G3P Elevated BM & LFE Issue

[marcl]

Reposting here also what I posted in the Dirac thread ....

I got the 3.3.3 Dirac and did a full calibration yesterday, 13 points. I basically used my same target curves I've been using for a few months to get a flat final response as measured in REW. Since our FW 2.5 (and the Beta we got a few months before) the BM bug was mostly fixed, so my small speakers should only measure about +3db in the bass rather than the +8 we got before. This is with Center Sub=LFE and BM going to the full range fronts.

Well ... I measured the result with the 3.3.3 and it looks like the BM bug is back!! The small speaker bass is elevated again! Here's a comparison of my last full cal with DL 3.2.3 and yesterday's full cal with DL 3.3.3. Screen shot from REW attached comparing Dirac 3.2.3 to 3.3.3.

Crazy! So I went back into Dirac and reloaded the project and double checked the target curves - especially to be sure their new bass boost curves were not applied. I rebooted the processor. I did a Backup Settings > Factory Reset > cold reboot > Restore Settings. Same result. With the full cal using DL 3.3.3, the BM bug is back!

And I verified there is no issue when using User EQ … it’s only with Dirac.

[ttocs]

May 19, 2022 10:53:47 GMT -5 marcl said:

Reposting here also what I posted in the Dirac thread ....

I got the 3.3.3 Dirac and did a full calibration yesterday, 13 points. I basically used my same target curves I've been using for a few months to get a flat final response as measured in REW. Since our FW 2.5 (and the Beta we got a few months before) the BM bug was mostly fixed, so my small speakers should only measure about +3db in the bass rather than the +8 we got before. This is with Center Sub=LFE and BM going to the full range fronts.

Well ... I measured the result with the 3.3.3 and it looks like the BM bug is back!! The small speaker bass is elevated again! Here's a comparison of my last full cal with DL 3.2.3 and yesterday's full cal with DL 3.3.3. Screen shot from REW attached comparing Dirac 3.2.3 to 3.3.3.

Crazy! So I went back into Dirac and reloaded the project and double checked the target curves - especially to be sure their new bass boost curves were not applied. I rebooted the processor. I did a Backup Settings > Factory Reset > cold reboot > Restore Settings. Same result. With the full cal using DL 3.3.3, the BM bug is back!

And I verified there is no issue when using User EQ … it’s only with Dirac.

<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button>

I can't help but wonder if some BM is sneaking into the Center Sub LFE output. Those BM traces for the Small channels look more like the LFE trace than the Left or Right traces. Can you disconnect the Center Sub Output and see what REW hears?

[marcl]

May 19, 2022 11:56:21 GMT -5 ttocs said:

May 19, 2022 10:53:47 GMT -5 marcl said:

Reposting here also what I posted in the Dirac thread ....

I got the 3.3.3 Dirac and did a full calibration yesterday, 13 points. I basically used my same target curves I've been using for a few months to get a flat final response as measured in REW. Since our FW 2.5 (and the Beta we got a few months before) the BM bug was mostly fixed, so my small speakers should only measure about +3db in the bass rather than the +8 we got before. This is with Center Sub=LFE and BM going to the full range fronts.

Well ... I measured the result with the 3.3.3 and it looks like the BM bug is back!! The small speaker bass is elevated again! Here's a comparison of my last full cal with DL 3.2.3 and yesterday's full cal with DL 3.3.3. Screen shot from REW attached comparing Dirac 3.2.3 to 3.3.3.

Crazy! So I went back into Dirac and reloaded the project and double checked the target curves - especially to be sure their new bass boost curves were not applied. I rebooted the processor. I did a Backup Settings > Factory Reset > cold reboot > Restore Settings. Same result. With the full cal using DL 3.3.3, the BM bug is back!

And I verified there is no issue when using User EQ … it’s only with Dirac.

<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button>

I can't help but wonder if some BM is sneaking into the Center Sub LFE output. Those BM traces for the Small channels look more like the LFE trace than the Left or Right traces. Can you disconnect the Center Sub Output and see what REW hears?

