Monitor Speakers and how to get the best sound out of them

Monitor speakers are a necessity for most of our recording. But I have always had a problem with the sound of speakers, and I explain why in this episode.

Speakers always exist in a space with  boundaries, even outside. Our control rooms tend to be small, and making a small room sound good for performing or listening is a challenge. The monitor speakers and the control room form a system. The acoustics of the room have a big impact on the sound.

I also compare the sound of headphones or earbuds with speakers, and provide some advantages and disadvantages of each.

And I talk about immersive formats, like Dolby Atmos and the increased requirements for the speakers and the room.

email: dwfearn@dwfearn.com
www.youtube.com/c/DWFearn
https://dwfearn.com/

Transcript

102                        Monitor Speakers                                             February 20, 2025

 I’m Doug Fearn and this is My Take on Music Recording

 The loudspeakers in your control room are a tool that is vital to the success of your recording. And the sonic characteristics of that control room have a major influence on how those speakers sound.

When I first heard the term “loudspeaker” when I was a kid, I thought that was about the most hilarious name I had ever heard. It was hard for me to take something called a loud speaker seriously.

And now, many decades later, I have different problems with speakers.

Of all the fine equipment we use in recording, there is no doubt in my mind that speakers are the worst at doing their job.

That’s generally true of all transducers, like microphones, headphones, ear buds, and cutting heads on disc lathes. We can make our electronics remarkably true and consistent, with minimal impact on the sound – compared to speakers.

Microphones are not quite as bad at their job, and we have learned to utilize their shortcomings and idiosyncrasies to our advantage in creating the sound we want.

But our monitoring systems are definitely the weak link.

And it’s not just the speakers themselves. It’s their interaction with the room they’re in that adds to the problem.

Of course, speakers are necessary in most recording situations. It’s the only practical way more than one person can hear what’s going on. And good monitor speakers in a good room can sound remarkable.

However, I have always had misgivings about the sound of speakers. They always sound like speakers to me, not the actual sound. There is something artificial about the sound you hear from speakers that reduces their credibility. I have never been fooled into thinking something was live when it was actually coming out of speakers.

I know they are a necessity. We simply have to learn how to get the best possible sound out of them.

There are hundreds of monitor speakers available today, or in the past. No two really sound the same. To a large extent, what we prefer is determined by our past experience.

And even a fairly crummy monitor speaker can be used. You just have to understand how what you are hearing translates into the real world of listeners.

How do you get the best possible monitoring in your control room?

First you have to determine what speakers work best for you. You may have always used a certain speaker and feel comfortable with it. You know how to record and mix using those speakers, based on how your recordings sound in a variety of listeners’ typical systems. Those may be the ultimate speakers for you.

A problem may arise, however, if you have other people using your studio for their projects. Your speakers may sound terrible to someone accustomed to the sound of another monitor speaker. Not much you can do about that except to have several monitoring systems available.

And if you do work in other studios and they do not have your preferred speakers, you might be able to bring yours along to the session.

Even the best speakers have frequency response anomalies that could easily be plus or minus 5dB. That’s about the best that speaker designers can achieve, although newer speakers are far better than most older designs.

If you were to apply 5dB of equalization on your mix, you would certainly hear that. You probably wouldn’t like it. But our speakers are doing that to us all the time.

There are a couple of ways speaker designers test and measure their speakers. One method is to go to a large, flat field and bury the speaker, face-up, so that the dirt or sand is even with the front of the speaker cabinet. Then a calibrated measurement microphone is placed a certain distance away, typically 1 meter or more. The speaker is fed with either discrete sine wave tones, a sweeping tone, or white noise. The resulting measurements are graphed to show the frequency response of the speaker.

In order for this measurement to be as accurate as possible, there can be nothing that reflects sound for a considerable distance away from the speaker being tested.

The goal is to eliminate the effect of a room, with its reflections, from the measurement.

