I have Cubase Pro 12 and anticipate a need to send out stems for mastering (TV, Instrumental, Vocals, etc.). We know that bouncing out stems separately will lose their inter-stem dynamics. I have developed a crude side-chain method using the entire bounced song as a trigger, but I am hoping there might be a better solution. If not, I’ll just bounce each stem out separately and make each one sound good 'nuff.
Hi @sled ,
I’m a bit in late, maybe, but I have your same problem.
Today I casually found this video by Anne-Kathrine Dern, a film composer based on LA; it could be the solution but I’m still trying to understand it well…
Please let me know if you have found a way, in the meanwhile.
Best,
Art
Using the full mix as sidechain is very slick if you ask me. 
You can probably also do this with bussing/sends though, no? Create a buss and send each track at 0db to that buss. Then send that buss to sidechain input. (I’m not at Cubase right now but I assume this is possible?)
I used to wonder the same thing, but I realized if you’re going to hand off stems to the post mixer, it is best to leave the dynamics decision to the post mixer. Because the post mixer has to deal not just with your music, but sound effects and dialog. That entire mix in of itself requires its own dynamics.
If you really need so, get in touch with the post mixer and let them know what sort of a mix you’re going for - but don’t expect too much because your music has to make way for other elements.
Yes, I think it is possible.
After seeing her video I tried to set up a real time version, where you don’t have to render an audio file. I use Cubase Pro.
Additionally to the one output bus that she uses you would need a second one and a Group Track.
The second output bus will be identical to the first output bus, ie. you can simply copy the channel settings of one to the other.
In the project you use Direct Routing with Summing Mode = On and…
- route every channel that went to the first output bus to the Group Track instead
- route every channel that now goes to the Group Track additionally to the second output bus- except for those that you don’t want to include in this mix
Then you set up the side chaining on the plugin(s) of the second ouput bus to the Group Track. The whole reason for using a Group Track is to be able to select it as the feed for the side chaining. That is not possible with an ouput bus.
Then, for the audio export you can queue two jobs. The first to export output bus one, the other to export output bus two.
Yeah, I personally don’t bother with any of this. To do so much work probably means the production has a small budget, and the end product will be butchered at the final mix anyway. It really is a waste of time.
I only have one group ‘Out’ with my mastering pluggins, and each file is exported through that. It sounds good enough, and in the nearly 15 year career, I had no complaints.
For top-tier mixing, the Spielbergs of this world have to cough up the cash and hire a scoring mixing engineer.
Yep. Many production houses in my neck of the woods don’t even know that they need to hire a score mixing engineer plus a music editor as well as a composer. They just assume the composer is the one that gives the final product. So not only do I have to wear many hats (compose exactly to sync, and mix) - I have to discount my fees as well. Alot of so-called “professional moves and steps” are just simply not worth the damn effort and time, and the miserable money you get paid for.
There is a huge disconnect between professionals who take for granted their top-tier pay, lecturing the smaller peasants about “how to be as professional as them in order to command more pay”. Well, is it the chicken, or is it the egg? I tried demanding more and ended up losing jobs to cheaper alternatives many times. I live in Asia, so that’s not surprising. We often never take CULTURE into consideration.
Oh yeah, of course. Do the best you can, but don’t go crazy about trying to deliver DUNE quality sound mix. Low-budget films will ask for stereo stems anyway. It really is not worth the hustle.
I have come across situations where my music, which I painstakingly mixed, was brutally EQed with the low bottom cut completely, and the high-mid also subdued in order to hear the badly recorded dialogue. The whole thing was further cheapened by amateur recordings of ambient sound.
Not saying that all films will end up like this, this was an extreme example, but yeah, these films with small budgets need work from the composer that is reflective of said budget, and there has to be a degree of realistic expectations from the production.
Hmmm, IMO, you can’t command more pay, unless the production has the budget to pay more.
For more pay, one has to knock on the doors of directors/producers who can provide such pay. Production companies don’t care if you can sidechain the compressor. They won’t know what you are talking about, anyway.
So, any rhetoric from YouTubers making such claims is just clickbait.
These YouTuber Composers outsource their mixes anyway; the whole thing is very disingenuous…
I wouldn’t be too discouraged about films taking cheaper alternatives. Small budget films end up ignored by the general public; don’t lose sleep over it. Stick to your guns.
I understand the Anne-Kathrin comes from the music-to-picture group of Cubase users. But this is actually helpful for people who want to create different mixes of their song, like full mix, acapella, and instrumental.
I’m all for YouTube Cubase teachers many of whom I follow and watch regularly. But having people stating (as per a previous post) that “do this trick to demand/achieve higher pay” is not rooted in reality.
However, if other type of creators are also helped by any of this, by all means.
Hi everyone,
and thank you very much for every answer.
I’m not a film composer but I was curious about this workflow because occasionally I have to export stems if I need to work with other musician/mixers, etc. and I still have to completely understand the correct way to do it (I also refer to this Dom Sigalas video regarding full projects backup that for me is much more important, at the moment).
What I’ve done, trying to follow AKD video, anyway, was:
-
exporting a rough mix with master effects
-
exporting the same mix with NO master effects
-
setting the second mix to no output but in solo defeat mode
-
sidechaining master effects to this second mix tracks
-
exporting stems with master effects active but sidechained to the second mix
-
Comparing the first rough mix with all the stems BUT here I don’t get the same output level so I’m missing something for sure but I don’t understand what and this is what makes me stubborn in trying to (understand).
