• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Does anyone else prefer a dipped midrange?

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,421
Likes
2,406
Location
Sweden
Again, Toole says there is a flaw with stereo. A ”flat”speaker preferred in mono will sound different in stereo. The solution according to Toole is to go multichannel. For stereo you are left with uncertanties in the 1-5 kHz region. There is nothing in Tooles book that contradicts this fact.
 

McFly

Addicted to Fun and Learning
Joined
Mar 12, 2019
Messages
903
Likes
1,861
Location
NZ
Also the most common two way speaker has a dip in the sound power around 1-3khz, due to the vertical ERDI cancellations, which could be why the 5 or 6+1 inch bookshelf speaker is such a mainstream combo. The BBC dip or midrange dip you speak of
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
Again. There is no foundation for your claim that a majority will prefer a mid dip. Therefore there are no studies which show this and most certainly not Toole.

I have been sensing a pattern of strong dunning Kruger effect from your posts lately, you might wanna bring the hostility down a notch and the friendliness up a notch.

Here is some foundation for orchestral pieces requiring a dip around 3KHz to sound more natural.


Typically these things should be sorted out in the production stage, but you can't really count on that sometime.
 
OP
D

Digby

Major Contributor
Joined
Mar 12, 2021
Messages
1,632
Likes
1,555
Toole has several times said that stereo is flawed. In the stereo triangle you will experience a dip around 2 kHz followed by a peak 3-4 kHz when using a speaker with flat frequency response. This effect is exaggerated in nearfield (and anechoic) conditions. The dip at 2 kHz can somewhat be filled with reflections from the room and the peak at 3-4 kHz can be damped with 1-2 dB dip.
Apologies, I missed this post the first time. This is the kind of useful information I am looking for. Thank you for the comment.

So, given what you said, a completely flat speaker is not ideal (for stereo) unless this dip & peak is accounted for. Would you say the common statement that we want a anechoically flat speaker (for stereo) is incorrect, or at least incomplete, if it isn't followed by stating the 3-4khz peak has to be accounted for. Another route is to find a speaker with this kind of dip built in, presuming it otherwise performs as it should.

I am listening in the nearfield, so this information coupled with this:


...may go most of the way to explaining what I'm experiencing. With classical recordings, it seems this hump may be exacerbated by another 2db or so (Linkwitz suggests a 4db dip).
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
Apologies, I missed this post the first time. This is the kind of useful information I am looking for. Thank you for the comment.

So, given what you said, a completely flat speaker is not ideal (for stereo) unless this dip & peak is accounted for. Would you say the common statement that we want a anechoically flat speaker (for stereo) is incorrect, or at least incomplete, if it isn't followed by stating the 3-4khz peak has to be accounted for. Another route is to find a speaker with this kind of dip built in, presuming it otherwise performs as it should.

I am listening in the nearfield, so this information coupled with this:



...may go most of the way to explaining what I'm experiencing. With classical recordings, it seems this hump may be exacerbated by another 2db or so (Linkwitz suggests a 4db dip).

It's just that these issues should be sorted out in the production stage, you can't expect the end user to play DJ in their free time.

the speaker could also be used in a multi-channel setup where this phenomenon would not exist. It's just much more logical to design a flat on-axis speaker and place some of the burden of good reproduction on good production.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,421
Likes
2,406
Location
Sweden
Apologies, I missed this post the first time. This is the kind of useful information I am looking for. Thank you for the comment.

So, given what you said, a completely flat speaker is not ideal (for stereo) unless this dip & peak is accounted for. Would you say the common statement that we want a anechoically flat speaker (for stereo) is incorrect, or at least incomplete, if the 3-4khz peak isn't accounted for.

I am listening in the nearfield, so this information coupled with this:



...may go most of the way to explaining what I'm experiencing. With classical recordings, it seems this hump may be exacerbated by another 2db or so (Linkwitz suggests a 4db dip).

There are several threads about this and referenced to Shirley et al, in Toole's book. I myself prefer that the on and off-axis response around 2 kHz should never have a dip, while the 3-4 kHz region should have 1-2 dB less. Voices tend to sound more natural that way, especially female voices, tenors etc. This should not be confused with "BBC dips" or construction errors due to bad pairing of woofer to tweeter crossovers. There is no research made that I am aware of that have tested such small variations in frequency response. A speaker with +/- 1.5 dB frequency response can be regarded as "flat", but within that range, you can have speaker sounding quite different depending on Q of dips and peaks and which frequencies they appear.
 

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
I have been sensing a pattern of strong dunning Kruger effect from your posts lately, you might wanna bring the hostility down a notch and the friendliness up a notch.

Here is some foundation for orchestral pieces requiring a dip around 3KHz to sound more natural.


Typically these things should be sorted out in the production stage, but you can't really count on that sometime.
Wow ... where did that came from? Please dont resort to ad-hominem attacks, if you disagree with me. If you disagree or think I am wrong, than please educate me and I am more than willing to listen and learn. If you feel indeed, I was out of line or not in compliance with the forum rules, please either PM me, so we dont derail this thread or report me to the moderators, so that someone independent can rule on it.


Back to the question / claim from the OP: " ... 9 out of 10 people prefer a mid dip ...": I still dont see any evidence, especially in Toole (except what Thomas_A quoted above), that a majority (9 out of 10) prefers a mid range dip. Even if, I am of the opinion, that the OP should provide evidence to his claim and not put the burden of proof to the forum members (incl. me and chide me for not providing evidence to his unproven claims).

The article you linked, refers to the "famous BBC dip" and as Thomas_A said is not to be confused with the preference studies by Toole, which find a preference to flat anechoic response. The BBC dip is not based on preference studies and therefore presents no evidence that any majority of listeners prefers a mid dip, but it comes historically from the non-flat FR baffle step response of "old" speakers formerly used by the BBC. Nowadays, we (hardly) dont see these flaws in speakers anymore (see Klippel data) so there is not necessity anymore to correct for the BBC dip. (Sidenote: The Denon Audyssey AVRs still correct for the BBC dip in their default "Reference" EQ settings, but it is widely recommended to switch this off due to modern speakers not needing this correction - see the various threads and the interviews with Audyessey Head of Development). Again, if you disagree, please educate and not attack me in person.

Yes, I agree with you that any recording flaws should be addressed at the recording stage.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
Wow ... where did that came from? Please dont resort to ad-hominem attacks, if you disagree with me. If you disagree or think I am wrong, than please educate me and I am more than willing to listen and learn. If you feel indeed, I was out of line or not in compliance with the forum rules, please either PM me, so we dont derail this thread or report me to the moderators, so that someone independent can rule on it.


Back to the question / claim from the OP: " ... 9 out of 10 people prefer a mid dip ...": I still dont see any evidence, especially in Toole (except what Thomas_A quoted above), that a majority (9 out of 10) prefers a mid range dip. Even if, I am of the opinion, that the OP should provide evidence to his claim and not put the burden of proof to the forum members (incl. me and chide me for not providing evidence to his unproven claims).

The article you linked, refers to the "famous BBC dip" and as Thomas_A said is not to be confused with the preference studies by Toole, which find a preference to flat anechoic response. The BBC dip is not based on preference studies and therefore presents no evidence that any majority of listeners prefers a mid dip, but it comes historically from the non-flat FR baffle step response of "old" speakers formerly used by the BBC. Nowadays, we (hardly) dont see these flaws in speakers anymore (see Klippel data) so there is not necessity anymore to correct for the BBC dip. (Sidenote: The Denon Audyssey AVRs still correct for the BBC dip in their default "Reference" EQ settings, but it is widely recommended to switch this off due to modern speakers not needing this correction - see the various threads and the interviews with Audyessey Head of Development). Again, if you disagree, please educate and not attack me in person.

Yes, I agree with you that any recording flaws should be addressed at the recording stage.

The article i shared has nothing about the BBC dip, it’s talking about a completely different topic and involves a different frequency range.

I’m just saying there is a certain argumentative tone to all your comments that doesn’t invite anyone to participate with you in any discussion.

None of us are experts, we’re all learning.
 

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
It is quite possible. What is the music you mostly listen to? Probably a full 80% of music I listen to is unamplified sonatas, concertos and such. Probably 2% of all music has vocals in it. If the music you listen to is mostly Rock, say, with 98% being vocal music, then it is quite possible this provides a different perspective entirely.


This isn't a particularly useful reply, you are bypassing the interesting things I say (with regards to a convivial discussion) and focussing on less interesting things. I said this was speculation on my part, I can only repeat it so many times. Can you provide this information as an excerpt from Toole, then we will have more to discuss:

"Size of room, level of soundproofing, distance from loudspeakers and so on. This may explain where some of the difference lies."

I would imagine types of music used would be of not insignificant importance too.


Yes, I enjoy it fine. No need to justify it to you or your preference to me, but I think there is something broader at play. I am willing to be wrong, but I am really scouting for information that may make sense of what I am hearing. I don't believe I am hearing it in isolation, but it may be part of a trend.

I'm engaging in a thought experiment/discussion. I should know better than floating such things on ASR, as they seem to go down like a lead balloon. Some people have a fantastic ability to bypass the main point and head straight to minor side points.

The main points of my OP are 1st paragraph and paragraph starting "can anyone offer".

Science, it seems, is not always considered as being the processes that lead up to new discovery and something becoming an established fact, rather than just the fact itself. I don't suggest anything so grand will come of what I am trying to figure out here, by any means, but I do feel there is sometimes a stark lack of imagination for what could be on this forum and it can make fruitful discussion like wading through treacle.

Cheers, Digby (English idiom dispenser).
As explained, I cant (copyright) and secondly dont know how to provide excerpts from a book I own in Kindle.

I dont remember seeing a picture of the Harman listening room in the book (at least not the Kindle version, but I might be mistaken).

Did you watch the video I linked? It shows the Harman listening room at ca. 10min into the video. Didnt I provide the info you asked for? If not, please clarify.
Also I can provide this (Google) as it is in the public domain - it details the setup of the Harman tests (room, procedures etc)

As for the music used for the Harman listening studies, yes Toole gives a list somewhere, but I dont have the time looking for it in the book. I only remember, it is a combination of many music styles of the time, so not only classical.

Please dont put the burden of proof onto me for your speculation (you said so, it was speculation yourself ( ..9 out of 10 prefer mid dip ... - thank you for the clarification, so we can put this aside).
 
Last edited:

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
The article i shared has nothing about the BBC dip, it’s talking about a completely different topic and involves a different frequency range.

I’m just saying there is a certain argumentative tone to all your comments that doesn’t invite anyone to participate with you in any discussion.

None of us are experts, we’re all learning.
Yes we are all learning ....

I interpreted this in your article as a reference to the BBC dip. Maybe I am mistaken.
1658594700659.png


and yes the second paragraph refers to "bad" recordings, which, as you correctly stated should be addressed during production.
1658594920990.png


However, I am confused then (honestly): Did you mean, by linking this article, to provide evidence contradicting the Harman studies (flat response prefered, also see the video at ca. 58:45min) or did you link it to make your point on the "bad" recording potential?
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,597
Likes
2,235
Measurements are required here, both before and after the "midrange dip" was applied.
Otherwise, the premise is just not-so-cleverly-disguised wanking (gotta appreciate the Brit idioms!).
 

Mart68

Major Contributor
Forum Donor
Joined
Mar 22, 2021
Messages
2,605
Likes
4,854
Location
England
It is quite possible. What is the music you mostly listen to? Probably a full 80% of music I listen to is unamplified sonatas, concertos and such. Probably 2% of all music has vocals in it. If the music you listen to is mostly Rock, say, with 98% being vocal music, then it is quite possible this provides a different perspective entirely.
I listen to pretty much everything except Classical. So yes, I agree, the perspective is different.
 

Vacceo

Major Contributor
Joined
Mar 9, 2022
Messages
2,627
Likes
2,748
Just a technical curiosity: the dip comes from traditional speaker designs and how crossovers work. That can be fixed with a crossover, right?

Does the dip happen in concentric/coaxial speakers too?
 

terryforsythe

Senior Member
Forum Donor
Joined
May 4, 2022
Messages
395
Likes
335
I just got a pair of Elac Uni-Fi Reference UBR62 speakers. They have a small bass/midrange dip, from about from 250 Hz - 700 Hz. To me vocals sound suppressed and overall the music is not as dynamic as I like. I used Dirac Live to get a flat response, and it sounds much better to me on good quality recordings. Some lower quality recordings do sound a little bright on the high end. For those I am using Dirac's default response curve, which is linear but tilted, starting about +3dB at 20 Hz and ending about -3dB at 20 kHz.
 
Last edited:

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,421
Likes
2,406
Location
Sweden
A countertenor with EQ applied:
1. Dip 2 kHz, peak 3-4 kHz

2. Peak 2 kHz, dip 3-4 kHz

With speakers in stereo triangle, which one is preferred?
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
4,765
Likes
3,703
Can anyone proffer other suggestions as to why one may preferred a dipped midrange, that is not related to personal preference.
Some speakers do this better than others and it depends partly on directivity. Crossover implementation, waveguide vs no waveguide, will make a difference.

Also, younger ears will tolerate the dipped midrange more whereas older ears will want more output there.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,421
Likes
2,406
Location
Sweden
#1 is clearer. Since I'm old as the hills and half deaf (I exaggerate slightly ;)) I would have expected #2 to be clearer. Not so.

Why?

Jim
It is definitely a "thinner" body of the voice. I think that is making it clearer. But unnatural sounding and tiresome in the long run, IMO. This goes also for female voices I've listened to. Why? I don't know really, other than I've stumbling on the 1-5 kHz region for more than 20 years of DIY speaker building. I have reached a good balance now.
 
OP
D

Digby

Major Contributor
Joined
Mar 12, 2021
Messages
1,632
Likes
1,555
A countertenor with EQ applied:
1. Dip 2 kHz, peak 3-4 kHz

2. Peak 2 kHz, dip 3-4 kHz

With speakers in stereo triangle, which one is preferred?
For me, the second. The first is certainly immediate, but it doesn't sound natural to my ears. Age wise I am about half way there, is this a large factor? I thought as people got older they lost the highs (about 8-10khz+), not that their tastes regarding midrange changes too.
 
OP
D

Digby

Major Contributor
Joined
Mar 12, 2021
Messages
1,632
Likes
1,555
This is all very complex, sometimes I wish my hobbies were simpler. Do you have any simple hobbies, Jim?
 
Top Bottom