Digital feedback reducers
David D. Berkowitz
applications in the signal path for acoustic guitars and whether it might be
a solution for a client with high stage levels.
--db
David D. Berkowitz
took out very little, but haven't had the opportunity to use it in a really
live application. --db
"David D. Berkowitz" <d...@berkowitzguitars.com> wrote in message
news:KFp3c.80750$6K.6...@nwrddc02.gnilink.net...
BOB URZ
"David D. Berkowitz" wrote:
We use them all the time in fixed applications. It really takes a PC
to properly set it all up. What problem are you trying to solve?
Bob
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----
Jason Lavoie
they look good. but setting them up without a computer would appear to
be almost impossible.
for less money than even a dfr11 you can get a behringer DEQ2496 which
is stereo, will do narrower notches from what I remember, plus you
get a ton of other stuff for free (like an RTA, SPL meter. etc.) and
those can be used without interrupting the filtered audio.
the DFR's might be my first choice for installs if the price came down
a bit. but live I'd never use one
Jason
David D. Berkowitz
setup: Single 12 wedge in front with dual twelve wedges angling out from
there and shoulder-high side-washes coming in from stage left & stage right.
Yea, I know its rediculuous. --db
"BOB URZ" <"sound(remove)"@inetnebr.com> wrote in message
news:404E2C17...@inetnebr.com...
David D. Berkowitz
Why would you not use one live?
--db
"Jason Lavoie" <jla...@engsoc.org> wrote in message
news:j0cs40ttqrofprtvh...@4ax.com...
BOB URZ
"David D. Berkowitz" wrote:
What he probably means is the meat of the device is accessed through
the GUI. You can set up a DFR to automatically set feedback filters.
You can get up to 12. You can lock them down or let them float.
BUT, it takes the GUI to go into the box to fine tune it. In the GUI
you can set the 1/3 octave or parametric, delay, limiter, DFR lock
down, DFR notch width and depth and system gain. The front panel controls of a
DFR-11 only allow you to bypass the unit or select a scene.
They work very well, but you need the laptop to get the most out of it.
If you not carrying a computer, another device with more front panel
controls might be a better choice. If you want one, i can shoot you
a price, but there not cheap. But there worth it for sound quality and
reliability.
David D. Berkowitz
Bob,
My impression when I auditioned the box at WashintonPro was that it took
out almost nothing from the signal while severely curtailing feedback. I
got 3/4 input gain with no feedback while standing right in front of the
monitor with a baritone acoustic fitted with a Fishman Matrix Natural I
pickup. Is this similar to the experience you have had?
My thanks for your responses,
David
BOB URZ
"David D. Berkowitz" wrote:
We mainly use these for fixed installs such as churches. We EQ the system using
the 1/3 octave then ring out the DFR filters with
the system microphones. I have also used one inserted in a channel
in a live sound situation. I was riding shotgun on the GUI to fine tune
what i wanted out of it.
The DFR starts at 1/10 octave filter at feedback detection, and can go as narrow
as 1/40th octave. So, it takes a narrow chunk of the signal.
This does translate into a more natural sound. There is no reason you cannot use
one on a monitor. Just be aware at a certain point,
thats all there is system gain before feedback wise. It can make a marginal
system much more stable, but it cannot make gold out of lead. I highly recommend
computer control or set-up if your using one.
The gui will also tell you the offending frequency if that interests you.
David D. Berkowitz
Bob,
I guess why I'm interested is that an acoustic guitar can suffer
terribly from poor eq. A regular 30 band eq takes too much out (relative to
gain before feedback) rametrics are better, but you're stuck with the
settings that work best for the response characteristics of that particular
instrument. The idea of being able to have very narrow filters that take
out almost nothing and the ability to set scenes for each guitar they use
seemed like a good way to handle it. --db
Steve White
> I guess why I'm interested is that an acoustic guitar can suffer
>terribly from poor eq. A regular 30 band eq takes too much out (relative
to
>gain before feedback) rametrics are better, but you're stuck with the
>settings that work best for the response characteristics of that particular
>instrument. The idea of being able to have very narrow filters that take
>out almost nothing and the ability to set scenes for each guitar they use
>seemed like a good way to handle it. --db
In my ceilidh band we have a set of scottish small pipes as lead instrument.
These are extremely quiet ... you could easily drown them out acoustically
by strumming an acoustic guitar loudly.
