My Thoughts on Audyssey MultEQ vs Dirac

[ansat]

I would like to start by stating that my comparisons and measurements are based on MultEQ XT. NOT XT32. But I felt that if Dirac could not hold its own against XT, then it would not be a competitor against Audyssey.

Today I received my USB Dayton Audio mic from Cross Specturm (This is my 2nd mic from them and I highly recommend anyone getting the mic to spend the extra 25 - 30$). I downloaded the Dirac 2 channel trail and had the pleasure of getting everything working on windows 8. I loaded up the latest version of REW and calibrated everything. I took a measurement of my room and the noise floor was a nasty 55db, however for this test that would not matter much. I decided to test everything upstairs as that is where the only remaining piece of equipment that has Audyssey lives. Enter the Denon 3808ci, equipped with the dynamic eq addon.

I dug out the calibration mic and proceeded to re-setup the denon as I had it set to manual eq previously. I ran all eight positions and then disabled the audyssey curve. I connected up the Dayton mic and started with the Dirac software.

Now before I get to deep in the software I want to make sure that I am clear that the version that will be packaged with the XMC will be unique to the XMC and will be a lighter version of the Dirac Live that is available on the website. The heavier version of the program was hinted to be available for purchase on a upgrade pricing model to be determined. But this is about dirac and Audyssey and to if Dirac can be a competitor.

When I loaded up the dirac software I was presented with a really nice interface that I think that anyone could operate. Everything was nice and clean and got straight to the point. Each point that the measurement tool wanted me to address was clearly labeled and easy to work with. The total setup time didn't take much longer then audyssey did. I calibrated all 9 points and was ready to get the REW measurements.



The measurements that I took can be found at the bottom of this post.

If you look to my target response above you will notice that this matches up with the averages below while Audyssey decided my lower frequencies needed to be adjusted as low as they were.


Since Dirac's claim to fame is in the time domain I got some impulse measurements as well. While Audyssey did pretty good on my left speaker, It seems to make it worse for my right speaker when you make the comparison. Dirac made a noticeable improvement in the measurements to Audyssey on both speakers and beat the Base measurements by a fair margin.

Left Base Impulse

Right Base Impulse

Left Audyssey Impulse

Right Audyssey Impulse

Left Dirac Impulse

Right Dirac Impulse


The other graphs in the REW file lead me to believe that by the numbers Dirac is superior to Audyssey MultEQ XT. Based on user reviews of xt32 I think its going to be a tight race and Audyssey has some pretty awesome competition. In the end, for me, If Dirac elects to keep the ability for the user adjustable curve in its port for the XMC I feel like I have my new room correction system. I hope this helps others out there feel more comfortable with Dirac and everything that they promised, my measurements proved to be accurate.

Tony

Edit
For easier reference.

Hello All,
Promised was a test result of audyssey vs Dirac vs nothing.

Test:
Setup:
Fronts A - Definitive Technology BP6

Receiver - Denon 3808ci Audyssey V. MultEQ XT
AMP - Sherbourn Zone Amplifier.


The Receiver Left and Right out are plugged into the zone amp’s bus 1 and 2.
Each speaker is running off of bridged mode on the amp.


I ran audyssey using all available locations (8) available and according to the instructions. 5 on the couch and three in front of the couch.


The audyssey results are as follows.


Speraker Config Check
Front - Small
Center None
Sub yes
surround a none
surround b none
surround back none


Distances:
Front Left - 7.3 ft
Front Right - 7.1 ft
Subwoofer - 11.6 ft


Channel Level Check
Front L - (-12db)
Front R - (-11db)
Subwoofer - (-7.5 db)


Crossover check
Fronts - 40hz

I calibrated according to Dirac instuctions with audyssey off.

I took a single measurement from the sweet spot with audyssey off then with Dirac on, then with Audyssey on (Dirac Disabled)

I used 1/1 smoothing in REW for all measurements Noise floor is high as my kitchen is right next to my living room. Noise floor in this room is 55db

Left Speaker Compare

Right Speaker Compare

Left Base Impulse

Right Base Impulse

Left Audyssey Impulse

Right Audyssey Impulse

Left Dirac Impulse

Right Dirac Impulse


REW FILE : ansatsusha.com/ansat/diracvaudyssey.mdat

Tony

sorry for the short response.
sent from my phone.

[Topend]

Nice comparison, thanks for sharing. I would be keen to read about a similar comparison once the XMC-1 is released.

Dave.

[hemster]

Thanks Tony for sharing your effort and your objective viewpoints.

Dirac sure looks to be a pretty solid offering and I'm looking forward to using it on my XMC-1 come February. Makes me wonder how much better the full version of Dirac would be.

[klinemj]

Great comparison. I am glad you did it. It does indeed support what Mikael told me about their effectiveness in time adjustment vs. Audyssey.

Mark

[TKO]

Sept 16, 2013 22:14:47 GMT -5 ansat said:

I would like to start by stating that my comparisons and measurements are based on MultEQ XT. NOT XT32. But I felt that if Dirac could not hold its own against XT, then it would not be a competitor against Audyssey.