Here ya go ... no difference.  When I turn the sub off and LFE plays to the fronts, the shape of the response is a little different because it plays through the subs up to 50Hz and then both fronts above that.  But you can see the left and center traces are identical whether the center sub is on or off.

[ttocs]

May 19, 2022 12:29:36 GMT -5 marcl said:

May 19, 2022 11:56:21 GMT -5 ttocs said:

I can't help but wonder if some BM is sneaking into the Center Sub LFE output. Those BM traces for the Small channels look more like the LFE trace than the Left or Right traces. Can you disconnect the Center Sub Output and see what REW hears?

Here ya go ... no difference.  When I turn the sub off and LFE plays to the fronts, the shape of the response is a little different because it plays through the subs up to 50Hz and then both fronts above that.  But you can see the left and center traces are identical whether the center sub is on or off.

<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button>

I'm confused, I think I need a refresher on how your system is routed. Are your Fronts used for some portion of the LFE range?

I'm just asking for the subwoofer to be disconnected from the Center Sub Output, and no other changes. Keep the Center Sub LFE Output active in the processor, but just disconnect the interconnect cable going to the subwoofer. Unless that's what you did and I'm not following correctly.

[marcl]

May 19, 2022 12:54:14 GMT -5 ttocs said:

May 19, 2022 12:29:36 GMT -5 marcl said:

Here ya go ... no difference.  When I turn the sub off and LFE plays to the fronts, the shape of the response is a little different because it plays through the subs up to 50Hz and then both fronts above that.  But you can see the left and center traces are identical whether the center sub is on or off.

<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button>

I'm confused, I think I need a refresher on how your system is routed. Are your Fronts used for some portion of the LFE range?

I'm just asking for the subwoofer to be disconnected from the Center Sub Output, and no other changes. Keep the Center Sub LFE Output active in the processor, but just disconnect the interconnect cable going to the subwoofer. Unless that's what you did and I'm not following correctly.

Sorry, I misunderstood ...  that's not what I did ... I changed Center sub to None.  Okay here it is with the sub still set to LFE but unplugged.  No difference.



And my Dirac measurement and target for the Fronts.




Refresher on how my Fronts are configured.  So BM goes to the Fronts .... small speaker bass should be identical to the Fronts.  LFE goes only directly to the subs.

[sealman]

May 19, 2022 10:53:47 GMT -5 marcl said:

Reposting here also what I posted in the Dirac thread ....

I got the 3.3.3 Dirac and did a full calibration yesterday, 13 points. I basically used my same target curves I've been using for a few months to get a flat final response as measured in REW. Since our FW 2.5 (and the Beta we got a few months before) the BM bug was mostly fixed, so my small speakers should only measure about +3db in the bass rather than the +8 we got before. This is with Center Sub=LFE and BM going to the full range fronts.

Well ... I measured the result with the 3.3.3 and it looks like the BM bug is back!! The small speaker bass is elevated again! Here's a comparison of my last full cal with DL 3.2.3 and yesterday's full cal with DL 3.3.3. Screen shot from REW attached comparing Dirac 3.2.3 to 3.3.3.

Crazy! So I went back into Dirac and reloaded the project and double checked the target curves - especially to be sure their new bass boost curves were not applied. I rebooted the processor. I did a Backup Settings > Factory Reset > cold reboot > Restore Settings. Same result. With the full cal using DL 3.3.3, the BM bug is back!

And I verified there is no issue when using User EQ … it’s only with Dirac.

<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button>

So this is a by product of the new 3.3.3 version of Dirac then and is not on the 3.2.3 version of Dirac?

[marcl]

May 22, 2022 14:36:25 GMT -5 sealman said:

May 19, 2022 10:53:47 GMT -5 marcl said:

Reposting here also what I posted in the Dirac thread ....

