Dirac Live Bass Control hardware requirements

[Chris Bolkan]

I am a long time lurker but this is my first post.

I am excited to try Dirac Live, especially with bass control or even better yet active room treatment when it becomes available. I've read a lot of the various threads about the (affordable) devices that currently support/will/hopefully will support Dirac with Bass Control and ART.

While looking at these DLBC and ART threads, I ended up on the Dirac site in the FAQ section and discovered that they want to see a minimum of a two octave overlap (subs can play one octave higher and mains can play one octave lower) at the crossover frequency in order for DLBC to work properly (or get better speakers). I had not seen this requirement discussed in any of the threads before.

My 2 channel system, which is also part of my home theater system, is comprised of ported pro audio mains (and surrounds) with a relatively high low frequency cutoff. The cutoff is pretty close to the frequency where one can detect location of low frequencies. The subwoofers are co-located with the mains so LF localization has not been a problem. The subs roll off naturally pretty quickly beyond 100Hz. So.....I do not have anywhere near the two octave overlap that the Dirac wants to see at the mains/sub crossover frequency.

Does this limitation preclude or severely limit Dirac with Bass Control (or ultimately ART) effectivity in my system? It's a pretty substantial investment to buy a device that supports Dirac live with Bass Control (or ART in the future) as well as buy the software, so it would be nice to know if my system stands to gain anything as it is configured.

Does anyone have experience running DLBC (or ART) with speakers that do not have a lot of overlap at the subwoofer crossover frequency?

Thanks! Chris

 

[kyuu]

Are you getting your speakers' and subs' frequency response from in-room measurements or from specs? Keep in mind that, in the low frequencies, the room dominates and usually you get bass reinforcement (along with modes and such of course). Can you tell us what speakers and subs you have?

Also, what Dirac FAQ are you looking at? Is it this guide?

 

[Chris Bolkan]

Hi kyuu,

I am aware of boundary and compression (for lack of a better term) reinforcement. My placement of the mains does not capitalize much on boundary effect for LF gain (in addition to their high LF cutoff). The mains are Tannoy V12s. I have not measured the mains in-room.

The subs are home brew infinite baffle type located in the basement under the living room. There are two manifolds co-located with the mains (so I can eventually implement stereo bass). You can see the outlets under the main speaker stands. Each manifold contains four 15 inch drivers. They do very well up to about 100hz where it seems something in the construction of the manifolds produces a fairly rapid roll off above 100hz. I have measured the subs.

The center, surrounds and height speakers are all Tannoy V8s.

Yes, that is the guide I read in the Dirac FAQ.

 

[kyuu]

You have an interesting and somewhat unorthodox setup. Cool though. Without measurements it's just about impossible to tell what kind of response you're getting from those subs, but I'd be highly surprised if it's insufficient to work for Dirac.

I'll refer to this image here:

Now, if I'm wrong hopefully someone more knowledgeable will come along to correct me. But my understanding is that the response of your mains just needs to fill in under the blue curve, and the subs the red, in order to integrate properly (with any bass management system, not just Dirac). So for a crossover at 100Hz as in the example picture, the mains don't need to have full output out to 40Hz. So for this example, if the mains were down -6dB at 100Hz, and then down -10dB at ~82Hz, that would be perfectly fine assuming the sub was doing its part.

Without measurements I'm just guessing, but I'd be willing to be bet those Tannoys have enough extension under the 70Hz rating (which I believe is its F3?), particularly after room gain, to meet up with the subs with an 80Hz crossover just fine. I mean those are 12" woofers aren't they?

If you're still unsure, you could run this all by Dirac support and see what they think. Perhaps they'll even give you a pass to return the license if it doesn't work out. I've heard of them doing that (no guarantees though of course).

 

[Chris Bolkan]

Thank you for your input Kyuu.

