IV anywhere on the horizon?

That's what most people who had the Axe fx II and now have the III said about the II also. I am sure sure when the IV is there many people will sell their III and say the IV is better and it can't get better than that. Then the V comes.....
I started on the Ultra 11 years ago, and you’re exactly right. I’m not sure if that is Fractal Audio’s amazing ability to go beyond what seems possible, or my lack of ability to imagine it, but it’s always ended in a win. The operative word is “needing” in that quote. I like having the flagship and have always been gigging enough that swapping to the next unit paid for itself… just a business expense. Post-pandemic, the live music scene locally has been very slow to recover. In this scenario, I probably won’t update unless something truly earth shattering is on board. I’ll have to really need it, which I can’t imagine at this point.

Literally, I won’t need it. Guitaristically, the meaning of that word is ambiguous, so you may turn out to be correct after all! In the meantime, the III-T is pretty darned awesome.
 
Software will eventually find a way to use all available resources. We don't know if Fractal's ability to be better is limited by CPU or other resources at this point. Or if that next step is too minimal to make it worth it or too big a jump to be able to do it. Too many unknowns, other than eventually we will see an AXE-FX IV.
 
Since I don't use the front of my Axe-FX III Mk2 very often (as the Editor is my choice of communications with it), I don't really mind if the Axe-FX III has a touch screen or not. I'd rather have better amps, cabs & tone than a touch screen any day. YMMV and that's ok. :)
My QSC TouchMix mixer has a touch screen, I’ve had it for 7 years with no problems. I’d like the IV to have the option to run it from an iPad or something similar. No built-in touch screen tho; it would make the unit too big.
 
Profiling is patented and I've been threatened more than once by Kemper's attorneys.

Using neural networks is possible but training a NN takes hours and requires hardware that a consumer modeler would not have (i.e. inference accelerators).

That said, my tests have shown that white-box modeling outperforms profiling and NN in terms of accuracy and aliasing. Whether that sounds "better" is subjective.
Suggestion. Instead of using a complicated NN to model an amp, use a simple NN to model the remaining error of the white-box model.
 
At this point it looks like the hardware is plenty fast enough. The more important thing is the software, which is always being updated and given away for free. We always have the latest greatest. Until the software requires better, faster, hardware then new hardware doesn't do anything for you. Those extremely difficult algorithms Cliff is working on in the software is where it's at.
 
Last edited:
Regardless of how long it takes for the AFX4 to arrive, when it does I'll have another good laugh at all the clueless reverb postings that'll be offering used AFX3 Mk2 Turbos for above AFX4 MSRP.
 
Regardless of how long it takes for the AFX4 to arrive, when it does I'll have another good laugh at all the clueless reverb postings that'll be offering used AFX3 Mk2 Turbos for above AFX4 MSRP.
Look what they sell for. A lot of foreigners buy them over msrp 🤷‍♂️
 
Back
Top Bottom