Post-DIRAC Live ART Calibration - Lack of Bass

[BG Extraordinaire]

Thank you for the warm welcome Kawauso! And thank you both for taking the time to give me your advice. I've been reading through the ART thread and plan to keep digging into the questions and experience shared there, as well as post any additional questions in that thread.

As a conclusion to my ART bass experience, I did do another fresh measurement run last night after cutting the gain on my subs, better matching the SPL from my mains and to have them hit the middle of the pre-measurement volume test screen as recommended. I'm not sure what I had failed to do previously, but this time the +11db curve sounded fantastic. Unfortunately, all of this was after potentially damaging one of the woofers in my left tower speaker To your point on the +10/-10, that was definitely incorrect and I believe sending a far-too-hot signal to my mains. It does sound obvious in retrospect, and I feel really dumb for this outcome. Although that tower isn't measuring as well as it did before, ART is compensating nicely. I'm hoping to be able to get a replacement woofer to remedy this situation I've gotten myself into. But at least now I know better, and hopefully this can be a cautionary tale for others new to Dirac and ART to not try fiddling with it and stick to the published guidelines.

 

[Oddball]

Thank you for the warm welcome Kawauso! And thank you both for taking the time to give me your advice. I've been reading through the ART thread and plan to keep digging into the questions and experience shared there, as well as post any additional questions in that thread.

As a conclusion to my ART bass experience, I did do another fresh measurement run last night after cutting the gain on my subs, better matching the SPL from my mains and to have them hit the middle of the pre-measurement volume test screen as recommended. I'm not sure what I had failed to do previously, but this time the +11db curve sounded fantastic. Unfortunately, all of this was after potentially damaging one of the woofers in my left tower speaker To your point on the +10/-10, that was definitely incorrect and I believe sending a far-too-hot signal to my mains. It does sound obvious in retrospect, and I feel really dumb for this outcome. Although that tower isn't measuring as well as it did before, ART is compensating nicely. I'm hoping to be able to get a replacement woofer to remedy this situation I've gotten myself into. But at least now I know better, and hopefully this can be a cautionary tale for others new to Dirac and ART to not try fiddling with it and stick to the published guidelines.

Sorry but just to clarify - did you actually blow the woofer or not? This would be a first on ASR so we would like to get closure on that. Is the woofer working at all or just producing distorted sound?

 

[BG Extraordinaire]

It's working, but the bass output appears to be reduced. At normal volumes, I can't audibly hear a difference in what I have listened to so far. However, I get distortion from that speaker now when turning up the volume past a certain point on bass-heavy content (loud, but not crazy levels - haven't measured what dB though). I'm including the measurements below that led me to this conclusion. However, I would love to be told I'm wrong and that there's some other cause of that that woofer distorting (perhaps just me still selecting poor settings on Dirac ART) and that this is an easy fix to get my speaker back to how it was performing previously. Let me know if there's anything else I can provide to give better clarity.

Here are the two mains (yellow right, purple left) Thursday night, when I ran my ill-advised hot-subwoofer measurements. These results are very typical to what I've experienced in prior runs, be it Dirac, Audyssey, or REW.

And here's Friday night, measuring after the incident I noted. Notice the reduced bass output on the left speaker. I remeasured a few times and continued to get that output. As mentioned previously, I've done REW measurements, but only of the stereo sweeps (no single speaker measurements).

When doing REW sweeps, I'm getting similar results in totality from both before (orange line) and after what happened Friday afternoon (purple line). There was a slight difference in MV on the AV10 between the two measurements, which muddies the picture a bit (not to mention the more appropriate trim levels from the AV10 after the Friday calibration). I think the REW sweeps look better than the Dirac measurements as these are post-ART, so the support from the other speakers is masking the impact. I don't usually run individual speaker sweeps when calibrating, so I unfortunately wouldn't have anything from REW to compare to before this happened.

Here's 10-200hz with the smoothing turned off

 

[Chromatischism]

What you're going through sounds like the exact same thing I went through with standard Dirac Live, so I don't think it is unique to ART. Several people coming from Audyssey have had trouble getting Dirac bass to sound as impactful as what they are used to.

Is just cranking up the bass the solution? I'm not sure. It wasn't for me. I'll follow this thread and hope the community can come to a conclusive reason for why this is happening.

 

[Oddball]

What you're going through sounds like the exact same thing I went through with standard Dirac Live, so I don't think it is unique to ART. Several people coming from Audyssey have had trouble getting Dirac bass to sound as impactful as what they are used to.

Is just cranking up the bass the solution? I'm not sure. It wasn't for me. I'll follow this thread and hope the community can come to a conclusive reason for why this is happening.

