Happy 10th Birthday!

Carsten,

Not a problem at all! I wrote a bit about mixing with Premiere in this blog post:

digitalfilms.wordpress.com/2017/12/02/audio-splits-and-stems-in-premiere-pro-revisited/

Here’s the short explanation. Let’s say your edits have VO, SOT, effects, and music. In Premiere you might arrange that across 4 tracks - 1) SOT, 2) VO, 3) SFX, 4 MUX. The levels and effect filters can be applied and adjusted on the clip and the track level. The track signals flow out through a master mix bus, where you can also apply effects and adjust the level.

In a more complex mix, you might have multiple tracks assigned to each group - 1-4) SOT, 5- VO, 9-12) SFX, 13-16) MUX. You can add submix tracks and route each group to a submix - S1) all SOT, S2) all VO, S3) all SFX, S4) all MUX. So now the signal flow is clip -> track -> submix -> mix out. At each stage I can adjust levels and add effects.
More importantly, I am constantly monitoring the output, so if I have to make an adjustment to an effect applied to a clip or a track I can hear the result of that change all the way through the signal chain. Likewise, I’m still working with individual clips, so I can still make picture or sound changed in the sequence.

Now let’s compare that to Final Cut Pro. I can make level changes and apply filters to clips and audio components of clips. I can’t go past that until I compound the sequence. Now I have the benefit of audio roles and subroles, which effectively act as tracks and submix buses. This is cool, because I can apply effects to the total, as well as the audio roles via lanes. But, I am only monitoring audio at that level.

For example, I have applied an EQ filter and multichannel compressor to the compound and and need to make audio changes to a clip within the compound. I have to open the compound. Once I do that the monitoring focuses on the output of the sequence within the compound, not the output after the compound. So I no longer hear the result of the EQ + multichannel compressor on the final mixed output. Therefore, I am unable to make any sort of interactive adjustments in the same way that I can in Premiere. I have to adjust the clip, step back out of the compound, review my change. If I don’t like it, then I have to repeat the process.

Now, on the plus side for FCP, is the roles concept. Ironically, FCP already has a roles mixer, but it’s just a summing bus without any manual adjustments. That is in play when you output a multichannel master file as a broadcast deliverable.

If I have to generate a master with a combination of channels - mix, mix-minus-VO, stems for all 4 groups, plus an M+E mix - then I can create that in a single output configuration using assigned audio roles. That configuration can also be saved as a preset.

The same multichannel output process in Premiere is more cumbersome, because it requires creating a new multi-channel sequence and making sure all of your tracks/buses are properly routed. So I feel that this process in Premiere is more prone to error.

However, here are some other audio features in Premiere and/or Audition, that would are lacking in FCP. Automatic ducking of music. That’s not a feature I often to use, but others want it. Writing fader automation for track levels - something FCP “classic” had. Finally remix. This is an Audition-only feature, but it’s AI-based music editing. Take a track, set a target length, and set parameters for more or fewer edit points, and Audition does the rest. The length won’t be perfect, but it will be in the ballpark.

I hope I’ve answered the question. Obviously not all of these features affect everyone’s workflow, but these are the sort of things people want when they talk about needing better mixing capabilities.

I've tried the FCP > Logic round trip quit a bit, and it sucks. Total fail compared to Pro to Audition.

I work broadcast TV and indie film formats. I am currently working a two our TV show series. Audio is the slowest, most time consuming part. I need a simply mixer with buses. FCP legacy had that, and I used it a lot. A hell of a lot. I'm really getting worn out hearing "I don't need a mixer so FCP doesn't need one" or "send it to a sound guy". Sorry, that's not good enough or workable for some of us.

The fact is FCP's growth and evolution has stagnated over the last few years. No software can sustain itself without constant improvement, or it gets overtaken by its competition. FCP is not a fully complete package yet. Why can't I put audio only clips directly into the primary storyline with one keystroke? Why can't I add video to an audio only Compound? These things make no sense. Configuring audio channels is very limited when there's no reason for that.

Before this I had a 20 year career in IT and worked for software companies. There's no reason programing wise that a roles-based mixer with buses doesn't or can't exist. Look how long it took to get professional level color tools? No, the color board doesn't cut it for complex work.

