Denon AVR-X3800H Review

[Narnian]

I have a S760H and was looking at this to upgrade but I think I'll wait for next year's model or the 2024 one. Seems smarter to just put more money into improving other components first.

I had a S760H and returned it for the X3800H before I saw the test results. However I have 60 days to try it (thanks Crutchfield!). The other was a little underpowered it seemed and the X3800H is perfect in that respect. It does sound fuller and a bit clearer to me.

Contact me in a month and I will let you know my thoughts. So far I am pleased with a number of the other features but am sad at the test results here. Unfortunately there was not another one available in my price range that had the feature set I wanted this time. They were out of the 3700H model.

If you see my other posts it is likely I will not be able to differentiate due to some hearing loss.

If I can get a 3700H or an open box 4700H I may return it as a result of Amir’s testing.

 

[peng]

Thanks for such a detailed review. My high hopes are crushed.

I am not that much of a technical person.

I have Rotel RB1590 power amplifier which has input sensitivity of 2.2 vrms. Looking at the measurements seems like it won’t be sufficient to drive my rotel. Is my understanding right?

No need to have your hopes crushed if your concern is whether the pre out voltage is high enough for the RB1590. Those Denon AVRs pre out can typically reach 4 V or higher before clipping. Your RB1590's distortion specs is 0.03% but they did not specify the load conditions. Assuming it is for rated output, that is, 350W then the 3800 is a decent match. Amir did not measure it beyond 2 V but you can see that at 2.2 V it should have no trouble staying above 80 dB SINAD, and that would be 0.01%.

Besides, the Rotel only needs 2.2 V for rated power output. Depending on your listening habit, distance and speakers you may need to push it to its limit anyway. Have you tried using an online calculator to estimate your actual power need, if not try it and you may be nicely surprised.

Peak SPL Calculator

 

[peng]

Were those alternate DAC chips available in the required volume in early in 2021 when they were forced to redesign the X3700H/X4700H/X6700H? It's a rhetorical question since I don't believe anybody knows, but there's an implication of malice in many of these "redesign" posts - it may not be a fair assessment.

Yamaha has been using the ES9006 and ES9026Pro and I have not read anything about them having to switch, so I guess at least one of those I listed should have enough quantity to meet D+M's requirements.

Here's just one vendor, for the ES9026Pro and the cheaper ES9010K2M used in D+M and Anthem's AVRs/AVPs:

https://www.mouser.com/ProductDetail/ESS-Technology/ES9026PRO?qs=yqaQSyyJnNhU2nm3v27xMw%3D%3D

https://www.mouser.com/ProductDetail/ESS-Technology/ES9010K2M?qs=sPbYRqrBIVkOHauV5xTLbw%3D%3D

 

[Bren Derlin]

Help me out on this. People keep mentioning the cost increase of the x3800H (some suggesting a 30% increase) over the x3700H.

Prices for both new upon release.
x3700H - $1,599 (you can still order this receiver new for $1,599 at Target)
x3800H - $1,699

I can somewhat understand the dismay over the $100 price increase if the audible performances are the same vs not being improved with the 3800H over the 3700H. But considering there are 4 subwoofer outs, that's a significant bump over the previous generation of 2. Dirac compatibility when it becomes available (2023?) - at an added cost. But it's there if you want it. 8K compatibility available on all HDMI inputs, instead of being limited to 1 or 2. ...if 8K is your thing, this could be a big deal. I have zero idea when 8K becomes mainstream. Those bitty bits of improvement could make the x3800H worth a $100 price bump.

If you're comparing the x3700H refurb at $1,200 vs. the $1,700 price for a new x3800H - this is ridiculous, as the x3800H doesn't have refurbs on the market as of yet. How about we wait until the x3800H refurbs hit the shelves at accessories4Less before complaining about this price differences. Jeez...

 

[Bren Derlin]

I had a S760H and returned it for the X3800H before I saw the test results. However I have 60 days to try it (thanks Crutchfield!). The other was a little underpowered it seemed and the X3800H is perfect in that respect. It does sound fuller and a bit clearer to me.

Contact me in a month and I will let you know my thoughts. So far I am pleased with a number of the other features but am sad at the test results here. Unfortunately there was not another one available in my price range that had the feature set I wanted this time. They were out of the 3700H model.

If you see my other posts it is likely I will not be able to differentiate due to some hearing loss.

If I can get a 3700H or an open box 4700H I may return it as a result of Amir’s testing.

You can't hear the difference, but you're considering exchanging it because of test results that don't affect you?

 

[Hobolicious]

No need to have your hopes crushed if your concern is whether the pre out voltage is high enough

It seems like you are quite knowledgable in the area. How do you think my Classe Audio CA-5200 would perform as amp for FL, FR and C in combination with the Denon X3800H? Specifications below:

 

[Descartes]

It seems like you are quite knowledgable in the area. How do you think my Classe Audio CA-5200 would perform as amp for FL, FR and C in combination with the Denon X3800H? Specifications below:

View attachment 240610

Have the same amp since it came out in 2010 and it works great regardless of the numerous numbers of speakers I throw at it!
It is dead silent just warms up a bit!

