How does changing phase by say, 40 degrees work?

C

composition4

New member
I know there are plugins for this type of thing, but how do they actually work?

I understand the science behind flipping the phase by 180 degrees... make the positive amplitude negative, and vice-versa. But how do you change by say, 90, or 40, or any other number?

Is it a time-based thing?
 
phase is a proponent of time. 180 degrees is half the cycle (360 being a full cycle). So by shifting the signal 180 degrees out of phase, you've effectively moved it half of it's cycle. 90 degrees would be half of half...or a 1/4 cycle. 45 degrees would be half of that....etc.
 
Yep.. i understand that applies for a fixed-wavelength signal..

but in a complex signal such as most of the stuff we record, it is made up of many different frequency/wavelengths and amplitudes. It's not constant..

Am I making sense?
 
well, first remember that in order for a waveform to be considered out of phase (by any degree) it has to involve a second waveform. A single waveform cannot be in or out of phase with itself, there has to be two.

so two exact wave forms playing together...and one shifted slightly out of phase, can still be considered 90 degrees out of phase. No matter if it's complex or not.

Now, perhaps you're talking about two completely DIFFERENT waveforms...that have differing/altering phase relationships (this is a GOOD thing...it's a natural occurence with EVERYTHING around us).

This is similar to how a phase effect works. With a phaser, a copy of the input signal is made and sent through an all pass filter resulting in a frequency dependent effect on the phase of each frequency. This resulting waveform also keeps the same amplitude as the original. However, because of the way this is done, this can cause certain frequencies to be 90 degrees out of phase, and others to be 45 degrees out of phase...depending on the settings. When this is combined with the unaltered wave, the phase differences between them now cause peaks and dips at various frequencies. The degree of phase shift is modulated periodically....so you really can't pinpoint and say the two waves are "40 degrees out of phase," because it varies continuously.

Again, this is how two different complex wave forms interact. They cause certain dips and peaks in the summed signal giving it it's certain characteristics.
Another good example would be acoustics in the room around you and why people worry so much about acoustically treating a room. Sounds reflect all around you and certain frequencies clash with eachother causing peaks and dips...some frequencies can even be cancelled 100% if they hit eachother 180 degrees out of phase! Other frequencies can be boosted 6-10dB just by interacting with another wave. It's all phase relationships.
 
oh.. i just thought it through.. of course. if i shift it by say, 10 samples, it will.. ok, gotcha.

so these phase shifting programs are just delays? how do they know how much to delay the signal to get it a certain amount out of phase?
 
composition4 said:
so these phase shifting programs are just delays? how do they know how much to delay the signal to get it a certain amount out of phase?
Click to expand...

not really, however that's how analog phasing was done with tape back in the day....playing two copies of the signal at slightly varying speeds.

phaser effects imploy different settings within the plugins. Rate, depth, feedback, min/max frequency values, intensity, spread...etc....all of these things alter the original signal and then combine it back with the original. The input signal is modulated with the help of an LFO.

stereo phasers can also be altered even more by shifting the phase of the Left and Right channels a smidge.
 
sorry... just to be clear, i'm not talking about a phaser effect... i'm talking about a fixed-rate phase shift.. somewhat like Betabugs PhaseBug. The phase shift is constant in that
 
i don't think it actually knows, i thought you had to put in a value by which you delay it, but maybe i'm wrong or haven't had experience in what your talking about. Alternatively you could just zoom in on the waveform and shift it back/forward or invert it completely, etc...
 
the Phasebug requires another audio waveform in order to be effective. With the stereo version it utilizes the other channel.

The numbers (in degrees) you are seeing are in reference to itself. It considers the original (left channel) to be 0 degrees, so if you shift it 180 degrees all it's saying is that it's 180 degrees out of phase with itself (still the left channel). Same with the mono channel version. But with that, you still need another channel. So that plugin is not a frequency dependent phaser that is modulated with an LFO....it's just shifting the wave form in time.
 
bennychico11 said:
The numbers (in degrees) you are seeing are in reference to itself. It considers the original (left channel) to be 0 degrees, so if you shift it 180 degrees all it's saying is that it's 180 degrees out of phase with itself (still the left channel).
Click to expand...
I have to admit to not understanding this either.

How does one "rotate" a non-repeating waveform by a specified number of degrees without first having a definition of cycle length? In order to be able to define a degree, one needs to first define the length of the cycle. A sine wave that's one degree out of phase from it's original means that it's shifted by a distance equivalent to 1/360th of it's wavelength, or 1/360th of one full cycle.

But in a case of a "normal" sonic waveform loaded with shifting fundamentals, harmonics, cancellations, etc., all in a non cycle-repeating form, how is the cycle length defined? Is there some arbitrary reference frequency (middle C? A440?) that's used as the baseline frequency/wavelength measure?

It's easy to talk about 180° and 0°/360° because those are special case situations that are not dependant upon any cycle length definition. 0°/360° is just another way of saying "identical", and 180° is accomplished by simple polarity inversion that requires no "rotation". But determining intermediate phase values of a random-amplitude waveform without being able to define it as a rotation through a base frequency menasurement is something I have yet to learn.

