EQ for Live Sound: Corrective vs Creative EQ

The Two EQ Domains

Every live sound system has at least two EQ stages: system EQ (applied to the output before amplification) and channel EQ (applied to each input on the mixing console). They serve entirely different purposes, and confusing them is one of the most common mistakes in live sound.

System EQ: Correcting the Room

System EQ compensates for the combined response of the loudspeakers, room reflections, and listener position. It is determined by objective measurement, not by listening. The process:

1. Measure the transfer function with pink noise through the system.
2. Identify deviations from flat response where coherence exceeds 0.6.
3. Apply broad parametric EQ corrections (Q = 0.5-2.0) to flatten the response.
4. Verify with before/after measurement.

The Coherence Threshold Rule

This is the most important rule in system EQ: only EQ where coherence γ² > 0.6.

When coherence is high, the measured response reliably represents the system's behavior. EQ corrections will be effective and predictable. When coherence is low (< 0.5), the response is dominated by room reflections, noise, or interference — the "system" is not causing the problem, and EQ cannot fix it.

Example: a 10 dB dip at 315 Hz with coherence of 0.3 is likely a comb filter from a nearby reflection. Boosting 10 dB at 315 Hz will not fill the dip — it will increase the level of the direct sound AND the reflection, maintaining the same cancellation ratio while wasting amplifier headroom. The fix is physical: manage the reflection with absorption or change the speaker/mic geometry.

Typical System EQ Corrections

Frequency Range	Common Problem	Typical Correction
200-400 Hz	Low-mid buildup (boundary coupling)	-4 to -8 dB, Q = 0.7-1.0
2-4 kHz	Presence peak (speaker horn response)	-2 to -4 dB, Q = 1.0-1.5
8-12 kHz	Air absorption / distance loss	+2 to +4 dB shelf (audience areas only)
80-120 Hz	Sub/main crossover dip	Fix with alignment, not EQ

Channel EQ: Shaping the Sound

Channel EQ is creative and subjective. Its goal is to make each source sound its best in the context of the full mix. Unlike system EQ, channel EQ is applied by ear, not by measurement.

Cut First, Boost Second

The oldest rule in mixing still applies: always try a cut before a boost. If the vocal sounds thin, cutting the competing instruments at 1-3 kHz is often more effective than boosting the vocal at the same range. Cuts preserve headroom; boosts consume it.

Common Channel EQ Moves

• Vocal: High-pass at 80-120 Hz, cut 200-300 Hz muddiness (-3 dB, Q=1), presence boost at 3-5 kHz (+2 dB, Q=1.5)
• Acoustic guitar: High-pass at 80 Hz, cut boominess at 200 Hz, sparkle at 8-12 kHz
• Kick drum: Sub at 50-60 Hz, remove boxiness at 300-400 Hz, attack at 3-5 kHz
• Snare: Body at 200 Hz, ring at 800-1000 Hz (cut if excessive), crack at 5-7 kHz

Graphic vs Parametric EQ

Graphic EQ (31-band or 15-band) was the system EQ standard for decades. Each slider controls a fixed-frequency band with fixed Q. The advantage is visual feedback — the slider positions approximate the correction curve. The disadvantage is limited resolution and band interaction.

Parametric EQ offers adjustable frequency, gain, and Q (bandwidth) per band. It requires fewer bands to achieve the same correction and avoids the band-interaction artifacts of graphic EQ. Modern digital consoles and processors universally use parametric EQ.

The AI diagnostic engine outputs parametric EQ recommendations (frequency, gain, Q) that can be directly applied to any digital processor.

When EQ Cannot Help

EQ is powerless against:

• Comb filtering: Periodic nulls from reflections cannot be filled with EQ — alignment or physical treatment is needed.
• Excessive reverberation: EQ changes the frequency balance of reverberant energy but not its duration. Acoustic treatment is the solution.
• Coverage problems: If a speaker does not cover an area, EQ cannot make it louder there. Speaker aiming or additional speakers are needed.
• Feedback: Narrow notch filters suppress specific feedback frequencies, but system EQ (broadband corrections) does not meaningfully improve GBF.

The SonaVyx Workflow

The Tune PA workflow integrates measurement-based system EQ into a guided process. After alignment (polarity → delay → level), the transfer function reveals what needs correction. The AI diagnostic suggests specific parametric EQ bands. Apply them, re-measure, and verify with before/after comparison. This measurement → EQ → verify loop ensures every correction is validated, not guessed.