|
|
Post by smfins on Aug 19, 2014 21:58:25 GMT -5
How the XMC-1 handles two subs is very important to me. I'm currently an Audyysey XT-32 user and it does an absolute amazing job with dual subs.
|
|
|
|
Post by igorzep on Aug 20, 2014 1:39:30 GMT -5
I think this is where I don't understand when you say Dirac won't be able to EQ both as one unit. I don't know how else the term "independent delay/phase/level/EQ" could be interpreted. And the problem is NOT ONLY in EQueing independently vs as a unit. It is only part of the problem. The other part is properly setting up delays and that is not as simple as setting it according to distances from the MLP to each of two subs. And so, as there is no automatic way to set them, there should be two part of the procedure - setting up an acoustical interaction, and it should be possible to do it manually (even if the second part is automatic) and the automatic part (the combined EQ) should respect this manual setting. Only solving both parts of the problem make it working flexibly and properly. Solving just one and not another make only visibility of something is gained. |
|
|
|
Post by sme on Aug 20, 2014 1:47:30 GMT -5
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 [emphasis added] as room wall are not perfect mirrors ResultVery 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. For those lacking context, my quote is from igorzep's discussion of how Double Bass Arrays are set up. I must ask here: What if the acoustic losses are frequency varying? In reality, I bet they do vary unless all your room boundaries are concrete and earth. That means you need frequency varying gain for the rear array to precisely cancel the wavefront from the front array. How do you do that without being able to use separate EQ curves? Stud walls often have mechanical resonances in the bass range that may introduce additional peaks and ringing in the response as the walls absorb and re-radiate energy at a later point in time. The same thing might happen with the rear array subs themselves if they don't have strong enough damping. Frankly, if I were ambitious enough to build an actual double bass array (how many woofers are people considering in these designs, anyway?), I'd probably look for a bass audio processor that offers a lot more than 2 outputs. Even now, I'm using a processor with 4 outputs (MiniDSP 2x4). Ideally, I would be able output a different signal to each woofer. Despite the theoretical appeal of the DBA, in practice it will likely still suffer from room problems, including the aforementioned mechanical resonances. In my view, being able to separately EQ the woofers maximizes the performance potential of any setup. |
|
|
|
Post by igorzep on Aug 20, 2014 5:54:18 GMT -5
I must ask here: What if the acoustic losses are frequency varying? In reality, I bet they do vary unless all your room boundaries are concrete and earth. That means you need frequency varying gain for the rear array to precisely cancel the wavefront from the front array. How do you do that without being able to use separate EQ curves? Stud walls often have mechanical resonances in the bass range that may introduce additional peaks and ringing in the response as the walls absorb and re-radiate energy at a later point in time. The same thing might happen with the rear array subs themselves if they don't have strong enough damping. In reality... In reality if you are in a room where whole wall has strong high Q acoustical resonance... and it is not specifically designed bass-trap... you are already in big trouble. I bet... with separate EQ you will not solve THE problem. What is more interesting, that by killing the reflection you are killing (or reducing) the problem even before there is a chance to excite this resonance, and so, the DBA still works well and not showing any problem. The only requirement for the DBA to work close to perfection is the rectangular room with not extremely much acoustic difference between opposite walls. If there is nothing to reflect then there is no frequency varying losses. Your fear is based on seeing the modal response of the room. But it is not because walls have varying losses, it is for completely different reasons. The real / commonly happening frequency dependent variation of losses across the sub range is tiny. Can you find just one case among those who did DBA who had even minor problem with it not working as 'advertised'? Frankly, if I were ambitious enough to build an actual double bass array (how many woofers are people considering in these designs, anyway?) If ceiling is low enough and width of the room is also not so big, the two pairs could be enough for 80Hz sub... Yes, it is a tiny room. Bigger room needs more drivers of course, or the highest frequency when it shows spectacular results will be lower. I'd probably look for a bass audio processor that offers a lot more than 2 outputs. And that would be overkill. Don't over-complicate things. Even now, I'm using a processor with 4 outputs (MiniDSP 2x4). Ideally, I would be able output a different signal to each woofer. Despite the theoretical appeal of the DBA, in practice it will likely still suffer from room problems, including the aforementioned mechanical resonances. It is not just theoretical. It works the same in practice. In my view, being able to separately EQ the woofers maximizes the performance potential of any setup. Will this potential be ever realized, if even without it you get negligible variation of frequency response across the whole room? I am perfectionist in everything. But I don't think you need to go that far and optimize something like the last dB of the in-room variation... For non DBA case and more than two subs placed 'unevenly' in the room - well, you probably need more than two outputs that each delayed differently, if those subs are different from each other then also probably you want one or two biquads to optimize that before applying common equalization. But if by 'separately' you mean each sub is EQued individually without looking at how it interacts with everything else... then it is far from maximizing potential. As it doesn't help in eliminating the main problem we are fighting with multiple subs - forming the room mode. |
|
bootman
Emo VIPs 
Typing useless posts on internet forums....