G.
 
Last edited:
SouthSIDE Glen said:
It's easy to talk about 180° and 0°/360° because those are special case situations that are not dependant upon any cycle length definition. 0°/360° is just another way of saying "identical", and 180° is accomplished by simple polarity inversion that requires no "rotation". But determining intermediate phase values of a random-amplitude waveform without being able to define it as a rotation through a base frequency menasurement is something I have yet to learn.

G.
Click to expand...
Inverting a wave form and a 180° shift (as does 360° or any degree) are not the same. Degrees (e.g. 180°) implies a time difference and Inverting does not. 180° shift and Inverting are often incorrectly interchangeably used. One cannot speak of degrees without knowing the frequency. A 360° degree shift implies time moving the wave one complete cycle based on the frequency.

I'm not clear about your use of the term rotate, other than a Sine Wave can be represented as a rotating phaser which is used in engineering.
 
Keiffer said:
One cannot speak of degrees without knowing the frequency.
Click to expand...
That's exactly my point.

As I understand it, the only time one can speak of "degrees of phase shift" is when one is referring to a wave of specific frequency, as with a pure sine wave, square wave, etc. or with a complex waveform that repeats itself at a defineable frequency.

But any time one moves to a complex *non-repeating* waveform that's an arbitrary mix of frequencies, any talk of "degree" of shift is meaningless, because there is no baseline reference or definition of "degree". One needs to have a set definition of wavelengh/frequency to be able to define the "length" of one degree of phase.

It seems to me at that point one can only talk in terms of time delay - e.g. "a 5ms shift" - and cannot talk of "x" degrees of phase.

As to the 180° vs. polarity thing, you are absoluetly right, which is what I was trying to say in an inside-out way. On a complex waveform, one can acheive the effect of "phase cancellation". This is often referred to 180° out-of-phase. But in reality, since on a complex waveform one cannot define it by degrees of phase - there is no rotation through phase because the wave is non-repeating - such cancellation is acheived through simple polarity inversion; as you say, a totally different mechanism. It is therefore a misnomer to refer to it as a 180° phase shift, since no "shift" (rotation) takes place.

G.
 
I see what you're getting at Glen, and I agree somewhat
I think, however, this is where it leads us into physics, wave theory and the complex plane. I had a brief discussion with cpl_crud about this a little while back here. Some math and a long read.
Basically I got that he was saying that you can't forget that we are dealing with complex wave forms where phase is depenant on angle and not time. It's just manifested as a time difference on the X/T graph.


With regards to the Phasebug, I was just basically reading from the manual and explaining how it interprets it...the programming could be a bit more complex than that.
 
bennychico11 said:
I think, however, this is where it leads us into physics, wave theory and the complex plane. I had a brief discussion with cpl_crud about this a little while back here. Some math and a long read.
Basically I got that he was saying that you can't forget that we are dealing with complex wave forms where phase is depenant on angle and not time. It's just manifested as a time difference on the X/T graph.
Click to expand...
Wow, great thread!

I guess I'm just going to have to read up on wave theory to understand how one can calculate degree of phase without first having a static frequency as a given value. Or, to put it in crud's terms, "how to determine the phase angle of a non-periodic waveform".

hmmm...sounds like a good Teoma search term to me ;)

I can easily picture it on the complex plane, but how that translates to real numbers I have yet to figure out

(BTW, don't EVEN get me started on Wikipedia...I've been holding my tongue on that subject for a few threads now, I don't want to crack just yet ;) :p )

G.
 
SouthSIDE Glen said:
I can easily picture it on the complex plane, but how that translates to real numbers I have yet to figure out
Click to expand...


me either...but I guess I never studied particle physics. I just press random buttons and it all works. :D
 
bennychico11 said:
I see what you're getting at Glen, and I agree somewhat
I think, however, this is where it leads us into physics, wave theory and the complex plane. I had a brief discussion with cpl_crud about this a little while back here. Some math and a long read.
Basically I got that he was saying that you can't forget that we are dealing with complex wave forms where phase is depenant on angle and not time. It's just manifested as a time difference on the X/T graph.


With regards to the Phasebug, I was just basically reading from the manual and explaining how it interprets it...the programming could be a bit more complex than that.
Click to expand...
Geez... this is way overly complicating this. The problem here is, with a single source (a must) how does one go about figuring phase differences with a complex wave like a distorted guitar. The key word is single source. Without it you don't get the frequency enforcement/cancellation due to time (distance) differences... stuff bouncing all over the place and arriving later and/or another mic some distance from the primary. One can really only concentrate on the fundamental frequency and it's harmonics. The rest is irrelevant.
 
SouthSIDE Glen said:
... but how that translates to real numbers I have yet to figure out
Click to expand...
It doesn't... one has to break the wave down into it's components and treat them individually. Complex waves are a generally, but not always, series of coherent sine waves.
 
You must log in or register to reply here.
Back
Top