Logic Pro 9 For or Against: Lowering Stereo Output Fader to Avoid Clipping

[MikeG]

In my previous mixing experience I have been simply lowering my Stereo Output fader (Output 1+2) when it is peaking. I have heard from some people that this is not a good idea as it's better to relatively lower all my faders rather than lowering my Stereo Output fader as I am just lowering a saturated signal rather than lowering the source of the saturation. I have always done it this way and it seems to work well for me but I would like to hear what others have to say about this?

I don't see an easy way in Logic to lower all my tracks relatively since I usually have automation on all tracks so it is so much more convenient to just lower the Stereo Output.

So any advice would be great to hear on this.

Thanks,
Mike

 

[CSeye]

Create an aux channel in the mixer. Name it Mix Out or maybe Bob.😀
Route the outputs of all channel strips to a bus. Set this bus as the input to the newly created aux.
Route the out of the aux to the main stereo out.

Now you can adjust the mix volume for the entire project before it reaches the main output.

 

[georgelegeriii]

Or simply add a gain plug-in in the first insert on your main stereo out...

you can adjust your master out to be a more reasonable level then, and also use any 3d party enhancers (like compression) that don't do internal 32 bit floating point math.

This essentially will do the same thing as the other suggestion

 

[Pete Thomas]

I don't see an easy way in Logic to lower all my tracks relatively since I usually have automation on all tracks so it is so much more convenient to just lower the Stereo Output.

For this very reason, many people don't use volume automation, they use automation on a gain plugin on each channel that needs it.

Additionally (since the implementation of region gain) I use region gains if it means I can avoid automation, which I try to do whenever possible.

I also try to avoid overloading the output by starting off my very first track of a project with a large headroom, either by recording with a lot of headroom or just pulling down the fader of the first track and build my recording/mixing round that.

 

[georgelegeriii]

Hey Pete, hows about a quick tutorial on Region Gain and how it is used for those of us that missed that in the update notes ;-))

Love the LUG, a place where you can learn something new almost every day!

 

[Pete Thomas]

Hey Pete, hows about a quick tutorial on Region Gain and how it is used for those of us that missed that in the update notes ;-))

Just select a region and add or subtract the gain from the Inspector. (I;m not sure which version this appeared in, I don't even remember the update notes I just stumbled across it one day.

The downside is there is no visual indication of which regions have gain adjusted until you select a region and look in the Inspector.

 

[Multispace]

As long as you don't put inserts on the Stereo Out fader (as inserts are pre-fader, you could overload these plug-ins), there's no reason not to lower the Stereo Output fader. Logic has a lot of headroom (around 1000 dB), so it's not true that you just lower a saturated signal as you wrote. The signal is not saturated, because of the headroom, so there's no quality-loss in lowering the Stereo Out fader.

 

[Peter Ostry]

I have heard from some people that this is not a good idea as it's better to relatively lower all my faders rather than lowering my Stereo Output fader as I am just lowering a saturated signal rather than lowering the source of the saturation.

The signal in the output channel is not necessarily saturated (distorted) just because it clips the channelstrip. Due to 32 Bit floating-point processing you have more room above 0 dB than you will ever need. If all single signals are ok, just pull the output fader down or use a Gain plugin to reduce the level. Otherwise, if you use dynamic processing in your tracks and submixes (compressors, gates etc), lowering the volume of all channelstrips would mess up all your perfectly adjusted thresholds.

However, ideally you should work well below digital zero all the time. And don't forget that some plugins may not be happy with hot levels. If you constantly get overloads towards the end of your mixing chain, rethink your gain staging. Don't concentrate on zero but rather go down a couple of dB from the beginning. There is no difference in sound if your tracks are 12 dB lower, just increase your monitoring volume until you finally bring the volume up to the desired level.

 

[MikeG]

Thank you everyone for the absolutely amazing information. This is just what I needed to hear.

Best,
Mike

 

[Antiphones]

Be careful of your instrument and audio track channel levels as they can clip individually if they're too high.

I've read many times that there is loads of headroom on individual channel strips and you don't need to worry if they go into the red when mixing.

It turns out this is not true. At least not in the latest Logic version.

I ran tests and there is significant distortion on audio channels which go more than a few db into the red. Since I discovered this, I never let my channels hit the red unless its very occasional and momentary.

