Axe-Fx III Firmware Release Version 12.09

I vaguely remember a pinnacle firmware on my AFX2...right around 12 or 13. Something happened right then that made things just click for me. Then they moved away from that for whatever reason, and some of the magic got lost for me, but when I commented on specifics, I got "You don't like it because that's the way the amps really sound."

We're back at one of those moments for me regarding the amp modeling.
 
This question must pop up after every firmware, so sorry if it's been asked dozens of times before.
Isn't the modeling something like 99.69% (or more; pick your percentage) accurate already? Basically indistinguishable in a A\B blind test from the real tubes. And hasn't this been the case for years?
Then how can a single version update provide such drastic changes to the sound/feel, such as in the high end here?
Or is my assumption wrong; that we're nowhere near 100% yet?
 
McillateA said:
This question must pop up after every firmware, so sorry if it's been asked dozens of times before.
Isn't the modeling something like 99.69% (or more; pick your percentage) accurate already? Basically indistinguishable in a A\B blind test from the real tubes. And hasn't this been the case for years?
Then how can a single version update provide such drastic changes to the sound/feel, such as in the high end here?
Or is my assumption wrong; that we're nowhere near 100% yet?
Click to expand...

I don't think of it in terms of percentages or accuracy at this point. More like each improvement Cliff makes is another step toward approaching infinity. You never quite get there, but you're always getting closer.
 
McillateA said:
This question must pop up after every firmware, so sorry if it's been asked dozens of times before.
Isn't the modeling something like 99.69% (or more; pick your percentage) accurate already? Basically indistinguishable in a A\B blind test from the real tubes. And hasn't this been the case for years?
Then how can a single version update provide such drastic changes to the sound/feel, such as in the high end here?
Or is my assumption wrong; that we're nowhere near 100% yet?
Click to expand...

Honestly, 99% of the time there's a firmware update I am wondering if my ears are broken after reading the threads. I seldom ever notice any difference. Typically, nothing sounds worse, so... yay? I just chalk it up to being a neanderthal when it comes to tone I guess :p
 
Joe Bfstplk said:
Falling in love all over again with the Hipower Jumpered in 12.09....

6CA7s, bump up the presence a tad, fatten up the bias a bit, bump up the hardness a bit, and boom! The top end is more delicate and less aggressive, but it's super detailed and sparkles like it's surrounded by phosphorescent pixie dust....
Click to expand...

Which bias parameter are you tweaking?
 
Morphine said:
Honestly, 99% of the time there's a firmware update I am wondering if my ears are broken after reading the threads. I seldom ever notice any difference. Typically, nothing sounds worse, so... yay? I just chalk it up to being a neanderthal when it comes to tone I guess :p
Click to expand...

There are plenty if things I don’t hear when I read them in the forum after updates - but, I also don’t spend hours and hours changing things like preamp tubes, speaker settings etc. if I hear something that I don’t like in a preset, I search out here to see if maybe one of those advanced parameters might be the answer.

That being said - I completely noticed some changes made lately- like the virtual capo, and in 12.09 - the difference in the high end. I really feel like 12.09 made a big difference for me and that it sounds amazing.

I might also be a tone-andrethal though. Lol
 
Last edited:
McillateA said:
This question must pop up after every firmware, so sorry if it's been asked dozens of times before.
Isn't the modeling something like 99.69% (or more; pick your percentage) accurate already? Basically indistinguishable in a A\B blind test from the real tubes. And hasn't this been the case for years?
Then how can a single version update provide such drastic changes to the sound/feel, such as in the high end here?
Or is my assumption wrong; that we're nowhere near 100% yet?
Click to expand...

I've been around since the beginning. I've seen since the beginning people claiming that it's indistinguishable from a real tube amp since then, but we have multiple generations along the way, with multiple firmware updates per generation. That just tells me that we wasn't there, and still aren't there. I absolutely love my AxeFx III, but there is still a missing 3d quality to the mids and highs.

There's a post from about six months or so ago from Cliff on TGP talking about how difficult it is to capture everything a tube amp has to offer with current DSP's. So we're obviously still not there 100% or even 99% or 95%, but obviously there good enough to get some gnarly sounds coming out of it. DB76598E-368B-4903-B24A-F10F1CB4B8BB.jpeg
 
There are SO many ways to hear a tube amp and they all sound different, too. So how do you hear the amp? ”amp in the room” through a 4x12? Through your studio monitors with the amp blasting out of earshot? Through a wedge on the floor? After you factor in a zillion different ways to hear a physical amp, they’ll ALL sound radically different from one another and I wonder if those different conditions can actually be modeled. Until people realize when you’re hearing a model and it’s not exactly what you’re used to, it doesn’t mean the model is inaccurate. I learned this after hearing just how close these models are to the actual amps when I ran both the model and the amp through the same cab block. Ever play at a professional session with a set of headphones on and wonder what in the hell happened to your beautiful amp sound? It’s still the same amp, but once again the listening medium changed.
To this point, I’m wondering just how much FAS is chasing its own tail. The models are THERE, at least the ones I gravitate to. I can‘t imagine the man hours that went into writing all of them. In my humble opinion (and that’s pretty damn humble), the user base would probably be better served if resources were more focused on:
New model creation
New Effects
Enhanced usability.
This is only evidenced by the now thousands of professional recordings where the Fractal is indistinguishable from the modeled targets.
 
Last edited:
Sorry if I’m just piling on here, but yes, the amps models sound even better now and it is noticeable. Which quite amazing considering how much I already loved the tones I was getting. These updates Cliff keeps putting out are making it impossible for me to get anything else done around the house.
 