We use a Behringer Feedback Destroyer on the monitors and this makes a
phenomenal difference to the amount of foldback we can get before the onset
of feedback. We leave the FOH signal uneffected so as to not alter the tone
and it works very well.
Cheers,
Steve W
L David Matheny
> Anyone played with a Shure dfr11eq or dfr22? I'm wondering
> about its applications in the signal path for acoustic guitars and
> whether it might be a solution for a client with high stage levels.
>
so I don't have the experience of the pro's here. But I use a
DFR11EQ when taping seminars, and I love it. I set it up with
some HF and LF rolloff and six fixed filters and four dynamic.
After setting up on location I can just clear the filters and then
train it while gradually raising the gain until all the fixed filters
have been set. It's inserted between an automixer (which feeds
the camcorders) and the PA amp. You'd be using it differently,
of course, but it can be useful even w/o a computer on location.
Bob Urz
Well, it seems you have two issues: EQ and feedback reduction.
In your case, the parametric might work better and be easier than the
graphic 1/3 octave (you can select one or the other). You can have a
different EQ setting and DFR setting for each preset. There are 10
presets. Only the first three are available from the front panel. The
rest need the GUI or remote control for access. The older DFR-11's did
not have the front panel scene controls.
Along with the EQ, you can use the DFR filters. So you can certainly
do what you want. You just need a PC to initially set it up.
The DFR filters are NOT for tone shaping, just feedback reduction.
The parametric or Graphic will do your voicing should you need to
EQ the instruments for best tone.
Joel Foner
> Jason,
>
> Why would you not use one live?
>
> --db
First - what is a feedback suppressor?
1) A set of narrow-band filters
2) Some programming that looks for things that look like feedback and place
a cut filter at that point
Part 1) Is more universally handled by a good parametric EQ (I only have
graphics around when there will be another engineer mixing - parametrics are
much more flexible and in particular for feedback tuning let you place the
notch exactly where you need it - far more feedback suppression with far
less impact on tone)
Part 2) Has a big "IF" in it... Can the programming determine what is
feedback without false positives, and can it fix the feedback as fast as you
can?
The answer to '2' is "kind of". The three biggest problems I've had with
them are:
- If you set the notches during sound check and lock them (to prevent the
other problems below), you'll often find that the feedback frequencies move
around with the venue full of people and during the show - now you've got
notches in the wrong places - cure worse than the disease
- If you let the system do its "dynamic feedback suppression" it will
regularly drop notches into the bass and mid-bass region when it hears
sustained tones that "sound like feedback".
- These systems usually can't pick out a "hollow ring" before things get to
something that sounds like feedback, and then usually don't act for between
0.5-1 second. This means that it will drop in a filter there, but you'll
hear a second of heavy squeal before it tries.
If you're listening during the show and can pick out the hollow ring before
it lets loose, you'll prevent the feedback long before the feedback
suppressor would have known about it.
Hope this helps,
Joel Foner
David D. Berkowitz
"Joel Foner" <joel....@fonerassoc.remove_me_to_reply.com> wrote in message
news:3vH3c.70088$vn.2...@sea-read.news.verio.net...
Thanks, everyone. Are there any two channel, five band parametric's out
there or will my client need two parametrics to handle the signals from his
stereo send on his guitar?
--db
BOB URZ
"David D. Berkowitz" wrote:
A DFR22 is two channel. But, what is the set up and signal routing?
You don't want to plug a guitar pickup directly into one.
You would need a pre amp that has a link or insert to plumb it in.
You say its a stereo guitar. Is it going through a stereo amp or direct into the
PA? How is the feed getting to the mixer and monitors?
Is that feed stereo?
David D. Berkowitz
> A DFR22 is two channel. But, what is the set up and signal routing?
> You don't want to plug a guitar pickup directly into one.
> You would need a pre amp that has a link or insert to plumb it in.
> You say its a stereo guitar. Is it going through a stereo amp or direct
into the
> PA? How is the feed getting to the mixer and monitors?
> Is that feed stereo?
>
> Bob
The guitar has two seperate pickups requiring different preamps and so
forth. They would be run into the board after a preamp/DI such as an ADL
300GDI. Then use an insert for the parametric. --db
A. & G. Reiswig
example, the Sabine higher-end stuff, the notch is narrower than you're
going to be able to get on a typical PEQ. I've used a Sabine 2020 with good
success and minimal impact on the overall tone.
George Reiswig
Song of the River Music
"Joel Foner" <joel....@fonerassoc.remove_me_to_reply.com> wrote in message
news:3vH3c.70088$vn.2...@sea-read.news.verio.net...
