Post your room modes correction PEQ settings

[daftcombo]

Hi,

Which settings do you use to deal with room modes?
Where is it effective: listening position only or wider area?
Do you use different settings for each channel?
Are the corrections different depending on the type of music and loudness?

 

[Absolute]

I use single point measurement of the modal EQ for my 25, 43 and 62 hz peaks and MMM measurements for absolute level EQ between peaks and dips and overall tonality.

Each channel separately to as smooth as they get, then combined with MMM to see if there's some interference between the channels I need to get rid of or use the sub to fill in

 

[daftcombo]

I use single point measurement of the modal EQ for my 25, 43 and 62 hz peaks and MMM measurements for absolute level EQ between peaks and dips and overall tonality.

Each channel separately to as smooth as they get, then combined with MMM to see if there's some interference between the channels I need to get rid of or use the sub to fill in

Thanks. Which gain & Q for those frequencies?

 

[Wayne A. Pflughaupt]

Thanks. Which gain & Q for those frequencies?

His settings won't be the same as what your room might need.

Regards,
Wayne A. Pflughaupt

 

[daftcombo]

His settings won't be the same as what your room might need.

Regards,
Wayne A. Pflughaupt

Haha yes, of course. I just want examples of values:

For me at the moment:

Focal Aria 906 in a 3,5m * 4,5m * 3m room, with the left speaker in a corner:

Speaker correction:
800 Hz, -3 dB, Q (proportional) = 4

Room correction:
42 Hz, -4,3 dB, Q = 2,7 (left speaker only)
81 Hz, -11 dB, Q = 5,7 (left speaker only)
110 Hz, -7 dB, Q= 10

 

[ernestcarl]

Which settings do you use to deal with room modes?
Where is it effective: listening position only or wider area?
Do you use different settings for each channel?
Are the corrections different depending on the type of music and loudness?

My answers to those questions will differ a little depending on which setup I'm using.

When generating and adjusting PEQs, I don't focus on just the modes, but rather the overall measured response. I use the moving microphone method (MMM) -- using the default calibration file with the microphone oriented generally towards the front mains speaker being measured. Approximately 60x60x30cm (LxWxH) is the cubic area I use for my single-seat office desk setup. I also take a few regular log sweeps.

Separate PEQs are created for each channel, as well as shared filters for channel volume level adjustments and for the 'house curve' EQ.

Corrections are universal and do not change depending on the type of music. However, I use JRiver's automatic internal loudness DSP which is set to begin at my relative internal reference volume level 80% (0.0dB) i.e. -10dBFS in the DAC. When the internal volume setting is at or above that set relative internal volume reference level, loudness DSP correction is bypassed.

I also prefer to always have JRiver's Adaptive Volume correction enabled (some people don't like it).

Notably, all DSP corrections -- but not xo and channel routing 'room correction' -- will be manually bypassed during raw measurements as well as volume level calibration.

Visualization purpose only! I don't have the PEQs entered in the miniDSP for this MCH setup.

DSP is all done internally inside JRiver to take advantage of its 64bit processing.

Filters and their Q listed (Right channel has same number of filters to Left)

The Internal Reference Volume Level doesn't have to be set at 100% (it's relative after all)

Just a rough visual preview of how loudness DSP compensation looks/works using pink noise. BTW, it's quite rare that I find myself disabling the loudness DSP as I've set it quite well enough that it just works almost perfectly with all media types.

Volume Level Normalization

Adaptive Volume Normalization

Initial subwoofer volume level is set high to fill in huge null between 50 to 90Hz -- there's also a gigantic null below ~25Hz so a high pass is applied at 23Hz even though the sub is capable of going down much deeper.

Sub's under the table (best compromise)! and can serve as a "leg shaker"

BTW, I'm under no illusion that those foam acoustic wedges do anything to mitigate room modes & nodes.

 

[daftcombo]

My answers to those questions will differ a little depending on which setup I'm using.

