Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Craig Smith]

I'm doing a kids stage production this week, and contemplating using a little compression -- most of the kids can hardly be heard, but once in a while you get one that yells into the mic, and I'd just like to tame them a little faster than my fingers can react.  I'll be using a first-gen Presonus 16.4.2 mixer for this.

1st: I've never played much with compression and I'm not sure what settings to use, so I've been researching the web, but they are all over the place -- some say it should be a slow attack, others around 5 ms, others around 1 ms, and Presonus' guide says .002 ms!  So I'm really confused. Suggested release times seem to be in the 40-60 ms range.

2nd: I know you shouldn't put compression into monitors, but I have to run monitors post-fader, but I'm thinking since my threshold will be really high and it should only kick in occasionally for a short period of time that it might be ok?

I also found a post here that suggested a limiter rather than a compressor, which seems like it would sound weird, but maybe that's the answer?

Thanks.

[Russell Ault]

I'm doing a kids stage production this week, [...] I'll be using a first-gen Presonus 16.4.2 mixer for this.

You have my sympathies.

1st: I've never played much with compression and I'm not sure what settings to use, so I've been researching the web, but they are all over the place -- some say it should be a slow attack, others around 5 ms, others around 1 ms, and Presonus' guide says .002 ms!  So I'm really confused. Suggested release times seem to be in the 40-60 ms range.

Console defaults for attack and release will probably be fine. The two most important controls on a compressor are threshold (i.e. how loud does something have to be before any compression is applied) and ratio (i.e. how much does a sound that goes louder than the threshold get turned down; a 3:1 ratio will take a sound that goes over the the threshold by 12 dB and take it down to 4 dB, resulting in a "gain reduction" of 8 dB). Concentrate on those two for now, and worry about the rest later (especially if you're short on time).

2nd: I know you shouldn't put compression into monitors, but I have to run monitors post-fader, but I'm thinking since my threshold will be really high and it should only kick in occasionally for a short period of time that it might be ok?

Let me go a step further: in theatre, traditionally, you should never put vocals into the stage monitors. If the kids can't hear each other, they need to sing louder, and vocal foldback isn't going to encourage that. Besides, post-fader monitors when you're actively mixing the microphones are going to cause all the same sorts of problems (i.e. inconsistent monitoring levels) that compression in the monitors would cause, so I guess don't worry about (but also, don't do it).

I also found a post here that suggested a limiter rather than a compressor, which seems like it would sound weird, but maybe that's the answer?

"Compressor" and "limiter" are typically describing two different uses of the same piece of equipment. In both cases you're simply taking a signal that goes louder than a certain amount (i.e. the threshold) and turning it down by a certain amount (i.e. the ratio). The difference is that compression tends to use lower ratios (say, around 3:1), and thresholds are set to give a more-or-less constant amount of gain reduction (on close-mic'ed vocals I tend to aim for around 6 dB of gain reduction on held notes). Limiters tend to use higher ratios (at least ~10:1—where a signal 20 dB over the threshold would be reduced to only 2 dB over the threshold—up to 100:1 or even so-called brick-wall infinity:1 ratios where the signal will basically not ever go over the threshold) but with thresholds set so that they should rarely (or, in the case of brick-wall limiting, ideally never) go into compression.

[...] but once in a while you get one that yells into the mic, and I'd just like to tame them a little faster than my fingers can react.

This to me sounds like a job for limiting. While compression can have a place in theatre (especially if you're using headset mics), it has a significant drawback: when a compressor is compressing, it's turning down the signal, and often you'll need to boost the signal back up to compensate for this. When you do that, you actually make feedback more likely, since the gain reduction of the compression will ebb and flow with the incoming levels, but the compensating boost (which is properly called "make-up gain") is constant. This can put you in a situation where there's no feedback when someone is singing (and the compressor is pushing the signal down), but the minute they stop things will start ringing.

On the other hand, limiting in this situation typically doesn't require any make-up gain because it's only there to cover surprises, not to be used constantly, so there's no significant effect on feedback. For this scenario, your best bet would be to send all the vocals to a sub-group and then put the limiter on that subgroup, as opposed to trying to limit each channel individually. This makes the limiting effect consistent regardless of how any given channel is gain-staged (especially useful if the kids are passing the mics around) and makes it easier to adjust if you find you need more or less limiting.

-Russ

[Keith Broughton]

I am currently working on my 3rd year with a community theater production with quite a wide variety of "talent" and ages.
Here is the compressor settings I use on an M7CL
ratio 10:1
knee 4
attack 10ms
release 100-150 ms
threshold- set to just get the compressor to start working with natural dialogue levels.
You should get much more compression when they yell or sing loud.
Limiter set on sub group of vocals.
No voices in the stage foldback monitors.
This may not be good for everyone but it works for me.

[Chris Hindle]

I'm doing a kids stage production this week, and contemplating using a little compression -- most of the kids can hardly be heard, but once in a while you get one that yells into the mic, and I'd just like to tame them a little faster than my fingers can react.  I'll be using a first-gen Presonus 16.4.2 mixer for this.

1st: I've never played much with compression ...........

