Denon AVR-X3800H Review

[oivavoi]

I just returned mine. Too many comments trying to convince SINAD is inaudible, but Denon has definitely done something or tweaked the unit’s curves without RC. Even PureDirect mode sounds different than my AKM 3700, which it should not. I just keep the 3700 for another year or so until they get their act together.

Did you do any blind comparisons, or do any kind of measurements? If puredirect indeed sounds audibly different in a way you could detect in a blind test it would indicate serious faults with your unit.

 

[anphex]

Wow, that was fast. Thought this AVR just came out a few days ago lol.
It's unfortunate that the performance got worse with the new generation. Wasn't Denon recently proactive on ASR and sending in gear by themselfes?

Then again, from what I read that plenty of people defend this backstep of an AVR just because it has Dirac. What good is Dirac when it's no audible improvement over XT32 and the sound purity is even worse than the 3700?

 

[Doodski]

The difference being - What Hi-Fi? based their findings on what you experience with real world usage - actually allowing the receiver to be used the way everyone who owns will be using it. Measurements are never any guarantee of what to expect from a perofrmance point of view. How are senses enjoy what a piece of home audio offers has to offer actually means something to the consumer. Lines on a graph are for would-be engineers, engineers and people with a science delusion.
It would take a complete fool to believe they senses are always going to agree with 'scope measurements.
Here's another thing to consider with measruements. In the late 1970's, Hitachi manufactured an integrated amplifier which measured 'perfectly'. But it was a flop because it apparently sounded very boring.

Pure gobbledygook from a person that apparently does not have a understanding of electronics. The circuitry is reacting to each electron passing through it and the designers are basically counting electrons when designing due to the ability to easily convert current and voltage to electron flow ~quantity. Electronics theory(s) has been around for a long time and it has been proven time and time again to be accurate. So when a layman comes along with big ideas about electronics it's amusing if not ridiculous.

 

[dougi]

Wow, that was fast. Thought this AVR just came out a few days ago lol.
It's unfortunate that the performance got worse with the new generation. Wasn't Denon recently proactive on ASR and sending in gear by themselfes?

Then again, from what I read that plenty of people defend this backstep of an AVR just because it has Dirac. What good is Dirac when it's no audible improvement over XT32 and the sound purity is even worse than the 3700?

Well, it doesn't have DIRAC yet. TBH, shouldn't DIRAC sound better with its full-range phase correction? Which I would have thought could be a real benefit for imaging with music in an AVR with imperfectly matched LCRs, for instance. For me, it certainly helps for 2.0 compared with the other room correction I have experience with (Room Perfect).

 

[NirreFirre]

I bet it is, take a look of their SINAD spec comparison:

AK4458 - 107 dB
PCM5102A - 93 dB


Amir, thank you very much for the timely review. Being that this is the first 2022 Denon that has the downgraded DAC chip measured and by you only, if you still have the unit can you please measure the preout/dac for up to 4 V (at least 3 V anyway) in both preamp and non preamp mode?

It will be very useful for those who plan on using it with amps with lower gain such as 25 to 26 dB Hypex based amps. Such amps will need at least 2.5 V to output > 200 W into 8 ohms. Thank you.

Agreed, it would be nice to have a set of measurements done for each category of equipment but I guess there's reasons Amir isn't strictly follow some checklist. Rebuilding, kids, life etc

As I have similar class D amps in the front and really need 1,5V+ the Onkyo RZ50 seems like a good option today with ~10dB better DAC-preamp section as well, Dirac included (I have a miniDSP to align my subs so can do without the DLBC), now and not as an unknown priced add on, hopefully next year, fully working HDMI2.1, some other niceities and imho better looking No DLBC, worse remote and some Eco mode bug limiting power with <~8ohm loads. Perhaps the 2023 RZ70/90 could have DLBC but who knows at these times.

 

[Peluvius]

I don't want to buy a new model that performs more poorly than the old model and I appreciate having an independent review to show this (thanks Amir).

I was planning to buy this for Dirac but will implement using MiniDSP now and keep my existing AVR.

I don't quite follow the "no audible difference" nonsense, why support a company that takes you for a fool?

 

[pseudoid]

Which one?

This Denon is the proverbial 'kitchen sink'; although seems to ONLY lack a USB audio interface!
Do you really need (or can use) 66 HDMI ins, 33 HDMI outs, 111 speaker outs, 99 Pre outs, 66 analog ins, 44 digital ins and 44 antenna ins?
The last 'kitchen sink' we purchased got warranty-replaced in the eleventh month for cracks but it also lacked a USB interface... yet will never require a software update!

 

[NirreFirre]

Some of these argument seems like "How many angels on a head of a pin"... I'd like to think that I can incrementally improve my audio..
So here is my deal, I'll be getting a X3800H tomorrow, and I've on order an Buckeye Purifi 3-channel amp.
So with even with the less than perfect measurements of the X3800H , it should still be head and shoulders better than what I have, a Onkyo NR656?
I fully admit I'm a dilettante audiophile, but am I making a mistake getting a X3800H?

