• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Dutch & Dutch 8Cs

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,204
Likes
16,985
Location
Riverview FL
where i'm wondering, is that the 8C dsp works under fir, while all bass management processes work with iir? or maybe, that 8C algorithm is a mixture of fir and iir like for Dirac??

I use AcourateDRC with a miniDSP OpenDRC-DI.

It uses IIR for the lows and broad strokes of levels.

FIR is used for higher frequencies with small adjustments, and for phase touchup.

The GUI display demonstrates this mix:

IIR: upload_2018-5-3_14-14-6.png

FIR: upload_2018-5-3_14-16-20.png

When the number of taps is limited (6144 here) IIR (biquads) is more efficient/effective at lower frequencies, in my understanding.

The flatline at the bottom of the IIR display is a display limitation, not the filter.
 
Last edited:

fredoamigo

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2018
Messages
604
Likes
998
Location
South East France
thank you...yes, that's why fir is never used for subwoofer processing... to my knowledge no algorithm uses it because there is too much latency with fir for bass.
I deduce that if dutch &dutch does external bass management it's necessarily in iir so their algorithm is not just fir .
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
Yes, that Martijn is a really smart guy! However, I don't think that the IIR vs FIR thing is the most important issue with the multisub method. The challenge with this method is twofold:

1) find optimum placement for the subwoofers
2) find optimum delay and frequency response etc for each sub

This is needed for the extra subs (in addition to the D&Ds) to cancel out modes etc in an effective way. So it's more a question of having good enough measurements, and then dial that in. The only program I know of which does that for free, with the goal of an even frequency response throughout the room, is Multisub Optimizer (MSO). Check it out online!

And then there is another approach ("source sink"), which is to use an extra subwoofer to play in anti-polarity, and thus get a faster decay of bass energy in the room. It's fairly new, so there is very little information on this method online.

Anyway: with the D&Ds, the main challenge with subwoofer integration is not what happens in the DSP in the speakers themselves - it's rather to get good and precise measurements, and knowing what to do with the subs that are connected.

As far as I understand it (which may not be very far) at least! :)
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,204
Likes
16,985
Location
Riverview FL
thank you...

The resulting filter yields an output at the preamp that looks like this with a sine log sweep (instead of flat lines - see below) for frequency amplitude and phase:

upload_2018-5-3_14-49-9.png


And an in room measurement at the listening position like this:

upload_2018-5-3_14-53-53.png


My apologies to Dr Toole for not (yet) following his advice for the slightly sloping response. I can't hear up high, and nobody who can has complained (even when prompted).

The speakers are MartinLogan dipoles (the worst speakers ever in Sean Olive's speaker shootout). I've not seen another listening position in-room measurement that throws a flat phase response at the listening position. Cones and domes will do it nearfield, but at a distance, the line always slopes sharply down, really down (like off the chart bottomless pit down).

The big dip at 50Hz is a phase anomaly (asymmetrical back of the room) that creates a cancellation between the "two speakers" at the listening position. I've eliminated that for the most part now with a 7ms delay on the right side subs. The 220Hz dip is (I think) a dipole cancellation from the wall behind the speakers.

Preamp output of sine log sweep without filter:

upload_2018-5-3_15-1-6.png
 
Last edited:

matthijs

Member
Joined
May 4, 2018
Messages
6
Likes
15
Depending on the Gnu licence, etc. maybe they have to publish the software somewhere..? Isn't that the way some of this open source licensing stuff works?

If you developed and sold a DSP system based on ALSA, for example, would you have to publish your part of the software?

No. libalsa is LGPL, which means the license only applies to libalsa itself, not to any software that uses it. Open source software we've written or modified can be found on github, but this doesn't include any audio stuff currently. In fact, linux isn't even part of the signal path unless network streaming (which is being worked on) is used. For the DSP we use a DSP. ;-)
 
Last edited:

andrew

Active Member
Joined
May 3, 2018
Messages
192
Likes
42
@oivaroi
I read today the very interesting comment from the designer ( Martijn) on gearslutz .
where i'm wondering, is that the 8C dsp works under fir, while all bass management processes work with iir? or maybe, that 8C algorithm is a mixture of fir and iir like for Dirac??



