Will we see improvements/new tech regarding speaker cabinets?

To me, this sounds like an array of impulse responses that are mixed on the fly.

Another company approached me about shooting such files for their software using a very specific methodology that I am not at liberty to discuss.

When the organization could not come across with a legal document that clearly stated who owned the intellectual property for those recorded files used in the array and where rights would revert should the company dissolve, go bankrupt, or be sold, I declined offering files in such a format. They also did not come across with sales projections, which would be important to know considering the gargantuan amount of work on my part per cab to generate what they wanted.
Sounds like the two notes guys
 
Marketing is an amazing thing.
Yup.

The Axe-Fx III (and II) actually capture the Volterra kernels when doing an IR capture (it's hidden in the firmware for possible future use). I've studied dozens upon dozens of them and the kernels above first order (the first order kernel is the linear IR) are so small as to be inaudible. The distortion from an amp is orders of magnitude greater even when using a clean amp.

The only significant nonlinear thing I've measured that speakers do is thermal compression (that we model already) and "cone cry" which sounds like sh*t.

Jay Mitchell is probably the leading authority on speaker design and he has stated pretty much the same thing.

I'm all for improvements but they need to be real improvements. I've sat here countless times comparing an IR to the actual speaker with a mic on it doing blind A/B tests and can NEVER tell the difference and I think my ears are pretty good.

I dug through my Matlab stuff and found this. This is an IR of a speaker taken twice. The first time the drive level is around 1W (in red). The second time the amp was turned way up, I would estimate at least 50W (in blue).
1W_vs_50W.jpg
As you can see the difference is extremely small. There's a small difference from 10 Hz and down which is way below the reproduction range of any system and a difference way up at Nyquist (24 kHz) but that's 100 dB down (!). Furthermore we don't know if the tiny differences are from the speaker or from the amp or the mic or the mic preamp.
 
Last edited:
I should add that speakers can and do distort (when Xmax is exceeded) but it's not a pleasant sound. Since the displacement of the cone is the inverse of the frequency the low frequencies are distorted which is the opposite of what you want when creating "pleasing distortion". Speaker distortion is flubby, flabby and farty. The Axe-Fx II and III can simulate that, if desired, using the Speaker Drive parameter in the Amp block. I always set it to zero.

There are probably some other modes that cause distortion but, again, these are dwarfed by the distortion of the amp. The only other significant one I've experienced is cone cry. Manufacturers go to great lengths to prevent it from happening. I have a speaker here that does it. Whenever I play a high F it cries and it's annoying.
 
There is also a video of him at the same show explaining it to the Sweetwater guy.


I just watched that video. Aside from effects like reverb, he talks about three things that this device brings to the table:

1) Reactive load (we have that in the SPKR page of the Amp block);

2) Thermal compression (we have that in the SPKR COMP parameter of the amp block);

3) Cone cry, which he says is a sub-harmonic of the note being played (sub-harmonics don't occur in nature — you have to synthesize them). You can hear a lower-octave note when he turns this on. But cone cry isn't harmonically related to the signal. It sounds like they're synthesizing an octave-divided note and calling that "cone cry" (we can do that with the Pitch block).
 
Yup.

The Axe-Fx III (and II) actually capture the Volterra kernels when doing an IR capture (it's hidden in the firmware for possible future use). I've studied dozens upon dozens of them and the kernels above first order (the first order kernel is the linear IR) are so small as to be inaudible. The distortion from an amp is orders of magnitude greater even when using a clean amp.

The only significant nonlinear thing I've measured that speakers do is thermal compression (that we model already) and "cone cry" which sounds like sh*t.

Jay Mitchell is probably the leading authority on speaker design and he has stated pretty much the same thing.

I'm all for improvements but they need to be real improvements. I've sat here countless times comparing an IR to the actual speaker with a mic on it doing blind A/B tests and can NEVER tell the difference and I think my ears are pretty good.

I dug through my Matlab stuff and found this. This is an IR of a speaker taken twice. The first time the drive level is around 1W (in red). The second time the amp was turned way up, I would estimate at least 50W (in blue).

