Dirac Live 1.2 for Mac

[dahlberg]

6 hours ago, Bob Stern said:

@Dahlbrerg:

 

Before your Dirac free trial expires, you might want to investigate Uli Brueggemann's Acourate.  It is designed to replace your present IIR active crossover, and you may prefer its sophisticated driver phase alignment algorithm.  FWIW, recording engineer Bob Katz is a fan.

 

Unfortunately, there is no free trial of Acourate.  To get an idea of how its room correction algorithm compares to Dirac using your existing active crossover, you can send your impulse response measurements (without Dirac) to the developer of Acourate along with a few music tracks, and he'll send back to you the music files processed with the Acourate room correction.  I don’t know whether there is a similar way to test Acourate's active crossover implementation.

 

http://www.audiovero.de/en/documents-and-tutorials.php

 

http://www.audiovero.de/en/acourate-test-for-free--.php

 

 

Well I'm looking at that as well. I also saw the video of Bob Katz and even if I have a lot of respect for his work the segment is over a year old and his impression is most likely even older so..... Lets for sake of argument say that his opinion is 2 years old, has Dirac developed over this time to the extent that he should try it again ? There is another a problem for me to use Acourate for crossover duties. I have been looking for some sort of interface that would bring out all channels needed (6) as i2s from the mac mini, my multichannel dac solution is not something I'm ever going to replace with some usb multichannel soundcard. Minidsp has something called the "usb streamer" that might work so I will look in to that. The no trial idea ??? The main  idea for me with Dirac is to achive linear phase and an improved impulse response, the frequency response is already quite good. We can leave "Acourate" out of this thread, there is another place for that.

[Bob Stern]

In my opinion, Dirac's literature appears deliberately obscure, but it appears to me that they only address frequency response, not time alignment (impulse response).  The only concession to the time domain is that their FIR filters have an asymmetrical impulse response with more post-ringing than pre-ringing, similar to setting the pre-ringing parameter in Audirvana's iZotope filter to 0.3 roughly.  But this has nothing to do with time alignment.

 

Incidentally, they brag they are unique in using a mix of IIR for low frequencies and FIR for high frequencies.  In my opinion there is no advantage to this.  IIR for the low end permits using much shorter FIR filters, thereby reducing computational load, but that is not an issue with modern computers.  I believe they originally adopted this approach for car audio applications where computing power was much more limited.

[dahlberg]

Hello again.

I'm still learning here (I think). I have joined the Acourate Yahoo group and I will try to find out more over there since there is not that much writen here (besides two articles I found that looked promissing). Regardless of any objections you might have to what Dirac is doing I will certanly try it out, no reason not to. Not right away though since there are a few other (non hifi) projects that need some attention first. Have to keep the Mrs happy to :-)

[Bob Stern]

I don’t object to Dirac, just to their marketing puffery.  I like to give their competitors some exposure once in a while because Dirac gets most of the attention on the CA forum.

 

Dirac is by far the easiest way to improve the frequency response of a room, which is why it is deservedly popular.

 

Room EQ Wizard does the same thing for free but is more complex to figure out and use than Dirac.

 

All the other free (e.g., DRC and OpenDRC) and paid (e.g., Acourate and Audiolense) alternatives require Windows (e.g., in Parallels or Virtual Box emulator on a Mac) to generate an FIR filter (impulse response).  The resulting filter can be loaded into a Mac music player that either supports convolution internally (HQ Player) or supports a convolver plugin (Audirvana, Pure Music).

 

The reason only Dirac offers a free trial is the implementation difference.  Dirac is a plugin with the frequency response corrections hidden inside.  Acourate and Audiolense are tools for generating an FIR filter that can be hosted by any music player software.  With Acourate and Audiolense, after you calculate the FIR filter, you no longer need to run the program.  Therefore, a free trial would enable you to calculate one or more FIR filters before the trial expires, and you'd never need to buy the program.

 

[Bob Stern]

rePhase is another free Windows tool for generating FIR filters, which I just discovered in this thread:

 

https://www.computeraudiophile.com/forums/topic/29737-rew-vs-acourate/

 

[dahlberg]

11 hours ago, Bob Stern said:

rePhase is another free Windows tool for generating FIR filters, which I just discovered in this thread:

 

https://www.computeraudiophile.com/forums/topic/29737-rew-vs-acourate/

 

Yes I know, I have been fiddling around with it enough to realise that I will run out of taps to soon using dsp's.

[flak]

On 3/27/2017 at 9:10 PM, Bob Stern said:

In my opinion, Dirac's literature appears deliberately obscure, but it appears to me that they only address frequency response, not time alignment (impulse response)...........................

 

Hi Bob,
I don't comment your words about Dirac using marketing puffery but I think it's important that some misconceptions are addressed, here a document by Dirac's Lars Johan Brännmark and Mikael Sternad about controlling the impulse responses and the spatial variability in digital loudspeaker/room correction:

http://diracdocs.com/ISEAT15_Brannmark_Sternad.pdf

 

It may appear obscure but certainly that's not deliberate
Thanks for reading it and please let us know If you'll find something that you consider inaccurate.

Flavio

[Bob Stern]

Thank you Flavio.  I read that document some time ago.

 

The potentially misleading part of the document is that it uses the term "controlling impulse response".  This is similar to terminology used by other companies to refer to time alignment.  As I pointed out above, Dirac merely uses this term to mean that the pre-ringing is less than the post-ringing, similar to setting the pre-ringing parameter in Audirvana's iZotope filter to approximately 0.3.

[wgscott]

11 minutes ago, Bob Stern said:

Thank you Flavio.  I read that document some time ago.

 

The potentially misleading part of the document is that it uses the term "controlling impulse response".  This is similar to terminology used by other companies to refer to time alignment.  As I pointed out above, Dirac merely uses this term to mean that the pre-ringing is less than the post-ringing, similar to setting the pre-ringing parameter in Audirvana's iZotope filter to approximately 0.3.

 

I thought it referred to doing this (which I have assumed to be time-alignment):

 

[Fitzcaraldo215]

3 hours ago, Bob Stern said:

Thank you Flavio.  I read that document some time ago.

 

The potentially misleading part of the document is that it uses the term "controlling impulse response".  This is similar to terminology used by other companies to refer to time alignment.  As I pointed out above, Dirac merely uses this term to mean that the pre-ringing is less than the post-ringing, similar to setting the pre-ringing parameter in Audirvana's iZotope filter to approximately 0.3.

I agree with @wgscott, in that I do not believe you are reading it correcty.

 

You seem to have some axe to grind, or possibly your own EQ under development or some measurements which debunk Dirac's technical claims. Otherwise, I just do not see where your negativity is coming from.

 

Perhaps, you are right and it is all overhyped BS, but have you any actual proof other than snipes and gripes at what Dirac says?

[jiminlogansquare]

Bob might be referring to time alignment between the individual drivers of a speaker. Some speakers are built with a rake in the front panel to make sure sound waves from the tweeter (at the top) reach the listening position at exactly the same moment as those from the woofers (at the bottom). Based on some minimal Googling, I think Mini DSP can duplicate that driver time-alignment effect in the digital realm, but based on my experience of using Dirac Live stereo, Dirac Live stereo does not. Dirac WILL delay an entire speaker's output to make sure both speakers are time-aligned, so in that sense maybe everyone is right. That probably oversimplifies things, but maybe it's what Bob had in mind. Am I right or wrong that this is different from the impulse response control that wgscott's picture illustrates?

 

(Addendum: Regarding the theoretical basis for why time alignment between high- and low-frequency drivers might be an issue that needs correcting, here's a relevant quote from Wikipedia's article on the speed of sound: "In dry air, the speed of sound increases by about 0.1 m/s as the frequency rises from 10 Hz to 100 Hz. For audible frequencies above 100 Hz it is relatively constant." So the woofers are putting out sound waves that move just a bit slower than the other drivers' sound waves. Learn something new every day!)

[Bob Stern]

I have no economic axe to grind.  I am merely a consumer

 

Furthermore, I owe Dirac an apology.

 

Bill Scott's impulse response measurement does evidence Dirac improves response in the time domain, not merely the frequency domain.

 

I just read a much older (2009) paper from Dirac:

http://www.dirac.se/media/12044/on_room_correction.pdf

 

This helped me better understand the 2015 paper that Flavio identified, especially section 3.4, which states that their filter tries to minimize ringing and that the filter is almost causal (i.e., almost all ringing is after the impulse), with pre-ringing permitted for only a brief time that they consider perceptually inaudible.

 

[jiminlogansquare]

Well, Bob, I tried.

[wgscott]

1 hour ago, jiminlogansquare said:

Well, Bob, I tried.

 

I think you are probably right, but it would be cool if this was capable of doing it in a frequency-dependent manner that allowed, for example, phasing the main speakers with a subwoofer, that would be a great achievement.  (If Dirac does this, it would be good for the marketing propaganda to make this explicit.)

[Kafy]

It would be surprise me that mini DSP can do something Dirac Live can't. For me, they use the same filters.
Regarding the time correction for the tweeter. it will always be dependent of the distance of listening. So what ever it is done by filter (L, R, C) or by physical offset of the tweeter, it is always a choice of the manufacturer which decide peoples will be at a "normal" distance of listening.
The only interest to make a time alignement by physical offset of the tweeter is it make possible to make a less complicated filter. Which can be a good thing because a filter always change many thing (gain different from frequencies, phase offset, ...).

 

[flak]

7 hours ago, wgscott said:

 

I think you are probably right, but it would be cool if this was capable of doing it in a frequency-dependent manner that allowed, for example, phasing the main speakers with a subwoofer, that would be a great achievement.  (If Dirac does this, it would be good for the marketing propaganda to make this explicit.)

 

Yes Wgscott... correcting those regions of frequencies that require it in a frequency dependent manner as you suggest is a good idea and that's what Dirac Live is actually doing

This allows the use of a subwoofer in a stereo system with the stereo Dirac Live two channel version even if it's placed in a different location.

The low frequencies from the subwoofer are corrected by Dirac Live exactly in a frequency dependent manner as you mention... needless to say some compromise is necessary for the frequencies in the crossover region 'cause some frequencies arrive from two different distances, only Dirac's Unison can be the ideal solution for that.

 

Flavio

[flak]

9 hours ago, Bob Stern said:

I have no economic axe to grind.  I am merely a consumer

 

Furthermore, I owe Dirac an apology.

 

Bill Scott's impulse response measurement does evidence Dirac improves response in the time domain, not merely the frequency domain.

 

I just read a much older (2009) paper from Dirac:

http://www.dirac.se/media/12044/on_room_correction.pdf

 

This helped me better understand the 2015 paper that Flavio identified, especially section 3.4, which states that their filter tries to minimize ringing and that the filter is almost causal (i.e., almost all ringing is after the impulse), with pre-ringing permitted for only a brief time that they consider perceptually inaudible.

 

 

Thanks Bob for going deeper into the subject, much appreciated!

By the way our site has been redesigned so your link to the 2009 document is not working,

this is one that does: http://www.diracdocs.com/on_room_correction.pdf

 

Flavio

[DSD256]

Dirac live support exists? No reaction on any request or questions for one week. i#m in trial phase...so would be nice  to get a feedback. Is there a working  email address for support?

thanks

[flak]

On 7/28/2017 at 0:44 PM, DSD256 said:

Dirac live support exists? No reaction on any request or questions for one week. i#m in trial phase...so would be nice  to get a feedback. Is there a working  email address for support?

thanks

Hi Thomas,

I hope you're satisfied with the answer that you received.

 

Good listenings, Flavio

[PowerQuality]

@flak I have a Dirac OSX licence. Is it possible to convert to Win licence as I am about to replace my playback Mac with a PC? 

 

Thanks 

[flak]

56 minutes ago, PowerQuality said:

@flak I have a Dirac OSX licence. Is it possible to convert to Win licence as I am about to replace my playback Mac with a PC? 

 

Thanks 

Hi PowerQuality,

 

yes, it's possible to convert to a Win license... just contact us from here:

http://helpdesk.dirac.se/index.php?pg=request

and we'll send you instructions and a download link.

 

Flavio

[PowerQuality]

@flak 

On 11/19/2017 at 6:48 PM, flak said:

Hi PowerQuality,

 

yes, it's possible to convert to a Win license... just contact us from here:

http://helpdesk.dirac.se/index.php?pg=request

and we'll send you instructions and a download link.

 

Flavio

 

Thanks for the response on help desk. But my licence is still not working with windows. I get a not valid license error when trying to validate. When I log in to Dirac online store it says I have an active stereo license with two remaining installations. 

 

It has been 5 days since I replied back to the help desk email, but no one has replied yet. 

 

Thank you. 

[flak]

Hi PowerQuality,

 

please send me a PM with your ticket number, I'll track it and follow up.

Thanks, Flavio

[Demetrios]

Hello Flak,

 

I performed new measures in my room this morning with 1.2 version.

Everything went fine and I created the filters.

But it's impossible to load them in the DAP or in Audirvana (with AU plugin) : it crashes immediately and program closes.

If I try to load "old" filters (before this new session) everything is fine.

What's happening ?

Thanks for your answer.

Regards.

Dem

 

[flak]

On 4/27/2016 at 6:57 PM, camott said:

 

Sorry, this plug-in cost is going to be in *addition* to the current cost for purchasing Dirac Live, or you are going to sell the plugin as a separate product with it's own price? If it's the former, I think you are going to have a lot of peed off customers ...

 

We certainly don't want peed off customers, the latest news are that not only all existing customers will receive the upgrade for free but also the new generation will use plugins instead of DAP, as a result they will be free of charge as well.