XMC bass management

[bootman]

Aug 18, 2014 18:34:39 GMT -5 Dan Laufman said:

If the EQ was tied together, you'd want it split... Can't win.

Actually the ideal setup with a true mono design would be to get each one as flat as possible (independent) then EQ the pair as one unit for the room.

Plus the original ask was for independent level and distance for dual mono vs the original one distance and level for both.
Since you did independant level, distance and EQ for stereo doing the same for dual mono was trivial, but in reality not the ideal way to do true dual bass management.

Now that said, if the way dual mono setup is done with firmware v1.1, what will dirac do with that?
Still EQ the pair as one or will it EQ each separately then the pair as one?

Or are we not explaining the question clearly enough?

[thecrusherk]

Aug 18, 2014 18:34:39 GMT -5 Dan Laufman said:

If the EQ was tied together, you'd want it split... Can't win.

Thanks for the reply Dan!  I actually can't think of a reason why I would want the EQ split at this time.  Perhaps there is some valid use case examples that I dont have experience with and would defer to the experts.  I am very happy that there is this discussion to learn.  I am amazed and happy with the speed and customization you did with the recent firmware update.  I have always found Emotiva to be willing to listen to the customer and to find appropriate solutions to problems and enhancement requests as possible.  I feel your passion and desire to create quality high performing audio products.  You have done this many times over and I am benefiting from it.   I hope you don't feel attacked as that is not my intent.  I am trying to suggest a change to the current design so that it can have consideration? Perhaps as an enhancement if the design allows for the EQ to lock (mirrored/copied) and unlock for those the need separate so everyone's wins?

Thanks again Dan for letting me bend your ear!

[bluescale]

Aug 18, 2014 18:34:39 GMT -5 Dan Laufman said:

If the EQ was tied together, you'd want it split... Can't win.

Of course you guys can win, Dan! You make an awesome product. We give you hard earned money. We give you suggestions on what could make said awesome product even more awesome. You make product awesomer. Even more people give you their hard earned money. Everyone wins!

Please don't take our comments for anything other than they are. We love our Emo gear, and as real world users, hopefully we've got valid feedback for you guys. in this instance, those of us who care about this feature feel you're most of the way there, but didn't quite hit it out of the park. Bootman's asked a few times how 2 subs will be handled by Dirac. If you guys can help us out with that answer, we'll have a better idea of whether or not this is a big deal or a minor quibble.

[klinemj]

Aug 18, 2014 18:55:09 GMT -5 bootman said:

Aug 18, 2014 18:34:39 GMT -5 Dan Laufman said:

If the EQ was tied together, you'd want it split... Can't win.

Actually the ideal setup with a true mono design would be to get each one as flat as possible (independent) then EQ the pair as one unit for the room.

Plus the original ask was for independent level and distance for dual mono vs the original one distance and level for both.
Since you did independant level, distance and EQ for stereo doing the same for dual mono was trivial, but in reality not the ideal way to do true dual bass management.

Now that said, if the way dual mono setup is done with firmware v1.1, what will dirac do with that?
Still EQ the pair as one or will it EQ each separately then the pair as one?

Or are we not explaining the question clearly enough?

I confess I have been trying to follow what it was vs. what it is vs. what was asked for, and I am totally confused. Can you elaborate. And, type slowly...I am not following at all.

Also, while we're at it...can you in very basic terms explain why (if having 2 subs is good and they appear to have done what you want for stereo subs) it would not be perfectly fine to use the setup as stereo subs instead of dual mono? I am no sub-junkie, so please bear with me as I learn.

Thanks,

Mark

[bootman]

Aug 18, 2014 20:41:14 GMT -5 klinemj said:

Aug 18, 2014 18:55:09 GMT -5 bootman said:

Actually the ideal setup with a true mono design would be to get each one as flat as possible (independent) then EQ the pair as one unit for the room.

Plus the original ask was for independent level and distance for dual mono vs the original one distance and level for both.
Since you did independant level, distance and EQ for stereo doing the same for dual mono was trivial, but in reality not the ideal way to do true dual bass management.

Now that said, if the way dual mono setup is done with firmware v1.1, what will dirac do with that?
Still EQ the pair as one or will it EQ each separately then the pair as one?

Or are we not explaining the question clearly enough?

I confess I have been trying to follow what it was vs. what it is vs. what was asked for, and I am totally confused. Can you elaborate. And, type slowly...I am not following at all.

Also, while we're at it...can you in very basic terms explain why (if having 2 subs is good and they appear to have done what you want for stereo subs) it would not be perfectly fine to use the setup as stereo subs instead of dual mono? I am no sub-junkie, so please bear with me as I learn.

Thanks,

Mark

I'll try.

