• We would like to remind our members that this is a privately owned, run and supported forum. You are here at the invitation and discretion of the owners. As such, rules and standards of conduct will be applied that help keep this forum functioning as the owners desire. These include, but are not limited to, removing content and even access to the forum.

    Please give yourself a refresher on the forum rules you agreed to follow when you signed up.

How do we know the amp models are very accurate?


the amps I owned over the years; Blackface Fenders, Plexis (actual Marshalls and a Metro), my Mark IV that I still own, all these sound and most importantly feel right to me in the Axe FX.

I have not really ventured into the SLO yet, although I owned one. But it felt a bit too sterile at the time.

To me it's more about dynamics, character of break up and gain, reaction to playing technique and style, soft touch versus digging in and letting the tone of the guitar shine through. In those regards the Axe FX isn't lacking the tiniest bit in my experience.

(What convinced me personally to take the plunge and order one was reading an interview with Cliff and his Cliff notes on the amps. It became clear to me that he has a deep insight into amp design and periodically brings amps out of storage to compare to the model, fine tune, rinse, repeat. He came across as a dedicated tone searcher first, an engineer second and a businessman third.)


Fractal Fanatic
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.


So many great replies up above with very valid points.

But to your question, without actually sitting down and A/Bing everything, a source of comfort should be marketplace competition. Don't you think Fractal would be called out by multiple reviewers if they fell short? How many comparison videos can we find debating which modeler does a better job of emulating amp X? It'd be nuts to think they all were off the mark. No, instead what we find is that, by and large, you see the modelers are "tightening their shot group" with these algorithms.

In the end though, are you happy with it? I am personally blown away.


Fractal Fanatic
"Accuracy" has never meant that much to me, whether it's amps or effects. The only thing that matters to me is being able to emulate tones that I need or want and the Axe III has delivered every time. Most of these tones require specific amps or, probably more accurately, amp types and not once has there been any that hasn't been found in the Axe III.

To attempt to answer the question posed though, I believe the result of FAS' commitment to producing the best modeling platform is a connection between player and device that can only be accomplished by being as accurate as possible. The biggest testament to the accuracy, in my opinion, is that not once playing through the Axe III has it ever felt that I was playing through a modeler.


To me accuracy is about whether they respond like tube amps do in general. Most modellers might sound close in a recording, but don’t respond under the fingers like tubes do.

What tubes mean to me is touch sensitivity, response to the volume knob, letting the wood of my guitar change the character of the sound, that undefinable gorgeous timbre that way back converted me from a Boss GT-6 within 3 days of buying it.

Fractal has all this, and that’s all I want. In fact, it even accurately reproduces potentially undesirable things like ghost notes and idiosyncratic interactions between tone knobs on some amps. Personally I don’t care for that level of accuracy, and that is what the FAS models are for. I remember a while ago there was even a discussion about whether amps with weird volume tapers should be modelled as-is or with the taper standardised 😄


I’ve a/b’d the axe 3 with my real Tucana, and it’s pretty darn right on after accounting for variance in pot tapers, etc.

But really, with anything else, beats me, and I don’t really care. They generally sound great, so it’s all good.
I have done thorough A/B testing between Axe3 (Cygnus) and my triple Rectifier with the 4CM method, and came to the conclusion that the Fractal model actually sounded a little better 😮. The tiny difference there was between the 2, were far less than between different revisions of the same physical amp, which actually kind of surprised me.

Personally I don't think YouTube videos doing a/b comparisons, are worth much more than hinting at something. As an owner of many tube amps, and a former "tube amp snob" - well, I guess I still am in many ways, the focus for me is not only how it sounds, but very much also how it FEELS to play on an amp/digital modeller. After Cygnus, Fractal got it right, which is actually quite amazing to me! I now get almost the same joy from playing on my axe3 setup (Axe3 -> Marshall 9200 tube Power amp -> 4x12 Mesa & Engl cabs), as playing on my physical counterparts to the preset I have dialed in, the Engl Savage 120 + Mesa Mrk V 90 watt. First time I've ever experienced that with a modeller 😮 (have owned both the Axe Ultra and the Axe2, never got those to work satisfyingly).


Power User
Without being able to play the reference amps Fractal used to model theirs, there's no way to know how close they get. With component variance the same knob settings are not going to produce the exact same results on a different amp. Fractal is also known to sometimes use an idealized version of something so it works in a more intuitive way, an example being say sliders of graphic EQ on some of the Mesa models.

Sometimes the model can be actually more to your liking. One of my favorite models in the Fractal is the Cornford MK50 model. I have played the real amp and would never buy one because it was insanely loud so completely impractical. With the Fractal that's not a concern.

All I know is I have been able to dial tones out of Fractal units that can match my tube amps when run through an equivalent output system at the same volume as measured with a decibel meter. An equivalent output system would be tube amp -> loadbox -> cab sims -> studio monitors or modeler and tube amp -> Fryette Power Station -> guitar cab. I don't own any amps that are directly modeled in my FM3 so this has been "take my favorite setting on my tube amp and see if I can replicate it on the modeler" type testing.

I picked up a Strymon Iridium recently because I wanted something more compact and simpler than my FM3 to put on my pedalboard. Putting the two head to head with the same IRs I can tell that the Iridium is not as complex sounding, tends to be cleaner with a distinct midrange push whereas the FM3 is more like what I expect a tube amp to sound like. Does that make the Strymon bad? Absolutely not as I was still getting tone/feel that I enjoyed playing. In the end that's all that matters. We should leave accuracy for the developers and focus on just getting tones that we enjoy out of these things.


