Saturation on individual instruments

[The Real Deanc2000]

Hello all.

I'm starting to learn about using saturation in mixing, but I was wondering one thing. If you want to use a little bit of saturation on an individual instrument, is it a send effect or is it an insert effect?

If it was a send, I would be able to conserve CPU by just using one saturation plugin, but if it's an insert, then I'd have to instantiate multiple plugins for each instrument.

Thanks.

[riemac]

Hello all.

I'm starting to learn about using saturation in mixing, but I was wondering one thing. If you want to use a little bit of saturation on an individual instrument, is it a send effect or is it an insert effect?

If it was a send, I would be able to conserve CPU by just using one saturation plugin, but if it's an insert, then I'd have to instantiate multiple plugins for each instrument.

Thanks.

You have to watch latency, when you use saturation as send effect. Most of the saturation RE's have latency.

[selig]

Hello all.

I'm starting to learn about using saturation in mixing, but I was wondering one thing. If you want to use a little bit of saturation on an individual instrument, is it a send effect or is it an insert effect?

If it was a send, I would be able to conserve CPU by just using one saturation plugin, but if it's an insert, then I'd have to instantiate multiple plugins for each instrument.

Thanks.

As long as the effect doesn't induce latency, it should work OK as a send. The problem I have with this approach is that you need two controls for saturation FX, a "drive" and a "level". The drive knob controls the color (and this would be equivalent to the "send" knob when in your example) and the FX Return then becomes the global "level". But what to do if you need more drive on one channel, but that ends up making the saturation effect too loud for that one channel? If you turn down the return, you turn down ALL channels feeding the saturation FX, when then changes the levels you've previously set on those channels. There would be no single setting that would account for all situations unless you just happen to get lucky and are OK with the global level.

The other advantage (over saving CPU) to using a send is getting the saturation to be more subtle by running it in parallel, but there are ways to accomplish that with inserts such as to use an effect that allows this, such as Pulveriser. In that situation, the "drive" is built into the device (Dirt), and the "level" is controlled by the Blend (dry/wet) control.

However, and this is a big one, there is one further consideration. Just as with a compressor on a bus vs individual compressors on each channel, using individual saturation gives a different effect than sending multiple signals to a single saturation or compression device. This is because both processes are non-linear with regards to level, and there is some interaction between the signals in the single device that will not exist in the individual devices. With each, the loudest instrument coming into the single device will set the tone for all. With compression, the loudest instrument will pull the level down on all the rest (a very desirable effect in some cases!), and with saturation the loudest instrument will saturate all instruments more than they would be saturated if feed individually (and there may be intermodulation distortion/saturation as well).

So the answer is that you will get different results with each approach, and the results with a single saturation device on a send will most likely not be the same as with individual saturation devices on each channel. Think of it as the difference between recording to analog multi-track and saturating each track vs sending the entire mix to a two-track recorder and saturating the entire mix. IMO, it's more subtle with the individual approach, and more "obvious" with the global approach - choose wisely, as sometimes you'll prefer the former while others you'll prefer the latter.

I suggest doing an experiment/comparison to train your ears to hear the difference, both with saturation and with compression, so that you'll learn which approach gives you the effect you're looking for in each situation.

Often times, especially with compression, I find I need a little on individual tracks (not all tracks), a little on a sub-mix (bus) especially with drums, and a little on the entire mix - each does something unique, and each can be more subtle (and thus less 'destructive') since you're doing a little bit at multiple stages rather than a lot at one stage. Make sense?

[ejanuska]

However, and this is a big one, there is one further consideration. Just as with a compressor on a bus vs individual compressors on each channel, using individual saturation gives a different effect than sending multiple signals to a single saturation or compression device. This is because both processes are non-linear with regards to level, and there is some interaction between the signals in the single device that will not exist in the individual devices. With each, the loudest instrument coming into the single device will set the tone for all. With compression, the loudest instrument will pull the level down on all the rest (a very desirable effect in some cases!), and with saturation the loudest instrument will saturate all instruments more than they would be saturated if feed individually (and there may be intermodulation distortion/saturation as well).

This is one of the most informative things I've read on this site. I never considered the effect of the loudest instrument pulling the others down.

You have to watch latency

As far as latency is concerned, I never noticed latency using REs for channels already recorded or using RE instruments MIDI note material.
How much latency are we talking here? It only a concern for poor performing computers?

[selig]

However, and this is a big one, there is one further consideration. Just as with a compressor on a bus vs individual compressors on each channel, using individual saturation gives a different effect than sending multiple signals to a single saturation or compression device. This is because both processes are non-linear with regards to level, and there is some interaction between the signals in the single device that will not exist in the individual devices. With each, the loudest instrument coming into the single device will set the tone for all. With compression, the loudest instrument will pull the level down on all the rest (a very desirable effect in some cases!), and with saturation the loudest instrument will saturate all instruments more than they would be saturated if feed individually (and there may be intermodulation distortion/saturation as well).

