Setting of front of house graphics

[Nick Richardson]

Looking for people's opinions and advice when it comes to how they set up graphic eqs on front of house.

It's something I'm very interested in learning about and want to know how different people set them up, so any input is gratefully received! :)

From talking to a few different people and some googling ive heard of three main ways (although there are probably more out there!) to set up a graphic equaliser on foh:

1. Trusting your ears and using music you know well to adjust until you believe it sounds correct.

2. Using a standard vocal microphone and "ringing out" the pa to find frequencies that easily feed back and then drawing them back.

3. Using rta software, calibrated microphone and pink noise to show frequencies and the adjust until the pa produces a "flat" response.

4. Same as number 3, but using the flattered pa as a starting point before adjusting again.

What other ways are there out there? What are people opinions on the ways mentioned? What is your preferred way to set a graphic on front of house?

Thanks in advance for any replies.

 

 

 

 

 

[dbuckley]

I tend to do 3, then do a little 1 to make it sound less obnoxious in the midrange. Which might be your option 4. That's my start point.

 

Then if I have to, EQ to get rid of ring. I'd rather put another graphic (or better still, a parametric) in a mic subgroup (and then, by type, so an EQ for radios on bodies with omni mics, and another for PCC160s or whatever) but if I cant do that then I'll tweak to main graphic to reduce ring.

[Bobbsy]

In the ideal world I'll have two equalisers.

 

One will be part of the permanent install and hopefully buried in a locked rack in a locked room. This one is set up--carefully--with an RTA and it's purpose is to make the room as flat as possible.

 

The second will be with the FOH mixer and used to make the show sound as he wants it--including ringing out the system for feedback reduction.

 

I realise that this isn't always possible so, with only a single EQ to play with, I'd be more or less option 4. I use SMAART to get a basic starting point (which almost always works on pink noise but sounds ghastly on real sound) then adjust from there knowing I have a repeatable starting point.

[Doug Siddons]

I tend to use option 1 then 2, then take a walk around the venue with the music running. I've never been really convinced that software is better than experienced ears when your setting up a system and taking it down every night, and EQ ing a room for different types of shows. It does have its place when you can't quite put your finger on that frequency which is annoying you

[J Pearce]

I try to do as little on the graphic as possible, and often use graphics on group inserts to remove feedback from a certain group of mics.

I often use a parametric EQ to set the tone of a system, and have experimented with dynamic EQ to try to balance out the perception issues documented in the equal loudness curves.

 

I once tried 3 methods on a 3 night run of a gig with a band that produced a reliably similar sound each night.

Night 1 was a quick remove the 4 worst squeals and pull down the 2k-4k area for a more relaxed tone.

Night 2 was a full Smaart RTA for a 'flat' system.

Night 3 was a tune by ear to known tracks.

 

The most natural 'transparent' sound with good vocal intelligibility was night 1.

Night 2 I had endless trouble getting the mix to settle. It's hard to quantify, but I suspect I was having issues due to phase response of a hacked about graphic EQ.

Night 3 sounded good, but I had trouble with intelligibility of vocals.

[Wilflet]

 

Night 2 I had endless trouble getting the mix to settle. It's hard to quantify, but I suspect I was having issues due to phase response of a hacked about graphic EQ.

 

 

Thats the problem with trying to do system set up with a 31 band graph.

 

And hence my normal aproach (we're onto what #6 or something here)

 

From predominantly a system techs point of view- system set up using measurement software and reference mic, but using the parametric eq within the processors (admittedly not all of you will have the benefit of fully configurable networkable processors) get the system flat. Leave graph flat, hand over to the bands engineer along with desk- if he wants to tweak anything for personal taste he can. has the advantage of if theres an engineer with questionable opinions about how it should sound then once he's gone flatten the graph and its back to the mixable starting point.

 

 

[Oldradiohand]

The (budget model) theatre I work in has a two channel graphic - it's in the two main stage monitor feeds! The FOH has a crossover for the subs but no graphic which suits me fine as I've never really liked the sound of graphics - all those overlapping bands adding differently phase shifted audio to itself. Gives me a reasonable and consistent sound to the auditorium and I can ring out the mics on an individual basis. The less processing the audio undergoes the better it sounds to me.

[Mark Payne]

Hmmmm

 

I don't use graphics on FOH either in digital or outboard form. Once you are using world class audio systems I think the less is more approach is the one to use.

 

If you pink the system then only do one side and copy over.. L/R comb filters occur otherwise.... unless you use non coherent pink sources for L and R.

 

Make sure you get a 3dB per octave LF contour increasing from 1K going down, by the time you arrive at 50hz you should be round 12dB up cf 1K. Make sure you know how to time align subs to tops using basic principles (get your laser out) and if possible using acoustic measurement. Systems for contemporary live music reinforcement are not flat. They have LF contour.

 

I live with the system contour in the HF that my trusted system provider had encoded for me. I tune low mid according to PA coupling (a function of rig size). With d&b you will be tweaking CPL, with L-Acoustics its LF Contour and/or Zoom Factor. The only real response measurement I am doing is to ensure top to sub alignment is optimised at the Xover. This is done by first aligning the impulse responses (I use Tuning Capture for this).... and then once you are within 1-2ms the rest is done by repeat measurement around the centre by tweaking relative delay by about 0.5ms at a time till you get the money. I use Tuning Capture for all of this.

 

I strongly suggest to all that you appreciate that LF and low mid measurement needs the measurement mic on the floor. If this is news to you then you have already been tuning PA systems tying to address the direct and reflected paths inherent in a mic stand and chasing comb filters that are not real. Stop doing that. My honest opinion is that in the real world of a live space we cannot learn anything worth knowing above 1K by measurement. There is so much that we cannot control in the measurement, our results cannot be tuned around. you are wasting your time with this. Low mid and LF ... yes, above 1K ... forget it. If it sounds nice, its nice.

 

I feel that the graphic is an answer to a question you should not be asking and a solution to a problem you don't have.

Most room tonality is generated by comb filters due to reflected energy or box to box interaction. You cannot EQ for this. If you steam more in, the comb will just take more out.

 

We do quite a lot of this on my 4 day master class at SFL.

 

Mark

[S&L]

my view from the bargain basement/bars/halls/small theatres working with bands and other musician assortments.

1. before I get near a stage, design out feedback - why have an sm58 when another mic behaves more reasonably. work out best placement of all items, wedges, FOH, mics etc - I know one fellow who will use in-ear monitors to further minimise these issues but I don't have the money to have those laying round and some musicians flat refuse and insist on wedges.

2. rudimentary ringing out of monitors - I rarely have any feedback issues with FOH. I have graphics on a couple of monitor channels to aid this process. IF I detect an issue with FOH feedback I may notch out a given frequency but it's rare and I dislike using such a blunt instrument to address what can just as easily be designed out.

3. play my test track through FOH speakers. I use hall and oats, she's gone but the point is to play something you know well and use repeatedly - my team loathe the song now but the know why I keep using it. I'm listening to hear the bass rounded but not boomy and the hi-hats penetrating and crisp.

4. walk the venue listening to the test track. checking the sides of the room as well as the middle, balcony seats etc and shape the graphic if needed to replicate what the test piece should sound like.

 

the above takes 5 minutes in a bar and 10 to 15 minutes in a theatre. I wish we had the budget for the proper kit but since I am getting plenty of compliments about the sound that I mix, I guess the low tech way is sufficient for my purposes.

[azlan]

my preferred way of doing things is to use a parametric EQ to tune FoH to sound good with a recording (something I know well and that is reasonably representative of the style of audio for the night) and then apply graphic EQ, ideally over individual subgroups/inserted over channels where possible, and then if needed, notch out a couple of frequencies with graphics over monitors if needed.

[Nocastus]

Generally method 1 for me. Ringing out is usually more important when setting monitor eqs than FOH, although obviously in some rooms there might be some attention to be paid to FOH ringing too. Software has never worked for me. The most important thing IMO is to remember that the acoustic will change when the audience come in: it will be less reverberant and the bodies (and the moisture from their breath) will absorb more highs and high mids than the empty room. So if it's too hi-fi sounding during soundcheck it's liable to be a bit dull and dead sounding with a crowd in the room. Of course if the band's management is listening to soundcheck you might need to make that a little soft sounding but be ready to lift the high mids a fraction for the show.

[RustyBrooks]

I'll use a combination of methods...

 

I'll design a system to put in so I can predict how it will respond and have the right tone into the room before ive turned it on, though this only really works with full scale touring line arrays, once this is done, I'll have a fairly predictable day to day response from a system, then on the first load in, normally to production rehearals, but some time into the first proper venue if we are using a rehearsal space not a venue, will use sys tune to capture and tune the system response, flattening the response of the room to a fairly neutral response that will work for the engineer, or me if im mixing, knowing how much low they need, whether going for an l acoustics prescribbed 12db lf contour, or, if working for some engineers, a flat down to 200hz system, allowing the engineer to put the weight in (tends to be the studio guys like this approach) then have a subtle check of the system with cd/voice etc to make sure we're in the right ball park, and that will be a go, at least for a one off. Then, if touring, do a sound check, make any other tweaks to sit the system to the band and check coverage and all that, with, this is the important bit for me, then taking a trace of the pa in an empty room before doors. This is then my stored reference for the system for that band/engineer pairing. I can use this reference to check how the room changes compared to sound check as people come and and compensate accurately, or make feel responses based on whats happening.

The BIG bonus of this approach is, for instance on a festival run where you don't get sound checks, even as headliners, you always get a chance to pink to the system, so by useing a calibrated soundcard, a 10 second burst of pink noise, can tell me, compared to the reference trace, how much gain difference there is between what I would like/need, and also any eq changes in the system to fit my curve (systune makes this very easy with the virtual eq) then, come change over, relase the mutes on the desk, and fingers crosed, the first tune sounds as it SHOULD.

 

Yes, easrs are good for hearing through some of that subjective waffle or for pulling some feel out of a dodgy room, or for making things sit right, but parametric eq's in a ssyetem controller, are far more repeatable, and reliable then your ears. Though there should always be a flattened 31band geq at the start of every show so you can just grab and remove stuff that comes at you mid gig..

[Mr.Si]

Looking for people's opinions and advice when it comes to how they set up graphic eqs on front of house.

 

 

I don't use one.