I got the 3.3.3 Dirac and did a full calibration yesterday, 13 points. I basically used my same target curves I've been using for a few months to get a flat final response as measured in REW. Since our FW 2.5 (and the Beta we got a few months before) the BM bug was mostly fixed, so my small speakers should only measure about +3db in the bass rather than the +8 we got before. This is with Center Sub=LFE and BM going to the full range fronts.

Well ... I measured the result with the 3.3.3 and it looks like the BM bug is back!! The small speaker bass is elevated again! Here's a comparison of my last full cal with DL 3.2.3 and yesterday's full cal with DL 3.3.3. Screen shot from REW attached comparing Dirac 3.2.3 to 3.3.3.

Crazy! So I went back into Dirac and reloaded the project and double checked the target curves - especially to be sure their new bass boost curves were not applied. I rebooted the processor. I did a Backup Settings > Factory Reset > cold reboot > Restore Settings. Same result. With the full cal using DL 3.3.3, the BM bug is back!

And I verified there is no issue when using User EQ … it’s only with Dirac.

<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button>

So this is a by product of the new 3.3.3 version of Dirac then and is not on the 3.2.3 version of Dirac?

I haven't totally tracked it down, but it is specific to my system where I use the large fronts for bass management and the symptom is that when the L/R play together the bass below 100Hz is boosted about +5db.  The crazy thing is, it will happen with Dirac 3.2.3 if I take new measurements (and with 3.3.3 with new measurements).  I have to investigate further.  And ... it seems no one else is seeing this issue, but you might not notice unless you measure after calibration.

[pepar]

So, I read the first two pages and this last one, and my takeaway is that I should contact Emo if I have this problem? That's kind of rhetorical ..

[webmst007]

Marc, LSC, ham,Keith and everyone
Firstly I'd like to thank you and all for the contributions I read here almost daily 😁
Marc - can I ask about your set up timing?
I have your Maggie's at zero and then the DWM woofers at least 1.07ms behind ( minimum minidsp delay according to them) and the LFE signal about the same.
But the low frequency sent from the main split via linked minidsp units goes to at least 2.14 ms delay. But I can't see how to correct for these delays in your chain.

[marcl]

Aug 8, 2022 14:30:48 GMT -5 webmst007 said:

Marc, LSC, ham,Keith and everyone
Firstly I'd like to thank you and all for the contributions I read here almost daily 😁
Marc - can I ask about your set up timing?
I have your Maggie's at zero and then the DWM woofers at least 1.07ms behind ( minimum minidsp delay according to them) and the LFE signal about the same.
But the low frequency sent from the main split via linked minidsp units goes to at least 2.14 ms delay. But I can't see how to correct for these delays in your chain.

Okay, here's my bat-$h!t crazy configuration, as it sits since March this year after I repositioned and realigned the DWMs using REW Alignment Tool:



There's latency in the two miniDSP units in play for sure.  But I worked with the measurements and the Alignment Tool before doing the Dirac calibration.  I got everything as good as I could get it and then ran Dirac.  The goal was to get each 3.7 and its respective DWM optimized first, then integrate them with the subs timing and level-wise.

Here's left, right and LFE with Dirac on:



The two DWMs are positioned slightly differently, with the right one a little more forward.

Updated scale drawing of the room with 3.7s in the Rooze configuration.

[webmst007]

