Axe-Fx III Firmware 29.01 Release

Bug ? When I assign Control Switch 5 to Fat in the Atomica High, I go to the controllers list in AE and see Control Switch 5 : Amp 1 (Treble Shift).
Shouldn't it say Fat Switch instead of Treble Shift ? (Though Fat might be a Treble Shift in a certain way ...)
 
lp59 said:
Bug ? When I assign Control Switch 5 to Fat in the Atomica High, I go to the controllers list in AE and see Control Switch 5 : Amp 1 (Treble Shift).
Shouldn't it say Fat Switch instead of Treble Shift ? (Though Fat might be a Treble Shift in a certain way ...)
Click to expand...
That kind of sounds more like an Axe Edit issue. Just curious if you did the old RANF after installing 29.01?
 
FractalAudio said:
https://www.fractalaudio.com/downloads/firmware-presets/axe-fx-3/29p0/axefxiii_dsp_rel_29p01.zip

Axe-Fx III Firmware Release Notes​


29.01

Improved intelligent pitch shifting in Pitch block.

Notes:
  • The Pitch Tracking parameter selects between Off, Chords and Notes. In Chords mode the tracking is optimized for playing multiple notes simultaneously and trades latency for stability. In Notes mode latency is reduced at the expense of stability. In Notes mode simple chords can be tracked but complex chords may present difficulty for the pitch tracker. The Tracking Adjust parameter allows for fine adjustment of the stability vs. latency trade-off.
  • The Pitch Tracking parameter is only available for chromatic shifting types (this includes whammy types). Diatonic shifting necessarily requires playing single notes and the algorithm is automatically set to Notes mode for these types.
Improved CPU usage for some types in Compressor block.

Reduced filtering of rear panel pedal inputs. This allows faster response to rapid pedal movements.

Fixed pop/thump when switching to/from certain Drive types.

Various other fixes and improvements.
Click to expand...
Thanks for the update! VC sounds great now!

And sorry for being nitpicky, but I think there is noticeable latency between pick attack and sound coming out of speakers. Don’t get me wrong, this is probably the best VC I have heard, but the lag is still there.

@FractalAudio I was wondering if there is a way to make it more responsive by using more CPU power, perhaps via an extra switch in the block to turn on for this purpose? (And no, tracking adjust did not help reduce the latency to acceptable levels)
 
oacikgoz said:
Thanks for the update! VC sounds great now!

And sorry for being nitpicky, but I think there is noticeable latency between pick attack and sound coming out of speakers. Don’t get me wrong, this is probably the best VC I have heard, but the lag is still there.

@FractalAudio I was wondering if there is a way to make it more responsive by using more CPU power, perhaps via an extra switch in the block to turn on for this purpose? (And no, tracking adjust did not help reduce the latency to acceptable levels)
Click to expand...

Why latency is unavoidable (physics):

https://wiki.fractalaudio.com/wiki/index.php?title=Pitch_block#Latency

EDIT: I edited the link, because the wiki page has been updated
 
Last edited:
My name is mud said:
I will have the Morpheus drop pedal at home in maybe one week, I never tried it even if 10 years ago it was considered one of the best drop pedal. It seems they bankrupt ?
Well we will see ! I ll tell you back if you are interested
Click to expand...
I have used it. The digitech drop tune is considerably better. Make sure you try a buffer in front of the Morpheus. It is a tone tainter.
 
jimmynorrisjr said:
I have used it. The digitech drop tune is considerably better. Make sure you try a buffer in front of the Morpheus. It is a tone tainter.
Click to expand...
Ah? For what I have seen they are pretty close . I owned the digitech drop quickly at home and wasn’t really …
It seems that the Morpheus doesn’t have all they weird things happening in clean tone . I will see, that’s just curiosity, I found one « cheap »

Most of the drop things are really ugly in clean tone

I also like that you can change tuning with your foot , and use it momentarily, does octaver