 

[Narnian]

You can't hear the difference, but you're considering exchanging it because of test results that don't affect you?

As Amir and several others pointed out I may not want to reward Denon for putting out a product that is a step backward in a major area of performance from its predecessor. I likely will not return it but may if I can get something better and cheaper. $ is always a good incentive. I have returned 3 AVRs to Crutchfield because of feature issues where they did not work as advertised but were complex to implement and maintain.

When I talked with the audiologist about my stereo purchasing and testing they said the cleaner the signal to your ear the more you can understand. Hearing is a lot more complex and clarity can be affected by things we may not be able to hear directly.

 

[peng]

It seems like you are quite knowledgable in the area. How do you think my Classe Audio CA-5200 would perform as amp for FL, FR and C in combination with the Denon X3800H? Specifications below:

View attachment 240610

Amir already answered your question in the first post if you looked at the SINAD vs output level graph lol..
At below 1.2 V, you can expect >=90 dB SINAD, that's 0.003% THD+N

You only need 1.4 V for the CA-5200 to output it's rated 200 W.

So while the AKM IC fitted Denon AVRs would theoretically be a better match, the X3800H should work well with the Classe amp too if you only listen with ears and brains.

 

[Madlop26]

So is the X3800H worth it? well the true is nobody knows for sure, to know it first step should be getting a statistical analysis of the population SINAD audibility threshold (stratify by age of course), but even if we get it, that will not answer the question for everybody, as this threshold is very likely to be individual.
What you need is a reproducible, reliable and practical way to measure your own SINAD threshold (tough), until then , I agree with Amir, as science improves we should be getting better SINAD (just in case) and not lower.

 

[Hobolicious]

Amir already answered your question in the first post if you looked at the SINAD vs output level graph lol..
At below 1.2 V, you can expect >=90 dB SINAD, that's 0.003% THD+N

You only need 1.4 V for the CA-5200 to output it's rated 200 W.

So while the AKM IC fitted Denon AVRs would theoretically be a better match, the X3800H should work well with the Classe amp too if you only listen with ears and brains.

Thanks for the reply. I have not changed my receiver for many, many years, but need to do so now as I am moving to a 7.2.4 setup.

I have not been in the HiFi «numbers» game for so long I have forgotten most of the lingo, and therefore not as easy to understand the graphs and numbers, so thanks for running through it.

 

[peng]

So is the X3800H worth it? well the true is nobody knows for sure, to know it first step should be getting a statistical analysis of the population SINAD audibility threshold (stratify by age of course), but even if we get it, that will not answer the question for everybody, as this threshold is very likely to be individual.
What you need is a reproducible, reliable and practical way to measure your own SINAD threshold (tough), until then , I agree with Amir, as science improves we should be getting better SINAD (just in case) and not lower.

If you search, you will find there are no shortage of articles and studies including statistical analysis from long time ago.
Here's an old one written by an electrical engineer, I like it because it is easy to read and only two pages.

https://worldradiohistory.com/hd2/IDX-UK/Technology/Technology-All-Eras/Archive-Wireless-World-IDX/80s/Wireless-World-1981-02-OCR-Page-0019.pdf

The thing is, if one tries hard enough, one may be able to hear the distortions of the X3800H under some conditions, but even then, it is very possible that one may prefer the higher distortions because they are the low order (2nd and 3rd) harmonics as shown in the FFT diagram. The known to be more offensive harmonics are at or below -110 dB.

I am a little concern about it's IMD too though again, one would have to have good hearing and listen very hard to be offended by the still low levels of IMD.

 

[pseudoid]

An ABX test is specifically designed to remove the subjective bias of the subject.

Not for a subject who gets anxiety-attacks when confronted with any type of 'tests'... ABX to colon-cancer to SAT/ACT!

Would you say the same about a TV released with a load of new features compared to the previous model but 'worse' measured infrared 'performance' (a difference your eyes won't detect)?

Depends.

One size does not fit all, even with Depends.
BTW: Have you tried to purchase an high-end, large-screen, OLED TV monitor that is just a "Monitor" (w/o a "load of features")?
Don't bother!

 

[Bren Derlin]

As Amir and several others pointed out I may not want to reward Denon for putting out a product that is a step backward in a major area of performance from its predecessor. I likely will not return it but may if I can get something better and cheaper. $ is always a good incentive. I have returned 3 AVRs to Crutchfield because of feature issues where they did not work as advertised but were complex to implement and maintain.

When I talked with the audiologist about my stereo purchasing and testing they said the cleaner the signal to your ear the more you can understand. Hearing is a lot more complex and clarity can be affected by things we may not be able to hear directly.

