Anyone want Improvements in Axe-Fx User Interface?

Graphics are cool in the beginning and can have that wow factor at the start. But eventually hampers productivity as one progresses into more in-depth editing.

As a main fractal user since 2008 ...I admit that at that time I would have liked to see skeumorphic GUI. I recall someone even posted a cool skeumorphic sample of a Triaxis with Axe-Edit. Cool!

I like great beautiful graphics! ....but not at the expense of the flow of work productivity.

After 12 years with Fractal, I find that their aims and goals with their developments and UI has been spot on for the discerning enthusiasts.

Axe-Edit has come a long way and is great for me. Further enhancements will only make it greater.
 
For me the biggest usability improvement opportunity is in the cab block. I just absolutely dislike the way that cab IR auditioning works, and trying to find the right sound. I like that I can select particular IRs in the cases I have particular ones I like to use, but when I'm trying to find a sound, it's just tedious to go surfing through the list of cabs because I'm effectively jumping through random variations in the sound.

Today I was fine-tuning some sounds using Logic's amp designer and their cab UI is just a breath of fresh air in comparison:

1605926843321.png

you can select a cab (not shown in this screenshot) and microphone combo, and then you get a grid where you can move the mic around for different tonal variations of that. It's just a lot easier to work with, though obviously far less powerful than, the Axe-FX cab block.

I hope that Fractal would consider adding this kind of cab UI for at least the factory cabs and perhaps their cab packs as an option. I except some users would prefer the full-featured cab block UI as is now (and I do not mind it in some cases) but having this kind of alternative UI would probably work better for me 90%. It's similar to how I just always use the "Authentic" tab of the amp models now, except for a small minority of cases where it's beneficial for me to tweak something that's not shown there.

Note, it's not about the cab graphic, I don't really care if that's there or not, but more about the ability to quickly audition different variations of a cab by just changing the mic model and then dragging around a dot in 2D space to change the distance / mic placement. I imagine under the hood they just have a few IRs for each cab with the different mics and then crossfade between them much the same way on the Axe you use the level controls in the cab block.
 
For me the biggest usability improvement opportunity is in the cab block. I just absolutely dislike the way that cab IR auditioning works, and trying to find the right sound. I like that I can select particular IRs in the cases I have particular ones I like to use, but when I'm trying to find a sound, it's just tedious to go surfing through the list of cabs because I'm effectively jumping through random variations in the sound.

Today I was fine-tuning some sounds using Logic's amp designer and their cab UI is just a breath of fresh air in comparison:

View attachment 75334

you can select a cab (not shown in this screenshot) and microphone combo, and then you get a grid where you can move the mic around for different tonal variations of that. It's just a lot easier to work with, though obviously far less powerful than, the Axe-FX cab block.

I hope that Fractal would consider adding this kind of cab UI for at least the factory cabs and perhaps their cab packs as an option. I except some users would prefer the full-featured cab block UI as is now (and I do not mind it in some cases) but having this kind of alternative UI would probably work better for me 90%. It's similar to how I just always use the "Authentic" tab of the amp models now, except for a small minority of cases where it's beneficial for me to tweak something that's not shown there.

Note, it's not about the cab graphic, I don't really care if that's there or not, but more about the ability to quickly audition different variations of a cab by just changing the mic model and then dragging around a dot in 2D space to change the distance / mic placement. I imagine under the hood they just have a few IRs for each cab with the different mics and then crossfade between them much the same way on the Axe you use the level controls in the cab block.
Yeah those graphical cabinet things are cool but the amount of data storage required is difficult for an embedded product.

To do it right you need hundreds of IRs per cabinet per microphone. You simply cannot store that much data on something that doesn't have a hard drive or a FLASH drive of some sort.

If they are NOT capturing thousands of IRs per speaker then the results are bogus. In the nearfield of a speaker the sound changes drastically just moving the mic a centimeter or less. Let's assume one centimeter resolution. That's 30 IRs for each distance from the speaker for a 12" speaker. Now if you want to vary the distance over 12 inches then it's 30 x 30 = 900 IRs per microphone. Now multiply that times the number of microphones (and any perturbations on the angle of the mic).

You can't sample at greater distances than that and crossfade. It's simply not how a speaker behaves in the nearfield. IOW if you take an IR at one location and then move the mic one inch and take another the location halfway between the two isn't the average of the two IRs. You have to sample at a fine resolution and then use a single IR.
 
