Denon AVR-X3800H Review

[peng]

If your auto-EQ can easily handle environments with this many variables—such as vintage speakers or
DIY passive subs—then it truly has the potential to become an excellent system.

Based on my systems in two rooms and one desktop, DLBC is better than Audyssey XT32+App overall, but Audyssey handled dips better most of the time, whereas DLBC seems to need some help from manual PEQ/BIQUADs. I was quite surprised by Audyssey's better performance with the dips, considering the fact that they don't do phase optimizing, only time alignment. I guess algorithm plays a larger part than I think, and could offset the limitations due to lack of phase optimization, partially. ARC Genesis is worse of the 3, its phase optimization looks impressive but not that effective, not for my HT room anyway.

Here's an example, the left channel has a couple of wide dips that no matter how hard I tried, DLBC couldn't do too much about it, much better without it but not good enough, with BIQUADs/minidsp, it is good enough to me. I would of course look better if don't put a tilt to the bass.

The right channel doesn't have the issue. Some rooms are just more challenging than others.

Left channel, Philharmonic BMR bookshelf DIY speakers, and two small subs, no smoothing:

 

[kawauso]

Based on my systems in two rooms and one desktop, DLBC is better than Audyssey XT32+App overall, but Audyssey handled dips better most of the time, whereas DLBC seems to need some help from manual PEQ/BIQUADs. I was quite surprised by Audyssey's better performance with the dips, considering the fact that they don't do phase optimizing, only time alignment. I guess algorithm plays a larger part than I think, and could offset the limitations due to lack of phase optimization, partially. ARC Genesis is worse of the 3, its phase optimization looks impressive but not that effective, not for my HT room anyway.

Here's an example, the left channel has a couple of wide dips that no matter how hard I tried, DLBC couldn't do too much about it, much better without it but not good enough, with BIQUADs/minidsp, it is good enough to me. I would of course look better if don't put a tilt to the bass.

The right channel doesn't have the issue. Some rooms are just more challenging than others.

Left channel, Philharmonic BMR bookshelf DIY speakers, and two small subs, no smoothing:

View attachment 493948

DLBC’s dips may appear more pronounced when you only measure at the MLP, but that could simply be the result of prioritizing seat-to-seat consistency and other parameters. If it sounds good, there’s no need to worry about it.

Earlier, I tried setting up XT32 using the Audyssey mobile app. Compared to DLBC, XT32 couldn’t boost the passive subwoofer as effectively, but the results were still reasonably good. With MultEQ-X, it should be possible to flatten the subwoofer response even without an external DSP.

That said… if you can use Dirac, Dirac is still the best option.

 

[peng]

DLBC’s dips may appear more pronounced when you only measure at the MLP, but that could simply be the result of prioritizing seat-to-seat consistency and other parameters. If it sounds good, there’s no need to worry about it.

I tried many positions, 5, 6, 7, 9, same thing.

Earlier, I tried setting up XT32 using the Audyssey mobile app. Compared to DLBC, XT32 couldn’t boost the passive subwoofer as effectively, but the results were still reasonably good. With MultEQ-X, it should be possible to flatten the subwoofer response even without an external DSP.

The way I did it with the MultEQ App to cure the nasty dips, I would lower the level by 2,3, even 5 dB across the board first, so I wouldn't have to rely on boost that much. Dirac tends to do that automatically, so I think Audyssey might have something in their Algo that just worked a little better with certain dips.

I do think Dirac is the best of the 3, or 4 if I count REW/minidsp, that's why I sold the Denon used in my 2ch system in the living room. If I eventually grab a Denon, then I would most likely use it in the HT room and move the Anthem to the living room and use it with Dirac/minidsp (then I would likely be the only one in the world who use use a 15 channel AVM that way but I would have no shame in doing it that way.

Managed to find one of the graphs I used to compare the 3 systems, iirc, it is left right and 1 sub (when DLBC was used, because it would have been difficult if possible at all to use two subs when I only had the DLBC license for the PC standalone version) combined, with 1/12 smoothing. So, in that comparison it wasn't a totally fair comparison and if two subs are used DLBC should have done better.

The black on is Audyssey with MultEQ editor app and Radbuddyssey. The sad thing about ARC G is that I couldn't even reduce the bump at about 52-60 Hz lower than it was at the time, dips weren't too bad but that was only achieved after hours of trial and error manual adjustments post calibration.

