Active Room Treatment (ART) by Dirac

[dlaloum]

For those of us on here, not yet on the ART bandwagon, but looking to get on there in the near future....

I contacted Tonewinner via their home (chinese) website, to ask about the rumoured AT600 AVP

Their response arrived this morning

1) The AT600 is under development
2) the AT600 will support ART

The Norwegian Tonewinner distributor, has the AT600 advertised along with an ETA of December 2025 (that's only 1 month away !)

I would take the Norwegian distributors ETA with a bit of a grain of salt, as we know the sort of delays that can affect things - but it would not be surprising to see this AVP (and possibly a related AVR?) get released in early 2026.

 

[ban25]

I'll be keeping an eye out for Black Friday deals on an AV20...

 

[Oddball]

I think I am in a similar situation - being set up in an open living space, with Kitchen, Dining, Lounge/Listening and general "play" area all in one big room


I'm chomping at the bit to try ART...



Yes - I think that is the next great frontier potentially - Both Trinnov and Sony have "Virtual speaker positioning / adjustment" in software...

With a trinnov style mic who knows what could be achieved!?! - ART type capabilities might be further enhanced/maximised if the system could precisely place the speakers in space (in cars, they apparently can adjust up to 4kHz!?! - due to it being a known/controlled environment)

I would definitely like to see a step-up from Dirac, but then not sure if they will go to the niche market. What they have is already world class. Perhaps if they make enough money on D&M and hopefully other brand's implementations of their current offering including ART.

3D remapping is a very interesting concept. Should improve sub optimal speaker placement vs Dolby specs, but hopefully might also improve asymmetrical positioning of the speakers in the room as e.g. room is open to one end and closed on the other. I could use the later for sure.

Never had a chance to play with Lyngdorf but their concept of taking measurements across the whole room is also interesting. Could be something in there as well...

 

[Fidji]

3D remapping is a very interesting concept. Should improve sub optimal speaker placement vs Dolby specs, but hopefully might also improve asymmetrical positioning of the speakers in the room as e.g. room is open to one end and closed on the other. I could use the later for sure.

it is great option - especially for side/rear surrounds and tops/height speakers.
For L-C-R not so much there it all becomes bit smeared, especially if you are out of MLP [makes sense as 3D remapping tries to create phantom source in the place where it expects to see a speaker]

From my perspective - especially with today's trend of increasing TV's that push Center out of ideal position - ability to employ Dual Center [one below and one above the panel] and then manage them in a way that they appear as one phantom source in center of TV would have also significant impact on immersion level.
I have tried it with my previous center and Center Height speaker and it was cool. Now I have Center that is optimised to be placed below the screen, so no need for this anymore.

 

[Oddball]

Dual centers could be useful as you note and I would like to see that on more gear. Still a bit sceptical about it as doing a perfect mono out of two channels that might be positioned differently in height, even with the same speakers, is a tall task. In my case, top center would be almost in line with the Atmos and not sure how that would fare. But never tried it - and I would like to. I still have pretty decent spare center in the storage.

Also not sure how many people will actually go for double the cost and installation issues. If you flip the right psycho acoustic switch, and don't stare at the bottom of a large screen, you get to where you should be for the most part - at least I do.

I did have a horizontal dual center setup at one point but remember it took professional calibration and quite a bit of time to do so. Also worked best at the main MLP, a bit smeared going left and right. After a short while reverted back to a single center and barred the door that was the issue as my cleaner could not keep the toe in required. It was just an experiment for the experiment sake. Might not be that relevant to vertically split centers, but they will also need to be cleaned .

 

[kawauso]

I would definitely like to see a step-up from Dirac, but then not sure if they will go to the niche market. What they have is already world class. Perhaps if they make enough money on D&M and hopefully other brand's implementations of their current offering including ART.

3D remapping is a very interesting concept. Should improve sub optimal speaker placement vs Dolby specs, but hopefully might also improve asymmetrical positioning of the speakers in the room as e.g. room is open to one end and closed on the other. I could use the later for sure.

