Microphone Placement

[Morgan]

Hi, I have a question regarding the microphone setup for XMC-1. I watched the video and the moderator said to back away from the microphone during the measurements so that the sound is not influenced by your presents. It would seem to make more sense to be in the normal listening position during the measurements in order to replicate your presents during playback. I'm confused.

[donh50]

"Presence".

The (one) problem is tiny changes in your position can significantly change the frequency response (magnitude and phase) the mic "sees". Move a hair, and the response changes -- a lot. The calibration algorithms assume a fixed position and at the normal listening position without you in the way -- i.e. the mic needs to be where your ears would be and the system will aim for its target curve without you there. Hard to do if your head is there, and if you move around you'll be introducing an unknown, changing variable to the measurements. There have been discussions about adding an absorber "dummy" to emulate a person but it is way more trouble than it's worth. I use a boom stand to position the mic and then lie or sit down behind my couch when doing the measurements or running room correction. Better protects my ears from the loud frequency sweeps as well.

HTH - Don

[Morgan]

Thanks Don, If the microphone measurements are indeed that accurate, then it makes sense. With other set ups, I simply put the microphone on top of my head and sat as still as possible during measurements. Has anybody ever tried placing a bunch of pillows, or similar in the chair?

[KeithL]

There is another way of looking at it.

You weren't there when the original recording was made.

So, if we could calibrate your room and system so perfectly that they could be an exact duplicate of the original room...

Then, adding yourself to your room will produce pretty much the same change as it would have if you'd been transported to the original room...

For example, when you're actually listening to the system, you won't also be standing between yourself and your front speaker, or standing behind yourself and causing stray reflections.

So, if you look at it that way, you are already "an invisible observer" who has been replaced by the microphone, and standing in the acoustic field would constitute another you in a different location.

Binaural recordings are often made using a dummy head with microphones in place of the listening mechanism inside the ears.

However, that is done to accurately record the sound that reaches the dummy's ears, and NOT because of how the dummy head affects room acoustics.

(In fact, recordings are often made in empty venues, even though filling most of the seats makes a huge difference in room acoustics.)

In practice, having you standing someplace else in the room is never going to contribute positively to the accuracy of the measurements, but is far more likely to reduce it.

If you stand someplace where you block one or another speaker then the measurements will simply be inaccurate.

And, if you happen to move around, you may produce inconsistent readings, which may significantly degrade the accuracy of the calculations.

(Very much as a long-exposure picture is blurred or jumbled if the subject or photographer moves while the camera shutter is open.)

The reality is that acoustic measurements aren't quite that precise... but it still matters...

At a very minimum you should make sure that you don't stand between the microphone and any speaker it's measuring.

You also don't want to block any reflections from reaching the microphone from the walls or floor.

You will also find that symmetry (specifically left/right symmetry) is critical when making corrections and measurements.

So... for example...

- leaving the room entirely would probably be best but may be impractical

- crouching down directly behind the microphone will provide a clear path from the speakers to the microphone (and also avoid blocking most wall reflections)

- staying directly behind the microphone will ensure that any inaccuracies you do introduce are symmetrical (left to right)

- crouching in the dead spot created by something large that blocks all sound is also acceptable (like behind a large couch)

- standing still while the measurements are taken will avoid jumbling the readings and producing odd and unwanted miscalculations

There have been various attempts to measure "exactly what you would hear".

Other companies have tried using some sort of headset with a microphone by each ear... and instructing the user to move to various listening positions.

It seemed to work reasonably well.. but imposes certain limitations (for example, you need to use "fast chirps" to minimize the effects of even small head movements).

However, doing it that way has a few drawbacks, and seems to offer no practical advantages over using a single microphone.

Also remember that, unlike with headphones, you cannot completely control which sound reaches which ear, as you could with headphones.

(Binaural recordings, made with dummy heads, only really work when listened to through headphones, where you can control exactly which sounds reach which ear.)

Feb 20, 2020 14:12:18 GMT -5 Morgan said:

Hi, I have a question regarding the microphone setup for XMC-1. I watched the video and the moderator said to back away from the microphone during the measurements so that the sound is not influenced by your presents. It would seem to make more sense to be in the normal listening position during the measurements in order to replicate your presents during playback. I'm confused.

[KeithL]

The difference even a slight movement makes will vary depending on the frequency being measured and the acoustics of the room itself.

