Has the Axe-FX II Achieved 99% amp modeling accuracy?

I'm someone who spends a lot of time in the studio with real amps, cabinets and the Axe-Fx. You guys seem worried about something but I'm not sure how many of you have actually done this A/B test in real life. Check this video I made a loooooooong time ago:



Same settings on the Axe-Fx amp sim and the real life Mark IV. An IR shot from that same mic-up used for the real Mark IV clip.

"Is it 100% there?" is not the real question. I'm sure if you did this test with two different Mark IVs you would be getting the same kind of difference as no two heads sound 100% alike either. The real question is "does the Axe-Fx sound just as real and authentic as the Mark IV?" to which the answer is hellls yeeessss it doeses!!1 It's not a matter of opinion. It can be measured and I've done it many times with Orange, Marshall, EVH and Mesa amps.

What the authencity comes down to is the IR's and IR shoot methods. This is why everyone is raving about IR's. You may have the best modeler out there but if you're not using authentic raw unprocessed clean IR's then that's likelier to make you sound unauthentic than something being wrong with the Axe-Fx. You can find authentic stock IR's also if you want to try it out.
 
ML SOUND LAB said:
I'm someone who spends a lot of time in the studio with real amps, cabinets and the Axe-Fx. You guys seem worried about something but I'm not sure how many of you have actually done this A/B test in real life. Check this video I made a loooooooong time ago:



Same settings on the Axe-Fx amp sim and the real life Mark IV. An IR shot from that same mic-up used for the real Mark IV clip.

"Is it 100% there?" is not the real question. I'm sure if you did this test with two different Mark IVs you would be getting the same kind of difference as no two heads sound 100% alike either. The real question is "does the Axe-Fx sound just as real and authentic as the Mark IV?" to which the answer is hellls yeeessss it doeses!!1 It's not a matter of opinion. It can be measured and I've done it many times with Orange, Marshall, EVH and Mesa amps.

What the authencity comes down to is the IR's and IR shoot methods. This is why everyone is raving about IR's. You may have the best modeler out there but if you're not using authentic raw unprocessed clean IR's then that's likelier to make you sound unauthentic than something being wrong with the Axe-Fx. You can find authentic stock IR's also if you want to try it out.
Click to expand...

ML SOUND LAB said:
The real question is "does the Axe-Fx sound just as real and authentic as the Mark IV?" to which the answer is hellls yeeessss it doeses!!1 It's not a matter of opinion. It can be measured and I've done it many times with Orange, Marshall, EVH and Mesa amps.
Click to expand...

This is the heart of the question I was asking. My question was worded to intentionally ignore the IR part of the formula, since a good IR can be created by someone with the right mics and equipment to perform the capture. Obviously, it takes an extensive level of experience with mic placement and room treatment to get a killer IR. The IR block is essentially a very complex bandpass filter, but certainly not a trivial part of the signal chain.... as we all know. Maybe there is more to be had in the future, but I'm guessing the technology is now very mature. IMHO, creating an authentic amplifier modeling algorithm is where the hard work of genius and innovation lies. BTW-- I love your IR's Mikko.
 
I guess what we're talking about here is essentially what the "Speaker" tab in the amp block is doing and making that fit the IR and amp. However I fear it's not that simple. I think different amplifier and cabinet combinations vary the impedance in ways that essentially could be simulated but that would require the IR format file to include some information on the speaker resonance of the actual cab that the IR was shot from. (which by the way I have for all the cabs I've shot) Whether the end result is realistic or not.. I have no idea.
 
ML SOUND LAB said:
I guess what we're talking about here is essentially what the "Speaker" tab in the amp block is doing and making that fit the IR and amp. However I fear it's not that simple. I think different amplifier and cabinet combinations vary the impedance in ways that essentially could be simulated but that would require the IR format file to include some information on the speaker resonance of the actual cab that the IR was shot from. (which by the way I have for all the cabs I've shot) Whether the end result is realistic or not.. I have no idea.
Click to expand...

A separate file containing the impedance data that could be loaded into the amp block spkr page would be cool.
 
barhrecords said:
A separate file containing the impedance data that could be loaded into the amp block spkr page would be cool.
Click to expand...
The extra step could slow things down a lot. Just imagine you having 5 IR's from difference cabs and comparing them and always needing to load different speaker settings for each of them. I would rather have it so that I could choose the Axe-Fx to automatically use the information based on if the IR includes speaker tab information. Plug'n'play and all that. :)
 
ML SOUND LAB said:
The extra step could slow things down a lot. Just imagine you having 5 IR's from difference cabs and comparing them and always needing to load different speaker settings for each of them. I would rather have it so that I could choose the Axe-Fx to automatically use the information based on if the IR includes speaker tab information. Plug'n'play and all that. :)
Click to expand...

Agreed that would be a better workflow.
 
ML SOUND LAB said:
I guess what we're talking about here is essentially what the "Speaker" tab in the amp block is doing and making that fit the IR and amp. However I fear it's not that simple. I think different amplifier and cabinet combinations vary the impedance in ways that essentially could be simulated but that would require the IR format file to include some information on the speaker resonance of the actual cab that the IR was shot from. (which by the way I have for all the cabs I've shot) Whether the end result is realistic or not.. I have no idea.
Click to expand...
 
Maybe in addition to the "speaker" tab having the resonance and "Q" data as separate entries, it would be possible to add a drop-down menu with the most common speaker models as choices. The resonance and "Q" data would be loaded automatically based on the speaker model. All the user would need to know when loading an IR is that the cabinet contained V30's or EVM12L's, or greenbacks, or creambacks, or, etc. etc....... I'm not trying to make work for Cliff, but based on the many questions I've seen here about the "speaker" tab parameters, it might help a lot of us sort out the right settings.
 
