This is the work flow we use this using PPro, AE and Audition. Fair warning: We're enthusiastic amateurs, It's not perfect, it's not quite finished, and there's certainly inefficiencies, but perhaps bits of it will prove useful to know.

Ingest

• Copy all the video (GH2) and audio (Tascam) files to the hard drive
• Run a BulkRename task (http://www.bulkrenameutility.co.uk) to rename them based on the file creation date.
• Sometimes - but rarely - two files have the same creation date right down to the second. When this happens we manually rename one of them, appending a 'b' to the datetime.

Editing

• Import all the video into PPro
• For each video file create one sequence each containing the video file
• Watch all the sequences and name them based on the camera angle, shot, etc.

Then we put together the edit using these sequences, not the source files. Note that we haven't bothered with transcoding before editing and at this stage we're just using the camera audio.

The idea of using nested sequences is so that technical colour grading, denoising and audio work etc. can be done in the nested sequence and these changes will just "propagate" through into the master sequence, wherever that shot is used. It also means if you are looking for a good area to find a good shot to noise profile for NeatVideo, or a quiet bit of the scene to grab an audio noise print, that you have the entire shot easily at hand.

Sometimes I will do a rough technical grade (in Colorista) or denoise (in NeatVideo) inside PPro while I'm editing. But this is really just to make sure it's going to work. The final work is done in AE.

When we've happy with the edit we sync up the audio from the Tascam. We place the tascam WAV files into the nested sequences containing the original video and sync them up by hand. Again, we only have to do this once in the nested sequence. Obviously we don't bother syncing the audio files in sequences that didn't make the final edit. Because the audio and video files are named by creation date it's relatively easy to pick the matching audio files.

Using nested sequences in this way isn't without it's problems. Adobe's engineers obviously don't expect it to be used this way, and you do lose some functionality. Because the files you edit with are sequences not "sources" you can't use the source window to view them, set in and out points etc. But honestly, I can't say I've missed that functionality.

Any dips, fades, transitions or colour correction we apply in PPro is all considered a draft. Most of these effects don't make the transition into After Effects properly so we leave the real work for there.

Edit Lock

Once we're happy with the edit we duplicate the PPro project and place it into a separate folder. If edit lock is correctly maintained the main project file and it's copy will always be identical.

We render out a guide video track with audio, and an uncompressed WAV.

Grading, Effects, Compositing, Onlining etc.

Once it's edit locked we import the PPro sequence into AE. AE muddles up the project quite a bit if you import with sound so we import the PPro poject into AE without audio and then add the guide WAV file (from the Edit Lock step) as a single layer.

I don't really have a set work flow once I get into AE, it depends on the project. But all the compositing, CGI and grading etc. is done in AE.

One set of scripts I find very useful are the DV Rebel Tools (http://ae.tutsplus.com/articles/news/dv-rebel-tools-for-free/). "Make Blank CC Layer" creates an adjustment layer for colour correction on top of the selected layer and already trimmed to the same length. "Make Thumbnail Comp" creates a composition of thumbnails of all your selected shots - this is very handy for visibly checking the colour balance across an entire scene.

Audio Post

This is the bit that's been the biggest pain. Ideally I'd like to be able to still treat the project as hierarchy of nested sequences and apply non-destructive changes, but the Adobe software doesn't want to play nice.

Another major problem with the nested sequences if you select "Edit Sequence in Audition" in PPro it will render nested sequences as single WAV files. Sometimes our projects have several levels of nested sequences (scenes assembled into a larger piece). Exporting to OMF does the same. It looks like AAF may just reference the original WAVs, but the AAF exports seem to be buggy and wouldn't open in a trial version of Nuendo.

Eventually after going round the houses a fair bit, we've started using the following. It works, but I'm not completely happy with it.

If any of the audio needs noise reduction we apply that at this stage in a separate audio editor. I apply the noise reduction to the original WAV or video file. Because we have the entire shots worth of audio (rather than just the bit that's made the edit) it's easier to find a quiet bit to get a decent noise print. I save the noise reduced version as a new WAV file and again I place this back into the nested sequence in PPro. I leave the original audio track(s) enabled but mute the volume. This way it will still be available in the DAW later should I want to check the original. (To be honest I would rather apply this noise reduction in the DAW, but I've found that when exporting to Audition or OMF Adobe will we re-render each bit of used audio as a new WAV file, so it's easier to do it once this way).

Then I have to workaround the inability to export nested sequences into Audition properly. Here are the steps:

• import the sequence - this time with audio - into After Effects
• use the "Unprecompose" script (http://aescripts.com/un-precompose/) on each of the subcompositions to create a flat project
• delete any layer that doesn't contain audio
• reduce the project
• export it as a new PPro project
• take a deep breath

Once it's like this I can then open this new PPro project and select "edit sequence in Audition". We don't render any of the clip effects, and we don't render a copy of the video out since we already have a guide video track from the Edit Lock stage.

Annoyingly this process of flattening the project will now mean that the audio is all over the tracks, so the first step is to select all the clips and select "Lock in Time", and then re-arrange them into a decent structure. Then it's the usual levels, EQing, etc.

Assembling

We export out the mix from Audition, and use it to replace the guide audio track in the "Grading..." After Effects project from earlier; then we render out the final video from inside of AE.

Thank you very much. Alot useful info in there!

