The Digital Doubt

[Sascha Franck]

Maybe a bad comparison, but imagine a replica painting on a wall next to an original. From a distance they may look the same, but the closer you get you start to see the color blends are not the same. Then you touch each "painting" and realize one has irregular texture and the other is smoother.

Yes, a bad comparison - simply because that kind of replica wouldn't even be close to component modeling. If that replica of the painting was component modeled, it'd use exactly the same materials (or at least some that'd feel exactly the same), so when touching it, the texture would be every bit the same, too.
Maybe google Wolfgang and Helene Beltracchi. They didn't copy famous paintings but did their own "in the style of" variations - and even some of the most acknolwedged experts had serious problems to detect they were fakes. One of those experts said that the painting he's seen in Beltracchi's house was one of the most beautiful works of Max Ernst ever and his widow agreed as well.

 

[Sascha Franck]

This is how it feels to me from experience.

In this discussion about compression, nothing is about feel. You can measure compression. Go ahead.

 

[Eagle]

Untill I see vids of people consequently identifying digital version from real ones (never seen one, and a bazillion showing the opposite)…and I’m diagnosed with earloss so my own observations no longer count (I can’t tell the difference between most captures and real through the same “make it louder gear”)…I’ll be navigating on the prefect that digital delivers on its promise.

You tell easily only if you’re the player. Literally the only time I ever had even a doubt was a power kemper with models made by me through the same cab without moving it. Volume matched and switched by a third party out of my sight.

 

[Boudoir Guitar]

A physical model is a physical model. Which, by definition, would allow them to be exactly the same.

I don't understand this. It is a digital model of an analog system and by definition a model is not the thing it is a model of -- that would be a copy, not a model.

The analog system is one that is highly predictable, though, so it is ripe for a very very accurate model to be made of it.

At the end of the day, I will go back to what I said in the other kind of similar thread. If a person has doubts about digital, the only way for them to resolve those doubts is to play a bunch of stuff side by side.

And to appreciate that it is impossible to separate out the physical and emotional experience one has, nor should one try to do so, really. I had a grand ol' time Friday playing my '56 Gibson guitar through a '52 Gibson amp, despite the amp being in an obvious need of a tune-up. Because '50s Gibson vibe for days. I wasn't sitting there thinking "Hmmm, I wonder if I can recreate this with my Axe Fx" because my guess is the sound wasn't actually all that great. But the experience was sublime.

 

[FuzzyAce]

So… the dynamics??

I think it's more than dynamics, that term could be taken different ways too I suppose. I think of dynamics maybe more as headroom thing, where there's room in an amp (or model) to fluctuate based on things like pick attack.
A tube amp is an electrical circuit where things are happening on different levels at different times, with real voltages hitting real transformers then into speakers. With a lot of amps (not all of course, depends on the design and output) when you hit it harder with pick attack, it responds in a dynamic way that I've yet to truly experience closely from a model. Some are better than others, and granted I haven't tried every modeler either. It's just an observation, not me saying a tube amp is better for every situation or player. It is a different experience though.

 

[Boudoir Guitar]

This is how it feels to me from experience. There will be a perfectly logical explanation for it.

I'm not going to argue with you about your experience -- I'm not doubting that at all. But your explanations for your experiences with digital gear are far, far, far from logical.

 

[Eagle]

Yes, a bad comparison - simply because that kind of replica wouldn't even be close to component modeling. If that replica of the painting was component modeled, it'd use exactly the same materials (or at least some that'd feel exactly the same), so when touching it, the texture would be every bit the same, too.
Maybe google Wolfgang and Helene Beltracchi. They didn't copy famous paintings but did their own "in the style of" variations - and even some of the most acknolwedged experts had serious problems to detect they were fakes. One of those experts said that the painting he's seen in Beltracchi's house was one of the most beautiful works of Max Ernst ever and his widow agreed as well.

But it is impossible to exactly replicate an existing work . No component modelling is more than a sample of the key elements. The aspects left will still have an effect however small. Multiply that by the number of components and its not irrelevant.

 

[mercifulfuzziness]

I think it's more than dynamics, that term could be taken different ways too I suppose. I think of dynamics maybe more as headroom thing, where there's room in an amp (or model) to fluctuate based on things like pick attack.
A tube amp is an electrical circuit where things are happening on different levels at different times, with real voltages hitting real transformers then into speakers. With a lot of amps (not all of course, depends on the design and output) when you hit it harder with pick attack, it responds in a dynamic way that I've yet to truly experience closely from a model. Some are better than others, and granted I haven't tried every modeler either. It's just an observation, not me saying a tube amp is better for every situation or player. It is a different experience though.

