[David Lawrence]

[David Roth Weiss] “I’m not yet sure what’s different or what’s the same. All I know at this point is both are powered by the same engine, and mysteriously Soundbite appeared ever so shortly after GET disappeared.”

Now if they can just get it working in reverse to generate accurate transcripts…

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[David Roth Weiss] “Let’s just say it’s a round #2 game changer. GET, which is Boris Soundbite’s no long available predecessor, changed that game well over a year prior using the exact same speech recognition technology. “

Absolutely! A friend in LA was a Beta Tester for GET so I learned about it a while ago, but missed the window when it was being sold by AV3. So glad it’s finally back on the market. I really could have used it last month.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Carsten Orlt] “Thought this was a brilliant idea to show what tracks really are. just a bunch of story lines hard connected to time.”

Carsten — Thanks for posting this clip. This illustrates an idea that Simon Ubsdell and I discussed in the Techniques forum here:

https://forums.creativecow.net/readpost/344/3938

One big problem I have with FCPX is the constraint of a single “primary” storyline. If we could have any number of user created storylines, that maintain a fixed relationship to time in the window space, then many of my problems in FCPX go away. I think the idea of user-definable virtual tracks that are media-type agnostic would be very powerful and interesting.

My other big issue is gaps and fixed ripple mode. I believe the position tool is a workaround, not a true solution. I’ll explain why later in another post.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeremy Garchow] “FCPXML? :-D”

LOL! not yet, but eventually, right?

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

Bravo Jeremy!

Thank you for this detailed and thoughtful post. There’s actually a lot we agree on, and a couple key places where we diverge. I’ll need some time to gather my thoughts and maybe a screen grab so hang tight. I really enjoy and appreciate this dialogue.

best,
David

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

No worries, Jeremy, take as much time as you need. I appreciate your thought and effort. It’s not like writing COW posts is the only thing we have to do all day 😉

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

Awesome example, Aindreas. Thank you!

[Aindreas Gallagher] “ultimately, on a personal level, FCPX feels like very good engineering metadata/database thought and very poor editing practise workflow thought.”

Agree 100%. I’ve been saying for some time now that FCPX’s metadata/database tools are excellent and I’d love to use them in my logging workflow. For those who missed it, here’s what I proposed in the FCPX Techniques forum.

I have no problem with skimming and keyword collections when I’m logging. But when it’s time to start cutting, I’m fully in Aindreas’ camp. I want to see bins and sequences as objects that I can spatially organize any way I want on the monitor.

Keyword collections and bins may have functional similarities, but keywords are words. Bins are objects. We think about words and objects in entirely different ways. As I read the comments, it seems that the debate here is between people who think abstractly, in terms of words, and people who think spatially, in terms of objects.

I think visually and spatially, hence the title of this topic. My editorial workflow efficiency depends on being able to very quickly take in a complete, holistic overview of my entire source, at any time. That includes bins and sequences I use as bins. Color coding, grouping random piles of objects and working in space; all of this is essential to how I work.

BTW, I’ve never said you can’t build a project like my example in FCPX. I’ve done it. My issue is that the constraints of the magnetic timeline make the process inefficient.

Another thing to notice about Aindreas’ screen shot — Typography matters in UI design. Look at how many lines of text appear on his screen. Because the text is rendered in a small, single pixel thick font, it’s possible to view significantly more information in a single glance than you can in FCPX. In fact, because the FCPX UI designers chose large, bold text, you see less than half the information on the screen, both horizontally and vertically.

FCPX is filled with typographic and graphic design decisions that make is visually harder to find information. What possible professional benefit is there to these visual design decisions?

This is great you guys. Please keep posting screenshots to demonstrate your workflows.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeremy Garchow] “I understand what’s in 2 but i cant see it are, so to clarify, are those just the little dialogue edits with overlaps (checkerboard)? Looks messed up, but sounds great? :). And 3 is a few or more left overs?”

Here’s a couple close ups so you can see what’s going on:

Section 2 (marked with In and Out points) is the edited segment
Section 3 is the scratch area. This area is where I experiment as I build in section 2

Zooming in closer:

This is the zoom level I typically work at for fine cutting. I have dedicated buttons on my mouse that make zooming in and out super fast.

