Creative Communities of the World Forums

The peer to peer support community for media production professionals.

  • Oliver Peters

    February 17, 2016 at 7:50 pm

    [Scott Witthaus] “A good reason to limit as much as possible….”

    What does that mean? You know in the normal production world and multiple client needs you generally have little control over these things.

    – Oliver

    Oliver Peters Post Production Services, LLC
    Orlando, FL
    http://www.oliverpeters.com

  • Oliver Peters

    February 17, 2016 at 7:52 pm

    [Jeremy Garchow] “If we added the proper data, the FCPXML worked just fine.”

    This isn’t Resolve and I’m not questioning the viability of FCPXML. Just that sometimes things don’t work and there needs to be a better solution than what there is now.

    – Oliver

    Oliver Peters Post Production Services, LLC
    Orlando, FL
    http://www.oliverpeters.com

  • Jeremy Garchow

    February 17, 2016 at 8:17 pm

    [Oliver Peters] “This isn’t Resolve”

    Then what is it?

  • Simon Ubsdell

    February 17, 2016 at 8:57 pm

    [Oliver Peters] “FCP 7 was far easier to work with than FCPX.”

    FCP 7’s brute force relinking was what made it one of the most powerful NLEs ever – if that’s what you needed.

    Simon Ubsdell
    tokyo-uk.com

  • Oliver Peters

    February 17, 2016 at 9:00 pm

    [Simon Ubsdell] “FCP 7’s brute force relinking was what made it one of the most powerful NLEs ever”

    Agreed. A versatile conform engine.

    – Oliver

    Oliver Peters Post Production Services, LLC
    Orlando, FL
    http://www.oliverpeters.com

  • Simon Ubsdell

    February 17, 2016 at 9:07 pm

    [Oliver Peters] “Agreed. A versatile conform engine.”

    Some very well-informed guy wrote this really excellent piece on “The NLE that wouldn’t die” which is well worth the read!

    Simon Ubsdell
    tokyo-uk.com

  • Andrew Kimery

    February 17, 2016 at 9:24 pm

    [Oliver Peters] “Agreed. A versatile conform engine.”

    FCP Legend certainly gave you enough rope to hang yourself with but if you knew what you were doing the amount of flexibility and options it offered was amazing.

  • Jeremy Garchow

    February 17, 2016 at 11:57 pm

    [Bill Davis] “but the IDEA that the pipe you cut for the project that needs to be 18 inches is suddenly 17 – and the system has to accommodate varying pipe lengths everywhere – well that’s a mess, isn’t it?”

    Is it? Aren’t databases designed to easily track and make sense of a big mess?

    Ranges are just points, and timecode is frame by frame metadata.

    Walter Soyka likes to point out, “The data model is the core of an application and it defines the limits of an application’s extensibility.” What does that mean?

    That means, in the case of FCPX, the “exposed” data model is FCPXML. You can export an FCPXML, and find a TON of information, and you can then use that XML to build/rebuild an Event, a timeline, a Library, a Project with all kinds of metadata attached to all of it.

    A database tracks records or entries, and clips are nothing but an entry or record. I can change the name or add different metadata to a clip in a Project, and it has a different name/metadata in the Browser, yet FCPX can track that record and the both/more sets data attached to it as every clip has a unique ID assigned by FCPX.

    In the case of FCPX, it seems like (and perhaps I am woefully wrong), just by looking at FCPXML we should be able to add new entires/records, and update a Project (point the “database” at new media/IDs, and say use “use that from this time to this time”). Similar to how I can add different metadata to an instance of the clip in the Browser or the Project, I should be able to import new records, point a Project to those records, find similar metadata (like timecode) and have the database rewrite the pointer to use that media. This leaves the original clip, metadata, Projects alone, I am only wanting to change one specific Project, not all the Projects or all the instances of the clip in the Library.

    In the case of needing relinking, It would be the case of making a new Project, unlinking the video of selected clips (basically, rewriting the database pointers to “empty”), adding new records, and telling FCPX to use the audio from the original clips, and the video from the newly rendered clips.

    FCPX already has this capability to read multiple atoms of a clip. If you export an FCXML of a multiclip, you will see that the data container lists every clip in the multiclip, even though I only have one angle, and perhaps only one channel of audio enabled in the Project. FCPX can already see multiple video components and multiple audio components from different clips, and display them as one clip in a timeline. Relinking a video pointer to a different piece of media should be as “simple” as literally updating the “database” entries or records.

    For whatever reason, the programming (or the set of instructions) to carry out these functions are not in FCPX, and I don’t know why. I am sure there is a great reason for it, or maybe there is not a great reason for it. I don’t really know. I do know that by observing what the FCPX data model can already do, relinking to a different video channel should be possible, it will just take a little programming to write all the variables and account for the unforeseen. Granted, none of this easy.

    I don’t know if you’ve ever relinked anything in Resolve (like relink Proxies to full res clips, or ungraded to graded) but the easiest way to do this is to add the full res clips to the media pool first. You import a database (another word for XML/EDL/Whatever) and you can define a set of parameters for Resolve to search for and relink the media. You can tell it to ignore file extensions (so you can change .mov to .R3D, or .mxf to .mov), but look for common items between the incoming database, and the existing database, like timecode (which is extremely useful and highly accurate metadata).

    You can then export the data of an FCPXML, import that into FCPX, and FCPX is pointed at all the correct media.

    I routinely get graded clips from Baselight, and have to jump through a lot of hoops to get back to FCPX with the newly graded media. I have lately been using Premiere for conform (So I leave FCPX, translate the timeline, import to Pr, watch the program in Pr and fix any problems, export an XML for Baselight, receive graded media, relink to the new video, watch the program to make sure it’s right, export an XML and translate to FCPXML, import and watch to make sure it’s right, fix any problems and output), but would like to stop doing that as it seems like it’s a crap load of unnecessary steps to go through all of that when I could simply just relink the media in FCPX on the way back (and hope that baselight hops on the FCPXML bandwagon).

    Anyway, sorry if this is boring, but pointing a database to new entires should be possible, even with FCPX.

    Thanks for reading,

    Jeremy

  • Michael Hadley

    February 18, 2016 at 12:13 am

    Totally agree.

    I did notice recently that I was able to relink a stock music WAV file with it’s .MP3 demo version. That seemed–don’t recall it could do that before. (But maybe it could).

    But yes, force relink would be fantastic.

  • Oliver Peters

    February 18, 2016 at 12:13 am

    [Jeremy Garchow] “I routinely get graded clips from Baselight, and have to jump through a lot of hoops to get back to FCPX with the newly graded media. …… …….but would like to stop doing that as it seems like it’s a crap load of unnecessary steps to go through all of that when I could simply just relink the media in FCPX on the way back (and hope that baselight hops on the FCPXML bandwagon).”

    And none of these steps were usually needed with FCP7, which is basically what Simon has been saying.

    – Oliver

    Oliver Peters Post Production Services, LLC
    Orlando, FL
    http://www.oliverpeters.com

Page 2 of 7

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy