Could Adobe (or someone else) adopt magnetic timeline features in a tracked timeline?

[Walter Soyka]

[Jeff Markgraf] “I don’t see a difference in the “absolute” time reference. The timecode (or frame count) of the sequence timeline in X is the reference. Just as in Avid or others. A clip exists at a particular linear time. The fact that a clip may be connected to another clip doesn’t change anything about its absolute position on the timeline – until you move it. Just like in Avid.”

Jeff, I’m curious what you think of my reasoning from a few years ago:
https://forums.creativecow.net/thread/335/41111#41309

Specifically:

[Walter Soyka] “… in a traditional timeline, you reposition clips in time directly. A timeline clip object’s temporal position is an inherent property. In FCPX, you reposition clips in time indirectly, because a clip’s position in time is a function of its relationship to its parent clips.”

I should also add “antecedent clips” to cover previous primary clips.

Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive   |   RenderBreak [blog]   |   Profile [LinkedIn]

[Michael Gissing]

[David Lawrence] “why haven’t we seen a trackless DAW yet?”

I think we sort of have. Doesn’t Fairlight allow multiple clips on the same “track,” layered on top of one another? A bit like the audition process in X?

Fairlight is utterly track based but it solved the overwrite problem by allowing clips to stack non destructively on a track. You always hear the top which means you can do crossfades to layers or hide alternates in place on the right track.

It is in fact a great model for how NLEs could have solved the overwrite issue. A trackless DAW would be an organisational nightmare. After all this is the point where we reorganise the dogs breakfast that comes from the edit. Tracks are so much more than just a way of displaying in the DAW world that no-one is thinking a trackless paradigm is worth toying with.

[David Lawrence]

[Michael Gissing] ”
Fairlight is utterly track based but it solved the overwrite problem by allowing clips to stack non destructively on a track. You always hear the top which means you can do crossfades to layers or hide alternates in place on the right track.

It is in fact a great model for how NLEs could have solved the overwrite issue. A trackless DAW would be an organisational nightmare. After all this is the point where we reorganise the dogs breakfast that comes from the edit. Tracks are so much more than just a way of displaying in the DAW world that no-one is thinking a trackless paradigm is worth toying with.”

Exactly right, Michael.

Most NLEs work in one of two edit modes – overwrite and insert. Generally speaking, overwrite covers the underlying media, insert ripples it in or out.

There’s no reason why there couldn’t be a third edit mode – layer. In layer edit mode, media might stack ala Fairlight, or there could be other UI possibilities.

The only NLE that I’ve seen attempt this is Sony Vegas which I guess makes sense given its roots as a DAW. But it doesn’t go very far with the concept. There’s so much opportunity for innovation in the NLE UI space. We’ve only scratched the surface.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
https://lnkd.in/Cfz92F
facebook.com/dlawrence
twitter.com/dhl
vimeo.com/dlawrence/albums

[Charlie Austin]

[Michael Gissing] “Fairlight is utterly track based but it solved the overwrite problem by allowing clips to stack non destructively on a track. You always hear the top which means you can do crossfades to layers or hide alternates in place on the right track.”

Logic X does the same. Also has “track stacks” which are kinda like FCP X compound clips but can be opened up in the timeline and work as summing stacks (re-routable) or Folders (a basic group) Pretty cool…

————————————————————-

~ My FCPX Babbling blog ~
~”It is a poor craftsman who blames his tools.”~
~”The function you just attempted is not yet implemented”~

[Jeff Markgraf]

Walter –

In short, I don’t agree with the reasoning.

From your previous thread:
In FCPX, the primary storyline itself is the clock that drives secondary storylines and clips. As Jeremy G said [link]: “The primary is time. Control the primary, you control time.”

Here’s where the fundamental disagreement begins.

X has a linear time clock, as does every other editor: the sequence time. Whether it starts at 00:00 or 1:00:00 or whatever, it goes forward. Period. Any clip on the primary by necessity starts at xx:xx and ends at xx:xx. Until you move it, whether by rippling the edit or positioning the clip with the P tool. Same as Avid or anyone else.

A connected clip also exists at an absolute time. The fact that it is “connected” means nothing in terms of where the clip lives on the linear timeline. Until you move it.

Many people edit by placing their “A-roll” on video layer 1, and put “B-roll” on video layer 2 or 3 or different clips on different layers, all above the A-roll on track 1. Some leave the clips in that arrangement. (Others (especially those of us who come from linear online) consider that sloppy and bring the clips into track 1 at some point to finalize the edit.)

If I decide to move a block of clips (call it a scene) to a new location, I manually select the clips on video 1, as well as any b-roll clips on video 2,3,4, etc., and move them as a unit to the new location. For the purpose of moving the block, the b-roll clips on 2,3,4, etc. have a time relationship that is tied to the A-roll on video 1. They are “connected” to the a-roll by design, inasmuch as I have selected all of them. Once the block is in its new location, all clips have new absolute relationships to the absolute time of the sequence.

Same thing in X. It’s just that the b-roll clips are connected clips, linked by default to the a-roll (primary). I simply don’t have to select all clips to move the block – just the primary. The connection is only relevant for the purposes of moving the block. Once in its new location, all clips once again have that absolute time relative to the sequence timeline.