This is one of the most informative things I've read on this site. I never considered the effect of the loudest instrument pulling the others down.

You can use this to your advantage - by compressing a vocal bus that contains the lead and BGVs, if you set it up right you can have only the lead vocal above the threshold. The result will be that the lead vocal will "push down" the BGVs when it enters, automatically making room for the lead. Also works for rhythm vs lead guitars, etc. It's an old trick I heard about way back in the 70s, before automation was common and when this type of "trick" was about as complex as things got for the most part!

You have to watch latency

As far as latency is concerned, I never noticed latency using REs for channels already recorded or using RE instruments MIDI note material.
How much latency are we talking here? It only a concern for poor performing computers?

It only takes one sample of delay on a parallel track to cause what is commonly called phasing issues (maybe better to simply say a "comb filtering" effect since both are caused by a very short delay). The result can be a hollow sound that can 'smear' the signal and create undesirable issues.

This is not the same thing as overall system latency, which affects all signals equally. The solution is to use devices that do not impart any latency, or use a device such as Normen's handy VMG-01 to compensate for this delay in such cases where it causes audible problems.



Sent from my iPad using Tapatalk

[Marco Raaphorst]

I love running saturation in parallel. Also try to phase invert the saturated signal which offers many super cool sounds!

[Bloma]

However, and this is a big one, there is one further consideration. Just as with a compressor on a bus vs individual compressors on each channel, using individual saturation gives a different effect than sending multiple signals to a single saturation or compression device. This is because both processes are non-linear with regards to level, and there is some interaction between the signals in the single device that will not exist in the individual devices. With each, the loudest instrument coming into the single device will set the tone for all. With compression, the loudest instrument will pull the level down on all the rest (a very desirable effect in some cases!), and with saturation the loudest instrument will saturate all instruments more than they would be saturated if feed individually (and there may be intermodulation distortion/saturation as well).

This is one of the most informative things I've read on this site. I never considered the effect of the loudest instrument pulling the others down.

You can use this to your advantage - by compressing a vocal bus that contains the lead and BGVs, if you set it up right you can have only the lead vocal above the threshold. The result will be that the lead vocal will "push down" the BGVs when it enters, automatically making room for the lead. Also works for rhythm vs lead guitars, etc. It's an old trick I heard about way back in the 70s, before automation was common and when this type of "trick" was about as complex as things got for the most part!

You have to watch latency

As far as latency is concerned, I never noticed latency using REs for channels already recorded or using RE instruments MIDI note material.
How much latency are we talking here? It only a concern for poor performing computers?

It only takes one sample of delay on a parallel track to cause what is commonly called phasing issues (maybe better to simply say a "comb filtering" effect since both are caused by a very short delay). The result can be a hollow sound that can 'smear' the signal and create undesirable issues.

This is not the same thing as overall system latency, which affects all signals equally. The solution is to use devices that do not impart any latency, or use a device such as Normen's handy VMG-01 to compensate for this delay in such cases where it causes audible problems.



Sent from my iPad using Tapatalk

This is cool. I've been using parallel compression and watching Andrew Scheps videos, and when I do it with sends I get the out of phase sound, so I've been using parallel channels to parallel compress, but I have to use the same REs with the effects off to get the same amount of delay. Selig, if you don't mind answering, with that plugin, would I be able to put a compressor on a send and use that VMG-01 so I can send as many tracks as I want to it? I don't fully grasp it yet, as in getting the right amount of delay compensation so every track could be sent theoretically to the same compressor and not be out of phase.

[ejanuska]

The solution is to use devices that do not impart any latency, or use a device such as Normen's handy VMG-01 to compensate for this delay in such cases where it causes audible problems.

I never knew about that VMG-01 device. It's a good idea.

[ejanuska]

This is cool. I've been using parallel compression and watching Andrew Scheps videos, and when I do it with sends I get the out of phase sound, so I've been using parallel channels to parallel compress, but I have to use the same REs with the effects off to get the same amount of delay. Selig, if you don't mind answering, with that plugin, would I be able to put a compressor on a send and use that VMG-01 so I can send as many tracks as I want to it? I don't fully grasp it yet, as in getting the right amount of delay compensation so every track could be sent theoretically to the same compressor and not be out of phase.

Does this sound right?
If using parallel (New York) compression, the compressed track will have some latency. Figure out how much latency the compressor adds with the VMG-01, apply that amount of latency to the unprocessed track with the VMG-01.

[Bloma]

Yeah that does sound right, and it's what i did, but I was still getting phasing for some reason.

[Stock Music Musician]

You can always induce delay on the original track by patching in the same effects used on the parallel track, but just turning them all the way down. E.g. you would use a compressor on both, but the original track would have a threshold set so that it doesn't trigger, or the Softube saturation knob would simple be at zero.

[Bloma]

True, yeah that's what I've been doing and it works, but I'm running a 2011 Macbook Pro, and CPU usage is pretty tight.