The Loudness War: A Game and Market Theory Analysis

Essay, 2012

7 Pages, Grade: 10.5/10


Introduction and History

The “loudness war” is the phenomenon in which loudness levels on commercially-released music recordings have been increasing over time. This is driven by the belief that, all else being equal, the louder of two recordings will be noticed more and preferred by a listener. The loudness war began as early as the 1950s and ’60s with 7" 45rpm singles. Jukeboxes playing these singles, which were commonly found in bars and other public places, often had only one preset volume level. Record companies tried to make their singles louder, or “hotter” than their competitors so they would stand out among the others[4].

The practice of making loud masters continued throughout the vinyl era, but it wasn’t until the Compact Disc became the dominant format that the loudness war really gained momentum. Unlike an analog recording, with digital audio, there is an absolute limit to the ampli- tude of a recorded sound wave, above which the wave is sharply truncated or “clipped.” This level is designated as 0 dBFS (decibels full-scale), and digital loudness levels are measured in negative dBFS (decibels below full-scale). When referring to dBFS figures, a distinction must be made between peak level, which is the level of the loudest points in a recording, and average or RMS level, which is the average loudness over the entire recording. In the 1980s, the early years of the CD, there was little incentive to create loud masters, but as CDs became more popular,

illustration not visible in this excerpt

Figure 1: A comparison between loudness levels from two versions of the Red Hot Chili Peppers’ song “Give It Away.” The top ver- sion is from the original 1991 album and the bottom from the 2003 “Greatest Hits” collection.[3]

recording engineers began using digital tools to push the loudness levels higher. For example, in 1987, a CD with an average level of −15 dBFS was considered loud. By 1994, the average level was roughly −12 dBFS, and in 2005 it reached −9 dBFS [8]. In human terms, the average CD in 1987 sounds about half as loud as its 2005 counterpart when played with the same volume setting.

Recording engineers cannot increase loudness levels indefinitely without compromising sound quality. In the 1990s, most recordings already had peak levels that reach 0 dBFS and cannot be made louder. In order to continue increasing average loudness, engineers resort to two tools: peak limiting and dynamic range compression. Peak limiting is the process of reducing the level of individual peaks in a recording, while dynamic range compression reduces the difference between loud sections and quiet sections in a recording, so that the recording as a whole sounds louder. Both of these techniques have negative effects on the audio fidelity. Peak limiting reshapes the loudest peaks, destroying the musical transients they represent, while dynamic range compression drastically reduces the contrast between loud and soft sections in a recording. Musicians and listeners alike have criticized these practices of “hyper-compression.” For example, musician Bob Dylan has stated:

“You listen to these modern records, they’re atrocious, they have sound all over them. There’s no definition of nothing, no vocal, no nothing, just like—static”[4]. But the loudness war continues on, and CD volume levels continue to increase.

Why the Loudness War?

One may ask why the record companies partake in the loudness war given the negative impact it has on sound quality. Isn’t recorded music supposed to be as clear and true to a live performance as possible? Simply put, the reason for the loudness war is that record companies believe that loud music sells better. In fact, there is evidence that in some cases listeners do associate loudness with quality [1]. There are several examples of commercially successful loud albums. The widely popular 1995 album by Oasis, (What’s the Story) Morning Glory?, had tracks reaching −8 dBFS RMS, which at the time was an astonishing level [4]. Today, many people listen to music primarily in the car or other noisy places, where louder music cuts through against the background noise. Record companies, especially today, tend to cater to this market of casual, “on-the-go” listeners and make heavy use of compression and limiting in order to make their album louder.

Let’s consider the loudness war as a two-player strategic game. The players are two different record companies, each producing a CD. Each record company has two strategies: release a loud CD (L) and release a quiet CD (Q). We can construct the following symmetric payoff matrix for this scenario.

illustration not visible in this excerpt

The values of a, b, c, and d for this game must satisfy the constraint b > d > a > c. This is because, under the belief that a louder album will sell more than a quieter album, a record company receives the maximum payoff when they produce a loud CD and their competitor releases a quiet CD. Conversely, the company releasing the quiet CD receives the minimum payoff because they are outsold by their competitor. If both release a loud album, we will assume that they will split sales evenly and receive the same payoff. If both produce a quiet CD, they still sell evenly, but their payoff is slightly higher because their CDs sound better than they would have had they been subjected to limiting and compression. Upon analysis of this payoff matrix, we see that no matter what the actual values of a-d are, (L, L) is the only Nash equilibrium in the game. This means that the case in which both companies release a loud CD is the only case in which neither can improve their payoff singlehandedly by switching their strategy. This result is an archetypal example of the famous Prisoner’s dilemma — a theoretical two-player strategic game in which the two players may not cooperate, even if it seems to be in their best interest[6]. For the loudness war game, Q can be thought of as the “cooperate” strategy and L the “defect” strategy. Even though (Q,Q) is better for both players than (L, L), they will, with high probability, both play L.

Further analysis of this payoff matrix reveals that in addition, L is an evolutionarily stable strategy — a strat- egy (call it S) with the property that if the majority of the population is playing S, no small group of “foreign invaders” playing a different strategy can thrive. The converse also holds: if the majority of the population is playing some strategy T , and S is an evolutionarily stable strategy, a small group of invaders playing S will thrive and eventually overtake those playing T. For the sym- metric payoff matrix above, the condition for L to be an

illustration not visible in this excerpt

Figure 2: A plot showing the decline of dy- namic range in commercial recordings over time as a result of the continued escalation of the loudness war.[9]

evolutionarily stable strategy is that either (i) a > c or (ii) a = c and b > d [2]. Since criterion (i) holds for the loudness war game, releasing a loud CD is evolutionarily stable. This explains why, despite the drawbacks with respect to sound quality, the vast majority of record com- panies and mastering engineers today choose to utilize heavy peak limiting and dynamic range compression to make loud albums.


Excerpt out of 7 pages


The Loudness War: A Game and Market Theory Analysis
Cornell University
INFO 2040: Networks
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
569 KB
loudness, game, market, theory, analysis
Quote paper
Drew Weymouth (Author), 2012, The Loudness War: A Game and Market Theory Analysis, Munich, GRIN Verlag,


  • No comments yet.
Read the ebook
Title: The Loudness War: A Game and Market Theory Analysis

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free