Questions for the audio post-production experts

[aqua_tek]

So, I'm in the process of trying to work on a few student films for reel purposes and I'm met with certain situations that I have doubts with.

This is mostly technical / format kind of stuff.

For example, I've been under the impression that you want to be working at 24 / 48 for video. The few packs I've been handed contain mostly of 16 / 44.1 recordings. I'm sure the bit depth doesn't really make much of a difference, but I'm wondering whether I should upsample these recordings to 48 khz, or is Live smart enough to resample it into project sample rate once imported. (does that make sense?)

Also, the few video clips i've received have no timecode burned in. Would this generate big problems in the long run?

I honestly need some post-production technical considerations review. Most of what I've found online is very vague, and doesn't talk about specifics. It's just typical "so you want to work in post production? then you need to use samples, record shit, and put it into a sequencer" type of shit, that's pretty much useless.

Funny thing is, I took a course in post-production for film about 2 years ago, and some of this was discussed briefly, but I continued to focus on music, so I've forgotten most of these considerations. Been searching around and can't find my class notes either.

[leedsquietman]

DV video works in 48 bit (actually 16/48 typically). Audio for DVD video has to be 16 or 24 bit, 48 or 96 Khz.

DVD-Audio (high quality surround audio but no video) is a high format audio format but not the same as putting audio on a video DVD http://en.wikipedia.org/wiki/DVD-Audio

If you need to sample rate convert, I suggest you use a different SRC to Live, which is not so good for sample rate conversion (most DAWs are not).

If you're on PC, download the free Voxengo R8Brain, or if you have Ozone, it has an excellent SRC. Not sure about Mac standalones (other than Ozone). Typically, if you use a good SRC upsampling to 48 Khz or 96 Khz should have virtually no detectable audio artifacts.

[LP]

If you use a Mac and have Logic, Soundtrack Pro is reasonably good at this.

I have to say though, Live is remarkably good at hosting different sample rates in the same session, when compared to other DAWs, in such a way that it kind of compensates for the frustration of not having OMF capabilities. I have used LIve for my last 3 film projects, in conjunction with Soundtrack Pro, and find it to be a very efficient pair.

You are going to certainly be using 48kHz for the video output.

Look at filmsound.org and creativeCOW.org

[leedsquietman]

Most other DAWS demand you convert all files to the same sample rate, so Live has a one up on many of them, but there can be a slight compromise of quality as a result when playing back files with different sample rates.

[3phase]

So, I'm in the process of trying to work on a few student films for reel purposes and I'm met with certain situations that I have doubts with.

This is mostly technical / format kind of stuff.

For example, I've been under the impression that you want to be working at 24 / 48 for video. The few packs I've been handed contain mostly of 16 / 44.1 recordings. I'm sure the bit depth doesn't really make much of a difference, but I'm wondering whether I should upsample these recordings to 48 khz, or is Live smart enough to resample it into project sample rate once imported. (does that make sense?)

Also, the few video clips i've received have no timecode burned in. Would this generate big problems in the long run?

I honestly need some post-production technical considerations review. Most of what I've found online is very vague, and doesn't talk about specifics. It's just typical "so you want to work in post production? then you need to use samples, record shit, and put it into a sequencer" type of shit, that's pretty much useless.

Funny thing is, I took a course in post-production for film about 2 years ago, and some of this was discussed briefly, but I continued to focus on music, so I've forgotten most of these considerations. Been searching around and can't find my class notes either.

better convert all the files you need yourself otherwise it will be a mess..


videoclips without timecode? its all about the timecode.. do they have at least black frames in front?

without timecode whatever you do will run out of sync after about 1-2 minutes.. in case frame acuracy is required...

with just an ambient soundtrack in the background it wouldnt matter much...

of cause you can reallign the audio file to the clip later.. but this makes extra work and might result in audio degeneration if done badly...

however..

maybe the dont want time aligned audio.. just content to align by themself in the edit..
That would be a quite usual request.

so blackframes as start position would be enough..


if you get timecode and there is the need of time alligned audio you should rewire ableton into a propper daw..

