How do we know the amp models are very accurate?

Because the "BRIT 800" in the AFX sounds just like both of my real JCM-800 2204 50W heads before I modded them lol...shrill and icepicky o_O...treble-peaking circuitry to the max....among other things.
My JCM-800 was modded to sound like a high gain Fender Bassman. Not on purpose but that was the result of messing with the tone shaping. If I remember it was only 3 caps that were changed.
 
How do we know the amp models are very accurate?

I think a more illuminating question will be: why do we care that they are very accurate? Not to ask in a facetious way, but I think it's important to understand. Especially when the measurements of the characteristics of the models are within the margins of error of the differences in actual hardware, then we can call them "accurate enough" and now we can get into the whole reason the amps and modelers exist in the first place, which is to get a guitar tone that fits the song, part, mood, whatever, and most importantly sounds good.

We want these models to be accurate because we like the sound of the actual physical amp being modeled. If the AxeFX can produce a guitar tone that we like just as much as a physical amp, for the same reasons, then what does it matter if it's 80% accurate, 90% accurate, 95% or 100% or whatever. Or even if it's a sound we literally cannot make with a physical amp. The end goal is a guitar tone. Get there however you desire, but focusing on whether or not it's identical to a physical amp is not seeing the forest for the trees.
 
Cliff made a statement recently that he uses measurements and doesnt only trust his ears.

Bottom line for me is that Fractal has been close enough to tubes without the shortcomings for years. Liking 18.03 firmware. The 15X firmwares had a sound I liked too.
 
I think a more illuminating question will be: why do we care that they are very accurate? Not to ask in a facetious way, but I think it's important to understand. Especially when the measurements of the characteristics of the models are within the margins of error of the differences in actual hardware, then we can call them "accurate enough" and now we can get into the whole reason the amps and modelers exist in the first place, which is to get a guitar tone that fits the song, part, mood, whatever, and most importantly sounds good.

We want these models to be accurate because we like the sound of the actual physical amp being modeled. If the AxeFX can produce a guitar tone that we like just as much as a physical amp, for the same reasons, then what does it matter if it's 80% accurate, 90% accurate, 95% or 100% or whatever. Or even if it's a sound we literally cannot make with a physical amp. The end goal is a guitar tone. Get there however you desire, but focusing on whether or not it's identical to a physical amp is not seeing the forest for the trees.

Yeah, once it is inside the range of tolerances for that amp model, it is 'close enough' to work with, and the refinements that follow are icing on the cake....

Code:
I never wanted to do this 
job in the first place!  I... 

I wanted to be...

A LUMBERJACK!

Leaping from tree to tree! 
As they float down the mighty 
rivers of British Columbia!
 
We want these models to be accurate because we like the sound of the actual physical amp being modeled. If the AxeFX can produce a guitar tone that we like just as much as a physical amp, for the same reasons, then what does it matter if it's 80% accurate, 90% accurate, 95% or 100% or whatever. Or even if it's a sound we literally cannot make with a physical amp. The end goal is a guitar tone. Get there however you desire, but focusing on whether or not it's identical to a physical amp is not seeing the forest for the trees.
I think it absolutely matters that amps are moving toward 100% - close is not good enough - which is what drives Fractal it seems. We can like many tones, sure, but if one is after a specific tone, from a specific real world amp that Axefx models, then 80/90% accurate is no where near close enough. This is born out again and again in threads where hairs are split on the specifics of given target tone by those who came to Fractal for the drive toward 100% accuracy on tones they know well from the real amp equivalent, or who have to reproduce those tones for those who know well - 80/90% close or an alternate sound that sounds great but is not the target tone won't due for many Fractal customers I suspect.
 
If you're gonna get quantitative about it, it will depend entirely on your scoring metrics and your statistical evaluation method of comparison.

Seriously though, when dealing with non-linear real world phenomena it is very normal to see some variance between models and real data. Not least because of the sparsity and noise in the measurements you are making.
 
which is what drives Fractal it seems
It drives Fractal because that's what customers are asking for.

He's said numerous times he'd prefer if the ideal FAS models were the only ones. But that's not what sells.

But you're right. Perfection is what drives Fractal's customers, and in turn they drive Fractal.

My personal opinion is that the AxeFX is good enough and has been for many, many years.

But I'm more than happy to reap the benefits gained from the perfectionists out there pushing Fractal to be even better.
 
This is super reverb preset and IR in the link below. I was testing a pedal i build that is plugged into the back of the axe fx unit, but never mind that here. To me, it sounds so like the real thing. Ofcourse you are NOT going to get the full room reverb feel through headphones, but if i was to put a microphone in front of the real amp OR use the axe fx 3 super reverb preset, both will come out in the videos, exactly the same, or very close. Thus, not all presets are created equal and satisfying, but a huge amount are pretty spot on. FRFR, or high quality head phones are great, but the final recording is where it is at. You can mostly nail a real amp with a mic in front of it when you capture audio. So all good. Its just so much easier to record using the axe fx unit than trying to perfect open air mic
 
Back
Top Bottom