Do you think modelers will get there in the next 10 years?

Agreed, mate.

I felt it might be instructive to sceptics here not only for the tone but the volume-knob and pickup behaviour.

When an experienced dude like Guido can't hear or feel any difference, you know you're doing OK. He's no shill for those who may still be sceptical; I'm honoured to be able to call him a bud and he's a cool, honest dude.
What I have found is that an electric guitarist is not just a guitar player but also a technician. On some level, they have to get good at sound engineering if they want to master electric guitar. This means being good at putting together rigs that help them best present their playing. Sure they can get help with their guitar tech/roadie along the way but they got to know some things. I find that if a guitarist is seeing no difference today between profilers and the real deal then their skills are probably much more rounded in both areas than in just one.

Which would you rather be?
  • Good at tube rigs only.
  • Good at profilers only.
  • Good at both.
If people own tube amps then I would suggest using a profiler with them and have a bigger sound for it. If you end up sticking to only one profile that you don't adjust then maybe you might want to get the real deal to put money back into that company for developing that system.

Anyway, the idea that profilers are taking money from tube amp developers or any amp developer for that matter is a big question in my mind. I tend to think that it has allowed more guitarists to play and develop their skills enough to increase potential buyers for the original hardware. Are amp companies really saying profilers are stealing our sales? That we have seen a decrease in the number of amps we sell? If this is true, then it should be true across the board for all amp developers and not just some.
 
ear-diagram-6.jpg


Answer to that question would have to be based on everyone having the same subjective hearing which isn't physically or scientifically possible.
 
What I have found is that an electric guitarist is not just a guitar player but also a technician. On some level, they have to get good at sound engineering if they want to master electric guitar.

You are WAY over-thinking this.

Let's try to simplify things:

1. Modelers CAN replicate tube amps.
2. Some people prefer digital amps
3. Some people prefer tube amps.
4. YOU think there is no difference between the two.
5. I, and others disagree with you.
6. You have trouble with folks who disagree with you.
7. You need to work on that.

Psssst: I got my acoustic engineering degree using one of these.
amp.jpg
 
You are WAY over-thinking this.

Let's try to simplify things:

1. Modelers CAN replicate tube amps.
2. Some people prefer digital amps
3. Some people prefer tube amps.
4. YOU think there is no difference between the two.
5. I, and others disagree with you.
6. You have trouble with folks who disagree with you.
7. You need to work on that.

Psssst: I got my acoustic engineering degree using one of these.
Overthinking something is when a subjective bias is pushing to favor one over the other. That is why those claims fail the Pepsi challenge. I don't overthink what I am hearing. That is why they are the same.

Why is it so hard to understand that if you make the claim that profiler hardware is lacking something or produces errors (like your digital artifacts claim) that to be objective you actually need to identify exactly what that is and why it can't do that?

If you are now appealing to an engineering degree then all you are telling me is that you should know this already.

All you want me to do is just join a side that wants to claim that profilers aren't there yet, are missing something, but can't identify what that is. I don't want to join a side that is trying to make an objective claim that turns out to be congruent with subjective causation. This is really what the Pepsi challenges are pointing at. That if you spend enough time working with profilers you can indeed make them sound no different to a tube amp.
 
I don't overthink what I am hearing. That is why they are the same.
Which is cool.
I didn't overthink my choice when comparing the two either.
Why is it so hard to understand that if you make the claim that profiler hardware is lacking something or produces errors (like your digital artifacts claim) that to be objective you actually need to identify exactly what that is and why it can't do that?
I don't actually have to have to do anything other than pronounce that I auditioned both side by side (extensively in my own home music room) and found that I prefer tube amps over modelers.

Why you are having such a hard time accepting others' opinions is just plain weird.
 
Last edited:
He's admitting that Pepsi tried real hard to emulate Coke but failed? Or that Pepsi and Coke are indistinguishable? Because not only can I tell them apart but actually I prefer Pepsi.....unless im eating chinese food, then it must be a cold can of coke. Anything else, Pepsi.
 
I don't actually have to have to do anything other than pronounce that I auditioned both side by side (extensively in my own home music room) and found that I prefer tube amps over modelers.

Why you are having such a hard time accepting others' opinions is just plain weird.
I have always accepted subjective opinions.

What I don't accept is when you make objective claims like this one. I have already pointed this out to you. So let's dispense with the false analogies about me addressing your subjective opinions. You claimed the hardware lacked processing power. Not just with some modelers... but all.

Digital aliasing is still a problem to varying degrees with every commercially available modeler/profiler.
The processing hardware required to do an optimum job is simply not cost effective for any production unit.

