Myths and facts about studio acoustics I: Analysis and Parameters
Italian Acoustician DONATO MASCI from Studio Sound Service introduces a series of articles that are based on his own measurements and appraisal of the many and varied rooms he has designed and built.
An acoustician’s work is quite delicate because you’re required to design environments that fit as much as possible with your clients' needs and yet it is mandatory to meet fully the acoustic requirements for any kind of room. I often feel like a tailor, sewing a client's demand for fashion while concealing his physical defects. The thornies of my tasks is dealing with requests from clients to scientific data, and that’s because it’s not exactly all scientific and that’s where the tricky part begins. Over the year I've developed working relationships with pros from other countries and I've been surprised to discover that many requests and market trends that I have always considered to be “Italian” are instead quite international. This is where I got the idea to write down a sort of compendium of these widely spread opinions and beliefs, comparing them with a statistical analysis of the acoustic parameters I measured in all the studios I worked in. The goal is to take stock of some conclusions, prove or disprove such opinions and “beliefs” and eventually to start a debate.
1.2 Myths to dispel: “beliefs”, market trends and ways of thinking
In my three-articles series, by using scientific criteria I will try to prove the following frequently encountered statements to be right or wrong:
1. You can't mix in rooms that are too big, too small, with a high ceiling, with a low ceiling.
2. Panels or tube-traps are enough to make a room a good mixing room.
3. I do not like to listen with the sub.
4. Some monitors sound too good to be used for mixing.
5. Big monitors are good for clients but not for mixing. They're just too big for that, they lack “definition”.
6. I don't want my monitors in the wall, it's not necessary and I can change them easily if I ever want to. Furthermore, they can be moved about in case I need to perform a “fine tuning”;
7. Nearfield monitors have much more “definition” than far-field ones;
8. Auto-calibration is useless if a room has good acoustic treatment;
9. I don't need acoustic treatment if I have auto-calibration.
1.3 Preliminary observations in recording studio acoustics
Let’s start with some preliminary observations on recording studio acoustics. In the audio community is now considered indisputable fact (even if some pros are still not aware of it!) that the control room has to “sound” as neutral as possible. To be more specific, AESTD1001.1.01-10 specifications represent a good guide line:
- the optimal reverberation time from 200Hz up is around 0.25s in 100m3 rooms and at lower frequencies can go up to 0.75s; for smaller rooms (or bigger) these optimal values are lower (or higher);
- frequency response has to be as flat as possible, better if within ±3dB range (even if many probably don't know that most “professional” monitors have a ±5dB range measured in an anechoic chamber);
- first reflections should be 15dB lower than direct sound.
The first thing sound engineers take a look at, maybe because the chart is easily understandable, is the frequency response. A lot has been said on the subject already in the pages of this magazine so I will not write any further on the subject. Many engineers actually only stick to these parameters, not knowing that, for non-treated rooms, frequency response is often quite flat (setting aside a physiological enhancement of lower frequencies, that can be easily corrected with the roll-off filter that is common to most monitors). In this case, the issue is that a flat response does not supply sufficient sound definition to a mix especially at low frequencies. The reason is that the monitor's direct sound as it reaches the listening spot is coloured by the reflected and reverberated sound that comes from the room. Reverberation time is a key factor. On the other hand, if this were particularly low it would make listening quite uncomfortable and quite different from the same sound reproduced in another more “normal” environment.
On average, among studios I visit on a preliminary basis before any drawings or treatment is done only a small portion have absolutely no treatment to begin with. The majority have self-made treatment solutions, based on what the Internet has to say on the subject, and most of the times these do not manage low frequencies correctly. Also, most clients treat their rooms with absorbent pyramidal panels et similar. What follows is that all such rooms are quite colored in frequency and fail to achieve what is required – a neutral sound. As shown in the pretreatment reverberation times charts frm my measurement of various rooms (Figure 1), the average is above the optimal value, considering that the rooms’ mean volume is 60 m3 – but the most important element is the variance: it's very high, mostly at low frequencies. In the post-treatment RT chart (Figure 2) of the same rooms, we see that:
- there is a substantial reduction in RT values, especially at low frequencies;
- the average value @ 63Hz is twice the value @ 500Hz – it was three times higher pre-treatment;
- variance is very reduced, although physiological discrepancies remain among rooms with different volumes;
- there are no rooms with RTs higher than 0.65s @ 63Hz.
