extracting the effect from the effect

kind of a weird question here. is it possible / how can one separate an unaffected signal, from the effected signal, after its been effected?

what i want to be able to do, is capture the essence of an effect, without hearing the original signal that later goes into the effect. so for example, you know how sometimes the flanger goes "boing", i want to record it doing that, but i dont want to hear the sound that was put into the flanger to begin with. or reverb, i want to send, say a clap, into a reverb, but i dont want to hear the clap, just the reverberated clap. i know this isnt always going to be possible, for effects such as a phaser, which is just a phasing stereo signal am i right? so impossible to get that phaser sound without putting a noise into it right? same with chorus? i know this is kind of weird, which is why i asked if its possible

i have previously recorded a flanger making the boing sound without sending a signal into it, but i think it was just some kind of feedback loop going on. which begs another question, can you hear effects without putting in a signal at all?

im speaking from a software point of view, but welcome thoughts on hardware as well

wet. dry.

if youre saying what i think youre saying, that would only apply to effects that have separate wet and dry signals. dont most effects only have a single wet/dry function?

right?

  On 2/20/2011 at 7:36 PM, ET said:

if youre saying what i think youre saying, that would only apply to effects that have separate wet and dry signals. dont most effects only have a single wet/dry function?

 

right?

years ago i managed to make the intro to Aphex's Milkman actually legible to the ear, by putting it through some filters and a few passes of harmonic rotation in Bias' Peak software.

it didn't glean a perfectly dry signal, but it made it a hell of a lot more legible.

what about not using the effect as an insert, but as an auxbus?

But a flanger normaly has wet and dry. As has Chorus, delay, reverb, etc...

It should work the same way like extracting vocals from a song you have the instrumental version from only that you do it the other way round. I don't remember for this way of doing it so

If I am understanding you correctly you want a signal that is 100% wet. I'm pretty sure most, if not all effects have this setting. Is there something I'm not seeing here?

  On 2/20/2011 at 8:09 PM, tonfarben said:

what about not using the effect as an insert, but as an auxbus?

But a flanger normaly has wet and dry. As has Chorus, delay, reverb, etc...

could you explain to me how i might do this? (ableton)

  On 2/20/2011 at 8:24 PM, Gocab said:

If I am understanding you correctly you want a signal that is 100% wet. I'm pretty sure most, if not all effects have this setting. Is there something I'm not seeing here?

ok guys read my first post again. i just want the sound of the effect, without any of the pre-effected signal coming through. like i said, imagine a clap with a big reverb, but without hearing the clap

the only thing i know that does something similar is some of the blue cats plugin effects.

do you see how it has a separate slider for dry and wet? you can pull the dry slider all the way down and the wet all the way up, and it kind of isolates the dry signal from the wet, but not completely, or not to my liking at least

  On 2/20/2011 at 7:38 PM, oscillik said:

years ago i managed to make the intro to Aphex's Milkman actually legible to the ear, by putting it through some filters and a few passes of harmonic rotation in Bias' Peak software.

 

it didn't glean a perfectly dry signal, but it made it a hell of a lot more legible.

You can decode Milkman using only a ring modulator, the effect that was used to modulate the voice in the first place. The reason being that ring modulation is linear in the manner that it modulates signals. Ring modulation is simply a multiplication of the volume domain so all you need to do is divide it by the inverse amount.

You may run into problems decoding ring mods if aliasing occurred during the initial mod process when the generated bands exceeded 22 kHz. In that case you will still be left with undesirable artifacts after you reverse the ring mod process.

what you call an effect is a signal processor. there's no sound of an effect.

put the effect you wanna use in a Return track and use the Send turning knobs to determine huw much effect you want

I do this for my reverb mostly, you have your dry sound and your wet sound in a different channel.

and if you just want your wet sound you link your Return track to an audio track

INSERT

AUX/SEND

exactly like you said - i want to send, say a clap, into a reverb, but i dont want to hear the clap, just the reverberated clap

exactly like you said - i want to send, say a clap, into a reverb, but i dont want to hear the clap, just the reverberated clap

you do that trough the aux/sends channel, then you'll have two tracks, affected and unaffected, just mute the unaffected one. done.

wicked, you guys are fucking awesome. thanks a lot!

ill let you know how it comes along

if something has only one dry/wet, 100% setting will do what you want, no send/soloing, turn the knob all the way up. Boom!