Yeah those graphical cabinet things are cool but the amount of data storage required is difficult for an embedded product.

To do it right you need hundreds of IRs per cabinet per microphone. You simply cannot store that much data on something that doesn't have a hard drive or a FLASH drive of some sort.

If they are NOT capturing thousands of IRs per speaker then the results are bogus. In the nearfield of a speaker the sound changes drastically just moving the mic a centimeter or less. Let's assume one centimeter resolution. That's 30 IRs for each distance from the speaker for a 12" speaker. Now if you want to vary the distance over 12 inches then it's 30 x 30 = 900 IRs per microphone. Now multiply that times the number of microphones (and any perturbations on the angle of the mic).

You can't sample at greater distances than that and crossfade. It's simply not how a speaker behaves in the nearfield. IOW if you take an IR at one location and then move the mic one inch and take another the location halfway between the two isn't the average of the two IRs. You have to sample at a fine resolution and then use a single IR.
Interesting, thanks for taking the time to write that up. The storage limitation is definitely an issue. I didn’t realize it would take that many IRs to get a good crossfade. I guess the only way would be to do it offline and be able to quickly preview on the AxeFX and then save the result in your preset.
 
Interesting, thanks for taking the time to write that up. The storage limitation is definitely an issue. I didn’t realize it would take that many IRs to get a good crossfade. I guess the only way would be to do it offline and be able to quickly preview on the AxeFX and then save the result in your preset.
Another possibility might be to live with less accuracy because of interpolation, but greater ease of use for folks who prefer the mic-moving paradigm.

OTOH, if the sound halfway between two mic positions isn't very related to the sounds at the original locations, then moving a mic visually wouldn't get you much more predictable results than semi-randomly auditioning IRs.
 
People get hung up on those people asking for like completely new UI with fancy features and the people asking for QoL changes get buried underneath that it seems
 
Another possibility might be to live with less accuracy because of interpolation, but greater ease of use for folks who prefer the mic-moving paradigm.

OTOH, if the sound halfway between two mic positions isn't very related to the sounds at the original locations, then moving a mic visually wouldn't get you much more predictable results than semi-randomly auditioning IRs.

As stated, accuracy is the whole driving force of fractal. Its the thing that wont get sacrificed. As such, we wont see this kind of thing unless something big happens behind closed doors.

I dont know what the plans are for Cliff moving forward, but Im excited to see them come to life.
 
Though I like the plugin graphics as I guess it provides a sense of psycological realism for me (whether or not actual tonal realism accompanies it is another matter), it definitely has the drawback of inconsistancy as pointed out by others - all the knobs are in different places and look different in every model - not so good for efficient tweaking.

I would not want, or expect a more graphical AxeEdit - the one we have suits the product's focus on precision and high tonal quality.

I was surprised that no-one reacted to the litle tidbit quoted below which seems to reveal an effort toward facilitating the often mentioned wish here to have greater ability to organize/categorize the longer model lists. This is good news!!

We are constantly investigating and working on usability enhancements and things like metadata, tags, etc. to streamline preset/IR management is a distinct possibility.
 
I'd love to see a wireless F/W upgrade app on my phone! Sell us a USB Wi-Fi dongle with an Android/ iPhone download!

That'd be cool...;)
 
I used Amplitube for a long time, mainly because of their movable mic and swappable speaker cabinet section. The Mikko plugin is a completely wonderful alternative, and far better for a Fractal user because of it's instant IR export function, among many other reasons. The amps in Amplitube were not right at all, and I hated it for that reason, in addition to the worst customer service, virtually no updates, and a lack of dynamic feel. Here's an example of the trade off between usability and looks: just try either of their officially endorsed Recitifers in Amplitube. They are so lacking in gain, it's unreal. I can't imagine how Boogie decided to license to them. I think their UI is really really nice looking, but their sound is lacking in direct proportion to the looks. I think they doth protest too much. It's like their trying to say, "Look how great you'll sound; see how good we look?" Not to mention the utter inability to approximate real world gain staging in Amplitube; the aspect of real world gain staging is another thing that makes the AxeFX the best thing on the planet. I'd much rather have everything that Fractal provides.

