Audyssey's Next Generation of Room Correction (MultEQ-X)

[GalZohar]

I don't understand why you would trchnically need to use feedback from REW into the Audyssey target curve. If your measurements are not flat post calibration, it is either because of a difference in how you measured, or because of something Audyssey decided to not correct. Using REW as described seems to me like something that you shouldn't be doing, and if it gives you any improvement then you should have probably measured differently when running the Audyssey calibration.

If feedback would be beneficial, then your method would have a measureable benefit even for a single point calibration. Maybe try that and see what you get.

 

[abdo123]

@HarmonicTHD

If i understood you correctly, when you add a filter to the target curve section of the calibration process you're actually adding the filter on the input (before the room correction filters) right?

I was going through the documentation earlier today and that's what i kind of understood alongside your comments.

 

[chych7]

I don't understand why you would trchnically need to use feedback from REW into the Audyssey target curve. If your measurements are not flat post calibration, it is either because of a difference in how you measured, or because of something Audyssey decided to not correct. Using REW as described seems to me like something that you shouldn't be doing, and if it gives you any improvement then you should have probably measured differently when running the Audyssey calibration.

If feedback would be beneficial, then your method would have a measureable benefit even for a single point calibration. Maybe try that and see what you get.

From all the Audyssey calibrations I've run in my system, it never seems to really flatten/meet the desired target curve, as measured by UMIK-1/REW. So the way I see it is this post calibration gives an opportunity to fine tune the correction beyond what Audyssey does by default. I've measured using the same locations as the Audyssey setup pattern, so that's not the reason. I think it boils down to (1) something not accurate with the Audyssey mic I have, or (2) Audyssey is not aggressive for mid-high freq corrections, which may be a good thing for most speakers. I have good directivity speakers (Philharmonic BMR), so I think I can EQ them more without running into issues.

I did try this at a single point calibration and it worked as intended (see the plots in my previous post).

 

[tjcinnamon]

From all the Audyssey calibrations I've run in my system, it never seems to really flatten/meet the desired target curve, as measured by UMIK-1/REW. So the way I see it is this post calibration gives an opportunity to fine tune the correction beyond what Audyssey does by default. I've measured using the same locations as the Audyssey setup pattern, so that's not the reason. I think it boils down to (1) something not accurate with the Audyssey mic I have, or (2) Audyssey is not aggressive for mid-high freq corrections, which may be a good thing for most speakers. I have good directivity speakers (Philharmonic BMR), so I think I can EQ them more without running into issues.

I did try this at a single point calibration and it worked as intended (see the plots in my previous post).

2 things:

It’s an “average” of multiple measurements (more complex than that but conceptually the same) so measuring at a single position will not fully align with the curve.

Many people don’t correct over 500Hz because a that point youre not correcting for the room; youre correcting for the speakers at that point.

So I don’t think the mic is massive issue. In Gene’s audioholics interview, they cover the mic tolerances.

 

[chych7]

2 things:

It’s an “average” of multiple measurements (more complex than that but conceptually the same) so measuring at a single position will not fully align with the curve.

Many people don’t correct over 500Hz because a that point youre not correcting for the room; youre correcting for the speakers at that point.

So I don’t think the mic is massive issue. In Gene’s audioholics interview, they cover the mic tolerances.

Yes I know of the averaging. I've measured the post Audyssey response with multiple measurements within the setup pattern, and it still shows the behavior of not achieving the target curve. I've played with below/above 500 Hz correction and I personally find the audio sounds better when Audyssey is run full range. L/R speaker matching is improved full range as well (confirmed with multiple averaged measurements of Audyssey on vs. off), which improves imaging. In my room there are interactions with the room and also furniture that clearly happens > 500 Hz, which shows up as dips in the response (if I measure nearfield in front of the speaker, they aren't there). For example, my seat's headrest is one source of issues. Thus, I think there could be some benefit in applying additional corrections.

