Originally Posted by iStorm
can answer this since I am going back through and re-reading the guide and I need clarification from the writer himself. Mr. Thomas, you briefly mention that it may be better to use your calibration 8 mic positions that do not go back behind the chairs, but do not mention why you feel that going behind the chairs shouldn't be done? The couch downstairs that has the Audyssey XT32 (using mobile app) has 4 chairs side by side in 1 row. Can you please expand on this on why it is not beneficial to do mic measurements behind the couch? I know you said to do trial and error (like everything else in this hobby), but I would like an explanation on this aspect please. I have done measurements using your calibration technique and also measurements the standard way that Audyssey wants it done, but do not discern too much of a difference in my room, especially since the distances/timings, etc are all based off of Mic position 1.
Thanks in advance, Mike!
I will have to go back to reread what I wrote, but I believe that I did explain why it probably isn't typically a good idea to go behind your main listening position with any of your microphone measurements during calibration. It's because no one is actually listening back there, and we typically want to EQ where we are
listening. I use words like "probably" and "typically" in most of my advice and explanations because I believe that it's better to generalize, rather than to speak in absolute terms, when we talk about this sort of thing.
For instance, you said that you have tried it both ways and can't really measure much difference. I would be curious whether you can hear
any difference, because those are two different things. But, I will leave that alone for a moment. I think that, in many cases, individual listeners might be able to hear a difference. And the general theory of going behind your chair with some mic positions, when no one is actually sitting there, just doesn't make any sense to me.
We know from long experience with Audyssey, that Audyssey usually (another generalization) works best when we keep the microphone pattern fairly compact. Even Audyssey came to realize that over time, and now recommends a much smaller microphone spacing than they did 6-8 years ago. The reason for that is the fuzzy-logic weighting system that Audyssey uses to average the results of the measurements from multiple (usually 8) microphone positions. The more we prevent anomalous measurements of random peaks and valleys from skewing the average, the better room EQ we generally get.
So, we try to concentrate our measurements in the area of the most important listening position(s), knowing that we can't realistically EQ the entire room. From long observation of Audyssey results, I would say that the larger area we attempt to EQ, the more likely Audyssey is to degrade our sound quality at the MLP (main listening position) rather than to improve it. So, we try to keep our measurement mic at about ear height, because that is where we are actually listening, and we try to keep the measurement microphone fairly close to just one, or at the most two, listening positions.
How does all of that relate to going behind a listening position with a couple of measurements? Well, why would we do that, if there is no one listening back there? Wouldn't that be more likely to generate exactly the kind of anomalous measurement information that we were trying to avoid? At best, it would be pretty unlikely to help, and at worst we would be much more likely to pick-up reflections from other surfaces (such as walls) behind our listening position.
This is one of those instances where owner's manuals are trying to cover every possibility, such as home theaters with multiple listening rows. And, as a result of trying to cover every possibility, they give bad advice to almost everyone. I think that the advice they give to everyone, to use two mic positions behind the main listening positions, is likely to end-up doing more harm than good. And, anecdotal reports from a number of users, in the form of both measurements and listening experiences, have tended to support that.
I also think that user manuals are rarely written by the engineers who actually designed the products. And, this is like the advice that Audyssey manuals used to give to keep measurements at least 3-4' apart. Users who followed that advice very often found themselves with relatively poor sound quality at the MLP.
The final thing that I will say about this is that measurements don't always provide an essential correlation to what we actually hear, even if we are fairly sophisticated at looking at FR graphs and impulse responses. For instance, a relatively modest looking boost at higher frequencies may translate into a shriller sound at some treble frequencies. The same thing is true for a very busy looking treble response.
We can often see a very jagged high-frequency response when we get the Audyssey microphone too close to a reflective surface, such as a seat back. That's why we have learned to put something absorbent over the seat back, during calibration. And, it is mostly higher frequencies that I would worry about, if I measured behind my listening position, because the higher frequencies are so dependent on reflections from any surfaces in the room. (I have also seen reports and measurements demonstrating that low-frequencies were adversely affected by measuring behind the MLP during calibrations.)
The Audyssey omnidirectional microphone can pick-up both early and late reflections, of high-frequencies, in a way that will influence the room EQ that it performs. And yet, our brains would often have ignored some of those early and late reflections that Audyssey "heard", and which it tried to "correct". In those cases, an unnecessary correction can actually degrade the sound quality. The Audyssey microphone and our brains simply don't "hear" sounds in the same way. The better we understand that fact, the better we can help Audyssey (or other forms of room EQ) to help us to achieve improved sound quality.
As a side note, I believe that the issue of high-frequency reflections, occurring at different points in space, is why some people end-up choosing not to use Audyssey (or other forms of room EQ) above certain frequencies. Depending on the room, and the speakers involved, it can sometimes take a fairly sophisticated effort to get a good full-range calibration. And, for some people, the full-range calibration might never be as satisfactory for them. YMMV again!
Ultimately, where room EQ is concerned, I believe that we all have to trust our ears, and our brains, which significantly influence what we hear. If a particular calibration (or setting) sounds better to us, then it is
better, because our real goal is improved sound quality. If we really can't hear any difference between two calibrations, then by definition, there really isn't
any meaningful difference where we are concerned.
Having said all of that, it is easy to understand why I consider so many of these questions to have a strong YMMV component. Different people, in different rooms, with different systems, will hear different things. The most that the Guide can do is to provide generalized advice. And then, it it up to individuals to decide if that advice is applicable to their specific circumstances. Measuring behind the MLP, during an Audyssey calibration, is a good example of that.
Not exactly how I had envisioned spending my 4th of July morning, but oh well!
Edit: I decided to provide a link to this post in Section I-B of the Guide. That way, people who are interested in having more detail, on why it may not be advisable to calibrate behind our listening chairs, can read the above explanation.