Article: Calibrating Desktop Speakers using Focus Fidelity Filter Designer

[dathzo]

18 hours ago, mitchco said:

Hi @dathzo Having spent more time with both DL and FF there are two differences that have come to light.

 

Dirac uses IIR filters at low frequencies. IIR filters cannot address the non-minimum phase behaviour at low frequencies that exists in virtually every room. This requires excess phase correction, which means a FIR filter. While Dirac does use FIR filtering, it is not at low frequencies. One can verify this by using a FIR filter designer to open up a Dirac correction filter, extract the excess phase response and observe no excess phase correction at low frequencies.

 

The other issue is the result of Dirac's averaging algorithm. More measurements for the calculation means a loss of correction resolution at the listening position. In this AES paper https://www.aes.org/e-lib/browse.cfm?elib=15154 it was noted via listener preferences that "measurements more localized to the primary listening seat correspond better with listeners’ subjective ratings, particularly below 300 Hz." My measurements (and listening tests) comparing Dirac with another DSP software that has full FIR filtering shows this as well: https://audiophilestyle.com/forums/topic/58546-article-dirac-live-2-digital-room-correction-software-walkthrough/?do=findComment&comment=1098275

 

Based on my tests, both measurements and listening, Focus Fidelity does not have these issues. 

Thanks a lot @mitchco! 

 

1. A follow-up question: how can you open the Dirac Live filters? The only file I see is the *.liveproject that I am not sure can be opened with an FIR filter designer software? Would be great when I can :)

 

2. Regarding the paper, as you said, it mentions "measurements more localized to the primary listening seat correspond better with listeners’ subjective ratings, particularly below 300 Hz". However, they still are doing averaging of 6 measurements, only that in one case (I below), the measured points are more scattered around the room than in case II (measurements around main listening position). I understand that Dirac has the option of a Tightly-focused image, which is similar to situation II. One can go even further and omit some of the measurements (or even measure the main listening position more than once) to get to a correction that gets closer to the preferred setting in Olive's paper? In that sense, I think the averaging "issue" may be avoided, but certainly not the lack of FIR filters at low frequencies...

 

 

[Focus Fidelity]

Hi @dathzo With the Focus Fidelity software you can select how many measurement points you'd like to use in a project between 1 and 10. You can cluster the points tightly to cover 1 seat of wider spacing to cover a sofa for example depending on your application. So there is room for experimentation.

 

[mitchco]

@dathzo the method of capturing a Dirac filter is in the post I linked. My complaint is not the number of measurements, but the averaging algorithm used. Again, looking at the link in my previous answer to you, one can see in the comparison chart a loss of correction resolution from 100 Hz to 800 Hz in the Dirac correction. And looks to be over-correcting at 90 Hz. As I said in my previous post, David's FFD is using a different algorithm for multiple measurements and does not have these issues. Cheers!

[dathzo]

1 hour ago, mitchco said:

@dathzo the method of capturing a Dirac filter is in the post I linked. My complaint is not the number of measurements, but the averaging algorithm used. Again, looking at the link in my previous answer to you, one can see in the comparison chart a loss of correction resolution from 100 Hz to 800 Hz in the Dirac correction. And looks to be over-correcting at 90 Hz. As I said in my previous post, David's FFD is using a different algorithm for multiple measurements and does not have these issues. Cheers!

Thanks @mitchco! Still learning and very far from your knowledge, so apologies if my questions seem trivial to you. I’m not familiar with the loopback sweep through jriver and hence the question.

 

indeed I see what you mean with the resolution and over correction, that at least it’s clear.

 

thanks again.-

[dathzo]

21 hours ago, Focus Fidelity said:

Hi @dathzo With the Focus Fidelity software you can select how many measurement points you'd like to use in a project between 1 and 10. You can cluster the points tightly to cover 1 seat of wider spacing to cover a sofa for example depending on your application. So there is room for experimentation.

 

Thanks a lot @Focus Fidelity. I have used Impala and it’s great for measurements. Extremely intuitive with a modern and fresh layout. Well done! 

[Markus8]

@Focus Fidelityhas just released a new version of Impala and Focus Fidelity Designer with clock drift correction feature. More at https://www.focusfidelity.com/clockdriftcorrection

Glad to see this evolving and looking forward to hearing feedback of others. 

[Confused]

11 hours ago, Markus8 said:

@Focus Fidelityhas just released a new version of Impala and Focus Fidelity Designer with clock drift correction feature. More at https://www.focusfidelity.com/clockdriftcorrection

Glad to see this evolving and looking forward to hearing feedback of others. 

I certainly plan to try it.

 

One problem is that I have a new sofa on order, so unless I find myself with lots of free time (which is unlikely at the moment) I will probably wait until my room is reconfigured slightly with new seating before remeasuring.

 

It is great to see updates like this, I look forward to experiencing the results!

[dathzo]

Hello David @Focus Fidelity.

Happy to see you keep investing resources on developing your software. Any news on the "digital crossovers, bass management and multiple subwoofer optimization.", as mentioned in the review?

You are closer to 2023 than most of us, so happy new year! :)

[Focus Fidelity]

Hi @dathzo Happy New Year to you as well. The current plan is some updates to the existing software followed by a Mac version. Crossovers etc will follow after that, I don't have a timeframe for those features right now.

[Confused]

For those that might be interested, Positive Feedback Review of Focus Fidelity Filter Designer:

(and the associated Impala measurement module)

 

https://positive-feedback.com/reviews/hardware-reviews/focus-fidelity-filter-designer/

 

[Confused]

@mitchco What do you think of the (how shall I put this?) slightly eccentric target curve used in the above linked review?

[Focus Fidelity]

Hi @Confused  The target curve was based on the Harman target but with the bump around 600Hz added to preserve the "character" of the speakers being used.

[Markus8]

@mitchco There have been some considerable new features and improvements to Impala & Focus Fidelity Designer over the recent years. 
Will you do an update of your review?

[mitchco]

@Markus8 Yup, I have been in touch with David. Great to see FFD on Mac! I hope to update the review, but currently don't have the time or an ETA. Will circle back when I do.

[Markus8]

5 hours ago, mitchco said:

@Markus8 Yup, I have been in touch with David. Great to see FFD on Mac! I hope to update the review, but currently don't have the time or an ETA. Will circle back when I do.

Thanks - keeping fingers crossed 😀