What is the difference between Chorus effects and Automatic Double Tracking?

[Jaidy]

After looking up Chorus effect and Automatic Double Tracking in Wikipedia, they both appear to be described as involving the same theory, that of adding a delayed and very slightly de-tuned signal(s) to the original.

Generally what is the difference, in terms of what they do to produce the final stereo signal?

To my ears they sound alike when applied to instrumental signals, while ADT seems to sound the best on vocals.

[Daimonicon]

A big diffirence between them. E.g doubble tracked guitars won’t be exact the same. No matter how tight they are no one can play exact the same. No matter if it has been done thousand times. You won’t hit the strings exactly the same. That’s what it is makes it human. It’s like a drummer. He won’t hit the snare with the same force on every hit. If that was possible it would sound like a robot and boring. A singer won’t hit the notes e.g in a chorus exactly the same with four takes. Instead they sing them a little different.With difference it will sound richer and bigger and more interesting. Also with doubble tracked guitars the mic position won’t be same or another mic(s) will be used if same amp is used. Or another cab or another amp and cab. If one use just the same it won’t sound like stereo. Delay is used to thicken up one sound source and to gel the sound to better fit in the mix. E.g ping pong delay is very common on vocals and guitar solos. Reverb is usually better for chorus and not lead song.

[seismic1]

As far as I'm aware, ADT works by generating a delayed version of the original signal (no feedback) to create a fatter/doubled version of the original signal. Chorus uses a delayed, modulated pitch-shift to create a shimmering effect.

[kvaca]

After looking up Chorus effect and Automatic Double Tracking in Wikipedia, they both appear to be described as involving the same theory, that of adding a delayed and very slightly de-tuned signal(s) to the original.

basically both techniques are using voice cloning and pitch shifting /PS/ to enhance sound...
but they do differ a lot:
ADT uses PS directly on delayed signal clones /originaly using 2 tape decks which were out of sync, today using PS dsp/ so there is no modulation usually involved, while Chorus uses periodical time modulation on his delay lines /see rate knob/...
and they differ in hearable results,too - ADT can only create beating sound artifacts while chorus always create slight movement-vibrato-like artifacts,depending on rate setting

Generally what is the difference, in terms of what they do to produce the final stereo signal?

they both are usable for stereo widening /mono-stereo/ or can be used in mono only, it depends what sound you are after

[mkruse]

To really understand the difference, you will have to go back to how these effects were created in the studio using tape decks, how they were used and what time period. These days a flanger or a chorus sounds very different to how flanging was done in the '60s.

Double tracking as you know, is recording the same part twice (by a human). As we cannot play computer perfect, each take will be slightly different (timing, tonality & pitch).

Automatic Double Tracking was achieved in the studio, by routing the vocal or instrument to a very short tape delay that was then mixed together in parallel. ADT is said to have been invented in '66.

ADT is a form or chorus by definition. It is a secondary copy of the original signal that is delayed and pitch modulated. The delay was created by the tape delay machine and the pitch was modulated, not by a LFO but by the inability of humans to perform each take the same. Actually, the tape machine would also affect pitch as tape often had wow and flutter, whereby the tape would slow down slightly (due to wear and tear) and then have to speed back up again.

Chorus as we know it today achieves the same effect using different technology (BBD); the Roland CE1 was released in '76 and there were also string machine synths that incorporated built in chorus. Chorus splits your input signal into two, delays one side and uses a LFO to slightly modulate the delay time which affects the pitch. Same effect, different technology. A tri-chorus has three separate delay lines. Dimension has multiple delay lines, but then does some trickery with the phase(?) which creates a 'motionless' sound that defines the '80s.

Flanger is the same as chorus except the delay time is much shorter (generally 1 - 5ms for flanging and 30ms for chorus) and you have regeneration. Regeneration is the feedback on your delay pedal/plugin. It increases the intensity of the effect as you send the (extreme) 1ms delay back into itself. The delay is modulated with an LFO which in turn affects the pitch.

Flanging was originally invented in in the '50s. The engineer would have two tape reels playing the same track. Then he/she would place a finger on one of the tape reel edges adding friction and slowing it down (and out of time with the other tape reel). At the engineers discretion, they would then take their finger off the tape reel and it would speed back up to the first reel thus creating the whoosh. The process would then begin again.

So to answer you question. Chorus and ADT are achieving the same effect using different technologies. Which one you use as a plugin, depends on what sound you prefer, and if the plugin in modelling certain hardware. A chorus will always sound like a pronounced effect, as they are modelling the analogue/digital boxes. ADT plugins may be much more discreet. An ADT plugin could also be designed by the developers to sound better at certain frequencies for voice, or to clean up any noise or other artefacts that may be present in chorus. That's probably why it's sounds better to your ears for vocals. Same effect, but designed by the developers for a different use and/or sound.

[spigmu]

With ADT you're taking one signal and trying to make it sound like two, not just a nicer/fatter/shimmery version of the one. A chorus generally is still keeping it as the sound of one, with the added niceness. A chorused 6 string guitar may sound like a 12 string but it won't sound like two guitars. A chorused voice doesn't sound like two voices, but that's what ADT wants to sound like, not like a single chorused voice. Generally they deal with different lengths of delay because of that. With a chorus if you go beyond the (very generally) 8 to 25 mill you stop sounding like a chorus (less will flange, more will double and then slap). I would say that an ADT's delay range starts at the top of the chorus's range and goes up until it starts to sound too much like a slap echo, which is well beyond the delay of a chorus. As well, an ADT's parameters are set to hone in on the sound of another voice, not just an affected voice, so it wants to achieve some sense of randomness to the extra voice, via the LFO waveform used for example, such as a "random" wave setting which won't be used much for a chorus, which generally wants its modulating to be consistent and repeating, which an ADT usually doesn't, because predictability makes it sound more like an effect on a singular sound, like a chorus or delay. Or even a long, slow rate on a modified square, triangle, etc wave so that the pitch is detuned but being delayed longer, so it's not the shimmering, musical beating of a chorus, but the sound of another voice sometimes up by 5 cents, sometimes down by 15.

Not that the two couldn't have settings in common, but since the goal is different they usually don't.

[mumpcake]

I have to agree with spigmu over mkruse regarding chorus times. 30ms is far too long. Most of the chorus settings I have liked have been between 8-14 ms.

According to this thread, the Diamond Halo Chorus has a trim pot which ranges from 7.5-20ms (and is set to 10ms at the factory).
https://www.thegearpage.net/board/index ... rs.814659/

Most of the articles I have read regarding ADT have said that is around 100ms (some say 50-100ms, some say 80-120ms, either way much longer than most choruses).