Laying Final Mix to Final Vision Without Re Encoding???

[Miker84]

Hi all,

I'm searching for a reliable method of laying back a final stereo mix to final vision that does not require re-encoding the video.
Typically this is a WAV and a QT ProRes video.

Here are the methods I've tried so far:

1. ProTools - Bounce To QT
Importing the hires video into PT, lining up the audio, checking broadcast audio levels and bouncing to quicktime.
Pitfalls - depending on the filesize, current network activity and the way the wind is blowing this often takes far too long and sometimes even crashes PT.

2. QT Pro
Outputting final mix wav, opening wav and vision in QTPro, copying audio, using 'add to movie' function, and 'save as self-contained movie'.
Pitfalls - again network speed and filesize tend to slow things down to a standstill. There are also issues if the video and audio durations are not exactly the same, and without a visual timeline (like Avid or PP) it introduces a lot of room for error.

PremierPro does a great job of speedily handling the video and audio assets across a network, however, is not an option as it re-encodes the video upon export.

Surely there is a third party app that is designed for this exact process.

What methods and applications is everyone else using?

Big thanks all,
Mike

[Stereo Missile]

I just bounce to QuickTime in protools and select "same as source" for the video output so that it doesn't reencode the video. Seems to work fine. Then I can transcode into a delivery format for previews etc in mpeg streamclip.

[papalou]

As for as I know, only Quicktime Pro does the job, all other software will re-encode.
We use Scratch for this here, because it's very quick (of course it depends on your computer and storage capabilities, but we can render at up to 70fps here), you can also try with Resolve Lite, which is free and can do it this very quick as well.

[Miker84]

Originally Posted by Stereo Missile

I just bounce to QuickTime in protools and select "same as source" for the video output so that it doesn't reencode the video.

You're right Stereo Missile, bouncing with 'same as source' checked does not re encode the video - It just seem to be flaky when working with with larger files like ProRes 422 or 4444.

Originally Posted by papalou

As for as I know, only Quicktime Pro does the job, all other software will re-encode.
We use Scratch for this here, because it's very quick (of course it depends on your computer and storage capabilities, but we can render at up to 70fps here), you can also try with Resolve Lite, which is free and can do it this very quick as well.

Thanks Papalou - I'm sorry I don't quite understand. Are you suggesting that Scratch and Resolve do not re encode the video?

Any other thoughts from anyone?

[papalou]

Scratch and Resolve will both re-encode, but :
- 2 generations of Prores 422HQ video encoding is not a problem, and there will be any noticeable difference.
- they're so quick that it will be much faster than the Quicktime Pro process or Protools' Bounce to Quicktime.

[tom_lowe]

Originally Posted by papalou

Scratch and Resolve will both re-encode, but :
- 2 generations of Prores 422HQ video encoding is not a problem, and there will be any noticeable difference.
- they're so quick that it will be much faster than the Quicktime Pro process or Protools' Bounce to Quicktime.

+1

ProRes, DNxHD and DNxHR are all designed with multiple encodes in mind. For example, if you received rushes in ProRes and then after grading you exported to ProRes again and there were visible signs of re-encoding, no-one would use it as it would make many workflows impossible. Broadcasters wouldn't want to deal with image sequences or the like.

The other option is GoPro Cineform which handles multiple encodes even better, but produces larger files.

[smurfyou]

The other option is ffmpeg, with or without a front-end. If you or your IT person can build batch files for it that makes it fairly easy.

[kcfroines]

Originally Posted by Miker84

2. QT Pro
Pitfalls - .....There are also issues if the video and audio durations are not exactly the same, and without a visual timeline (like Avid or PP) it introduces a lot of room for error.

When I do this I bring the picture into the Protools timeline, then consolidate the audio to be the exact same length before adding it in Quicktime Pro.

[Miker84]

papalou & tom_lowe - great info, this makes sense. I'll look at all of these apps as options.

smurfyou - I've been reading about ffmpeg - this could be an option too - I've got some great IT support

kcfroines - thanks Yes this is an option, but slows workflow down evne further when a high volume of work is being performed.

A massive thanks for the help all