Comparison of DRCs: Dirac Live for Studio, IK Multimedia ARC System 3 and Sonarworks Reference 4 Studio edition

[richard12511]

Yes, that's why it is often better to EQ for a linear direct sound instead of a predefined target curve above transition frequency.

I've honestly never tried this, but it makes sense. Are you using a collection of gated responses? I'm pretty confident that the 8351s already have a linear direct sound, so I'm not sure how necessary it would really be. I'm not even sure what Dirac does above the transition frequency, but I know it has 3 different frequency zones(at least) for which it uses very different EQ strategies. The frequency responses of Dirac full range vs limited range correction (using MMM) are identical above 1kHz. It takes 5 seconds or so to switch between profiles with my DDRC88A, and with that switch time, it's really difficult to hear any difference at all, if there is even any.

 

[SIY]

Sonarworks offers a ready-made preset that works well for stereo hi-fi, as I've personally tested. Dirac gives nothing.

Saying it repeatedly doesn’t make it true. It remains a lie.

 

[richard12511]

@thewas you're using LS50 Meta? If so, then your situation is probably similar to mine with the 8351s, in that EQing to linear sound above 1kHz or so doesn't really do anything. I imagine that strategy would be more beneficial for speakers with non neutral on axis responses.

There is a region between say 300-1000Hz that does seem to be still affected by the room(but not dominated). I'm interested to try it, though I'm not sure I can get all the way down to the transition frequency with gated measurements.

 

[txbdan]

Are the posts above, as I understand it, intended to justify Dirac? But Dirac is a bad solution for the home, since it only gives the user a bare tool, without understanding how to use it correctly. The client is forced to change the frequency response curve at random and search for info on the Internet

Sonarworks offers a ready-made preset that works well for stereo hi-fi, as I've personally tested. Dirac gives nothing.

Do you really think that the type of person who A) is into high fidelity audio, B) identifies issues with room acoustics, and C) purchases room correction software is incapable of some googling and moving of little dots around? They LOVE that stuff.

 

[txbdan]

I've also subjectively evaluated Sonarworks, IK ARC, and DiracLive a couple of years ago. I posted about it on here somewhere. I also chose to pay up for Dirac Live. I run it on my Mac hosted in SoundSource. Works perfectly.

 

[dominikz]

Are the posts above, as I understand it, intended to justify Dirac?

Not mine, no. I have no reason to justify any of these DRC tools as I neither own any of them, nor do I have any personal gain in doing so. Nor do I feel there is any need to justify Dirac Live at all - it is a good tool that can be used with great results, if applied correctly. It is more complicated than some, but far easier to use than others.

But Dirac is a bad solution for the home, since it only gives the user a bare tool, without understanding how to use it correctly.

Here I strongly disagree. Any of the DRCs I tested could be said are a 'bad solution' if applied incorrectly. But can also be great solutions if applied correctly for the use case at hand. In my testing, it was easiest to get good results quickly with Dirac.
Sonarworks also doesn't work equally well in all systems - and the user still needs to know at least when to apply the B&W curve and when to use the other options.
Also, you may find there's quite a lot of arguments by leading professionals in audio whether it is beneficial at all to apply full-range room EQ correction. I'm afraid it is not a black and white situation at all - which is definitely unfortunate for a non-expert end user.

The client is forced to change the frequency response curve at random and search for info on the Internet

I understand this sentiment. But having *any* hard-coded target will not be a good solution that fits all speakers, rooms and listening setups.

Sonarworks offers a ready-made preset that works well for stereo hi-fi, as I've personally tested. Dirac gives nothing.

As I said, I can understand the appeal of having a simple UI with few pre-selected options and therefore avoiding decision paralysis. But it is also limiting if you want to optimize further, which is why there are other tools on the market that target a different segment of end users.

 

[dominikz]

Dirac limits the number of points at which you can manually change the frequency response curve. That is, you will no longer have a SOFT, Smooth curve, as in the above preset. Not to mention the fact that there is a specification with exact data for this curve, the curve is developed on the basis of acoustic calculations, you cannot "manually, roughly, by eye" make a curve with identical parameters

Also, regarding this, maybe you missed it, but you can add *a lot* of points to the curve in Dirac, and you can also import standard (or any, really) curves prepared externally as text files (including B&W). Area between the points is smoothed using what looks like some type of logarithmic approximation. Here's an example with a Harman curve with +6dB bass boost below transition frequency + following the speaker in-room steady-state measurement above it:

 

[tecnogadget]

I think you can tailor the house curve on Dirac trough a txt. If that’s the case then a solution for precision is given and you don’t need to graphically change the curve with the cursor.

 

[richard12511]

I think you can tailor the house curve on Dirac trough a txt. If that’s the case then a solution for precision is given and you don’t need to graphically change the curve with the cursor.

Indeed you can. Attached is an example text file I've used.

 

[thewas]

I've honestly never tried this, but it makes sense. Are you using a collection of gated responses? I'm pretty confident that the 8351s already have a linear direct sound, so I'm not sure how necessary it would really be.

With your 8351 you don't really need to do your own gated responses as you can trust the manufacturer specs and low pair deviation.

@thewas you're using LS50 Meta? If so, then your situation is probably similar to mine with the 8351s, in that EQing to linear sound above 1kHz or so doesn't really do anything. I imagine that strategy would be more beneficial for speakers with non neutral on axis responses.

Exactly, I only EQ above transitition frequency my old LS50 but not my Meta, https://www.audiosciencereview.com/...n-room-measurements.13540/page-18#post-674447

 

[vavan]

the whole "soft smooth curve" is likewise nonsense.

I had no problems loading/importing custom target curve in dirac with about 120 points from 20 to 19kHz which is quite smooth in my book

you've somehow latched onto a non-issue.

There's some agenda here and I am getting curious about what it is

fwiw I put that user on my blacklist/killfile on some other forums several months ago

 

[ai studio]

hi .
How to arc 3
Will run on windows in sound system

I tried
equalizer apo

But one can only control the volume it does not show METER
working

 

[dominikz]

hi .
How to arc 3
Will run on windows in sound system

I tried
equalizer apo

But one can only control the volume it does not show METER
working

IMHO the easiest way to have ARC System 3 VST running systemwide is to follow the instructions from IK Multimedia available on this link.

 

[ai studio]

IMHO the easiest way to have ARC System 3 VST running systemwide is to follow the instructions from IK Multimedia available on this link.

Ok thank you .

 

[ai studio]

Hi dominikz
I installed successfully
Is it worth working in settings in mode
default
or
sharp
?


I felt sharp
It feels too much
"Clean and neutral as in headphones

I feel but maybe I'm wrong
Maybe it's good
That it's not perfect for mix .

But it made me a wonderful fix in the room
I have acoustics
I'm still missing a few panels
But I also have good GIK Trap bass catchers
And a room treated
But there were still problems
That arc it arranged.

I felt at frequencies at low that it gave a little too much
"not realistic "
I had to cut.

 

[dominikz]

Is it worth working in settings in mode
default
or
sharp
?

If you're asking about the filter settings, with IK Multimedia ARC System 3 I typically used 'Natural' phase and 'Sharp' filter type in my tests, as I found those settings best addressed the room modes in LF in my rooms. From your screenshot it looks like the 'default' filter sharpness is doing a pretty good job

Based on your screenshot may I suggest to also try limiting the correction to just 55Hz-250Hz? That should cover the main room-induced peaks and dips without changing the on-axis response of the loudspeaker.

 

[ai studio]

If you're asking about the filter settings, with IK Multimedia ARC System 3 I typically used 'Natural' phase and 'Sharp' filter type in my tests, as I found those settings best addressed the room modes in LF in my rooms. From your screenshot it looks like the 'default' filter sharpness is doing a pretty good job

Based on your screenshot may I suggest to also try limiting the correction to just 55Hz-250Hz? That should cover the main room-induced peaks and dips without changing the on-axis response of the loudspeaker.

hi @dominikz
I can do it

Do you think it works better?
For Mix

I did a TEST like you said
Up to 250
Sounds good

Frequencies to the end
Sounds more open
And hears small subtleties
And more "clean"
But maybe it's not real and it's an illusion and it can fool me in the mix.?

Based on your experience with the software

What can you recommend.

Or is it just trial and error.
And try to mix in 2 modes
And decide which is more appropriate.?

Thank you very much for your help.

 

[dominikz]

hi @dominikz
I can do it

Do you think it works better?
For Mix

I did a TEST like you said
Up to 250
Sounds good

Frequencies to the end
Sounds more open
And hears small subtleties
And more "clean"
But maybe it's not real and it's an illusion and it can fool me in the mix.?

Based on your experience with the software

What can you recommend.

Or is it just trial and error.
And try to mix in 2 modes
And decide which is more appropriate.?

Thank you very much for your help.

Here I'd like to say that I'm not a professional mixing engineer so take my advice on this with a grain of salt

IMHO any reasonably flat loudspeaker should be OK to make tonality-related mixing decisions. Your pair seem to be within +/-3dB >250Hz measured in-room, so it seems to me they are probably just fine. If you want to correct above e.g. 500Hz in your shoes I'd probably try and look for anechoic measurements of your speakers and create an EQ profile based on that.

If you're happy with how other music sounds on your speakers, you're likely to aim for the same tonality when mixing on them. Flattening the entire FR with auto-EQ like ARC may or may not be 'more true to the source', as it can also make things worse if it 'fixes' certain/peaks dips above transition frequency that might be related to perceptually benign reflections (and not loudspeaker direct sound issues). This would require more analysis for one to be sure.
Therefore when using well-engineered loudspeakers I typically just limit EQ correction to the range below the Schroeder/transition frequency to just fix issues resulting from room modes, and leave the rest as-is.

 

[dasdoing]

@ai studio I would use "linear" for mixing because hearing the real transients is important when dailing in compressors and limiters

 

[ai studio]

Ok thanks for the tips friends

Thanks for the tip and information
dominikz


Thanks for the tip and information

dasdoing

I will test TEST linear

@ai studio I would use "linear" for mixing because hearing the real transients is important when dailing in compressors and limiters

Here I'd like to say that I'm not a professional mixing engineer so take my advice on this with a grain of salt

IMHO any reasonably flat loudspeaker should be OK to make tonality-related mixing decisions. Your pair seem to be within +/-3dB >250Hz measured in-room, so it seems to me they are probably just fine. If you want to correct above e.g. 500Hz in your shoes I'd probably try and look for anechoic measurements of your speakers and create an EQ profile based on that.

If you're happy with how other music sounds on your speakers, you're likely to aim for the same tonality when mixing on them. Flattening the entire FR with auto-EQ like ARC may or may not be 'more true to the source', as it can also make things worse if it 'fixes' certain/peaks dips above transition frequency that might be related to perceptually benign reflections (and not loudspeaker direct sound issues). This would require more analysis for one to be sure.
Therefore when using well-engineered loudspeakers I typically just limit EQ correction to the range below the Schroeder/transition frequency to just fix issues resulting from room modes, and leave the rest as-is.