""Credit to whom credit is due - to my knowledge Todd Welti of Harman (whom I had the pleasure of meeting recently) first introduced the concept of cancelling modes with multiple subwoofers. It was a real eye-opener for me at the time and it works really well. It's not for everyone though, because it's difficult to set up and you have to be willing to place several subwoofers around the room. The performance is there though!

You can find more information here: https://www.harman.com/sites/default...multsubs_0.pdf

With the 8c the idea is that the two 8c's act as subwoofers themselves and you have to add only two more subs to effectively get a 4-sub setup. The 8c's have a subwoofer output, but right now it's not supported by the Dutch & Dutch app on lanspeaker.com. It's one of the many features on the back-log. When it's done, you will indeed be able to adjust basically any useful parameter, such as EQ's, crossover types, delay etc.""

Optimising the support for subwoofers that augment the 8c seems like being a valuable development. That said, right now there are an increasing number of subs with built-in DSP (e.g., SVS) and external devices that can be used to manage subwoofer EQ; e.g., mini-DSP. The missing bit is that it’d should be possible to delay the main speaker in relation to the subwoofer output of the 8C to account for the common situation in which subs are further from the listening position and have in-built delay. Such a set-up would allow one to feed the 8c with a digital input with the subwoofer or AES thru output sent, say, to a mini-DSP that controllers the multiple subs and still have everything time aligned.
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
Optimising the support for subwoofers that augment the 8c seems like being a valuable development. That said, right now there are an increasing number of subs with built-in DSP (e.g., SVS) and external devices that can be used to manage subwoofer EQ; e.g., mini-DSP.

Good point. For people who already have a functioning and well-implemented dsp/bass/subwoofer system, that part of the D&D's isn't the most important thing. It would for example be possible to just keep the "front part" of the D&D's, and cross the sub 100 hz part over to external subwoofers completely. That would arguably be a superior solution on paper, as it would create less vibrations inside the speaker cabinet. It would also put the subwoofers closer to the floor, which could be a good thing (but I'm not sure if that's important or not).

What the D&D package does offer, though, is the possibility for the D&D's to really be a complete solution in itself, with no other DSP thingies required, and deep enough bass as is. That's quite attractive to me, at least. It has always been a goal for me to make it simple. Given that they implement the subwoofer capabilities, and make it roon ready, this will be all I need for getting really amazing sound: A roon core running on my computer of choice. The 8Cs. A single extra subwoofer configured as a "sink" on the other side of the room, connected with a long xlr-cable to the 8Cs, which gets the job of cancelling excessive bass energy in the room. And voila. Nothing more needed. Based on my initial listening, I think it will be very hard to improve on such a setup for conventional two-channel stereo. And the beauty of it: It's so simple! Almost no cables, no box salad, no thinking about gear matching. Audiophile zen. Me like.
 
Last edited:

andrew

Active Member
Joined
May 3, 2018
Messages
192
Likes
42
What the D&D package does offer, though, is the possibility for the D&D's to really be a complete solution in itself, with no other DSP thingies required, and deep enough bass as is. That's quite attractive to me, at least. It has always been a goal for me to make it simple. Given that they implement the subwoofer capabilities, and make it roon ready, this will be all I need for getting really amazing sound: A roon core running on my computer of choice. The 8Cs. A single extra subwoofer configured as a "sink" on the other side of the room, connected with a long xlr-cable to the 8Cs, which gets the job of cancelling excessive bass energy in the room. And voila. Nothing more needed. Based on my initial listening, I think it will be very hard to improve on such a setup for conventional two-channel stereo. And the beauty of it: It's so simple! Almost no cables, no box salad, no thinking about gear matching. Audiophile zen. Me like.

Yes that’s a compelling offering. How, though, would one reliably setup a sub as “a sink”?
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
Yes that’s a compelling offering. How, though, would one reliably setup a sub as “a sink”?

It's a method which is fairly new on the scene. The only one I know who has implemented it here is @dallasjustice (if I'm not mistaken?). It was described in this paper, where it received higher ratings than any of the other methods for bass management under test: https://www.researchgate.net/public...reference_of_modal_control_in_listening_rooms

Of course, it's only one study. But the idea is inherently reasonable to my mind: that too much energy buildup in the bass in the room may be negative. When bass energy decays faster, it's easier to perceive transients and dynamics in the music. This will be always be less of a problem in large rooms though.

