Active Room Treatment (ART) by Dirac

[thcdru2k]

Im new to dirac. Thinking about if I could do this active crossover with the dirac PC Suite:

Is this possible to crossover with the PC suite alone, no additional hardware such as minidsps?

Pretty sure you need minidsp or trinnov to implement a custom crossover like that

 

[Punchy_Power_Cat]

Pretty sure you need minidsp or trinnov to implement a custom crossover like that

I don't need anything "custom" just a pretty standart Linkwitz-Riley Crossover around 500Hz. From the screenshots I have seen there is crossover function in the dirac PC suite. I just don't know how it is handling the speaker inputs. Can you plug multiple USB sources to one setup do does it all have to go to one dedicated USB source?

 

[Daka]

Have we reached consensus what's the best settings for Dirac ART?
Which speakers should support which groups and how big support ?

Im trying one suggestion - to not use speakers for subwoofer group support - and to use support speakers near the group - front - C, LR, FH - rear - Surround, RH
Two subs I have in the left and right corners - when I tried splitting them - it sounded noticeably worse - less impact.
Would adding 1-2 smaller subs in the rear make it better? Or its rather not worth it?

 

[PGAMiami]

Have we reached consensus what's the best settings for Dirac ART?
Which speakers should support which groups and how big support ?

Im trying one suggestion - to not use speakers for subwoofer group support - and to use support speakers near the group - front - C, LR, FH - rear - Surround, RH
Two subs I have in the left and right corners - when I tried splitting them - it sounded noticeably worse - less impact.
Would adding 1-2 smaller subs in the rear make it better? Or its rather not worth it?

The answer is unfortunately not that simple. Engineering in a way is the art of making smart compromises. The general rule should be not to over task a speaker by asking it to do more than it can, both in terms of spl and frequency. Large fronts can support subs but I wouldn’t ask them to do this below their crossover frequency, or even 40 hz. Even center and surrounds can support but above 80 or 100hz for most of these. Adding rear subs can make a very big difference as Dirac ART will use these as both actuators and absorbers, similarly to a double bass array or Trinnov Waveforming. In my case I use two sub groups. One for the front and one for the rear subs. All subs support all speakers but the amount of support is adjusted so the fronts do more for LCR and the rear do more for the rear surround.

 

[90sNintendoKid]

I correct full range in Dirac ART. I do use Magic Beans True Target which creates per speaker target curves and the Unified Sound Field versions. I live the way my systems sound and I've done this in three different rooms and I always get good results.

 

[Flak]

 

[tifune]

Dirac audio device, is that the PC/Mac audio output?

 

[Foggy1956]

I correct full range in Dirac ART. I do use Magic Beans True Target which creates per speaker target curves and the Unified Sound Field versions. I live the way my systems sound and I've done this in three different rooms and I always get good results.

This sounds interesting, is there a link on how to integrate the two?

 

[90sNintendoKid]

This sounds interesting, is there a link on how to integrate the two?

If you search for Magic Beans True Target you'll find the site. There is a Dirac export and details in the Discord for Dirac ART.

 

[mib382]

Hi! I'm a new guy here. Just wanted to thank this community for providing so many real world measurements, which reassured me that it's now actually possible to get high quality bass in a small room (without space, weight and money required for actual effective bass absorption). It was a dream of mine to have a somewhat cinema-level sound at home, but dealing with low frequency issues the classical way was never really an option for me (always moving, always renting), so I didn't even bother with subs. But recently I learned that Dirac released ART and that got me interested. Fast forward hundreds of forum pages and a few days of reading, I bought 2 Kef R3 Meta's, 2 SVS 1000 Pro subs, Denon X3800H and put my Yamaha MSP5s that I already had to the rear channels. A couple evenings of setup and measurements and now long-lasting dream is fulfilled. I still have a hard time believing what I'm hearing... What a blast.

Below are some pictures and measurements. I have a ~6.4 x 3.2 x 2.8m room. Support ranges: Front (KEF) - 50 - 150hz, Back (Yamaha) - 70 - 150hz, Subs - 20 - 150hz.

1/6 smoothing. Range managed by ART is, for all intents and purposes in a real room, flat. The PITA frequencies (150-500hz or so) are +- 5db, so I'm very happy. The rest of the range is also +- 5, so I think it's a success, especially with the consistent decay. I can't treat the ceiling, so +- 5 I think is perfectly fine.

Oh, and I have +8db low / -3db high target in Dirac, which is reflected on the graphs. It's just how I like it.

Regarding bass decay, I'd appreciate some feedback. It grows steadily as the frequency goes down, but there are no sudden peaks. Honestly, what I hear is insane to my ear already and I'm not sure whether I should bother (or even can) making it any better with the room I have.

In any case, I just wanted to share a little.
Cheers!

 

[Mort]

Fast forward hundreds of forum pages and a few days of reading, I bought 2 Kef R3 Meta's, 2 SVS 1000 Pro subs, Denon X3800H

That's a pretty good bang-for-the-buck haul for a few hundred forum pages read. Well done.

