(SOLVED) Feedback Send/Return FW21.01

TubeAmpNut

TubeAmpNut

Inspired
Not sure if this recently broke with 21.01 or earlier, but I noticed my patches where I send a signal back using the feedback send/return seem broken (the feedback blocks). Would anyone mind cooking a simple patch with feedback through a phaser or pitch block then back into the delay and verify you are feeding the signal back as external repeats?

thanks
 
Feedback blocks are working correctly from what I can see.

BK-Amps said:
Would anyone mind cooking a simple patch with feedback through a phaser or pitch block then back into the delay and verify you are feeding the signal back as external repeats?
Click to expand...
This doesn't make sense to me.

Instead of asking us to create a preset, share the one you are having problems with. It means nothing if we create a preset that works, but the developers can fix the problem if your preset demonstrates a true problem.
 
I worked it out. I had phase invert enabled on the delay (2290 style) and it was cancelling itself out in the feedback when it got back to the delay since it wasn't stereo which negated any feedback at all..
 
GlennO said:
+1, but it's the return block. That gets me every time. There must be a better way.
Click to expand...
it used to be 100% by default, but there were many complaints about possible damaged speakers when the return was added somewhere sub-optimally, creating instant feedback.

with the current implementation, the user must ramp up the knob which could clue into feed back as it's being turned up, and at a lower level.

if the block is added correctly, 100% default could work. but it could be added incorrectly leading to the issue.
 
chris said:
it used to be 100% by default, but there were many complaints about possible damaged speakers when the return was added somewhere sub-optimally, creating instant feedback.

with the current implementation, the user must ramp up the knob which could clue into feed back as it's being turned up, and at a lower level.

if the block is added correctly, 100% default could work. but it could be added incorrectly leading to the issue.
Click to expand...
Yes, I'm familiar with the history. There are ways to address it without confounding the user though. For example, only turn it down upon insertion when there is a feedback loop. 99% of the time the send/return pair are inserted as a row wrap where there is no danger of feedback at the time they are inserted. Or simply display an alert that notifies the user the return value has been turned down for their safety.

I consider myself fairly familiar with the Axe-FX, but this trips me up frequently. This often comes up on the forum as a source of confusion, so I'm not alone. I'm just suggesting maybe there's a way to improve this.
 
chris said:
it used to be 100% by default, but there were many complaints about possible damaged speakers when the return was added somewhere sub-optimally, creating instant feedback.

with the current implementation, the user must ramp up the knob which could clue into feed back as it's being turned up, and at a lower level.

if the block is added correctly, 100% default could work. but it could be added incorrectly leading to the issue.
Click to expand...
Hmm, makes me wonder if the feedback situation could be automatically detected by analyzing the connection graph (before inserting) and/or monitoring levels live (automatic feedback suppressor).
 
Greg Ferguson said:
Laughing … such a delicate way of saying they screwed up. :)

I have screwed up enough times to treat the return block with kid gloves.
Click to expand...
whenever I come across it, I have to chuckle, imagining Cliff possibly yelling at some junior coder: "default it to 0!, so they can't blame us fur blowin up their speakers!" 🤣
 
Ther
sprint said:
whenever I come across it, I have to chuckle, imagining Cliff possibly yelling at some junior coder: "default it to 0!, so they can't blame us fur blowin up their speakers!" 🤣
Click to expand...
There is no winning. 0% gets complaints, 100% got them, 50% would too.

I understand why for each case, and I understand why Fractal opted for 0%, because, in the long term, it is safest for the gear. Let everyone else sort it out.
 
Greg Ferguson said:
Ther

There is no winning. 0% gets complaints, 100% got them, 50% would too.

I understand why for each case, and I understand why Fractal opted for 0%, because, in the long term, it is safest for the gear. Let everyone else sort it out.
Click to expand...
yup - understandable

I got the itch to make a Tom Scholz hyperspace preset a long while ago and was going to post a version of it for comments, but changed my mind with the thought that: hmmm - this thing could take off on somebody n I'll be in hot water!
 
Last edited:
sprint said:
yup - understandable

I got the itch to make Tom Scholz hyperspace preset a long while ago and was going to post a version of it for comments, but changed my mind with the thought that: hmmm - this thing could take off on somebody n I'll be in hot water!
Click to expand...
Been in the same spot.
 
sprint said:
I got the itch to make a Tom Scholz hyperspace preset a long while ago and was going to post a version of it for comments, but changed my mind with the thought that: hmmm - this thing could take off on somebody n I'll be in hot water!
Click to expand...
Yeah, that's why a configurable hard limiter setting (global by output # and/or an option in the output block) would be great to have in general: https://forum.fractalaudio.com/threads/output-block-hard-limiter.174260/.

Or similarly, a clamp that brings any out-of-range level way down to some ~reasonable~ level above nominal (e.g. +12 dB).

This wouldn't stop the offending feedbacking signal from growing and getting crazy, but it would limit the final output level.

EDIT: And/or: a limiter/clamp in the Return block - which could keep the FB in check to some degree.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom