Main EQ Curve questions
Old 18th August 2010
  #1
Gear addict
 
parkay909's Avatar
 

Thread Starter
Main EQ Curve questions

What curve should I be shooting for with my mains? I have a white instruments EQ I am going to hook up but I figure I am probably not going to want to tune my room to flat. Any thoughts? I will be using the mains for music.
Old 18th August 2010
  #2
SAC
Registered User
 

You are limited to EQing the DIRECT sound from the speakers, not the interaction of the room and the direct sound. In other wors, perhaps not very meaningful at this point, but nevertheless more precise, you can only EQ minimum phase signals - not non-minimum phase signals.

Thus you want to employ a method of measuring the speaker response that affords an effectively anechoic response, such as TEF or a well controlled and gated (windowed) time response.

But, in any event, EQ is not the proper manner by which to treat the broad interaction between the direct signal and the various other real and virtual sources in a bounded space.

For this, you address issues in the time domain, which 'magically' resolves the frequency domain anomalies resulting from the interaction of various non-minimum phase signals.
Old 18th August 2010
  #3
Gear addict
 
parkay909's Avatar
 

Thread Starter
more specifically once I have the room tuned physically as close as possible to flat what am I shooting for with my main EQ?

At all of the large studios I have worked at they have a White 4700-2 EQ strapped across their mains. When they call in certain experts to "tune" the monitors they are pretty secretive about the curves they are trying to match which are not flat. I am wondering what I would be trying to shoot for after my room is treated and I have tweezed speaker placement, etc.
Old 18th August 2010
  #4
Gear Guru
 
DanDan's Avatar
Normal

It is common to use a 'House curve' in professional studios. Digital processors are sometimes used with several curves. One for TV, one for CD, one for Download perhaps.
There are many of us, Bruel and Kjaer included who subscribe to a curve which is very broadly speaking +3 around 100Hz, -3 around 10KHz, and falling. This has been found to enhance the translation of mixes to the outside world. Two examples of it:-
Understanding RTA at studiotips - tips on studio design, acoustics, and wiring. A simple view.
The Bruel and Kjaer QR2011 test record. P258 Acoustic Techniques by Everest.
Intended for home Hi Fi tweaking but the same curve.

For what it's worth I am the best mixing engineer on the planet, I use this curve, (broadly speaking using S3A onboard eq.) and IMHO this curve delivers uncannily.

DD
Old 18th August 2010
  #5
SAC
Registered User
 

To repeat, the EQ is used ONLY to EQ the direct signal - the signal as it leaves the speaker, not the direct signal after it has superposed (combined) with any additional source in the a bounded space.

You used to see SR rigs with 30+ channels of EQ in a dedicated roadcase, now you see several only for direct signals. For that matter, Don Davis, while with Altec developed the first 3rd octave EQ and he was among the first to identify the limitations of EQing non-minimum phase signals!

We have no idea what the response of the speaker is, how it was designed and if it was designed to accommodate any particular spatial loading or if there is selectable shelving intended to help compensate for this nor how the speakers are mounted! All basic functions that substantially effect such a response.

We don't even have a baseline response! Free field (as if someone is actually going to suspend their speaker from a piece of rigid mono-filament in the middle far from any boundaries) or otherwise!

We worked extensively with B&K when the were in the field demonstrating their TDS units. No one presets an EQ and assumes it to be valid for any purpose without a baseline response and the knowledge of the spatial loading of a speaker.

Again, even after you have the best world class room available, you measure the direct signal response at the listening position (using either a TEF with its tracking filters) or a time based impulse response windowed to exclude early reflections and convolve its FR. You then can EQ the direct signal of the speaker relative to the listening position.

EQing the signal is the LAST step after room tuning, and it is integral to the room treatment process. At that point the speaker's response is measured at the listening position(s), and the non-minimum phase regions that can be EQed are done so to the degree possible for a flat response. And then they are not changed.