Bruce Sokolovic said:
There are SO many ways to hear a tube amp and they all sound different, too. So how do you hear the amp? ”amp in the room” through a 4x12? Through your studio monitors with the amp blasting out of earshot? Through a wedge on the floor? After you factor in a zillion different ways to hear a physical amp, they’ll ALL sound radically different from one another and I wonder if those different conditions can actually be modeled. Until people realize when you’re hearing a model and it’s not exactly what you’re used to, it doesn’t mean the model is inaccurate. I learned this after hearing just how close these models are to the actual amps when I ran both the model and the amp through the same cab block. Ever play at a professional session with a set of headphones on and wonder what in the hell happened to your beautiful amp sound? It’s still the same amp, but once again the listening medium changed.
To this point, I’m wondering just how much FAS is chasing its own tail. The models are THERE, at least the ones I gravitate to. I can‘t imagine the man hours that went into writing all of them. In my humble opinion (and that’s pretty damn humble), the user base would probably be better served if resources were more focused on:
New model creation
New Effects
Enhanced usability.
This is only evidenced by the now thousands of professional recordings where the Fractal is indistinguishable from the modeled targets.
Click to expand...
It's not mutually exclusive. There has been just as much work on those items.
 
The "compromises" in the Axe are more than worth it for the awesome effects and ease of use, plus tweakability.
In regard to the accuracy of the modeling and how much cpu power it really takes, I have been modeling and running wav audio through LTSpice since it came out in the early 2000s.
With the full precision spice engine you will be lucky today to get over 500uS/S simulation speed with a non-linear tube circuit with 5 preamp stages and two power amp tubes in class AB even with the fastest 8 core intel overclocked. It's more often going to chug far below that speed.
When I started on a single core AMD 1400mhz cpu, I would render 10 second files out overnight or sometimes a full day at 10uS/S on big peaks, but much faster on tails, over 50uS :D
It goes much faster if you turn the gain down!
Given that FASpice can get even remotely close as they do to the real thing on current DSPs is an engineering miracle. There will always be new tricks and algorithms that will improve accuracy and that's what you get with these major firmware revisions.
 
Bruce Sokolovic said:
There are SO many ways to hear a tube amp and they all sound different, too. So how do you hear the amp? ”amp in the room” through a 4x12? Through your studio monitors with the amp blasting out of earshot? Through a wedge on the floor? After you factor in a zillion different ways to hear a physical amp, they’ll ALL sound radically different from one another and I wonder if those different conditions can actually be modeled. Until people realize when you’re hearing a model and it’s not exactly what you’re used to, it doesn’t mean the model is inaccurate. I learned this after hearing just how close these models are to the actual amps when I ran both the model and the amp through the same cab block. Ever play at a professional session with a set of headphones on and wonder what in the hell happened to your beautiful amp sound? It’s still the same amp, but once again the listening medium changed.
To this point, I’m wondering just how much FAS is chasing its own tail. The models are THERE, at least the ones I gravitate to. I can‘t imagine the man hours that went into writing all of them. In my humble opinion (and that’s pretty damn humble), the user base would probably be better served if resources were more focused on:
New model creation
New Effects
Enhanced usability.
This is only evidenced by the now thousands of professional recordings where the Fractal is indistinguishable from the modeled targets.
Click to expand...

Cliff made reference to a patent he's waiting on that will change the whole ballgame, at least that's what it sounded like he was saying. He was intentionally being vague but it sounded like he had something in mind to help users design their own custom virtual amps. At least that's what I think he was getting at.
 
deathbyguitar said:
Cliff made reference to a patent he's waiting on that will change the whole ballgame, at least that's what it sounded like he was saying. He was intentionally being vague but it sounded like he had something in mind to help users design their own custom virtual amps. At least that's what I think he was getting at.
Click to expand...
I really hope he outlives me and doesn’t retire. If the axefxIII never got another update I still see it as an immense value. The fact even more innovation goes into it, sometimes weekly, is astounding.
 
FractalAudio said:
It's not mutually exclusive. There has been just as much work on those items.
Click to expand...
Please don't ever stop the improvements to the amps and all these other areas. If you stopped now, as you once said, "where's the fun in that?". I for one hear a very noticeable improvement in this FW (and usually all the others too). The 6CA7 tube additional sounds great, put into a 6550 Plexi with a tad increase in grid bias, pre-rola greenback IRs, and the tone sounds great to me. I have had a hell of a time previously taming the high-end as I really like how brightness adds bite to the low strings on my guitar and chords, but simultaneously reducing the ice-pickiness or shrill that comes along with that has been very difficult. With this new firmware I can now get both. With some tweaks my plexi sounds the best it ever has to my ears. I am was very happy previously with tone, so please, keep going!
 
guitarzen said:
Please don't ever stop the improvements to the amps and all these other areas. If you stopped now, as you once said, "where's the fun in that?". I for one hear a very noticeable improvement in this FW (and usually all the others too). The 6CA7 tube additional sounds great, put into a 6550 Plexi with a tad increase in grid bias, pre-rola greenback IRs, and the tone sounds great to me. I have had a hell of a time previously taming the high-end as I really like how brightness adds bite to the low strings on my guitar and chords, but simultaneously reducing the ice-pickiness or shrill that comes along with that has been very difficult. With this new firmware I can now get both. With some tweaks my plexi sounds the best it ever has to my ears. I am was very happy previously with tone, so please, keep going!
Click to expand...

Cliff is a man on a mission. He's not gonna stop for shit, not until he's taken the technology as far as he can. The unit's only been out for 2 years. We've got some crazy times ahead of us,
 
You must log in or register to reply here.
Back
Top Bottom