[BlackMath]

if you need timecode alignment definitely use a DAW that excels at that because you or the film editor will have a lot of probelms with sync.

If you can't try putting a line up click in the beginning and end of the locked picture BEFORE it comes to you would help. Meaning get your editor to render a QT movie with a scratch audio file with the clicks in place.

I love Live and it was great fun doing the score and design for this http://www.aliceandthewhitehair.com

but it would have saved days of work if we had devised a workflow that used TC. But it is possible to do it without but you pretty much end up mixing the movie in the editing bay

[LP]

Ableton DOES allow for a timecode-based timeline (or else this would all be completely moot!) and the video file will be a certain frame-rate (check the QT settings) and you can assign the video file to be the tempo master, so the issue of drifting out of sync is not an actual problem, so even if the time-code is not burned-in, it doesn't mean that there isn't time-code sync.

This may not SEEM like something Ableton is good at, but it actually isn't too shabby.

It may just be that many users spend more time in the Session view than the Arrangement view, so they don't know it as well.

[Ajbbklyn]

What is it exactly that you're doing on these films?

Music to picture?
Re-recording (mixing) the soundtrack?

If the former, there's no need to worry about time code. You'll record/mix your music track(s) to the picture and export an audio file from Live. The video editor will simply take your music track and drop it on the timeline of the NLE. Your file output should be 16 bit/48kHz, wav or aiff depending on whether PC or Mac. As others have mentioned, it's a good idea to use a separate program for bit depth/sample rate conversion.

If the latter, i.e., mixing dialogue/music/efx/narration - then Live is not the application to use. This is where Avid/Pro Tools or Final Cut/Logic would come into play.

In the pre-digital days, music recordings were recorded "wild" (not in sync) to the picture and subsequently transferred simultaneously to
1) 35mm mag stripe to be edited as part of a sync roll of picture & track; and
2) a safety copy on ¼" tape with neo-pilotone sync signal (done on a Nagra)

In the digital age, everything runs "to speed" because there are no moving parts. The editor will be able to match "frame" (that point on the timeline that coincides with the clips you were provided with) and drop your music in the appropriate position.

[aqua_tek]

Thanks for the tips guys!

Yes, since what I'm going to be doing is mixing and sound design, one of my main concerns was sync. I guess I need to read up some more on this, but I wondered if and how not having timecode burnt in would generate problems. So, basically, no matter how accurately I try and do FX spotting, it will end up drifting? Should I ask these people to try and send me a copy with timecode burnt in then?

I'm still not understanding it very deeply, so I'll keep on looking for info around. Also, how do the black frames come into play here?


Ugh... sorry about the plethora of questions. Too much time focusing on just making music has really made me forget about all this stuff. Now trying to pick up the pieces. Maybe I should look for a book on the subject


Anyways many thanks again! Looking forward to reading more insightful comments.

[Hidden Driveways]

If you're working on student films, chances are that not only do they know absolutely nothing about audio, they likely know very little about video as well. The second you start throwing around terms like "burnt timecode," "blackframes," and "up-click" (say huh?), then they'll likely pee a little bit in their pants and wish they'd gone to accounting school. There will be a lot of on all sides.

Unless you're working on a longer project (say over 15 minutes in length), then a lot of this stuff will likely not come up. If it's a longer format film, then I would still go for it, just deal with each isolated issue one at a time as they pop up. I use the creativecow forums for this.

Reading books on this stuff is one thing, but getting your hands dirty and doing it is another. One problem I find with "audio post" is that there are 50 million names for everything. People make up their own names for technical stuff, use the names among their group of people in their production circles, and eventually have no idea that no one else knows what they're talking about. I read about this stuff every day, and every day I see 1000 new technical terms. I mean, even the name "audio post" isn't standard. There are like 50 different terms for it.

