• Welcome to ASR. 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!

Active Room Treatment (ART) by Dirac

Did you try sideways? I get the best response from both my subs if I face the wall which is studs over block and I think that extra gap matters maybe cause my subs are long and facing out is way out from the block for SBIR. Aesthetics suck though and one sticks way out in a walkway and both show their butts which I don’t like. I catty-cornered the small one and went sideways with the big long one, they measure almost as good as butt out.
You make excellent points. Will definitely try sideways. But yeah, aesthetics suck and I'm losing that battle to my wife. Worth a try though.
 

I use per speaker target curves that have +10 dB bass rise. Flat 20-70Hz the a slope downwards to 320Hz which happens to be the transition region for my room. This slope downwards curve sounds great in my room. Dialogue is clear, and voices maintain their natural character.

ART Front Left Target Curve and Predicted Response.
View attachment 501198

ART Center Target Curve and Predicted Response
View attachment 501199

ART Front Right, Target Curve and Predicted Response
View attachment 501200

ART LCR Actual Response
View attachment 501201
A downward slope above 70Hz on a boosted curve seems to sound the best indeed. I reloaded my previous boosted curves also with the bass flat to 70Hz instead of 90Hz.
 
Looks like all of the ART-ists are pretty much set

So just not to get bored suggest we help the ones that are yet to implement ART. And to make it more fun, with budget in mind.

On my part I am not the best person to help out as have oversized system. But from what I have gathered, 2 subs adequate for the room (aka end up max 3/4 of gain after Dirac setup - middle of the bar) and 2 bookshelves with F3 at 80hz would do nicely. Let’s say target SPL is -10 to reference. Above that is where could get better but also more expensive. Channel count aside - that is obviously what people need to determine themselves.

I discounted towers in my suggestion even though I have essentially 9 towers for my beds. I feel that they are not necessary if people don’t want to push SPL to reference.

This is just a rough concept, hopefully other members can chime in to make it usable.
 
Looks like all of the ART-ists are pretty much set

So just not to get bored suggest we help the ones that are yet to implement ART. And to make it more fun, with budget in mind.

On my part I am not the best person to help out as have oversized system. But from what I have gathered, 2 subs adequate for the room (aka end up max 3/4 of gain after Dirac setup - middle of the bar) and 2 bookshelves with F3 at 80hz would do nicely. Let’s say target SPL is -10 to reference. Above that is where could get better but also more expensive. Channel count aside - that is obviously what people need to determine themselves.

I discounted towers in my suggestion even though I have essentially 9 towers for my beds. I feel that they are not necessary if people don’t want to push SPL to reference.

This is just a rough concept, hopefully other members can chime in to make it usable.

I like ART-ists.

My budget tip is building ported or sealed DIY subs with large PA drivers combined with Crown XLS amps. Building these subs obviously requires some basic woodworking skills but 2x Crown XLS1502/2002/2502 are very quiet and can feed 4 (dual) subs with more clean headroom than you'll ever need.

Bookshelves can be sufficient but their woofers should be large enough to keep up with these subs from 80Hz up.
 
Another update on my ongoing ART odyssey....

So I had an afternoon available for measurements and adjustments... first I wanted to try the Denon X4800 on its own with my speakers - and yes it can handle them without untoward behaviour (unexpected! but very much appreciated... and justifies the additional expense over the X3800...)

Of course the amplification change then required a full set of fresh Dirac measurements (13 point)... and then a series of configurations and tests...

Just so people can see my starting point ... here is a measurement without ART, with Dirac Live RC (the base product, which is equivalent to other EQ systems on the market like Audyssey, ARC, etc....)

The setup measured here is 5.2.4 - I deployed the 2nd sub in an attempt to improve the low end...

1767392017028.png


Here was my first ART measurement - using all defaults from ART
1767393148310.png

OK it is better... but honestly not all that much to write home about (the subjective improvement was audible, the objective charts less so!)

