Question for the v11 beta testers

I was of the impression that even a great IR was basically a complex EQ analysis of a speaker cab with frequencies being run through it. I totally accept the fact that I may be wrong, so please do educate me rather than just blow me off.

I don't know what you mean by complex EQ analysis, But look into linear system theory and linear systems characterization. A loudspeaker is mostly, yes mostly described as a linear time invariant system. If you capture an impulse response using a log sweep (see Farina papers) then you can see the non-linear distortion is at such lower level than the linear part that it's essentially irrelevant. The nebula uses volterra kernels, which you can obtain from a log sweep (if you construct it right). I actually built one myself but didn't make much of a difference in the final output, i.e. the linear part completely dominates - makes sense, as when the non-linear response starts becoming significant, the loudspeaker is so far into the limits of its mechanical excursions that it really sounds unusable - not what you normally hear when playing through a guitar cabinet. So again, no the "technology" is not limited as far as emulating a guitar loudspeaker.
 
I don't know what you mean by complex EQ analysis, But look into linear system theory and linear systems characterization. A loudspeaker is mostly, yes mostly described as a linear time invariant system. If you capture an impulse response using a log sweep (see Farina papers) then you can see the non-linear distortion is at such lower level than the linear part that it's essentially irrelevant. The nebula uses volterra kernels, which you can obtain from a log sweep (if you construct it right). I actually built one myself but didn't make much of a difference in the final output, i.e. the linear part completely dominates - makes sense, as when the non-linear response starts becoming significant, the loudspeaker is so far into the limits of its mechanical excursions that it really sounds unusable - not what you normally hear when playing through a guitar cabinet. So again, no the "technology" is not limited as far as emulating a guitar loudspeaker.

I have come to the same conclusions. I made measurements at varying excitation levels and the non-linear components were insignificant until you got to the point that you were on the verge of destroying the speaker. FWIW, the IR capture utility in the Axe-Fx uses the Farina technique so I can measure the high-order components right on the unit (it's just not exposed to the user).

I believe Jay Mitchell also came to this conclusion in his research.

What I have found is that thermal compression is somewhat noticeable and measurable. This is modeled by the Motor Drive parameter.
 
That's what I was gonna say :lol

But seriously ... where is this rule written that an exactly matched tube amp is the best possible guitar tone? We've had the ability, even in the Standard/Ultra, to modify amps to our individual tastes beyond what is available in the real world. So in my book, we have amps that are better than 100% realism right now. As Cliff improves dynamics and matching, we just have the ability to start with a closer match and benefit from some additional tonal subtleties.
 
I have come to the same conclusions. I made measurements at varying excitation levels and the non-linear components were insignificant until you got to the point that you were on the verge of destroying the speaker. FWIW, the IR capture utility in the Axe-Fx uses the Farina technique so I can measure the high-order components right on the unit (it's just not exposed to the user).

I believe Jay Mitchell also came to this conclusion in his research.

What I have found is that thermal compression is somewhat noticeable and measurable. This is modeled by the Motor Drive parameter.

All this math talk is "almost" better than actually playing the AxeFx II
 
I can't even spell "fiziks" I thought we were talking about playing guitar? and what was the topic of this thread anyway?

Oh and B.T.W. ...........nice vocabulary guys:encouragement:
 
Cool! I learned something about IRs today. That's why I love this forum. Back to the original point though - FW 11b is the shit. You will love it. Make it happen!
 
I have come to the same conclusions. I made measurements at varying excitation levels and the non-linear components were insignificant until [...]

Cliff, I love it when you share some of the technical stuff with us. Ever considered giving a public lecture (Ted talks etc), or would that come too close to spilling the beans on the secret stuff under the hood?
 
Cliff, I love it when you share some of the technical stuff with us. Ever considered giving a public lecture (Ted talks etc), or would that come too close to spilling the beans on the secret stuff under the hood?
You just described Axe fest man! Come on out!
 
I can't even spell "fiziks" I thought we were talking about playing guitar? and what was the topic of this thread anyway?

Oh and B.T.W. ...........nice vocabulary guys:encouragement:
I agree. Very eloquent.


Dibs on Fiziks for a band name! Including derivatives like Fizzikkal!
 
... where is this rule written that an exactly matched tube amp is the best possible guitar tone?
^^^ This. Tube amps (some of them, anyway) are just the best guitar tones most people have heard. The tone you could achieve was limited by the physical reality of what the amp was capable of.
 
Last edited:
^^^ This. Tube amps (some of them, anyway) are just the best guitar tones most people have heard. The tone you could achieve was limited by the physical reality of whta the amp was capable of.
Exactly what i thought !
We don't need to copy real amps anymore, the AxeFX easily achieves tones that real amps/setups couldn't approach !
 
I don't know what you mean by complex EQ analysis, But look into linear system theory and linear systems characterization. A loudspeaker is mostly, yes mostly described as a linear time invariant system. If you capture an impulse response using a log sweep (see Farina papers) then you can see the non-linear distortion is at such lower level than the linear part that it's essentially irrelevant. The nebula uses volterra kernels, which you can obtain from a log sweep (if you construct it right). I actually built one myself but didn't make much of a difference in the final output, i.e. the linear part completely dominates - makes sense, as when the non-linear response starts becoming significant, the loudspeaker is so far into the limits of its mechanical excursions that it really sounds unusable - not what you normally hear when playing through a guitar cabinet. So again, no the "technology" is not limited as far as emulating a guitar loudspeaker.

Cliff, give this man a job already....
 
I have come to the same conclusions. I made measurements at varying excitation levels and the non-linear components were insignificant until you got to the point that you were on the verge of destroying the speaker. FWIW, the IR capture utility in the Axe-Fx uses the Farina technique so I can measure the high-order components right on the unit (it's just not exposed to the user).
One aspect of multi-speaker cabs that I think IRs have trouble with is the fact that the way the drivers interfere with each other is different for each ear in your head. One ear will hear one flavor of comb filtering, and the other hear will hear a different flavor. IMO, this difference contributes a "3-dimensional" quality to some real cabs that is part of the sonic difference between IRs and some cabs.
 
Back
Top Bottom