After Apple released the 10.0.1 update for FCP X I decided to take the application for another spin, and actually edit something that I would normally cut in Final Cut Pro v.7. This wasn’t an actual paying gig, but more of a simulated “real” session that closely represents the typical job that crosses my desk. While this is no way a comprehensive review, I thought I’d share my initial impressions.
First, a bit of background. I installed 10.0.1 on an 8 core Mac Pro that is connected to Xsan shared storage and has the Blackmagic Decklink Extreme 3D card installed. That last point is somewhat moot considering FCP X does not yet support external monitoring in any way (Apple claims that’s coming in “early 2012”), nor can I capture from tape without using FCP 7 or a third party utility. The new FCP release does include official support for Xsan volumes, which is a very welcome addition. The project I built was a 30 second “image” promo for a major network that I’ve done a lot of work for over the years. I used elements from their promo graphics package along with show clips, music, sound effects, and generated text. The fake spot highlights the new fall lineup.
After first launching FCP and importing media, I noticed how fluid and fast the application feels. Clips scrubbed quickly, the playback was responsive, and there was no lag that I could detect. Opening and closing events felt instantaneous, and the window of clip thumbnails loaded very fast. Apple has touted the lack of render bar as a headline feature (it’s been replaced by what looks like a render “dial” in the dashboard in the center of the screen). The point is that you never have to stop down to wait while FCP takes over and processes filters and effects. The system keeps its responsive feel by only rendering when the mouse pointer is idle- you can keep an eye on this with the Background Tasks window.
Speaking of rendering, a huge hurdle has been cleared by this entirely new code base: true multicore rendering on everything I threw at it. Editors who work in FCP 7 and keep an eye on the Activity Monitor know that certain rendering tasks barely use more than one core, which is a shame when you’ve dropped some serious cash on a Mac Pro. FCP X handles this much more efficiently.
Oh and there is no Save command under the File menu. You don’t save. If you quit the application, it doesn’t ask you if you want to save your changes. And if the application crashes (as it did twice this week for me) you relaunch and pick up right where you left off. FCP X tracks and saves your every move, and I never lost a single edit all week. It’s a pretty liberating feeling, actually. Editors on any platform have been trained to keep fingers near the Command and S keys and use them frequently to avoid losing work. Apple doesn’t even give you that option. Combine that with an unlimited undo queue and you’re talking about some serious benefits.
In terms of actual editing, Apple has upended much of what an experienced editor is accustomed to. The traditional insert and overwrite edit functions have taken a back seat to “append” and “connect” commands, which correspond to a “storyline” instead of a traditional track based timeline. The overwrite function doesn’t even get a visual button onscreen, which I found somewhat shocking considering that’s the most common type of edit in every earlier iteration of Final Cut Pro. I started to grow accustomed to this after some practice, and for quick stringouts of shots I found myself skimming, marking, and editing clips together pretty quickly. The append edit is nice in that you don’t need to worry about where your playhead is in the timeline- it will always tack the shot onto the end. I’m still wrapping my brain around the concept of the primary storyline and connected clips, so the jury is out on that until I can really cut something of substance.
I’ll round out the “good” section with two items that are new but I haven’t had reason to test yet: XML I/O and audio stem support. We had good reason to believe that XML was on its way to FCP, and the door is now open for third parties to support OMF, EDL, and interoperability with color correction tools like DaVinci’s Resolve. It’s worth noting, however, that the XML language used in v.10 of Final Cut Pro is not the same as the one used in earlier versions. Philip Hodgetts can do a much better job of explaining why in his blog post. As for audio stems, it’s a useful tool as long as you are mindful of metatagging media on the front end. You can export separate audio files for voiceover, music, sound effects, etc. for delivery to audio mix or broadcasters that require it in a more automated way with stems support. FCP X will attempt to figure out what an audio item is during import, but otherwise it’s up to the editor to make sure it’s all properly tagged. I’m not convinced this is better or more productive than the “old school” way of laying out your tracks in a traditional timeline.
As you’re probably aware by now, the entire notion of what a “project” is in Final Cut Pro X is radically different than before. Your project file is no longer a container that references all your sequences, bins, and clips for a given project, but rather a sequence. One sequence. That’s it- you start a new project, and that project is just the one timeline you’re working on. Want to duplicate your sequence to make an alternate cut? New project. Need to work on a series of four promos for one show? Four projects, times the number of versions of each cut, each with a unique name, and I presume unique render destination. Multiply this out amongst a host of edit suites tied into an Xsan and working with dozens (if not hundreds) of clients in a year and this starts to look like an organizational nightmare.
FCP X’s handling of source media faces similar issues. As I mentioned before, bins are out and events are in, and these events live in a location of the editor’s choosing on local or networked drives. You tag and store media by keyword in what’s essentially one big pot of catalogued events. That’s all fine if you’re sitting at home cutting your independent documentary with all your project media sitting on one drive, but with multiple rooms and multiple projects concurrently in progress on shared storage and your brain starts to implode. Additionally, there are many occasions where I’m working with a client and I don’t want them to see or have access to media from another session I worked on the day before. It’s more than an inconvenience; it’s a potential minefield with disastrous consequences when you’re dealing with clients who are competitors with each other.
Another big issue for me is the use of UI eye candy that looks good in a demo, but can get old fast in day to day work. The timeline layout in FCP X is an exercise in wasted space. I don’t have a need for big rounded bubble-like clips, and even on a large Cinema Display I have to start scrolling up and down to see my storylines and connected clips. Setting the clips to their smallest settings helps a little, but it still feels inefficient. Some would argue that I could collapse clips together when I start to get a timeline that’s too tall, but that introduces another setback from a visual standpoint, especially with audio. If you have split audio edits where the audio track from clip B starts under clip A and you collapse them into a storyline, the edit point in the timeline reflects only video edit- the audio edit is hidden. Not good, and the more complex the timeline the greater the need to collapse and hide that important information.
Speaking of audio, there is in my opinion a need for more control than what FCP X provides. When making any type of edit (append, connect, overwrite, insert) you’re given the option of All, Audio Only, or Video Only. It’s an all or nothing affair. What if I have a source clip with four channels of audio and I only want the one track from the boom mic? I suppose I have to cut them all into the timeline and then go in and turn off the three unwanted tracks. That would get old fast when dealing with hundreds of clips.
Overall, it sure feels like one of Apple’s main objectives when designing FCP X was to save you from yourself. The number of options available to you in FCP 7 and earlier could be seen as overwhelming, but I thought Apple did an admirable job of streamlining most workflows with things like Easy Setups that grouped the particular nuts and bolts settings together in selectable groups. My hands feel tied in FCP X, at least in its current form.
I can’t, however, deny the powerful potential in this software. There really is a lot to like from a technical standpoint, and certain aspects of the workflow are really compelling. A few times this week I found myself wishing that I was able use X in my current job for the speed that it provides in quickly skimming and locating good source clips, as well as the powerful rendering engine that would greatly cut down on the time I spend waiting for the machine to catch up.
I’ve said this before, but FCP X is not finished. It feels like a stable beta release, but one where many features are missing or incomplete. If I was forced to make a decision today to switch from FCP 7 to something else, the answer would be clear: X is not an option. But a year from now? Two years from now? Things could be entirely different. The question is whether or not the pro editing community can hold on that long.