I am trying to print stems by recording busses to audio tracks in Nuendo. I do this occasionally to get my VSTis in audio form. It all works great until I try it and expect my AUX EFX to print, too.
For example, I have strings broken down to Long and Short. They are bussed to groups Lng Str & Sht Str respectively. In turn theses buss to an overall strings group for added flexibility.
I have setup up audio channels to allow printing stems on the fly.
All works if I insert EFX on the group channel insert; however, I do not run reverbs and delays this way. So, my stems record dry.
Is their a way to get the AUX EFX to print to the audio tracks. I want to do this for say a whole mix (i.e Drums, Bass, Orchestral etc)
It’s a script with which you can automate soloing and rendering tracks. It only works on windows. (I think)
The other option is to make a couple of outputs in connections (f4) and don’t set the output so that they’re “phantom outputs”. You can use those like busses in Pro-tools. So just send the audio (through an aux) to output 5 for instance and set that output as the input of an audio channel.
Sorry, I don’t understand your routing. Are you saying that you are using the reverb as a send and you want to record this separately from the dry signal?
So I assume that you are sending to the reverb from either the individual string tracks, or from the string group(s). The signal goes to an FX track (or “aux”) that only holds the reverb.
If that’s the routing, then simply have that FX track send to the audio tracks on which you record your stems. You can send from one FX track to several audio tracks simultaneously.
The problem here is going to be if you’re using one reverb for several sources. If you do this and record to several stems at the same time you’ll have all stems receive all reverb instruments - i.e. the string stem will have the reverberated drums on them. So in this case you’ll have two options:
Print each stem separately using solo/mute. Now the process (time) of recording stems will be multiplied by the amount of stems you have to record of course. Not so convenient.
Set up several different reverbs, one for each instrument group (stem). More to deal with. Not so convenient.
I’m curious about what the purpose is of these stems. I work in post mostly, but I receive stems frequently from composers. One thing I absolutely hate is when they print reverb to the stems. For a great deal of post setting up a reverb for a piece of music is a no brainer and takes very little time. But for me the benefits are that I get control over the audio. I have lost count of how many times I’ve received stems where the instruments were either completely drenched in or with unequal amounts of reverb.
So I’d definitely give some consideration to what your usage is of the stems.
The reason I set up the audio tracks in Nuendo to print stems was because I saw the video below (see near the end where he is showing his reverbs as inserts and printing to audio tracks in DP):
I wanted to be able to print stems (especially for VIs) in a single pass within my sessions to include any AUX EFX I may be using. To give me a quick way to get midi to audio or stem out a session with all EFX processing in one pass for mixing etc…
I am currently working on a video project in which the client chose to use parts of all three my temp music. My approach was to stem out each of the sessions and import them into the new video session - to align my music with how they edited the temp tracks. My goal was to retain my mix less any mix buss processing I had applied. I was trying to stem everything within my current session on the audio tracks - hence the post.
I used the Export Audio function and chose batch and exported all of my groups and EFX channels simultaneously, which at the end of the day works well; however, I have all dry tracks (with the exception of those with any insert EFX - typically I put all my delays and reverbs on AUXs) with EFX tracks in the new session. They equal my mix. I think I was just trying to keep things in tact for a time saver.
I am still relatively new to this and am trying to learn at light speed the best of How Tos.
Lydiot, with you primarily working in post, is the approach to stem within a session probably not a good one? There may be times where my files go to another mixer etc.
Print dry stems… use the same send settings when you mix
In the future, templates are your friend. build them. You get all this headache out of the way once and then you can just be creative.
Like… (just making this up)
Instrument A, get’s 3 sends. 4 tracks in total, All route to Instrument A Buss.
Instrument A buss has no ouput.
Set up an Audio Track that has Instrument A buss as it’s input and put it into INPUT MONITORING.
So everything you hear from instrument A will be coming through that monitor and will be a separate recording.
Instrument B, maybe made up of two drum machines, a few sends, a few audio tracks… All go to instrument B buss with no output. Again, set up your audio track to input monitors Instrument B.
rinse and repeat
So when your job comes in, you can spend all that time just jamming and being creative… while staying in the technical constraints of what you know you’ll need later. At any moment the client can say, hey can I get splits? and you just Record down all your busses. The audio files that you hand off will equal your mix when stacked.
Nuendo has a great batch exporting feature… however, I always prefer to re-record back into the system. When you know you’re laying off a final, sometimes you pick up on little things you missed when you were not in that headspace. When you find one of these spots… you stop, fix, then pick up the recording where you left off. At the end, just bounce all the parts to one file.
No I think it’s fine to do within a session. But you might want to consider the following two options:
1: Use Nuendo 5’s ability to do several mixdowns from busses simultaneously (audio mixdowns, or “bounces” as PT people say). This would alleviate the need to add many more tracks, route internally to them arming them etc. It seems like it would be more convenient.
