The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
feedback destroyer Utility Software
Old 16th March 2014
  #91
Lives for gear
Quote:
Originally Posted by Samc View Post
The real, hard fact of the matter is that these units cannot, and do not prevent feedback, they can only stop it after the fact. They can't take preemptive action or anticipate anything. They can only react...doesn't matter how quick they are they're still only reactive.
Well they set filters. So are you saying that the setting of filters cannot prevent feedback? What about dialing in you system before hand with GEQ or PEQ. It's the same thing.

Quote:
Are people using them on all open mics or are they just using them across the main L R outputs? Because it would seem that the latter would be less efficient and more destructive. Plus what happens when the synth or guitar player plays some pure tones?
So as JR said above, per channel is the least distructive way. However using it across a bus will cut feedback and usually at very little loss of sonic quality. In as many cases as not less loss than other methods involving EQ. These units use very narrow filters. You can measure them all day but human hearing cannot integrate them unless you stack a bunch of them together. But even then that is likely a smaller cut than from a single band of a 1/3rd octave EQ.
Old 16th March 2014
  #92
Lives for gear
Quote:
Originally Posted by Samc View Post
The real, hard fact of the matter is that these units cannot, and do not prevent feedback, they can only stop it after the fact. They can't take preemptive action or anticipate anything. They can only react...doesn't matter how quick they are they're still only reactive.
Well they set filters. What they really are is parametric EQs with fixed narrow band filters that have their center frequencies set by measurement. So are you saying that the setting of filters cannot prevent feedback? What about dialing in you system before hand with GEQ or PEQ. It's the same thing.

Quote:
Are people using them on all open mics or are they just using them across the main L R outputs? Because it would seem that the latter would be less efficient and more destructive. Plus what happens when the synth or guitar player plays some pure tones?
So as JR said above, per channel is the least distructive way. However using it across a bus will cut feedback and usually at very little loss of sonic quality. In as many cases as not less loss than other methods involving EQ. These units use very narrow filters. You can measure them all day but human hearing cannot integrate them unless you stack a bunch of them together. But even then that is likely a smaller cut than from a single band of a 1/3rd octave EQ.
Old 16th March 2014
  #93
Lives for gear
Quote:
Originally Posted by nedorama View Post

ART and Peavey make 31-band EQs that do have lights that show the offending frequency(s), which may be more helpful to be able to quickly grab the offending fader. ...
JR could say better than me ( as he designed it) but these units are showing you just the 1/3rd of an octave of bandwidth in which the feedback is occurring and not the single frequency that is causing the feedback. I also believe that what these devices actually measure is only peak material and not necessarily feedback at all.

And even if the measuring is a 1/3rd of an octave, when you move a slider you are making a cut that affects an entire octave width. With a PEQ or a FBS you are likely cutting only a 1/10th of an octave or less. So while a 31 band graphic will stop the feedback it chunks out rather unnecessarily a large section of what would otherwise be useful material.
Old 16th March 2014
  #94
Lives for gear
 

Quote:
Originally Posted by dboomer View Post
So as JR said above, per channel is the least distructive way. However using it across a bus will cut feedback and usually at very little loss of sonic quality. In as many cases as not less loss than other methods involving EQ. These units use very narrow filters. You can measure them all day but human hearing cannot integrate them unless you stack a bunch of them together. But even then that is likely a smaller cut than from a single band of a 1/3rd octave EQ.
For that reason I do not use the system EQ for feedback 'abatement'...I use the parametric EQ and the volume fader of the offending channel and if you're skillful and vigilant you can be less destructive than any box strapped across the master bus.

Cutting a frequency on the system EQ to rectify a ring on the tom channel would be a little stupid. How do these boxes handle long, sustained notes?

And I can sometimes hear these boxes working...regardless of how narrow the cuts are I can often hear the after effects.
Old 16th March 2014
  #95
Lives for gear
At last ... We are in total agreement

But I think think it is far more common to see some EQ device across a bus used to kill feedback in practice. And in my book, feedback being a deal-breaker, must be stopped before you need to worry about tone. Of course it would be preferable to maintain tone when possible but most live sound is really just doing damage control.

How they work on sustained tone is exactly the same way your parametric works on sustained tone because they are parametric EQs. And weather or not you can hear them is the same as your parametric depending on how both of them need to be set in order to prevent/destroy feedback.