It is worthy of note that in this setup, all speakers were both being supported by, and in turn were supporting (or attempting to) all other speakers - all available filters (98) were used. - I have no way of knowing whether it needed more filters...

As a contrast, here is my latest 5.2.4 setup, after adjusting the speaker groups, adjusting frequency limits for support, and removing the heights from support duties...

1767392318697.png

I suspect the most impactful change I made was to remove the heights from support duties, which instantly freed up about 40 filters - depending on the subsequent configuration (I messed with various support groups, etc... and with subs split into seperate groups vs included in a single group) - the configurations used between 51 and 63 filters... - My suspicion is that as a configuration rule, you should ensure that you NEVER use up all the filters - leave at least 1 spare (so treat the 98, as 97 available actual)


Here however is roughly the same configuration, but with only 1 sub - so 5.1.4

1767392398435.png


Is there an advantage to having 2 subs... (when you already have 4 full range mains) - yes there is, but it is only marginal - and honestly well below the threshold of audibility. (there may be some justification at high SPL's... these measurements were taken at my standard listening level which is 75db).

So my previous theory, that with full range mains, more than 1 sub is not really needed, has been shown to be true...

Caveats: - the mains don't cover the 20Hz to 30Hz range (1 set is limited to circa 28Hz, the other to 40Hz) - and the subs I have although specced to 24Hz, really are effective only to around 26Hz.

So if I wanted to fill out the low end below 30Hz, and clean up the decay below 40Hz, I would most likely need a pair of subs as a minimum, with capabilities down to the mid teens (15Hz spec, would probably give me 18Hz actual...) - and given what I am hearing subjectively, I feel no need for additional depth or punch... even with only the one sub configuration, the results are exemplary, and audibly better than my setup has ever been....
This is a result that many custom HT builds, with room treatments etc... would be very happy with, and I have it in a shared open living space!

So my take aways for those seeking value for money performance:

1) Get full range speakers if at all possible (space, layout, WAF all factor in, I know - but in an ART setup, every full range main counts as an additional Sub in the mix!)
2) Don't use all the filters available - make sure the demand of your support schema requires less than the maximum available filters (this value may vary with differing processors, on the current generation D&M's it is 98.... so you should ensure that maximum filters used is no more than 97)
3) If you have full range mains, you may be able to get away with no sub.... but in any case, as per my included measurements 1 x 10" sealed sub does the job admirably!!
4) is more is better.... is false! - as can be seen with the default dirac measurement above... this was a 5.2.4 setup using the Dirac ART defaults, and it is absolutely outperformed by the later single sub 5.1.4 configuration...

Front Mains have 10" woofers (sealed), Surround has 4 x 5" woofers (ported), Subs are 10" (sealed)
 
Looks like all of the ART-ists are pretty much set

So just not to get bored suggest we help the ones that are yet to implement ART. And to make it more fun, with budget in mind.

On my part I am not the best person to help out as have oversized system. But from what I have gathered, 2 subs adequate for the room (aka end up max 3/4 of gain after Dirac setup - middle of the bar) and 2 bookshelves with F3 at 80hz would do nicely. Let’s say target SPL is -10 to reference. Above that is where could get better but also more expensive. Channel count aside - that is obviously what people need to determine themselves.

I discounted towers in my suggestion even though I have essentially 9 towers for my beds. I feel that they are not necessary if people don’t want to push SPL to reference.

This is just a rough concept, hopefully other members can chime in to make it usable.

For me mono bass sucks, I don’t do TV or movies just music. I think a budget way to get full range speakers with good room correction would be a Denon 3800, ART, 2 80Hz (ish) F3 stand mounts and 2 cheap subs that dig low mainly. Boom, bad ass stereo. Making subs work with little stand mounts is tough enough, ART makes it easy plus gives you the benefit of stereo subs real quick if you want. Easy enough to have a preset for mono (just messed with that yesterday) if that’s better for the TV? Add a center and surrounds as the budget allows if you like 5 channel music, or if TV is your jam.
 