The idea is to set the signal to the sub to play with sufficient delay and low enough amplitude, so that it plays in anti-phase at the exact moment that the bass wave from the "source" (in this case the D&D's) hits it. The idea is that the bass wave from the sink only cancels the incoming waves, so you never hear the sink sub itself.

So it requires quite some measurements and adjustment...
 
Last edited:

fredoamigo

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2018
Messages
604
Likes
998
Location
South East France
Optimising the support for subwoofers that augment the 8c seems like being a valuable development. That said, right now there are an increasing number of subs with built-in DSP (e.g., SVS) and external devices that can be used to manage subwoofer EQ; e.g., mini-DSP. The missing bit is that it’d should be possible to delay the main speaker in relation to the subwoofer output of the 8C to account for the common situation in which subs are further from the listening position and have in-built delay. Such a set-up would allow one to feed the 8c with a digital input with the subwoofer or AES thru output sent, say, to a mini-DSP that controllers the multiple subs and still have everything time aligned.

Hi...I don't think a configuration with 2 bass management dsp could work very well... it would be like having two watches on your arm, but none of them would give the right time...
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,051
Likes
12,150
Location
London
Thanks for posting that article O very interesting, once Dutch&Dutch implement their subwoofer integration I will be experimenting with the sink method which ( naively probably) to me makes the most sense.
Keith
 

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
9285F5C1-0E25-4818-95ED-26E81FC65886.jpeg
D58DA99B-8F0D-4E3F-9651-78AE580363A2.jpeg
It probably won’t work in the way you are hoping it will. The source/sink requires one sub on frontwall and one sub on rear wall. When setup right, it will eliminate all length modes. It won’t work if you don’t have a sub on the frontwall.

It takes a little effort to set it all up. It is correct that the anti-phase sub is not time aligned to the seated position. It’s actually a little slow. The sub behind the seated position should be delayed so that the front sub (plane wave) meets the rear sub wave behind seated position. The rear sub should be in opposite electrical polarity from front sub. Once you have things setup correctly with delay and electrical polarity, you need to generate pink noise in REW and switch to a Hann window using RTA, while measuring both subs together. Then you slowly adjust the phase knob on the rear sub until the bass is flat. It’s amazing to see it in action. When it’s flat, you can then calculate the correct delay for the rear sub by add the rear sub phase rotation together with delay you already set for rear sub. I used Acourate Convolver to set it up.

Even tho the rear sub is not perfectly time aligned, overall time domain (decay) is greatly improved. So the net effect is much tighter and articulate bass than any other mono sub arrays.

Currently, I don’t use this method. I’m using 4 subs. All of them are in stereo configuration. The front subs and rear subs are cascaded by frequency to get the smoothest response. It’s kind of best of both approach because you can get smooth bass with DSP EQ and sub position (eliminate boundary interference). All 4 subs are time aligned to seated position.

So I’d say source sink is the best method if you have 2 subs and a nasty rectangular room.

The above REW screenshots are measurements of the source/sink without any EQ applied.
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
View attachment 12492 View attachment 12491 It probably won’t work in the way you are hoping it will. The source/sink requires one sub on frontwall and one sub on rear wall. When setup right, it will eliminate all length modes. It won’t work if you don’t have a sub on the frontwall.

It takes a little effort to set it all up. It is correct that the anti-phase sub is not time aligned to the seated position. It’s actually a little slow. The sub behind the seated position should be delayed so that the front sub (plane wave) meets the rear sub wave behind seated position. The rear sub should be in opposite electrical polarity from front sub. Once you have things setup correctly with delay and electrical polarity, you need to generate pink noise in REW and switch to a Hann window using RTA, while measuring both subs together. Then you slowly adjust the phase knob on the rear sub until the bass is flat. It’s amazing to see it in action. When it’s flat, you can then calculate the correct delay for the rear sub by add the rear sub phase rotation together with delay you already set for rear sub. I used Acourate Convolver to set it up.

Even tho the rear sub is not perfectly time aligned, overall time domain (decay) is greatly improved. So the net effect is much tighter and articulate bass than any other mono sub arrays.

Currently, I don’t use this method. I’m using 4 subs. All of them are in stereo configuration. The front subs and rear subs are cascaded by frequency to get the smoothest response. It’s kind of best of both approach because you can get smooth bass with DSP EQ and sub position (eliminate boundary interference). All 4 subs are time aligned to seated position.

