Re: New blog post from Philip Hodgetts. Worth the read.

[Jeremy Garchow] "Show me someone who says that FCPX isn't faster, and I'll show you someone who's spent about 15 minutes with it, or is trying to make it work like other software. FCPX is incredibly fast at organizing huge swaths of data, extremely easily. You can save those sorts for later, or not. I can't tell you how much faster it is, I have never timed it, but from experience, it is simply faster."

I agree! I wasn't comparing the FCPX workflow to the FCP7 workflow; I was just saying that preparation takes time upfront, but makes the process better and faster overall. Editing without reviewing (and logging) the footage is like chopping a tree down with a dull axe. You'll see earlier signs of progress, but it'll take a longer time and more effort to get the tree down.

I'm pro-metadata. It's my third favorite feature in FCPX (after the imaging engine and the skimmer). It's not that useful for me, because I'm not doing creative editorial very often anymore -- I'm finishing and adding effects -- but I would have loved to have had it when I was doing creative editorial every day. I had to rely on tons of subclips and ridiculous naming conventions to get a poor approximation of the sort of metadata that's available today in FCPX. These features would have made my life a lot easier.

Whether we tag clips with great discipline or just try to commit them to memory, the part about actually associating metadata (either formally or informally) is vital to the process and still requires someone to actually watch and log the footage.


[Jeremy Garchow] "FOr instance, a current project I am working on consists of about 30 hours of interviews on two camera (so double that in real time). We have to go through the 30 hours of footage when we receive a paper edit from the client. I would much rather feed it in to a system than have me go through and sort all 30 hours of material."

What are you calling a paper edit? Are we talking about different things?

I used to get paper edits with reels and timecode of selects. I didn't have to sort 30 hours of material to assemble the paper edit; I just opened up the reels, set the in and out point, and added it to the program.


[Jeremy Garchow] "I think you might be misunderstanding, This isn't shot selection. That is my argument. The shot selection comes later, which I still have to watch and organize while creating 1.5 hour paper edit assemblies from 30(60) hours of interviews. FCPX will not help you find the right shot. That is the job of the editor, but as you watch the footage and tag it, it will help you "refind" the right shot later. This time saving is then compounded across the project (which is why it makes it hard to run a stop watch and find out just how much faster it is). And THIS here is another reason why metadata is important (and makes things easier/faster)."

Again, I don't think we're using the term paper edit the same way. The whole point of the paper edits I used to get was to produce a stringout of selects and reduce editorial time by shifting it to production management time.

I agree that metadata pays dividends in re-finding the right shot later in the edit. It's letting computers manage data, which is what they're good at -- it's faster, but it's not new or fundamentally different from the old-fashioned way of writing down timecodes. The difference between searching FCPX and searching a good XLS log or DAM is minimal; the difference between searching anything and not having formal metadata to search in the first place is huge.

There's a lot of metadata that never makes it out of the logger's head and into the notes (whether paper or electronic). Metadata is not a substitute for actually watching and knowing the footage. Knowing the footage isn't a responsibility I'd want to give up as a creative editor.

When I'm finishing someone else's creative edit, I generally don't care -- I rarely even have access to anything but the edit and maybe some handles.


[Jeremy Garchow] "Also, as the project gets older, and more time passes (and perhaps more projects are worked in the meantime) a data search and sort becomes more and more valuable and allows me to reacquaint myself with the footage much more quickly than sorting through a ton of bins/sequences."

Again, agreed -- but managing metadata is not new. The ability to do that directly within the NLE is. I do think this helps us be faster, but I don't think it quarters the amount of time needed to do a project.


[Jeremy Garchow] "Look at it this way. Does tapeless acquisition save you time, or does it simply time shift the acquisition and subsequent archive of the footage? Would you rather go back to tape? Where is saving time important?"

An ironic example. I used to use capture time to get familiar with the footage.


[Jeremy Garchow] "If that also includes giving me more time to be more creative as I'm doing less monkey work, then call it what you want."

The time crunch trend that Philip was referring to goes somewhat along these lines: as soon as you deliver "good enough" in less time than you needed before, you set a new, harder expectation for delivery times. Cutting out monkey work will not free up time for you to do better work for long; it'll shorten the amount of time you're allowed before final delivery is expected.

It's a vicious cycle, and you can't break it from within. The only solution is to step outside the cycle, manage the process, and guide expectations better.

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