And of course you can ride gain with a fader but in effect that is just a filter set to 10 octaves wide.
Old 16th March 2014
  #96
Lives for gear
 

There is a hard to accept reality for some folks that will fight to the end over a given purest protocol to manage high quality sound reinforcement. The reality is most of us, for many reasons, prefer a pink noise/reference mic based system to adjust FOH speakers for room anomalies. Smaart, Fuzz Measure or Sabine power Q all feature the ability to quickly identify system hot spots in a given room and that process absolutely is step one for an efficient gain structure procedure. My personal preference is to set no more that 4 carefully shaped parametric bands to remove subject hot spots indicated by subject RTA analysis and I will try to run a quick double check after the crowd arrives. These tools along with feed back suppression can be a big help for novice and less "gifted" folks twisting knobs and push/pulling faders to quickly identify performing hot spots. Very few high end pro systems will have feed back suppression deployed how ever most folks that follow these threads do not fall in that category.
Hugh
Old 16th March 2014
  #97
Lives for gear
Quote:
Originally Posted by dboomer View Post
JR could say better than me ( as he designed it) but these units are showing you just the 1/3rd of an octave of bandwidth in which the feedback is occurring and not the single frequency that is causing the feedback. I also believe that what these devices actually measure is only peak material and not necessarily feedback at all.

And even if the measuring is a 1/3rd of an octave, when you move a slider you are making a cut that affects an entire octave width. With a PEQ or a FBS you are likely cutting only a 1/10th of an octave or less. So while a 31 band graphic will stop the feedback it chunks out rather unnecessarily a large section of what would otherwise be useful material.
Don - agree these aren't perfect, but for a band to have instead of regular 31 band EQs, having a light to at least indicate which band to pull down vs. guessing can help. And yes, also agree a graphic EQ is a butter knife compared to a more surgical parametric EQ.
Old 16th March 2014
  #98
Lives for gear
 

You need a full-time sound-person that knows what to do in your case, with all the extra sound you have (sequencing and all) theres no way around it really. The only small-room gigs that you can get away with no sound guy is when you have a band with just a few performers and you don't need much in the PA/monitors to start with. And the way to do that is to have a stage level where you don't need anything but singing mics in the monitors, thus, singing mics and two mics on the drums (snare and kick) is about all for the FOH.
Old 16th March 2014
  #99
Lives for gear
 
Electrolytic's Avatar
 

^ the easiest, cheapest and most reliable system. just play at a sensible level with dynamic expression and it will sound better for it.
Old 17th March 2014
  #100
Lives for gear
 

Quote:
Originally Posted by hughshouse View Post
Very few high end pro systems will have feed back suppression deployed how ever most folks that follow these threads do not fall in that category.
I haven't taken a poll but I bet that most people here don't have one of those boxes…

Every system in every room/space have a threshold and in most cases all the sound person need to do is turn the volume down to a level that is below the threshold and they would't be faced with this dilemma in the first place…this is science not voodoo.

This is where knowledge and experience counts, there can be no excuse for not learning our craft properly, the solution is not always in a new box or in some 'trick' method we read about on the internet.
Old 17th March 2014
  #101
Lives for gear
 

Quote:
Originally Posted by nedorama View Post
Don - agree these aren't perfect, but for a band to have instead of regular 31 band EQs, having a light to at least indicate which band to pull down vs. guessing can help. And yes, also agree a graphic EQ is a butter knife compared to a more surgical parametric EQ.
You should not use the system EQ to suppress feedback…find the offending channel and deal with the problem there!!!

Using the main system EQ to suppress feedback is just as silly as strapping a feedback killer across the main output bus.
Old 17th March 2014
  #102
Lives for gear
 

Quote:
Originally Posted by dboomer View Post
But I think think it is far more common to see some EQ device across a bus used to kill feedback in practice.
Anybody who does this doesn't know what they're doing.

Quote:
How they work on sustained tone is exactly the same way your parametric works on sustained tone because they are parametric EQs.
I think you misunderstood my question; can it be fooled into activation by a soloing guitar for example? My parametric can't be fooled because I'm operating it and I know the difference between a screaming guitar and feedback.

Quote:
And of course you can ride gain with a fader but in effect that is just a filter set to 10 octaves wide.
Yes, but I'm only riding one channel and unlike the box I can decide when, where and how to ride…especially if I know the song.
Old 17th March 2014
  #103
Lives for gear
Quote:
Originally Posted by Samc View Post
You should not use the system EQ to suppress feedback…find the offending channel and deal with the problem there!!!

Using the main system EQ to suppress feedback is just as silly as strapping a feedback killer across the main output bus.
Who said anything about these being a system EQ? Not me.
Old 17th March 2014
  #104
Lives for gear
Quote:
Originally Posted by Samc View Post

I think you misunderstood my question; can it be fooled into activation by a soloing guitar for example? My parametric can't be fooled because I'm operating it and I know the difference between a screaming guitar and feedback.
They can, some handle it better than others.