So I’d say source sink is the best method if you have 2 subs and a nasty rectangular room.

The above REW screenshots are measurements of the source/sink without any EQ applied.

Thanks! Very very useful. Those plots look good.

But why do you think it won't work? With the D&Ds, each speaker functions as a subwoofer, firing from 10 cm into the back wall to avoid any Allison dip. When this functionality gets implemented, Martijn wrote on gearslutz that all bass below 100 can be routed to mono, so it effectively becomes one homogenous bass wave from the front wall. Shouldn't it be possible to use that as "source" (as long as they're not placed very far apart at least), and use a single sub on the back wall as sink?
 

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
Thanks! Very very useful. Those plots look good.

But why do you think it won't work? With the D&Ds, each speaker functions as a subwoofer, firing from 10 cm into the back wall to avoid any Allison dip. When this functionality gets implemented, Martijn wrote on gearslutz that all bass below 100 can be routed to mono, so it effectively becomes one homogenous bass wave from the front wall. Shouldn't it be possible to use that as "source" (as long as they're not placed very far apart at least), and use a single sub on the back wall as sink?
My understanding of what you are saying is that you believe you could do he source sink with only one sub. And the “plane wave” would be the R/L speakers.

If the source of bass is not against the frontwall it wouldn’t be a “plane wave.” You would get much less smooth bass. Also the 2 speakers wouldn’t have consistent bass.

Remember the D&D as well as Kii audio perform the cardiod bass gimmick only down to 100hz or so. The bass most folks are trying (unsuccessfully) to control is <100hz.

To the extent the D&D have bass below 100hz, it’s Omni directional.
 
OP
D

Dialectic

Major Contributor
Forum Donor
Joined
Sep 26, 2017
Messages
1,740
Likes
3,098
Location
a fortified compound
Remember the D&D as well as Kii audio perform the cardiod bass gimmick only down to 100hz or so. The bass most folks are trying (unsuccessfully) to control is <100hz.

To the extent the D&D have bass below 100hz, it’s Omni directional.

Having heard the results in the bass when my 8Cs' boundary compensation was not dialed in, I suspect that the 8Cs are doing something novel below 100 Hz with respect to front wall reflections. I had thought that the adjustment in the speakers was just an easy-to-set EQ feature.

I now think it's doing something more sophisticated than EQ alone. It may be correcting phase problems that result from using an omni bass source next to the front wall.

If I'm wrong, perhaps the D&D folks will correct me.
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
My understanding of what you are saying is that you believe you could do he source sink with only one sub. And the “plane wave” would be the R/L speakers.

If the source of bass is not against the frontwall it wouldn’t be a “plane wave.” You would get much less smooth bass. Also the 2 speakers wouldn’t have consistent bass.

Remember the D&D as well as Kii audio perform the cardiod bass gimmick only down to 100hz or so. The bass most folks are trying (unsuccessfully) to control is <100hz.

To the extent the D&D have bass below 100hz, it’s Omni directional.

All true. But with the D&D, the bass wave effectively comes from the front wall, because the subwoofers on the back of the speakers fire backwards to the front wall, and are to be placed only 10 cm from the wall. Shouldnt this work?
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,051
Likes
12,150
Location
London
Around 60Hz I believe but effectiveness does decrease, the 8C can be placed as lose as 10cm to the rear wall, it will no doubt take some experimentation, but just two front speakers and a rear ‘sink’ sub would be pretty much ideal.
Keith
 

dallasjustice

Major Contributor
Joined
Feb 28, 2016
Messages
1,270
Likes
907
Location
Dallas, Texas
Give it a try and see if it works. I doubt it will.

I’ve never seen any compelling measurements which demonstrate modal reduction below 100hz for any cardiod speaker. I can see it’s merit in reducing lateral boundary interference (100-300hz common sidewall interference range). But Kiiaudio and D&D have terrible floor bounce; one step forward, one step back. I bet the kiiaudio BXT module will eliminate the floor bounce. Maybe it can lower the bottom frequency for the cardiod response too.

I’m still waiting on some high resolution LF measurements which compare a normal full range speaker to cardiod speaker. There are tons of measurements, tests and mathematical simulations which demonstrate subwoofer efficacy. Subwoofers are way more effective (performance and cost).
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,051
Likes
12,150
Location
London
And necessary if you main speakers only go down to 60Hz.
Keith
 
Top Bottom