Calibrating Input Level for Plugins

[discondition]

The axe fx models are calibrated for those inputs specifically, and the level on those inputs is also auto-conpensated when you adjust the input gain in the I/O settings, so no, this issue doesn't apply to this hardware (on some other hardware modelers it does though).
Obviously if reducing the input gain sounds better to you, you're free to do that anyway.

Ah, I think you're talking about the Input Sensitivity (A/D) controls.
I was talking about Input 1 Gain, which isn't auto-compensated and does change the amount of signal going into the front of the signal chain.

I have many guitars that I HAVE to lower the input with or the Axe screams at me it is clipping the input.

Yes, the Axe compensates but there is good reason to adjust it.

The clipping would be Input Sensitivity (A/D) not set low enough for that guitar, which has nothing to do with Input 1 Gain.

At the end of the day, if it sounds good, it is good

 

[DLC86]

h, I think you're talking about the Input Sensitivity (A/D) controls.
I was talking about Input 1 Gain, which isn't auto-compensated and does change the amount of signal going into the front of the signal chain.

Yep, sorry, I meant input sensitivity, but anyway no input gain adjustment should be necessary to make the amp behave accurately on the axe fx.

 

[Aleksi]

This just in (NDSP Discord):

 

[Aleksi]

It's not perfect, but much better than "gotta gain stage, bro. It's industry standard." This will get you at least in the ballpark.

This also vindicates my petty crusade for leaving the QC input gain at 0. The goal post will be moved to "it's an artistic choice, bro," but whatever.

 

[Jkist7]

It's not perfect, but much better than "gotta gain stage, bro. It's industry standard." This will get you at least in the ballpark

I do appreciate how transparent NDSP have been. They are known for a lot of negative things, but at least the way they've handled this particular subject has left me impressed.

 

[Orvillain]

This just in (NDSP Discord):
View attachment 18024

I don't think he gets it.

The entire point is... different audio interfaces have different dBu levels corresponding to their 0dB Hi-Z input.

YOUR INSTRUMENT INPUT IS NOT THE FRICKIN' SAME AS MY INSTRUMENT INPUT! NOR HIS, NOR HERS, NOR ANYONE'S!

Unless we have the same audio interface of course

What we need to know is what the models are designed for in terms of input voltage or input in dBu.

 

[Aleksi]

I don't think he gets it.

The entire point is... different audio interfaces have different dBu levels corresponding to their 0dB Hi-Z input.

YOUR INSTRUMENT INPUT IS NOT THE FRICKIN' SAME AS MY INSTRUMENT INPUT! NOR HIS, NOR HERS, NOR ANYONE'S!

Unless we have the same audio interface of course

What we need to know is what the models are designed for in terms of input voltage or input in dBu.

I think he actually gets it. You have to consider who he's communicating to. There's people on there asking how to set their input gain to zero. This advice will probably get most users closer than telling them to crank the gain just below clipping.

Neural has also told us what the values actually are.

Wouldn't a decent solution be to provide a list of input gain values for several of the most popular audio interfaces? That way you could cater to a large group of users.

 

[ivanmancic]

The more I see people (including plugin developers) misunderstand this issue, the more convinced I am that this requires an industry-wide solution, where every audio interface with an instrument (guitar) input would have to come calibrated out of the box. Then plugin developers would have a hard time f-ing it up.

 

[dronerstone]

The more I see people (including plugin developers) misunderstand this issue, the more convinced I am that this requires an industry-wide solution, where every audio interface with an instrument (guitar) input would have to come calibrated out of the box. Then plugin developers would have a hard time f-ing it up.

I believe interface input gain "calibration" isn't the main issue here.

 

[Orvillain]

Wouldn't a decent solution be to provide a list of input gain values for several of the most popular audio interfaces? That way you could cater to a large group of users.

@MirrorProfiles has already been doing this.

 

[Orvillain]

I believe interface input gain "calibration" isn't the main issue here.

Agreed.

The issue here is modelling standards and practices, and the differences between their measuring equipment. This issue is nothing to do with audio interfaces per-se.

 

[Aleksi]

@MirrorProfiles has already been doing this.

