Why gaps are great!
Following on from David Lawrence's Position Tool thread where gaps got quite a bit of bashing, I thought I would stand up for them - at least in one respect - by doing a comparison between FCPX, Media Composer and FCP Legacy.
There's an editing task that I am sure everyone has to cope with on a very regular basis and that is opening up a space in the timeline so that new material can be inserted, a task that quickly gets quite messy when you have lots of overlapping video and audio clips at the point where you want to open up your space.
In Media Composer - until the arrival of track tool, borrowed directly from FCP - this was always a pretty clunky experience. You would have to do something like inserting filler across all tracks resulting in destructive behaviour - i.e. overlapping clips would get split in two at the edit point - that you would then have to go in and fix.
A different option that I would use but not many othner editors seemed to favour because it was quite fiddly was to use a complicated asymmetric trim, trying to pick up all the right transitions and were possible trim out filler to make the space you needed. And it sure wasn't quick.
With the advent of the "track tool" it is now possible to select all clips forward from the selection point - which tracks get picked up is determined by the track selectors. It's an OK solution but you still have to make sure you've picked up the right overlapping clips and deselect the ones you don't want to move, in addition to making sure all your tracks selectors are turned on. Also as this is effectively a Segment Tool function, you can't use the number pad to enter a specific offset for your move, you have to drag and watch the offset display, which is awkward.
As in the new Media Composer, you can use the track tool for this with the added advantage that you can type in an offset rather than having to rely on dragging alone. (Incidentally, I notice a lot of people talking about using the TTT shortcut for selecting all tracks forward - but the much quicker way is to use Shift/T.) There is still the issue of having to select and/or deselect clips that you don't want included in the move and with a complex timeline that can be genuinely tiresome, especially if you have to scroll vertically to check what's happening offscreen. Adding to or removing from the selection involves switching back to the arrow tool and using the Cmd key, so quite a lot of extra keystrokes/clicks. It's OK, it works, but it's not the most elegant thing in the world.
Go to the point where you want to open up the space and Insert Gap (Opt/W). Done. Reliably every time. Unless you have some very anomalous clip connections, everything will move or stay put exactly as you'd expect it to. The default gap is 3 seconds but if you want longer you can select the gap, type Ctrl/D and enter a new duration, or you could trim the gap. For me this is a genuinely elegant and efficient improvement over any other way of solving this non-trivial editing task.
Whatever else I think about FCPX - and I have tons of reservations - I really like the simplicity/sophistication of this one aspect.
At the same time, I believe it is based on a very traditional model of how editing works that goes all the way back to film editing. To accomplish this same task with film you would add an equal amount of spacing/slug in between your shots and into your mag tracks in order to push everything along and keep it all in sync.
For me this is a very solid and intuitive model (based on an actual physical process) that I am happy to see explicitly implemented in an NLE. (As I've mentioned before, Media Composer's use of "filler" is very, very close in conception to FCPX's use of gaps but without the added power that comes with the latter implementation.) The idea of gaps having a graphical representation and a "physical" reality is for me a fundamental strength not a weakness.
Note that, despite my provocative headline, I am not actually trying to extrapolate from this one feature to make any other claims about the magnetic timeline. I just wanted to point out that here at least it is providing a really fast, accurate and elegant solution to a long-standing editing issue.
Gaps are great.
The Sony Xpri system we edit on now has something similar. Any bit of black space between clips on a track can be manipulated like a clip, so to create more space in the center of an edit, I can simply ripple one end of the gap and make more space.
Gaps are great.
So are tracks.
FCX. She tempts me, abuses me, beats me up, makes me feel worthless, then in the end she comes around, helps me get my work done, gives me hope and I can't stop thinking about her.
Evening Magazine,Seattle, WA
In my Avid days, I ALWAYS operated with sync locks ON. 99% of the time I use overwrite, and am editing to a VO or music track. And if I insert something, I can't ever recall desiring anything to shift out of place with another track. I would even want my background music track to be split in two so that the start, AND the finish portions are intact. I'll then roll back one half and find a point to marry them back together creatively. Having sync locks on in Avid was the best way to protect from losing sync. With sync locks on you could insert the new clip, automatically splitting ALL tracks. No need to do any filler bs. Yes, you would usually have to repair or delete a few pieces. But it was just rolling them back out over a split edit or marrying split parts back together. Hardly worth fussing about. I also used to begin a sequence by inserting half an hours worth of filler before a frame of empty audio. Since Avid had the growing timeline back then. It gave a nice space to move things around with the move (red arrow) tool. I could also zoom out and marquee a bunch of clips just like FCP or shift T tool, and physically open up a workspace. It's not cumbersome on Avid or FCP classic.
But glad ya like gaps. I'd rather have a source window amongst other things.
[Bret Williams] " In my Avid days, I ALWAYS operated with sync locks ON. 99% of the time"
I do too - I should have made this clear.
Check out this video for an overview of my comparison - I do think FCPX has the edge in this if nothing else.
[Bret Williams] "But glad ya like gaps. I'd rather have a source window amongst other things."
