[Completed Beta] Room Correction for Individual Channels

[slartibartfast]

It’s in beta stage so it’s algorithm will changes before public release.

Hmm we will see. If I do it manually I get similar results because I don't know the correct method to successfully achieve room correction by measuring individual channels.

 

[Smartplug]

Hmm we will see

Remember when stereo rc was in beta it had poor sound? Public release fixes those and most agree having rc made sound better than none.

 

[Smartplug]

What happen with auto timing/latency that supposed to get release?

 

[slartibartfast]

What happen with auto timing/latency that supposed to get release?

I was thinking the same.

 

[Smartplug]

@WiiM Support @WiiM Team What happen with auto timing/latency that supposed to get released? Thanks!

 

[slartibartfast]

Comparing stereo to L/R RC here are the results of L/R




and here are the results of Stereo

 

[MCG555]

Like I suspected Room Correction using individual channels produce too much bass. I only tried with a subwoofer. First two results are from individual channels and third is stereo.
View attachment 11564

View attachment 11565

View attachment 11566

Thanks for sharing!

 

[slartibartfast]

Thanks for sharing! There must be clearly an error in L/R as there is no correlation between curve and written value.
E.g. it shows +12dB at 41Hz in the curve but +1.5 dB in the table.
In stereo they do correlate.

It does show +8.6dB at 43Hz though. Still doesn't quite add up.

 

[MCG555]

Comparing stereo to L/R RC here are the results of L/R

View attachment 11568
View attachment 11569

and here are the results of Stereo

View attachment 11570
View attachment 11571

As I thought. The modes together in stereo add much more weight than in individual setup.

 

[MCG555]

It does show +8.6dB at 43Hz though. Still doesn't quite add up.

True, saw it afterwards and corrected/deleted it. Sorry. I saw this behavior before and think it‘s correct. It relates to the „sharp“ Q factor.

 

[slartibartfast]

I haven't seen many reports of success or failure using this feature. Is anybody else trying it?

 

[slartibartfast]

As I thought. The modes together in stereo add much more weight than in individual setup.

What I don't understand is the 15dB dip between 50 and 100Hz with the individual channels measurements. The Subwoofer crossover is set to 85Hz.

 

[Wiimer]

What I don't understand is the 15dB dip between 50 and 100Hz with the individual channels measurements. The Subwoofer crossover is set to 85Hz.

This is a silly question, is there test audio coming from the sub when measuring individual channels?

 

[slartibartfast]

This is a silly question, is there test audio coming from the sub when measuring individual channels?

It's not a stupid question, I thought exactly the same. I could feel the speaker moving but whether it was moving as much as it should I am not sure. It didn't sound as though it was.
I sent feedback and WiiM say they saw something in the logs so let's see what happens next.

 

[slartibartfast]

This is a silly question, is there test audio coming from the sub when measuring individual channels?

After I changed the sub crossover I seemed to get more sensible results. Results also seem better if EQ is manually turned off before performing the Room Correction. This is with 80Hz sub crossover and 40-4000Hz correction range.

 

[Maurizio]

Hi to all
I'm not signed up for the Beta Test for the per-channel EQ, but I'd like to give my opinion. I tried to do the per-channel EQ before the Beta was even mentioned.
How? Easy:
1) All balance on the left, measurement, curve saved as Auto L
2) Same procedure with the right channel, saved as Auto R
Listening using not Stereo, but L+R
Here, in my opinion, is the result of the test compared to the Stereo EQ.
In the Stereo RC the bass becomes too cut
In the RC per channel, with my method, the bass is not cut. It is very similar to listening without RC, but it has gained definition.
The quantity or presence of bass, as I said, is very similar to the uncorrected sound, but it has become less boomy. The bass has solidified, it is more solid and incisive without any boomy effect.
This is not however, always in my opinion, the only benefit. All the notes and instruments have gained clarity and perceptibility. For example, the cymbals, especially when struck with brushes, are more perceptible.
They stand out clearly and are not mixed up or overwhelmed by the other instruments without becoming too present or sparkling.
Channel separation and soundstage, width and depth, also benefited.
If I remember correctly, @Wiimer too tried my method and came to the same conclusion as me.
I'm curious and eager to try, when it's final, the per-channel RC that WiiM will implement to see if there will be any differences.

 

[slartibartfast]

Hi to all
I'm not signed up for the Beta Test for the per-channel EQ, but I'd like to give my opinion. I tried to do the per-channel EQ before the Beta was even mentioned.
How? Easy:
1) All balance on the left, measurement, curve saved as Auto L
2) Same procedure with the right channel, saved as Auto R
Listening using not Stereo, but L+R
Here, in my opinion, is the result of the test compared to the Stereo EQ.
In the Stereo RC the bass becomes too cut
In the RC per channel, with my method, the bass is not cut. It is very similar to listening without RC, but it has gained definition.
The quantity or presence of bass, as I said, is very similar to the uncorrected sound, but it has become less boomy. The bass has solidified, it is more solid and incisive without any boomy effect.
This is not however, always in my opinion, the only benefit. All the notes and instruments have gained clarity and perceptibility. For example, the cymbals, especially when struck with brushes, are more perceptible.
They stand out clearly and are not mixed up or overwhelmed by the other instruments without becoming too present or sparkling.
Channel separation and soundstage, width and depth, also benefited.
If I remember correctly, @Wiimer too tried my method and came to the same conclusion as me.
I'm curious and eager to try, when it's final, the per-channel RC that WiiM will implement to see if there will be any differences.

You would be an ideal person to try the beta. Go on you know you want to. It is easy to return to official firmware.

 

[sNc]

What should be done with subwoofer if I use it? Left / right should be done together with sub?

 

[slartibartfast]

What should be done with subwoofer if I use it? Left / right should be done together with sub?

I think you would have to include the subwoofer.

 

[ArthurStereo]

Hi to all
I'm not signed up for the Beta Test for the per-channel EQ, but I'd like to give my opinion. I tried to do the per-channel EQ before the Beta was even mentioned.
How? Easy:
1) All balance on the left, measurement, curve saved as Auto L
2) Same procedure with the right channel, saved as Auto R
Listening using not Stereo, but L+R
Here, in my opinion, is the result of the test compared to the Stereo EQ.
In the Stereo RC the bass becomes too cut
In the RC per channel, with my method, the bass is not cut. It is very similar to listening without RC, but it has gained definition.
The quantity or presence of bass, as I said, is very similar to the uncorrected sound, but it has become less boomy. The bass has solidified, it is more solid and incisive without any boomy effect.
This is not however, always in my opinion, the only benefit. All the notes and instruments have gained clarity and perceptibility. For example, the cymbals, especially when struck with brushes, are more perceptible.
They stand out clearly and are not mixed up or overwhelmed by the other instruments without becoming too present or sparkling.
Channel separation and soundstage, width and depth, also benefited.
If I remember correctly, @Wiimer too tried my method and came to the same conclusion as me.
I'm curious and eager to try, when it's final, the per-channel RC that WiiM will implement to see if there will be any differences.

Thank you for your work I have a question, do you use target curve?