Another method is to test the speaker in an anechoic chamber. This is the ultimate dead room. The anechoic chamber has sound-absorbing material many feet think on the walls, ceiling, and floor. These chambers take up a lot of space and they are expensive to build.

The other characteristics of the speaker, like harmonic distortion can be measured in the same setup. And the way the speaker sounds as you move away from directly in front of it can also be plotted.

In any audio equipment, the goal is to have a perfectly even level at all frequencies that our hearing can detect, conventionally from 20Hz to 20kHz. When graphed, this looks like a flat line with no variations. That’s where the term “flat” response comes from.

It is relatively easy to make our electronic tools, like mic preamps and outboard gear, consoles, power amps to drive the speakers have perfectly flat response from 20Hz to 20kHz.

It’s much more challenging to achieve that with any transducer, including speakers.

Why? Well, it is an electro-mechanical system to convert electrical energy into sound waves in the air. Not easy to do over a 10-octave range.

In the early days of loudspeakers, the technology was so novel that people were delighted to hear anything reproduced from a record or a radio broadcast. Those speakers from the 1920s had very limited frequency response, maybe from around 200Hz to 2kHz. That was considered hi-fi at the time. And the frequency response deviated widely over that range.

One of the problems was the it was not possible for any speaker to cover more than a few octaves with relatively good flatness. Beyond a certain low or high frequency, the speaker would not reproduce much of anything.

The solution was to use more than one speaker, often just two. The low frequency speaker was called the “woofer,” and the high-frequency one was called a “tweeter.” That added two more ridiculous-sounding terms to my vocabulary as a kid.

But you couldn’t just drive both speakers at once. There was a frequency range where the sound would come out of both the woofer and tweeter, which didn’t sound good at all.

The solution was to use a simple equalizer, called a crossover network, to filter the audio and send the appropriate frequencies to each speaker.

Now we have introduced another potential problem. All equalizers have phase shift, which can make the music sound strange. And that applies to the crossover network as well.

As speaker manufacturers got better at crossover design, the sound improved. But the transition from one speaker to the other was still often audible.

Many monitor speakers utilize just two speakers, or “drivers” as they are often called: a woofer and a tweeter. They are called “two-way” speakers.

Typically, the woofer covers a range from somewhat below 100Hz, up to around 1.5kHz. The tweeter takes over from there on up. The lowest frequency the system can reproduce depends on many things, primarily the size of the speaker and the size of the enclosure. Most near-field monitors are small. They can sit on the meter bridge of the console, or on dedicated stands. They were designed to be only a few feet from the listener, at most. Most of those speakers go down to around 50 Hz. Better designs might extend the response down to 40Hz, but the response starts dropping off before that limit is reached.

That’s 4 to 6 octaves.

It’s better to think in terms of octaves, since that is they way our hearing works.

The tweeter covers the range above 1.5kHz. Like the low end, the tweeter starts to roll off after some point, often around 20kHz for better systems. That’s less than a 4-octave range.

Some speakers are designed with three drivers: a 3-way system. That middle-range speaker reduces the burden on the woofer and tweeter.

3-way systems need to have two crossover networks. That doubles the potential for strange things to happen to the sound, versus a 2-way system.

Personally, I prefer 2-way systems. But you might prefer the characteristics of a 3-way system. Like most things in our business, it is very subjective.

Most music still has some content below 40 or 50Hz. Some genres rely on that bottom octave from 20 to 40Hz for their sound. And you need to hear those lowest frequencies to make sure there are no problems going on in that bottom octave or two.

You could install bigger speakers that go down that low. Or you could simply install a sub-woofer, which will handle that lowest octave.

We take advantage of the fact that our hearing is largely insensitive to the direction a low-frequency sound is coming from. We can’t tell where the sound originates below around 50Hz or so. We can fill in that bottom octave with yet another speaker that covers only those frequencies. And it does not have to be located near the main speakers to be effective. That’s yet another crossover.

