Did you hear my interview with Dave Swallow? You should. In our conversation it was obvious that he has a ton of valuable information to share, but that information is kind of hidden in his book. Let’s dig around in the dirt and pull out the gems.
Your job as an engineer is to get the best out of what you are given, even when you are given the worst.
Bored.
Swallow spends a lot of time covering information that you can find in the user manuals for mixing consoles and outboard gear. It overshadows his personal experiences with mixing, which were the parts that I found interesting AND that I couldn’t find somewhere else. If I were to edit this book, I would cut everything but the personal experience.
I found the summary of audio basics to be too brief, and disconnected from the practical goals of the book. It would probably be better served with footnotes and a glossary.
WTF Bro?
Some parts of Live Audio are just plain misleading.
For example, the section on hearing loss prevention is misleading because he says, “You can’t wear earplugs while you are mixing because then you won’t be able to hear what you are doing,” then later mentions that molded earplugs are “worth the money.” The truth is that many artists and engineers work with molded ear plugs and IEMs and enjoy slower hearing loss and a longer career because of it.
I do really appreciate this point, though: “Don’t put anything down your ear.” It reminds me of a story I heard from a friend about new military recruits that had access to medical facilities for the first time and would excessively clean their ear canals, leaving them dry and irritated. Ear wax is normal.
Some of Dave’s comparisons of point source and line arrays are also misleading.
- Hardly anyone uses a pure line array with all speakers parallel anymore. So the arrays he refers to are really two different versions of a coupled point source array, with different speaker types.
- You can aim any speaker array poorly. Swallow writes, “Line arrays tend to create strong sound reflections off the rear wall, which muddies the sound in shallow rooms.” That doesn’t make sense. You can point a point source array at the back wall and make problems just as easily as you can with a line array.
- When Swallow writes, “Line arrays require a very high ceiling because they must be tall to work properly,” I think, wow, that’s way too general. Firstly, most line arrays have hardware available so that they can be ground stacked. Secondly, he is really discussing the physics of array length versus frequency response, so his offhand reference to “tall” is useless. Lastly, Swallow suggests avoiding line arrays for outdoor events because, “As much as they have the ability to not lose many decibels over a long distance, the actual power of the wavefront that is formed seems to be fairly weak, rendering the sound pretty useless in anything over a slight breeze.” Is it just me, or is that confusing? If it’s not losing decibels then why would the wavefront be weak?
- In another example Swallow writes, “…with the point source system, you get more volume down at the front”; why? I want more explanation!
I understand that Swallow is introducing these concepts in a way that won’t be overwhelming, but you can’t just skip to the end and lead the reader to think there is nothing more to know, that things are always one way. My advice is to take his rules of thumb with a grain of salt.
Another confusing subject is the environmental effects of humidity and temperature. Swallow’s comments on directional transmission are helpful, but he makes a big deal about temperature increasing the speed of sound and having some effect on high frequencies. I read it several times and didn’t understand what he was trying to say. Here are three important things to remember about temperature and humidity from Sound Systems: Design And Optimization:
- The speed of sound in air is only slightly temperature dependent. A 1% change in the speed of sound occurs with either a 5°C or 10°F change in temperature.
- As humidity rises, high frequency transmission improves.
- The high frequency loss rate is highest around room temperature and increases as temperature either rises or falls around this standard.
And one more thing that is just plain wrong: “To measure a room’s frequency response, you use a real-time analyzer (RTA).” Why is this wrong? Please read this.
Meeting Ear To Ear
And then, a saving grace! Remember in my article on controlling feedback onstage when I condemned “ringing out” stage monitors? Swallow totally agrees! See? I’m not crazy. The world is crazy!
This procedure…is extremely annoying for anyone else trying to work in the room and is considered by most industry professionals to be extremely amateurish. Instead, you should understand what the frequencies do in terms of how boosting and cutting each frequency on the graphic affects the overall sound…Do not sacrifice the sound of the system just for a little more volume.
Also, Swallow’s explanation of why graphic EQs are pretty much useless is right on the money. I’ve already beaten that subject to death on Sound Design Live.
Hey, remember in my interview with Philip Graham when he challenged my assumption that dynamic mics are more forgiving than condenser microphones on vocals? Well, read it and weep Philip:
A dynamic mic can be much more sympathetic, more forgiving to a bad mic technique…
There are some sections that a good editor should have just cut. For example, in the section on using an SPL meter Swallow describes the fast and slow response settings: “If you have a fast reading, you’ll see the numbers changing very quickly on the display, whereas with a slow reading, the numbers change much more slowly.” Really? You just wasted seven seconds of my life. I can’t get those back now.
Awesome!
So where are the good parts in this book?
The value of Live Audio is Swallow’s personal experience. For an example in this section on how he uses Pitch Shift:
With La Roux, I use +4 on the left side and a delay of 14ms; on the right side I use -14 and a delay of 4ms. When combined with the vocal, it gives a very unique sound and sinks the vocal into the music while still keeping it loud enough to be heard.
