Audyssey Room EQ Review

[John Galt]

...and here come the trolls ...

Who, exactly, are you referring to?

 

[Alice of Old Vincennes]

Who, exactly, are you referring to?

I was also wondering

 

[amirm]

I respect you immensely Amir, but would you consider a REW measurement ‘clean’ if I simply held the calibrated mic in my hand with associated body reflections?

Sure. Indeed what I showed was done that way!

The main domain to EQ are low frequencies and as I explained, the wavelengths are so long that your arm and body are not seen. It is like holding a hair in front of a light bulb. It doesn't block any of the light.

 

[John Galt]

Sure. Indeed what I showed was done that way!

The main domain to EQ are low frequencies and as I explained, the wavelengths are so long that your arm and body are not seen. It is like holding a hair in front of a light bulb. I

Sure. Indeed what I showed was done that way!

The main domain to EQ are low frequencies and as I explained, the wavelengths are so long that your arm and body are not seen. It is like holding a hair in front of a light bulb. It doesn't block any of the light.

Would this not affect higher frequencies? I thought the chart went into relatively high frequencies.

 

[amirm]

Would this not affect higher frequencies?

It can but no correction at high frequencies should be done with high resolution measurements. Our hearing ERB (discrimination bandwidth) is quite poor so what looks like comb filtering on a graph, is not audible that way. See: https://www.audiosciencereview.com/...ds/perceptual-effects-of-room-reflections.13/

In addition, if you just move the mic one inch the response changes a ton at high frequencies.

For these reasons, any proper EQ system heavily filters mid to high frequencies and as a result, nullifies any effect of your body.

 

[John Galt]

It can but no correction at high frequencies should be done with high resolution measurements. Our hearing ERB (discrimination bandwidth) is quite poor so what looks like comb filtering on a graph, is not audible that way. See: https://www.audiosciencereview.com/...ds/perceptual-effects-of-room-reflections.13/

In addition, if you just move the mic one inch the response changes a ton at high frequencies.

For these reasons, any proper EQ system heavily filters mid to high frequencies and as a result, nullifies any effect

It can but no correction at high frequencies should be done with high resolution measurements. Our hearing ERB (discrimination bandwidth) is quite poor so what looks like comb filtering on a graph, is not audible that way. See: https://www.audiosciencereview.com/...ds/perceptual-effects-of-room-reflections.13/

In addition, if you just move the mic one inch the response changes a ton at high frequencies.

For these reasons, any proper EQ system heavily filters mid to high frequencies and as a result, nullifies any effect of your body.

Would my body be less significant than the back of my couch, or wall? At what point would you think the measurements would be affected?

 

[SIY]

@amirm did you acquire the impulse response with and without? I did a pile of measurements on the Dirac system, and it would be interesting to see the contrast, if any.

 

[amirm]

@amirm did you acquire the impulse response with and without? I did a pile of measurements on the Dirac system, and it would be interesting to see the contrast, if any.

I do not. I have no use for it.

 

[Senior NEET Engineer]

@amirm did you acquire the impulse response with and without? I did a pile of measurements on the Dirac system, and it would be interesting to see the contrast, if any.

Here is one comparison: AVS Forum

 

[bobbooo]

Sure. Indeed what I showed was done that way!

The main domain to EQ are low frequencies and as I explained, the wavelengths are so long that your arm and body are not seen. It is like holding a hair in front of a light bulb. It doesn't block any of the light.

Holding the mic has another possible problem in that it likely won't be 100% stationary, which could affect the response (at all frequencies). I think it would have been better to just use the cardboard stand provided to make sure this isn't an issue, plus this is probably how the majority of consumers would take the measurements as most people don't have mic stands/tripods (and if they do they might not be compatible as you found out for yours), and would just follow the Denon instruction manual.

 

[Promit]

Call me contrary but I actually think Amir should have performed an elaborate dance while holding the microphone through each set of measurements. Surely this would be the optimal way to get the best possible measurements for Audyssey.

 

[amirm]

Holding the mic has another possible problem in that it likely won't be 100% stationary, which could affect the response (at all frequencies).

The whole point of multi-measurements as performed by Audyssey is to not care about any specific location let alone my hand moving a millimeter or two! They are spatially averaged with weighting. No small difference survives that transformatinon.

I already showed you proof of this by making my REW measurements at entirely different microphone location than Audyssey (and even a different microphone) yet its modal correction is exactly the same:

No calibration is useful anyway if it relies on your head being in a vice.

You all need to listen to me. Measurements can be a huge trap in acoustics. The proof is in the pudding anyway with post EQ sounding excellent. If what I did was wrong how did this happen?

