JBL, Bob Hodas re: EQ of Monitors

[Gary Flanigan]

I've seen many discussions here regarding the folly of trying to
make up for room deficiencies by eq'ing one's monitors, and yet
a new series of speakers from JBL have exactly this capability
built in, and in a current Mix article Mr. Hodas concludes by
recommending this practice (after all room treatments have been
applied).

Have I misunderstood, are they clueless, or is this an area where
honest folks may disagree?

Thanks

[Phil Brown]

>I've seen many discussions here regarding the folly of trying to
>make up for room deficiencies by eq'ing one's monitors, and yet
>a new series of speakers from JBL have exactly this capability
>built in, and in a current Mix article Mr. Hodas concludes by
>recommending this practice (after all room treatments have been
>applied).
>
>Have I misunderstood, are they clueless, or is this an area where
>honest folks may disagree?

While I cannot speak for Mr. Hodas the Mix article has to do with existing
rooms that can't be changed much if at all. Obviously getting the room right by
design is paramount. I was involved in the construction of two mastering rooms
with Bob and we shot the room as we went along the construction process with
false walls to determine the best placement. Not everyone has that luxury. We
also were very carefull with the reflectence/absorbance pattern on the walls to
get the desired result at the monitoring position. The results were two
absolutely terrific sounding rooms that required no EQ and only a little bass
trapping in one room, surely a great result. The rooms are at Paul Stubblebine
Mastering in San Francisco and can be seen at www.paulstubblebine.com. I've
been an engineer for more than 30 years and the are the best sounding mastering
rooms I've ever been in. Bob's a genius as far as I'm concerned and anyone
would do well to use him. Thus endeth the commericial.
Phil Brown

[hank alrich]

Gary Flanigan wrote:

> I've seen many discussions here regarding the folly of trying to
> make up for room deficiencies by eq'ing one's monitors, and yet
> a new series of speakers from JBL have exactly this capability
> built in, and in a current Mix article Mr. Hodas concludes by
> recommending this practice (after all room treatments have been
> applied).

> Have I misunderstood, are they clueless, or is this an area where
> honest folks may disagree?

Folks may disagree, nor not, but here is the key to what you're reading:

"after all room treatments have been applied".

Once you do that you're not trying to fix a broken room; issues are far
less severe and a room can be "seasoned to taste" with EQ.

--
ha

[Phil Brown]

>Once you do that you're not trying to fix a broken room; issues are far
>less severe and a room can be "seasoned to taste" with EQ.

Sorry, but I couldn't disagree with "seasoned to taste" more. We're talking
about monitoring here, not producing a pleasent listening experience.
Monitoring should aspire to show what is actually going on with the program
material. This means warts and all. "Season to tast" is dangerous. What if your
taste runs to, oh, I don't know, lots of bass. You EQ the bass up, sounds good
to your taste, client takes it home, very bass light. Dangerous.
Phil Brown

[hank alrich]

Phil Brown wrote:

> >Once you do that you're not trying to fix a broken room; issues are far
> >less severe and a room can be "seasoned to taste" with EQ.

> Sorry, but I couldn't disagree with "seasoned to taste" more. We're talking
> about monitoring here, not producing a pleasent listening experience.

We're also talking about imperfect systems. One person's accurate is
another's painful and adjustments will be made to accomodate reality. I
postulate that the big troubles should be ameliorated by room
treatments, and the little troubles and overall spectral balance
addressed with EQ.

> Monitoring should aspire to show what is actually going on with the program
> material.

Since there is no such thing as a perfect monitoring environment, and
since individual hearing appreciation varies from one human to the next,
I have always found that adjustments made with EQ to monitoring rigs
reflect the preferences of whomever does the adjusting. Get Hodas into
it and you have a precise target, but one that is still amenable to user
preference.

When we have perfect output transducers we might be able to move beyond
this. But even given the comparative perfection of say, mic pres, people
have their likes and dislikes and invoke those while working.

> This means warts and all.

