Wish Faster Bypass/Mixer/Multiplexer ramps

B

Bakerman

Axe-Master
Except when controlling a volume block's volume parameter, any method of muting and unmuting a signal has a fade of 50 ms or more.

HZVwp5F.png


This has audible effects that aren't always ideal. The swell can be noticeable when bringing a new tone/side/effect/harmony/etc. into the mix. Engaging or bypassing a pitch shift before an amp causes sputtering from both notes briefly hitting the amp.

On the Axe-FX II bypass and level adjustments all used ~15 ms ramps, except for the mixer which had no auto-damping and only used whatever modifier damping was set. This sounds better in the scenarios mentioned above. A parameter to set ramp time might be nice for the multiplexer and (with channel switching, not level modifiers) mixer blocks.
 
Ha! +1, it's like Mesa boogie's channel switching *pop, or the reverb tank delay when changing channels. They did it to avoid painful circuit switching sounds, but that's in a mechanical box. Shirley there is a way to cut signal instantly and switch over in the digital realm, sub 1ms. A ramp can be sub 1ms, a 1ms crossfade would sound just as instant as you could get.
 
shatteredsquare said:
Shirley there is a way to cut signal instantly and switch over in the digital realm, sub 1ms. A ramp can be sub 1ms, a 1ms crossfade would sound just as instant as you could get.
Click to expand...
Any ramp shorter than 30 ms or so would cause an audible artifact.
 
Rex said:
Any ramp shorter than 30 ms or so would cause an audible artifact.
Click to expand...

10-15 ms is fine I think. The Axe-FX III uses ~15 ms for amp channel & preset changes (with additional silence between the fades). Things mentioned above work nicely on the Axe-FX II around 15 ms too.
 
Last edited:
Bakerman said:
10-15 ms is fine I think. The Axe-FX III uses ~15 ms for amp channel & preset changes (with additional silence between the fades). Things mentioned above work nicely on the Axe-FX II around 15 ms too.
Click to expand...
That's 15 ms each way, right? For a total 30 ms fade-in/fade-out?
 
Rex said:
That's 15 ms each way, right? For a total 30 ms fade-in/fade-out?
Click to expand...

15 ms each way, but none of these examples are actually a 30 ms process. You'd either just be muting/unmuting or have 2 signals that could crossfade, or it's something like amp/drive channel changes which have additional silence (~35 ms for amp, ~15 ms for drive) between the fades.
 
Bakerman said:
15 ms each way, but none of these examples are actually a 30 ms process. You'd either just be muting/unmuting or have 2 signals that could crossfade, or it's something like amp/drive channel changes which have additional silence (~35 ms for amp, ~15 ms for drive) between the fades.
Click to expand...
I can understand the shorter times for crossfades. I suspect the longer times are necessary when an entirely new waveform has to be calculated and used to replace the original.
 
Rex said:
I can understand the shorter times for crossfades. I suspect the longer times are necessary when an entirely new waveform has to be calculated and used to replace the original.
Click to expand...

I'm not sure what you mean by "the longer times". The amp and drive blocks already use 15 ms ramps on channel switching. There's extra silence between the ramps, but this wish isn't about improving amp/drive channel switching speed.
 
Rex said:
Any ramp shorter than 30 ms or so would cause an audible artifact.
Click to expand...

Wut. The only artifact is caused by the audio turning on when the waveform is off of the zero crossing. All the DAWs do automatic crossfades now, it only needs to be a few ms ramp up to avoid the pop of the speaker jumping from it's zero point to where the waveform is trying to pickup off of zero.
 
shatteredsquare said:
Wut. The only artifact is caused by the audio turning on when the waveform is off of the zero crossing. All the DAWs do automatic crossfades now, it only needs to be a few ms ramp up to avoid the pop of the speaker jumping from it's zero point to where the waveform is trying to pickup off of zero.
Click to expand...

Crossfading "existing" audio data is different from "live" crossfading.
 
  • Like
Reactions: Rex
I could understand that much latency if switching a processing module, so you don't waste the resources, it's not like a double clutch transmission where the next gear is already loaded up. But where it's just an audio path change, no reason for that much gap, if it's just to silence a pop.
 
Bakerman said:
Except when controlling a volume block's volume parameter, any method of muting and unmuting a signal has a fade of 50 ms or more.

HZVwp5F.png


This has audible effects that aren't always ideal. The swell can be noticeable when bringing a new tone/side/effect/harmony/etc. into the mix. Engaging or bypassing a pitch shift before an amp causes sputtering from both notes briefly hitting the amp.

On the Axe-FX II bypass and level adjustments all used ~15 ms ramps, except for the mixer which had no auto-damping and only used whatever modifier damping was set. This sounds better in the scenarios mentioned above. A parameter to set ramp time might be nice for the multiplexer and (with channel switching, not level modifiers) mixer blocks.
Click to expand...

What are you using to get that much latency though, axe edit or the front panel is slower than midi direct or an fc.
 
You must log in or register to reply here.
Back
Top Bottom