So what's the point in sending the signal to other track with effects and muting the one with no effects if you can just turn the dry/wet knob?

because even at 100% wet, you'll have the source signal coming through on most fx.

  On 2/20/2011 at 11:54 PM, Adam Beker said:

So what's the point in sending the signal to other track with effects and muting the one with no effects if you can just turn the dry/wet knob?

better control of the effect i think? also sounds better when layering the wet and dry then to be fucking around with the dry/wet knob wich can fuck up your sound.

and saves cpu if you use the same effect on different channels and stuff

I'm a n00b tho so could be wrong

Yeah I don't think people are getting it, we be talking about HEARING ONLY THE EFFECT and NOT ANY OF THE SOURCE MATERIAL TO BE AUDIBLE/NOTICEABLE in the end result. I'm going to look into this and try the routing ideas mentioned previously but I wouldn't surprised if this is almost impossible save for extremely rare VSTs.

the whole idea about having an independent "affected" track (trough aux/sends) is so that you get able to apply further effects solely on this track without effecting the "original unaffected" track.

let's say you have a reverberated snare and a kick compressing it trough sidechaining (to get that "sucking" effect on the snare), if you apply the compression to the original snare, you'll not ear the attack of the snare when the kick punches at the same time, but if you have applied it solely on the reverberated snare, it will work like a charm.

lol, i actually never tried that

it kind of depends on the effect. If you want the sound of a flanger without there being source material, turn the feedback all the way up and put a short noise burst through it, or just continuous white noise if you want less of the boing and more airey flanger noise.

P.S. effects don't generate sound on their own(in most cases), they depend entirely on a source signal. That's what effects are.

^ word. this whole concept blows me mind. i don't quite understand the point but maybe running white noise through an effect IS what you want?

ET is 3/4ths troll & 1/4ths troll but I would think the idea could work and could sound interesting. Upon thinking about it a bit more it does seem to be nearly impossible for most FX.

  On 2/21/2011 at 2:01 AM, sneaksta303 said:

because even at 100% wet, you'll have the source signal coming through on most fx.

You'll have the same signal coming trough anyway. That's how effects work. You'll just have two tracks from which one is muted.

  On 2/21/2011 at 2:13 AM, Ranky Redlof said:

  On 2/20/2011 at 11:54 PM, Adam Beker said:

So what's the point in sending the signal to other track with effects and muting the one with no effects if you can just turn the dry/wet knob?

better control of the effect i think? also sounds better when layering the wet and dry then to be fucking around with the dry/wet knob wich can fuck up your sound.and saves cpu if you use the same effect on different channels and stuffI'm a n00b tho so could be wrong

I'd agree about better control of the effect. I use send with reverb so i could eq dry and wet separetly myself. But in this case it's pointless because ET only wants wet. You can get it by just turning the dry/wet knob. Also, how dry/wet knob can fuck up your sound???

I don't really understand anything anymore. If ET wants a sound of an effect without hearing the source material at all that is imposible because as I said there's no sound of an effect. It's a signal procesor not generator. Try runing silece trough an effect. If ET wants just a sound of an affected track that's what for is dry/wet knob. Btw if you run white noise through an effect you'll get white noise with that effect.

I guess ET is just troling.

Dunno, when I put down the dry on my delay effect and put up the wet. What happens? The original tone is gone. I only can hear the delayed tones getting through. No need for auxsend and auxreturn. I use buzz. Is it different in Cubase/logic/ableton/all the other vst hosts? It should not be! I don´t get why thi9s thread needs two pages. Enlighten me, please...