Newbie question. What setting should I have for Gain so that I do not get distortion or clipping of WAV

7 posts / 0 new
Last post
paulcopeland's picture
Joined: March 1, 2012


I am working on my first composition with AM and want to get it right the first time.

I have a very simple setup of 8 FilePlayers.

I want to control the gain output and panning output via the Metasurface.

I am using the SLimiter_1 for the gain, as I want to turn it from 0 to maximum gain via Metasurface.

This is then being routed to the P2Mixer_1 for panning.

The only knobs I will be changing are INPUT GAIN (SLimiter_1) and Pan 1 and 2 of P2Mixer_1

Is the following setup okay? Are the setting for SLimiter_1 Threshold, Ceiling, Hold and Release correct for straight throughput of the WAV file without any effect being added except for Gain?

What is the maximum Input Gain number for SLimiter-1 that I can use that won't give me any distortion or clipping of the WAV file?

Should I use SLimiter for volume control, or is another contraption better?

Many thanks for your help and I look forward to hearing from the experts soon :)



Winslow17's picture
Joined: December 29, 2010

I am no expert, but one thing I can point out is that in the View drop-down menu up top, you can select "Show Contraption I/O Activity Monitors," and this will make the I/O ports of your contraptions, as seen in the Patcher window, light up green when active - and red when they are overloaded and potentially clipping. When these indicators are turned on, they do tax the CPU, and you may see some degradation in the Properties window (the one you've grabbed in your post.) But once you play around with AM, and these indicators, a bit, you will get a feel for it all and you should be able to have them off most of the time.

I find that in general, I set contraption volume controls at around 1 o'clock. There are exceptions, but that seems to be a workable, if entirely unscientifically determined, setting.

paulcopeland's picture
Joined: March 1, 2012

Hi Winslow17

Many thanks of ryour quick response.

I will look into your suggestion.

Best wishes and continue to create some great pieces with AM



I still wonder what the setting should be for Input Gain: on both contraptions for no clipping and no distortion. Should the SLimiter Input Gain andP2 Mixer gain be the same and at what number?

Spectro's picture
Joined: June 23, 2009

In general, if your source sound file(s) are already gain optimised or maximised (ie meters hit or approach maximum if the FilePlayer output is directly connected to the Limiter input at 0dB gain or directly to SoundOut for example), then any additional gain applied above 0db in any part of the signal chain (ie Limiter Input gain) will probably introduce clipping. (Note that the mixer contraptions can't actually apply "gain" as they usually only go up to 0dB) If however the wavs are lower in overall level, then the gain applied in the chain may be greater, but by how much is a matter for experimentation and dependent on the source material you are using...

Also, typically, gain (or more strictly, cut) levels would be controlled at the mixer which is the contraptions primary purpose (input channel gains would be used to set some kind of aesthetically acceptable balance between the different sound sources) and output channel gain would be set to provide a decent but not overwhleming (ie clipped) mix signal to the limiter. Either the Master or channel gain controls will be reduced by a small amount (around -3dB) for each additional (full scale or maximised) sound source.

Personally speaking, I would not use the Limiter Input gain as a variable gain control as you seem to be intending as this may alter the dynamic profile of your material depending on the limiter settings and amount of inout gain applied. The limiter should probably come last in the signal chain (before SoundOut) as a defence against clipping and typically in this context, its Input level could be adjusted to optimise sound levels if your sound sources aren't maximised or full scale...

Of course, all of the above isn't a set of strict rules or prescriptions, but suggestions based on the way I tend to operate in AM. If I were to take the liberty to make one more suggestion, (and here I am, from your questions, assuming that its is not *just* that you're new to Audiomulch) I would say that, you would probably benefit from checking out a few tutorials on sound mixing basics. Maybe a google search along the lines of "Basics of sound mixing" and "audio limiting" would probably yield some useful results.



Ross B.
Ross B.'s picture
Joined: April 11, 2009

> What is the maximum Input Gain number for SLimiter-1 that I can use that won't give me
> any distortion or clipping of the WAV file?

Short answer: whatever setting leads to the limiter GR meter (Gain Reduction) never changing from 0.

Slightly longer answer: it depends on how many input signals are being mixed into the limiter and what their mixer levels are. 0dB will give unity gain, which means whatever goes in comes out. But you'd also need the threshold set above 0dB, since any signal above the threshold will be limited.

I recommend you study the help file for the limiter and understand how the various gain controls are related. To avoid clipping you at least need to understand the relationship between input gain, threshold, and ceiling. You can deduce most of this by watching the meters on the limiter (they show the post-gain input signal level, the amount of gain reduction, and the final output level). As long as there is no GR then the limiter is not acting, and as long as your ouput level is below 0dB it won't be clipped at AM's output.

paulcopeland's picture
Joined: March 1, 2012

Hii Spectro and Ross.

Many thanks indeed for your very thorough answers to my questions.

I will now wander off and do some homewok :)

Thank you again and keep up the great work.


Ross B.
Ross B.'s picture
Joined: April 11, 2009

Hi again Paul,

One thing you need to consider is that the peak output level of a set of mixed tracks is highly dependent not only on how loud the tracks are, but also on how they blend together. An extreme example would be two tracks with gaps, where track only has sound where the other one has gaps -- in this case you don't need to account for the loudness of the combined mixed signal, because the two never really mix together to make a louder sound.

Therefore, gain settings that *guarantee* that there will never be any clipping/limiting/gain reduction no matter what the input tracks are, are likely to be much lower than settings that don't clip most of the time, or even never clip in practice.

So I guess the question is: do you need to know the theoretical worst-case settings, or something that works in practice?