Posts: 9,358
|
Post by bootman on Aug 20, 2014 7:44:54 GMT -5
I think this is where I don't understand when you say Dirac won't be able to EQ both as one unit. I don't know how else the term "independent delay/phase/level/EQ" could be interpreted. And the problem is NOT ONLY in EQueing independently vs as a unit. It is only part of the problem. The other part is properly setting up delays and that is not as simple as setting it according to distances from the MLP to each of two subs. And so, as there is no automatic way to set them, there should be two part of the procedure - setting up an acoustical interaction, and it should be possible to do it manually (even if the second part is automatic) and the automatic part (the combined EQ) should respect this manual setting. Only solving both parts of the problem make it working flexibly and properly. Solving just one and not another make only visibility of something is gained. That is not what I asked so I will re ask it.  I asked how do you know that Dirac won't "respect the independent settings" (as you put it) of the bass setup as it is now available in V1.1? I asked this question of Emotiva here about 4 times now in different ways hoping someone would understand what I'm asking. (this one is now five?) No one has answered that yet so we just don't know. If it does, then I think we are set which is what I also pointed out a few times. Also how do we know Dirac won't do a two part procedure? We know the trial version only does .1 but the XMC was .2 in the truest sense from the beginning of its design. It would be disappointing if this wasn't accounted for in the XMC's version of Dirac since it may severely limit dual sub placement in a room with Dirac engaged. (dual up front in stereo and dual stacked should work, anything else who knows for sure) So far we have ZERO official information in this regard so everything up to this point is speculation. I want to put an end to speculation but until Emotiva tells us how Dirac handles dual subs, we will have to wait a bit longer to clear the air. |
|
|
|
Post by igorzep on Aug 20, 2014 8:27:30 GMT -5
I asked how do you know that Dirac won't "respect the independent settings" (as you put it) of the bass setup as it is now available in V1.1? replied that subs will get independent EQ. This means they will not be even EQued as a unit. It is the only way I can interpret the answer. And past experience of Emotiva people answering in such a way means there is no hope they have thought this out and we cannot expect more than what was said if they are not explicitly enough mentioned it. I asked this question of Emotiva here about 4 times now in different ways hoping someone would understand what I'm asking. (this one is now five?) No one has answered that yet so we just don't know. How many times did I... and bluescale did... and some others... They prefer just not to give clear answers until they have something positive to say about it. So, no answer pretty much mean negative one. Also how do we know Dirac won't do a two part procedure? We know the trial version only does .1 but the XMC was .2 in the truest sense from the beginning of its design. Trial version is a Dirac product. It does only what Dirac does - the EQ. No Bass Management (although it equalizes subwoofer channel the BM should be done elsewhere). It is also what they are licensing to the licensees. Bass Management is done by them, not Dirac, and we should not expect it be done by Dirac, or make any conclusion from other products. Datasat have no problem supporting two subwoofers. But what capabilities they have - I don't know... And I wouldn't call the "Stereo sub" concept of Emotiva a "truest sense .2". It is not. It is not a bass management, it is just (limited by frequency) bi-amping in it's truest sense. Do you call bi-amped active speakers as 1.1 speaker, or the pair of such speakers as 2.2? I don't and I don't know anyone who is. So far we have ZERO official information in this regard so everything up to this point is speculation. I want to put an end to speculation but until Emotiva tells us how Dirac handles dual subs, we will have to wait a bit longer to clear the air. Let's wait. I am tired already going rounds and rounds. |
|
|
|
Post by bluescale on Aug 20, 2014 12:16:59 GMT -5
Trial version is a Dirac product. It does only what Dirac does - the EQ. No Bass Management (although it equalizes subwoofer channel the BM should be done elsewhere). It is also what they are licensing to the licensees. Bass Management is done by them, not Dirac, and we should not expect it be done by Dirac, or make any conclusion from other products. Datasat have no problem supporting two subwoofers. But what capabilities they have - I don't know... The reason, I believe, that this won't be an issue with Datasat is because it's almost always installed by professionals. Honestly, most people don't know how to set delay on a sub correctly (I know I didn't until very recently). As igorzep points out, Dirac isn't designed to do bass management. It's designed to just do the EQ piece. If it's presented with two separate channels to EQ, I don't see why it would think to do otherwise. If Emotiva or Dirac have put in fancy processing to first determine the sub mode, it sure would be nice if someone could confirm that for us. flak - if you're around, you sure could make some of us breathe a bit easier with some information. |
