Dirac Live Bass Management?

[ad_fletch]

Darn, so over $4k for multiple sub control. Not (yet) for the likes of me I guess....

 

[markus]

Latest talk about dirac bass management:

The usual marketing show from these guys. Unnecessary lengthy and old but good basic introduction from Jacob what Dirac Live Bass Control is. Bass management is something else but it's also part of Dirac's processing when you own DLBC.

My 2 cents:
08:42 – "It is not possible to control the variance using minimum phase filters". That's incorrect and easy to test using MSO.
30:37 – "The result is still not quite what we would like it to be... They also don't add up well in the cross over region". Thanks for proving that "sounds good to me" is just the usual audiophile self-deception only good for marketing or why has Dirac Live been considered "state-of-the-art" in the past when they themselves have been "doing it wrong" all those years?
32:00 – "For stereo content it can be expected low frequency material is correlated". And that's exactly why you should NOT apply differing all-pass filters to different channels, especially speaker channels that are used for phantom imaging.
35:54 – Best moment in the whole video. 3 guys having no freaking idea what that Swedish guy is talking about
56:38 – "Ideally it [the impulse response] should be a Dirac impulse". In that case you would have removed the room response completely, i.e. you've created an anechoic chamber. Certainly NOT what we want.

 

[Dathzo]

Maybe I'm misunderstanding what you're asking but if you have 2 subs why wouldn't you want Dirac Live to correct them? That's a frequency region where EQ can be most beneficial.

Hi Markus,

Sorry, maybe confusing indeed, I do want the subs corrected: I own a NAD C658, where I can have different configurations. For example:

Configuration 1

a) L and R speakers connected to balanced preamps
b) Subs 1 and 2 connected to Sub outputs 1 and 2
c) With BluOS, I would select 2 subwoofers so the C658 will bo the bass management and establish the crossover
d) DIRAC will see the system as 2.2, hence it will measure and correct L, R, Sub1 and Sub2 channels

Configuration 2

a) L and R speakers connected to balanced preamps
b) Subs 1 and 2 connected to unbalanced preamps
c) With BluOS, I would select no subwoofers so the C658 will send full signal to speakers and I would use the LPF on the subs themselves to get only the low frequencies
d) DIRAC will see the system as 2, hence it will measure and correct L, R channels (subs would be playing during the sweeps of course and they will be corrected)

My question is if DIRAC can handle well Configuration 2, and time-align the low frequencies when only doing L and R sweeps (with subs playing of course). The reason I even think about Configuration 2, is i) it has been suggested that when having subs connected to different outputs, DL does not do a good job and ii) to remove the distortion caused when Bass Management is active in the C658.

Thanks.-

 

[markus]

Hi Markus,

Sorry, maybe confusing indeed, I do want the subs corrected: I own a NAD C658, where I can have different configurations. For example:

Configuration 1

a) L and R speakers connected to balanced preamps
b) Subs 1 and 2 connected to Sub outputs 1 and 2
c) With BluOS, I would select 2 subwoofers so the C658 will bo the bass management and establish the crossover
d) DIRAC will see the system as 2.2, hence it will measure and correct L, R, Sub1 and Sub2 channels

Configuration 2

a) L and R speakers connected to balanced preamps
b) Subs 1 and 2 connected to unbalanced preamps
c) With BluOS, I would select no subwoofers so the C658 will send full signal to speakers and I would use the LPF on the subs themselves to get only the low frequencies
d) DIRAC will see the system as 2, hence it will measure and correct L, R channels (subs would be playing during the sweeps of course and they will be corrected)

My question is if DIRAC can handle well Configuration 2, and time-align the low frequencies when only doing L and R sweeps (with subs playing of course). The reason I even think about Configuration 2, is i) it has been suggested that when having subs connected to different outputs, DL does not do a good job and ii) to remove the distortion caused when Bass Management is active in the C658.

Thanks.-

Option 1 appears best.

Does the 658 allow DLBC with 2 independently addressable subs? If yes then do that. If not then make sure DL treats both subs as a single sub (cluster) and not as 2 independent subs.

 

[Sal1950]

AVR-5 and AVR-10 are a max of one independent subwoofer output.

I can see little value in paying $499 for DLBC if you only have one sub channel.
DLBC appears to be a very powerful piece of software with some amazing capabilies but to have value it must be combined with like capable gear.
Outside of the extreme high end of processors only a couple besides D-M products have discreet multi sub channels.
DLBC is going to require a whole new line of gear.
Buyer Beware, there have been many AVRs with 2 sub out connections but they only have 1 channel of control.

