Calibrating Input Level for Plugins

[valimaki]

I cannot believe this is not all over the place. It was one of my first thoughts when I got my firsts audio interface and I started using plugins 15 years ago or so, and it has been driving me crazy for so many years. You can read countless articles and watch tons of videos and no body talks about it. So glad I finally found actual technical info about it.

Anyway, I have a Clarret 4Pre USB and I am using Amplitube 5 max + Tonex max. I don't know what would be the optimum level to hit IK Multimedia products.

After doing the calibration I ended with the preamp gain level I have been using for some time in the instrument input of my interface for single coil guitars, it is a relief to know that I could dialed on my own following my instincts. But I have a problem, if use a humbucker at that level my interface starts clipping on peaks. The input indicator goes from average green with single coil, to average orange with humbucker, and peaks end in read with humbuckers. I have measure how many dbs I have to go down to avoid clipping and I am adding the gain later in my DAW, this way I can totally see how using humbuckers hits hardder amplitube. Is anyone having a similar problem? I would prefer to not have to touch my input level knob in my interface ever after calibration, I am considering configuring two inputs one for single and one for humbuckers.

 

[the swede]

If I were to use plugins for amp modeling again, through an interface, this method bookmarked. Only have one guitar so that’s that. Still, I don’t like it…. For me it’s either a Stomp or something like the Iridium (which has a really nice input… at least it “feels” like so). So tired of messing around. I know for a fact that the last few years have been focused on “messing around” instead of recording a song or playing for fun. I’m a homeplayer with no time over. Maybe it’s time to go back to a 1x10 combo and 3-5 pedals

 

[James Freeman]

Maybe it’s time to go back to a 1x10 combo and 3-5 pedals

You laugh, but this thought never stops haunting every modeler owner.

 

[MirrorProfiles]

After doing the calibration I ended with the preamp gain level I have been using for some time in the instrument input of my interface for single coil guitars, it is a relief to know that I could dialed on my own following my instincts. But I have a problem, if use a humbucker at that level my interface starts clipping on peaks. The input indicator goes from average green with single coil, to average orange with humbucker, and peaks end in read with humbuckers. I have measure how many dbs I have to go down to avoid clipping and I am adding the gain later in my DAW, this way I can totally see how using humbuckers hits hardder amplitube. Is anyone having a similar problem? I would prefer to not have to touch my input level knob in my interface ever after calibration, I am considering configuring two inputs one for single and one for humbuckers.

Humbuckers SHOULD hit the amp harder, that’s exactly the point - you want the model to react just like the real amp would.

My recommendation would be to set the interface gain to 0 and record with more headroom for all pickups, and then boost by a fixed amount in the plugin (vs changing gain levels and using different amounts of plugin input gain).

If 12dBu isn’t enough headroom, try 14 or 16 dBu

 

[the swede]

You laugh, but this thought never stops haunting every modeler owner.

Laughing at myself…

 

[SillyOctpuss]

You laugh, but this thought never stops haunting every modeler owner.

I did it. Kinda. I just integrated my stomp with a jvm and a bunch of pedals. I've never been happier.

 

[MirrorProfiles]

modellers don’t replace a real amp, they’re a different thing with their own pro’s and con’s.

electric guitar as an instrument is intrinsically tied to the amp and cab - it’s part of the instrument itself and you perform and react to the whole chain as one. modelling is an awesome compromise but it ain’t the real thing and shouldn’t be treated in that way IMO.

 

[valimaki]

Humbuckers SHOULD hit the amp harder, that’s exactly the point - you want the model to react just like the real amp would.

My recommendation would be to set the interface gain to 0 and record with more headroom for all pickups, and then boost by a fixed amount in the plugin (vs changing gain levels and using different amounts of plugin input gain).

If 12dBu isn’t enough headroom, try 14 or 16 dBu

The problem is the SNR, it is better to use the preamp in the audio interface as much as possible. So I think I am going with a compromise raising the input in the interface but with enough headroom for all my guitars, and then adding a fixed amount in the plugin/daw (6db). That way it is pretty much plug and play for all my guitars.

 

[MirrorProfiles]

The problem is the SNR, it is better to use the preamp in the audio interface as much as possible. So I think I am going with a compromise raising the input in the interface but with enough headroom for all my guitars, and then adding a fixed amount in the plugin/daw (6db). That way it is pretty much plug and play for all my guitars.

are you having issues with signal to noise that is caused by the preamp? Unless something is seriously wrong I don’t think a couple of dB is going to make any perceivable difference, I think most of the attitude of “tracking as hot as possible” is born out of reading things online and worrying about stuff that isn’t actually causing any sort of issue. Recording at 16 bit on some old ass gear? maybe I’d track hotter. 12dBu is already too hot for my liking, absolutely no qualms with tracking at 20dBu and having a bit more breathing space before clipping.

My guess is you’ll have higher levels of RFI/EMI as well as other things before the noise of your converters and preamp become an issue.

 

[Orvillain]

The problem is the SNR, it is better to use the preamp in the audio interface as much as possible.

No, it is better to have good gainstaging, which is what this entire thread is about.