For me mono bass sucks, I don’t do TV or movies just music. I think a budget way to get full range speakers with good room correction would be a Denon 3800, ART, 2 80Hz (ish) F3 stand mounts and 2 cheap subs that dig low mainly. Boom, bad ass stereo. Making subs work with little stand mounts is tough enough, ART makes it easy plus gives you the benefit of stereo subs real quick if you want. Easy enough to have a preset for mono (just messed with that yesterday) if that’s better for the TV? Add a center and surrounds as the budget allows if you like 5 channel music, or if TV is your jam.
I would tend to agree. Cheap subs are a personal choice. Unfortunately all decent subs are not cheap. For the movie crowd they should look at the spec and figure out how much they want to stretch the budget. Placing subs closer to MLP tends to save money in that respect.
 
Another update on my ongoing ART odyssey....

So I had an afternoon available for measurements and adjustments... first I wanted to try the Denon X4800 on its own with my speakers - and yes it can handle them without untoward behaviour (unexpected! but very much appreciated... and justifies the additional expense over the X3800...)

Of course the amplification change then required a full set of fresh Dirac measurements (13 point)... and then a series of configurations and tests...

Just so people can see my starting point ... here is a measurement without ART, with Dirac Live RC (the base product, which is equivalent to other EQ systems on the market like Audyssey, ARC, etc....)

The setup measured here is 5.2.4 - I deployed the 2nd sub in an attempt to improve the low end...

View attachment 501381

Here was my first ART measurement - using all defaults from ART
View attachment 501387
OK it is better... but honestly not all that much to write home about (the subjective improvement was audible, the objective charts less so!)

It is worthy of note that in this setup, all speakers were both being supported by, and in turn were supporting (or attempting to) all other speakers - all available filters (98) were used. - I have no way of knowing whether it needed more filters...

As a contrast, here is my latest 5.2.4 setup, after adjusting the speaker groups, adjusting frequency limits for support, and removing the heights from support duties...

View attachment 501383
I suspect the most impactful change I made was to remove the heights from support duties, which instantly freed up about 40 filters - depending on the subsequent configuration (I messed with various support groups, etc... and with subs split into seperate groups vs included in a single group) - the configurations used between 51 and 63 filters... - My suspicion is that as a configuration rule, you should ensure that you NEVER use up all the filters - leave at least 1 spare (so treat the 98, as 97 available actual)


Here however is roughly the same configuration, but with only 1 sub - so 5.1.4

View attachment 501384

Is there an advantage to having 2 subs... (when you already have 4 full range mains) - yes there is, but it is only marginal - and honestly well below the threshold of audibility. (there may be some justification at high SPL's... these measurements were taken at my standard listening level which is 75db).

So my previous theory, that with full range mains, more than 1 sub is not really needed, has been shown to be true...

Caveats: - the mains don't cover the 20Hz to 30Hz range (1 set is limited to circa 28Hz, the other to 40Hz) - and the subs I have although specced to 24Hz, really are effective only to around 26Hz.

So if I wanted to fill out the low end below 30Hz, and clean up the decay below 40Hz, I would most likely need a pair of subs as a minimum, with capabilities down to the mid teens (15Hz spec, would probably give me 18Hz actual...) - and given what I am hearing subjectively, I feel no need for additional depth or punch... even with only the one sub configuration, the results are exemplary, and audibly better than my setup has ever been....
This is a result that many custom HT builds, with room treatments etc... would be very happy with, and I have it in a shared open living space!

So my take aways for those seeking value for money performance:

1) Get full range speakers if at all possible (space, layout, WAF all factor in, I know - but in an ART setup, every full range main counts as an additional Sub in the mix!)
2) Don't use all the filters available - make sure the demand of your support schema requires less than the maximum available filters (this value may vary with differing processors, on the current generation D&M's it is 98.... so you should ensure that maximum filters used is no more than 97)
3) If you have full range mains, you may be able to get away with no sub.... but in any case, as per my included measurements 1 x 10" sealed sub does the job admirably!!
4) is more is better.... is false! - as can be seen with the default dirac measurement above... this was a 5.2.4 setup using the Dirac ART defaults, and it is absolutely outperformed by the later single sub 5.1.4 configuration...