|
|
|
Post by sme on Aug 21, 2014 3:21:07 GMT -5
Can you find just one case among those who did DBA who had even minor problem with it not working as 'advertised'? Admittedly, no because I haven't yet taken the time to read the links you sent. You seem fairly knowledgeable on the subject, so for now, I'll take your word for it that it works pretty consistently. That's interesting and encouraging. One drawback I see is that one may completely lose the benefit of room gain in the lower frequencies. If I go to the trouble and expense to install 8 woofers or more, I want to hear bass down into the single digits! I'm thinking you could apply an all-pass of some sort to the rears so that they play more in phase with the fronts as the frequency drops below the one corresponding to the longest room. Here again, having independent filters would be useful. This may be the root of the confusion. I absolutely *do not* mean that the subs can be successfully EQed individually without looking at how they interact in the room. What I'm saying is that in most instances it is *possible* to achieve better results using independent EQ curves by optimizing them together for the best combined response. You may be right that this is not necessary in a DBA arrangement or in some other configurations that are nearly symmetric. For everyone else, I believe separate filters can be helpful. I have already acknowledged that, unless Dirac has the capability to optimize separate EQ curves for the best combined response, it should instead optimize a single EQ curve applied to both subs. The exception to this rule would be when using the system in stereo bass mode. I would kind of expect Dirac to insist on this behavior if they are in fact as knowledgeable about room correction and acoustics as they say they are. However, there's always the possibility that, by virtue of having never implemented this technology in a consumer processor with multiple sub outputs (I'm ignoring Datasat here), the Dirac engineers merely hadn't given it enough thought. Either way, Dirac has not yet been released for the XMC-1 so there is still time to correct this issue if it exists. |
|
|
|
Post by flak on Aug 21, 2014 6:27:51 GMT -5
Trial version is a Dirac product. It does only what Dirac does - the EQ. No Bass Management (although it equalizes subwoofer channel the BM should be done elsewhere). It is also what they are licensing to the licensees. Bass Management is done by them, not Dirac, and we should not expect it be done by Dirac, or make any conclusion from other products. Datasat have no problem supporting two subwoofers. But what capabilities they have - I don't know... The reason, I believe, that this won't be an issue with Datasat is because it's almost always installed by professionals. Honestly, most people don't know how to set delay on a sub correctly (I know I didn't until very recently). As igorzep points out, Dirac isn't designed to do bass management. It's designed to just do the EQ piece. If it's presented with two separate channels to EQ, I don't see why it would think to do otherwise. If Emotiva or Dirac have put in fancy processing to first determine the sub mode, it sure would be nice if someone could confirm that for us. flak - if you're around, you sure could make some of us breathe a bit easier with some information. Hello Bluescale, yes... not always but I'm around if the XMC-1 reports to DLCT that there are two subs (i.e. 7.2 speaker configuration), then the DLCT for Emotiva will treat the dual subs as two separate subwoofer channels. Besides impulse and frequency response correction for each individual channel, the adjustments that will be made that affect their relative settings are level and delay adjustments (in order to align them with each other at the listening position). Bass management, crossover setting, or up-mixing from other source formats is out of the scope of Dirac. I hope this helps clarifying the situation, Flavio |
|
|
|
Post by bluescale on Aug 21, 2014 9:58:29 GMT -5
yes... not always but I'm around if the XMC-1 reports to DLCT that there are two subs (i.e. 7.2 speaker configuration), then the DLCT for Emotiva will treat the dual subs as two separate subwoofer channels. Besides impulse and frequency response correction for each individual channel, the adjustments that will be made that affect their relative settings are level and delay adjustments (in order to align them with each other at the listening position). Bass management, crossover setting, or up-mixing from other source formats is out of the scope of Dirac. I hope this helps clarifying the situation, Flavio Thanks for the prompt reply, Flavio. This is exactly what some of us thought. Are level and delay adjustments set in the XMC-1 distance and trim settings, or do they become part of the Dirac proprietary filters? If they are part of the filters, I guess the solution would be to run the old mono setting. If they are set within the XMC-1 UI, it'll be very important for Emotiva to address this. |
|
|
|
Post by rswood on Aug 21, 2014 11:01:49 GMT -5
Received my XMC yesterday and during set found some software issues listed below.