Also there was a member here who collected and compared multiple Audyssey mics and found significant deviations among them, so I wouldn't 100% trust the mic either.

 

[HarmonicTHD]

@HarmonicTHD

If i understood you correctly, when you add a filter to the target curve section of the calibration process you're actually adding the filter on the input (before the room correction filters) right?

I was going through the documentation earlier today and that's what i kind of understood alongside your comments.

Almost. MultiEQX does not distinguish between filters derived from REW for room correction or filters one might define for a house / preference curve.

In my view there are two possibilities now with the most recent MultiEQX update.

A) use the automatic room measurements and define a preference curve in MultiEQX (eg -1dB roll off or whatever one fancies). This is how it has always worked.

B) New alternative approach for those not wanting to use the automatic measurement but want to use REW instead. Roughly speaking one keeps Audyssey off, then do your REW sweeps, define the preference curve in REW and calculate the filters (in REW). Export and import the REW filters into MultiEQX (without having defined any filters in the Target curve menu. The imported REW filters will show up as individual Biquad filters inside the Target curve MultiEQX menu. No additional Preference curve definition needed in MultiEQX as this is already considered by the REW filter calc.). Then write the filters to the AVR. (and cross check the result with another REW sweep if so desired).

Probably needs some screenshots to better document and avoid confusion but that is all I was able to pull together for the time being.

 

[GalZohar]

From all the Audyssey calibrations I've run in my system, it never seems to really flatten/meet the desired target curve, as measured by UMIK-1/REW. So the way I see it is this post calibration gives an opportunity to fine tune the correction beyond what Audyssey does by default. I've measured using the same locations as the Audyssey setup pattern, so that's not the reason. I think it boils down to (1) something not accurate with the Audyssey mic I have, or (2) Audyssey is not aggressive for mid-high freq corrections, which may be a good thing for most speakers. I have good directivity speakers (Philharmonic BMR), so I think I can EQ them more without running into issues.

I did try this at a single point calibration and it worked as intended (see the plots in my previous post).

Did you try this with a single point calibration though? That is the only easy way to eliminate the effect on averaging methods. If Audyssey decided not to fix something due to variations between the different positions, it'll appear as if it missed something that could be corrected when in fact it was intentional. Unless you rule out the measurement locations and averaging methodology, you don't really have any proof to your claim, which I find quite odd - If Audyssey couldn't correct something the first time around, why would adding a peak/dip in the target curve make it correct it? At least according to how Audyssey is expected to work, getting that extra correction with modifying the target curve will only correct things that Audyssey could correct, but decided not to, and you set the target curve to bypass Audyssey's decisions. If I'm right, then it's probably just better to revise your mic positions when running Audyssey calibrations. If you have the app you can probably try various combinations of positions relatively easily, and see the effect.

I don't see where you posted results that were specifically titled as single point Audyssey calibration compared to a single-point REW sweep?
Also you'd need to make sure the difference isn't purely because of a difference in the microphones or their positions even with just a single point.

 

[MarkI]

The top "before" graph is what happens when you put real distances into the AVR, and that's what I expect you to be getting now. (A bit better, because your speaker distance spread is smaller - he gets a 800us spread, while you'd have 600us).

If you put real distances multiplied by 0.875 into the AVR (or MultEQ-X does it for you because you give it real distances), I would expect to see results like the bottom graph. (The 100us grid lines represent match the resolution of the adjustment, so bottom graph is pretty much as good as it gets - all speakers within one step).

Hello KMO,

Sorry I didn’t respond a few weeks back. I got caught up but wanted to get back and thank you for taking the time with your in-depth and very accurate explanation.

I did spend some time with the new calibrated mic to verify the time alignment that MultEQ-X is landing in my Marantz AV. The results are below. The first one is using the front left as the default input and the second is the right front speaker as the default input. As far as I can tell and probably why so many people have noticed an immediate and better result with how their system sounds it’s because MultEQ-x is getting the time alignment damn near perfect.