 

[MirrorProfiles]

worth remembering what the noise floor of a typical 9V pedal, amplifier is, let alone the interference a pickup can’t avoid in the room.

Not saying low noise isn’t important but focussing on that isn’t going to have any real impact on tone, there is lower hanging fruit that will make a difference. It’s probably takes more work to record a DI that is too quiet than it is record one too hot

 

[pipelineaudio]

You laugh, but this thought never stops haunting every modeler owner.

Not even once

 

[MirrorProfiles]

Fuse Audio Labs:

The F-59 is matched at 0 dBFS = +18 dBu just like all our other analog modeled plugins

The pedals are matched at 0 dBFS = +14 dBU

Hope that helps.

Best,
Ray

Gotta respect the companies that give a straight answer with no bullshit. 50% off their plugins this weekend, I've only used the pedals so far, which I can absolutely recommend).

Fuse Audio Labs | Audio Plugins and Mobile Audio Apps

Fuse Audio Labs - Audio Plugins for Recording, Mixing and Mastering

fuseaudiolabs.de

 

[SwirlyMaple]

Fuse Audio Labs:

The F-59 is matched at 0 dBFS = +18 dBu just like all our other analog modeled plugins

The pedals are matched at 0 dBFS = +14 dBU

Hope that helps.

Best,
Ray

See, IK Multimedia? See how easy that was? C'mon, you can do it. I believe.

 

[bikescene]

Fuse Audio Labs:


Gotta respect the companies that give a straight answer with no bullshit. 50% off their plugins this weekend, I've only used the pedals so far, which I can absolutely recommend).

Fuse Audio Labs | Audio Plugins and Mobile Audio Apps

Fuse Audio Labs - Audio Plugins for Recording, Mixing and Mastering

fuseaudiolabs.de

Great info.

I just came across this forum through this thread.

The calibration for the Fuse sim, the Softube, and Mercuriall stuff earlier in the thread is very helpful.

It looks like you guys already thoroughly explored levels for the Helix and FM3, but I'll make sure measurements on my own time too. I have a Fluke 289 lying around that should still be in calibration.

 

[James Freeman]

Plugin manufacturers need to wake the hell up, this is a real issue.
One line in the manual can increase modeling accuracy a lot for those who want it and change nothing for those who go by ear.
This information is crucial for proper gain staging ESPECIALLY guitar amplifier plugins where GAIN is everything.

Why can't everyone provide proper dBFS and dBu numbers in the manual like UAD?

 

[pipelineaudio]

Wait, did UAD actually do something right for once?

 

[MirrorProfiles]

^ x1000000000000

Funny how a lot of that studio gear is purposefully designed to operate in a linear way across a wide dynamic range. Obviously a lot harder to achieve on older tube gear where they definitely have more non linearities and sweet spots, but even on much of the UAD lineup, the differences with +/-5dB aren't going to be anywhere near as drastic as a guitar amp where they are absolutely NOT designed to be patched into a mixing desk and fed any manner of different signals.

Guitar amps expect a guitar (or pedals) to be plugged straight in. The design is based around that, no matter how it ends up being used. For a guitar amp emulation to have ANY kind of success it needs the correct input signal, and there is no way around that.

Guessing is absolutely not good enough, and it doesn't help when I'm trying to use an emulation of a $20,000 amp that I have no experience of using to gauge if its performing correctly.

Wait, did UAD actually do something right for once?

I mean, emulation wise, they're still the benchmark right? Some other companies get close but I think basically every developer out there has a ton of respect for UAD's emulation quality and the depth they model complex behaviours and weird interactions. There are still many behaviours that most other emulations simply miss, that UAD have been incorporating for a long time now.

 

[laxu]

I mean, emulation wise, they're still the benchmark right? Some other companies get close but I think basically every developer out there has a ton of respect for UAD's emulation quality and the depth they model complex behaviours and weird interactions. There are still many behaviours that most other emulations simply miss, that UAD have been incorporating for a long time now.

UA is well liked for how their things sound. Everything else, like features, software/firmware support, Windows support, pricing...that's a lot more iffy.

 

[pipelineaudio]

I mean, emulation wise, they're still the benchmark right?

I don't know about "still" as likely many would say "never were". Similar to the line 6 2203 situation, the units modeled were hillariously chosen, pretty sure the "1176" was a very specific Purple Audio MC76 rental unit for instance

Some other companies get close but I think basically every developer out there has a ton of respect for UAD's emulation quality and the depth they model complex behaviours and weird interactions.

I'm not aware of any developer who says this. I AM aware of many developers who take issue with the poor coding quality of these things since day one, their nonsense claims about their card vs native cpu use which was utterly and ultimately decimated when it could finally be shown with a plugin that ran on both that not only did the card dongles NOT save you CPU, but it cost you more cpu to run the same number of instances on the card than just straight native.

And of course, most famously, when almost universally seen as one of the best if not the best coders in the business offered to fix UAD's horrible and unusable drivers for free and they instead said their customers could just suck it and buy the next version of the dongle card that came out

There are still many behaviours that most other emulations simply miss, that UAD have been incorporating for a long time now.

Anything audible? Can you name it?