Of course it means warts and all.

> "Season to tast" is dangerous. What if your
> taste runs to, oh, I don't know, lots of bass. You EQ the bass up, sounds good
> to your taste, client takes it home, very bass light. Dangerous.

One who is either ignorant enough or inexperienced enough to do that is
not one I am considering in this discussion. Such an individual should
stay away from monitoring system EQ until they can understand their own
hearing well enough to deliver substantially translatable mixes in a
variety of settings.

After one has experience at this one calibrates one's own work in a
given environment against known good work, and right there, whatever
one's preferences are as manifested in EQing a system, one has a
reference against which one can compare mixes.

But you know all this, and I know you know it. <g>

--
ha

[Bob Olhsson]

"hank alrich" > wrote in message
...
> After one has experience at this one calibrates one's own work in a
> given environment against known good work, and right there, whatever
> one's preferences are as manifested in EQing a system, one has a
> reference against which one can compare mixes.

This is and isn't true. Great sounding rooms can send one's work into far
left field even when known good work sounds wonderful. It's also important
that clients can offer input based on what they hear without having to
listen to "trust me" from the engineer.

I think the place to begin is audio that both measures and sounds good.
There's plenty that meets one of those criteria but not much, in my
experience, that meets BOTH . EQ. is perfectly fine when it's correcting
colorations in the speaker. It's a problem when it colors the speaker in an
attempt to overcome a coloration in the room.

--
Bob Olhsson Audio Mastery, Nashville TN
Mastering, Audio for Picture, Mix Evaluation and Quality Control
Over 40 years making people sound better than they ever imagined!
615.385.8051 http://www.hyperback.com

[Scott Dorsey]

Gary Flanigan > wrote:
>I've seen many discussions here regarding the folly of trying to
>make up for room deficiencies by eq'ing one's monitors, and yet
>a new series of speakers from JBL have exactly this capability
>built in, and in a current Mix article Mr. Hodas concludes by
>recommending this practice (after all room treatments have been
>applied).
>
>Have I misunderstood, are they clueless, or is this an area where
>honest folks may disagree?

The JBL speakers you refer to are speakers that have some low end
EQ built into them to compensate for different boundary placement.
This _does_ work, and it can effectively compensate for different
distances between the speaker and the wall. Mackie and NHT do
similar things.

All this does is mean that you don't need to position the speakers
as precisely as you otherwise would with respect to the rear wall,
and it's a useful thing because it gives you some flexibility in
placing.

EQ can fix the things that EQ can fix, and it can't fix the things
that EQ can't fix. There are some things that can be helped with
changing EQ or speaker voicing. Room modes aren't those things.
Third-octave EQ is pretty much always a bad idea, but EQ itself is
not inherently bad. It's just one tool of many, and it's one that
people seem obsessed with using to do things that it's not good at.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arny Krueger]

"Phil Brown" > wrote in message


>> Once you do that you're not trying to fix a broken room; issues are
>> far less severe and a room can be "seasoned to taste" with EQ.

> Sorry, but I couldn't disagree with "seasoned to taste" more. We're
> talking about monitoring here, not producing a pleasant listening
> experience. Monitoring should aspire to show what is actually going
> on with the program material. This means warts and all.

So far so good.

>"Season to taste" is dangerous.

Methinks you vastly overestimate the levels of accuracy possible with
loudspeakers or any electroacoustic transducer.

Every known transducer, whether headphones, earphones or loudspeakers, adds
some kind of *taste* of its own to the music you play through it.

So, the issue is not uncolored versus colored. Rather its all about colored
this way versus colored that way.

Furthermore, every listening environment adds considerable color of its own.
Therefore, intelligently applied *seasoning* has an excellent chance of
neutralizing some undesired undesirable added flavoring from the
transducer/environment interface.

> What if your taste runs to, oh, I don't know, lots of bass.

Even if one likes lots of bass, there is still a concept of too much bass
or too little bass.

Let's say that my monitor/room combination provides exactly what I consider
to be tasteful amount of bass with a neutral recording. Then, whenever I use
these monitors in this room I'm going to be happy, if the recording has the
exact right amount of bass. I'll add or remove bass from recordings that
have too little or too much base. Net result, a steady stream of recordings
with the right amount of bass.

Let's say that my monitor/room combination provides less bass than what I
consider to be tasteful amount, with a neutral recording. Then, whenever I
use these monitors in this room I'm going to want to add bass to suit my
tastes. Net result, a steady stream of recordings with too much bass.

Let's say that my monitor/room combination provides more bass than what I
consider to be tasteful amount, with a neutral recording. Then, whenever I
use these monitors in this room I'm going to want to remove bass to suit my
tastes. Net result, a steady stream of recordings with too little bass.

>You EQ the bass up, sounds good to your taste, client takes it home, very
bass light.

Not a problem if the room has the amount of bass I like with a neutral
recording.

>Dangerous.

Smart, for the reasons given above.

What's dangerous is a monitoring system with dead spots or blind spots. The
classic example is a monitoring system that has no deep bass. How can one
reasonably tune something by ear, if you can't hear whether its there or
not? A person who likes a lot of deep bass is going to more intolerant of a
monitoring system that lacks deep bass than someone who really doesn't care
for a lot of deep bass.

[hank alrich]

Bob Olhsson wrote:

> "hank alrich" wrote...

> > After one has experience at this one calibrates one's own work in a
> > given environment against known good work, and right there, whatever
> > one's preferences are as manifested in EQing a system, one has a
> > reference against which one can compare mixes.

> This is and isn't true. Great sounding rooms can send one's work into far
> left field even when known good work sounds wonderful. It's also important
> that clients can offer input based on what they hear without having to
> listen to "trust me" from the engineer.

****, I never tell a client that! If they don't trust me already they
shouldn't be here. <g>

> I think the place to begin is audio that both measures and sounds good.
> There's plenty that meets one of those criteria but not much, in my
> experience, that meets BOTH . EQ. is perfectly fine when it's correcting
> colorations in the speaker. It's a problem when it colors the speaker in an
> attempt to overcome a coloration in the room.

Well stated.

--
ha

[Ethan Winer]

Gary,

> are they clueless, or is this an area where honest folks may disagree? <

Probably a little of both!

A lot of great points were made so far, and I'll add a few of my own.

All untreated rooms have peaks and deep nulls. Adding bass traps and other
treatment helps a lot and must be the first line of attack, but some amount
of peaks and nulls will still remain. Even a $1 million dollar control room
has them. EQ can sometimes tame the peaks a little, but it can't solve deep
nulls because that requires too much boost for most amps and speakers to
handle without distorting. Nulls as deep as 35 dB are not only common, but
typical.

There are two kinds of peaks and nulls - modal and non-modal. The
significant difference is that modal peaks and nulls stay at the same
frequency, and non-modal peaks and nulls do not. If the frequency of a peak
or null changes as you move six inches forward or back in your chair, you
cannot solve that with EQ. Any EQ you apply to fix one place will make
things worse somewhere else nearby.

In one room I tested, the "crossover" point from modal to non-modal was
around 150-160 Hz. Below that frequency the peaks and nulls were related to
the room dimensions and did not change around the room. However the
*magnitude* of the peaks and nulls changed a lot, so no one amount of boost
or cut you apply would be correct for more than one tiny area.

Above 150-160 Hz the peak and null frequencies were related mainly to the
listener's distance to the rear wall. EQ cannot solve this at all because
it's a moving target.

It makes sense that peaks and nulls transition from modal to non-modal as
you go higher in frequency because a room is basically three simple bandpass
filters in parallel. As you go farther from the center tuned frequency,
there's less resonance, and by the third or fourth "mode harmonic" simple
acoustic interference dominates. So it's not that the crossover from modal
to non-modal is always around 150 Hz, but rather that it depends on the room
dimensions.

