Hello,

The idea would be to implement what Thuneau does in its phase arbitrator plugin:
www.thuneau.com/arbitrator.htm
www.thuneau.com/Loudspeaker%20Phase%20Ar...r%20User%20Guide.pdf

This is also the method used by Grimm Audio (Hypec) for their LS1 loudspeaker:
www.grimmaudio.com/pro_loudspeakers_ls1.htm
www.grimmaudio.com/whitepapers/speakers.pdf

This plugin would let the openDRC correct the phase behavior of conventional loudspeakers (the better time aligned they are, and the most symmetrical their acoustical slopes, the better the result of course).

With this plugin, putting the openDRC in front of a conventional IIR digital crossover (such as a minidsp, a dcx2496, or any IIR crossover device) would turn it into a fully linear phase capable crossover (and also allow for a cool volume control, which is already a good thing on its own )

It looks like the most sensitive way to achieve linear phase: IIR slopes and FIR phase compensation.

With that many taps available it should be easy enough to implement the inverse allpass needed, even in the lowest frequency range (sealed box 12dB/oct bass rolloff phase compensation for example).

What do you think?

Hello,

We don't have any plans for such plug-ins but that doesn't mean that it's not an interesting one.

Using the OpenDRC (Custom IIR + Custom FIR), nothing prevents you from running an external software to compute the phase linearization. All the processing is there. We can't build a plug-in for every single application out there so the concept of the OpenDRC is to leave end users with the ability to build such custom solution (i.e. as you're looking for).

Hoping this info helps

DevTeam

I don't think that you would need any special plugin. The current FIR plugin should work well for phase linearization:)
It's more about FIR filter development. Target response (magnitude) will be allpass and the FIR filter will correct phase only. Did some tests with Crumboo's Scilab code and it works well!

thanks guys

I read Crumboo's threads, but it is a little bit over my head for now (will have to dig a little bit further...).
Where can I find Crumboo's current Scilab code? I did not find any link in his threads.

Hi, send me an email and I can send you the code. It's however not very well commented, so I hope you can figure out how it works. Also, you need to use measured data as input (a .txt file), and this data should be without excess phase (i.e. without the delay caused by distance between the speaker and microphone).

After some thoughts and work, here is a software for linearization impulse generation, as well as FIR filters and EQ...

sourceforge.net/projects/rephase/

To generate impulse files for the openDRC choose the 48000Hz sampling frequency, 6144 taps, and IEEE754 file format.

The documentation is lacking (to say the least...) but should improve with time.