But why would you insert one on a guitar channel? If you only had it inserted into a vocal channel the level of that guitar would have to be so loud as to actually become feedback in that vocal channel.
Old 17th March 2014
  #105
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by dboomer View Post
JR could say better than me ( as he designed it) but these units are showing you just the 1/3rd of an octave of bandwidth in which the feedback is occurring and not the single frequency that is causing the feedback. I also believe that what these devices actually measure is only peak material and not necessarily feedback at all.
You called...?

Yes my FLS invention is pretty simple and just detects which band of say a 31 band GEQ is playing loudest at any given moment and illuminates that one LED. So yes this will indicate for loud musical notes as well as feedback. FLS is an aid for human sound engineers for when they hear feedback occurring to help them know which EQ band it is in so they can quickly eliminate it.

Note: I used a variant of those FLS LEDS with sweepable narrow notch filters inside a monitor console to give visual feedback about which direction to tune the notch filter to capture the feedback when occurring. In that same console I used another variant on FLS where an LED on each input channel would indicate which input was loudest, to help identify which mic the feedback was coming in on. Again these are just visual aids to help a human sound engineer manage his world.
Quote:
And even if the measuring is a 1/3rd of an octave, when you move a slider you are making a cut that affects an entire octave width. With a PEQ or a FBS you are likely cutting only a 1/10th of an octave or less. So while a 31 band graphic will stop the feedback it chunks out rather unnecessarily a large section of what would otherwise be useful material.
Not to veer too much, I have put FLS feedback LEDs on GEQ as small as 5 band. Before Sam tells me how bad that is for feedback management, I will note that it was appreciated by the entry level system operators to see where their sound is loudest. A 5 band EQ is better than no EQ.

Ideally we maximize our GBF from using high quality mics with the proper pattern, quality monitor speakers, and proper mic placement. In the real world stuff happens and modern tools can help us deal with it.

JR
Old 17th March 2014
  #106
Lives for gear
 
loujudson's Avatar
DBX 224 is the best one for me. I only use them for lavalier and spoken word gigs, settiing as many freqs as needed before it becomes a problem. the DBX has much narrower nothes, and has 12 ffixed and 12 v=that can be set to go away after a settable time.

They trigger AT -9 SO YOU NEED TO GIVE THEM ENOUGH SIGNAL SO THAT THEY CAN WORK FAST.
Old 17th March 2014
  #107
Lives for gear
 

Quote:
Originally Posted by dboomer View Post
But why would you insert one on a guitar channel? If you only had it inserted into a vocal channel the level of that guitar would have to be so loud as to actually become feedback in that vocal channel.
In the case it was strapped across the main stereo bus…I was under the impression that this is where most people insert them.
Old 17th March 2014
  #108
Lives for gear
 

Quote:
Originally Posted by nedorama View Post
Who said anything about these being a system EQ? Not me.
Where else do you normally expect to see "regular 31 band EQs"?

I don't usually see them inserted on input channels...
Old 17th March 2014
  #109
Lives for gear
 

Quote:
Originally Posted by JohnRoberts View Post
Yes my FLS invention is pretty simple and just detects which band of say a 31 band GEQ is playing loudest at any given moment and illuminates that one LED. So yes this will indicate for loud musical notes as well as feedback. FLS is an aid for human sound engineers for when they hear feedback occurring to help them know which EQ band it is in so they can quickly eliminate it.

Note: I used a variant of those FLS LEDS with sweepable narrow notch filters inside a monitor console to give visual feedback about which direction to tune the notch filter to capture the feedback when occurring. In that same console I used another variant on FLS where an LED on each input channel would indicate which input was loudest, to help identify which mic the feedback was coming in on. Again these are just visual aids to help a human sound engineer manage his world.
Quote:
Not to veer too much, I have put FLS feedback LEDs on GEQ as small as 5 band. Before Sam tells me how bad that is for feedback management, I will note that it was appreciated by the entry level system operators to see where their sound is loudest. A 5 band EQ is better than no EQ.
Nothing wrong with any of this because with this system the sound person makes the decision on how best to deal with the feedback, plus this system can help the sound person learn how to identify feedback/frequencies by ear.

With a regular box that operates in the background the less knowledgeable will not learn anything or even know what's going on under the hood.
Old 17th March 2014
  #110
Lives for gear
Quote:
Originally Posted by Samc View Post
Where else do you normally expect to see "regular 31 band EQs"?

