Hearing what you want to hear
We sometimes front-load our expectations into what we believe people will say or what a stereo system should sound like. I can’t count the number of times I’ve walked into a room full of loudspeakers and prejudged their performance before the music started playing. Often, I am surprised, both pleasantly and otherwise.
The problem with preloaded expectations is we have to work past them to get to the core of what’s really there—yet, it’s often those very expectations that drove us to try something new in the first place.
When I am told what to expect from a piece of audio gear or new technology, the results can go one of two ways: I am happily rewarded or sadly disappointed. The problem with this process is we can often miss the underlying truth blurred by our preconceived notions.
It’s not always possible to audition new gear without the burden of expectations but, when we get the chance, it’s likely to give us a more honest result.
Hearing what we cannot hear
As a design team, we work hard at extending our amplification product’s bandwidth beyond that which humans can hear. We do that because it sounds better: more open, extended, transparent.
Yet, how could that be true? If we cannot hear above 20kHz (and most of us not above 12kHz), then why should it matter that our preamplifier extends to beyond 60kHz?
Spoiler alert. The answer is phase shift, something we humans are quite sensitive to. In an analog circuit with a maximum bandwidth of 20kHz, phase shift occurs well into the audible band. Take, for example, this simple graph below. It’s a simulation of phase shift from a 1kHz signal from both a high pass (bass roll off) and a low pass (high frequency roll off) function. Note how much phase lag and lead occur away from the center frequency.
To get phase shift out of the audible band we need to extend the roll off point well beyond that which we can hear.
The point of all this is simple. Sometimes we’re looking so hard in one place we ignore what’s actually happening in quite another.