reffer to title
I assume you are talking about reverb as is used in music and vocals.
Reverb is technically an echo of sorts. Back in the old days, there were two main ways of generating reverb. Both used basic electromagnetics to generate the reverb effect. One used hanging metal plates that would be vibrated by an electromagnetic driver using the audio signal. This would cause the plate to vibrate, and those vibrations would be converted back into an audio electrical signal, which was then mixed in with the original signal to create the reverb effect. The second type functioned similarly, but used springs instead of plates.
Plate reverb units had a reputation for sounding better, but they were bigger and clunkier and were therefore often used mostly in recording studios. Spring reverb units could be made much smaller and were often built into amplifiers, and were much more common for stage use.
There was also echo effect that, in the old days, was achieved using a tape loop, usually with two separate tape heads. The delay between the write head and the read head created the echo. Then again, sometimes folks just used big echo chambers. The band U2 recorded one album in an old castle to take advantage of the natural reverb effect of the large rooms and stone walls.
These days, all of the above are often simulated digitally, and you can buy effects boxes with all three built into their software.
One of the cool things about reverb is that there are so many cool different ways to do it. ECG covered the primary two vintage approaches, plate and spring reverbs. But various other shapes were used, including drums and goodness knows what. Basically, studio engineers would take a pair of transducers (the things that turn electrical signals into vibration and vice versa) and stick them on just about anything in different places to find new sounds.
What is reverberation to begin with? The electromechanical versions described above were imitations of something that already existed in reality. I’m sure you’ve heard the effect big rooms with hard walls have on sound; just go to any stairwell or cathedral for great examples. To get a sense of what the room is doing, clap your hands sharply, once, and listen to the room’s response. Go to different places and try it: you get different results.
What’s happening? The sound is radiating outward and hitting lots of reflective surfaces and bouncing: sometimes bouncing straight back, other times bouncing off other surfaces before it returns. And the sound that returns often hits something behind you, bouncing it back out again. The effect can continue for quite a while. So, a reverb is just a very complex type of echo.
Complex in three ways: number, time, filtration, and feedback.
Number & Time: you get lots of different echos from different distances, and the time depends on the distance.
Filtering: each time the sound reflects, it gets affected by the reflecting surface and shape of that part of the room. Affected as in having EQ applied, with some frequencies subdued and others reinforced.
Feedback: sending the results back into the process.
And that’s just how many digital reverbs work: they implement a large number of different delays, with different EQ curves, and with various amounts and kinds of feedback. Another factor is what’s called “early reflections”. These are the first bouncebacks that hit your ears after the original sound, and they’re the ones your brain uses for locating sounds in space, especially when the reflections are different in each ear. For example, if the early reflections are faster on the left than the right, the sound will mentally seem to be in the left hand side of a room. (It can seem like the sound source is to the left, or you the observer is to the left, depending on how it’s done.) Fiddling with the parameters, you can also place a sound source far back in a room, or up close. (Oversimplification: quiet reverb for up close, loud reverb for way back.)
And there’s another really nifty way of doing reverb, using a math trick called “convolutions”. What you do is place a sharp sound source (like a starter pistol) in one place and a stereo mic in another, in a room whose reverb you want to imitate. Fire off the pistol, record the result. You use that recording as the “convolution filter kernel”, meaning, you use it as the control to all delaying and feeding back (in a digital filter program or plugin).
So, you go to Carnegie Hall, set up a recorder in the audience, fire a starter pistol on the stage and record it. Later, back in the studio, play your guitar and feed it through this software with this recording used as the configuration “kernel”, and the result is pretty darn close to how you’d sound, on the stage at Carnegie Hall!
Which, in my case, would sound like someone who needs more practice.
Back in the old days, they actually had reverb chambers: dedicated rooms with hard walls, usually only two of which were parallel, a loudspeaker at one end, and a mike at the other. Here are some pictures, including, apparently, modern-day ones.
ECG’s explanation was a very good one.
One distinction I would like to see made, and it isn’t always, is to keep the definition of reverb distinct from that of what is most commonly called “delay.”
At least in my guitarist/vocalist world, this is important. It’s the distinction between the sound of your voice in a cave or large room (reverb) vs. the sound of your voice coming back to you in distinct, repeating (and usually dwindling) iterations. The latter I guess you’d get if you shouted into a canyon with multiple hard surfaces that your voice bounces off of at different times.
The classic musical example, of course, is the fabled “Sun Studio Echo” on the Sun label recordings of Elvis Presley, Carl Perkins, Jerry Lee Lewis and others. This was achieved, as ECG notes, by feeding the guitar and/or vocal signal through a tape recorder and mixing the recorded signal from the playback head back into the main one. By varying the position of the tape head or the speed of the transport, the elapsed time between the initial attack and subsequent repeats could be changed.
In more recent times, U2’s The Edge used this technique extensively on his guitar, especially in the band’s early days.
With reverb, by contrast, you tend to hear the attack of the original note only once, with a sort of soft aura of sound hanging in the air around it — rather than a repeat of that note’s initial attack.
Side note…
I had the opportunity for a “behind-the-scenes” tour of the Rock and Roll Hall of Fame Library and Archives in Cleveland a year or so ago thanks to a friend of my son’s, who was working there at the time. (An absolutely amazing experience, by the way.)
He told us that this one particular flat-file drawer contained the original architectural blueprints for Gold Star Studios in Los Angeles, where Phil Spector made many of his classic recordings.
I said, “Well, then somewhere in here there has to be a representation of Gold Star’s echo chamber” — which, of course, was legendary for its unique sound.
And sure enough, a couple of massive sheets later, there it was! I got a real kick out of that.
One more fun reverb/echo story…
Legendary bluesman John Lee Hooker made the first recordings of his career in the late 1940s in Detroit. Producer Bernie Besman ran the output of his vocal microphone into a small speaker, and then (with a set of very long wires) ran that speaker down the hall to a studio restroom, where he placed it in a (hopefully unused!) toilet. He then put a microphone on that speaker and mixed its output back into the main vocal signal — producing an interesting echo effect.
modern day reverberation chambers are useful for testing the efficacy of sound absorption materials. I used to work for a test lab, and the main lab elsewhere in the state had a reverb room for this sort of testing.