 

[chych7]

Did you try this with a single point calibration though? That is the only easy way to eliminate the effect on averaging methods. If Audyssey decided not to fix something due to variations between the different positions, it'll appear as if it missed something that could be corrected when in fact it was intentional. Unless you rule out the measurement locations and averaging methodology, you don't really have any proof to your claim, which I find quite odd - If Audyssey couldn't correct something the first time around, why would adding a peak/dip in the target curve make it correct it? At least according to how Audyssey is expected to work, getting that extra correction with modifying the target curve will only correct things that Audyssey could correct, but decided not to, and you set the target curve to bypass Audyssey's decisions. If I'm right, then it's probably just better to revise your mic positions when running Audyssey calibrations. If you have the app you can probably try various combinations of positions relatively easily, and see the effect.

I don't see where you posted results that were specifically titled as single point Audyssey calibration compared to a single-point REW sweep?
Also you'd need to make sure the difference isn't purely because of a difference in the microphones or their positions even with just a single point.

Tough crowd here. Alright I just did a single point calibration experiment, Audyssey on/off and with/without REW correction. Front speakers only, subwoofer disabled. Receiver is Denon X3600, speakers are Philharmonic BMR v2s.

To start with, this is what the Audyssey mic looks like at the MLP position in my setup (the mics are pointing up, the camera angle shows it tilted). The UMIK-1 is right next to it for comparison (but not present during calibration). When I'm taking REW measurements, the UMIK-1 is placed confidently within 1 inch of the Audyssey mic position, and the Audyssey mic is removed from the setup (tripod remains for the sake of keeping things as consistent as possible).

Ok, so on to the measurements. Here I am looking at overall smoothness of the response and how well the left/right channel match (which impacts imaging). I tried to quantify the average left/right SPL error with linear and log weighting (the log weighting will weight bass mismatch more).

First we have the Audyssey off measurement, showing typical room effects and significant bass frequency mismatch:

Next we have the Audyssey measurement, showing great improvement of bass modes and significant improvements to left/right channel matching. However, strangely there is a high frequency rise and some peaks/valleys in the mid/high freq response. In the MultEQ-X "Design Target Curve" tab, I have "Theater high frequency roll-off 1" enabled, so it's very strange I'm getting this rise. I've run probably a couple dozen calibrations with different configurations and speakers, and I've always seen this post measurement. I suspect my Audyssey microphone is not well calibrated. I have ordered a calibrated Audyssey microphone so I can determine if that's the issue (will check this weekend). I do not believe my UMIK-1 is miscalibrated; I've tested it against a different UMIK-1 and they matched well (and I am using the 90 deg cal file).

Below is the Audyssey correction after putting in filters from REW. The target curve I used here was a flat response for a "bass limited speaker" with a 6 dB/oct roll-off at 30 Hz. This provides additional smoothening of the response, and some more improvements in left/right matching for higher frequencies.

So my conclusions are that Audyssey is (1) trying to achieve some strange non-flat target curve, which may be a microphone problem, and (2) there is some potential correction left on the table that Audyssey could correct, but doesn't.

 

[jhaider]

You’re placing the mics right in front of a large reflective headrest?

 

[chych7]

You’re placing the mics right in front of a large reflective headrest?

Yes, that would capture how I actually listen to my system with the headrest behind my head...

There's some difference; green is above the headrest, blue is as I showed it with the headrest behind it. This is without Audyssey, right channel. Hard to tell what's due to the headrest and what's due to the position change.

 

[Webninja]

Yes, that would capture how I actually listen to my system with the headrest behind my head...

There's some difference; green is above the headrest, blue is as I showed it with the headrest behind it. This is without Audyssey, right channel. Hard to tell what's due to the headrest and what's due to the position change.

View attachment 211668

Most of you have far more experience with Audyssey and I don’t have multiEQ-X (yet), but I’ve read as much as I could on mic position and I have read the mic needs to capture the bounce from the rear wall.