Today I received my USB Dayton Audio mic from Cross Specturm (This is my 2nd mic from them and I highly recommend anyone getting the mic to spend the extra 25 - 30$). I downloaded the Dirac 2 channel trail and had the pleasure of getting everything working on windows 8. I loaded up the latest version of REW and calibrated everything. I took a measurement of my room and the noise floor was a nasty 55db, however for this test that would not matter much. I decided to test everything upstairs as that is where the only remaining piece of equipment that has Audyssey lives. Enter the Denon 3808ci, equipped with the dynamic eq addon.

I dug out the calibration mic and proceeded to re-setup the denon as I had it set to manual eq previously. I ran all eight positions and then disabled the audyssey curve. I connected up the Dayton mic and started with the Dirac software.

Now before I get to deep in the software I want to make sure that I am clear that the version that will be packaged with the XMC will be unique to the XMC and will be a lighter version of the Dirac Live that is available on the website. The heavier version of the program was hinted to be available for purchase on a upgrade pricing model to be determined. But this is about dirac and Audyssey and to if Dirac can be a competitor.

When I loaded up the dirac software I was presented with a really nice interface that I think that anyone could operate. Everything was nice and clean and got straight to the point. Each point that the measurement tool wanted me to address was clearly labeled and easy to work with. The total setup time didn't take much longer then audyssey did. I calibrated all 9 points and was ready to get the REW measurements.

<portion deleted for brevity>

The other graphs in the REW file lead me to believe that by the numbers Dirac is superior to Audyssey MultEQ XT. Based on user reviews of xt32 I think its going to be a tight race and Audyssey has some pretty awesome competition. In the end, for me, If Dirac elects to keep the ability for the user adjustable curve in its port for the XMC I feel like I have my new room correction system. I hope this helps others out there feel more comfortable with Dirac and everything that they promised, my measurements proved to be accurate.

Tony

Thanks Tony,

appreciate the analysis and the measurements.  Looks like Dirac is a very able competitor to Audyssey, ARC, and the others.

Just to offer a word of caution though - be very careful with trying to correct nulls, especially the one that shows up around 52-55Hz.  This looks like a frequency boost of 5 to 12 dB.  It may correct that null at that specific listening position while using >10 times more power but the person two feet to the left on the sofa gets this incredible peak of muddy and overwhelming bass.  From my experience I would stick to +3dB as the maximum boost for nulls in the frequency range < 300Hz.

Cheers.

[geebo]

Thanks for posting, Tony. I'm a bit surprised by Ausyssey's bass roll-off. But you seem to have confirmed what I'm hearing with the Dirac trial. A real improvement all around.

[ansat]

TKO,

I didnt do any corrections to the curve for the comparison, both audyssey and dirac corrected the null that exists in my room. Given the simplified design of dirac there is no limits that can be set for how much gain the filter adds. This can be done however by setting the individual target curves for each speaker and putting dips in the curve to make up for it. I dont believe that this functionality will remain in the version that will be given out in the xmc. For reviewing purposes I went with the mindset of set it and forget it as 99 percent of users will use this method.


Tony

sorry for the short response.
sent from my phone.

[Jim]

Sept 17, 2013 7:11:28 GMT -5 geebo said:

Thanks for posting, Tony. I'm a bit surprised by Ausyssey's bass roll-off. But you seem to have confirmed what I'm hearing with the Dirac trial. A real improvement all around.

I've read that bass seems to be where people have the most complaints about Audyssey. Granted, low frequencies are the hardest to tame, but still. The complaint that I've read more than a few times is that the bass was worse with Audyssey on (ie, more roll-off).

I've experienced that too, where I've had to try to tweak it with the Audyssey Pro kit, to try to get it flatter.

I'm impressed with the DIRAC results, they look promising.

[geebo]

Sept 17, 2013 7:10:55 GMT -5 TKO said:

Sept 16, 2013 22:14:47 GMT -5 ansat said:

I would like to start by stating that my comparisons and measurements are based on MultEQ XT. NOT XT32. But I felt that if Dirac could not hold its own against XT, then it would not be a competitor against Audyssey.

Today I received my USB Dayton Audio mic from Cross Specturm (This is my 2nd mic from them and I highly recommend anyone getting the mic to spend the extra 25 - 30$). I downloaded the Dirac 2 channel trail and had the pleasure of getting everything working on windows 8. I loaded up the latest version of REW and calibrated everything. I took a measurement of my room and the noise floor was a nasty 55db, however for this test that would not matter much. I decided to test everything upstairs as that is where the only remaining piece of equipment that has Audyssey lives. Enter the Denon 3808ci, equipped with the dynamic eq addon.

I dug out the calibration mic and proceeded to re-setup the denon as I had it set to manual eq previously. I ran all eight positions and then disabled the audyssey curve. I connected up the Dayton mic and started with the Dirac software.

Now before I get to deep in the software I want to make sure that I am clear that the version that will be packaged with the XMC will be unique to the XMC and will be a lighter version of the Dirac Live that is available on the website. The heavier version of the program was hinted to be available for purchase on a upgrade pricing model to be determined. But this is about dirac and Audyssey and to if Dirac can be a competitor.

