Isolated saturation doesn't sound good

[sambaji]

Distortion appears to be psychoacoustically similar to loudness in that the as you add it is appears to sound good but on more objective listens it actually sounds bad - loss of dynamic range, clarity, transients and over-all bloated weight to the whole thing.

Your loudness analogy is a good one. What I often find with saturation/subtle distortion is that while the additional frequencies of overtones give more density/richness to a mix, they also fill up the sonic space potentially leading to less openness in the sound when applied to the whole mix, similar to an artist leaving little negative space on their canvas for their subject to breath. That likely explains in part why a lot of heavy metals mixes (e.g., Slayer's Angel of Death) don't sound great to my ear production-wise. The overloading distortion often fills up the sonic space clouding the non- or less distorted signals to some degree. Combine that with over-compression, and it can sound like the mix has gone through the meat grinder, which perhaps is the effect some of these bands are going for--character over clarity?

[cthonophonic]

Isolating the fuzz through phase inversion can be useful if you want to manipulate it independently of the dry signal - by filtering, gating, delay etc. Because it's harmonically related to (but distinct from) the input, you can let it function as a sort of ghostly double. Like any kind of distortion, it's easy to overdo, but it has it's place.

[sambaji]

From Izotope's post on ever-processing,

The new harmonics we get from saturation make drums, synths, and vocals sound livelier and glue together various elements in the mix. This is why many mixers turn to analog-modeled saturation plug-ins to warm up their digital tracks. But it's also easy to go overboard.

As you crank up the saturation dial, the mix will begin to feel crowded and fatiguing. To test whether you’ve gone too far, play the mix from start to finish. If an annoying, hard-edged buzz prevents you from making it all the way though, you’ve gone too far.

To illustrate this point further—ever put an Instagram filter on an image and find the intensity of colors offends your eyes? Well, photographers have a term for that: over-saturation!

[Ah_Dziz]

Listening to just the added signal (through phase inversion) isn't going to give you much info. Like cthonophonic mentioned it can be used for neat effects. I sometimes will smash a signal through some distortion, then extract just the added signal, then felted that and whack it through a fast compressor and into a nice thick plate reverb.
Aside from special cases like that I tend to just distort things that need it. As a rule of thumb I try not to drive any nonlinearities at the mix level past 1-2 dB. If you don't have access to the source (you're Mastering someone else's work or trying to polish up old material) you can use a bit of distortion to add info that may not exist in a a track for whatever reason. Those highs can then be processed further with eq and such to get a better tonal balance. Similarly isolating the bass and adding some distortion for better translation on small speakers is a good old trick and the basis for lots of "bass enhancement" plugins.
At the end of the day I wouldn't worry about what the distortion sounds like in isolation because that's not the context in which anyone will hear it. If you are going to use nonlinear processors on a while mix make sure they are properly up sampled and be sure to A/B check often that you aren't getting out of hand.

[bmanic]

A cool little trick you can do to minimize the bad artifacts, even if your saturation plugin doesn't support oversampling, is to first pre-EQ the signal, before the distortion/saturation, then post-EQ it in the exact opposite direction an equal amount. For instance using a high-shelf EQ to remove 12dB from the highs.. then put it back after wards in equal amounts. This can help change the amount and pattern of aliasing (though very much depending on the saturation shape itself of course) and thus minimize harshness.

That does not work, as aliasing would still occur and you would boost the aliasing as well. It does help if you don’t boost the highs after the distortion though, as the distortion will add highs on its own. The principle of the Aural Exciter...
What would help as well is to split the sound into two bands and only distort the low band...

True, but it DOES work for a lot of different saturations because the series and strength of the harmonics is dependent on how hard you hit them. Take for instance a guitar amplifier, most people know where the "line" between semi-clean to crunch happens, and it happens quite suddenly, within about 6dB of input drive increase.

This is the line you can "dance" around with the EQ trick. Forcing your lower octaves of audio material to hit much harder into the distortion, producing longer louder chains of harmonics, whereas you try to keep the highs from hitting this same line. Thus you reduce how far down and loud the aliasing happens.

Try it for yourself. You can start with a hard clipper. It will sound drastically different.

Also worth noting, the phase around the EQ will warp the phase of the harmonics as well, which in turn has it's own sound (unless you use a linear phase EQ).