The fact that I like gaps doesn't mean that I don't have very major issues with a huge number of other aspects of FCPX - without OMF for example I simply can't begin to think of using it professionally.
I just think it's useful and interesting to see where the design advantages might lie rather than assuming it's all equally a piece of junk.
I just thought I would follow this up with a video showing off the three different methods:
I should have mentioned that in Media Composer I always work with locked tracks - trying to do this task without locked tracks would be hideous in the extreme.
I think it is clear that the FCPX method wins hands down - unless I've missed a trick with the two otgher NLEs which is no doubt a possibility ;-)
its looks like to me, in the fcp-x one , all your audio went out of sync ? The dialouge, and fx tracks are now with a differrent clip? Surely thats not what you wanted ?
Neil Goodman: Editor of New Media Production - NBC/Universal
[Neil Goodman] "its looks like to me, in the fcp-x one , all your audio went out of sync ? The dialouge, and fx tracks are now with a differrent clip? Surely thats not what you wanted ?"
No, everything stayed exactly where it was meant to. As it would, because connected clips stay connected where they are meant to be. You need to look closely at the clip connections to see what's happening. I have a dialogue clip that bridges the edit point that is attached to a clip after the edit point as I wanted to show how this worked as well.
Maybe I made the example too complicated for it to be immediately obvious how efficiently this works - though in the real world I usually deal with situations a lot more complex than this. But I have tried out lots of different scenarios for this and it behaves entirely predictable and accurately each time.
For me it's a "jawdroppingly" elegant solution to a genuinely complex editing task that the other two NLEs don't come close to matching.
[Simon Ubsdell] "I think it is clear that the FCPX method wins hands down - unless I've missed a trick with the two otgher NLEs which is no doubt a possibility ;-)"
[Simon Ubsdell] "For me it's a "jawdroppingly" elegant solution to a genuinely complex editing task that the other two NLEs don't come close to matching."
I don't know about Avid, but in your FCP7 example, you're working a lot harder than you have to. Here's a better technique:
1) zoom out to reveal your tracks first
2) TTT - make sure you select forward in front of any overlapping clips
3) A - use command-select to select any clips you missed (you can drag to rubber band multiples) or deselect any clips you don't want.
Here's an example:
While the gap technique with FCPX works well for that one particular case, what happens if you wanted some of that dialogue and effects audio to stay put? Good luck with that. "Jawdroppingly" elegant quickly becomes jawdroppingly frustrating.
In my example, I can easily select or deselect anything I want. It's fast, simple, and far more flexible most of the time.
I find the magnetic timeline gets in the way more often than it helps. The track select tools are easy and powerful once you get the hang of them.
I just use ctrl-V to split all my tracks, then "t" or insert edit to move all else over, then I heal as needs be. I can do exactly the same thing in Avid. The truth of the mater is that unless each edit is very self contained, an insert is going to through something off. magic timeline or not. For my work, splitting and healing is generally ideal because it keeps everything in sync. The music will generally need readjustment, but it would require it in the magic timeline as well.
[Chris Harlan] "I just use ctrl-V to split all my tracks, then "t" or insert edit to move all else over, then I heal as needs be. I can do exactly the same thing in Avid. The truth of the mater is that unless each edit is very self contained, an insert is going to through something off. magic timeline or not. For my work, splitting and healing is generally ideal because it keeps everything in sync. The music will generally need readjustment, but it would require it in the magic timeline as well."
Yep, another perfectly useful technique. The bottom line is it's just not that big a deal. Certainly not worth ditching all the benefits of tracks, open space and industry compatibility for the one or two special cases where ripple and collision avoidance are a nice convenience.
I mean we're all pros, right? It's our job to handle that stuff.
[David Lawrence] "The bottom line is it's just not that big a deal. Certainly not worth ditching all the benefits of tracks, open space and industry compatibility for the one or two special cases where ripple and collision avoidance are a nice convenience.
This is the crux right here, right? All of this trackless insanity exists because this kind of edit is slightly more awkward to make than other kinds of edits? That's madness.
[Chris Harlan] "This is the crux right here, right? All of this trackless insanity exists because this kind of edit is slightly more awkward to make than other kinds of edits? That's madness."
Absolutely! This is what I've been saying since day #1. To force 2-million users to abandon what they know and to retool and retrain because of clip collisions on the timeline does not rank as "efficient" where I come from.
Here's a post I made on this subject way back on Aug 5th:
[Geoff Dills] "A huge problem with X is it's NOT intuitive to anyone who has ever edited on any other piece of editing software. You have to rethink, relearn and struggle to use it. It may be more intuitive to a novice, but even there I think it is a steep learning curve to acquire the deep toolset hidden from view on initial contact."
Very good point Geoff. This is in fact a major issue with FCP X that's been overlooked by most, because it's overall efficiency is typically discussed in a vacuum, without any consideration of the costs of both time and dollars retraining the huge base of existing editors to whom traditional track-based timelines are completely intuitive.
We know there were 2-million FCP 7 users, because that's Apple's own number. So, there are 2-million users to factor-in, in terms of time and money for retraining. And, retraining on other NLEs would actually need to be factored into the equation too, as Apple's decision making process ultimately requires retraining for every current user of FCS 3 on one NLE or another.
