Denon AVR-X3800H Review

[Rockman2]

Makes it harder to answer with certainty...

Martin Logan are also known for some of their 4 ohm speakers (stats) dropping down to 1 ohm at the high end (!!)

So the question is, are they a "true" 4 ohm - staying above 3 ohm throughout the frequency range - in which case you should be fine

Or, are they the sort of "4 ohm" speaker that drops down to below 2 ohm (like a number of ML designs, and my own Gallo's) - if it is the latter, then the answer might be try them, but be ready to potentially require external amps.

You don't need a heap of power for surround - but if they are low impedance, even 30W @2ohm might be a stretch from an AVR. (Luckily an X4 has a better power supply than an X3... so you might be OK...)

I read about the amp section in the Cinema 50 was discrete amps for each channel and a bigger power supply. Don't know if that would help with my surrounds being 4 ohm. But again I am new at the technical side maybe the 4800 has the same thing. But you thought would be to go with the 4800?

 

[dlaloum]

I read about the amp section in the Cinema 50 was discrete amps for each channel and a bigger power supply. Don't know if that would help with my surrounds being 4 ohm. But again I am new at the technical side maybe the 4800 has the same thing. But you thought would be to go with the 4800?

If you have "difficult" speakers then every little bit helps! - the larger power supply in the X4 series would help

I would argue that the "value" bargain at the moment is the X4700 as it has the same larger power supply, and Auro3D - the only thing it misses out on is the multisub integration that the X4800 has.... also at the moment there are bargains to be had, with X4700's selling for less than X3800's

My gut feeling is go for the larger power supply and the bargain of the moment, unless you are running multisubs - in which case the newer X4800 will probably be the go-to.

Personally I am waiting for Onkyo to play its hand - they are due to release their RZ70/RZ90 flagships - which should compete against the X4800. And I have a bias against Audyssey, after about 14 years of disappointment with it, and being very happy with my current Dirac based setup. (yes it was an older version of Audyssey, I'm sure the updated version in the current models does better, and the Denon's were on my shortlist.... )

 

[Rockman2]

If you have "difficult" speakers then every little bit helps! - the larger power supply in the X4 series would help

I would argue that the "value" bargain at the moment is the X4700 as it has the same larger power supply, and Auro3D - the only thing it misses out on is the multisub integration that the X4800 has.... also at the moment there are bargains to be had, with X4700's selling for less than X3800's

My gut feeling is go for the larger power supply and the bargain of the moment, unless you are running multisubs - in which case the newer X4800 will probably be the go-to.

Personally I am waiting for Onkyo to play its hand - they are due to release their RZ70/RZ90 flagships - which should compete against the X4800. And I have a bias against Audyssey, after about 14 years of disappointment with it, and being very happy with my current Dirac based setup. (yes it was an older version of Audyssey, I'm sure the updated version in the current models does better, and the Denon's were on my shortlist.... )

Thanks for the info. One more thing I did not mention is these surround speakers are pretty efficient at 91db if that makes any difference. I do have 2 subs but they are also martin logan and have ARC room correction so I can adjust them individually. I also had an Onkyo TXNR 5010 that was great just did not pas 4k so I sold it for the Marantz sr8015. But the Onkyo sounded just as good as the 8015

 

[Fidji]

Makes it harder to answer with certainty...

Martin Logan are also known for some of their 4 ohm speakers (stats) dropping down to 1 ohm at the high end (!!)

So the question is, are they a "true" 4 ohm - staying above 3 ohm throughout the frequency range - in which case you should be fine

Or, are they the sort of "4 ohm" speaker that drops down to below 2 ohm (like a number of ML designs, and my own Gallo's) - if it is the latter, then the answer might be try them, but be ready to potentially require external amps.

You don't need a heap of power for surround - but if they are low impedance, even 30W @2ohm might be a stretch from an AVR. (Luckily an X4 has a better power supply than an X3... so you might be OK...)

Impedance decrease with frequency is valid only for electrostatic speakers of the range and is driven by the principles of their construction. On top it is gradual decrease, that is not extremely demanding on the amp. Of course you do not want to have 20W tube amp connected, but there are lot of people connecting Benchmark, Hypex and Purifi amps to them. What also helps, that the bigger ones have active bass section, so even easier for the amp.

I personally drive them at mid-gain setting of Purifi amp, and it is more than enough of power.

Motions are just normal AMT speakers, they could come from any other manufacturer as well.

I read about the amp section in the Cinema 50 was discrete amps for each channel and a bigger power supply. Don't know if that would help with my surrounds being 4 ohm. But again I am new at the technical side maybe the 4800 has the same thing. But you thought would be to go with the 4800?

Those CI/HT Motion Series MartinLogan speakers are absolutely benign, relatively high sensitivity - AMT + 8in woofer. If you properly Bass Manage them, e.g. at 80Hz I do not see any potential issue driving even lot of them with AVR.

