Compressor Types
Optical (Opto)
An optical compressor uses a light source and a photocell to control gain reduction. When the signal gets loud, the light gets brighter; the photocell detects the brightness and applies corresponding gain reduction. The physics of light-to-resistance conversion creates a naturally smooth, slow response — optical compressors are inherently gentle.
Character: Smooth, musical, forgiving. The attack and release are program-dependent — the compressor responds differently depending on the material. It’s difficult to make an optical compressor sound harsh.
Classic hardware: The LA-2A is the most famous optical compressor. It has two controls: peak reduction (threshold) and gain. That’s it. No attack knob, no release knob, no ratio knob. The circuit does what it does, and what it does happens to sound excellent on vocals, bass, and anything where you want transparent, musical leveling.
Typical use cases: Vocals, bass, acoustic guitar, any source where transparency matters more than aggressive control.
FET (Field-Effect Transistor)
A FET compressor uses transistors to simulate the behavior of a tube compressor, but faster. FET designs can respond to transients extremely quickly — faster than any other compressor type.
Character: Aggressive, punchy, fast. FET compressors can catch transients that other designs miss. They add a characteristic “edge” — a forward, in-your-face quality that’s useful on drums, vocals, and guitars.
Classic hardware: The 1176 is the definitive FET compressor. It has continuously variable attack and release (both very fast at their extremes), selectable ratios, and a distinctive sound that ranges from subtle to aggressive depending on settings. The “all-buttons-in” mode — engaging all four ratio buttons simultaneously — produces a heavily distorted, crushed sound that’s become a creative effect in its own right.
Typical use cases: Drums, aggressive vocals, electric guitars, anything that needs impact and energy.
VCA (Voltage-Controlled Amplifier)
VCA compressors use a voltage-controlled amplifier as the gain reduction element. They’re precise, predictable, and flexible — the most “neutral” compressor type.
Character: Clean, accurate, transparent. VCA compressors can do almost anything — from gentle bus compression to aggressive peak limiting. They don’t add as much harmonic character as optical or FET designs, which makes them ideal when you want compression without coloration.
Classic hardware: The SSL bus compressor (built into SSL mixing consoles) is the most iconic VCA design. It’s the standard for mix bus compression — the “glue” compressor that holds a mix together.
Typical use cases: Mix bus, master bus, any situation where you want precise control without added character.
Tube (Variable-Mu)
Tube compressors use vacuum tubes as the gain reduction element. As the input level increases, the tube’s gain naturally decreases — compression is an inherent property of the tube, not an external circuit controlling it. This produces the smoothest, most gradual compression of any design.
Character: Warm, thick, colorful. Tube compressors add harmonic saturation (primarily even-order harmonics — the “warm” ones from Chapter 6). The compression is slow and gentle, with a natural soft-knee behavior.
Tube compressors are also called Vari-Mu compressors, because the tube’s gain varies with the signal level (mu is the Greek letter for gain). This is the most “analog” of all compressor types — the gain reduction is a natural property of the tube itself, not a circuit controlling a separate element.
Typical use cases: Master bus, vocals, entire mixes where you want warmth and character.
When the Type Matters (and When It Doesn’t)
The differences between compressor types are real — but they matter less than good technique. A well-set VCA compressor will outperform a poorly-set FET compressor on any source. The type adds flavor; the settings do the work.
That said, when you have a choice:
- Transparent leveling → optical
- Punch and impact → FET
- Precision and flexibility → VCA
- Warmth and character → tube
These are starting points, not rules. Many engineers use “wrong” compressor types to great effect — a slow optical compressor on drums for smooth, vintage-feeling control, or a fast FET on a vocal for aggressive presence.
Serial Compression
One compressor doing heavy work sounds worse than two compressors each doing half the work. This is serial (or multi-stage) compression: placing two compressors in sequence, each doing moderate amounts of gain reduction, instead of one compressor working hard.
Why it works: a compressor that’s reducing 12 dB on peaks is working hard — the sound of the compression itself (the pumping, the artifacts, the distortion) becomes audible. Two compressors each doing 6 dB of reduction sound more natural, because each one is operating within its comfortable range.
