Voxengo Premium Membership - All Voxengo Plugins For a Fixed One-Time Fee
Forums     Plugins     PHA-979 Some confusing words in the manual


Hi, just new to phase vs. time adjustment topic.  It's said in PHA-979's user guide that "What is meant by the phase shift here is simultaneous shifting of all frequencies across the active frequency range of the signal by the given value in degrees.  This is achieved by linear-phase design."

What confuses me is "linear-phase design".  Acording to Wiki, "Linear phase is a property of a filter, where the phase response of the filter is a linear function of frequency.  The result is that all frequency components of the input signal are shifted in time (usually delayed) by the same constant amount, which is referred to as the group delay.  And consequently, there is no phase distortion due to the time delay of frequencies relative to one another." Then "linear-phase shift" equals to moving the whole audio clips forward or backword on the timeline, right?

But PHA-979 shift different frenquencies in a signal by the same phase degree, that's equal to shifting by different time amount.  Then here comes the puzzle.  Do Voxengo and Wiki use the same words but with different meanings?

And quoted from a post here "Since all frequencies get shifted for the same amount of deg. you won't hear any obvious change in sound because phase-relations between frequencies stay untouched." Maybe we can't hear any obvious change in sound.  But to my understanding, this does change the phase-relations between frequencies.  Since each frenqucncy is shifted by the same degree, that equals to shifting/delaying by different time amounts.  Before shifting, two frequencies may both start from 0 deg at the same time point, but after phase shift, while one frenquency starts from 0 deg, another is already at 90 deg at the same time point.  Can this be called "no phase-relation change"?  My understanding acording to Wiki, this does cause phase distortion.

Some insight please.  Thanks.


I think of the action of the PHA-979 in a certain way which may be of assistance in your understanding.  I understand that Aleksey Vaneev will correct me if I am wrong.  For the most powerful and precise presentation of recorded sound, you usually want the sound to be firing directly at you from the speaker. (You would not normally wire your speakers in 180-degree reversed phase-polarity.  This would put the "thrust" of the sound into the cabinet, not out at you.)

For my purpose, imagine the "phase direction" of the sound coming from your speakers as if it were the light coming from a handheld flashlight.  In an ideal world, that flashlight would point the sound (the light) directly at you Between mic placement, electronics/speakers, and acoustic bouncing within the studio, just to name some obvious ones, there is never just one "phase direction" (flashlight direction) captured in any track, and mixes make it even more diverse, of course.  Much like the "cone" of white light coming from a flashlight, directionality becomes splintered into many directions, but PHA-979 allows you to choose the phase-direction that sounds best to you, by "re-pointing the flashlight" to achieve optimum clarity and musical power--or a more diffuse sound, if you wish.  So, in my example, PHA-979 is not (just) about moving the time-line, although you can and should use that feature, if you need it.  Rather, it is about "aiming the direction" of the music recorded on a track to best align (or not-align) with your other parts of the mix, and of course you can make a complete mix more presentable, more powerful, or more diffuse, as you choose.

It is my opinion that "linear phase" operation is used to preserve the purity and predictability of the phase adjustment that the PHA-979 provides, and no doubt PHA-979 would not really exist without it.  I have been consistently amazed at the good, musical choices made at Voxengo, regarding the small colorations that the implementation of linear phase can introduce.  Designing a plug-in often involves a lot of compromises, due to Mother Nature's Physics And Math.  No free lunch in Math.  But the one you get can taste real good.  Thanks, Aleksey.


Thanks for your kind reply.

But what confuses me here is the technical terminology which Voxengo uses to explain PHA-979's features.  It seems that what Voxengo refers to is different from or even contrary to the conventional meaning of the terminology.


I have never found a User's Manual that I was completely happy with.  Scientific minds seem to speak a different version of the language from the rest of us.  Maybe I can offer a further notion for your consideration.

A typical loudspeaker phase-integrity-test graph might display phase-deviation from "perfect" on say, the vertical axis and a frequency range, low to high, on the horizontal axis.  Due to things like dispersion, driver and crossover issues (to say nothing of cabinet vibration "souring" the picture), one can see that the line which should rest at zero-degrees off from perfect is... not perfect.  What we would hope is a straight line at the zero-deviation point from left to right, for all frequencies, is not "all that".  That is an example of non-linear operation, i.e. "different frequencies shifted by different degrees", or in my flashlight idea, a "crooked" flashlight, not aiming straight at all frequencies.

By using linear phase design, "re-centering" the phase anomalies does not change the shape of the deviations on the graph, only their vertical position as you moved the entire set of deviations up and down.  Without a linear phase design, the shapes of the deviations would *themselves* be altered further (different amount of correction for different frequencies), The perception of the sound would be altered in its character, not just its directionality (power thrust).  Stated another way, Voxengo has avoided an unwanted "tone control" effect.

This topic was last updated 180 days ago, and thus it was archived.  Replying is disabled for this topic.