Question 1

What are the three parameters of a parametric EQ?

Accepted Answer

A parametric equalizer filter has three controls: Frequency (the center frequency of the filter), Gain (how much to boost or cut at that frequency, in dB), and Q (quality factor, which determines the bandwidth). Frequency sets where the filter acts. Gain sets how much change is applied. Q determines how wide the affected area is: Q of 0.5 affects about 2.5 octaves, Q of 1.0 affects about 1.4 octaves, Q of 2.0 affects about 0.7 octaves, and Q of 10 affects about 0.14 octaves. A low Q creates a broad, gentle shape. A high Q creates a narrow, surgical shape. SonaVyx AI diagnostic recommends specific frequency, gain, and Q values based on measured deviations from the target curve.

Question 2

What is the difference between parametric, graphic, and shelving EQ?

Accepted Answer

A parametric EQ offers continuous control of frequency, gain, and Q for each filter band. It is the most flexible and is used for precise corrective EQ. A graphic EQ has fixed center frequencies (typically 1/3 octave spacing: 31 bands from 20 Hz to 20 kHz) with adjustable gain only — the bandwidth is fixed. It provides less precision but is intuitive for visualizing the overall EQ shape. A shelving EQ boosts or cuts all frequencies above (high shelf) or below (low shelf) a corner frequency. It is used for broad tonal adjustments: adding warmth (low shelf boost) or reducing brightness (high shelf cut). For sound system tuning, parametric EQ is preferred because of its surgical precision. SonaVyx recommendations use parametric EQ parameters.

Question 3

How many parametric EQ bands do I need for system tuning?

Accepted Answer

For most system tuning, 6-12 parametric bands per speaker zone are sufficient. Allocation: 2-3 bands for broad tonal corrections (low shelf, high shelf, and a mid parametric), 2-4 bands for correcting room-related peaks (room modes, boundary reflections), 2-3 bands for speaker deficiencies (dips or peaks in the speaker response), and 1-2 bands reserved for feedback notching if needed. Professional system processors (Lake, BSS, Symetrix) typically offer 12-24 bands per output. Using more than 12 bands for corrective EQ usually indicates an underlying problem that EQ cannot solve (room modes, reflections, speaker placement). SonaVyx AI diagnostic typically recommends 3-8 filters for a well-set-up system, or 8-12 for a challenging room.

Question 4

What Q value should I use for different types of corrections?

Accepted Answer

Broad tonal corrections (overall brightness, warmth): Q of 0.5-1.0 (2-3 octave width). These shape the overall spectral balance. Mid-range corrections (presence, nasal quality, box resonance): Q of 1.0-3.0 (1/2 to 1.5 octave width). These address specific coloration. Room mode corrections: Q of 2.0-6.0 depending on the mode width. Room modes are narrower at lower frequencies. Feedback notching: Q of 8-30 (very narrow). The goal is to suppress a single frequency without audible tonal change. Speaker resonance: match the Q to the resonance width by measuring the transfer function — the measured peak Q guides the corrective Q. As a rule: use the lowest Q that addresses the problem, to minimize artifacts and interaction with adjacent frequencies. SonaVyx AI diagnostic suggests Q values matched to the measured problem bandwidth.

Question 5

What is the difference between peaking, notch, and bell filters?

Accepted Answer

These terms are often used interchangeably but have subtle differences. A peaking or bell filter boosts or cuts around a center frequency with a smooth Gaussian-like shape. It is the standard parametric EQ type. A notch filter is a very narrow cut (high Q, typically 10-30) designed to remove a single frequency. Technically, a true notch has infinite attenuation at the center and very narrow bandwidth. In practice, a notch filter is a bell filter with high Q and negative gain. SonaVyx EQ recommendations use bell/peaking filters for tonal correction and flag when a notch filter is recommended (for feedback suppression or removing a specific tonal artifact). Both types are standard in any professional DSP processor.

Question 6

How does a high-pass filter differ from parametric EQ?

Accepted Answer

A high-pass filter (HPF) removes all frequencies below its cutoff frequency, with a slope defined by the filter order (6, 12, 18, or 24 dB/octave). It is not an EQ in the traditional sense — it is a protection and cleanup tool. Use an HPF to: protect small speakers from low frequencies they cannot reproduce (prevents cone over-excursion and distortion), remove low-frequency noise (wind, HVAC rumble) from the signal, free up system headroom by eliminating inaudible sub-bass content. Set the HPF just below the lowest useful frequency of the speaker. A parametric EQ shapes the response; an HPF eliminates the bottom end. Most system processors apply the HPF before the parametric EQ. SonaVyx diagnostic may recommend an HPF if low-frequency energy below the speaker capability is detected.

Question 7

What are all-pass filters and when would I use one for system tuning?

Accepted Answer

An all-pass filter passes all frequencies at the same level (flat magnitude response) but rotates the phase. It has no effect on the frequency response but changes the time-domain behavior. Uses in system tuning: rotating the phase of one speaker to better match another at crossover, compensating for the phase rotation of a loudspeaker crossover, aligning the phase of a subwoofer to the main system when delay alone is insufficient. The parameters are the same as a parametric EQ (frequency and Q), but instead of gain, the filter provides phase rotation (90 degrees for first-order, 180 degrees for second-order at the center frequency). SonaVyx phase display is essential when setting all-pass filters because you are adjusting phase rather than magnitude.

Question 8

What is filter interaction and how do I avoid it?

Accepted Answer

When multiple parametric EQ filters overlap in frequency, they interact. Two cuts at nearby frequencies can create a deeper combined cut than intended. A boost near a cut can partially cancel the cut. Adjacent filters with high Q values can create ripple in the response. To minimize interaction: space filters at least one octave apart when possible. Use the minimum number of filters needed. Apply broader corrections first (low Q), then add narrow corrections. Check the combined EQ curve on the DSP display before applying. Always verify the actual acoustic result with a measurement after EQ, because the interaction of EQ with the room response is not always predictable from the EQ curve alone. SonaVyx AI diagnostic accounts for filter interaction in its recommendations.

Parametric EQ Fundamentals: Frequency, Gain, Q, and Filter Types

Frequently Asked Questions