Maybe my setup is acting differently. I’m not having a problem panning something into the 7.1.4 track - it’s where that 7.1.4 track is placing it’s content in the “renderer”.
Let me see if there is way I can reproduce it with the panners linked.
I re-read your initial solution and now I get what you are doing by having objects sent to the 7.1.4 which is not an object. I think that works fine - no different than just having the output of the track itself be an object.
I guess my question is more centered around the reverb/upmix issue though. Consider when you have a 7.1.4 reverb - how does the output of that get sent to to the renderer?
When I take the FX bus/group and make that bus into objects with the menu item in the ADM window, that’s when stuff gets strange.
Does that help explain?
Yes.
What I did was as I said send from the source mono audio track to a group track. On that group I used Stratus 3D and the group is 7.1.4. Then I can make that group into an object.
Because the source audio track is panning correctly it’ll feed signal just fine into the group and Stratus show how the input signal follows perfectly in all channels. The renderer then shows the same signal following correctly:
This I’m less sure about.
EDIT: I might have been wrong about what I see. There’s actually maybe a bit too much signal in the top channels so I stand corrected I think. Hard to tell with reverb and I can’t listen to it right now (only look).
Maybe this can help - I did a quick screen shot captured right when the audio click happens:
I am sending the output of the test click (stereo) to my “OBJECT BED”
I see the audio stereo signal going into objects 22 and 23 (which is correct) and the panning should send it just to the left and right places in space, but as you can see it is getting sent to the Top Left and Top Right channels also (in the Renderer)
Doesn’t matter if it’s direct, or a bus, or…
My concern is that any 7.1.4 bus is not getting sent to the renderer correctly.
Incidentally - what happens in your scenario that you took a screenshot of if you take your audio source and say pan it to the center on the floor going to the reverb - but bypass all of the reverb effects inside Stratus? Basically sending it directly thru the 7.1.4 bus.
I would expect the renderer to look like this:
Seeing the same as you. I think I was too tired to think straight yesterday, and before (several days ago) I was focusing on the input signal into my effect tracking my panning (which is fine) and ignored the output, and the output was ‘obscured’ because of the reverb reflections. Bypassing the verb shows me what you see.
Actually the inputs aren’t lighting up for the top channels so it’s not clear to me just where the error is…:
I also went ahead and created a Quad audio track with the signal generator on it and then sent that into the same 7.1.4 Object track. I then took three images; first is with no panning done really which means we would expect nothing in the top, and then just L/R/Ls/Rs, but we get the top four anyway;
Then I tried the “Scale” function. I restricted “Depth” so that front-to-back would be shallow. It works with the exception of the top again so on the 2D plane we see signal in sides only;
And then finally I dropped the “Height” parameter down to 0%, which should reduce top channels to zero signal, at least it seems that way intuitively. As can be seen in the 3D image showing panning it’s all flat, yet the output in the renderer shows signal still in all four top channels…;
I think it’d be good to get some input on this from Steinberg at this point, because I’m not really sure what the intended functionality is.
This first Atmos mix is for music streaming only, not music for post. So really not a special effect in my case, just some bass reinforcement. Translating well on headphone, though without a legit dolby atmos binaural downmix I am relying on DearVR Monitor.
How does this work during automatic downmixing? I think I’ve seen in some specs that the LFE channel just gets dropped rather than folded into the mains if moving to stereo for example (not Atmos-specific necessarily). If that’s the case then anything you add to the LFE goes missing.
And if you filter content to low-end only for the LFE then you wouldn’t really need to pan that as an object since localization of low frequencies is very “low” anyway, no?
Genuinely curious since I don’t normally mix music in 5.1 and up…
Thanks for taking the time to prove I’m not going crazy! I’ve tried all of the different settings as well - with the same results.
Somehow the mapping is wrong between the VST Multipanner out and the Object panning distribution - the signal is as expected.
I don’t know if it’s on the Nuendo side, or possibly the DolbyAtmos Renderer plugin side.
The issue isn’t as apparent when using a 7.1.4 reverb, but when upmixing it really creates a mess. Also, you should be able to use a L-R stereo child bus of the 7.1.4 Object bus as an output, and not have audio show up in the height channels.
Now you know why I’m pulling my last bit of hair out. I did submit a ticket to Steinberg/Yamaha months ago but…
Well, Atmos Beginner to Beginner, for this I took my lead somewhat from Alan Meyerson who is a movie score mixer these days. He does warn and practices not sending bass management to the LFE channel since that is not what it is for. But I think he puts things there occasionally. I had my bass sounding fine without it, then just reinforced it a bit so that it will sound good either way but just a bit fuller on a bigger surround speaker system. Still learning.
1 Like
I tested from a current session the action of sends going to FX that I have setup in a 7.1.2 bed.
Here are my notes (tested with the reverb off, just the receive dry signal):
With pan set for bottom only on track, the send:
-Post fader will follow the panning properly
-Pre-fader still acts as post fader, follow panning only. BUT does follow fader volume.
With main pan sent to top speakers only:
-Goes to all upper channels regardless of panning. Also, even thought being received in a 7.1.2 bed, it shows up in rear surround upper channels, too.
Also, a different issue I noticed:
Object size control -no difference between 0% and 100% for main output, pans identically instead of getting more coverage at 100% as expected. UPDATE: This is now working properly, guess I had a glitch going on. Object size very much affects the channels the object shows up in as one would suspect. Both in the bed and routed to a 7.1.4 group. Presumably other group configs, too.
I’ll check around with some Pro Tools guys on the Dolby Atmos Production Suite (DAPS) to see what they say. Recommended group on Facebook to join: Atmos Mixing Professionals
Oh, just realized, if I remember correctly, in Pro Tools you can only have an object be a mono or stereo track (input. Will pan properly to surround beds and surround master). Something to keep in mind if you are asking questions there since most members by far are using Pro Tools for mixing on a Mac.
My friend is a scoring mixer down in LA. He never uses the LFE for his music mixes he sends to the mixing stage. Sometimes they also send optional tracks which are meant to be sent to the LFE if the film mixer decides to. BTW the splits they send to the mix stage now numbers on average 64 mono tracks. Strings 5.0, Basses LCR etc.
1 Like
sounds like a good way to go for tv/film delivery. But for music streaming, I think a little bass management in the LFE is OK. Still learning, but if consumer has the sub capability, I like to give them something if it makes sense in the music.
I would probably do that too. The Capitol Records engineer says when he’s mixing he sets up a bed track for the sole purpose of accessing the LFE. He uses objects for everything else.
1 Like
Yes, that is what I do. I’ll make a copy of the track I want to “LFE” and then send to a bed if it isn’t already.
Bump for Timo and other Steinberg people. I really think you should comment on this.
@TimoWildenhain
@Thorsten_Marx
PS: Nuendo 11 Atmos “Object Beds” - Nuendo - Steinberg Forums
Hello, thanks for raising this topic.
We appreciate that you make use of the multi-object feature. In the current version there is indeed a z-axis offset error when using 3-D multi-objects. It causes the undesired audio on the top speakers.
A fix is ready and will come with the next update.
Thorsten
3 Likes
Fantastic!
Thanks Thorsten!
Hey All, Thorsten -
I parallel Mattias - Fantastic!
It’s also great that my sanity remains intact.
Let me know if you need me to test anything.
Tad Michael Wheeler