Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums Creative Community Conversations FCPWorks sessions from NAB posted.

  • Bill Davis

    April 21, 2015 at 11:17 pm

    FWIW, I spent a good chunk of my time talking with Phil Hodgetts at NAB and one thing he mentioned that surprised me was that instead of the “classic” timecode approach of a black burst generator and running coax everywhere – he tested just bringing a basic GoPro on-set to use its internal clock as the primary timekeeper in a lumberjack environment.

    The point is that reference time isn’t what it used to be.

    Are we that far from our cameras “sensing” the presence of an iWatch on set and letting each device track that time hack internally?

    Lumberjack uses clock based “real time” because that supports its mission. But I couldn’t help thinking how if devices in general can tap into global time precisely, then the old TC system could actually become obsolete? No?

    Just food for thought.

    Know someone who teaches video editing in elementary school, high school or college? Tell them to check out http://www.StartEditingNow.com – video editing curriculum complete with licensed practice content.

  • Oliver Peters

    April 21, 2015 at 11:59 pm

    [Bill Davis] “But I couldn’t help thinking how if devices in general can tap into global time precisely, then the old TC system could actually become obsolete? No?”

    Well maybe, but I don’t see that being implemented across the board anytime soon. As far as accuracy, take an iPad for example. When it is disconnected from the internet or a phone data network, it does not maintain perfect accuracy. Fine for taking notes in something like Lumberjack, but not accurate enough for sound sync.

    Take any current camera or sound recorder. These cost a lot of money, yet do not maintain perfect sync on their own in “free run” without periodically having to be externally jam synced.

    – Oliver

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

  • Michael Gissing

    April 22, 2015 at 12:07 am

    [Bill Davis] “You can try to come up with all the ” it can’t be THAT much better” rationalizations you like, but serious people in very professional situations are starting to report precisely that. It IS a much better workflow – and one that produces bundles of time savings and workflow efficiency. ”

    Did I say that? I am asking if there is proper quantifiable comparison of workflows. It is great if an editor finds something is faster only to discover the assistant is spending hours on extra work to make that possible. Again not saying this isn’t desirable but motherhood statements about speed need to be both quantified and qualified.

    As Oliver points out there is a shift on logging from post to location. For some workflows like doco are just not feasible. I started watching the Indie clip also posted and I am impressed by the location to post workflow and that a lot of logging can be semi automated from location logs. Brilliant but again is this just a shift of work rather than some quantum leap in efficiency. Show me the data not the confirmation bias.

    Also you have not addressed my concerns that were also mirrored in Mike’s presentation that some editors are not bothering with using Roles or don’t understand. I see the potential but I also realise that if editors can deliver a job to post finish with multiple files named “untitled” then asking for them to do more metadata work up front is likely to fall on deaf ears. In that case hand over to post is worse and requires more time post edit. Efficiencies really matter to me because I work in a faster turnaround low budget doco world. Any shunting of track management to me affects my bottom line so I really want Roles to work.

  • Michael Gissing

    April 22, 2015 at 12:15 am

    [Bill Davis] “But I couldn’t help thinking how if devices in general can tap into global time precisely, then the old TC system could actually become obsolete? No?”

    In the file world, time of day is excellent. In the old tape world, it was a problem with prerolling. So in theory yes. I watched a good vid from the Deneke people at NAB. It might be worth tracking it down as they show how a distributed TC and sync system needs a single ultra stable clock and the ability for all devices to re jam after power down.

    Camera’s like RED are notorious for needing re-jamming after battery changes. On multi cam drama good code management makes a difference. Auto sync systems that just rely on mod matching guide to hero will not be reliable in many circumstances and will also need distributed guide audio to record on cameras so that a system like Pluralize can work best. Might as well distribute timecode from a proper stable master and jam. Much easier.

    I have a Blackmagic 4k camera which doesn’t support timecode but does time of day. I have a simple system of a single clapper and calculating offset to sync for a whole days rushes. It works but with more than one camera it suddenly becomes messy because the devices require manually setting rough time.

  • Michael Gissing

    April 22, 2015 at 12:22 am

    [Mathieu Ghekiere] “Hi Michael,…well, some things to think about:

    I think about this sort of thing a lot so thanks for the perspective. I too see Roles as a thing with much greater potential. Part of my concern doing doco post is how often that work is not done. The penalty is to me not the editor in spite of the fact that sorting Roles up front should be more efficient for the edit as well.

    So many editors just want to get editing and sort later. I am questioning the assumption that X is always more efficient or faster. The work has to be done and more than ever with resolve supporting metadata like shot descriptions, I can see it will flow into grade and assist in finding shots with the same grade so I am keen for editors to go there with Roles. So far my limited experience is that they are not doing it upfront and feel a bit annoyed when they need to do it later. Alternatively they just pass messy AAFs on to me.

  • Oliver Peters

    April 22, 2015 at 12:24 am

    [Michael Gissing] “and will also need distributed guide audio to record on cameras so that a system like Pluralize can work best.”

    Unfortunately many cameras record audio out-of-sync by + or – 1 to 2 frames.

    – Oliver

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

  • Michael Gissing

    April 22, 2015 at 12:34 am

    [Oliver Peters] “Unfortunately many cameras record audio out-of-sync by + or – 1 to 2 frames.”

    True Oliver, especially when using external recorders like Ninja or Ki Pro. The problem is the video is delayed. Audio is fine.

    The solution is to feed through the camera, not direct into the recorder and the camera compensates on the embedded HDMI or SDI out. Same with the timecode. The camera needs to be the jam source not the external recorder.

  • Oliver Peters

    April 22, 2015 at 12:46 am

    [Michael Gissing] “True Oliver, especially when using external recorders like Ninja or Ki Pro. The problem is the video is delayed. Audio is fine.”

    Actually I’ve seen it direct in the camera. I recently cut a video shot with two different models of Canon DSLRs and one had internal audio sync that was two frame off from the video, while the other was dead-on.

    – Oliver

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

  • Michael Gissing

    April 22, 2015 at 12:57 am

    Ah DSLRs are another matter.

  • Aindreas Gallagher

    April 22, 2015 at 10:18 pm

    [Oliver Peters] “in the presentation, it wasn’t clear to me whether Mike was talking about organizing tracks into DME groups to send on to the mixer or whether he was talking about exporting a final version with DME stems.”

    but it is to pro-tools right? he says they demanded EDL and stuff – but he references pro-tools destination quite a few times. That whole production was tailed with an avid matched edit in unmarked cars the whole time apparently? Agnostic devil’s advocate: the thing you can’t get away from is that all the items on the timeline carry persistent identity.

    as in every time we dump VA objects into a timeline in ppro etc, all they’re left knowing is the track they’re in. X objects kind of do have persistent actionable sets of characteristics? although X fails on matchback duplicates stuff (does it still?) You’d think they get lost in the metadata. You’d suspect X sucks on a lot of common or garden label colour stuff as well. It feels like proof of a demanding intellectual cleanliness more than anything.

    I’m not sure what apple won by being as metadata calvinist as they did? It looks a lot like an outcome sea of metadata objects inside odd timeline rules with constrained matchback links to the original sea of metadata in the library. Apple seemed really, really determined to prove a point to themselves.

    https://vimeo.com/user1590967/videos http://www.ogallchoir.net promo producer/editor.grading/motion graphics

Page 2 of 5

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