Denon AVR-X3700H AVR Review

[jpqpi]

In order for this to be an objective fact based discussion, here are some differences I have found between the TI PCM5102A and the AKM4458 DACs (from datasheets):

1. THD + N
AKM4458 = -107dB
PCM5102A = -93dB

2. Stop band Filter attenuation
AKM4458 = -80dB
PCM5102A = -60dB

Taken from manufacturers datasheets:
https://www.akm.com/content/dam/doc.../audio-dac/ak4458vn/ak4458vn-en-datasheet.pdf
https://www.ti.com/lit/ds/symlink/p...https%3A%2F%2Fwww.ti.com%2Fproduct%2FPCM5102A

It seems to me that (and please correct me if I am wrong here):
1. there is a large difference in THD + N ratio and this is quite early in the DAC -> PREAMP -> AMP signal chain so it wont be the final THD + N figure.
2. Amir in his other reviews considers 60dB attenuation to be poor

 

[peng]

1. there is a large difference in THD + N ratio and this is quite early in the DAC -> PREAMP -> AMP signal chain so it wont be the final THD + N figure.

Agreed, in the signal chain there are at least one NJM8080G and there is one NJU72343 ICs upstream/downstream of the DAC chip per channel. For Marantz, there is the HDAM as well.

In terms of THD+N:

NJM8080G - https://pdf1.alldatasheet.com/datasheet-pdf/view/1221310/NJRC/NJM8080G.html
0.0005%, 1 kHz

AVR-Audio Video Reciever-Build Quality: Part V - HomeTheaterHifi.com

Good News! Marantz Says No More AVR LSIs for You! You might recall my article on Audio Video Receiver Build Quality announcing the downgrade of the analog section of almost every AVR and Pre Pro coming from the major Far Eastern manufacturers. The article discussed how multi-channel audio...

hometheaterhifi.com

NJU72343 - https://www.nisshinbo-microdevices.co.jp/ja/pdf/datasheet/NJU72343_J.pdf (Page 3)

0.0004% - f=1kHz, VIN=1Vrms, Volume=0dB, BW:400Hz-30kHz
0.0006% - f=10kHz, VIN=1Vrms, Volume=0dB, BW:400Hz-30kHz

So on surface, the NJM8080G opamp would be the bottleneck when the AK4458 was used. With the PCM5102A in the signal chain, it would likely be the bottleneck but given that each chip's specified typical THD+N has tolerances, it probably would depend on the specific DUT. As others mentioned, there likely won't be audible differences if based on THD+N only, but if you only want to look at things on the objective, quantifiable side only, you would stick with those fitted with the AK4458.

 

[ieatfishburritos]

Just thought I'd post to share that my 9 month old 3700 has experienced a failure and will be shipped off for repair. TV was on during the daytime, nothing loud, and sound stopped working. factory reset or such doesn't recover. zone 2 still works. after factory reset it freezes at the first speaker setup test tone "can you hear sound from the left front". If I exit out of setup mode and manually into audyesee setup it will again freeze as well starting with the subwoofer pre-out. Can't set levels with test tones manually as well.. Seems the main zone has an issue.

 

[amper42]

Just thought I'd post to share that my 9 month old 3700 has experienced a failure and will be shipped off for repair. TV was on during the daytime, nothing loud, and sound stopped working. factory reset or such doesn't recover. zone 2 still works. after factory reset it freezes at the first speaker setup test tone "can you hear sound from the left front". If I exit out of setup mode and manually into audyesee setup it will again freeze as well starting with the subwoofer pre-out. Can't set levels with test tones manually as well.. Seems the main zone has an issue.

Hopefully, you didn't have to ship it to PANURGY OEM for service. There are so many horror stories of receivers coming back worse than they were when shipped in. Lots of new scratches, parts rattling inside the chassis after the repair and broken front doors. If that happens you have to send it back again. Have fun!

