Active Room Treatment (ART) by Dirac

[dlaloum]

But this is a 2 channel number right, will is do this with 5 channels full range and 4 heights is the Q?

AVR's are always power supply constrained.... if you hit it with a bunch of full range speakers and have impedances that dip below 3 ohm in the bass, most AVR's will get into trouble - they haven't got enough current to cope.

So yeah the number of channels, as well as the speaker specs (impecance) etc... all factor in.
if the speaker was previously crossed over before the power amp circuit (so you weren't using its bass capabilities) then most likely in a full range config, it will be more demanding.

But it really depends on the speakers! Taking my front L/R off the AVR and putting them on an external power amp cleaned up all the channels - the AVR was not coping with the 1.6ohm minimum impedance... (and yes I run my fronts full range - they are specced down to 24Hz)

 

[DutchJay]

AVR's are always power supply constrained.... if you hit it with a bunch of full range speakers and have impedances that dip below 3 ohm in the bass, most AVR's will get into trouble - they haven't got enough current to cope.

So yeah the number of channels, as well as the speaker specs (impecance) etc... all factor in.
if the speaker was previously crossed over before the power amp circuit (so you weren't using its bass capabilities) then most likely in a full range config, it will be more demanding.

But it really depends on the speakers! Taking my front L/R off the AVR and putting them on an external power amp cleaned up all the channels - the AVR was not coping with the 1.6ohm minimum impedance... (and yes I run my fronts full range - they are specced down to 24Hz)

Well, I'm driving Arendal center and some monitor audio apex 10 speakers with it, they don't dip below 4 ohms, so prob will be ok.

Cheers

 

[kawauso]

I tested Dirac ART in a setup consisting of 5 Denon AVR-A1H units handling the base layer, 4 subwoofers, and 8 top-layer (height) speakers.


Sub-20Hz output is rolled off, which is a limitation of ART at this stage.
The frequency response became smoother, distortion decreased, and low-frequency RT60 was significantly shortened.


While I cannot claim an objective improvement in sound quality over Bass Control, the differences were clearly audible.
ART also produced more stable and repeatable results, requiring less manual adjustment compared to Bass Control.


REW measurement data (.mdat) is available here:
https://drive.google.com/file/d/1vsJfFTi2eHLE4mlAcyGhY6LPSiLvPpGJ/view?usp=sharing

 

[DutchJay]

Another subjective review after listening to my go to test traks, movie demos and standard movie clips: so this is what REAL tight bass sounds like . Much more impact and definition, therefore it doesn't overshadow/bleed into the other frequencies. Nicely done Dirac!

 

[kawauso]

I noticed a visible dip around 90Hz in the frequency response under Dirac ART.


When I adjusted the number of support speakers included in the bass group, the response changed noticeably.


It would be ideal if the default configuration provided the optimal balance, but further evaluation is ongoing.

 

[Gadgety]

I tested Dirac ART in a setup consisting of 5 Denon AVR-A1H units handling the base layer, 4 subwoofers, and 8 top-layer (height) speakers.


Sub-20Hz output is rolled off, which is a limitation of ART at this stage.
The frequency response became smoother, distortion decreased, and low-frequency RT60 was significantly shortened.

I noticed a visible dip around 90Hz in the frequency response under Dirac ART.

When I adjusted the number of support speakers included in the bass group, the response changed noticeably.


It would be ideal if the default configuration provided the optimal balance, but further evaluation is ongoing.

Very interesting to see the curves, even if the initial one didn't look that impressive. Detailing the changes, from what, to what, in terms of your support groupings would be interesting to know, as well as info about your room. I've seen photos of two of your set ups and browsed through your posts to understand a little about your system.

The way I understand ART it thrives on more support speakers, scaling it's performance. Variation in their positions enables the system to both sort out what is "noise" in the processing of the measurements, and also pull more correction levers in the correction.

Did you say you're using 5 x Denon AVR-A1H units? That's a whole lot of channels. What makes you need that many?

 

[Gadgety]

My initial (10 minute) subjective impressions were really interesting. The first thing I noticed with ART is that the bass now sounds like it's coming locked in from the center channel. Swapping back to DLBC only, the bass feels spread out and slightly uneven. This was surprising because I was already quite satisfied with DLBC's integration of my 3 sub setup (versus what I could tweak with MSO/Audyssey previously), but ART is producing a bass image that reminds me of listening through headphones. Outside of imaging, the bass sounds less flabby and, again, more headphone-like.

I need to try this out more with music and movies, but I'm pretty excited so far that even with smaller Atmos/surround speakers, I can tell a subjective improvement. Maybe the number of support channels helps more than the size of each channel? And I wonder, would ART correction run out of steam if the volume increases beyond what the support channels can handle without distortion?

Anyways, I will have to do the wife listening test later to see if even she can tell. I'd also be curious to try measuring the ART changes with REW/UMIK-2, but I'm not sure what the easiest way would be to run measurements through the AV20.

Thank you for sharing your initial impressions. Would be interesting if you tried material, particularly bass, that is explicitly not mixed in the center channel, such as Simon & Garfunkel's Mrs. Robinson, to notice what happens then.

As for ART performance Dirac states somewhere that performance scales with the number of support speakers. Interesting it uses even your limited woofer range surround speakers, which according to the manufacturer present a typical in-room bass response of 90 Hz (-6dB). They are within the range.

 

[kawauso]

Very interesting to see the curves, even if the initial one didn't look that impressive. Detailing the changes, from what, to what, in terms of your support groupings would be interesting to know, as well as info about your room. I've seen photos of two of your set ups and browsed through your posts to understand a little about your system.

