Glossary: Sound Quality (SQ)

SOUND QUALITY (SQ) TERMS

Accuracy

A problematic concept taken literally, but sometimes meant to indicate believability. See Realism, Believability, and the absolute sound.

Bass

Lower musical frequencies, from approximately 20 Hz to 200 Hz.

Believability

When we judge audio equipment we are fundamentally asking if the rendition of sound is “believable” in the sense that real instruments in a real performance space might sound like this. Believability integrates the instrumental and vocal sounds captured, along with the effects and mixing and mastering processing applied to the recording. See also: the absolute sound.

Black Background

By ‘black background’ we generally mean the myriad elements of noise and distortion that mask or blur the rendition of small signals (e.g. reflections in a concert venue or harmonics from an instrument). So, when a speaker or DAC or amplifier is introduced into a reference system and resolution or ambience or inter-transient silence or depth of image or soundspace rendition are increased, these phenomena can fall under the ‘black background’ term.

Bright

Refers to an elevated level of treble, generally somewhere in the range between 4 kHz and 10 kHz. See Frequency Balance.

Definition

See Resolution

Dynamics

The ability of the device we are discussing to produce soft and loud sounds without distortion. Dynamics are affected by timing: when a device doesn’t respond to an input as quickly as the input requires, the transition from soft to loud or back can be delayed leading to a “heavy” or “slow” or “soft” sound. Well executing dynamic timing leads to terms like “punch”, “drive”, “quickness”, and “pace”.

Frequency

Musical instruments and voices produce sound by vibrating (strings, instrument bodies, vocal chords, oscillators, horns, drum skins, etc). These vibrations can be characterized by their frequencies. Frequencies are measured in terms of cycles per second or Hz (1 Hertz is 1 cycle per second). A “low” frequency would be for example the low string (E1) of a bass guitar, which is 41 Hz. A “middle” frequency would be A4 on a piano keyboard, which is 440 Hz. A “high” frequency would be the top note of a piccolo, C8, which is 4186 Hz. An important thing to understand when using instrumental examples is that all acoustic instruments and voices vibrate (resonate) at the fundamental frequencies mentioned above and at multiples of the fundamental. So, a piano will resonate at 440 Hz (fundamental) when A4 is played and at 880 Hz (second harmonic) and 1320 Hz (third harmonic) and 1760 Hz (fourth harmonic) and so on. So, the sound of real instruments extends well above their highest fundamental tone. See: Bass, Midrange, Treble.

Frequency Response

When discussing sound quality, we often mention a set of terms related to frequency response. Frequency response is the output level of the device (speaker, amp, DAC, etc) for each relevant frequency (generally 20 Hz to 20 kHz, but possibly higher). A signal of constant level is fed into the device at each frequency, and we measure the output. Since the input is level with frequency, we generally want the output to be level with frequency or “flat”. But there can be cases where we do not want exactly flat response, particularly with speakers where on-axis and off-axis measurements may differ and, e.g., flat on-axis response may not sound accurate.

Frequency Balance (overall)

We often characterize the frequency response of a device using the term frequency balance. When we are speaking about overall balance, we usually mean the basic shape of the frequency response curve: is it tilted up in the treble or up in the bass or scooped (depressed) in the middle or rolled off in bass and treble?

Frequency Balance (octave to octave)

We may discuss frequency balance in octave-to-octave terms. Octaves are just a doubling of frequency. So the octave above 41 Hz extends to 82 Hz, and the octave above 440 Hz extends to 880 Hz. Octave-to-octave frequency balance is a useful way to communicate if local regions of frequency response are smooth and even or bumpy or peaky. The smoother an octave and the next one are, the more instruments in that range sound right, because the fundamentals and harmonics are in balance. It helps to understand the the harmonic character of an instrument is how we know a guitar from a cello or a piano from a clarinet.

Midrange

Middle frequencies of musical instruments and voices, generally from approximately 300 Hz to 3000 Hz.

Noise Level

This term usually does not refer to audible noise in the sense of noise you hear explicitly as you would with wind noise in a car. Rather, reviewers and audiophiles estimate the very low-level noise of audio devices by observing how the device affects small signals (instrumental overtones or venue reflections). Sometimes a low noise level is referred to as a “black background”.

PRaT (Pace, Rhythm and Timing)

While it is easy to think of musical sounds in what is called the “frequency domain”, meaning in terms of the musical notes with their frequencies (e.g. A4 on piano has a fundamental of 440 Hz), we also need to think of the output of audio equipment in terms of time. Almost all audio equipment has some amount of time distortion, meaning that certain tones that should have occurred at time t=X will completely or partially occur at t=X+.01 seconds or t=X+.2 seconds. This timing error leads to observational terms like “blur” and “overhang” and “slowness” and “softness”. As a way of summarizing how well a device limits time error, some reviewers use the term PRaT (Pace, Rhythm, and Timing) to capture the accuracy or lack thereof in the time dimension.

Realism

The idea that reproduced music sounds, or does not sound, as it would when performed. A difficult concept because of the listener’s generally limited knowledge of the original performance environment, and the heavy use of studio techniques in which the music is not performed in a real space at one time. See: Believability.

Resolution

By visual analogy, resolution is the ability of a device to produce separate aural images of closely spaced objects (could be closely spaced in location or in time). A high resolution sound field has clarity, depth, and little blur. We perceive resolution as definition of sounds.

Soundstage

The reason for having stereo (2 channel) or potentially even more channels is to present the music as occurring in a 3-dimensional space. Soundstage refers to the 3-dimensional presentation of performers, primarily in the left-right and front-back elements of their positioning on a virtual stage. We often refer also to the overall dimensions of the virtual stage that is presented, articulating whether it is wide or narrow and shallow or deep.

Soundspace

This term refers to the sense that an audio device gives of the size and shape of the overall space in which the virtual performers appear. Remember that the instruments and voices resonate to make sounds. These sounds travel out from the instruments and if the performance were in a large concert hall, the reflections would occur perhaps 1 or 2 seconds after the initial sound from the instrument made its way to your ears. The time delay of the reflections is used by your ear/brain to sense the size and shape of the concert hall. If the performance were in a small club, the reflections would occur more quickly and your ear/brain would sense the smaller venue.

the absolute sound

This term refers to the sound of real instruments and voices played in a real space. The idea of the absolute sound is to create a reference or a standard for observationally objective evaluation of the sound of audio equipment. If the sound of audio equipment believably resembles the absolute sound, we judge that it is performing well. If the sound does not believably resemble the absolute sound, then we judge that it is performing poorly. Since nothing is perfect, these evaluations must necessarily comprehend the tradeoffs that all real audio equipment make. Listeners can learn what the absolute sound means by attending concerts, especially those involving acoustic or limited-amplification instruments and voices. Without such a reference communication between listeners is rendered limited if not impossible. See: Believability.

Tilt

See Frequency Balance.

Transparency

Transparency is used to indicate the sense that a device can transmit a signal faithfully. In music reproduction, this means the device gets closer to what we imagine happened at the live (studio or concert) event. Transparency is mostly a combination of resolution and naturalness. We add the naturalness criterion because at times, practically speaking, there are some artificial distortions that can seem to enhance resolution at the price of naturalness.

Treble

Upper frequencies of instruments and voices, generally from approximately 4000 Hz to 20,000 Hz or higher.

Warm

Refers to somewhat elevated mid-bass and lower midrange, generally somewhere in the range from 80 Hz to 400 Hz.