 

[Doro89]

This is a review and detailed measurements of Denon AVR-X3700H 9.2 channel 8K Home Theater AV Receiver (AVR). It is kindly loaned to me by a local member and costs US $1,199.

Visually, if you have seen one AVR, you have seen the look of the AVR-X3700H:

View attachment 76014

The input selector is plastic and while rotating it feels fine, the plastic feel is a bit jarring because you expect it to be metal. AVRs though are universally used with remote controls so this is just a minor nit.

The back panel has the typical connectors including the single new "8K" capable HDMI Input:

View attachment 76015

The 3700H is lighter than its higher end siblings which I personally appreciated as I lugged it around.

In use, the unit was quite robust, never shutting down no matter how much I pushed its amplification in my standard tests. Other AVRs can be very sensitive, constantly shutting down when pushed, often requiring a power cycle. Not here.

The 3700H does run warm and I will show the thermal imaging of it later.

There is a lot of anticipation for this review as I gave its last year mode, the AVR-X3600H my best rating of any AVR tested. Alas, I reviewed that late in its design cycle so availability is very poor. Hope is that this replacement unit doesn't regress in measured performance while adding features. So let's get into that.

Note: Denon engineering was kind enough to review these measurements and confirm that they match their expectations.

AVR DAC Performance
As usual, when we have pre-amplifier output functionality on an AVR, we feed the unit digital 1 kHz perfect sine wave and see what it outputs in analog domain (amplifiers shut off):

View attachment 76021

Ah, that is a sign of relief. SINAD which is the sum of noise and distortion is in the same ballpark of AVR-X3600H:

View attachment 76020

One to two dB difference is to be expected so no cause for concern. I also tested Coax input (not shown) and it had the same performance.

My standardized tests use 2 volts output but since you can't turn off the amplifiers beyond 2 or all channels, let's see what the performance is like for other output levels before the amplifiers clip and drag down the performance of the DAC subsystem:
View attachment 76022

As you see, peak performance with the amplifiers off is around 1.1 volts with SINAD of 101 dB which is excellent for an AVR. With the amps on, you are OK up to 1.4 volt output before it nose dives. So when selecting an external amplifier for channels beyond fronts, make sure it can output its maximum power at or below 1.4 volts (usually specified as "sensitivity").

Signal to noise ratio is about the same as 3600H as well:
View attachment 76023

32-tone test signal which simulates "music" shows more intermodulation distortion than the 3600H:
View attachment 76024

Here is the 3600H for reference:

The 3700H though evens the match by producing near perfect linearity (accuracy) score:

View attachment 76025

Back to intermodulation distortion, here is how it varies with level:

View attachment 76026

Jitter performance is the same as 3600H in both good and bad ways:

View attachment 76027

Coax/Toslink inputs generate a ton of jitter and spurious sidebands (tones). Levels are very low though so more of a visual distraction than what is audible. HDMI achieves a lower noise floor which indicates the other inputs suffer from random/broadband jitter. However, the HDMI jitter profile is different (blue) showing that better care could be put in there to isolate the DAC from what is around it.

The filter performance is decent:
View attachment 76028

The attenuation here is better than what we get in the higher end AVR-X6700H which is an interesting "reversal of fortunes:" (label on graph is wrong -- should say AVR-X6700H)

Poor attenuation usually makes our broadband distortion+noise versus frequency worse. So good thing the filtering is better in 3700H:

View attachment 76029

HEOS Streaming Performance
Here is a quick test which shows streaming performance to be similar to HDMI/Coax inputs:

View attachment 76044

AVR Amplifier Measurements
I have been testing Denon AVR amplifiers using CD input which I have found to not be digitized allowing us to see the true performance the amplifier rather than any processing blocks. Question has been raised as to what happens when you turn on signal processing such as bass management. Here is the answer:

View attachment 76030

The answer is naturally, the AVR will digitize such analog signals the moment you do that. As it should. Fortunately the Pure Direct button overrides that as show in red graph. Digitization is at high sample rate of 96 kHz which is nice. Note that level changes a bit so be careful if you are doing AB tests.

