Small jobs and big jobs with FCPX
I've read a lot on here where people have said FCPX is good for small jobs, but not big ones.
Typically a lot of our jobs are 30sec - 4mins. So I would consider these 'small jobs' or are people referring to the amount of footage?
We do a lot of multicam shoots and edit multiple films for clients. Our next job is 7 films from a one day shoot (how to's). Is this what people are referring to? More complex jobs like this?
Or more full length productions? And/or workflow between say AE/Colour etc?
Just trying to get my head around why some people think FCPX is not suitable for 'big jobs'
[Olly Lawer] "Just trying to get my head around why some people think FCPX is not suitable for 'big jobs'
Some of the comments may relate to the relative lack of collaborative workflow on larger projects. There are also some issues if you have projects that are longer than 15-20 minutes, when the system can start to bog down, but you can work round this by editing larger projects in smaller sections.
I've done a few shows with 30-40 hours of source material with no problems.
Workflow for grading in FCPX is better with DaVinci Resolve, Color is EOL so if you were to go with FCPX I would recommend learning it.
Thanks for your perspective, Steve. This is something I'd love to hear more about. I have a job coming up that seems tailor made for FCPX. 26 episodes, very fast turnaround, digital delivery. But it is a 30 minute show. So, is longer than 20 minutes really a wall here? To the extent that I have played with this software I feel it could help me with tight turnaround times. Anybody out there with experience beyond a 15 or so minute length?
And yes, I did read the part about doing longer programs in smaller sections. I'm interested in what happens when you put them together.
I think the question that has yet to be fully resolved, is:
A) Whether you should build small segments as compound clips in the Event browser and then combine them into a single timeline as a Project? ... Or
B) Build sections as individual Projects and combine them there in the Project library as compound clips? ... Or
C) Build sections as individual Projects and combine them there in the Project library by copy & pasting and duplicating?
The concern is whether one of the other is more likely to cause bloat and sluggishness.
For example, in a feature film, would you cut each scene as it's own compound clip in the Event? (Probably creating a dedicated Event just for these sequences.) I'd love to hear whether anyone has had enough seat-time to determine this yet.
Oliver Peters Post Production Services, LLC
Oh, boy. Thanks Oliver. Maybe on the next go round.
So is it the lack of compounding or creating multiple events that bog down FCPX?
What I mean is, when you come to compose the entire edit, does it then slow down? Thus by managing each section by events, your delaying this process until right at the end?
Or would you render out each event/section as prores or uncompressed then bring back into FCPX? Or is that just samantics and it has the same bogging down effect?
[Olly Lawer] "So is it the lack of compounding or creating multiple events that bog down FCPX?"
I'm not sure if I know the answer, but from previous posts, bloat seems to be related to things like markers and other clip-based metadata.
My concern is that the more Projects you have, the slower things are to open, because these have to be buffered into RAM to permit skimming. This is true whether in the Event or Project library. So if you have many, complex, long sequences "exposed", it takes a long time for the app to fully open. This can be mitigated by placing stuff into folders and only opening the one folder containing timelines you want to work on. Or by using Finder "hiding" techniques (or Event Manager) to show/hide pertinent Events and Projects. For example, hiding some of the program sections that you aren't working on at any given time.
[Olly Lawer] "Or would you render out each event/section as prores or uncompressed then bring back into FCPX?"
I would NOT do that if you can avoid it, because you lose all metadata tracking. Of course, if you mean export by "reels" - as in a feature - then yes. Just not as an interim step, unless there are tons of effects and compositing.
BTW - the best way to work is with the least amount of effects applied. Save effects for the end. Leave everything unrendered if at all possible. Let the app render on export (actually the fastest way). You will notice that when Projects load in the Project browser, all effects are first filled and buffered to the "filmstrip" view as NOT rendered, then the app follows up by linking to the render files as the last step in this process.
Oliver Peters Post Production Services, LLC
Thanks for your input!
[Oliver Peters] "My concern is that the more Projects you have, the slower things are to open, because these have to be buffered into RAM to permit skimming. This is true whether in the Event or Project library. So if you have many, complex, long sequences "exposed", it takes a long time for the app to fully open. This can be mitigated by placing stuff into folders and only opening the one folder containing timelines you want to work on. Or by using Finder "hiding" techniques (or Event Manager) to show/hide pertinent Events and Projects. For example, hiding some of the program sections that you aren't working on at any given time."
(Sorry if this is a bit off topic.)
I've brought up a question with Simon a couple of times as to whether or not the amount of 3rd party plug-ins loaded will make a difference with app speed other than the initial boot-up of FCPX (I have a ton in my "Motion Templates" folder). There doesn't seem to be an answer to this yet. Have you done any testing with this or read anywhere if someone has?
If there is a connection to the amount of loaded plug-ins (similar to as you mention with loaded Events and Projects), I suppose quitting and rebooting FCPX and re-naming the "Motion Templates" folder temporarily or moving some plug-ins out would help. My guess is that the plug-ins wouldn't affect anything without being applied, but I really have no idea. If there is any truth to this, then a 3rd party "Plug-in Manager" a la "Event Manager" would be very cool.
[Dave Gage] "Have you done any testing with this or read anywhere if someone has?"
I haven't tested this or heard anything one way or the other. I do know that certain filters and/or combination of filters in FCP 7 can cause instability just by being there. I'm not sure the same structure is there in FCP X that would permit this. At least, so far I haven't noticed anything.
