Audyssey XT32 vs Dirac Live

[Dj7675]

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.

IMO Dirac is much better at following the target curve than Audyssey. This is a blessing and a curse.. in my opinion there is not a one slope fits all for all rooms/speaker combinations. Depending on the dispersion of speakers and the amount of absorption/treatment in a room, a given room can sound bad with the wrong amount of slope. My main theater has quite a bit of treatment. I have tried a 1dB/octave slope many times and have always been disappointed even thought the graph looked very pretty. As a result, simply cutting off correction to around 500hz was the solution and it sounded very good. For full range correction, with a room curve at 0 dB at 200hz, and -1dB at 10khz sounds very good (With +6dB boost from 200hz to 20hz). I’m still on the fence though if it sounds better than just using Dirac below 500hz, and fixing a few speaker flaws with some targeted manual EQ above the transition area. I still think that may sound best but it now at least sounds good to me with a room curve that better matches my room and speakers.
My reason for posting is I think that one reason some may have an issue with Dirac above the transition is Dirac is doing what it is told to do which may audibly may not sound good...not necessarily because of Dirac but because the target curve people are using may not be a good fit for their room/speakers.

 

[GalZohar]

IMO Dirac is much better at following the target curve than Audyssey. This is a blessing and a curse.. in my opinion there is not a one slope fits all for all rooms/speaker combinations. Depending on the dispersion of speakers and the amount of absorption/treatment in a room, a given room can sound bad with the wrong amount of slope. My main theater has quite a bit of treatment. I have tried a 1dB/octave slope many times and have always been disappointed even thought the graph looked very pretty. As a result, simply cutting off correction to around 500hz was the solution and it sounded very good. For full range correction, with a room curve at 0 dB at 200hz, and -1dB at 10khz sounds very good (With +6dB boost from 200hz to 20hz). I’m still on the fence though if it sounds better than just using Dirac below 500hz, and fixing a few speaker flaws with some targeted manual EQ above the transition area. I still think that may sound best but it now at least sounds good to me with a room curve that better matches my room and speakers.
My reason for posting is I think that one reason some may have an issue with Dirac above the transition is Dirac is doing what it is told to do which may audibly may not sound good...not necessarily because of Dirac but because the target curve people are using may not be a good fit for their room/speakers.

Sounds right, although same can be said for Audyssey.

 

[hmt]

IMO Dirac is much better at following the target curve than Audyssey. This is a blessing and a curse.. in my opinion there is not a one slope fits all for all rooms/speaker combinations. Depending on the dispersion of speakers and the amount of absorption/treatment in a room, a given room can sound bad with the wrong amount of slope. My main theater has quite a bit of treatment. I have tried a 1dB/octave slope many times and have always been disappointed even thought the graph looked very pretty. As a result, simply cutting off correction to around 500hz was the solution and it sounded very good. For full range correction, with a room curve at 0 dB at 200hz, and -1dB at 10khz sounds very good (With +6dB boost from 200hz to 20hz). I’m still on the fence though if it sounds better than just using Dirac below 500hz, and fixing a few speaker flaws with some targeted manual EQ above the transition area. I still think that may sound best but it now at least sounds good to me with a room curve that better matches my room and speakers.
My reason for posting is I think that one reason some may have an issue with Dirac above the transition is Dirac is doing what it is told to do which may audibly may not sound good...not necessarily because of Dirac but because the target curve people are using may not be a good fit for their room/speakers.

I think it depends highly on your speakers directivity. A smooth directivity (or maybe even constant) yields in a very good EQability. When your directivity shiws th typical problems around 2KHz (crossover in many speakers) audyssey with the default BBC dip on might soud better.

 

[luft262]

Based on the opinions and comments I've read in some threads here recently there seem to be people with very strong negative opinions about the sound that Dirac Live produces. I have an Onkyo TX-RZ50 on order that I can still cancel if I wanted to go with a Denon X3700 instead. Did I make the wrong choice by choosing the receiver with Dirac?

Dirac is probably as good or better than Audessey, but the Denon AVR probably has a more accurate DAC and AMP so it's a tradeoff.

 

[Dougey_Jones]

Dirac is probably as good or better than Audessey, but the Denon AVR probably has a more accurate DAC and AMP so it's a tradeoff.

Really though?

Denon x3700:

Onkyo TX-RZ50:

Even in PreAmp mode, the Denon gains what, a few dB of SINAD? And that was before the DAC chip debacle over @ Denon. Dirac more than makes up for that difference IMO.. Seeing everyone downvote the RZ50 as a "Poor" performer while the x3700 is "Great" has really made me question the groupthink exhibited on ASR sometimes.

