Print

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

[brian maddox]

A recent corporate gig with a room full of people using microphones with no idea how to use them properly started my brain to thinking...

How do i deal with the situation of a handheld Q&A mic being passed around and one person eating it and shouting while the next person holds it at their waist. 

I know, i know...  Compression.  But...

Too much and my S/N and GBF plummet.  Not enough is just, well, not enough.  What i really need is a Variable Ratio compression.  That is, the higher the signal, the higher the compression ratio, so that my Signal-to-Noise stays reasonable but i can still get a hard limiter at the point that someone is eating the mic.

Oh, and i want it to be fairly transparent.  And THAT's the tricky part.

And that got me to thinking about my experience in using my DAW for mixing and the EXTREME amounts of compression i could get away with and have it still sound fine, simply because the processing was NOT in real time.  That is, the computer could "look ahead" and see what was coming and ramp the compression up BEFORE the loud signal even "got" there.  Kinda like an Attack setting of -20mSec.  By doing this, the plugins can set themselves up for the impending compression, compensate for the inherent high end loss, compress the loud section, and then return to normal.  All without generating any significantly audible artifacts.  [i'm over simplifying, but you get the idea]

And that made me wonder.  In THIS scenario [corporate SR] adding 20 mSec [arbitrary number...  it could probably be less] of overall mixer latency would be Totally Fine.  i don't think anyone would even notice.  And if you gave the Mixer/Computer 20 extra mSec to see the loud sound coming, could you achieve the same result i have seen in my DAW?

I understand this is a thoroughly academic discussion.  Mixer manufacturers are battling to REDUCE latency, and i understand all the reasons why.  I'm just wondering if what i'm thinking makes sense, and if it does is there some way for me to leverage something outboard to accomplish this?

and...  discuss...

[Scott Helmke]

Maybe play with sidechain compression and some delay on the audio path?  IE mult the signal into two channels, use one as the sidechain source and the other as the sidechain audio plus delay.

[Riley Casey]

In the rather specific instance of corporate Q&A mics we're dealing with several phenomena associated with mic handling skills and lack there of.  The mic eater is killing the mic pre and at the same time achieving 12 or more db of LF boost from proximity, maybe even popping the occasional pppP.  The hold the mic in the lap contestant needs 20db more gain right off the bat and sounds thin even with the bass boost you wouldn't dare apply anyway.  Multing the mic to two channels set up for the appropriate gain and EQ and even a dynamic EQ insert then riding up the fader on the appropriate channel as the mic either rises to the chin or sinks to the waist seems like the best bet.  Guessing wrong and cross fading isn't any worse than frantically dialing gain and EQ during the first words.  Thats what works for me when the opportunity arises.  Not sure the delay trick makes a smart enough channel do be of benefit without even going to the people who back off on the mic after the first syllable of close talking scares the crap out of them.

[Bob Faulkner]

Was thinking the dbx Driverack 260 (maybe there's others) that can do what you are needing to accomplish.  dbx called it "Automatic Gain Control"; though, it is on the output stage of the signal flow, but may provide some utility to boosting the quiet people.  I've never really used it, but have seen it used.

[Mike Caldwell]

Around 2000 XTA came out with the C2 a stand alone digital compressor and it had a "look ahead" attack setting.
I just looked them up to refresh my memory and the look ahead time appears to be only at the most 60us as in micro seconds.

Like Scott said with a digital mixer and some creative signal and side chain routing it can probably be done.

Never used one but I always thought it was an interesting unit.

[Nathan Riddle]

MBC + parallel compression seems to me that would help a lot.

I'd ride the hpf and fader in addition

The M7 has a lookahead in their MBC, what console?

[Mike Caldwell]

Was thinking the dbx Driverack 260 (maybe there's others) that can do what you are needing to accomplish.  dbx called it "Automatic Gain Control"; though, it is on the output stage of the signal flow, but may provide some utility to boosting the quiet people.  I've never really used it, but have seen it used.

The DBX Zone Pro units have the AGC function as well.

[Keith Broughton]

It's very hard to correct bad mic use with electronics.
In a Q/A mic situation I set the channel gain so no clipping if mic is eaten.
Then set comp/limiter at 20 or 30:1 with 10 ms attack and 100 ms release. Threashold at normal speaking volume to get 3 db compression. Lot's of high pass!
It's not  a perfect solution but gives you the fighting chance to grab a fader.

[brian maddox]

In the rather specific instance of corporate Q&A mics we're dealing with several phenomena associated with mic handling skills and lack there of.  The mic eater is killing the mic pre and at the same time achieving 12 or more db of LF boost from proximity, maybe even popping the occasional pppP.  The hold the mic in the lap contestant needs 20db more gain right off the bat and sounds thin even with the bass boost you wouldn't dare apply anyway.  Multing the mic to two channels set up for the appropriate gain and EQ and even a dynamic EQ insert then riding up the fader on the appropriate channel as the mic either rises to the chin or sinks to the waist seems like the best bet.  Guessing wrong and cross fading isn't any worse than frantically dialing gain and EQ during the first words.  Thats what works for me when the opportunity arises.  Not sure the delay trick makes a smart enough channel do be of benefit without even going to the people who back off on the mic after the first syllable of close talking scares the crap out of them.

Mr. Casey i picked you at random to reply to, so i'm not singling you out i promise. 

For Clarification:  I'm not looking for a trick to accomplish this with existing mixer technology.  No existing mixing console that i know if is going to do what i'm talking about.  I'm talking about a lookahead compressor that is using 10-20 milliseconds of "looking ahead" in order to produce some crazy smooth result.  This is purely an academic exercise, not a practical one.  I certainly know all the practical tricks to use to mitigate the situation and make it "ok" to most people [including my clients].

What i want to do is replicate the results i could get in any DAW.  That is, 40-50 dB of gain reduction smoothly and with no significant artifacts.  I can do this on a mixdown quite easily [since the processing is NOT in real time], which is how we end up with modern albums having 1 dB of dynamic range.  but i digress.... 

Now, perhaps a software plugin solution could be used?  Perhaps Waves or similar?  I have a feeling such a plugin does not exist [Let's advertise our plugin has MAXIMUM LATENCY!].  But i can dream.

Again, i am expecting this to introduce 10-20 mSec of latency and i don't care.  Any more than moving the speakers 20 feet farther away from the audience would be a big deal either.  With modern processing this could probably be accomplished in a lot less time than that, but the latency is not the really big issue.  The big one is the ability for a compressor to not have to REACT, but be able to act PROactively.

[Johannes Halvorsen]

The big one is the ability for a compressor to not have to REACT, but be able to act PROactively.

Wouldn't a delayed signal with a compressor side chain triggered by an undelayed copy of the same signal achieve much of this...?

If what is needed is a variable compression ratio as well, it should be possible to implement this in something like MAX if there exists no of-the-shelf compressor plugins that has this functionality.

https://cycling74.com/products/max/

[ProSoundWeb Community]