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

[Acerun]

Yeah, they just told me to send it back for a replacement. In their words "that's really weird!"

 

[pmar4015]

I would add to this: I do my first three positions at center of MLP, where my left ear is and where my right ear is... and then on the last step of MultEQ-X, I select all three of those positions for the Trim source so that it averages based on those three measurements. Gives me more consistent channel trim results almost every time when checking post-cal with a meter.

Wow. What a good idea

 

[Acerun]

Any reason not to disable the cutoff mode in Audessey MultiEQ - X if using high quality subwoofers? What's the best setting there for 2 subs that go down to 14 on paper?

 

[Reverend Slim]

Any reason not to disable the cutoff mode in Audessey MultiEQ - X if using high quality subwoofers? What's the best setting there for 2 subs that go down to 14 on paper?

Look at the measured rolloff and then change the EQ limit on the low end to make it generally line up to it. Shouldn't be an issue. Just try different values on the filter screen until the predicted rolloff lines up at about the same slope as the natural rolloff.

Also, in case anyone hasn't noticed, the 1.6 update just released and now allows import of target curves from text files. So instead of importing filter data, you can just import the curve you want it to correct to. I played around with importing Harman, Harman trained listener, steady state response, Olive-Welti, and a few other curves last night, and it works better than importing filters.

 

[Acerun]

Look at the measured rolloff and then change the EQ limit on the low end to make it generally line up to it. Shouldn't be an issue. Just try different values on the filter screen until the predicted rolloff lines up at about the same slope as the natural rolloff.

Also, in case anyone hasn't noticed, the 1.6 update just released and now allows import of target curves from text files. So instead of importing filter data, you can just import the curve you want it to correct to. I played around with importing Harman, Harman trained listener, steady state response, Olive-Welti, and a few other curves last night, and it works better than importing filters.

So use Auto and set the low roll off? Is that what you're saying? I'm not at home but here's the offline screenshot showing basically flat to 20...Although the company swears that this sub goes down to 14. Maybe there is usuable bass down to 14, I don't know how to interpret whether the drop off lower than 20 is still usable or not.

 

[pmar4015]

Any reason not to disable the cutoff mode in Audessey MultiEQ - X if using high quality subwoofers? What's the best setting there for 2 subs that go down to 14 on paper?

I dont see any harm in disabling the cut-off mode for the subwoofers, found in the ‘Design Target Curves” tab, provided you limit the eq filters (specifically if any filters provide for any boosting) to the detected cut-off frequency (you don't want to apply any boost to a sub below its natural roll-off.

Note: The eq filters parameters/limits are found in the “Filters Setting” tab

 

[pmar4015]

So use Auto and set the low roll off? Is that what you're saying? I'm not at home but here's the offline screenshot showing basically flat to 20...Although the company swears that this sub goes down to 14. Maybe there is usuable bass down to 14, I don't know how to interpret whether the drop off lower than 20 is still usable or not.

View attachment 298617

I would go with what MultEQ-X has detected……just because your sub may go down to 14Hz anechoically doesnt mean your room will support such

 

[HarmonicTHD]

Look at the measured rolloff and then change the EQ limit on the low end to make it generally line up to it. Shouldn't be an issue. Just try different values on the filter screen until the predicted rolloff lines up at about the same slope as the natural rolloff.

Also, in case anyone hasn't noticed, the 1.6 update just released and now allows import of target curves from text files. So instead of importing filter data, you can just import the curve you want it to correct to. I played around with importing Harman, Harman trained listener, steady state response, Olive-Welti, and a few other curves last night, and it works better than importing filters.

I also noticed the update but didn’t get around to check it out.

Would you mind sharing the text files for the Harman curves here or provide a link?

So far I have used filters for their approximation.

 

[Reverend Slim]

So use Auto and set the low roll off? Is that what you're saying? I'm not at home but here's the offline screenshot showing basically flat to 20...Although the company swears that this sub goes down to 14. Maybe there is usuable bass down to 14, I don't know how to interpret whether the drop off lower than 20 is still usable or not.

Based on that measurement, I would just leave it as it is. You're still getting some usable sound below 20Hz, and the predicted rolloff is slightly better, so it's still correcting a bit in that region. Should be fine. Depending on the sub, it likely has its own limiter to prevent over-excursion anyway.

 

[pmar4015]

I also noticed the update but didn’t get around to check it out.

Would you mind sharing the text files for the Harman curves here or provide a link?

So far I have used filters for their approximation.

Harmon Curve attached

Harmon.txt

drive.google.com

Additional curves attached

A collection of speaker target responses in .csv/.txt format

In order to use REW's automatic EQ feature, I started collecting and converting every speaker target that I could find into the necessary .txt format. In this thread I want to share my work so far, as well as ask you guys for target responses that I may have missed. Comparison: So...

urldefense.com

 

[HarmonicTHD]

Harmon Curve attached

Harmon.txt

drive.google.com

Additional curves attached

A collection of speaker target responses in .csv/.txt format

In order to use REW's automatic EQ feature, I started collecting and converting every speaker target that I could find into the necessary .txt format. In this thread I want to share my work so far, as well as ask you guys for target responses that I may have missed. Comparison: So...

urldefense.com

Thanks appreciate it.

