Independent Recording Network home Page Glen Stephan - The SouthSIDE of the Tracks
News & Features Contributor's Columns Online Resources Online Marketplace IRN Members Network E-mail links, RSS feeds and Mailing lists

EnginEARing Part 2: Listening Without Bias - Volume
By Glen Stephan on July 14 2008 04:00 PM | Permalink | Author Info
The first step to developing an “engineer’s ear” (Part 1) is to learn to listen objectively, to have an open mind regarding what we listen to. There are a few inherent biases we need to first overcome: the volume bias, the music genre bias and the “professional” recordings bias. This article will deal with part one, the volume bias. We’ll cover parts two and three, the music genre bias and the commercial recording bias in the next EnginEARing article.

One listening bias that’s part physical and part psychological that we must overcome early on is the bias of relative volume. When we compare music that sounds good at one volume and suddenly increase the volume – not to ear-bleeding levels, but just so that it’s somewhat louder – we almost invariably think that the louder playback sounds better. There are both physical and psychoacoustic reasons for this that have to do with things like the inverse proportions of power and wavelength, Fletcher-Munson response curves and other technical whatzits. We'll talk more about these at another time, for now the exact explanations are not so important as the concept itself.

What is important is that we remember that in the above volume example, the source music recording remained the same. Nothing happened to change the quality of the recording, and in fact the quality did not change. Yet we perceived an increase in quality. That perception is an illusion caused by the increase in volume. The recording has not changed; only our perception of it has changed.

This illusion is widely known amongst retail sales staff for places that sell consumer audio equipment, and is often deceptively used to sell audio gear. Two old-time tricks of the sales trade involve selling loudspeakers and selling equalizers.

In the loudspeaker gambit, it is not uncommon for a store to stock a “home” brand that includes at least one model of  “high efficiency” speaker. High efficiency in this case basically means that it will sound louder at a certain amplifier volume setting than a similar but lower efficiency speaker at the same setting. It is easy to make a high efficiency loudspeaker that has an OK sound for a relatively cheap cost, certainly much cheaper than it costs to make a great-sounding loudspeaker that may not be as efficient. But when that salesperson hits that button that switches from the high quality, lower efficiency speaker to the lower quality, high efficiency speaker, the unsuspecting customer with the untrained ear is often immediately hooked, and the sales person has sold them on a lesser-quality speaker that is more profitable to him and his company.

An even more common gambit is the use of volume to sell aftermarket equalizers. Ever notice that when you see a functioning graphic equalizer on display at your local consumer dealer, it almost always has all it’s sliders at or above the center zero gain level? Nobody ever actually cuts any frequencies; they all just boost them. Why? Well other than the fact that many sales folk just don’t have any better ears than their customers do, the fact is every slider on a graphic EQ is a volume control that is boosting the volume of a given frequency band. All the staff needs to do to sell an EQ is to set the EQ to boost volume in a not-unpleasant way. Then they turn the EQ bypass switch on and off a few times to show the difference between the equalized signal and the non-equalized signal, and the difference in volume will again set the hook.

The trained ear knows not to be fooled by changes in volume. The trained ear knows that the recording needs to be listened to in detail at each individual volume, and that apple vs. apple comparisons can only be made at the same apparent volume. The trained ear knows that comparing two things at two different volumes is an unfair apples and oranges comparison.

The trained ear knows to take control of the volume and play back each loudspeaker in that listening test at the same apparent volume in order to get a proper comparison of the actual sounds of the speakers. The trained ear also knows to take control of the volume so that the EQ is outputting at the relatively same volume that it’s inputting, so that the sound is being shaped without being affected by the illusion of volume. And the trained ear knows to compare two different studio recordings at the same playback volume if they are to make a fair judgment, and not to let the illusion of volume affect their analysis of the recording.

When comparing two recordings, two versions of a recording, or the same recording on different signal chains, we must recognize the influence that playback volume has and a) mechanically eliminate that influence, b) separate the volume from the other characteristics of the recording and listen past the volume to the music itself, or c) take control of the volume and use it to our advantage to see what in the recording works at low volumes and what works at louder volumes.

It's important to note here that it's a bias, a trick that only works when compared to a lower volume. If the listener just listens to the song at the one higher volume, with nothing to compare it to, the extra volume means nothing and the mix will sound just as good or bad as it does at lower volumes.

Finally, it is important to note that we are talking about the overall volume of an unchanged mix here. Sure, when mixing, if we raise the volume or gain on one instrument but leave the other instrument the same volume, there is a real and an important change that is not an illusion. That’s not what we’re talking about here; we’re talking about overall listening volume of the mix or final recording itself.

So we must learn to recognize overall mix volume immediately and understand how that might affect our perceptions, and not be seduced by it’s ways. If we really want to listen to something with an engineer’s ears, we need to a) listen to it at all volumes and listen to how volume does affect what we hear, and to make sure that what we hear at volume X is as pleasant as what we hear at volume Y, b) do not be fooled into thinking that louder is automatically better, because it only seems better in a quick comparison, not on an extended listen, and c) that when we are comparing two sources or two mixes, we have to compare them at roughly equal apparent volumes to ensure we are comparing on a level playing field.

We need to practice listening to some of our favorite CDs or home recordings, switching back and forth between differing playback volumes and pay attention to the difference a change in playback volume can make. Take mental notes of the differences that strike us. We should practice listening “past” the playback volume, focusing our attention on the various components of the sound rather than to the volume itself.

When we find we can focus just about as easily on the details of the sound at a playback volume about that of a quiet conversation as we can at a playback volume rivaling that of a Harley going down the street, and we can recognize how the bias trick can try to fool us even though the mix has not actually changed, we will have taken the first very large step to developing our critical listening skills.

Next up: Part 3: Listening Without Bias - Genre and Professionalism

Recent Entries RSS 2.0 Newsfeed:Most Recent Columns from Glen Stephan

2017 GJS/SouthSIDE Multimedia Productions and the Independent Recording Network