Never had a chance to play with Lyngdorf but their concept of taking measurements across the whole room is also interesting. Could be something in there as well...

It’s an interesting idea, but remapping alone can’t completely remove the unnaturalness in the soundstage. Just like how a phantom center can feel slightly off, remapping tends to leave a subtle sense of “disconnection” or unevenness. Honestly, I’m not sure how much real demand there is for something like that.

If the goal is a truly natural and seamless spatial experience, I think it would eventually require a huge number of physical speakers. That said, expanding ART’s operating range up to around 300 Hz—or even 500 Hz—seems perfectly reasonable, as long as it’s used within the range where it still feels natural.
More importantly, though, I really wish ART could go down to 10 Hz. Many users are struggling with how it behaves below 20 Hz. I understand it’s technically difficult, but even partial control down there would already be a big improvement.

 

[Oddball]

It’s an interesting idea, but remapping alone can’t completely remove the unnaturalness in the soundstage. Just like how a phantom center can feel slightly off, remapping tends to leave a subtle sense of “disconnection” or unevenness. Honestly, I’m not sure how much real demand there is for something like that.

If the goal is a truly natural and seamless spatial experience, I think it would eventually require a huge number of physical speakers. That said, expanding ART’s operating range up to around 300 Hz—or even 500 Hz—seems perfectly reasonable, as long as it’s used within the range where it still feels natural.
More importantly, though, I really wish ART could go down to 10 Hz. Many users are struggling with how it behaves below 20 Hz. I understand it’s technically difficult, but even partial control down there would already be a big improvement.

I would still like to see more assignable channels on the next D&M generation. Not that difficult, just requires more coding and perhaps some adjustments to the boards. More channels would also not hurt.

Best for audio are obviously large audio transparent screens, but then to get close to the quality of today's flat panels they cost much more. There is absolutely nothing in the PJ space that can get same image quality as TCL 115" max - which is approx $15K if you can still get it as last year's model.

For selfish reasons not that focused on subsonic issue, but it is a very fair point. Storm has custom PEQ that can be applied to that range, and one can also work with sub DSP to optimize, but still would be good to see some level of integration that is better than current one. I see delay skyrocketing around 20hz - could be my setup, but also see that as potentially why Dirac does not want to go into it as really difficult and would ruin their pretty graphs. These are humongously long waves and potentially there is no good way of handling them with the current software - without some specific hardware requirements, or more advanced mic.

My DBA could not handle delay well above 30hz as positioning was not perfect, but even then wonder what it could do below 20hz as only 2 out of 4 subs go there. Some members have bass arrays working to 10hz but in 8 or more sub setup and some on wall positioning. This would be Trinnov like implementation.

 

[kawauso]

I would still like to see more assignable channels on the next D&M generation. Not that difficult, just requires more coding and perhaps some adjustments to the boards. More channels would also not hurt.

Best for audio are obviously large audio transparent screens, but then to get close to the quality of today's flat panels they cost much more. There is absolutely nothing in the PJ space that can get same image quality as TCL 115" max - which is approx $15K if you can still get it as last year's model.

For selfish reasons not that focused on subsonic issue, but it is a very fair point. Storm has custom PEQ that can be applied to that range, and one can also work with sub DSP to optimize, but still would be good to see some level of integration that is better than current one. I see delay skyrocketing around 20hz - could be my setup, but also see that as potentially why Dirac does not want to go into it as really difficult and would ruin their pretty graphs. These are humongously long waves and potentially there is no good way of handling them with the current software - without some specific hardware requirements, or more advanced mic.

My DBA could not handle delay well below 30hz as positioning was not perfect, but even then wonder what it could do below 20hz as only 2 out of 4 subs go there. Some members have bass arrays working to 10hz but in 8 or more sub setup and some on wall positioning. This would be Trinnov like implementation.

The pre-outs on the A1H and A10H are fixed, but the speaker outputs are quite flexible. I actually find this really convenient — being able to assign 12 height speakers depending on Auro or Dolby formats works very well for me.
I believe the AV10 also allows flexible pre-out assignments similar to the A1H, right? Do you find any particular limitations there?