 For many years, the monitor speakers I used were Altec 604s, or later versions. The 604 is a two-way system, using a 15-inch woofer and a horn speaker for the high frequencies. The horn is mounted in the center of the large low-frequency speaker. That arrangement is called coaxial.

This was the standard studio monitor speaker for decades. Were they perfect? Of course not. But everyone in the business was accustomed to those speakers and they did an excellent job.

The 604 reproduced lows down to 40Hz very well, and the response was down only slightly at 20Hz. The highs went up to around 15kHz very well and then gradually rolled off.

One advantage of the 604 was precise sound location. And they were highly efficient speakers, so they could be played very loud with only a moderately sized power amplifier.

The 604 was a large and heavy speaker.

The 604 was supplanted by the UREI 811 in the 1980s, but that style of monitor speaker lost favor soon after that.

Most studios went to smaller, more modern speakers, which worked well. But to this day there are times I wished I still had the 604s, just because they were such a dependable, known-quantity to me.

Most modern speakers include the power amp to drive them right inside the speaker cabinet. That is convenient, and eliminates long cables between the amplifier and the speaker. The resistance of the speaker lines can degrade the sound.

There is no need for a conventional crossover network. The audio can be split into highs and lows and then fed into individual amplifiers to drive each speaker. The low frequencies require a lot more power to drive their speaker than the highs. That results in smaller power requirements overall, which reduces the heat produced.

But self-powered speakers have a downside. The amount of heat that can be dissipated is limited by the size of the speaker enclosure. You don’t want cooling fans inside your speakers. One solution is to use digital amplifiers to drive the speakers, usually what are called “Class D” amplifiers.

I like the sound of high-resolution digital audio, but I don’t like the sound of Class D amplifiers. But they are very efficient, which reduces their size, weight, and the heat produced. They are also much cheaper to manufacture. But I have not heard a Class D amplifier that did not have distortion to my ear, especially at higher volume.

My ideal monitor system would be a two-way system with a passive crossover, driven by a good-sounding amp, ideally a vacuum tube amplifier.

I prefer the sound of soffit-mounted monitor speakers. That is, the speakers mounted flush with a surrounding area that effectively extends the front panel of the speaker. In my large studio, I designed the control room around the soffit-mounted Altec 604s. They were above the mixing position, angled down and in, so that the sound converged on the mix engineer.

In my first control room, the speakers were mounted similarly, but hung on chains. That free air space around those speakers was not helpful to the sound. The same speakers mounted in a soffit were much more uniform in frequency response and the localization was better.

That’s not a popular approach these days where most speakers are placed on speaker stands or on the console or desk. That’s the way my current monitors are placed. I wonder if the sound would be improved by soffit mounting. That’s an expensive experiment.

It is a frustrating fact of studio life that as you find the best recording equipment, each improvement reveals defects in everything else.

For example, when I upgraded to better converters, I started to hear unacceptable distortion in most of my condenser mics. That became even more obvious when I installed better monitor speakers.

That’s one example. There are other things better equipment will reveal. For many people, it is of no consequence. But it matters to me. And I suspect there are a lot of other engineers who have found the same thing. This is one reason why I think the D.W. Fearn products are so successful.

The sound of all speakers is influenced by the room they are in. It’s really a system, with the room affecting many aspects of the sound you hear. The most obvious is frequency response.

In a previous episode called “Improving Acoustic Spaces,” I discussed why some rooms sound better than others, and offered some suggestions on how to improve the sound of your studio.

Those problems are also a factor in your control room or other monitoring environment. Remember, the larger the room, the better it can sound, as a general rule. Control rooms tend to be smaller than studios, and that makes taming the sound of a control room even more challenging than making a studio sound good.

You might want to listen to that episode on “Improving Acoustic Spaces,” if you want some ideas on how to improve the sound of your control room.

As a gross example, imagine a control room that is 10x10x10 feet in size, with typical drywall on the walls and ceiling. Add your speakers.

That room will sound truly awful. No one would use such as space, since everything is wrong with it. The same dimensions mean that there will be a huge peak in the frequencies determined by the 10-foot dimension.