For the rest of these tests I will be using CD input in pure-direct with speakers set to large.

Here is our usual dashboard with 1 kHz tone again:

View attachment 76031

Denon once again shows that its amplifiers are above average even though many channels are stuffed in a "mass market" product:

View attachment 76032

And within all AVRs tested:

View attachment 76034

I wish the dynamic range was 6 dB better to clear CD at 16 bits (which the 3600H achieved):
View attachment 76033

Crosstalk is a bit disappointing but Ok from audibility point of view:

View attachment 76035

Power output versus distortion and noise which is one of the most important metrics into 4 ohm shows respectable results:

View attachment 76036

Allowing distortion to grown to 1%, we get even more power, continuous or peak:

View attachment 76037

That's good bit of headroom which is nice to have in any amplifier since peaks in music can be momentary.

Switching to 8 ohm load we get to compare our measurements against the company's spec: (label is a typo: should say 3700H)

View attachment 76038

Testing frequency sensitivity with full power sweep we get:

View attachment 76040

This is a bit less tidy than 3600H but still far better than many other amps, specially the switching/Class D ones.

Note that as is typical, power output drops at 20 Hz (orange) which is where you need it most.

AVR Power Scaling with Channels
There is a shared power supply in all of these AVRs so as you turn on more channels, power you get from each channel drops:

View attachment 76039

AVR Heat & Thermal Analysis
Denon & Marantz AVRs tend to run warm especially if used without the ECO mode. The 3700H doesn't get as warm as its larger brothers which is good:

View attachment 76041

Can't figure out why the hot channels are in the middle even though the speaker terminals are all the way to the right. So perhaps that is how they are wired.

Conclusions
What a sigh of relief that the Denon AVR-X3700H performns on par with the later year mode, 3600H. The other "2020 year" Denon AVRs we have tested have had worse performance, leaving the 3600H as the best performing AVR until now. So buy the 3700H with peace of mind knowing that it performs quite well (for an AVR). Of course be mindful of what other features the units above it have which you may want, top of which is more power.

I am happy to add Denon AVR-X3700H to my recommended list.

-----------
As always, questions, comments, recommendations, etc. are welcome.

Get this: money is so tight that I not only have to mow my own lawn, but had to fix the darn riding mower myself!!! Sweating in hot sun, fixed the rusty connections on the solenoid to get it to start reliably. Please rescue me from this misery so I can find other people do such work by donate what you can using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Hello, thank you for all your reviews.
So I see that some equipment like an AVR X8500H has two results on the spreadsheet/index of all the equipment tested. 1st is with HDMI IN and RCA OUT which rated good. 2nd is with HDMI IN and Speaker Out which rated not so good about 20db difference.
Does it mean if I let’s say have that AVR and have speaker wire connected I will not get best results?
Does it mean that the best way to go is to have an amplifier connected to preouts on the AVR?

 

[GalZohar]

Hello, thank you for all your reviews.
So I see that some equipment like an AVR X8500H has two results on the spreadsheet/index of all the equipment tested. 1st is with HDMI IN and RCA OUT which rated good. 2nd is with HDMI IN and Speaker Out which rated not so good about 20db difference.
Does it mean if I let’s say have that AVR and have speaker wire connected I will not get best results?
Does it mean that the best way to go is to have an amplifier connected to preouts on the AVR?

Supposedly your amplifier will need to have better performance than the receiver's built-in amplifier, otherwise you could get even worse results.

 

[Doro89]

Supposedly your amplifier will need to have better performance than the receiver's built-in amplifier, otherwise you could get even worse results.