ART will not sound like Audy - thanks god for that. Impactful is a complex word. In Audy that is more like distorded.

We do have a consensus as far as ART bass levels are concerned. Turn it up to to +12dB. That's max. If not up to your taste than I guess other solutions might work better.

 

[kawauso]

It's working, but the bass output appears to be reduced. At normal volumes, I can't audibly hear a difference in what I have listened to so far. However, I get distortion from that speaker now when turning up the volume past a certain point on bass-heavy content (loud, but not crazy levels - haven't measured what dB though). I'm including the measurements below that led me to this conclusion. However, I would love to be told I'm wrong and that there's some other cause of that that woofer distorting (perhaps just me still selecting poor settings on Dirac ART) and that this is an easy fix to get my speaker back to how it was performing previously. Let me know if there's anything else I can provide to give better clarity.

Here are the two mains (yellow right, purple left) Thursday night, when I ran my ill-advised hot-subwoofer measurements. These results are very typical to what I've experienced in prior runs, be it Dirac, Audyssey, or REW.
View attachment 498681

And here's Friday night, measuring after the incident I noted. Notice the reduced bass output on the left speaker. I remeasured a few times and continued to get that output. As mentioned previously, I've done REW measurements, but only of the stereo sweeps (no single speaker measurements).
View attachment 498682

When doing REW sweeps, I'm getting similar results in totality from both before (orange line) and after what happened Friday afternoon (purple line). There was a slight difference in MV on the AV10 between the two measurements, which muddies the picture a bit (not to mention the more appropriate trim levels from the AV10 after the Friday calibration). I think the REW sweeps look better than the Dirac measurements as these are post-ART, so the support from the other speakers is masking the impact. I don't usually run individual speaker sweeps when calibrating, so I unfortunately wouldn't have anything from REW to compare to before this happened.
View attachment 498684

Here's 10-200hz with the smoothing turned off
View attachment 498685

It does indeed look like one channel is behaving oddly. With an AV10, NCx500, and Ascend ELX Towers, I don’t see this as a configuration or connection that would normally lead to speaker damage.
In my own experience, I once nearly ran into trouble when connecting an A1H to a miniDSP HTx and speakers, due to a gain-setting mistake that caused the signal to clip by exceeding the miniDSP’s 4V maximum input. Fortunately, the speakers were unharmed.

 

[kawauso]

For example, even if the L/R speaker trims after Dirac calibration on an AV10 end up at +10 dB,
the maximum master volume of the preamp should be limited to 0 dB or below in that case.
I don’t believe the AV10 would output a clipped signal.

 

[Miniyouuuu]

I have no direct Dirac ART experience but in my case…
-before moving to Umik2 microphone, Dirac filters where inaccurate, soundstage was narrower. After changing the microphone, Dirac improved dramatically. Timing accuracy is important, and Umik1 and other low cost microphones are not accurate enough for this matter, resulting in worst Dirac implementation.
-bass reduction improves clarity. Dirac tries to get the tighter bass possible at any cost but you have to get used to the new sound of your system. First impression could be: where is my bass??

 

[Dovydas]

There's no difference between Umik-1 and Umik-2 for Dirac. Dirac engineer mentions it himself in this youtube video. Sorry, can't find quickly an exact time where he says that, but it's definitely in the first half of the video.

 

[Miniyouuuu]

The accuracy of the inputs makes no differences on the results... what an irony!
https://www.audiosciencereview.com/forum/index.php?threads/umik-1-vs-umik-2.36912/post-2475994

If you see that post you can observe minor discrepancies between mics. None of these are crucial but on a dual channel sum, minor phase differences are shown.

Because linearity in Dirac is more important than precission, Umik1 is perfectly capable as mic for Dirac, but I don't agree about the results they provide. Having timing issues in the inputs for a software that tries to fix phase and delay is important.

Maybe not for getting night&day results, but you pay 900$ for a software and >3k$ in gear and you don't want the better input accuracy for your measurements?
We set our systems to get closer to our desirable target curve and all the quality we get from the DSP will depend on the precission of the inputs we provide. I was happy with my Umik1, now I'm happier with my Umik2. Is not gimmick or snake oil. The Umik2 is more precise, and its measurements from one day will affect my everyday DSP processed sound everyday.

 

[peng]

I have no direct Dirac ART experience but in my case…
-before moving to Umik2 microphone, Dirac filters where inaccurate, soundstage was narrower. After changing the microphone, Dirac improved dramatically. Timing accuracy is important, and Umik1 and other low cost microphones are not accurate enough for this matter, resulting in worst Dirac implementation.
-bass reduction improves clarity. Dirac tries to get the tighter bass possible at any cost but you have to get used to the new sound of your system. First impression could be: where is my bass??

