Offline bouncing 24 FPS material to 23.976 audio files - Workflow question

[polyh3dron]

Hi all,

I'm trying to take my delivery prep stem session, which is natively 24 FPS, and offline bounce it to create 23.976 FPS audio files for various deliverables.

I understand that the traditional method of creating deliverables in a different frame rate is to import the original audio into a session with the desired frame rate and sample rate convert the audio in the import menu, but I'm looking to natively offline bounce, and I have attempted two methods; both which yielded undesirable results:

Method 1: 0.1% Pulldown (Film to NTSC) Audio setting in session settings activated, and session frame rate changed to 23.976 FPS prior to bouncing. The problem with this method is that while the audio has had the pullup applied properly and does not drift against 23.976 FPS picture, the end timecode of my bounce is actually incorrect, and my audio ends up being incorrectly time stamped, making this option not viable.

Method 2: Keep Session Settings at 24 FPS with no audio pullup, and select the 0.1% Pulldown (Film to NTSC) sample rate setting in the offline bounce menu. The problem with this method is that when attempting to import this audio into a session to check it against my 23.976 video, Pro Tools sees these audio files not as 48000Hz files, but as 47952Hz files and warns me that while I can add them to my 48000Hz session directly, they will not play at the proper speed. After getting rid of these nagging dialog windows, the files of course play at the proper speed (they are also time stamped properly and have the correct in and out points!), but I can't deliver 47952Hz audio files.

As a workaround for this project, I'm just reconsolidating the 47952KHz audio files I've made, but this is taking an eternity and is far from ideal.

Having exhausted these two options for the reasons I've just laid out, is there any viable workflow to offline bounce 24 FPS material to 23.976 FPS audio files or vice versa within Pro Tools? Or is converting existing audio files on import the only way this can work properly?

PS. The "Film to NTSC" and "NTSC to Film" descriptors are the wrong way around in the offline bounce menu's sample rate settings, which I also learned the hard way.

[minister]

Originally Posted by polyh3dron

Hi all,

I'm trying to take my delivery prep stem session, which is natively 24 FPS, and offline bounce it to create 23.976 FPS audio files for various deliverables.

I understand that the traditional method of creating deliverables in a different frame rate is to import the original audio into a session with the desired frame rate and sample rate convert the audio in the import menu, but I'm looking to natively offline bounce, and I have attempted two methods; both which yielded undesirable results:

You say, "delivery prep stem session". Does this mean you are handing this off to be mixed? Or, are you finished and want to create final MIX and Stems @24fps and 23.976fps? I am thinking it is this latter option.

I print everything at the working frame rate. Then, I do what you call the "traditional method". Works every time. Always bang on in sync. Pretty darn fast. Why not use this method?

In my experience, the "Film to NTSC" menus are correct in the Import & SRC menu.

[polyh3dron]

Originally Posted by minister

You say, "delivery prep stem session". Does this mean you are handing this off to be mixed? Or, are you finished and want to create final MIX and Stems @24fps and 23.976fps? I am thinking it is this latter option.

I print everything at the working frame rate. Then, I do what you call the "traditional method". Works every time. Always bang on in sync. Pretty darn fast. Why not use this method?

In my experience, the "Film to NTSC" menus are correct in the Import & SRC menu.

By a delivery prep stem session I mean a session that references my recorder stems, with bussing to offline bounce all my required deliverables: printmaster, DMEs, folddowns, etc. Was trying to figure out how to bounce out a set of deliverables at 23.976 FPS from my 24 FPS stems. My longplay stems only exist as a clip group that references the original stems, that clip group then gets split out back to reels with pullups, still as clip groups referencing the original stems. I have bounced out my 24 FPS deliverables in reels, but need to bounce out 23.976 FPS deliverables in a longplay. I'm trying to avoid bouncing out 24 FPS longplay deliverables that no one needs and instead bounce out longplay 23.976 FPS deliverables directly from the 24 FPS stems. Trying to reduce the amount of steps in the process.

It's the offline bounce menu's sample rate settings for pulling down or up that have the descriptors the wrong way around, and it's the offline bounced files with the pull down sample rate setting that actually get flagged as 47952 Hz files.

[minister]

Hunh. I see. Not sure, especially given the clip groups workflow....

Off the top of my head, you could dump those 24fps deliverables in Zynaptic (Prosoniq) Time Factory II and do the pull in there offline.

Side question: Reels with pullups were necessary in the analog(ue) film days. Do we need pullups in the digital reels days?

[dr sound]

Originally Posted by minister

Side question: Reels with pullups were necessary in the analog(ue) film days. Do we need pullups in the digital reels days?

Yes for DCP deliverables, they require Pullups.

[polyh3dron]

Originally Posted by minister

Hunh. I see. Not sure, especially given the clip groups workflow....

Off the top of my head, you could dump those 24fps deliverables in Zynaptic (Prosoniq) Time Factory II and do the pull in there offline.

Side question: Reels with pullups were necessary in the analog(ue) film days. Do we need pullups in the digital reels days?

My objective here is to skip having to make 24FPS longplay assets, since these are not in my delivery specs.

And yes like Marti said, pullups are standard for features, and when you do them right like with the clip group method I laid out, it ensures sample accurate, perfect joins in a DCP with no ticks or other anomalies, and if the stems get updated destructively, the delivery prep session referencing them will also be updated without having to punch multiple generations of stems.

[smurfyou]

EDIT: several posts were added before my reply

[Farhoof]

If I remember correctly the pull up/down only marks the files with a certain sample rate. Importing them into a new 48kHz session does the actual conversion so you will always need to do this 2 step process. In my experience it's easier to export as working framerate and import with the target conversion applied at the import, not the export. So this in-between version, either long play or weird sample rate, is unavoidable.