So just to be clear. I'm not talking about the audio files themselves clipping during recording or the audio channels clipping while recording - this would obviously be a problem. I'm talking about taking a track with a cleanly recorded signal or soft synth and turning up one these individual channels during a mix so that it hits the red on its own meter. Its simply not true that there is lots of headroom.

Don't believe me? Try it yourself with the latest version of Logic. Take a cleanly recorded keyboard track routed to the output faders. Use a sustaining sound like a clean sounding keyboard pad so you can clearly hear the difference when distortion begins. Then as it plays back, turn up its channel strip fader (using a plugin if necessary to boost the level) until its into the red on its individual channel (not the master channels which obviously need to be keep out of the red). Now listen as you raise the level into the red - it doesn't go far until very obvious distortion is heard. Keep the output faders in the green.

I can reproduce this reliably.

I say this simply because as I said, I've read lots of times that there is huge headroom on individual channel strips. It might be true that this is the case on some versions of Logic. But its not the case on the current version.

Good practice - keep your meters out of the red - across the board.

 

[Peter Ostry]

Take a cleanly recorded keyboard track routed to the output faders. ... Then as it plays back, turn up its channel strip fader ... until its into the red on its individual channel ... Now listen as you raise the level into the red - it doesn't go far until very obvious distortion is heard.

This would be a very bad bug, actually it would mean that 32 Bit floating point processing does no longer work. But I cannot confirm your findings. I tested with Logic 9.1.5 and 9.1.7 with the following setup:

Testoscillator with 1 kHz and -1 dB.
4 Gain plugins, each automated to go to +24 dB and back to 0 dB.
The oscillator is sent to an Aux Channelstrip.
The Aux has also 4 Gain plugins, automated for 0, -24, 0 dB.

This means, the output of the Testoscillator goes 96 dB up while the Aux compensates this for the output. Will we get distortion?

​

As expected, the output remains almost stable and there is neither audible distortion nor visible sinus distortion according to the oscilloscope of Spectre.

I can reproduce this reliably.

I can't. Not at all.

You may hear distortion in the sound itself, whis is built-in and gets audible when the volume raises. I expect this from many good keyboard sounds. This is one of the reasons why they are able to cut through a mix.

I've read lots of times that there is huge headroom on individual channel strips.

Just to clean up our wording - there is no "headroom" in digital world as it exists in analog processing. Digitally, zero is the end. But floating point processing shifts the working range. This means you will hardly reach zero, maybe somewhere at 1,200 or 1,500 dB, who knows ...

Good practice - keep your meters out of the red - across the board.

Absolutely good advice for several reasons.

 

[Antiphones]

Thanks Peter, for providing the screenshots and tests. Its clearly working as you would expect for you. I can't explain why its not the same here, but I can assure you it is not. If I can find the time I'll do a screenshots of my own with mp3s demonstrating the distortion that happens when you push channels into the red.

You may hear distortion in the sound itself, which is built-in and gets audible when the volume raises.

No I'm very careful about that sort of thing. I'm not saying I'm not missing something here, but I've been engineering professionally for 17 years and I'm well aware of volume differences and how they can affect the sound. The distortion is like a fuzz box when was way into the red, I'm not talking about anything subtle here. Its get worse the farther into the red you push it. The routing is direct into the L/R output faders not via a bus so it wasn't overdriving an effect on a bus or anything like that. I reproduced this same effect with more than one source, it isn't just one keyboard sound.

So, I'm sure you are correct about 32 bit floating point. But if I get distortion here when hitting the red, because of some bug in Logic, the same may happen for others. So I thought it was worth mentioning.

I am human and hence open to human error 🙂 But I am a really experienced engineer and I've been using Logic for 8 or 10 years, so I don't think I'll have missed anything obvious here - eg anything else that could obviously be causing the distortion like a plugin. I have demonstrated this to another engineer friend who argued vehemently against the possiblility until I demonstrated it to him. He was as surprised as I was!

 

[Pete Thomas]

Thanks Peter, for providing the screenshots and tests. Its clearly working as you would expect for you. I can't explain why its not the same here, but I can assure you it is not.

I believe you, but I've just done a quick test and everything is fine here.

I added 4 x 16dB gain inserts on the out buss, followed by one gain at -96dB. The mix sounded exactly the same as without the gains.

Warning to all. If trying this at home, do not attempt to see what happens if you bypass the last -96dB gain plugin.

 

[Peter Ostry]

If I can find the time I'll do a screenshots of my own with mp3s demonstrating the distortion that happens when you push channels into the red.

This would be fine.

