Is there research on room correction EQ as it relates to listener preference?

[JohnBooty]

The following two things are well-researched and regarded as fact.

I don't think there's much debate about these two items... if there is, this isn't the thread for it.

1. From Sean Olive's research at Harman, we know there is a strong positive correlation between accurate sound reproduction (tonal correctness for both direct and reflected sound) and listener preference.
2. Rooms will always introduce their own peculiarities due to room modes, reflections, absorptions, etc. As far as measurements are concerned, room EQ is a must for an accurate response at the listening position.

Here's my question: has there been any research to correlate listener preferences and room correction EQ?

I think we would all agree that large nulls and peaks due to room modes are never desirable and should be dealt with.

However, beyond that, I'm not sure that completely "removing the room from the equation" should always be the ultimate goal. This can creates an unnatural, headphones-like effect. After all, our brains always perform a lot of "room correction", which is why we e.g. recognize a familiar sound or voice no matter the acoustic environment.

I'm not sure where I stand on this issue myself. I'm curious if research has been done on the matter.

 

[pozz]

Do you mind restating your question? I don't completely follow you.

There's plenty that EQ can't do when it comes to room acoustics (reverb times, reflection direction, delay and strength). Toole wrote a lot about the ineffectiveness of the older third-octave measurement and EQ method, and what new technology is capable of, for example. His final recommendation was to use corrective EQ only up to the transition region so that effectively, you are EQing the room response, not the speaker response. The assumed goal was to achieve smooth low-end across a wide area in your room. Non-corrective EQ-to-taste out of scope and considered a form of tone control.

Are you just looking for studies about the effectiveness of different types of room EQ?

 

[JohnBooty]

Do you mind restating your question? I don't completely follow you.

Yeah, I struggled with how to word it... and it could use refinement to put it mildly. Let me retry.

(clears throat)

We know accurate speaker FR and dispersion is generally preferred by listeners, thanks to research at Harman.

But has similar work been done to correlate room correction with listener preferences?

 

[pozz]

We know accurate speaker FR and dispersion is generally preferred by listeners, thanks to research at Harman.

But has similar work been done to correlate room correction with listener preferences?

This is where I have trouble.

Room correction messes with FR directly. So the preference studies relating to FR (and adjoining factors like phase) are the most relevant, unless you mean a comparison of different room correction systems/approaches.

 

[Wes]

I suspect there are several questions rolled up inside the broader one.

1. What is the variation in listener preferences among listeners?

2. What is the variation in listener preference for a single listener based on type of music?

and...

How do listener preferences vary based on age, gender, noise exposure, cultural and developmental histories?

 

[DonH56]

Yeah, I struggled with how to word it... and it could use refinement to put it mildly. Let me retry.

(clears throat)

We know accurate speaker FR and dispersion is generally preferred by listeners, thanks to research at Harman.

But has similar work been done to correlate room correction with listener preferences?

I think all that would show is how close it comes to meeting its target curve, and perhaps how well it handles time alignment (which ends up being the same thing, of course). That is, how closely the correction implements the desired response in the real world.

Anecdotally there have been posts around various audio fora discussing how they prefer one or another room correction program, but my fly-by view is it comes down to the user interface and target curve implemented. One thing I liked about MCACC and Dirac Live over YPAO and Audyssey at the time (few years ago) was the former allowed the user to customize the target curve. I think the latest Audyssey versions have added that capability, have not followed YPAO, and don't know about ARC (Anthem Room Correction, I still think Audio Research Corporation) or other programs.

Note RC can help with accurate FR but dispersion is harder to tackle; that is more about the speakers and the room.

IMO - Don

 

[sigbergaudio]

Properly applied EQ below the Schroeder frequency is so obviously better sounding that I'm not sure one would even need a study?

Full range EQ is different matter.

 

[andreasmaaan]

Olive et al did a small, somewhat limited study comparing various room correction products, which involved only one room, one speaker system, and one sample of listeners.

He wrote about it here.

The full paper is available here (AES membership required I believe).

 

[JohnBooty]

This is where I have trouble.
Room correction messes with FR directly.

This is of course true, but I suppose that's where things probably get into psychoacoustics.

Our brains seem to be constantly performing their own "room correction" by trying to ignore the acoustical effects of our surroundings to some extent. This is why we can easily recognize a loved one's voice, the growl of a dangerous predator animal, or some other familiar sound no matter what our acoustical surroundings are.

I can recognize the jingle of my car keys whether I'm in a recording studio, a cave, or an open plain. Even though those three situations would measure very differently.

So I was curious whether or not room correction above the Schroeder frequency -- even if it resulted in a more measurably correct frequency response at the listening position -- might actually not even be preferable. If removing the room from the sound too thoroughly might actually defeat the illusion of realism, etc. After all, if removing the room entirely was really our goal, we'd skip speakers altogether and listen exclusively on closed-back headphones, right?

It turns out that Toole did in fact cover this to some extent, and only recommends room correction below the Schroeder frequency. ‍ I'm an idiot. Though, I'm not clear if he based this on blind trials with a large sample of listeners.

I should just go suck it up and pay my $59 for the book.

Sorry to pose a question already answered in the book.... but I do truly appreciate the thoughtful replies that have now pointed me in the right direction. Thanks, guys.

 

[Pio2001]

Properly applied EQ below the Schroeder frequency is so obviously better sounding that I'm not sure one would even need a study?

That is not widely accepted. Many professional who design and built home theater (but not all) say that equalizing room modes must be avoided, and used only in the most desperate cases.

Even in high fidelity, I (and Floyd Toole) prefer a complete and accurate equalization of room modes, while other people say that corrections must be soft and broad, and that one should not try to correct completely the frequency response of room modes.

It turns out that Toole did in fact cover this to some extent, and only recommends room correction below the Schroeder frequency. ‍ I'm an idiot. Though, I'm not clear if he based this on blind trials with a large sample of listeners.

Floyd Toole says that if a speaker has the same frequency response variations in all directions of space (the two directivity index bottom curves in spinoramas are straight lines), then it is possible to equalize the speaker so that its frequency response is flat.
From a room eq point of view, it looks like a room correction was made, but towards an optimal, personalized target curve, that is the one that fits that speakers in this room, from this listening position.
However, if the speakers' frequency response behave differently according to the directions (the two directivity index are not straight), then equalization is hopeless. It will never sound right.

 

[JoachimStrobel]

I think you need to separate the 2ch vs Mch RoomEQ. For Mch, RoomEQ should and is enrolled into distance, time and volume equalization, and then comes the Room curve which convolves everything again and almost makes it a must. In particular as a Mch system for sure will not want to deal with the room and neither do you want that.
For 2 channel, where speakers are often placed in a good triangle, the room is your synthetic Mch system, giving you the illusion of space and what else. This is the same argument as saying certain music sounds better through a Bluetooth boom box than through a good Stereosystem.
(I think Toole writes somewhere about his MCh home setup with simulated concert hall ambience - sure, for that you certainly need to apply RoomEQ first to start from scratch)

 

[Mnyb]

Hmm room EQ above shröder , my amateur thoughts .

Maybe above the transition one should eq the speaker not the room ? As our brains does it own thing.

Some few speakers with modern directivity features in their designs is much more eq friendly than older designs ?
Here I think the largest fail is , with some old style speakers correction of the reflected sounds disturbs the direct sound to much no acceptable compromise can be found. But with some speakers more room EQ can be applied before any adverse effects sets in.

Transition frequency ( shröder ) is not an absolute , mild room EQ slightly above may work too ?

 

[thewas]

Olive et al did a small, somewhat limited study comparing various room correction products, which involved only one room, one speaker system, and one sample of listeners.

He wrote about it here.

The full paper is available here (AES membership required I believe).

Here are the presentation slides which are linked also in his blog
https://drive.google.com/file/d/0B9...EyNC00ZDcyLWEzZTAtMGJiODQ1ZTUxMGQ4/view?hl=en

 

[sigbergaudio]

That is not widely accepted. Many professional who design and built home theater (but not all) say that equalizing room modes must be avoided, and used only in the most desperate cases.

Even in high fidelity, I (and Floyd Toole) prefer a complete and accurate equalization of room modes, while other people say that corrections must be soft and broad, and that one should not try to correct completely the frequency response of room modes.

You are having two discussions here, one is correction vs no correction, the second is type of correction.

I would argue that anyone saying "no correction is better than room correction (broad or accurate) below the Schroeder frequency" in a normal home / living room situation has been living under a rock for a while. Do you have any references to these professionals you speak of?

 

[sfdoddsy]

As mentioned above, there have studies about preference for different Room EQ products.

But I suspect the answer is a bit simpler.

If you agree with Point 1 on an objective basis, and if Room EQ gets you closer to Point 1, then by definition Room EQ is to be preferred.

And the particular Room EQ that gets you closest (objectively) is thus the preferred one.

Thus, since spinoramas are the accepted objective standard, we should do spinoramas of Room EQs.

 

[ernestcarl]

You are having two discussions here, one is correction vs no correction, the second is type of correction.

I would argue that anyone saying "no correction is better than room correction (broad or accurate) below the Schroeder frequency" in a normal home / living room situation has been living under a rock for a while. Do you have any references to these professionals you speak of?

There is one particular acoustician with a youtube video (he looked like he had at least a hundred or more in his channel) posted in a thread here somewhere a while back where he says EQ is just the worst thing you can do and the only way to treat a room properly is with his very expensive acoustic products. I know his face (long-haired dude) but can’t for the life of me remember his name or company.

 

[ernestcarl]

I found him:

He seems to have found business success, too.

 

[sigbergaudio]

Doesn't appear to be a 100% objective source.

 

[Pio2001]

I would argue that anyone saying "no correction is better than room correction (broad or accurate) below the Schroeder frequency" in a normal home / living room situation has been living under a rock for a while. Do you have any references to these professionals you speak of?

Maybe I overinterpreted some statements by professionals.

The first source was a company that I can't remember, that was saying that they first do everything that they can with acoustic correction, and hopefully it is enough, but in some cases, electronic correction must be added on top.

The other one is Roland Delacroix aka thxrd. His job is to design and build rooms. His point of view
is that equalization is ineffective at correcting problems : https://www.homecinema-fr.com/forum/post180033096.html#p180033096

 

[sigbergaudio]

Maybe I overinterpreted some statements by professionals.

The first source was a company that I can't remember, that was saying that they first do everything that they can with acoustic correction, and hopefully it is enough, but in some cases, electronic correction must be added on top.

The other one is Roland Delacroix aka thxrd. His job is to design and build rooms. His point of view
is where equalization is ineffective at correcting problems : https://www.homecinema-fr.com/forum/post180033096.html#p180033096

We may need to determine what use case or situation we are discussing.

If you are to build a money no object studio, I would agree that obviously you would treat the room as much as possible, and potentially to the extent that limited correction is needed.

For a normal living room, that simply is not feasible unless you have no aesthetic sense and no partner. There are a number of visually acceptable measures that can be taken, and I would recommend everyone to do it to the extent possible for the mid/highs. But below 100hz it's pretty difficult without massive bass traps.

That equalization is ineffective can be objectively disproved by measurements, so I'm not sure what to do with that statement. I tried translating the french page and it seems to talk about the problem of only being able to correct for a small listening space, and that is a valid argument. This can be helped by adding one or multiple subwoofers, which will even out the response.