After I built the subs, I used a little stand alone Audyssey MultiEQ box to measure them. This was 8-10 years ago. I knew the response from the drivers in the manifolds was going to be a crapshoot, but I figured a lot of drivers would not have to work very hard to produce lots of output and it should be relatively low distortion. I figured any FR variations could be corrected with equalization.

I ran a bunch of sweeps with the Audyssey box and got what I know about the subs currently. The Audyssey software would measure the response then as far as I understood would correct the response and graph that too. (I have no idea of the filtering they used for the display of the graphs, but it gave me an idea of what I had in each scenario.)

The first sweeps were the left and right manifold close mic'd right at the mouth of the floor grates at a low level where the room would not affect the readings.

The black line was without Audyssey doing its automatic thing and the red line was 'corrected"

Then I moved the mic to the listening sweet spot and ran the scans again:

There was too much roll off on the low end for the Audyssey SW to "grab" the low frequencies and boost them up. I was told the SW looked for what it thought was a natural roll off for the speakers so as not to damage them from over boosting. So I built a little first order shelving network to boost the LF output from the speakers so the SW could "grab" the signal and work with it. This was the output close mic'd at a low level:

The shelving network enabled the SW to EQ the subs down quite a bit lower. I know they can go lower yet because of the available displacement, but this seemed like a a good compromise. Then I ran the shelved response at the listening sweet spot:

Then I ran a final sweep with both subs running in this configuration at the sweet spot:

It's interesting that both subs have a more ragged response than each individually, which is one of the things that DLBC and ART claim to correct.

I ultimately ended up not using the Audyssey box in the system at all and bought a little 2x4 mini dsp box and did PEQ filters to flatten the subs. With the 90Hz crossover at 24 db/octave from the processor, I reasoned the mains had about a 48 db/oct total roll off going on with their natural roll off added to that of the processor, so I threw an additional 24 db/octave LP on the subs with the Minidsp just to try and match up the slopes and that is how I have used it for a long time. I recently removed the minidsp box altogether from the system just running the processor 24db/octave crossover on the mains and subs and I actually prefer the sound, so go figure. The subs are pretty ragged un-eq'd but they still sound good to me, just different.

Anyway, with the advances of DLBC and ART, I am thinking it is time to revisit my system and see if I can get better overall sound by the benefits of impulse response correction and hopefully reducing room nodes at low frequencies. I do have a few of those! I just don't understand well enough how important a lot of overlap is at the sub crossover frequency in the function on the DIRAC SW.

 

[Chris Bolkan]

You have an interesting and somewhat unorthodox setup. Cool though. Without measurements it's just about impossible to tell what kind of response you're getting from those subs, but I'd be highly surprised if it's insufficient to work for Dirac.

I'll refer to this image here:

View attachment 389080

Now, if I'm wrong hopefully someone more knowledgeable will come along to correct me. But my understanding is that the response of your mains just needs to fill in under the blue curve, and the subs the red, in order to integrate properly (with any bass management system, not just Dirac). So for a crossover at 100Hz as in the example picture, the mains don't need to have full output out to 40Hz. So for this example, if the mains were down -6dB at 100Hz, and then down -10dB at ~82Hz, that would be perfectly fine assuming the sub was doing its part.

Without measurements I'm just guessing, but I'd be willing to be bet those Tannoys have enough extension under the 70Hz rating (which I believe is its F3?), particularly after room gain, to meet up with the subs with an 80Hz crossover just fine. I mean those are 12" woofers aren't they?

If you're still unsure, you could run this all by Dirac support and see what they think. Perhaps they'll even give you a pass to return the license if it doesn't work out. I've heard of them doing that (no guarantees though of course).

Oh and yes, My thinking was that as long as there is enough overlap in frequency response to create the curve you have shown above, whether it is from natural rolloffs or electronic filter networks or a combination of both, everything should work out. That is why my lack of understanding about the requirement for 2 octaves of overlap. Dirac seems pretty insistent on that unless I misinterpreted what two octaves of overlap means. I suppose I am OK if I look at it through the lens of do the speakers have ANY output in the crossover zone as opposed to being flat and octave above and below. I'm so confused.

