What Richard meant when he said that computers cant distort audio?

[salvakkpooo]

He said it in a few recent interviews.

[pcock]

possibly that past the digital clip point, audio sounds absolutely terrible, whereas if you overdrive an neve channel strip, or a tape machine, the harmonics that come off it are pleasing to the ear.

[Guest skibby]

yeah, to simulate distortion is really hard for plugins to do. the best digital distortion simulations will bring an i7 to its knees. so most dist plugins need to cheat, therefore not getting that close.

[khov]

but he's into haswell/hecker, seems pretty computerish distorsion if you ask me

[psn]

I think the context was that analogue emulation works OK in software for certain things, but not distortion.

[IOS]

Floating point format and processing chain; technically speaking, all calculation is done within 0.0 and 1.0, so it's possible to apply several processes without fear of distorting the signal.

 

Bob Katz explains it in "Audio Mastering":

"In a floating point system, you can break all the rules: floating point can literally ignore the individual levels in the chain. It's possible to drop the signal level by 100dB, store the signal as a floating point file, then open the file, raise the gain 100dB and get back the original signal, with little or no deterioration. Or vice versa"

[ZoeB]

In a digital mixer, you can get as close to the maximum volume as you like without quite touching it, and it sounds literally as perfect as possible. As soon as you touch it, it clips in a harsh, brittle way. By comparison, with an analogue medium such as tape (or even vacuum tubes before you get as far as a digital recording, I've noticed), as you creep ever closer to the maximum volume, the sound gets distorted in a mathematically complicated way that colours it in a way that's often considered pleasing to the ear, adding in various harmonics and the like. There are vintage multitrack tape machine emulating plug-ins and nonlinear summers (analogue mixing desk emulating plug-ins) that try to reproduce this effect, to add interesting, subtle distortion that slightly colours the sound. I have no idea how authentic they sound, not having recorded to tape in a long time, and even then it was Compact Cassette, a low quality medium with a noise floor equivalent to a roughly 6-bit depth. The plug-ins don't break down like tape machines do. They do often contain DRM, or require replacing as you update your hardware, OS, DAW, etc. Everything has pros and cons. None of this is as important as actually making music, of course, and your average person on the street won't care all that much that you spent a lot of time, effort and money making something sound slightly better... but it is fun playing with all this gear, real or virtual. :)

[ZoeB]

Floating point format and processing chain; technically speaking, all calculation is done within 0.0 and 1.0, so it's possible to apply several processes without fear of distorting the signal.

 

Bob Katz explains it in "Audio Mastering":

"In a floating point system, you can break all the rules: floating point can literally ignore the individual levels in the chain. It's possible to drop the signal level by 100dB, store the signal as a floating point file, then open the file, raise the gain 100dB and get back the original signal, with little or no deterioration. Or vice versa"

 

This sounds like an odd thing to say. Bit depth is important, not whether you're using floating point arithmetic or integers. A bit depth of 4 is generally pretty lousy, regardless of whether you're counting from -7 to +8 in increments of 1, or 0 to 1 in increments of 0.0625. A bit depth of 16 is much nicer, regardless of whether you're counting from -32767 to +32768 in increments of 1, or 0 to 1 in increments of 0.00001525878. But hey, I'm a simple creature and like programming with integers. :)

[Guest yikes]

here is a lovely article on this

http://www.soundonsound.com/sos/feb10/articles/analoguewarmth.htm

[IOS]

 

Floating point format and processing chain; technically speaking, all calculation is done within 0.0 and 1.0, so it's possible to apply several processes without fear of distorting the signal.

 

Bob Katz explains it in "Audio Mastering":

"In a floating point system, you can break all the rules: floating point can literally ignore the individual levels in the chain. It's possible to drop the signal level by 100dB, store the signal as a floating point file, then open the file, raise the gain 100dB and get back the original signal, with little or no deterioration. Or vice versa"

 

This sounds like an odd thing to say. Bit depth is important, not whether you're using floating point arithmetic or integers. A bit depth of 4 is generally pretty lousy, regardless of whether you're counting from -7 to +8 in increments of 1, or 0 to 1 in increments of 0.0625. A bit depth of 16 is much nicer, regardless of whether you're counting from -32767 to +32768 in increments of 1, or 0 to 1 in increments of 0.00001525878. But hey, I'm a simple creature and like programming with integers. :)

 

 

Bit depth is important indeed, and that's why you won't see FP processing happening in anything less than 32bits. For instance, Reaper lets me choose between 32bit or 64bit FP as the bit depth for uncompressed audio files.

 

It may sound odd probably because the FP calculation can be performed 'internally' (within a DAW); but audio converters use fixed point arithmetic so whenever the FP processor "meets the real fixed-point world" (as Bob Katz puts it), the signal must be regulated to normal levels. So when RDJ says 'computers can't distort audio', he implies 'computers can't distort audio, when working within a DAW using floating point processing'.

