TOPIC:

Dear All,

While releasing the upcoming DDRC-88A, we're receiving similar tech support tickets about how to customize more than 1 sub channel? (e.g. Atmost setup)
We thought that we'd summarize the info here in this FAQ.

minidsp.desk.com/customer/portal/article...8a-and-multiple-subs

Hoping this clarifies.

DevTeam

Hi devteam,

Will the software equalize each sub separately or are they treated as one?
Why does a channel need to be specifically marked as "subwoofer"? Is there a difference in how a sub is measured or equalized or both?

Thank you!

One clarification: Atmos is height channels.

I'd be really curious to see field results of Dirac multisub EQ. This application could be killer if it is capable of eq'ing each one separately in order to optimize them as a single system.

Since you are setting up a channel specifically for a sub, does that mean that Dirac will set level and delay for each sub and then measure the combined subs as one? I don't see how it could work if it does not work this way, because if calibrated individually, the combined response would be different. Please advise. If it does this, then I think I am ready to order.

Keep in mind that both Dr. Geddes' software and Harman SFM measure the subs individually and optimize the in room response (using slightly different optimization criteria) based on those measurements. To my knowledge, nether one requires a combined measurement. Though I could be wrong about that. Markus?

If Dirac can be deployed similarly, that would be really cool!

DS21 wrote:

Keep in mind that both Dr. Geddes' software and Harman SFM measure the subs individually and optimize the in room response (using slightly different optimization criteria) based on those measurements. To my knowledge, nether one requires a combined measurement. Though I could be wrong about that. Markus?

If Dirac can be deployed similarly, that would be really cool!

Correct, optimizations based on Geddes and SFM use the complex sum of single sub responses. Not sure if Dirac is capable of doing that. Let's see what information devteam has for us.

Dear All,

Some good questions/brainstorming going on here..

When it comes to Dirac Live, the sub checkbox tells DLCT to treat that channel slightly differently. It will have a different sweep and a different delay detection algorithm, due to its low frequency. It does as well apply an extra 10 dB alignment gain in the filters since LFE source channel requires that from the equipment. (If we would not apply this extra gain, we would counteract the equipment's intentional addition of 10 dB.)
Each sub is treated separately though so the depending on your setup, it might actually be beneficial to use 1 x SUB in, 1 x SUB out (split to multiple subs). That's a guess for now on.. Early 2015 (post parties and big dinners), it's best that we figure out some time for our technical writer to put together some extensive app note on that interesting topic and include measurements. I'm also looking forward to hearing the experience from the community.. Each system being different, I'm sure that we'll all learn from it.

DevTeam

Thanks devteam.

So Dirac takes care of speaker delay and it doesn't have to be set upstream. Is this also true for gain?
Could the 10dB boost be made optional? Most people will probably use bass management upstream hence the 10dB LFE boost would have been already applied. Or am I just misinterpreting what the "extra 10 dB alignment gain in the filters" does?

@ Markus,

Sorry for the delay, just back at office today after our holiday break.
Yes, indeed it will take care of delay/level (gain). The optional boost is indeed a good trick. Let us talk to the Dirac team to see if that's possible trick for the future GUI since that would be part of their setup. We'll report shortly when we hear back from them. We're both getting ready for CES show (miniDSP & Dirac) so appreciate your patience as we work things out.

Post XMAS, a little brainstorming later, one trick we were thinking would be to include some level of matrixing post Dirac in our DSP structure. Combined with some delay, one could: Use one Dirac filter block doing overall Dirac filtering of the LFE signal (i.e. for both speakers) AND do time alignment inside for each dedicated sub (most typical cause of comb filtering issue). The DDRC-88A indeed has more power than the nanoAVR so we do have more flexibility. Just a thought needing a bit more brainstorming/feedback from real world experience at this time.

DevTeam


DevTeam

devteam wrote:

@ Markus,

Sorry for the delay, just back at office today after our holiday break.
Yes, indeed it will take care of delay/level (gain). The optional boost is indeed a good trick. Let us talk to the Dirac team to see if that's possible trick for the future GUI since that would be part of their setup. We'll report shortly when we hear back from them. We're both getting ready for CES show (miniDSP & Dirac) so appreciate your patience as we work things out.

Thanks for taking this into consideration. Most people probably use the DDRC-88A with an AVR and bass management, so if Dirac boosts the subwoofer channel, this bass managed subwoofer channel would be 10dB too hot.

devteam wrote:

Post XMAS, a little brainstorming later, one trick we were thinking would be to include some level of matrixing post Dirac in our DSP structure. Combined with some delay, one could: Use one Dirac filter block doing overall Dirac filtering of the LFE signal (i.e. for both speakers) AND do time alignment inside for each dedicated sub (most typical cause of comb filtering issue). The DDRC-88A indeed has more power than the nanoAVR so we do have more flexibility. Just a thought needing a bit more brainstorming/feedback from real world experience at this time.

DevTeam


DevTeam

Yes, an additional delay (and gain) stage after Dirac filtering would be very desirable. Something like this:

Markus,

Diagram is similar to what we had in mind by using a Matrix mixer so you can grow your sub count to however many you want. (i.e. combine tick box)

DevTeam

Just resurrecting this thread and wondered if things had moved along any further with the matrix idea you put forward?

I've read the FAQ on the first post but I'm still confused. Does it EQ all subs as one or separately?

If you define single subwoofer channels in the DDRC-88A they will be equalized separately. I would recommend connecting all subs to a single channel though. Then equalize the whole sub cluster as one.

Skylinestar wrote:

I've read the FAQ on the first post but I'm still confused. Does it EQ all subs as one or separately?

It, quite properly, EQs all subs as one, which is how it should be done (so that the EQ takes account of the subs' interaction with the room and also with each other, which is what happens when you are listening to them).

There is only one subwoofer output on the 88A so if you have multiple subs you have to be a little creative if they are at different distances from MLP, or (not usually a great idea) different electrically or acoustically. (If the subs are equidistant and identical then a simple Y-cord is all you need). You will need to set levels and distances (delays) individually for each sub and then EQ them as one. To do this, the simplest way seems to be to use a miniDSP 2x4. Fortunately, AVS member AustinJerry has written a definitive Guide on how to accomplish this and you can download it by clicking here .