The way I understand ART it thrives on more support speakers, scaling it's performance. Variation in their positions enables the system to both sort out what is "noise" in the processing of the measurements, and also pull more correction levers in the correction.

Did you say you're using 5 x Denon AVR-A1H units? That's a whole lot of channels. What makes you need that many?

Actually, I’m using a single AVR-A1H in a 5.4.8 configuration (5 base layer channels, 4 subwoofers, and 8 height speakers).
The mention of 5 AVR-A1H units was a mistake. ‍


As for support speaker configuration, it seems that more is not always better.
The default setting didn’t sound bad, but it introduced a dip in certain frequencies, so I'm still experimenting.


Right now, having only the subwoofer group as the support seems to produce the best result.

 

[Gadgety]

Thank you for the clarification. Now that more users will have ART it'll be interesting to learn about how it's configured and works in different set ups, so your detailing of the support configuration is intriguing. What are your room dimensions?

 

[Daka]

Fun discovery: when playing stereo music - all speakers specified in support groups will play also.

Must admit yesterday after applying ART - it didn't sound particularly outstanding - perhaps thought the difference will be more apparent without need to A/B.
Today though after initial hour with music and doing A/B with BC - I definitely hear the difference - considerable difference. Feels like impossible I did enjoy BC earlier.
But what it did to atmos music is outstanding. Normally I would enjoy it to some extent but still stereo would feel cleaner, tighter, less muddy. Well now with ART there is no mud in atmos tracks - the way sound blends now from all the channels - makes them disappear altogether and you feel being in a big room/hall rather your room with multiple speakers. Did not expect that

 

[Gadgety]

Fun discovery: when playing stereo music - all speakers specified in support groups will play also

That's the MI-part of MIMO and is fundamental to how ART works. It'll engage the resources it has been allocated as support, within their limits. I believe this is also why it's useful to allocate speakers to a support group where the other speakers in that support group have similar bass capability, or else they may limit the more capable speakers in that support group, and hence the overall support capability.

 

[mxmzd]

Thank y'all for the first impressions. Setup doesn't sound very straightforward so far.

I will probably wait until BF hoping for a deal, bc I already have DIRAC Live + BC. I've got KEF R7 Meta in the Front, KEF R5 in the Rear and 4 15" Subs, so it should be a quite ideal setup I hope.

 

[Curvature]

Group Settings
Group 1 (L/R):

• Group 1 → –6 dB (going higher can cause directivity issues)
• Group 3 → –6 dB (going higher can cause directivity issues)
• Group 4 → –20 dB (I tested between –18 and –22 dB; more than that can also affect directivity)

Group 2 (Center):

• Group 1 → –6 dB (going higher can cause directivity issues)
• Group 4 → –20 dB (I tested between –18 and –22 dB; more than that can also affect directivity)

Group 3 (Surrounds):

• Group 1 → –6 dB (going higher can cause directivity issues)
• Group 3 → –12 dB (a bit more than L/R since the surrounds are further from the sub, need more help)
• Group 4 → –18 dB (sometimes +12 dB can work better, for example with bass-heavy content like cars, coming from the rear, you might ear engine noise from the back surround but also from the front subwoofer which is destroying the localization and unwanted)

Group 4 (Subwoofer):

• Group 1 → –18 dB

What do you mean by "directivity"? Do you mean localization, where the sounds appear to come from?

 

[Curvature]

I noticed a visible dip around 90Hz in the frequency response under Dirac ART.


When I adjusted the number of support speakers included in the bass group, the response changed noticeably.

Do I understand the graph correctly, that you obtained the best result with only a single support speaker? (The red trace.)

 

[Frage]

What do you mean by "directivity"? Do you mean localization, where the sounds appear to come from?

Yes

 

[Curvature]

Yes

Thanks. Is the effect on localization mostly negative or more of a difference in kind? Are the placements of sound primarily impacted, their clarity, envelopment or sense of reverberation?

Curious about what to expect when changing those settings.

 

[jtatknox]

I noticed a visible dip around 90Hz in the frequency response under Dirac ART.


When I adjusted the number of support speakers included in the bass group, the response changed noticeably.


It would be ideal if the default configuration provided the optimal balance, but further evaluation is ongoing.
View attachment 479955

Thanks for these measurements. Check out your waterfall plots for the real difference - it's not so much FR but decay.

Check out this parallel thread, a different user posted:

R

Post in thread 'Dirac ART is now running on beta FW for Denon Xx800H AVRs!'

Oct 2, 2025

Ok, so I finally did some listening tests and as picky as I am, there are some issues that I have with ART (atleast in my room). But there are also some really good things that I have to write about it first.

I have not taken my time to remeasure everything. I have used my latest dirac file and switched from DLBC to ART.

So the first thing that I have noticed is the "tighter" bass. Some songs with low end bass around 20-30hz sound much better to me. That is truly amazing. However, I already liked the result of DLBC and the addition of 2 extra subwoofers (switching from 2 to 4) was...

• R4pture

• 

 

[jtatknox]

If anyone is doing measurements and has time... I would love to see the relationship between support level and the bass decay in a waterfall plot. In other words, we are told -18 dB is ideal but it would be fun to see the relationship between bass reverberation in the room and self-support level in a multi sub group.

 

[Oddball]

I can already see that for demanding users and bigger systems, ART will be a nice toy to play with for a while...

 

[kawauso]

Do I understand the graph correctly, that you obtained the best result with only a single support speaker? (The red trace.)

I'm still learning how ART works, but in my current setup:

• Group 8 (the subwoofer group) is supported by L, C, R, SL, and SR.
• All other groups (1–7) are supported only by the subwoofer group.

There are four subwoofers, each placed in a corner of the room.