Denon AVR-X4700H 2020 AVR Review

[amirm]

@amirm, can you please measure the outputs at 1.0 or 1.2 volts ?

That is in the review graphs:

Just look at the X axis. It starts at 0.4 volts and goes up to 4+ volts.

 

[SIY]

Would you agree that we have seen a trend where quality has taken a back seat to corporate profits in many industries?

That depends on your definitions. For the things that are important to me, I can get fantastically better computing, sound, video, phone, cameras, recorders, cars, refrigerators, stoves, and amazing automation in cheaper and cheaper products than ever before. The cost over time is lower than ever before, as well.

Real quality, in the engineering sense, drives higher profits. That's why we do it.

 

[Skylarlove1999]

Again, they already have a design with good engineering (3600H). Unless that was designed by another company, there is no added cost there.

Yes you are absolutely correct. I am just talking in general about the added costs of hiring accomplished experienced engineers. I mean you said yourself maybe somewhat tongue and cheek that maybe Denon stumbled into good results with the 3600 than by any conscious engineering decisions. Not trying to argue with your premise for the 3600 and the differences between it and the 4700. Hopefully they can find the issue and resolve it.

 

[peng]

I doubt the HDMI board differs between models but it may differ between model years.
The AVR-3600H multi-tone measurements for the COAX input not the HDMI inputs
https://www.audiosciencereview.com/...ds/denon-avr-x3600h-av-receiver-review.12676/

View attachment 70022

The 4700H has HDMI measurements and not COAX in so there is no apples-to-apples comparison.
Perhaps the other measurements are indicators but 3600H HDMI input may not be issue free.

- Rich

Did you realize we are saying the same thing basically? I am proposing two possible reasons for why the 4700 measured worse than the 3600, one being the tolerance of the volume chip being very wide, and second the different HDMI boards, and yes they are between different model year, in fact different generations! The 3600 has HDMI 2.0, and only 4 K capable. The 4700 has the 8 K capable HDMI 2.1 board, again that's different generations and is the first available so may even be buggy for a while.

 

[Gedeon]

That is in the review graphs:

Just look at the X axis. It starts at 0.4 volts and goes up to 4+ volts.

Thanks. It's through HDMI or analog input ?

 

[amirm]

Thanks. It's through HDMI or analog input ?

HDMI.

 

[RichB]

Did you realize we are saying the same thing basically? I am proposing two possible reasons for why the 4700 measured worse than the 3600, one being the tolerance of the volume chip being very wide, and second the different HDMI boards, and yes they are between different model year, in fact different generations! The 3600 has HDMI 2.0, and only 4 K capable. The 4700 has the 8 K capable HDMI 2.1 board, again that's different generations and is the first available so may even be buggy for a while.

Largely, except for the elephant in the room.

To me, like the Arcam's weird behavior in the significant variance in SINAD measurements that seemed to be affected by whether only the front two speakers were selected or with more speakers selected (i.e. non none..) So it's not clear cut to me if such weird measurements mean anything bad audibly speaking.

What you call weird, I call atrocious.
Perhaps @amirm will establish the precise criteria that could relegate the performance issue to a somewhat unique use case.
However, at this point, the conservative view is that in the normal HDMI use case, this device is deeply flawed.

It may well be that this issue can be corrected with a firmware update.
But as it stands today, it is rational (since there is little need for 8K) to buy an AVR-X3600H over the AVR-X4700.

- Rich

 

[peng]

That's a good point. I will need to investigate this as it may be that more than one input is driven in HDMI mode even though I disable the other channels in software.

But I think RichB got the 3600 mixed up in terms of model year. The 3600 has the 4K HDMI board, and is a 2019 model, the 4700 has the brand new HDMI 2.1 board 8K ready, definitely a different board and obvious the earliest version, bug prone, perhaps.. As to your point about ....more than one input is driven in HDMI mode.............that seems to be a separate issue than RichB's.

 

[peng]

Largely, except for the elephant in the room.


