-
Layback of audio stems 29.97 to 25 fps conversion
I did a “Better” quality Compressor conversion of a 1080i2997 project (94 minute feature length) to create an “international” 25fps (PAL) version ( original frames play at 100% duration, no simple 4% offset), and while it took a week of processing, it looks okay. I carried the stereo full mix audio with it and it plays back in sync.
However, international requirements normally require a number of audio stems, music, FX, Mix-Minus and so on, which the audio mastering house provided — ostensibly for the 29.97 version, but at 48K.
Attempting to lay back these 48KHz audio files in a 25fps timeline results in massive sync issues — by the end of the show, the mix-minus is 6 seconds short — which leads me to believe its the old .1% pulldown problem. “Stretching” the audio clip by applying a 99.90% speed remap does place the tail pop in the right place, so one thinks problem solved, right? Export a new wav or aif, re-import that, and you’re ready to master, one would think? Not so fast. FCP either ignores such a “small” remap, or something, because the new wav or aif still plays back short, even though the graphical representation of the waveform is dead on in-sync. In other words the “sync-pop” burst visually lines up at the end, but the actual sound is still 6 seconds early. Quite perplexing. But get this — even the “100%” audio track that was created and plays back in sync as part of the exported 25fps Quicktime video exhibits the same behavior if you export it as an independent aif or wav file — it is also suddenly about 6 seconds short. What a wonderful world we live in, right?
I have resorted to trying to learn Soundtrack Pro on-the-fly, which is not a good idea… but applying a process “Time stretch” seems to help, although the accuracy is somewhat hit-and-miss. Doing a strict mathematical .1% stretch came up 3 frames short on my first pass, but that was a significant improvement that yieled an offset error that can be dealt with. At least I hope it does, because I’m still not out of the woods.
The down and dirty, though appears to be that FCP treats all audio files, regardless of whether they are 48KHz or whatever, as 29.97, and will apply .1% pulldown, which is death and doom when dealing with a whole-frame timeline like a true 24.000 (film) or 25.000 (PAL) project.
Correct me if I’m wrong or if I have missed something that Apple advised somewhere deep in a white paper lost in antiquity. I’d love to know that its something that they’re even aware of….
you know, so it doesn’t show up in i-Cut.1., sorry FCPX.
jPo
You mean “Old Ben”? Ben Kenobi?
-
9 Replies
-
Mr. Owens,
Have you tried the Pull up/Pull down feature in STP instead of the traditional calculated time stretch which you mentioned above? It works… I used it to go from true 24fps audio from a mag track to 23.98 video, and it worked like a charm first time. And, I coached a complete newbie though it’s use recently, though I can’t say I remember the parameters.
In any case, I’ve only used it a few times, so I have to feel my way through it and thus can’t give you precise instructions. However, I’m confident that it’s a good area for you to explore in STP, and if I could get a complete newbie to use it successfully, I’m sure you can work your way through it.
Hope this helps…
David Roth Weiss
Director/Editor/Colorist
David Weiss Productions, Inc.
Los Angeles
https://www.drwfilms.comPOST-PRODUCTION WITHOUT THE USUAL INSANITY ™
A forum host of Creative COW’s Business & Marketing and Apple Final Cut Pro forums. Formerly host of the Apple Final Cut Basics, Indie Film & Documentary, and Film History & Appreciations forums.
-
[David Roth Weiss] “Have you tried the Pull up/Pull down feature in STP”
In fact, that is how this situation has been resolved. I used an “absolute frame” pull-up factor of 100.1 per cent stretching 142409 frames to 142551+1 frame (probably an artifact of the Compressor process) and it syncs to- the-frame on playback. Interestingly, the visual waveform on the FCP timeline now shows the tail pop as being about 6 seconds late…. but it actually plays correctly. So no one thinks there’s anything wrong with FCP’s frame handling?
And, I’m surprised to find that STP is not completely useless. I’m applying the stretch to the Full mix as well, because there’s a ton of phase-flanging now between the two mixes, and contrary to popular belief, there are still some things that are unacceptable for broadcasters. Not many, but a few.
Thanks for the advice, its the right way to go.
jPo
You mean “Old Ben”? Ben Kenobi?
-
[Joseph Owens] “I’m surprised to find that STP is not completely useless.”
I was too!
