(shortcut) What does DIRAC really do?
Jan 11, 2016 at 9:29 AM Thread Starter Post #1 of 13

Giogio

Bluetooth Guru
Joined
Nov 13, 2014
Posts
1,419
Likes
366
I just wanted to post a link to the thread I have opened in Computer Audio, because I think the subject applies to both sections and I would like to reach also the members who only check this section.
To avoid double threads anyway, this is meant just as a notification/shortcut, but the discussion will be there.
 
I hope this is ok.
 
Jan 12, 2016 at 12:07 AM Post #2 of 13
Dirac is one of many digital room correction programs.
 
There are also several competitors, both proprietary and open source.  The algorithms that Dirac uses are proprietary.
 
Whether Dirac gives better results than alternatives is debatable.  But it is one of the easiest to use.
 
The output is a bunch of FIR and PEQ settings to load into the DSP of your choice.
 
Jan 12, 2016 at 3:44 AM Post #3 of 13
Yes, Dirac uses I think a mixture of FIR and IIR filters, can do parametrics etc.  It of course does a sweep to get the impulse response of the system.  The normal way is to measure several spots around the listening area. You then can also shape the target curve for the correction.  Which means beyond the correcting of the in room response you also get to shape the general trend to your liking. 
 
I have used it and it works very well.  Most definitely the easiest software I know of for this purpose.
 
Jan 12, 2016 at 6:35 AM Post #4 of 13
Sorry, I forgot to post the link to the other thread, I did not mean this to be a double post :frowning2:
I have edited now.
So, next readers, please go to the other thread, sorry!
 
I have anyway doubts about your reply, I am posting the question there, maybe you can answer me in the other thread
biggrin.gif
 
 
Jan 12, 2016 at 2:04 PM Post #5 of 13
not sure if this topic has any reason to exist if you want to keep the dialog on the other one(and I agree it should stay on 1 single topic).
 
Jan 12, 2016 at 2:59 PM Post #6 of 13
My concern was that I could not understand which section was better, and then also that some users may only check one of the two sections, which is a waste of answering potential.
So I just wanted to put here an announcement that there was this thread on another section, which actually fits in this section too.
 
I myself would close this thread without deleting it, but I live the decision to you.
Thanks :)
 
Jan 15, 2016 at 10:47 AM Post #7 of 13
  My concern was that I could not understand which section was better, and then also that some users may only check one of the two sections, which is a waste of answering potential.
So I just wanted to put here an announcement that there was this thread on another section, which actually fits in this section too.
 
I myself would close this thread without deleting it, but I live the decision to you.
Thanks :)

 
This is why we have a search function...so people don't have to post pointer threads all over.
 
Jan 15, 2016 at 12:24 PM Post #8 of 13
To search for something you must either know that it exists (I saw a thread once, where was it?) or know what you want to search for (let me see if there is something interesting about Dirac).
I did not think about being found through searches, but about being seen through "new thread" for people browsing only ONE of the two sections, or through notifications for people being subscribed to only ONE of the two (I cannot imagine how somebody interested in Computer Audio would not be interested in Sound Science, but I am sure some people interested in Sound Science are not interested in Computer Audio).
Anyway, what's done is done, apparently Castleofargh agreed with letting this here for linking purposes and as eternal testimony of my mistake. Which I am grateful for.
For the rest, what's done is done.
 
Jan 15, 2016 at 4:35 PM Post #9 of 13
it's not that I agree and more that it changes nothing.
to me this topic is the old dog I don't have the strength to bring to the vet one last time even though it stopped walking like a week ago. I'm sure it's gonna go, but one more day seems somehow significant to him.
deadhorse.gif

 
 
biggrin.gif

 
Jan 15, 2016 at 4:40 PM Post #10 of 13
Why is relevant to headphones? Perhaps speakers in a compensated room but never mind. Seen this mainly being used a studio application for setting up monitor speakers for studio environments.
 
Although I suppose a dedicated music room could benefit.
 
Jan 15, 2016 at 7:29 PM Post #11 of 13
Originally Posted by castleofargh /img/forum/go_quote.gif
I'm sure it's gonna go, but one more day seems somehow significant to him.

That was deep. I am moved.
 
  Why is relevant to headphones?

Try the XTZ Headphone Divine with and without the Dirac, and then tell me if you still do not understand why.
There is no substantial difference in the reasons why someone would want to record the sound of speakers with a room analyser to compensate for their (and the room's) limitations, or why someone would want to record the sound of headphones with a dummy head to apply frequenca/impulse response correction.
And as we are discussing in the thread about "original sound", apparently some people goes till the point of doing anatomical tests on their ears to apply further corrections.
There is a lot which can be done to headphones, with Dirac and other DSP...
This may be the last day of a dying dog, but it surely is not the last day of my enthusiasm for DSP.
I do not care if Dirac is the best and I am not even trying to claim that. It is the first I use which improves the sound so dramatically. Till now I was sure that DSP were just toys to add effects to the sound and make it more wahwah.
Dirac completely changed the was I listen to music, brought to life sounds in the high range which I never knew they where there in those recordings (and I use EQ, but apparently it was never enough).
It may be something old for more expert guys, but it was a revelation to me.
 
Jan 15, 2016 at 8:58 PM Post #12 of 13
  That was deep. I am moved.
 
There is no substantial difference in the reasons why someone would want to record the sound of speakers with a room analyser to compensate for their (and the room's) limitations, or why someone would want to record the sound of headphones with a dummy head to apply frequenca/impulse response correction.
And as we are discussing in the thread about "original sound", apparently some people goes till the point of doing anatomical tests on their ears to apply further corrections.
 

 
I'm not so sure about that...
 
I use OpenDRC now and have used Dirac before, so am pretty familiar with the techniques, which mainly consist of:
 
1. PEQ
 
Yep, this will obviously work.  But you don't need a DSP engine to do it.
 
2. Phase adjustments
 
This is really tricky because the geometry is so small.  I've tried measuring and correcting this a few times but it just ended up sounding weird and...well, phasey, like one of those headroom expansion products.
 
3. FIR/IIR
 
I skeptical if the juice is worth the squeeze unless you were trying to map it to your specific ear geometry.  Also, the Dirac Live algorithms are designed for a room correction scenario and rely upon multiple measurements taken across a listening position and that's where they made their reputation.  $400 is a lot to pay for software that wasn't even originally designed to be used with headphones.
 

Users who are viewing this thread

Back
Top