 

[Krobar]

I can see little value in paying $499 for DLBC if you only have one sub channel.
DLBC appears to be a very powerful piece of software with some amazing capabilies but to have value it must be combined with like capable gear.
Outside of the extreme high end of processors only a couple besides D-M products have discreet multi sub channels.
DLBC is going to require a whole new line of gear.
Buyer Beware, there have been many AVRs with 2 sub out connections but they only have 1 channel of control.

I mentioned it a couple of posts ago, it is $325 for the AVR-5 to go from no Dirac to Full Dirac with Bass Control.

Agree that Dirac BC is probably only worth it if you have multiple subs normally (The single sub discount is not significant) but the AVR5 price is a different matter.

 

[Kal Rubinson]

Outside of the extreme high end of processors only a couple besides D-M products have discreet multi sub channels.
DLBC is going to require a whole new line of gear.
Buyer Beware, there have been many AVRs with 2 sub out connections but they only have 1 channel of control.

That describes the D-M (and most other brands) which only allow independent level and delay settings for the 2 sub outputs. So far as EQ (Audyssey) is concerned, there is only one sub channel.

 

[Sal1950]

That describes the D-M (and most other brands) which only allow independent level and delay settings for the 2 sub outputs. So far as EQ (Audyssey) is concerned, there is only one sub channel.

Sorry Kal but you have me confused with your post.
Many AVR's have included 2 sub out connections but they only provided the exact same signal to both. This just removed the need for the user to buy a Y cord. My older 7701 was this way and I believe your old 8801 or 8802 was the same.

The later D-M products have 2 separate channels driving the 2 connections as in my 7703 and the AV8805 you currently own, and yes they do provide independent level and delay settings to the 2 sub outputs. If I understood Jacob from Dirac correctly during the first 10 minutes he explains how DLBC does basically the same. As he mentions just past the 7:00 minute mark "adding flat to flat wouldn't = flat. The sub FR correction and integration with the rest of the speakers is done later with all subs combined as a whole.

 

[Kal Rubinson]

Sorry Kal but you have me confused with your post.
Many AVR's have included 2 sub out connections but they only provided the exact same signal to both. This just removed the need for the user to buy a Y cord. My older 7701 was this way and I believe your old 8801 or 8802 was the same.

Sorry, I did not mean to. Yes, that is correct.

The later D-M products have 2 separate channels driving the 2 connections as in my 7703 and the AV8805 you currently own, and yes they do provide independent level and delay settings to the 2 sub outputs.

Agreed.

If I understood Jacob from Dirac correctly during the first 10 minutes he explains how DLBC does basically the same. As he mentions just past the 7:00 minute mark "adding flat to flat wouldn't = flat. The sub FR correction and integration with the rest of the speakers is done later with all subs combined as a whole.

Well, I cannot recall that exact statement but I do not believe your rendition of it is correct. DLBC has three levels (at the current time) and what you have described is the intermediate one. The 3rd level corrects each sub individually but it requires an environment in which each of the subs can be addressed and EQ-ed individually. That is not the case in many AVRs and prepros.

 

[Sal1950]

That is not the case in many AVRs and prepros.

Yep, as I mentioned earlier, DLBC is mostly going to require a whole new generation of gear with the possible expection of those currents priced in the 5 figure range. Looks to be a extremely powerful piece of software, in the main designed for TOTL products. They aren't going to be including the leading versions of Dirac Live and DLBC in under $2k AVRs, they are more appropriate to Trinnov level kits.

 

[Kal Rubinson]

They aren't going to be including the leading versions of Dirac Live and DLBC in under $2k AVRs, they are more appropriate to Trinnov level kits.

Or a PC.

 

[mumford]

Or a PC.

A 5 years old PC is not just a PC, if you can convince people to pay $18K for one.

 

[Dj7675]

A 5 years old PC is not just a PC, if you can convince people to pay $18K for one.

Trinnov and StormAudio are expensive for sure. But for what they do in regards to the number of channels, EQ flexibility, features, and support etc are pretty amazing.

 

[jhaider]

The later D-M products have 2 separate channels driving the 2 connections as in my 7703 and the AV8805 you currently own, and yes they do provide independent level and delay settings to the 2 sub outputs. If I understood Jacob from Dirac correctly during the first 10 minutes he explains how DLBC does basically the same.