 

[chych7]

Really though?

Even in PreAmp mode, the Denon gains what, a few dB of SINAD? And that was before the DAC chip debacle over @ Denon. Dirac more than makes up for that difference IMO.. Seeing everyone downvote the RZ50 as a "Poor" performer while the x3700 is "Great" has really made me question the groupthink exhibited on ASR sometimes.

The 4 ohm load performance of the RZ50 amp is what really kills its reputation. However from real world examples of people using it with 4 ohm speakers and not having issues, this may be overblown. Nevertheless, still not well proven that Dirac is actually better than Audyssey.

 

[luft262]

Really though?

Denon x3700:

View attachment 198051

Onkyo TX-RZ50:

View attachment 198052

Even in PreAmp mode, the Denon gains what, a few dB of SINAD? And that was before the DAC chip debacle over @ Denon. Dirac more than makes up for that difference IMO.. Seeing everyone downvote the RZ50 as a "Poor" performer while the x3700 is "Great" has really made me question the groupthink exhibited on ASR sometimes.

That's a good point and I'm glad you pointed that out. However, how do the two amps compare?

Really though?

Denon x3700:

View attachment 198051

Onkyo TX-RZ50:

View attachment 198052

Even in PreAmp mode, the Denon gains what, a few dB of SINAD? And that was before the DAC chip debacle over @ Denon. Dirac more than makes up for that difference IMO.. Seeing everyone downvote the RZ50 as a "Poor" performer while the x3700 is "Great" has really made me question the groupthink exhibited on ASR sometimes.

Yeah, but look at the difference in the AMPs.

So you lose 10+ dB on the amp side, another dB or two on the DAC side, a bit of linearity, and on top of all that the Onkyo isn't recommended by Amir.

It's personal preference, but I don't think Dirac Live would be worth losing all of that. Room correction is great, especially below 500Hz, but Audyssey is still very good room correction. If you really want more advanced features they now offer Audyssey MultEQ XT32 and a more advanced microphone. In my particular case I also have a Umik-1, REW, and SVS subs that have built in PEQ functionality so the better amp and dac in the Denon are more attractive to me than Dirac Live. My old AVR was an Onkyo, albeit without Dirac, but I much prefer my Denon AVR all around.

Either AVR has its pros and cons, but the Denon is the better package IMHO and that's why so many people praise it around here. I agree, though, that looking closely at the data can show that various products are actually much closer in performance than it may at first appear and sometimes features are more important than performance. For example, I chose SVS subs over HSU despite the HSU digging deeper, because I wanted the features so I get what you're saying.

 

[Dougey_Jones]

That's a good point and I'm glad you pointed that out. However, how do the two amps compare?

Yeah, but look at the difference in the AMPs.

View attachment 198062
View attachment 198064

So you lose 10+ dB on the amp side, another dB or two on the DAC side, a bit of linearity, and on top of all that the Onkyo isn't recommended by Amir.

It's personal preference, but I don't think Dirac Live would be worth losing all of that. Room correction is great, especially below 500Hz, but Audyssey is still very good room correction. If you really want more advanced features they now offer Audyssey MultEQ XT32 and a more advanced microphone. In my particular case I also have a Umik-1, REW, and SVS subs that have built in PEQ functionality so the better amp and dac in the Denon are more attractive to me than Dirac Live. My old AVR was an Onkyo, albeit without Dirac, but I much prefer my Denon AVR all around.

Either AVR has its pros and cons, but the Denon is the better package IMHO and that's why so many people praise it around here. I agree, though, that looking closely at the data can show that various products are actually much closer in performance than it may at first appear and sometimes features are more important than performance. For example, I chose SVS subs over HSU despite the HSU digging deeper, because I wanted the features so I get what you're saying.

I hate to say this because it's going to sound a little snobbish, but, everyone touts the x3700's performance in PreAmp mode which assumes external amplification and then "makes hay" about the amp performance of the RZ50 as if a LOT of people aren't going to use external amplification.. I haven't used the onboard amplification in a receiver in over a decade and won't be going back, so that "10dB" swing in the comparison is actually a 2-3dB swing ASSUMING you have one of the original x3700's with the good DAC's and not a new one.

 

[luft262]

I hate to say this because it's going to sound a little snobbish, but, everyone touts the x3700's performance in PreAmp mode which assumes external amplification and then "makes hay" about the amp performance of the RZ50 as if a LOT of people aren't going to use external amplification.. I haven't used the onboard amplification in a receiver in over a decade and won't be going back, so that "10dB" swing in the comparison is actually a 2-3dB swing ASSUMING you have one of the original x3700's with the good DAC's and not a new one.