 

[kyuu]

Yeah I don't believe Dirac meant you need flat frequency extension two octaves below and above the crossover, that would be pretty nuts and disqualify a lot of systems. Like probably most of them. Your posted measurements from Audyssey are very smoothed and as typical with Audyssey the "results" graphs are aspirational rather than reflecting reality. Despite that, I think they still show enough to conclude that you are almost certainly OK to use Dirac.

I still think you should confer with Dirac directly if you want to be absolutely sure, and possibly ask if you can try it out before committing. If you are able to use a PC as a source, you can definitely use a trial version of Dirac on the PC to see how well it does.

You might also consider investing in a UMIK-1 and learning how to use REW to take proper measurements.

 

[Chris Bolkan]

Thanks kyuu, I will contact Dirac and see if I can get clarification regarding the one octave above and below the crossover frequency. I do have a UMK-1 (not the USB version), an external ADC that I used with it and very old copy of REW on a laptop. I found the whole setup cumbersome to use. I had to fart around with ADC levels out of the ADC into the computer and frankly I just gave up using it. Like the newer correction SW, I am sure a new version of REW and a USB version of the UMK-1 mic might be less fiddly and easier to set up and use. It's been a lot of years. I will get the USB version of the mic and update the software and give REW another try.

 

[Chris Bolkan]

I contacted Dirac. It took a while to get an answer because the initial person who looked at my question did not know the answer and ran it up the food chain. Here is the correspondence:

Hi!

I am currently a Dirac live customer running Dirac on an NAD T-758. My system is getting long in the tooth and I have been looking to upgrade once ART becomes more available. I believe my room would benefit significantly.

I will most likely be looking at the Denon/Marantz family of hardware......something that can support multiple subwoofers.

I have been trying to learn as much as I can about ART on Audio Science Review and links from there as well as your own FAQs.

While looking at this FAQ document on your site https://mehlau.net/audio/dirac-live-2-avr-crossover/ I see that you want the subwoofer and mains to have two octaves of overlap. (subs can play one octave higher and mains can play one octave lower than the crossover frequency.)

My system is unable to do this. My mains -3dB point is spec'd at 70 Hz. My subs roll off pretty quickly above 100Hz. The mains are ported and have a higher roll off slope than the subwoofers which are infinite baffle. The FAQ says to get better speakers if the system cannot overlap by two octaves. I like my speakers and the sound of the system and do not want to replace them, but also need to know if this lack of overlap precludes successful implementation of ART.

There is some detail about my system in this link where I asked on ASR, but no one really seemed to know the answer if ART should benefit my system. https://www.audiosciencereview.com/...ive-bass-control-hardware-requirements.56588/

I appreciate any feedback you can provide. I am excited to take my system to the next level of bass control if it looks like ART will work.

Best regards,

Chris Bolkan


First reply:

Flavio Fellah commented:

Hello,

thanks for reaching out and for your interest in Active Room Treatment.

b.t.w. this page isn’t ours: https://eur01.safelinks.protection....1EW+fUtlGNDzD9EgYAXssXYWzrMMD5Ps4=&reserved=0

I have no sufficient hands-on-experience with ART to answer your question so I’ve escalated your ticket,

it may take time though… sorry!

Kind regards


Then the answer from who he forwarded the question to:

Glenn Liljestrand commented:

Hello Chris,

Sorry about the wait, and thanks for the interesting questions. You have built a very cool audio rig there! ART would work fine with the diversity it gets between the two subs and between the main speakers, respectively. A larger frequency range overlap would allow more diversity to be leveraged also between subs and main speakers.

Bass Control will also work but obviously be slightly limited in the choice of crossover points, due to the limited frequency range where both subs and main speakers can operate.