When I loaded up the dirac software I was presented with a really nice interface that I think that anyone could operate. Everything was nice and clean and got straight to the point. Each point that the measurement tool wanted me to address was clearly labeled and easy to work with. The total setup time didn't take much longer then audyssey did. I calibrated all 9 points and was ready to get the REW measurements.

<portion deleted for brevity>

The other graphs in the REW file lead me to believe that by the numbers Dirac is superior to Audyssey MultEQ XT. Based on user reviews of xt32 I think its going to be a tight race and Audyssey has some pretty awesome competition. In the end, for me, If Dirac elects to keep the ability for the user adjustable curve in its port for the XMC I feel like I have my new room correction system. I hope this helps others out there feel more comfortable with Dirac and everything that they promised, my measurements proved to be accurate.

Tony

Thanks Tony,

appreciate the analysis and the measurements.  Looks like Dirac is a very able competitor to Audyssey, ARC, and the others.

Just to offer a word of caution though - be very careful with trying to correct nulls, especially the one that shows up around 52-55Hz.  This looks like a frequency boost of 5 to 12 dB.  It may correct that null at that specific listening position while using 100 times more power but the person two feet to the left on the sofa gets this incredible peak of muddy and overwhelming bass.  From my experience I would stick to +3dB as the maximum boost for nulls in the frequency range < 300Hz.

Cheers.

Good words of caution but Dirac does, in fact, limit boost. This is from the manual:

"The filter design algorithm has a hard-coded maximum gain value that limits the boost at any frequency."

It doesn't say what that maximum is however.

Also, I was told that very often Dirac can correct for nulls by phase adjustments or something (I don't remember exactly) as mentioned by the Dirac representative at Emofest. But there is more than just boosting and cutting going on here.

[bootman]

Outstanding work ansat.
Bravo.

The next step is to compare results with x32 but this comparison with xt is a fair one given the competition.

Example; Marantz AV7701 also has XT at a similar price point as the XMC.

[ansat]

Geebo
When I get home I will confirm / deny the phase alteration.

Tony

sorry for the short response.
sent from my phone.

[ansat]

Bootman

I have no access to anything with xt32. We will have to rely on someone else with the gear available. Unless someone in st louis has something that I can use.

Tony

sorry for the short response.
sent from my phone.

[bootman]

Sept 17, 2013 9:02:33 GMT -5 ansat said:

Bootman

I have no access to anything with xt32. We will have to rely on someone else with the gear available. Unless someone in st louis has something that I can use.

Tony

sorry for the short response.
sent from my phone.

This is a valid cause.
Surely someone from that area can help out!

[jdskycaster]

Tony,
Thanks for taking the time to do this comparison. The Dirac results look promising and although everyone's outcome will differ depending on their specific room characteristics this is still very useful info. I am also currently using XT but recently had a chance to audition an AVR with XT32 in my HT just for the sake of comparison and to see if upgrading would be worth it.

I have to say that in my room I did not notice a major improvement with XT32. The low end was a bit better (was able to run the subs a little less hotter than I currently am with XT) but not enough of a difference to justify the upgrade for me. In your room maybe XT32 would provide better improvement. It would be interesting to see the results if you can get someone to loan you one for the sake of direct comparison.

JD

[ansat]

14 days left on the trial. I would need something set up soon.

Tony

sorry for the short response.
sent from my phone.

[LCSeminole]

Tony,

Thanks for your insight on the DIRAC trial. Hopefully I'll be feeling up to playing around with the trial version this weekend.

[geebo]

Sept 17, 2013 20:23:20 GMT -5 LCSeminole said:

Tony,

Thanks for your insight on the DIRAC trial. Hopefully I'll be feeling up to playing around with the trial version this weekend.

My trial ends this weekend so you can pick up where I left off.

But seriously,I found it very easy to learn and use and it has a much more interactive feel to it than other systems I've used. I'm also very happy with the results I've gotten with it. I'm going to miss it.

Will you be using the multi-channel or stereo version?

[LCSeminole]

The multi-channel version has peaked my interest. Once I have an XMC-1 I have a feeling the full version will be calling me with a deal I can't refuse.

[ansat]

Tko, geebo

Looking at my 51 hz

Taking off the 1/1 smoothing and digging in deep the 50 - 60 hz range of the spl in the Dirac measurement. The base was 58 db @ 51hz and dirac increased it to 64db. Looking across the whole spl it appears that there is a fixed limit of 6db of gain that dirac is willing to add. Audyssey's max appears to be 8db according to the graph but in many cases audyssey did a lot of attenuating. At 51 hz there was no change to the phase of the speaker along with any other changes that my mic and REW can detect put in place by Dirac.

[geebo]

Sept 17, 2013 21:00:37 GMT -5 LCSeminole said:

The multi-channel version has peaked my interest. Once I have an XMC-1 I have a feeling the full version will be calling me with a deal I can't refuse.

I'll be very interested in the multi-channel results. I'd try it but there's just not enough time left to devote to it.