Unless I have a misunderstanding of the issue; the "keyframe time" is the time reference to the original clip rather than the timeline itself. For example, if my original clip is 20 seconds long and I trim 5 seconds from the beginning and 5 seconds from the end, I am left with an object 10 seconds long. If I then place a keyframe at the beginning and another one at the end of the object, the beginning key frame will have a time of 5 sec and the ending keyframe will have a time of 15 sec. No matter where the object is moved along the timeline, these keyframe times will remain the same because they are referencing the original untrimmed clip.
2- keyframe time marker cursor "disappeared"
This one is interesting, and I need to conduct more testing to determine what exactly the system is doing when a clip is cut or trimmed. What I have observed is when the red marker disappears, if you lengthen the object, you may find another "hidden" keyframe. So, the red marker jumps to a keyframe that is invisible because it resides in the trimmed or cut area of the clip. More testing is needed because I'm not sure if this is the only reason why the red marker disappears or just one of the reasons.
Note that I considered it a bug because I understood that it was a reference to the main timeline of the video to be produced. But also thinking that if it is specific to the clip in relation to the original video then "it is not a bug!". In the manual I did not find what the reference is for this keyframe time marker when the icon turns blue! On the other hand, about 3 years ago when I migrated from Vegas Movie Studio to MMS 2023, I opened a topic involving the program's functionality questioning whether the program stored information about the time interval of the clip in relation to its original video, as the VMS informed. In the posts no one said if it did! Now with this reference of yours, I consider it welcome because if this is the case, it solves this other issue! So I thank you and if we meet I'll buy you a coffee!
Regarding the second item about the marker disappearing, it shouldn't disappear exactly as I show at the beginning, since it's a normal feature. I don't remember exactly, but I think that in version 17 they fixed a bug where at some point, in relation to version 16, the initial marker of the clip in the key frame also disappeared and this seems to have been fixed.
@AAProds As per the previous post, it was clear that the beginning of the object in the key frame is directly related to its source video. So answering your question, it does not necessarily mean that it is the start of the object.
If you create 2 key frames in a clip, added at the 1st and last position of this object. Then you cut the object in half and play around within the SPR timeline via go to start, go to end. You know what's happening in the main timeline. Now take the 1st clip and invert its position with the second clip and drag them a few seconds to the right in the main timeline. Then try to navigate within the SPR timeline and you will conclude that you get lost in the navigation logic! Whereas when cutting the clip, the VPX should restrict navigation in the SPR only between the key frames of the selected object in question.
. . . . I don't remember exactly, but I think that in version 17 they fixed a bug where at some point, in relation to version 16 . . . .
The times you see with reference to the image below, for VPX 16, are:
= the timeline cursor position
= end of movie, this may be greater than the end of the last object in the movie after editing.
Keyframe times are relative to the start of the object
Cut object at 1 sec
If you cut an object at for example 1 sec and add keyframes to the second section they will still be referenced to the start of the complete object, it does not matter whether the part before the cut is present or not, or moved.
The only time the Preview monitor times and the Keyframe times will be the same are when the object starts at time 0 and you do not move the second part to the right in the movie.
AFAICS there is no bug in VPX 16 or in the video clips you have shown.
Hi @johnebaker in a simple little project and you get around the issue by softening it. But this is inconsistent... let's see...
--->Scenario Description:
In a normal video editing workflow, a user adds a 10-second clip to the timeline. During this phase, the user also enables keyframe mode and sets 3 keyframes at time points: "0 seconds, 1 second, and 9 seconds", then saves the project.
Later, upon further review, the user decides that the segment between "0 and 1 second should be discarded". The timeline cursor is positioned at the 1-second mark, and the user performs an operation to "discard the section from 0s to 1s", keeping only the remaining part of the clip in the project. The project is saved again.
Now, to examine the modified clip more closely, the user uses the dedicated "keyframe navigation buttons", which are represented by arrow icons (next/previous keyframe).
Here’s what happens:
1. The user positions the timeline at the 1-second mark and presses “go to next keyframe” → the program jumps to 9s (correct). 2. From the 9s keyframe, the user presses “go to previous keyframe” → the program goes back to 1s (correct). 3. From the 1s keyframe, the user presses “go to previous keyframe” → "the program jumps back to 0s", which is "part of the discarded interval".
---
Why This "Is Clearly a Bug" in VPX16 and 17:
---> 1. "Unexpected and Illogical Behavior":
* Once a clip segment has been removed from the timeline (0s–1s), the editor should no longer "retain or reference" keyframes located within that deleted segment. * Navigating to a frame that has been explicitly discarded "defies user expectations" and the logic of the editing process.
---> 2. "Failure to Update Internal State":
* It appears that when trimming or discarding a clip segment, the program does "not properly remove or invalidate keyframes" that fall within the discarded region. * As a result, the keyframe at 0s remains accessible, even though it's no longer part of the visible or active timeline.