Front Mains have 10" woofers (sealed), Surround has 4 x 5" woofers (ported), Subs are 10" (sealed)

Did you by chance take the temperature of your 4800 running loud for a long time with the amps working?
 
I would tend to agree. Cheap subs are a personal choice. Unfortunately all decent subs are not cheap. For the movie crowd they should look at the spec and figure out how much they want to stretch the budget. Placing subs closer to MLP tends to save money in that respect.
You said budget, my choice would be upfront budget for decent subs and upgrade the standmounts next. You would think most folks would be at least starting with a pair of speakers. Budget aside it for me would be R3 Metas and those subs that look like washing machines :D I don’t like much bass boost but I like good bass loudly. So clean with ART, it really is amazing.
 
For me mono bass sucks, I don’t do TV or movies just music. I think a budget way to get full range speakers with good room correction would be a Denon 3800, ART, 2 80Hz (ish) F3 stand mounts and 2 cheap subs that dig low mainly. Boom, bad ass stereo. Making subs work with little stand mounts is tough enough, ART makes it easy plus gives you the benefit of stereo subs real quick if you want. Easy enough to have a preset for mono (just messed with that yesterday) if that’s better for the TV? Add a center and surrounds as the budget allows if you like 5 channel music, or if TV is your jam.
Something I observed during my latest ART setup/test session - adding the center into the support mix, even at a relatively low support level (I used -5db) - had an immediate and noticeable positive impact - even for stereo where it functions only as a support speaker.

I think it is worth while considering each speaker (and the number of speakers planned) from a room/support perspective - which is why I would suggest where possible to go for full range mains - from a support perspective, you instantly add an additional sub to your setup - a pair of full rangers effectively act as 2 (smallish typically) subs...
 
Did you by chance take the temperature of your 4800 running loud for a long time with the amps working?
Only by hand feel - it is staying nicely cool - no more than warm to the touch - very happy with that!

It is in an open air placement.
 
You said budget, my choice would be upfront budget for decent subs and upgrade the standmounts next. You would think most folks would be at least starting with a pair of speakers. Budget aside it for me would be R3 Metas and those subs that look like washing machines :D I don’t like much bass boost but I like good bass loudly. So clean with ART, it really is amazing.
My Gallo TR1's are going cleanly down to around 26Hz, in a compact 10" package... (used ones go for circa AU$500) and they are a low distortion design.

But I don't listen loud...
 
Something I observed during my latest ART setup/test session - adding the center into the support mix, even at a relatively low support level (I used -5db) - had an immediate and noticeable positive impact - even for stereo where it functions only as a support speaker.

I think it is worth while considering each speaker (and the number of speakers planned) from a room/support perspective - which is why I would suggest where possible to go for full range mains - from a support perspective, you instantly add an additional sub to your setup - a pair of full rangers effectively act as 2 (smallish typically) subs...
I see your point. Perhaps we should branch out suggestions based on expected SPL. At moderate levels anything could potentially help. But at reference levels it really challenges your support speakers.

Members might be looking at our graphs but not realize how different they would look if done at 105/115dB reference level.
 
I see your point. Perhaps we should branch out suggestions based on expected SPL. At moderate levels anything could potentially help. But at reference levels it really challenges your support speakers.

Members might be looking at our graphs but not realize how different they would look if done at 105/115dB reference level.
I probably should do a test run at higher SPL's.... something for another afternoon...
 
Something I observed during my latest ART setup/test session - adding the center into the support mix, even at a relatively low support level (I used -5db) - had an immediate and noticeable positive impact - even for stereo where it functions only as a support speaker.