(Typo. Harman from Harman Kardon)

 

[Acerun]

Based on that measurement, I would just leave it as it is. You're still getting some usable sound below 20Hz, and the predicted rolloff is slightly better, so it's still correcting a bit in that region. Should be fine. Depending on the sub, it likely has its own limiter to prevent over-excursion anyway.

There is a switch for a limiter recommend for HT use. Recommend off for music. I don't play my system very loud.

 

[Acerun]

Based on that measurement, I would just leave it as it is. You're still getting some usable sound below 20Hz, and the predicted rolloff is slightly better, so it's still correcting a bit in that region. Should be fine. Depending on the sub, it likely has its own limiter to prevent over-excursion anyway.

When I switched to manual it looks even better...looks flat to 14.

 

[pmar4015]

Ok but just to be clear it appears your result is due to almost 8 db of boost (in terms of its peak) from ~14Hz to ~20Hz. You are OK driving your sub that hard (or having to rely on the limiter)? I agree with Reverend Slim…..I would leave it at its default. Why push your sub so hard when the audible range for a human doesn’t begin until 20Hz? Seems like a lot of wasted power/potential distortion for very little gain (understanding, in my view, the grossly overhyped benefit of subsonic base……which you are already benefiting from to a certain degree if you leave the setting at its default)).

 

[Reverend Slim]

When I switched to manual it looks even better...looks flat to 14.

But I would bet good money that if you actually measured the post-cal result down to 14Hz, you're not actually getting that response. Audyssey's predicted response can be a bit... optimistic. Especially where narrow dips that may well be nulls are concerned, in which case you may be dumping power into something that you may not hear in practice, but that would resonate like mad.

That said, if you like the way it sounds and aren't putting your subs in danger with crazy levels, roll with it! We can nitpick things to death, but all that matters at the end of the day is if you like the way it sounds.

 

[Ruspamen]

Hello, i read that audyssey applies a maximum boost of 9db. I wanted to know if there was a way to limit the amount of boost applied to filters. Thank you

 

[pmar4015]

Hello, i read that audyssey applies a maximum boost of 9db. I wanted to know if there was a way to limit the amount of boost applied to filters. Thank you

Not that I am aware of. Headroom expansion allows the user to go above+ 9db (with a corresponding cap) but I am not aware of a way to limit the correction within the range of +0db - +9db

 

[Lbstyling]

So.

I bought multi eq-X when it released about a year ago.

Having designed speakers and such over the years, I wanted:

-More control from Audessey so I could play around with it with my active DIY speakers.

-Better quality data. Accurate graphs without having to use REW to confirm.

-Control over the EQ cut off frequencies to try 500hz cut off with my L+R speakers that have very good off axis control already (KEF meta)

- Bass extension below the silly 30hz cut off safety limit impose by the android app.

When I got it, it was clearly an improvement over the basic app.

However....

I have to say the latest 1.5 update has blown my mind! It is another world better with the 9db EQ limit now removed for the bass.

I have a 18inch JBL 2242h sub that can slam, but i used to keep having to add 4 to 6 db to the bass shelf, yet it never hit all the notes. Now..... I'm finding detail in bass on all kinds of tracks. It's a joy to listen to all the different bass notes on music tracks.

This latest update is what audessey should have always been!

It is more fiddly to get right vs DIRAC, but given dynamic EQ, it's now better than DIRAC for me.

 

[OCA]

So.

I bought multi eq-X when it released about a year ago.

Having designed speakers and such over the years, I wanted:

-More control from Audessey so I could play around with it with my active DIY speakers.

-Better quality data. Accurate graphs without having to use REW to confirm.

-Control over the EQ cut off frequencies to try 500hz cut off with my L+R speakers that have very good off axis control already (KEF meta)

- Bass extension below the silly 30hz cut off safety limit impose by the android app.

When I got it, it was clearly an improvement over the basic app.

However....

I have to say the latest 1.5 update has blown my mind! It is another world better with the 9db EQ limit now removed for the bass.

I have a 18inch JBL 2242h sub that can slam, but i used to keep having to add 4 to 6 db to the bass shelf, yet it never hit all the notes. Now..... I'm finding detail in bass on all kinds of tracks. It's a joy to listen to all the different bass notes on music tracks.

This latest update is what audessey should have always been!

Did they add an option to turn off extra surround/height speaker boost of Dynamic EQ?

 

[LeoB]

No, they did not. They told me not long ago that it's up to manufacturer to implement this feature in the AVR firmware.