Doesn't matter how good the sound engineer/programmer if he's working with limitations in processing
due to costs.
Those aren't subjective opinions. You were claiming the hardware isn't good enough. That they lack processing power.

I pointed out that what was causing artifacts was stacking pointlessly high sample rates like 96 kHz. This does not provide any benefit to the audio quality, compared to 44.1 kHz.

Profilers have blocks and the block totals can't exceed the processing power limits of the rig. The software will cut you off from using more than the system can handle. Some people found a bug by stacking 96 kHz blocks. It caused lagging and the fidelity of the system was compromised in various ways. It was corrected with new patches.

None of that added up to any limitations in processing for optimal performance across every commercially available modeler/profiler as you claimed.

If you want to strawman my position then have it but it's been burning since the last time you tried that. Trying to tell me what I am saying, instead of just addressing what I am saying. If I want you to be my spokesperson then I'll ask you.
 
I pointed out that what was causing artifacts was stacking pointlessly high sample rates like 96 kHz. This does not provide any benefit to the audio quality, compared to 44.1 kHz.
Since you're all about science and objectivity, allow me to explain to you why 96kHz is in fact not just ALL marketing bs.
I apologize in advance if you already know this, and it's a bunch of mansplainin' to you. :giggle:

In general, we've defined the audible audio range roughly between 20Hz and 20kHz, right?
Typical red book audio CD quality is encoded in 16 bit, 44.1kHz. The 44.1kHz has to do with doubling the 'allowed maximum frequency', in this case ~22kHz, AKA the 'Nyquist frequency'.
You would think 'no problem, 22kHz is already outside the audible range, no harm, no foul'.
Here's the kicker. In order to do the digital to analog translation, one needs a sloped anti-aliasing low-pass filter. And since you want your audible signal to extend to at least 20kHz, this filter needs to be fairly steep, to do roughly -96dB (max dynamic range at 16 bit) within about 2kHz.
And mind you, this 2kHz portion isn't even an octave or something in musical terms, but just a portion of it.
Now... in an ideal world, such a filter would leave the audio untouched below 20kHz and then start with its slope. However, such filters do not live in an ideal world. So they introduce artifacts in the frequency response.
And the steeper you want your filter to be, the heavier these artifacts. Which means in our 44.1kHz situation, that in say the range of 14kHz to 20kHz (or maybe 16kHz to 20kHz), this will mess with your audio AND be inside the audible range.

Now if you extend your sampling frequency to 96kHz, it means your Nyquist frequency resides at 48kHz, and suddenly you have the option to use a more gentle anti-aliasing filter, that can do its slope not over a mere 2kHz, but over ~28kHz (from 20kHz to 48kHz). Or, it can still be steep, but start later, from say 40kHz to 48kHz. Mind you, it's still 4 times less steep AND it starts at DOUBLE the maximum audible frequency, meaning any artifacts it will then introduce under 40kHz, will be safely away from the audible range.

Now whether hardware can deal with such high sample rates, frequencies, and whether it's beneficial for reproducing a guitar's signal, which is already limited by itself, save the harmonics, is a whole different discussion.

But I just wanted to present some objective reasoning why 96kHz exists.
A very steep filter that doesn't muck up your audible signal isn't easy to design, thus costly.

More info:
https://en.wikipedia.org/wiki/44,100_Hz
https://en.wikipedia.org/wiki/Nyquist_frequency
 
Serious question: how does Axe/Kemper perform for a variety of rolled back volume tones? As some may or may not know, with a sufficiently dynamic tube amp you can get a myriad of tones in real time by simply manipulating the guitar's volume knob and altering one's touch on the strings. Do the modelers translate this somewhat, or at all? Not to mention being able to generate different pitches of harmonic feedback by changing one's position to the cabinet. Do the modelers allow one to interact within the amp's feedback loop in this way?

In my experience (which is just my AxeFX III, I’ve never owned another modeler but have used a bunch of amp sim plug-ins) it really depends on the amp and how much gain you’ve got on it. I mean, cranking the shit out of the gain on a Dual Rec dialed in for a tight, chugging metal tone won’t give you the sweetest tone when rolled back in real life, so it shouldn’t be expected to happen in a modeler.

Last night I was working on my Jeffrey Weidlant preset and got some really cool tones when rolling down the volume and switching pickups. That real tube-y, almost Strat kind of sound. My Friedman presets get used for this stuff all the time as well and I often layer stuff and only roll the volume down/switch pickups (I have Fishman’s in one guitar, switching between Modern and Classic is a pretty big difference and works great for layering shit).

My tube amp experience mainly falls under Mesa/Peavey heads (Dual Recs/MKIV/5150/XXX) and the AxeFX is pretty damn spot on with the volume knob interaction I had with those amps. I’m not sure I’d have the same love for the AxeFX if it didn’t do that as I’m constantly fidgeting with the volume knob in a live setting.

Feedback isn’t any different, it’s still audio feeding back into your pickups. If you’re loud enough, you’ll get it. Not sure how it is for the FRFR guys, but plugging into a power amp/cabinet feels like playing a regular amp to me in every sense. I’ve certainly got a lot more experience playing through a half-stack than I do a modeler as I’ve only had my AxeFX for just over a year, opposed to the 25 years I spent playing through tube amps.
 
It's these experiences that teach the player that they HAVE to learn to use their pickup volumes..very quickly or else you'll lose control of your rig lol.

Actually, that ties into what I was saying in my previous post in regards to how modelers interact with the volume knob. I said I‘m always fidgeting with the volume knob in a live setting (haven’t played live with the AxeFX yet) and it was because exactly that, not wanting to lose control of the rig, but in more ways than just feedback. My first head was a Carvin Legacy head and Vai was really adamant about using the volume knob to dial back the gain. He didn’t even use a footswitch to switch to a clean sound for years, he’d just roll the volume back. Back then, I wasn’t even using a boost for leads, I’d just roll my volume up and have all the volume/distortion I needed, then roll it back down a bit for rhythms.

Then when I switched the Mesa’s, it was the same deal only a bit more refined; I had specific lead channels I’d switch to, but even for the rhythm stuff, I’d ride the volume knob throughout the song and use it dynamically to add in more distortion (you’re not getting a big volume drop through a cranked Dual Rec half stack by turning the volume knob from 10 to 7) as the song progressed.

Still doing that same thing now with the AxeFX, but since I don’t record at gig volume, feedback isn’t the concern, it’s just tweaking the amount of gain/distortion. Same way I’ll ride a fader to automate a vocal or compressor setting.
 
Since you're all about science and objectivity, allow me to explain to you why 96kHz is in fact not just ALL marketing bs.
I apologize in advance if you already know this, and it's a bunch of mansplainin' to you. :giggle:

In general, we've defined the audible audio range roughly between 20Hz and 20kHz, right?
Typical red book audio CD quality is encoded in 16 bit, 44.1kHz. The 44.1kHz has to do with doubling the 'allowed maximum frequency', in this case ~22kHz, AKA the 'Nyquist frequency'.
You would think 'no problem, 22kHz is already outside the audible range, no harm, no foul'.
Here's the kicker. In order to do the digital to analog translation, one needs a sloped anti-aliasing low-pass filter. And since you want your audible signal to extend to at least 20kHz, this filter needs to be fairly steep, to do roughly -96dB (max dynamic range at 16 bit) within about 2kHz.
And mind you, this 2kHz portion isn't even an octave or something in musical terms, but just a portion of it.
Now... in an ideal world, such a filter would leave the audio untouched below 20kHz and then start with its slope. However, such filters do not live in an ideal world. So they introduce artifacts in the frequency response.
And the steeper you want your filter to be, the heavier these artifacts. Which means in our 44.1kHz situation, that in say the range of 14kHz to 20kHz (or maybe 16kHz to 20kHz), this will mess with your audio AND be inside the audible range.

Now if you extend your sampling frequency to 96kHz, it means your Nyquist frequency resides at 48kHz, and suddenly you have the option to use a more gentle anti-aliasing filter, that can do its slope not over a mere 2kHz, but over ~28kHz (from 20kHz to 48kHz). Or, it can still be steep, but start later, from say 40kHz to 48kHz. Mind you, it's still 4 times less steep AND it starts at DOUBLE the maximum audible frequency, meaning any artifacts it will then introduce under 40kHz, will be safely away from the audible range.

Now whether hardware can deal with such high sample rates, frequencies, and whether it's beneficial for reproducing a guitar's signal, which is already limited by itself, save the harmonics, is a whole different discussion.

But I just wanted to present some objective reasoning why 96kHz exists.
A very steep filter that doesn't muck up your audible signal isn't easy to design, thus costly.

More info:
https://en.wikipedia.org/wiki/44,100_Hz
https://en.wikipedia.org/wiki/Nyquist_frequency

I felt like I was trying to run up a steep, sandy hill when reading that. Took me a few starts and I had to exert more power than usual to accomplish it, but I got through it. Thanks for that, it’s rare someone goes into detail on that stuff and I actually take it all in. Excellent post!
 