All in all, I think Dirac Live will work great compensating your system. Building custom audio rigs is hard to get right, and Dirac Live is more than capable to compensate out the linear imperfections.

Best regards, let me know if there’s anything else!

// Glenn

I am pleased with the answer! I also bought and received the USB version of the UMK-1 and will be downloading the latest version of REW and re learning that tool to do system measurements.

 

[Chris Bolkan]

I've been thinking about what Glenn from Dirac said and I think it may be sinking into my thick skull why an overlap is very beneficial (even though not mandatory like the guidance document led me to believe). It must be that they can fiddle with all speakers relative to the subs in the overlap region (which is probably a critical FR area for room nodes in most systems) to help null peaks and bring up low spots. In my system only the subs will be able to to work with/against each other to smooth the response below the crossover and only the mains and surrounds will be able to work with and against each other to smooth above the crossover with maybe a small overlap region for all of the speakers to help with smoothing. With a broad overlap, they can use ALL of the speakers to help smooth the response. I sort of feel like well duh.........why didn't I get that at first? But I also see no reason why ART should not work well in systems without a ton of overlap at the crossover frequency with more subs being better than less.

 

[b**tlover]

Hi kyuu,

I am aware of boundary and compression (for lack of a better term) reinforcement. My placement of the mains does not capitalize much on boundary effect for LF gain (in addition to their high LF cutoff). The mains are Tannoy V12s. I have not measured the mains in-room.

The subs are home brew infinite baffle type located in the basement under the living room. There are two manifolds co-located with the mains (so I can eventually implement stereo bass). You can see the outlets under the main speaker stands. Each manifold contains four 15 inch drivers. They do very well up to about 100hz where it seems something in the construction of the manifolds produces a fairly rapid roll off above 100hz. I have measured the subs.

The center, surrounds and height speakers are all Tannoy V8s.

Yes, that is the guide I read in the Dirac FAQ.

View attachment 388865View attachment 388866View attachment 388867

Hi, Chris! I am excited by your setup, especially cause I plan to use Tannoy pro speakers (VX or VXP12 as LR, Tannoy PCI line for CC and Atmos in-ceiling) for my planning media lounge with Atmos. And hope to get Dirac ART to manage it too. Please, share the info about your whole equipment? How did you find V12's for music and HT? What speakers applied as surrounds? As far as I understood NAD T-758 is managing your HT? No external amps? Is it enough to feed your V12's? How do you power your amazing subs?

 

[Chris Bolkan]

Hi b**lover,

Thank you for the kind words.

I am very happy with the sound of the Tannoy pro audio speakers for both 2 channel listening and as theater speakers. I do not have an ideal room for listening or theater, but I am not disappointed in the sound. There are almost always things I like better and things I don't like as well when I listen to "higher end" speakers, but I haven't changed in well over 10 years. I cannot imagine a better bang for the buck speaker. When I decided to try and outfit my room with multiple speakers, more speakers cost more money, especially if you want them all voiced the same as the mains. Used Tannoy pro audio speakers were very inexpensive for what I got in terms of performance. I bought them all on EBAY over time and have just added more speakers as Dolby evolved. I was very particular though when buying because used condition can vary a lot.

I have always gravitated towards speakers that approached sound reproduction from a theoretical point, line or plane perspective. Each has a unique character as to how they fill a room with sound, and the Tannoy speakers do a good job at representing a point source, at least from the listening spot and image wonderfully.

Currently my mains are V12s and the center, surrounds and height channels are all V8s. They can play very loud without strain in my space and they are pretty small (relatively speaking) so they tick a lot of my boxes. One thing is for sure. These speakers have WAY too high of a LF roll off to enjoy listening without LF augmentation. They will sound cheap and tinny on their own and I am sure no one would consider them if they heard them without LF augmentation. However, I think they more than hold their own when paired with good woofers.

