Ongoing Beta Beta Testing: Independent L/R Room Correction with Subwoofer

[slartibartfast]

Looks more like Harman than B&K.

-Ed

It does indeed

 

[slartibartfast]

with rew ..filter a 3ms 330hz 0r 2-25.ms.. for see
..

I have no idea what that means

 

[canard]

I have no idea what that means

This is one of the main and essential functions of rew...

you need it... ;-)

Impulse Responses

 

[slartibartfast]

This is one of the main and essential functions of rew...

you need it... ;-)

Impulse Responses

I am measuring frequency response as you can clearly see.

 

[dominikz]

I think the Harman filter was generated using the previous beta L/R algorithm and the B&K was with the new algorithm.
View attachment 18000
View attachment 18001

At first glance the filters look like we should be seeing a significant difference around 20Hz (which we do) and about 3-5 dB difference between 150Hz and 250Hz (which we don't).
I assume this is the combined L+R measurement in both cases, right?
Do you by any chance have individual channel measurements in REW? To compare just the left channel response with B&K vs Harman filters loaded? Since the filters were calculated per-channel that might be informative.

 

[slartibartfast]

It does indeed
View attachment 18002

Strange thing is when I measured the response of the same B&K L/R filter on the day the new algorithm was introduced the result was different and more similar to a B&K curve. I feel like I am being gaslighted by the app.

 

[EddNog]

Strange thing is when I measured the response of the same B&K L/R filter on the day the new algorithm was introduced the result was different and more similar to a B&K curve. I feel like I am being gaslighted by the app.
View attachment 18013

This does look more like B&K.

-Ed

 

[slartibartfast]

This does look more like B&K.

-Ed

So it is a mystery what happened between 4 March and now. The same filter is acting differently.

 

[slartibartfast]

I performed two more room corrections. One Harman and one B&K. After saving saving them as Custom EQs and measuring again the results have changed again. The Blue and Green curves at the bottom are both B&K, one of them is the same EQ that previously looked like Harman. The Orange curve is a Harman generated with the old algorithm and the Red curve is Harman generated by the new algorithm. The new Harman curve is 2-3dB higher than the old one at low frequencies.
So somehow the act of generating new Custom EQs has corrected the way the other saved EQs perform.

 

[EddNog]

I performed two more room corrections. One Harman and one B&K. After saving saving them as Custom EQs and measuring again the results have changed again. The Blue and Green curves at the bottom are both B&K, one of them is the same EQ that previously looked like Harman. The Orange curve is a Harman generated with the old algorithm and the Red curve is Harman generated by the new algorithm. The new Harman curve is 2-3dB higher than the old one at low frequencies.
So somehow the act of generating new Custom EQs has corrected the way the other saved EQs perform.


View attachment 18015

The universe really works in mysterious ways sometimes. ¯\_(ツ)_/¯

-Ed

 

[dominikz]

I performed two more room corrections. One Harman and one B&K. After saving saving them as Custom EQs and measuring again the results have changed again. The Blue and Green curves at the bottom are both B&K, one of them is the same EQ that previously looked like Harman. The Orange curve is a Harman generated with the old algorithm and the Red curve is Harman generated by the new algorithm. The new Harman curve is 2-3dB higher than the old one at low frequencies.
So somehow the act of generating new Custom EQs has corrected the way the other saved EQs perform.


View attachment 18015

Good to see that now at least it seems to work as expected! Still probably worth opening a ticket regarding the issue you saw before.

BTW I personally wasn't able to reproduce the issue you had, sorry - running the new RC with different target curves provides the expected results, and my older manual individual channel PEQ presets seems to be loaded correctly as well:

Perhaps it is because in the last days I already used the new RC a few times for testing? No way to be sure...

 

[slartibartfast]

Good to see that now at least it seems to work as expected! Still probably worth opening a ticket regarding the issue you saw before.

BTW I personally wasn't able to reproduce the issue you had, sorry - running the new RC with different target curves provides the expected results, and my older manual individual channel PEQ presets seems to be loaded correctly as well:
View attachment 18020
Perhaps it is because in the last days I already used the new RC a few times for testing? No way to be sure...

I did open a ticket earlier. Mine was working yesterday morning, I posted in this thread showing the difference between the old and new B&K results. Somehow it stopped working later

 

[Sampetrini]

Filled out the Google form but don’t seem to have this feature on my latest version of the app.