(UPDATED for new FW!) EVH amp vs Axe Fx, QC and TMP...

[fazer]

guessing he’s basically saying “a stopped clock is right twice a day” - at certain (very different) settings you can make the Helix 2204 sound pretty close, but that doesn’t make it a good representation of that circuit.

I don’t think Line 6 have anything to cop out of on that one - it presumably sounds like their reference amp. The issue is with the reference amp rather than a mistake with the algorithm. If there was a “mistake” it’s that the circuit wasn’t verified/checked before including it, or that they purposefully chose it over other units despite it not being the right circuit

I believe Line 6 used a modded JCM800 so, no, they have nothing to cop out of as such. Your explanation was spot on though. I’m not a big fan of these sound-alike videos because they seem to want to give the impression certain models are perfect but sounding good at a particular setting doesn’t make the whole model a good representstion, as you say. And feel is important - you only get that from playing, not listening.

@GuitarJon I wasn’t trying to accuse you of anything here - I enjoy your videos! But it’s been an issue for a while now, especially on TGP rather than here: “Something wrong with the model? Wait - can you tell the difference in a blind test? No? Nothing can be wrong then!”

 

[GuitarJon]

In my opinion: if something sounds the same, it also usually feels the same! Latency aside of course. So if you can dial something in to sound and feel very close, there's not a lot more to it imho. I get it, you guys want more accurate models with the "stock" or "noon" positions but that's another discussion imho.

 

[SeeD]

“Something wrong with the model?

Yeah, something is wrong with the TMP model, and maybe it`s getting fixed in the near future. And that is a good thing

 

[laxu]

I’m not a big fan of these sound-alike videos because they seem to want to give the impression certain models are perfect but sounding good at a particular setting doesn’t make the whole model a good representstion, as you say. And feel is important - you only get that from playing, not listening.

GuitarJon's videos are IMO the right way to compare this stuff. Minimize the number of variables as much as possible by using the same DI tracks, the same cab sims, matched volumes etc. It's a comparison of "can these modelers sound like my preferred reference tone", not a comparison of whether any modeler is going to sound spot on at every setting possible.

I think you can evaluate the accuracy of those models vs the real amp to some degree by looking at the settings needed to reach the reference tone. For example when I had the QC, to make its JTM45 model sound and respond like the JTM45 in the Fractal FM3 I had at the time, I had to use some pretty weird settings whereas the Fractal model operated much closer to how I'd expect to configure a real JTM45.

"Feel" is thrown around more often nowadays because the audible differences between real amps and amp modelers have shrunk to very small. "Feel" is something that nobody can debunk because there's no way to perceive what you "feel" when playing a particular amp with the rest of your gear, your playing style etc.

I've seen some people on The Other Place adamantly say that e.g Universal Audio pedals "feel" better than a Fractal or Strymon, yet my own experience is that I have been able to make my Fractal sound and feel very much like some tube amps I've owned when used in the closest like-for-like signal chain. I certainly did not experience some superior "feel" in the UA Dream or Ruby I've tried, but I haven't put them head to head with what I own.

But I can't say "you are feeling it wrong" because I can't experience what these other people claim. It could be that I am less sensitive to "feel" than others, or it could be that those other people are just not as good at getting results that work for them out of the same devices I have.

"Feel" is the new "I have golden ears that can hear details you can't" because the only way to verify that would be to put these people in a blind test. Most people fail these, just like many people fail picking the real amp from modelers in GuitarJon's videos. Yet they confidently make claims like "this is the real amp because X and Y, tube amps are so superior" only for their choice to be one of the modelers. So in the end they just liked that sound better.

"Do I like how this device sounds/feels and its features" is a more constructive way to approach these things. If it sounds/feels good to play, it's good.

 

[the-trooper]

"Feel" is the new "I have golden ears that can hear details you can't" because the only way to verify that would be to put these people in a blind test. Most people fail these

 

[fazer]

In my opinion: if something sounds the same, it also usually feels the same! Latency aside of course. So if you can dial something in to sound and feel very close, there's not a lot more to it imho. I get it, you guys want more accurate models with the "stock" or "noon" positions but that's another discussion imho.