Once you get used to OLED picture quality, projection screens can feel a bit lacking. Still, sometimes size matters more than absolute image quality, and LCD panels have become surprisingly affordable lately.

As for the low end, I’d love to see Dirac push further. Even if it can’t go all the way down to 10 Hz, making it effective around 15 Hz would already help a lot.
Come on Dirac, please make it happen.

 

[Oddball]

All agreed and wise words as usual Kawauso. We want and expect more from Dirac. The only way is forward and they still have ways to go, although we are grateful for the mighty gift of ART in our small D&M universe.

I don't think that you can get 2 sets of independent surrounds or 2 centers though - out of the box. From what I understand it will be measuring the primary set and then applying the same settings to the other set of speaker/s. I never tried as never had a need for that. My room, as my rack, are full .

But it would be interesting if people tried to adjust delays and volume post calibration on a "clone" setup like second surrounds or more Atmos speakers, and report if there is ability to use separate EQ for the clone pairs.

 

[Oddball]

Oh well, I am watching Hobbit 2 and have just tried to make it more "wet" than +12 dB ART setting. Did not go through the entire movie yet but it seemed that more meaningful wet presentation came only after getting my subs up in their gain by some 15-20%. Subs are now nearly maxed out due to the one weak one in the link, but still have some dB to go.

All I can say, it was still ART, but it was not that controlled, so I guess it is what people say wet vs dry is. Not sure what are the other implications, just focusing on the low end.

 

[kawauso]

Oh well, I am watching Hobbit 2 and have just tried to make it more "wet" than +12 dB ART setting. Did not go through the entire movie yet but it seemed that more meaningful wet presentation came only after getting my subs up in their gain by some 15-20%. Subs are now nearly maxed out due to the one weak one in the link, but still have some dB to go.

All I can say, it was still ART, but it was not that controlled, so I guess it is what people say wet vs dry is. Not sure what are the other implications, just focusing on the low end.

Quick note on routing and presets that might be useful.

If you’re not using outputs like Surround Back or Front Wide, you can repurpose those terminals for another set of speakers—for example, a second center or an alternate surround pair—and switch between them. If every terminal is already in use, a simple speaker selector will do the trick.

On the A1H the pre-outs are fixed, but using them often frees up the corresponding speaker outputs. For instance, if you use pre-outs for Front L/R, the built-in Front terminals can be reassigned to heights. That makes it possible to re-route outputs without extra hardware.

Each setup can have its own EQ. Within one Dirac preset you can also use multiple measurement slots—for example, Slot 1 for the main position, Slot 2 for one meter behind, Slot 3 for two meters to the right—each can be remeasured and optimized separately.

If Speaker Config 1 & 2 plus the three Dirac slots (six total) still aren’t enough, you can save and reload profiles via USB. I even keep a USB with a “no-Dirac / loud” preset for movie nights, then switch back to my usual profile afterward.

 

[smygolf]

Oh well, I am watching Hobbit 2 and have just tried to make it more "wet" than +12 dB ART setting. Did not go through the entire movie yet but it seemed that more meaningful wet presentation came only after getting my subs up in their gain by some 15-20%. Subs are now nearly maxed out due to the one weak one in the link, but still have some dB to go.

All I can say, it was still ART, but it was not that controlled, so I guess it is what people say wet vs dry is. Not sure what are the other implications, just focusing on the low end.

In theory.....

Do a DLBC calibration, measure with REW
Then do a ART calibration, measure with REW

Then play around with support/lvl:s until the waterfall starts to get worse/longer decay.
One thing that just might work, have a target with +4dB slope in the bass.
Then after calibratio just raise the volume on all subs in the receiver.

If anyone has more insight, plz let us know

 

[kawauso]

Oh well, I am watching Hobbit 2 and have just tried to make it more "wet" than +12 dB ART setting. Did not go through the entire movie yet but it seemed that more meaningful wet presentation came only after getting my subs up in their gain by some 15-20%. Subs are now nearly maxed out due to the one weak one in the link, but still have some dB to go.