This is gold! I had never thought about using pitch shifting in live sound before! Why can’t the entire book be like this?
When we get into the chapter on mixing is when Live Audio starts reading like great advice and less like a bad user manual. His suggestions for equalization, dynamics, and effects processing are very helpful. Also, the section on microphone technique is useful and gave me a lot of god ideas to try. Especially his five-mic array for piano.
I really appreciate Swallow’s suggestions for critical listening. I generally try to work really fast and have been guilty of trying to fix everything with EQ. Swallow reminds us throughout the book to walk around the venue and stage when there is a problem we cannot identify.
His best tip on soundchecks is to finish the soundcheck on the first song of the set so that everything will be set to go right out of the gate. Brilliant.
I’m glad Swallow talks about panning and stereo because I’ve long wondered why anyone would ever setup a stereo sound system when only a small portion of those people in the center will hear in stereo (see What’s wrong with stereo?). Swallow answers this question by explaining that while yes, stereo panning will produce a different result in each seat of the audience, it also produces a sense of spaciousness that everyone enjoys. And now I remember that Bob McCarthy said pretty much the same thing in our interview. Basically, stereo is here to stay because people like it and there are ways to accomplish it without sabotaging your system optimization.
Conclusions
I can see how this book might be good for a stage manager or technical director who doesn’t have experience in pro audio and may want an overview of how it all works. But if you are a sound engineer, skip to the sections with personal experience and definitely read the mixing chapter. Because honestly, Live Audio is not based on science. Swallow has toured the world making music, so when he says that something works or doesn’t, it’s because he’s been there and done that.
Hi Nathan – I’m not weeping over that! of course a dynamic mic _can_ be designed to be more forgiving. My argument was simply that there’s nothing inherent about condenser mics that makes them inappropriate to use in live designs – assuming the designer deals with issues like proximity effect, consistent polar pattern across all frequencies, and pop filtering. Granted there are more dynamic mics on the market that explicitly tackle these issues.
I’ve been hearing more and more from users that the clarity of my mics means that in practice they hear themselves much better in monitors at the same volume. The “presence” eq of standard vocal dynamics may (or may not) be useful in the main mix, but it’s not necessarily helpful to a singer trying to hear themselves. Note that “clarity”, just like “pop resistance”, isn’t _necessarily_ an attribute of the mic’s basic technology – but I’ve found it easier to build mics with that quality using condensers.
Anyway, I’d put that opinion of Swallow’s in the “regurgitated folklore” category. I’ve heard any number of singers popping the hell out of SM58s, and they are also extremely unforgiving of singers moving more than a few inches away.
Hi Philip! Thanks for responding to my attacks. 😉
I completely agree. Why would you ever want lower quality sound? You can make an Edwina sound like an SM58, but not vice-versa.
Let me present you with this idea: 13 years ago I was interning in recording studios and remember overhearing conversations about recording drums. People would say that they still liked tracking drums to tape because of the transient response. I got the impression that the response was slower and thus more pleasing to the ears. Do you think this is the same quality that some people respond to with some dynamic mics? If dynamic mics, in general, track audio waveforms less accurately, then it’s kind of like wearing beer goggles. Everyone looks sexy.
Interesting. I think what tape does is more along the lines of compression. I don’t know exactly what the attack and release qualities are, and it probably is frequency variant like a multi-band, but clearly there’s something about the “natural” compression qualities of tape that those engineers thought was really good for drums… at least for their preferred genre. Classical engineers never much liked that aspect of tape and in general much preferred the truer linearity of digital when it became available. And in that respect it is kind of like mics: each one definitely has particular qualities of EQ and resonance and even some compression (directionally, in 3 dimensions!) that will make it “good” or “bad” for any given source, and for particular applications. The same inaccuracy can be beer goggles in one setting or frosted glass keeping you from seeing your true love in another.
Right on. Welcome to Audio Myths Debunked by Philip and Nathan.
Generalizing is generally useless. It’s fun to talk about gear preferences, but somehow you end up creating rumors. Let’s leave that to the audiophiles. 😉
Although I’m not certain as to how serious you are about what Dave writes, I believe sound, to a professional, is highly scientific. The largest limiting factor for most is money. Only the 1% ers. have the budgets necessary to acquire any and every tool that they desire. I run other peoples million dollar rigs and I am usually sent in with little input on what I would prefer to use. The client wants a line array, that’s what they get. If they want a point source system, that is what they get. The client may dictate heavily what will be used even if it goes against conventional science. In the end our job is to make the best of what we have with the science we know.
Based on some of your responses, I am lead to believe that you either don’t know the science, or are one of many who are sold on a particular system type regardless of what would truly be the best performer? This is surmised by the comments as to why a ground stacked line array would not be acceptable and how does a line array not do so well in outdoor windy situations. Few other thoughts of yours are factored, but not important for this discussion.
