Dirac Live with Bass Control

felty99

Active Member
Well done - great result by the sound if it and another confirmation that we are on to something here I feel!
I'm actually quite impressed with how close Dirac v2 got to the user-validated measurements with REW, but I know a lot of others aren't. v1 did seem to get a lot of flack for being inconsistent and over-correcting to the standard Dirac target curve so everything sounds the same, but if you know what you're doing then it's a hugely powerful and flexible system. Moreso with the imminent BM update. I only use it correct the frequency response of the Sub and Front L&R below 500Hz whilst tracking the FR of my speakers, and integrates everything really well.

As others have pointed out here and other threads, the latest version does seem a lot more bombastic in the bass so that is another confounding factor when trying to assess any clear impact of these changes.
 

lexicon

Well-known Member
I'm actually quite impressed with how close Dirac v2 got to the user-validated measurements with REW, but I know a lot of others aren't. v1 did seem to get a lot of flack for being inconsistent and over-correcting to the standard Dirac target curve so everything sounds the same, but if you know what you're doing then it's a hugely powerful and flexible system. Moreso with the imminent BM update. I only use it correct the frequency response of the Sub and Front L&R below 500Hz whilst tracking the FR of my speakers, and integrates everything really well.

As others have pointed out here and other threads, the latest version does seem a lot more bombastic in the bass so that is another confounding factor when trying to assess any clear impact of these changes.
Yours are the best delay readings I have seen as it in fact allowed more delay than required by 1.5ms. If you look at mine you'll see it only allowed 2.87ms delay. Inconsistent at best.
 

felty99

Active Member
Yours are the best delay readings I have seen as it in fact allowed more delay than required by 1.5ms. If you look at mine you'll see it only allowed 2.87ms delay. Inconsistent at best.
Yes, that why I said I was impressed:). I don't know if it makes any difference but I don't use a dedicated centre channel. Didn't Dirac move from taking the centre as the baseline delay in v1 vs the Sub in v2? Probably a red herring though.
 

neo_2009

Member
will be interesting how close those corrected curves correlate to REW. With the Arcam I always thought they were wildly optimistic and were nowhere near the real result.
The "predicted" corrected curves are accurate, but they represent an average of all measure predictions. In order to get an exact measure on Dirac and REW, you would have to perform a single-point measure and correction in Dirac, and measure with REW in the same exact position.

Please see this thread, where AustinJerry made the above test: The Dirac Official Thread - AVSForuns
 

lexicon

Well-known Member
Yes, that why I said I was impressed:). I don't know if it makes any difference but I don't use a dedicated centre channel. Didn't Dirac move from taking the centre as the baseline delay in v1 vs the Sub in v2? Probably a red herring though.
Only Dirac knows. Yes they did make that change as far as I can tell. It's a step in the right direction.
 

nkatz

Active Member
I don't understand why all the fussing about with manually setting delays; isn't BM supposed to do that automatically?
 

lexicon

Well-known Member
I don't understand why all the fussing about with manually setting delays; isn't BM supposed to do that automatically?
All explained at the start of this thread. I am not convinced that DLBM will fare any better than basic Dirac yet as it seems to use the same delays but we shall see - at the moment I‘m trying to get my head around the new REW alignment tool (no manuals or training vids cover that feature as far as I can find) so I will try to answer that question but it’s going to take some time. A lot to play with and take in.
 
Last edited:

nkatz

Active Member
I am not convinced that DLBM will fare any better than basic Dirac...

Not that your investigative activities aren't very interesting, but lots of us have been dying to know how close DLBM comes to fulfilling its claims.

Why not just let it do its automatic thing, which ought to be a lot quicker, or at least easier, than all of your various measurements.
 

lexicon

Well-known Member
That makes two of us but apart from the complexity in working out what this is doing and how it works, I don’t feel there would be much value in me approaching this from a purely subjective viewpoint.

I’d like to do it right and for any results I provide to be backed up by measurement if possible to confirm what I am hearing over some test familiar material.

To assist me........

I have elicited some help from a REW expert in the states who has already determined that my choice of crossover at 40hz mains is unlikely to produce a good result in my room. He seems to know a load about how to analyse the results from REW graphs using the REW alignment tool.

It seems at this point I need to determine the right Xover settings before I can look again at timings.

Progress is being made and you can see what I am trying to achieve by looking at this thread:-


Alas I cannot hog the AV system all the time as it’s in my main living room and my partner insists on a plethora of quiz programmes most afternoons and complains at constant REW and Dirac sweeps 😖😁
 

hestepare

Member
Final iteration of the FFT app method is up:

 

lexicon

Well-known Member
Final iteration of the FFT app method is up:

Many thanks for that, great stuff. For use with non Lyngdorf processors I would substitute amplifier delay for “all individual speaker delay settings in ms other than the sub”.