All I can say, it was still ART, but it was not that controlled, so I guess it is what people say wet vs dry is. Not sure what are the other implications, just focusing on the low end.

“So you wanted to raise the LFE rather than stop at +12 dB — nice move.
Even if the low end isn’t perfectly controlled, that’s okay. Loudness covers up a lot of sins.

 

[smygolf]

Loudness covers up a lot of sins.

Best comment Ever !!

 

[Oddball]

Quick note on routing and presets that might be useful.

If you’re not using outputs like Surround Back or Front Wide, you can repurpose those terminals for another set of speakers—for example, a second center or an alternate surround pair—and switch between them. If every terminal is already in use, a simple speaker selector will do the trick.

On the A1H the pre-outs are fixed, but using them often frees up the corresponding speaker outputs. For instance, if you use pre-outs for Front L/R, the built-in Front terminals can be reassigned to heights. That makes it possible to re-route outputs without extra hardware.

Each setup can have its own EQ. Within one Dirac preset you can also use multiple measurement slots—for example, Slot 1 for the main position, Slot 2 for one meter behind, Slot 3 for two meters to the right—each can be remeasured and optimized separately.

If Speaker Config 1 & 2 plus the three Dirac slots (six total) still aren’t enough, you can save and reload profiles via USB. I even keep a USB with a “no-Dirac / loud” preset for movie nights, then switch back to my usual profile afterward.

Have not tired, but as I said, the limitation apparently was to distance, volume and EQ settings of the clone channel. This is just what I heard, not what I verified as never had a use case for that. If D&M has "assignable" channels then all the better but also not sure why that knowledge is not more popular. You are rising the right point though, as usual.

 

[Oddball]

“So you wanted to raise the LFE rather than stop at +12 dB — nice move.
Even if the low end isn’t perfectly controlled, that’s okay. Loudness covers up a lot of sins.

Well, this is where it becomes wet or dry. I did not get into dying up the ART within its own parameters, as this was the most obvious one.

As you move your sub output (at the sub level) from what it was calibrated with ART, you do start getting the decay and fat bass. With +12dB curve on Dirac and +10dB on subs, it gets really fat on Hobbit 2. But not too hot by traditional non-ART means, at least not for the bass-heads. At the same time it seems like there is a trade-off as the whole thing is not so composed as with the ART at +12. There is some loss of detail as well...

I guess people will need to choose the wet or dry presentation they want. They are both available and very adjustable. BTW no point to REW this, it's really the ear or preference level adjustment.

Edit: I was testing at reference levels. Given ART does not have loudness adjustment it might be different at other levels.

 

[fzst]

So, yesterday I finally had some time and motiviation a to pull the mic out again to confirm the applying of the target curves via measurements. Turns out, there really was something wrong. Frankly, I have no idea what happened, my best guess would be that one of my subs didn't work during the last couple of days (maybe the auto-on feature didn't trigger..?) or maybe the something happened while I calculated a new filter (same Storm TC just with the infrabass checked) ? That's less likely though since I didn't change the original Storm TC preset w/o infra checked since I want to compare the two. No idea but the result is that I now have way more bass pressure, with a way thicker bass with the Storm TC. I guess that's gonna teach me again, never to listen to something before I cross checked via REW

Anyhow, that's how the Storm TCs look:

The waterfall actually doesn't even look that different (BTW: No idea what the shenanigans are at around 50Hz, maybe my floor vibrating or maybe outside noise):


The difference is clearly visible in the decay chart though:


So (now at least) the TC is actually working as it should and there is not really a measurable difference in bass level between the two, I still hear a difference though (albeit not as stark as before whe sth wasn't working right).

Anyhow, I didn't have to listen to those two curves for long to know that DLBC had no chance with this TC. ART is just better in every way... sure the sometimes a little too dry/pronounced upper bass is still there as well as the slightly less "full bodied" deep bass compared to DLBC but the difference is pretty much neglible with this TC applied and ART just sounds better - full stop. There is more detail and separation, better connection to the mains and first and foremost: more punch, the bass is just insanely tactile and in a way DLBC doesn't do when I simply raise the bass level up there.