4:00 for what I was saying about clean tone

 
Last edited:
jimmynorrisjr said:
Even the drop tune has latency, and it has been one of the best of all time. I think the VC in the new software is actually better.
Click to expand...
Having just compared them, I have to disagree. The VC is much, much improved - and there really isn't much in it - but for me, the Drop feels more 'immediate'. Only slightly mind you.
 
oacikgoz said:
Thanks @yek . Yes I am aware of all that, latency is a function of cpu utilization among other things. so latency can be decreased, whether you should or not depends on a lot of other parameters I don’t have access to. Mine was a question on whether there is more headroom for that kind of improvement.
Click to expand...
Latency is NOT a function of CPU. Reread those posts.
 
FractalAudio said:
Latency is NOT a function of CPU. Reread those posts.
Click to expand...
I seem to remember to have read an article a while ago that said it is possible in some way to detect pitch from a transient without needing to wait for a full wave cycle... Is this kind of algorithm applicable to the axe fx or has some drawbacks or... did I just dream about it and confused it for a memory? 😅

EDIT: probably I didn't dream about, I think it was an article by Joel De Guzman, cuz he briefly talks about it in his latest one (I think it's what he calls "pre-pick/pluck detection"):

https://www.cycfi.com/2024/09/pitch-perfect-enhanced-pitch-detection-techniques-part-1/

EDIT2: nope, that seems a different thing... But still quite interesting
 
Last edited:
DLC86 said:
I seem to remember to have read an article a while ago that said it is possible in some way to detect pitch from a transient without needing to wait for a full wave cycle... Is this kind of algorithm applicable to the axe fx or has some drawbacks or... did I just dream about it and confused it for a memory? 😅

EDIT: probably I didn't dream about, I think it was an article by Joel De Guzman, cuz he briefly talks about it in his latest one (I think it's what he calls "pre-pick/pluck detection"):

https://www.cycfi.com/2024/09/pitch-perfect-enhanced-pitch-detection-techniques-part-1/

EDIT2: nope, that seems a different thing... But still quite interesting
Click to expand...
I think my old Antares hexaphonic MIDI setup may have even been using the transient to detect pitch. But, that's only detecting the pitch, there's a lot more going on with creating real-time harmonies, virtual capo, etc (methinks).
 
DLC86 said:
I seem to remember to have read an article a while ago that said it is possible in some way to detect pitch from a transient without needing to wait for a full wave cycle... Is this kind of algorithm applicable to the axe fx or has some drawbacks or... did I just dream about it and confused it for a memory? 😅

EDIT: probably I didn't dream about, I think it was an article by Joel De Guzman, cuz he briefly talks about it in his latest one (I think it's what he calls "pre-pick/pluck detection"):

https://www.cycfi.com/2024/09/pitch-perfect-enhanced-pitch-detection-techniques-part-1/

EDIT2: nope, that seems a different thing... But still quite interesting
Click to expand...
Even if you could detect the pitch instantly (and you can't) that won't reduce the latency. Pitch detection isn't the source of the latency. You MUST buffer up "grains" of data and then play them back at a different speed. You can't play the grains back until you've acquired them.

Our pitch detection is actually quite fast and robust, doesn't even need two full cycles to determine the pitch. It also does transient detection. But all that is irrelevant.
 
FractalAudio said:
Even if you could detect the pitch instantly (and you can't) that won't reduce the latency. Pitch detection isn't the source of the latency. You MUST buffer up "grains" of data and then play them back at a different speed. You can't play the grains back until you've acquired them.

Our pitch detection is actually quite fast and robust, doesn't even need two full cycles to determine the pitch. It also does transient detection. But all that is irrelevant.
Click to expand...
I only ask this out of pure curiosity and fascination that any of this can be done at all. When you say "play them back at a different speed" that sounds like it's in the analog domain. Is this happening after the d/a conversion? If so, what stops you from converting while still in the digital realm and then play back at "new normal" speed?
 
You must log in or register to reply here.
Back
Top Bottom