Now for the bad side of STP… I tried to do “the full treatment” on a 40-min. doco with STP back in April, meaning 100% of the sound design and 100% of the mix. However, STP crashed so often that I had to give up on the concept completely. I did a complete ground up reinstall of all software too, and even broke the project into 10-min. reels, but nothing I could do made the crashes cease.
David Roth Weiss
Director/Editor/Colorist
David Weiss Productions, Inc.
Los Angeles
https://www.drwfilms.comPOST-PRODUCTION WITHOUT THE USUAL INSANITY ™
A forum host of Creative COW’s Business & Marketing and Apple Final Cut Pro forums. Formerly host of the Apple Final Cut Basics, Indie Film & Documentary, and Film History & Appreciations forums.
-
Sync drift has been covered here a lot. In a nutshell, FCP stamps a frame rate on an audio file. Why? Ask Apple.
The problem is the frame stamp is done when the audio file(s) are imported into FCP and is based on your default FCP project, not the frame rate of your current sequence. Why? Ask Apple.
The fix is to delete the files completely from your drive, change the file name and copy back onto your Mac. Change FCPs default project to PAl 25 fps and reimport the files back into FCP. Now, drop those files into your 25 fps sequence and they won’t drift.
The big question is why stamp a frame rate? I can see absolutely no advantage and I don’t know why Apple choose to go to the trouble of adding a frame rate when none is ever required. The NTSC world is already an awful mess of frame rates, so why oh why add this dogs breakfast to audio is beyond me. (*rant mode off*)
-
Now Michael if you could just explain why STP crashes so darned much your rant would not be in vane.
David Roth Weiss
Director/Editor/Colorist
David Weiss Productions, Inc.
Los Angeles
https://www.drwfilms.comPOST-PRODUCTION WITHOUT THE USUAL INSANITY ™
A forum host of Creative COW’s Business & Marketing and Apple Final Cut Pro forums. Formerly host of the Apple Final Cut Basics, Indie Film & Documentary, and Film History & Appreciations forums.
-
I have only used STP once and it didn’t crash for the three minutes I used it. With two Fairlights in hand, I doubt if it will get a second run
-
Michael,
I don’t have even one DAW, so I tried to use STP just to see if some of it’s finer attributes might actually be helpful. And, I can honestly say that had the darned thing not crashed every time I blinked, it’s sub-frame audio editing capabilities and it’s great library and auditioning features would have been quite useful and much speedier than what I had to do in FCP. But alas, I was not able to get past the crashing, which defied even my troubleshooting and technical sleuthing abilities.
As I’ve shared with you before, years ago, like you, I was a location sound mixer here in Hollywood for seven or eight years when I was in my twenties, and I was also doing sound design before that job title existed. I was doing the job on mag track and 1/4 inch tape in those days, so for me doing it on any system today is no real sweat.
In any case, the sound design job I did in April was the first I’ve done for someone else in years, and it was actually fun. And, though my finished product was ultimately seamless, I know I could have done the same job, or better, about fives times faster and ten times easier had even STP worked properly. And, with a “real” DAW, who knows???
David Roth Weiss
Director/Editor/Colorist
David Weiss Productions, Inc.
Los Angeles
https://www.drwfilms.comPOST-PRODUCTION WITHOUT THE USUAL INSANITY ™
A forum host of Creative COW’s Business & Marketing and Apple Final Cut Pro forums. Formerly host of the Apple Final Cut Basics, Indie Film & Documentary, and Film History & Appreciations forums.
-
I use STP for audio work on a project I do once a year, and it performs surprisingly well.
My setup has all the content media, and all caches, installed on the media array and not the default setting of the startup disc. Or maybe I have just been lucky.As far as the time stamp issue…perhaps this has to do with interoperability with Avid/Protools, which also do this?
Of course, in Avid you must specify a project frame rate, where in FCP anything goes….so timestamping can give you umm….interesting results. -
[Michael Gissing] “The fix is to delete the files completely from your drive, change the file name and copy back onto your Mac. Change FCPs default project to PAl 25 fps and reimport the files back into FCP. Now, drop those files into your 25 fps sequence and they won’t drift.”
Interesting. I could not for the life of me figure out where FCP was getting the 29.97 frame rate when I was clearly in a 25 fps project. Hmmmmm. Its almost worth re-opening the project to see if the preview waveforms would line up correctly… nawwwww……
jPo
You mean “Old Ben”? Ben Kenobi?
Reply to this Discussion! Login or Sign Up