Within the intricate aspects of the music world, the term frequency response pops up all the time. Whether you’ve heard of it or not, it is quite important for many reasons. It affects headphones, studio monitors, headphone amplifiers and DACs, and even the acoustics of a bathroom. It can dictate a lot of decisions that a producer or engineer can make. Believe me, I’ve been there. Well, fear not, we’re going to break it down together in the quest to understand this complex subject.
What is frequency response?
Frequency response is the range of frequencies that a component can reproduce. We measure this using the decibel (dB). Let’s step back and understand that the human ear can only hear any frequency between 20Hz and 20kHz. Say you’re listening to a sound through a pair of headphones. You’re only hearing the range of frequencies those headphones can reproduce. Frequency response measures how well those headphones reproduce audible frequencies. It also measures any changes are occurring as the signal passes through them.
Refer to the frequency response graph above. You will see the response curve of my own pair of headphones. This is measured by magnitude (gain or output) and frequency (kHz). This gives you a sense of where frequencies can lay along a spectrum with the maximum output being 0dB. I know, this is all very nerdy to some, but this is important stuff.
Why is frequency response Important?
An audio component with good frequency response is able to reproduce all the tones of a sound. It’s also able to reproduce that sound with minimal to no changes to the signal at all. This means that this unit is a reliable reference to use for your mixes, depending on your needs. Sometimes imperfection is what you may be after. For instance, a pair of bass heavy headphones, or a pair that sounds more bright. The only way to know any of this is through the measured frequency response of that component. Whew, okay let’s move on.
Flat Frequency response and phase
So, after all that there is still A LOT more to understand about this topic. Let’s dive into what flat response and phase are and how this can be applied to real life situations.
We’ve talked about how frequency response is measured in magnitude and frequency. Refer to the above graph. You can see the peaks and valleys flowing throughout the spectrum. Resonant frequencies are the notable humps on the chart. These can be a huge problem seeing as they can cause tones to become overly harsh or extremely dull. In a sense, a plus or minus curve of 3dB is within the lower limits of what humans can hear. Any deviation above or below that 3dB limit will make a noticeable difference.
As you have already assumed, the lesser amount of these resonant frequencies is best. This is why a more flat frequency response is ideal in an audio component. Unfortunately, this is very hard to achieve. You can see this in the unrealistic flat red line on 0dB on the graph above. In the audio world, it’s somewhat unrealistic to find a perfectly flat system. Rest assured that using a pair of studio monitors with a more flat response will still be much more useful. Especially for mixing and production.
How this applies to the studio
I can speak from experience here in my own productions. I had a pair of cheap headphones that I used for sound design in my songs for a period of time. After extensive research, I found a nice pair of headphones with a flatter response. My life significantly improved from that day forward in a variety of ways. My mixing workflow felt smoother. I started feeling much more confident. I knew I was relying on a system with good frequency response. It’s important to note that I still use those cheap headphones as a reference when needed.
Phase
Phase is quite an overlooked aspect of accurate sound reproduction. Phase is the depth and presence of a given component. If the phase of a set of studio monitors is incorrect, they will sound muddy and unpleasant. Even if those studio monitors have an acceptable frequency response curve. Thankfully, you can achieve proper phase-align with simple placement of your studio monitors. Try to position them about 8-12 inches from a wall or other hard surface. Also adequate positioning of a studio monitor subwoofer is a must if you use one.
All this becomes more and more complicated as more components are added to the equation. So far, we’ve explored what affects a pair of headphones or studio monitors can have on the signal.
Adding a DAC (digital-to-analog converter) or an amplifier can further alter the frequency response of the source. Consider looking into a DAC and Amp combo if you really want to get fancy. Some audiophiles out there debate about having a high quality external DAC separate from a good amplifier. I have found that it doesn’t make too much of a difference but it’s an interesting subject to read up on.
Where do sounds live in a mix?
