creating the Visual Previews). So, from LRT's perspective, one or more XMP files has "changed" since they were loaded, and thus it wants to save before allowing deflickering.

I had assumed that all the loading that is going on when I click "Reload" would finish before LRT starts calculating the Visual Previews, but I'm thinking not (if I jump the gun and click Visual Previews before all that loading is done).

The only thing that goes against my theory is that I always check the Visual Preview-ed movie before deflickering, and I would have expected to have at least one frame unedited (missing the XMP file settings), but I never have. What I haven't been looking for is a number with a red background, I'll be looking carefully now.

What do you think?

The problem is, that if you do the syncing in Lightroom only and then go to LRTimelapse, LRT will recalculate some values, that Lightroom doesn't upgrade.

The intended workfow is: even if you don't animate any keyframes, do the regular workflow. In LRT create only one keyframe (first in sequence). Then edit that keyframe for the whole sequence in LRT. Save Metadata for that keyframe. In LRT reload, then do "Auto Transition" - this will bring the settings of the first keyframe to all other images (that's what you do with sync in LR) - now you can do Visual Previews and Visual Deflicker as usually.

The trick is, that the LRT wokflow is designed that it's all the same, whether you use one keyframe only (no anmiation) or a couple of keyfames. Whether you do holy grail or not. It's always the same procedure. Which makes it easy to learn and you don't make any errors.

Ah, that explains why LRT always creates a single keyframe on the first photo.

Thanks for all this info, that's really helpful. I'll alter my workflow accordingly.