But then again, many choose to do most of their configuring by simply following the 'oft cited' methods.
Old 18th August 2010
  #6
Gear Guru
 
DanDan's Avatar
Repeat

The use of White, and other 1/3 octaves has diminished. It has been superceded by parametric equalisers from Meyer and Digital units by Sony, XTA and so on.
There are reasons for this but let's not lose track of the OP's question.
The B and K curve is almost identical to the studiotips one. There is a family of curves in Newells book. All have HF rolloff. There was a genuine question here, not asked often enough IMHO. Flat doesn't work well for us Mixers.
I often find the eq on powered speakers is sufficient to get the 'slope' needed.
Eq at the speaker or adjusting the crossover, same thing, is perfectly legitimate.
If a room is skewed, a slight skewing of the source can make the combined reproduction system better. Many look down on the use of Eq without regard to reality. In live sound we absolutely rely on system Eq in all it's guises. Why not in a studio? Crossover's are Eq's and they are adjustable for good reason. And by the way REW = Room Eq Wizard....
Notwithstanding all of that, the B and K and others' curve is there for translation purposes. There is a similar principle involved in the cinema X curve. The why in both cases is not fully explained or understood, certainly by me. However as a working mixer I can state with certainty, the sloped curve works.
A 'house curve' is one custom generated to persuade a particular set of speakers to deliver approximately the B and K response in that specific room. It is not generic in any way.

DD
Old 18th August 2010
  #7
SAC
Registered User
 

Oh, sorry... Never-mind what you can EQ and what you can't (but which far too many still try to do) , and never-mind the response of the units you are attempting to optimize, just plug in a generic 'house curve'.

I knew I was making this acoustics stuff too hard.
Old 19th August 2010
  #8
Lives for gear
 
Dange's Avatar
 

Just out of interest, what would be some examples of types of spatial loading on the loudspeakers? Whether it's sitting in a modal null/peak or SBIR for example?
Old 19th August 2010
  #9
Gear addict
 
parkay909's Avatar
 

Thread Starter
What Newell book are you referring to? This is good advice so far and thank you.

mainly from what I see in the forums people are looking to make their rooms/monitors as flat as possible. I know that in most of the studios I have worked at their mains (all Augspurger designs) are not tuned to be flat but have a curve. I am trying to find out what curve a pair of monitors like that would be tuned to.
Old 19th August 2010
  #10
Lives for gear
 
Weasel9992's Avatar
 

Quote:
Originally Posted by SAC View Post
Oh, sorry... Never-mind what you can EQ and what you can't (but which far too many still try to do) , and never-mind the response of the units you are attempting to optimize, just plug in a generic 'house curve'.

I knew I was making this acoustics stuff too hard.
All SAC is saying is that the only place to put an EQ is directly behind the speakers in the monitoring chain, so we need to know more about the space in terms of its response. He's right.
Old 19th August 2010
  #11
Gear Guru
 
DanDan's Avatar
Welcome

You are welcome park.
Here's the Bruel and Kjaer. Application Notes - 1979-1970 - Brüel & Kjær
Note it is 'domestic' orientated, but strangely pretty much the same curve appears all over the place. e.g. at studiotips and many other books.
Don't quote me but Boner may have been involved in early research on this.
The book, which is brilliant, is
Recording Studio Construction by Philip Newell

Hi Frank, we were posting simultaneously. I don't quite understand what you said, but to be clear, the B and K curve is the target curve, i.e. a measured response frequency response in the room.
It can be persuaded to happen by Speaker Processors, Crossover adjustment, Level/Eq controls on active monitors, Parametric or Graphic Equalisers.
It matters little what curves or manipulations happen in those electronic realms as long as the target translation curve is achieved.
For a working Mix engineer, the difference is night and day, the results remarkably fast and accurate.
Best Regards, DD
Old 19th August 2010
  #12
Gear addict
 
parkay909's Avatar
 

Thread Starter
i think we are starting to hone in on the answer I am looking for. I am not looking for acoustical advice on an existing space.

