Room correction with Acourate

Agree. I outlined my learning curve, but now it seems very simple to do a second and third room. Maybe you can make fewer errors.

This has been my biggest concern with Acourate. I have emailed Uli @ acourate asking how Acourate can get a proper measurement of the entire listening space based on a single point measurement.

Please report back.

Same thoughts here.

Still, I dont fully understand the trouble of the Roon Team to make that compatible?
Is this an idelogical problem?
Even if JRiver did that?
And I pay $18 a year for JRiver.

Architectural limitation: a plugin in the classical single-box environment has the user interface integrated with the processing, cannot be separated. That’s not the Roon model.

@brian comments here.

Thanks. Thats understandable now.
On that basis the Dirac VST plugin is more of a dead end.

Here is a discussion about Acourate, Dirac and Audiolense. Its some 3 years old, maybe a bit outdated.

A Dirac Review on Stereophile:

A bit more on Single vs Multipoint Measuremnet.

I picked up Mitch’s book, spun up Acourate and worked out my office system today.

Thanks for the tutorials @AndersVinburg and @rovinggecko (and of course @Uli_Brueggeman for producing the software).

It is a somewhat weird room because it’s connected to three other spaces, but the speakers are a pair of near-field monitors pointed at a single listening position, so at least that part is simple. Like I suspect most people will do, I used a UMIK-1 and ASIO4ALL.

Hearing the slow log sweep make its way through the room, swelling dramatically as it hit the room modes, was really interesting. With REW, the test sweep is only a few seconds long, so you don’t get such a tangible depiction of the room modes during the measurement.

The improvement is very clear. Uncorrected, if I close my eyes, the speakers sounded a bit like point sources. With the correction, the imaging is a lot more coherent, and of course the bass sounds a lot more even handed.

I think I want to go back and do more iteration after I’ve spent a few days with this, but so far I’m happier with this result than what I was getting out of my work with REW, and it only took a couple of hours of following instructions and a little bit of iteration.

1 Like

fyi, i never heard back from uli/audioverve about my question re. single point vs multi point measurements … guess they don’t want to sell a license.

Question from the Acourate User Group:


I’m looking into Accurate and Dirac.

Is there a basic difference of where Dirac uses an approach to measure and correct for multiple spaces in a room because as they put it, "a perfect correction in one single point most always implies degradation everywhere else.”? And does Acourate focus only on one ideal listening position?

Does the Dirac multi position approach produce a more even, and compromised, response? Does the Accurate approach produce a more ideal single listening position at a greater compromise to the rest of the room?

This specifically could apply to my room, I think, because it is a large workspace where I’m moving around a lot.



Yes, there are basic differences between Dirac and Acourate.

Acourate mainly uses the measurement at the listening position. This results in the best correction at thee sweet spot. But. because Acourate applies a different approach in the calculations the result is not only valid for a single position. So you can walk around and you will notice that the sound at other positions is not getting bad. The Acourate calculation avoids over-boosts by principle…

Dirac, Audyssee, Lyngdorf use different calculations by averaging multiple frequency responses. They need to do this to to also avoid over-boosts.

Despite the typical single point measurement Acourate allows you to do multiple measurements and to combine them by functions like

  • averaging the frequency responses
  • calculating the max. envelope for a bundle of frequency responses
  • averaging multiple pulse responses

The last method produces very good results for the beamforming. method. This method is a good approach to get quasi anechoic results.

It is possible for you to test Acourate. See So you can do a single point measurement and then listen to the result at the sweet spot but also by walking around in your room.

  • Uli

I like to add a further remark to the topic:

We have now learnt for a long time that the frequency response is an important indicator for the quality of a system. And indeed it is.
Now we learn that at different positions the frequency response is different, there will be even big differences.

So it is time to learn more: even with a corrected system all these frequency responses are different, maybe even very different. Despite the correction !
And more important: it does not matter if the filter is made by a single point measurement or by multi-point measurements. The applied filter will improve, hopefully, the playback. But at different positions the frequency responses will be different.

So the challenge is to find a filter which produces a good result. And the filter shall not create a very bad result at a room position.
In the meantime everyone should know that correcting deep dips in the frequency response causes an overboost and this directly leads to bad results. So in any way an overboost must be avoided.

But there is another important point:
we must relax a bit more and we must not only focus on the frequency response !

A frequency response describes a steady-state behaviour. There is good example to get an understanding. take a nice music track (you love it because it is so nice) and calculate the frequency response of the track. Now reverse the track and calculate another frequency response. Both frequency responses are fully identical! But of course the track played backwards sounds somehow different [by magic? :slight_smile: ] and usually it does not sound very pleasant.
This example tells us, that music is not steady-state. Time is a very important factor in the game.

Acourate uses algorithms which also analyse the time behaviour of the speakers. Thus a single-point measurement is sufficient in normal listening environments.
The success of Acourate now for more than 10 years can IMHO be considered as a good indicator. Otherwise a multi-point measurement procedure would have been implemented. The underlying maths is not mystic.

  • Uli

Thanks for sharing…

Thanks for all your Input here.
I got Accourate and your writings here made it possible for me to use Accourate without bigger troubles.
My first results sound promising.

I’m not seeing any sample rate for Acourte. Sorry if I’m missing something here. Does it do up to 768kHz?

Highest filter you can generate at this moment is 384 kHz.
At this moment, on version 1.9.8, you can select:

1 Like

Well, that’s better than the Dirac solution in this instance, as well, isn’t it, being that Dirac only does 192kHz? I believe that’s what I read, though I admit I haven’t read too deep into Dirac.

Correct - Dirac only supports 192 and has to come after Roon or HQPlayer and thus effectively makes software based upsampling questionable at best. Acourate filters can be run inside Roon before or after upsampling.

For Dirac it also depends how you apply it. For instance embedded in minidsp it uses a (much) lower rate (96 or 48 kHz).