Loudness War Research
This page aims to collect as much research as possible on the subject of the Loudness Wars, both academic and informal. The goal is to provide a resource of objective evidence about the driving beliefs behind the Loudness Wars, and their real-world impact.
Clearly, more research on this subject is needed. If you would like your research to be included here, please contact Ian Shepherd.
NEW – Listeners DON’T prefer heavy compression – Naomi Croghan
From the conclusion:
“Contrary to a “louder is better” mentality, the current findings suggest that “louder is better… to a point.” Louder, more compressed, music samples were preferred when dynamic-range compression was applied to a moderate extent. However, high levels of compression were either detrimental to music quality or had little effect“
In other words, “louder is better – but too loud is worse”
The full article is available on the Acoustical Society of America website, here.
Loudness has no effect on sales – Earl Vickers
Earl Vickers has published an AES paper on the subject. Vickers is the first to admit that further research is needed, but his conclusions are striking. Here are a few:
- Loudness is not correlated with sales figures
- Loudness has almost no affect on listener’s preferences when comparing different songs
- Listeners tend to dislike the side-effects of hyper-compression, and prefer more dynamic music
- Content trumps loudness, especially on the radio
Read more on Vicker’s web-page:
Less compression sounds better; no effect on sales – Dave Viney
Several of the results cited by Earl Vickers are drawn from research by London College of Music researcher Dave Viney. We are delighted to be able to host the full text of his dissertation here.
Excerpts from the Conclusions:
The key finding, in the context of this project, has to be that, based on the sample of 30 CD single tracks from recent months, there is no evidence of any significant correlation between loudness (& implied compression) and commercial success – this is based on both actual loudness measurements and assessments of professional producers & engineers between which there is significant correlation.
In addition, Viney finds that
“A secondary but supporting finding is some evidence from correlations between panel assessments and commercial data that recordings with little ‘processing’ & ‘compression’ sound ‘more pleasant’ & ‘above average quality’ and are more commercially successful”
Please contact Dave directly for more information.
Listeners don’t prefer loud music – Sound On Sound Magazine
SOS recently published the results of some informal testing on listeners perception to absolute level. You can read a full account of their results here:
Interestingly, the examples used for testing by SOS only varied in absolute level – by up to 6dB – meaning they all had the same degrees of compression and limiting applied. Nonetheless, the main conclusion was that there was no statistically significant preference for loud music.
We would be very interested to see the results of a similar experiment where lower-level songs also benefited form wider dynamic range !
Loudness research at Anglia Ruskin University (ARU), The University of West London (UWL) and Cambridge University (CU)
“Loudness and dynamic range research is currently being conducted by Dr Rob Toulson and Bill Campbell (ARU), Justin Paterson (UWL), and Professor Brian Moore & Dr Michael Stone (CU). This holistic study covers many factors related to loudness and dynamic range in commercial music. It aims to develop a complete understanding of loudness, its manipulation and implications in relation to human auditory perception, signal attributes, and the wider implications of the cultures of sound creation and consumption.
Anglia Ruskin University is leading the technical strand of the project, evaluating signal attributes and artefacts given a number of dynamic range manipulation techniques. Cambridge University Psychology Department is the world-leading research unit in psychoacoustics and loudness; this strand of the project is to evaluate listening patterns with respect to perceived quality, loudness and listener fatigue. The University of West London’s role is to evaluate cultural listening trends and music production & distribution methods, and the creation of quantified test recordings. Each project-strand’s output forms significant inputs into the adjacent research strands, thus facilitating the first complete study into loudness and dynamic range. The full-scale project is dependant on an Engineering and Physical Sciences Research Council (EPSRC) grant, which is currently in application; however, work is already being undertaken at all three universities.
Formal project supporters and contributors include: BBC, Prism Sound, DTS, TC Electronic, TF Pro and The Pleasurize Music Foundation, with many more top industry professionals from around the world prepared to support and contribute to the work once it fully commences.”
Two relevant publications can be found here:
Campbell, W., Toulson, E. R. & Paterson, J. 2010 – The effect of dynamic range compression on the psychoacoustic quality and loudness of commercial music, Proceedings of Internoise 2010 Conference, Lisbon, June 2010
Stone, M.A., Moore, B.C.J., Fullgrabe, C. & Hinton, A.C., 2009. Multichannel Fast-Acting Dynamic Range Compression Hinders Performance By Young Normal-Hearing Listeners In A Two Two-Talker Separation Task. Journal of the Audio Engineering Society, 57(7/8), 532-546
In Europe, the EBU‘s Loudness Group has been researching loudness for over two years, and has issued detailed recommendations for loudness in broadcasting (EBU R 128) plus specifications for a new loudness meter to allow this to be implemented.
These recommendations have already been adopted by both France and Belgium along with 20 different vendors and highly affordable plugin versions of the meter are already available.
If use of R128 becomes widespread, the “Loudness War” could quickly become an irrelevant footnote in music production history. For more information, click here
LRA hasn’t decreased through the “war” – Emmanuel Deruty (via Sound On Sound)
Detailed, incisive analysis of the Loudness Wars using the EBU 3342 definition of “loudness variation” gives a surprising result – there’s no evidence that average dynamic range has decreased over the course of the “war”.
To some extent the distinction is semantic – the article observes that over-use of limiting and compression tends to cause…
“…reduced crest factor, envelope modifications… and in the worst cases, distortion. Common sense suggests that although there is nothing wrong with these characteristics as such, they shouldn’t be on virtually all records“
Nonetheless it’s a fascinating read – highly recommended:
I posted my thoughts in response to the article here: