How to create a feedback loop in Reason (Noob Question)

[mravdirato]

Hey peeps,

Can I ask how to create a feedback loop in Reason? I got one of my local drumming heroes who happens to be a Reason User invited me to his studio to recreate it using a Combinator. He wants the sound source to be drum loops. I was thinking maybe I can use a Dr. OctoRex as the source for loops. Any suggestions on how I can do it? Thanks!

[deeplink]

Question unclear.

I could explain what a feedback loop is, and how to do it - but it will bare no significance here and I cannot see what the use is yet.

Maybe instead you want to create a beat reapeater / looper of some kind?
Or your drummer mates wants to listen / monitor to his drums while playing?

[mravdirato]

Hey,

(I knew it, the question was unclear ugh sorry XD)

Anyway, I think this will be easier.

He wants to emulate this, only the audio source is drum loops inside Reason:

https://www.instagram.com/reel/Coc2B-_J ... o9Qf1xr4V4

And then emulate this pedal:

[deeplink]

It's difficult to exactly model such a device, the easiest route is just to use Scream on its "feedback" distortion setting.

Otherwise to go a step further, route the output of the Scream to a spider audio merger and splitter and back into the Scream - use a line mixer in the signal chain to control the amount of feedback.

This is dangerous though and you could blow a speaker, so have a limiter or clipper on your master.

[mravdirato]

It's difficult to exactly model such a device, the easiest route is just to use Scream on its "feedback" distortion setting.

Otherwise to go a step further, route the output of the Scream to a spider audio merger and splitter and back into the Scream - use a line mixer in the signal chain to control the amount of feedback.

This is dangerous though and you could blow a speaker, so have a limiter or clipper on your master.

Thanks! I know this is a long stretch but can you show it through a screenshot? I really suck at like text instructions XD

[deeplink]

Here you go;

[selig]

Ah, you want to do sourceless/"no input" effects. This is one area where analog has a huge (light speed actually) advantage over digital.
For something to feedback in the digital world you need AT LEAST one sample of delay because you can't feed something back until after it happens. So the first sample is read, THEN routed back to the input but isn't actually processed until the second sample occurs. Already you are one sample behind, and that's the best case scenario since some processes will impart additional latency on the signal. In some case this doesn't matter but in many it does because that latency is added with each loop.
Other differences: Analog electrical signals travel at the speed of light, so feedback is near instantaneous with an analog circuit. Also, there is noise, which can sometimes help stimulate a response in the system (aka, "get things going"). And finally, analog systems are going to clip with high levels such as you get with feedback, and this is a big part of how you control runaway feedback in an analog system.

I'm not saying it can't be done, just that it's not going to be the same and possibly not as useful or controllable IMO. Because of this I've never bothered to mess with it in Reason. That said, a clipper in the feedback path (and a high pass filter as well) is all but essential in such experiments, so I'd start there for sure!

But IMO it's the imperfections of analog that make that technique so interesting - because of this it can sound very different with each different setup/routing you use, so there's tons to explore right there.

[Jac459]

Ah, you want to do sourceless/"no input" effects. This is one area where analog has a huge (light speed actually) advantage over digital.
For something to feedback in the digital world you need AT LEAST one sample of delay because you can't feed something back until after it happens. So the first sample is read, THEN routed back to the input but isn't actually processed until the second sample occurs. Already you are one sample behind, and that's the best case scenario since some processes will impart additional latency on the signal. In some case this doesn't matter but in many it does because that latency is added with each loop.
Other differences: Analog electrical signals travel at the speed of light, so feedback is near instantaneous with an analog circuit. Also, there is noise, which can sometimes help stimulate a response in the system (aka, "get things going"). And finally, analog systems are going to clip with high levels such as you get with feedback, and this is a big part of how you control runaway feedback in an analog system.

I'm not saying it can't be done, just that it's not going to be the same and possibly not as useful or controllable IMO. Because of this I've never bothered to mess with it in Reason. That said, a clipper in the feedback path (and a high pass filter as well) is all but essential in such experiments, so I'd start there for sure!

But IMO it's the imperfections of analog that make that technique so interesting - because of this it can sound very different with each different setup/routing you use, so there's tons to explore right there.

I am not really sure you need this sample of delay for all implementations. You could add more latency and output after no? Or you could process the beginning of the first sample with the beginning of the second sample.

Bitwig has an audiorate modulator that can be fed by anything including the own output of the instrument on which you put it. That can create very interesting effects and I never really noticed delay. You can also use this modulator to feed a dc output to regenerate the sound with whatever effect/transfo you want in between.