Lets say for the sake of argument I already own a nice acoustically tuned room built out by a big name designer. I am going to hire one of the "name" room tuners to come in and tune my big honking mains. In this hypothetical room I have (fill in the name of a custom speaker designer here) monitors in soffits powered by big huge Bryston amps and a white 4700 EQ. When this room tuning expert comes in to adjust this EQ what sort of curve will they be trying to match the EQ/SPEAKERS/CROSSOVER/AMPS to? I know for a fact they will not be trying to get me flat response on my mains.

I think DanDan's answer is pretty close to what I am trying to find out. I know of two different well known "room tuners" who have different curves they will try to match your speakers to and that is also partly why the studio owners may pick one or the other. For instance I have heard that one well known (and very well liked I might add) room tuner tunes your mains to sound like a big set of NS-10s and that is why an other tuner (also well known and well liked) was chosen at this facility.
Old 19th August 2010
  #13
Gear addict
 
parkay909's Avatar
 

Thread Starter
Or to be more specific what response curve are they trying to get the speakers to produce? I am not looking for an answer like "set the 100hz slider to +3 dB" because obviously every speaker/room combination will require different settings on the EQ and room treatments to get the desired result. I am trying to find out what the desired end result coming out of the speakers should be, and the answer is not flat response.

People don't pick NS-10s because they are flat, nor do they buy big custom mains because they are flat. They want good mix translation, loud, and little fatigue. Flat is good if everyone is going to listen to your mixes in anechoic chambers or outer space or whatever... Wait, in space they won't hear it at all!
Old 19th August 2010
  #14
Film, post, and cinema room EQs are a different subject altogether, so sticking to music, traditional rooms focusing on modern R&B, hip hop, and rap often have Augspurgers with a little curve to them. Guys like Coco are known for tuning rooms like this. It's really a genre-specific thing, and something subjectively pleasing to that style and helpful in translation with that unique set of circumstances.

However, for the vast majority of music mixing situations, as flat as possible within the limits of the system is the thing you should aim for. Sure, a Coco room will have an audio bling factor, and people might dig the hyped sound, and it could be helpful for marketing, but unless you're in a big room with Augspurger mains catering to that specific market, I'd skip the "magic" tunings.

Still, with a main monitor tuning in a large, traditional music control room, the last word should be upon listening to a wide variety of music on the EQ'd flat response. There may be certain anomalous things that the measurements don't do justice to, and a minor tweak here or there can make a system sound subjectively more "right." This is still a practical compromise however.

That said, I think that particular old control room paradigm is past its prime. Monitors are better these days (even big ones - Quested, ATC, Dynaudio, PMC all have mains that can be accurate and detailed), and no longer are you limited to 1/3 octave EQ. No EQ in the monitor path is the goal, but if you absolutely need some due to real-world factors, as few bands as possible of parametric is a better choice than 1/3 octave. We have an xta (crossover with EQ) on our Boxer T5s in our music control room, but I wouldn't let an EQ anywhere near my monitor chain in my mastering room.
Old 19th August 2010
  #15
Gear addict
 
parkay909's Avatar
 

Thread Starter
Smile

AND JAY IS THE WINNER (so far)!!!! Give that man a cupie doll and a cigar!!!!

You dropped a few of the names I was hinting at. Ultimately when I do build my room I will want that hyped "bling" sound (I am listed as a "rap" engineer on allmusic) but hiring Mr. CoCo or Mr. Augspurger is a little out of my budget and my question is mainly academic at this point. Obviously nobody is going to publish their special room "curves" but I am looking for some good references. I wish I had done a plot of the room I was working in before I left!

Any more advice on where to start or where someone may have published some of the "secrete recipes" for me to study while I am still in the design (dream) phase would be wonderful.

