• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Do Room EQ ruin stereo soundstage and speaker coherency ?

Duke

Addicted to Fun and Learning
Manufacturer
Forum Donor
Joined
Apr 22, 2016
Messages
797
Likes
1,791
Location
Princeton, Texas
Does Audyssey measure and room-correct each channel separately?

If so, seems like the typical result would be a different first-arrival frequency response from each speaker, which I would expect to degrade imaging.
 
Last edited:

Galz

Member
Joined
Aug 8, 2021
Messages
40
Likes
14
Does Audyssey measure and room-correct each channel separately?

If so, seems like the typical result would be a different first-arrival frequency response from each speaker, which I would expect to degrade imaging.

It does correct each speaker separately. Now I'm not expert, but if you measure properly then you should get correction results as symmetric as your uncorrected response, so if you lack symmetry in the correction I suppose you also lacked symmetry without correction anyway, so imaging shouldn't be any worse than without correction?

If first-arrival is symmetric but reflections aren't, that might exactly be why some are recommending to limit the correction frequency, which seemed to have helped for me (but I have only XT which is not as good as XT32 which could be the reason I needed to limit it to improve the results).
 

Duke

Addicted to Fun and Learning
Manufacturer
Forum Donor
Joined
Apr 22, 2016
Messages
797
Likes
1,791
Location
Princeton, Texas
It does correct each speaker separately. Now I'm not expert, but if you measure properly then you should get correction results as symmetric as your uncorrected response, so if you lack symmetry in the correction I suppose you also lacked symmetry without correction anyway, so imaging shouldn't be any worse than without correction?


The first-arrival sound is by far the most important for imaging, and imo sacrificing symmetry in the first-arrival sound in favor of symmetry in the in-room sound would not be conducive to good imaging.

Presumably the first-arrival sound from each of a stereo pair of loudspeakers is identical, up to the limits of production tolerances, but the in-room sound from each of a stereo pair of loudspeakers is virtually guaranteed to be different because rooms are virtually never acoustically symmetrical.

It is generally not possible to EQ a speaker's in-room response independent of the first-arrival sound. So to the extent that each speaker of a stereo pair is getting a different EQ in order to achieve the desired in-room response, the first-arrival sounds from the two speakers become increasingly dissimilar in frequency response.

At frequencies (in particular mid and high frequencies) where one speaker's first-arrival sound is louder than the other's, the image will be pulled towards the louder speaker.

I welcome correction from anyone who knows more about this than I do; I have zero experience with Audyssey.
 
Last edited:

Kal Rubinson

Major Contributor
Industry Insider
Joined
Mar 23, 2016
Messages
3,423
Likes
5,301
Location
NYC/CT
I welcome correction from anyone who knows more about this than I do; I have zero experience with Audyssey.
AFAIK, Audyssey does not take any inter-speaker correlations into account. It is, possibly, one reason why some users prefer to limit its corrections to the sub-Schroeder range.
 

Galz

Member
Joined
Aug 8, 2021
Messages
40
Likes
14
AFAIK, Audyssey does not take any inter-speaker correlations into account. It is, possibly, one reason why some users prefer to limit its corrections to the sub-Schroeder range.
While theoretically it sounds like it makes sense, I'm not 100% sure if this is the main reason to limit the frequency. Most explanations I found for why to limit it are more due to the lack of the correction system to be able to properly correct the higher frequencies, even when considering only a single speaker, plus those higher frequencies generally don't really need correction anyway, at least nowhere near what the low frequencies need.
 

Soundmixer

Active Member
Joined
Mar 8, 2021
Messages
210
Likes
109
The problem that you are describing is caused by Audyssey's tendency to assign slightly different distances to each of the front channels during calibration. Manually setting the front left and right channels to exactly the same distance value will solve this problem.

If I am not mistaken, Audyssey does not use the actual measured distance to determine distance. It uses arrival time to the microphone, so if we are not perfect in the setup of the microphone in relation to the speakers, then the distance calibration won't be identical between channels. WE determine whether Audyseey will give the correct distance. If we don't, Audyseey compensates for that.
 

Weeb Labs

Active Member
Joined
Jun 24, 2020
Messages
276
Likes
559
Location
Ireland
If I am not mistaken, Audyssey does not use the actual measured distance to determine distance. It uses arrival time to the microphone, so if we are not perfect in the setup of the microphone in relation to the speakers, then the distance calibration won't be identical between channels. WE determine whether Audyseey will give the correct distance. If we don't, Audyseey compensates for that.
In principle you are correct but regardless of microphone position relative to the speakers, Audyssey will almost always fail to assign identical distance values to each of the front channels. The acoustic data is simply too noisy to reliably produce measurements accurate to the second decimal place. At frequencies such as 10KHz, even a mismatch of 0.09 metres comprises several full wavelengths of latency and so is very destructive to the perception of spatial qualities.
 

Galz

Member
Joined
Aug 8, 2021
Messages
40
Likes
14
In principle you are correct but regardless of microphone position relative to the speakers, Audyssey will almost always fail to assign identical distance values to each of the front channels. The acoustic data is simply too noisy to reliably produce measurements accurate to the second decimal place. At frequencies such as 10KHz, even a mismatch of 0.09 metres comprises several full wavelengths of latency and so is very destructive to the perception of spatial qualities.
I'm not sure exactly how it measures the distance, but although it was tedious, I always eventually manage to get it to measure equal distances. I wonder if having a better way to level the microphone would make it easier. If I just re-measure without moving the mic usually I get the exact same result as the previous measurement.
 

Kal Rubinson

Major Contributor
Industry Insider
Joined
Mar 23, 2016
Messages
3,423
Likes
5,301
Location
NYC/CT
OTOH, the distances are, in fact, unimportant. It is the delays consequent to those distances because other factors, e.g., processing latencies in individual channels, demand compensation as do physical placement disparities.
 

Galz

Member
Joined
Aug 8, 2021
Messages
40
Likes
14
OTOH, the distances are, in fact, unimportant. It is the delays consequent to those distances because other factors, e.g., processing latencies in individual channels, demand compensation as do physical placement disparities.
For identical speakers amplified by identical amplifiers I'd expect the processing should be the same, though, so I would expect accurate mic placement to provide equal distances (and therefore delays) for both left and right speakers. I found this wasn't easy to achieve in practice but was possible after several attempts to place the mic. I highly doubt that difficulty was due to differences in processing times.
 

Soundmixer

Active Member
Joined
Mar 8, 2021
Messages
210
Likes
109
In principle you are correct but regardless of microphone position relative to the speakers, Audyssey will almost always fail to assign identical distance values to each of the front channels. The acoustic data is simply too noisy to reliably produce measurements accurate to the second decimal place. At frequencies such as 10KHz, even a mismatch of 0.09 metres comprises several full wavelengths of latency and so is very destructive to the perception of spatial qualities.


In a room full of reflections, do you really believe our ears can hear that mismatch? I would agree with you if we were in an anechoic chamber, but not in a room full of reflections.
 

Soundmixer

Active Member
Joined
Mar 8, 2021
Messages
210
Likes
109
For identical speakers amplified by identical amplifiers I'd expect the processing should be the same, though, so I would expect accurate mic placement to provide equal distances (and therefore delays) for both left and right speakers. I found this wasn't easy to achieve in practice but was possible after several attempts to place the mic. I highly doubt that difficulty was due to differences in processing times.


I physically measured the distances from the MLP to each of my speakers to make sure they were physically aligned BEFORE I ran Audyssey. I then put an "X" at that central point where I put the microphone. Each time I put the microphone in that X, I got the same distance measurements over and over again. It does take practice and some level of pre-planning to get reliable results.
 

Weeb Labs

Active Member
Joined
Jun 24, 2020
Messages
276
Likes
559
Location
Ireland
In a room full of reflections, do you really believe our ears can hear that mismatch?
Demonstrably. Reflections are irrelevant to this particular scenario, as both direct and indirect dispersion are subject to the same additional latency. As explained, we are not simply dealing with the natural soup of phase shifts which occur within a typical room here but rather many full wavelengths of intra-channel latency.

The effect is similar to that of moving significantly off-centre between two speakers and worsened considerably by the fact that both channels retain exactly the same SPL in this case, which together with arrival time would ordinarily comprise two of our most critical modes of localisation.
 