What external amp are you using and why? Unless it's a Benchmark it's probably not much better than the amp in the denon avr...

 

[luft262]

I hate to say this because it's going to sound a little snobbish, but, everyone touts the x3700's performance in PreAmp mode which assumes external amplification and then "makes hay" about the amp performance of the RZ50 as if a LOT of people aren't going to use external amplification.. I haven't used the onboard amplification in a receiver in over a decade and won't be going back, so that "10dB" swing in the comparison is actually a 2-3dB swing ASSUMING you have one of the original x3700's with the good DAC's and not a new one.

Here is also what gets me. People say Dirac is better than Audessy and that's fine, but that's subjective. The DAC and AMP in the Denon are objectively better. If your subjective opinion of Dirac is so high that it overrides that difference maybe just get the Onkyo and stop trying to justify its lower objective performance. It goes back to what I said about features. Sometimes getting the features you prefer is worth objectively lower performance if the difference is marginal or less important to you.

 

[Dougey_Jones]

What external amp are you using and why? Unless it's a Benchmark it's probably not much better than the amp in the denon avr...

I'm not really interested in your opinion of my external amplification as it doesn't change the argument I made above, but they're clearly listed in my signature. The PS Audio Trio's are ICE Power based, and I've swapped the modules to current 2022 versions (200ASC). Two channels bi-amping the center channel and one channel each for L/R. The rest of the channels are handled by a Rotel 5-Ch.

People usually tout the x3700's performance in preamp mode, so it makes sense to at least compare apples to apples when talking about other receivers that have pre-outs. Maybe that should be a focus of Amir's future testing methodology for AVR's.

 

[peng]

Really though?

Denon x3700:

View attachment 198051

Onkyo TX-RZ50:

View attachment 198052

Even in PreAmp mode, the Denon gains what, a few dB of SINAD? And that was before the DAC chip debacle over @ Denon. Dirac more than makes up for that difference IMO.. Seeing everyone downvote the RZ50 as a "Poor" performer while the x3700 is "Great" has really made me question the groupthink exhibited on ASR sometimes.

You have a point, but about groupthink, I can say the same about Dirac vs Audyssey. If you look at it in more detail, you might consider the reasons why there seem to be much more DL supporters than Audyssey's. To name a few of the possible reasons, DL has a much better user interface, its default target curve is more bass biased vs Audyssey's "flat" one, Audyssey won't EQ the BBC dip unless you disable MRC, so naturally for people who just run the REQ/RC once without doing any tweaking DL has a significant edge, but for those who spent time to get the best of either, the gap may be smaller than you think. Audyssey also is suffering from having a longer history, that is, there are many more people from the earlier versions still remembering, or suffering from the poorer performance of those versions that also ran and many are still running, on less capable processors.

If groupthink wasn't affecting people, the apparent gap between the DL vs Audyssey might be smaller too, just like Denon vs Onkyo in terms of pre out SINAD.

Back to Onkyo, the difference in SINAD for the pre out may be only 2 dB but I would caution that if you look at the actual curve instead of a single point at 2 V, Denon vs Onkyo gap is bigger than 2 dB over the important 50 mV to 1.2 V range.

 

[dlaloum]

You have a point, but about groupthink, I can say the same about Dirac vs Audyssey. If you look at it in more detail, you might consider the reasons why there seem to be much more DL supporters than Audyssey's. To name a few of the possible reasons, DL has a much better user interface, its default target curve is more bass biased vs Audyssey's "flat" one, Audyssey won't EQ the BBC dip unless you disable MRC, so naturally for people who just run the REQ/RC once without doing any tweaking DL has a significant edge, but for those who spent time to get the best of either, the gap may be smaller than you think. Audyssey also is suffering from having a longer history, that is, there are many more people from the earlier versions still remembering, or suffering from the poorer performance of those versions that also ran and many are still running, on less capable processors.

If groupthink wasn't affecting people, the apparent gap between the DL vs Audyssey might be smaller too, just like Denon vs Onkyo in terms of pre out SINAD.

Back to Onkyo, the difference in SINAD for the pre out may be only 2 dB but I would caution that if you look at the actual curve instead of a single point at 2 V, Denon vs Onkyo gap is bigger than 2 dB over the important 50 mV to 1.2 V range.

Sure, but when it came down to it - I had to decide between the Onkyo family or the Denon X3700....

My previous experience having been with Onkyo/Integra and Audyssey XT32 - the Audyssey was consistently disappointing - with my system sounding better without EQ than with Audyssey activated.

