Main/monitor output volume control

[Skutov]

Hi all,

I don't know how easy this would be but could we get a volume control for the main and monitor outputs. We use OBS as a live video switcher for school assemblies being sent out round the school (via youtube) and into our theatre space (via a projector window and the monitor audio output), having the ability to adjust the main and monitor output volumes would give us a great deal of flexibility with our setup.

Ideally if they could be accessible through the Elgato Streamdeck software that would be even better as that's what we use as out main control panel.

Thanks,

// Skutov

 

[AaronD]

You don't want to use OBS as a DAW! Use a real DAW instead. If you still need to stream or record audio in OBS, pipe the finished soundtrack from the DAW into OBS as its only audio source at all, to pass through unchanged.

I like this one:

https://ardour.org/

But there are lots more to choose from, with not much correlation between price and features, so you might as well pick a good free one.

 

[Skutov]

You don't want to use OBS as a DAW! Use a real DAW instead. If you still need to stream or record audio in OBS, pipe the finished soundtrack from the DAW into OBS as its only audio source at all, to pass through unchanged.

I like this one:

https://ardour.org/

But there are lots more to choose from, with not much correlation between price and features, so you might as well pick a good free one.

We're aware of that option, we've also got a full digital sound desk as well, but we're trying to keep the streaming system as simple as possible as it'll be used on a weekly basis not always by people familiar with the sound desk, also the audio inputs are arriving over NDI so it's difficult to route them out into a DAW or our sound desk.

Personally, Reaper is my DAW of choice but Ardour is pretty good too.

 

[AaronD]

Don't simplify it to the point of not working.

That's a VERY common mistake, and I think it stems from lumping together the setup and running complexities. Those are two completely different things. A given task has an irreducible complexity, but you can *move* that complexity around. You can make it easy to set up and hard to use, or you can spend a boatload of effort during the setup phase to make it easy to use.

Is there a utility somewhere, possibly outside of OBS, that can receive NDI and split the audio and video? Video goes to OBS and audio to the DAW.

Or maybe the Audio Monitor plugin can send the audio of each NDI feed to a separate channel of the DAW.

Or perhaps the Audio Monitor plugin can do do the entire job that you're looking for, but I'd still not want to use OBS as a DAW like that. Use a real DAW.

Audio Monitor

Plugin for OBS Studio to add Audio Monitor dock and filter. It allows you to put the audio of a OBS source to an audio device by adding the Audio Monitor filter on the Source in OBS Studio. This can be useful in different use cases...

obsproject.com

---

For an example of *moving* complexity away from runtime and into the setup, I have a rig that automates audio changes in the DAW based on scene changes in a silent instance of OBS.

The DAW takes Open Sound Control (OSC) messages, and the Advanced Scene Switcher plugin can send them:

Advanced Scene Switcher

This plugin will allow you to automate various tasks using "Macros". Macros consist of a list of conditions under which a list of actions will be performed. Examples and guides can be found in the wiki. Feel free to contribute! Currently...

obsproject.com

Read the documentation for your DAW to see what it wants. This one, if I remember the formatting correctly, sets the 2nd send level of the 1st channel strip to 0dB. That 1st strip is actually a 20kHz sinewave generator, and the sends go to the sidechains of some gates. Then I can use the timing controls of those gates to create a fade while the control from OBS is just on/off.

 

[Skutov]

I completely understand where you're coming from but I think you're also missing my point a little.

In an ideal world the audio would go into our sound desk and be mixed there, no question. But in this specific setup we are trying to keep the runtime and bootup complexity to a minimum, OBS is doing 99% of what we need it to do here, the audio mixing isn't doing anything fancy, just separate the two tracks that are coming in over NDI and controlling their levels, all contained within one program that we can train other, non-technical, staff members on as needed. We have a streamdeck with encoders on it, already have 2 knobs with volume controls for the 2 channels of audio, just want to add a master knob alongside so the live volume can be easily adjusted as needed without affecting the stream.

 

[AaronD]

I think I do understand what you're trying to do, and I'm saying it's not possible. Or if you do find a way to do it, it'll be practically unusable anyway.

I think you're focusing too much on the installation complexity and trying to minimize that, without realizing yet that the goal itself is irreducibly complex. So by moving complexity away from the installation, you're ADDING it to the operation! You can't get rid of it.

Take a serious look at going the other way: Be more than willing to make the installation complicated, with every part there for a known and understood reason, SO THAT the operation really can be dirt simple.

And don't be married to anything. If you bought something expensive because you believed the ad for it but didn't really understand yet (even if you thought you did), be more than willing to end up with a rig that not only doesn't use it but CAN'T even use it.

Home - The XY Problem

xyproblem.info

You seem to be better at the *asking* part of that than a lot of people, providing enough details to see the real problem, but I think your difficulty may lie in *accepting* a solution that completely scraps and reworks everything.

Sunk cost - Wikipedia

en.wikipedia.org

---

As soon as you go beyond the basic, stereotypical, bedroom streamer, OBS's audio becomes a nightmare. So pop it out of there and into a tool that is actually made to do exactly what you want it to do.