FR obviously is just one thing, where Dirac wins big is impulse response, and now of course with ART, FR should do a little better and decay is supposedly much better based on yours and others posted graphs. On that note, Anthem ARC G is probably the best if we rely on AVSF posts that are based on subjective views.

 

[OCA]

It’s just an ordinary, slightly small Japanese-style bedroom, and the listening distance is about 1.5 meters.
Here are the Dirac MDAT files measured at the MLP and at an off-MLP position (with no PEQ applied).

If your auto-EQ can easily handle environments with this many variables—such as vintage speakers or
DIY passive subs—then it truly has the potential to become an excellent system.

By the way, I’ve tested many speakers in this bedroom — Sierra-LX, UBR62, HDI-1600, and LS50 Meta.
Among them, the 4425 with PEQ applied is the one I like the most.

In my JBL room downstairs, I have four more 4425s, and they show very similar characteristics.
There, I insert the PEQ between the AVR pre-DAC and the power amplifier.

Bedroom.mdat

drive.google.com

These are intriguing speakers. In a small room, the bass should build up significantly, unlike the high frequencies, which typically diminish due to out-of-phase reflections bouncing from every direction. I don't have time to review any more data today, totally swamped trying to wrap up the release on schedule. I sort of sabotaged myself by committing to launch it tomorrow evening.

 

[peng]

These are intriguing speakers. In a small room, the bass should build up significantly, unlike the high frequencies, which typically diminish due to out-of-phase reflections bouncing from every direction. I don't have time to review any more data today, totally swamped trying to wrap up the release on schedule. I sort of sabotaged myself by committing to launch it tomorrow evening.

OCA, its Saturday too, how about stop reading posts, grab another coffee or Vodka and focus on your A1 launch lol. I think kawauso is still having fun with ART anyway..

 

[kawauso]

I tried many positions, 5, 6, 7, 9, same thing.

The way I did it with the MultEQ App to cure the nasty dips, I would lower the level by 2,3, even 5 dB across the board first, so I wouldn't have to rely on boost that much. Dirac tends to do that automatically, so I think Audyssey might have something in their Algo that just worked a little better with certain dips.

I do think Dirac is the best of the 3, or 4 if I count REW/minidsp, that's why I sold the Denon used in my 2ch system in the living room. If I eventually grab a Denon, then I would most likely use it in the HT room and move the Anthem to the living room and use it with Dirac/minidsp (then I would likely be the only one in the world who use use a 15 channel AVM that way but I would have no shame in doing it that way.

Managed to find one of the graphs I used to compare the 3 systems, iirc, it is left right and 1 sub (when DLBC was used, because it would have been difficult if possible at all to use two subs when I only had the DLBC license for the PC standalone version) combined, with 1/12 smoothing. So, in that comparison it wasn't a totally fair comparison and if two subs are used DLBC should have done better.

The black on is Audyssey with MultEQ editor app and Radbuddyssey. The sad thing about ARC G is that I couldn't even reduce the bump at about 52-60 Hz lower than it was at the time, dips weren't too bad but that was only achieved after hours of trial and error manual adjustments post calibration.

FR obviously is just one thing, where Dirac wins big is impulse response, and now of course with ART, FR should do a little better and decay is supposedly much better based on yours and others posted graphs. On that note, Anthem ARC G is probably the best if we rely on AVSF posts that are based on subjective views.

View attachment 493976

ARC Genesis looks underwhelming to me based on this graph. There are still many sharp peaks left over, and honestly, my manual PEQ seems to work better.
(In my KEF system, when aiming for maximum headroom, I sometimes turn Dirac off and manually tame only the subwoofer with gentle-Q filters.)

OCA, its Saturday too, how about stop reading posts, grab another coffee or Vodka and focus on your A1 launch lol. I think kawauso is still having fun with ART anyway..

I’m literally enjoying ART in my 4425 bedroom system right now — currently listening to some classical music.

 

[OCA]

OCA, its Saturday too, how about stop reading posts, grab another coffee or Vodka and focus on your A1 launch lol. I think kawauso is still having fun with ART anyway..

I need a pause and a quick distraction from time to time. I am coding at the uncharted limits of my iq level these days and concentration fades regularly. It's tough to be old.

 

[peng]

ARC Genesis looks underwhelming to me based on this graph. There are still many sharp peaks left over, and honestly, my manual PEQ seems to work better.
(In my KEF system, when aiming for maximum headroom, I sometimes turn Dirac off and manually tame only the subwoofer with gentle-Q filters.)