So here I was looking at an X3700 at circa $2400 or an Integra 3.4 at circa $1700 (Aus$) - take another chance on Audissey getting it right (knowing that with the app and adjustable target curves, and the ability to disable the MRC, it might finally get it right) at a price premium, or take a chance on Dirac, based on consistent positive reviews... and if I found it disappointing, I could easily resell the Integra at the same price I paid for it (or at a minor loss) - and then purchase the Denon.

From where I am sitting, this was a really easy decision - the difficult part was waiting for stock to arrive (!! 6 months !!).

After more than 10 years of disappointment with EQ systems (Audyssey XT then XT32) - I finally have a setup where the EQ actually achieves what it promised to achieve all those years ago - quite a dramatic improvement!

If someone wants to donate a current generation Denon for back to back comparison against the Integra, I'm happy to do it - but right now, my vote goes to the Onkyo/Integra/Pioneer AVR's as the best value for money on the market today. (assuming one is seeking good speaker/room EQ)

Power amp issues are not a concern, as I already know that my speakers cause trouble (low impedance/high current) for most AVR amps... and I therefore already have external amps for my L/C/R. - And my amps provide full rated power @1.4V... so any of my shortlist could do the job easily and well (shortlist was Integra 3.4, Onkyo RZ50, Pioneer LX505, Denon X3700).

Chances are prices will rise, and Onkyo / Denon pricing will end up neck&neck over the next few months - had that happened 6 months ago, my decision would have been more difficult.

 

[Dougey_Jones]

You have a point, but about groupthink, I can say the same about Dirac vs Audyssey. If you look at it in more detail, you might consider the reasons why there seem to be much more DL supporters than Audyssey's. To name a few of the possible reasons, DL has a much better user interface, its default target curve is more bass biased vs Audyssey's "flat" one, Audyssey won't EQ the BBC dip unless you disable MRC, so naturally for people who just run the REQ/RC once without doing any tweaking DL has a significant edge, but for those who spent time to get the best of either, the gap may be smaller than you think. Audyssey also is suffering from having a longer history, that is, there are many more people from the earlier versions still remembering, or suffering from the poorer performance of those versions that also ran and many are still running, on less capable processors.

If groupthink wasn't affecting people, the apparent gap between the DL vs Audyssey might be smaller too, just like Denon vs Onkyo in terms of pre out SINAD.

Back to Onkyo, the difference in SINAD for the pre out may be only 2 dB but I would caution that if you look at the actual curve instead of a single point at 2 V, Denon vs Onkyo gap is bigger than 2 dB over the important 50 mV to 1.2 V range.

How big a SINAD difference between 50mv-1.2v? If it’s not >6-8db then I probably still don’t care. You make a good point about peoples bad past experiences with Audyssey. That’s definitely the case with me, I don’t think I’ve ever left it enabled on five separate receivers over the years because it always made things worse.

 

[GalZohar]

As far as I understand, untweaked high frequency correction will normally do more harm than good, and even after tweaking it might not improve the result, and that's rather general and not specific to Audyssey. I guess with Audyssey, since any tweaking requires some annoying tinkering with bad interface and paying extra, most people just leave the default, complain it sucks and live with it or move on. I think having a paid app (be it the 20$ or 200$) and having such a poor interface on the 20$ app was really hurting them even though if you are really into getting a good result this annoyance won't stop you.

 

[peng]

How big a SINAD difference between 50mv-1.2v? If it’s not >6-8db then I probably still don’t care. You make a good point about peoples bad past experiences with Audyssey. That’s definitely the case with me, I don’t think I’ve ever left it enabled on five separate receivers over the years because it always made things worse.

I thought I posted the following comparison before but here they are again to save time searching:

To see the difference below 100 mV we would have to compare the graphs in their corresponding DAC datasheets, so I apologize for using 50 mV, but in Amir's graphs you can see that at 1 V, it looks like 93 dB vs 101 dB and at 600 mV it looks like 88 dB vs 98 dB. So it is within your 6 to 8 dB range. I suppose we can agree that even a 10 dB difference is not likely an audible issue since the Onkyo's 88 dB at 600 mV is still excellent, in my opinion. Would be great to see if at 100 mV or less it would drop below 80 dB because at that point I would begin to be concerned, that's of course just me.

I only mentioned the better pre out SINAD performance of the Denon (since you mentioned it..) to make my case that while I agreed with the "groupthink" effect you cited, that the same mentality may also apply to the DL vs Audyssey claims by forum users who often seemed to have focused too much on that one feature too, to the point that because the AVR has DL (without DLBC), it will "sound better" than Denon, Yamaha's etc., regardless of bench test results.

