-
Tedious track based editing
-
[Walter Soyka] “This is why I keep blabbering on with Franz about clip relationships …”
Also, thanks for that.
Franz.
-
Jeremy, I’m really surprised here. I’m not criticizing FCPX and I’m not saying that you can’t use it to make the edits you want. I honestly don’t see where we disagree.
I am just trying to explain how it is that FCPX’s editorial is relationship-driven in away that traditional timelines are not.
The FCPX magnetic timeline has a different underlying data structure (parent/child hierarchy) than a traditional timeline (EDL). I’m arguing that both timelines express their data models in the tools they offer the user and the way timeline maneuvers work. I’m arguing that FCPX exposes its internal data model in its external UI. I’m arguing that because clip relationships underpin FCPX’s representation of a timeline, FCPX automatically reflows (and shows you those magnetic preview reflows) every time you make (or consider) an editorial change, and that it does so according to a set of rules that are a consequence of that parent/child data structure. I’m arguing that there is no parallel for this behavior in traditional timelines, because they lack this higher-order construct relating clips back to the beginning of the timeline.
That’s not a value judgment — if anything, I actually think the engineering of this data model is beautiful. I’m not trying to offer an opinion on whether it’s good or bad, or easy or hard — just that it’s different, how it’s different, and why it’s different, in terms of that data model.
What exactly are we arguing about? Do you think that clip relationships don’t drive the magnetic timeline?
[Jeremy Garchow] “Are we (the royal we) really saying that FCP7 has no rules? I’d call BS on that one, I just think that people are so familiar with them, they don’t feel like rules and they have somehow morphed in to…..Legend.”
Not saying that at all. I am saying that FCPX has rules which have no Legendary equivalent (even in ripple mode).
[Jeremy Garchow] “I have to remember to not stomp on my floor too hard today. It is saving me from certain death at every step and I should be more respectful of that fact. The problem is, I’m not sure if I can convince my dog of the same…”
Dude, it’s not just the floor. The crust of the earth prevents you from getting sucked into a hot swirling molten iron core of doom. Unless your canine companion is Clifford the Big Red Dog, I think you’ll be ok. 🙂
[Jeremy Garchow] “Perhaps I am simply misunderstanding. You are saying that I can’t put a clip where I want it because it’s always going to pull to the left? THis is crazy. I would not be here if this is actually how FCPX worked.”
Jeremy, I am not saying that at all. I promise, I fully understand how the position tool works, and I agree that it lets you put a clip wherever you want.
If you want that clip way out on the right somewhere, moving it with the position tool (really maybe more a mode than a tool as you illustrate with your top/tail trim) automatically creates a gap (or gaps) for you that blocks leftward reflow, thereby preserving position. FCPX cannot have a clip hanging off in absolute time, related to nothing; everything must be related back to frame 0 via these blocking objects, be they clips, gaps, or anchors, that come before it.
This bit of engineering is the heart of what what enables all the cool magnetic features. Without this, there is no magnetism. The gaps must exist, and clips’ temporal positions must be relative to their parents or antecedents, in order to satisfy this requirement that is inherent in the data model.
[Jeremy Garchow] “If I remove the offending transition that is 5 seconds down the timeline from where I am trying to ripple trim, certain things move, certain things stay put. So tell me how this represents more exact time than relative time? FCP7 moves certain things that I have no control over unless I correctly select a myriad of connection points, and delete certain offending transitions. All of this is mitigated in FCPX at a glance with the connection points (or parent and child or whatever)”
I’m not saying that absolute time is “more exact” than relative time. I don’t even know what that means. I have said over and over here that FCPX, by virtue of its data structure, defines the relationships between clips as the editor works, and in doing so, eliminate the need for the editor to interpret the relationships by remembering or reading the timeline and then to make complex pre-maneuver selections and modifications before finally making the edit itself.
If you drag some clips, see the preview reflow, and you don’t like what a magnetic move will do, what do you do? You use FCPX’s tools to change the relationships between the clips! You move a clip into or out of the primary. You change a connection point. You use the position tool. Then you attempt the maneuver again. The reflow rules are all still the same, but the redefined relationships result in a different reflow with no complex selection on your part.
I actually think all of this is really cool!
I think the magnetic timeline could be enhanced with David Lawrence’s suggestion of multiple primaries, which would still allow local magnetism/relativity and global absolute positioning if you wanted it, or allow you to work with FCPX exactly as it is today if you only wanted to use a single primary. Really, as he has said, the best of both worlds. Legendary and innovative at the same time.
I also think that FCPX’s unique timeline representation poses some challenges for workflow and interchange, but there’s been a lot of progress on that front in the last year and a quarter, and that has nothing to do with my opinions on how the mechanics of the magnetic timeline actually work anyway.