Its get worse the farther into the red you push it. The routing is direct into the L/R output faders not via a bus so it wasn't overdriving an effect on a bus or anything like that.

It may sound like a silly question but just to get sure - you did not overload the output channelstrip, did you?

 

[AntibalasHiFi]

Logic's headroom is 6db per channel. If you go more than 6db into the red, you will hear distortion. Less than 6db and you're fine. That said, if you see red early on then you probably want to lower all your tracks before you start doing volume automation, otherwise you'll run into problems down the road.

 

[Antiphones]

It may sound like a silly question but just to get sure - you did not overload the output channelstrip, did you?

No the output channel strip is well below peaking.

 

[Peter Ostry]

Logic's headroom is 6db per channel.

Wrong. You can go up over 1000 dB without distortion. But your plugins don't like that and it is generally not a good idea. And of course you must not overload the output, in this channelstrip 0 dB or a little less is the limit.

 

[georgelegeriii]

I've also done the test of taking a test signal, running it way into the red, and then lowering the output fader to compensate for the overloaded level on output, and got NO distortion.
Try this: take an instrument channel, instantiate a test oscillator, and crank the output to +6, then add a gain plug-in, and add 24 DB more, for a total level of +30 DB.

Send it to a bus, add a gain plug-in and take off the 30 DB you added, so the level is below 0 db. Now take that bus and record it to a new track. Be careful you have no output set to this track, or you will make a feedback loop.

Record a second of the tone (I did 1k) and then zoom in and see if the waveform is distorted in any way (it won't be).

That is a 100% verifiable test.

Now, you could also try this using a 3d party plug-in after the oscillator and gain plug-in, and see what happens to the test signal. If it comes back distorted (the sine wave will not be smooth) the plug-in does not have a 32 bit floating internal resolution, or they are doing something else to the signal in their emulation of whatever the process they are trying to do.

That's my best, "simple guy" understanding of this whole issue. Now I hope a real technical person can jump in and add to this, so this internal headroom issue can be "put to bed"

 

[Peter Ostry]

That's my best, "simple guy" understanding of this whole issue. Now I hope a real technical person can jump in and add to this, so this internal headroom issue can be "put to bed"

Actually it is pretty simple for the part we should understand:
Wikipedia: "Floating point"

"... whereas a floating-point representation ... with seven decimal digits could ... represent 1.234567, 123456.7, 0.00001234567, 1234567000000000, and so on. The floating-point format needs slightly more storage ... so when stored in the same space, floating-point numbers achieve their greater range at the expense of precision."
​

See what they do? It's all just numbers. If we leave a certain (mathematical) working range they shift the decimal point and continue calculating as before, just with another exponent.

Regarding the "precision" argument people can discuss if floating point or fixed point is better for audio, but there is no discussion about the headroom of floating point calculation. Of course there is a ceiling somewhere but it is very far away. Since floating point arithmetics permanently shift the head up and down in such a big space, it makes no practical sense to talk about the room above the head.

---

Maybe the following is a better explanation for musicians:

You have a small keyboard with 32 keys. This means you can play 3 octaves, ok? If you try to play higher you will miss the keys and finally fall off the stool because there are no keys up there. But wait, this is a "floating point keyboard"! You simply press a button and shift the octave range up. Amazing, you play as before, but it sounds higher. If you don't need more than 32 keys simultaneously, with octave transposition you can cover much more octaves than there are keys.

Compared to floating point arithmetic, the 32 keys are the available decimal digits and the octave transposition button is the exponent. The higher you set the exponent, the higher goes the sound, but the keys remain as they are.

Now, who would say, that a g''' is not playable on a small electronic keyboard because there is no key for it? Just shift the octave. And who says that Logic can't handle +300 dB because there is some miraculous limit at zero or +6 dB? This is not the case, Logic shifts the arithmetical working range, so to say. Not even the output channel clips - what we hear is the the clipping of the D/A conversion.

---

Here is an interesting video that shows a practical usage of floating point calculation. We cannot do this in Logic because we don't have this export option. But Logic does it itself: When we freeze tracks, Logic writes a 32 Bit floating-point file to disk which is used for playback. If we unfreeze, Logic deletes this file and starts to calculate the track in realtime again.

---

As usual we must be aware that this is a technical discussion. We should not mix all in the red because we can. Some plugins do not like that. We don't get optical feedback about the levels. Finally we have to bring all down anyway. And more general: this is bad behavior 😉