Plate, spring and hall reverb: differences, uses and settings

Reverb recreates the reflections of a space around a sound. The main types differ by origin: room, chamber and hall imitate real places, while plate and spring come from mechanical devices. In practice, a plate flatters vocals and snare, a hall suits pads and ballads, a room glues a drum kit together, and a spring colours guitars and dub. The settings that make the biggest difference are pre-delay, decay time and EQ on the return.

What is reverb, exactly?

In a room, the direct sound reaches your ears first. Next come the early reflections: a handful of distinct echoes from the walls, floor and ceiling that tell the brain how big the space is and what shape it has. Finally, late reflections pile up until they blur into a diffuse tail that slowly fades away.

The length of that tail is measured as RT60: the time it takes for the reverb level to fall by 60 dB once the sound stops. Most plugins call it Decay or Time. A small furnished room drops below one second; a large concert hall often goes past two.

Keep these three stages in mind and any reverb plugin becomes easier to read: every control acts on one of them.

What are the main types of reverb?

Room. It recreates a space of a few dozen square metres. Its signature lies mostly in the early reflections: it puts an instrument "in the room" without lengthening the sound. Used subtly, it brings separately recorded tracks closer together.

Chamber. From the late 1940s, studios built dedicated echo chambers: a loudspeaker plays the signal into a hard-walled room and microphones pick up the result. The sound is dense and natural, softer than a plate. Chamber plugins follow the same idea.

Hall. Think of a concert hall: density builds gradually and the tail is long and wide. It is ideal for giving depth to pads, piano or strings, and it can easily swamp a busy mix.

Plate. A large, thin metal sheet stretched in a frame is set vibrating by a transducer, and pickups capture the vibrations. The best-known early commercial model, the EMT 140, dates from 1957. With no walls, a plate has no clear early reflections: density is instant and the top end is bright, which explains its popularity on vocals and snare.

Spring. The signal sets one or more springs vibrating. Made popular by organs and then by guitar amps, this reverb has an unmistakable metallic character, with an audible bounce on sharp attacks. You choose it for its colour, not for realism.

There are also derived effects: gated reverb, cut off abruptly by a noise gate and closely associated with 1980s drum sounds; reverse reverb, which swells in before the sound; and shimmer, which pitches the tail up one or two octaves to build airy textures.

Algorithmic or convolution reverb: what is the difference?

A convolution reverb applies an impulse response (IR) to the signal: a recording of how a real space or device responds to an impulse. The realism is striking, as you are effectively borrowing a specific church, plate or booth. The trade-off is that the IR is fixed: you can shorten it, filter it or delay its start, but you cannot resize the room with a knob.

An algorithmic reverb computes reflections using a network of delays, filters and feedback paths. It does not copy a space; it builds one. Every parameter can be adjusted freely, the tail can be modulated to avoid metallic ringing, and the CPU load stays light.

Neither is better in absolute terms. Convolution shines at placing acoustic instruments in a believable space; algorithmic reverbs lead for creative spaces, long modulated tails and fine-tuning in the service of a mix.

Which reverb settings really matter?

Names vary from plugin to plugin, but you will almost always find the same controls. In the order to set them:

1. Type and Size. Pick the space first: size affects the spacing of reflections and the density.
2. Pre-delay. The gap between the direct sound and the start of the reverb. Between 10 and 40 ms, it keeps a vocal intelligible by letting the attack through before the space; at zero, the source moves further back in the room.
3. Decay (RT60). The length of the tail. Short for a fast, dense song; longer for a slow tempo or a sparse arrangement.
4. Early reflections and diffusion. More early reflections make the space easier to identify; high diffusion smooths the tail, low diffusion makes it grainier.
5. Damping and high cut. In a real hall, highs die away faster than lows. Taming the top of the tail makes the reverb more natural and less sibilant.
6. Low cut. Cutting the bottom of the return, often somewhere between 150 and 300 Hz depending on the source, stops the reverb from thickening the low end of the mix.
7. Modulation. A little movement in the tail prevents metallic tones and widens the image; too much and it sounds out of tune.
8. Mix (Dry/Wet). 100% wet on an aux send; set by ear when used as an insert.

Insert or aux send: how should you route reverb?

On an aux send, a single reverb set to 100% wet receives a share of several tracks. The benefits: a shared space that glues the mix, one place to EQ the return, and a lighter CPU load. It is the standard approach when mixing.

As an insert, the reverb sits directly on the track and the Mix control sets the amount. Keep this for cases where the space is part of the sound itself: a guitar through a spring, a special effect on a single vocal, an automated reverb throw on one word.

A common setup is to prepare two or three returns: a short space (room or short plate) for cohesion, a longer one (hall or long plate) for depth, and perhaps a colour effect.

How do you time reverb to the tempo?

A quarter note lasts 60,000 ÷ BPM milliseconds. At 120 BPM, a quarter note is 500 ms and a bar of 4/4 lasts two seconds.

• Decay: a tail that dies just before the next downbeat lets the groove breathe. On a snare at 120 BPM, aiming for roughly a quarter or a half note is a good start.
• Pre-delay: a small rhythmic value works well, for example a 32nd note (62.5 ms at 120 BPM).

These numbers are starting points; the final call is made by ear, in the full mix.

Which reverb suits which instrument?

Treat these values as guides, not rules. Modern algorithmic reverbs such as NAUPLY Lontana Reverb (coming soon) bring several of these characters together in one plugin, from a small room to a vast space. If you are choosing a reverb on a Mac, see also our guide Reverb plugin on Mac.

Which mistakes should you avoid?

• Setting reverb in solo. It always sounds right on its own; judge the amount in the full mix.
• Leaving the return unfiltered. An unfiltered return muddies the lows and makes the highs hiss.
• Stacking too many spaces. A different reverb on every track blurs the stage: two or three shared returns work better.
• Putting space under the bass and kick. The low end loses definition.
• Ignoring mono. Some very wide tails partly cancel in mono: check the mix on a single speaker.
• Drowning the vocal. If the words become harder to follow, raise the pre-delay or shorten the decay before turning the level down. Our vocal chain guide explains where reverb sits among the other processors.

Key takeaways

1. Choose the type for the space you want: room for cohesion, plate for brightness, hall for depth, spring for colour.
2. Set pre-delay to keep the attack clear, then set decay to suit the tempo.
3. Filter the return: lows cut, highs damped.
4. Use a few shared aux sends rather than one reverb per track.
5. Always judge in the full mix, and check in mono.

Frequently asked questions

What is the difference between plate and hall reverb?

A plate comes from a mechanical device: its density is instant, its tone bright, and it has no clear early reflections. A hall imitates a large room: density builds gradually and the tail is longer and more enveloping. A plate pushes a vocal forward; a hall sets it back in the space.

Should reverb go on an insert or an aux send?

On an aux send in most cases: one reverb at 100% wet shared by several tracks gives a coherent space that is easy to EQ. Use an insert when the space is part of a single track's sound, such as a spring on a guitar.

What pre-delay works for vocals?

Somewhere between 20 and 60 ms is a good starting point: the attack of each syllable comes through before the reverb, so the lyrics stay clear. The denser the song, the more a longer pre-delay helps the vocal stay in front.

Is convolution reverb more realistic?

For reproducing a specific space or device, yes: it uses a recorded impulse response. But it is less flexible. A good algorithmic reverb provides a convincing space with far freer controls, which often matters more when mixing.

To go further, browse all our space effects on the Reverb and delay page.