1. Setup, preset 1, dual mono subwoofer, level adjustment, right subwoofer does not play test tone to adjust level (when using stereo subwoofer, adjustment for left and right test tones work and can be adjusted).
2. Dual Mono Subwoofer, there is no option in the menu to adj
|
|
|
|
Post by rswood on Aug 21, 2014 11:20:27 GMT -5
Sorry about the missing information on my last post, problem with my computer.
Received my XMC yesterday and during set found some software issues listed below.
1. Setup, preset 1, dual mono subwoofer, level adjustment, right subwoofer does not play test tone to adjust level (when using stereo subwoofer, adjustment for left and right test tones work and can be adjusted).
2. Dual Mono Subwoofer, there is no option in the menu to adjust the sum of both subwoofers, hence the name Dual Mono. why is there no option to adjust the level of both speakers combined?
3. Stereo Subwoofer set up, EQ setup, Left subwoofer, filter 1, 25hz, example 3db to hot, step down to level adjust, adjust level meter shows level drop, works great. Same process for Right sub drop down to level adjust, the value changes but the level does not, I moved the level down 40 clicks still no change in volume of frequency being played.
4. Mono subwoofer setup, play test tone to change level, only left sub plays test tone. I thought both subwoofers were supposed to get the same information?
I called this information in and I believe they were aware of some of these issues but I wanted to write them down because I am not sure I was explaining it very well over the phone.
|
|
|
|
Post by igorzep on Aug 21, 2014 12:01:57 GMT -5
One drawback I see is that one may completely lose the benefit of room gain in the lower frequencies. If I go to the trouble and expense to install 8 woofers or more, I want to hear bass down into the single digits! I'm thinking you could apply an all-pass of some sort to the rears so that they play more in phase with the fronts as the frequency drops below the one corresponding to the longest room. Here again, having independent filters would be useful. This is something that is definitely possible - only what is needed it seems is LR2 or reverse LR2 (can't say which) all pass (rotating the phase by 180 degrees for one of the arrays. Depending on what is needed (forward or reverse) it could be computationally quite intensive (very long FIR as frequencies are very low)... either way - not impossible. I absolutely *do not* mean that the subs can be successfully EQed individually without looking at how they interact in the room. What I'm saying is that in most instances it is *possible* to achieve better results using independent EQ curves by optimizing them together for the best combined response. You may be right that this is not necessary in a DBA arrangement or in some other configurations that are nearly symmetric. For everyone else, I believe separate filters can be helpful. You don't, someone who read this will... So I am clarifying. Considering the possibility for optimizing, the primary one comes from the delays, sure some EQ could help optimizing more, but you should need it a lot. As I said - the ideal case if there are a few (like two at max) biquads allowed in addition to the delay. There should not be a need for more if you properly (i.e. - spacialy) optimizing. Everything above that should be common EQ. If you feel that you need a lot of separate EQ then this only means the delays were not optimal. The optimal system is the one that needs minimal EQ. |
|
|
|
Post by AudioHTIT on Aug 21, 2014 13:13:19 GMT -5