As you can see the difference is extremely small. There's a small difference from 10 Hz and down which is way below the reproduction range of any system and a difference way up at Nyquist (24 kHz) but that's 100 dB down (!). Furthermore we don't know if the tiny differences are from the speaker or from the amp or the mic or the mic preamp.

I guess we'll just have to keep repeating it every couple of years or so :)
From 2013:
https://forum.fractalaudio.com/threads/question-for-the-v11-beta-testers.73859/page-3#post-904477
 
Yup.

The Axe-Fx III (and II) actually capture the Volterra kernels when doing an IR capture (it's hidden in the firmware for possible future use). I've studied dozens upon dozens of them and the kernels above first order (the first order kernel is the linear IR) are so small as to be inaudible. The distortion from an amp is orders of magnitude greater even when using a clean amp.

The only significant nonlinear thing I've measured that speakers do is thermal compression (that we model already) and "cone cry" which sounds like sh*t.

Jay Mitchell is probably the leading authority on speaker design and he has stated pretty much the same thing.

I'm all for improvements but they need to be real improvements. I've sat here countless times comparing an IR to the actual speaker with a mic on it doing blind A/B tests and can NEVER tell the difference and I think my ears are pretty good.

I dug through my Matlab stuff and found this. This is an IR of a speaker taken twice. The first time the drive level is around 1W (in red). The second time the amp was turned way up, I would estimate at least 50W (in blue).
View attachment 44467
As you can see the difference is extremely small. There's a small difference from 10 Hz and down which is way below the reproduction range of any system and a difference way up at Nyquist (24 kHz) but that's 100 dB down (!). Furthermore we don't know if the tiny differences are from the speaker or from the amp or the mic or the mic preamp.

That answers a question that had been on my mind (the impact of volume on IR capture) and that illuminates it very clearly.

Anyone want to take bets on "cone cry" being 2018's guitar related buzzword ?
 
There's really no need for me to back up Cliff but I'll chime in on the OP.

I know IRs are great and can sound extremely close to a mic'd up cab but I have to feel like its not as good as it could be sometimes.

Extremely close is quite an understatement. You can shoot an IR many different ways but if you do it accurately there's no one betting money in A/B comparisons. I work with real cabs and IR's multiple times a week and sure don't "feel like its not as good as it could be sometimes". Have you been in a studio control room and compared the same microphone position to an IR shot of it? I highly recommend people doubting IR technology to experience this.

This is the marketing statement:

Dynamic Speaker Modeling
UA’s groundbreaking Dynamic Speaker Modeling faithfully emulates speaker breakup, drive, and cone cry — an essential ingredient to authentic tube amp tones. Using the Speaker Breakup knob in the OX app, you can control the harmonics and sonic complexity that occurs only at certain frequencies and volume levels on various speakers — far beyond a static Impulse Response system — giving you the same sonic phenomenon that occurs with a real speaker cabinet.

So... isn't this just 4 IR's per cabinet with speaker breakup, drive and "cone cry"? UA has quality products for sure but they haven't stated anything new even in this sales pitch except "cone cry" which is something I think they made up. But damn... now I want my cones to cry.
 
Yup.

The Axe-Fx III (and II) actually capture the Volterra kernels when doing an IR capture (it's hidden in the firmware for possible future use). I've studied dozens upon dozens of them and the kernels above first order (the first order kernel is the linear IR) are so small as to be inaudible. The distortion from an amp is orders of magnitude greater even when using a clean amp.

The only significant nonlinear thing I've measured that speakers do is thermal compression (that we model already) and "cone cry" which sounds like sh*t.

Jay Mitchell is probably the leading authority on speaker design and he has stated pretty much the same thing.

I'm all for improvements but they need to be real improvements. I've sat here countless times comparing an IR to the actual speaker with a mic on it doing blind A/B tests and can NEVER tell the difference and I think my ears are pretty good.