I have tried to convert this too....please let me know if I have messed up!


Getting your Sub/Sat Timing correct

Delay (ms) works conversely to Distance (ft or cm). Different Processors will generally use one or the other (Delays in ms or Distance in ft/cm.) so you need to select and use the correct method.

Reducing delay between sub and your other speakers means "making the sub start sooner relative to all your other speakers".

You do this either:-

1. If your processor works in ms delays - by reducing the ms delays on all speakers except the sub by the same amount, which really means you are “reducing their delay time as compared with the sub“

or:-

2. If your processor works in ft/cm distance - by increasing the sub “distance“ setting in ft or cm. If the sub is farther away, the processor will take into consideration that the sound it emits will have to travel a longer distance and will have to start sooner when it does its calculations.

So adding distance (ft or cm) to the sub setting is the same as reducing the delay (ms) setting of all of the other speakers.


Increasing delay between sub and your other speakers means "making the sub start later relative to all your other speakers".

You do this either:-

1. If your processor works in ms delays - by increasing the ms delays on all speakers except the sub by the same amount, which really means you are “delaying the sub as compared with all your other speakers“

or:-

2. If your processor works in ft/cm distance - by decreasing the sub “distance“ setting in ft or cm.



Takes a while to get your head around that!
 
Last edited:

hestepare

Member
Looks good to me! Looking at it now, I'm not even sure I got it right, even if I spent quite a while making sure yesterday.

I might suggest that the following claim needs a qualification: "assuming that active subs are inherently slower than passive speakers…"

Reducing delay between sub and your other speakers means "making the sub start sooner relative to all your other speakers".
 

hestepare

Member
A theoretical point that – if I got this right – goes a long way to explain why subs are so damn slow:

"It is commonly thought that there will always be a phase difference of 180° between the outputs of a (second order) low-pass filter and a high-pass filter having the same crossover frequency."

From wikipedia

I take this to mean that using second-order filters, the sub will by definition be one cycle off at the crossover frequency. Makes sense for my crossover, which is at 100 Hz. B&W claim that their sub only has a delay of "less than 1 ms", but I can only get a good result at 10.8 ms delayed fronts. That's one full cycle + slightly less than 1 ms. Using a different order crossover will not give the same results.

Edit: disregard this, I got it wrong.
 
Last edited:

lexicon

Well-known Member
Some interesting stuff on DLBM on my Nirvana thread guys.

I have established that DLBM has eliminated a big dip between mains when using a 40hz Xover making the use of this Xover now viable. This seems to indicate that the claims that it corrects phase may be justified.

I am also trying to come up with an idiots guide on how to measure and correct timing differences between L&R mains and mains and sub.

You can also see a REW comparison of BM off and On to get an idea of what it is doing.

 

ggwoodland

Active Member
Lexicon - I admire your pursuit of the science behind DL BM but I would be interested to know if the bass does indeed sound better. I'm hoping this will come to the NAD but it will be a paid option on the NAD and I wondering whether it will be worth 'shelling out the cash' for the upgrade - George
 

lexicon

Well-known Member
Lexicon - I admire your pursuit of the science behind DL BM but I would be interested to know if the bass does indeed sound better. I'm hoping this will come to the NAD but it will be a paid option on the NAD and I wondering whether it will be worth 'shelling out the cash' for the upgrade - George
I haven’t had enough time with this yet but yes I am seeing a noticeable improvement in the bass integration.

The bass just seems smoother and more integrated.

As a result there is a cleaning up of the sound field.

Sounds seem to be moving around the sound field more noticeably but how much of that is due to the new processor is hard to say.

I will be able to provide more insight as I am able to clock up more listening time.

My testing so far (in my room) also seems to be showing that BM will help to alleviate dips in more challenging rooms caused by phase differences between the L&R, the location of which cannot always be changed.

As a result, more crossover settings that did not produce good results before may become more effective giving us more choice.
 
Last edited:

lexicon

Well-known Member
This is a good demonstration of how DLBM has resolved a phase issue created by my room between L&R speakers. On their own, the L&R REW traces look very good and if I were to just look at the average of the two without bringing the effect of phase into the equation the combined average trace would look even more hunky dory.

However once you bring phase into the mix the picture changes. You can see what "old dirac" was doing (Green trace) and perhaps explained why I was having to resort to measures such as tweaking timings in an attempt to ameliorate this issue to get things to sound better.

With BM applied (Purple trace) however and all the BM calculations complete, you can see the improvement.

