There will be a whole new version of Dirac this year, allowing to produce VST plugins to be used in 3rd party software as Jriver or Roon or maybe HQP.
Up until now you had a “Asio” driver as product to use with your software.
I asked Flavio at Dirac and he granted me access to that new software in Beta status.
Now I have to run through the room measurement procedure to create these plugins.
Whenever I find the time i do that and report it here.
Looks very promising.
1 Like
Ah ok. Yes their VST plugin was always planned - I believe the beta has been out there for a while. Won’t help us with Roon, not any time soon anyway.
I thought maybe a version was coming out that would allow exporting filters.
Oh well, stand down excitement. 
In Accurate you can also use different measurement techniques, like beamforming.
Yes, I read about the beam forming technique, very interesting. (It uses several measurements along a straight line from the speaker, and when you add them the direct sound is reinforced while the reflections are averaged away.)
But that is the opposite of the multi point approach used by Dirac and others.
I’m still unclear on which is best. Or why Acourate does not recommend it (Uli and Mitch don’t argue against multi point, just don’t mention it at all). I have one hunch, that Acourate can do a good job from a single measurement because it pays great attention to the response over time. Acourate reports the correlation of signals within different time segments: 0-10 ms, 10-20, 20-80, 80-plus. And it reports predicted changes to those. So I imagine that allows it to do the best possible job. It doesn’t address differences between different seating positions, but neither does Dirac or anybody else: multi point averaging will create an average correction, if there are huge differences they remain. Or maybe Acourate does address the difference by the time delay analysis, I don’t know. But averaging will just produce an average correction.
In my room, post-Acourate the differences across seating positions were within 3 dB, mostly, and the response that Acourate created for the central seat was a good average.
This is all my hunch.
I posted some links to Dirac ‘papers’ up above somewhere. I understand from those that their multipoint measurement is more than just averaging.
A few snippets from the second Dirac paper (linked again here so the source is clear), a bit out of context but gives a hint as to the discussion:
If we build an inverse filter based on a simplistic average model of the original transfer function measured in different listening positions, we might end up correcting for issues late in the tail of that model that are not in accord with perception. Consider the following plot taken from a rather poor, but large, listening room, using a very good speaker, measured in 12 different positions corresponding to two different seats in a sofa. The plot is a Fourier magnitude response based on a simple delay-and-average response, i.e. the impulse responses were aligned in time and then averaged (i.e. beamforming), before taking the Fourier transform.
And
The naïve averaging led us to compensate for issues that are (a) nonexistent in any single listening position and (b) psycho-acoustically irrelevant. It is well worth noting that correcting for something inaudible may lead to something audible.
And
Based on the same twelve measurements we used to produce the plot above, we can form estimates of more psycho-acoustically relevant representations by using different types of time- frequency analysis.
Yes, my excitement is gone.
That .dll format of the VST Plugin is not compatible with Roon.
Works perfect with JRiver.
Are there any plans to make that Compatible?
Does anybody know about that?
It’s been discussed by the Roon team, but it’s not officially on the roadmap as far as I’m aware. That said I’m hopeful…
My hunches may be all wrong.
Will have to do more reading.
In between listening.
It’s not a ‘light’ read (not for me anyway), and there’s plenty I can’t quite follow, but it’s really interesting all the same.
Some people hold such incredible knowledge of their specific fields, most of which is pretty much gobbledegook to anyone else. I didn’t know most if this stuff even existed a few years ago, yet clearly some people have been studying it for decades.
So that means waiting ( a year or so) or spending another $500 for new software plus time to learn how to run it.
Does not sound compelling.
I believe Brian wrote about this, that the VST model is designed for an architecture where the server has a GUI. Is the GUI. Doesn’t fit the Roon model which separates processing and GUI.
if I am forgiven for harping, this Is another example of the new world vs. classical computers that I wrote about.
I have no emotions in that.
My Question is what to buy: Accourate or Audiolense?
Some say Accourate, some Audiolens.
Maybe canvass who uses what here - I think there are more Accourate users.
Neither are very appealing in UI/ease terms, and are Windows only. I’m desperately holding off hoping someone will bring out a Dirac style offering. But I’m thinking it’s unlikely and may just have to man up and follow in the footsteps of @AndersVinberg.
I found, sonically, that acourate works much better for me than Dirac.
Ease of use is nice, but the result is more important for me. I found that the ease of use in Dirac restricted me too much.
Acourate is nit that daunting, when you follow the steps listed above it is pretty monkey see, monkey do.
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.