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

[wjp007]

To use Rat, you are supposed to copy/export the .ady file from your smart phone or tablet to your PC running Rat, and use Rat to open the .ady file and edit it. You seem to be doing exactly that except I don't understand what you meant by "When I pull that into Ratbuddyssey.."



If that's the edited file that you send back to the smartphone or tablet running MultEQ Editor App that the target curve looks normal if you have only edited the range from about 300 Hz down from what I can see.

Are those bookshelf speakers?

So what I did is open a text editor and added the room curve to the .ady file. I looked for the following string "customTargetCurvePoints" and added the room curve for both left and right (BTW, I'm using the 6dB Harman curve). As shown in this thread:

https://www.avsforum.com/threads/announcing-ratbuddyssey-a-tool-for-tweaking-audyssey-multeq-app-files.3006886/page-24

My speakers are bookshelf, but I am editing the full curve. Looking at other posts, I believe what is shown in the ipad is the reference curve setting. While I'm not sure, I believe when I set the AVR to flat, the high frequency doesn't get rolled off. I will run some test later today to verify with REW.

 

[peng]

So what I did is open a text editor and added the room curve to the .ady file. I looked for the following string "customTargetCurvePoints" and added the room curve for both left and right (BTW, I'm using the 6dB Harman curve). As shown in this thread:

https://www.avsforum.com/threads/announcing-ratbuddyssey-a-tool-for-tweaking-audyssey-multeq-app-files.3006886/page-24

My speakers are bookshelf, but I am editing the full curve. Looking at other posts, I believe what is shown in the ipad is the reference curve setting. While I'm not sure, I believe when I set the AVR to flat, the high frequency doesn't get rolled off. I will run some test later today to verify with REW.

That should work too, but I would just use Rat's table to enter the data. To me it is better because you can do the other stuff on one page. If you only want to edit the curve, just enter the cuts and boosts (avoid if possible), once done, save the file with a name of your choice, example: test1.ady and then send copy (or import it from the App) the file back to your smart device, open it and you should see the editor results.

I have posted one example before based on my last edited curve, but there are plenty posted on AVSF. Here's my play file I did recently:

 

[jhaider]

I have no idea what you're on about, but the topic of discussion was to compare Audyssey correction to the target curve. The only way to make it an apples to apples comparison is to use the same 8 measurement points and apply the same averaging and same smoothing as the target curve graphs.

Doing that tells you much less about what the system is actually doing than sampling properly, i.e. a spatial average covering the same area as the calibration samples but with randomly chosen sample points. Have you ever tried sampling the sound field properly? You can see my Marantz, Denon, and Bryston reviews for examples of doing it right.

This “must use same points” thing makes no sense at all.

 

[FrantzM]

Have you tried what's in my thread on Audioholics.com yet?

No

 

[wjp007]

Esteemed colleagues, I ran some tests comparing MultiEQ to Dirac and I'm thoroughly confused. I'm trying to decide between buying a full DIRAC license vs getting MultiEQ X. I did buy the ipad MultiEQ app.

I didn't have enough time today, so I didn't actually take multiple mic position measurements which I know is a problem. In any case, I thought I could learn something in the relative differences between the two if I had the mic in the same position for the tests.

For all of the measurements, I kept the mic position in the main center position I used for EQ. The room curve I am using is the Harman 6dB curve. For DIRAC setup I used a calibrated USB mic, for MultiEQ, I used the mic that came with the receiver. For the tests below, I used the same calibrated USB mic.

It's clear from below that DIRAC is producing the closest to the target room curve.
Orange - No EQ
Blue - MultiEQ reference setting
Purple - MultiEQ flat setting
Dark Red - DIRAC

I was trying to understand what the actual EQ that was being applied so I tried feeding the preamp output of my AVR into the MIC input of my sound card input and the curves below are what I measured. What the heck? MultiEQ doesn't seem to do anything below 200 Hz. And why is Flat called Flat, it's anything but Flat? It's more like kill the bass and enhance the treble, almost a reverse typical room response. Dirac is doing what it's supposed, and MultiEQ is kinda similar to Dirac above 1kHz. What could I be doing wrong in my setup? I checked my speaker config and I had them set to large with no crossover and no sub.

DIRAC - Dark Blue
MultiEQ Flat - Light Blue
MultiEQ Reference - Green

I'm going to keep testing until figure this out. I'll probably re-run the EQ setup and mark the mic locations so I can run the post EQ measurements more accurately.

 

[Reverend Slim]

I was trying to understand what the actual EQ that was being applied so I tried feeding the preamp output of my AVR into the MIC input of my sound card input and the curves below are what I measured. What the heck? MultiEQ doesn't seem to do anything below 200 Hz.

Very odd results. What AVR is this on? And are you sure you didn't have a curtain set somewhere in the workflow to limit equalization?

 

[FrantzM]

Sorry about my density and my lack of resolve ..
Let's suppose the following: I measure with the microphone at 8 positions and I obtain a curve/file .ady file. In that file I would have the Audyssey reference curve? Right? Since that is how Audyssey is set-up; of course I would remove the midrange dip, LFC is off, Dynamic Volume off, but Dynamic EQ is on. I would forcibly set the crossover at 80 Hz, rather than the 40 Hz , Audyssey keeps in recommending to me.. That is the file I have. Let's suppose I call it AUDISOUND_1.ady.

How do I apply an Harman/Toole reference curve to it? with ratbudyssey? .. So far I use my fingers or some kind of pen to "draw" a curve. on my iPad.

Thanks in advance

Peace.

 

[chych7]

I didn't have enough time today, so I didn't actually take multiple mic position measurements which I know is a problem. In any case, I thought I could learn something in the relative differences between the two if I had the mic in the same position for the tests.

I think there's your answer. This may not be worth looking into further until you do a proper calibration with multiple mic measurements. The room correction may not do much intentionally if it isn't given enough input measurements (which could cause overfitting/non-generalization of the fitted curve).

Also does your receiver support Audyssey XT32, or just XT?

 

[Chromatischism]

Doing that tells you much less about what the system is actually doing than sampling properly, i.e. a spatial average covering the same area as the calibration samples but with randomly chosen sample points.

If you choose random points then you are using data that you didn't give to Audyssey. Expect to see differences in the result.

 

[Chromatischism]

Also does your receiver support Audyssey XT32, or just XT?

This is key. It is XT32 that enables a high level of resolution in the bass region.

 

[Dj7675]

If you choose random points then you are using data that you didn't give to Audyssey. Expect to see differences in the result

My 2 cents... When testing whether an automated EQ is effective or not, you should be looking at 2 things (and not trying to measure the exact points measured by Audyssey).
1-Spatially averaged response of the measured listening area. Whether you do a bunch of single point measurements and average them, or whether you do an MMM measurement over the entire area. The goal is to provide good response of the area and should be measured as such IMO).
2-Also interesting would be to see what it does at each seat. A spatially averaged measurement around the head of each listening location to see how the system does with seat to seat variation.

 

[Chromatischism]

I can see the first point but not the 2nd - unless you measured those other seats and fed that data to Audyssey.

My point was that you should not expect to match Audyssey's predicted response if you're not measuring the same points.

 

[jhaider]

If you choose random points then you are using data that you didn't give to Audyssey.

So what? That's actually the damn point.

If the room correction system doesn't provide substantially the same result when you sample different points in the same area, then it's an unstable POS that nobody should ever use under any conditions.

Really, let’s apply some thought instead of talking points. What’s the point of a room correction system? Is it to change the response at specific chosen points? If so obviously measure at those points. However, if the point is to modify the response over an area, the way to test that is sample the area randomly.

 

[chych7]

Within the bubble of the measurement points, there's an expectation that the room correction algorithm would provide even response. Thus I would agree it makes sense to sample different points in the bubble to test how well the algorithm generalized. I'm planning on using the 9 point cube+MLP center Dirac tight focus for measurement (both for Dirac and Audyssey), and then verifying with REW at 5 points (all within the bubble): MLP, in front of MLP, left high of MLP, left low of MLP, right high of MLP, right low of MLP. Seat headrest is in the way for anything behind.

I'm also planning on making this a speakers only comparison, disabling the sub. The PC version of Dirac doesn't have access to the receiver/sub outs, so it wouldn't be fair vs. Audyssey. In any case I use a miniDSP/MSO for sub EQ. I'm more interested in the full range performance of Dirac vs. Audyssey, especially for tonality, soundstage, and imaging. On measurement metrics, I'm thinking:

• Tonality: how accurately the target curve is achieved in SPL vs. freq; also gives insight into correcting room modes for bass freqs
• Imaging: how well the left and right speakers are matched in SPL vs. freq and Phase vs. freq
• Soundstage: RT60 decay times (not really sure about this one though)

Any thoughts on how to measure for these factors? I'm also going to recruit my wife to do some blind A/B comparisons.

 

[Dj7675]

Within the bubble of the measurement points, there's an expectation that the room correction algorithm would provide even response. Thus I would agree it makes sense to sample different points in the bubble to test how well the algorithm generalized. I'm planning on using the 9 point cube+MLP center Dirac tight focus for measurement (both for Dirac and Audyssey), and then verifying with REW at 5 points (all within the bubble): MLP, in front of MLP, left high of MLP, left low of MLP, right high of MLP, right low of MLP. Seat headrest is in the way for anything behind.

