• 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!

How importend is impulseresponse of a speaker for the stereowide?

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
Have speakers with better impulsresponse a better stereowide? And if so, how do you distinguish between other failures that lead to greater stereowide? Many questions in my mind.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Have speakers with better impulsresponse a better stereowide? And if so, how do you distinguish between other failures that lead to greater stereowide? Many questions in my mind.

To my mind, it makes more sense to look at the radiation pattern of the speakers. The impulse response view doesn't provide enough obvious difference in information to make any kind of strong assumptions in that context.
 
OP
tomtoo

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
To my mind, it makes more sense to look at the radiation pattern of the speakers. The impulse response view doesn't provide enough obvious difference in information to make any kind of strong assumptions in that context.

Thx! If you have any information, it would be very helpfull. See i have a ongoing discussion about this. Iam your opinion, but i want more information.
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
I agree with @ernestcarl: the (anechoic) impulse response won't tell you anything about the spatial properties of the speaker.

In terms of radiation pattern, in general, speakers with wider directivity will tend to have the widest stereo image (assuming a reflective listening room).

Psychoacoustically, horizontal reflections, particularly lateral and contraleteral sidewall reflections, are by far the most significant factor in "apparent source width", which is essentially what I think you're calling stereo width.

This implies that a speaker with wide horizontal directivity, placed in a listening room in which sidewall reflections are not absorbed, is likely to produce the the widest soundstage.
 
OP
tomtoo

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
I agree with @ernestcarl: the (anechoic) impulse response won't tell you anything about the spatial properties of the speaker.

In terms of radiation pattern, in general, speakers with wider directivity will tend to have the widest stereo image (assuming a reflective listening room).

Psychoacoustically, horizontal reflections, particularly lateral and contraleteral sidewall reflections, are by far the most significant factor in "apparent source width", which is essentially what I think you're calling stereo width.

This implies that a speaker with wide horizontal directivity, placed in a listening room in which sidewall reflections are not absorbed, is likely to produce the the widest soundstage.

Thx! Let me explain. I have a friend that is nearly obsessed by the thinking that speakers with better impulse response give greater stereo wide. I dont know, for me its possible, maybe, maybe not.My personal opinion is exactly what you say. Radiatonpattern and roomreflexions have the biggest influence. But problem for me is how big is the influence of better impulseresponse? Someone did a deeper look?
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Thx! Let me explain. I have a friend that is nearly obsessed by the thinking that speakers with better impulse response give greater stereo wide. I dont know, for me its possible, maybe, maybe not.My personal opinion is exactly what you say. Radiatonpattern and roomreflexions have the biggest influence. But problem for me is how big is the influence of better impulseresponse? Someone did a deeper look?

I don't know of any study linking impulse response (or any time-domain properties) with stereo width specifically, but there's been a lot of research of thresholds at which phase distortion/group delay becomes audible and, generally speaking, most passive speakers (and most active speakers) stay below even the most stringent experimentally derived thresholds for loudspeakers/music.

I can link you to some of these studies if you like?
 
OP
tomtoo

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
I don't know of any study linking impulse response (or any time-domain properties) with stereo width specifically, but there's been a lot of research of thresholds at which phase distortion/group delay becomes audible and, generally speaking, most passive speakers (and most active speakers) stay below even the most stringent experimentally derived thresholds for loudspeakers/music.

I can link you to some of these studies if you like?

Oh that would be very nice!!!
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
OP
tomtoo

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany

onion

Senior Member
Joined
Mar 5, 2019
Messages
338
Likes
379
A cross-talk cancellation system (XTC - that 'cancels' the sound of the left speaker at the right ear and vice versa) performs better with 'clean' impulse response as opposed to one marred by reflections. By performing better, I mean that the 3d soundstage becomes much better defined. To my simple mind, that suggests that a 'better impulse response' does improve 'stereo-wide'. In my case, that is more to do with the room than the speakers

Screenshot 2021-01-20 at 19.37.52.png
Screenshot 2021-01-20 at 20.43.45.png
 

Attachments

  • Screenshot 2021-01-20 at 19.38.08.png
    Screenshot 2021-01-20 at 19.38.08.png
    72.4 KB · Views: 95
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
A cross-talk cancellation system (XTC - that 'cancels' the sound of the left speaker at the right ear and vice versa) performs better with 'clean' impulse response as opposed to one marred by reflections. By performing better, I mean that the 3d soundstage becomes much better defined. To my simple mind, that suggests that a 'better impulse response' does improve 'stereo-wide'. In my case, that is more to do with the room than the speakers

View attachment 107423View attachment 107429

I just finished an experimentation with linearizing the phase of my KH120s manually using rePhase, and my previous suspicion about the audible effect has been confirmed -- at least it has finally convinced me personally -- increased clarity and additional "brightness" colouration noticeable in mid- and high-frequency transients. The overall effect is subtle, though, for the vast majority of content. But there are certain tracks/types of music content where it's definitely more obvious such as this one: https://www.amazon.com/Re-Foc-Rodrigo-Y-Gabriela/dp/B0000BXBY8

Interestingly, it also had the side-effect of visibly improving the impulse views seen in REW a little bit as well.

However, quite frankly, I am still unconvinced of its effect with regard to stereo-width perception -- perhaps the slight improvement in clarity gives this supposed side 'impression' indirectly. The measurements and width perception I experienced with my KH120s were already quite excellent even well before the time-domain correction -- and my listening tests were done in the nearfield at 0.8 - 1 meter.
 

dasdoing

Major Contributor
Joined
May 20, 2020
Messages
4,209
Likes
2,675
Location
Salvador-Bahia-Brasil
the effect of linearizing the phase of a SPEAKER can't be making too much diference at LP since there will be delayed signals all over the place.
linearizing the phase of the LP has an audible effect on the image...I wouldn't call it wider though, the center image is more centered, which in consequence makes the side image clearer
 
Last edited:

BenB

Active Member
Joined
Apr 18, 2020
Messages
284
Likes
446
Location
Virginia
I just finished an experimentation with linearizing the phase of my KH120s manually using rePhase, and my previous suspicion about the audible effect has been confirmed -- at least it has finally convinced me personally -- increased clarity and additional "brightness" colouration noticeable in mid- and high-frequency transients. The overall effect is subtle, though, for the vast majority of content. But there are certain tracks/types of music content where it's definitely more obvious such as this one: https://www.amazon.com/Re-Foc-Rodrigo-Y-Gabriela/dp/B0000BXBY8

Interestingly, it also had the side-effect of visibly improving the impulse views seen in REW a little bit as well.

However, quite frankly, I am still unconvinced of its effect with regard to stereo-width perception -- perhaps the slight improvement in clarity gives this supposed side 'impression' indirectly. The measurements and width perception I experienced with my KH120s were already quite excellent even well before the time-domain correction -- and my listening tests were done in the nearfield at 0.8 - 1 meter.

If you're interested in a completely isolated test for phase distortion, that's matched in frequency response to hundredths of a dB, please see my post in this thread:
https://www.audiosciencereview.com/forum/index.php?threads/phase-distortion-abx-testing.18709/

In my experience, the vast majority of differences attributed to phase linearization are actually attributable to magnitude differences.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
If you're interested in a completely isolated test for phase distortion, that's matched in frequency response to hundredths of a dB, please see my post in this thread:
https://www.audiosciencereview.com/forum/index.php?threads/phase-distortion-abx-testing.18709/

In my experience, the vast majority of differences attributed to phase linearization are actually attributable to magnitude differences.

Magnitude is the same in my measurements. It did not change. Nada. Zero.

1611255212595.png


Actually, there is a small difference visible only if you apply FDW.

1611255217361.png
 
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Magnitude is the same in my measurements. It did not change. Nada. Zero.

View attachment 107586

Actually, there is a small difference visible only if you apply FDW.

View attachment 107587

I have to say, based on what we know about the audibility of group delay vs the audibility of amplitude changes, wouldn't you agree that the latter is the far more likely source of the perceived differences in this case?

Also, have you verified using blind ABX or similar that the differences are audible?
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
I have to say, based on what we know about the audibility of group delay vs the audibility of amplitude changes, wouldn't you agree that the latter is the far more likely source of the perceived differences in this case?

Also, have you verified using blind ABX or similar that the differences are audible?

Perhaps if it were only isolated at 2kHz. The HF brightness I speak of goes way up in the 'presence' and 'brightness' region of the frequency response.

Sorry, not a blind ABX listening test -- which I know would make a lot of people happy -- I really don't care about convincing anyone else. What I used was jus a simple matter of disabling and enabling of the impulse .wav file convolution or whatever you call it using the space bar key via JRiver.

1611256922664.png
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Perhaps if it were only isolated at 2kHz. The HF brightness I speak of goes way up in the 'presence' and 'brightness' region of the frequency response.

Sorry, not a blind ABX listening test -- which I know would make a lot of people happy -- I really don't care about convincing anyone else. What I used was jus a simple matter of disabling and enabling of the impulse .wav file convolution or whatever you call it using the space bar key via JRiver.

View attachment 107589

Yes, no need to convince anyone :) Just worth noting that what you're hearing (if indeed a result of differences in phase response) would be a major anomaly in light of the established science.
 

Tim Link

Addicted to Fun and Learning
Forum Donor
Joined
Apr 10, 2020
Messages
722
Likes
638
Location
Eugene, OR
This question relates to something I've been thinking about - cabinet diffraction effects on stereo imaging. I've been contemplating an experiment using some 1 1/8" full range drivers mounted at the end of long PVC tubes pointed straight at me to see if there's any audible stereo imaging benefit to ultra minimal cabinet diffraction design. I was thinking of listening to these fairly nearfield with woofers time aligned and perhaps behind my shoulders to keep treble from bouncing off of them, while seated relatively far away from sidewalls. Maybe some treatment on the ceiling. It wouldn't play very loud and would be too awkward a setup for normal use but it does seem like it would create an almost zero diffraction situation for the entire treble range and allow me to hear the best possible stereo imaging that can be achieved from removing cabinet diffraction. Then various diffractive elements could be added until a noticeable degradation in imaging occurs. BenB - your experiment matches up with others I've read about where people have a hard time hearing phase distortion if frequency response is held constant. That's been my personal experience too. When it comes to diffraction, where an echo signal is superimposed on the original it gets more complicated. Playing an echo through the speaker and listening to it in anechoic space or over headphones is far more noticeable and bothersome than if the same timing and level of echo is created by an actual surface at another location in a room we are in. Our brain filters that out so we don't notice it in the same way as a problem. But it still can change our perhaps confuse perception of the sound's direction.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
People who know how to analyze the measurements I got in REW can freely examine the mdat files. I have limited understanding in what to traces to look for exactly here. FDW 6 cycles applied to minimize file size. I got tired after several iterations with rePhase and so did not completely linearize the sub bass.

*One thing I did not mention is I blocked the ports just for the sake of matching my sealed sub. The effect was actually a small increase in the group delay instead! :eek: I was too lazy to take it out... and flattening the phase fixed that issue anyways.
 

Attachments

  • L+R sum Minimum Phase.zip
    649.9 KB · Views: 92
  • L+R sum Linear Phase.zip
    648.8 KB · Views: 79
Top Bottom