[Brisbot]

This is going to seem kind of random and only somewhat on topic, but one way I like making use of distortion is having another side channel dedicated to the distortion, and bandpassing (with an eq or two) the distortion in places that sound nice, which I guess basically means I'm bandpassing it where the harmonics generated sound good! I've begun doing this recently and have been getting very nice results, especially in the mid-upper registers!

I've also noticed that distortion can sound VERY different depending on the plugin so. The best distortion I've heard is probably Harmor's distortion. I have no clue what they did but it sounds fantastic. I wish I could use Harmor as an effect, but sadly the distortion itself is built into the synth. Besides that I use Ohmicide.

ALRIGHT back on topic! I wish I knew more. Could someone point me somewhere where I can read in depth about distortion so I can understand what you're talking about?

[chim]

He meant that it doesnt harmonically distort in a musically pleasing way, one of many little occurances in analog chains. I beg to differ after hearing plugs from plugin alliance, nomad factory and softube, but I'm not a huge analog fetischist. I think you are a bit justified to be a hardass on this when it comes to amping guitars as a mic'ed tube amp is pretty tough to emulate, but it doesn't mean you can't get a good sound out of a digital/software chain - plenty of harmonic distortion options among high end plugs, although they tend to be a bit CPU intensive.

[IOS]

Bah, can someone post a link to the RDJ interview? Must check the context.

[Mag]

anyone has a link/wall of text that drunk aphex posted on mu-ziq forums long time ago? i can't find it atm

he gone pretty detail about differences of analog and digital, maybe that would help

also just want to read it one more time with the knoledge I have today

[IOS]

Found it: http://www.groove.de/2014/12/25/25-questions-for-aphex-twin/

 

This one? >> "The reason I prefer to work with analogue synths is – for me it’s like a mathematical thing when you come down to it. Basically a computer can’t do distortion, everything on the computer just sounds perfect, which is nice if you want to make perfect tracks, but if you don’t, then you’re kind of stuck."

 

So,

• a computer can't 'do' distortion (i.e., it doesn't sound good enough, there are better alternatives for audio distortion than a puter), which is what RDJ suggests
• a computer cannot distort audio (i.e., using floating point processing), which is what this topic title suggests

 

Oof, that interview.. "I like using analogue synths / for me it's a mathematical thing / that's why I don't use the computer, it cannot do distortion / but then, I hired this dude to write music software (because synths can't do mutation)" Go figure

[Guest skibby]

http://mercuriall.iks.ru/cms/?p=190

[Bechuga]

He also said chemtrails are a government conspiracy. He says a lot of things.

[ZoeB]

He also said chemtrails are a government conspiracy. He says a lot of things.

 

Indeed.

[Brisbot]

ZoeB is teaching Reason 101 this semester! 3 credit hours.

[awepittance]

i never really was obsessive about analog distortion until i played a set off of someone's custom made 4-channel tube mixer and tube amplifier running to 2 18" subwoofers and tops. I was literally in the red for about 75% for the set and it still sounded glorious, after that I was sold on the concept

[may be rude]

im surprised skibby didnt mention his excellent vst phase voltage module, which employs an innovative and cpu-efficient way of implementing a highly customizable distortion effect. it can range from a slight luster to utter speaker mayhem. i particularly like eqing after it to shape very interesting sounds.

[Mag]

Skibby and very honest: thx, definetly checking em out!

[MisterE]

This is going to seem kind of random and only somewhat on topic, but one way I like making use of distortion is having another side channel dedicated to the distortion, and bandpassing (with an eq or two) the distortion in places that sound nice, which I guess basically means I'm bandpassing it where the harmonics generated sound good! I've begun doing this recently and have been getting very nice results, especially in the mid-upper registers!

 

I've also noticed that distortion can sound VERY different depending on the plugin so. The best distortion I've heard is probably Harmor's distortion. I have no clue what they did but it sounds fantastic. I wish I could use Harmor as an effect, but sadly the distortion itself is built into the synth. Besides that I use Ohmicide.

 

ALRIGHT back on topic! I wish I knew more. Could someone point me somewhere where I can read in depth about distortion so I can understand what you're talking about?

 

 

 

you're doing the bandpassing before the distortion, right? because if you bandpass after the distortion, you're actually removing lots or most of the harmonics.

[hoggy]

maybe you can make some kinds of distortion by putting a very amplified signal into an effects box for example - like if a signal was too loud going into a flange pedal, maybe it would distort in an unexpected way? so by overdriving different boxes, if you see what I mean - and like Zoe was saying about tape distortion too - it distorts the signal in a different way

 

or for example overdriving/damaging a speaker and recording it

[Ragnar]

graphical waveshapers like mwaveshaper would prob sound hella different if the sampling rate was infinity/computers didn't think in those terms right? That's what I've been led to believe. It said an interesting thing that symmetrical waveshaping can only generate even harmonics, while asymmetrical can make odd and even.

 

Wasn't RDJ album done on a fuckin computer. Because in that case if fingerbib was digital then digital is just as gud