And I am old school, I am building a Heathkit crossover for when i finally get around to building my mains... Already bought the Brystons and the White 4700.
Old 20th August 2010
  #16
SAC
Registered User
 

And I made the mistake of thinking you were interested in knowing what to do rather than who did what.

And now you ask "where to start?"....

You have received the same answer several times already...

You can only EQ the direct signal.

And if measured "in" the room (meaning the direct response is not isolated by either tracking filters (TDS), windowing or other techniques effectively 'isolating' the direct signal from other signals - for example, reflections), you can effectively EQ only those response regions that are "minimum phase" - 'simply' meaning, that the measured response is not comprised of superposed (combined) direct and indirect signals from the room and the speaker.

Any regions where the measured response is non-minimum phase - meaning it is a result of the superposition of direct and indirect signals (e.g. reflections, etc.) - EQ will not have the desired effect as there is no 1:1 correlation between phase and amplitude.

And yes, you ideally want a parametric EQ with selectable and variable Q, as the regions that can be EQs don't cooperate and coincide with fixed EQ, regardless of the partitioning..

A precise definition of minimum phase is a detailed mathematical concept involving positive real transfer functions, i.e., transfer functions with all zeros restricted to the left half s-plane.

Likewise, there is a 1:1 relationship between the magnitude and phase as is provided by the Hilbert transform.

I apologize if this seems like technobabble, as the concept of minimum phase is technically much more complex than that mathematically. Nevertheless it is correct.
Old 20th August 2010
  #17
Gear Guru
 
DanDan's Avatar
Target

park, I am still not sure that you have understood what I am saying.
The +3 at 100 sloping gradually to -3 at 10K and so on is the result. That is the goal, the ambition. It is not the settings for the eq. If you were to play pink noise and view it on an RTA you would see the slope described. It is pictured at Understanding RTA, studiotips, and various forms of it are described in Newell.
It is not necessarily an on axis response, more likely a spatial average over a considerable listening area. Much work is done in large control rooms. PMC for instance have a distinctly rising on-axis response, but they intend their speakers to shoot over and beyond heads.
What is 'flat' and where would be an interesting debate.
One of the reasons for the HF roll off is the difference in usual levels between studio mixing and domestic listening. At elevated SPL's say 85dBC, let's call it average, with peaks to 95, the ear becomes close to flat. Thus we hear the HF (and LF) a lot better than at say 65- 75dB. Without a HF rolloff the treble would take the head off you.
In Mastering, lower levels are typical so this effect is less at play.

DD
Old 20th August 2010
  #18
Lives for gear
 
Dange's Avatar
 

Quote:
Originally Posted by DanDan View Post
What is 'flat' and where would be an interesting debate.
You could start from basic psychoacoustic theory - Equal-loudness contour - Wikipedia, the free encyclopedia *



Interestingly 'flatness' appears to vary with intensity of sound, not massively but there is a variance.


*It is Wikipedia I know! But this thread isn't a research paper so I think I'm safe to reference it
Old 20th August 2010
  #19
Gear addict
 
parkay909's Avatar
 

Thread Starter
Main EQ Curve questions

DanDan your on the money as well (I'm not looking for "magic" EQ settings I know there are many factors that influence). Dange also on the money. This is great info for me to start with! Thank you all!
Old 20th August 2010
  #20
Gear addict
 
parkay909's Avatar
 

Thread Starter
Main EQ Curve questions

SAC thanks for the input as well. Once again my question is mainly academic as I am trying to better understand the overall objective (other than flat) that a room tuner may go for (how they would "hype" or "bling" the mains).
Old 20th August 2010
  #21
Gear Guru
 
DanDan's Avatar
Opposite

park, I don't know what those guys do but hype and bling would be the opposite to what I do. My goal is translation. I have spent so many hours listening to mixes in three different rooms and mentally averaging. Spare me. The curve eliminates that, for me this is fact. Nowadays when I reference on my second system, in another room, it is almost boringly right first time.

