EQ and Compression Order: Why Signal Chain Matters

Department of Audio Engineering | Level: Intermediate | Duration: 35 minutes

Objectives

Understand the technical differences between Comp→EQ and EQ→Comp signal chains
Identify when each order produces better results on vocals
Apply the EQ→Comp→EQ sandwich technique for maximum control
Recognize frequency-dependent compression artifacts and how to prevent them

1. The Core Question

Does it matter whether you compress before or after EQ? Yes — measurably so. The order changes how the compressor reacts to frequency content, which affects the naturalness of the dynamics and the clarity of the tone.

The reason is simple: a compressor responds to level. If certain frequencies are louder (e.g., proximity effect boosting 100-200 Hz, or sibilance peaking at 4-10 kHz), the compressor reacts to those frequencies, not just the overall vocal performance.

2. Compression Before EQ (Comp→EQ)

Vocal → [Compressor] → [EQ] → Mix Bus

What happens:

The compressor sees the raw signal — including problematic frequencies
If sibilance is loud (4-10 kHz peaks), it may trigger compression on those transients
The compressor “pumps” on problematic frequencies rather than controlling overall dynamics
EQ afterward shapes the already-compressed signal

When to use:

When the vocal is well-recorded with minimal frequency problems
When you want the compressor to react to the full, natural signal
When using gentle compression (2:1 ratio, slow attack) for “glue”

Risk: Frequency-dependent pumping. The compressor doesn’t know you don’t want it reacting to the 200 Hz proximity hump — it just sees level.

3. EQ Before Compression (EQ→Comp)

Vocal → [EQ] → [Compressor] → Mix Bus

What happens:

Corrective EQ removes problems first: cut proximity at 200 Hz, tame sibilance at 6 kHz
The compressor receives a cleaner, more balanced signal
Compression responds to the musical performance, not frequency artifacts
Result: more natural, transparent compression

When to use:

When the vocal has noticeable frequency problems (proximity, room resonance, harshness)
When you want the compressor to respond to dynamics, not frequency peaks
When recording quality is variable

This is generally the safer default — fix the frequency problems before asking the compressor to manage dynamics.

4. The Sandwich: EQ→Comp→EQ

Vocal → [Corrective EQ] → [Compressor] → [Tonal EQ] → Mix Bus

This is the professional standard for a reason:

First EQ (corrective): High-pass filter at 80-100 Hz, cut mud at 200-300 Hz, notch room resonances. This is surgical — you’re removing problems, not shaping tone.
Compressor: Now reacts to a clean signal. Set ratio (3:1-4:1 typical for vocals), threshold to catch ~6 dB of gain reduction, medium attack (10-30ms) to preserve transients, medium release (50-100ms).
Second EQ (tonal): Now shape the sound creatively. Boost air at 10-12 kHz, add presence at 3-5 kHz, warm up the low-mids. This EQ sits after compression, so your boosts won’t trigger the compressor.

Why it works: Separation of concerns. The corrective EQ prevents compressor artifacts. The compressor manages dynamics on a clean signal. The tonal EQ shapes the final character without affecting dynamics.

5. Frequency-Dependent Artifacts to Watch For

Problem	Cause	Fix
Pumping on plosives	Low-frequency bursts triggering compressor	HPF before compressor (EQ→Comp)
Sibilance amplified	Compressor reduces body, sibilance stays	De-esser before compressor, or EQ cut at 6 kHz first
Dull after compression	Fast attack squashing transients	Slow attack (15-30ms), or parallel compression
Inconsistent tone	Compressor reacting differently to quiet vs loud sections	Use 2 stages of gentle compression instead of 1 heavy stage

6. Dynamic EQ: The Modern Alternative

Dynamic EQ combines EQ and compression in one processor. Each EQ band only activates when the frequency exceeds a threshold — like a compressor that only works on specific frequencies.

Use case: Sibilance that varies throughout the performance. A static cut at 6 kHz would dull the entire vocal, but a dynamic EQ cut only activates when sibilance exceeds the threshold.

This doesn’t replace the Comp→EQ question — it’s a specialized tool for frequency-dependent dynamics problems.

Practice Exercise

Exercise 1: A/B the Order

Take a vocal recording. Set up two parallel chains:

Chain A: Compressor (4:1, -6 dB GR) → EQ (boost 3 kHz +3 dB, cut 250 Hz -4 dB)
Chain B: Same EQ → Same Compressor

Listen to both. Where do you hear the difference? Focus on:

Low-frequency consistency (proximity effect handling)
Sibilance level
Overall “naturalness” of the compression

Exercise 2: Build a Sandwich

Set up: HPF at 100 Hz + cut at 250 Hz → Compressor (3:1) → Presence boost at 4 kHz + Air at 12 kHz. Compare against a single-EQ-then-compress chain.

Key Takeaways

Order matters — the compressor responds to whatever frequencies are loudest in the input
EQ→Comp is the safer default — fix problems before compression for more natural results
EQ→Comp→EQ sandwich is the professional standard — corrective first, tonal last
There is no universally “correct” order — it depends on the recording, genre, and intent
Dynamic EQ is a modern tool for frequency-dependent dynamics problems
The goal is always: control dynamics without destroying the natural character of the performance