Active Room Treatment (ART) by Dirac

[pogo]

two more SVS SB 1000s

When I look at the step response of SVS subs, I think they are less suitable for ART: Link

 

[napfkuchen]

When I look at the step response of SVS subs, I think they are less suitable for ART: Link

Can you elaborate on that in simple terms?

 

[pogo]

ART expects a system that plays as precisely as possible. The swing-out behaviour of the SVSs is counterproductive, regardless of whether the sub outputs a correction signal or the regular channel signal.

 

[napfkuchen]

Hmm, it's a shame, I would actually like to stick with a subwoofer with such compact dimensions. We'll see until the update for my receiver actually appears.

 

[Curvature]

ART expects a system that plays as precisely as possible. The swing-out behaviour of the SVSs is counterproductive, regardless of whether the sub outputs a correction signal or the regular channel signal.

This is nonsense.

The timescale of bass signals is in the hundreds of milliseconds and multiple seconds as you get to the lowest frequencies. It takes a long time for the ear to acquire and register a bass pitch (tens of milliseconds).

ART does not need or "expect a system to play as precisely as possible". This is obsessive tweaker thinking. Impulse response correction at the bass frequencies has a comparatively easy job because of the timescales at work.

Step response can be derived from frequency response and vice versa.

 

[Keith_W]

This is nonsense.

The timescale of bass signals is in the hundreds of milliseconds and multiple seconds as you get to the lowest frequencies. It takes a long time for the ear to acquire and register a bass pitch (tens of milliseconds).

Umm. I have a lot of respect for you, such that I think that I probably misunderstood you. A 20Hz signal = 20 cycles per second = each cycle 1/20 second = 50ms. Not in the hundreds of milliseconds range. Could I ask you to clarify please?

 

[pogo]

We'll see until the update for my receiver actually appears.

There will certainly be an improvement, but an optimum will sound different. Then it's better to do without new additional subs and replace the existing subs with ones that play precisely. But even a booming, imprecise reproduction can be pleasing.

 

[Curvature]

Umm. I have a lot of respect for you, such that I think that I probably misunderstood you. A 20Hz signal = 20 cycles per second = each cycle 1/20 second = 50ms. Not in the hundreds of milliseconds range. Could I ask you to clarify please?

Sure. My fault for assuming that it was clear that I was talking about real room measurement and correction within the context of ART.

Here's a measurement from a 2001 Genelec paper, Low-Frequency Modal Equalization of Loudspeaker-Room Responses. Notice the time is in seconds. You can see the same phenomenon in your own room measurements, using REW or the like.

 

[Curvature]

You can see the same thing in the Diract ART advertisements, but the room they represent is unusually absorbent.

 

[markus]

ART expects a system that plays as precisely as possible. The swing-out behaviour of the SVSs is counterproductive, regardless of whether the sub outputs a correction signal or the regular channel signal.

Define "precisely" and "counterproductive swing-out behaviour" at subwoofer frequencies (<80Hz).

 

[napfkuchen]

So no need to worry about step response?

 

[Keith_W]

So what is the difference between ART and a VBA?

 

[markus]

So what is the difference between ART and a VBA?

VBA tries to kill "length" modes, ART tries to minimize (the effects of) lower frequency reverberation within the listening area.

P.S. Uli presents an updated view on VBA at https://www.audiovero.de/acouratefo...611048d36ab4eb76beee1234554a26&start=30#p1637

 

[pogo]

Define "precisely" and "counterproductive swing-out behaviour" at subwoofer frequencies (<80Hz).

It should be clear which one sounds better/precise and is more suitable for co-optimisation in a room.

 

[markus]

View attachment 365644


View attachment 365648
It should be clear which one sounds better/precise and is more suitable for co-optimisation in a room.

Any verification that this is not just a measurement artefact or misconfiguration/defective?

Edit: Looks like all SVS subs show the same behavior so it's probably worth exploring what's causing this. Looks like some kind of processing within the unit to me (subsonic filter?). Anyway, group delay seems fine. Don't see why it wouldn't work with ART.

 

[Curvature]

So no need to worry about step response?

In any practical sense, not really. From a technical perspective, yes. Dirac's claim is time and phase correction, so it corrects step response anyway. How well it can do so has limits that are to do with the accuracy of measurements.

 

[pogo]

Even Dirac does not turn an SVS sub into an 'precise' Arendal, Neumann, B&W, ...sub.
That would be nice

 

[markus]

Even Dirac does not turn an SVS sub into an 'precise' Arendal, Neumann, B&W, ...sub.
That would be nice

That doesn't matter as everything is already contained in the measurements.

 

[Curvature]

Even Dirac does not turn an SVS sub into an 'precise' Arendal, Neumann, B&W, ...sub.
That would be nice

Dirac Live 2 Digital Room Correction Software Walkthrough

Dirac 2 is the latest version of Dirac’s Digital Room Correction (DRC) technology. In this article, I walk through the steps of using Dirac to optimize the response of the Purifi SPK4 demo kit that I recently reviewed here on Audiophile Style. Can we make an already great sounding speaker sound b...

audiophilestyle.com

You can also check out @mitchco's book and other articles.

 

[IamJF]

The subwoofer needs to be LINEAR so a correction algorithm can be done. So it needs to be loud enough to not compress/make noise with ports etc. - then ART should be able to do a useful correction.

From a technical perspective, yes. Dirac's claim is time and phase correction, so it corrects step response anyway. How well it can do so has limits that are to do with the accuracy of measurements.

There will be limits for low frequencies otherwise delay and processor useage would be to extensive.
Just build a bunch of closed volume woofers and be happy ...