Fortunately, it does sound reasonably good, at least for movies, the nature of movie tracks is such that it's not that easy to detect a few sharp bumps and dips as shown on the REW/FR graphs, so to me the main goal is to try and supress those as much as possible. It is really a pity that ARC G put so much restrictions on users who are experience and like to do manual adjustments. That wouldn't not likely change any time soon because users don't complain, for whatever reasons (I think I know the reasons but...)

This one is more recent, look a little better (only when using the psychoacoustic smoothing magic) but those bumps are very stubborn, so still there.

I’m literally enjoying ART in my 4425 bedroom system right now — currently listening to some classical music.

Given the well behaved FR and decays, I can imagine that.

 

[kawauso]

Fortunately, it does sound reasonably good, at least for movies, the nature of movie tracks is such that it's not that easy to detect a few sharp bumps and dips as shown on the REW/FR graphs, so to me the main goal is to try and supress those as much as possible. It is really a pity that ARC G put so much restrictions on users who are experience and like to do manual adjustments. That wouldn't not likely change any time soon because users don't complain, for whatever reasons (I think I know the reasons but...)

This one is more recent, look a little better (only when using the psychoacoustic smoothing magic) but those bumps are very stubborn, so still there.

View attachment 494164




Given the well behaved FR and decays, I can imagine that.

The 4425 has a somewhat distinctive on-axis frequency response, but once I apply a PEQ curve tuned by ear, it sounds remarkably good.
Originally, I was looking for surround speakers for my JBL system, but I wasn’t very fond of the 4338. While searching online, I found that the 4425 had excellent DI data, so I decided to try them. I ended up liking them a lot.
After that, I purchased two restored pairs of 4425, and moved the first pair into the bedroom.
As you can see, the tonal balance doesn’t change much vertically or horizontally, which makes them perfect for bedroom use.

 

[kawauso]

The "fan" epithet is , correct. I have both Audyssey paid Apps: the Smart device IOS App and the Windows-PC App, MultEQ-X. I also used MSO and after literally years of fiddling with these, REW and U-Mik-1, I obtained good results. I have been satisfied and posted about that.
Then, I tried @OCA Evo Express...
In less than 30 minutes, to my ears, (You cannot be more subjective than that), and using only the microphone that came with the AVR... the results were better than what I had with MultEQ-X + MSP + miniDSP 2x4 HD + UMik +hours of measurements.
It took me a bit more, about 2 hours, to reach better subjective results with Acoustica, I prefer the level of control of Acoustica.
And these scripts are free...

I have not experienced a Dirac application. I know these are not inexpensive, in general far more expensive than Audyssey's. I am yet to see a solid report that proves , not opine, that DIrac is superior to Audyssey MultEQ-X. It is usually a toss-up... In the meantime, the best I have gotten from my, albeit, modest system is from @OCA free Scripts. I am thus looking for his Dirac-ART- competitor scripts.
An aside: As an IT (Information Technology) professional and a long time audiophile, I can marvel and wonder about the level of knoweldge, In acoustics, psycho acoustics, Internal working of the some AVR, mastery of TCP/IP , Windows Shell commands and working, as well as the dedication and polish of @OCA scripts to achieve such consistently and increasingly good ( if I am to follow his scripts progression through the years). Let's not forget his Generosity. Those are free., as in: "you just download them" Nothing else required, except a PC which you can just ditch immediately after having run the scripts.

The difference to my ears are profound and unmistakable.

I plan one day to measure the before and after to "see" what is going on. I am currently too happy enjoying my system, to bother. The intellectual part of me, wants, needs some objective proofs.. but the music lover and audiophile are busy , just enjoying music and great sound... so these will come in a time of boredom or satiety.

to @OCA: I for one , am waiting for that Dirac competitor

@kawauso ... Once you have the Marantz, give Evo Express a try and report, It may not take more than 30 minutes, even with 2 subwoofers... We'll accepts your subjective, evaluations

A1 Evo Express didn’t work well at all in my bedroom.
Even OCA said my environment is too unusual and basically gave up.
But Dirac has no issues in this room, and the XT32 app version works fine.
MultEQ-X would probably work even better.

If it’s working well in your environment, could you share your MLP measurements for L, R, and L+R (mdat)? I’d like to see them for reference.