---> 3. "Negative Impact on User Experience":
* From a client or editor's perspective, this is a "serious UX flaw". Navigating back to discarded frames can:
* Confuse the user; * Cause unnecessary doubt about whether the cut was effective; * Introduce fear of data corruption or unexpected behaviors in complex projects. * In a professional editing environment, this type of behavior "undermines trust in the software".
This stance "lacks technical validity", for the following reasons:
* The behavior clearly contradicts standard editing logic: a removed segment "should not retain active navigable metadata". * If the editor allows users to cut content, but still lets them navigate back into discarded content, "it creates a functional inconsistency" and defeats the purpose of the cut. * Any functionality that behaves in a way that "breaks the user’s logical flow or introduces invalid states" should be treated as a bug.
VPX is considered a professional and good editor! So I hope that new editions will also bring corrections, which is exactly the case here!
. . . . From the 1s keyframe, the user presses “go to previous keyframe” → "the program jumps back to 0s", which is "part of the discarded interval" . . . .
This behaviour has been present in MEP/MMS/VDL and VPX, for 15+ yrs, ie least as far back as Movie Edit Pro 16, I do not have an early version of VPX installed so cannot test. Similar behaviour is seen in other non-destructive video editors.
The behaviour is, IMO, correct when the video object has been cut and the unwanted portion removed from the timeline and the object is selected. The editor records the physical (real) start and end locations of the video clip, and the visible start and end ie when trimmed.
The key frame positions are recorded based upon the physical (real) start of the video clip which does not change when you remove the start of the clip.
It does not matter where you put the key frames if you split the object between a pair of key frames the key framed effect is not broken and the behaviour of the previous/next key frame buttons is still the same - see video below.
It makes no sense for the deleted or inactive interval to become visible as a haunting of the active navigation. I understand that the behavior of the "previous" arrow icon relative to the first keyframe should simply be visually disabled (and effectively so in this case). This would solve the situation quickly, as suggested to the Magix programmer. So if you want to restore what was hidden, simply reactivate the interval as shown in the video above. But I'll check the behavior in some other editors I have and post it here.
In a quick situation, Vegas Movie Studio 17 does exactly the same logic, that is, when eliminating the unnecessary interval, it makes it impossible to navigate the area in the keyframe of the eliminated keyframes, and finally it does not move or disappear from the timeline, obeying the purpose of discarding by the user and the keyframes are eliminated. (** a button to navigate to the last and the first is always welcome, that is, without having to go through the intermediate keyframes **).
I can think of three solutions to this issue. The first is easy to implement by all of us editors of video. The other two require some programming enhancement.
Instill in your personal workflow, the habit of cutting and trimming clips/objects prior to adding any effects. This is typically my workflow, hence I wasn't aware of the issue until @Roberto-Goncalves brought it to our attention. This solution doesn't require any programming.
If an effect(s) is in place and the object is subsequently cut in two, then have the system add key frames at the end of the fist clip and the beginning of the second clip. The effect doesn't change, just keyframes added at these points. Likewise, if an object with an effect(s) is trimmed, automatically add keyframes to the trim point. Again, this doesn't change the effect, it just adds a keyframe(s). This solution does require programming but I don't think it would be difficult to implement.
If the enhancement in 2 is implemented, then make the keyframe navigation arrows smarter so that they stay within the clip being edited rather than hopping to keyframes that are hidden due to cutting or trimming.
Note: In my testing of this issue, I believe I've come across a couple of potential issues as it relates to key frames which I will share in this post once I complete my video capture and edit to help explain the phenomena.
In Magix it seems previously placed keyframes keyframes are retained in 'memory', so if the length of the clip is then restored to it's orig length after being previously trimmed the keyframes reappear,
In Vegas if a clip is trimmed any keyframes before the trim point are lost, the existing keyframes adjust to fit the new length even if the clip is the restored to it's previous length.@Roberto-Goncalves
This is the key point -
3. From the 1s keyframe, the user presses “go to previous keyframe” → "the program jumps back to 0s", which is "part of the discarded interval".
---
Why This "Is Clearly a Bug" in VPX16 and 17:
---> 1. "Unexpected and Illogical Behavior":
* Once a clip segment has been removed from the timeline (0s–1s), the editor should no longer "retain or reference" keyframes located within that deleted segment. * Navigating to a frame that has been explicitly discarded "defies user expectations" and the logic of the editing process.
I agree, Magix shouldn't jump to an empty part of the timeline referencing the previously trimmed part of the clip it's now ghosted keyframes.,
Vegas works like your VMS, after the clip is trimmed, using the arrow buttons to navigate the keyframes on the keyframe timeline will only let you navigate within the trimmed clip.