When generating and adjusting PEQs, I don't focus on just the modes, but rather the overall measured response. I use the moving microphone method (MMM) -- using the default calibration file with the microphone oriented generally towards the front mains speaker being measured. Approximately 60x60x30cm (LxWxH) is the cubic area I use for my single-seat office desk setup. I also take a few regular log sweeps.

Separate PEQs are created for each channel, as well as shared filters for channel volume level adjustments and for the 'house curve' EQ.

Corrections are universal and do not change depending on the type of music. However, I use JRiver's automatic internal loudness DSP which is set to begin at my relative internal reference volume level 80% (0.0dB) i.e. -10dBFS in the DAC. When the internal volume setting is at or above that set relative internal volume reference level, loudness DSP correction is bypassed.

I also prefer to always have JRiver's Adaptive Volume correction enabled (some people don't like it).

Notably, all DSP corrections -- but not xo and channel routing 'room correction' -- will be manually bypassed during raw measurements as well as volume level calibration.

View attachment 73520
Visualization purpose only! I don't have the PEQs entered in the miniDSP for this MCH setup.

View attachment 73521
DSP is all done internally inside JRiver to take advantage of its 64bit processing.

View attachment 73535
Filters and their Q listed (Right channel has same number of filters to Left)

View attachment 73523
The Internal Reference Volume Level doesn't have to be set at 100% (it's relative after all)

View attachment 73525
Just a rough visual preview of how loudness DSP compensation looks/works using pink noise. BTW, it's quite rare that I find myself disabling the loudness DSP as I've set it quite well enough that it just works almost perfectly with all media types.

View attachment 73531
Volume Level Normalization

View attachment 73532
Adaptive Volume Normalization

View attachment 73526
Initial subwoofer volume level is set high to fill in huge null between 50 to 90Hz -- there's also a gigantic null below ~25Hz so a high pass is applied at 23Hz even though the sub is capable of going down much deeper.

View attachment 73528
Sub's under the table (best compromise)! and can serve as a "leg shaker"

BTW, I'm under no illusion that those foam acoustic wedges do anything to mitigate room modes & nodes.

Thank you for the extensive explanation and EQ settings.

I think you are over-thinking the thing. You have a LOT of EQ settings, with some that contradict each other, for example:

could be replaced by 73 Hz, Gain = 5.0, Q = 2.0

and many others.

But if you like the sound, all good

 

[ernestcarl]

Thank you for the extensive explanation and EQ settings.

I think you are over-thinking the thing. You have a LOT of EQ settings, with some that contradict each other, for example:

View attachment 73553

could be replaced by 73 Hz, Gain = 5.0, Q = 2.0

and many others.

But if you like the sound, all good

Definitely can be refined! I have to admit, with the extremely narrow curves at one area in particular. hehe! t’was hastily redone from REW generated EQ filters. Even with MMM, doing five channels manually was testing my patience — I thought about re-adjusting each channel individually yet again prior — but laziness decided that was already good-enough.

Life would be easier with something like Dirac! Although you do submit even more of your freedom... Manually inputing each filter and adjusting here and there feels a bit old now.

 

[daftcombo]

My correction is not perfect neither. It seems like it misses a bass boost between 20 Hz and 200 Hz and a low shelf high shelf starting at 4 kHz as @amirm suggests in this Audyssey review:

source: https://www.audiosciencereview.com/forum/index.php?threads/audyssey-room-eq-review.12746/

 

[ernestcarl]

It seems like it misses a bass boost between 20 Hz and 200 Hz and a low shelf starting at 4 kHz

High shelf starting at 4kHz, you mean, I think...

As long as HF is not fatiguing over time and at higher volume levels, shelving may not be necessary. Before when I had the KH120 in a different room (and house) and where it was situated much further away from the MLP, I actually applied a 1-2dB boost there. Current near-field setup at less than a meter(!) makes that type of boost too fatiguing.

 

[daftcombo]

High shelf starting at 4kHz, you mean, I think...

Yes of course, thank you.

 

[daftcombo]

I just tried the Audissey curve. Sounds great but I don't think the Aria 906 can handle it. Will need a sub for sure!