XMC firmware V1.0

Two options:
1) Stereo bass with independent level, distance and EQ
2) Dual mono bass with one single level distance and EQ.

With the choices above what do you do if you have a dual sub setup that are not both up front?
Example: one sub in the front left corner and one behind your couch to your right?
Option A doesn't help. yes you can set each sub individually but it is stereo bass. Not all base info is equally going to both subs.
Option B doesn't help because you can't set independent level and distance to your subs to begin with. (remember one set of level and distance.)


XMC firmware V1.1

Two options:
1) Stereo bass. Same as above.
2) Dual mono bass with dual level, distance and now EQ.

This was rather simple to implement because all that was done was that the mono signal was routed to both outputs but they kept the stereo adjustability.

However here is my question.
If I have my dual subs setup like in the example above how do I get a flat response with dual EQ?
Notice I didn't mention level or distance. I do need those to be separate.

While I can EQ each separately, do you think that after doing that they will TOGETHER measure flat in the room without additional EQ?

Just think about it for a bit.

Hopefully it starts to make sense.

There are other sub EQ systems out there (Audyssey SubEQ HT) that does do this with a dual sub out.
They independently setup up each sub (not 100% sure about EQ) then EQ the set together for best in room response.

Now the only question that remains is this.
We have the ability to EQ each sub but not the ability to EQ the set together in the room.
Will Dirac do this for us?

[geebo]

Aug 18, 2014 21:23:23 GMT -5 bootman said:

Aug 18, 2014 20:41:14 GMT -5 klinemj said:

I confess I have been trying to follow what it was vs. what it is vs. what was asked for, and I am totally confused. Can you elaborate. And, type slowly...I am not following at all.

Also, while we're at it...can you in very basic terms explain why (if having 2 subs is good and they appear to have done what you want for stereo subs) it would not be perfectly fine to use the setup as stereo subs instead of dual mono? I am no sub-junkie, so please bear with me as I learn.

Thanks,

Mark

I'll try.

XMC firmware V1.0

Two options:
1) Stereo bass with independent level, distance and EQ
2) Dual mono bass with one single level distance and EQ.

With the choices above what do you do if you have a dual sub setup that are not both up front?
Example: one sub in the front left corner and one behind your couch to your right?
Option A doesn't help. yes you can set each sub individually but it is stereo bass. Not all base info is equally going to both subs.
Option B doesn't help because you can't set independent level and distance to your subs to begin with. (remember one set of level and distance.)


XMC firmware V1.1

Two options:
1) Stereo bass. Same as above.
2) Dual mono bass with dual level, distance and now EQ.

This was rather simple to implement because all that was done was that the mono signal was routed to both outputs but they kept the stereo adjustability.

However here is my question.
If I have my dual subs setup like in the example above how do I get a flat response with dual EQ?
Notice I didn't mention level or distance. I do need those to be separate.

While I can EQ each separately, do you think that after doing that they will TOGETHER measure flat in the room without additional EQ?

Just think about it for a bit.

Hopefully it starts to make sense.

There are other sub EQ systems out there (Audyssey SubEQ HT) that does do this with a dual sub out.
They independently setup up each sub (not 100% sure about EQ) then EQ the set together for best in room response.

Now the only question that remains is this.
We have the ability to EQ each sub but not the ability to EQ the set together in the room.
Will Dirac do this for us?

Just one small correction:

XMC firmware V1.1

Three (not two) options:
1) Stereo bass. Same as above.
2) Dual mono bass with dual level, distance and now EQ.
And
3) Dual mono bass with one single level distance and EQ.

[bootman]

Ah cool because that would be the correct one to pick if you have only one sub. ( option #3 )

Maybe that is what they should eventually call it?

[bluescale]

As usual, Bootman did a great job of breaking it down. The only thing I'll add is the reason why many of us want to have dual subs that are by set up in stereo. A stereo configuration routes data from the left channels to the left subwoofer, and data from the right channels. This can work well in some circumstances, generally when both subs are located up front by your mains, and that co-location yields a good in-room response. From the beginning the XMC-1 has allowed for this configuration.

Most people find they get the best in-room response if they have multiple subs placed in a semi-random configuration around their room to address modes and nulls. In that case, you want things set up the way Bootman describes, and that's what we're trying to influence with our suggestions.

[bluescale]

Aug 18, 2014 21:45:47 GMT -5 bootman said:

Ah cool because that would be the correct one to pick if you have only one sub. ( option #3 )

Maybe that is what they should eventually call it?

It would also work well it your subs are co-located.

[bootman]

True stereo sub is a feature that separates the XMC from the competition.
Option #1 should never change.

[bootman]

Aug 18, 2014 21:53:54 GMT -5 bluescale said:

Aug 18, 2014 21:45:47 GMT -5 bootman said:

Ah cool because that would be the correct one to pick if you have only one sub. ( option #3 )

Maybe that is what they should eventually call it?

It would also work well it your subs are co-located.

You mean stacked subs?
yes that would work for that setup.
The flexibility shown so far is great.
Just one more tweak (or confirmation on what Dirac does) to make it perfect.

[bluescale]

Aug 18, 2014 21:55:56 GMT -5 bootman said:

Aug 18, 2014 21:53:54 GMT -5 bluescale said:

It would also work well it your subs are co-located.

You mean stacked subs?
yes that would work for that setup.
The flexibility shown so far is great.
Just one more tweak (or confirmation on what Dirac does) to make it perfect.

Exactly. If you're going for max SPL, stack the subs. If you're looking for best in room response, find the ideal locations and plink 'em down there.

[thecrusherk]

Since Bootman made reference to Audessey SubEQ HT

audyssey.zendesk.com/entries/20953442-SubEQ-HT-vs-MultEQ-XT32

Chris´s first comment:

"MultEQ XT32 is the flagship version of our technology to measure and correct room acoustical problems. Sub EQ HT is a method we came up with to deal with multiple subs. If you only have one sub then it's not in use. The idea is to first measure each sub separately, then apply delay and level settings so that the two subs are now time and level aligned. Then we ping them once more as "one" sub to derive the room correction filter."

[bborzell]

For what it's worth, I have one sub in front left and the other back right. If I set them up as stereo, I get a test signal through each. If I set them as dual mono, only the front plays the signal. The right rear is silent. I just switched them back to stereo and declared victory.

[sme]

So how about if Emotiva added an option to tie together the EQs? Any changes made to the EQ for one sub would be automatically copied to the other. I imagine this would require a relatively simple tweak to the code compared to the effort required to make dual-mono subs work as they do now. Such a change would eliminate the current inconvenience of having to copy settings between the two when separate EQs are not desired.

There are many good reasons to EQ a pair of subs identically, even when using different level and distance settings. That doesn't mean that it's the "right" approach and that EQing the subs independently is the "wrong" approach. The reality is that EQing the pair as one is "better" because it's easier. I know for certain that Audyssey MultEQ XT32 does not EQ subs independently when optimizing response. I doubt Dirac has the sophistication to optimize the combined response of separately EQed subs. (Let us hope Dirac on XMC-1 provides a mode to EQ a pair of subs as one while allowing different distance and level settings.) I don't know enough about Trinnov to comment either way.

Nevertheless, I am very confident that I have achieved better results across my listening area using separate sub EQ curves on my playback equipment, in my (asymmetric) room, and with help from my measurement system and software tools. I don't expect many people to spend as much time on this as I have. For this approach to be useful to a wider audience, it may need to be automated. I'm not yet certain how difficult it is to automate the optimization and obtain a good result. I may study this problem more intently at a later date, but for now, I have other things to work on.

Even still, I can think of other use cases for separately EQed subs. If the two subs are not identical, the separate EQ can be used to try to distribute the load better to obtain more headroom. Another possibility is to use a pair of subs (not necessarily of the same type) in a bi-amped configuration. Let one sub handle deep bass and another mid bass. There are a few interesting reasons to want to do this. Unfortunately, I don't think the XMC-1 has (nor necessarily will have) the option to crossover one sub to the other, but who knows?

[igorzep]

Aug 18, 2014 18:34:39 GMT -5 Dan Laufman said:

If the EQ was tied together, you'd want it split... Can't win.

Can win, but only one way - let EQ be TIED together (in EQ as an absolute requirement), but NOT EQUAL (at least in the delay/polarity/level as a minimal requirement). As it was requested initially here, further explained in details here and here. And here is a suggestion for something that is as flexible as possible and that will fit just everyone beating and throwing out a need for any external BM device that is still (for Dirac case) required to properly manage two optimized subwoofer signals.

I suggest everyone interested to re-read it.

Aug 18, 2014 21:54:10 GMT -5 bootman said:

True stereo sub is a feature that separates the XMC from the competition.
Option #1 should never change.

True stereo sub is an oxymoron. As subwoofer is something that plays below the woofer, i.e. frequencies that are not localizable. And the term bass-management is also built on assumption you are working with non-localizable frequencies. Studies show us that sub with a crossover set @ less than 100Hz (80Hz typically) and 4-th order slope is not localizable. And "stereo" assumes localizability... But let the feature live for audiophiles who setup their system by religion instead of science. If it brings money for Emotiva, I am happy. I have no problem with it as long as it doesn't break it for another group of people. And now (well, not exactly now, but at the time the Dirac is out) it is broken as independent EQ of two subwoofers is not what we need.