I don't usually see them inserted on input channels...
Monitor channels, for one, which is usually where feedback can happen...
Old 18th March 2014
  #111
Lives for gear
 

Quote:
Originally Posted by nedorama View Post
Monitor channels, for one, which is usually where feedback can happen...
My advise is still the same; find the offending channel and deal with it there.

This will be more efficient and less destructive.
Old 18th March 2014
  #112
Lives for gear
Quote:
Originally Posted by Samc View Post
Nothing wrong with any of this because with this system the sound person makes the decision on how best to deal with the feedback, plus this system can help the sound person learn how to identify feedback/frequencies by ear.

With a regular box that operates in the background the less knowledgeable will not learn anything or even know what's going on under the hood.
But it only shows which "band" has the loudest signal. That's better than picking the wrong band but that is far different from identifying the single Hz frequency that is causing the feedback. A good FBS can identify a single frequency and drop a 1/10th to 1/70th of an octave filter directly on it. That Peavey graphic is almost one full octave wide.
Old 18th March 2014
  #113
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by dboomer View Post
But it only shows which "band" has the loudest signal. That's better than picking the wrong band but that is far different from identifying the single Hz frequency that is causing the feedback. A good FBS can identify a single frequency and drop a 1/10th to 1/70th of an octave filter directly on it. That Peavey graphic is almost one full octave wide.
I'd think we've beat this all to death by now...

When I designed the sweepable notch filters for use inside the monitor console that I did years ago, I wrestled with how narrow to make the notch filters. Acoustic feedback is a relatively narrowband phenomenon, but.... in the real world the mic can get bumped changing the path length slightly, and the room can heat up and get more humid, changing the speed of sound traveling over the feedback path, also altering the feedback frequency so making the FBK filter overly narrow could allow the feedback to start up again later when conditions change enough.

Not trying to be argumentative, just sharing that it is not always a narrower is better decision.

I will concede my 5 band GEQ with FLS was not narrow enough to be a professional tool.

JR

PS: Don how was Frankfurt?
Old 18th March 2014
  #114
Lives for gear
Quote:
Originally Posted by JohnRoberts View Post
))

Not trying to be argumentative, just sharing that it is not always a narrower is better decision....
Agreed, but that works both ways. But (at least with some units) they either expand the bandwidth when called upon or set additional filters nearby.

Funny though, when they do this they are often accused of changing the sound even though a GEQ changes it much more.


Btw no Frankfort for me this year
Old 18th March 2014
  #115
Lives for gear
 

I have always beiieved in detecting and dealing with any possible "Standing Waves" that a given room may present and this process should always be the first set up chore. Untreated Standing Waves, if present, will corrupt any and all subsequent alterations to individual channels as well as two mix distribution. I currently use Fuzz Measure to detect and parametric EQ to reduce hot FOH and to a lesser extent floor wedge frequencies. I totally agree with John Roberts about moving room anomalies that can require small EQ adjustments: From that point out mic placement, DB restraint and performance technique along with individual channel compression will go a long way in preventing feed back of any kind. Pushing any SR system beyond it's limits is a sophomoric exercise that is a dead give away of novice behavior.
Hugh
Old 14th April 2014
  #116
Here for the gear
 

Hey guys im trying to decide between the Sabine 2410/2400, Sabine SM-820 and the Shure DFR22. Does anybody have any experience with these units that could let me know what would do the job best? My sole purpose is to use it for only vocals not the main mix. The Shure DFR22 and the Sabine 2410/2400 only take line level input, i don't mind adding a preamp to the setup if they would do the job best.
Old 17th April 2014
  #117
Here for the gear
 

Suggestions for Feedback Destroyer Units

I work for dbx and have run some very vigorous tests on all of the main feedback destroyers (old and new). I won't give specific numbers, but I will give some general guidelines.

Invest in the newer versions of each product!
Each brand (dbx,Sabine,Peavey,Behringer, etc.) has made significant improvements in their latest feedback destroyer products. You'll get much better performance this way. Yes, even the Behringer has improved to a useable state. My only beef with it is the interface. The newer products I am referring to are: dbx DriveRack PA2, Sabine FBX2410, Peavey Feedback Ferret D, Behringer FBQ2496, Shure DFR22.

For the following test results, I'm referring to the newest version of feedback destroyer products from each brand.

Test 1: Amount of "Spectral Damage" (how much the filters cut into your mix) A good feedback destroyer's filters usually won't even be audible, contrary to some peoples' comments on these forums. They also do not add noise to your mix. This spectral damage measurement is dependent on the depth and width of the filters. Most brands are pretty good in this category, with only Sabine having noticeably wider filters than the others.

