[Bouke Vahl]

[Benny Boo] “But I don’t think you read the question right. I am looking to export in 1920×1080”

I did read the question right, but you did not state your delivery format was 1920×1080.

If that is the case, change your timeline settings to 1920×1080, go trough it again and reformat if needed, then do whatever you think you must do, and be happy.

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

afaik, it’s not possible. (But perhaps I’m overlooking something.)
Did you not try to export one single line before putting in effort (way too much as PP is not a caption / subtitle tool)?

Good news, if you can export ANYTHING metadata style, it can probably be converted, but no guarantees that will be free.
Oh, SRT is a subtitle format, not a caption format. There is a difference.

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

There are about 20 questions you did not ask but are more important.
What kind of scaling is applied on what kind of content being one of them.
(Look it up, there are at least 20 scaling algorithems, each with their own pros and cons.)
Then, are you required to deliver a specific resolution? If so, edit in that resolution.

What are you doing? Putting your puking cat / baby on Youtube?
How many people in your audience have a 4K monitor?

You do know that 99.X % of the money made in the movie industry was shot on LESS than 4K?

Did you do a test, looked at it, and decided what looks best?

Sorry to sound harsh, but really, you should have asked tons of questions before touching the cam.

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

[Job ter Burg] “When those 48.048 sound files are played back at 48 kHz, the 0.1% pull down”

If this is the case, I stand corrected of course.
Do you happen to have one of those files to study? (Just for hobby, I very much doubt this is still used nowadays.)
Both samplerate and byterate should reflect 48048.
Of course there is no way for me to check what is really happening unless there is tone that should be at a very exact frequency, but still, it’s fun. (Yes, I am a nerd.)

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

[Job ter Burg] “and recording sound at 48048Hz”

No, huge thinking mistake. The audio is in fact recorded as 48000Hz, but flagged as if it was recorded at 48048.
Same as shooting 50 fps and put it frame by frame on a 25 timeline, you get slomo.
Hence it’s very easy to ‘undo’ the slomo. (It’s just some 12 bytes in the Wave header.)
(This is what I wrote you might have done 15 years ago…)

[Job ter Burg] “And my point is: the TC rate does not necessarily tell you what the framerate on set”

Yes and no. If you see the framecounter go from 23 to 00, of course it can be 23.976 as well as 24, so this is relative meaningles.
But normally the metadata inside the clip will tell you shooting rate, and that most of the times corresponds with TC rate. (High fps excluded, that varies per manufacturer.)
And yes, of course setting up the project to the right FPS is very important.
(but imho more to avoid blended frames then for syncing…)

[Job ter Burg] “Anyway. I don’t think we’re in disagreement.”

Me neither, just trying to educate the masses, not you ☺
(And avoid doing actual work…)

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

Hi Job,

[Job ter Burg] “If the cam was running at 23.976 and the sound as running at 48K, picture and sound should sync up in a 23.976 project. Timcode counts, but does not change speed.”

It should always sync, no matter the cam framerate and sound sample rate.
BUT, to calculate the TC from the BWF timestamp, it IS important to know the speed.
23.976 is a non-drop format, (Or, regular TC24 as you call it) where 01:00:00:00 is NOT a wall clock hour, but 0.1% less (Thus 3.6 seconds)
A BWF recorder does nothing different in recording the files at any speed. Only thing that changes is a tiny bit of metadata in the file (Bext entry SPEED=XXX.XXX that has not to be used).
But, of course the LTC in/output does change with the setup, and that is crucial to match the cam.
Again, nothing else is different. (Ok, in the old days you could fool Avid by setting the sample rate wrong to speed up/down for a video shoot intended for film or vice versa, but that workflow went the way of the Dodo. You might have done such a thing 15 years ago.)

But still it’s unclear if the OP has problems with the start sync, or problems with drifting sound. If the latter, there are numerous things that could cause the problems.

And you know this, as on import of a BWF file you’ll be asked what speed to interpret it, and you see the TC numbers change every now and then. (Not on 24 / 25 / 30 / DF, if the start is at a round second, but otherwise it does.)

Now, (at least on my system), things are different with AMA Link. Then the project speed is always used.

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

The file is fine, it’s just not a Wave file.
Tell Philipp to send you a new file if you can’t recover it with Audacity. (I can’t, or it was very overmodulated to start with.)

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

Differences between ProRes and DNxHD:
(Not all, but first that comes to mind)
Prores is VBR, DNxHD CBR
Prores is ALWAYS (at least) 10 bits, DNxHD only on the ‘x’ flavours
Prores is resolution independent, DNxHD has a flavour per resolution, and that is a limited set

Now, there is a (cheap or free) solution to get ProRes out of your PC PP, based on a trick that it hooks into the frameserver PP seems to have.

But, DNxHD is no punishment, so if you have a decent flavour, just use it.

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

again, if you did not touch the director’s sequence, you’re not to blame.
Perhaps a trip to the attic?

Bouke
http://www.videotoolshed.com

[Bouke Vahl]

“delete audio track” will only remove that audio track with metadata from that sequence.
Nothing else should be affected, so if this is the case, something else has happened.

Bouke
http://www.videotoolshed.com