Effect used.

Posted: Wed Feb 01, 2006 1:18 pm

Ok, now that you've gotten me thinking about phasing vs. flanging, two approaches come to mind...

The net result - what the listener hears with "Flanging" is something that's done as if by magic, originally created in a way that did not involve today's real-time processing; and it's tougher to pull off. (Hint: it involves early as well as late information, as the second copy passes against the common/null/in-phase material).

With Flanging, it's the entire sonic spectrum being affected, and when done correctly, it's a very full, deep effect, across the entire bandwidth. You can have the second (flanged) track actually begin early or late - doesn't matter which - but it's these crossing points (various freqs being accented/reduced) as the timing offset changes that make the sound we're all talking about. It's often done in a very slow sweep, to let one hear the beautiful and "Trippy" result. And, with the forwards/backwards capability of creating the effect offline, one can get some nice effects as the material goes in and out of "phase" - slowly, going forward and backward in time.

It's a tougher sound to pull off - and again, because it involves late AND early timings should you want to create that effect properly. For example, if you start the second (copy) track EARLY, it's not something you can actually pull off in real time or live performance, obviously. (Unless you're some kind of time traveller who can go back in time, even by a few milliseconds.)

With something as simple (in todays world) as a varispeed, the flanging could be very subtle, as an engineer would cue up two machines with the same start point - locked to each other, for example. Then by turning OFF the sync lock, and slowly adjusting the varispeed (up or down), the real "Flanging" effect begins.

What I'm saying is that good flanging can move in either direction; the "early" (new) track begins early and the effect sweeps in one direction as it falls back in time, and after it falls far enough behind, it starts to go back the other way.

So, when creating a true "flanging" effect, it's all about making one copy of the track go back and forth against a common null point. Of course, good software can emulate this trick, and I can think of a few way to do this with copies of wav files that run at different speeds, and crossfade them at will.

PHASING, on the other hand, (at least as I recall the stomp boxes and rack units called "Phasers"), were a much simpler frequency-based delay circuit - going back and forth, but always just going from "Null" to delayed, and never (obviously) starting early. Due to the kind of circuitry, this was just a swirly, thinner-sounding wash. (Related to, but different than "Chorusing")

Also, very often much of the "Phase-shifter" box effect is usually applied to one instrument alone; not the whole mix-as done with tape in the studio.

As affordable bucket-brigade delay chips came on the market, they did get closer to real "Flanging", but they didn't have the bandwith of real tape, and again, when you do an effect like this in real time, you can only go from "Null" to "delayed", never having the option of "Early" to "Null" to "Delayed". (There's that time travel issue again.)

So, in spite of all the misnomers floating around out there, for my own reference, I always think of true flanging as something done in the studio by pro's, and phasing as something done on the fly with a simple, repetitive cycling box, with boring predictable results.

I could be all wrong on this, but that's how I think it goes...

Give me REAL flanging anyday. Thumbs Up

Posted: Wed Feb 01, 2006 1:29 pm

JoeH wrote:

Take a few moments with google and search for "Roy Thomas Baker" - you'll be amazed at what's out there. (RTB was teaching a master class in 2004, no idea what he's up to lately, but certainly worth a look-see.)

He recently produced The Darkness' latest album "One way ticket to hell.... and back"

Posted: Wed Feb 01, 2006 1:53 pm

JoeH wrote:

Ok, now that you've gotten me thinking about phasing vs. flanging, two approaches come to mind...

The net result - what the listener hears with "Flanging" is something that's done as if by magic, originally created in a way that did not involve today's real-time processing; and it's tougher to pull off. (Hint: it involves early as well as late information, as the second copy passes against the common/null/in-phase material).

With Flanging, it's the entire sonic spectrum being affected, and when done correctly, it's a very full, deep effect, across the entire bandwidth. You can have the second (flanged) track actually begin early or late - doesn't matter which - but it's these crossing points (various freqs being accented/reduced) as the timing offset changes that make the sound we're all talking about. It's often done in a very slow sweep, to let one hear the beautiful and "Trippy" result. And, with the forwards/backwards capability of creating the effect offline, one can get some nice effects as the material goes in and out of "phase" - slowly, going forward and backward in time.

It's a tougher sound to pull off - and again, because it involves late AND early timings should you want to create that effect properly. For example, if you start the second (copy) track EARLY, it's not something you can actually pull off in real time or live performance, obviously. (Unless you're some kind of time traveller who can go back in time, even by a few milliseconds.)

With something as simple (in todays world) as a varispeed, the flanging could be very subtle, as an engineer would cue up two machines with the same start point - locked to each other, for example. Then by turning OFF the sync lock, and slowly adjusting the varispeed (up or down), the real "Flanging" effect begins.

What I'm saying is that good flanging can move in either direction; the "early" (new) track begins early and the effect sweeps in one direction as it falls back in time, and after it falls far enough behind, it starts to go back the other way.

So, when creating a true "flanging" effect, it's all about making one copy of the track go back and forth against a common null point. Of course, good software can emulate this trick, and I can think of a few way to do this with copies of wav files that run at different speeds, and crossfade them at will.