Test 2: Average Suppression Time (for ideal feedback)
Most brands are pretty good here, with the Behringer unit being noticeable slower in "Setup mode" (each unit calls this a different thing). In "Performance Mode," dbx and Shure come out on top as the fastest. Keep in mind, that in "Performance Mode," when you have music playing through the feedback destroyer, it may take them a while to lock down on the feedback (they're trying their best to distinguish between the music and feedback so they don't set filters on the music).

Test 3: Average Error in Feedback Tone Frequency Estimation
This is the difference between where the filter gets sets and where the actual feedback frequency is. dbx and Sabine come out on top here, with frequency error less than 1 Hz. This is one category that if you buy an older unit from any brand, this error will be significantly more.

Test 4: Filters Falsely Set on Music
This is the hard part for the feedback destroyer box, but again, all of the brands have come a long way here compared to the older (90's) products.
dbx, Sabine, and Shure come out on top here. The Peavey and Behringer performance are still respectable though.

I hope this information is useful. It is as objective as possible, and based on hardware tests. There's too much supposition and too many unsupported blanket statements being thrown around about feedback destroyers, so I feel I have to set the record straight.

Now in answer to the original question (and this part may be a little biased), I would recommend the dbx DriveRack PA2 for your feedback destroyer. Why?
1. The PA2 feedback destroyer is greatly improved over past dbx products and performed well in all of the possible tests.
2. The mobile/computer app lets you see the feedback destroyer filters in real time. It helps to get this visualization. You could also see them on the Shure, but you need an RS-232 connection.
3. Like people have said, most important is to set your gear up correctly. After that's done, second most important is to tune your room correctly to get rid of any resonances in your system (there are always some). For that job, the PA2 has a killer Automatic room EQ system (but you'll need an RTA mic). You would definitely want to tune your room/system before setting any feedback filters. In most cases, this gives you more gain before feedback than the feedback destroyer is going to.
4. PA2 is $500 (meets your $ criterion) and gives you a lot more than just a feedback destroyer, but it serves well as just that (with nice app) if you want.

my 2 cents.
Old 17th April 2014
  #118
S21
Lives for gear
 
S21's Avatar
 

Thanks for the info. It is always good to hear from people who have spent significant time with a wide range of products.

The PA2 sells for about $850 where the original poster resides. Is it good at catching live feedback "pre-ring" or does it generally just allow an audible ring to occur and then limits this quickly?
Old 18th April 2014
  #119
Here for the gear
 

All the (good) boxes that I know of are doing some sort of fft analysis on the signal, so the feedback has to be present in order to be detected/eliminated. There's always a tradeoff between catching feedback quickly and making sure you don't mistakenly set filters on program material. Every brand has the "setup" mode, which will generally catch feedback at the lowest levels they can, because it'll set a filter on any suspicious feedback-like stuff since it doesn't have to worry if it's music.

In "performance" mode, every box will be more hesitant to set a filter (so it doesn't notch out the vocals, etc.). I've done the most "performance" mode testing with the PA2, and in that mode it will catch the feedback very quickly if there is a lull in the program material (pause in vocals, etc.). But if you're sending a full mix, heavily compressed with no quiet spots, all of those other strong peaks in the spectrum may prevent the box from detecting the feedback tone quickly.

If you're asking if these boxes can detect feedback before it happens, not normally. The two things I can think of are:
1. I think one of the Behringer boxes will send out clicks when in "setup" mode to help excite any potential feedback areas. You could also do this by clapping/tapping the mic while setting up. The algorithms in all of the boxes are good enough now to know that a clap isn't feedback, although speaking into them during this setup mode will probably set a false filter.
2. Your feedback will occur first where your system has the biggest resonance. Best way to prevent feedback from ever happening is to take care of that.

There are some Karaoke boxes I know about that implement a small pitch shift on the signal to prevent the feedback from every solidifying on a solid frequency. They do, to some extent, prevent the feedback before it happens. But this solution isn't really acceptable for pro audio (when feedback is strong enough to take off, it just keeps shifting in pitch and never dies out.
Old 18th April 2014
  #120
Lives for gear
Hi Mawg ... Welcome aboard

I'm curious how you go about measuring the setting of false filters. How did you determine what was false?

Also, I think there is a correlation between the speed of setting filters and the avoidance of setting false filters. I expect speed is a trade-off for accuracy.

Just for the record I should say that I was the product manager for the Peavey Ferret D, worked at Sabine, and was product manager for the StageScape mixer from Line 6 (which uses an even more advanced method of distinguishing feedback from music, but is not available as a stand alone product).
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump