• 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!

Our left and right ears hear differently. Best EQ approach?

theyellowspecial

Active Member
Forum Donor
Joined
Sep 21, 2020
Messages
253
Likes
274
So, I've always noticed I have a slight left ear dominance when it comes to listening in stereo. I've compensated in most situations by increasing/decreasing LR gain to center the image. However, after purchasing Dirac I started wondering what the best approach would be, or what the best approach would be in general. After some initial Googling I soon came to learn that our left ears are music/song dominant. I'm not sure I ever knew this in the past. At least it explains my ears/brain are OK and the phenomenon is normal. But it begs the question: How do we best approach this phenomenon in our stereo and home theater setups?

EDIT: A couple studies for regarding left-right auditory dominance (laterality) in humans:

https://www.sciencedaily.com/releases/2004/09/040910082553.htm
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3294255/
 
Last edited:

North_Sky

Major Contributor
Joined
Oct 31, 2019
Messages
2,741
Likes
1,553
Location
Kha Nada
That's a good reason to have a Balance control on our preamp.
 

beefkabob

Major Contributor
Forum Donor
Joined
Apr 18, 2019
Messages
1,636
Likes
2,073
Sounds a bit pseudo scientific to me. Get a stereo that plays music accurately to the recording. Don't aim for something beyond unless it pleases you to play around with esoteric concepts that have little to do with the real world.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
I've compensated in most situations by increasing/decreasing LR gain to center the image.

Try moving the right speaker closer to you an inch at a time.

Changing the timing (in my opinion) has a more immediate effect than adjusting the relative volume.

Maybe your right ear is "slow". (I just invented this malady, so take it however you like).

---

Does the same sensation occur with headphones?
 
Last edited:

gvl

Major Contributor
Joined
Mar 16, 2018
Messages
3,427
Likes
3,981
Location
SoCal
Not to hijack the thread but sort of related, for REW measurements and EQ, does it make sense to do it separately for L/R channels or better do it combined?
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
Not to hijack the thread but sort of related, for REW measurements and EQ, does it make sense to do it separately for L/R channels or better do it combined?

I measure left and right AND combined.

They may not sum as expected, depending...

Left and right in red, both playing in green. An interesting and unexpected hole develops around 48Hz.

1601843314331.png


Reason: The standing waves from the two speakers around that frequency go about 180 degrees out of phase (cancellation) at the listening position in this asymmetrical at the rear room.

1601844007742.png


256 - 77 degrees = 179 degrees difference at 48.1hz
 
Last edited:

frangle

Member
Forum Donor
Joined
Feb 26, 2020
Messages
55
Likes
30
On one hand you could argue that the ear is just the last flawed transducer in the reproduction chain and correction for this could be a benefit. On the other, it is followed by a pretty powerful "DSP" anyway. But who knows what parameters that uses for correction.

And how would your internal DSP interpret a sound system corrected to have a flat response for both ears when everything else you hear is not corrected externally?
 
OP
T

theyellowspecial

Active Member
Forum Donor
Joined
Sep 21, 2020
Messages
253
Likes
274
Sounds a bit pseudo scientific to me. Get a stereo that plays music accurately to the recording. Don't aim for something beyond unless it pleases you to play around with esoteric concepts that have little to do with the real world.
Sorry, I should have linked some relevant studies. I updated the OP. It looks like what we're discussing is "laterality", (the dominance of one side of the brain in controlling particular activities or functions, or of one of a pair of organs such as the eyes or hands, according to the dictionary).

Try moving the right speaker closer to you an inch at a time.

Changing the timing (in my opinion) has a more immediate effect than adjusting the relative volume.

Maybe your right ear is "slow". (I just invented this malady, so take it however you like).

Does the same sensation occur with headphones?
Yes, perhaps the answer lies in the time domain. The problem is that if I post-EQ with Dirac I will be back to where I started. After all, the microphone will measure SPL as a single ear. Maybe the solution is a stereo microphone with compensation filters taking into account auditory laterality.

Yes, perhaps my right ear is slow, but then I suspect in general everybody's right ear is slow, at least according to the studies on auditory laterality.