Trying to do a professional production job without the professional controls is like chasing windmills. The professional tools are the way they are for very good reasons. Just give up and do it that way. Not necessarily with expensive gear, but with a set of controls that already "just works", even if it's multiple apps.

I have a hard time believing that even a few more *minutes* (that's a lot) of automated startup time would be a problem. You're supposed to have the media gear on, warmed up, and waiting, long before you actually need it. That time is used for last-minute troubleshooting if needed.
If you're first giving it power at the last possible panicked moment, then you've got something seriously wrong with your procedures! And you should expect your operators to wing it with critical functions missing, completely at random and without warning.

---

That said, I did have a mobile "guerrilla concert" rig a few years ago, where we really did have about 10 minutes between dumping the bus's air brakes and the first note of a live song from a full band, including the unloading, finding power in a place we've never been before, marking a makeshift stage with speaker stands, etc. That was great fun! But I also did a LOT of prep work beforehand - starting about *two years* before the first show that really worked - to build a rig that was specifically designed to do that.

It also had a custom battery backup for a Pi that I buried in the 6U double-sided rack (12U if you're good at playing tetris inside), so we could just yank the power when we were done, toss it back in the bus, and go to the next stop while the Pi shut down gracefully on battery and then the battery disconnected to avoid further drain. That last part turned out to be a deal-killer for all of the off-the-shelf UPS hats that I looked at. Otherwise I would have bought and used one of those for cheaper.
I still have the electrical schematic, printed circuit board files, and source code for the firmware that went into that custom UPS. All done by me in about a month if you include cheap shipping time.

A big part of what made such a short setup time work, in addition to *lots* of testing the automation beforehand, was to parallelize everything that possibly *could* run in parallel. Set up the "B-chain" first (amps, speakers, everything downstream of the mixer), while the Pi is booting up and running its startup script, so that those two things are ready at about the same time. Then play some canned music from the Pi, controlled remotely from a laptop that had also connected to the rig's embedded WiFi by then (take the plastic case off an old D-Link router so it fits in a rack tray, and cable-extend the antennas to the outside of the rack), both to test the "B-chain" and to attract and/or create some excitement in the audience that was often already there and watching us set up.

Meanwhile, the musicians are setting up their "A-chain" rigs and plugging them into their assigned channels on the mixer and 8-ch headphone amp, so when I'm done with the quick check of each speaker, I take the laptop around the "stage" and tap/strum/activate each individual thing while watching the meters on the remote-control laptop, starting with the faster things to set up while the slower things are still working on it. Everything's still muted while I do that, except the canned music. Then I find a spot in the audience to mix from, which is the band's cue to come back on stage, and I unmute the band and fade down the canned music, which is their cue to start playing live.

Here's the mixer that I used for that:

Behringer | Not Found

www.behringer.com

Expand "Software" on the right side, and download the version of X-Air Edit that works on your system. It's completely free and is designed to be the complete user interface for that mixer. And the controls work offline too, so you can play around and see what's there. The mixer is controlled by OSC, as noted above, as is the X32 that it was pared down from, so you can automate either one from Adv. SS if you want.

If you must have physical faders for it:

Behringer | Product | X-TOUCH

www.behringer.com

I've done that too, with a crossover Cat-5 from that to a Pi that was also running the app, in addition to passing through the data for the physical faders. The Pi then had a single network connection to the XR18 on stage.
That was an interesting rig too, because it also had a "postage stamp" of an analog board in back as well, that I plugged a phone into for a jukebox and everyone else was welcome to use too, while I shutdown the digital controls and put them in a closet. The XR18 was still in charge, with the analog board going to some of its inputs, with limiters in the XR18 so that they couldn't blow tweeters anymore! I just loaded that scene before I packed up. The analog board could also take a couple of mics in addition to the jukebox phone, so a smaller event could run with just that.

The "guerrilla concerts" always had some sort of "mini-drama" in addition to the live music, and I've also done some full-on theatrical productions with that rig. Things like "The King and I", "Annie Get Your Gun", etc. Those were fun too!

• More characters than we had mics, and more characters than the 16 channel strips in the mixer too, so lots of double-patching and some scene-switching mid-show, while the backstage crew swapped the mics around.
• Sound effects from the same remote-controlled jukebox that runs the canned intro music for the "guerrilla concerts".
• Sometimes terrible mixing position in order to sell more good seats. Do most of the mixing in rehearsal, and only *tweak* live, and not too much because you're hearing it wrong in such a bad position. Trust what you did in rehearsal, despite the acoustics changing with an audience, compared to an empty room.

 

[korsynx]

I came here to add a request for a main audio output meter, and ability to add plugins to the master.

Personally I find it ridiculous that such a basic and necessary feature is missing after so long. It seems it's been requested on various sites for at least 4+ years.

This is more important than any fixes for video or UI imo, it prevents OBS from being useful for most serious uses including basic streaming and recording.

Please add this feature asap.. thanks. long time user here. Working with multiple inputs now, and being unable to see the summed output and limiter it is a major problem.

thanks

 

[AaronD]

I came here to add a request for a main audio output meter, and ability to add plugins to the master.