​

As others have said, your setup will be unbalanced in that the amps performance can't be utilized by any AVR but the X3800H seems to loose on more than ~10dB of SINAD. It depends on more parameters for sure but getting a more expensive unit than say, a Onkyo RZ50 or Integra variant, that lacks DL(BC) today (and will cost more as an add on), proven HDMI2.1 features etc seems a bit odd.. but 4 ind sub outs could be something you value so may not be an obvious choice. Outside the TX-NR7100 has pre outs for the front channels and units have begun (Finland) or will begin selling in November (Sweden) so that seems you're getting a lot for reasonable money (the X3700H

 

[Triliza]

If you are using parameters that have no audible effect to market your product as more high end because of it, what is the difference between that and the BS they try to sell cables with?

Is SINAD going to become the new Skin Effect?

As already said, better SINAD on a product shows that the designer put some real effort to optimize the build, took their time with the layout, placements of components, power supply, cables etc. Audible or not, who would you prefer, or even support for their efforts? And not to forget that this is a forum for people that like innovations and excellency on audio gears, why should we be content with such performance. In any other hobby's forums (cars, motorbikes, projectors etc), you would get the same reaction, nobody would be happy if a new car model had worse performance and was more expensive that the previous model, even if you'd never tell the difference in practice.

If I have a choice, I'd never support a product with less than 96 SINAD, I don't care if it's audible or not, should be doable even for AVR I'd like to think.

 

[Roland68]

Coincidentally, the AVR market will be 2.8 billion USD in 2028. I doubt the market for DAC’s will be that big.

Currently, sales for AV receivers are said to be around $2.2 billion.

What they don't realize is that this statistic also includes all cheap AV complete systems/home entertainment with amplifiers from LG, Harman Kardon, Sony, Yamaha, Inkel Corporation, Pyle etc., which are sold by the millions.
Take a look at electronics discounters, there are 10-20 of these systems on a "real" AV receiver.
In addition, there are all the very cheap and also pseudo AV receivers from Auna, Pyle, Inkel Corporation and many others, which are sold in some countries for as little as $100 and in unbelievable quantities.

If the type of AV receivers we are talking about here on the forum and in this thread account for 25% of the above sales, then that should be a lot.

 

[radix]

The 3800 seems to have quite a few more procesing options (imax, dts hd, 360, etc.). I wonder if the updated processing SW might be accounting for the change in sound from the 3700.

 

[TonyJZX]

dirac relies on the silicon being software programmeable for it right?

I cannot imagine dirac relying on some fixed silicon function.... put another way, it's like saying you need AMD processors to run MS Office 365... it would sound funny to me to think dirac implemenation results in lower overall system performance because its inherent in their software?

 

[Dj7675]

The 3800 seems to have quite a few more procesing options (imax, dts hd, 360, etc.). I wonder if the updated processing SW might be accounting for the change in sound from the 3700.

In order to make a conclusion that without any EQ engaged there is a change in sound between the 3700/3800 would require some measurements or blind testing level matched. There would need to be some basis to state this as fact. I haven’t seen any yet. And would be surprised if they actually sound different under most circumstances. If the basis is subjective reviewers or having a single user say the highs are exaggerated or it is lacking bass, it doesn’t hold much validity IMO.

 

[beagleman]

I see something, that to me is a bit more than annoying.

As soon as something is tested and has less than very good measurements, many all just pile on, having NO intention of buying said item, nor having ever heard the unit.
In fact I did not see Amir even mention having played music or sound on this receiver.

This objectivism thing has gone out of control.
I feel that many in here have literally no idea if something will sound good, great or mediocre, but merely parrot what Amir posts and what his measurements reveal.

 

[Karls_speaker_works]

I see something, that to me is a bit more than annoying.

As soon as something is tested and has less than very good measurements, many all just pile on, having NO intention of buying said item, nor having ever heard the unit.
In fact I did not see Amir even mention having played music or sound on this receiver.

This objectivism thing has gone out of control.
I feel that many in here have literally no idea if something will sound good, great or mediocre, but merely parrot what Amir posts and what his measurements reveal.

I was just looking at the distortion measurements of my speakers at 93dB SPL. They range from 0.5% to 2% in the critical frequencies for hearing (peaks at over 100% distortion at low frequencies - lol.) Even in the critical midrange frequencies, this is 100 to 400 times as distorted as the DAC in the X3800H. They say you can't compare these measurements directly, but I'd like to know how I would "hear" something 200x smaller than the typical distortion produced by a loudspeaker? I just have to shrug.

 

[Peluvius]

I see something, that to me is a bit more than annoying.

As soon as something is tested and has less than very good measurements, many all just pile on, having NO intention of buying said item, nor having ever heard the unit.
In fact I did not see Amir even mention having played music or sound on this receiver.

