Active Room Treatment (ART) by Dirac

[Oddball]

This is from MDAT ART file, that somebody posted here [or on AVSforum - impressive stuff and room acoustics better than 90% of rooms out there..
But if you look closely there is visible mismatch between ART and above ART frequencies - EDT looks completely different, like fivefold. Energy of bass fundamentals is long gone, there are still harmonics flying around for next half second. It makes LFE feel shallow, even when there is objectively high SPL signal there.

This correlates to my listening experience - while all the systems were clearly capable and bass was clean, punchy and tight, with lot of addictive qualities, there was something off - sheer weight and size were just not there - I have my reference tracks and clips and e.g. with fast panning bass heavy scenes in Top Gun or F1 you did not get this hit of air pressure that should be there. It reminded me of those “audiophile” 2 way speakers with mid bass bloom, designed for first impression.

What you actually want id linear increase of decay times below 100Hz, it can be freely double at 20Hz, once you remove ringing and resonances. My target is to be at 0.5-0.6s at 20Hz [which is below first room mode so no standing waves, just pressure build up]. Is a thing called Bass Ratio and it is how best classical concert venues are build - it creates envelopment and size.

View attachment 512055


View attachment 512054

Just for comparison - this is my room, with waveforming [dotted] vs ART [bold]

View attachment 512069

Thanks - a bit clearer now. Will need to digest in peace as now between (delayed) flights .

Couple of thoughts though. There are many different ART setups and perhaps not fair to just take a random one. I don't even know how was this ART system set up vs your waweforming. I will not volountier my MDAT file though as really had busy schedule and still not happy with graphs - although it all sounds really good.

In my room decay seems to creep up starting with 50hz or so and I do get into 0.6ms range at hottest points, and definitively at 20hz. I would expect that to happen in most systems that use aggressive shelves, aka +8dB or more. So the bass pressure level seems fine to overwhelming (aka +12dB shelf), and rest of the range is clear and crisp.

I did experience 2 Trinnov theaters with wave-forming and they were really impressive, but not really a fair comparison to my multi-purpose highly irregular "room" - they were dedicate large HT rooms with all the bells and whistles.

 

[kawauso]

Here is what my LFE-channel support assignment looks like in my JBL room (LR and four subwoofers).
Surprisingly, it seems that quite a lot of higher-frequency content is being assigned to the subwoofers.

 

[Frank207be]

Here is what my LFE-channel support assignment looks like in my JBL room (LR and four subwoofers).
Surprisingly, it seems that quite a lot of higher-frequency content is being assigned to the subwoofers.
View attachment 512177

Same here.

Now I see the subs injecting peaks up to -2dB at 150Hz and -8dB at 160Hz towards my main speakers. With the previous LPF support from subs to mains quickly rolled off after 90Hz with only subwoofer 3 as exception doing -15dB at 110Hz . No wonder that I noticed such dramatic difference

 

[ban25]

Just got the AV30 setup in my system and completed my first 13-point ART calibration. I took the opportunity to make a couple of tweaks to the placement of my main speakers, namely moving them apart a further 8" and out from the wall a further 4".

I'll do REW measurements this weekend, but subjectively this is the best the system has ever sounded by far!

 

[Oddball]

Just got the AV30 setup in my system and completed my first 13-point ART calibration. I took the opportunity to make a couple of tweaks to the placement of my main speakers, namely moving them apart a further 8" and out from the wall a further 4".

I'll do REW measurements this weekend, but subjectively this is the best the system has ever sounded by far!

View attachment 512256

Congrats on your new AV30 and looks like you are already enjoying it.

Also nice to see 98 filters - AV10 has "only" 94.

 

[Frank207be]

Here two screenshots from the 106Hz LPF vs 200Hz LPF on my 4 subwoofers. I had to recalculate the previous one with the lower LPF but the result stays the same.

I didn't use the curtain at 1.7kHz with the new calibration to tame the Focal sparkle.

Edit: a simulation of both 106 and 200Hz LPF in WinISD:

 

[kawauso]

Here two screenshots from the 106Hz LPF vs 200Hz LPF on my 4 subwoofers. I had to recalculate the previous one with the lower LPF but the result stays the same.

I didn't use the curtain at 1.7kHz with the new calibration to tame the Focal sparkle.

Edit: a simulation of both 106 and 200Hz LPF in WinISD:

Between the two setups, the upper limit of the subwoofer support range is different,
and in the latter setup it looks like more is being allocated to the subwoofers. Is the LPF setting different as well?