Denon AVR-X3800H Review

Denon website price is $1,459, not too bad. https://www.denon.com/en-us/category/av-receivers/ That would have been a good price for the 4800 but very mediocre for the 3800.

www.audiosciencereview.com

 

[tsammyc]

I bought the whole Dirac package including ART on the Black Friday sale and spent a few hours setting it up. The system is 5.4.2 with 20 yr old Wharfedale Diamonds. The two subs are a 20 yr old nearfield SVS PB-12/2 and a SVS 3000 Micro. Before Dirac, I alternated between the PB 12-2 for HT and the 3000 Micro for music. This time I ran them combined in a sub group with DLBC.

This was my first experience with Dirac, having used Audyssey and the $20 app previously. Dirac was buggy. It wouldn't find my X3800H on the network of three mesh routers unless the Mac was connected to the same mesh node as the X3800H. Even after that, it couldn't find the X3800H on the network unless I put in both IP address and a port number.

While running Dirac, I got occasional messages regarding a disconnect from the licensing server. After measurements, I used the ART's defaults and it resulted in a rubbish curve on REW that made no sense. Turned everything off and on again, rebooted the Mac and ran DBLC (no ART) and it produced a pretty flat curve. Re-ran ART after that and got a good curve but no bass. I moved the bass group down from 20Hz of 12Hz and got the frequency response chart below which looked acceptable. The previous Audyssey curve is in the chart for reference.

Movies actually sound great. Previously Audyssey would annoyingly rattle my roof tiles, windows and other things in the room. Everything now sounds tight and more realistic. Happy camper as it's the best the room has sounded in the last 20+ years. Was doing room correction with test tones, a Behringer Bass Feedback Destroyer and a Radioshack SPL meter in the early 2000s. This is much better.

 

[amper42]

Rattling Roof Tiles? Oh my... you must have found just the right frequency.
Maybe replace paper with Boral TileSeal HT underlayment next time?

 

[kawauso]

I bought the whole Dirac package including ART on the Black Friday sale and spent a few hours setting it up. The system is 5.4.2 with 20 yr old Wharfedale Diamonds. The two subs are a 20 yr old nearfield SVS PB-12/2 and a SVS 3000 Micro. Before Dirac, I alternated between the PB 12-2 for HT and the 3000 Micro for music. This time I ran them combined in a sub group with DLBC.

This was my first experience with Dirac, having used Audyssey and the $20 app previously. Dirac was buggy. It wouldn't find my X3800H on the network of three mesh routers unless the Mac was connected to the same mesh node as the X3800H. Even after that, it couldn't find the X3800H on the network unless I put in both IP address and a port number.

While running Dirac, I got occasional messages regarding a disconnect from the licensing server. After measurements, I used the ART's defaults and it resulted in a rubbish curve on REW that made no sense. Turned everything off and on again, rebooted the Mac and ran DBLC (no ART) and it produced a pretty flat curve. Re-ran ART after that and got a good curve but no bass. I moved the bass group down from 20Hz of 12Hz and got the frequency response chart below which looked acceptable. The previous Audyssey curve is in the chart for reference.

View attachment 494440
Movies actually sound great. Previously Audyssey would annoyingly rattle my roof tiles, windows and other things in the room. Everything now sounds tight and more realistic. Happy camper as it's the best the room has sounded in the last 20+ years. Was doing room correction with test tones, a Behringer Bass Feedback Destroyer and a Radioshack SPL meter in the early 2000s. This is much better.

There is no option in ART that allows you to set the range down to 12 Hz, so you may be confusing it with a DLBC setting.
When using both the SVS PB-12/2 and the SVS 3000 Micro, it might work better to split the subwoofer groups by dragging them apart.
Try setting the support range of the SVS 3000 Micro to 50–150 Hz, and the PB-12/2 to 20–150 Hz.
Also, enable the Infra Bass option by checking the box in the lower-left corner.
Do this for all groups.

 

[tsammyc]

There is no option in ART that allows you to set the range down to 12 Hz, so you may be confusing it with a DLBC setting.
When using both the SVS PB-12/2 and the SVS 3000 Micro, it might work better to split the subwoofer groups by dragging them apart.
Try setting the support range of the SVS 3000 Micro to 50–150 Hz, and the PB-12/2 to 20–150 Hz.
Also, enable the Infra Bass option by checking the box in the lower-left corner.
Do this for all groups.

