Unnatural sounding 'fizz' on sustained notes

[FractalAudio]

I didn't want to get into this but here we go...

The plot below is a real Deluxe Reverb (green) vs. a QC (magenta) vs. an Axe-Fx III blue. This is a simple harmonic overtone test using a 1kHz sine wave (shifted up/down by one semi-tone to separate the traces) into the DUT.

If you look at the green trace (real amp) you'll see the overtones are strong all the way out to 20kHz. The Axe-Fx III's overtones are very similar in amplitude. The QC's overtones roll off starting before 10kHz. This makes the fizz less noticeable but it's less accurate.

As an aside you can see the significantly better SNR of the Axe-Fx III vs. the other two devices. Also note that there are no sidelobes on the QC at the fundamental frequency. This is due to the lack of power supply modeling.





If I were to guess I'd say the QC is using arctan or tanh to model the grid clipping. This results in smoother clipping with less overtones. We use a high-order function that more accurately models the forward-biased diode behavior of a grid.

 

[Dave Merrill]

I didn't want to get into this but here we go...

The plot below is a real Deluxe Reverb (green) vs. a QC (magenta) vs. an Axe-Fx III blue. This is a simple harmonic overtone test using a 1kHz sine wave into the DUT.

If you look at the green trace (real amp) you'll see the overtones are strong all the way out to 20kHz. The Axe-Fx III's overtones are very similar in amplitude. The QC's overtones roll off starting before 10kHz. This makes the fizz less noticeable but it's less accurate.

As an aside you can see the significantly better SNR of the Axe-Fx III vs. the other two devices. Also note that there are no sidelobes on the QC at the fundamental frequency. This is due to the lack of power supply modeling.


View attachment 148580


If I were to guess I'd say the QC is using arctan or tanh to model the grid clipping. This results in smoother clipping with less overtones. We use a high-order function that more accurately models the forward-biased diode behavior of a grid.

Remarkable accuracy with the Axe, very cool.

Do i understand correctly that for folks who prefer a more airbrushed overdrive tone without rolling off high end, the thing to try is lower Master Bias Excursion?

 

[Rex]

I didn't want to get into this but here we go...

The plot below is a real Deluxe Reverb (green) vs. a QC (magenta) vs. an Axe-Fx III blue. This is a simple harmonic overtone test using a 1kHz sine wave (shifted up/down by one semi-tone to separate the traces) into the DUT.

If you look at the green trace (real amp) you'll see the overtones are strong all the way out to 20kHz. The Axe-Fx III's overtones are very similar in amplitude. The QC's overtones roll off starting before 10kHz. This makes the fizz less noticeable but it's less accurate.

As an aside you can see the significantly better SNR of the Axe-Fx III vs. the other two devices. Also note that there are no sidelobes on the QC at the fundamental frequency. This is due to the lack of power supply modeling.


View attachment 148580


If I were to guess I'd say the QC is using arctan or tanh to model the grid clipping. This results in smoother clipping with less overtones. We use a high-order function that more accurately models the forward-biased diode behavior of a grid.

On this Thanksgiving Day, I give thanks that you’re willing to break it down and lay it out like this.

 

[FractalAudio]

Remarkable accuracy with the Axe, very cool.

Do i understand correctly that for folks who prefer a more airbrushed overdrive tone without rolling off high end, the thing to try is lower Master Bias Excursion?

That may or may not help. There is a parameter to adjust how hard the grids are driven but it is not exposed to the user (FWIW that parameter is nearly at maximum for the Diamante Fire model. That amp absolutely obliterates the grids). There's also hidden code to use softer clipping models but, again, not exposed to the user.

Power tube grid clipping is nasty stuff but it helps cut through the mix. When listening in isolation using IRs, especially IRs made with dynamic mics (i.e. SM57), that fizz can be quite noticeable. As you can see from the plots the fizz is in the upper treble region.

There are several things you can try:
Use a different IR that doesn't accentuate this region.
Turn down the High-Cut Frequency on the IR.
Reduce MV.
Increase Power Tube Grid Bias.
Increase Transformer Matching and/or Speaker Impedance so that the plates clip before the grids (plate clipping is smoother).
Use a different amp model that relies more on preamp distortion.