I've been an FCP fanboy since v.1 and an Apple cert'd trainer since v.4. But my patience is wearing thin at this point. My personal goal is to give FCP until Oct of this year. If I'm still using silly workaround for things all other NLEs have, I'm going to Resolve.

3-4 million seats sold according to the last time Apple made that public. And at $300 a pop, they've made a fortune on it so far. My best inside source has led me to believe the FCP dev team is taken off to work on macOS and iOS and other projects often. The Motion dev team is sizable but isolated and I've no clue how they can do so little with Motion in 10 years.

Rant over... y'all have fun with this discussion. Interesting to see everyone's unique perspectives.

FCP, to do roles based filtering and psudo-mixing, you have to Compound, and then you lose access to your edits. Go back inside the compound and you lose access to your mixing. Unacceptable workaround, IMHO.

Kartsen,
Depth-map composites. Hmm... Do you really want your composites to looks like bad Zoom calls? But yes, tracking - absolutely. The irony is that the toolkit that developers use to add tracking callout effects is from Apple. The gold standard is Resolve, Mocha, and/or Flame. So it would be great to have something internal in FCP. After all, there's tracking with masks in Resolve, Premiere Pro, and Media Composer. High time FCP had something at least as good without depending on 3rd parties.

Thanks Oliver, makes sense.
And it makes sense now that I don’t miss it because I never mix w buses. I always mix on the clip level and only use roles to create stems and apply a master LUFS or level optimise filter. And the reason I never mix w buses is that a bus is always time based. What I mean is that the bus volume keyframes do not know what content is at that time and therefore if I edit the content they stay on the time they were initially created. So if eg I keyframe the music volume I would have to redo or move them every time I change the edit regardless if the change is in that section of the timeline. (This probably also the reason why Apple never went ahead with the role mix part: it can only be time based)  The only thing that I can't do is to apply a filter per role/bus. But that's easily solved but copy-paste. Even if I change my mind and re-adjust say a voice eq I can easily copy paste again. But being able to do that on a global level, kind of like an adjustment layer, w a bus could maybe sometimes be handy. Then again if you have many speaking voices managing all those buses/roles could get confusing. Makes more sense for me to stay with adjusting at the clip level.
I can see why the extra functions you mention like auto duck etc might be interesting for some but again when you would never use them you don't miss them .

Carsten - yes, mixing in tracks for level change is indeed time-based. In fact, I’ve altered my mixing style over the years because of that.

I used to write volume level changes as track automation (riding the fader live to automatically create keyframes). These days I adjust per clip and then use the track for an overall adjustment (up or down a few dB overall). However, I usually apply filters only to the track. Some filters don’t work properly on a clip, because they need some lead-in time. This is common for noise reduction filters.

Typically a voice track would get some “repair” filters (de-ess, noise reduction, etc), general EQ, and light compression for additional volume leveling. Then I’ll add mastering compression and limiting on the master mix bus. This works the same in FCP, except you have to compound the sequence in order to apply a filter to an entire role. In short, my use of tracks/roles is a bit more of a polishing step and not purely a mixing issue.

Of course, I stick with simpler mixes inside the NLE. More complex projects typically get sent out to our go-to ProTools mixer and he can deal with it

While compounding audio ‘kind of ‘creates busses it falls apart when I have to “dig “ back into the compound clip components and lose processes that have been applied to the compound while working inside it. Especially ambient effects which have to be compounded to get reverb tails!

Once there’s multiple levels of compounding things get very confusing. When that happens I wish I was back in an audio app. But the lure of getting it all done in one place is powerful...

ltf3 wrote: Especially ambient effects which have to be compounded to get reverb tails!

Reverb tails at clip level are possible by applying a 'Hold' (speed change) to the clip (you need to possibly detach the audio from the video). Length can be adjusted by lengthening the hold. To only have a reverb 'out' I use keyframes to change the 'dry' and 'wet' parameter of the reverb just before the hold.

There is nothing that can't be done on clip level and compounding the edit should only be done for mastering to avoid unnecessary back and forth between compound and its content. At least that's how it works for me.