AudioMatt

So let me just say this. I'm not a moron. I've read many a book on sound synthesis and for years I never found a comprehensive understandable example of a Phaser. A lot of descriptions refer to a "network of allpass filters" which is not specific at all and does not in any way help me build one.

Anyway, I finally suffered by doing it the long way and building the filters, calculating the coefficients, and experimenting with the signal flow myself and I believe **THIS** is the phaser I grew to know and love. Please correct me if I'm wrong. The sound I'm going off of is the TC electronic fireworx Phaser module.

Maybe I'm being self centered here but I'm guessing there are at least a few people like me so including this in the examples or bleep or blop or whatever would probably be useful to them.

So let me just say this.  I'm not a moron.  I've read many a book on sound synthesis and for years I never found a comprehensive understandable example of a Phaser.  A lot of descriptions refer to a "network of allpass filters" which is not specific at all and does not in any way help me build one.  

Anyway, I finally suffered by doing it the long way and building the filters, calculating the coefficients, and experimenting with the signal flow myself and I believe **THIS** is the phaser I grew to know and love.  Please correct me if I'm wrong.  The sound I'm going off of is the TC electronic fireworx Phaser module.

sharing-doesnt-phase-me-8-stage-phaser

I was looking forward to some swooshing and tinkering, but allpassbiquad (the lowest level gen patcher) appears to be missing :(

Hey
I was looking forward to some swooshing and tinkering, but allpassbiquad (the lowest level gen patcher) appears to be missing :(

doh!!! replaced the file. should work now.

doh!!!  replaced the file.  should work now.


niiiiiooooiiiiiiooooiiiice ! Lots of vocal/throat formants in there, almost like a rich wah.

(I like the TC guitar fx myself, long been a fan)

Perhaps you can explain (in n00b terms, of course) why you prefer hand-built allpass filters at the sample level to, say, simple delay lines? What I mean is, would you get the same rich swirly-ness using Max's own allpass? Did you get hold of a schematic for your favourite phasor?

niiiiiooooiiiiiiooooiiiice ! Lots of vocal/throat formants in there, almost like a rich wah.
(I like the TC guitar fx myself, long been a fan)

The reason I built the allpass by hand is that I don't have any idea what "delay" and "linear gain" (the controls on [allpass~] have to do with frequency and q. Perhaps someone smarter could explain. So I went to 

, figured out how to calculate the biquad coefficients of an allpass filter, and used the biquad gen~ example and factored out some unnecessary math.

I threw the tanh at the end to make it a tad little less sterile. Could use some nonlinearity.

While a simple delay line is technically a type of allpass filter, I don't believe a delay works the same as this filter. As far as I can tell, a simple modulated delay line would make a flanger. Flangers have notches that, when plotted on a logarithmic graph, get closer together as they went up (like a hair comb). A Phaser's notches, on the other hand, would be equally spaced out.

The reason I built the allpass by hand is that I don't have any idea what "delay" and "linear gain" (the controls on [allpass~] have to do with frequency and q.  Perhaps someone smarter could explain.  So I went to  RBJ's cookbook , figured out how to calculate the biquad coefficients of an allpass filter, and used the biquad gen~ example and factored out some unnecessary math.  

I threw the tanh at the end to make it a tad little less sterile.  Could use some nonlinearity.

While a simple delay line is technically a type of allpass filter,  I don't believe a delay works the same as this filter.  As far as I can tell, a simple modulated delay line would make a flanger.   Flangers have notches that, when plotted on a logarithmic graph,  get closer together as they went up (like a hair comb).  A Phaser's notches, on the other hand, would be equally spaced out.  

thanks for taking the time to reply; I've seen several references to the Cookbook you mention - some of the equations were a little intimidating for me. But there's obviously a secret sauce in there, because I tried recreate your algorithm using vanilla allpass filters and LFOs, and - as you say - I wasn't getting the same degree/depth of control using "delay" and "gain" parameters. Nice sounds, but not as rich as your phasor.

Ah, ok, the difference between comb and phasor are well explained too.

Hey
thanks for taking the time to reply; I've seen several references to the Cookbook you mention - some of the equations were a little intimidating for me. But there's obviously a secret sauce in there, because I tried recreate your algorithm using vanilla allpass filters and LFOs, and - as you say - I wasn't getting the same degree/depth of control using "delay" and "gain" parameters. Nice sounds, but not as rich as your phasor.

Brendan, it's actually pretty easy. It might be a cool project to translate that page into gen~. Gimme a few days.

Edit: Also, I did the same thing with Allpass~ and thought, "NO THAT SOUNDS NOTHING LIKE JIMI HENDRIX"

Brendan, it's actually pretty easy.  It might be a cool project to translate that page into gen~. Gimme a few days.

Edit:  Also, I did the same thing with Allpass~ and thought, "NO THAT SOUNDS NOTHING LIKE JIMI HENDRIX"


[Sharing doesn't phase me] 8 Stage Phaser (works now)