In the case of the Helix I would just like a ‘stock’ 2204, as the model name implies. A stock 2204 is more versatile than the modded version in Helix so, no, there is more to it than just being able to copy a sound at a particular setting.

And latency can, of course, be an issue and I guess I must disagree with you that something that sounds the same will also feel the same - it might, but in my experience it might not. Is it all down to latency? I don’t know - it’s possible a small amount of latency can just make something feel off when it isn’t possible for us to identify latency as the cause. Unless I have misunderstood something you record your base guitar output into a DAW for re-amping so I guess feel doesn’t enter into the equation for you, at least as far as making the comparison video goes.

The comparison videos are useful in showing that a model can, at least, sound indistinguishable from a real amp from the perspective of an external listener but they certainly don’t prove anything about how I might perceive the difference between playing and experimenting with the real amp versus the model for myself.

 

[Lysander]

The comparison videos are useful in showing that a model can, at least, sound indistinguishable from a real amp from the perspective of an external listener but they certainly don’t prove anything about how I might perceive the difference between playing and experimenting with the real amp versus the model for myself.

You turn knobs, and experience changes in sound. "Feel" cannot be quantified.

There's a reason why most players fail ABX tests of modelers vs amps, or why shootout videos are filled with confident "oh, X is definitely the amp, those transients give it away" comments... which end up being (hilariously, for me) wrong every time.

Same goes for latency, which is a metric people obsess over these days and, and is completely irrelevant for most of the top-tier products out in the market today. No, Steve Vai, you cannot sense 2ms.

 

[MirrorProfiles]

I get it, you guys want more accurate models with the "stock" or "noon" positions but that's another discussion imho.

To be clear, the discussion about Helix's 2204 model has nothing really to do with noon positions. Its the fact that the amp itself is not a 2204 circuit, its something else. I was able to make the Archon model sound VERY close to my Krank Revolution, but it doesn't make the Archon a model of a Krank. Try and match the Brit 2204 model with the gain lower, where the bright cap would be more in the circuit - it simply will not sound like the real amp without using extra processing.

I think you can evaluate the accuracy of those models vs the real amp to some degree by looking at the settings needed to reach the reference tone. For example when I had the QC, to make its JTM45 model sound and respond like the JTM45 in the Fractal FM3 I had at the time, I had to use some pretty weird settings whereas the Fractal model operated much closer to how I'd expect to configure a real JTM45.

All it takes is a different load for the amp, El34's vs KT66's (for instance), different bias/wall voltages etc and 2 amps can both be accurate even with different knob positions (and thats not taking into parts tolerances yet which is obviously also a big deal). Given the nature of how Neural DSP make amp sims, you can be pretty certain their models are accurate to the unit in their test conditions. I'm pretty certain the times I've struggled to make their models sound just like my amps has been down to that, the differences are in the same areas I'd expect.

Totally agree on feel, often my impression of one thing feeling better than another goes away when I have them matched in tone and volume. It becomes very hard to tell one over the other. The input impedance does have a big influence though, if a different input is used across different modellers, it can make one feel more crisp and up front and another more muffled/soft by comparison. There is so much more to feel though and its largely hard to quantify. Our brains are quite good at telling us when something is off, though.

 

[GuitarJon]

In the case of the Helix I would just like a ‘stock’ 2204, as the model name implies. A stock 2204 is more versatile than the modded version in Helix so, no, there is more to it than just being able to copy a sound at a particular setting.

And latency can, of course, be an issue and I guess I must disagree with you that something that sounds the same will also feel the same - it might, but in my experience it might not. Is it all down to latency? I don’t know - it’s possible a small amount of latency can just make something feel off when it isn’t possible for us to identify latency as the cause. Unless I have misunderstood something you record your base guitar output into a DAW for re-amping so I guess feel doesn’t enter into the equation for you, at least as far as making the comparison video goes.

The comparison videos are useful in showing that a model can, at least, sound indistinguishable from a real amp from the perspective of an external listener but they certainly don’t prove anything about how I might perceive the difference between playing and experimenting with the real amp versus the model for myself.

Subjective indeed. I can only speak for myself and how I experience the differences from my pov.