I wrote this up recently, including graphs of LF response versus placement
in the room:

www.realtraps.com/art_modes.htm

--Ethan

[Roger W. Norman]

"Gary Flanigan" > wrote in message
om...
> I've seen many discussions here regarding the folly of trying to
> make up for room deficiencies by eq'ing one's monitors, and yet
> a new series of speakers from JBL have exactly this capability
> built in, and in a current Mix article Mr. Hodas concludes by
> recommending this practice (after all room treatments have been
> applied).

Garner what you can from the information, apply your experience, and come up
with your own conclusions about Bob's thoughts. As I see it, you have two
possibilities, one being fixing the room to require as little as possible in
terms of EQ, or simply EQing the speakers output to somehow match the
deficiencies in the room. Some might say that both exercises are equal, but
I'd think that the standard method of making the room right is preferable to
making the speakers take up the slack.

But in today's modelling world, one has to question whether either one is a
viable solution since one could ostensibly digitally recreate any room.
That's if you believe in that sort of thing. How about simply being able to
take a recording made in a specific room and working with it as it is rather
than taking your (not you specifically) 12' X 8' bedroom and making it Abbey
Road?

It's not the application of technology that makes something worth having
recorded. It's having something worth having been recorded in the first
place, and maybe, just maybe, it's the ability to distinguish between the
two.

And maybe I'm full of ****. That possibility always exists.

--


Roger W. Norman
SirMusic Studio

>
> Have I misunderstood, are they clueless, or is this an area where
> honest folks may disagree?
>
> Thanks

[Roger W. Norman]

Certainly Hank knows that, Phil. Not to try to blow smoke up your butt, but
look Hank up and you'll find an entirely different person than deigns to
answer here in that he's so innocuous. In particular, look him up in terms
of Austin, Texas music. He's a singular individual and a tremendous
resource, just as you are.

--


Roger W. Norman
SirMusic Studio

"Phil Brown" > wrote in message
...
> >Once you do that you're not trying to fix a broken room; issues are far
> >less severe and a room can be "seasoned to taste" with EQ.
>
> Sorry, but I couldn't disagree with "seasoned to taste" more. We're
talking
> about monitoring here, not producing a pleasent listening experience.
> Monitoring should aspire to show what is actually going on with the
program
> material. This means warts and all. "Season to tast" is dangerous. What if
your
> taste runs to, oh, I don't know, lots of bass. You EQ the bass up, sounds
good
> to your taste, client takes it home, very bass light. Dangerous.
> Phil Brown

[ScotFraser]

<< or simply EQing the speakers output to somehow match the
deficiencies in the room. >>

This is not possible, as the room deficiencies create time problems, i.e. a
resonance will ring well past the point of the initial excitation. EQ doesn't
work on that.

<< Some might say that both exercises are equal,>>

That would be an incorrect statement by the some who say such.

<< but
I'd think that the standard method of making the room right is preferable to
making the speakers take up the slack.>>

EQ will help frequency response errors in the speakers themselves, but can't
begin to deal with the time-based effects of room acoustics.

Scott Fraser

[Phil Brown]

>EQ will help frequency response errors in the speakers themselves, but can't
>begin to deal with the time-based effects of room acoustics.

Which become more and more important as we get into more surround material.
Phil Brown

[Chris Hornbeck]

On Sun, 18 Jul 2004 14:39:50 -0400, "Roger W. Norman"
> wrote:

> As I see it, you have two
>possibilities, one being fixing the room to require as little as possible in
>terms of EQ, or simply EQing the speakers output to somehow match the
>deficiencies in the room. Some might say that both exercises are equal, but
>I'd think that the standard method of making the room right is preferable to
>making the speakers take up the slack.

Yeah, the size of room bandaid available by speaker EQ is limited.
One important issue is that the speaker's direct sound always arrives
*first* and our hearing gives that a priority.

Room sound, arriving later, is heard comparatively more forgivingly.
Electronic EQ can only affect both identically, so ya gotta choose.