I've contemplated for years a parameter that allows switching between "Authentic" and "Smooth" power tube clipping because this topic rears its ugly head now and then. But in the end I always stick to the "accuracy is paramount" philosophy.

 

[Jim Amsden]

Hard to know for sure, but this might be a psycho-acoustic effect. Typically tube amps, especially tube amps with no master volume, are turned up really loud to get any clipping/saturation. But that time our ears are sort of naturally “ducking” the fizz/intermodulation distortion, simply because the distorted guitar is loud enough to overwhelm it, and the overall amplifier doesn’t have any gain left to amplify the intermodulation distortion.

But modelers mostly have both master volume, amp block levels, and output/playback levels. The good news is that lets us get a nicely saturated tone with power amps, preamps and/or drive blocks at lower, sustainable volumes. What we often do to compensate for that is add more gain/saturation which looses dynamic range and introduces more intermodulation distortion, making the OP’s problem worse. Then if you’re going through a cab block or IR into a FRFR, and you don’t use quite a bit of high-cut, that will make the fizz even louder because you’re effectively listening to a close-mic’d speaker.

So I don’t know if Fractal Audio devices are creating more intermodulation distortion than real amps or other modelers. I’ll try to compare to Quad Cortex, Helix, and a Hot Rod Deluxe this weekend, busy cooking now. But I suspect and trust that Cliff has done a fantastic job accurately modeling tube amps and all their warts. As others have said, it’s these quirks that give the amps color and help cut through the mix.

Try running the modeler and amp at the same levels with the same amount of saturation, obtained mostly from the power amps. I suspect you’ll find they sound pretty similar. Also try putting your modeler into the power amps input of your guitar amp to isolate just the effect of the amp.

 

[Rex]

I've contemplated for years a parameter that allows switching between "Authentic" and "Smooth" power tube clipping because this topic rears its ugly head now and then. But in the end I always stick to the "accuracy is paramount" philosophy.

As long as the accuracy is available, why not? Fractal has a long history of providing ways to deviate from accuracy. That’s been a wonderful creative tool.

 

[duzie]

I didn't want to get into this but here we go...

The plot below is a real Deluxe Reverb (green) vs. a QC (magenta) vs. an Axe-Fx III blue. This is a simple harmonic overtone test using a 1kHz sine wave (shifted up/down by one semi-tone to separate the traces) into the DUT.

If you look at the green trace (real amp) you'll see the overtones are strong all the way out to 20kHz. The Axe-Fx III's overtones are very similar in amplitude. The QC's overtones roll off starting before 10kHz. This makes the fizz less noticeable but it's less accurate.

As an aside you can see the significantly better SNR of the Axe-Fx III vs. the other two devices. Also note that there are no sidelobes on the QC at the fundamental frequency. This is due to the lack of power supply modeling.


View attachment 148580


If I were to guess I'd say the QC is using arctan or tanh to model the grid clipping. This results in smoother clipping with less overtones. We use a high-order function that more accurately models the forward-biased diode behavior of a grid.

This is your passion
Let it fly !

 

[Dave Merrill]

That may or may not help. There is a parameter to adjust how hard the grids are driven but it is not exposed to the user (FWIW that parameter is nearly at maximum for the Diamante Fire model. That amp absolutely obliterates the grids). There's also hidden code to use softer clipping models but, again, not exposed to the user.

Power tube grid clipping is nasty stuff but it helps cut through the mix. When listening in isolation using IRs, especially IRs made with dynamic mics (i.e. SM57), that fizz can be quite noticeable. As you can see from the plots the fizz is in the upper treble region.

There are several things you can try:
Use a different IR that doesn't accentuate this region.
Turn down the High-Cut Frequency on the IR.
Reduce MV.
Increase Power Tube Grid Bias.
Increase Transformer Matching and/or Speaker Impedance so that the plates clip before the grids (plate clipping is smoother).
Use a different amp model that relies more on preamp distortion.

I've contemplated for years a parameter that allows switching between "Authentic" and "Smooth" power tube clipping because this topic rears its ugly head now and then. But in the end I always stick to the "accuracy is paramount" philosophy.