If I were you (and in this case I am you because I live in NYC and help students with their movies), I would just do it. Take on their projects, mix them with Live, and see how it goes. Don't let someone else tell you Live isn't a good tool for a given application. That's something you should determine on your own. Personally, I mix to picture and export audio WAV files from Live at 24-bit 48 kHz (don't export the video - it often looks like garbage). I've done a lot of projects this way and so far no one has complained. In fact, I'm told that my stuff sounds good and works well.

Just take it bird by bird, man. Bird by bird. See, that doesn't really mean anything, but you could say that in a technical conversation and 74% of producers, directors, and editors will nod their head in agreement with you.

[LP]

Absativley what Hidden Driveways says.

I am in the same boat, I help students finish audio on films every day, and while Soundtrack Pro is set up to kind of idiot-proof the process, and has a great spectral editor, the films I have delivered audio post done in Live have worked out just fine, with no more hassle (or birds...) than ones done in Logic.

[leedsquietman]

I personally just find it easier to import .wav files into Vegas and use that. Even the cut down Vegas Movie Studio is good enough for most personal video projects, so long as you don't need Blu-Ray. Cubase also has more options for exporting video with timecode etc, more video file formats.

I think Live's inbuilt video is still more geared towards a bit of visuals for Live performance. I have to say that I have seen some surprisingly good and deep stuff done with it, which would have probably been done in half the time or less in Vegas but it's a case of priorities. Do we want Ableton to add things such as more video features and surround sound, giving it some leverage in the audio to picture market?, or do we want it to concentrate on more audio and MIDI centric features, such as session view automation, proper dual monitor support (session on one page, arrangement on the other), bezier curves, 64 bit and most importantly, vastly improved stability ?

[SubFunk]

^^^ yep.

one big downer IS that you can't time stamp in live.

makes it for professional video / audio work pretty much redundant in my eyes.

and not having a proper big SMPTE display at hand is simply annoying the hell out of me on top of it.

someone made a proper arrange time display with M4L, maybe a SMPTE could follow up... still no time stamping, though.

but i see the urge to use live for that, because it is so great for sound design, i like to use it for that and a rough sketch and then accurately fine tune the 'time editing' elsewhere to the video / movie. (soundtrack pro / logic)

[aqua_tek]

If you're working on student films, chances are that not only do they know absolutely nothing about audio, they likely know very little about video as well.

Yeah I've started to notice this lately.

Don't let someone else tell you Live isn't a good tool for a given application. That's something you should determine on your own. Personally, I mix to picture and export audio WAV files from Live at 24-bit 48 kHz (don't export the video - it often looks like garbage). I've done a lot of projects this way and so far no one has complained. In fact, I'm told that my stuff sounds good and works well.

Oh no doubt. As SubFunk said, I just talk about live because it's just so second nature for me already, it's very easy to do my sound design in it, and I'm sure it'll be fine for smaller projects.

But in no way am I looking at it as the go-to DAW for all post-production tasks. In the end, I'd probably be using Live for the sound design side, and Logic / Soundtrack for the mixing, spotting, etc. Just need to figure out a good workflow going from one to another.

Would be cool to be able to test it within FCP, but I don't have that package. Just logic and soundtrack from the Logic Studio package. Don't think it would be necessary though. As some of you mentioned, prob best to just bounce the final .AIFF or .WAV and have them add it in.

And leeds, I agree that Live should not suffer from any more bloating.


Thanks again.

[aqua_tek]

So I decided to take on a little side project: I took one of my favorite childhood flicks, and redesigned one of the most memorable, most sound-design intensive scenes.

In this case, I took the original TRON, and redid the sound for a short segment of the light cycle scene.

I uploaded it to vimeo and Youtube. Not sure how long it'll be up, since I'm almost sure I'll be hit with some sort of copyright infringement thing, but here are the links:

http://www.vimeo.com/10249865
http://www.youtube.com/watch?v=eG2J0RegypE

Did it all in live. All effects are native. The only third party thing i used was Zebra 2 for some of the sound layers.

I apologize for the crappy video quality. Exported directly from Live and it was a not-so-great quality source vid in the first place.




Comments are very welcome... actually, I encourage them.