... And I wouldn't call the "Stereo sub" concept of Emotiva a "truest sense .2". It is not. It is not a bass management, it is just (limited by frequency) bi-amping in it's truest sense. Do you call bi-amped active speakers as 1.1 speaker, or the pair of such speakers as 2.2? I don't and I don't know anyone who is. Correct, no one calls bi-amped speakers .1 or .2, but those nomenclatures were specifically designated for subwoofers and LFE information. What would you call a pair of speakers, with a pair of subwoofers other than 2.2? You might argue that they are just 'stereo', but that doesn't include the possibility that the subwoofers could carry LFE, or be configured for mono - something not likely with a pair of traditionally bi-amped speakers. I think using the X.2 (or X.x) terminology is less ambiguous (7.2, 9.2.2, etc). I know you don't see the need for 'stereo' subs, but the capability is here so shouldn't we document it properly? |
|
|
|
Post by urwi on Aug 21, 2014 13:34:38 GMT -5
So Dirac treats two subs as two independent speakers and doesn't optimize the combined response in any way?? Not good. if the XMC-1 reports to DLCT that there are two subs (i.e. 7.2 speaker configuration), then the DLCT for Emotiva will treat the dual subs as two separate subwoofer channels. Besides impulse and frequency response correction for each individual channel, the adjustments that will be made that affect their relative settings are level and delay adjustments (in order to align them with each other at the listening position). |
|
|
|
Post by bluescale on Aug 21, 2014 14:26:40 GMT -5
So Dirac treats two subs as two independent speakers and doesn't optimize the combined response in any way?? Not good. It's important to keep in mind that Dirac isn't designed to do that. That's precisely why we need Emotiva to get the bass management for dual subs nailed down. |
|
|
|
Post by igorzep on Aug 21, 2014 14:33:58 GMT -5
Correct, no one calls bi-amped speakers .1 or .2, but those nomenclatures were specifically designated for subwoofers and LFE information. What would you call a pair of speakers, with a pair of subwoofers other than 2.2? You might argue that they are just 'stereo' They are. but that doesn't include the possibility that the subwoofers could carry LFE, or be configured for mono - something not likely with a pair of traditionally bi-amped speakers. The "Stereo sub" is quite specific configuration of the XMC-1. It doesn't include "the possibility". There is another one, it is called "Y-splitter". Well.. now there is a third one. But somehow out of all the first one, that is "just stereo", and not the one specifically designed for subwoofers and LFE information was called "the truest". I think using the X.2 (or X.x) terminology is less ambiguous (7.2, 9.2.2, etc). I know you don't see the need for 'stereo' subs, but the capability is here so shouldn't we document it properly? Unfortunately X.2 terminology is made too ambiguous today... Some manufacturers even declare internal Y-splitter as X.2. It is a pure fraud IMO. |
|
|
|
Post by AudioHTIT on Aug 21, 2014 14:51:00 GMT -5
... Some manufacturers even declare internal Y-splitter as X.2. It is a pure fraud IMO. And that I agree with, but just because some misuses a term doesn't mean you shouldn't use it at all. So I ask, if not 7.2 how would you identify the XMC-1's channel/sub configuration? (and no fair saying you would redesign it not to have stereo subs) |
|
|
|
Post by igorzep on Aug 21, 2014 15:29:49 GMT -5
So I ask, if not 7.2 how would you identify the XMC-1's channel/sub configuration? (and no fair saying you would redesign it not to have stereo subs) By number of unique outputs it is definitely 7.2, but the X.2 part in incomplete. It is something like having All Wheels Drive without locking differential - you kind of have it... but it doesn't work as you would expect from AWD. |
|
|
|
Post by AudioHTIT on Aug 21, 2014 17:23:34 GMT -5
So I ask, if not 7.2 how would you identify the XMC-1's channel/sub configuration? (and no fair saying you would redesign it not to have stereo subs) By number of unique outputs it is definitely 7.2, but the X.2 part in incomplete. It is something like having All Wheels Drive without locking differential - you kind of have it... but it doesn't work as you would expect from AWD. OK, now you're giving my AWD car a bad time; it gets me through the snow so I'm happy. |
|