@Johnny_Moneto I’ll try your way for sure; hope I’ve understood it correctly. Thank you for taking the time to try it and replying! (I use Cubase Pro too, still v.14)
Art
Here a screenshot from a minimum setup example, showing the principle.
On the left is an audio track that represents your project.
On the right is one Group Track called PreMaster. All signals are routed to this group, it acts as the final summing bus for the full mix. It is routed to the first output bus, here: Stereo Out.
Mastering plugins will have to be placed on the output busses. As an example I loaded Frequency2. It is easy to have the same settings on both busses as you can simply copy & paste the settings of one bus to the other.
We see the UI of Frequency2 from the second output bus, here named Instrumental. The side chain input for band 5 of the EQ is the PreMaster group channel.
As you can see the band 5 reacts to a signal, even though no audio runs through it.
For the audio track we use Direct Routing with Summing Mode = On in order to send the desired signals to both, group channel PreMaster and output bus Instrumental. The latter one of course only on the channels that are supposed to be part of the alternative mix.
Art,
Yea, I watched about 10 minutes of her video. I like that she bounced a dry stereo mix and then applied the 2bus master processing and sidechains to the stems using a ghost track to trigger those effects on those stems. I’m thinking that if you performed a null test of the original 2Bus Stereo Processed Mix with the sum of all the processed stems as described in her method, you will probably find that they do not cancel out, but probably not enough to make a difference. I also consulted our friend ChatGPT. It laid out a more Group centric approach, but you can check that out yourself. BTW, ChatGPT also said the following, “Music Supervisors often prefer clean stems without interdependence. They want flexibility to rebalance in post.” Then there is Colt Capperrune’s take on the subject: https://youtu.be/nr4KWBs86-A?t=563
Hope this helps.
HI @Johnny_Moneto ,
I tried you settings and it’s very useful; I had to use the classic Cubase compressor as no one other internal dynamic VST has side chain option apart from Black Valve, that I still don’t know; but for a test purpose it’s ok and everything was ok.
I had a super simple setup with drum, bass and an organ playing chords and I used Instrumental bus to export an alternative mix (only bass and drums) and I got it; obviously having the same compressor on a different number of tracks give a minor final level in the alt mix but I guess I could put a second compressor with different settings to compensate, is it correct?
When I tried AKD way, on the opposite, I could I had a compressor and a Neutron 3 elements on the master bus but as I saw compressor activity in the compression level meter I couldn’t see no apparent activity in Neutron and this confused me even more…
That said my last question is: if I follow Dom’s video, just for a backup purpose, and I export each single track with his full path, including master fx, when I compare the sum of all tracks with the output stereo mix stem I have different output level, is it correct or I’m missing something once again?
Thank you so much for your detailed answer and I hope I don’t bother you with all my question…
Best,
Art
Hi @sled, thank you for replying; I actually did a null test after trying AKD method but It seemed to me I had a bigger difference in my results compared to her and I can easily guess I did something not correctly.
Thank you for this video, very useful to me, especially at 9’50” where Colt confirm that single stems sum level is a bit a different from the stereo mix one and this, at the end, is very logic and Colt explains it very well.
So, to resume, understanding what one’s is printing, how and why is fundamental to do a good job; I only have to understand side chain way correctly but this thread was really very useful to me!
Best,
Art
One final question, after a number of tests:
what’s the purpose to set side chaining to export stems in Master/Group/Sends (CPSM) mode that, played together, don’t give the same level of full rough mix when I can simply obtain the same result exporting them with insert+strip (CSP)? 
I think the whole idea is not only about levels (anyone, correct me if I’m wrong). It’s about compression, clipper, and so on, (from the main final mix bus where you apply the mastering effects) to react to each stem as though each track is not exported individually but as part of the entire mix, as such, having the ‘glue’ effect achieved to the stems, the same way it would sound to the stereo file.
I suppose compression will keep rogue transients in line.
In addition, I am sure Cubase 15 now exports side-chained plugins with the sidechain bounced correctly.
Check this guy out. I think he is using your trick but explains a little more in depth, hope it’s of some help: Sidechain Stem Mastering: The Complete Guide
HI @somecomposer ,
I was ‘focused’ on levels as AKD said that there’s kind of a rule for exporting stems to film/media composer but I am note one, and what’s important to me - at the moment - execept some rare occasions, is to have a really complete backup go my projects once they are done, more like in the Dom Sigalas video I linked above.
I’m still on 14 but I’m planning to upgrade, even I’m still following all ‘bugs’ reports from users (the one about Project Logical Editor is not very inviting); there are features, on the opposite, I really would appreciate to have, like new kind of folder/group track and others.
I’ll do for sure even the sjidechaining one it’s not exactly a trick of mine 
Thank you for answering,
Art
Hey, it’s a real pleasure. I hope I was somewhat helpful.
I hear you. I didn’t watch the whole video by Dom, but I suspect he is probably bouncing everything in audio. His channel is a real gem. I regularly watch the guy plus a few other great instructors. It’s an ongoing thing with technology, I keep learning.
Yeah, same here, I never switch to a new Cubase version until a year to a year and a half later, until the version has been stabilized.
I’m still on 14, no hurry to swap, but I must admit, the folder to group AND the video exporting capabilities that they added are amazing additions!!
Good luck with your music!
Absolutely!
Yes, he does; very interesting video. And I’m still learning too…
Curiosity can be dangerous and I sometimes fell for it
Gettin’ older I’ve learned to be a bit more patient as there is hardly anything so indispensable one can’t wait a bit longer…
Thank you! Best wishes to you too!
Art