How huge? It could be 40dB. And there could be 40dB nulls, too. You would only have to move your head slightly to completely change the balance of the mid-range frequencies.

What if you had an infinitely large control room, with no reflections at all? That would make the frequency response more even – given the limitations of the speakers. But it would not sound good. We need some degree of sound reflection in the room to bring the music to life.

Those are examples to illustrate the extremes of possible control room design.

The room size and proportions are fundamental in determining the sound of the space. Those have to be decent to begin with.

And then the proper acoustical treatments must be used to make the room sound good for listening to music.

No matter how well-designed a control room is, it is not going to sound excellent at every point in the room. Good design focuses the best-sounding area at the mixer’s position.

A typical control built without consideration for proportions and treatment could easily have a 10dB variation in frequency response, both peaks and nulls. The sound will change as you move about the room, sometimes over a very short distance. Where is it right? Perhaps nowhere. But even if there is a sweet spot, it is likely to be narrow and not really perfect.

If you go into some classic old control rooms of the 1950s and 60s, the sound in there is pretty awful. And yet they put out classic records we still enjoy today. A good control room makes our job much easier, but a mediocre room can be made to work if you recognize the limitations.

Sometimes you can tell a bit about the monitoring environment where something was mixed by the odd sound. There may be deficient mid-range frequencies, due to the engineer scooping out the middle to compensate for the lousy room. Or the bass may be out of proportion, either too loud to too soft, as the engineer tries to make it sound right.

An engineer who works in the same room day after day learns what must be done to make the recording sound good in typical listening environments. The mixer might know that he has to boost the lows more than what sounds right in the speakers, in order to compensate for a room that is too bassy. Or vice-versa.

There are a couple of other factors that influence the sound of the monitoring system. One is the reflection of the sound off the mixing console or desk in front of the engineer. This was identified many years ago when large-format consoles were in every control room. The sound was degraded by the reflection from the hard surface of the console. Even though there are lots of knobs and buttons that might provide a bit of diffusion, a console is mostly an acoustic mirror. That reflection can affect the sound, and make the monitoring position more critical both horizontally and vertically. There isn’t much you can do about that in a control room with a large console. Glass in the windows can also confuse the sound for the mixer.

Today, many studios do not have a console. But they do have a table or desk for the computer displays, keyboard, mouse, and perhaps a small controller. In order to minimize those reflections, the area in front of the mixer should be as small as possible, perhaps only large enough for the keyboard and mouse. The computer displays can be on floor-mounted stands.

That is uncomfortable to most engineers, but it may be a way to improve your monitoring if it is otherwise good to begin with.

Some control rooms have equalizers on the speaker system to attempt to make the frequency response flatter. I see a major flaw in this approach if the room is already compromising the monitoring accuracy. Perhaps in a superlative room, equalization might even out some minor problems inherent in the speakers. But that does not apply to most of us.

In the 1970s I was involved with a scientist who developed one of the first analysis and corrective equalization system for control room monitoring. My studio was an early test bed for this technology because it was conveniently nearby for the inventor.

Over a period of a couple of years he went through several iterations of his system. I would take some downtime in the studio and donate it to him for experimentation. I liked the concept and was interested in what might result.

We evaluated my control room and applied the necessary equalization to achieve flatter frequency response. I was excited about this and tried to convince myself it was an improvement.

But it felt very uncomfortable to me. My engineers had the same reaction. We thought we would get used to it, but we never did. After a few weeks, I took the eq out and we all felt a lot better.

I thought I was an outlier in this experience, but in talking with many engineers, they all tell me they don’t like the sound of equalized monitors. They take the eq out after a while.

And what is the problem we’re trying to solve with eq? Is it to compensate for room resonances? That will only help in one precise location. If you move your head some distance, the sound will change, with or without equalization. I don’t like that, but your experience may be different.

