Bug on sequences Dopesheet

megametx

I'm currently using last stable 4.2, I noticed some problem with sequences like this video
if I do some change on sequences "15/liq" as video ( or sometime moving bones,changing slot opacity ) that not directly affected bugged sequences "3/liq" it would randomly making a new sequences key with abnomal frame rate, sometimes messing with other premade key on dopesheet
The problem also persist on other project, which also mean it would happen randomly with random bugged sequences, on more complicated project with more sequences, this bug can even cause multiple bugged sequences at once, it should be more clear at long sequences set to play Once
I still dont know how to reproduce it, this one sequences "2/Comp_" bellow is example of abnormal framerate and sequences key (play Once, FPS:50)

Misaki

megametx We apologize for the inconvenience! We would like to reproduce and look into this issue, so we would appreciate it if you could send us a Spine project that reproduces the issue via email. : contact@esotericsoftware.com
Please include the URL of this forum thread in the email so we know the context. Then we can take a look at what's wrong.

Spinebot

It seems like you are experiencing an issue with sequence keys in the Dopesheet where changes to one sequence can unexpectedly affect another, creating new sequence keys with abnormal frame rates. This type of issue can be challenging to diagnose without specific reproduction steps. However, based on similar past issues, it is recommended to ensure that you are using the latest version of Spine, as many bugs related to the Dopesheet have been addressed in recent updates. If the problem persists, consider documenting the steps leading to the issue as precisely as possible, as this can help in identifying and resolving the bug.

megametx

Misaki I had sent it

Misaki

megametx Thank you for sending us the Spine project file. I also appreciate the video you provided showing the entire editor screen. This was very helpful.
I noticed one thing when I watched your video. Before deforming the mesh, the sequence key button is orange:

And as a result of deforming the mesh in this state, since Auto key is turned on in your case, I think that changes that had not yet been recorded were converted to keys, resulting in additional keys being added to the sequence.

As for why the sequence keys turn orange, it seems to happen when you change the Timeline FPS in the playback view.
The tricky part is that the change doesn't happen immediately. Rather, it appears to change to orange when you select the sequence attachment on top of it.
There may be other reasons why the sequence button turns orange, but I haven't found any other cases yet.
If this gives you a clue, could you tell us what operations you were performing before this issue occurred?

megametx

Thank you for your detailed solution, i doubt this would have something to do with Timeline FPS, So how can I fix it without turning off Auto Key ? or what should i do with Timeline FPS to solve this ?
I'm on Windows 10

Misaki

megametx First of all, I believe this is a bug that needs to be fixed, so I would like to understand how you encountered the problem.
At least on my end, I have confirmed that changing the Timeline FPS reproduces the problem 100% of the time. Specifically, it can be reproduced by following these steps:

Align the timeline bar with a frame where no sequence keys are recorded.
Change the Timeline FPS in the Playback view.
Change the selection.
Select the sequence attachment again to confirm that the sequence key button is orange.
Without moving the timeline bar, move any element to trigger auto key, and a key will be added to the sequence attachment.

As a workaround, moving the timeline bar when the sequence key button is orange will resolve this issue. Once you move the timeline bar, the key button will return to green, and moving bones or attachments will not automatically add keys to the sequence.

I suspect that if you have never changed the Timeline FPS, there may be other operations that are recognized as changes to the sequence. Therefore, I would like to identify operations that are mistakenly recognized as changes to the sequence. Could you recall what operations you were performing when you encountered this issue?