Thanks for the airbrushing rundown

I remember when i first heard videos of an early Axe, probably the original, it had more of that airbrushed sound (says my notoriously unreliable audio memory). I thought it was amazing, like a record. I couldn't get my actual amp to sound like that. I remember using the term "grown-up tone".

Things are much more nuanced today, both the Axe itself and our perception of tone.

Personally, other than the non-trivial time and effort investment it would take on your end, i don't see a downside to an Airbrush switch or parameter. Best of both worlds. It would also let folks experience the pluses and minuses of those real-world imperfection for themselves, rather than having to debate the value of fidelity every few months.

 

[Countryclarke]

Ya tell em why QC sucks balls.

 

[Jim Amsden]

QC definitely doesn't suck. I have QC and FM9 in front of me at the moment building similar patches for both with an A/B switch to switch between them. They're different for sure. I like them both, including the differences. I can get great tones for any purpose out of both. I'm loving the new Soldano 100 plugin model in QC 3.1 released today. Its great to have the options.

I did a quick test of similar QC and FM9 patches to compare fizz/intermodulation distortion. To my ear, they're more similar than different. So I don't think there's a fizz/intermodulation distortion issue with FM9 8.0.2.

I find the biggest difference between QC and FM8 tend to be in the cab models used. I haven't tried the same IR in both, but I suspect the difference in tone would not be that significant. But I actually like the differences and that's why I use both for different purposes.

 

[Countryclarke]

QC definitely doesn't suck. I have QC and FM9 in front of me at the moment building similar patches for both with an A/B switch to switch between them. They're different for sure. I like them both, including the differences. I can get great tones for any purpose out of both. I'm loving the new Soldano 100 plugin model in QC 3.1 released today. Its great to have the options.

I did a quick test of similar QC and FM9 patches to compare fizz/intermodulation distortion. To my ear, they're more similar than different. So I don't think there's a fizz/intermodulation distortion issue with FM9 8.0.2.

I find the biggest difference between QC and FM8 tend to be in the cab models used. I haven't tried the same IR in both, but I suspect the difference in tone would not be that significant. But I actually like the differences and that's why I use both for different purposes.

I was being facetious.
There’s so much involved with what makes up a “tone”. I really have a hard time thinking any of these flagship units can’t get you what you’re looking for in the end. It’s just the process that people have preference over. Fractal can be a little crazy sometimes with the amount of tweaks available and how the interact with each other. SIC, for example, is just one very simple tweak that can make a MASSIVE difference in what you’re hearing and feeling.
I recently switched my Div/13 amp model from 6v6 power tubes to el84 and it drastically changed how “flubby” the low end was. I could have probably changed 6 different things to accomplish a similar goal. Now the other 2,861 tweaks available are likely going to be affected by the switch to a different power tube, so these little micro interactions are crucial and can pile up quickly if you’re not careful and don’t have intent to your tweaks.

 

[Brewce]

I've contemplated for years a parameter that allows switching between "Authentic" and "Smooth" power tube clipping because this topic rears its ugly head now and then. But in the end I always stick to the "accuracy is paramount" philosophy.

I'm in the "Why not both?" camp, which may be for some wrong reasons and just a hunch...

I would almost say that the Ultra clip in this genre is still "unsurpassed" (there were a few)...
https://www.jocce.net/80s_1.mp3
(for another: https://forum.fractalaudio.com/threads/80s-landau-sounds.18154)

Not because I want that sound every time, but because it kind of sounds "plastic" in a most desirable way.

I don't nearly know all the factors that go into this (maybe it's mostly how he perfected mids and cuts? Or a Tyler guitar mid boost?), but it makes me want a "plasticity" control or something ("Plastic Fantastic"...? ).

Maybe the "plastic" extreme values could even be a newish sound in ways, or just "like very old modelers, but better."
Every sound fits something best...?

Anyway, if it could be gradual, I suspect everyone can be happier for the options. If you only want to do one, keep it where it is -- I have an Ultra. But it does seem enough people would like it too. I think I understand why you could lean both ways and have done so.

If I knew how to make this same clip with the current modeling, it would be easier to compare and I might not even want the Ultra sound -- but at the moment I'm still doubting that. And more importantly, I choose it not because I dislike real amp sounds, but because there's something magical here that fits this style very well (sure, many things may play into it, like the excellent playing, FX, and production to my senses -- it's just KILLER!).

