We do not get to talk to guest speakers let alone have multiple events to ‘train’ them.
Clearly they think that they are excellent doing what they do.
Else they would have learned by now and somebody with access to them would have told them.
So why do they do what they do if they do not think it is somehow good?
Does anybody try to do things poorly on purpose?
Most speakers try to do their best. But most are not true professionals.
Some think they are. Some think they are greater than sliced bread and have egos to match.
And some are just old and set in their ways.
And then there are the ‘pros’ who refuse to use a mike at all.
Those are usually loud enough and rarely the problem.
The worst problem I recall was a guy who was an ‘emoter’
He thought bending over whispering then bolting upright and yelling was somehow going to help us understand what he said.
That guy was all over the stage, at times like a gymnast of sorts with his movements.
I could not understand half the words as they were way too soft.
And the prancing around was a distraction.
I have no idea what he talked about.
I almost think he was a charity case our pastor gave a chance to talk because he felt sorry for him.
OK I watched some videos on upward compression, none of them with live sound in mind
and the implementations of the upward compression plugins use all the mentioned method to avoid the raise of noise and extremly low sounds
low and ambient sound gets mostly raised very strong and this is the use most of the videos trigger
to get the room and ambience up
BTW. My Experience over decades is that doing automatic level changes in a live system, especially when the situation is not well prepared before (good microphone practice, low background noise, perfect tuned sound system and so on) is very risky. Especially if you allow the automatic to raise the volumn will almost always cause feedback. Ok, I do not have articles or academic studies to prove that, only my experience.
Yet they have automatic feedback killers which work just fine.
Combine the automatic DR squashing and EQ following, with the automatic feedback killer and still no problems.
Should be no need for automatic feedback killing but always nice to have just in case:)
“Yet they have automatic feedback killers which work just fine.“
Do they?
I’ve never met one that’s anywhere close to good enough to use ‘live’. They all are too slow and require the feedback to be uncomfortably loud before they do anything useful.
And that’s before you get on to how it deals with intentional feedback, or long notes that sound like feedback.
I’ve never met one that’s anywhere close to good enough to use ‘live’. They all are too slow and require the feedback to be uncomfortably loud before they do anything useful.
And that’s before you get on to how it deals with intentional feedback, or long notes that sound like feedback.
Imagine if Hendrix had had one ?
EXACTLY!!!
Just last night I actually had a show, 8 piece band, three horns.
I’m sure the trumpet solos would have triggered any “feedback killer”.
Not a hint of feedback during the sound check or the show and no FBK in use.
If your going to or think you need to use an FBK at all, put it inline, activate maybe
4 or 5 filters at the most, push up the system gain till it rings/feeds back, let it grab that one keep pushing till it grabs a couple more, bring the system gain back down
to the normal level and then lock all the filters so it will not keep hunting and notching.
That’s my normal tutorial to people who either have no other EQ on a system than an FBK
or can never figure out how to do it by ear with an EQ.
@MikeC
+1 from my side. I never wanted to use such tools. In fact it is not that complicated to find the most critical frequencies manually. The only advantag of such tools are the narrow notch filters that are used. Ifi want any feature in any actual digital desks is a replacement of that graphical EQs with a multiband Notch filter, lets say 6-8 filters with a Q-Factor between 20 and 60. classical graphic EQs are like axes used for brain surgery.
Especially in a fixed environment of a venue it should be possible to set up the system once, so automatic feedback killer should not be required at all.
“If I want any feature in any actual digital desks is a replacement of that graphical EQs with a multiband Notch filter, lets say 6-8 filters with a Q-Factor between 20 and 60. classical graphic EQs are like axes used for brain surgery.“
This! I wonder if it could be added to SQ and above…
The filter I recall, which I suggested for my previous church to use, monitored the frequencies in use and notched whatever ones were triggering feedback at that moment. They changed/adapted in RT as needed.
It is clear how automatic feedback suppression works. Nevertheless I dont like it. One reason is that you can find that frequencies manually too, especially in static systems. The second is that this algorithm cannot distinguish between a real, unwanted feedback and a longer tone from an instrument. This algorithm has no musical sense. It is like AMM. They are more or less worthless for musical program. No automatic system will replace a well educated sound engineer with trained ears.