The effect is noticeable on headphones but I should point out the effect is very slight in all stereo reproductions. My speaker plane is roughly 8' away and the centered image is maybe 3" left dominant along that plane. On headphones the effect is less noticeable than speakers.
 
Last edited:

Vasr

Major Contributor
Joined
Jun 27, 2020
Messages
1,409
Likes
1,922
Not to hijack the thread but sort of related, for REW measurements and EQ, does it make sense to do it separately for L/R channels or better do it combined?

No perfect answer but...

I recommend measuring separately for generating filters (like any commercial room eq system out there) and optionally together for detecting any cancellations (as @RayDunzl illustrated) or for verifying the correction. The phase cancellation is better fixed with room treatment or speaker positioning than EQ. You can also try to fix it by providing a small delay difference between the two that isn't enough to make the phase difference between the two audible but enough to remove some subtractive or additive effects. The latter may some times move the cancellation to a different point in the spectrum. Often, when one is using a sub, such tricks can be used to move the nulls into the sub frequency range so that the two main speakers are not radiating at that problem area.

Problem with measuring both is that it only applies when there are identical signals coming out of both. Most of the content one would be listening to wouldn't be correctly represented by this (unless you are playing mono through both channels) so such cancellations wouldn't necessarily happen to such an extent. If you correct both based on combined measurement by feeding identical signals to both for such cancellations , then you would be unnecessarily over-correcting for any content that is skewed towards one speaker or the other. The symmetrical problem is far less frequent.

The only time I do two different speakers at once is in a speaker+sub crossed over situation. But separately for each such crossed-over speakers.
 
Last edited:
  • Like
Reactions: gvl

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
My speaker plane is roughly 8' away and the centered image is maybe 3" left dominant along that plane.

Maybe your sweet spot isn't quite where you think it is.

Here, both speakers playing, acoustic reference is the left speaker, microphone is 3" to the left of my "confirmed" acoustic center.

A clear impulse is seen from each speaker, and enough to make the "image" skew slightly left of "center", when my head is in that 3 inches off location.

The right speaker is "late".

1601853191496.png


When the microphone is moved back to the acoustic center, the timing (and image) is improved:

1601853299977.png
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
DRC helps by reducing left vs right frequency response variation. No matter how careful I am with setup (I use a Stabila laser measure), I usually end up having to manually adjust the gain of one of the DRC filters using a pink noise track to get the center image right.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
Maybe the imaging is correct to begin with and adjustment is only required to suit an incorrect predisposed notion of otic symmetry.

https://www.hearingreview.com/inside-hearing/research/remarkable-history-right-ear-advantage

Ok, but...

"Note that to detect the onset of a word you need only the primitive symmetric auditory system common to all animals who need to know that a sound has occurred and from which direction it is coming. To recognize an actual word, however, you need a much more complicated asymmetric auditory system."
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,747
Likes
16,182
As known in the modal region we rather correct room and above we correct loudspeakers, thus my experience is that above the transition frequency good (=with no large pair deviation) loudspeakers don't really show significant differences when measured carefully (exact middle or MMM method) at the listeners position and even if they showed differences correcting them differently would mean making them having different direct sound which is not recommendable as in this region we perceive more the direct sound.
In the modal region the sum of L+R often shows somewhere a dip due to a room acoustic related phase rotation which can usually only be treated well with a different positioning which is though often not possible (FIR filters can usually also not help there).
Correcting each driver there separately can enhance the summed dip and because also most signals in that region tend to be recorded rather mono and also our listening separation is very insensitive there, I tend to correct that region usually as a L+R sum, especially with setups that are quite SPL limited.
 

Absolute

Major Contributor
Forum Donor
Joined
Feb 5, 2017
Messages
1,084
Likes
2,125
On my Samsung S8 there's a hearing-test (in the settings under audio) that aims to find out the difference between ears and then compensate for them to provide you with a normalized EQ so both ears hear the same. It turns out that I have some area around 2 kHz where my right ear is less sensitive than the left, probably due to one of the times I ripped my eardrum during boxing.