Last edited:

Soundmixer

Active Member
Joined
Mar 8, 2021
Messages
210
Likes
109
Demonstrably. Reflections are irrelevant to this particular scenario, as both direct and indirect dispersion are subject to the same additional latency. As explained, we are not simply dealing with the natural soup of phase shifts which occur within a typical room here but rather many full wavelengths of intra-channel latency.

The effect is similar to that of moving significantly off-centre between two speakers and worsened considerably by the fact that both channels retain exactly the same SPL in this case, which together with arrival time would ordinarily comprise two of our most critical modes of localisation.

Simple question. How does this change if you use a measuring tape? Is this anechoic data, or data gleaned from ordinary rooms? I would be interested in the demonstrable data that supports this.
 

Weeb Labs

Active Member
Joined
Jun 24, 2020
Messages
276
Likes
559
Location
Ireland
Simple question. How does this change if you use a measuring tape? Is this anechoic data, or data gleaned from ordinary rooms? I would be interested in the demonstrable data that supports this.
I’m afraid that I don’t understand your question or how it relates to the explanation that I have provided.
 

Soundmixer

Active Member
Joined
Mar 8, 2021
Messages
210
Likes
109
I’m afraid that I don’t understand your question or how it relates to the explanation that I have provided.

Okay, let's simplify this question. I would love to see the data that supports your comment. Can you provide it here?
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,549
Likes
1,086
Location
Canada
At frequencies such as 10KHz, even a mismatch of 0.09 metres comprises several full wavelengths of latency and so is very destructive to the perception of spatial qualities.

While it is going to have a psychoacoustic effect, absolutely no doubt, “destructive” seems very unlikely. I can hear a change of half a millisecond when playing with delays manually in my semi-treated room — though only barely. I usually make manual adjustments of 0.1 to 0.2 ms increments. In a normal living room and at greater distances, the exercise would be much more difficult. Being able to detect a change does not necessarily mean that perception of spatial qualities of the original recording automatically becomes degraded or compromised severely. If that were so, stereo or even multichannel surround content wouldn’t sound right unless our heads were permanently locked into one position.
 

Weeb Labs

Active Member
Joined
Jun 24, 2020
Messages
276
Likes
559
Location
Ireland
While it is going to have a psychoacoustic effect, absolutely no doubt, “destructive” seems very unlikely. I can hear a change of half a millisecond when playing with delays manually in my semi-treated room — though only barely. I usually make manual adjustments of 0.1 to 0.2 ms increments. In a normal living room and at greater distances, the exercise would be much more difficult
To be clear, the effect produced by an Audyssey channel distance mismatch is not subtle. It is very pronounced and far beyond what I would consider to be the threshold of typical confirmation bias or placebo. I am fairly certain that I could successfully demonstrate it via in-room recording and will look into the possibility this evening.

I don't entirely discount the possibility that the AVR's distance setting is modifying parameters other than simple delay but if not, then my earlier explanation would seem to be the only logical candidate.

If that were so, stereo or even multichannel surround content wouldn’t sound right unless our heads were permanently locked into one position.
Regarding stereo content, I would contend that it doesn't.

Okay, let's simplify this question. I would love to see the data that supports your comment. Can you provide it here?
I would suggest that you have a look at this paper.
 

Duke

Addicted to Fun and Learning
Manufacturer
Forum Donor
Joined
Apr 22, 2016
Messages
797
Likes
1,791
Location
Princeton, Texas
I would suggest that you have a look at this paper.


VERY INTERESTING! The detection threshold for trained listeners under the test conditions corresponds to a path length difference of a little less than 1/10th of an inch, if my math is correct.
 
Last edited:

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
1,614
Likes
3,723
Location
Canada
If stereo imaging requires equal distances as precise as within 1/10th of an inch, or even 1 inch, to the listener, then the obvious conclusion must be that stereo is unlistenable and broken for effectively all listeners. While I welcome such a conclusion, since I do think that stereo is crap, I've never even once heard stereo imaging completely disintegrate because I moved my head slightly.

I would suggest that you have a look at this paper.

A detectable difference doesn't necessarily imply a significant effect on stereo, though.
 
Last edited:
Top Bottom