Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Don Boomer]

Well going down that path, the on/off switch will also stop feedback.

The goal would be to notch out the feedback and nothing else.  Using even a single filter from a 31 band EQ is likely a full octave wide or more and even wider if you use two filters.  So yes, sucking out everything from here to there will stop the feedback but it will carve a whole lot more desirable frequencies with it.  .There is a much better way

[John Roberts {JR}]

Well going down that path, the on/off switch will also stop feedback.

The goal would be to notch out the feedback and nothing else.  Using even a single filter from a 31 band EQ is likely a full octave wide or more and even wider if you use two filters.  So yes, sucking out everything from here to there will stop the feedback but it will carve a whole lot more desirable frequencies with it.  .There is a much better way

Not to quibble but 31 band GEQ are generally 1/3rd octave wide. 31 one octave wide bands would be so wide they would overlap and interact horribly.

Note: One of my long running unresolved rants is about the lack of precise definition and agreement about what exactly is the shape or response of a 1/3 octave wide boost/cut EQ, but it is generally not a full octave wide when frequency centers are spaced that close together. 

Even a one Hz wide notch will also remove some program, unless that cut exactly matches the level build up from the feedback resonance/repeats (not likely).

JR

[Tim McCulloch]

Not to quibble but 31 band GEQ are generally 1/3rd octave wide. 31 one octave wide bands would be so wide they would overlap and interact horribly.

Note: One of my long running unresolved rants is about the lack of precise definition and agreement about what exactly is the shape or response of a 1/3 octave wide boost/cut EQ, but it is generally not a full octave wide when frequency centers are spaced that close together. 

Even a one Hz wide notch will also remove some program, unless that cut exactly matches the level build up from the feedback resonance/repeats (not likely).

JR

And this brings us full circle, back to Bink's graphic EQ shootout a decade ago.  This is where we got to take a look at filters behaving badly.... particularly adjacent band interactions and how that varied depending on filter topology, even among models from the same manufacturer.

The text is still around on the interwebs, somewhere, but the pictures are long gone.  Too bad, it's worth a new discussion.

[Steve M Smith]

Not to quibble but 31 band GEQ are generally 1/3rd octave wide. 31 one octave wide bands would be so wide they would overlap and interact horribly.

Also, if you were to play some music through a system with all of the EQ at 0dB then just pull out a single frequency, I doubt that many people would notice.

In fact, at a Peavey training day many years ago, the engineer took out every third frequency and adjusted the gain to compensate and no one in the room could tell the difference but when he moved just one frequency positive, you would have to have had tin ears not to notice.

I'm not sure what point I'm trying to make now as those two sentences seem to contradict my previous posts.  Oh well, I know what I mean... I think!


EDIT:  I think what I really mean is that we can argue about the technicalities all day long, but if it works, that's the end of the argument. Practicality trumps theory every time!


Steve.

[Don Boomer]

Not to quibble but 31 band GEQ are generally 1/3rd octave wide. 31 one octave wide bands would be so wide they would overlap and interact horribly.

JR ... I'm shocked.  Take a look at the 30 or so units we measured at Bink's last party.  Almost all of them where an octave wide and some very popular ones were even wider.  The 1/3rd octave part is the centers of adjacent filters.

[Scott Holtzman]

One of the things I really like about iPad apps and the PEQ's is you can create a narrow and deep notch and do a quick "search and destroy" of the offending frequency.  An offending instrument mic or monitor send can utilize the PEQ to surgically excise the offending frequency. A good ear should be able to know what octave to start in.

I also am not deluded that the UI representation of the PEQ curve is reality. 

[Michael {Bink} Knowles]

...you can create a narrow and deep notch and do a quick "search and destroy" of the offending frequency.

One of the reasons I like digital PEQ so much is because I don't have to sweep around to find the right frequency. When I hear something I think can be fixed with a parametric filter then I first go to a spectrograph to see what its frequency is. No more searching for me; it's just a bit of minimal and gentle "destroy".

-Bink

[John Roberts {JR}]

JR ... I'm shocked.  Take a look at the 30 or so units we measured at Bink's last party.  Almost all of them where an octave wide and some very popular ones were even wider.  The 1/3rd octave part is the centers of adjacent filters.

There is disagreement about how to characterize boost/cut bandwidth. Some topologies exhibit apparently different bandwidth for different amounts of boost cut. I stand by my suggestion that full octave wide boost/cuts placed on 1/3rd octave centers will obviously interact.

I suspect some of those did.

JR

 

[John Roberts {JR}]

One of the reasons I like digital PEQ so much is because I don't have to sweep around to find the right frequency. When I hear something I think can be fixed with a parametric filter then I first go to a spectrograph to see what its frequency is. No more searching for me; it's just a bit of minimal and gentle "destroy".

-Bink

While academic at this point, I designed a monitor console with some sweepable notch filters in the sends. I used a variant on my FLS invention to advise with three "goldilocks" LEDs by the tuning control that showed if the signal was above, below, or in the bandpass of the notch filter.

Back in the day people used the old Mentor that had a 1/3 octave display to help people identify which slider to yank on their non-Peavey GEQ.

Modern digital desks can provide this capability and more... Not only should it identify a peak but pre-tune or snap the EQ to that frequency. While I still think there may be another way to skin that cat.

IEM should also make feedback in monitors less of a problem in the future.

JR 

[Don Boomer]

I stand by my suggestion that full octave wide boost/cuts placed on 1/3rd octave centers will obviously interact.
 

They are supposed to interact.  How else would you get to a frequency that was not the center? I'm away from my files, but I believe that all combined.

Of course having parametric filters solves that problem doesn't it.

[ProSoundWeb Community]