What you call weird, I call atrocious.
Perhaps @amirm will establish the precise criteria that could relegate the performance issue to a somewhat unique use case.
However, at this point, the conservative view is that in the normal HDMI use case, this device is deeply flawed.

It may well be that this issue can be corrected with a firmware update.
But as it stands today, it is rational (since there is little need for 8K) to buy an AVR-X3600H over the AVR-X4700.

- Rich

You are right about the 32 tone test though, that he used coax for the 3600. If he had used coax on the 4700, the results might have been more comparable but we'll never know now.

 

[demoncamber]

You certainly have enough people on the case, managing optics anyway

- Rich

Hopefully I get at least one persons attention

 

[Baff]

The 3600 has the 4K HDMI board, and is not a 2019 model

The 3600 is a 2019 model. Or, do you mean the 4K HDMI board?

 

[Baff]

But as it stands today, it is rational (since there is little need for 8K) to buy an AVR-X3600H over the AVR-X4700.

It is not only about 8K. It is also about 4K/120Hz, which will be very relevant in a few months.

 

[peng]

The 3600 is a 2019 model. Or, do you mean the 4K HDMI board?

Oh, it was a typo, of course the 3600 is a 2019 model with a 4K/HDMI2.0 HDMI board. The 4700 is the real 2020 model with an 8K/HDMI2.1 HDMI board. Thank you again.

 

[peng]

First off, let me just say I'm always looking forward to your next review. I actually preordered the x4700!
So after reading all of this I decided to send it as well as my own feelings to a lot of the inside people in hopes of getting some sort of explanation, and if I get a reply I'll be posting them here.[MOD: email addresses deleted] all have something to say.

Thank you, and I think there is a good chance your effort would pay off, based on their history of having reacted to the critique by hometheaterhifi.com and Audioholics.com by finally upgrading the HDAM modules, replacing the LSI volume control chip with the NJU MSI chip and upgraded the DAC chip by a notch, and just added a preamp mode to the 2020 models. So it is time they listen to ASR and do something about the 2020 models inexplicable slip back from what they gained in the X3600H.

 

[kokishin]

Oh, it was a typo, of course the 3600 is a 2019 model with a 4K/HDMI2.0 HDMI board. The 4700 is the real 2020 model with an 8K/HDMI2.1 HDMI board. Thank you again.

I'm not contradicting you. Just pointing out that the new HDMI 2.1 board has a single HDMI 2.1 40GHz input and two HDMI 2.1 40GHz outputs, one of which supports eARC. Should be an even more interesting design challenge in the future, when all HDMI inputs and outputs support 48GHz.

 

[peng]

Largely, except for the elephant in the room.


What you call weird, I call atrocious.
Perhaps @amirm will establish the precise criteria that could relegate the performance issue to a somewhat unique use case.
However, at this point, the conservative view is that in the normal HDMI use case, this device is deeply flawed.

It may well be that this issue can be corrected with a firmware update.
But as it stands today, it is rational (since there is little need for 8K) to buy an AVR-X3600H over the AVR-X4700.

- Rich

I hope so, but the issue appear to be very much higher noise level than the X3600H so I doubt (again love to be wrong) FW could fix it, not completely anyway. The more I think about it, the more I think that new HDMI board is the real culprit for the high noise. If you look at the analog input and coax measurements, the numbers look a lot better, still not as good as the 3600 but at least clearly better than the AV7705.

 

[peng]

I'm not contradicting you. Just pointing out that the new HDMI 2.1 board has a single HDMI 2.1 40GHz input and two HDMI 2.1 40GHz outputs, one of which supports eARC. Should be an even more interesting design challenge in the future, when all HDMI inputs and outputs support 48GHz.

Noted, and as Amir pointed out, high noise (noisier than his famous reference dongle) was the issue and it may not matter which inputs/outputs were used. THD alone was at -100 dB, that's not bad at all, so the 8K HDMI 2.1 board may be the one part that generated much higher noise level than the 4K HDMI 2.0 board.