You can compare it with cars. All that automatic gear, ABS, traction control and so on will not help if there is an untrained person on the drivers seat. First you have to know how the physics works before you can try to find the limits. Automatic systems can fail and then the person on the desk is in more trouble as without them if this person in not well educated. Thats it.
does not matter whether tone or freq or something else if that puts the system into feedback
WHEN there is feedback the system EQs out that frequency
you could certainly design it to return to flat slowly if you did not want to keep that EQ active
not using automatic feedback killer would not stop the system from squealing whether tone or other freq causing it
so the choice is automatic feedback killer or just pull down the faders when it happens and hope it is safe to slide them back up
of course the experts here would have rung out the room and EQed the mikes and done sound checks to ensure feedback would never be possible no matter what.
but for small venues with less experienced operators AFK is useful
Hmm, in fact such devices do not EQs out the feedback frequency WHEN there is a feedback but rather AFTER the feedback starts. It needs some response time. Depending of this response time the feedback can be really painful.
But as I said „I dont like such devices“. So do what you want and be happy.
…and do not forget that if you tell „less experienced operators“ that there is an automatic device which will compensate operation errors ( which is in fact one of the intentions behind such devices) they will feel free to do operation errors.
My approach is more that I use good sounding components which fits the needs, place them carefully, tune the EQ of the system to eliminate remaining problems of room (if required), use good microphones, EQ them carefully, use carefully some dynamics, only if required. And thats it. Thes simple steps brought me success over decades, so I dont think that I will change that strategy.
Strange.
My UPS blocks transients so my pc never blinks.
A good servo control of feedback should stop it dead so fast that nobody could even hear anything.
Moreover, a properly designed auto feedback killer would let you set a level under the feedback trigger point that it would maintain as the highest the signal could reach. So not even the risk of a tiny click when the feedback would have happened because it would never happen at all. In short it stops the feedback BEFORE the problem happens, not afterwards.
Like it or not, sooner not later, mixers will have AI built in that not only fix feedback issues BEFORE they happen, but will automatically adjust the entire show based on how you describe you would like it to sound. Your job will be limited to setting up, tearing down, and maybe turning on the power switch too.
A good servo control of feedback should stop it dead so fast that nobody could even hear anything.
None of the feedback killer units or software can tell the difference between
feedback and a sustained note on a guitar, horn, keyboard ect.
They just look for a peak in the audio spectrum that jumps up over the base line average level, they digitally identify the frequency and apply cut at that frequency.
They first have to see/hear the peak frequency before they can do anything about it, leaving everyone hearing feedback till it may get reduced.
A good operator can almost sense feedback before it starts and take care of it before
anyone notices, short of someone sticking a mic right into a stage monitor!!
Feedback killers are nothing new, Altec Lansing had one in the early 70’s designed for
speech reinforcement systems that would shift the feedback frequency a few hz in an attempt to eliminate it, analog by the way.
You can believe that. Perhaps the ones you have seen have that flaw.
It IS possible to architect design and build an auto feedback killer withOUT that defect.
I have seen feedback the instant someone flipped a switch or pushed a button.
Good luck with any operator sensing that.
I have seen feedback situations where short of a panic button there is no way it can be stopped quickly short of power off.
Feedback killers may not be new but new ones should be far better than the ones folks here seem to think act like all of them.
I have seen feedback the instant someone flipped a switch or pushed a button.
Good luck with any operator sensing that.
True, like if someone disengages a 30db pad switch or like my example someone sticking a mic into a monitor, those are what test your reflex skills in hitting the right mute button.
It *IS* possible to architect design and build an auto feedback killer withOUT that defect.
NOT A CHANCE…there is no way a feedback killer is going to know the difference between 440hz feedback or a horn player holding an A natural note.
Maybe in post production software you could sample a burst of feedback and the software could scrub through the entire audio file looking what sounded like the sample, but that’s not going to happen in anything close to real time.
You can believe that. Perhaps the ones you have seen have that flaw.
It *IS* possible to architect design and build an auto feedback killer withOUT that defect.
Lets look at this closer…The ones I have seen are like the ones everyone else here has seen either in a stand alone unit or as a function as part of a DSP.
They all have some form of sensitivity adjustment, filter release times, live and locked filters and none are or will be smart enough to know the difference between music and feedback.
You then say it “IS possible” to me that says none exist, have you seen one?
Or is this yet another thing you have read about and talk about but have zero practical knowledge about?