The bottom line is, if FCP X is actually more efficient for newbies than those with previous training, a cost-benefit analysis would indicate an overall break even point would only be reached when approximately 2-million new users have adopted FCP X.
Will there ever be 2-million totally new users who'll pickup FCP X more efficiently, to balance the scales, (if it is in fact more intuitive for newbies)? If not, then it naturally follows that the overall efficiency of FCP X and the ratio of its overall costs to overall benefits would be suspect.
David Roth Weiss
David Weiss Productions, Inc.
Don't miss my new Creative Cow Podcast: Bringing "The Whale" to the Big Screen:
POST-PRODUCTION WITHOUT THE USUAL INSANITY ™
Creative COW contributing editor and a forum host of the Business & Marketing and Apple Final Cut Pro forums.
[David Roth Weiss] "Will there ever be 2-million totally new users who'll pickup FCP X more efficiently, to balance the scales, (if it is in fact more intuitive for newbies)?"
At $299/seat vs $999/seat, it'll actually take a lot more than 2-million users to make up the difference.
[Chris Harlan] " I just use ctrl-V to split all my tracks, then "t" or insert edit to move all else over, then I heal as needs be. I can do exactly the same thing in Avid."
But not if there are dissolves in the way - same thing applies in both FCP and MC. You'd have to take the extra step of removing the dissolves first. (Incidentally why not map the Ctrl/V function onto just plain V? You'll never look back, especially for ease of adding edits on the fly.)
As I mentioned in my reply to David L., it's not that I don't know how to do this stuff in FCP and MC - there are plenty of ways of approaching it. The point I wanted to make is that for all the scenarios of this that I come up against FCPX is the clear winner. On this one point alone, I hasten to add.
I notice that you like me do quite a bit of complex short form spots/trailers where the audio really stacks up in highly complex ways - this is exactly the kind of situation where the FCPX way of opening gaps would be far preferable. If only we could find ways of coming to terms with all the other currently unworkable nonsense!
[David Lawrence] "1) zoom out to reveal your tracks first
2) TTT - make sure you select forward in front of any overlapping clips
3) A - use command-select to select any clips you missed (you can drag to rubber band multiples) or deselect any clips you don't want."
I'm sorry - I obviously didn't make it clear what I was talking about (certainly not the first time!). What I was describing and illustrating was exactly what you have described and exactly how I would deal with this situation. My focus was actually more on the audio side where things tend to get even messier (more elaborate!) than your video example and where I think FCPX particularly scores the advantange.
Here's why I think the track tool method in FCP is laborious - let's take a typical example of something that I deal with.
To simplify let's have one track of video and 16 tracks of audio (typically I will have more like 24 but I know this is unusual).
Let's also assume that half of those audio tracks "overlap" the selection point and let's assume that half of them have dissolves on at one end.
So here are the keystrokes:
1) Hit T three times (incidentally you can simplify this by hitting T once and then using the Shift key as you make your selection - I find this a lot smoother and quicker).
2) Hit A to start deselecting your tracks.
3) Click eight times to deselect the overlapping clips you don't want included in the selection.
4) Click eight more times to deselect the dissolves.
And you're just about there.
Except that you almost certainly need to zoom out, zoom back in, scroll horizontally and scroll vertically to make sure you've made the appropriate selections/deselections.
I make that well over 20 actions to achieve the desired result in this case. Bear in mind that in my situation there will be more tracks to take care of and more actions.
This being the case I really don't think the one action Opt/W (Insert Gap) FCPX method comes off too badly!
I have yet to try it out over hundreds and hundreds of different edit situations (I can't use FCPX professionally yet for reasons that should be obvious) but I find that this technique does actually work 100% accurately 100% of the time.
The only wrinkle I could see affecting it is if you had some very eccentric clip connections in the first place.
In the case of a music overhang where of course some "healing" is always going to be required whatever method you use, I suppose I would make an edit in the music clip, attach the end section to where I wanted it to end and proceed with the insert gap method as normal.
I hasten to add that I am in no way an FCPX apologist - I do think there are major issues with everything not least the magnetic timeline (but most of all the lack of essential interchange tools that for me are a non-negotiable must-have for which no dodgy third party workaround is going to cut the mustard, now or ever).
That being said, I wanted to applaud the one thing that I have been able to find that would make my day-to-day editing life substantially smoother. (Maybe it's because I've been stuck on Media Composer for the last three weeks that this feature seems incredibly appealing right now.)
Also I'm not saying I have any special difficulty making any of this work in either Media Composer or FCP Legacy, it's just that I resent the work involved and the consequent disruption to the speed the creative process (that crucial thing called "flow"). A few extra seconds is still a few extra seconds - individually they represent a hiccup, collectively they add up to significant loss of time over an editing day and believe me I have to perform this operation a lot!
[David Lawrence] ""Jawdroppingly" elegant quickly becomes jawdroppingly frustrating."
"Jawdropping" in this case was more in the nature of what we Brits like to call "a joke". I'd stick by elegant though, if that's OK with you ;-)