I have a JVM 410 HJS and an Axe FX III. I connected the JVM and Axe FX using 4CM. Due to the plethora of connections on the Axe III, I also put a XiTone MBritt on Output 1. I can create presets that mix and match components. I can use amp modeling and send it to my 1960 AHW or I can use the Marshall pre-amp and send it to the XiTone. I can quickly and effortlessly move between different configurations by switching a preset.

I spend a lot of time flipping back and forth through presets that use all the possible combinations of components available to me. Running a Axe III model through the tube power section of my JVM to the 1960 AHW gets some great tones. But if I dial up a JVM HJS Model and send it to my 4x12 it is going to sound different. There is aliasing and the compression is different. Not bad. Just different. It actually sound pretty bad ass. It is clearly there when you can strike a chord, let it sustain and flip between presets to A/B the differences.

The XiTone has a solid state power amp and 1x12. I would say that this is where I notice the biggest difference. Running a model through the tubes of the JVM power section to a 4x12 gives a completely different feel than an IR, solid state amp, and a 1x12.

I am adept with tube amps. I am adept with modelers. I chose both.

Looking forward to Axe FX FW 16 getting out of beta to compare this latest update. The ring leaders on the fractal forum claim it is a significant advancement.
 
I have a JVM 410 HJS and an Axe FX III. I connected the JVM and Axe FX using 4CM. Due to the plethora of connections on the Axe III, I also put a XiTone MBritt on Output 1. I can create presets that mix and match components. I can use amp modeling and send it to my 1960 AHW or I can use the Marshall pre-amp and send it to the XiTone. I can quickly and effortlessly move between different configurations by switching a preset.

I spend a lot of time flipping back and forth through presets that use all the possible combinations of components available to me. Running a Axe III model through the tube power section of my JVM to the 1960 AHW gets some great tones. But if I dial up a JVM HJS Model and send it to my 4x12 it is going to sound different. There is aliasing and the compression is different. Not bad. Just different. It actually sound pretty bad ass. It is clearly there when you can strike a chord, let it sustain and flip between presets to A/B the differences.

The XiTone has a solid state power amp and 1x12. I would say that this is where I notice the biggest difference. Running a model through the tubes of the JVM power section to a 4x12 gives a completely different feel than an IR, solid state amp, and a 1x12.

I am adept with tube amps. I am adept with modelers. I chose both.

Looking forward to Axe FX FW 16 getting out of beta to compare this latest update. The ring leaders on the fractal forum claim it is a significant advancement.

I ended up with two AFXIII units and had one with 15.1 and the other with 16 beta. I definitely had to redo the gain and master volume settings, but I didn't find the differences to be earth shattering. In fact I didn't like the 16 beta at first but I stuck with it after figuring out that it seemed to need higher gain and master settings to get the sounds I wanted from it.
 
I have a JVM 410 HJS and an Axe FX III. I connected the JVM and Axe FX using 4CM. Due to the plethora of connections on the Axe III, I also put a XiTone MBritt on Output 1. I can create presets that mix and match components. I can use amp modeling and send it to my 1960 AHW or I can use the Marshall pre-amp and send it to the XiTone. I can quickly and effortlessly move between different configurations by switching a preset.

I spend a lot of time flipping back and forth through presets that use all the possible combinations of components available to me. Running a Axe III model through the tube power section of my JVM to the 1960 AHW gets some great tones. But if I dial up a JVM HJS Model and send it to my 4x12 it is going to sound different. There is aliasing and the compression is different. Not bad. Just different. It actually sound pretty bad ass. It is clearly there when you can strike a chord, let it sustain and flip between presets to A/B the differences.

The XiTone has a solid state power amp and 1x12. I would say that this is where I notice the biggest difference. Running a model through the tubes of the JVM power section to a 4x12 gives a completely different feel than an IR, solid state amp, and a 1x12.

I am adept with tube amps. I am adept with modelers. I chose both.

Looking forward to Axe FX FW 16 getting out of beta to compare this latest update. The ring leaders on the fractal forum claim it is a significant advancement.
I also have tube amps and an AXE FX III. They both have their place in my arsenal. The versatility of the AXE FX III in a cover band when a I need big fat crystal cleans to heavily overdriven tones is perfect for my needs.

Patiently waiting on the Cygnus FW 16 update!!!

I run my AXE FX III in to a Fryette LX II and out to traditional guitar cabs. I think that combo sounds best.

I have tried solid state power amps and FRFR speakers.
 
Back
Top