Calibrating Input Level for Plugins

[Sascha Franck]

some profiles have too much gain i have to decrease the gain on interface

Don't do that. Use a gain plugin or the input trim of the plugin or the gain pot of the amp.
And yes, required input levels for Tonex captures are all over the place. Still wouldn't use my input gain to adress the issue.

 

[Orvillain]

Umm it does improve things

Turning up your preamp gain does not reduce or improve the signal to noise ratio.

"Record as hot as possible without clipping" is massively out of date advice, belonging to the era of:

• 16-bit converters
• Noisy early ADCs (no longer a problem)
• Analogue tape workflows

In modern systems:

• 24-bit ADCs
• around 120–130 dB dynamic range
• DAW noise floors effectively irrelevant

Preamp gain does not improve the signal-to-noise ratio of the source; it only prevents downstream noise from becoming dominant. But that simply isn't really a concern when plugging into the Hi-Z input on your audio interface, and running your signal into a guitar amp plugin. The effect of ADC noise is so pitifully minimal, that it simply isn't a concern.

At least, this is my understanding of it.

 

[Melodioso]

If you have to add that much gain, I’d add a sizeable amount of it at the preamp. RME are stepped gain anyway, so it’s easy to know what your headroom is and recall any value.

BUT ALSO, more importantly. Ghost Note has sort of copy and pasted data a bit haphazardly and his info isn’t thoroughly checked. If you check the manual for the UFX IIIView attachment 58208

I can’t remember from memory, but I feel like RME are one of those companies that post their specs a bit weird. I’d double check those values are inclusive of the 8dB preamp gain - you’d either just need to leave it at minimum, or you’d need to add another 8dB to reach 13dBu headroom.

Most RME interfaces use 13dBu=0dBFS as their instrument input headroom. I know there are some outliers so it’s worth checking.

The specs for the RME UFX III if written out like most other manufacturers do would read:

Maximum input level, Gain 0 dB: +29 dBu (but minimum selectable Gain is 8 dB).

So if you use this particular interface with NDSP plugins, your input gain in TotalMix would be at +17.0dB (with Instr. selected).

 

[nitrateaudio]

And in case you don't want to lose headroom, you keep the input trim at zero and add those 16.8dB in the digital domain.

FWIW.. ^^ this is what I started doing recently. I think it's making a difference. It takes a little getting your head around it as I am used to doing it the way the sim vendors suggest.

 

[AlbertA]

Turning up your preamp gain does not reduce or improve the signal to noise ratio.

Of the input, of course not, but mostly we refer to the signal relative to ADC noise fixed noise floor.

"Record as hot as possible without clipping" is massively out of date advice, belonging to the era of:

• 16-bit converters
• Noisy early ADCs (no longer a problem)
• Analogue tape workflows

In modern systems:

• 24-bit ADCs
• around 120–130 dB dynamic range
• DAW noise floors effectively irrelevant

For the most part yearh - but when it does matter is when applying tons of gain in the digital domain, as its the case with amp modeling, which is mostly the context we are talking about. I demonstrate that in an audio clip on my post I linked to.

Preamp gain does not improve the signal-to-noise ratio of the source;

Correct - you need to address the source to improve the source SNR, like reduce EMI in the environment, play harder, raise the pickups, etc, at least until the thermal noise limit.

it only prevents downstream noise from becoming dominant. But that simply isn't really a concern when plugging into the Hi-Z input on your audio interface, and running your signal into a guitar amp plugin. The effect of ADC noise is so pitifully minimal, that it simply isn't a concern.

At least, this is my understanding of it.

As said, above, yeah mostly not a concern until you apply enough enough digital gain such as with amp modeling.

 

[Orvillain]

Of the input, of course not, but mostly we refer to the signal relative to ADC noise fixed noise floor.


For the most part yearh - but when it does matter is when applying tons of gain in the digital domain, as its the case with amp modeling, which is mostly the context we are talking about. I demonstrate that in an audio clip on my post I linked to.


Correct - you need to address the source to improve the source SNR, like reduce EMI in the environment, play harder, raise the pickups, etc, at least until the thermal noise limit.


As said, above, yeah mostly not a concern until you apply enough enough digital gain such as with amp modeling.

Let's just be clear....

When people say "record as hot as possible without clipping" it is because they believe they are reducing the amount of noise in their ultimate signal. Which is just not true in the grand scheme of things. I think you're getting caught up on minutia because you know more than the average berkshire hunt who talks about this stuff.

Which is why I took the general statement on it.

 

[AlbertA]

Let's just be clear....

When people say "record as hot as possible without clipping" it is because they believe they are reducing the amount of noise in their ultimate signal. Which is just not true in the grand scheme of things. I think you're getting caught up on minutia because you know more than the average berkshire hunt who talks about this stuff.

Which is why I took the general statement on it.

Gotcha

 

[yeky83]

When people say "record as hot as possible without clipping" it is because they believe they are reducing the amount of noise in their ultimate signal. Which is just not true in the grand scheme of things.

Eh, it does though.

 

[MirrorProfiles]

Eh, it does though.