At 20 Hz the wavelength of a sound wave is about 50 feet... so moving a few inches makes almost no difference...

At 10 kHz the wavelength of a sound wave is just over one inch... so, if there are cancellations occurring, peaks and nulls may be as little as 1/2" apart...

Practical reality falls somewhere in-between...

We are most sensitive to cancellations and irregularities in the critical midrange region...

Those are the frequencies that our brains predominantly use to determine sound stage location and such.

Those are also the frequencies that are most easily corrected or addressed by room treatments and room correction software...

And the wavelength at those frequencies is on the order of several inches to a few feet...

In practice, if you hold very still, and avoid moving more than about an inch, you could probably achieve reasonable success with a microphone mounted on your head...

However, that's asking quite a lot, and taking an unnecessary risk... with no benefit in return.

A bunch of pillows will tend to prevent any reflections from below... but that really shouldn't make a significant difference.

(Just make sure they don't settle and allow the microphone to move during the measurements.)

I generally recommend that, if you don't have a stand, sitting the microphone on the back of your listening seat, on top of a folded up towel, will work pretty well.

You also DO NOT want the microphone down on the seat... where the seat back will both add unnatural reflections and block the sound from the rear speakers.

In theory, if your seat has a high back, which normally blocks sounds from the rear of your head, that would be correct place.

However, in practice, you'll end up with the rear surrounds boosted way up, and the treble frequencies in the rear surrounds, that are better blocked by the chair back being boosted even further.

As a result, it may sound OK sitting down, but the rears will sound overly sharp when you stand up, and for everyone else in the room.

And, when a sharp high pitched sound occurs in the rear, the echo of that sound from the front wall will be unnaturally loud.

And you DO NOT want to set the microphone on something like an empty box or upturned waste basket...

Doing that will add all sorts of odd pseudo-reflections at the resonant frequencies of the space inside your "stand".

(If you must do something like that then stuff your box or bucket solidly with several blankets to help prevent it from "booming".)

Feb 20, 2020 14:55:37 GMT -5 Morgan said:

Thanks Don, If the microphone measurements are indeed that accurate, then it makes sense. With other set ups, I simply put the microphone on top of my head and sat as still as possible during measurements. Has anybody ever tried placing a bunch of pillows, or similar in the chair?

[Morgan]

Thanks Keith,it's obvious that you know a whole slew more about this stuff than I do, so I'm convinced. Sounds like I'm going to put the mic on a stand at ear level hit the button and leave the room. Let's just hope that I'm quick enough. One other question that wasn't covered in the videos, when I'm taking measurements 2 through 8 on my chair, what distance should I be moving the mic up/down - left/right - front/back? Thanks again, Paul

[donh50]

Feb 20, 2020 14:55:37 GMT -5 Morgan said:

Thanks Don, If the microphone measurements are indeed that accurate, then it makes sense. With other set ups, I simply put the microphone on top of my head and sat as still as possible during measurements. Has anybody ever tried placing a bunch of pillows, or similar in the chair?

"Accurate" is not really the issue, though they are pretty good. It is measuring the frequency response so you have to look at the wavelengths and then typically the 1/4- to 1/2-wavelengths as that is the distance it takes to change the signal. Going from a negative peak to positive peak takes just 1/2 wavelength. At 100 Hz, a wavelength is about (1127 ft/sec) / (100 Hz) = 11.27 feet. Lot of leeway to move around. At 1000 Hz, only about 1 foot, and at 10 kHz, only about an inch or so, so moving more than about half an inch can completely change the response at 10 kHz, and an inch or two of wiggle will impact much of the upper vocal band. If you could sit perfectly still that would help, but that's pretty hard to do -- even if you (try to) control your conscious movements, stopping bodily functions like breathing and heart beats can be detrimental. So your body moves a little, and that upsets the readings.

As for absorption, your body will absorb some sound energy, varying with frequency, hair, clothing (or lack of), where and how close the mic is to you, etc. etc. etc. You aren't in the studio, for that matter, so if you want to match the recording as it was mic'd, then you shouldn't be there when calibrating, either.

A decent boom mic is about $20 and will get you much better results. If running Dirac Live, put it right at your ears for the crucial first ("sweet spot") measurement, then follow the directions for the other points. If it is just you, you can take fewer measurements or take all nine in a smaller volume. Taking more points with greater spacing, as on a long couch, will provide a better average for all but probably a little worse for the person in the sweet spot. But taking a few measurements means it can compensate for those peaks and valleys caused by small movements so I would not take just one (cue potato chip commercial).