I'm also planning on making this a speakers only comparison, disabling the sub. The PC version of Dirac doesn't have access to the receiver/sub outs, so it wouldn't be fair vs. Audyssey. In any case I use a miniDSP/MSO for sub EQ. I'm more interested in the full range performance of Dirac vs. Audyssey, especially for tonality, soundstage, and imaging. On measurement metrics, I'm thinking:

• Tonality: how accurately the target curve is achieved in SPL vs. freq; also gives insight into correcting room modes for bass freqs
• Imaging: how well the left and right speakers are matched in SPL vs. freq and Phase vs. freq
• Soundstage: RT60 decay times (not really sure about this one though)

Any thoughts on how to measure for these factors? I'm also going to recruit my wife to do some blind A/B comparisons.

Sounds like a fun project. A couple of other random thoughts..
-Instead of worrying about positions where ears are not, I would focus the points within the bubble around the ears where the listeners will be. That is what will be heard. It would be interesting to see what the EQ system does at other points but where the listeners ears are would seem to be of the most important. And I would consider doing a spatially averaged measurement in a small area where the listeners head would be instead of a single point.
-It may be beyond the scope of what you want to try to learn, but do a limited correction of both in case you like uncorrected better above say 500hz.

 

[Chromatischism]

So what? That's actually the damn point.

If the room correction system doesn't provide substantially the same result when you sample different points in the same area, then it's an unstable POS that nobody should ever use under any conditions.

Really, let’s apply some thought instead of talking points. What’s the point of a room correction system? Is it to change the response at specific chosen points? If so obviously measure at those points. However, if the point is to modify the response over an area, the way to test that is sample the area randomly.

We are obviously talking about different things.

Often people want to compare their results to the Audyssey predicted response. THAT is when you have to use the exact same measurement points. People usually don't, then complain that Audyssey is "making things up". Hence, the disclaimer on the correct procedure.

 

[thorvat]

We are obviously talking about different things.

Often people want to compare their results to the Audyssey predicted response. THAT is when you have to use the exact same measurement points. People usually don't, then complain that Audyssey is "making things up". Hence, the disclaimer on the correct procedure.

For the sake of following your logic let's assume here we are testing a room EQ system which is based on a single sweep. Would you, in that case, suggest we test how well it has done it's job using a single sweep from the same point or using a spatially averaged measurement over the area of the listening position?

 

[Fidji]

For the sake of following your logic let's assume here we are testing a room EQ system which is based on a single sweep. Would you, in that case, suggest we test how well it has done it's job using a single sweep from the same point or using a spatially averaged measurement over the area of the listening position?

TRINNOV is working that way. For RoomPerfect from Lyngdorf - you would need to run randomly around the room ....

 

[peng]

Sometimes it just takes one to post a response without trying to understand the point the original poster made to things in a convoluted back and forth, and I believe this is one of those case here.

If I understand Chromatischism correctly, he suggested using the same 8 points ref. post#413) that @chych7 might have used in his DL/Audyssey comparison when using REW to find out how well each performs in terms of actual results (averaged) versus the corresponding target curves.
Note: I assume while he said 8 positions, he meant the number of positions chych7 used to run DL and Audyssey so it was 6 then it would be the same 6 positions for his REW measurements.

So I assume he wanted to an apples to apples comparison if @chych7 is going to do a comparison of the results he got from using the two REQ software. I don't think he is trying to use such a comparison to establish one system is more effective than the other in general or overall. If he does, that would probably deserve a separate discussion or debate.

I hope @Chromatischism, will clarify that one more time to hopefully avoid people trying to argue the point(s) he might have never made in that one liner post#415.

Edit: Just realize I missed his post#456, confirming what I thought was the case.

 

[jhaider]

We are obviously talking about different things.

Often people want to compare their results to the Audyssey predicted response. THAT is when you have to use the exact same measurement points.

We’re talking about the same thing, and no.

People usually don't, then complain that Audyssey is "making things up". Hence, the disclaimer on the correct procedure.

The “correct procedure” is the one I’ve outlined above, not aping measure points.

If otherwise, Audyssey be an unstable POS. It is not, per my published measurements of two processors. Have you looked at that data, or tried doing things properly yourself? It shows the argument to make more effort for inferior data is not only dumb, but also unnecessary.

Where people get into trouble is running a cal and then trying to “confirm” the results with a single point measurement.

If I understand Chromatischism correctly, he suggested using the same 8 points ref. post#413) that @chych7 might have used in his DL/Audyssey comparison when using REW to find out how well each performs in terms of actual results (averaged) versus the corresponding target curves.

We all understand that incorrect position. There is no miscommunication. Have you tried taking confirmatory measurements properly - same area, random sample points? Have you read the Geddes and Blind sound power measurement paper?