Sequencer improvements 2019 from an Artist not a programmer

General Feedback & Requests
1

markers - please allow me to move them ( now if i make a mistake it seems I have to delete all my markers and start over ???)

name]/add info to markers - now they are just lines with no info for why i put it there

THE SOLO BUTTON - the most important button in After Effects - I would hate to use AE without it - your solo button is hidden in 2 menu clicks and is really unusable AND everything comes back on when you hit play ??? this makes no sense to people coming from After Effects world - when you shut stuff off in the outliner it stays off so you can focus on what you want to focus on

another button to make ALL BLueprints play in Sequencer without going into pie - please just make it EASY to use - I tried the going in and clicking the run construction script and turning it on in Sequencer and it didnt work

lock render setttings to the Sequence please so I dont have to continually find what that Sequence should be rendered at

Conversion tool for MM to pixels so I can set up the width and height of my frame in the editor in pixels so I know where my edges are for my custom screens and projection mapping jobs where it has to be pixel perfect - somehow After Effects lets you do this effortlessly

button in Sequencer to stop auto save while I am working in it - make it a strong color - or at least not autosave when I am playing the sequence - i dont want to have to go shut it off every time I go into Sequencer

Select layers -“MAKE SUBCOMP” - so I can fold stuff into other Sequencers easily

get rid of CUT TRACK - can someone tell me why this is needed ? - why do i not need that in premiere or After Effects ? How is this a benefit except to make every one confused and mess up renders cus it gets switched ALOT - just have the camera be the camera - why make it complicated - If there is actually a benefit can someone tell me what it is ?

there are at least 5 things in the preferences that should be just along the top of the Sequencer UI - there is a huge swathc of empty UI use it - put functionality there instead of hiding it -
like the motion presets - frame curves - frame all - etc

make the preset motions way easier to get to and allow them on Keyframe interpolations - curves - etc - used now in the blending stuff

make the pre and infinity motions - specific to each thing - transforms - rotations etc - I’d prefer if they where specific to xyz each has its own

lock -0 axcess from happening so if you want to go before 0 you have to uncheck something - it is a constant headache to try and get stuff at 0 where it should be just a flick of your hand
again no other software Premier or After Effects or any 3d really allows going before 0 easy - its way more on the 10-20% usage I see in game stuff but - its really annoying for most of the other time . so let the advanced users uncheck it to make it work and the people who are use to it never happening dont have to struggle with it - like all premiere and after effects users

make a thinner UI option- your UI is so fat it takes up alot of screen space even on my 2 4k monitors

its really challenging to copy and paste keyframes I don’t get it - sometimes it takes like 3 or 4 tries to get it to work

duplicate layers with all keyframes - maybe make those spawnables ?

when you change the filmback settings - now you have to hit save on the Sequencer timeline why ? Why not just have stuff autosave - this is one of the only things that I found does it so its just confusing to anyone learning - It took me a long time to figure out why my camera wouldnt hold the settings - I dont get this coming from the Production world - when you make a change it gets saved . You can always undo but I’ve lost so much work cuz stuff doesnt autosave - materials to - there you have to hit apply - save and then when you close window it asked you to save again - its very unituitive and makes no sense . Are there really alot of people creating stuff that they dont want to save ??? I dont get the reason why autosave isnt institued in every area - I understand compiling can take time and you don’t want to do that over and over but can’t you still save it without compiling it ?

Allow more than one Sequencer window open in tab - sometimes the breadcrumbs are lost or I am not making them right cuz its very confusing - I’ve been using After Effects since version 1 and I’m very confused alot on how Sequencer works

add procedural noise module so I can just throw a block onto the timeline with actions that will make my “not characters”/static meshes or blueprints move quick - easy - anyone can use in seconds - sinewaves , audio reactions, math functions, reference Signal from Greyscale Gorilla plugin -instant secondary animations that anyone can use in minutes of learning your software - no blueprint skills needed - for the artist

hope someone is listening :slight_smile:

2

This sounds really whiny and this is not after effects. Asking a bit more professionally may help your cause.

3

4.22 should add the ability to add names and colors to your markers, as well as changing which time they’re on via the properties panel. This feature was an experimental feature that was added as part of another feature so its featureset continues to evolve as people find new and interesting use cases for them.

As far as I know there is no support for any solo-like features. You can disable tracks (which prevents their evaluation) but this is a non-temporary change - they stay that way until you re-enable them. How would you expect this feature to work? If you solo something and then go in game, do you expect it to stay soloed or do you expect all active tracks to re-enable themselves?

I don’t know what this means.

This is something I’d personally like to look at in the future, but these things take time. As an intermediate workaround you can configure the movie rendering pipeline through Python and kick off a render using a Python script. An example is included in the /Engine/Plugins/MovieScene/SequencerScripting/Content/Python folder iirc.

I don’t know how plausible this is. Sequencer is designed for editing level sequences which are intended for use in full 3d worlds designed to target playback on a wide variety of screen resolutions (from phones to 4k screens). If you want to be doing things in pixel space it sounds like you want to be in UMG and not Sequencer. Can you explain your use case for this more?

Understandably frustrating but autosaving in the middle shouldn’t cause any lasting issues (such as saving objects in the wrong location).

Good idea!

The cut track specifies which camera is active. It is common in game cinematics that you may want to animate multiple cameras for the duration of the cinematic. By having a separate track that chooses which camera is “active”, it allows a developer to easily and quickly adjust which camera is active - for example if your art director comes over and says “can we cut to a shot of his face here for dramatic effect”, instead of finding the camera that is currently active, cutting its section into two (which makes it harder to manage later) and inserting a new camera and positioning it at his face you can instead just slice your Camera Cut track, take your new section and assign it to the pre-existing camera that faces him. Later when your art director changes his mind and decides it’s not worth the effect, you simply delete the section you made and everything is good to go again.

Can you post an example of your Sequencer layout? The toolbar already tends to be pretty crowded on most people’s screens once they have opened the Curve Editor. What are “motion presets”? Pressing F should be the keyboard shortcut for focusing on selection.

Same question as above.

I believe they are already like this, though I’m not sure it’s expressed very well in the current curve editor. This is something I’d like to improve for 4.23.

Do you mean zero time? Does enabling “Keep Cursor in Playback Range while Scrubbing” help?

IMG_20190331_mstsc_143539.png
IMG_20190331_mstsc_143539.png716×340 92 KB

I don’t think this will prevent you form pushing sections/keys outside of the playback range, but there’s no technical reason you can’t put your sequence wherever. Would having this option prevent you from moving edited data outside the playback bounds (and not just the cursor) address the concern you have?

If you can find reproduction steps please leave a bug report via the bug reporter.

I don’t know what this means.

I think you are talking about a couple of different systems. The editor supports an auto-save feature which is global, and, as you’ve previously noticed, it requires stopping what you’re doing while all assets that have been dirtied get saved. In Sequencers case, it sounds like you are working with a Spawnable. Spawnables store a template inside the sequence which describes what the properties are set to. However, when you’re changing the settings in the Details Panel on a spawnable you see in the level, that is actually a copy of the template that has been spawned into the world, so the settings you are changing are being applied to the copy and not the original. When the Spawnable is destroyed (by evaluating to a section where the Spawn track is false), the Sequence is saved, or the button below is pressed the settings are copied from this duplicate onto the internally stored template so that they will be used the next time something is spawned.

IMG_20190331_mstsc_144919.png
If you can come up with a specific example where settings are being lost (such as changing it on a spawnable and then opening a new sequence) please use the aforementioned bug reporter so we can take a look at it. In general though, if you save the sequence (either manually, autosave, or when closing the editor) the changes you’ve made to the Sequence asset should be saved.

Materials are different. They have separate Save and Apply buttons (though Save implies Apply iirc). Applying a material change can be a very expensive time consuming operation. If you use material instances and have a “master” material for most of the materials in your game, when you apply it needs to recompile all of those shaders which can result in thousands of shader compile operations. It can be beneficial to have more control over when this happens as you might want to change your material, confirm that it looks good in the preview window and then only have it applied when you save the asset when you go to lunch for example.

It is pretty common to make changes to things and not want to save them. Saving them will prompt you to check out the asset from source control, which in some source control providers prevents anyone else from working on them. If you need to change a material to test something, but someone else needs to update it to do real work on it, if you check it out you keep them from doing their work, and you don’t actually care about saving your test change anyways.

This is unfortunately something that would be very technically challenging to implement and leaves lots of problematic design questions open; What if two sequences both try to animate the same object and they’re both playing at once? If you autokey something which Sequence does it get autokeyed into? If two sequences are trying to control the camera, what happens?

The breadcrumbs are relative to the master sequence you have open. Each time you dive into a shot or subsequence track, the breadcrumbs are pushed into. If you open a sequence from the Content Browser (or via clicking on the button in the Level Sequence Actor) it is considered a new master sequence and your breadcrumbs are cleared. Sequences can contain other sequences (nestled, and multiple times) and a sub-sequence can be shared between other sequences so it really depends on how you opened that sequence, ie: was it opened in the context of a parent (by a shot/subscene track) or was it opened in isolation (level sequence actor, content browser)

Sequencer is unfortunately not set up for blending arbitrary data and this would be quite complicated. For example, if you wanted to apply noise to a Transform Track, you would have to make it aware of a Transform track so it could not apply noise to the scale functions (for example), or to apply different noise to the rotation, etc. But then if you wanted to apply noise to a Camera FoV track, now it has to be aware of that. There is a limited implementation of this in the form of the Camera Shake track which uses a premade asset that describes some mathmatical parameters about generating noise.