Interaction between room correction filters and DAC filters

[Davide]

Hi all,

I am new to the forum and since I believe it is an excellent meeting place for competent and passionate people I have signed up to submit a question.

I am in the process of adopting room correction software (I tend to Dirac) but a doubt has arisen about the interaction that it can have with the digital filters of the DACs.

As we know, DAC filters determine effects on frequency, phase and echo, factors on which room correction software go to act.

For this reason I spontaneously ask myself what effect the different types of filters (at least Sharp and Slow) have on the filters generated for the room correction.

My skills in this area are not so thorough, however I am sure that the effect of two filters in series is something very complex to determine, especially if the filter that is upstream is modeled on the basis of what is downstream.

These effects are probably negligible on hearing, however I think it may be constructive to discuss them.

From my point of view it would be nice to determine which is the best DAC filter to carry out the measurement process for room correction, in order to prevent unexpected effects from loopback triggering...

So I would like to leave the floor to whoever is competing on the matter.

 

[AudioStudies]

Is the doubt yours alone, or from something you have read? I would think the Dirac would correct to match the room irregardless of what is fed from a given digital filter. If you have read something on this issue, I am curious where you found it.

 

[dc655321]

Hi all,

I am new to the forum and since I believe it is an excellent meeting place for competent and passionate people I have signed up to submit a question.

I am in the process of adopting room correction software (I tend to Dirac) but a doubt has arisen about the interaction that it can have with the digital filters of the DACs.

As we know, DAC filters determine effects on frequency, phase and echo, factors on which room correction software go to act.

For this reason I spontaneously ask myself what effect the different types of filters (at least Sharp and Slow) have on the filters generated for the room correction.

My skills in this area are not so thorough, however I am sure that the effect of two filters in series is something very complex to determine, especially if the filter that is upstream is modeled on the basis of what is downstream.

These effects are probably negligible on hearing, however I think it may be constructive to discuss them.

From my point of view it would be nice to determine which is the best DAC filter to carry out the measurement process for room correction, in order to prevent unexpected effects from loopback triggering...

So I would like to leave the floor to whoever is competing on the matter.

Welcome to the forum.
Assuming all else is equal, use the "sharp" filter.

Room correction typically targets the low end of the audible spectrum (eg: 20-200Hz), where a typical residential room exerts the strongest influence.

The reconstruction filter in a DAC is a low-pass filter with a transition region at the very top of the audible/ultrasonic region (eg: > 20000Hz). This filter is not present to produce audible effect(s), but rather to interpolate between samples and suppress ultrasonic content. The effect of room correction filters in series with the DAC filter is fairly trivial to calculate (convolution, assuming LTI filters).

TL;DR
Unless for some odd reason your room correction filters target the upper end of the audible spectrum and your DAC filter is "boutique", there is little to be concerned about here.

 

[JoachimStrobel]

I have a NanoAvr with Dirac. The NanoAvr has a build in DAC, and passes the Diraced signal on via HDMI too. I feed that HDMI via an de-embedder into an ancient Parasound DAC.
The Dirac correction measures the same and corrects the same regardless if I use the NanoAVR‘s Dac or the Parasound DAC as output. The Parasound sound slightly better, with and much better without correction than the NanoAVR. But with correction enabled it sounds better overall. The means that Dirac levels the DACs. That makes since as the NanoAVR‘s DAC is of low quality while the Parasound Dac2000 once had a place in a Stereophile ranking.

 

[DonH56]

The DAC's filters essentially work above audible range so the chances of them affecting room correction programs is essentially zilch. Especially considering the impact the room itself has along with speaker response.

IMO - Don

 

[Davide]

Well guys, thanks for the replies.
Mine was a personal doubt, nothing derived from other information.
Either way, my Dirac automatically sets up to around 22 kHz, so there is a certain band in which both filters directly overlap.
Actually at the direct level I don't think there are any particular complexities, my question was more related to the ringing effects generated on the rest of the band (both in amplitude and in time domain).
But I did some tests and in fact the difference with the various filters cannot be determined. Or rather, if I repeat the calibration N times (with the same DAC filter), I have seen that each generates a slightly different correction filter, probably due to the manual positioning of the microphone or noise levels... therefore I recognize that the DAC filter is negligible in this process (hearing speaking).
Out of curiosity, I tried to ask Dirac support and they suggested me to use the Super Slow Roll Off (AK4493), and once the calibration was done, modify the DAC filter to find the one I like best.

 

[Willem]

Try with Dirac correction limited to e.g. only below 500 Hz. Not because of interaction with the DAC filter but because room correction usually does not work well above the lower frequencies.

 

[Kal Rubinson]

The NanoAvr has a build in DAC,

Mine doesn't. HDMI only.

 

[JoachimStrobel]

Mine doesn't. HDMI only.

There are two versions