 

[kawauso]

Just got the AV30 setup in my system and completed my first 13-point ART calibration. I took the opportunity to make a couple of tweaks to the placement of my main speakers, namely moving them apart a further 8" and out from the wall a further 4".

I'll do REW measurements this weekend, but subjectively this is the best the system has ever sounded by far!

View attachment 512256

Congratulations on your AV30 purchase and getting ART set up. It sounds like your first impressions were very good.
I also noticed the in-room response of the Blades. The mids and highs look perfect even without correction.

 

[Frank207be]

Between the two setups, the upper limit of the subwoofer support range is different,
and in the latter setup it looks like more is being allocated to the subwoofers. Is the LPF setting different as well?

Yes, the first scheenshot is with the old 106Hz LPF and the second one is with the 200Hz LPF.

 

[Foggy1956]

Apologies for not looking through all the info, conflicting info on the web. Should Dirac live and DLBC be run prior to running
ART? Familiar with Dirac live, new to DLBC and ART.

 

[Chromatischism]

Apologies for not looking through all the info, conflicting info on the web. Should Dirac live and DLBC be run prior to running aa
ART? Familiar with Dirac live, new to DLBC and ART.

According to others, it will be a waste of time because the ART module is self-contained and will not use information from DL/BC.

 

[smygolf]

Apologies for not looking through all the info, conflicting info on the web. Should Dirac live and DLBC be run prior to running
ART? Familiar with Dirac live, new to DLBC and ART.

Nope, just do the sweeps and then chooce ART instead of DL/DLBC

 

[Foggy1956]

According to others, it will be a waste of time because the ART module is self-contained and will not use information from DL/BC.

But doesn't ART only work on frequencies up to 150hz, what happens if I'd like to equalize up to say 500hz?

 

[Oddball]

But doesn't ART only work on frequencies up to 150hz, what happens if I'd like to equalize up to say 500hz?

When you choose ART it will engage all layers of Dirac software, so you will be able to do EQ up to or beyond 500hz.

 

[Foggy1956]

When you choose ART it will engage all layers of Dirac software, so you will be able to do EQ up to or beyond 500hz.

Thanks, that's what I was hoping, love learning about new toys.

 

[dlaloum]

My intention is to eventually have mains that are flat to 20hz at reference level and then add 2-4 additional subs for ART to work with. Should be best of both worlds.

Nothing succeeds like excess

 

[dlaloum]

Congrats on your new AV30 and looks like you are already enjoying it.

Also nice to see 98 filters - AV10 has "only" 94.

Would be nice to get an understanding of what is driving those filter # limits and why the AV10 has fewer than the AV30 - even though they share the same DSP.... (is it perhaps caution due to the additional channels the AV10 has over the AV30?... reserving processing bandwidth for the additional channels?)

 

[middlemarch]

Would be nice to get an understanding of what is driving those filter # limits and why the AV10 has fewer than the AV30 - even though they share the same DSP.... (is it perhaps caution due to the additional channels the AV10 has over the AV30?... reserving processing bandwidth for the additional channels?)

I'm wondering if you've hit on it. My A1H ever only shows 92 available. It has the same number of channels as the AV10, however, so maybe not? I've been thinking it has to do with the number of channels actually in use. For that I'm maxed out at 9.4.6. But it seems others run that configuration but have more than 92 available? It's confounding to me...

 

[middlemarch]

On a different topic, something that has been driving me crazy. It has me going back to my original reluctance to try Dirac again after my DDRC24 experience, all revolving around lack of software maturity, more specifically, error handling.

So I've been merrily trying to optimize my ART configuration (the "open door" tune is working great, by the way, kitty is happy his potty won't be disturbed). This of course involves making tweaks and then loading them into the receiver over wifi. I've been having a devil of a time with downloads failing with an error saying "Activating Filter Request Timed Out", although a green text box says it loaded fine.

In fact, it didn't. It erases the chosen slot but nothing is put in it's place. Then I'll get another error "Msg 40" which says the device configuration has changed, when in fact nothing has changed, then crashes back to the "select receiver" screen. Once this happens there's no more talking to the receiver, the only solution is a power on reset (pulling the plug from the wall, going to standby doesn't do it).

I've even had one instance where all three slots were wiped out, checking Setup:Audioirac Live showed only "OFF", but using the Options function showed the previous residents of those slots. But I couldn't actually select any of them.