I dug through my Matlab stuff and found this. This is an IR of a speaker taken twice. The first time the drive level is around 1W (in red). The second time the amp was turned way up, I would estimate at least 50W (in blue).
View attachment 44467
As you can see the difference is extremely small. There's a small difference from 10 Hz and down which is way below the reproduction range of any system and a difference way up at Nyquist (24 kHz) but that's 100 dB down (!). Furthermore we don't know if the tiny differences are from the speaker or from the amp or the mic or the mic preamp.

In my measurements I achieved the same results as Cliff. As long as the speaker is driven within its specifications, the result remains virtually identical visually, differences from IRs that are not audible except for the loudness with which they have been captured.
Comparing recordings of microphoned cabinet or the IR of it, audible differences are just as difficult to detect.
Compression artifacts (especially in the bass range) and undesired partial oscillations only occur when the speaker is driven outside of its specifications (or if it`s broken), i. e. the diaphragm is no longer able to vibrate properly, and that`s what the speaker compression knob will simulate, similiar to what the OX box will do, i assume...
Here is a picture:
RED: Speaker was driven with +20db far above its load capacity.
GREEN: Speaker operates within its specified watts

speaker.jpg
 
Last edited:
I think the non technical/engineering types are just used to the "there's always room for improvement" attitude. Through the past decade amp/cab modeling has just gotten so good that we are close to splitting hairs. Because of FAS, people have been groomed to think there's always room for improvement though. In the end I can't tell the difference between a cab mic'ed up and the same IR of that setup. There comes a point where we are going to be "there".
 
There's really no need for me to back up Cliff but I'll chime in on the OP.



Extremely close is quite an understatement. You can shoot an IR many different ways but if you do it accurately there's no one betting money in A/B comparisons. I work with real cabs and IR's multiple times a week and sure don't "feel like its not as good as it could be sometimes". Have you been in a studio control room and compared the same microphone position to an IR shot of it? I highly recommend people doubting IR technology to experience this.

This is the marketing statement:

Dynamic Speaker Modeling
UA’s groundbreaking Dynamic Speaker Modeling faithfully emulates speaker breakup, drive, and cone cry — an essential ingredient to authentic tube amp tones. Using the Speaker Breakup knob in the OX app, you can control the harmonics and sonic complexity that occurs only at certain frequencies and volume levels on various speakers — far beyond a static Impulse Response system — giving you the same sonic phenomenon that occurs with a real speaker cabinet.

So... isn't this just 4 IR's per cabinet with speaker breakup, drive and "cone cry"? UA has quality products for sure but they haven't stated anything new even in this sales pitch except "cone cry" which is something I think they made up. But damn... now I want my cones to cry.
Truth!
 
To be fair, I’m thinking that I remember seeing engineering plots that compared the performance of overdriven solid state components versus tubes that were used to show how imperceptible the differences were between the two. Yet people still seem to be able to hear the differences.

Maybe it’s not a fair comparison relative to what is being discussed here, but I guess I’ve grown skeptical about this kind of comparison. But if folks are directly comparing the mic’d cab in a studio with an IR of the same and studio producers can’t tell the difference, it feels like the objective of emulating a mic’d cabinet has been achieved.
 
The Axe-Fx III (and II) actually capture the Volterra kernels when doing an IR capture (it's hidden in the firmware for possible future use). I've studied dozens upon dozens of them and the kernels above first order (the first order kernel is the linear IR) are so small as to be inaudible. The distortion from an amp is orders of magnitude greater even when using a clean amp.

I certainly appreciate your response and overall I do understand where you are going with your reply. I do not doubt the amount of time you've put into this, along with the rest of your research for this machine we all love. But I'm not afraid to admit that I'm certainly no electrical engineer and have no idea exactly what most of that means and that the only kernels I'm aware of are the ones that turn into popcorn. I'm always open to learning more though so I'll take away what I can.

It sounds like they're synthesizing an octave-divided note and calling that "cone cry" (we can do that with the Pitch block).

I definitely agree with this and thats the first thing I thought of when I actual heard it in the video, it sounds like a synthesized note just added in there, unnaturally.

I have never heard of the term cone cry and only assumed they made it up when I first heard the term.