If I try to match perceived volume on a set of headphones to center the image, I usually need about 3-4 dB more oomph on the right ear, but if I change the balance on the speakers while sitting dead-center the phantom image goes far right if I add just half a dB. It seems headphones and individual hearing is not transferable to speakers.
 

tuga

Major Contributor
Joined
Feb 5, 2020
Messages
3,984
Likes
4,281
Location
Oxford, England
I have astigmatism on my right eye. Should I correct the image on a display/TV or by wearing glasses/contact lenses?

In other words, what is the point of correcting/compensating for ear imbalance on playback if one hears "imbalanced" live the rest of the time? Unless the plan is to correct ear response with a hearing aid.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
On one hand you could argue that the ear is just the last flawed transducer in the reproduction chain and correction for this could be a benefit. On the other, it is followed by a pretty powerful "DSP" anyway. But who knows what parameters that uses for correction.

And how would your internal DSP interpret a sound system corrected to have a flat response for both ears when everything else you hear is not corrected externally?


My ears are flawed, I could not honestly claim otherwise, nor have I.

My internal DSP seems to be working well enough, though, with the bandwith it has available to work with.

Any live sounds in the room, above the transition frequency (which most are), I should expect to be transmitted to my ears from the source without any particular modification in frequency response, or "flat". (fact check for that below)

Everyday sounds in the room do not excite the low frequency room nodes to a noticeable degree. Full range music does, and at reference levels, causes some dissatisfaction and demands a little attenuation in that area, if you can consider 20dB hot "a little".

For the rest of the audio band, I will, for now, entertain the thought that my speakers might be a little deficient, and allow the microphone to suggest some changes, and turn the automated DSP loose to calculate a adjustment according to the psychoacoustic principles it follows (AcourateDRC) to that area.

My stated "goal" is to make the equalized sonic RTA at the listening position match the raw electrical RTA of the recording being reproduced. If frequency response is the dominant factor in how something sounds, I figure if I can match peak levels across the band, it's a useful and relatively achievable target.

The automated DSP will also make some timing adjustments, again, most obviously to the bass region.

Matching waveforms between that received at the listening position versus the waveform present in the source won't work due to the remaining room interaction (reflection, echo, phase), else I would attempt to closely match the signal's wave at the mic.

"Flat" EQ achieves the stated goal, and presents the recorded signal to the listening position, with only a minimum of obvious errors, I still have a mono bass hole at 48Hz (which usually disappears when the material has the bass recorded in stereo), and a dip at the frequency calculated for "dipole bounce off the wall behind", also present with the monopole speakers, which surprised me to some degree.

If you think the highs are too bright with a "flat" EQ, consider the natural roll-off inherent in any reasonably well recorded piece of music. There seems to be enough attenuation in the highs already present in the recording.

If not, I have a setting saved on the DEQ2496 to provide a "scientifically acceptable based on preference testing" 10dB slope from 20Hz to 20kHz,

I find that slope accentuates the bass again, and don't use it, but it is there to be recalled if somebody should demand it.

---

Dissipation of high frequencies in air:

1601969559108.png


http://www.sengpielaudio.com/calculator-air.htm

Ok, so at my 3m listening position my "flat" EQ could be argued to be 0.39dB "hot" at 10kHz vs what's on the disc or in the file or streamed.

Below 4kHz less than 0.078dB and becoming even more less with decreasing frequency.

Maybe I'll fix that sometime.
 

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
753
Likes
625
Location
Herts., England
I have astigmatism on my right eye. Should I correct the image on a display/TV or by wearing glasses/contact lenses?

In other words, what is the point of correcting/compensating for ear imbalance on playback if one hears "imbalanced" live the rest of the time? Unless the plan is to correct ear response with a hearing aid.

Not sure they are comparable.

You only look at one screen but we listen in stereo and from the information our brain creates a phantom image.

Does creating a phantom image have an equivalence in real life? If not then I would guess it would be appropriate to adjust EQ for unbalanced hearing damage just as much as you might do for an unbalanced room.
 
Top Bottom