At no time am I required to connect a clip. I can happily cut only in the primary, as long as I separate my audio from the primary. But that process seems unnecessarily restrictive.

a clip’s position in time is a function of its relationship to its parent clips.

Again, only for the purpose of moving the block of clips described by being connected to the primary.

One can like or dislike this default behavior. One can like or dislike the default behavior of clips moving vertically to overlap rather than overwriting or blocking. But except during the process of moving a block of clips, the clips have an absolute relationship to the timeline, whether connected or primary. Just like any other NLE.

That’s my thinking on this.

[Jeff Markgraf]

Michael and David –

Yes, I understand that Fairlight is track-based. I was looking for comparison of the overlap/layering feature to the “trackless” concept. Turns out it’s not a very good comparison.

I think audio is fundamentally different from video in terms of how we have to work with it. It’s perfectly natural to hear and process multiple audio sources simultaneously, in parallel. One audio track does not by design “cover” or cut out another track. In video, except as a specific key or compositing effect, it does. So audio MUST have discrete tracks, even if they are ultimately collapsed into 2 or 5.1 tracks for presentation.

I think a better way of looking at audio in X to acknowledge its “always on” mix down to a stereo (or 5.1) output. If i wanted to, I could put SOT and nat sound and music and effects on random tracks in Avid and end up with the same stereo output. Not that I’d want to!

So I get the objection that X sort of places clips in a somewhat random order, more so as the timeline gets crowded. And the rules for what goes up and what goes down seem inconsistent (I don’t think they are, but the logic is complex.) It would be nice to have the ability to lock clips in vertical positions in certain cases, if only as an aid to organization.

But as has been said many times, roles and subroles take care of the organization for output (like AAF to ProTools). And the timeline index makes it easy to turn specific roles on and off to see what’s what and where. Customizable colors for roles would be nice.

Since it’s really only upon output to a mixer that the organization traditionally accomplished with tracks is necessary, and it’s taken care of with roles and X2Pro, I guess I don’t see the problem. While I’m editing, I really don’t care too much where a particular clip lives as far as vertical orientation. No matter how high or low a “whoosh” effect appears in the timeline’s vertical space, if it’s attached to the head of a clip to accompany a transition, it will stay with that clip if and when it gets moved. Someone (Charlie?) has suggested/toyed with the idea of creating specific vertical areas to corral certain kinds of clips, using dummy tracks as barriers. That might be a nice option for Apple to figure out. But hardly a deal-breaker.

I’ve got a 3 minute sales reel in front of me with lots of sound and lots of clips. I can pretty easily see my groups of VO, SOT, mx and sfx. Same with the movie I’m cutting in my off hours. In either case, when they go to sweetening, the mixer will have completely organized track groups via roles and X2Pro. And if he’s happy, I’m happy.

[David Lawrence]

[Jeff Markgraf] “I think audio is fundamentally different from video in terms of how we have to work with it. It’s perfectly natural to hear and process multiple audio sources simultaneously, in parallel. One audio track does not by design “cover” or cut out another track. In video, except as a specific key or compositing effect, it does. So audio MUST have discrete tracks, even if they are ultimately collapsed into 2 or 5.1 tracks for presentation.”

It seems to me that’s a narrow way to think about time-based media.

Both audio and video happen in time. Both audio and video can be mixed and processed in various ways. Whether audio or video cuts, covers or mixes is entirely dependent on the type of content being authored. It has nothing to do with the inherent properties of the media itself.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
https://lnkd.in/Cfz92F
facebook.com/dlawrence
twitter.com/dhl
vimeo.com/dlawrence/albums

[Charlie Austin]

[Jeff Markgraf] “Someone (Charlie?) has suggested/toyed with the idea of creating specific vertical areas to corral certain kinds of clips, using dummy tracks as barriers. That might be a nice option for Apple to figure out. But hardly a deal-breaker. “

I’ll sometimes do this to put a “ceiling” above music i’m cutting. I really don’t do that too much anymore though. The timeline behavior is just second nature now, I don’t really notice it…

————————————————————-

~ My FCPX Babbling blog ~
~”It is a poor craftsman who blames his tools.”~
~”The function you just attempted is not yet implemented”~

[Jeff Markgraf]

David –

Well, I didn’t say the two types of media were inherently different. I was talking about the way we typically, and practically, work with them.

By default we mix, or add, audio to other audio. Process it, etc., but not so much hard cutting except for specific effect.

By default, we cut video serially, with some blending (hard or soft) for some specific effect. Yes, video tracks, or layers, often run in parallel, but unless we’re keying or some such thing, it’s usually cuts and dissolves.

Thant’s why I think the UI experience for each is naturally different, to accommodate the different needs in practical use.

[Michael Gissing]

I just think it would be smart to allow an editor to change the display in X to show clips arranged by sub roles and roles to then allow automated audio processing on each, just like a track, bus and mix display on a DAW.

It is after all just a user display and would give back a DAW like organisation and signal processing without changing anything during the edit process. To not do so is basically saying that X is not a finishing tool which seems strange when so much of the approach with X seems to be to finish in the one ecosystem. Otherwise why would they have taken so long to sort xml and not embrace any of the standard interchange formats like OMF, AAF, AES31, EDL etc?