Simply put, I can hear very well, not physically perhaps, but certainly mentally. We don't listen in the same way as humans If I mix in a flat environment i.e. speaker+room combo, the mixes sound dull outside in the real world.
An example, my S3A's are +5 at LF, -2 at HF, and another -2 at the 6K shelf. They are nowhere near dull sounding to me or anybody who comes here. In fact they are quite revealing. The measured response is.... well you guessed it, the classic.

Those White Eq's were typically adjusted using pink noise and an RTA. Some RTA's could store and average, but it was also done by eye and ear.
Things have moved on in many ways but it would be foolish to presume that it is all better. Just listen to the records done in those old scenarios and listen to the current crop for instance. Better?

One reason why opinions differ is the shifting sands of time. Over time, domestic systems and professional systems seem to be drifting towards 'flat'. There is an intuitive assumption that this must be better, speaker manufacturers like to publish that flat line, the public 'understand' a flat line. Power Response in actual rooms both pro and domestic is another matter entirely though. In my experience and opinion, strangely, everything and nothing has changed. Modern domestic environments are full of hard reflective surfaces. Electronic Bass boost is everywhere. It is difficult for clarity to survive. In the old days it was carpets and furry wallpaper. LOL.

DD
Old 20th August 2010
  #22
Gear addict
 
parkay909's Avatar
 

Thread Starter
Main EQ Curve questions

And 703 board!
Old 20th August 2010
  #23
Gear addict
 
parkay909's Avatar
 

Thread Starter
Main EQ Curve questions

The spot I mainly worked at had Augsperger mains with a White 4700 and JBL 12dB/octave crossover. When I asked the studio owner why they didn't have a more modern crossover/eq/digital DSP flavor of the month processor he replied "Don't you think they sound great (as is)?"
Old 20th August 2010
  #24
I don't want people to think a CoCo tuning is all about subjective hype. Some of it is just using his experience with the behavior and limitations of those systems under those conditions to adapt the tuning to the application. He has his bag of tricks that he has found to work in practical application over years of serving that niche.

The mains in rooms like that are often for impressing the client or spot checking low-end extension, and the mixer will tend to use the close fields of his/her choice for most of the critical listening. Technical perfection is great, but as a studio owner you also have to take into account practical concerns. I'm more of a minimalist flat-tuning kind of guy, but I can certainly appreciate the market pressure to appeal to your client base so they keep writing checks!
Old 20th August 2010
  #25
SAC
Registered User
 

There is much more to the subject of "Equalization" than simply loudness curves or the settings on a particular brand EQ.

This topic suffers from much of the same issues that that confuse a discussion of reverberation and reverberant sound fields, as most are not aware of the definition of reverberation and mistake what is associated as ‘reverb’ with the delay some effects knob on an amp says it is (but is not!).

The fundamental issue that few are addressing is that changing the setting on an EQ unit does not necessarily result in a 1:1 change in the corresponding phase of the signal.

A precise definition of minimum phase is a detailed mathematical concept involving positive real transfer functions, i.e., transfer functions with all zeros restricted to the left half s-plane.

Likewise, the relationship can be determined empirically by examining the relationship between the amplitude and phase of the signal by examining the Hilbert transform between the impulse and doublet response and examining the relationship for regions of a 1:1 relationship which indicates a minimum phase relationship.

To ignore this fundamental relationship is to twiddle knobs thinking one is making a desired difference when instead they are destructively modifying relationships elsewhere.

I know few are interested in understanding this aspect, as after all, I am constantly reminded that folks simply want answers and results. So with that, I'll bow out of this topic.