A bit like the Super reverb compared to the JMT45 in the helix?

It seems like the SR is a bit more forgiving? Where the JMT45 lets everything come through whatever you do on it?

 

[Eagle]

I'm not going to argue with you about your experience -- I'm not doubting that at all. But your explanations for your experiences with digital gear are far, far, far from logical.

Maybe I am not describing it correctly but it feels smaller, lacking in interactions that make it feel compressed. Have you ever played an amp that feels compressed but doesn’t really sound it?

 

[AFKAEjay(retired)]

You tell easily only if you’re the player. Literally the only time I ever had even a doubt was a power kemper with models made by me through the same cab without moving it. Volume matched and switched by a third party out of my sight.

There you added a powersection, even though SS…that’s not equal comparisson.
And the observation remains…I’ve never seen anything where someone consequentially identifies digital.
And…when I capture something…I can’t tell the difference…IF I keep the components equally…and only replace analog ones by digital ones.
Try profiling a preamp and A/B through the same poweramp and cab.

 

[FuzzyAce]

Maybe google Wolfgang and Helene Beltracchi. They didn't copy famous paintings but did their own "in the style of" variations - and even some of the most acknolwedged experts had serious problems to detect they were fakes. One of those experts said that the painting he's seen in Beltracchi's house was one of the most beautiful works of Max Ernst ever and his widow agreed as well.

Not a good comparison either, because those are still actual paintings and not "prints".

 

[Sascha Franck]

I don't understand this. It is a digital model of an analog system and by definition a model is not the thing it is a model of -- that would be a copy, not a model.

Whatever, I think that's semantics (which I rather not get into, given that english isn't my native language). So, of course, a model is *not* exactly the same, but it's supposed to deliver exactly the same functionality.

The analog system is one that is highly predictable, though, so it is ripe for a very very accurate model to be made of it.

That's pretty much all I wanted to say anyway. There's not too much voodoo in amps. Whether things have been nailed yet (or haven't, for the matter) could obviously be debated, but at least on paper, it should be absolutely doable to create a model of an amp that could not be distinguished from the real deal at all anymore (apart from there being some latency, which is something that we may never be able to completely solve - but that's certainly not what people are getting in arms about).

 

[Sascha Franck]

Not a good comparison either, because those are still actual paintings and not "prints".

Of course. But they could be.

 

[Sascha Franck]

But it is impossible to exactly replicate an existing work

It is very possible in the audio world, at least as soon as the input signal is the same.

 

[Eagle]

There you added a powersection, even though SS…that’s not equal comparisson.
And the observation remains…I’ve never seen anything where someone consequentially identifies digital.
And…when I capture something…I can’t tell the difference…IF I keep the components equally…and only replace analog ones by digital ones.
Try profiling a preamp and A/B through the same poweramp and cab.

I’ve done this too with my VH4 and Axe VH4 back through the power stage and same cab. Easily identifiable by feel.

 

[FuzzyAce]

A bit like the Super reverb compared to the JMT45 in the helix?

It seems like the SR is a bit more forgiving? Where the JMT45 lets everything come through whatever you do on it?

I wouldn't say one is necessarily more forgiving then the other, that would depend on the player, but they are definitely different in harmonic content and gain structure.

 

[Eagle]

It is very possible in the audio world, at least as soon as the input signal is the same.

You don’t accept the digital and analog are inherently different? Putting something through ADDA is loosing information and modelling even more.

 

[FuzzyAce]

Of course. But they could be.

Like I said, I'm ok with modeling being different. I think it's awesome in a way because new sounds are being created because of the technology.

 

[Boudoir Guitar]

You don’t accept the digital and analog are inherently different? Putting something through ADDA is loosing information and modelling even more.

To be frank, an electric guitar, as far as signals go, is not very precious or special. ADDA can handle it just fine.

I do accept that the experience of someone who approaches the experience with an assumption about what the ADDA process is doing to their guitar signal MIGHT have the experience colored by that assumption.

 

[AFKAEjay(retired)]

I’ve done this too with my VH4 and Axe VH4 back through the power stage and same cab. Easily identifiable by feel.

Still…my own observations…and every blind test I can find online..give different conclusions. So there’s no soul behind this keyboard you can win for the dark side