It’s definitely pertinent to point out where certain tones might sit in a particular mix. Look at the graphs and charts from earlier. There you’ll see that those same peaks and valleys all have a small range of frequencies that they impact. Refer to the graph below. Compare what potential resonant frequencies that a component will have an effect on. This will tell you what areas are most inaccurate:
- Bass = 20hz – 521Hz (Sub 16Hz – 32Hz) (Low to Mid Bass 32Hz to 521Hz) (Kicks 60Hz to 250Hz)
- Mid = 521Hz – 2048Hz (vocals, pianos, guitars, the bulk of the perceivable sound, etc.)
- Treble = above 2048 Hz (Cymbals- 8,000Hz to 16,000Hz, bells, highest perceivable sounds)
You can use the information in this section as a reference to any given system. That way you’ll know if it will have an exaggerated bass, very present mids, or super airy highs. Alright, now we are ready to dive in even deeper and get into Fourier Analysis.
Fourier analysis
We’ve talked about a lot so far. In a nutshell, we know that nonlinear responses can severely affect our source sounds. But, in the less rudimentary way of looking at all this, we need to look deeper, literally speaking. A nonlinear response also has an impact on the entire family of sounds and instruments in the mix. To understand this, we need to look at Fourier analysis.
Fourier analysis and the Fourier Transform is the process of decomposing a sound into its constituent sine or cosine waves. In layman’s terms, Fourier analysis is the sum of a series of sine waves. Sine waves can visually appear in many shapes like a square, a triangle, or a chaotic mess that’s known as white noise. If you saw all these sine waves in all their unique forms in a linear manner, you would be looking at a Fourier analysis.
Why does Fourier analysis matter?
When you play a note on any instrument, it will bleed out multiple harmonic frequencies. If you have a great number of instruments playing notes, they each give off harmonic frequencies, each one of them unique in their own respect.
This is where Fourier analysis becomes interesting in the terms of frequency response. A nonlinear response not only alters the reproduction of a single sound. It also affects every single little sound in the mix. Mind-blowing right? If a more flat frequency response didn’t seem that important before, I’m sure it does now.
Headphone Application
This is where we should begin talking about the real world application of all we’ve learned. There are things you might want to consider when looking into different components. For the sake of simplicity, we will focus on headphones in particular.
Say you’re researching a pair of quality studio headphones. You say to yourself, “Well, I read this one article on frequency response and I learned so much from it. Let’s apply what I learned from that article to my research.” Well, consider yourself ahead of the game.
Try to keep in mind that a pair of headphones with a flat frequency response of at least 50Hz-17kHz +/-3dB is ideal. Why? That frequency range is well within the realm of what humans can hear. The +/-3dB is an adequate area of disparate fluctuation as we talked about earlier. If the +/-xdB variation is not stated by the manufacturer after the frequency range, this is a major red flag.
Also, aim for headphones with a THD (Total Harmonic Distortion) below 1%. This will help to avoid any unwanted distortion. A low THD will tame harmonic frequencies without destroying their fundamental noise floor.
Next, you’ll notice that higher end headphones will have an impedance measurement. Impedance is the effective resistance of an electric circuit or component. This is measured in ohms (Ω). All you’ll need to know here is the higher the impedance, the chances are you’ll need an amplifier. For example, a pair of headphones with 32Ω don’t need an amplifier to power them. A pair with 80Ω or 250Ω most likely will need an amp. Without an amplifier, the higher impedance headphones might lead to a muffled output.
What to keep an eye out for
The last thing to keep in mind is the fact that cheap headphones will favor bass over mids. This leads to a more inaccurate representation of the source sound. Although, as I stated in my personal story earlier, this may be what you’re looking for in certain scenarios.
Most music nowadays is heavily influenced by bass. It’s not a horrible idea to use a pair of bass heavy headphones for the sake of reference. You will potentially have to compensate in the mixdown though. Also, many times manufacturers will use treble (high end) frequencies to emphasize clarity. This usually only means increased loudness, not necessarily more detail. These are a couple of examples of an unavoidable affect that certain components can have.
Summary
Okay, we’ve covered a lot here. We’ve learned that frequency response is a huge factor in the music world. All in all, frequency response is our friend, or our foe. If our goal is to understand music in its most natural form, then this is surely essential. Moving forward we can keep what we’ve learned about frequency response in our back pocket.