 

[Pizzaiolo]

Is there a consensus yet on whether it makes sense to have all the subs support all the overhead channels? I doubt very much, if any, bass content is being sent to them. And if I remove sub support to those channels I free up 24 cross terms that could be used for my base channels. Would I gain much by increasing the amount of support among the base channels?

After investing ~20 hours watching videos (interviews w/Joss from Dirac, Storm, Joe N Tell, Mad Audio, GH Fishing, Jason Fallacaro, etc.), 20 hours reading, and another 20 hours tweaking ART settings, this is what I have finally settled on. Most interesting things I learned:

(1) Support levels made almost no difference in my room with my speakers. I kept levels at -18 db for all speakers except subs (-24 db) and fronts supporting center (-12 db).
(2) High Frequency settings also made minimal difference. I kept at 150 Hz for all.
(3) Lo Frequency -- I set the cutoff based on 3 factors: (i) manufacturer specs, (ii) 3rd party testing, and (iii) Dirac measurements.
(4) Corrected spread was the most helpful metric in tweaking
(5) Jason Fallacaro's videos on YouTube, specifically videos #13 and #14, were the most helpful among everything I have seen or read.
(6) Biggest surprise -- the MVP speaker turned out to be my top right rear. I had enormous nulls across all of my speakers from 60-80 Hz, but somehow the location of my top right rear apparently is providing significant boundary gains that evened out the nulls superbly! Never expected a top speaker could provide such effective support to a sub, center, or front!

I have finally reached a set-and-forget state and am thrilled with the results.

Attaching my matrix of support settings and a graphic of my room showing size and layout for context.

I hope this is useful.

 

[Oddball]

Hi! I'm a new guy here. Just wanted to thank this community for providing so many real world measurements, which reassured me that it's now actually possible to get high quality bass in a small room (without space, weight and money required for actual effective bass absorption). It was a dream of mine to have a somewhat cinema-level sound at home, but dealing with low frequency issues the classical way was never really an option for me (always moving, always renting), so I didn't even bother with subs. But recently I learned that Dirac released ART and that got me interested. Fast forward hundreds of forum pages and a few days of reading, I bought 2 Kef R3 Meta's, 2 SVS 1000 Pro subs, Denon X3800H and put my Yamaha MSP5s that I already had to the rear channels. A couple evenings of setup and measurements and now long-lasting dream is fulfilled. I still have a hard time believing what I'm hearing... What a blast.

Below are some pictures and measurements. I have a ~6.4 x 3.2 x 2.8m room. Support ranges: Front (KEF) - 50 - 150hz, Back (Yamaha) - 70 - 150hz, Subs - 20 - 150hz.

1/6 smoothing. Range managed by ART is, for all intents and purposes in a real room, flat. The PITA frequencies (150-500hz or so) are +- 5db, so I'm very happy. The rest of the range is also +- 5, so I think it's a success, especially with the consistent decay. I can't treat the ceiling, so +- 5 I think is perfectly fine.

Oh, and I have +8db low / -3db high target in Dirac, which is reflected on the graphs. It's just how I like it.

Regarding bass decay, I'd appreciate some feedback. It grows steadily as the frequency goes down, but there are no sudden peaks. Honestly, what I hear is insane to my ear already and I'm not sure whether I should bother (or even can) making it any better with the room I have.

In any case, I just wanted to share a little.
Cheers!

Glad you love it. It is natural that decay grows beyond 50hz. It might not in some rooms, but it does give that "wet" bass feeling that some like. Think you are done with ART.

Dirac Live, well that part could get better. Perhaps you could try with some manual adjustments to the curve to iron out the peaks, and perhaps fill in dips gently?

 

[Oddball]

After investing ~20 hours watching videos (interviews w/Joss from Dirac, Storm, Joe N Tell, Mad Audio, GH Fishing, Jason Fallacaro, etc.), 20 hours reading, and another 20 hours tweaking ART settings, this is what I have finally settled on. Most interesting things I learned was:

(1) Support levels made almost no difference in my room with my speakers. I kept levels at -18 db for all speakers except subs (-24 db) and fronts supporting center (-12 db).
(2) High Frequency settings also made minimal difference. I kept at 150 Hz for all.
(3) Lo Frequency -- I set the cutoff based on 3 factors: (i) manufacturer specs, (ii) 3rd party testing, and (iii) Dirac measurements.
(4) Measured spread was the most helpful metric in tweaking
(5) Jason Fallacaro's videos on YouTube, specifically videos #13 and 14, were the most helpful among everything I have seen or read.
(6) Biggest surprise -- the hero in my room was my top right rear. I had enormous nulls across all of my speakers from 60-80 Hz, but somehow the location of my top right rear was providing significant boundary gains. It evened out the null superbly!

I have finally reached a set-and-forget state and am thrilled with the results.

Attaching my matrix of support settings and a graphic of my room showing size and layout for context.

