Audio Measurement: The Blessings And Curses

There’s no question: the single most important judgement tool for audio professionals are the ears. There is actually no measurement instrument in the world that is as versatile, fast, convenient and intuitive. And after all, music is made for ears, not measurement instruments. But there is also no instrument that can be as imprecise, mood-dependent and subjective as the ears. Thus, audio engineers and equipment developers alike need a complementary tool of judgement, to overcome these disadvantages. This gap is filled by a variety of audio measurement tools available, like spectrum analyzers, loudness meters, oscilloscopes and many more.

Here on The Science of Sound, we will rely heavily on measurements as we dig deep into the guts of our beloved audio tools and processes. But it’s important to be very clear and mindful about what we can and cannot achieve by relying on measurements. So here are 4 blessings and 4 curses of audio measurement.

As a research & development engineer, a big part of my job is measurement. Although the ears always come first, I need to rely on measurement to back up my ear’s judgements and find a technical explanation of what I’m hearing. That way I can replicate it if it’s a desired quality, or fix it if it’s a problem. After adjusting to what I found out by measurement, I’d switch off the instruments and concentrate on listening again, to find out if what I measured was what I heard, or if there’s anything left I’m missing. Then the loop starts again. There is no way around either one of these tools.

But also for you as a user of studio equipment, measurement can be an equally important tool for your daily work. Especially when you are evaluating new equipment and workflows or when trying to improve your skills. Let’s start with 4 blessings of measurement when working with audio.

Measurement helps you verify what you hear

No matter how well we train our hearing, results will always be very subjective. Especially when things get really detailed and we are starting to get tired after a long session, it becomes hard to distinguish subtle features. The absolute classic in this regard is loudness. When making complex adjustments, for example with EQs and compressors, it is all too easy to get distracted by slight loudness differences. This way, an objective judgement of what our tool of choice actually does to the sound can become very hard. Careful measurement and adjustment in this regard can help tremendously to make sure our ears are not fooling us too much.

Measurement helps identifying problems that you’d otherwise overhear

While working with audio, we regularly shift our attention to certain aspects of the sound we are tweaking. For example, when recording some tracks that will get heavily processed later, we might need to spend some extra attention on noise issues. Otherwise, later processing could amplify noise & hum that was originally just quiet enough so we didn’t hear it. And as we’re not permanently in full attention mode for the really subtle issues, it is always good to have a visual safety net that puts possible problems right in front of us. I like to have a high-resolution spectrum analyzer always-on so I can immediately see from the corner of my eye if there’s something unusual going on.

Measurement helps you better understand and set up your studio

Especially when incorporating new tools in the studio, measurement is absolutely essential. First when calibrating the tools to work together as intended. And later when evaluating the new tool and finding out how the audio reacts to it. A good example is examining the filter characteristics of a new EQ or the control behavior of a compressor. This is also handy when comparing different tools to better understand their exact differences and thus their preferable application areas.

Measurement can make you learn and work faster

If you don’t have golden ears yet, but even if you do, nothing beats a good real-time spectrum analyzer to quickly find that annoying resonance and get rid of it immediately. Or a good peak & RMS meter to bring up an initial ballpark estimate for compressor settings that are then refined by ear. Having a visual reference can be a great guide to get better & faster results. And being faster means having more energy left to really go into detail.

But not all that glitters is gold. There’s a whole other side to it, so let’s have a look at the 4 curses of measurement.

Measurement might distract you from actually listening

It is all too easy to get seduced into mixing for the eyes rather than the ears. So be careful and switch off the instruments once in a while, especially for the final touches. Don’t forget: in the end, it’s only the listener’s ears judging your work!

You might not measure what you think you do

This is not that much the case for the at-work kind of measurements like realtime spectrum analysis or level metering, but rather when measuring for calibration and evaluation purposes. There might always be some additional effects that can distort the measurement that you are making. For example, measuring the impulse response of a modulating reverb effect can – depending on the measurement method – reduce the high frequency content of the reverb tail. Also, harmonic distortion can alter the frequency response measurement of an EQ you are evaluating.

Trivial measurements won’t take you very far

When things get more complex, it is often hard to separate different features like frequency response and distortion. So with most equipment, you might only get a snapshot of a device’s actual behavior. This is especially true with dynamic tools like compressors. To get a rather complete idea of what a specific device does to your audio, you’ll have to make multiple measurements of different aspects. And you’ll need some good knowledge about the inner workings to put all the pieces together in order to fully understand what’s going on. But in fact, going that far will rarely be necessary if you’re not a developer.

Audio Measurement is hard!

As you can guess from the above, making reliable measurements that reflect what’s actually going on is very hard. It’s easy to make mistakes, and it’s equally easy to overlook something that can make a result unusable or misleading. Making good measurements requires a very deep knowledge about the method and the device or phenomenon you’re after. And it is most important to always be very mindful about what you are not measuring!

So now you have an overview about the fascinating topic of audio measurement. If I scared you off with the latter points, don’t be afraid! We will do lots of very advanced measurements here on The Science of Sound, and I will always be very specific and honest about what we can and cannot learn from these measurements. The goal is to enhance your awareness about the different features that your gear provides and establish useful new mindsets for thinking and working with audio.

How do you use measurement tools in your daily work? Has it helped you in the past, or did it totally mislead you? Share your opinion in the comments!