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
[Franz Bieberkopf] “That’s pretty much my understanding of it. So I guess I understand it. I just have a nagging suspicion that creeps up when people start talking about relative time and absolute time. Maybe I can’t conceptualize the distinction and Einstein was right.”
Let’s drop the words relative and absolute for a moment.
My thesis is that in an FCPX timeline, as you travel up and to the right in a timeline, the clips you encounter have the position in time that they do not as an intrinsic property in terms of a common clock, but rather in terms of the clips that exist before them, down and to the left, all the way back to The Singularity (frame 0).
A traditional timeline has no such construct, and clips’ positions in time are treated by the data model and modified by the tools as intrinsic properties of those clips objects themselves, all in terms of the same common clock.
Am I still spouting crazy talk?
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
[Walter Soyka] “This is why I keep blabbering on with Franz about clip relationships …”
[Franz Bieberkopf] “Also, thanks for that. Franz.”
For clarity, I am the one doing all the blabbering. You are asking insightful questions, and I am delivering 10,000 word dissertations in which I still fail to adequately express my reasoning about how the FCPX data model is directly expressed in the timeline.
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
[Craig Seeman] “I think we’re looking at a few more releases of “catchup” features. At some point they may introduce features that may fit facility needs. Also “maturity” itself is changing as the definition evolves. Actually Adobe Anywhere might be one of the bigger indicators rather than anything by Apple . . . yet.”
Agreed. One of the reasons I continue to pay attention to FCPX is because I’m curious to see how the paradigm evolves. I have doubts about its scalability, but I’d love to be proven wrong. I think the next major release and the approach to multi-channel audio will tell a lot.
Adobe Anywhere really looks like a game-changer. Apple likely also has something up its sleeve, but if Anywhere lives up to its promise, I think Adobe’s use of open standards and ability to work with existing infrastructure will give it a huge advantage.
[Craig Seeman] “Apple often isn’t the first with a feature. They’re revisioning of things usually works. Time will tell. That’s the nature of attrition battles. I think it will come from “young” adaptors maturing along with FCPX itself.”
I’m hard pressed to think of any examples of Apple winning attrition battles. It’s true they often lag behind with features until they get them right (iPhone copy and paste), but historically, that strategy has only worked for them in markets they created and already owned.
Ironically, in 2011, Apple was well positioned to own the NLE market too. But because they decided to take this radical new approach, they’re now starting completely from scratch in a market that’s not only mature, but highly competitive. On top of that, by burning their high end professional users with FCPX’s botched marketing and release, they’ve blown the trust of decision makers who determine what gets used in the future. The damage from that in itself will take years to repair. The bar is high and I don’t see attrition making much of a difference.
Maybe they’ll come up with a future release so mind blowingly awesome everyone will *have* to use it. It’s totally possible, but I’m skeptical. As you say, time will tell.
_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl -
[Walter Soyka] “For clarity, I am the one doing all the blabbering. You are asking insightful questions, and I am delivering 10,000 word dissertations in which I still fail to adequately express my reasoning about how the FCPX data model is directly expressed in the timeline.”
Walter, I think you’re doing an outstanding job describing the FCPX data model and how it impacts and is expressed through the UI.
I’m really enjoying this thread. It reminds me of the good old days of 2011. 🙂
_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl -
[David Lawrence] “Agreed. One of the reasons I continue to pay attention to FCPX is because I’m curious to see how the paradigm evolves. I have doubts about its scalability, but I’d love to be proven wrong. I think the next major release and the approach to multi-channel audio will tell a lot.”
For me, I’m looking at the data management. The Event Project paradigm needs serious management features and I’d expect it to be a bit more than Event Manager X. I can’t help but think that’s going to happen. More generally there’s a lot of metadata that’s there or at least the pieces are there, that aren’t handled well or at all so far. I’d almost say it’s more Bento than FileMaker. In this case though they have potential in the Advanced areas of the Inspector that they simply aren’t using. Also a powerful database has a lot of customization to display that data. FCPX is still rigid in that regard. I do think, given some of the bits, that they will get there though.
Assuming they hit those points (certainly my assumption), I think it will be a very powerful NLE for the “new” facility (whatever qualifies as a facility a couple of years down the road).
[David Lawrence] “I think Adobe’s use of open standards and ability to work with existing infrastructure will give it a huge advantage.”
For the time being. Apple’s not one to simply match somebody else’s functionality. One might guess that all these data centers they’re building is just for iCloud. It maybe be iCloud is to whatever is coming (I think) as is iMovie is to FCPX (although may not take that the way I intend it).
[David Lawrence] “I’m hard pressed to think of any examples of Apple winning attrition battles.”
In many respects this is new to them but they are making changes to their business model in some areas recently. I think this is what they anticipate though.
[David Lawrence] “burning their high end professional users with FCPX’s botched marketing and release,”
That’s why I think it’ll be attrition. It’s a much easier sell to those who haven’t been burnt. There’s a lot of scare tissue.
[David Lawrence] “Maybe they’ll come up with a future release so mind blowingly awesome everyone will *have* to use it.”
All it needs for a beachhead is to be compelling in a couple of areas to be at least one tool in the toolbox. At $300 if it’s a great “plugin” it’ll be in the tool chest. Then it’ll take some time to exploit the beachhead.
-
[Walter Soyka] “Am I still spouting crazy talk?”
Makes sense to me.
Franz, here’s how I think if it:
Absolute time = fixed external temporal frame of reference = the window space — all media objects placed in this window are fully independent of one another in space. Objects can be placed anywhere. Empty space is empty.
Relative time = internal temporal frame of reference = the primary storyline in FCPX — all media objects must be connected to or contained by the primary. Editing the primary storyline by default changes the spatial and temporal positioning of connected objects. The position tool can be used to override this behavior by automatically inserting gap objects to hold space. Space is never empty. Space must always be filled with an object because the magnetic timeline always pulls objects to time zero on the left.
This leads to weird UI feedback, like Aindreas mentioned. The visual feedback’s correct when you think about what the timeline is doing. But the premise of the timeline itself is debatable.
_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl -
[Walter Soyka] “What exactly are we arguing about?”
Gravity.
[Walter Soyka] “What exactly are we arguing about? Do you think that clip relationships don’t drive the magnetic timeline?”
Clip relationships drive every single edit, FCPX or not. In my FCP7 screengrabs, I seriously messed up the relationship of my clips by ripple trimming that clip. That wouldn’t happen in FCPX (I mean, of course it could happen, but not as easily).
[Walter Soyka] “If you want that clip way out on the right somewhere, moving it with the position tool (really maybe more a mode than a tool as you illustrate with your top/tail trim) automatically creates a gap (or gaps) for you that blocks leftward reflow, thereby preserving position. FCPX cannot have a clip hanging off in absolute time, related to nothing; everything must be related back to frame 0 via these blocking objects, be they clips, gaps, or anchors, that come before it. “
OK.
I guess I think you are trying to point this out as some sort of a problem. I see what you’re saying now. This happens in FCP7 too, it’s just that the relationships are less clear and they are “implied” by the edit, and when the timeline gets more complex, you have to constantly remind yourself exactly what those relationships are. FCPX shows you those relationships (most of the time) and also gives you control if you’d like it (some of the time, audio needs more help).
I never need a clip hanging off in “absolute time related to nothing”. I don’t understand this metaphor. David Lawrence uses it a lot, too. I am sure it is my fault.
A clip is always involved in some way with something in the edit. At the very least, it is related to time. X allows for this.
[Walter Soyka] “If you drag some clips, see the preview reflow, and you don’t like what a magnetic move will do, what do you do? You use FCPX’s tools to change the relationships between the clips! You move a clip into or out of the primary. You change a connection point. You use the position tool. Then you attempt the maneuver again. The reflow rules are all still the same, but the redefined relationships result in a different reflow with no complex selection on your part.”
But that is the very nature of editing. An edit creates relationships all over the place.
There are many things that you simply don’t have to do in X that you do in 7.
You sometimes have to move clips up/down tracks, you have to lock certain things in place while moving others, you have to make room by selecting/deselecting certain clips and not others, you have to delete transitions to make a ripple, you have to select a video/audio target tracks before adding it to the timeline, you have to select the exact right number if clip edges to avoid some sort of collision. You are constantly redefining relationships in FCP7 as well.
From you: “While nothing will happen in a traditional timeline that the user doesn’t explicitly order, the FCPX magnetic timeline has a system of rules that guide operations during timeline maneuvers.”
So does every timeline everywhere. There are always rules. Plenty of crap happens in a “traditional timeline” that isn’t “ordered”.
-
[David Lawrence] “The visual feedback’s correct when you think about what the timeline is doing. But the premise of the timeline itself is debatable.”
David,
I’ll have to remember this line as a delicate way of expressing myself when I don’t like a cut.
[David Lawrence] “Franz, here’s how I think if it …”
[Walter Soyka] “Let’s drop the words relative and absolute for a moment.”
So,
I think the thing is that I’ve always had some sort of timecode-based “primary storyline” (yeesh – that term) in my head when I work with timelines. So the concept has always been there and now Apple have taken it, emptied it of any sort of preloaded (useful) template, and asked editors to fill it as they begin. If that really is all there is to it, then I’ve got it. In fact, I’ve always had it.
Relative Time is Absolute Time without the Time.
Franz.
Reply to this Discussion! Login or Sign Up