Sidechain/FX gate solution to squeals / note stops sustaining?

[jesussaddle]

I'm interested in delving into something. I am struggling get a sustaining tone at reasonable volume in my room. I know that if I increase the actual volume I can get more sustain (which solves the problem in a way since the sustain can be maintained long enough for me to enjoy it, and when I decide I'm ready to stop the note I can mute the strings. But I'd like to think of a way to suppress the squeals in result in a preset called Still Blue (preset 121 or 120) that has a high bright cap setting) when I boost the drive of the amp or do other things to increase input into the amp (drive or compressor block).

I was watching a Fabfilter video on Pro G (Gate / Expander ) in which sidechaining is used. Meanwhile I watched a video on the C$ck Blocker pedal:



This would be a "4 cable method" (cable 1 is from guitar to blocker, cable 2 is to instrument input of axe fx, cable 3 is from axe fx loop out to blocker fx loop INPUT, cable 4 is from blocker back to axe fx loop in).

But apparently the circuit from cables 3&4 would suppress the signal noise from the amp, whereas the circuit from cables 1&2 would suppress (gate) the notes, which doesn't really help me, since the squealing is too loud to choke off - it would glitch and pop as it approached the threshold since it is just about as loud as the notes.

There should be a way to detect when the "note" stops based on the synth block since it needs to stop sending a synth tone. Can it be used to trigger a gate in the AXE fx? (At the moment I only own a II but I'm curious about the rest of the product line actually, since I am considering them all for travel.)

 

[Bakerman]

I'm not sure what you're picturing the synth block doing to help with noise gating.

The Axe equivalent of that pedal's 4-cable connection style would be gate/expander block after amp block with sidechain parameter set to Input 1.

In the Axe-FX it's generally less crucial to have the gate post-amp. The amp block just amplifies noise reaching it, rather than generating its own additional noise like a real amp. Still, there can be a difference in sound when the gate is working. Before the amp block with enough distortion happening, you'll hear the tone clean up before the level drops. After the amp, it will just sound like you're reducing amp level while the tone pretty much stays distorted (if threshold/ratio set high enough).

 

[jesussaddle]

I'm not sure what you're picturing the synth block doing to help with noise gating.

The Axe equivalent of that pedal's 4-cable connection style would be gate/expander block after amp block with sidechain parameter set to Input 1.

In the Axe-FX it's generally less crucial to have the gate post-amp. The amp block just amplifies noise reaching it, rather than generating its own additional noise like a real amp. Still, there can be a difference in sound when the gate is working. Before the amp block with enough distortion happening, you'll hear the tone clean up before the level drops. After the amp, it will just sound like you're reducing amp level while the tone pretty much stays distorted (if threshold/ratio set high enough).

I see what you mean about post amp gating.

In its present form the synth block will do nothing, but I meant that I imagine that somewhere in the code base Cliff may have figured out when to suppress this particular type of amp / feedback squealing since it would be soft and unstable enough to present a problem - when the data is taken from its original audio form and put into the form used by the synthesis algorithms. [There are types of feedback that sound just like a sustaining note in terms of "shape", but have different harmonic content - this is not what I have now, that is unwanted here] . I imagine the synth block detects note and chords (seemingly very very well) and all it would need to do - similar to gating a signal that falls below a threshold (but we're talking about a "threshold" of what looks like a plucked or picked or hammered note in its transient versus what looks like a very very slowly (relative to note attack by finger or pick) building event - a feeding back signal - I imagine its possible to somehow recognize that feedback often "looks" different than a note, in terms of there being a very different (and more gradual) transient... Could be I'm mistaken.

This particular preset I'm working on has piercing, latent feedback, and its a "swell" like waveform. In the 4 cable method the first pair is before the amp, right after the guitar. I imagine that if the gate were setup in such a way as to not open for a "swell" like waveform, it might be possible to suppress this if its unwanted - as a guitar player sometimes I love feedback. It would be could to have settings in the Axe FX that optionally could suppress it - for example there are cases when you can keep sustain going and feeding back at roughly the same level as your note - Nazareth Love Hurts solo for example seems to have feedback in there somewhere. There is also a kind of feedback that happens after the note clearly drops off (guitarists will sometimes try to keep the note sustaining by wiggling the string, then the note will end, and then the feedback will begin.

 

[jesussaddle]

What I decided was to try and use Fabfilter's Pro G, since it has an internal sidechain feature that essentially allows one to trigger just by a portion of the frequency spectrum so I can keep it from opening and closing sporadically. (Also it has a dual threshold, so that the decibel level on which the gate closes is lower than the threshold on which the gate opens. (Nicer would be the ability, as in Gullfoss, to have an "ends or center" filtering mechanism, so I could trigger on "all but" a certain frequency range, as well as triggering on a certain frequency range. Can't be perfect.)

 

[Bakerman]

The gate/expander block has low cut & high cut parameters for the detector. You could also use EQ a parallel row and use that as sidechain.

Separate rise/fall thresholds could be done by auto-engaging a volume/filter boost in a sidechain row via envelope, with the modifier otherwise doing nothing e.g. a balance control that stays centered. This works due to autoengage on/off thresholds differing by 5%. Normally this would limit the threshold difference to 4 dB but if you assign envelope to its own gain parameter (use FracPad) you can work around this by reducing gain as envelope value rises.