I would respectfully suggest that this is not the time to do your learning.
Grab a couple of mics, and a couple of friends / associates and do your basic learning in a non-show situation.
In theater. stuff happens REAL fast when kids are involved.
Your best (safest) bet is to ride faders and no vocals in the monitors until you get a good hand on how everything interacts with each other.
Been there, done that, a very long time ago.
R&R with a bunch of Prima Donna's on stage is easy compared to Kids Theater....
Chris.

[Eric Snodgrass]

Kids performing in a show can have a huge vocal dynamic range - they usually go from quiet speech (especially when they are unsure about their lines) to screaming lyrics during songs. 
I find myself using a 3:1 ratio and a threshold that allows the compressor to kick in during the singing (or loud dialogue).  Then I use some make-up gain (up to 5db) on the compressor to get the levels up on the kids, especially during those dialogue scenes.  If the mic in question is shared between performers I typically end up tweaking the threshold to accommodate the difference between performers. 

[lindsay Dean]

I would also like to add an elementary comment," thanks we'll be here all week". don't forget  the low cut on all the vocal Mics

[Luke Geis]

I go with a simple rule. The vocals should always be under compression with normal speaking and a reduction that sets between -3db to -6db. How you get there is the question. For spoken word a ratio of 4:1 is a good starting point. The knee setting will change when the compression starts. I like a soft knee for those who are less predictable or more dynamic and a slightly sharper knee for those who are more consistent or less dynamic. The threshold sets when the compressor will do its thing and is relative to the knee setting. A soft knee and the threshold will need to be higher ( less ) before compression starts. A sharper knee will require a lower setting before compression starts. The goal is to get between -3db to -6db of compression that is always engaged when the person is under normal performance. So speaking vs singing will be TOTALLY different. So you will have to play with the settings in that regard. With this general rule being used, it will make it so that spurious, loud signals are squashed down to roughly -12db. This isn't an extreme amount but is enough to take the edge off.

The release and attack settings are more to taste. A short attack is generally best with spoken word as it will just always be in there. A long attack is good if you need the consonants and the initial attack of the words to get through, a setting more useful for singing. The release is simpler to set. Basically, you want the release to be set so that all compression is released by the time the next word is sung or spoken. I shoot to have it be as long as it can be so that it is only just released as the next word is said. You don't want the release to be so long that the compressor packs up on the words. If the compressor is still releasing as the next word is said, the compressor will re-compress, packing down on the next word perhaps too much. You want each word to have its own compression envelope.

As you can imagine compression settings are not very esoteric and require a constant eye. One setting is not going to work for both spoken word and singing. You will typically need one setting for each and that setting will need to be adjusted throughout the performance as the dynamic of the individual changes.

[Craig Smith]

Thanks very much everyone, and thanks for your sympathies.  I agree, a performance is never a time to learn; I had intended to "play" more in years past when I had better equipment and did sound for better performers and was more in touch, but it never seemed to work out.  I had experimented a little with compression on subgroups but forgot all about that.

I've ended up doing a lot of these types of shows but I'm not sure how I can really prepare for them given the variety and unpredictability of the performers.  A few more details on this one: it will be a musical of sorts, so speaking and singing, with almost 100 performers, with everyone having a line and only a handful with any measureable performing skill.  Most sound will come through area mics, and I will likely be on the edge of feedback the whole time trying to pick them up.  Luckily they are getting some wireless mics for key performers, so that will make some things better. 

I will have 2 dress rehearsals but it may be hard to set levels to anything really useful.  I had planned to just use a pretty high threshold and low compression.  The default attack is 10 ms and release is 150 ms on this mixer.  I had even thought about trying to use a gate to keep out spurious noise (and maybe so I don't miss a cue), but I don't have much experience with that either and the mixer has limited control over that.  So not sure it's worth trying to do anything. 

I do like the idea of keeping the vocals out of the monitors -- I've been keeping them low in recent years for this show but I felt guilty for the good singers so usually turned them up a little.  I think I will keep the area mics out of the monitors completely and maybe only put some of the wireless mics in.  I only use reverb on the very best singers and of course never in the monitors.

I had hoped to ring out the system but this mixer doesn't have any truly parametric or high Q EQ so I don't know if that's a good idea.

Thanks again for all the advice, I'll try to digest it all and maybe play a little on the first dress rehearsal just to see.

[lindsay Dean]

You might want to look at getting or renting if available, some
Audio-Technica PRO44 at $100
these are plate Mics with guards you can put them spaced along the lip of the stage.
And turn the one up closest to person about to speak.
they work really well

[Jean-Pierre Coetzee]

My favorite compressor settings to just slap on spoken word that might be all over the place is to have a very high ratio, probably around 10 or higher, make the knee as soft as it will go and set a very slow attack and release time, in the range or 80-90ms attack and 450-800ms release. Then set it to be limiting maybe 1 or 2 dB when they are speaking softly, 3-6 when speaking normally and will pretty much hard limit if they shout.

The knee pretty much makes the compressor act with a variable ratio with the ratio increasing from nothing when not compressing to whatever ratio you set when you are hitting the compressor hard. I know a lot of people don't like to use a knee but to fix someone or something that has dynamics all over the place I like to you it quite "soft".

[ProSoundWeb Community]