For what it’s worth, for those who desire the ability to evaluate such a response (without bothering to understand the underlying concepts), there are a few analysis platforms which provide such analysis, and can even generate precise parametric equalization settings for the minimum phase regions. But one wonders as to the value of discussing such aspects as it is a bit more involved than the ETC response, and such a discussion is predicated upon understanding some of the component relationships thereof, especially as we all know how questionable some here feel about the discussion of such an 'advanced' and 'esoteric' function…
Old 20th August 2010
  #26
Off the topic of specific tunings and re: room EQ in general...

At the risk of oversimplification, you can't fix a time domain problem in the frequency domain. EQ's indeed address minimum phase issues but are powerless in other areas.

Even if you make it look perfect to the measurement mic, it will likely be orders of magnitude worse almost everywhere else. If you can put your head in a vise and nobody else in the room needs to listen, then maybe...

Another practical, though limited and simplified example: imagine a response with a 12 dB hole in it, say at 100 Hz just for fun. Well, there's probably a node at your measurement or listening position. The physical wavelength of that 100 Hz component corresponds to some distance between boundaries and a reflection is causing destructive interference.

Let's say you decide to fill in that hole at that particular location with EQ. The easy part to imagine is that as you add 12 dB hoping it will fill in at that location (which you actually can be pretty sure it won't, but that's a different story) you might not be surprised to find 12 dB too much 100Hz at a bunch of other locations in the room (not quite as simple as that, but nonetheless).

The part people often forget is that you happen to have a problem at that frequency, and if you go adding 12 dB, you are exciting the room even more at that problem frequency. You are over-exciting problem modes, and additionally, if there is trouble with ringing, it will be ringing even more. As you try to fix the problem, it grows.

Again, an oversimplification, but perhaps a simple look at the problem to help non-acousticians to begin thinking in new directions. If you now want to progress into Hilbert transforms, don't let me stop ya...
Old 20th August 2010
  #27
Lives for gear
 
jhbrandt's Avatar
 

+1

Jay said it all. Check please!

Cheers,
John
Old 20th August 2010
  #28
Gear Guru
 
DanDan's Avatar
Broken Record

I think there are two issues here. One is the room the other the speaker system. Many of us know it is almost valueless to consider one without the other. However, I would like to leave a thought on the table.
The is not verbatim, from Newell.

In Elstree 71, mixing Clockwork Orange.
Large really full range KEF's 2 M from the mix position.
The sound was accepted as full range, great.

The main, calibrated, normally used, full range monitor were then Eq'ed to sound 'the same' as the KEF's.
The slope was -3dB per octave starting at 2Khz. So -3 at 4K, -3 at 8K, -9 at 16K.

The mains were quite far away, big cinema mixing room, but quite dead.

It could be said that Cinema mixing or Surround mixing has very different requirements and goals.
That different?

One would have thought that proximity effects would have enhanced the bottom and top on the KEF's requiring similar electronic adjustment on the mains. The opposite happened. What effect is at play here?

Let's not forget B and K and almost any text book concerning CR monitoring. There is almost invariably a suggested HF roll off, it is only the extent that varies.

I don't think the aim is to hype, as I said my aim would the opposite.
In other words, place a veil over the sound, so that the mix has to be very well done to make it's way through. That's the essence of translation.

Plse note my sticking to the original question is in no way intended to contradict or diminish other input here. The 'Target' response I speak of would clearly be much better achieved using Room Treatment and Crossover/Level adjustments.

DD
Old 20th August 2010
  #29
Lives for gear
 
Weasel9992's Avatar
 

Fantastic post, Jay. thumbsupthumbsupthumbsup
Old 21st August 2010
  #30
Gear addict
 
parkay909's Avatar
 

Thread Starter
Main EQ Curve questions

Thanks again everyone!
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
ruy / Post Production forum!
11
Jesse Peterson / Post Production forum!
2
Yiannis / So much gear, so little time!
0
stevens119 / Low End Theory
11

Forum Jump
 
Register FAQ Search Today's Posts Mark Forums Read

SEO by vBSEO ©2011, Crawlability, Inc.