Measurements from listening position (one measurement, not MMM) before any EQ:

Right speaker:

Left speaker:

Now the curve applied is in blue:

and I added the room correction PEQ directly in JRiver:

(Sorry Q don't show)

Anyone else tried it?
https://www.audiosciencereview.com/...des-correction-peq-settings.14716/post-457623

 

[Wayne A. Pflughaupt]

Filters and their Q listed (Right channel has same number of filters to Left)

You have 7 filters there on the left channel in a space of about 1.1 octave. That’s like – wow. Many people won’t use more than that for their entire frequency range! Looking at the list I can see that some of them are essentially cancelling each other out.

Here’s a good read, a case-study on full-range EQ:
Spridle’s Experiment

You might also want to wade through my epic “Minimal EQ” article.

You can model the filters in REW and see what they are doing. I’ll try to find time in the next day or two to do that and post some graphs.

Regards,
Wayne A. Pflughaupt

 

[Absolute]

Thanks. Which gain & Q for those frequencies?

Ah, too many to remember! About 5 peqs below 200 hz per speaker, I think.
Due to the M2 needing quite a lot of PEQ-bands to be flat, I have limited amount of bands left for EQ. Still working with finding the optimal settings.

Here's the preliminary MMM result from LP;

 

[ernestcarl]

You have 7 filters there on the left channel in a space of about 1.1 octave. That’s like – wow. Many people won’t use more than that for their entire frequency range! Looking at the list I can see that some of them are essentially cancelling each other out.

Here’s a good read, a case-study on full-range EQ:
Spridle’s Experiment

You might also want to wade through my epic “Minimal EQ” article.

You can model the filters in REW and see what they are doing. I’ll try to find time in the next day or two to do that and post some graphs.

Regards,
Wayne A. Pflughaupt

Hi, Wayne! Thanks for the feedback. The 7 PEQs for the left channel were generated using REW's equalizer module. Will read your link and see what can be improved. Oks!

 

[ernestcarl]

I just tried the Audissey curve. Sounds great but I don't think the Aria 906 can handle it. Will need a sub for sure!


Measurements from listening position (one measurement, not MMM) before any EQ:

Right speaker:
View attachment 73642


Left speaker:
View attachment 73640


Now the curve applied is in blue:
View attachment 73633



and I added the room correction PEQ direclty in JRiver:

View attachment 73636

Anyone else tried it?
https://www.audiosciencereview.com/...des-correction-peq-settings.14716/post-457623

The Aria is still a small bookshelf speaker, so it would struggle with that kind of bass boost at higher volume levels. At moderate or lower listening levels it probably would be able to handle your boost and bass extension fine.

I've deliberately set my xo higher at 120Hz with the KH120s since it audibly improves the clarity of the mids when listening loud. Yeah, it's preferrable to let the sub do the heavy lifting, esp. in the sub bass.

 

[daftcombo]

The Aria is still a small bookshelf speaker, so it would struggle with that kind of bass boost at higher volume levels. At moderate or lower listening levels it probably would be able to handle your boost and bass extension fine.

I've deliberately set my xo higher at 120Hz with the KH120s since it audibly improves the clarity of the mids when listening loud. Yeah, it's preferrable to let the sub do the heavy lifting, esp. in the sub bass.

I am not even sure a big bass boost like that is useful since I keep seing @amirm perfom his subjective listening tests of bookshelf speakers with a room mode correction in the form of a notch and no bass boost / treble shelf and enjoy it like that!

 

[daftcombo]

I guess the answer to my own question is here:
https://www.audiosciencereview.com/...-is-your-favorite-house-curve.2382/post-85237
Thanks @Cosmik !

 

[dasdoing]

I use linear phase filter which corrects magnitude (red) and phase (green)

 

[ernestcarl]

I guess the answer to my own question is here:
https://www.audiosciencereview.com/...-is-your-favorite-house-curve.2382/post-85237
Thanks @Cosmik !

I don't view the bass boost as a "correction" that has to be forcibly applied, but more like a preference template to test and try out for yourself in your room(s) and with your speakers.