I'm a pretty new Axe FX owner, but for the few amp models I'm pretty familiar with in the real world, it's dead on to my experience, and that's worth everything to me. I think skeuomorphism can be a problem in that it can be a distraction from what's important. In the case of IK, I think they just poured all of their resources into looks, spent time hounding rock stars and engineers for endorsements, and left the sound to rot by not updating any models (instead selling new versions of old ones, at a snail's pace). The point of the AxeFX for me is very much to forget about the AxeFX, and my guitar, and just focus on the music itself. The AxeFX is so good that I don't need to worry about it constantly, tweak it all night long, and realize at the end of the night I've not spent any real time practicing or playing. I could spend forever working with it, but only out of joy and experimentation, because of it's world of features, which is awesome, but for me, it's a trusty tube amp, reliable and ready to go at 3 a.m. with no compromise in sound or feel. With Amplitube I really did fall into a bad cycle of constantly trying to find something that felt right without significant problems, and it was just demoralizing. It took a long time before I could even consider owning an AxeFX, but it really changed things for me, and, having spent so much of my time using pretty programs that sound lifeless, I'd jump on a command line modeler if it sounded better. To that point, I don't think the AxeFX UI looks bad; it's exactly what you need, right in front of your face, getting you to all the important aspects of your tone right away. I think it's some of the same thought process behind Cliff telling his fellow engineer not to label the FAS IRs with specific mic placement details: he just wanted it to be based on using your ears. That's the basis for all good audio engineering, and, by extension, guitar tone; just move knobs and sliders until it doesn't suck. Then, if you really value the sound itself as an art, you can craft it ad infinitum like the best engineers to create something sublime.

And that gets into the question of, if you only need your ears, why does the accuracy of models matter? Well, it only depends on how much you trust the designs amp designers, and how much you think their work is a link to good tone. I'm not an engineer, but I totally trust that all of the great designers, and the modders, are at the very least a good starting point. Obviously Cliff is one of those designers himself, not only with his only FAS models, but with the very specific suggestions he provides to correct for the inherent drawbacks of others' designs. I'm in no position to judge either way, but I trust Rupert Neve too, based on how much of all good sound we know comes directly or indirectly from his ideas about how it should be. I don't at all think about IK that way, quite the opposite.

All this is a very long-winded way for me to say, substance over style, every time.
 
If you only need your ears, why does the accuracy of models matter?

An amp model doesn't have to sound "accurate" to sound good, but by the same token, it's the pursuit of accuracy that's been the driving force behind the sonic improvements that we enjoy from today's top modelers.
 
In the nearfield of a speaker the sound changes drastically just moving the mic a centimeter or less.
This is why IR selection, for me, is an almost insurmountable process of auditioning each and every one of thousands of IRs. Filename descriptions don't tell me anything about how an IR sounds. Hitting the down/up arrows on my keyboard in Axe Edit is as fast as any other method I can think of. Moving a graphic of a mic with my mouse doesn't really seem more efficient, even though it is cool.

Still, I'm a huge fan of skeumorphic GUI for plug-ins, phones, etc. Not so much a fan of flat designs (iOS 7 and above). Axe Edit has a nice 3D physical look to the knobs, switches, etc., so I'd argue that it is at least semi skeumorphic.
 
Last edited:
Just spent time with AT5...mainly the fenders and vox’s and I’ve got to say - they’ve come a long way. I could really get great tones with it. The Fender 2 collection is really good. I could go back and forth with Axe and the differences were subtle. Now, the ease of use is way better with Axe, and until I get a super computer that could do 64 buffer with 4 instances, it’s not really worth the hassle. But I’m surprised at how far it has come.
 
Just spent time with AT5...mainly the fenders and vox’s and I’ve got to say - they’ve come a long way. I could really get great tones with it. The Fender 2 collection is really good. I could go back and forth with Axe and the differences were subtle. Now, the ease of use is way better with Axe, and until I get a super computer that could do 64 buffer with 4 instances, it’s not really worth the hassle. But I’m surprised at how far it has come.
+1 - Upgraded to A5 last week - really nice - Running the 3 Amps in Parallel is cool (2 different pitch shifts L and R into their own distorted amps + a dry "un pitch shifted" distorted amp in the middle - sweet! Feel is great running at 64 samples on my recently upgraded Mac (8Core/16Gb) - multiple instances needs128, but still feels good.
 
Back
Top Bottom