Some things to note:

I use volume curves to get subframe accuracy with audio transitions. I frequently use two-frame dissolves as well. You’ll see one at 3:33:14. I create them with a single click and can easily set them to start, center or end of edit. I find these tools to be more accurate and responsive than the tools in FCPX. Technically, FCPX may be more accurate, but the gratuitous UI chrome and animation makes it feel sloppy. BTW, did you know you can’t do a two-frame dissolve in FCPX? Three is the minimum. (feature request?)

Using these tools and techniques, I create seamless audio right in the timeline. I rarely need to make any edit adjustments by the time I go to Soundtrack Pro for sweetening.

The second track is room tone.

Maybe I’m not being clear about what I mean when I say “checkerboarding”. I’m referring to a process, not necessarily how the timeline looks. Checkerboarding is actually an old-school term that refers to CMX linear days. The CMX operator would program in the EDL and because the tape decks were computer-controlled and frame accurate, it was possible to “checkerboard” all the material from the source tape to exactly where it needed to be on the record side. This made assembly efficient because you never needed to swap back and forth between source reels. When I was at LucasFilm, we made laserdiscs that were used as frame stores for database-driven multimedia projects. Each of the 54,000 frames was either a single-frame edit or a segment of moving stills. We used checkerboading a lot.

In the context of NLEs when I talk about checkerboard workflow, what I mean is the process of building a sequence in a totally random access manner. Starting at the end, middle or beginning; working backwards or from the center. It doesn’t matter, it’s completely fluid because that’s the nature of the digital medium. NLEs made checkerboarding a completely natural way to work in video.

[Jeremy Garchow] “and you can do this in FCPX before you hit the timeline. With the favorites sorted you can make a compound clip (in the browser) of just the favorites and make one clip if you want to, and that clip is then editable in to the timeline, and can be broken apart at anytime in the timeline (or opened in another timeline, and necessary parts copied and pasted). To this compound clip, you can add/remove/reorder whatever you want, including more layers of video/broll/audio.”

See my response to Jeff regarding favorites. I’m not sure this would work with my method because I don’t know what I want until I’m in the process. Agree about compound clips, but I think jumping in and out of context would break my flow and slow me down. I feel like my efficiency depends on having everything available all at the same time in the same place. I’ll give it a try and let you know.

[Jeremy Garchow] “Alternatively, keeping the favorites as separate clips in the browser, you can easily skim to the next clip. Skimming, is awesome. I didn’t like it at first, seemed a bit out of control, but I learned to control it.”

I like skimming 🙂

[Jeremy Garchow] “I will craft a detailed response when I’m in front of FCPX and can provide more screen grabs. “

Looking forward!

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeff Folland] “Go through clip by clip like you would in the timeline. Set ins and outs and mark sections as favorites and tag them with various tags like ‘must use’. ‘interesting’ etc. Think of the tags as your markers.”

FCPX’s metadata and tagging features are one of its greatest strengths. They’re so interesting in fact, that I think FCPX would make an excellent logging and preflight assembly tool. If there were some XML translation utility, I’d be using it today. I wrote about a logging workflow like you describe here in the FCPX Techniques forum:

https://forums.creativecow.net/readpost/344/4418

[Jeff Folland] ” It’s like you already have a rough edit of the must use stuff in the event window.”

It’s not even close. I don’t work with a script, I make everything up as I go along. I may know something is “must have” but I have no idea how or where I’ll use it until I start building the piece. And as I build, I may change my mind several times. I never know what the finished piece will look like until it’s done. It’s a creative, organic process. The process is what defines the workflow.

[Jeff Folland] “You could still use a gap clip to create a scratch section where you shuffle stuff around. “

You’re still in ripple mode. Only now you have keep track of not just the stuff you’re cutting, but also everything in the space in-between. How is that a better way to work?

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Michael Gissing] “David, this edit methodology is similar to how I approach audio editing on a DAW with the difference that I have source at the end and the edit is at the beginning. You method however, allows for ripple edit which I don’t use. “

Michael —

I’m not surprised to hear this. I think understanding audio-centric workflow helps make the constraints of the magnetic timeline obvious. For editors who work like I do, it’s inefficient for anything beyond the most basic editorial tasks.

You’re correct that the reason I keep source at the front and edit from the end is because of ripple. I’ll often shift-delete gap space as I’m cutting down dialogue. The important thing is ripple is a tool. I only use it when I need it. It’s not the permanent state of my timeline workspace.

Imagine cutting dialogue or music on a timeline that’s in fixed ripple mode — the only way to hold time apart being with explicitly-placed spacer objects. Audio people would never accept this. I think it’s a deal-breaker for most advanced video people as well.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl