TOPIC:

dreite wrote:

I'm still unclear on what the Dirac Live "proprietary filter structure" is. I suppose the mystery is the objective, but I'd still like to better understand this implementation.

Is it simply a combination of linear-phase and minimum-phase corrections in addition to time delays?

Dave.

Ahahah.. Dave, we'd guessed that you'd be knocking at the door wondering "How do these guys do it?"..
There is indeed some mystery behind it (and some pretty strong NDA to make sure it stays that way). So as much as we love to share knowledge, that's unfortunately not something we can share as far as how we implemented inside the Sharc..
Did you have a read at the technical paper here? That's the best open information you'll find: www.dirac.se/media/12044/on_room_correction.pdf

DevTeam

Hey Guys,

Thanks for the answer! Much appreciated. I totally agree that the years of scientific knowledge and testing that have gone into these solutions is an amazing amount of work. I didn't intend to undermine these achievements by any means. Sorry if that is how it came across, I for one understand the long hours in getting something like this finalised and.... that is the exact reason for the costs on these streamlined and intelligent filtering codes/units. And as i said, Dirac indeed sounds good.
Its good to know the filters are the same. So anyone that Trials the PC/Mac software based solution can be assured that will be the sound available to them with the DSP based solution. Obviously this may not be the case in other hardware based Dirac solutions from other companies with the different levels of implementing their code.

It's totally understandable that the software creators of these filtering techniques protect their code heavily. In this day and age anything digital code is almost expected to be free as it takes no physical form. So when you dedicate your life to these solutions ( as most have in this field ), it's only fair to make sure any income streams from your IP are protected.

Thanks Guys, Love your work in bringing the DIY and startup Guys usefull tech at reasonable prices.

The things I want to know is how Dirac differentiates between speaker response and reflected response, and how it corrects each. And how it handles it when the speaker is in the wall or ceiling and does the room response correction limit itself to low frequencies?

DEQX, I would say, isn't that different, as it is speaker impulse response correction. They had always said they wanted to get into room impulse response correction, but apparently never felt they had properly cracked that problem and so they didn't release it. Audyssey always tried to imply that it did impulse response correction, but I don't see how with the power limitations involved. Plus it never really sounded that good except on in-wall or low quality speakers.

@john
I mentioned this above really - Dirac (or anyone similar) aren't going to explain the details of their solution that give other people insight on how to copy their filters. The algorithms developed for averaging and isolation of what they correct and the final creation of the filters would be beyond most peoples knowledge or understanding.
If you want to know more then i guess you could look into some research into filtering mathematics and psychoacoustics.
There are limitations to what is actually possible of course. You can't stop a reflection from happening but if you can get a picture of the rooms resonances and interactions with the speakers then you might be able to minimise their effect in a particular position in the room.
Looking at it like that shows their is still much to be gained in getting a more "3d" picture of the room e.g. mic arrays and controlled sound emission into the room. And then of course what and how to use all that data. If you worked that out... would you just tell someone on a forum.

Well, first, I am curious because I know more or less how I would approach it. Secondly, you can't just throw out common concepts and then have someone rush out and turn it into a workable, sellable alternative. And third, yes, if I were confident in my product and the difficulty in getting the math right, I wouldn't have any problem at all sharing the general approach.

Audyssey had some pretty fatal flaws to it because they didn't understand psychoacoustics or target curves.

You can choose what frequency range the magnitude response correction operates over. By default it's full range (or whatever the algorithm estimates "full range" means for your speakers). I don't think the algorithm specifically differentiates between speaker and room response - if you run it and measure the result, that's when you get impressed. What it does is look for commonalities across the listening area (or volume, to be more precise).

Yeah, I agree in that common concepts are far from final code.
I too would love to know what things are actually happening. Call me a pessimest but nowadays it's damn common practice in retail to be fed misleading information.
And, let me say, Im far from any expert in this.
Maybe Time windowing snapshots could be employed to ascertain speaker response (limited in freq though) separate to room reflections and their cancellations? Analysing distortion levels of certain freq areas and decay rates?
I don't know enough about psychoacoustics and how they could be applied to the measurements either...
Working out common points of response in phase and magnitude and harmonics around an area could give a picture of how to correct other areas to provide balance inside the room?
Either way it's a great deal of optimisation and testing of the calculations.

Press release:

www.prlog.org/12301981-minidsp-and-dirac...ection-solution.html

Hi all,
Just wanted to say: Like most miniDSP customers I understand that digital audio doesnt have to be expensive to be good and I really enjoy efficient, clear-headed design. For anyone who is science driven, miniDSP is a beautiful oasis where empirical data meets efficient technology, unencumbered by marketing nonsense.

miniDSP products have a nice little niche which I'm sure the company knows how to manage far better than any of us. Their marketing/product model is perfect for me - I love the way they do business, across the board!

Devteam:
I am interested in the DDRC but would like to hold out for a Multichannel unit.
Is there any timeframe and price target for this yet? I don't want to get a stereo version and have to sell it in less than a year to purchase a multichannel. Unfortunately $825 is a lot of money for me .

Thanks,
Moto

devteam wrote:

Yes there is indeed support for the Apple remote on this hardware. It's not something we can provide to current VOL-FP though. Different firmware and unfortunately can't be upgraded (need to be returned to us). As for the remote sensitivity, OT so let's not dwell too much here, we can have a look at it. Please start another thread. Thanks.

OK, I've posted over on the thread I did start about this - thread here

Ta

NoDelusions wrote:

Hi all,
Just wanted to say: Like most miniDSP customers I understand that digital audio doesnt have to be expensive to be good and I really enjoy efficient, clear-headed design. For anyone who is science driven, miniDSP is a beautiful oasis where empirical data meets efficient technology, unencumbered by marketing nonsense.

miniDSP products have a nice little niche which I'm sure the company knows how to manage far better than any of us. Their marketing/product model is perfect for me - I love the way they do business, across the board!

Thanks for your nice comments and your support! We'll keep pushing the limits to please the whole community.

Devteam:
I am interested in the DDRC but would like to hold out for a Multichannel unit.
Is there any timeframe and price target for this yet? I don't want to get a stereo version and have to sell it in less than a year to purchase a multichannel. Unfortunately $825 is a lot of money for me .

Thanks,
Moto

Can you at least tell us if this is going to happen in the near future?

MX48 wrote:

Devteam:
I am interested in the DDRC but would like to hold out for a Multichannel unit.
Is there any timeframe and price target for this yet? I don't want to get a stereo version and have to sell it in less than a year to purchase a multichannel. Unfortunately $825 is a lot of money for me .

Thanks,
Moto

A big chunk of change for this unit is the Dirac license. I think this is a great 2-channel solution.

For a great multi-channel preamp check out the coming XMC-1 from Emotiva - with 7.2 Dirac. It should be released by month's end.not cheap either.

I am planning to hold out for object based surround such as Atmos if I can. Then I could use the MiniDSP Dirac multichannel after it.