The only people that this truly matters to are the tube purists who look for any excuse to defend their personal choices. The internet is their soapbox.

I have a bunch of tube amps, and I don't care one bit if I can dial in my digital stuff to sound "exactly" like them. Just like I don't sweat the fact that my Deluxe Reverb doesn't sound anywhere close to my Boogies. There are certain sounds I am after, and I can get them from either platform.
Well, some people care about great accuracy because what they're after is a given tube amp tone, which not every amp or amp sim can offer. I do not think this is solely about "tube purists who look for any excuse to defend their personal choices". That seems way reductionist. And certainly Cliff cares a lot about accuracy provided the constant updates.

Geoff L

New Member
For people who don't really play tube amps and don't have access to play a Friedman, vintage Fenders and Marshalls, etc., how do we know that the amp models are extremely close to the amps that they're modeling? The amp models sound very good obviously, but do they sound almost dead on to the real deal?
Rarely do 2 tube amps of the same model sound identical. That can change with what cab you run it through, your lead or wireless, the guitar and pickups. The AXE allows you to modify your amp model without having to a) visit an amp technician and b) irreversibly damage your amp. At this point all the pluses out weigh the negatives.


Another thing to consider. Once you wander off the authentic controls page, you've thrown accuracy out the window, and entered modded amp territory. Certainly you don't have to do this, but the ability to tweak up the models to get them to do exactly what you want is what many of us are paying Fractal to be able to do.


Power User
Funny that many say it does not matter while Fractal has doggedly pursued a very high degree of accuracy to the originals for over a decade now and which is regularly applauded and precisely measured by many Fractal users. It does indeed seem to matter a lot as that accuracy is one of the common yardsticks by which modellers are measured and which consistently establishes Axefx as "Top of the Heap"
It does matter, but not in the sense of Can I make the 2204 in the Axe Fx sound exactly like my early 80's 2204. And that's because I could buy another 2204 and not get that to sound exactly the same as my early 80's 2204. When it comes to the accuracy of the Axe FX I mainly look for, Does it feel like my old 2204, does it have the same tonal characteristics of my old 2204, does it vibe like my old 2204. Then i know i will be able to get a tone I am happy with but just turning knobs like i did on my old 2204.

But I am sure if I A/B'ed them one would most definitely sound different than the other, just as if i A/B'ed two tube amps of the same model, and some may perceive this is better or worse. And the one you think sounds better may vary from the one i think sounds better. So accuracy when it comes to comparing tube amps vs modeling is a very deceptive term.
Bottom line, if I am going for a certain tone and I know such and such amp is the amp used to get that tone, I know and trust that if i select that same amp model in the Axe FX i will most certainly be able to capture the general vibe and intent, although it may not be the "exact" same tone.
People can take "accuracy" way to far when speaking about modeling, to an unrealistic point.


A real amp will sound different as you move around the room relative to the cab, move cab positions within the room, tubes age, tubes get warm, etc. so not sure a real amp sounds like a very good reference ! If the reference sound keeps changing how can you judge the axefx against it?


A real amp will sound different as you move around the room relative to the cab, move cab positions within the room, tubes age, tubes get warm, etc. so not sure a real amp sounds like a very good reference ! If the reference sound keeps changing how can you judge the axefx against it?
How can Cliff claim that Cygnus JCM 800 is more accurate in Cygnus than previous firmware revisions?

Sure, we don't have the exact same amps he does. But many variables can be eliminated. And you often see trends as well -- one being how Cygnus is supposed to have more "thump" than previous firmwares, more dynamics across many amps, ECT ECT.

Which is presumably more accurate. Not a case of "yes accurate" or "no it isn't", but more so in a specific manner imo.

I believe some of these general trends can be perceived by end users even without cliffs amps. But surely a lot of uncertainty comes with that, yeah. Pot tolerances is something to keep in mind among other variables.

But sure. Of course cliff is a better position than the rest of us for that considering he has the specific amps. And surely not all variables can be eliminated yeah.
Last edited:


Power User
The whole "just use your ears" thing has never really added up to me. Sure ears are important but if it's just that, then its twiddling knobs and hoping to land on the sound in your head - and given the number of knobs / combinations of knobs, the chances of aimlessly stumbling on that sound is unlikely. I think most people start playing guitar because they were inspired by some player or other which translates into a playing skillset quest and a tone quest which translates into buying / selling / buying gear to match the target tone(s), and, of course measuring progress along the way and predetermining if possible (as the OP requests) how close a given piece of gear can carry one to the target tone. Learning about the "standard" amp tones and how they came to be is also essential in my experience so far (ie if Brian May inspires one's playing, check out resources like "5 Watt World" to understand the history of the Vox AC30...). Learning / understanding more about the mechanics of various amp flavours, and tools like eq, compression + various fx also tremendously improves my ability dial in tones. It's so much more than just listening for what sounds good, though I do think the intention of the this statement often includes these other considerations - but without articulating those related items, the "just use your ears" statement does not seem very meaningful imo.
I've been saying this! 😁

It drives me bonkers to hear the phrase "just use your ears."


Obviously, they are accurate enough and have been for years. The ease of use and built in effects make it more practical as well.

And you don’t actually demand absolute accuracy, otherwise you wouldn’t be asking this question.


Power User
I never asked them, but I'm pretty sure that even my bandmates don't care at all what the model is after that I play with in a certain song. Ok, maybe the bass player, she plays the guitar in another band, but she never asked. Once you nailed perfect candidates for clean, crunch and solo, you don't need to search any longer. Nice to have hundrets of options though, but that is just for fun once you are done with your gotos.
Top Bottom