Extremely close is quite an understatement. You can shoot an IR many different ways but if you do it accurately there's no one betting money in A/B comparisons. I work with real cabs and IR's multiple times a week and sure don't "feel like its not as good as it could be sometimes". Have you been in a studio control room and compared the same microphone position to an IR shot of it? I highly recommend people doubting IR technology to experience this.

To be clear, the whole point of this thread was not to just bash IRs and doubt them overall, because I recognize their purpose and abilities and all that they add to what we do. I have no spent time in a studio comparing them, no. Certainly not as much as someone like yourself. I am only going off of my own personal experiences and what I have heard. This wasn't just to tout the UA Ox and to try and bash the Axe for not having these abilities but merely to bring it up as a topic of discussion because UA is not known for just putting out crap for the sake of sales and maybe they are onto something here. As I said, I don't know as much about IRs as I'd like so I'm making no hard claims here. I know this is the business of what you do along with many others and have a lot invested into it so it is not to put that down in any way. Its all in the sake of improvement.

But if you do have any references of good quality IR vs real speaker A/B comparisons, I'm all for listening. Some of the results online certainly vary.

Also in regards to "no one betting money in A/B comparisons", I don't doubt that either but people have been saying that about the Axe in general for years now and improvements continue to be made. Again, not doubting you or your experience but if every argument ended in "well no one can tell in A/B comparisons" then what else is there to be done? Heck just with the power of the new unit, Cliff has said there is an improvement in the modeling just because the same corners didn't have to be cut in the algorithms, so I am curious if that will benefit the speaker cab side of things as well.

In my measurements I achieved the same results as Cliff. As long as the speaker is driven within its specifications, the result remains virtually identical visually, differences from IRs that are not audible except for the loudness with which they have been captured.

I believe I understand the purpose of these graphs and understand how they are different, especially with Cliff explaining the first one and then one with the amp driven harder. But how to do these correlate to someone playing, as in if I am playing very softly on a clean patch versus chugging through a 5150? Does the speaker not react differently to these playing dynamics in the way that a static IR won't produce on the other end? Thats a genuine question, not just to argue with you.

Again, the point of this thread was out of curiosity and not to show the Ox and insult Fractal for not having all the technologies in their system. Heck, there is a 3 page thread on how dim the third bar on the new logo is o_O..so I figured this has to spark up some more useful conversation haha.
 
Really interesting discussion, but a question arises: could the method to capture IRs hide some phenomena that occurs only on some circumstances?

I mean, an IR is almost always captured with a sine sweep which is just a single frequency being reproduced by the speaker at a time, but this would not detect for example some irregular motion of the cone that might occur when the speaker tries to reproduce more complex sounds.
Could be interesting to compare IRs of the same cab+mic configuration captured using different test tones (white noise, pink noise, double sine sweep, music program) to see if the frequency response remains the same across all of them.

But maybe someone has already done this kind of tests or maybe my knowledge is too limited to talk about these things and my argument is just "bovine stools" :)
 
But if you do have any references of good quality IR vs real speaker A/B comparisons, I'm all for listening. Some of the results online certainly vary.
Where's the online results that vary? Here's one comparison:


I believe I understand the purpose of these graphs and understand how they are different, especially with Cliff explaining the first one and then one with the amp driven harder. But how to do these correlate to someone playing, as in if I am playing very softly on a clean patch versus chugging through a 5150? Does the speaker not react differently to these playing dynamics in the way that a static IR won't produce on the other end? Thats a genuine question, not just to argue with you.
I believe this tells you that the speaker responds linearly with changing volume/dynamics, whether you're playing on a clean patch or chugging through a 5150. Meaning, you get the same response, just different volume.
I think the term "speaker breakup" causes us guitarists to think that there's some significant overdrive, "dynamic" thing happening at the speaker level. But authoritative figures like Jay Mitchell and Celestion have let it known that the cone breakup modes result in a linear response, completely captured by IRs. Other non-linear parts are already modeled by the Axe-Fx like thermal and over-excursion as Cliff mentioned. And cone cry, naw cus it sounds nastay lol.
 
Back
Top Bottom