Hi Marc
your setup always gives me pause for thought :-)
So I'm going to go back and review for a moment if I may.
the signal information we are dealing with comes in 3 forms:
L/R main signals - (size = LARGE) which means full range including bass frequencies.
LFE signal - movie and encoded material provided via SUB OUT (CENTRE).
Bass frequencies from all other channels chosen as SMALL - output via SUB OUT (CENTRE).
the 2 mini dsp units you're employing have only 2 inputs each - hence the setup you have designed.
The issue I have is the time alignments - e.g. the BM signal from the L/R transferring through 2 MiniDSP's to the Outlaw subs in relation to the 3.7's.
The starting delay is 2.14ms (that's from MiniDSP website).
If we allow for additional delays by virtue of the crossover, PEQ, and attenuation in MiniDSP No.1 plus the additional attenuation in MiniDSP No.2, the final delay could be 3.5ms or higher. (I do not account for any PEQ phase shifts here).
The delay for the CENTRE SUB OUT information in the same Outlaw subs is probably less than half that - at around, say, 1.5-1.8ms at a guess.
Note -this signal includes bass information from all the channels marked "SMALL."
I would be fascinated to test out that theory in your system Marc by way of timing test signals via specific channels measured in REW.
From a PA background - I would not like to work with timing issues like that in a large sound re-enforcement setting.
So as a preference, I would probably sacrifice some source material and some EQ for a cohesive time-aligned array - but that may not be the case in your home setup of course.
AND it's worth noting the Maggies setup may help to "Cloud" The timing issues based on your Rooze configuration.

So I have some options to consider (although you may have already been down these routes yourself many times :-)
does the bass sound overall "tighter and leaner" if you disconnect the 2 minidsps from each other completely?
This stops BM from L/R getting to the Outlaws.
If you try ATMOS material how does the sound compare?

Next - Have you considered an alternative setup using these 2 units to remove the timing issue?
In this alternative I consider using 1 MiniDSP for each of the main L and R channels separately.

1. Connect the L/R inputs directly to input 1 on each of the miniDSP's and remove the splitters from the L/R circuits.
2. Connect SUB OUT to a splitter (freed from L/R duites) and connect 1 of it's 2 outputs to input 2 on both miniDSP's respectively.

So now each miniDSP has a main speaker and a feed from the SUB OUT (CENTRE) as the 2 inputs.

Connect miniDSP output 1 on each unit (L and R channels) to the NORD amplifer L and R inputs to drive the Maggy 3.7's.
Connect miniDSP output 2 on each unit (L and R channels) to the Crown XLi800 (L and R channels) to drive the DWM woofers.
Connect miniDSP output 3 on each unit to the Outlaw Subwoofers (Front and Back).

This is where there are 2 options now for the Outlaw Subwoofers:
Option 1 - Either connect the Outlaw Subs to output 3 on each unit directly.
Option 2 - SUM the MiniDSP outputs 3 by using the spare splitter (in reverse) and then split the single output signal again to feed both Outlaw subs.

Option 1 feeds only R MAIN or L MAIN BM channel bass to either the front or rear sub (plus the sub out signal of course).
Option 2 feeds both the L and R MAIN BM channel bass to both Outlaw subs (plus the sub out signal - which replicates your current setup only without the time alignment issues).

You would need to try both options - I would start with option 1 to test it - but -more likely- you will find option 2 fuller.
If you do explore option 2 - I would start by removing the SUB OUT feed from the routing table for one of the output 3 channels in one of the miniDSP's.
This way you don't "double up" the SUB OUT voltage (BM from various channels plus LFE signal) routing to the Outlaw Subs when the signal is summed and then split between them.

This setup should align all outputs in the time domain and in so doing:
Resolve the timing between the 3.7's and the DWM's.
Resolve the timing issue between the SUB OUT signal through the Outlaws and the 3.7's and the DWM's.
Resolve the timing issue between the L/R BM signal being fed to the Outlaws and DWM's.

Another option is to replace the 2 miniDSP's with a 3 in 6 out DSP unit - the miniDSP DDRC-88A would do the job and provide 1 interface + dirac etc plus it's consistent with the interfaces you're used to using already (always a bonus).