I have had my NAD T-758 since when it first came out as the budget ATMOS option. It has been OK, but after reading the scathing review and lowest performance ratings of ANY processor on this site, I honestly can't wait for it to be gone. I would much prefer an objectively decent measuring processor, especially since I don't have the luxury of separate two channel and theater systems, but for now it is what I have.

I do not drive any speakers directly with the NAD. I have a little "amplifier closet" that contains separate amps. The NAD is used only as a pre-amp. I am a big fan of old Mackie M-1400 pro audio amplifiers. I bought all of mine used on EBAY over time. I am a fan because they are beautiful fully cross coupled complimentary circuitry which is music to my eyes, have nice big toroid power transformers, are spec'd nicely and can drive ANY load without a hitch. I think of them as very powerful high end audiophile amps with loud obnoxious fans. Mackie must have had a big contract to produce laboratory amplifiers at one time, because the M-1400 schematic shows a way to jumper and bypass all of the input circuitry (filters, EQ and so forth) and have just a super clean amp. It requires disassembly of the amp to do the mod but it is worth it and I have done it to all of mine. Plus its a good time to replace the main filter caps and give the amps a good cleaning and lookover before putting them into service.

For the subwoofers, I am not as picky circuit wise, so am using a newer Mackie FRS-2800 amp. It has enough power to drive the subs and is very stable into any load. It is operating off 240V since it is doing the most work driving the subs.

To power everything, I ran a dedicated 30A 240V circuit into the amp closet, and I found this very cool industrial server sequencer on EBAY that I wired to use the trigger signal out of the processor to sequence all of the amps so they don't turn on all at once and blow the breaker. Here's a pic of the amp closet.

Did this answer your questions?

 

[Chris Bolkan]

Hi again b**lover,
It occurred to me that I did not answer one of your questions. I am fairly certain that the NAD would drive all of the speakers without external amplification in most all "normal" listening situations and even play quite loud. Especially the mains since they are about 100 dB efficient. However, I am also certain that there are times I have driven my overall system to levels that the NAD simply could not have supported. It's fun to do that sometimes. The Mackie amps are capable of supplying the full rated power of the speakers without clipping.

 

[b**tlover]

Hi again b**lover,
It occurred to me that I did not answer one of your questions. I am fairly certain that the NAD would drive all of the speakers without external amplification in most all "normal" listening situations and even play quite loud. Especially the mains since they are about 100 dB efficient. However, I am also certain that there are times I have driven my overall system to levels that the NAD simply could not have supported. It's fun to do that sometimes. The Mackie amps are capable of supplying the full rated power of the speakers without clipping.

Thank you for detailed info! what are dimensions of your room? As far as I learn NAD T758 is about 110W on 8 Ohm. Since V12 sensitivity is 97dB means your front speakers are able to do 117 dB from receiver each. How it is in real? I explain my question, want to understand if 150W/ch will be enough for room about 42 sq.m (about 452 sq feet)

 

[Chris Bolkan]

Hi b**lover,

My "room" is our main living space that includes the living room, dining room and kitchen. It is all open. The living room part is about 24 X 24 feet but the dining room attached puts the back wall at maybe 35 feet? with Kitchen and entry off to the side. The width from the Kitchen/entry to the other dining room wall is also about 35 feet. So now that I think about it, the entire space is about 35 X 35 with a walled office in one corner of the square. It's a pretty big volume total and the living room ceiling is vaulted.

I suppose what is most important to your question is that the couch is about 12 feet from the speakers. I have never run the speakers off the T-758 but if it is capable of an honest 110WPC, 2 channels driven, I would guess it can get to within 3-5dB of the Mackie amps I am using because they are rated at a clean 250wpc into 8 ohms and somewhere around 300 at clipping.

So this is all guess, but it seems like 150wpc will get to within 3db of Mackie amps into 8 ohm before clipping. I doubt very much that my speakers have ever seen clean peaks of more than 150 watts while I have owned them. It would just be too loud (for me) at 12 feet from the speakers.