It was the support range that I altered to 12-150Hz from 20-150Hz that seemed to restore the low and infra bass. I did not enable the infra bass option. If I split the subs up into individual groups how do I decide the range? You suggested 50-150Hz for the SVS 3000 Micro. It's rated by SVS to 23Hz and I have verified fairly high SPLs on a 25Hz test tone. At 30Hz, it is very loud. Should I still set to 50-150Hz? The PB Plus 12/2 is in a 16Hz tune and has strong output on a 14Hz test tone. Should I still set it to 20-150Hz?

This graph shows the difference between setting the sub group (two subs) to 12-150 (blue line) vs setting it 20-150 (red line). With the sub group set to 12-150 I'm still getting a rolloff below 20Hz, but it's not as severe. I have tactile transducers that go down to 8Hz, so infra bass isn't that critical to me.

 

[kawauso]

It was the support range that I altered to 12-150Hz from 20-150Hz that seemed to restore the low and infra bass. I did not enable the infra bass option. If I split the subs up into individual groups how do I decide the range? You suggested 50-150Hz for the SVS 3000 Micro. It's rated by SVS to 23Hz and I have verified fairly high SPLs on a 25Hz test tone. At 30Hz, it is very loud. Should I still set to 50-150Hz? The PB Plus 12/2 is in a 16Hz tune and has strong output on a 14Hz test tone. Should I still set it to 20-150Hz?

This graph shows the difference between setting the sub group (two subs) to 12-150 (blue line) vs setting it 20-150 (red line). With the sub group set to 12-150 I'm still getting a rolloff below 20Hz, but it's not as severe. I have tactile transducers that go down to 8Hz, so infra bass isn't that critical to me.
View attachment 494747

This behavior is typical when the Infra Bass option is not enabled.
Please set the SVS 3000 Micro to 50–150 Hz.
40–150 Hz also works, but there will be almost no difference.

Fine-tuning the support range should be done later while listening to the actual sound.
Also, the subwoofer channel is essentially LFE-only, so in order for the subwoofer settings to take effect, you must apply the same settings to all other groups as well.
In ART, there are no options such as “12–150 Hz” besides simply enabling Infra Bass.
If you’re unsure, please show me a screenshot of Group 1 (LR).

 

[Ciobi69]

I got some trouble setting Dirac on the denon , I am using the pc app, shortly after when I continue it opens this on my denon and they want to use the app in my phone , but I got the umik connected to the pc and basically I can’t do the calibration , I don’t want to do it on my phone Dirac app , does anyone know the problem

 

[kawauso]

I got some trouble setting Dirac on the denon , I am using the pc app, shortly after when I continue it opens this on my denon and they want to use the app in my phone , but I got the umik connected to the pc and basically I can’t do the calibration , I don’t want to do it on my phone Dirac app , does anyone know the problem

I’m not really sure about that issue.
I’ve never used the Dirac app for smartphones.
If you’d like, you could ask in the Dirac ART thread — someone there might know what’s going on.

 

[Denon545]

I got some trouble setting Dirac on the denon , I am using the pc app, shortly after when I continue it opens this on my denon and they want to use the app in my phone , but I got the umik connected to the pc and basically I can’t do the calibration , I don’t want to do it on my phone Dirac app , does anyone know the problem

Are you running the Dirac software on your PC? If so, make sure it connects to your Denon. If not, you need to download it to your PC and make sure your license is activated. I would turn off WiFi and Bluetooth on your phone so that it isn’t trying to connect as well.

 

[Ciobi69]

Are you running the Dirac software on your PC? If so, make sure it connects to your Denon. If not, you need to download it to your PC and make sure your license is activated. I would turn off WiFi and Bluetooth on your phone so that it isn’t trying to connect as well.

Ok I will try that, yes the software it’s on the pc , it connect to the denon , but when I have to measure in the mic and speaker test section it goes on the denon Dirac app on the avr saying I got to open on my phone , I will try to disconnect the WiFi on my phone too tomorrow thanks! Will keep updated

 

[kawauso]

Ok I will try that, yes the software it’s on the pc , it connect to the denon , but when I have to measure in the mic and speaker test section it goes on the denon Dirac app on the avr saying I got to open on my phone , I will try to disconnect the WiFi on my phone too tomorrow thanks! Will keep updated

Is your setup connected like PC → AVR → display?
During measurements, please connect your PC directly to the display without passing through the AVR.