I hope this is useful.

View attachment 517710

View attachment 517711

Interesting story. Looks like ART really knows what to use and when. Hope your top right rear is good for the support in that range as it is still a pretty intense in that range, especially if you support the subs. Did you look at the filter to see how stretched are they?

 

[Pizzaiolo]

Interesting story. Looks like ART really knows what to use and when. Hope your top right rear is good for the support in that range as it is still a pretty intense in that range, especially if you support the subs. Did you look at the filter to see how stretched are they?

I looked at corrected spreads, which showed a significant improvement. I haven't been able to read filters properly, as clicking the checkbox throws all 13 of them on the screen. Any tips on how I can effectively view the filter for that speaker?

 

[middlemarch]

Well, what I did...

Without thunderous support one way or the other. Since I had real limitations in terms of moving support speakers around with my default configuration I bit the bullet and removed sub support for all my overheads.

I then beefed up some of the front row support to each other, but reduced support levels to 5db, this from the Storm Audio suggestions. Dropping the sub support to the heights did not change anything that I normally listen to so I call that a win. Reducing support levels in the front row did spread the sound stage a bit, so kept that.

I then set front left and right as support to front heights, surrounds as supports to middle heights, and rear surrounds as supports to rear heights. Didn't hurt anything.

Finally, added front left and right (Studio 2s) as supports to my subs, but at 12dB and 50Hz, and my center as well, also at 12dB. My center is a 426Be, which I call my 426 Hemi (cause it's a beast). It seems to do fine at 60Hz support to the subs, four 6.5" ported woofers will do that I guess. In any event I've now managed to use 92 of my 92 available cross terms.

Finally, after much back and forth, I've decided to use full range correction for the time being. Partly because it seems to sound smoother, and partly so I can boost the treble a bit to help my hearing.

So I think I'm done for a bit. Very happy overall. And have to admit to being a bit stunned that all the tweaking really didn't change the overall sound very much over the default, certainly not night and day. The tweaking does make me feel better, but I think I'm probably just playing into my own placebo effect. Target curves are the main useful tool.

As an aside, I've had a number of back and forths with Dirac support about why I only get 92 cross terms versus the 94, 96 or 98 other D/M users report. So far just evasive answers, which I don't understand. Seems like this is a known by someone somewhere, why be evasive about it? Whatever.

 

[Oddball]

I looked at corrected spreads, which showed a significant improvement. I haven't been able to read filters properly, as clicking the checkbox throws all 13 of them on the screen. Any tips on how I can effectively view the filter for that speaker?

Spreads are probably the way to go if you want to go into the deep. Although it does not necessarily mean that it will sound better if spreads are tighter. That is up to you to decide. Spreads are across the range, so might end up being one battle after another.

FIlters just show you what they are intending to do, with ART being pretty precise in predicting the response. If you check the speaker, you will see all the filters that are going to be deployed for that particular speaker. If that particular channel is supported by 13 speakers, then you will see their individual contribution. Not sure if you would ever want that in 13 ch setup. It would though depend on your setup and what you are trying to achieve.

 

[JRowland]

Any tips on how I can effectively view the filter for that speaker?

To the right I have my mouse on front right, it is the bright red box, it highlihlghts the bright red filter line showing my right front speaker. If you move your mouse around each speaker trace sort of lights up in the view.

 

[mib382]

Glad you love it. It is natural that decay grows beyond 50hz. It might not in some rooms, but it does give that "wet" bass feeling that some like. Think you are done with ART.

Dirac Live, well that part could get better. Perhaps you could try with some manual adjustments to the curve to iron out the peaks, and perhaps fill in dips gently?

I played with EQ a little bit and I think I managed to even out a couple zones. But overall, due to seat to seat variation, EQ'ing isn't doing very wel. If I only was doing EQ for a single spot in the room - sure, but for multiple seats, consistent peaks to bring down are not that common. I think here I'd really need to treat the room more seriously, especially the ceiling, but no can do.

Here I have psycoacoustic smoothing. Yellow - before EQ, green - after. Relatively to the bass elevation the difference gets too big. I need to address that.

 

[Pizzaiolo]

So I think I'm done for a bit. Very happy overall. And have to admit to being a bit stunned that all the tweaking really didn't change the overall sound very much over the default, certainly not night and day. The tweaking does make me feel better, but I think I'm probably just playing into my own placebo effect. Target curves are the main useful tool.

As an aside, I've had a number of back and forths with Dirac support about why I only get 92 cross terms versus the 94, 96 or 98 other D/M users report. So far just evasive answers, which I don't understand. Seems like this is a known by someone somewhere, why be evasive about it? Whatever.

I was also surprised at the lack of audible impact from my tweaking. But it speaks to the strength of ART's default performance, which should be appropriate for at least 80% of users. Glad you have reached an end point with which you are satisfied.

Re cross terms, my Marantz AV 20 provides 96 terms, FYI. I suspect Dirac doesn't track implementation by manufacturer.