Vegas film workflow?

Shenan wrote on 4/21/2010, 6:18 PM
Hello All,

I haven't posted here in a long time, but I'm just starting to edit in Vegas again. I'm using 9.0c. I have a question for you all about what is the best workflow for material originating on film, using double-system sound. I'm sorry if this has already been covered previously or elsewhere, but I couldn't find the info with a search here or on google.

Keep in mind that I'm not planning on going back to film, so I'm mostly concerned about the initial prep of syncing sound and creating subclips to use for editing. The film is 16mm at 24fps, telecined to 29.97fps, and the sound was captured on a digital recorder at 30fps in mono (left channel only).

Here are the specific questions and what I'm planning on doing. Please correct me if I'm wrong.

1. How to slow down the 30fps sound to 29.97fps to match the video.

I'm guessing that this is done automatically if my project properties are set to 29.97 and I drop in the 30fps audio files and then render. I guessed this based on the fact that there is a project audio setting for resample and stretch quality. The help file says this about it: "Choose a setting from the drop-down list to determine the accuracy with which audio files will be resampled to match your project settings."

2. How to turn the left-channel only audio to two channels?

I'm just right-clicking on the audio events and choosing Channels -> Left Only.

3. How to sync audio and video?

I have yet to really get into the thick of this, but it looks like I will just have to slide the audio in relation to the video until either the audio waveform peak of the clapboard clapping to the visual of the first non-blurry frame of the clapboard after it closes, or by what looks good in the shots where there is no clapboard.

4. How to create synced subclips?

It looks like after I sync the sound and video of one take, I would select both tracks, select a region, and create a subclip from the timeline. Will this automatically put the audio and the video together in the subclip, or do I need to group them before creating the subclip, or something else?

5. General process

After synching one take, should I then delete the audio event, drop in the audio of the next take and repeat the process, or is there some more efficient way to do this, such as dropping the audio for each take into a separate track and then create the subclips by selecting the video track and its matching audio track?

Thank you all in advance for the help!

Shenan

Comments

farss wrote on 4/21/2010, 6:36 PM
1) Assuming the audio was recorded on an audio recorder you don't need to do anything. Vegas will ignore the fps metadata and work from the sample rate.
2) You're doing it the right way.

3) As you said, line up the clapper. Can help if both heads and tailes of the shot were slated.

4) Not too certain about this. I'd simply use Grouping.

5) Bit confused here. If there was multiple takes shot there'd be multiple audio recorded as well. I hope there's visual and audio slates or what a task.


I have to say Vegas might not be the best axe for this. Movie shoots today have systems that electronically slate vision and sound. I've had some files from the Diva recorders and sorry to say, Vegas really cannot use all the goodness the high end location gear brings to the table. Which is a friggin shame because Vegas is such a great audio application in many ways.

One tip. Projects like this can rapidly become HUGE and hard to manage. It can pay dividends to split the project up into reels. Obviouslt also choose one fo two approaches. Lock vision and then think about audio ot lock audio and then worry about vision.

You could create DV proxies with location sound and work with those rather then the full telecined original and then swap out for the final render.

Whatever you do test, test, test. No matter how well intentioned any advice is anything can kill your workflow and you may not find out about the problem until way down the track.

Bob.
PerroneFord wrote on 4/21/2010, 7:08 PM
Bob covered the highlights...

But I just HAVE to ask... why on EARTH did you telecine to 29.97 for? That workflow has been dead for going on 7 years. All these NLEs can do 24fps timelines. You should have just done straight DPX/Cineon scans of the film and matched them frame for frame to the audio.

Now you are going to have a mess on your hands with 29.97.
Shenan wrote on 4/21/2010, 7:13 PM
Thank you very much for the quick reply, Bob!

Just to add a little bit about the project for clarification, this is just a short film for school. It's just one reel of 16mm - 11 minutes of footage. Also no fancy equipment (so no high end goodness :-) - it was just a Fostex FR2-LE(?) that recorded 48k/16bit broadcast wav files. All 11 minutes of film footage are in a single quicktime file - Prores 422 at 720x486.

About my question 5 - I have my 11 minutes of video on the timeline. I'm going to add the audio for one shot or take. I sync that up and create a subclip. Then do I remove that audio from the 11 minute timeline, drop in the audio for the next shot or take and repeat the process? Or can I put the audio for each shot or take in a separate audio track so that they don't overlap? It's all a linear 11 minutes of video so each shot and take is one after the other, while the audio is all in individual wav files for each shot or take, and the audio clips are longer than each video take because we rolled sound before camera.

Excellent advice about testing, testing, testing. Thanks again!

Shenan
Shenan wrote on 4/21/2010, 7:21 PM
Perrone,

The answer is... I don't know! :-) That's just the default that they do. Like I added in my reply to Bob, it's just a student short, so I'm trying to figure this out as I go along. I think it may have just been a ursa diamond rank telecine. From checking out the lab's website, it seems like they offer the option of transferring the film at full-frame 30fps (no pulldown) which can then be slowed down to 24fps on the NLE, but I didn't even think about it. I didn't think it would bring any advantage if I ended up going back out to DVD, but now that you mention it, I guess DVDs can handle 24fps as well, right?

I have Another question about this:

"1) Assuming the audio was recorded on an audio recorder you don't need to do anything. Vegas will ignore the fps metadata and work from the sample rate."

Cool! Should I then leave the project default of 44.1kHz for the audio, even though the audio recorder recorded at 48kHz? DVD will probably be the highest quality that I'm going to output to.

I should have posted this question here a couple of weeks ago! :-)

Thanks!