[marcl]

Aug 9, 2022 4:34:25 GMT -5 webmst007 said:

Hi Marc
your setup always gives me pause for thought :-)
So I'm going to go back and review for a moment if I may.
You information we are dealing with comes in 3 forms:
L/R main signals - (size = LARGE) which means full range including bass frequencies.
LFE signal - movie and encoded material provided via SUB OUT (CENTRE).
Bass frequencies from all other channels chosen as SMALL - output via SUB OUT (CENTRE).
the 2 mini dsp units you're employing have only 2 inputs each - hence the setup you have designed.
The issue I have is the time alignments - e.g. the BM signal from the L/R transferring through 2 MiniDSP's to the Outlaw subs in relation to the 3.7's.
The starting delay is 2.14ms (that's from MiniDSP website).
If we allow for additional delays by virtue of the crossover, PEQ, and attenuation in MiniDSP No.1 plus the additional attenuation in MiniDSP No.2, the final delay could be 3.5ms or higher. (I do not account for any PEQ phase shifts here).
The delay for the CENTRE SUB OUT information in the same Outlaw subs is probably less than half that - at around, say, 1.5-1.8ms at a guess.
Note -this signal includes bass information from all the channels marked "SMALL."
I would be fascinated to test out that theory in your system Marc by way of timing test signals via specific channels measured in REW.
From a PA background - I would not like to work with timing issues like that in a large sound re-enforcement setting.
So as a preference, I would probably sacrifice some source material and some EQ for a cohesive time-aligned array - but that may not be the case in your home setup of course.
AND it's worth noting the Maggies setup may help to "Cloud" The timing issues based on your Rooze configuration.

So I have some options to consider (although you may have already been down these routes yourself many times :-)
does the bass sound overall "tighter and leaner" if you disconnect the 2 minidsps from each other completely?
This stops BM from L/R getting to the Outlaws.
If you try ATMOS material how does the sound compare?

Next - Have you considered an alternative setup using these 2 units to remove the timing issue?
In this alternative I consider using 1 MiniDSP for each of the main L and R channels separately.

1. connect the L/R inputs directly to input 1 on each of the miniDSP's and remove the splitters from the L/R circuits.
connect SUB OUT to a splitter (free'd from L/R duites) and connect 1 of it's 2 outputs to input 2 on both miniDSP's respectively.

So now each miniDSP has a main speaker and a feed from the SUB OUT (CENTRE) as the 2 inputs.

Connect miniDSP output 1 on each unit (L and R channels) to the NORD amplifer L and R inputs to drive the Maggy 3.7's.
Connect miniDSP output 2 on each unit (L and R channels) to the Crown XLi800 (L and R channels) to drive the DWM woofers.
Connect miniDSP output 3 on each unit to the Outlaw Subwoofers (Front and Back).

This is where there are 2 options now for the Outlaw Subwoofers:
Option 1 - Either connect the Outlaw Subs to output 3 on each unit directly.
Option 2 - SUM the MiniDSP outputs 3 by using the spare splitter (in reverse) and then split the single output signal again to feed both Outlaw subs.

Option 1 feeds only R MAIN or L MAIN BM channel bass to either the front or rear sub.
Option 2 feeds both the L and R MAIN BM channel bass to both Outlaw subs.

You would need to try both options - I would start with option 1 to test it - but -more likely- you will find option 2 fuller.
If you do explore option 2 - I would start by removing the SUB OUT feed from the routing table for one of the output 3 channels in one of the miniDSP's.
This way you don't "double up" the SUB OUT voltage (BM from various channels plus LFE signal) routing to the Outlaw Subs when the signal is summed and then split between them.

This setup should align all outputs in the time domain and in so doing:
Resolve the timing between the 3.7's and the DWM's.
Resolve the timing issue between the SUB OUT signal through the Outlaws and the 3.7's and the DWM's.
Resolve the timing issue between the L/R BM signal being fed to the Outlaws and DWM's.

Another option is to replace the 2 miniDSP's with a 3 in 6 out DSP unit - the miniDSP DDRC-88A would do the job and provide 1 interface + dirac etc plus it's consistent with the interfaces you're used to using already (always a bonus).