Chris Hornbeck

[Kurt Albershardt]

Chris Hornbeck wrote:
>
> One important issue is that the speaker's direct sound always arrives
> *first* and our hearing gives that a priority.
>
> Room sound, arriving later, is heard comparatively more forgivingly.

And: Making sure that the reflected sound arrives *sufficiently later* is one of the primary goals of room design. Early reflections confuse the ear/brain system in its quest to sort out direct from reflected sounds.

[Bob Cain]

ScotFraser wrote:

> << or simply EQing the speakers output to somehow match the
> deficiencies in the room. >>
>
> This is not possible, as the room deficiencies create time problems, i.e. a
> resonance will ring well past the point of the initial excitation. EQ doesn't
> work on that.

Depends on the nature of the equalizer. An FIR equalizer
attacks both time and frequency domain problems equally.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

[Chris Hornbeck]

On Sun, 18 Jul 2004 23:27:32 -0700, Bob Cain
> wrote:

>ScotFraser wrote:
>> This is not possible, as the room deficiencies create time problems, i.e. a
>> resonance will ring well past the point of the initial excitation. EQ doesn't
>> work on that.
>
>Depends on the nature of the equalizer. An FIR equalizer
>attacks both time and frequency domain problems equally.

You've intimated here that you're working on a no-room black box.
Any titillating details?

Inquiring minds,

Chris Hornbeck

[Ethan Winer]

Roger,

> 12' X 8' bedroom and making it Abbey Road? <

Not possible. Not even with really good acoustic treatment! :->)

Read through some of the articles on my personal and company's sites. A 12x8
bedroom has 20-35 dB nulls all over the place. What amp/loudspeaker can
handle 35 dB boost at 80 Hz and not blow up or at least distort? More to the
point, that 80 null becomes 90 Hz when you lean back in your chair. I'm not
aware of any EQ that senses where your ears are and adjusts the curve to
suit.

--Ethan

[Ethan Winer]

Bob,

> An FIR equalizer attacks both time and frequency domain problems equally.
<

It is possible to design a DSP algorithm that can reduce ringing. The
problem is it works only in an ivory tower. The correction needed to counter
time-based effects caused by a room is so *incredibly specific* that it
works only for an area about one cubic inch. You can't even get both of your
ears into the same "corrected zone."

--Ethan

[Bob Cain]

Chris Hornbeck wrote:

> On Sun, 18 Jul 2004 23:27:32 -0700, Bob Cain
> > wrote:
>
>
>>ScotFraser wrote:
>>
>>>This is not possible, as the room deficiencies create time problems, i.e. a
>>>resonance will ring well past the point of the initial excitation. EQ doesn't
>>>work on that.
>>
>>Depends on the nature of the equalizer. An FIR equalizer
>>attacks both time and frequency domain problems equally.
>
>
> You've intimated here that you're working on a no-room black box.
> Any titillating details?
>
> Inquiring minds,
>

Nope. I do have some ideas, though. The big problem with
rooms is that they are so multi-modal. You can find a near
perfect solution for any one point in the room but you don't
have to move far to find a point where that solution is
utterly contrary. Speakers have the potential of having a
solution that varies too much with dynamic conditions of
temperature, humidity and air pressure (not really sure
about the last one.)

I do have a solution for a multi-measurement system which
will find the optimum equalziation FIR in a least mean
squares sense over a set of measurements but have yet to
apply it to rooms or speakers, just mics at this point,
although its application to surround feed encoding is in the
offing. Not sure what it will do for a room but the real
solution is first to damp the room properly regardless. The
wilder the room is, of course the less likely any kind of
equalization will apply over many points in it and the more
likely that any perturbation like a human body or the
position of a chair will throw off the correction signifigantly.

I was just trying to say that the distinction between time
and frequency domain correction only exists if you are using
highly constrained types of equalizers, such as minimum
phase or linear phase ones. If you take an approach with
the degrees of freedom of a finite impulse response
equalizer the distinction vanishes.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein