Audyssey XT32 vs Dirac Live

[HarmonicTHD]

No the default Audyssey curves will try and make the response flat at the listening location, with some high freq roll-off for the reference mode. In my system I don't see that high freq roll off and it sounds bright (my Audyssey mic may not be correctly calibrated for high freqs, I'm not entirely sure why I don't see it), and the mids/upper bass get recessed. The target curve I use is just a -1 dB/dec target, like in all of the plots Amir shows for speaker reviews, red arrow in this example. So for example I would put in something like:

100 Hz: +2.5 dB
1 kHz: 0 dB
10 kHz: -2.5 dB

And use REW/UMIK1 to verify the results and adjust it as needed. Actually I now just use MultEQ-X and just put in a -1 dB/dec tilt, makes it easier (but I also keep the additional high freq roll-off, to compensate whatever my mic seems to be doing incorrectly).

Thanks appreciate it. Yes I also always used REW/Umik just to double check what Audyssey is doing. Will dive into it again sometimes next week.

 

[GalZohar]

If you found Dirac off better than on then there was probably something wrong with your calibration? Differences aren't supposed to be so great that one is better on and another better off? Did you use the same target curve?

 

[chych7]

If you found Dirac off better than on then there was probably something wrong with your calibration? Differences aren't supposed to be so great that one is better on and another better off? Did you use the same target curve?

I doubt it, I used REW/UMIK-1 to verify the results and Dirac did what it was supposed to do, fit the target curve. I also checked the impulse response, and indeed Dirac fixed it in the way it was designed to. Like I said, this is not an uncommon finding, other people (including Amir) limit Dirac to <500 Hz (just for bass), because it doesn't sound good for mid-high freqs. You can read more details about my methodology, target curves assumed, and measurements in the MultEQ-X thread.

 

[HarmonicTHD]

BTW. Does anyone know if someone ever did such a thing as a controlled comparison test between the Dirac and Audyssey?

So we would have some real facts and not have to rely on personal perceptions.

I thinking along the lines of taking an Onkyo (Dirac) and a Denon (XT32) AVR. Establish a baseline FR using REW. Then let each System do its automatic adjustment. Do another REW check and compare how each system adjusted amplitude and Phase etc. or sth like that ….

I could not find anything like it but that doesn’t mean it does not exist … so if it does please point me in the right direction.

 

[GalZohar]

=

BTW. Does anyone know if someone ever did such a thing as a controlled comparison test between the Dirac and Audyssey?

So we would have some real facts and not have to rely on personal perceptions.

I thinking along the lines of taking an Onkyo (Dirac) and a Denon (XT32) AVR. Establish a baseline FR using REW. Then let each System do its automatic adjustment. Do another REW check and compare how each system adjusted amplitude and Phase etc. or sth like that ….

I could not find anything like it but that doesn’t mean it does not exist … so if it does please point me in the right direction.

I don't remember which thread it was but there was a comparison and after aligning the target curves to be as similar as possible, calibrating both to a single position, Audyssey and Dirac had a nearly overlapping frequency response measurement, with some phase differences in some areas.

 

[chych7]

=


I don't remember which thread it was but there was a comparison and after aligning the target curves to be as similar as possible, calibrating both to a single position, Audyssey and Dirac had a nearly overlapping frequency response measurement, with some phase differences in some areas.

That's literally me, like a day ago posting about this, in this same exact thread. All you need to do is click the links and read.

Audyssey XT32 vs Dirac Live

Well, at least it's supposedly better than EQing each individually and independently. But yeah, seems like there are much better ways to do it, but they require additional hardware. It might be possible to get reasonable results with a Y splitter and some DSP built-in to the sub for delay/phase...

audiosciencereview.com

Would be great if other people can take the time to do such an analysis.

 

[GalZohar]

I think it's very strange that the frequency response is almost the same yet one sounds better disabled while the other sounds better enabled...

 

[chych7]

I think it's very strange that the frequency response is almost the same yet one sounds better disabled while the other sounds better enabled...

There's way more to how we perceive sound than just plain frequency response/SPL, especially when considering time domain effects, reflections, and phase. Dirac certainly does something different with phase and created more L/R errors in the mid range, as was shown in my analysis. Clarity plots were also different in the upper freqs; whatever Dirac did reduced direct sound and increased reflections, which reduced the clarity response.

 

[anotherhobby]

It seems like an impossible task to compare Audyssey to Dirac in a useful way, but maybe somebody will come up with a method that's meaninful. I have two systems that are vastly different, with one running Dirac Live on a miniDSP Flex while the other is Audyssey XT32 on a Denon AVR-X4300H. I do happen like both a lot, but I don't know how to compare them in a way that would be meaningful to other people.

miniDSP Flex w/ Dirac Live
Topping PA5 powering Revel M105s
Crown XLS1500 powering two 8" Dayton Ultimax based subs I built
miniDSP handles 24dB/oct LR crossovers at 80 Hz
No additional EQ, Dirac Live is the only correction used
*Treated 115 sq/ft office and used for near field listening at 3 feet

Denon AVR-X4300H w/ Audyssey XT32
Internal amps powering Focal Aria 936's
Two JL Audio e112 Subs, first corrected with MSO filters on an miniDSP 2x4
Denon handles crossovers at 80 Hz
*Untreated 1000 sq/ft room and used in a TV/living room at 11 feet

It is clearly not useful to compare the setups as they are now since everything is vastly different between them, and they are obviously in different rooms. However, I really like both of them for how each corrected for the room it's in. I like Dirac's software UI and process a little better than the new MultEQ-X software, but I got good end results from both. Also, after a lot of testing and tweaking with each, I ended up preferring full range correction on both for different reasons (I have waffled back and forth on this one, but that's where I am now).

I have the time and interest to try and perform a valid comparison, but I'm at a loss for how would one design a valid and meaningful test using a config people would care about.

When I say a valid and meaningful test, I'm mostly meaning setup and configuration stuff. For example... With Audyssey, I was able to get the best sub response across 2 seats by using a miniDSP and Multi-Sub Optimizer first, and then running Audyssey after that. Since Audyssey XT32 only has one sub correction curve that's applied to both subs, getting better results with MSO isn't surprising, and the results were substantially different. However, I was able to get better sub results in my office with Dirac all on it's own and no MSO EQ. I tried it both ways. Since Dirac Live corrects only R/L, and the subs are crossed with the mains at 80 Hz on those channels, it corrects the L main with the L sub and the R main with the R sub. So while not exactly independent sub correction, it does effectively EQ each sub on it's own. When I used MSO the results were odd and I didn't like it, but I also think a lot of that could be the room and near field use case.

I feel like no matter how I set a test up, people would be like "well that's not really valid/useful/interesting/etc because you set up the subs not the way I would use them bla bla bla." And then do we care about how well each one was able to be tweaked to it's best and most optimal curve (adding parametric EQ in the Audyssey software, or making changes in the Dirac curve)? It seems that is more of a test of how good I am personally with dialing in each one, but in the end the ability to tweak those things does make a difference. Would we only care about base/default curves? But is that valid if most people want to tweak it for different results? Adding in MSO throws in a ton of variables, but Audyssey sucks at more than one sub, and Dirac Live thinks sub's are just the low end of your R/L channels, so I don't even know where to start there.

In other words, I think due to the level of complication and the number of options involved in the setup/config of each system, I can't see how I could actually do a valid and meaningful test for anybody other than myself. I'd say the same of other people as well, but maybe somebody has different physical gear that makes this easier to test in a more broad way for others.

 

[HarmonicTHD]

That's literally me, like a day ago posting about this, in this same exact thread. All you need to do is click the links and read.

Audyssey XT32 vs Dirac Live

Well, at least it's supposedly better than EQing each individually and independently. But yeah, seems like there are much better ways to do it, but they require additional hardware. It might be possible to get reasonable results with a Y splitter and some DSP built-in to the sub for delay/phase...

audiosciencereview.com

Would be great if other people can take the time to do such an analysis.

Thanks and sorry I must have overlooked your links to your detailed analysis in the other thread. I like it a lot and thanks for your work - just what I was thinking too.

I didn’t fully digest all your results yet and will have a more detailed look over the weekend. Will also dig out my REW measurements before and after Audyssey correction. Especially if I also see the boost in the highs you showed in the post 463. I can’t get my head around yet why the LR FR responses are virtually identical btw Dirac and Audyssey but the combined response shows the boost in the highs for Audyssey. The results from hobbyuser do not exhibit the high boost. Also I remember some months ago seeing some results from penh, which also did not seem to be boosted in the highs if I remember correctly. Mhhh - will certainly have a look into my results, but can’t remember nor do I perceive a particular bright sound.

Also phase seems interesting. If I see correctly (currently on my iPhone traveling), Audyssey has an almost identical phase response in comparison to the original whereas Dirac does some change around upper bass if I recall correctly.

Anyhow. Thx again for your work and the insights.

 

[chych7]

When I say a valid and meaningful test, I'm mostly meaning setup and configuration stuff. For example... With Audyssey, I was able to get the best sub response across 2 seats by using a miniDSP and Multi-Sub Optimizer first, and then running Audyssey after that. Since Audyssey XT32 only has one sub correction curve that's applied to both subs, getting better results with MSO isn't surprising, and the results were substantially different. However, I was able to get better sub results in my office with Dirac all on it's own and no MSO EQ. I tried it both ways. Since Dirac Live corrects only R/L, and the subs are crossed with the mains at 80 Hz on those channels, it corrects the L main with the L sub and the R main with the R sub. So while not exactly independent sub correction, it does effectively EQ each sub on it's own. When I used MSO the results were odd and I didn't like it, but I also think a lot of that could be the room and near field use case.