Yes, but I'm thinking of an "official," straight from the plugin company list, that actually makes it into the hands of customers when they buy a plugin.

 

[Orvillain]

Yes, but I'm thinking of an "official," straight from the plugin company list, that actually makes it into the hands of customers when they buy a plugin.

Right. So he's been approaching the plugin companies to get the details, and quite a lot of them will not give him a straight answer. So ... good luck with that!!

 

[Aleksi]

Right. So he's been approaching the plugin companies to get the details, and quite a lot of them will not give him a straight answer. So ... good luck with that!!

Hey, I never said it would happen, but it would be a decent solution

 

[Orvillain]

And in case people aren't getting it...

The reason this matters is because at the moment people are "using their ears" or "setting the interface preamp gain to just below clipping" and they are not getting a reflection of how those amp models would sound in the real world, because they are hitting it with too much signal.

 

[dronerstone]

And in case people aren't getting it...

The reason this matters is because at the moment people are "using their ears" or "setting the interface preamp gain to just below clipping" and they are not getting a reflection of how those amp models would sound in the real world, because they are hitting it with too much signal.

Which, in turn, might recoil and result in poor reviews - hence it's in their own best interest to reveal the numbers.

Unless... all the amps were modeled with different levels, and the manufacturer wants to avoid the awkwardness (but might get flak later on anyway).

 

[ivanmancic]

Unless... all the amps were modeled with different levels, and the manufacturer wants to avoid the awkwardness (but might get flak later on anyway).

This is exactly why I mentioned the need for a DI input level standard.
Those companies that have been using UA interfaces to model gear are usually much closer to sounding "right" for many users (especially since Focusrite and UA are quite popular choices and similar in input level).
Were they to use i.e. an Antelope Zen Go Synergy Core that I have when modeling, even with the best algorithms they'd end up with overly hot mess that sounds bad on any other interface.
It's not interfaces fault, I agree, but we are in desperate need for standards which would make things easier for users. Both in terms of software and hardware.
I do consider forum members here to be advanced users, but what about an average Joe that knows nothing about dBfs and dBu stuff?
YMMW is not a good design philosophy.

 

[MirrorProfiles]

I think he actually gets it. You have to consider who he's communicating to. There's people on there asking how to set their input gain to zero. This advice will probably get most users closer than telling them to crank the gain just below clipping.

Neural has also told us what the values actually are.

Wouldn't a decent solution be to provide a list of input gain values for several of the most popular audio interfaces? That way you could cater to a large group of users.

I actually don’t think he gets it because a QC is 15dBu and the plugins are 12.2dBu. So leaving things at 0 in that case would be undergained.

Hard to do an all encompassing list because that would involve way too many interfaces and there’d still be a chance people are using them wrong. Likewise getting a straight answer from a company isn’t easy because sometimes you have to go around a lot of people to find an answer from a developer rather than customer service.

The idea with what I did was to provide the means for people to work it out for themselves. If you look through the comments on the video/facebook etc, the vast majority of people would rather just be told and don’t want to look it up for themselves.

 

[Aleksi]

I actually don’t think he gets it because a QC is 15dBu and the plugins are 12.2dBu. So leaving things at 0 in that case would be undergained.

Hard to do an all encompassing list because that would involve way too many interfaces and there’d still be a chance people are using them wrong. Likewise getting a straight answer from a company isn’t easy because sometimes you have to go around a lot of people to find an answer from a developer rather than customer service.

The idea with what I did was to provide the means for people to work it out for themselves. If you look through the comments on the video/facebook etc, the vast majority of people would rather just be told and don’t want to look it up for themselves.

I assumed he meant to leave QC at zero when using the QC standalone.

 

[MirrorProfiles]

I assumed he meant to leave QC at zero when using the QC standalone.

• 0db (no extra gain on your interface or gain knob at minimum) on Instrument inputs and Hi Impedance inputs. Plugin Input gain at 0

that only makes sense for SOME interfaces ,most of the time it’ll need some adjustment on the plugin. The QC is one that’ll need a boost, it doesn’t use the same internal level as the plugins.

But yeah, for a HW modeller they’re all calibrated for their own input because they know their own specs