Wow ... thank you for giving this so much thought and responding with all these options!  Most people just roll their eyes and hug their turntable 

So a couple things .... first, this arrangement - convoluted as it ended up - evolved over the last three years or so with many intermediate steps.  Like, I had a Y cable for the subs ... then one miniDSP just for the subs ... then added one DWM just for the center with the CC5 ... and so on.  Every step of the way was guided by a problem I wanted to solve.  Each problem was either audible, measurable, or both.  So each potential solution could be evaluated pretty objectively.  I think I can accurately say there was no audible problem that could not be confirmed with a measurement; but there were measurable problems that I could not necessarily hear conclusively.  The current configuration sounds better than the system has ever sounded .... 2-channel, multichannel and Atmos music ... 2, 5.1, upmixed and Atmos video ... it all sounds great and measures really well too.

I read your analysis and recommendations in detail and I have considered some of these ideas.  One I hadn't thought of was using one miniDSP per channel.  But there are a couple things that have a significant bearing on the options:

1. Bass Management does not go to the Center Sub output.  I have Center Sub=LFE so ONLY LFE goes to that output.  With L/R Fronts set to Large, BM (summed to mono) goes to L/R Fronts. 
2. The goal of #1 is to be sure LFE goes only to the subs, because LFE sent to any of the Maggies has too much dynamic range and causes diaphragm slap.  The other goal of #1 is to have all speakers in the room playing "Maggie bass" down to 40Hz, with the subs only playing below 50Hz.  This is especially important because the surrounds and tops cross over at 100Hz and the center CC5 only goes down to 200Hz.  I wanted BM to come from the 3.7s.
3. But I wanted the L/R channels to have the subs play below 50Hz.  In order to get BM to play to the L/R Fronts I had to have them set to Large.  Normally that would not send L/R to the subs.  Using the Enhanced Bass feature would send L/R to the subs below a configured crossover ... but, I would have no control over that sub level or the combined response of L/R and subs ... Dirac would not "see" this combined response for calibration purposes.  So that's why I did the split.
4. I have thought of solutions that send L/R to miniDSP and then feed the 3.7s, DWMs and subs.  Fundamentally - and maybe this is more philosophical than anything - I can't bring myself to run the full L/R Fronts through an A/D/A cycle in the miniDSP.  Emotiva has a robust DAC implementation and I think running the L/R through the miniDSP would compromise that.  I have much less concern doing that with signals that only drive the DWMs and subs.

So then for the delays.  Yeah there's latency through the miniDSPs and I think the latency may be different for the 2x4 Balanced and the 2x4HD ... and then I cascaded them for the L/R sub feed!  Holy cow!  And yes, the Rooze setup makes the timing more complex because the direct sound comes from a bounce off the left and right walls.  Placement of the DWM for each 3.7 can't be done absolutely by distance.  So I experimented with placements of the DWMs based on their measured effect combining with their respective 3.7 in the range from 50-200Hz.  The goal was to fill in modal cancellations that Dirac could not fix.  And finally I used REW Impulse Response measurements and the Alignment Tool to optimize the position and delays of the DWMs.  Note that the DWMs did not end up positioned symmetrically, and their delays are different.  This is what I derived from measurements .... and I believe one explanation is the fact that the 3.7s have tuning discs that purposely cause resonances in the panels to occur at different frequencies for the left and right.  the DWM positions and delays ended up set to smooth out the response of the resonances and room modes. 

Your suggestion to run the L/R through one or two miniDSPs would have allowed me to adjust the delays much more precisely.  The 3.7s being the "faster" speakers would have had the most delay so the DWMs could catch up to them and then the subs catch up to the sum of them.  But like I said, for now I resisted doing the A/D/A again on the 3.7s.  It would simplify the whole process to use a miniDSP with more inputs and outputs, but still split out the L/R to the mains.

Thanks again!