I feel like no matter how I set a test up, people would be like "well that's not really valid/useful/interesting/etc because you set up the subs not the way I would use them bla bla bla." And then do we care about how well each one was able to be tweaked to it's best and most optimal curve (adding parametric EQ in the Audyssey software, or making changes in the Dirac curve)? It seems that is more of a test of how good I am personally with dialing in each one, but in the end the ability to tweak those things does make a difference. Would we only care about base/default curves? But is that valid if most people want to tweak it for different results? Adding in MSO throws in a ton of variables, but Audyssey sucks at more than one sub, and Dirac Live thinks sub's are just the low end of your R/L channels, so I don't even know where to start there.

In other words, I think due to the level of complication and the number of options involved in the setup/config of each system, I can't see how I could actually do a valid and meaningful test for anybody other than myself. I'd say the same of other people as well, but maybe somebody has different physical gear that makes this easier to test in a more broad way for others.

Yes I agree it's a very difficult thing to compare the two room correction approaches. One thing I did was to completely avoid the whole subwoofer part, and only focus on the speaker response (subwoofers disabled). That will at least simplify some of the complexity.

But yes, target curves is another complexity, what to use? And then what about the mic setup pattern? There are lots of scenarios that could go into such a comparison, but starting with some reasonable assumptions may get some reasonable (but not 100% conclusive) results. For me personally, I setup the assumptions in a way that was important to me, because ultimately I was deciding if I wanted to go with Dirac or stay with Audyssey. I hope whatever findings I had is useful to others, but results may vary considerably across different equipment, rooms, and initial assumptions.

 

[chych7]

I can’t get my head around yet why the LR FR responses are virtually identical btw Dirac and Audyssey but the combined response shows the boost in the highs for Audyssey. The results from hobbyuser do not exhibit the high boost. Also I remember some months ago seeing some results from penh, which also did not seem to be boosted in the highs if I remember correctly. Mhhh - will certainly have a look into my results, but can’t remember nor do I perceive a particular bright sound.

May be my eyes but I'd think the Audyssey looks a little bit elevated in the highs in the LR curves. The LR phase error plots also show that the Audyssey response has lower error at upper frequencies, which would also mean the summed response is stronger (more in phase). But I do suspect my Audyssey mic is not calibrated correctly, and contributes to the boosted highs.

 

[Dougey_Jones]

There's way more to how we perceive sound than just plain frequency response/SPL, especially when considering time domain effects, reflections, and phase. Dirac certainly does something different with phase and created more L/R errors in the mid range, as was shown in my analysis. Clarity plots were also different in the upper freqs; whatever Dirac did reduced direct sound and increased reflections, which reduced the clarity response.

I could be wrong here, but due to the types of filters used, doesn't Dirac also attempt to time-align the sound where Audyssey just does FR correction?

 

[chych7]

I could be wrong here, but due to the types of filters used, doesn't Dirac also attempt to time-align the sound where Audyssey just does FR correction?

They both are using time domain corrections (FIR filters). Audyssey is less aggressive in using it from what I can tell, applying it more to lower frequencies and being lighter on it on higher frequencies.

You can look at the impulse response charts I had:

(full range)

Audyssey's Next Generation of Room Correction (MultEQ-X)

We all understand that incorrect position. There is no miscommunication. Have you tried taking confirmatory measurements properly - same area, random sample points? Incorrect position, aping measure point? He's only talking about an apples to apples comparison in a specific sense. You are...

audiosciencereview.com

(higher freqs)

Audyssey's Next Generation of Room Correction (MultEQ-X)

We all understand that incorrect position. There is no miscommunication. Have you tried taking confirmatory measurements properly - same area, random sample points? Incorrect position, aping measure point? He's only talking about an apples to apples comparison in a specific sense. You are...

audiosciencereview.com

It looks like Audyssey is improving impulse response decay time, but isn't trying to change the phase like Dirac is.

 

[SimpleTheater]

Looks like the Audyssey vs DIRAC argument is going to start all over again after the Scientific Audiophile just posted the microphone fix for both of them.

 

[HarmonicTHD]

Looks like the Audyssey vs DIRAC argument is going to start all over again after the Scientific Audiophile just posted the microphone fix for both of them.

Would you have a link please? Google wasn’t of much help.

 

[SimpleTheater]

Would you have a link please? Google wasn’t of much help.

Fix for microphone issues link.

 

[chych7]

Check todays date guys.

 

[Dougey_Jones]

Fix for microphone issues link.

Can't believe I just watched that.

 

[HarmonicTHD]

Oh gee. Nicely done. ;-)