 

[thepiecesfit]

In an interesting development, I've brought this to the attention of Audioholics on their Youtube channel on the video 'Denon Virtual Press Event - New 8K Ready X-Series AV Receivers' who seem to have some influence with Denon engineers. Some bold claims there I'll be looking forward to their alternative testing methods.

Audioholics: #######, I've read some of his test reports on dacs and don't agree with his test methods. I also know the engineers that design Denon gear and the 4700h is not a downgrade to the 3600h. When I get caught up I will measure and review and newer Denon receiver to illustrate that.

 

[peng]

In an interesting development, I've brought this to the attention of Audioholics on their Youtube channel on the video 'Denon Virtual Press Event - New 8K Ready X-Series AV Receivers' who seem to have some influence with Denon engineers. Some bold claims there I'll be looking forward to their alternative testing methods.

Audioholics: #######, I've read some of his test reports on dacs and don't agree with his test methods. I also know the engineers that design Denon gear and the 4700h is not a downgrade to the 3600h. When I get caught up I will measure and review and newer Denon receiver to illustrate that.

I frequent that site too and while I respect Gene, I have to say while the two may have different test protocols, the important point is consistency. Amir measures gear the same way, using the same AP so the result between gear are highly comparable barring exceptions due to various reasons, as there always could be exceptions. If AH is to demonstrate, they would have to measure not only the 4700, but also the 3600.

 

[kokishin]

In an interesting development, I've brought this to the attention of Audioholics on their Youtube channel on the video 'Denon Virtual Press Event - New 8K Ready X-Series AV Receivers' who seem to have some influence with Denon engineers. Some bold claims there I'll be looking forward to their alternative testing methods.

Audioholics: #######, I've read some of his test reports on dacs and don't agree with his test methods. I also know the engineers that design Denon gear and the 4700h is not a downgrade to the 3600h. When I get caught up I will measure and review and newer Denon receiver to illustrate that.

Below is a straight copy/paste (no formatting) from the comments section of:

Audioholics reply seems harsh and uncalled for: "ASR seems to be on a crusade to thrash as much gear as possible using an Audio Precision. If it's not a $20 USB dongle, it's likely not going to get a glowing review from them."

I suppose if AH had discovered the issue, they'd be crowing about it?

COMMENTS

thepiecesfit
thepiecesfit
2 days ago
Audio Science Review recently reviewed the Denon X4700H and it measured considerably worse than last years X3600H. Can Audioholics back this up in some future review? Would love to see more sites doing objective measurements.
https://www.audiosciencereview.com/forum/index.php?threads/denon-avr-x4700h-2020-avr-review.14188

1


Audioholics
Audioholics
1 day ago
ASR seems to be on a crusade to thrash as much gear as possible using an Audio Precision. If it's not a $20 USB dongle, it's likely not going to get a glowing review from them.



SDX9000
SDX9000
1 day ago
@Audioholics, please explain this statement in detail.
Note that this ASR with the "crusade to thrash as much gear as possible" gave a very favorable recommendation to X3600H. Now when the latest and more expensive X4700H showed substantially lower performance in ASR testing, Denon should step up and defend their reputation. Receivers are all about audio quality over HDMI, so this is not a small joke.

+ here is an example of a manufacturer that does step up for their reputation: https://www.audiosciencereview.com/...eaker-measurements-take-two.11323/post-425237

Denon, your move.



Audioholics
Audioholics
18 hours ago
@SDX9000 I've read some of his test reports on dacs and don't agree with his test methods. I also know the engineers that design Denon gear and the 4700h is not a downgrade to the 3600h. When I get caught up I will measure and review and newer Denon receiver to illustrate that.



SDX9000
SDX9000
8 hours ago
@Audioholics I am looking forward for an alternative point of view and especially a dispute for why ASR measurements might be misleading in this case.
Must say it is risky to make claims about X4700H performance without measuring it. There are plenty of examples of products with great claims and even designs that are crippled by poor quality control or some mess-up on the road from engineering prototype to a store shelf. Hope Denon can resolve any issues, because everything but ASR measurements seem to be great with X3700 and X4700.