 

[JiveTurkey]

I feel like I've reentered a conversation from 2018

 

[FuzzyAce]

GuitarJon's videos are IMO the right way to compare this stuff. Minimize the number of variables as much as possible by using the same DI tracks, the same cab sims, matched volumes etc. It's a comparison of "can these modelers sound like my preferred reference tone", not a comparison of whether any modeler is going to sound spot on at every setting possible.

I think you can evaluate the accuracy of those models vs the real amp to some degree by looking at the settings needed to reach the reference tone. For example when I had the QC, to make its JTM45 model sound and respond like the JTM45 in the Fractal FM3 I had at the time, I had to use some pretty weird settings whereas the Fractal model operated much closer to how I'd expect to configure a real JTM45.

"Feel" is thrown around more often nowadays because the audible differences between real amps and amp modelers have shrunk to very small. "Feel" is something that nobody can debunk because there's no way to perceive what you "feel" when playing a particular amp with the rest of your gear, your playing style etc.

I've seen some people on The Other Place adamantly say that e.g Universal Audio pedals "feel" better than a Fractal or Strymon, yet my own experience is that I have been able to make my Fractal sound and feel very much like some tube amps I've owned when used in the closest like-for-like signal chain. I certainly did not experience some superior "feel" in the UA Dream or Ruby I've tried, but I haven't put them head to head with what I own.

But I can't say "you are feeling it wrong" because I can't experience what these other people claim. It could be that I am less sensitive to "feel" than others, or it could be that those other people are just not as good at getting results that work for them out of the same devices I have.

"Feel" is the new "I have golden ears that can hear details you can't" because the only way to verify that would be to put these people in a blind test. Most people fail these, just like many people fail picking the real amp from modelers in GuitarJon's videos. Yet they confidently make claims like "this is the real amp because X and Y, tube amps are so superior" only for their choice to be one of the modelers. So in the end they just liked that sound better.

"Do I like how this device sounds/feels and its features" is a more constructive way to approach these things. If it sounds/feels good to play, it's good.

To me, what Jon's and other videos reveal is that a majority of these different tone platforms (when listened back to back) are each satisfactory in a mix. But... an important part of inspiration is how the player feels a tone at the time of tracking. Thankfully re-amping a DI is so much easier and instantaneous through plugins and other means -- but they don't always feel right in the moment if playing live through them. So track with whatever feels right, run a separate DI track if necessary, and then change (or not) in post production. Playing out live is totally different, and many still use tube amps because they feel different and respond quicker imo, although modelers are quickly closing that gap.

 

[FuzzyAce]

I feel like I've reentered a conversation from 2018

 

[fazer]

GuitarJon's videos are IMO the right way to compare this stuff. Minimize the number of variables as much as possible by using the same DI tracks, the same cab sims, matched volumes etc. It's a comparison of "can these modelers sound like my preferred reference tone", not a comparison of whether any modeler is going to sound spot on at every setting possible.

I think you can evaluate the accuracy of those models vs the real amp to some degree by looking at the settings needed to reach the reference tone. For example when I had the QC, to make its JTM45 model sound and respond like the JTM45 in the Fractal FM3 I had at the time, I had to use some pretty weird settings whereas the Fractal model operated much closer to how I'd expect to configure a real JTM45.

"Feel" is thrown around more often nowadays because the audible differences between real amps and amp modelers have shrunk to very small. "Feel" is something that nobody can debunk because there's no way to perceive what you "feel" when playing a particular amp with the rest of your gear, your playing style etc.

I've seen some people on The Other Place adamantly say that e.g Universal Audio pedals "feel" better than a Fractal or Strymon, yet my own experience is that I have been able to make my Fractal sound and feel very much like some tube amps I've owned when used in the closest like-for-like signal chain. I certainly did not experience some superior "feel" in the UA Dream or Ruby I've tried, but I haven't put them head to head with what I own.

But I can't say "you are feeling it wrong" because I can't experience what these other people claim. It could be that I am less sensitive to "feel" than others, or it could be that those other people are just not as good at getting results that work for them out of the same devices I have.