You already know about my discomfort with most speakers. But I want to broaden the picture somewhat and talk about the fundamental flaw with the loudspeaker concept.

For all the reasons I have mentioned, and more that I have not, there has never been a speaker system that fooled me into thinking something was live in the room. I’ll bet you are not fooled, either, if you listen carefully to the contrast between the real sound and what comes out of the speakers. It’s not the same, and at our level of technology, it never will be.

This is not a fatal flaw. Most often, the purpose of a recording is not to reproduce the sound that occurred in the studio or in a performance space. It is to convey the intent of the music creators in a way that is effective and satisfying for the listener. We bend reality often. We can’t give the listener the true experience, but we can give them something better, in most circumstances. What we do isn’t “real,” but it is effective.

That is a good thing. And it takes the burden off of our imperfect speaker systems, as long as we know how the ultimate listener is going to experience what we recorded.

I have talked only about speakers so far, but there are other ways of hearing music. Namely, headphones or ear buds. I will lump those two together and just refer to them as headphones.

One key difference is that we have eliminated the contribution of the listening room when we wear headphones.

And the sound is channeled directly to our ear drums, through a negligible amount of air, compared to listening to a speaker.

I vividly recall the first time I heard stereo music through headphones. It wasn’t that I never heard headphones before that, but they were simple headphones, more like a telephone handset, with limited frequency response and a “tinny” sound.

The first quality headphones I bought were Koss Pro4s. The company still makes a variation of this headphone. By modern standards, they were not very good. But they were a breakthrough in the 1960s.

I found a cheap old turntable someone had thrown out. It had its own electronics but I built my own phono preamp and fed it into a power amplifier I built from a kit. But I only had one speaker. I wanted to hear what stereo sounded like.

I built a second phono preamp and replaced the mono cartridge with a stereo one. The preamps could drive the headphones directly.

I put a stereo record on the turntable. I am pretty sure it was Herb Alpert and the Tijuana Brass, a fine recording and good choice for my introduction to stereo.

The sound I heard totally amazed me. I could picture where the various instruments were, and I could feel the sense of space in the studio. It changed my life.

When I finally bought a pair of AR3 speakers and built a stereo power amplifier, I was disappointed in the sound. The headphones sounded better to me. And headphones still sound better to me to this day, 60 years later.

Headphones are a different experience. They have their advantages and disadvantages. I would not be comfortable mixing entirely with headphones. I find them excellent for listening to details. I always use them for any editing.

For mixing I try to stick with the speakers as long as I can, since it is more challenging for me to get things to sound best in the speakers. And arranging a stereo image is best done on speakers, since there is considerably less separation between left and right in speakers, compared to headphones, where the separation is infinite.

If I want to listen to music for enjoyment, I always listen on headphones. They focus my attention. I hear more details. It is a more compelling experience for me.

If you like your music loud with the bass thumping your chest, you won’t get that experience with headphones.

I know that most people prefer speakers. And they get excellent results. I am not trying to talk you into headphones. But they work best for me.

And judging by what I see in the real world, a significant percentage of listeners will hear your work on headphones. That argues for giving them a great experience while listening to your work with their headphones.

I should also mention immersive formats, like Dolby Atmos, which use many discrete speakers spread around the room to achieve a different listening experience. I have limited experience with Atmos, but I have heard my projects mixed in Atmos and heard them in an excellent control room. The experience is amazing. There is something about the Atmos process that makes details easier to hear, even if listening with less than the full complement of speakers.

An Atmos room puts an even greater burden on the speakers and the control room acoustics. The amount of space required, and the cost of 17 speakers puts an Atmos control room out of the reach of most people. And I am OK with that. It is a cool concept, but one that is likely to have only a minor impact on how I record or how I listen. Time will tell if immersive formats have staying power. I hope that they do.

That’s my take on loudspeakers and how we use them in our recording.

Thanks for listening, commenting, subscribing, and give me your feedback and ideas for future episodes.

 This is My Take on Music Recording. I’m Doug Fearn. See you next time.