Not a Number
I’ve gone back and forth on writing about today’s topic for several years which might have been frustrating to some because it addresses the number one question I get as Audio Director for North Point. In some ways the popularity of this question surprises me because there is so much more to what our team does at North Point and why we have been successful. I guess I’m just naive, but I would think most of my brethren would be more interested in the other stuff. Apparently, that’s not the case, though, so the main question I am always asked is, “How loud do you mix your music?” or some other variation on that.
I get it, though. I used to ask the same question to other engineers. I still do sometimes. But I think it’s a dangerous question to ask. It’s one thing to just be curious, but what I’ve found most of the time is the question typically comes from a FOH engineer attempting to justify their right opinion that their music is currently not loud enough. This is a bad idea.
The reason I know it’s a bad idea is because I used to follow the same reasoning for asking. Don’t feel bad if that’s where you’re coming from, because a lot of us have been there including myself. Over time I’ve just come to believe what Andy Stanley said in his Comparison Trap series at North Point applies to this: There’s no win in comparison.
The loudness issue is a hot-button issue in today’s church, but it’s also much trickier than we want to believe. Robert Scovill dug in on this topic a bit in a post he did for the Sound Pro Live that I suggest you take a look at. Loudness in live sound is not a cut and dry thing that is easily measured in an empirical sense. There is a very personal and emotional side to this because loudness is ultimately a perception and not a number. In my experience, no matter how hard I’ve tried, I’ve found it is very difficult to win an emotional debate by using a number.
So before I go any further, let me just answer the question once and for all for those of you who aren’t interested in looking at this loudness thing any further and will disregard the rest of my opinions set forth in this lengthy post.
We mix music on Sundays where we feel it should be, and that determination is something that was arrived at over time and still continues to adapt to current tastes. We are much more concerned about making the mix engaging than thinking about loudness. We know if it’s too quiet it doesn’t engage, and if it’s too loud people shut down. Finding the sweet spot for engagement is more art than science, but it’s not so complicated to get in the ballpark over time by achieving consistent mixes and using measurement tools to confirm things are in the ballpark.
Sorry I don’t have an emotionally satisfying, quantifiable answer to the big question. I mentioned earlier this year at Guru’s that I’ve stopped giving out SPL levels when people ask this question, and I have a few reasons for that. Read on if you’re interested.
One reason is purely technical. We use the latest version of Smaart for monitoring SPL levels, and we also use a calibrator from Cirrus Research to routinely calibrate our mic and ensure a consistent measurement. I rarely find someone with a calibrated measurement system when I visit other churches or talk with other church techs. In fact, most guys don’t even understand that this is something that needs to be calibrated. If you’re not using a calibrated reference, then any numbers we share would be unequal which makes my measurements meaningless next to yours. But there’s more to this.
SPL, in my opinion, is problematic on its own for measuring loudness because as I mentioned, loudness is a perception. SPL stands for Sound Pressure Level which basically amounts to a measurement of the pressure of sound waves in the air. While this is a component of the loudness perception, it fails to account for the frequency content of sound.
So what about weightings? Doesn’t that account for this?
Not really. SPL weightings basically employ a filter to measure only part of the sound at the measurement mic. Since it’s a filter, it is simply removing portions of sound from the measurement. Weightings have no concept of frequency balance, and this is a key component of loudness. If you don’t believe me, play some of your favorite music at the loudest level you would like to listen to and measure the SPL level. Then take a 1k sine wave and turn it up so it measures at the same level as your music. My guess is you can’t even get the 1k tone up to the same level before you can’t take it anymore.
Loudness is complicated. While frequency balance is at play in the example above, dynamics are also at work. A pure sine wave has ZERO dynamics. Guess what? Our ears aren’t generally fans of steady-state sounds, and if you’re like me that’s something you wish portions of the recording industry would remember. Consequently, our tolerance to steady-state sounds is less than more dynamic material which in turn affects our loudness perception.
Did you know that a big reason why we have OSHA noise regulations is to protect workers from the noise in factories where machinery is constantly running? It makes me wonder why we are ever comparing live performance loudness against those regulations because I don’t believe they’re anywhere near the same thing. It’s true that sustained exposure to loud sound(s) of any type can be damaging, but I think this is something researchers are still figuring out. OSHA has different numbers than NiOSH which differ from European standards; there is no consistency on this. Plus, my guess is this is hard to study because how do you find the point where damage occurs without inflicting damage? Anyone here want to volunteer for that?
And by the way, OSHA only applies to workers. There are currently no regulations in the United States to protect audiences and congregations from high SPL levels. Think about it. While I want to create a safe and engaging audio experience for those who come into the room, is it only an engineer or band or whoever’s fault if someone voluntarily stands around in a room exposed to sound they think is too loud? Workers, on the other hand, sometimes don’t have a choice in what they’re exposed to.
So here’s a great quote my friend Jeff Sandstrom tweeted. If you’re unfamiliar with Jeff, he mixes FOH for Chris Tomlin, Passion Conferences, and Passion City Church.
