Skip to main content
A
Anonymous
October 17, 2017
Answered

Premiere Pro Mismatch - Export is 3 Secs Longer Than Sequence Duration

  • October 17, 2017
  • 3 replies
  • 8396 views

Hello everyone,

I have an issue here with PP CC 2015 - or maybe I do not understand frame rates / time codes correctly.

Here's the PP project info:

> PP sequence setup is 23.976p

> all footage used in the sequence was shot at 23.976p

> timeline in the sequence shows a total duration of <b>00.39.59.19</b>

> when I export the video as 23.976p, it then shows a duration of 40 mins 2 secs in MediaInfo - a full 3 secs difference

> VLC player shows a duration of 40 mins 2 secs as well when playing the file yet the time code that was burned into the exported file shows 0.39.59.19 on last frame...... (?)

When I import that PP video file into Pro Tools into a 23.976p timeline it shows a duration of 0.39.59.19, but when I click on video properties I see 40 mins 2 secs 19 frames...

What am I not understanding here ? Shouldn't these SMPTE timcode values refer to the actual length of the sequence depending on the frame rate setting ?

Why does the total duration of my export differ from what was stated in the sequence timecode ?

Thanks !

    This topic has been closed for replies.
    Correct answer Meg The Dog

    It's the difference between a 23.976 frame rate and real time.

    24 fps (23.976) video is always non-drop timecode. There is no standard for 24fps drop frame code. And within Premiere Pro, your are reading the total running time based on non-drop frame timecode - so it is giving a time readout duration based on frames - which does not match real time - since the frames are running slightly slower than real time (23.976 fps as opposed to true 24fps).

    Over the period of 30 minutes, this slow frame playback is making the show length run longer in real time by about two seconds per 30 minutes of program length.

    This why in NTSC 29.97 and 59.94 video there is drop and non-drop timecode. Drop frame timecode drops timecode numbers at specified intervals over time to compensate for this difference - that the video rate is slightly slower than real time. By dropping a timecode number, the duration amount displayed increases on each dropped timecode number.

    In drop frame timecode, when you have a duration of 30 minutes, it coresponds to real time 30 minutes.

    Here's the math:

    If you had 30 minutes of footage running at true 24 fps - 24fps X 60 seconds = 1440 frames in a minute, 1440 x 30 minutes = 43200 frames in the 30 minute show. (So non-drop TC, which counts the frames without regard to the true playback rate, displays a duration of 30 minutes - 43200 total frames, divided by 24(fps) = 1800 seconds = 30 minutes).

    But since the video, is, in fact, running slightly slower, at 23.976, those 43200 frames take longer to play out over time by .024 frame per second of real time.  So 43200  (your total show amount of frames) divided by 23.976 (the playout per second frame rate) = 1801.801802 seconds, which is 30 minutes, 1.8 seconds.

    Avid used to ( and maybe still does) come with a calculator to make this conversion and to help make the frame count = the desired running time. Both Premiere, and legacy FCP do not - I don't know if FCP-X does.

    It should be noted that within commercial spot production, with durations of 30 or 60 seconds, this whole mess is less apparent. It only shows up when you work with program length material. This is why they drop frame timecode was invented.

    It doesn't sound like you are aiming for a target show length, like a network time slot that require 28:30:00 on the frame, but if you were, a solution is to tighten up the edit by 1.8 seconds over 30 minutes

    MtD

    3 replies

    A
    Anonymous
    October 17, 2017

    @ ACP: thank u for confirming that PP does not show real time as time code values, ergo not adjusting for the 0.024 drop.

    Unfortunately I do need this to stay exactly under 40 mins, and believe it or not, there are no 3s to cut.

    PP was just the edit, this 3s offset has trickled down to all other apps (sound, score, grade) and all of them do display the time code as well as u explained - and all of them referenced the PP video.

    Def cutting the next one in Avid if they still have correct display..

    what is the point of displaying time code duration if it is incorrect... ?

    M
    Meg The Dog
    Inspiring
    October 17, 2017

    As I said, there is no drop frame option for timecode at the 23.976 frame rate - on any system. You only have one choice for timecode at 23.976:

    So you have to have a way of determining what the realtime length is going to be - there are on-line calculators available to do the math - you can google search for them.

    If I have to edit program length material at 23.976, I create a second sequence at 29.97 Drop Frame and nest the first sequence (the 23.98) into the 2nd as needed during the edit to check program length.

    MtD

    A
    Anonymous
    October 17, 2017

    @ ACP: yeah.... my point was:

    in 2018 u have to create a 2nd sequence in PP in order to get correct time code duration.... ?!?

    wow. just wow.

    the simple math that u provided above could be integrated into PP in 5 mins, and then simply show correct time code as per the chosen fps for that timeline... or at least a second timestamp that is correctly adjusted for 23.976... because what they call "23.976" is actually 24 fps

    unreal I cannot trust a time code display - no matter what the frame rate is. that is just a parameter in an equation.

    M
    Meg The DogCorrect answer
    Inspiring
    October 17, 2017

    It's the difference between a 23.976 frame rate and real time.

    24 fps (23.976) video is always non-drop timecode. There is no standard for 24fps drop frame code. And within Premiere Pro, your are reading the total running time based on non-drop frame timecode - so it is giving a time readout duration based on frames - which does not match real time - since the frames are running slightly slower than real time (23.976 fps as opposed to true 24fps).

    Over the period of 30 minutes, this slow frame playback is making the show length run longer in real time by about two seconds per 30 minutes of program length.

    This why in NTSC 29.97 and 59.94 video there is drop and non-drop timecode. Drop frame timecode drops timecode numbers at specified intervals over time to compensate for this difference - that the video rate is slightly slower than real time. By dropping a timecode number, the duration amount displayed increases on each dropped timecode number.

    In drop frame timecode, when you have a duration of 30 minutes, it coresponds to real time 30 minutes.

    Here's the math:

    If you had 30 minutes of footage running at true 24 fps - 24fps X 60 seconds = 1440 frames in a minute, 1440 x 30 minutes = 43200 frames in the 30 minute show. (So non-drop TC, which counts the frames without regard to the true playback rate, displays a duration of 30 minutes - 43200 total frames, divided by 24(fps) = 1800 seconds = 30 minutes).

    But since the video, is, in fact, running slightly slower, at 23.976, those 43200 frames take longer to play out over time by .024 frame per second of real time.  So 43200  (your total show amount of frames) divided by 23.976 (the playout per second frame rate) = 1801.801802 seconds, which is 30 minutes, 1.8 seconds.

    Avid used to ( and maybe still does) come with a calculator to make this conversion and to help make the frame count = the desired running time. Both Premiere, and legacy FCP do not - I don't know if FCP-X does.

    It should be noted that within commercial spot production, with durations of 30 or 60 seconds, this whole mess is less apparent. It only shows up when you work with program length material. This is why they drop frame timecode was invented.

    It doesn't sound like you are aiming for a target show length, like a network time slot that require 28:30:00 on the frame, but if you were, a solution is to tighten up the edit by 1.8 seconds over 30 minutes

    MtD

      E
      excited_Genie16B8
      Legend
      October 17, 2017

      I think the reported duration is not something you need to worry about.  Your sequence is coming out correctly.

      Terms & ConditionsAccessibility statement
      Using the community Terms of use Privacy Cookie preferences Do not sell or share my personal information ad choicesAdChoices