MMM approach and a new calibration app (magic beans)

[Keith_W]

Yes. I am on the beta program. What would you like to know?

 

[GXAlan]

Yes. I am on the beta program. What would you like to know?

Does it work for you?

 

[Keith_W]

Does it work for you?

The answer is rather complex. The best brief answer I can give is: the concept is sound, the correction works as intended ... but this solution does not work for me.

I posted the result of my experiment in this post where I demonstrated @joentell's method indeed does correct the nearfield MMM to flat. However, it also produces a rising frequency response at the MLP. This is the opposite behaviour of any other speaker, where flat nearfield = falling Harman-like frequency response at the MLP. The reason why is because of the particular characteristic of my speaker, it has horns > 500Hz and a conventional woofer and subs below. Another experiment described in this post explains why my speakers behave this way.

BTW, when I did those experiments, I was not on the beta program. I was able to replicate the method based on the description of the concept. I posted a workflow for Acourate users earlier in the thread. The app automates the process a little and produces the target curve for you, but you do not need the app if you pay attention to the method and think about how to implement it in whatever DSP software you are using. A friend of mine got it working with Audiolense, but the problem with Audiolense is that it can not import target curves. Nor can it manipulate curves like you can in Acourate and REW. So he had to derive his target curve in REW, and then manually create a target curve in Audiolense using its target curve editor.

I still think this method is brilliant but it is a shame that my speaker characteristics prevents me from implementing it. If I get new speakers I will be sure to try it again.

 

[joentell]

The answer is rather complex. The best brief answer I can give is: the concept is sound, the correction works as intended ... but this solution does not work for me.

I posted the result of my experiment in this post where I demonstrated @joentell's method indeed does correct the nearfield MMM to flat. However, it also produces a rising frequency response at the MLP. This is the opposite behaviour of any other speaker, where flat nearfield = falling Harman-like frequency response at the MLP. The reason why is because of the particular characteristic of my speaker, it has horns > 500Hz and a conventional woofer and subs below. Another experiment described in this post explains why my speakers behave this way.

BTW, when I did those experiments, I was not on the beta program. I was able to replicate the method based on the description of the concept. I posted a workflow for Acourate users earlier in the thread. The app automates the process a little and produces the target curve for you, but you do not need the app if you pay attention to the method and think about how to implement it in whatever DSP software you are using. A friend of mine got it working with Audiolense, but the problem with Audiolense is that it can not import target curves. Nor can it manipulate curves like you can in Acourate and REW. So he had to derive his target curve in REW, and then manually create a target curve in Audiolense using its target curve editor.

I still think this method is brilliant but it is a shame that my speaker characteristics prevents me from implementing it. If I get new speakers I will be sure to try it again.

Get those KEF Blades 2 Metas we joked about.

The directivity on those horns are crazy. You did a lot to fix the mismatch with your crossover changes though.

 

[Stefan H.]

Yes. I am on the beta program. What would you like to know?

Hi Keith,
What is the speaker configuration?
7.2.4?
What Receiver do you use?
Audyssey or Dirac?
Is the imersiv sound bubble better than before?
KR Stefan

 

[joentell]

Hi Keith,
What is the speaker configuration?
7.2.4?
What Receiver do you use?
Audyssey or Dirac?
Is the imersiv sound bubble better than before?
KR Stefan

@Keith_W has a fully active 2.2 channel setup with lots of DSP and an atypical loudspeaker. It's one of the reasons I asked him to participate in the private beta testing.

 

[Keith_W]

Hi Keith,
What is the speaker configuration?
7.2.4?
What Receiver do you use?
Audyssey or Dirac?
Is the imersiv sound bubble better than before?
KR Stefan

Take a look at this thread for details about the system. I don't think the Magic Beans correction will change the nature of your immersive sound bubble, that is an issue of speaker timing and positioning. It will change the tonality to be more accurate, though.

 

[joentell]

Take a look at this thread for details about the system. I don't think the Magic Beans correction will change the nature of your immersive sound bubble, that is an issue of speaker timing and positioning. It will change the tonality to be more accurate, though.

I would agree that timing and levels has more to do with immersion. Generally speaking, correcting speakers to have the same tonal balance can help with immersion if they are dissimilar. For example, if a constant sound pans around to other speakers and the tonality changes, it can take away from the experience. Tonal changes can be caused by different speakers being used, a speaker's distance from the listening position, and proximity to nearby surfaces or boundaries.