Oliver Peters Post Production Services, LLC
I've been doing local episodic television in segmented talk show format (half-hour) with FCPX for several months. I actually consider these small projects because they are low maintenance. There's not a lot of demand on the color, so long as it's balanced correctly and isn't too hot or cold. Lighting is simple and standard. The audio is run through the Logic compressor to level it out, easy-peasy. Basic lower third IDs. This isn't Emmy winning stuff I'm doing here. So even an hour long one of these I would still consider a small project.
However... I'm thinking of putting together some local Emmy contending episodes in documentary format, and also breaking out to try my hand at short movies. These would fall under big projects because I will be using more tools to create more storytelling. More color grading. More audio (backing tracks or sound fx). More complex titling. VFX, maybe (might do some sic-fi-ish shorts). The shorts may only be 7 minutes, but they'll require more stuff, and I'd call it a 'big' project. For that I'll move over to Adobe's suite for the goodies.
So in my case, when I say big and small projects, it's less about the length of the finished project as it is about how much effort I'm going to put into whatever I do create. Even a dynamo :30 spot could be a big project.
Hope that helps! And good question.
Thanks. Bobby. I understand how shorter can be more complicated. It doesn't look like FCPX is going to work for me at this point. I will be very curious to see how it further develops.
[TImothy Auld] " It doesn't look like FCPX is going to work for me at this point"
A free trial and a copy of 7toX will prove if you can work with the kind of projects you usually do. Just open up the largest one you have handy in FCPX. Cost: $10.
All NLEs (especially FCP7) suffer from some slowdown with larger project sizes. Avid has always done better with them, thanks to it's file and metadata management, but the others have always been much of a muchness.
FCPX is capable of handling more RAM, and should in the future handle larger projects better. For now, I find it quite capable of the kind of work most people do (including most documentary work).
Edit Out Ltd
FCP Editor/Trainer/System Consultant
EVS/VT Supervisor for live broadcast
RED camera transfer/post
[Ben Holmes] "All NLEs (especially FCP7) suffer from some slowdown with larger project sizes. Avid has always done better with them, thanks to it's file and metadata management, but the others have always been much of a muchness."
I did a lot of 2 hour shows in FCP7 (even longer actually), and while it slowed a bit, it doesn't seem to be anywhere near what people report here with FCPX.
Some go these programs had 30 layers, all had color correction, effects, etc. I think we're talking a different thing altogether in FCPX. In it's defense, it does try to load up everything in the known universe, and the database is doing a lot more...
Insight Productions Corp.
"Criticism may not be agreeable, but it is necessary. It fulfills the same function as pain in the human body. It calls attention to an unhealthy state of things." - Winston Churchill
[Mitch Ives] "and the database is doing a lot more..."
Doing more than what? FCP 7? Maybe. More than Media Composer? No way.
In the Avid world, there is a concept called "building the pipes". This basically means that the software has to have complete "look-ahead" forward and backward on the timeline, including mixes, effects, etc. Anytime you make a change this has to be rippled in both directions so that anywhere an editor puts the playhead, the software is ready to play the complete timeline.
On older systems, doing this on a feature film was very cumbersome. Nubus machines were glacial with a 2 hour timeline. There was a console command (later turned into an actual user commend) called "toggle play length". The user could invoke this command plus a number, which limited the length of play. So PL+1 meant you could hit play (fwd or rev) and the timeline would play for 1 minute and stop. Using this made trimming in a 2 hour timeline very fast by comparison and tolerable when working with a director while making adjustments to an entire film. I'm sure all NLEs suffer from this design issue, but it's just that machines have become much faster, so it's far less obvious or show-stopping.
In the case of FCP X, it makes better use of RAM, but the UI design also requires more RAM for the equivalent function in a less "pretty" NLE. As a result on large projects, performance gains could be a wash. Loading an entire FCP X project in the browser and having it be skimmable is equivalent to FCP "legacy" doing its "preparing video display" thing when loading a large timeline.
I also wonder if keywording causes slowdowns. For instance, if I use multiple descriptions as keywords, every instance of this on a clip places a "subclip" into a separate collection. You might have the same clip in several different collections and each of these collections might only be populated with a few clips. In a traditional NLE, this info might not be a "subclip" but rather simply a text description in a user-defined column within the bin. Less taxing on the database. Not a big deal for a basic commercial, but could add up exponentially on a feature film-sized production.
Oliver Peters Post Production Services, LLC
Somewhat related to this topic - how would one solve my problem fast in FCPX:
- I'm going to do 220 (two hundred and twenty) versions of a given film
- What is dynamic in each version is 1 audio-clip at the start of each film as well as 1 video-clip at the end of each edit
Could the above even realistically be done in FCPX? I wouldn't call it a "breeze" to do in After Effects / Final Cut Pro but there won't be any issues solving it there. From my experience it takes a few seconds for FCPX to simply create and open a sequence…
Just a thought on managing larger projects.
I see that FCPX creates and pulls in Events and Projects from two designated folders in the Movie folder (in Lion at least anyway).
After playing around with it for a bit and following some tutorials with some pre-made Events and Projects, I noticed that FCPX slowed down somewhat.
I couldn't see any easy way to load up FCPX without bringing in all the Events and Projects in that folder, so I created two new folders and pulled in the unwanted Events and Projects in finder.
Now when FCPX boots it is much quicker and only brings in the Events and Projects that remain in the two designated folders.
I'm thinking that, for larger projects at least, it could be an idea to create several Events and Projects as I think has been alluded to in this thread, and then pull them out of the default Event and Project folders essentially segmenting them off into different projects as I would with FCP 6.
Of course if all these Events and Projects need to come together to form a final edit, then it would simply be a case of pulling them back into the default folders so FCPX see's them when it loads up.
This seems like a slightly messy method of managing a large project, but one at least that may yield a better performance on FCPX.
Happy to hear other methods tho.