Aug 18, 2014 20:41:14 GMT -5 klinemj said:

I confess I have been trying to follow what it was vs. what it is vs. what was asked for, and I am totally confused. Can you elaborate. And, type slowly...I am not following at all.

Also, while we're at it...can you in very basic terms explain why (if having 2 subs is good and they appear to have done what you want for stereo subs) it would not be perfectly fine to use the setup as stereo subs instead of dual mono? I am no sub-junkie, so please bear with me as I learn.

Read the above links in the first place. It is a good summary of history (not replied by anyone from Emotiva unfortunately)... Also a tittle following to those who still do not understand the problem.


A good example - the DBA concept (Double Bass Array). How it is configured 'as designed':
Sub A:
delay = the distance from the front array to the main listening position
phase (polarity) = 0 degrees
level = X dB
Sub B:
delay = delay of sub A plus the distance from the front array to the back array
phase (polarity) = 180 degrees
level = level of sub A minus some small number representing acoustic losses as room wall are not perfect mirrors
Result
Very flat response that is equal across the room. Done pure acoustically (we only used the delay/phase/level). We need only a very little electronic EQ here applied to the combined response.

Now what happens if we run it through the completely "independent" process:
Sub A:
delay = the distance from the front array to the main listening position
phase (polarity) = 0 degrees
level = Y dB
EQ = a lot of modal problems, so a lot of EQ
Sub B:
delay = the distance from the front array to the main listening position
phase (polarity) = 0 degrees
level = Z dB
EQ = a lot of modal problems, so a lot of EQ
Result
Response flatness is about equal to what we had with single equalized sub. May be better, may be worse. But newer even close the the properly set DBA as described above even without a single biquad applied. Response is heavily non-flat at any other distance from the front/back walls than the MLP.

While the DBA is quite idealistic case but it is a) pretty possible realistic config that some enthusiasts are building b) something similar, just not of so impressive and easily explained comparison, will happen with any other not-so idealistic config when locations of the subs are specifically optimized for even response across a large enough listening area. The process still should be
1) optimize the in room response acoustically (with sub placement and delay/level/phase in the first place, not necessary but may be a few biquads in addition {one or two, but not more, or you are doing something wrong if you want more at this stage}). The target is - even response across the room (or listening area), no wide heavy dips.
2) EQualize what was achieved above as a tied unit electronically (with Dirac/PEQ/whatever).

As the DBA example was mentioned in the original feature request, and the way it is implemented now in XMC-1 (independent EQ of two subs) does not allow to configure it properly together with Dirac (assuming, as was replied by Keith with Dirac it will do just that)... I can all rights to say that what was implemented is not what was asked.

Aug 18, 2014 21:23:23 GMT -5 bootman said:

There are other sub EQ systems out there (Audyssey SubEQ HT) that does do this with a dual sub out.
They independently setup up each sub (not 100% sure about EQ) then EQ the set together for best in room response.

They got the process right, but failed at the first step of acoustic integration. The process of finding the delay/phase/level is incorrect (try to apply it to DBA as an exercise). And so the end result is also incorrect and impossible to fix as the Audyssey filters are not editable after calibration and calibration cannot be done against the manual sub settings.

[bootman]

Igorzep, can you explain how your DBA setup applies to those with two box subs?
You keep bringing up DBAs but in all honestly that is a very rare setup and should be treated as such if the setup routine doesn't really work that well with normal box subs that >99% of enthuthiast have.

So can you clarify the dual box setup routine?

I think if Dirac treats the dual subs as one unit after setting up levels and distance independently, we are good to go for 99.99% of all home applications.
The other 0.01% will usually find a custom solution anyway don't you think?

[rswood]

Igorzep,

I have a few questions and please excuse me as one of us, English is not our first language.
In your DBA example Sub A is a box Subwoofer, yes/no.
Front Array= MAIN speakers. Back Array= Rear speakers?
Sub A Phase (polarity)= 0 degrees Do you mean this is set by the polarity switch on the sub?
Sub B Phase (polarity) = 180 degrees is this set by the polarity switch on sub? Is this assuming sub B is in the back of room? Or is the 180 degree polarity set on sub B wherever it is placed?
Sub B Level same as Sub A minus What? And how do you calculate? Are both Sub A and Sub B level matched at MSP?

You are I believe way out in front of most people, me included, in audio theory.

[bootman]

He means one of these setups.

www.avsforum.com/forum/155-diy-speakers-subs/837744-double-bass-array-dba-modern-bass-concept.html

Double bass array.

[rswood]

Aug 19, 2014 14:18:03 GMT -5 bootman said:

He means one of these setups.

www.avsforum.com/forum/155-diy-speakers-subs/837744-double-bass-array-dba-modern-bass-concept.html

Double bass array.

Thank you! I think 99.99 percent is too low.