BOB URZ
"A. & G. Reiswig" wrote:
> I disagree somewhat. If you look at the bandwidth of the notches on, for
> example, the Sabine higher-end stuff, the notch is narrower than you're
> going to be able to get on a typical PEQ. I've used a Sabine 2020 with good
> success and minimal impact on the overall tone.
>
> George Reiswig
>
Comparing a parametric and a DFR filter is like comparing apples
to oranges. A parametric is more a Macro type filter generally used for voicing
EQ. A DFR filter is more a Micro type filter with a VERY narrow band device used
for feedback suppression, NOT a general voicing EQ. The shure's DFR is 1/10 to
1/40th octave. That's pretty
narrow. That's more like trying to take the fly off the elephants butt
sort of EQ. The parametric is more used for turning the elephant from
pink to grey.
You have to distinguish between an analog controlled parametric
and a digital one. If your trying to control very narrow notches
with a analog controlled parametric, it is more prone to drift
than a digitally set unit using a DSP is.
David D. Berkowitz
> "A. & G. Reiswig" wrote:
>
> > I disagree somewhat. If you look at the bandwidth of the notches on,
for
> > example, the Sabine higher-end stuff, the notch is narrower than you're
> > going to be able to get on a typical PEQ. I've used a Sabine 2020 with
good
> > success and minimal impact on the overall tone.
> >
> > George Reiswig
> >
>
> Comparing a parametric and a DFR filter is like comparing apples
> to oranges. A parametric is more a Macro type filter generally used for
voicing
> EQ. A DFR filter is more a Micro type filter with a VERY narrow band
device used
> for feedback suppression, NOT a general voicing EQ. The shure's DFR is
1/10 to
> 1/40th octave. That's pretty
> narrow. That's more like trying to take the fly off the elephants butt
> sort of EQ. The parametric is more used for turning the elephant from
> pink to grey.
>
> You have to distinguish between an analog controlled parametric
> and a digital one. If your trying to control very narrow notches
> with a analog controlled parametric, it is more prone to drift
> than a digitally set unit using a DSP is.
>
> Bob
That sounds like for my client that a DSP might be the thing, with all the
aforementioned caveats. I'm trying to take out as much of the offending
specific frequencies. The narrower the notch, the more natural the signal
will be. Part of my client's problem is that their stage levels are really
inappropriate for a high end acoustic instrument. That their stage levels
overwhelmingly wash over into the audience is the problem. My client would
like to use my guitar live, but standard EQing takes too much out. --db
Kurt Albershardt
>
> George Reiswig wrote:
>
>
>> I disagree somewhat. If you look at the bandwidth of the notches on, for
>> example, the Sabine higher-end stuff, the notch is narrower than you're
>> going to be able to get on a typical PEQ. I've used a Sabine 2020 with good
>> success and minimal impact on the overall tone.
>
>
>
> to oranges. A parametric is more a Macro type filter generally used for voicing
> EQ. A DFR filter is more a Micro type filter with a VERY narrow band device used
> for feedback suppression, NOT a general voicing EQ. The shure's DFR is 1/10 to
> 1/40th octave. That's pretty
> narrow. That's more like trying to take the fly off the elephants butt
> sort of EQ. The parametric is more used for turning the elephant from
> pink to grey.
>
> You have to distinguish between an analog controlled parametric
> and a digital one. If your trying to control very narrow notches
> with a analog controlled parametric, it is more prone to drift
> than a digitally set unit using a DSP is.
Does anyone make a moderately-priced cut-only analog parametric or notch filter these days? I had an Ashly SC-68 years ago and I've had several occasions recently where it would have been my first choice tool.
Joel Foner
> Comparing a parametric and a DFR filter is like comparing apples
> to oranges. A parametric is more a Macro type filter generally used for
> voicing EQ. A DFR filter is more a Micro type filter with a VERY narrow
> band
> device used for feedback suppression, NOT a general voicing EQ. The
> shure's DFR is
> 1/10 to 1/40th octave. That's pretty
> narrow. That's more like trying to take the fly off the elephants butt
> sort of EQ. The parametric is more used for turning the elephant from
> pink to grey.