The big difference is how the subwoofers are combined. Audyssey basically level matches the two subs and sets the delays so they're "time aligned" and then applies EQ. As I understand DLBC, Dirac uses levels and maybe delays (not sure there) as part of the optimization process rather than constants.

Subjectively, in our old house I don't think DLBC was necessarily better than my manual optimization. It also has a current shortcoming in that it assumes all the subs are similar in capability. A manual optimization can make good use of subs with disparate capability. However, it also took like 2 consecutive hours from setting up the mike to listening, and not hours of tweaking over months.

 

[Dj7675]

I haven’t read all of this thread yet, but I found this document informative.

https://www.stormaudio.com/wp-content/uploads/2021/02/Bass-Control-in-Live_20200417.pdf

A snippet from the document summarizes what DLBC does pretty well.

 

[Dathzo]

Option 1 appears best.

Does the 658 allow DLBC with 2 independently addressable subs? If yes then do that. If not then make sure DL treats both subs as a single sub (cluster) and not as 2 independent subs.

Thanks again. There is no DLBC yet in the C658.
I would need to split the signal from sub out 1 into my 2 subs then. That is not going to solve the distortion issue, but as you say, may get me the best sound quality with the tracks that don’t distort

 

[Sal1950]

As I understand DLBC, Dirac uses levels and maybe delays (not sure there) as part of the optimization process rather than constants.

That's about the way I understand it also.
The new Dirac Live Complete and Live Bass Control are some powerful pieces of DRC software, that's for sure.
And they dang well should be when you begin to take their cost into consideration.

Subjectively, in our old house I don't think DLBC was necessarily better than my manual optimization.

I believe that, the needed numbers are a known, you just need to plug them into the calculations to get the answers.

Does the set-up wizard of DLBC require a 20 point measurement around the MLP as shown at the beginning of the video???

 

[Krobar]

DLBC does seem to send different filters for each subwoofer, not just gain / delay (Some beta firmwares allow you to see the Dirac cals). I suppose this is how they intend to support things like independent Tactile/Transducer output in future.

 

[jhaider]

And they dang well should be when you begin to take their cost into consideration.

If you compare the cost vs the hours of time in front of the audio system that can be spent listening rather than messing with subwoofers, it’s a high value purchase - especially to working professionals with young children!

I believe that, the needed numbers are a known, you just need to plug them into the calculations to get the answers.

I’ve been doing multisubs since at least 2006, and not my experience at all. Plug and play levels and delays have never given me optimal results. Lots of experimentation with relative levels, delays, and polarity is needed.

I’m sure there are sophisticated modeling programs that tier 1 CI installers either commissioned themselves or have access to through a supplier (e.g. JBL Synthesis) that can spit out plug and play exports that mostly work well for the mid six figure rooms they design. DLBC is the closest alternative for the rest of us right now.

Does the set-up wizard of DLBC require a 20 point measurement around the MLP as shown at the beginning of the video???

I think I measured 8 or 12 points. No different than Dirac pre DLBC.

 

[Sal1950]

I’ve been doing multisubs since at least 2006, and not my experience at all. Plug and play levels and delays have never given me optimal results. Lots of experimentation with relative levels, delays, and polarity is needed.

As have I since around 1990, with dual HSU 7' tall subs. That's why DSP software such as Audyssey and Dirac has been has been such a blessing to us. Getting things working decently with a SPL meter and a tape measure isn't brain surgery but I didn't say it was easy either. It does have a steep learning curve and is quite time consuming. Then again you said
you were able to get a superior result manually than with Dirac, it was just a lotta work.

If you compare the cost vs the hours of time in front of the audio system that can be spent listening rather than messing with subwoofers, it’s a high value purchase - especially to working professionals with young children!

Ah, but we're not all "working professionals" earning high dollar salaries.
Maybe for you and your friends an extra $500 to 700 dollars for the full boat TOTL Dirac is nothing to be concerned over, but to many blue collar working stiffs or retiree's, that's a hefty sum.

I think I measured 8 or 12 points. No different than Dirac pre DLBC.

I was just curious and the video was a bit vague on the subject but they did show 20 measurement points?
What does it say in the DLBC manual?
I don't believe there's any really solid answer with the best results coming from a variety of approaches and needs, but that "variety" is hard to get across to the setup layman.