Once you set the mic up, move away when you take the readings, making sure you are not in line of sight from the mic to any speaker. I usually sit on the floor with my notebook on a piece of cardboard (don't want carpet fuzz getting sucked into it!) between the back speakers to minimize the effect of my "being there". I used to run out of the room, but (a) half the time I didn't make it and had to start over, and (b) I found it made essentially zero difference in the results if I was just out of the way. Plus it took twice as long as I waited until I was sure one measurement set was done -- my room is isolated so it's hard to hear the tones from outside the room.

HTH - Don

Edit: Missed the other posts; sounds like Keith and I are in agreement, and you've got a plan, go for it!

[Morgan]

I actually ordered a mic stand for under $12 with shipping from Amazon with Prime, so NOT using a mic stand would be foolish.

Keith, regarding Dirac Live, back in Oct. of 2016 you posted "The $99 "upgrade" version of Dirac Live for Emotiva is going to give you exactly the same results as the one we give you for free IF YOU LEAVE EVERYTHING SET AT THE DEFAULTS and use our microphone. The difference you get for your $99 is more options and more things you can change to suit your preferences.". Has any updates to Dirac changed that, or is that still the case?

[Jack]

I just received my refurbished XMC-1 and know nothing about Dirac. Do you need a computer to use it? If so, will a Mac laptop work? Thanks!

[Deleted]

Feb 20, 2020 14:55:37 GMT -5 Morgan said:

Thanks Don, If the microphone measurements are indeed that accurate, then it makes sense. With other set ups, I simply put the microphone on top of my head and sat as still as possible during measurements. Has anybody ever tried placing a bunch of pillows, or similar in the chair?

Small request. Can you take a selfie when you have the mic ontop of your head and post the picture here?

[SticknStones]

Feb 22, 2020 14:17:57 GMT -5 Jack said:

I just received my refurbished XMC-1 and know nothing about Dirac. Do you need a computer to use it? If so, will a Mac laptop work? Thanks!

I would start here

You have to download Dirac from the Emotiva page, not Dirac. Hopefully, this is in your box and instructions otherwise you will need to call them on Monday to get your key. Good luck, Have fun, and congratulations from all of us on the lounge.

[Morgan]

Feb 22, 2020 14:22:47 GMT -5 @shimei said:

Feb 20, 2020 14:55:37 GMT -5 Morgan said:

Thanks Don, If the microphone measurements are indeed that accurate, then it makes sense. With other set ups, I simply put the microphone on top of my head and sat as still as possible during measurements. Has anybody ever tried placing a bunch of pillows, or similar in the chair?

Small request. Can you take a selfie when you have the mic ontop of your head and post the picture here?

Thanks, you've been very helpful!

[rogerlons]

I use a spare mic cable to get myself farther from the mic, and thus farther from any reflections that could be bothersome. It doesn't quite get me "out of the room", but it does get me another 15' away - which puts me so far from the speakers as to be nonexistent. Then stand very still.

If you don't happen to have a spare mic cable just kicking around, you can get one on Amazon for around $10.

[millst]

I'd probably argue that there is no point in correcting anything that changes greatly due to minute changes in microphone position. Are you going to calibrate your system for a 1ft cube and then lock your head in a vice after? The point of moving the microphone around during measurements is find the find the common correctable issues in your room/setup.

[donh50]

Feb 26, 2020 0:48:02 GMT -5 millst said:

I'd probably argue that there is no point in correcting anything that changes greatly due to minute changes in microphone position. Are you going to calibrate your system for a 1ft cube and then lock your head in a vice after? The point of moving the microphone around during measurements is find the find the common correctable issues in your room/setup.

Yup. Many folk, including Dr. Toole, believe correction should only be applied below the Schroeder Frequency (many references; here's one: www.linkwitzlab.com/rooms.htm ) or essentially only through the upper bass region in most rooms (perhaps 300~500 Hz). Direct sound usually dominates over that and, assuming you like your speakers, no point in correcting it.

IME/IMO - Don

[Deleted]

Get a USB extension cable.
Connect the Mic to the laptop, set-up the Mic, go to an adjacent room and let it run.

Same with Audyssey or Emo-Q, get an extension cable if needed and get out of the room.
Don't over-think it, they're the scientist you're not.