 

[volkan]

in practice i have never heard the noise reduces when you increasing the gain on the interface it is never happened to me i have tried dozens of interface and plugins during 20 years
Sorry guys but
This idea sounds really stupid to me(increasing gain on interface reduces SNR)
it doesnt make any sense

 

[MirrorProfiles]

in practice i have never heard the noise reduces when you increasing the gain on the interface it is never happened to me i have tried dozens of interface and plugins during 20 years
Sorry guys but
This idea sounds really stupid to me(increasing gain on interface reduces SNR)
it doesnt make any sense

It’s really easy to do the same test yourself if you need to

 

[volkan]

It’s really easy to do the same test yourself if you need to

i have already done the test for years i say never heard reducing noise

 

[yeky83]

Post in thread 'Calibrating Input Level for Plugins'

Thursday at 8:37 AM

This is simply not true.

Your interface has a fixed noise floor. That's literally what the signal-to-noise-ratio is. When you turn up the preamp, you also turn up the noise. You're not improving anything by recording as hot as you can. Even plugin devs do not suggest that.

Umm it does improve things

• AlbertA

• 

 

[MirrorProfiles]

Post in thread 'Calibrating Input Level for Plugins'

Thursday at 8:37 AM

This is simply not true.

Your interface has a fixed noise floor. That's literally what the signal-to-noise-ratio is. When you turn up the preamp, you also turn up the noise. You're not improving anything by recording as hot as you can. Even plugin devs do not suggest that.

Umm it does improve things

• AlbertA

• 

Yes, that’s all shown in my example. There is a point where the improvements become extremely redundant.

 

[Whizzinby]

I just wanted to chip in to say that I used to do the "set gain just below clipping" and often struggled to get clean amp sims to sound clean.
When I changed to the zero gain on interface method I noticed that as well as getting clean tones more easily I also found my tone to be clearer for whatever reason. There was less apparent "mud" to the sound at least to my ears.

This was me as well. I was like WTF does this TK Imperial have so much gain, isn’t it a clean amp?

Once I dumped the interface levels, all my plugs sounded wayyyyy better.

 

[Orvillain]

Eh, it does though.

Only in esoteric situations. For most people's use cases, it makes very little difference. Try it. Record a signal at 0dB, +5dB, +10dB, +20dB, etc. Then normalize in the DAW, and flip phase. You'll get them all to null.

Julian Krause has a video on this where he does exactly that. linear gain alone does not change SNR.

 

[AlbertA]

Only in esoteric situations.

It's not esoteric, hear my example - just a low output pickup single coil guitar into a lead type of amp in the digital domain which will have plenty of gain in the digital domain. Is that really esoteric?

That's because after a bunch of digital gain, you are elevating the ADC noise floor and start competing with the guitar noise floor and worse the ADC noise floor is Wideband noise which will start to be noticeable in that context.

Julian Krause has a video on this where he does exactly that. linear gain alone does not change SNR.

Julian literally says the same thing I'm saying right here: https ://youtu.be/beXVfl1TSD0?si=gJAvPioGgg2Pz6Hn&t=428

 

[Orvillain]

It's not esoteric, hear my example - just a low output pickup single coil guitar into a lead type of amp in the digital domain which will have plenty of gain in the digital domain. Is that really esoteric?

That's because after a bunch of digital gain, you are elevating the ADC noise floor and start competing with the guitar noise floor and worse the ADC noise floor is Wideband noise which will start to be noticeable in that context.




Julian literally says the same thing I'm saying right here: https ://youtu.be/beXVfl1TSD0?si=gJAvPioGgg2Pz6Hn&t=428

Yes, if you record a signal very quietly and then apply a lot of gain later, you raise everything: signal, source noise, and ADC noise. That part is not in dispute.

The real question is whether ADC noise is ever the limiting factor in a modern guitar DI into interface into amp sim workflow. In practice, it almost never is.

Modern interfaces have ADC noise floors around -110 to -120 dBFS. A real electric guitar DI, especially a low output single coil, typically has a noise floor much higher than that, often around -70 to -80 dBFS once pickup noise, hum, cables, and EMI are included. That means the guitar is already 30 to 40 dB noisier than the converter.

When you add digital gain, the relative relationship does not change. The guitar signal and its own noise remain dominant, and the ADC noise stays buried underneath. Digital gain does not reshuffle which noise source is dominant.

The old record as hot as possible advice made sense with 16 bit converters, early ADCs, and tape workflows. In modern 24 bit systems with roughly 120 dB of real dynamic range, it no longer applies in the same way.

This is not an argument for recording extremely quietly. The sensible modern advice is simply to record at a reasonable level with plenty of headroom and avoid clipping. For guitar DI, peaks around -18 to -12 dBFS are perfectly fine. Recording hotter than that does not improve sound quality or signal to noise ratio, it just reduces headroom.
If a system is linear and time-invariant, gain multiplies signal and noise equally. Therefore, gain cannot improve the signal-to-noise ratio of the source.

 

[volkan]

how to send a sin wave and calculate the amount of gain we added on the interface (for the electric guitar)
somebody tell please
how do we use "tone generator plugin" to create sin wave on logic pro for electric guitar

 

[LiCoRn]

hey @volkan,
this Gost Note Audio guy from YT suggests a precedure to exactly figure out what you want using two methods:
1- a loudness meter plugin (MLoundnessAnalyser in his video, link to his video section)
2- buy their product/box that happen to generate a calibrated signal generator to implement @MirrorProfiles method

PS: Gost Note Audio is on the opposite side of the debate, and advocates against setting interface gain to zero/minimum (the famous you are all wrong video)