For rooms I selected on the basis of a better subjective quality of listening experience (Figure 3) this trend is even more noticeable. As it shines through the AES guidelines and as shown on these charts, optimal RTs depend on the room's volume. Therefore, there's no answer to the question of “what is a control room's ideal RT?” if its volume is not known because the factor that is most related to sound definition is the ratio between direct vs. reverberated sound energy that reaches the listener.
1.4 More acoustic parameters in recording studios: energy vs. RT
Let’s now look at more acoustic parameters in recording studio and particularly energy vs RT. For all reasons given above, I found it interesting to examine control room acoustics through “energy” parameters, usually common in architectural acoustics (theaters, auditoria, etc.) and defined by ISO3382 standards. Unfortunately, in the literature there is no way to find (or at least, I have not found yet!) their optimal values for recording studios. Most common energy parameters are Clarity (C50 or C80) and Definition (D50), that relate the “constructive” sound energy (the type that comes within 50 or 80ms of the direct sound and makes it “stronger”) to the overall one. It is easy to understand why these parameters (used for big spaces where it is necessary to strengthen some of the first reflections in order to “amplify” direct sound of a source from a stage) are definitely not usable for control rooms. In control rooms it is usual to kill first reflections that have a tendency to colour the direct sound. Furthermore (and this is a common problem and matter of discussions in architectural acoustics too) these parameters do vary quite a bit, even from close positions and are affected by reflections on the line of the integration limit (50 or 80ms). This makes them often unusable. Instead, I found the analysis of the Center Time or Barycentric Time (Ts; It's called Barycentric because of its analogy with a solid center of mass) that quantifies the time required for energy to reach the measuring spot as if this energy were “packed” in a single reflection as particularly interesting. What is remarkable is that Ts assumes very similar values for treated rooms and is even more similar for quality selected rooms. There is no substantial variation based on a room's volume, therefore it is, in my opinion, an absolute parameter that defines after how much time you achieve the average sound energy at different frequencies. In the charts at Figures 4, 5 and 6, it's clear that this parameter, for quality selected rooms, tends to an average value independently from the single room's volume.
1.5 Preliminary conclusions and first dispelled myths
All of the above given, the great issue in non-treated rooms, or particularly in badly treated rooms, are low frequencies. Back to our myths then...
1. You can't mix in rooms that are too big, too small, with a high ceiling, with a low ceiling. – Small variations in the size of mixing rooms can be accepted. What matters is that the Ts complies with average values observed in quality selected studios. However, smaller rooms will be problematic because of stationary waves that concentrate in a short range of frequencies and often cause peaks in the Ts parameter. Very large rooms might suffer a loss in definition, caused by the distance between the monitors and the listening spot but I've never designed rooms big enough to actually have this problem.
2. Panels or tube-traps are enough to make a room a good mixing room – False. To make efficient full range acoustic correction a large quantity of absorbent material is required to handle low frequencies; the depth of premade mobile acoustic panels is simply not sufficient.
3. I do not like to listen with the sub and (4.) Some monitors sound too good to be used for mixing – To me, such claims derive from the fact that some rooms are often so “colored” that the listener prefers monitors that do not go too low. For the same reason many do not like listening with the sub, along with the fact that the sub's release time is in some cases too long and may easily effect Ts values (making them longer on low frequencies).
In the next article, I’ll look at loudspeakers in mixing rooms and more myths to be dispelled.
Donato would like to thank his business partner Francesca Bianco for reviewing the text and his colleague at Studio Sound Service, Valentina Cardinali, for her patience, daily collaboration and useful thoughts.