FCP-X: Thinking Differently?

[Jeremy Garchow]

[Chris Harlan] “I’m sorry. How can this be confusing to you?”

Yeah, it’s not. You select the audio and tell it what you need them to be (or what channel they should be on, or channels they should be on).

I don’t know, I have been working with the “pool of media” instead of compartments of media idea for a long time now, so assigning metadata to things to define them doesn’t scare me.

[Andrew Richards]

[Herb Sevush] “Which works fine when building consumer oriented devices, but does not work fine when building specialized commercial software. “

According to Chris, they would have followed that philosophy for the entire lifespan of FCP as well though.

Best,
Andy

[Andrew Richards]

Be careful. That word is the third rail in these parts.

Best,
Andy

[Geoff Dills]

Don’t I know. 🙂

Best,
Geoff

[Herb Sevush]

Yes they would have and they did, leading to the situation now where they just dumped a category leading Ap to replace it with … something else.

Did they hit the sweet spot for a few years — yes; but even a blind squirrel … etc.

I am beginning to think that X is to FCP7 as Motion is to Shake.

With Motion they were all a flutter with how it was going to be revolutionary with it’s use of “behaviors” which were going to take the drudgery out of compositing. Simplifying post for everyone.

It was different, it was novel, it wasn’t too bad, and I use it to this day, although it ushered in no revolution and is a huge step back from Shake. And the stupidest thing about it are the dammed “behaviors” which any professional compositor could have told them was moronic. But then who would want professional feedback?

Herb Sevush
Zebra Productions

[Michael Hancock]

[Andrew Richards] “3-5) This where metadata shines. Your track assignments in a fixed track timeline are immutable. With roles assigned as metadata for each audio clip, you get to assign output based on role and not have to put any effort into maneuvering your clips into different tracks to meet an output requirement. At least that’s how Apple stated it would work in their FAQ. Further, you can quickly select and disable audio channels by searching the timeline using the same metadata. “

You don’t have to put output into patching your audio to the proper tracks, no, but now all of your effort goes into properly tagging your audio so you can use the metadata accurately. Seems like it’s just trading one load of work and necessity for detail for another.

I wonder, too, what happens when you have a piece of audio that can be categorized as both Music and SFX, or VO and NatSound? Do you do a general assignment, then go through the timeline and change the metadata on a clip by clip basis? Is this even allowed? With tracks, you just put it on the right one to assume that “role” in your sequence.

Here’s what I would rather see – tracks + metadata with track patching based on that metadata. So you can assign Tracks 1 – 4 as Dialogue. If you choose, you patch manually. Or, you assign your audio with Dialogue to auto-patch to 1-4. Now you can cut freely and your audio will automatically patch. If you need to use part of that dialogue as a sound effect you can simply hardpatch it or move it down to the SFX track. The metadata auto-updates. This way you can still search your sequence but have a visual, easily read timeline of what exactly is VO, NatSound, SFX, Music, etc… The best of both worlds, and it makes editorial faster because you aren’t messing around with patching tracks – you assign it via metadata. But it still keeps things organized and fits current workflows.

—————-
Michael Hancock
Editor

[Andrew Richards]

[Herb Sevush] “I am beginning to think that X is to FCP7 as Motion is to Shake. “

That’s a solid parallel you’ve drawn there. I too remember the giddiness about how they thought behaviors made keyframes obsolete. Yeah, not so much…

Best,
Andy

[David Lawrence]

[Andrew Richards] “1) Isn’t this a perfect use case for compound clips? You cluster your audio into a nest and it acts as a unit, but you can explode it to make fine adjustments at will. “

The problem with compound clips is that they rob you of your piece’s overall context. Any adjustments must be made inside their own little island. If you break them apart to see context, you lose any transitions.

[Andrew Richards] “3-5) This where metadata shines. Your track assignments in a fixed track timeline are immutable. With roles assigned as metadata for each audio clip, you get to assign output based on role and not have to put any effort into maneuvering your clips into different tracks to meet an output requirement. At least that’s how Apple stated it would work in their FAQ. Further, you can quickly select and disable audio channels by searching the timeline using the same metadata. “

It’s fantastic in the abstract and would be a wonderful as an additional tool in the usual kit. But I don’t believe it’s a replacement for the flexibility of the spatial model. Especially for audio. I do a ton of multitrack mixing in FCP as part of my normal workflow.

Question: How much work have you done with Soundtrack Pro or Pro Tools? Do you think a trackless magnetic timeline model would be more efficient for these applications?

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

[Timothy Auld]

And who comes up with the naming conventions? Would the metadata tags be dialog 1, dialog 2, sfx 1, music 1. And if so then what’s the difference from what we do now? This is a serious question. I know
there are a fair number of intelligent people out there who excited about this but I don’t understand how
it would work. And I would like to.

bigpine

[Jeremy Garchow]

[Michael Hancock] “You don’t have to put output into patching your audio to the proper tracks, no, but now all of your effort goes into properly tagging your audio so you can use the metadata accurately. Seems like it’s just trading one load of work and necessity for detail for another.”

Yeah, but you don’t have to do this right away. You can bring in the footage and edit. You can then tag your footage later. you can use separate metadata in the timeline, and it won’t effect the original names of the clips. It’s fall down easy, and so fast.

[Michael Hancock] “Do you do a general assignment, then go through the timeline and change the metadata on a clip by clip basis? “

Sure. This is the beauty. THat clip doesn’t have to be either or, it can be both. Think of a submix through metadata alone. Sweet.