Shenan
Former user wrote on 4/21/2010, 7:31 PM
He won't have a mess on his hands. He has film with pulldown which has been a standard for decades. Not everybody offers DPX scans and for a school project, that was probably not in the budget.

If your audio takes are short, you probably won't have a sync issue. If they are long (as in minutes) then you will. Your audio would need to be slowed down a bit to sync to the 29.97 transfer. But since your whole film is only 11 minutes, then this probably won't be an issue.

Dave T2
Former user wrote on 4/21/2010, 7:32 PM
Change the project audio to 48. All audio for TV should be at 48. 44.1 is the standard for music CDs.

The 30fps transfer is IF you shot at 30fps. Not for shooting and slowing down to 24.

People shoot at 30fps sometimes in standard def in order to keep a frame to frame relationship. The 30fps is slowed to 29.97 without need for any frame/field maneuvering.

Dave T2
PerroneFord wrote on 4/21/2010, 8:11 PM
He didn't offer that this was a school short until AFTER I posted. But even at that, if the option existed to bring in the footage at 24p (assuming the film was shot at 24p) that would have been my first choice.

Honestly, It's just nice to hear some film school somewhere is actually still USING film!
farss wrote on 4/21/2010, 8:49 PM
"Honestly, It's just nice to hear some film school somewhere is actually still USING film!"

I hear there's more 16mm being shot now than ever.

Bob.
Former user wrote on 4/22/2010, 4:13 AM
"I hear there's more 16mm being shot now than ever."

Yeah, people still want that film look.

Dave T2
Shenan wrote on 4/22/2010, 5:24 PM
Yes, the budget was pretty low. The 30p transferred and slowed down to 24p is something that the lab's website mentions. It would be a frame to frame transfer (1 frame of film = 1 frame of video), so I would imagine that could have worked too, but I wasn't sure if there would be other issues and I wanted the most common workflow, which seemed to be the 29.97.

Speaking of which, the issue with the sync is why I asked about slowing down 30fps audio to 29.97fps. I was told that with FCP you need to slow the audio .01% so that they match, but I know that Vegas is pretty flexible and adaptable with differing formats, so I wondered if it was really necessary to do manually in Vegas, or if it was an automatic process.

About film, yeah, I'm glad too, because if nothing else, it teaches discipline in shooting. We have to stick to one 400ft roll of 16mm film for this film, and a certain number of hours for the entire shoot, and even less hours with the actors. It was a good experience. I met someone that's currently at USC who told me that they have largely converted to HD video except for the intro classes and the cinematography track.

Thank you everyone again for your help!

Shenan
Shenan wrote on 4/22/2010, 5:37 PM
I've run into a little bit of a workflow problem:

I'm splitting the long video event into individual shots, lining up the audio, grouping them, and creating subclips. The problem is it creates a separate video and audio subclip, not one integral event, so the sync doesn't keep. This means it's starting to look like I will need to keep all the subclips on the timeline and move them around as I need them.

Is there a fix to "glue" the audio and video into a single subclip, or am I pretty much stuck working like this, with everything always on the timeline?

Thanks!

Shenan
Former user wrote on 4/22/2010, 6:09 PM
Vegas will not slow down the audio automatically. You will need to do that manually as well.

Dave T2
farss wrote on 4/22/2010, 6:51 PM
What I'd suggest you do is render out to new AVI files.
You haven't mentioned what the film was physically transferred to though. Generally this is done to Digital Betacam which is 10bit YUV and the Sony YUV codec matches that so you could use that. If it's only SD then you should get reasonable playback performance without needing a RAID. I know this means more work upfront but you then have sound locked to vision. That codec does support 24/48KHz if you want to keep the sound quality high as possible.

One thing I noticed from your previous posts. You seem to plan to pick the best vision take of a shot and then match the best audio take. If so I should say this may not work well at all. I don't really understand why but man it can really stick out that something is wrong between speech and vision, even when you cannot see the face it can jar. You might be fine but just take a break and have another look at it before deciding to commit.

Bob.
Shenan wrote on 4/22/2010, 7:09 PM
Thanks Dave and Bob.

Bob, I didn't mean to give that impression of what I'm planning to do, as I know it wouldn't work right. I'm going to sync the video with the same take's audio. I want to do everything by the book, so will do the .01% time stretch to have perfect sync.

Regarding what it was transferred to, my computer is not the most powerful, so I asked them to put it on Prores 422 codec, which seemed to be the most basic they offered. And it is SD video.

Is there no way to permanently lock separate video and audio into an event/clip that can serve as source for a project without having to render it out first?

Thanks!

Shenan
farss wrote on 4/22/2010, 7:49 PM
"Is there no way to permanently lock separate video and audio into an event/clip that can serve as source for a project without having to render it out first?"

NO!
Man I wish there was. Vegas will only assume sound and vision from the one file should be in sync. It's not really locked either, it is simply Grouped but on the upside if you do manage to Ungroup them and slide them out of sync Vegas will turn the event pink and give you an indication of how out of sync they are.

One way I get around this is to add event level markers from the trimmer labelled "Sync". Works quite well. Slip and slide the audio until it's in sync. RClick, open in trimmer and hit M. Do this for both the vision and sound event without moving the playhead / cursor.

Now some good news. That Sony YUV codec I mentioned is quite light to decode, perhaps even lighter then DV. Less compression = less CPU = More disk i/o bandwidth. Give it a test, see how it work, only take a few minutes. Your other choice would be Cineform but I've zero dramas with that Sony YUV codec, goes into and out of AE just fine as well.

Bob.

Shenan wrote on 4/22/2010, 8:28 PM
Thanks again, Bob. I will try some of these things tomorrow.