Yes, that’s what I meant. Better performing amp.
I am trying to figure out, how can I have my setup. I have Nvidia Shield Pro for all my movies and flac/hi res music that’s stored on local NAS. Doing tons of research it seems all avrs are crappy for music. So I started to look for separates (topping, hypexetc), but no affordable dacs have hdmi.
My goal is to have 5.1 or 5.1.2 at most. I want to use same system for stereo music. Speakers that I am getting are not easy to drive. LCR Lumina II, 4ohm, 86db, 30-150w.

So my options are only to have a AVR and deal with not so perfect audio quality or get htpc which will send video to TV and audio to usb dac and amps, but then I can’t use other sources like ps4 etc? Anything else can be done to maintain high audio performance?

 

[Doro89]

Supposedly your amplifier will need to have better performance than the receiver's built-in amplifier, otherwise you could get even worse results.

also, let’s say I get denon x3700h. For movie performances stand point it’s great, but for music not so great unless I get an amp. But another problem here is that if I do get an amp I am forced to have an amp that has 5+ channels because AVR can’t enable/disable each channel independently, it’s either all or nothing.

 

[HarmonicTHD]

also, let’s say I get denon x3700h. For movie performances stand point it’s great, but for music not so great unless I get an amp. But another problem here is that if I do get an amp I am forced to have an amp that has 5+ channels because AVR can’t enable/disable each channel independently, it’s either all or nothing.

Where do you get the idea these are not great for music? What do you base this statement on?

 

[Doro89]

Where do you get the idea these are not great for music? What do you base this statement on?

if you look at the measurements. Using speaker wire as output on lets say denon X8500H, has SINAD score 85. At approximately what point the SINAD score becomes inaudible? As far as I understood anything better than 80db is good? So 85 would be just above? not ideal? Can you hear the difference between 80db SINAD and 100db listening to good quality stereo music?

 

[delta76]

also, let’s say I get denon x3700h. For movie performances stand point it’s great, but for music not so great unless I get an amp. But another problem here is that if I do get an amp I am forced to have an amp that has 5+ channels because AVR can’t enable/disable each channel independently, it’s either all or nothing.

I believe you can choose Pure Direct (or whatever the mode is called on Denon), and it's basically 2.1 setup for listening to music. You only need stereo amp.

 

[Doro89]

I believe you can choose Pure Direct (or whatever the mode is called on Denon), and it's basically 2.1 setup for listening to music. You only need stereo amp.

So to be clear. I have denon. I have let’s say topping or any other hypex or purifi amps that is connected to LR preouts. Sub and surround connected to the AVR speaker out. When I watch movie all channels are working including Front LR? When listen to music in 2.1 all speaker outs on AVR go to sleep, and preouts are bypassing directly to the amp?

 

[delta76]

So to be clear. I have denon. I have let’s say topping or any other hypex or purifi amps that is connected to LR preouts. Sub and surround connected to the AVR speaker out. When I watch movie all channels are working including Front LR? When listen to music in 2.1 all speaker outs on AVR go to sleep, and preouts are bypassing directly to the amp?

You can connect the LR to preout => amp => speakers, and other channels directly to the amped channels of the avr. Not saying you have to. The denon should sound just fine and room correction would make a bigger impact than adding an amp.

Of course if you want to use amp, or amps, you have several options.

 

[Doro89]

You can connect the LR to preout => amp => speakers, and other channels directly to the amped channels of the avr. Not saying you have to. The denon should sound just fine and room correction would make a bigger impact than adding an amp.

Of course if you want to use amp, or amps, you have several options.

Well that’s the question. Whether worth it having a good sounding stereo amp and connect it to preouts on Denon? Or denon will degrade the quality somehow?

 

[delta76]

As I said, the denon should sound just fine. You can always add an amp, or a dac with SOTA measurements but you will likely not notice a difference. Better save your money and invest in room correction

 

[Doro89]

As I said, the denon should sound just fine. You can always add an amp, or a dac with SOTA measurements but you will likely not notice a difference. Better save your money and invest in room correction

Got it thank you for help!

