ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1]   Go Down

Author Topic: Compression for a Livestream  (Read 448 times)

Stephen Swaffer

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2408
Compression for a Livestream
« on: September 14, 2020, 01:35:18 pm »

I am breaking this out at Brian's suggestion from this thread:  https://forums.prosoundweb.com/index.php/topic,174247.0.html

I know that broadcast engineers have this figured out-but a lot of us have been pushed into "broadcasting" via the web with the current situation.  I am making progress-but I am sure it is painful at times to those listening.  I'll try to keep it short, but give enough info for discussion.

The program:  Auditorium is from 1865, it would rather not have a PA.  Our service is traditional-but that means different things to different people.  Main instruments are piano/organ. We do have an "orchestra"-brass/flute/violins varies by service-they are mainly "extra".  Most specials are more of the old time gospel quartet style-though often with larger groups-multiple voices on each part.  We have a 40-60 voice choir.

We use a QU-32 with a "sub mix" to the livestream.  It is mostly post fader to follow the main mix (mixed by another tech) with a few tweaks.  Piano is not really needed in the mains (micing is really more for platform monitors and livestream) so it is hotter in this mix.  I also feed an orchestra mic and an ambient mic (both pre-fade since they are not really used in the main mix).  I also add a little reverb to the piano in the livemix-I have considered adding to other sources but haven't found that necessary.  Right now, I mix in the same room using sound isolating headphones-not really ideal.

I have not used compression a great deal (heavily influenced by a pastor years ago that hated compression-mainly because he had mastered the use of the microphone and knew what he wanted) so I am still learning the ropes.  I don't want to mess with the dynamics of the main mix-but I need to compress the livestream mix.  On the QU, I can compress the live stream mix.  The challenges I am facing are, with different pianists, the piano can drive the compression-killing vocals.  Also, from time to time, we do have instrumentalist use guitars with pick ups and that tend to throw things off. I understand Mac's aversion to bus compression because of this-but currently it is the only real option I have, I think?

Questions:  Would a main mix that is compressed "properly" eliminate or minimize the need to compress the submix-without killing dynamics?  As it is, we rarely adjust piano volume in main mix-and I am fairly certain pastor would not want that to change much (incidently, he is also choir director-it is critical for that).  Perhaps if we upgrade the QU to an SQ-7, then I can move the Qu-32 to a dedicated livestream mixer, compress channels individually there and we can get the desired results?  (This upgrade is already being discussed).
« Last Edit: September 14, 2020, 04:18:29 pm by Stephen Swaffer »
Logged
Steve Swaffer

brian maddox

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2628
  • HeyYahWon! ttsss! ttsss!
Re: Compression for a Livestream
« Reply #1 on: September 14, 2020, 03:30:26 pm »

I am breaking this out at Brian's suggestion from this thread:  https://forums.prosoundweb.com/index.php/topic,174247.0.html
...
Questions:  Would a main mix that is compressed "properly" eliminate or minimize the need to compress the submix-without killing dynamics?  As it is, we rarely adjust piano volume in main mix-and I am fairly certain pastor would not want that to change much (incidently, he is also choir director-it is critical for that).  Perhaps if we upgrade the QU to an SQ-7, then I can move the Qu-32 to a dedicated livestream mixer, compress channels individually there and we can get the desired results?  (This upgrade is already being discussed).

And since i'm referenced, i'll not hesitate to weigh in...  :)

So i grew up in your church. Not literally, but i might as well have. So i'm pretty familiar with what you're up against. Here's my thoughts...

In churches such as yours, Dynamics is one of the main tools in the Music Director's toolkit to convey emotion. Some of this is endemic to the style of music, but some of this is also just practical.  The Director is usually working with amateur musicians and vocalists who possess, at best, average ability and experience. It is much easier to instruct someone to play/sing "louder" than to have them play/sing with "more intensity/feeling". The latter is just much more difficult to convey as a Director and much more difficult to achieve as a performer. So your Music Director is ALWAYS going to want to have access to a significant amount of Dynamic Range for the In Room Experience TMto convey musically what they're trying to convey.

Of course one of your additional challenges is that The In Room ExperienceTM is not completely dictated by your Main Mix. That is, you don't have complete control over what the people in the room are experiencing as there is a LOT of sound coming acoustically off the stage. At best, all you're trying to do with your Main Mix is provide reinforcement to those things that may need a little bit of help getting over the already existing ambient level in the room.

There's a third issue here which is that the In Room ExperienceTM provides a much wider "Mix Window" than your stream experience. What i mean by that is that the balance between different elements can vary pretty widely from seat to seat and from moment to moment precisely BECAUSE so much of what is being heard is coming from different locations. The reason this is actually MORE tolerable for most people is that we as humans have a remarkable ability to "focus" our attention on a sound source we want to listen to and "tune out" one we don't, particularly if those sound sources are coming from different locations.

So here is how i imagine this likely plays out in your church. There are times when the piano [or organ or flute or whatever] plays quite softly and perhaps the "orchestra" or a singer or group ends up being quite a bit louder. Then there are other times when the piano is playing quite loudly and, at least objectively, is quite a bit LOUDER than the other musical elements. All of this is "tolerated" in the room because subjectively it feels dynamic and energetic and "real" and the listener can still hear what they wish to hear just by "focusing" on that thing.