This objectivism thing has gone out of control.
I feel that many in here have literally no idea if something will sound good, great or mediocre, but merely parrot what Amir posts and what his measurements reveal.

I can understand how you would feel that way. My objection here is to the fact that a company which promotes itself as representing technical excellence has engineered a downgrade into a next gen product. They did this knowingly and quite deliberately. Whether you take objective measurements into account when selecting gear is entirely up to you, my own view is that it forms a relevant part of the buying/assessment process.

Perhaps if they promoted the fact that there was a technical downgrade to this model and no audible variance from the previous model I would feel a bit less passionate about it. I think a reasonable person would expect a new model to be an improvement over the last and it isn't and I am disappointed by this.

Electronics already play such a small part in the overall audio reproduction value chain (Room correction aside) the least manufacturers can do is improve or at least maintain technical standards.

I had a pre-order in for a 3800 which I have cancelled.

As to whether or not this is a "pile on", well we would agree on that, you are in a forum which focusses on engineering and technical aspects of reproduction equipment....

 

[Karls_speaker_works]

I had a pre-order in for a 3800 which I have cancelled.

I'll take if off your hands. Lol. I can't believe that people are doing so over a parameter that they would never know was different if they hadn't seen the measurement - it would be funny if it wasn't so painful to my engineering soul.

 

[beagleman]

I can understand how you would feel that way. My objection here is to the fact that a company which promotes itself as representing technical excellence has engineered a downgrade into a next gen product. They did this knowingly and quite deliberately. Whether you take objective measurements into account when selecting gear is entirely up to you, my own view is that it forms a relevant part of the buying/assessment process.

Perhaps if they promoted the fact that there was a technical downgrade to this model and no audible variance from the previous model I would feel a bit less passionate about it. I think a reasonable person would expect a new model to be an improvement over the last and it isn't and I am disappointed by this.

Electronics already play such a small part in the overall audio reproduction value chain (Room correction aside) the least manufacturers can do is improve or at least maintain technical standards.

I had a pre-order in for a 3800 which I have cancelled.

As to whether or not this is a "pile on", well we would agree on that, you are in a forum which focusses on engineering and technical aspects of reproduction equipment....

But at the same time, I think some have lost sight of the realities of what is audible or not.

I totally get the measurement thing. I am into it mostly with speakers.

Having relatively recently gotten a few bargain priced older (2015 about) AVRs, that would be bashed here for sure, I found miraculously they sound just fine and completely do nothing at all to offend me in ANY way!

So I sense a huge disconnect between the bashing and what is actually audible.

 

[GaryH]

I was just looking at the distortion measurements of my speakers at 93dB SPL. They range from 0.5% to 2% in the critical frequencies for hearing (peaks at over 100% distortion at low frequencies - lol.) Even in the critical midrange frequencies, this is 100 to 400 times as distorted as the DAC in the X3800H. They say you can't compare these measurements directly, but I'd like to know how I would "hear" something 200x smaller than the typical distortion produced by a loudspeaker? I just have to shrug.

Exactly. In practical circumstances -87 dB DAC/amp THD is provably inaudible, with just a bit of math. In order to be audible and result in the JND of 0.1 dB (the minimum 'just noticeable difference' the human ear can detect) over the loudspeaker's distortion, this -87 dB THD would have to be combined with a speaker THD less than -70 dB (worst-case, in-phase, combining to -69.9 dB, see the calculator at the bottom here). That's 0.03%. I don't think any speakers consumers will use will have THD this low. And of course this is not even considering absolute distortion hearing thresholds and the psychoacoustics of masking which will make the actual audible transparency threshold even more lenient. Insisting that vanishingly low audio electronics THD is needed for transparency is not only absurd, it's demonstrably false in practice, because the transducer is invariably by far the weakest link distortion-wise in the audio reproduction chain.

 

[Karls_speaker_works]

But at the same time, I think some have lost sight of the realities of what is audible or not.

I totally get the measurement thing. I am into it mostly with speakers.

Having relatively recently gotten a few bargain priced older (2015 about) AVRs, that would be bashed here for sure, I found miraculously they sound just fine and completely do nothing at all to offend me in ANY way!

So I sense a huge disconnect between the bashing and what is actually audible.

Earlier in the thread another was clinging to their 52 dB SINAD AVR rather than be thoroughly INSULTED that Denon lowered their performance to only 15x better than their current unit - which DOES have distortion that should be detectable (though probably isn't since they seem happy with it.) Heck, I am perfectly happy with my 2010 model Pioneer VSX-820 AVR. If it had more than 5 amplifiers and did the other 20 things that the 3800H does, I wouldn't care to replace it. This entire thread is so painful for me to read. The information is valuable, but many just aren't sophisticated enough to interpret it. I may be a snob whereas I'm an electrical design engineer, but on the other hand I would hope that an engineers opinion would carry a tiny bit of weight? Not on the internet.