2: Don’t print reverb to the stems. I’m willing to bet money that most other mixers will prefer it dry. If you want to give the mixer an idea of what you’re going for, as well as having some reverb available if there’s little time to mix, just print the reverb separately to its own reverb stem. Obviously that’s not ideal, but it might be useful in some cases. But at any rate, I’d say don’t print the verb on the instrument stems…
KDS, to clarify, when you say BUS, you mean GROUP track , correct?
SO your using GROUPs as a pass through ( well I guess thats a buss alright )
I will try this , I have been using groups to recieve audio tracks, and Groups point to stereo out…
But this is not most efficient, I would rather have the GROUP buses as you describe above point to a GROUP stem
I need a way to render stems for finaly delivery without having to go back and solo at the audio track level…
Maybe its batch export -
As for reverb, I automate at the track level with sends, via send fx channels, and inserts if unique reverb is needed in isolation for a track at track level
I know there are quite a few people that prefer this workflow. I use it myself 70% of the time or more on PT.
The one thing about it though is that if you want to be 100% sure everything is correct then you have to audition the final file which will take as long to do as the file is long. So for those who prefer to do that exporting faster than real time, importing and auditioning is actually faster.
I’m absolutely not saying I disagree with your workflow, and I think it’s a great time saver in many cases, but for those that do disagree with you with the argument that Nuendo’s punch-ins and x-fades and bounces (consolidate region in PT language) can’t be trusted clearly the best option is “offline-faster-than-realtime” mixdown…
I think you have that right. The difference in nomenclature between US/Britain PT/Nuendo etc is kind’a annoying at times…
Yes, batch export would be one way once you’ve set up the “buses”. However, you could do as KDS said as well; once your (stem) buses are set up they don’t have a direct “output”, but instead the signal that they get and sum is used as an input in the corresponding audio tracks that are set to input monitor. So whenever you want to render the stems you don’t have to solo at all, just put all those tracks into record and go. Perhaps if we write it more in detail it’ll be clearer:
-Kick
-Snare
-HiHat
-Tom L
-Tom R
-Overheads
…all go to “Drum Group”; “Drum Group” has NO output selected
-Violas
-Violins
-Cello
-Dbl Basses
…all go to “String Group”; “String Group” has NO output selected -etc
“Drum STEM” gets “Drum Group” as input and put into input monitoring mode. “String STEM” gets “String Group” as input and put into input monitoring mode.
When it’s time to actually record your stems you simply put “Drum STEM” and "String STEM" into record mode and hit record. You shouldn’t have to solo any tracks at all.
As me and others pointed out before, there are a couple of things to note about reverb:
If you have a reverb preset that you like and you apply it to different types of instruments to make them sound like they’re in the same environment;
Strings send to “My Hall Reverb”
Woodwinds send to “My Hall Reverb”
Orch Percussion send to “My Hall Reverb”
and“My Hall Reverb” is ONE FX track that routes out to the main output…
then you can’t print a stem for Strings with Reverb of the strings only without soloing the strings (or “String Group”). If you print a reverb stem with all of them then the wet/dry ratio can’t be controlled well later when working with the stems. In other words if you bring up the string stem only in the mix they’ll be less wet. If you want them to have the same level of reverb then bring up that printed reverb, which then brings up the reverb for the other instruments as well, because they were all printed together…
So the solution is to either do multiple passes of the reverb, soloing each instrument group feeding it, so you have several reverb stems, one for each group of instruments… Or you use more reverbs, one per instrument group, and route them along with the audio tracks “one level down”, so that you have the reverb on the same “level” as “violas”, “strings” and “dbl basses”, and it too feeds the “String Group”.
Don’t bother with the reverb. Pretty much all mix engineers I know that are good will have good reverb at hand in their DAWs. They prefer a dry sound and don’t want to deal with “premixed” reverb.
Lydiot, thanks for the detail, ah ha , i would opt for the KDS method, so the final stem tracks are not group tracks but audio tracks ( because groups dont have a record function)…so its audio(s) out to group category, group category out to nothing, audio(s) in from group categorys all to master “stereo out” ( your not getting into multiple stereo outs then ST 1 ST 2 ST 3…?..)
As far as reverb I would opt for more dedicated FX channels.
So your a post professional ? I am a composer who does post ( until I pick up larger budget films ) so I have full control. I use waves platinum, I believe they are about to go to 64 bit.
Curious if you use wavelab? Is it worth the 500 in your opinion ( since I already have waves platinumn) and I had wavelab about 4 years ago so I did like the tools for leveling tracks for CD production, …i suppose I could drop the stems in and level, polish them in Wavelab… and oh yeah I have the nuendo dolby encoder collecting dust on a shelf.