I haven't heard the x3800H personally (yet), so I don't know if it's audibly (perceptibly?) any better or worse than the x3700H. If the audible difference is literally negligible from a human perspective, then it's not a step backwards in a major area of performance that affects anything other than a highly sensitive recording device.

But really - if you don't need 4 subs out, nor Dirac, and you're not heavily into 8K video processing (who really is at the moment?), then getting the x3800H over the x3700H is moot anyway. Save your money, and get the x3700H refurb for $1,200.

 

[GaryH]

If you search, you will find there are no shortage of articles and studies including statistical analysis from long time ago.
Here's an old one written by an electrical engineer, I like it because it is easy to read and only two pages.

https://worldradiohistory.com/hd2/IDX-UK/Technology/Technology-All-Eras/Archive-Wireless-World-IDX/80s/Wireless-World-1981-02-OCR-Page-0019.pdf

Do you have the rest of that article?

 

[voodooless]

Not for a subject who gets anxiety-attacks when confronted with any type of 'tests'... ABX to colon-cancer to SAT/ACT!

For those people, the M-Scaler is made…

 

[Narnian]

I haven't heard the x3800H personally (yet), so I don't know if it's audibly (perceptibly?) any better or worse than the x3700H. If the audible difference is literally negligible from a human perspective, then it's not a step backwards in a major area of performance that affects anything other than a highly sensitive recording device.

But really - if you don't need 4 subs out, nor Dirac, and you're not heavily into 8K video processing (who really is at the moment?), then getting the x3800H over the x3700H is moot anyway. Save your money, and get the x3700H refurb for $1,200.

If I was starting over and knew what I know now I would likely have gotten a 3700 refurb. However Crutchfield has been very good (and patient) with me. I have been buying from them for years and their support is superb. They have earned my respect and my $. Especially for the 3 exchanges I have done the last few months.

If in the next 60 days one pops up I can exchange it again.

I did testing with a 1700H (returned because I could not pass optical inputs to Zone 2) with an external DAC and could not tell the difference. I have another one coming in I will test with the 3800H for grins. I suspect I will not hear anything different. The DAC is going to be used in the library (aka family room overflowing with books) with an Aiyima A07 amp.

 

[Karls_speaker_works]

I haven't heard the x3800H personally (yet), so I don't know if it's audibly (perceptibly?) any better or worse than the x3700H. If the audible difference is literally negligible from a human perspective, then it's not a step backwards in a major area of performance that affects anything other than a highly sensitive recording device.

And I hate to keep harping on it, but AKM had a fire in October of 2020 that stopped all production of the DAC used in the X3700H, but using a little math, it seems unlikely that a X3700H produced since early-mid 2021 performs any better than the X3800H.

I.E.
AKM DAC: 104 dB spec - 98 dB measured; difference = 6dB.
TI DAC: 93dB spec - 87 dB measured; difference = 6dB.

I see no way to infer any notable engineering issues doing a simple comparison, so we can assume the X3700H sold for the past 14 months or more performed essentially the same as the X3800H. Testing of the X4800H will confirm if the design is truly limited by the new DAC as it appears or whether there's still some engineering "hygiene" that can be improved. But buying an X3700H assuming better performance is a nothing more than a roll of the dice.

Do you have the rest of that article?

Haha, I had the same question, I was reading intently, and then "hey, where's the next page". Lol.

 

[GaryH]

Haha, I had the same question, I was reading intently, and then "hey, where's the next page". Lol.

Found the full issue:
https://worldradiohistory.com/UK/Wireless-World/80s/Wireless-World-1981-02.pdf
(The missing last page of the article is on page 20 of this pdf.)

Quotes:

D.E.L. Shorter of the BBC Research Department compared the sound quality of a number of systems for which the measured harmonic spectra were known. Comparing amplifiers having different harmonic spectra he found that the just perceptible distortion was 0.4% in one instance

If it is intended to be ultra-critical our 'well below the detectable distortion' value of 0.13% could be accepted as the desirable target

Those correspond to -48 dB and -58 dB respectively (rounding the latter down to 0.1% would be -60 dB).

 

[Dj7675]

In general I don't think we need to worry much about the 10 dB headroom because Dirac, like other RC software typically would avoid boosting anyway. And most if not all such high magnitude net boosts will be in the bass range, where people have very high tolerance to distortions. In terms worst case scenario, it is a valid point regardless.

Not sure sure... I have 2 systems with Dirac... flipping between Dirac on/of is a significant drop in volume. I had read that it allows for 10dB of boosts. I had never really noticed different volume levels much with Audyssey enabled/disabled. But it is very significant volume drop when Dirac is enabled. If you have to turn up your volume to get to the same volume without Dirac enabled, it would seem all of the noise/distortion would be boosted as well. I need to play a test tone with REW with Dirac enabled/disabled to know exactly how many dB the difference actually is.