IOW, you don't always want a real amp sound...
OTOH, if half of the new III clips would become the plastic sound, I would understand your frustration...
On another hand, I also think it is cool to have your own history in the units, like (if not a control) a dropdown that somewhat represents Ultra / Axe-Fx II / pre-organic III sounds too.

Maybe the above can add an extra argument to things... -- or I'm 70% off the mark as usual...

 

[Dave Merrill]

Thanks for the airbrushing rundown

I remember when i first heard videos of an early Axe, probably the original, it had more of that airbrushed sound (says my notoriously unreliable audio memory). I thought it was amazing, like a record. I couldn't get my actual amp to sound like that. I remember using the term "grown-up tone".

Things are much more nuanced today, both the Axe itself and our perception of tone.

Personally, other than the non-trivial time and effort investment it would take on your end, i don't see a downside to an Airbrush switch or parameter. Best of both worlds. It would also let folks experience the pluses and minuses of those real-world imperfection for themselves, rather than having to debate the value of fidelity every few months.

FWIW, i * think * this is the first Axe video i saw, by @Scott Peterson. I haven't watched again now, but if it is that video, it might be an example of that older style Axe texture.



(I didn't tag him here because I'm not actually sure that SP is the same person as the current member by that name.

EDIT
Confirmed and tagged. Hi Scott! You don't know me, but you're my original Fractal gateway guy. Cheers!)

 

[skolacki]

As long as the accuracy is available, why not? Fractal has a long history of providing ways to deviate from accuracy. That’s been a wonderful creative tool.

Gotta agree with you. I love the accuracy but on some of my tones it’d be nice to dial it down a bit. I’m pretty damn happy as is though, no complaints.

 

[Rex]

(I didn't tag him here because I'm not actually sure that SP is the same person as the current member by that name.)

Let’s see... PRS in hand, smooth head, Laura Wilkie Band, Living Water... yeah, it’s him.

 

[Dave Merrill]

Let’s see... PRS in hand, smooth head, Laura Wilkie Band, Living Water... yeah, it’s him.

Post updated

 

[Dino334]

Authentic and Smooth option? Be kinda cool if ya did! Alright, alright, alright

 

[hippietim]

Lots of crispy stank on this one too. Tubes are not always warm and purring. Sometimes they clip hard and nasty.

That's glorious.

 

[Smilzo]

I've contemplated for years a parameter that allows switching between "Authentic" and "Smooth" power tube clipping because this topic rears its ugly head now and then. But in the end I always stick to the "accuracy is paramount" philosophy.

I don't know if it's doable, but as musician I Like to have parameters that let me colour the sound. I wish some "macros" parameters, like "fizziness" "stiffness" "Smoothness" "Brightness" "Dynamics" that affect one or more virtual component/behaviour. Say, "Stiffness" at 50% is authentic. Boosting the parameter make the amp "stiffer", decreasing less stiff. Not everybody is an engineer and know or want to know the electronic component(s) to alter.

So I envision 3 folders in the amp: first the amp authentic control (gain, eq, bright, etc...), a second folder with macros (that allow the player intuitive and fast adjustment), and last the advanced electronic folder (with bias, hardness, transformer ratios, histeresis,...).

@FractalAudio what you think? Doable? There are many ways to implement those macros, each with pros and cons...

 

[Cuiken]

You'd probably be happier with a Quad Cortex. Our modeling stresses accuracy. The "fizz", as you call it, is caused by power tube grid clipping and has a somewhat disassociated quality.

Woa Woa Woa! Let's not get drastic now. I love the pure clean tones on my FM9 and find them to be way more 'substantial' somehow that those on the Neural plugins I have been playing with. The FM9 has incredible definition to the sound along with all the high order shimmery harmonics that, I think, other modellers lack. The Neural plugins aren't bad at all and do have some of the shimmer but, IMO, lack the string definition. I'm only guessing, but I assume it'd be the same on the QC.

High gain stuff on the FM9 is incredible too. I'm only having difficulty dialing in some mid gain crunch tones that work when played throiugh headphones or at 'bedroom' volume levels through an FRFR (when the 'fizz' or power tube grid clipping becomes more obvious).