"Feel" is the new "I have golden ears that can hear details you can't" because the only way to verify that would be to put these people in a blind test. Most people fail these, just like many people fail picking the real amp from modelers in GuitarJon's videos. Yet they confidently make claims like "this is the real amp because X and Y, tube amps are so superior" only for their choice to be one of the modelers. So in the end they just liked that sound better.

"Do I like how this device sounds/feels and its features" is a more constructive way to approach these things. If it sounds/feels good to play, it's good.

Feel is thrown around nowadays because it’s real, at least to some of us. Maybe it’s not for everyone. Yes, it’s easy to fool someone with a blind test. A real amp played in person has NO digital at all to enter into the equation. Even real amps tend to be played through IR’s for these tests, hence digitising them, then reproduced via a digital platform. If you are satisfied that these tests prove beyond doubt there is no difference between a particular model and the equivalent amp in all use cases that’s fine but I don’t agree.

 

[GuitarJon]

Feel is thrown around nowadays because it’s real, at least to some of us. Maybe it’s not for everyone. Yes, it’s easy to fool someone with a blind test. A real amp played in person has NO digital at all to enter into the equation. Even real amps tend to be played through IR’s for these tests, hence digitising them, then reproduced via a digital platform. If you are satisfied that these tests prove beyond doubt there is no difference between a particular model and the equivalent amp in all use cases that’s fine but I don’t agree.

It's fine to not agree. We can't all agree! Sometimes that happens :)

 

[Orvillain]

I do think feel is real. But it is hard to quantify and describe. Because ultimately, there is one variable you cannot really equalize and that is the player, their technique, and their style. Feel is the result of various micro interactions between all of the various articulations and ways of playing - legato, picking strength, figure pressure on the fretboard, vibrato, tremolo picking, hand position around the pickups area, etc.

The interaction between how the amp is designed to distort, and how you play to create the signal that hits the front of the amp... I think that is roughly where "feel" lives.

 

[gearJunkie]

Sag, compression, latency (digital & distance from monitoring source, sound pressure level, floor coupling, room dynamics, individual "sensitivities", etc. all contribute to "feel". Many of these contributors are outside the scope of any model and more dependent on the monitoring solution and environment.

Feel is a selfish performance and enjoyment factor important to the player, but irrelevant to the (external) listener. IMO it makes sense to focus on the listening (external) experience for these kind of comparisons to minimize immeasurable subjectivity.

In summary, feel is important but difficult if not impossible to measure/perceive by anyone other than the performer.

 

[laxu]

Feel is thrown around nowadays because it’s real, at least to some of us. Maybe it’s not for everyone. Yes, it’s easy to fool someone with a blind test. A real amp played in person has NO digital at all to enter into the equation. Even real amps tend to be played through IR’s for these tests, hence digitising them, then reproduced via a digital platform. If you are satisfied that these tests prove beyond doubt there is no difference between a particular model and the equivalent amp in all use cases that’s fine but I don’t agree.

I never said it wasn't real, I'm saying that it's a thing that is impossible to say "this is the way it is" because our perception might be different with no way to prove it. Just like I cannot know if I see a particular shade of the color green the exact same way as someone else (color blind people excluded), feel is hard to quantify and is heavily dependent on so many factors.

You can take any tube amp, add a compressor pedal in front and you have changed how it feels. But if the comp is subtle enough, to the outside listener it might sound the same as before while to the player it feels very different.

In a fully digital system, I can just move a virtual mic a bit and the perception of "feel" changes because the overall EQ profile changed. "Trebly and not a not a whole lot of low end" is perceived as "immediate and tight", "lots of bass with less treble" is "loose and smooth".

 

[Lysander]

In summary, feel is important but difficult if not impossible to measure/perceive by anyone other than the performer.

Honestly, i'd be less picky about this issue if people started using "preferences" instead of "feel". Because, yes, using hardware you prefer to use will make you perform better, but that has zilch to do with some unquantifiable magic sonic dust that won't show up in measurements / ABX tests.

 

[Orvillain]

some unquantifiable magic sonic dust

That is your strawman of what feel is. It isn't what people are saying feel is.

 

[Lysander]

That is your strawman of what feel is. It isn't what people are saying feel is.

Can you define "feel" for me?