[bmanic]

It’s funny, when I was first learning the ropes of audio engineering, there was sort of a push to make sure everything was recorded as cleanly as possible with as much isolation as possible. Any saturation was specifically added as an effect. Pushing things into distortion was frowned upon. Now we love it, but I agree that in some cases it’s used to excess. It can sound great on a bass drum, but too harsh on a drum bus, especially if there’s a lot of cymbals or even a bright snare. As usual, use your ears.

Saturating mixing desk channels has _always_ been a thing, at least in my experience. It was the way to deal with poky snares and kick microphones.. percussion in general.

Tape has been pushed hard ever since it was invented, for the same reasons. Especially in the days where noise was a bigger baddie than distortion. Everything got maximized, within tolerance (of course trying to avoid complete fuzz), to minimize noise.

So not sure what era of recording you are talking about.

Perhaps some genres like classical music are and have been the exception.. but that is all I can think of.

[bmanic]

Saturation is distortion. I get that analogue gear imparts distortions and those that work ITB want those analogue sounds. But just adding distortion with no strategy or specific goal is just looking for a crappy sounding mix.

I've tried many of the distortion tools out there and yes they do affect the sound and generally in a crappy way. Its distortion!

Now I no longer call these tools and techniques "saturation" - I call them what they are: distortion. this helps keep me focused from doing damage to my audio with these tools.

Distortion appears to be psychoacoustically similar to loudness in that the as you add it is appears to sound good but on more objective listens it actually sounds bad - loss of dynamic range, clarity, transients and over-all bloated weight to the whole thing.

You can't "objectively listen" to something that is completely subjective. Your whole post in fact demonstrates that subjectivity with a great example of cognitive bias. You explain that bias very clearly in your post.

You say: "Now I no longer call these tools and techniques "saturation" - I call them what they are: distortion. this helps keep me focused from doing damage to my audio with these tools."
.. which justifies it as "damage" and thus you've cognitively decided that it is "bad" when in fact it is just another color/mood/state for audio.

Having said that, I do understand what you are getting at, and depending on the genre of music or the mood one is going for, I completely agree with you!


Distortion is a natural phenomenon though. You could even call it "excitation", causing the sudden appearance of complex additional harmonics related to the fundamental frequency.

Here are a few examples of "distortion":

Plucking the string of a guitar with a plectrum. The harder you pick, the more "distortion".
Hammer hitting the piano string
Violin bow strokes on a violin string
Rock hitting the pond of water

You get the idea. Distortion is one of THE fundamental parts of audio and can't in my opinion be described as either bad or good. It just is. ..and it's all subjective, always. Even the "bad" stuff that I personally dislike (mainly aliasing when unintended and sometimes intermodulation distortion if it's too abrasive/disconnecting).

[MogwaiBoy]

Listening only to the 'delta' signal doesn't give you the full picture, obviously. Just use it to indicate where/when things are happening, not "this is what saturation sounds like".

[excuse me please]

I think this article explains it well:

https://www.soundonsound.com/sound-advi ... distortion

[jens]

Its what it sounds like in the mix that matters, not when it's isolated.

Perhaps, but that's not even what the OP meant.

What he meant with "isolation" was the end-result of a phase-cancellation test, which - as bmanic pointed out already in the first reply - does not really represent what the saturation actually is and sounds like - i.e. this thread is the product of a flawed thinking that in the meantime has been corrected/solved.

And thus, basically: move along folks, nothing to see here.

[gentleclockdivider]

Chebbyshev distortion to the rescue
Here's a chebby shev polynomial shaper I made in reaktor , perfect for bringing out certain harmonics

Download here , you need reaktor
https://www.native-instruments.com/en/r ... l/1781422/

[plexuss]

Chebbyshev distortion to the rescue
Here's a chebby shev polynomial shaper I made in reaktor , perfect for bringing out certain harmonics
Download here , you need reaktor
https://www.native-instruments.com/en/r ... l/1781422/

Thanks! This is awesome. I plan to check out your other Reaktor products! thanks for sharing. Also, thanks to the mods for cleaning up this thread!

Your chebby poly shaper is very cool and sounds great. It's also a good learning tool to hear how various harmonics affect the sound and simultaneously see the transfer function, which can be useful knowledge in learning about distortion.

You can't "objectively listen" to something that is completely subjective. Your whole post in fact demonstrates that subjectivity with a great example of cognitive bias. You explain that bias very clearly in your post.

I never said nor implied I was "objectively listening" so the whole premise of your post is bunk.