Magix is not updating the keyframe timeline after the clip is trimmed, it's still looking & finding the now irrelevant part of the clip & it's keyframes.
I went to evolve and see that "without even eliminating" the divided clip, in the first interval the VPX16 when eliminating the effects, as the SPF interface shows, its keyframes are no longer visible, but "their position is not eliminated"! Simple as that! See if this happens on your machines and post here.
In this video @Roberto-Goncalves is restoring the clip to it's orig length by dragged the end of the object. If this was done in VMS those keyframes would not reappear.
In this Video he is using the Undo button so those keyframes obviously reappear.
This is not a true or correct comparison of the two programs.
Could be, the 'leading' video editor, used to, and may still, show the same behaviour as MEP/MMS/VDL and VPX, however there is the option to turn this off.
Going back to the editors we are concerned with
. . . . It makes no sense for the deleted or inactive interval to become visible as a haunting of the active navigation . . . .
and
. . . . Magix shouldn't jump to an empty part of the timeline referencing the previously trimmed part of the clip it's now ghosted keyframes . . . .
An important point is that this behaviour occurs only while that object is selected and the Combine effects curve with object length option isunchecked.
Select another object or section of the split object, the KF timeline is related to the now selected object/section.
Vegas, and presumably VMS, behaves in a different way, where you have to click the Pan/Crop event button of the individual split sections to see the selected sections KF timeline
. . . . In Vegas if a clip is trimmed any keyframes before the trim point are lost, . . . .
True.
However, a new keyframe with the correct parameter value is added at the start and end of the appropriate sections of the Split clip
. . . . the existing keyframes adjust to fit the new length even if the clip is the restored to it's previous length . . . .
Depending on your view, the VP and MEP/MMS/VDL/VPX behaviours are either an unwanted behaviour, ie 'a bug', or a benefit.
Actually the situation with VMS17 is more complicated as how the keyframe position behaves can vary with how you adjust the length of a clip. Dragging the beginning of a clip to trim it gives different keyframe positioning to cutting it.
Anyway what would be ideal with any editor would be the behaviour that was in my old Serif Movie Plus where you could tell a keyframe how to behave by choosing the Keyframe Mode.
@johnebaker When you cut a clip that has keyframes inserted, but you don't delete it from the work, and then "delete the effects", what do you expect? I believe that the effects are eliminated! But you discover that they were not eliminated and worse: they cannot even be recovered because the SPR interface does not allow it! What is the benefit?
. . . . cut a clip that has keyframes inserted, but you don't delete it from the work, and then "delete the effects", what do you expect? I believe that the effects are eliminated! . . . .
The simple answer Yes and No, the behaviour is shown below and is what I would expect
The logic is:
A single reference of the KF's are 'recorded' for the entire clip
After the cut/split, a second reference is recorded, each section has its own reference to the KF's
Delete the KF's from one of the sections removes the KF's from its reference, the other sections KF reference is not affected.
Note that within the original question, it involves a complex project with intervals with several keyframes, hoping not for a miracle, but for something productive, agile that eliminates ghosts and maintains cleanliness and referential integrity. In the case that you expose in one timeline you keep the ghosting and in the other you eliminate all effects, but at what cost? No more ghosting in this functionality! The ghosting of interpolations is enough to smooth the video! anyway I also appreciate your inspiration HCH!
. . . . it involves a complex project with intervals with several keyframes, hoping not for a miracle, but for something productive, agile that eliminates ghosts and maintains cleanliness and referential integrity . . . .
Complexity of the project makes no difference to the behaviour exhibited, nor does the behaviour affect the project.
Referential integrity is maintained, as I previously commented 'An important point is that this behaviour occurs only while that object is selected . . . . ' it does not affect the behaviour of other objects on the timeline in any way.
Personally I prefer the behaviour exhibited, it tells me that what I see of the media object, is a section that has been trimmed from a longer object and there are more keyframes outside the visible part which can be recovered if necessary.
An additional behaviour which may be useful is setting the effect KFs Combine effects curve with object length option on, then, when you split the object, both parts get the 'full' effect.
This was a great exercise in exploring how VPX handles keyframes. And we could continue to debate whether this is a bug or a feature. Perhaps, as I suggested in my previous post, the system automatically add keyframes when an object is cut or trimmed (Point 2) could be considered an "enhancement" to the process. 🙃
In my testing of keyframes, I do think I have found an actual bug, unless someone can explain the logic. See the video below but the following is an outline of the workflow:
Place object on timeline
Cut object in half creating 2 objects
With the first object, add an effect with beginning and ending keyframe
Test to ensure the effect works as expected
Test keyframe arrow buttons to ensure red curser doesn't disappear
Move object to another part of the timeline
Ending keyframe disappears!
To reveal the keyframe, the object has to be lengthened