Would you say the next-gen Axe is years away?

If only the end result (tone) matters, maybe yes, maybe no, I don't know.

But that's not the approach of Fractal. White box modeling is so much more valuable, scientifically speaking.
May not matter to some, but it matters to me.

If all tube amplifiers would vanish tomorrow, capture tech would still reproduce a selection of their tones.

That's nice, but Fractal's amp models would still provide insight in how tube amplifiers actually work and why they sound the way they sound etc.

Capture tech is one or more meals. White box modeling is the cookbook, the kitchen and the ingredients.

Adding NAM is, like Cliff wrote, just a means to have access to sounds of amps that have not been modeled by FAS.

White box modeling sits on another tier because it explains the “why” behind the tone. It shows the behavior of the circuit, the interaction of stages, and the reason an amp reacts the way it does. That level of depth is what lets Fractal recreate the full experience of an amp, not only the snapshot of a single setting. Anyone who has spent time with the amp block knows how much range and realism that gives you.

At the same time, NAM still has plenty of value. If it didn’t, Cliff wouldn’t bother putting it in the box. Simple as that. NAM shines when you want a fully baked tone where someone already dialed the amp, picked the boost, picked the cab, placed the mic, hit the console eq, hit the master eq, and handed you the final sound. Tweaking those chains can feel overwhelming if you hate deep diving into parameters. Plenty of players in the Fractal world prefer to load something that already sounds finished. The whole paid preset market exists because of that. People even pay others to design tones for them. Matt and Carter literally travel to artists to fine tune their rigs because not every player wants to be an engineer.

So yes, the Fractal box is the full kitchen with every tool and every ingredient. Still, not everyone wants to be a chef. And even the best cooks pick up a street vendor hot dog once in a while because it hits the spot with zero effort.

Both approaches have strengths. They solve different needs. And soon we get both in one unit. When that happens, I’ll bet my guitar collection plenty of people who swear they don’t care about NAM will be singing its praises once they hear how handy it is to have both paths in the same workflow.


What paper? Interested.

Leo Gibson is a good place to start for those that are interested.



 
White box modeling sits on another tier because it explains the “why” behind the tone. It shows the behavior of the circuit, the interaction of stages, and the reason an amp reacts the way it does. That level of depth is what lets Fractal recreate the full experience of an amp, not only the snapshot of a single setting. Anyone who has spent time with the amp block knows how much range and realism that gives you.

At the same time, NAM still has plenty of value. If it didn’t, Cliff wouldn’t bother putting it in the box. Simple as that. NAM shines when you want a fully baked tone where someone already dialed the amp, picked the boost, picked the cab, placed the mic, hit the console eq, hit the master eq, and handed you the final sound. Tweaking those chains can feel overwhelming if you hate deep diving into parameters. Plenty of players in the Fractal world prefer to load something that already sounds finished. The whole paid preset market exists because of that. People even pay others to design tones for them. Matt and Carter literally travel to artists to fine tune their rigs because not every player wants to be an engineer.

So yes, the Fractal box is the full kitchen with every tool and every ingredient. Still, not everyone wants to be a chef. And even the best cooks pick up a street vendor hot dog once in a while because it hits the spot with zero effort.

Both approaches have strengths. They solve different needs. And soon we get both in one unit. When that happens, I’ll bet my guitar collection plenty of people who swear they don’t care about NAM will be singing its praises once they hear how handy it is to have both paths in the same workflow.




Leo Gibson is a good place to start for those that are interested.




Yeah the fully baked in sound doesn't do it for me but I'm just one consumer. The market for NAM is pretty large right now.
 
A good capture pack will contain multiple captures for different channels and gain/tone settings. You pick the one that gets you closest to your target and make small tweaks from there. You can also get captures of just the amp (DI) with nothing else baked in then use your own cabs to taste. Most people using Nam/ToneX etc prefer it this way.
 
Yeah the fully baked in sound doesn't do it for me but I'm just one consumer. The market for NAM is pretty large right now.

I’ve got no experience with NAM so far, but what if you think of it as your scenes in the next gen units? Maybe you could built your scene, create a NAM profile of it, load this into your NAM block and would have the whole FAS Modeling CPU free for use in parallel to that. This way you could create some really wild things that don’t effect cpu that much. 🤷‍♂️
 
Machine-learning still seems a fundamentally poor idea to me for most applications. Sledgehammer to crack a very big nut is so wasteful compared with figuring out how to open the nut intelligently, and getting the edible part out in one piece. I watch colleagues wasting time on AI solutions on a daily basis, and to me, it reminds me of why I think modelling the fundamentals works better than profiling, no matter how sophisticated the profiler might be.

(Many of those colleagues seem to think AI is adding to their intelligence, when all it really does is expose their lack of it.)
 
Machine-learning still seems a fundamentally poor idea to me for most applications. Sledgehammer to crack a very big nut is so wasteful compared with figuring out how to open the nut intelligently, and getting the edible part out in one piece. I watch colleagues wasting time on AI solutions on a daily basis, and to me, it reminds me of why I think modelling the fundamentals works better than profiling, no matter how sophisticated the profiler might be.
Perhaps in many exploratory use cases this would be true but I disagree for profiling. It's one of the areas where it obviously shines. I'm not going to get into a this is better than that debate since both profiling and modelling have their teeth dug deep in the industry suggesting both are viable solutions. To me, I can tell the difference right away between a good model and good profile. One isn't better than the other but there is something to the profiles that make me feel like I'm playing a real amp versus a great sounding model of an amp.
 
I have the complete opposite experience: Every profile/capture/whatever I've played has lacked something that Fractal's modeling has.
Try some of Amalgam's V2 captures. I am using ToneX profiles I don't know if there is a difference between Nam and the proprietary profiles from ToneX though.
 
Nothing is free. Running the NAM capture takes CPU just like the Fractal amp models do. Amp models run on a dedicated core currently so they contribute very little to the reported CPU usage now.
Shouldn’t matter much on the new gen since according to Cliff, it’ll be like a super-computer…….biggrin.gif
 
Of course more power always be welcome, I’m very, very happy with my Turbo III.

I never took me thinking about axe IV.
However, I’m wondering how good would be if one next firmware bring to us:

. Diferent modifiers on each Chanel to the same parameter;
. Global blocks with name;

All the best, folks!
 
I don’t like AI. FAS has been an oasis of sanity.

Back to the OP’s original question, seeing that there are some recent quotes from Cliff about what the next gen will have, that probably means it is in some level of development.
 
Last edited:
Well, TBH, I can’t really get any money for the Tonex LOL! I like all three but I can’t really afford to keep em all.
Especially not now with the plug just being released and BF deals. BTW, I tried some of your presets on ToneNet or from one of your posts and really enjoyed them.
 
Back
Top Bottom