Calibrating Input Level for Plugins

[James Freeman]

Note: Please don't quote this entire post, I may want to edit and add information or clarify things.
You can just reply without hitting the reply button, Thank You!

My method of calibrating your Audio Interface and Helix Native to match Helix hardware.
The result is more accurate gain from Helix Native equal to the Hardware, the last thing you want is to unknowingly set your vintage Telecaster to be as hot as a shredding axe with 18v EMGs, which a LOT of people mistakenly do.


1. Generate a 1kHz sine wave test tone in your DAW.

The free DD SigGen VST works great: https://plugins4free.com/plugin/2801/
Allows 0.01dB level adjustments while holding Shift, very accurate.

2. Send the sine tone to a free output on your Audio Interface and adjust level to read 500mV AC with a Volt Meter.

Send the sine to a free unused output on your audio interface, ie. NOT to your studio monitors unless you need a headache.
If your audio interface has 1/4" outputs use a regular guitar cable or RCA cable with a 1/4"-to-RCA adapter on the other end.
Set your Volt Meter to AC reading and adjust the output from your DAW so the 1kHz signal reads 500mV AC RMS on the other end of the cable.
The meter doesn't have to be true RMS because we are measuring pure undistorted sine wave.

3. Find the threshold on Helix Floor hardware (you can skip this step).

This supposed to be exactly the same for all HX/Helix hardware, since I already measured it you don't have to.

Insert the other end of the 1/4" cable into the Guitar Input of your Helix hardware and add a single Hard Gate to the signal chain.
Find the Open level that the 500mV signal opens the Hard Gate, on my Helix it is exactly -15.3dB.

Hard Gate settings: Close: 0.0dB, Hold: 10ms, Decay: above 200ms.

4. Loop back the 1/4" output into the Instrument Input of your Audio Interface.

If you did step 3, pull out the 1/4" from your Helix and stick it in the Instrument Input of your Audio Interface.

Your audio interface should have an Instrument button so turn it On, and set the Gain on ZERO.
Wait... you set your gain above zero? NOOB, don't talk to me ever again.

Joking aside, the only consistent setting (ie. NOT Random) of the Gain knob is zero, on my Focusrite Interface I almost hit clipping with my EMG81 guitar when I play hard with gain at zero, so I just keep it there.

The 500mV AC signal should still be running from your DAW and you should see the input LEDs light up (not shown in the picture).

5. Calibrate Helix Native Input Gain.

First, add a Hard Gate to your chain and set the Open Threshold to the exact same level we found in step 3 with the HX hardware.
I have found that my Helix Floor opens at -15.3 dB with a 500mV AC RMS signal, this should be true to all HX/Helix hardware if the Guitar Input electronics are exactly the same across all SKUs.

Now, find the exact Input level by moving the little white slider in Helix Native until the Reduction bar opens/closes, you can use the mouse wheel for finer control.

This input level is your calibration value to that specific input on your audio interface with the preamp gain set to zero on that input, this should match the HX hardware and all the models should behave exactly like on the hardware.
The patches you create now with Native should accurately translate to any HX/Helix hardware without the need to re-tweak the Drive or Level setting of Amps and Pedals and vice versa.

I hope you find this useful.

 

[James Freeman]

"Exclusive Content"





EDIT:
All-In-One dBu Calculator:

https://drive.google.com/file/d/1yg3FPZPMTpNwaVd1IKfuy2zKMygpk_Bl

 

[TheTrueZoltan!]

Calibrating the audio interface's instrument input is crucial for using amp modelling plug-ins. See this thread on KVR also, where it is suggested to calibrate the input to 1VAC=0dBFS. Very useful.

 

[James Freeman]

Here's what I think, how can a plugin manufacturer guarantee any kind of accuracy if every audio interface is different and every user randomly adjusts how hot the guitar?

 

[JiveTurkey]

For old times sake

 

[James Freeman]

For old times sake

LOL, that's not Helix specific though.
Take any other plugin like Amplitube or ToneX for example, you have absolutely no idea how hot the amp was captured, what output level the audio interface produced, how hot the input gain was, etc. it's totally random.

The actual voltage your pickup and gain accuracy of the model loses any meaning without calibration.

 

[JiveTurkey]

LOL, that's not Helix specific though.
Take any other plugin like Amplitube or ToneX for example, you have absolutely no idea how hot the amp was captured, what output level the audio interface produced, how hot the input gain was, etc. it's totally random.

The actual voltage your pickup and gain accuracy of the model loses any meaning without calibration.

Agree. But I have never seen this discussion come up with any other amp plug-ins. Like; ever. Native is the most sensitive of them all from all I have ever read.

 

[James Freeman]

Agree. But I have never seen this discussion come up with any other amp plug-ins. Like; ever. Native is the most sensitive of them all from all I have ever read.

I don't know why more people are not aware of this 'problem' and why nobody discusses this.
People usually adjust the audio interface input gain so all their guitars have roughly the same level and headroom, that means a vintage Strat will be just as hot as a 18v EMG shredding axe.

The second problem is manufacturers, they simply cannot claim accuracy form plugins without 1:1 gain matching to the modeled amps, the customers are using a plethora of different audio interfaces with completely different gain amounts.

I have suggested a small 1/4" dongle with a AAA battery (like a wireless transmitter) that will output a 500mV AC 1kHz test tone to calibrate and minimize this randomness of audio interfaces for people who care about accuracy from their VST plugins.

 

[TheTrueZoltan!]

Did the test again and my active DI box is actually 1dB too hot. The instrument input of my audio interface (PreSonus Quantum 2) has to be boosted by 12dB to have the optimal level for Helix Native.