I am not suggesting that the Onkyo is not an excellent processor for use with external power amps because I consider it is. That being said, I would still prefer Denon's (the ones that have the AK4458 DAC chip) because of the two subouts and SubEQHT feature. When Onkyo includes DL3 DLBC in their AVRs, I would likely be ready to switch.

 

[Dougey_Jones]

I thought I posted the following comparison before but here they are again to save time searching:

To see the difference below 100 mV we would have to compare the graphs in their corresponding DAC datasheets, so I apologize for using 50 mV, but in Amir's graphs you can see that at 1 V, it looks like 93 dB vs 101 dB and at 600 mV it looks like 88 dB vs 98 dB. So it is within your 6 to 8 dB range. I suppose we can agree that even a 10 dB difference is not likely an audible issue since the Onkyo's 88 dB at 600 mV is still excellent, in my opinion. Would be great to see if at 100 mV or less it would drop below 80 dB because at that point I would begin to be concerned, that's of course just me.

I only mentioned the better pre out SINAD performance of the Denon (since you mentioned it..) to make my case that while I agreed with the "groupthink" effect you cited, that the same mentality may also apply to the DL vs Audyssey claims by forum users who often seemed to have focused too much on that one feature too, to the point that because the AVR has DL (without DLBC), it will "sound better" than Denon, Yamaha's etc., regardless of bench test results.

I am not suggesting that the Onkyo is not an excellent processor for use with external power amps because I consider it is. That being said, I would still prefer Denon's (the ones that have the AK4458 DAC chip) because of the two subouts and SubEQHT feature. When Onkyo includes DL3 DLBC in their AVRs, I would likely be ready to switch.

Thanks for clarifying, I appreciate it. Is SubEQ HT roughly equivalent to DLBC?

Another thing that isn’t addressed frequently is the fact that DL with the Onkyo allows you to use a calibrated UMIK-1 where XT32 (afaik?) leaves you stuck with an inferior mic situation. That’s certainly worth something to me.

 

[peng]

Thanks for clarifying, I appreciate it. Is SubEQ HT roughly equivalent to DLBC?

Another thing that isn’t addressed frequently is the fact that DL with the Onkyo allows you to use a calibrated UMIK-1 where XT32 (afaik?) leaves you stuck with an inferior mic situation. That’s certainly worth something to me.

I don't know enough about DLBC yet, but I know SubEQ HT works very well for me, for two subs anything even two very different subs. DLBC should be better if only based on the fact that is is new, whereas SubEQ HT is getting long in the tooth.

Regarding the Audyssey mic, I did read an AVSF post where someone compared several mics and show some significant enough difference, iirc only one particular mic so I am not sure that is typical or an outlier scenario. I have experienced with about at least 2 Denon and 2 Marantz mics so far and found no significant discrepancies or accuracy related issues that my REW (numerous, probably hundreds over the years) plots have identified. I could probably use the Umik-1 mic with REW to verify the results of several Audyssey auto calibrations using the mic positions, with 1/12 or 1/24 smoothing I am confident the results, or variations will show that it would really be an audible issue. I may do it for fun but may have to limit mic positions to 3, otherwise it will be too time consuming.

Overall, in my brief experiments with DL3, I do prefer DL3 but it is difficult to really do a proper comparison because my DL3 is a PC two channel version, whereas my XT32 runs on a 7.1.4. I am still trying to find a way to compare it in the 7.1.4 system by limiting it to two channels, but cannot think of a way to avoid the huge time delay between switching between the two, so any such comparison will be again limited to REW plots only.

 

[Steve Dallas]

I would not say Dirac is necessarily inherently better than XT32 at least in terms of FR. I cannot speak to nebulous things about how either affects stereo separation, etc. YMMV.

EDIT: Some explanation of the chart is in order. XT32 is correcting the subs with its and Denon's bass management. The Dirac signal path does not include subs. XT32 is correcting to 1KHz. Dirac is correcting to 600Hz. Therefore, the area of interest is roughly 70 to 600Hz.

You can decide which is better in this region, but bear in mind that both can be improved with repeated iterations of measurement and tweaking. It appears I need to spend more time with Dirac in this room.

Most of the variation you see above 1KHz is caused by high wind and construction noise at the time the measurements were taken. The noise floor was unusually high on this day.

 

[chych7]

DLBC is probably much better than SubEQ, but you pay a serious $$$ for the upgrade (and not even possible on RZ50 currently). Going the route of miniDSP and Multi-Sub-Optimizer is probably a better and more cost effective solution for either receivers.