ITU-R BS.1116-3, “Methods for the subjective assessment of small impairments in audio systems”
Dr. Floyd Toole, “The Measurement and Calibration of Sound Reproducing Systems”
Dr. Sean Olive, “The Subjective and Objective Evaluation of Room Correction Products”
These standards and best practices describe various specifications required for accurate sound reproduction. For each specification there is a “range of operation” usually specified with a “tolerance.”
For example, for in-room frequency response, there is specification for “operational room response curve” with a specified ±3 dB tolerance. See section 2.4 on page 6 of the EBU Tech 3276 standard:
The operational room response curve is an important criterion in the evaluation of the mutual influence of the loudspeaker and the listening room, and hence for the assessment of the listening conditions. It corresponds well with the subjective assessment of reproduced sound. For stereophonic reproduction, the close matching of the room response of each channel is important.
We have taken the operational room response curves from each standard plus best practices based on subjective listener preferences and overlaid onto one chart below:
This is the “range of operation” for room response curves each having a ±3 dB tolerance. When your loudspeakers in your room are calibrated, we start with these four operational room response curves to determine which one sounds best to you.
Once one is selected, we fine tune the response, within tolerance, to best match your personal preference. This may take two or three iterations to finalize your specific operational room response curve. Once set, it should sound good with a wide range of music and media content.
In addition to frequency response, here are a few summarized target specifications from the standards and best practices:
Reverberation time is separated into two measures. One is above the room’s transition frequency and one below. This specification is for broadband reverberation time above the room’s transition frequency to 8 kHz.
We use REW’s Topt reverberation measurement for small room acoustics above the transition frequency to determine how absorbent or lively the room sounds relative to industry specifications.
The specification has a range of operation and a tolerance. In addition to the standards, best practices have shown listener preferences for a more lively room. The range also depends on room volume. Typical ranges are 250 ms for a damped room to 500 ms for a lively room, again depending on room volume.
Referring to the reverberation specification above, we can see there is an allowable rise in decay time at frequencies below 100 Hz. E.g. 300 ms at 63 Hz. Due to small room acoustics below the rooms transition frequency, room resonances are prevalent and audible in almost all rooms. To make it more challenging, long wavelengths below 100 Hz are difficult to damp using passive acoustic treatment. We use specialized DSP software to control room resonances.
The default specification is for an equilateral triangle with the speakers toed in pointing at ones ears. This will ensure you are listening to the same stereo image as the artists, engineers and producers intended.
There is a range and tolerance for this specification.
Reference listening level specification
The reference level for critical listening is a range of operation depending on musical content, how it was recorded, mixed and mastered. For highly compressed music, an average listening level is approximately 77 dB SPL C weighting with slow integration using a calibrated Sound Pressure Level (SPL) meter. For uncompressed, wide dynamic range content, 83 dB SPL. The EBU and ITU standards specify an 82 dB SPL A weighting.
The listening level is important as the final tonal balances for mixes and masters are set in the studio and mastering lab while listening at reference level. Our ears hear the “flattest” at this sound pressure level. So listening at this level, you are listening as the artists intended.
Our ears frequency response is non-linear, so at lower levels, our ears are less sensitive to bass. This is why we use loudness controls like JRiver for example to compensate so the music sounds full at low listening levels.
Early reflections are defined as reflections from boundary surfaces or other surfaces in the room which reach the listening area within the first 15 ms after the arrival of the direct sound. The levels of these reflections should be at least 10 dB below the level of the direct sound for all frequencies in the range 1 kHz to 8 kHz.
One can design and setup a sound reproduction system to minimize early reflections. In unavoidable circumstances, passive acoustic treatment can be used to absorb or diffuse early reflections.
The step response graph illustrates a 3 way active system using linear phase digital crossover with driver time alignment. All drivers are perfectly aligned in the time domain (i.e. the vertical step). The result is that all frequencies from all drivers are arriving at one's ears, all at the same time. A "textbook" minimum phase response following the target perfectly. This is accurate sound reproduction in the time domain.
The standards include operational loudspeaker specifications for:
Frequency response: ±4 dB from 40 Hz to 16 kHz. ±3 dB for directional angles in the range +10° +4 dB +30°.
Directional characteristics: The directivity index should be within the following limits over the frequency from 250 Hz to 16 kHz: 4 ≤ D ≤ 12 dB Large, frequency–dependent variations are undesirable, especially in the high–frequency range (f > 500 Hz).
Distortion: at 90 dB SPL in the frequency range from 250 Hz to 2 kHz. With respect to this level, no harmonic distortion component shall exceed the following values: –30 dB (3%) for 40 Hz < f < 250 Hz –40 dB (1%) for 250 Hz < f < 16 kHz.
Dynamic range: The maximum operational sound pressure level, measured using a sound level meter set to flat response and RMS (slow) at the reference distance of 1 m is: Leff–max ≥ 108 dB
Separate low–frequency loudspeakers: The overall requirement is that the locations of the separate bass loudspeakers must not be audibly apparent.
See standards for a complete list of loudspeaker operational specifications. One can compare existing loudspeakers to the specifications or use as reference when purchasing new loudspeakers.
The gold standard for measuring consumer loudspeakers is the ANSI/CTA 2034-A, “Standard Method of Measurement for In-Home Loudspeakers.” This is a free download available to anyone who is interested in the state of the art in loudspeaker measurements.
“This standard describes an improved method for measuring and reporting the performance of a loudspeaker in a manner that should help consumers better understand the performance of the loudspeaker and convey a reasonably good representation of how it may sound in a room based on its off-axis response and how this response affects the consumer’s experience. It also describes how to use this directivity data to estimate the in-room frequency response that more recent research has shown correlates well to subjective listening preferences of consumers.”
Consumers should insist on published loudspeaker reports as described in the standard from loudspeaker manufacturers. With "spinorama anechoic measurements" and "estimated in room frequency response" charts, consumers can make informed choices based on specifications that subjectively correlate to how accurate the speakers might sound in your listening room.
We recommend purchasing well designed loudspeakers with smooth on and off axis frequency response for accurate sound reproduction.
Well-designed loudspeakers should only require calibration below 500 Hz to smooth low frequency room response and remove room resonances.
accurate sound delivered worldwide
© 2020 Accurate Sound Reproduction Services