 

[James Freeman]

Did the test again and my active DI box is actually 1dB too hot. The instrument input of my audio interface (PreSonus Quantum 2) has to be boosted by 12dB to have the optimal level for Helix Native.

A huge difference between the two devices.
Did you test with the suggested 500mV AC RMS and -15.3dB value in the Hard Gate block?

 

[Digital Igloo]

Here's what I think, how can a plugin manufacturer guarantee any kind of accuracy if every audio interface is different and every user randomly adjusts how hot the guitar?

We can't. All we can do is request people use interfaces with 1M Ohm Hi-Z inputs, preferably those with no gain control. Or, y'know, Line 6 Helix/HX boxes.

 

[TheTrueZoltan!]

A huge difference between the two devices.
Did you test with the suggested 500mV AC RMS and -15.3dB value in the Hard Gate block?

Yes, that's what I did. The active DI box (BSS AR-133) - phantom powered by the audio interface - is a lot hotter than the instrument input of the Quantum. If I remember correctly, the Babyface Pro I had before the Quantum had an even lower input level.

 

[James Freeman]

We can't. All we can do is request people use interfaces with 1M Ohm Hi-Z inputs, preferably those with no gain control.

Different audio interfaces will produce a different input DBFS with zero gain and exactly the same analog signal on the input, it's a total crapshoot.
Analog-to-dBFS calibration is almost mandatory if you want consistency and accuracy.

This is a must watch:

 

[TheTrueZoltan!]

We can't. All we can do is request people use interfaces with 1M Ohm Hi-Z inputs, preferably those with no gain control. Or, y'know, Line 6 Helix/HX boxes.

That's true. But... the information that is given in the manual ("The Optimal Input Range Indicator provides a guideline for where you should keep your input level, between -36 dB to -12 dB.") is misleading. When the level is calibrated to match the Helix hardware devices - using the aforementioned method - high and even medium output pickups can (and should) easily achieve much higher levels to get the most accurate results.

 

[Elric]

Agree. But I have never seen this discussion come up with any other amp plug-ins. Like; ever. Native is the most sensitive of them all from all I have ever read.

Agree. The vast majority of plugins you can 'earball' it if you know how it is supposed to respond for the model and input you are feeding it. Native will be completely off if not set just right. IME.

 

[Elric]

Did the test again and my active DI box is actually 1dB too hot. The instrument input of my audio interface (PreSonus Quantum 2) has to be boosted by 12dB to have the optimal level for Helix Native.

As someone who also uses a Quantum 2 but is way too lazy to bother with this just for one plugin, this is excellent info. Will at least be willing to play with the level on it.

 

[James Freeman]

The vast majority of plugins you can 'earball' it if you know how it is supposed to respond for the model and input you are feeding it.

How accurately can you earball though?

In my experience even a 3db difference will feel off with a real amp if you are familiar with it and your guitars.
The high/low inputs on a Marshall/5150 are 6db apart and it feels like a different amp especially when you compensate with the Drive knob that has a bright cap on it, you end up with a different tone and tightness to the amp.

Plugins with multiple amp models like Native or Amplitube are no different than real amps in that regard, input gain matters and has to be accurate to get authentic results.

 

[SwirlyMaple]

I don't know why more people are not aware of this 'problem' and why nobody discusses this.
People usually adjust the audio interface input gain so all their guitars have roughly the same level and headroom, that means a vintage Strat will be just as hot as a 18v EMG shredding axe.

The second problem is manufacturers, they simply cannot claim accuracy form plugins without 1:1 gain matching to the modeled amps, the customers are using a plethora of different audio interfaces with completely different gain amounts.

1000x this ^^^ If a plugin isn't doing something with its input levels so that a low-output single coil is providing a lot less signal than a hot humbucker, that plugin is inaccurate.

Native's sensitivity with this is a sign that it's actually treating differing input levels in an accurate way. This is exactly what is necessary for sensitive dynamics and volume knob interaction, for example.

It's not a knock against Native, though some seem to think so. It's a sign it's probably behaving more realistically than the rest. If a plugin is numb to this reality, that means it's not handling dynamics or signal levels very accurately.

PS Great method James, very clever! We just need to make sure the people's voltmeters are true RMS and they know how to use them in A/C mode, otherwise all bets are off!

 

[James Freeman]

It's not a knock against Native, though some seem to think so. It's a sign it's probably behaving more realistically than the rest. If a plugin is numb to this reality, that means it's not handling dynamics or signal levels very accurately.

Exactly.
How can it be a knock if I'm sharing a way to remove randomness and make Native match the hardware hence the real amps they modeled?

Other software only modelers cannot be calibrated without information from the manufacturer, the "1vAC Peak = 0dBFS" 'rule' from SPICE cannot be assumed.
It would be great if manufacturers provided Analog-to-dBFS calibration values to their guitar VST plugins, at least the more serious ones, so studios and semi-professionals can get the most out of the plugin.
Put this Analog-to-dBFS number in the user manual, only good things can come of it, and for those who don't care nothing has changed.

Don't tell me a professional recording studio doesn't have a FLUKE around?
A 20$ multimeter from Amazon will do just as well.

PS Great method James, very clever! We just need to make sure the people's voltmeters are true RMS and they know how to use them in A/C mode, otherwise all bets are off!

I thought everyone has one? That's a mandatory tool in any toolbox.

The DVM doesn't have to be true RMS, because we are measuring pure sine function, any meter with mV AC reading will do just fine.
The tolerance difference between the two non-true RMS meters pictured below is less than 0.1dB, I'll take that over blissful randomness any day.

 

[Los-D]

My method of calibrating your Audio Interface and Helix Native to match Helix hardware.

Great job as always ! First TGF bookmark!

EDIT: quoting subject only