I recline my couch all the way when doing Audyssey or REW. Maybe try that, could be the cause for what you’re seeing.

 

[chych7]

Most of you have far more experience with Audyssey and I don’t have multiEQ-X (yet), but I’ve read as much as I could on mic position and I have read the mic needs to capture the bounce from the rear wall.

I recline my couch all the way when doing Audyssey or REW. Maybe try that, could be the cause for what you’re seeing.

The rear wall bounce still is captured, the headrest isn't 100% blocking everything (in fact it's probably doing very little on bass freqs, which is where most of the room correction is needed). If I had to wager, I'd guess that the 1-2 kHz dip is probably due to the headrest. On a more fundamental level, shouldn't the calibration be done in a way that best captures the way you listen to the system and where your ears are situated, including effects of furniture? Room correction also applies to furniture, IMO.

 

[KMO]

I did spend some time with the new calibrated mic to verify the time alignment that MultEQ-X is landing in my Marantz AV. The results are below.

I'd say that's almost too perfect, and you're just seeing impulses that REW has aligned - are you sure the timing reference was enabled so it doesn't do that? I'd check by trying to get it to show something non-aligned, if it hasn't already. (Don't need to change any settings, just try the mic in a different position).

 

[Chromatischism]

Next we have the Audyssey measurement, showing great improvement of bass modes and significant improvements to left/right channel matching. However, strangely there is a high frequency rise and some peaks/valleys in the mid/high freq response.

Forgive me if you've provided this before but I can't find it; what speakers are these again? I'd suggest you're still on the brighter side unless that's near-er field or it's for lower-volume listening, then it's a great result. . But it has to be a product of the speakers.

Tough crowd here.

Thank you for taking the time to walk us all through your process. It's much faster to ask questions than it is to answer them comprehensively.

 

[abdo123]

You’re placing the mics right in front of a large reflective headrest?

Your head is going to be infront of a large reflective headrest so what’s the point of this comment?

 

[chych7]

Forgive me if you've provided this before but I can't find it; what speakers are these again? I'd suggest you're still on the brighter side unless that's near-er field or it's for lower-volume listening, then it's a great result. . But it has to be a product of the speakers.


Thank you for taking the time to walk us all through your process. It's much faster to ask questions than it is to answer them comprehensively.

Speakers are Philharmonic BMR v2, they are quite flat if I measure near field and they generally measure well (recently measured with Klippel on Erin’s audio corner). I usually put a -1 dB/dec tilt on my target curve for Audyssey, otherwise the post corrected result is on the brighter side. I didn’t apply the tilt for this particular experiment.

 

[Webninja]

The rear wall bounce still is captured, the headrest isn't 100% blocking everything (in fact it's probably doing very little on bass freqs, which is where most of the room correction is needed). If I had to wager, I'd guess that the 1-2 kHz dip is probably due to the headrest. On a more fundamental level, shouldn't the calibration be done in a way that best captures the way you listen to the system and where your ears are situated, including effects of furniture? Room correction also applies to furniture, IMO.

Good point, I’m going to try it both ways this weekend to see what REW shows for my room.

 

[Chromatischism]

Speakers are Philharmonic BMR v2, they are quite flat if I measure near field and they generally measure well (recently measured with Klippel on Erin’s audio corner). I usually put a -1 dB/dec tilt on my target curve for Audyssey, otherwise the post corrected result is on the brighter side. I didn’t apply the tilt for this particular experiment.

I'm not sure how to explain the curve in that case. I'm using controlled directivity speakers, but not narrow (Buchardt S400 MKII, +/- 50-60 degrees) and I have what'd you call a textbook tilt, so maybe it has something to do with that ribbon. Now, that is without Audyssey. I normally don't tell it to go full-range. I'll have to experiment with that when I get MEQX and see what it does.

 

[jhaider]

Your head is going to be infront of a large reflective headrest so what’s the point of this comment?

Two ears and a brain work differently than a microphone.