By contrast, a Broadcast mix is in most ways the exact opposite of everything i just said. Dynamics are NOT your friend [especially in a streaming situation], mostly because we've been normalized to have broadcast sound being heavily compressed from decades of Radio and TV exposure. Since the sound all comes from one location [often EarPods or worse] the listener's ability to "focus" on one element over another is severely impaired. And since there is no shared "in room experience" it's very difficult to translate the energy in the room to the listener. All these combine to narrow that "Mix Window" i referred to above to not much more than a Narrow Slit that you have to somehow force the entire experience through in a way that will be meaningful and enjoyable.

The good news in all of this is that with the Broadcast Mix you DO have full control over what is being heard. Feedback concerns go out the window. The Amount of ambience that is heard is more or less completely under your control.  And you CAN control dynamics in ways that your Music Director would never tolerate for the In Room Experience.

Okay great, but how do you go about creating this mix? The simplest [notice i didn't say easiest] way is to simply use another mixer for your broadcast mix so that you can tailor ALL the settings to meet the needs of a broadcast. But that of course requires money for another mixer. It also requires another PERSON to operate it, which for most churches is the much larger challenge. You can also choose [as you have] to create a Post-Fader mix from your main mixer so that you can have SOME degree of separate control without all the complexity and headache of a dedicated Broadcast Mixer.

Which brings me finally back to your specific situation. The issue you've come across is caused precisely by the limitation of trying to serve two needs with one console. Here's what i would do if i were in your situation.

1. I would split the "problem child" inputs [piano and maybe those guitar inputs] into two separate inputs so that i could compress one channel heavily to use for the broadcast and leave the other channel uncompressed to use for the Main Mix.

2. I would try to sneak a little compression into ALL the other inputs just to tame the dynamics a little. No more than 3:1 ratio with the threshold set so that you're only tickling the compression on the loudest passages. I'd say at max maybe 6dB of compression. Any more than that and it'll start to affect the Main Mix too much and your Music Director will frown.

3. i would try to get my broadcast mix to have two layers of compression using either a Matrix output after the Broadcast Mix bus or something similar.  [i don't know your console well enough to know how easy/hard this is, but for others playing along at home...] The first layer of compression I would set to somewhere around 3-5:1 compression with a threshold set so that you're seeing about 3-6 dB of compression during the music segments of the service. The second layer would then be more of a "limiter" layer that would be set more like 20:1 with the threshold set so that you're hitting it at the loudest times of your service.

4. I would also pay close attention to the levels of my "speaking" mics relative to everything else. You're likely to need to boost your speaking mics at least 10dB or so on your post-fader sends to get the speaking portions at a more listenable level for broadcast. In fact, i suspect this one change may be more effective than all the other changes combined as it make it MUCH easier to create a listenable Broadcast Mix without constantly finding yourself running out of headroom.

5. After all of that i would very carefully adjust the output level of the Broadcast Mix Bus so that it's at the maximum input level of whatever it is feeding WITHOUT overloading it. This will be entirely contingent on what your mixer is feeding and can be trickier than it looks as often the software meters in OBS and other solution are less than optimal.

I apologize for the length of this, but i find it helpful to understand the larger picture BEFORE i decide what steps to take to help solve a problem. Hopefully that is useful to you as well.

tl;dr Turn the speaking mics up a LOT. Split the Piano input and Compress the snot out of it. Try to double compress your Mix outputs and Pay close attention to the levels going INTO your streaming device.


Logged
"It feels wrong to be in the audience.  And it's too peopley!" - Steve Smith

brian maddox
bdmaudio@gmail.com

'...do not trifle with the affairs of dragons...

       ....for you are crunchy, and taste good with ketchup...'

Caleb Dueck

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1346
Re: Compression for a Livestream
« Reply #2 on: September 14, 2020, 04:00:06 pm »

Perhaps if we upgrade the QU to an SQ-7, then I can move the Qu-32 to a dedicated livestream mixer, compress channels individually there and we can get the desired results?  (This upgrade is already being discussed).

Brian already posted a lot of great info on how to use what you have, and touched on the fact that what is really needed are two consoles and mixer-people.  Assuming this is the route you go, which I agree is the best - the question is which console to use on which mix. 

I would look closer at where to use each console.  If the needs in the room really are that minimal - the QU could handle that fine.  Broadcast has more layers of compression, multi-band compression, routing, etc - where an SQ (or even Avantis, I don't know if that would be severe overkill) may work better. 

Yes it means another room (Dante?), a pair of studio monitors, some acoustic treatment, and another mix engineer.  As long as you have the space, other than the console - it's not a huge cost.  If you have enough experienced techs to handle mixing, that should be great. 

I don't have much experience with mixing for broadcast, but the dusty college days classes of recording and mastering share the same basic core - creating a quality, 'dynamic' mix where everything is louder than everything else all the time - but never goes above -0.1 dB FS.   ;D  In addition to individual channel compression and bus (multi-band) compression - a good multi-band compressor on the output bus before the final gain stage/limiter really helps.  I bet there are a lot of articles and YouTube videos on mastering that you could glean a bunch of great tips from.
Logged
Experience is something you get right after you need it.

ProSoundWeb Community

Re: Compression for a Livestream
« Reply #2 on: September 14, 2020, 04:00:06 pm »


Pages: [1]   Go Up
 



Page created in 0.055 seconds with 22 queries.