A line array has a weak wave front formation because it is made up from many different elements that appear to sum together as one, but actually do not . A line array is not a true line source, as it’s elements are spaced too far apart to create a coherent, cylindrical wave front. Consequently, when outdoors and windy, sections of the line arrays sound are pushed around, severely impacting it’s performance. In contrast, a true point source system creates a coherent, uniform wave front that cannot be separated from itself and maintains superior performance in windy situations. In essence, a point source system is everything a line array system is not and visa versa.
The inverse square law explains why a point source system is louder up front than at the rear of a listening space. The science is easy for this one. The Inverse Square Law states that: For every doubling of distance you lose -6db in SPL. This changes with a line array system to a degree. A properly deployed line array beats the inverse square law by 3db to make a -3db loss for every doubling of distance instead. The only problem is that the -3db loss is only valid for frequencies in which the line arrays is long enough to actually have control over. The minimum length to have any control over a usable frequency range is about 10 feet. The array length to have full directional control down into the lowest octaves would be too long to support. Sadly most arrays do not even start to beat the inverse square law until more than 4 boxes are used. Unfortunately that small number of boxes results in little control over a limited bandwidth. Several boxes more, roughly 8-16, would be a more realistic number to have control over a larger usable frequency range. Performance is improved as the number of boxes increase. This answers why ground stacking is not ideal for line array elements. Most ground stack systems cannot go more than 4 boxes high. At this point you have limited control over a limited bandwidth and at least one, possibly even two boxes ( depending upon venue and setup ) may be pointing at nothing of interest.
An RTA is a great tool if used correctly. For system optimization, I would say it’s a highly valuable tool, that without it, consistent optimization could not be done repeatedly. The RTA when employed to time align and make the system linear is the only tool that will work. A linear system doesn’t mean it will sound good though. It simply means that in that room, the system is behaving in a linear fashion. The human ear does the rest to make the system ” sound ” good. At the very least it will do two things better than any human can using nothing but science. It makes it possible to properly align mains, subs, fills and delay lines and it will get you to a pre-determined sonic point more consistently. During a show it is only useful as a point of reference as your ears get tired or to pinpoint possible problems.
As for mixing in stereo, you don’t disagree, nor do you agree with it. I have heard many an engineer ask why would you do that? My answer is this. When you stand facing a person, you relate their position to you and you hear the voice you are looking at in the center. Now when you are looking at a band on a stage you see the band spread across it. You associate the guitar on one side, the sax and keyboards on the other perhaps? So where do you want to hear them? Right up the middle, or would you kinda like to hear them as you would the recording where they are spread apart? Taking it one step further. Any time you have the same audio signal coming from two sound sources that are spread apart at the same time and amplitude, you get comb filtering. This nasty thing we call comb filtering, would then happen to everyone not equidistant from each speaker. If we pan the instruments in the stereo field and apply some vocal ” tricks “, it will help reduce comb filtering because the instance of identical frequencies being reproduced simultaneously at equal amplitude by each speaker, will be reduced. Lastly the human brain is pretty smart. If you hear the guitar and it is coming from the side of the stage that the guitar is on, it shouldn’t be hard to figure out why. Also if you play pre-recorded media that was not mixed well for mono playback, it may also make it sound even worse. Higher intelligibility can be noticed by panning and utilizing some audio foolery on vocals.
One last trick that many don’t know about is cross panning. It really only applies to smaller stages. Using the guitar again as an example, lets say that the guy is blasting that thing to high hell. If you can beat him with the PA and want to even his beamy 4×12 out over the stereo field, you can pan his guitar to the opposite side allowing his instrument to fill out the whole area with level and again reduce comb filtering in the PA side of things. Anytime you can reduce the symmetrical acoustic reproduction content of the PA, the lower the level of comb filtering will be present. It will also increase localization of the particular instrument while also leaving a hole up the middle for vocals to pop through.
Thanks for detailing all of your experience Luke. This is great!
Line arrays are rarely suitable ground stacked unless they are aiming upwards or they are Anya or Renkus Heinz IC2. Whenever there is a straight section of a line array it will beam increasingly with frequency. This beam goes above the audiences head unless the ground is raked. Linearrays suffer from transient smear also.
Thanks Abe!
YOU SAID: “With La Roux, I use +4 on the left side and a delay of 14ms; on the right side I use -14 and a delay of 4ms. When combined with the vocal, it gives a very unique sound and sinks the vocal into the music while still keeping it loud enough to be heard.
This is gold! I had never thought about using pitch shifting in live sound before! Why can’t the entire book be like this?”
MY COMMENT: As the Beach Boys engineer during the 60-70’s, I was using this technique for concerts all over the world. Used an imported tape delay machine from Phillips for producing the effect as it was before the invention of digital delays. I’ve written about it for years. You just haven’t been reading the right books!
Thanks Stephen. The Beach Boys, wow!