I don't believe the Umik-2 would make such big differences, there's no technical reasons for that, but since you experienced such issues (such as "inaccurate filters" when using the Umik-1, then I think there may be other reasons. If the Umik-1 is that bad, Dirac Live would not have recommended its use.

 

[alvigg]

The interview makes clear that both work equally well.

 

[Dovydas]

I'm not going to pretend being an expert here, but from what I understood, calibration file is what makes both mics even and Dirac does not use higher sample rate for measurements, so as peng has mentioned, no technical reason for the difference.

 

[alvigg]

I'm not going to pretend being an expert here, but from what I understood, calibration file is what makes both mics even and Dirac does not use higher sample rate for measurements, so as peng has mentioned, no technical reason for the difference.

That is confirmed in the interview as well.

 

[peng]

The accuracy of the inputs makes no differences on the results... what an irony!
https://www.audiosciencereview.com/forum/index.php?threads/umik-1-vs-umik-2.36912/post-2475994

If you see that post you can observe minor discrepancies between mics. None of these are crucial but on a dual channel sum, minor phase differences are shown.

Because linearity in Dirac is more important than precission, Umik1 is perfectly capable as mic for Dirac, but I don't agree about the results they provide. Having timing issues in the inputs for a software that tries to fix phase and delay is important.

Maybe not for getting night&day results, but you pay 900$ for a software and >3k$ in gear and you don't want the better input accuracy for your measurements?
We set our systems to get closer to our desirable target curve and all the quality we get from the DSP will depend on the precission of the inputs we provide. I was happy with my Umik1, now I'm happier with my Umik2. Is not gimmick or snake oil. The Umik2 is more precise, and its measurements from one day will affect my everyday DSP processed sound everyday.

I understand there are differences, just wonder if such would actually result in the audible differences such as soundstage you mentioned, I would think even moving one's head a little while listening to music would have make much more timing differences than the two mics would do. Also, ART is more for the 20-200 Hz range, does the timing accuracy between the two that significant in that LF range?

 

[Chromatischism]

ART will not sound like Audy - thanks god for that. Impactful is a complex word. In Audy that is more like distorded.

I have no "distortion" from Audyssey, so I'm not sure what would go wrong to cause that.

 

[Oddball]

I have no "distortion" from Audyssey, so I'm not sure what would go wrong to cause that.

I might have used a wrong word - better one would be less fidelity, at least in my system that is not simple. Much better phase alignment overall, less decay and better FQ response with ART.

As to distortion, you could get to that if you push Audy system too far. Either with running big speakers as full range or feeding them LFE via LFE distribution. Audy was doing exactly what it was told but as bass content in soundtracks varies it was getting out of hand at times so the speakers would start to distort. Especially audible with port chuffing, less so with sealed but still noticeable if you know the speakers well. Obviously you could dial back and not experience these issues. With ART this really does not happen as works completely differently.

 

[Chromatischism]

I might have used a wrong word - better one would be less fidelity, at least in my system that is not simple. Much better phase alignment overall, less decay and better FQ response with ART.

Sure phase alignment can make an audible difference if it would otherwise cause frequency response issues (dips, etc). But if you don't have those issues, you wouldn't hear a difference.

Contrary to some opinions (maybe due to believing in marketing) Dirac bass does not sound better than Audyssey bass when all else is equal. ART may be different, but whether it's better will depend on you. Audyssey doesn't do something similar to make a comparison.

 

[Oddball]

Sure phase alignment can make an audible difference if it would otherwise cause frequency response issues (dips, etc). But if you don't have those issues, you wouldn't hear a difference.

Contrary to some opinions (maybe due to believing in marketing) Dirac bass does not sound better than Audyssey bass when all else is equal. ART may be different, but whether it's better will depend on you. Audyssey doesn't do something similar to make a comparison.

Don't want to make a long post out of this but generally agree with you. Bass control was always much better for smaller systems, but Audy offered a whole bunch more to tweak with bigger systems. It remains questionable as to how wise and beneficial it was to spend so much time on Audy, but it is a fact.

I also don't want to sound like Dirac salesman (which was at times implied), but you either try ART or you stay in the silent, or vocal, and uniformed opposition. Not much more to say about it.

 

[AudioJester]

All of us with Denon and Marantz gear can easily compare Audy and Dirac ART.
For me its not even close, everything else being equal - in my untreated, accoustically terrible multipurpose room. Look at all the measurements people have posted with ART, Audy can never achieve that.