I’d really stress this though: Even if you have complete control now there’s a benefit to not marry a reverb with its source(s), i.e. the string section with it’s reverb blended in, when you do stems. And the reason is that you might come back to it later, or something might change and somebody else may end up doing the final re-recording mix of the actual program, and at that point with different monitoring fresh ears etc the feeling might be that there’s too much reverb - and then you’re stuck.
I’m really not a fan of composers merging their reverb with their instrument stems. It really ties my hands.
Well, now you’ll have to clarify what your tracks are used for I think… Well, first things first; no, I don’t have Wavelab and never used it. If you do music for post-production though and already have Waves Platinum then I don’t see much use in having Wavelab as well. Just seems like a waste of money to be honest.
Back to “polishing” though; The reason I ask what you use your tracks for is because some composers mixers will mix a track and slap a bunch of stuff on the master output to master it. The problem is though that when you then print the stems.
Either you solo and export mixdown, meaning that every stem will run “through” the mastering chain on the master - but independently. If you do this then your plugs on the master will work completely differently. Just consider the very different content in the stems in terms of transients and energy distribution (frequency) and how that’ll effect multiband compressors, limiters etc.
Or you simply skip that, and output the stems without that mastering stuff on there.
In either of the above cases the stems won’t sum to the full mix you delivered, and any engineer that takes your stems won’t get that final result when lining them all up with no gain.
Part of this problem is nomenclature and practicality. To many “stems” are the same as “splits”, whereas to others it’s not. And in one case (the former) it should sum to exactly recreate the full mix you delivered, where as in the latter it does not.
So that’s why I’m curious what type of post your compositions are used for. In some cases this isn’t an issue. In some cases it is.
Well, now you’ll have to clarify what your tracks are used for I think… Well, first things first; no, I don’t have Wavelab and never used it. If you do music for post-production though and already have Waves Platinum then I don’t see much use in having Wavelab as well. Just seems like a waste of money to be honest.
Back to “polishing” though; The reason I ask what you use your tracks for is because some composers mixers will mix a track and slap a bunch of stuff on the master output to master it. The problem is though that when you then print the stems.
Either you solo and export mixdown, meaning that every stem will run “through” the mastering chain on the master - but independently. If you do this then your plugs on the master will work completely differently. Just consider the very different content in the stems in terms of transients and energy distribution (frequency) and how that’ll effect multiband compressors, limiters etc.
Or you simply skip that, and output the stems without that mastering stuff on there.
In either of the above cases the stems won’t sum to the full mix you delivered, and any engineer that takes your stems won’t get that final result when lining them all up with no gain.
Part of this problem is nomenclature and practicality. To many “stems” are the same as “splits”, whereas to others it’s not. And in one case (the former) it should sum to exactly recreate the full mix you delivered, where as in the latter it does not.
So that’s why I’m curious what type of post your compositions are used for. In some cases this isn’t an issue. In some cases it is.
In recent cases, My stems go straight to the director ( not a post professional or post shop ) who then renders my stems with digital video and delivers to the film distributor per spec’s specified by the film distributor…( so my reverb settings have final say AT this point in time subject to change if it goes back to post ( back to your point) )
This is a film score with sound design, music, and dialog stems.
You made some excellent points that I need to absorb, should improve my end product.
On the master stereo out, I only have Dither, a slight low end rumble EQ cut, and a very sublte limiter to creat a brick wall at about -3db ( but now second guessing that approch because each stem will behave in a different manner when isolated via stereo out)
Compressors and other gizmos are at the audio track level or group level
How then, do you acheive stems that a post shop can drop in and get the same mix I deliver, other than take out all plug ins and let the post shop mix/master which I think is your answer ?
Most things have been covered here but I will add:
Look at the new Control Room features added in Nuendo 5 which let you set up multiple monitor paths so you can be monitoring all your individual stems either together or individually. Very helpful when mixing so you can hear what the mix sounds like at the stem level and then batch export will print the stems when you’re happy - http://www.steinberg.net/uploads/pics/N5_audiomanipilation_440x480_01.jpg
Ah ha, so a more “efficient” and elegant way to solo by stem - so if you choose to monitor "together " that would = the stereo out master mix - correct, or is this bypasing ST master out sum … I need to dig into this
If I am understanding you guys, “Yes”. You can solo your stems in Control Room without “breaking” any paths in the actual mix. It’s monitoring only. If you end up doing what KDS suggested, and recording your stems onto stem tracks as was shown before, then I believe you can let your song record (stems) and solo various stems in CT without actually “breaking” that recording… I hope you know what I mean.
If you have dynamic master out buss processing of any kind then you run into problems at the stem level. With the Control Room functions you can choose to monitor the sum of all your stem busses so you have a realistic idea of how things are sounding at the stem level when summed together (or you can monitor those individually if needed). One example of “real world” use is you can be making mix decisions on stems in context of the summed stem busses.
?