Many professionals, and much more qualified people than me, have written extensive papers on this subject. The debate is not about rather it exists or not, it most certainly does, but if the loudness war has actually resulted in better preference to listeners. Perhaps I should begin by explaining what exactly the ‘loudness war’ is.
Recordings over the decades have gotten louder and louder in post-mixing volume. That is the short version. If you have someone in an isolated test listen to one sample of a song, and then the same sample at a higher volume, they will tend to prefer the louder version. There is science to this, since the human ear does not perceive sound on a linear scale, increasing the volume on a linear scale will allow us to hear certain nuances at certain frequencies that we could not have heard before. Being able to hear more nuances is a good thing, and this is the basis for the logic behind the loudness war.
Imagine you are a record company. You want to sell the most records make the best sounding records possible. Knowing that human perception causes us to prefer louder music, you begin to demand that your recording/mixing engineers create tracks with a volume that is higher than before. So they up the volume. A competing record company observes you have higher volumes than theirs and thinks that is the reason you are selling records better than them, so they increase the volume on their tracks even more so that people perceive their music to sound even better. A third record company, then a fourth repeats this, this continues for decades.
Now imagine you are a radio broadcasting company. You know the same thing the record company knows, that people tend to like louder music better. You want the most viewers possible, so you tell your DJ to set the outgoing volume to be higher so that people will think it sounds better. A competing radio company knows the same thing. They think the reason more people are tuning into your station and not theirs is that they prefer the louder volume you are running, so they turn up the volume even louder to make it sound even better. This also goes back and forth for decades.
Hence loudness ‘war.’
If it is true, that people prefer louder music, why is this a bad thing?
http://i3.tinypic.com/8f0z2hu.gif
To an extent, increasing the volume will help bring out the harder to hear frequencies creating a better perceived sound, but the sound quality begins to degrade after a certain level. This picture is just a raw example, I don’t know what song it is or where it is from (thanks google images for making plagiarism so easy), but it shows two versions of the same waveform, the second has drastically increased in volume. Without knowing anything about sound, notice how the first waveform has a certain shape and level of dynamic contrast (some parts are loud, some are soft). The second example loses both of these things. The reason this happens is that the recorded volume is being digitally increased beyond a point that the original sound retains its harmonic properties. In other words, the person sang at a certain volume into a microphone, which was then recorded onto a computer. On the computer, the engineer, following commands from his superiors (or being a knowing contributor to the loudness war) turns the volume up to the level that is desired, regardless of what it sounds like or what the waveform looks like. He then uses filters and equalization to ‘fix’ it and remove the negative effects caused by the volume increase. Sure, it sounds ‘good’ but the greater the increase in volume, the less of the original sound is left intact. In other words it destroys the intent of the artist in favor of the record companies desire to sell more records. This was never their intention, but over the years and years this has been going on we have been brought to where we are now, with over produced, hyper compressed, near robotic sounding music.
This has caused much grief in the audio engineering community, since they of all people know what is happening yet records that take these practices to the extreme continue to top the charts. It certainly is not the only reason, or even the most influential reason, but the damage has been done.
People tend to prefer louder versions of the same song, but what if you play two different songs with different volumes? Anyone who has done a science experiment in high school knows that this is too many variables, you can only test one variable at a time to prove that that variable is indeed causing the difference in results, with all other variables controlled. This is where it gets confusing, since songs never have to compete with themselves for success; they have to compete with other songs. It’s very difficult to prove that one song had success over another because it was louder, when the most important factor in music selection is genre preference, however you can’t prove statistically that it is NOT the reason either. I will take advantage of this not being a professional or official paper to say, that this is why the loudness war is a load of crap.
However, there are certain situations where the loudness war does exploit human perception in a more correct way. The most obvious one is live concerts. Perhaps you have not been to a concert with multiple artists or have not been to one at all, but as the night goes on and the artist playing is more well-known than the one before it, live sound mixers will often turn the volume up. The first artist was probably already well above the human threshold for hearing loss, but the louder a sound is, the easier it is to have an emotional response to it. Having an emotional response will make it a more memorable experience, therefore giving the audience the perception that the loudest, and headlining, artist was the one who put on the best show. I’ve been to a lot of concerts where the sound guy purposely caps an artist’s volume to leave more headroom for the artists performing after, even though the artist on the stage currently is requesting more sound. The sound guy doesn’t care about the opening band, and usually neither does the audience, so the headlining artist always gets the best show and everyone is happier, except the little guy who opened for them.
Another situation is far more common, tuning the radio. The end user has control over the power that the amplifier is using to drive the speakers, but the end volume is decided by this and the outgoing volume set by the radio station. Radio stations all have different volumes, sometimes changing even song to song to try and combat the effects of the loudness war or make them even worse. While a listener is changing from station to station, the ones that will stand out are the louder ones. Even if it causes just a few seconds longer of lingering before the user continues to search for something they want to hear, the loudness war was won. As stated above, the end decision will always be made based on genre preference, but in the situation where two radio stations have similar catalogs of music, they feel that they need to be louder to be more easily noticed by the listeners.
What we can take away from all this is that the loudness war was never the factor that determined what was successful, yet record companies still drive mix volumes ever higher. It’s impossible to disprove that loudness effects musical preference between two different songs, but it is also impossible to prove that it is, because given two samples at different volumes a person will always choose the song they like better as the one that sounds better (assuming comparable quality). In conclusion the loudness war is a war that cannot be won, but they won’t stop fighting it until we are all deaf by the age of 30.
Metalmalarkey's Blog 