What are people doing for large (i.e. > 1hr) projects? One giant render, or multiple renders and a 'join' afterwards.

If it's the later which codecs / containers are good for that (that can avoid another encoder pass)?

And would the audio be handled in similar sized chunks also? If so, I guess this places this some limits on J/L cuts over scene changes. And if not, it seems to me there could be sync issues arising at the 'join' stage.

Has anyone worked with photoshop elements organizer to organize ingested originals and/or use it for further (non premiere) edits for example in sony vegas pro.

This organizer seems to be more resposive than lightroom for video pics, you would have photoshop elements 11 for pictures and could use pro editors of your choise...

In Europe you find at the moment latest premiere elements 11 for 50 EUR at amazon.

http://helpx.adobe.com/pdf/elements-organizer_reference.pdf

My workflow as of today:

1. Shoot Sync sound via ART Duo Pre and Sescom GH cable (everything mounted on a rig with XLR inputs in line with shoulder plate)

2. Edit natively in Premiere CS6 (cuts and cross fades only); there used to be a problem on Windows where sound in the MTS wasn't seen, it's ok now.

   a) Prep Edit: Assign meaningful clip names (Seq1Slate2Take3) in Project Mgr and Rename footage to match (Reveal in Explorer/Make Offline/rename in Explorer/Replace Footage)

   b) Edit each script sequence in its own...well Sequence

   c) Pull all (nested) Sequences together in a master sequence (final cut); this is Picture Lock. Optionally: checkerboard pattern

   d) Duplicate nested sequences on (video above, audio below). Hide/mute original tracks. Replace (video duplicate) with After Effects composition. Looking for a similar audio workflow to Audition.

3. All picture post in AE (Edit Original). AE project to 32bpc. Sound post in Audition

4. In Adobe Media Encoder, import Premiere project and render out master to RGB codec (Prores4444, DNxHD444, Cineform444, RGB Uncompressed). This is the only time I render (forever and a day).

@radikalfilm, sounds similar to my current workflow, minus my inclusion of neatvideo in AE. ive been rendering out the denoiser in PP via bridge and saving before jumping back into AE for color work. for some reason if i try to denoise and color edit at the same time i run into hiccups. do you always use media encoder? i use PP if outputting small format for vimeo.

I have old Macbook Pro and I'm using Premiere CS6. Because my computer is pretty slow I think the best way is to transcode .mts files to Apple ProRes 422 Proxy using Clipwrap.

Question: How would you replace edited ProRes Proxy files in timeline with original .mts files before exporting into final media?

@vstardust

In the project window just right click each file and select replace media. Make sure that your sequence is set up for original files and not for proxy ones.

@dansyver

Create new sequence and don't select any of the presets, go to settings tab and from the

Editing mode: choose Custom.

Also manually select your framerate in Timebase drop down menu, If you shoot 24p in GH2 choose 23.976fps here.

Type in your frame size, 1920x1080 or whatever your footage is

Pixel aspect ratio: you should probably choose "square pixel" unless you want to de-squeeze anamorphic footage.

Fields: Select progressive or interlaced.

Audio should be at 48kHz. You can select your audio display format here, default is "Audio samples"

The bottom part is for preview settings. Those are the files that PPro creates when you render something on the timeline. When you do the final render of your project, you have the option to use those previews, so it should probably be set to be the same as main footage.

Preview file format: choose "Quick time (desktop)"

Codec: select "Apple Prores 422HQ" and type in proper frame size for previews, by default it will follow the main sequence frame size.

Now save your preset, and it will appear on the bottom of sequence presets menu. That's it.

@inqb8tr

Thanks man!

@MagicMountainMan That's what I would do myself, but I wouldn't denoise every shot - only the ones that need it. So if every shot doesn't need denoising, then I would edit the avchd first, then denoise particular shots, then render out to DNxHD. Bring it back as a single clip and cut all the splices and use a second track for crossdisolves, then xml it out to Davici.

This is my workflow, I know that there are no "right" ways or wrong ways, each person has its own methods...so I guess I'm sharing my way of workflow for web video, if anyone knows/uses more efficient one, please chime in :)

1. Import footage on a video-only disk
2. 5D2RGB batch the footage - ProRes 422 HQ, 709, Broadcast Range, MOV
3. Edit the entire footage in Premiere, audio and all.
4. Effects or titles via dynamic link
5. Export the video-only footage to animation MOV
6. Grade in Resolve
7. Export from Resolve QuickTime MPEG4 MOV
8. Adjust the new timeline in Premiere, export with audio to h264

I'm sure that I'm doing some bad steps here, most likely with mov conversions...so I'm open to suggestions. :)

@inqb8tr : thanks for the advice...

Is there any Premiere export setting I should use to bring the best possible quality to Resolve?

Is there any Premiere export setting I should use to bring the best possible quality to Resolve?

Well that's not just Premiere specific thing, and there is no special thing that you can do in Premiere to improve your pipeline. It depends on original footage, and what you want to do with it in post/delivery. For 8bit 4:2:0 AVC footage, such as GH2, you should be ok with 4:2:2 intraframe 10-bit codec for some resonable color grading, such as ProRes HQ or DnxHD for example. QT Animation is lossless at 100% and robust but it is more demanding on hardware.

I use ProRes HQ from PPro to Resolve, and sometimes ProRes 4444 if I know that I'll grade the hell out of it.