[webmst007]

Hi Marc
You are very welcome.
To your points:
I absolutely get why the 3.7 signals are not going through the minidsp.
I'm in a similar position in fact, using a 3in 6out (dbx venue360) just for 4 subs.
That unit has proved useful in some modest PA setups but I've not trusted it in main electrostatic stereo setups. That view has been supported by ASR's Venue360 test results.
On that subject - the new miniDSP flex has very good performance test results which could be another option for you long term. I don't think 2 of those replacing the 2x4 units you have currently would degrade your Maggie's signal much at all. Maybe you could try that at some point?
If that were the case then all drivers could be time aligned.
I think the gains in that setup - whereby your signals are all cohesive - could outweigh any slight change to signal through an additional A/D/A filter.
So the EE in me is curious now to test that theory out on your existing setup but that would require a rewire 😁👍☕🥳👌🤓
My own process was starting with 3 sub outputs from the RMC1, with the 360 taking all 3 and mixing to 4 subs laid out in a Harmon pattern.
I didn't get great results so I've played with room treatments for quite a while now.
Then I migrated to a single sub out (like you) and calibrated the subs using REW and MSO which has been a lot of fun and is still a work in progress.
However, like you, the integration of the 2 front subs into the mains ( DIY 2metre electrostatic panels - I build various panels and designs) as "LARGE" is my goal at this point - at least until Emotiva get subs working properly.
The reason I mentioned the Flex is it may solve some sub issues and it's got Dirac in it. I'm considering one to replace my 360 just for the subs anyway.
But that does not address main integration and that's why I've been looking at various either new 3 or 4 ins and 8 outs DSP options or units I've used previously for larger arrays. The 88A is definitely an option I'm looking at.

Marc I still think you should try experimenting with your current setup. 🥳🥳🥳👍🤓🌟🌟🌟🌟🌟

[marcl]

Aug 9, 2022 18:36:41 GMT -5 webmst007 said:

Hi Marc
You are very welcome.
To your points:
I absolutely get why the 3.7 signals are not going through the minidsp.
I'm in a similar position in fact, using a 3in 6out (dbx venue360) just for 4 subs.
That unit has proved useful in some modest PA setups but I've not trusted it in main electrostatic stereo setups. That view has been supported by ASR's Venue360 test results.
On that subject - the new miniDSP flex has very good performance test results which could be another option for you long term. I don't think 2 of those replacing the 2x4 units you have currently would degrade your Maggie's signal much at all. Maybe you could try that at some point?
If that were the case then all drivers could be time aligned.
I think the gains in that setup - whereby your signals are all cohesive - could outweigh any slight change to signal through an additional A/D/A filter.
So the EE in me is curious now to test that theory out on your existing setup but that would require a rewire 😁👍☕🥳👌🤓
My own process was starting with 3 sub outputs from the RMC1, with the 360 taking all 3 and mixing to 4 subs laid out in a Harmon pattern.
I didn't get great results so I've played with room treatments for quite a while now.
Then I migrated to a single sub out (like you) and calibrated the subs using REW and MSO which has been a lot of fun and is still a work in progress.
However, like you, the integration of the 2 front subs into the mains ( DIY 2metre electrostatic panels - I build various panels and designs) as "LARGE" is my goal at this point - at least until Emotiva get subs working properly.
The reason I mentioned the Flex is it may solve some sub issues and it's got Dirac in it. I'm considering one to replace my 360 just for the subs anyway.
But that does not address main integration and that's why I've been looking at various either new 3 or 4 ins and 8 outs DSP options or units I've used previously for larger arrays. The 88A is definitely an option I'm looking at.

Marc I still think you should try experimenting with your current setup. 🥳🥳🥳👍🤓🌟🌟🌟🌟🌟

Interestingly, a friend of mine just got a Flex to use with his Mini Maggie desk system.  So I'll hear that at some point and see how it works for him.  I do think it's unlikely to degrade the main signal.

But doing Dirac style processing outboard of the XMC-2 seems like it would not work as well as using Dirac in the XMC-2, in particular because my system is 7.1.4 and I want it all to work well together.  I even had experimented with doing PEQ in the miniDSP and always went back to letting Dirac do the whole job.  

Yeah ... the EE in us is always curious and always thinking of a new experiment.  My ideas generally arrive between 3 and 4am 

[webmst007]

Yes Marc - I agree about use of the processor-based Dirac as the primary and only EQ as a principle starting point.
I only EQ add for the 4 subs which I evolve on MSO as a result of REW readings.
Those filters I add to the Venue360.
That has proved interesting but not necessarily successful compared to a more progressive manual approach.