 

[amirm]

Call me contrary but I actually think Amir should have performed an elaborate dance while holding the microphone through each set of measurements. Surely this would be the optimal way to get the best possible measurements for Audyssey.

 

[Music1969]

Measurements can be a huge trap in acoustics.

Are you referring to flat (measured) bass response not sounding good in a room?

Or other things also?

 

[amirm]

Are you referring to flat (measured) bass response not sounding good in a room?

Or other things also?

Well, that is just one thing. As soon as the wavelength of audio becomes close to width of your face, then one microphone is not measuring what two ears are hearing. Measurements can quickly become useless than way unless great care is taken. One of these days I will write the REW tutorial and explain these things.

 

[bobbooo]

The whole point of multi-measurements as performed by Audyssey is to not care about any specific location let alone my hand moving a millimeter or two! They are spatially averaged with weighting.

I already showed you proof of this by making my REW measurements at entirely different microphone location than Audyssey (and even a different microphone) yet its modal correction is exactly the same:

No calibration is useful anyway if it relies on your head being in a vice.

You all need to listen to me. Measurements can be a huge trap in acoustics. The proof is in the pudding anyway with post EQ sounding excellent. If what I did was wrong how did this happen?

I'm talking about (slight) movements of your hand, and so the mic, during each measurement, not between measurements. The former could affect the measured response. The result may have been good holding the mic, but it might be even better using the provided (guaranteed stationary) stand. And then there's the inevitable expectation bias that as you know can have an effect on perceived sound quality. I don't think anyone's saying what you did was wrong, just maybe not optimal, or how most users would measure their systems (and so not as representative of the results they would get).

 

[amirm]

I'm talking about (slight) movements of your hand, and so the mic, during each measurement, not between measurements.

That is exactly what I explained. The system absolutely does not care if you have the mic 5 inches from your ear or 5.1 inches. Indeed it makes sure it doesn't care by averaging measurements across many feet end to end.

The result may have been good holding the mic, but it might be even better using the provided (guaranteed stationary) stand.

Might? I am explaining that the system is designed to not care about such differences. If you don't believe me, then you are arguing contrary to how the system is working. Please read the paper on its design:
Objective Function for Automatic Multi-position Equalization and Bass Management Filter Selection
Sunil Bharitkar1, Chris Kyriakakis2
1Audyssey Laboratories, Inc., and University of Southern California, Los Angeles, CA 90064.
2Audyssey Laboratories, Inc., and University of Southern California, Los Angeles, CA 90064.

The spatial measurements are much farther apart than movement of my hand. By definition then their weighted sum will obliterate the micro mic movements.

If you don't believe this, then conduct your own test. Do the calibration with the mic stand and then move the mic 1/4 inch left or right and measurement again. Report what difference it made.

Until then, please, please don't use your intuition on this stuff. This is how those long FAQ threads on AVS went wrong.

 

[bobbooo]

That is exactly what I explained. The system absolutely does not care if you have the mic 5 inches from your ear or 5.1 inches. Indeed it makes sure it doesn't care by averaging measurements across many feet end to end.


Might? I am explaining that the system is designed to not care about such differences. If you don't believe me, then you are arguing contrary to how the system is working. Please read the paper on its design:
Objective Function for Automatic Multi-position Equalization and Bass Management Filter Selection
Sunil Bharitkar1, Chris Kyriakakis2
1Audyssey Laboratories, Inc., and University of Southern California, Los Angeles, CA 90064.
2Audyssey Laboratories, Inc., and University of Southern California, Los Angeles, CA 90064.

View attachment 59835

The spatial measurements are much farther apart than movement of my hand. By definition then their weighted sum will obliterate the micro mic movements.

If you don't believe this, then conduct your own test. Do the calibration with the mic stand and then move the mic 1/4 inch left or right and measurement again. Report what difference it made.

Until then, please, please don't use your intuition on this stuff. This is how those long FAQ threads on AVS went wrong.

I think you've misunderstood what I'm talking about. I'm talking specifically about slight accidental movements of your hand (and so the mic) positioning during each individual measurement i.e. during each chirp.

 

[amirm]

I think you've misunderstood what I'm talking about. I'm talking specifically about slight accidental movements of your hand (and so the mic) positioning during each individual measurement i.e. during each chirp.

And once more, that is exactly what I addressed in my posts. There is no misunderstanding on my part whatsoever.

 

[QMuse]

I respect you immensely Amir, but would you consider a REW measurement ‘clean’ if I simply held the calibrated mic in my hand with associated body reflections?

Before speculating further about body reflections in the context of measurements for room EQ you may want to check this video.