I think it is worth while considering each speaker (and the number of speakers planned) from a room/support perspective - which is why I would suggest where possible to go for full range mains - from a support perspective, you instantly add an additional sub to your setup - a pair of full rangers effectively act as 2 (smallish typically) subs...

I noticed the center support made a nice difference as well. A lot of the guides recommend against it and folks say it here too, I don’t agree but I don’t tell folks how to live either. I’m with you on full range mains and surrounds (the more support you have available the more tools you have at your disposal to make it whatever you want), my response up there ^^^ with stand mounts was to Oddballs game of build a budget ART set up a few posts back.
 
Also just some other food for thought. Deleted a whole bunch of Dirac/REW files as we needed some space on the Mac, but this should still be somewhat indicative although not the same cal file. The first one is with subs support engaged and the second one is subs off but other ART group support engaged.

Seems like a little messy thing without the subs. By traditional means having +/- 5dB band is great, but then subs really tighten that up.


Screenshot 2026-01-02 at 20.44.57.png
Screenshot 2026-01-02 at 20.52.10.png
 
My Gallo TR1's are going cleanly down to around 26Hz, in a compact 10" package... (used ones go for circa AU$500) and they are a low distortion design.

But I don't listen loud...
My little Klipsch is +10db at 20Hz, it was $600 new on sale. For what I consider a normal sized room it seems to me bass is somewhat cheap to obtain. Making it work is the trick.
 
I noticed the center support made a nice difference as well. A lot of the guides recommend against it and folks say it here too, I don’t agree but I don’t tell folks how to live either. I’m with you on full range mains and surrounds (the more support you have available the more tools you have at your disposal to make it whatever you want), my response up there ^^^ with stand mounts was to Oddballs game of build a budget ART set up a few posts back.
I've recomended similar things before - a used (relatively) high end floorstander from 10 or 15 years ago, can typically be picked up for 1/10th the price of an equivalent new speaker...

I'm currently using a pair of 1980's vintage B&O Penta Towers, which cost me AU$500... they were circa AU$12k new... - and have been refurbished at some point much more recently (so surrounds are in good order)

There are loads of similar examples - but it does require substantial effort in researching vintage speakers... but flagship models of 20 to 40 years ago, typically have excellent performance even by current standards, and have very limited resale value due to having been forgotten. (especially from brands that have disappeared or gone dormant....)

I chose the Penta's due to their tower/column format making them easy to fit into our living space, without consuming a lot of space, while also meeting everyones "lifestyls" appearance criteria.

If it fits into people's space I would recommend vintage 1980's Boston Acoustics speakers like the A400/200/150... Or the Revox Agora-B active speakers of the era... fans of Klipsch have the Klipschorns, La Scala, Forte etc... lots of vintage gems out there from B&W, Bose (yes...), Wharfdale, KEF, AR, Gale - and reading period reviews of those will provide references to their competitors! - Many of these will have had a refurb since new... or will be due for one - and that may need to be factored into the price.

Some years back, I put together a vintage system for a friend on a tight budget - a Quad 33/303 preamp/power, Gale 402 speakers, FM3 tuner... all of that cost less than AU$1000 at the time - matching the available performance new would have cost at least 5x that sort of money.
(but this is getting somewhat tangential to the thread!)
 
@Flak Could you please comment on the impact of distortion (THD) on ART decay control?
 
I see your point. Perhaps we should branch out suggestions based on expected SPL. At moderate levels anything could potentially help. But at reference levels it really challenges your support speakers.

Members might be looking at our graphs but not realize how different they would look if done at 105/115dB reference level.

100% agree on this one.

ART works miracles all the way from quiet stereo listening to high octane movie watching in a huge variety of room sizes and layouts. You just need enough cone area, power handling in either speakers or subs to reach the SPL and (currently limited) infrasonic extension of choice.

Other limitation is your and your partners willingness to have all that equipment "polluting" the living room.
 
Back
Top Bottom