Distortion is a natural phenomenon though. You could even call it "excitation", causing the sudden appearance of complex additional harmonics related to the fundamental frequency.
Here are a few examples of "distortion":
Plucking the string of a guitar with a plectrum. The harder you pick, the more "distortion"., Hammer hitting the piano string, Violin bow strokes on a violin string, Rock hitting the pond of water
You get the idea. Distortion is one of THE fundamental parts of audio and can't in my opinion be described as either bad or good. It just is. ..and it's all subjective, always. Even the "bad" stuff that I personally dislike (mainly aliasing when unintended and sometimes intermodulation distortion if it's too abrasive/disconnecting).

This is completely wrong. That is not what distortion is, even a beginner audio engineer knows this. You might be good for tips and tricks but you fail when it comes to facts and science.

What you are describing are the additional harmonics or vibration modes caused by increasing the energy vibrating the sound source (eg string, reed, etc). This forms the source audio, the funamental and all harmonics generated by the sound source. Those harmonics are not distortion. Distortion is any change to the source audio after the fact. In our case caused by using software tools to add more harmonics to the source audio.

Look at it this way using the example of a vibrating string: a string is not capable of creating a pure sine wave. A string can get close to it in a highly controlled apparatus. Even still there will be harmonics. Therefore the vibration of the string is the SOURCE and comes with it's own set of harmonics. One can not say those harmonics are "distortion" because nothing has yet distorted the sound from the string. I know this is obvious to a lot of people but it appears there are some involved in audio engineering that don't understand these basic ideas and in turn post inaccurate and false information that doesn't benefit anyone.

Some examples:
"Audio distortion: With respect to audio, distortion refers to any kind of deformation of an output waveform compared to its input, usually clipping, harmonic distortion, or intermodulation distortion (mixing phenomena) caused by non-linear behavior of electronic components and power supply limitations." - wikipedia

"Distortion: Some form of corruption — wanted or unwanted — of the source signal. "- Sound on Sound glossary

[jens]

This is completely wrong. That is not what distortion is, even a beginner audio engineer knows this. You might be good for tips and tricks but you fail when it comes to facts and science.

This thread just got heavily moderated - and personally I certainly think it needed a good dose of that - and here you are at it again full throttle with your petty insults and your unwarranted vitriol...

Is that a Freudian beg for a ban?

[MogwaiBoy]

I agree, a plectrum hitting a guitar string isn't "distortion", but let's be civil and not get personal.

[Hink]

Waveform plot showing the different types of clipping. Valve overdrive is a form of soft limiting, while transistor clipping or extremely overdriven valves resemble hard clipping.

The word distortion refers to any modification of wave form of a signal, but in music it is used to refer to nonlinear distortion (excluding filters) and particularly to the introduction of new frequencies by memoryless nonlinearities.[33] In music the different forms of linear distortion have specific names describing them. The simplest of these is a distortion process known as "volume adjustment", which involves distorting the amplitude of a sound wave in a proportional (or ‘linear’) way in order to increase or decrease the volume of the sound without affecting the tone quality. In the context of music, the most common source of (nonlinear) distortion is clipping in amplifier circuits and is most commonly known as overdrive.[34]

Clipping is a non-linear process that produces frequencies not originally present in the audio signal. These frequencies can be harmonic overtones, meaning they are whole number multiples of one of the signal's original frequencies, or "inharmonic", resulting from general intermodulation distortion.[35][36][37] The same nonlinear device will produce both types of distortion, depending on the input signal. Intermodulation occurs whenever the input frequencies are not already harmonically related. For instance, playing a power chord through distortion results in intermodulation that produces new subharmonics.

"Soft clipping" gradually flattens the peaks of a signal which creates a number of higher harmonics which share a harmonic relationship with the original tone. "Hard clipping" flattens peaks abruptly, resulting in higher power in higher harmonics.[38] As clipping increases, a tone input progressively begins to resemble a square wave which has odd number harmonics. This is generally described as sounding "harsh".

Distortion and overdrive circuits each 'clip' the signal before it reaches the main amplifier (clean boost circuits do not necessarily create 'clipping') as well as boost signals to levels that cause distortion to occur at the main amplifier's front end stage (by exceeding the ordinary input signal amplitude, thus overdriving the amplifier) Note : product names may not accurately reflect type of circuit involved - see above.[39]

A fuzz box alters an audio signal until it is nearly a square wave and adds complex overtones by way of a frequency multiplier.

https://en.wikipedia.org/wiki/Distortion_(music)