Bottom line is, I'm now ready to abandon DLBC and fully commit to finding the perfect settings and TC with ART. Now I have a different problem though: too much bass I think I'll probably end up at a almost flat curve again - at least for music but this aggressive kick with ART is a lot of fun indeed

I also gave the infra- option another go and I had a little bit better luck this time:

However, in my room more bass <20Hz also means more decay and raised GD:


I'll compare the two and see if I actually hear a difference.

I also find it a bit odd that Storm’s TC curves differ between LCR and SLR,

I think I've read somewhere that Storm does roll off the highes of the SLR more to give the surrounds a more "spacious" sound... I guess since I have dipoles, this is probably counterproductive?

The default Dirac TC might actually make more sense.

Maybe you're right. I'm usually just not a fan of too much bass >50Hz but with ART, the bass is noticeably cleaner, so it's not really bothering me that much. I also want to try the Harman TC but when loading into Dirac, it looks like the Storm TC follows the high roll off of my speakers better...

Harman:

Storm:

Maybe I'll go back to just using the sliders after all...

 

[fzst]

Quick note on routing and presets that might be useful.

If you’re not using outputs like Surround Back or Front Wide, you can repurpose those terminals for another set of speakers—for example, a second center or an alternate surround pair—and switch between them. If every terminal is already in use, a simple speaker selector will do the trick.

On the A1H the pre-outs are fixed, but using them often frees up the corresponding speaker outputs. For instance, if you use pre-outs for Front L/R, the built-in Front terminals can be reassigned to heights. That makes it possible to re-route outputs without extra hardware.

Each setup can have its own EQ. Within one Dirac preset you can also use multiple measurement slots—for example, Slot 1 for the main position, Slot 2 for one meter behind, Slot 3 for two meters to the right—each can be remeasured and optimized separately.

If Speaker Config 1 & 2 plus the three Dirac slots (six total) still aren’t enough, you can save and reload profiles via USB. I even keep a USB with a “no-Dirac / loud” preset for movie nights, then switch back to my usual profile afterward.

Thx for the tip, btw: is there a way to circumvent Dirac's configuration error when trying to load filters onto the other preset where I haven't taken the measurements in? Dirac always tells me that I cannot use the project for the other speaker configuration even though it's the same... Do I really have to take new measurements to get filters onto preset 2?

 

[Oddball]

Thx for the tip, btw: is there a way to circumvent Dirac's configuration error when trying to load filters onto the other preset where I haven't taken the measurements in? Dirac always tells me that I cannot use the project for the other speaker configuration even though it's the same... Do I really have to take new measurements to get filters onto preset 2?

From what I understand you need to set your AVR/AVP into the speaker setup 1 or 2 before you try to load filters, even if they are the same configuration. If you are in perset 1 and trying to load preset 2, you get the error. Probably a reason to raise a ticket with Dirac, but I got used to it.

 

[fzst]

Once you get used to OLED picture quality, projection screens can feel a bit lacking. Still, sometimes size matters more than absolute image quality, and LCD panels have become surprisingly affordable lately.

Indeed, today's Oleds are stunning!

As for the low end, I’d love to see Dirac push further. Even if it can’t go all the way down to 10 Hz, making it effective around 15 Hz would already help a lot.
Come on Dirac, please make it happen.

Agreed, to about 15Hz would be enough imo - it would cover 99% of all content out there (maybe they could implement an easier work around for those who want to use BEQ with shakers though as well). The current 20 Hz version leaves kinda wanting indeed since A: there is quite a bit of content that actually reaches this low (considering that it actually starts to roll off even earlier, somewhere around 22-23Hz) and B: this makes kind of a mess around the whole 20Hz- region with the HPF applied and the rise in GD, decay and in some cases even FR - it would be nice seeing them pushing this firmly into inaudible levels...