So whilst before BM the only options available to correct issues like this would have been:-

1. Play with timings in an attempt to effect a better phase match (difficult and complex)
2. Move the L&R speakers in the room (often not really possible or ideal)
3. Raise the crossover above the dip and allow the sub to fill in the hole

Now, after the adjustments to phase have been calculated and applied across all speakers we no longer have to compromise by finding less that ideal ways to fix the problems.

I, for one, am impressed at these findings :)

This seems to account for the fact that pre BM the 100hz Xover setting always sounded the best in my room.

BM V NO BM at 40hx L&R only.PNG
 

Wull

Well-known Member
Pretty cool results. I wander what individual REW sweeps would look like now. Interested to know how DLBM addresses the Phase issue?

So to be clear, the above the results are from running Dirac, you've done nothing else before hand?
 

lexicon

Well-known Member
Pretty cool results. I wander what individual REW sweeps would look like now. Interested to know how DLBM addresses the Phase issue?

So to be clear, the above the results are from running Dirac, you've done nothing else before hand?
Indeed.

I would imagine that the methodology or algorithms used to achieve that are a carefully guarded (and ever improving?) secret!

No - that’s just Dirac + Dirac BM, nothing else. Although I did adjust levels at my MLP first as they were out but mainly between fronts and rears, only by 1db at front L&R.

Assuming it achieves the same result across multiple speakers and subs then it should make speaker/sub integration much more effective and consistent.
 

ggwoodland

Active Member
Lexicon - thanks for investigating this - looks good. If and when DL BM comes to the NAD I will probably purchase as the NAD locks the settings post Dirac and Manny from JL Audio was disappointed he couldn’t do his phase tweaks on my JL sub. This appears to do what a few people were doing manually. Btw what crossover settings has DLBM set now - I thought you had some strange values for your Atmos speakers - george
 

lexicon

Well-known Member
Lexicon - thanks for investigating this - looks good. If and when DL BM comes to the NAD I will probably purchase as the NAD locks the settings post Dirac and Manny from JL Audio was disappointed he couldn’t do his phase tweaks On my JL sub. This appears to do what a few people were doing manually - george
I don’t think you’ll need to do that any more after BM. I haven’t entirely completed my evaluation of this but it seems likely that the default timing is probably now optimal. I think it probably needed BM to “make it so” as Jean Luc would say :) .

From what I have discovered you need to correct both timing and phase together at all frequencies to achieve the desired result. I presume that’s what BM was developed to achieve.

It was a big help when using Dirac V1 though in my experience - not being able to adjust delays post Dirac would have been a real PITA!
 
Last edited:

lexicon

Well-known Member
Lexicon - thanks for investigating this - looks good. If and when DL BM comes to the NAD I will probably purchase as the NAD locks the settings post Dirac and Manny from JL Audio was disappointed he couldn’t do his phase tweaks on my JL sub. This appears to do what a few people were doing manually. Btw what crossover settings has DLBM set now - I thought you had some strange values for your Atmos speakers - george
Yes, it’s clear the automatic calculation of Xover levels needs further work. Dirac are aware.
Apart from anything else, running a couple of in-ceilings at 27Hz could cause them damage. For now just set your own Xovers at sensible levels. A 80hz universal is likely to be a good place to start for all non mains.
 

Wull

Well-known Member
Indeed.

I would imagine that the methodology or algorithms used to achieve that are a carefully guarded (and ever improving?) secret!

No - that’s just Dirac + Dirac BM, nothing else. Although I did adjust levels at my MLP first as they were out but mainly between fronts and rears, only by 1db at front L&R.

Assuming it achieves the same result across multiple speakers and subs then it should make speaker/sub integration much more effective and consistent.

That's great to hear. Looks like this version of Dirac could be the one to go for :)
 

lexicon

Well-known Member
I am advised by Dirac that the name of this product has been changed to Dirac Bass Control so I have altered the title of this thread accordingly.
 

AdamAttewell

Well-known Member
@lexicon I presume you are measuring at just one seating position?

I do wonder how Bass Control compares to Multi Sub Optimiser as with MSO you can EQ for multiple seating locations which is very useful for us that have multiple rows or seats.
 

The latest video from AVForums

Podcast: The Best TVs and Projectors of 2020, plus AV, TV Show & Movie News & Reviews

Latest News

What's new on Netflix UK for December 2020
  • By Andy Bassett
  • Published
Warner Bros 2021 movies coming to HBO Max and cinemas simultaneously
  • By Andy Bassett
  • Published
What's new on Disney+ UK for December 2020
  • By Andy Bassett
  • Published
4K Streaming impacts carbon emissions says Royal Society
  • By Andy Bassett
  • Published
AVForums Podcast: 2nd December 2020
  • By Phil Hinton
  • Published
Top Bottom