 

[peng]

Well that’s the question. Whether worth it having a good sounding stereo amp and connect it to preouts on Denon? Or denon will degrade the quality somehow?

You are asking a very popular question but unfortunately it is hard to give a yes/no answer without qualifications.

I compared most of my amps, preamps and avrs over the years and have never been able to tell which one, or combinations sounded better or even different when never matched and switched between the duts quickly. So I used REW to see if the mic could tell a difference. The most extreme one I compared was a case of a Denon AVR-X3400H in pure direct mode versus a Cambridge Audio preamp and Parasound A21 combination. Based on specs and available measurements of comparable devices, I think is is reasonable to assume both have THD+N and FR, below the threshold of audibility. Frequency response are arguably the easiest to tell if different. Since most FR bench measurements used resistor test loads so I thought may be if real speakers are used, the difference would be more obvious. Yet all my REW graphs showed difference of within +/- 1 dB using 1/48 or psychoacoustic smoothing.

Below is one example, take a look and do you think the AVR degrade the "sound quality"?
Note that at the SPL for that test, my Halo A21 amp would certainly be running in class A mode too, yet I really couldn't say it sounded better, but then again that's just me. Having said that, most reviewers and audiophiles would probably consider 0.01% THD+N (80 dB SINAD) below the threshold of audibility.

If you look at the Denon, Marantz and Yamaha AVRs tested on ASR so far, their SINAD at 5 W are mostly in the mid 80 range, very similar to that for the integrated and power amps measured.

 

[Doro89]

You are asking a very popular question but unfortunately it is hard to give a yes/no answer without qualifications.

I compared most of my amps, preamps and avrs over the years and have never been able to tell which one, or combinations sounded better or even different when never matched and switched between the duts quickly. So I used REW to see if the mic could tell a difference. The most extreme one I compared was a case of a Denon AVR-X3400H in pure direct mode versus a Cambridge Audio preamp and Parasound A21 combination. Based on specs and available measurements of comparable devices, I think is is reasonable to assume both have THD+N and FR, below the threshold of audibility. Frequency response are arguably the easiest to tell if different. Since most FR bench measurements used resistor test loads so I thought may be if real speakers are used, the difference would be more obvious. Yet all my REW graphs showed difference of within +/- 1 dB using 1/48 or psychoacoustic smoothing.

Below is one example, take a look and do you think the AVR degrade the "sound quality"?
Note that at the SPL for that test, my Halo A21 amp would certainly be running in class A mode too, yet I really couldn't say it sounded better, but then again that's just me. Having said that, most reviewers and audiophiles would probably consider 0.01% THD+N (80 dB SINAD) below the threshold of audibility.

If you look at the Denon, Marantz and Yamaha AVRs tested on ASR so far, their SINAD at 5 W are mostly in the mid 80 range, very similar to that for the integrated and power amps measured.

View attachment 218785

That’s exactly the answer I was looking for.
Thank you! No reason to spend $$ for technical numbers.
One more question is if x3700h will be able to comfortably drive speakers with nominal 4ohm? Nominal means as a starting point? I doubt I’ll ever use past 100w/ch.
I’ve heard denon run hot. Will 4ohm speaker be a concern?

 

[delta76]

That’s exactly the answer I was looking for.
Thank you! No reason to spend $$ for technical numbers.
One more question is if x3700h will be able to comfortably drive speakers with nominal 4ohm? Nominal means as a starting point? I doubt I’ll ever use past 100w/ch.
I’ve heard denon run hot. Will 4ohm speaker be a concern?

It should. Nominal means usually, but impedance will dip at certain frequencies. Higher power means bigger dip. It should be fine unless you play very very loud

 

[delta76]

Also regarding heat, just make sure you have enough space. If not, think about active cooling, a 140mm quiet fan should do just fine. 3700h can withstand quite high temp but as always, lower temp means longer lifespan