Now that you know compressor types, serial compression becomes a strategy: a fast FET (like an 1176) catching transients in front of a smooth optical (like an LA-2A) handling overall leveling. The FET does the quick work; the optical does the steady work. Each operates in its sweet spot, and the result sounds more natural than either one working hard alone.
Signal flow showing two compressors in series: FET (fast, catching transients) feeding into optical (smooth, leveling), vs single compressor doing 12 dB.
This is a technique, not a rule. You don’t need two compressors on everything. But when one compressor sounds strained, splitting the work can help.
Parallel Compression
Parallel compression — sometimes called “New York compression” — blends a heavily compressed signal with the original uncompressed signal. You get the consistency and density of compression plus the dynamics and transients of the dry signal.
The setup: duplicate the track (or use a send), compress the duplicate aggressively — high ratio, lots of gain reduction, fast attack. Then blend it in underneath the original. The compressed copy fills in the quiet parts (bringing up details and sustain), while the uncompressed original preserves the transient peaks and dynamic range.
Signal flow: original splits into dry path and heavily compressed path, both blend at output. Dry preserves transients, compressed adds density.
The result is louder, denser, and more present — but with more natural-feeling dynamics than you’d get from compressing the single signal the same amount. You’re getting the benefits of heavy compression without the artifacts.
Most modern compressors include a wet/dry mix or blend knob that does this internally — dial in heavy compression, then pull the mix knob back to blend the dry signal in. Same concept, simpler routing.
Parallel compression works on individual tracks, on buses (particularly drums), and on the mix bus. It’s one of the most versatile techniques in mixing.
Bus Compression
Bus compression — also called glue compression — is compression applied to a group of tracks via a bus (Chapter 11). Instead of compressing each drum track individually, you compress the drum bus. The compressor responds to the combined signal, which means it reacts to the loudest element (usually the kick or snare) and subtly pushes the whole group together.
The mix bus is the bus that all other buses feed into — the final output before mastering. Mix bus compression is the most debated topic in mixing. Some engineers swear by it. Others never use it. The key is understanding what it does: a gentle compressor on the mix bus (1.5:1 to 3:1, slow attack, auto or medium release, 1-3 dB of gain reduction) makes the mix feel more cohesive — like all the elements belong in the same room. Push it too hard and the mix pumps unnaturally.
A caution: bus compression affects everything downstream. If you add bus compression to your drum bus, every drum processing decision you make afterward is colored by that compression. Put bus compression on early and mix into it, or leave it off until the end — don’t add it halfway through.
Stereo Compression
When compressing a stereo signal (a bus, the mix bus, or a stereo track), you have two options:
Stereo-linked (the default): Both channels are compressed by the same amount, determined by the louder of the two. A loud hit panned left compresses both left and right equally. This preserves the stereo image — the balance between left and right stays consistent.
Dual-mono (two independent compressors): Each channel is compressed independently. A loud hit panned left only compresses the left channel. This can create subtle stereo movement — the image shifts slightly as different elements trigger different amounts of compression on each side.
For most mixing and mastering, stereo-linked is what you want. Dual-mono can be useful for creative effects, but it can also cause the stereo image to wander distractingly. The Fairchild 670’s stereo behavior falls somewhere between these two modes, which is part of what gives it its distinctive character on the mix bus.
What to Practice
- Compare compressor types. If your DAW or plugin collection includes different compressor models, compress the same signal with each type — optical, FET, VCA. Match the gain reduction amount. Listen for the differences in character, transient response, and harmonic content.
- Set up parallel compression on drums. Send your drum bus to a return track. Put a compressor on the return with aggressive settings (8:1 or higher, fast attack, moderate release, 10+ dB of gain reduction). Blend the crushed return underneath the dry drums. Notice how it adds density and punch without squashing the original.
- Try a wet/dry blend. On a vocal track, set a compressor to heavy compression (8:1, 8-10 dB gain reduction). Pull the mix knob to 30-50%. Compare this to a lighter compression setting at 100% wet. The parallel approach often sounds more natural at similar perceived density.
