Do Room EQ ruin stereo soundstage and speaker coherency ?

[Duke]

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.

My guess is that the "detection threshold" under the test conditions is probably not the same as what "stereo imaging requires..." to generate a credible stereo image.

 

[Weeb Labs]

My guess is that the "detection threshold" under the test conditions is probably not the same as what "stereo imaging requires..." to generate a credible stereo image.

Yes. The test conditions under which the greatest discerning ability was measured utilized pure sine waves in order to produce a best case scenario. The findings of this study indicate 2.5mm to be the smallest discernible ITD value but they do not make pronouncements regarding the point at which stereo imaging breaks down.

In the present real world scenario, we are dealing with deltas of approximately 90mm, which are just under four orders of magnitude greater than the minimum threshold findings of the study.

 

[Sancus]

My guess is that the "detection threshold" under the test conditions is probably not the same as what "stereo imaging requires..." to generate a credible stereo image.

Right, I agree that's probably correct. Given the detection thresholds are very low, it seems like stereo imaging must still function reasonably well even if there are detectable distance differences. So to determine how much difference people can tolerate on average would probably require controlled testing of that specific question.

There's also going to be other variables like the signal/material and the loudspeaker dispersion. FWIW, my personal anecdotal experimentation indicates that it's easy to make stereo images collapse with 1 foot distances/1ms delays, but under 6"/0.5ms doesn't tend to cause too much of a problem.

That said I haven't done that much testing in a while since I Auro3D upmix everything these days, and due to use of the center the image does not collapse in that way anymore.

 

[Soundmixer]

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

Thank you for this, but there is a problem. This paper does not address the arrival times of multichannel or immersive systems where there can be anywhere from 8 channels (5.1.2), 11 channels (7.1.4), 13 channels (7.1.6), or 16 channels (9.1.6). Components with these channel configurations usually have some kind of room correction.

I must agree with Sancus on this one.

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.

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.

I have to agree with both of these gentlemen.

Regarding stereo content, I would contend that it doesn't.

Keeping this in context about Audyssey, how many stereo components actually have it? Secondly, was this gleaned from headphones or speakers? What speakers or headphones were used? What size was the room, and are any room treatments deployed in the room? While this paper is big on conclusions, it is not big on fine detail about the equipment and room the experiment was conducted in.

 

[ernestcarl]

Right, I agree that's probably correct. Given the detection thresholds are very low, it seems like stereo imaging must still function reasonably well even if there are detectable distance differences. So to determine how much difference people can tolerate on average would probably require controlled testing of that specific question.

There's also going to be other variables like the signal/material and the loudspeaker dispersion. FWIW, my personal anecdotal experimentation indicates that it's easy to make stereo images collapse with 1 foot distances/1ms delays, but under 6"/0.5ms doesn't tend to cause too much of a problem.

That said I haven't done that much testing in a while since I Auro3D upmix everything these days, and due to use of the center the image does not collapse in that way anymore.

I'm just using JRiver's stock surround upmixer (probably similar to the old dolby pro-logic IIx). At the leftmost corner seats of my couch. Now, the image certainly shifts to the left, but I wouldn't say stereo imaging perceptually collapses destructively. It actually still sounds great. Overall spectral and SPL balance between channels may partially explain why this is the case...

 

[ddlooping]

Hi all

I hadn't realised the importance of setting the L/R distances correctly for stereo listening until very recently.
As advised in several threads/articles on the subject, I used the following track to test my system "phantom center"...

The singer is supposed to appear precisely in the center of the soundstage, with little to no "spreading/smearing" into the left and right sides.
I was gobsmacked to discover that setting a wrong distance, down to 10cm or less, could "stretch" the singer from left to right as opposed to clearly defined in the center.
What's interesting is that the spread of the phantom center due to incorrect distances might actually be more pleasing to some.

 

[Koo]

Audyssey does not recommend changing the distances. They are simply calculated from delays and do take into account reflections too. Since Audyssey FIR filtering corrects the signal also in time domain by aligning the on axis sound with the reflections, it's critical not to mess with the measured delays or else you could get very messy results. I tested this myself very unscientifically by creating a copy of my measurements and changed the delays. Switching the profiles takes a pretty long time so the A/B test isn't very good but to my ears it sounds better with the original profile that has the Audyssey measured distances.

 

[puppet]

L/R amplitude differences can shift the image as well.

I wouldn't be as hasty to dismiss the results of the paper posted earlier. If one has the ability to delay a tweeters response compared to the mid range driver in the same loudspeaker, one can hear the change in timbre caused by the phase relationship between the drivers. I'd image the same condition is at work with individual loudspeakers placed at dissimilar distances to the primary listening spot or as we move from left to right in front of an equally spaced pair. One reason I don't understand EQing a rooms response to death. For my money, just work to "somewhat" limit the obvious response peaks. Anymore than that (creating a newer "factory frequency response") is asking for other problems. Work to treat the room if it's that bad.

 

[Holmz]

…
Regarding the subwoofer; this is to be expected. The DD18's onboard DSP introduces additional latency to the signal path, which is what the AVR's distance values represent. In the case of your passive speakers, the only source of latency is their physical distance from the microphone.

And also any group delay gets ball up in there as well.