Then Dirac sees those 4 as a single sub and adds filters and phase/delay accordingly.
Marc
I'll post readings and results later and we can discuss more 😁🤪🤪🤪🤓🤓🧐🧐🧐😁😂

[marcl]

Is anyone using Dirac with a configuration either Center Sub None or Center Sub LFE, with Bass Management going to Large Fronts?

Are you using the latest Dirac 3.4.4?

[marcl]

Just saw this posted on the Emotiva FBook Group.

I guess I didn't know that the high pass is always 12db/octave, and the setting is selectable 12 or 24 only for the low pass.

Also, it seems to imply the Enhanced Bass setting is just for when you have large fronts playing 2.0 channel content.  Does it not work for multichannel?  Does it not work if you have no subs or center sub set to LFE, and bass management going to large fronts and Enhanced Bass enabled?


Bass_Management_and_XMC_RMC_Processors.pdf (83.37 KB)

[fbczar]

Mar 20, 2023 16:25:07 GMT -5 marcl said:

Just saw this posted on the Emotiva FBook Group.

I guess I didn't know that the high pass is always 12db/octave, and the setting is selectable 12 or 24 only for the low pass.

Also, it seems to imply the Enhanced Bass setting is just for when you have large fronts playing 2.0 channel content.  Does it not work for multichannel?  Does it not work if you have no subs or center sub set to LFE, and bass management going to large fronts and Enhanced Bass enabled?


View Attachment

I asked Keith sometime ago if the G4 processors might have a variable filter and he said there were no plans for that functionality. 

[hsamwel]

Feb 22, 2023 9:54:21 GMT -5 marcl said:

Is anyone using Dirac with a configuration either Center Sub None or Center Sub LFE, with Bass Management going to Large Fronts?

Are you using the latest Dirac 3.4.4?

Not many use these kind of configurations as they are MUCH harder to get right in a room.

Most use a single output for two or more subs LFE+BM.. Or multiple subs through a MiniDSP or similar
device still with single output LFE+BM from the processor.

When you have Dirac BC then I guess you could configure a system differently with better results
more easily.

On a different note.. I have totally changed my listening room, with new furiture and some changes
in the speaker placements. Added a ceiling absorbant that also work as a anti reflect for light as I have
a white ceiling (for my projector).
When I measured my room now with 13p I got a lot worse bass readings than before.
70hz and below pretty much the same. But 70-150hz I got really low readings.
On the previous measurement from april 2022 I used 9p tightly focused.

I think because of this I got a slight lower volume calibrated. Before I had pretty loud watching atmos
at -26dB. Now the feeling is I can raise it to about -20dB for the same result.
The bass got tighter again because I simply have run a ”wrongly calibrated” for a long time.
The strange thing is that I feel I got deeper bass now and more even.
Don’t know if this is my room changes or if the new Dirac does it differently?

[hsamwel]

Mar 20, 2023 16:25:07 GMT -5 marcl said:

Just saw this posted on the Emotiva FBook Group.

I guess I didn't know that the high pass is always 12db/octave, and the setting is selectable 12 or 24 only for the low pass.

Also, it seems to imply the Enhanced Bass setting is just for when you have large fronts playing 2.0 channel content.  Does it not work for multichannel?  Does it not work if you have no subs or center sub set to LFE, and bass management going to large fronts and Enhanced Bass enabled?


View Attachment

Enhanced bass is duplicating the bass from your large fronts to the subs.
LFE sub is pretty much not having any sub as this is an own channel.
The Enhanced bass is somewhat working with the BM as it gets its HP filter from
the BM settings of your front speakers when set to small.
So how could this work without any subs? 

This feature seems REALLY pointless in a multi channel setup so I guess 2.0 would be correct.
In any normal setup that is.. Yours is excluded!