After days of back and forth I 'think' I may be getting somewhere. I've noticed that this happens when trying to load a new filter over my network when I'm a few rooms away from the receiver. This means a couple of hops over my mesh network. My receiver is hardwired into my base mesh node, which is in the same room as the receiver. Once I've restored communications via the POR, I can then load filters to my hearts content, as long as I'm in the same room as the receiver!

So my suspicion is that the Denon or the Dirac software on my laptop is somehow extremely picky about the timing, and the extra hops via the mesh nodes throws one or the other for a loop. I know how to use a network analyzer and could watch the TCP-IP traffic, but I'm just not that motivated. Funny that it can download all the data involved with the filters seemingly OK, but bombs when trying to activate it?

What is clear is that both the Denon and Dirac software are lacking in the error handling department and it's pretty sad that you have to resort to pulling the wall plug to get things going again, but here we are.

I have a support ticket into Dirac, but suspect it'll just be the old finger pointing I experienced with them and MiniDSP, each blaming the other. Or my network.

So I guess I was wondering if any of the rest of you have had any similar experiences or if it's just me. At this point I'm back to my perception of the Dirac software being very brittle. I feel like I'm dancing on egg shells every time I try to do something, should just be happy that I can get anything done, I suppose. But criminy, I paid $800 smackers for this stuff. It's the most expensive piece of software I've EVER purchased, by a country mile. You'd think for that you'd at least be able to get to a support ticket on their website. But I had to dig through the ASR threads to find where FLAK had posted the email address for their help desk. Try finding that on their website.

 

[dlaloum]

On a different topic, something that has been driving me crazy. It has me going back to my original reluctance to try Dirac again after my DDRC24 experience, all revolving around lack of software maturity, more specifically, error handling.

So I've been merrily trying to optimize my ART configuration (the "open door" tune is working great, by the way, kitty is happy his potty won't be disturbed). This of course involves making tweaks and then loading them into the receiver over wifi. I've been having a devil of a time with downloads failing with an error saying "Activating Filter Request Timed Out", although a green text box says it loaded fine.

In fact, it didn't. It erases the chosen slot but nothing is put in it's place. Then I'll get another error "Msg 40" which says the device configuration has changed, when in fact nothing has changed, then crashes back to the "select receiver" screen. Once this happens there's no more talking to the receiver, the only solution is a power on reset (pulling the plug from the wall, going to standby doesn't do it).

I've even had one instance where all three slots were wiped out, checking Setup:Audioirac Live showed only "OFF", but using the Options function showed the previous residents of those slots. But I couldn't actually select any of them.

After days of back and forth I 'think' I may be getting somewhere. I've noticed that this happens when trying to load a new filter over my network when I'm a few rooms away from the receiver. This means a couple of hops over my mesh network. My receiver is hardwired into my base mesh node, which is in the same room as the receiver. Once I've restored communications via the POR, I can then load filters to my hearts content, as long as I'm in the same room as the receiver!

So my suspicion is that the Denon or the Dirac software on my laptop is somehow extremely picky about the timing, and the extra hops via the mesh nodes throws one or the other for a loop. I know how to use a network analyzer and could watch the TCP-IP traffic, but I'm just not that motivated. Funny that it can download all the data involved with the filters seemingly OK, but bombs when trying to activate it?

What is clear is that both the Denon and Dirac software are lacking in the error handling department and it's pretty sad that you have to resort to pulling the wall plug to get things going again, but here we are.

I have a support ticket into Dirac, but suspect it'll just be the old finger pointing I experienced with them and MiniDSP, each blaming the other. Or my network.

So I guess I was wondering if any of the rest of you have had any similar experiences or if it's just me. At this point I'm back to my perception of the Dirac software being very brittle. I feel like I'm dancing on egg shells every time I try to do something, should just be happy that I can get anything done, I suppose. But criminy, I paid $800 smackers for this stuff. It's the most expensive piece of software I've EVER purchased, by a country mile. You'd think for that you'd at least be able to get to a support ticket on their website. But I had to dig through the ASR threads to find where FLAK had posted the email address for their help desk. Try finding that on their website.

Although I do have a mesh setup, all of my Dirac configuration connectivity has been done from within the same room - and therefore from the same primary mesh AP... to which the AVR is also connected...

I have never experienced any of these kinds of network issues - but as I said, my use case is very unlikely to test latency/hops issues like yours.
I also think that the vast majority of users will be connecting within the same room - and therefore the testing of this type of issue has been very very limited...

The issue might potentially be ameliorated by connecting either AVR and/or PC to a hardwired connection.... even if connected to seperate AP's in the messh, with a WiFi backhaul, you would eliminate some hops in the process... (just a thought)