Agreed mathematically. There are plenty of filter types, especially in
digital units - it's all DSP math at that point. In non-extreme situations a
narrow parametric with a few dB of cut is not all that different in terms of
impact though. If you've got a 1/10th octave filter cut by 3-6 dB will it
sound "different" than a 1/40th octave filter cut by the same amount? In a
studio yes. I've never done the experiment but I'm not sure it will be
audible in most venues. If the cut is extreme, like 10 dB+ will it get more
audible? Sure. I've found that I'm no longer needing extreme notches, which
is why I didn't make a big deal out of the difference between the filter
types.
I guess the other piece of info that fits in here is that the Sabine 3100
series units that I have look like graphic EQs, but if you hook up a
computer to them you have access to a dozen parametric/notch filters per
channel - stuff you don't get from the face plate knobs. I'm used to using
them in that mode, and so the option is there to have the same filters
around without using the automation. In other words, with these units you
can leave the auto-feedback suppression on, in which case it uses the
filters, or you can turn off the automation and use those same filters
yourself. ("Luke, use the force...!") It's really the automation I was
commenting on getting away from more than the filter type - sorry for any
confusion from writing while tired.
The other thing which probably should have been a comment in an entirely
separate thread was the graphic vs parametric rant. I found that after I
switched to all parametrics that things sounded noticeably cleaner, and that
I had less feedback issues from the get-go - part of reducing my interest in
the automated feedback suppression. Was this from hearing better through
working with parametrics (forces you to think about filter width and center
point in a far different way than graphics do)? Maybe. Was it from
compensation curves that use fewer and broader filters rathern than a bunch
of overlaid graphic adjustment filters? Maybe / probably. As they say, your
mileage may vary - just passing along that when I switched to a more
parametric based strategy I found it less and less interesting to put up
with the side effects of auto-feedback suppression and found that the really
narrow notches of these units didn't seem to be needed any more.
Regards,
Joel
L David Matheny
>
> > Jason,
> >
> > Why would you not use one live?
> >
> > --db
>
> First - what is a feedback suppressor?
> 1) A set of narrow-band filters
> 2) Some programming that looks for things that look like feedback
> and place a cut filter at that point
>
> Part 1) Is more universally handled by a good parametric EQ
> (I only have graphics around when there will be another engineer
> mixing - parametrics are much more flexible and in particular for
> feedback tuning let you place the notch exactly where you need it -
> far more feedback suppression with far less impact on tone)
>
notches which do less damage to the sound quality or tone.
> Part 2) Has a big "IF" in it... Can the programming determine what
> is feedback without false positives, and can it fix the feedback as
> fast as you can?
>
> The answer to '2' is "kind of". The three biggest problems I've had
> with them are:
> - If you set the notches during sound check and lock them
> (to prevent the other problems below), you'll often find that the
> feedback frequencies move around with the venue full of people
> and during the show - now you've got notches in the wrong places -
> cure worse than the disease
>
I've never noticed this being a problem, but I'll watch in the future.
The dynamic or floating filters should catch some of this, unless you
decide to use all fixed because of the false-positive problem below.
> - If you let the system do its "dynamic feedback suppression" it will
> regularly drop notches into the bass and mid-bass region when it hears
> sustained tones that "sound like feedback".
>
This is an important point. I was retraining a Sabine unit used at church
(for the pastor's wireless mic) while the organist was rehearsing, and it
used all the filters immediately because of false positives. Frequency
sweeps will do that, too. You must train the unit using pink noise or
impulse noise or voice, then lock the filters before playing music. The
DFR's do seem to work more naturally for spoken word than for music.
> - These systems usually can't pick out a "hollow ring" before things
> get to something that sounds like feedback, and then usually don't act
> for between 0.5-1 second. This means that it will drop in a filter there,
> but you'll hear a second of heavy squeal before it tries.
>
> If you're listening during the show and can pick out the hollow ring
> before it lets loose, you'll prevent the feedback long before the
> feedback suppressor would have known about it.
>
Well, nothing is perfect. They work best when properly "trained" ahead
of time. You can allow some room for ringing by using a bit more gain
while training and then backing off a bit.
Jason Lavoie
<d...@berkowitzguitars.com> wrote:
>Jason,
>
>Why would you not use one live?
>
>--db
>
the lack of front panel control on the DFR's would make it less than
ideal for live use.
the behringer is cheaper, and esier to use, plus it doesn't require a
laptop in order to access all its features.
in a fixed install the use of a laptop is acceptable.. you do it once,
and then the users have access to the preset buttons and that's all
they need.
in live use you need to adjust the settings every gig (yes, every
single gig they will need to be at least slightly different)
Jason