Input Voltage/Gain Calibration for Guitar Plugins

[Ilya-v]

I am going to test and verify (again) that adding gain on the audio interface is quieter then boosing in the daw after the adc.
I am sure I already tested that and found that the noise floor boosts by the same amount in both cases, but I can remember wrongly.

[Burillo]

you should also keep in mind that results you obtain on your interface may not be applicable to other interfaces. if i get time, i might run a few tests on my hardware, and see what i get.

[michaeltn86]

Michael you seem to completely ignore or didn't understand the calibration procedure.
Did you understand the reason why I'm trying to calibrate to external 1vAC Peak?
.

well, I could have understood wrong, as English is not my mother language. as I said, I'd be happy to be corrected.

But, here's why calibration doesn't work:
Even though u calibrated through a sine wave, how ur picks ups / guitars work with their input / output level differs depending on the chain.

For instance, I have a MOTU interface. and if I calibrate the guitar through it, the sound will differ than have an DI / PREAMP and connecting into line level at the same exact peak level.

it's the same behavior as pedals. even if the pedal is from the same company and the same materials, it'll still differ. it depends on the chain.

Surely the way u are thinking is nice, to have a "controllable" and "measurable" level, but if u are not calibrating with ur guitar, then it's a kind of wasted effort in my opinion. if u consider how ur chain works, it gets worst.

In my experience, finding the best ratio of noise / guitar level is the best approach, and then just find a level that works nice with the amp sim.

every developer are different, and some "program" their vst to receive their "input level" differently, so that's why saying "most vst" guitar amp sims is not healthy in my opinion. I surely would like a step by step "guaranteed" way of getting good sound through amp sims, but, I have low hopes on that.

in my experience, the worst problem for guitar amps sim are:

-not properly setup guitar

-poor electronics in ur guitar

-too peaky signs depending on how "strong" u play the guitar. for instance, I am a heavy handed guitar player, so I need a compressor before hitting my converter (1-4 db GR max works for me). this is really really nasty, as the input from the amp "naturally" compress and saturates depending on its gain, making ur guitar sound more natural, while in the amp sim, the signal is just too clean.

-too hot signal going in the interface. In my experience with different interfaces, I found out that MAX -10db peak level is preferable than higher level. I'm not sure why, but sometimes the converter starts clipping, and the sound gets nasty (even if u are not hitting 0 dbfs). Then gain stage with digital gain plugin.

-too hot signal going into the amp sim. Amps react different depending on how loud ur input level is and how u hit them. It also depends on how the developer developed the amp sim. Even though u calibrated the level, it will vary from amp to amp, from developer to developer, and from the exact amp modeled (it's pretty difficult to have exact clones, even from the same manufacturer).

-string gauge. If u want a thicker sound, use a thicker string gauge.

-3-5K nastiness. u need to EQ or multi band compression

-the bad playing. when someone is playing bad and /or is out of tune and you are hearing from "far away", it doesn't bother u too much. But get closer to the sound, and it's painful. an Amp sim is an AMP close to ur ear, and that definitely hurts

That said, gain staging is a priority. I understand ur point, but the way u stated, and how u are suggesting to work is not healthy in my understanding. it's just too many variables to have a "perfect recipe".

I am a firm believer that every new plugin should have a in depth tutorial. especially plugins to guitar players, as guitar players, in general, are not very bright. I have experience on the "not very bright" because I'm one. ;]

I would love to be corrected though.

[michaeltn86]

I also forgot to say, that "today", and in my opinion, almost all guitar amp sims are good sounding. But how they work are totally different.

also, in a mix context, a good sounding amp sim might not sound as good. and a not so good sounding amp sim might sound the perfect choice.

I was shocked hearing Van Halen's guitar playing of "Panama" in an isolated tracking. that totally blowed my mind and it was last week.

BTW, I also would love to know how ur guitar sounds with ur method. Please post some samples !

[michaeltn86]

In AmpliTube, the setting of the input gain on the Hi-Z input of your interface (or on iRig HD 2 for example) has to be made right in the same way you would properly set the gain of any preamp in the world (mic or instrument) - exactly like the ABC rules of audio gain staging for digital audio: set the input gain so that the acquired signal on the digital recorder or DAW is as close as possible to 0dBFS without ever clipping. This ensures the best possible dynamic range and SNR.

That’s the basic rule, and it’s the best practice to set the input level of AmpliTube. We ensure doing this you get the most realistic experience from the models that are finely calibrated to exactly these conditions.

Hi Peter, sorry to bother. I'm no pro and not a developer like you (IK), but does IK really suggest to record as close as possible to 0 dBFS? And, what's the explanation for that (besides the ones u mentioned - This ensures the best possible dynamic range and SNR)?

I was rereading this thread and I am a little surprised about this info.

In my experience, and I would love to be corrected, that's not a very good advice.

It really made me curious to test a IK interface and try to record as close to 0dbfs.

thanks in advance for the explanation.

[Burillo]

different people recommend different standards. for example, Lexicon Omega manual states that your guitar should be peaking at -6dB. some people say it should be "as close to clipping as possible" (like IK and a number of other developers). some people recommend setting up input levels so that hard playing registers at around 0VU (this is what i do). now we got another one - minimum gain!

[michaeltn86]

I might be completely wrong, but the advice from IK was shocking to me.

thanks for sharing the lexicon omega info too. Well, I'm not a successful person, so the wrong one might be me haha

I also try to go for the 0vu level...

[Burillo]

to be fair, i wouldn't expect 0VU vs "maximum before clipping" to have many differences. the advice regarding 0VU is usually justified with the fact that -18dB RMS (aka 0VU, loosely speaking) level has the lowest THD by virtue of being "reference level" (i.e. level at which hardware is expected to operate). so, SNR would probably be about the same, but maybe the "maximum before clipping" would introduce a tad more distortion (which the OP probably wouldn't care about as he has EMG pickups...).

with the "minimum gain" advice however, you're entering a territory of intentionally raising the noise floor, which will be more noticeable. so i would say that even if IK's advice might turn out to be slightly wrong, it'll probably be far less wrong than what OP is recommending.

[Ilya-v]

In my experience, finding the best ratio of noise / guitar level is the best approach, and then just find a level that works nice with the amp sim.

This thread is about NOT doing exactly that.
Please try to read and understand the first post.


Burillo:
My intention was not to maximize snr but calibrate to 1vac peak so that the simulated amps receive emulated 1v when there is actual 1v at the input, even if it raises the noise floor.
Yet the discussion steered to the snr part of it even if it's uncertain if true.
I have several guitars with semour duncans, single coils and an emg equipt guitar, please don't make quick assumptions.

[michaeltn86]

hi iilya, ok, I must be not understanding it.

I put a logic project, a very simple one, on dropbox in a post in this topic. it has the raw files from my recording chain.

if u understood ur explanation correctly, I should boost the level through my daw to as close as 0db as possible and then load an amp sim. is that correct? If so, can you please kindly edit any of tracks in the project (I can transform to wav if u are not a logic user) according to ur suggested 1st post and please send it to me? it can be a pm or a post in this thread. please use the amp sim that u see fit, no additional eq / compression /etc is needed. just a simple gain stage and load of amp sim of ur choice.

I also be happy if u can pm or post any guitar sample that u can make. if u can do that I would like the DI and the project so I can check it here too.

To be completely honest, I love to learn more regarding amp sims, as I can't play through an amp regularly. so I have real interest and curiosity towards the subject. thanks!

edit: here's the wav files https://www.dropbox.com/sh/n6ufwrx5g8ot ... 8sh-a?dl=0

[Ilya-v]

Michael did you completely ignore the 1vAC peak sine wave at the input part?
Please read slowly and understand every word of the first post, there are even pictures of exactly what I'm doing to calibrate in the second post.

[dimitar]

Guys, I know its too much to post a video here....self promotion and so on.
The truth is simple. Just keep it clean and low. Trust your ears.
In this track the solo guitar is hitting around -4 -6 in Reaper dB meter.
No big science here. Every guitar has a different voltage output. You cant make it as a standard...!

[Burillo]

My intention was not to maximize snr but calibrate to 1vac peak so that the simulated amps receive emulated 1v when there is actual 1v at the input, even if it raises the noise floor.

this is exactly why i'm commenting. if you were offering a way to make up the difference between "properly calibrated" (i.e. best SNR) input and 1VAC virtual reference, i wouldn't even be in this thread - it would've just been an interesting curiosity for those that want additional realism.

i don't care what is your intention, it's the side effect of raising noise floor that i'm worried about, and that's why i'm commenting - so that the people who don't know any better don't see all of this complicated magic and electronics jargon and think "oh, this guy definitely knows what he's talking about" and follow your advice. you tell people how to achieve "realistic" voltage levels in the box - great. but you're also giving bad gain-staging advice, and that is a big no-no in my book.

[Ilya-v]

Burillo do you assume the gain knob on the audio interface does not raise the analog noise floor?
Have you actually tested it?

[michaeltn86]

well, I will try to read again tonight, not the brightest one here. and then try again.

kvass amps sounds great to my ears, but don't react well to my playing =/ that's also a big playing factor to me.

kvassa bass amp on the other hand sounds great to me in my crap playing :] and I have it!!!

cheers