LZRBeam said:
Maybe in addition to the "speaker" tab having the resonance and "Q" data as separate entries, it would be possible to add a drop-down menu with the most common speaker models as choices. The resonance and "Q" data would be loaded automatically based on the speaker model. All the user would need to know when loading an IR is that the cabinet contained V30's or EVM12L's, or greenbacks, or creambacks, or, etc. etc....... I'm not trying to make work for Cliff, but based on the many questions I've seen here about the "speaker" tab parameters, it might help a lot of us sort out the right settings.
Click to expand...
The problem is that the 'correct' values for Q or LFR are not solely dependent on the speaker(s); so your suggestion is essentially no more accurate than what the AxeFX does now; which to have those values set for typical application/type. Go back and reread Cliff's posts on this topic regarding the physical effect that is being accounted for here; Jay Mitchell has also posted on this here and elsewhere a number of times... you guys are really making a way too much of this entire topic, IMHO. This is a 'tuning' parameter not a key tonal component. Get the core components properly in place and then adjust this by ear and feel. Sure it would be great to have a measurement taken to have this fully plug and play for 'accuracy' but this entire discussion seems a bit like navel gazing to me. In some ways being able to manually correct this and not have to live with a cabinet that has a funky resonance value is an example of something that makes AxeFX > Reality.
 
I agree with @zenaxe that we're making a huge deal about this. We're once again assuming that "the real thing sounds better because it's real" but in real life you're stuck with that one impedance setting which is why some amp+cab combinations sound bad in real life.

The impedance difference is IMO most noticable when palm muting and even then I set things differently depending on guitar tuning etc. which is something I can't do with real amps.
 
zenaxe said:
This is a 'tuning' parameter not a key tonal component. Get the core components properly in place and then adjust this by ear and feel.
Click to expand...

It sort of matters more when I plug my Matrix amp into a guitar cabinet. Other than that, in a live mix especially, it's almost irrelevant. And if it feels right, it's good, doesn't have to be "right". Come to think of it, I've never played into a perfect real tube amp+cab where I didn't want to change something.
 
ML SOUND LAB said:
Check this video I made a loooooooong time ago:
Click to expand...

People with golden ears immediately give lots of reasons why the real amp is superior, although mostly those reasons are quite poetic, like depth, tightness, whatever. I once took these samples from Soundcloud, added some random drum track, mixed up pieces of samples and asked a bunch of golden ears to tell which is which. Quite predictably, they failed. Until one of them remembered this video. After that, they could tell more reliably, of course. :)
 
Personally, I feel like EQ wise it has, but feel and response it hasn't. That being said, I really love the feel and response of the AxeFX II.

Clark's explanation makes a lot of sense as well. I'm curious what you think on the feel.
 
RKF said:
The number of people in a loud bar that will know that your modeled Plexi is only 99% there...

...Zero
Click to expand...


Agreed... number of people in a loud bar who give a shit about your tone in general ? ...zero
Guitarists - we are a self serving bunch !
 
I feel amp models are really really close to being spot on. If only the interaction between the transformers and a real cab you're using could be modeled somehow, we'd be at the 99%.

IRs though are a weak link in the chain. I've been recording for the past 3 months using a Mesa 4x12 Traditional cab, and the sound I get by miking it up is significantly different to the sound I get through an IR I shot of it using the built-in tool in the Axe.

I tried refining it using live Tonematch, to no avail. It is great for late night playing and silent recording, but layer tracks with the IR and you've got a painful mixing job to do, layer tracks with a real cab and they just smooth out.

Here's a comparison I shot last month:

https://www.dropbox.com/s/llmkfp7r1sgraf2/SandmanComp.mp3?dl=0

First part is real cab, second part is an IR I shot of the same exact cab through the same exact microphone, placed at the same spot, same mic pre, back into the AxeFx input 2. I'll let you be the judge

Getting an amp is on my plans before the end of the year, I'll make some further comparisons when I get it.
 
If we make a blindtest with "Amp A" and "Amp sim B" none of us would tell which one is the "real thing".

This conversation has been talked before and it always pops up again. So let's end it now.

Here is the right answer:
Axe FX II sounds 100% real. :)
 
Alex EShadow said:
I feel amp models are really really close to being spot on. If only the interaction between the transformers and a real cab you're using could be modeled somehow, we'd be at the 99%.

IRs though are a weak link in the chain. I've been recording for the past 3 months using a Mesa 4x12 Traditional cab, and the sound I get by miking it up is significantly different to the sound I get through an IR I shot of it using the built-in tool in the Axe.

I tried refining it using live Tonematch, to no avail. It is great for late night playing and silent recording, but layer tracks with the IR and you've got a painful mixing job to do, layer tracks with a real cab and they just smooth out.

Here's a comparison I shot last month:

https://www.dropbox.com/s/llmkfp7r1sgraf2/SandmanComp.mp3?dl=0

First part is real cab, second part is an IR I shot of the same exact cab through the same exact microphone, placed at the same spot, same mic pre, back into the AxeFx input 2. I'll let you be the judge

Getting an amp is on my plans before the end of the year, I'll make some further comparisons when I get it.
Click to expand...

In the comparison, Seems there was a gain difference at some point.... The second half just sounds louder.
 
You must log in or register to reply here.
Back
Top Bottom