“Don’t wanna hear “it’s too loud” anymore. Philly Orchestra just finished Brahms 1 with no PA at 102 dB. Stunning!”
So far I’ve said loudness depends on sound pressure, frequency content, and dynamics, but it’s still more complicated. Loudness also depends on music styles, mix styles, personal tastes, the volume(think space) of the venue, and probably a whole slew of other things beyond that. Think about it. You know as well as I do that the same people who think those electric guitars and drums are too loud might have no problem listening to the Philly Orchestra as mentioned in the quote. Loudness is very subjective.
Here’s one last example. We recently had a Night of Worship that featured music from our student ministry. Our campus pastor came up after the first song to welcome everybody and based on what he had heard during the first song mentioned in an encouraging fashion that we were going to be a bit louder that night.
So, were we really louder that night? Depending on how you understand and interpret loudness, he might have been right and he might have been wrong. I would agree that it felt a bit louder than our typical Sunday. However, according to my SPL meter, we were actually at a lower SPL level than on Sunday for most of the night and definitely on the first song he made the comment after. So what was happening at the Night of Worship?
For starters, the music was a little different stylistically. It wasn’t a big departure, but there were a lot of heavily compressed tracks coming from the stage that needed to be pushed, and the band dynamic was a little different as a result. The band was also visually more active on stage–oh, yeah, did I forget to mention that what we see affects our hearing perception….
I also made some slight mix style changes that amounted to the drums and bass being a little more aggressive. Overall mix moves were often a little more exaggerated than I typically make as well. Then there was my 2-bus compressor sitting on the left/right that I leaned into a little bit more than usual, and by a little I’m talking maybe a 1-1.5 dB.
The songs, arrangements, energy of the band, lighting, and mix decisions gave everything a different feel than what we are normally doing. In the end, these small and in many cases subtle changes resulted in a perception of louder sound at a lower SPL level. I even had someone come up to me after the event telling me with a sense of awe that he thought the system was going to blow up any minute. I didn’t bother mentioning the absurd amount of headroom I still had left in the system.
Are you starting to see that this loudness thing can be quite complex?
Here’s the bottom line I’m trying to get at. Loudness is much more complicated than a simple SPL measurement. It depends on a number of factors that in the end ultimately ride on a lot of personal and subjective opinions. So do you really think I can just throw out a number, and everyone will have an instant understanding of what an experience in our room is like?
Why do you calibrate at 1K vs 250?
I’ve never heard of calibrating to 250 Hz before.
1k is a standard reference tone that also happens to be right in the middle of our range of hearing.
One question, though… given that SPL measurements give an incomplete picture at best and are downright misleading at worst, why do you measure SPL at all?
Good question, Jeff. I can think of a couple reasons.
1.) Safety. I think it’s always a good idea to have an idea of where you’re at in terms of noise exposure. While I think there’s still research to be done, that doesn’t mean what’s already out there is completely wrong. Knowing where we’re at and that we’re safe has been a good reference to have when dealing with “too loud” complaints. It rarely satisfies an argument, but it can help get at the real issue which is they didn’t like how loud it was.
2.) Consistency. I know approximately where our SPL levels should be based on particular styles of music. I check SPL against a stored RTA trace taken with a long average, and those two together help maintain consistency when I’m mixing as well as help our other FOH engineers address issues.
I know sometimes it’s easy for me to get caught up in the moment of mixing a live band and overdo things a bit. The reverse sometimes happens when using virtual soundcheck where things feel louder in the empty room than they will need to be when the seats are full.
If I just mixed things to the way “I” feel every week, the levels would probably be a little inconsistent because my mood changes just like everyone else’s. The meter helps me reorient myself to the week-to-week average. I don’t live by the meter and go by my gut most of the time, but I believe an SPL reference helps me make more objective decisions at times when dealing with something that can be very subjective.
250Hz is the typical frequency of a Bruel & Kjaer Pistonphone calibrator. It is more commonly used in acoustical labs rather than in the field due to higher precision( and $) than the transducer based calibrators(B&K’s 4231 uses a feedback microphone to enhance stability) you typically see. Most of the ones I’ve seen are also either 114 or 124dB SPL. Obviously you wouldn’t use A weighting with it.
That’s interesting. I’m going to check that out. Thanks for the info, guys!
Nice work Dave — and thanks for referencing my SPL blog. As you note, this is a very complicated issue that standardized regulations appear to want “dumb down”. It’s not black and white if for no other reason than the fact that SPL is not one consistent level throughout the listening area in even the best designs. So, then how do you accurately legislate it? If we rely solely on the number, and you the mixer are within the regulation at the mix position, but not so in a couple of other locations in the room — who is at fault? You the mixer or the system designer integrator? As you can see — you can get sucked down a deep rabbit hole very quickly with this topic — See ya at WFX Dave where I’m certain this conversation will continue with great fervor.