PHASING, on the other hand, (at least as I recall the stomp boxes and rack units called "Phasers"), were a much simpler frequency-based delay circuit - going back and forth, but always just going from "Null" to delayed, and never (obviously) starting early. Due to the kind of circuitry, this was just a swirly, thinner-sounding wash. (Related to, but different than "Chorusing")

Also, very often much of the "Phase-shifter" box effect is usually applied to one instrument alone; not the whole mix-as done with tape in the studio.

As affordable bucket-brigade delay chips came on the market, they did get closer to real "Flanging", but they didn't have the bandwith of real tape, and again, when you do an effect like this in real time, you can only go from "Null" to "delayed", never having the option of "Early" to "Null" to "Delayed". (There's that time travel issue again.)

So, in spite of all the misnomers floating around out there, for my own reference, I always think of true flanging as something done in the studio by pro's, and phasing as something done on the fly with a simple, repetitive cycling box, with boring predictable results.

I could be all wrong on this, but that's how I think it goes...

Give me REAL flanging anyday. Thumbs Up

I think you're just describing different applications of the same basic principle. Two signals out of phase is two signals out of phase, essentially, regardless of any of that other stuff.

Moving a track "forward" in time to knock it out of phase is mathematically the same as moving it backwards by the same amount. It's still two tracks with a given amount of time between them.

Now if you really were mashing a pencil on a tape reel, you'd be CHANGING the amount of phase change between the two signals somewhat randomly, but to me it's still just a variation on the same effect, not a new effect in principle.

To make an analogy, "room" and "cathedral" are not two effects, they're two different ways of doing "reverb" that sound completely different.

I'm still not convinced, though I agree with your preference to expensive, better sounding flangers over cheap stomp boxes.

Really, I think "phasing" is a bad term anyway. Chorus effects are kind of the same thing and use phase differentials as well. So unless I trip up, I almost always call what we're talking about flanging anyway.

~S

Posted: Wed Feb 01, 2006 6:01 pm

You're right, Shotgun, it's mostly about terminology and perception, no matter how I described its creation.

The basic principle is SUPPOSED to be the same thing, I suppose....The net result of the "flange" effect I'm describing from the old days with tape is (Duh!) Phase Shifting - cancellation or reinforcement is what you hear.

In theory, that's what a "phaser" should do too. But of course, we all know the net result is different, because with good audio, there is no free lunch.

So, maybe although the goal is the same, the terms have taken on new subjective meanings based on the way they each perform? (I still think there's more to be had from the early phase shift you get from tape, because your ear is tracking the EXISTING audio, and when the new (early) audio begins to come in, it's sweeping in a different direction (up?) than it is when it's going back out of time (Down?) It's one of those things (like good visual effects in movies) that you know it when you hear it, and the cheap-o imitations are just that. Never as good.

The not-so-subtle difference here, including the full bandwidth you get with tape (and not chips or stomp boxes) separates the toys from the professional gear. To my ears, it's really obvious.

And once again to restate the obvious; flanging seems to be done more on the entire track, soup to nuts, while phasing is often applied to one instrument or track at a time, usually within a mix.

Posted: Thu Feb 02, 2006 8:44 am

If anyone cares, I just looked this up in Zölzer's Digital Audio Effects book. Regardless of what they sound like, there are implementation differences between the two.

A phaser can be implemented as summing the output from one or more notch filters with the dry signal. Another method uses allpass filters instead of notch filters. They can also use a feedback loop for added effect. This can also be done fairly easilly with analog filters.

A flanger is usually implemented as a delay line / comb filter, where the delay time is modulated with a very low frequency signal. This is a decent approximation of what the mechanical effect actually does.

Posted: Thu Feb 02, 2006 8:52 am

mpd wrote:

If anyone cares, I just looked this up in Zölzer's Digital Audio Effects book. Regardless of what they sound like, there are implementation differences between the two.

A phaser can be implemented as summing the output from one or more notch filters with the dry signal. Another method uses allpass filters instead of notch filters. They can also use a feedback loop for added effect. This can also be done fairly easilly with analog filters.

A flanger is usually implemented as a delay line / comb filter, where the delay time is modulated with a very low frequency signal. This is a decent approximation of what the mechanical effect actually does.

And there ya go. When in doubt, read a book. Nice work MPD.

~S

Posted: Thu Feb 02, 2006 9:05 am

I could readily hear the differences between the 2 (phasing vs. flanging), and apparently so could enough engineers and producers during the 70s. After all, Eventide offered the Instant Phaser AND the Instant Flanger...
Flanging is reproduced by "bucket-brigade" type delays modulated by an LFO (sometimes a VCO), and there IS pitch-shifting involved as a result.
It is also referred to as a Doppler shift (Doppler noted that train whistles seemed to change pitch as the train changed distance in relation to the listener)...
Phase shifting is created by a notch filter being swept by an oscillator, and I've never noticed a pitch shifting with that. Phasing is a much simpler effect to generate. There is a world of difference between the sound of my ancient Maestro (by Oberheim) Phase Shifter and my ADA Flanger, and upon opening them up, a world of difference in their guts!

This topic sponsored by: