Differing amp model output levels

I’m still not sure why the idea of normalizing is so polarizing here.
I don't think it is... I think there are a few people in this thread who seem to think it's simple and the others (myself included) who don't.

It's not the first time the subject has been raised... We've tried to demonstrate why it's not practical.

I'm done debating. If it happens, great - I don't think it's a bad idea - but I would be extremely surprised to see it happen.
 
no, because the resulting Level number doesn't/shouldn't matter if it creates the amount of signal you want to hear. the delay doesn't see numbers, that one is +7 and another is -5, "so one is clearly louder by 12dB." it just takes the signal level the same way we hear it.

i haven't had time to load that preset yet, but what you're describing shouldn't be happening, if i'm understanding correctly.

I’m sure your right, that’s my understanding as well. Now I’m starting to wonder if there’s some phase irregularity or something else which is making things hard to balance. I’ve given up on numbers and meters and just using my ears for now.
 
I don't think it is... I think there are a few people in this thread who seem to think it's simple and the others (myself included) who don't.

It's not the first time the subject has been raised... We've tried to demonstrate why it's not practical.

I'm done debating. If it happens, great - I don't think it's a bad idea - but I would be extremely surprised to see it happen.
Simple or not... we both agree it would be nice as an option.
 
How would you normalize two patches with two different amps so that you could use them in the same song?
You’d adjust output levels. I’m just asking that that be automated. AS AN OPTION.

With the output VU meters.
After you're done tweaking a patch, check the output level at the output block and get it as close as you can to 0.
Rinse and repeat.

Also, why 0db? IT's good practice to do so because it improves sigal to noise ratio.
 
Just to be clear, I didn't mean normalizing in the sense that a process is applied to raise waveforms to a target of 0dB, although that's an interesting idea in itself. I just mean getting the outputs close to the same level when the block has defaults (all controls at noon or whatever). Same as when the presets are done, they seem to go through them and get them on pretty much the same level. That evidently seemed like a step that was worthwhile to someone, loudness accuracy be damned. So I suppose I should have stuck with the term leveling for what I was trying to describe.

There seems to be an argument for accuracy here at times that I just don't think applies in this case: SPL is dependent on the amp and its settings when you're standing there in front of it. But one of the main points of this box is that it is more like hearing the amp from the control booth, where SPL from the amp can be dealt with such that a cranked Twin can be quieter than a 5-watt Pignose with the volume low.

As I said before, it's a nice-to-have IMO. I think default settings work better across amp models compared to earlier versions of the AFX2 firmwares, where defaults tended to be a mixed bag at best with even the gain and TMB controls. So since they are closer to the ball park, I thought it would be nice to not blow my monitors out switching from one amp to another when trying them out. That's all.
 
but yes it bugs me also when i scroll to jazz chorus and its almost dead quiet.

still i am trying explain, the whole idea would not make things any easier....
I've experienced this personally while beta-testing preset packs. I'd flip through a set of presets that were remarkably well-leveled on one guitar, only to find big discrepancies in level with another guitar. Different pickups with different output levels will drive the amp model very differently, taking them into differing amounts of compression and distortion.

It's possible to level presets pretty well — for one guitar. But if your guitar is different from the guitar that the presets were leveled on, they won't be level anymore...and you'll be right back in the same situation of having to adjust levels to compare amp models.
 
There's also the question of what volume you level the presets at. If you take a clean preset and a lead preset, and make them both sound like they're at the same level at moderate volumes, the clean tone will always be louder than the dirty tone when you crank up to gig volume.

And if you dial in both tones to read 0 dB on the VU meter, the clean tone will be perceived as quieter than the lead tone, because it has a higher crest factor and less upper-mids.
 
It's possible to level presets pretty well — for one guitar. But if your guitar is different from the guitar that the presets were leveled on, they won't be level anymore...and you'll be right back in the same situation of having to adjust levels to compare amp models.
[... the clean tone will be perceived as quieter than the lead tone, because it has a higher crest factor and less upper-mids.

Yes, the issue we are all pointing to is NONLINEARITY of amplifier output vs. input (besides other parameters and even human perceptual factors).

I'm *guessing* that FAS's relative amp loudness might be constrained by the dynamic range of the amp at minimum settings versus fully dimed and hit with a hard input so as not to get anywhere near digital clipping. FAS's cleaner amps do seem to be quieter at low-moderate settings, and it's likely that they are more linear and will get relatively louder as settings are cranked: witness a Twin Reverb. Highly compressed/saturated amps have a lower slope or a roll-off of output relative to input. Thus for processing sake their internal "volumes" may be set to be as high as possible without digital clipping to maximize quantization resolution. Or there is some entirely other valid explanation only FAS knows!

But regardless: even if FAS used some method to set them "close" at "default" settings with an assumed certain input level, all the other factors still affect the perceived/actual volume: hotness of guitar PUs, finger/picking attack, bass cut, etc...
 
Last edited:
Yes, the issue we are all pointing to is NONLINEARITY of input vs. amplifier output (besides other parameters and even human perceptual factors).

I'm *guessing* that FAS's relative amp loudness might be constrained by the dynamic range of the amp fully dimed and hit with a hard input so as not to get anywhere near digital clipping. FAS's cleaner amps do seem to be quieter at low-moderate settings, and it's likely that they are more linear and will get relatively louder as settings are cranked: witness a Twin Reverb. Highly compressed/saturated amps have a lower slope or a roll-off of output relative to input. Thus for processing sake their internal "volumes" may be set to be as high as possible without digital clipping to maximize quantization resolution. Or there is some entirely other valid explanation only FAS knows!

But regardless: even if FAS used some method to set them "close" at "default" settings with an assumed certain input level, all the other factors still affect the perceived/actual volume: hotness of guitar PUs, finger/picking attack, bass cut, etc...
You’re on the right track, but it’s not about maximizing internal “volumes.” Internally, Fractal gear has more than 1400 dB of dynamic range. :)
 
These kinds of things always seem to come down to “why can’t my personal use case be the default?”.

People here have debated for years over what the default settings of the models should be (noon, zeroed, common model-specific settings, etc.) Different people use the same amp models in wildly different ways.

Then there’s the issue of different instruments, which can mean wildly different EQ/output from one guitar to the next that will further wreck any attempt at establishing a baseline.

Beyond those variables, there’s just the matter of normalization itself. AFAIK (could very well be wrong) any kind of automated normalization process is just going to be some sort of global limiting, not at all the same result as manually adjusting output level. That would be a huge step in the opposite direction from the kind of realism that the amp models aim for.

It would be really nice for sure, and I’d be happy to be wrong. But I don’t think there’s any way to account for so many different use cases, and even if there was I don’t know of an automated solution that wouldn’t have a negative impact on sound overall.
 
Last edited:
let me dumb this way down. could we simply have just the few amp models which are really quiet brought up a bit to sort of match the rest? just so their levels are not so extreme when scrolling thru the models. thank you
 
Back
Top Bottom