Personally I find it ridiculous that such a basic and necessary feature is missing after so long. It seems it's been requested on various sites for at least 4+ years.

This is more important than any fixes for video or UI imo, it prevents OBS from being useful for most serious uses including basic streaming and recording.

Please add this feature asap.. thanks. long time user here. Working with multiple inputs now, and being unable to see the summed output and limiter it is a major problem.

thanks

I agree that not having a master, really at all, is at the very least disconcerting. That signal must exist, so why not have it available to mess with? (actually 6 of them exist, one for each track)

But, if you remember what OBS was originally for - a bedroom streamer with one mic and a possibly 5.1 game, and that's it - you really don't need a master for that. And it doesn't take much more to cause all the headaches that come with being a real DAW, which the OBS devs understandably don't want to mess with. In fact, it seems that they've realized their mistake, in already going too far in that direction, and want to back up, which I also understand. Talking about having barely anything at all in OBS proper, and exposing a plugin API for the community to go to town with, for just one possibility...

So, if you're doing more than that stereotypical bedroom streamer, you should really consider taking all of your audio out of OBS and processing it in a dedicated tool instead, like a DAW or a physical console. OBS then, only knows about the final result to pass through unchanged. If there's something that first appears inside of OBS, like an NDI feed or a video soundtrack, send it unchanged to the Monitor and NOT the main Output, or use the Audio Monitor plugin or some other way to get it from OBS to the DAW/console in raw form, let the DAW/console handle it, and then send the final mixed and mastered result back to OBS to pass through unchanged.

Yes, you're running multiple apps that way, but like I showed above, they can be connected and automated with a carefully thought through and thoroughly tested one-time setup. Then you run the script that you made to start everything - you don't start each thing by itself anymore - and it "just works".

In the pro world, video and audio are completely separate things anyway. Silent and blind, respectively. They're processed by separate toolchains that barely have any connection to each other except for a common clock, and combined at the last moment to send out. So the workflow that I described is actually quite common. Pay attention to a sports broadcast, for example, and notice that the picture and sound aren't really coordinated with each other, except for lip-sync on the handful of shots where you actually see the person speaking.

 

[AaronD]

I guess if you really must have audio controls *inside* of OBS, you could set up some "dummy" sources that don't go anywhere, just to have those controls, and then use the Advanced Scene Switcher plugin to read the "dummy" controls and relay their values to the external DAW.

Again, one-time careful setup, and write a script to *always* set it up that way, and then it "just works" as long as you use that script and don't try to start it manually.

 

[korsynx]

Hey, I'm a professional producer. I didn't ask for opinions on the merits of my request.

I asked to add a master audio output track to OBS. So we can see the level and add plugins.
I'm not interested in other ideas.

Devs please add this feature as soon as you can! Thanks.

 

[korsynx]

I guess if you really must have audio controls *inside* of OBS, you could set up some "dummy" sources that don't go anywhere, just to have those controls, and then use the Advanced Scene Switcher plugin to read the "dummy" controls and relay their values to the external DAW.

Again, one-time careful setup, and write a script to *always* set it up that way, and then it "just works" as long as you use that script and don't try to start it manually.

Hello Aaron. I think you are missing the point here.

There are many ways to record. I am aware of the possibilities.

OBS is a software that is meant to make recording and streaming video and audio on a PC easier.

Adding a master fader is not an unreasonable request, as you seem to believe.

In fact it is a necessary feature, since OBS is meant to support multiple audio input sources.

We need to see the summed output to make sure the mix of audio inputs is not clipping, at the very least.

Ability to add plugins so we could add a limiter would be great.

Devs please take note. Thanks

 

[AaronD]

It's not unreasonable, as I said before:

I agree that not having a master, really at all, is at the very least disconcerting. That signal must exist, so why not have it available to mess with? (actually 6 of them exist, one for each track)

But I also said:

...if you remember what OBS was originally for...

It grew out of that, without ever breaking backwards compatibility, and it's now a mess. Seemingly simple things, that would actually *be* simple if it weren't a mess, can be amazingly hard.

I've inherited a few software projects like that myself. Usually because management was stuck on the idea of, "the previous thing with changes," and had been for years before I got there. When I looked at it, I saw an original good design underneath a giant pile of band-aids. I made the change that they wanted this time around, and of course something else didn't work because it relied on the EXACT behavior that didn't happen anymore because of that change. I chased that down and fixed it, which broke something else because it was also looking for exact behavior that didn't happen anymore because of the first fix, etc. By the time it finally all worked again, I had spent enough time on it that I could have completely rewritten all of it from scratch to the new spec instead, and it was even more of a mess because they never let me do that.
The software world calls that "technical debt".

I would not be surprised if OBS were the same way. I've seen a little bit, in other conversations, of the devs wanting to do that complete rewrite, at least of the audio system, but they're getting hung up on what the new spec should be: how broad or narrow, where to set the limits in what directions, etc. The one thing that everyone says, is that OBS should never be a DAW, and I completely agree with that! So if you need a DAW's functionality, either now or "nickel-and-dimed over time", you'll never get it from OBS.