 

[dlaloum]

Thanks for the info. One more thing I did not mention is these surround speakers are pretty efficient at 91db if that makes any difference. I do have 2 subs but they are also martin logan and have ARC room correction so I can adjust them individually. I also had an Onkyo TXNR 5010 that was great just did not pas 4k so I sold it for the Marantz sr8015. But the Onkyo sounded just as good as the 8015

Ahh - My Integra DTR 70.4 was of the same generation, the integra version of your 5010 I believe - but it let out its DSP "magic smoke" ... I still have it, and would happily repair it if the replacement board was reasonably priced (pending.... chances are they will ask me for more than the whole unit is worth.... in which case it will go to landfill - sadly)

I opted for the current "baby" model the Integra DRX3.4 - and when I tried it with my low impedance speakers, it didn't sound good - but with a set of external amps driving L/C/R it sounds just as good as the old 70.4.... and it handles my surrounds just fine. - It has less channels than the RZ50 or the X3800/X4800, but my current setup is still 5.1 - and Dirac was a definite step up on Audysey on the old AVR. (immediately noticeable! much clearer dialog and midrange)

Yes the speakers being efficient (91db) - reduces the load on the amp... it is a good thing!

Yeah I would aim for a X4xxx or RZ50 just so you have the extra current in the PSU... you may not need it.... but it is nice to have!
On the other hand if you have a dealer that can provide a trial where you can return/swap it over - then try the X3xxx (or Integra DRX3.4) and see whether the less powerful models can handle the load ....

Given you don't need the sub integration as you already have that within the subs themselves - I would go for the Onkyo/Integra AVR's with the built in Dirac licence - mostly because I have been impressed with its performance vs Audyssey. (on the other hand, if you want Auro... then Denon it is... or Maybe Marantz)

 

[Fidji]

Thanks for the info. One more thing I did not mention is these surround speakers are pretty efficient at 91db if that makes any difference. I do have 2 subs but they are also martin logan and have ARC room correction so I can adjust them individually. I also had an Onkyo TXNR 5010 that was great just did not pas 4k so I sold it for the Marantz sr8015. But the Onkyo sounded just as good as the 8015

As an advice - if you have Dynamos [I assume] - skip ARC and let Audyssey do its work, even more so, if you have more of them.
ARC will try to adjust each sub individually, AUdyssey then perform SUBEQ and you might end up with suboptimal result with lot of distortion.

 

[Brambo67]

Can anyone explain?
I mentioned already I don't particularly like Sound United's choice to implement lower grade material (DAC) in more expensive boxes. Even though it offers upgradeability to DIRAC (thé non-plus-ultra when it comes to room calibration according to many forum members), 4 subwoofer outputs, etc. etc. Still, 'we' all agree the older Denon AVC-X8500h is a perfect example how an AVR should perform.

Now I don't know much about the impact of DAC when the signal finally has reached the speaker but if 'm correct the DAC sits before the amp in the chain, which of course is before the speaker and the cable connecting to the speaker. The 8500, when tested by Amir showed this as amp-performance:

What I understand from it is the signal, which arrived from the exquisit AKM AK4490 DAC with SINAD no less than 102 dB (!) for Channel 1, is being amplified to send to the speaker (which ever that may be) with SINAD of a little below 82 dB.

Underneath we have the 3800:

It's Channel 1 Digital to Analog Conversion (DAC) is very frustratingly poor with SINAD of just below 87 dB. When this gets amplified, the signal which is ready for the speaker to bring to me, the listener (> 50 yrs old, for my age considerably good hearing, still as proven in test, but of course not able to hear what I could say 20 or even 30 years ago) it is done so with 86dB for Channel 1. This is a 'whopping' 4dB better than the highly praised 8500 can do. So I'm getting kinda lost here... I feel I need to get rid of my faultless x4700PCM and sell before anyone in this world knows to stay away from Denon and Marantz PCM DAC receivers, but at the same time there is no alternative... Of course I can buy the only available 2nd offer in Europe for a 3600 (with the same AMP-impact on the signal) or spend like € 20k on Tinnov, Storm and some amps.

Am I getting this completely wrong or is at the end of the line in this example the signal from the 3800 less distorted compared to the 8500 before it hits the speaker?

 

[delta76]

Can anyone explain?
I mentioned already I don't particularly like Sound United's choice to implement lower grade material (DAC) in more expensive boxes. Even though it offers upgradeability to DIRAC (thé non-plus-ultra when it comes to room calibration according to many forum members), 4 subwoofer outputs, etc. etc. Still, 'we' all agree the older Denon AVC-X8500h is a perfect example how an AVR should perform.

Now I don't know much about the impact of DAC when the signal finally has reached the speaker but if 'm correct the DAC sits before the amp in the chain, which of course is before the speaker and the cable connecting to the speaker. The 8500, when tested by Amir showed this as amp-performance:
View attachment 241089
What I understand from it is the signal, which arrived from the exquisit AKM AK4490 DAC with SINAD no less than 102 dB (!) for Channel 1, is being amplified to send to the speaker (which ever that may be) with SINAD of a little below 82 dB.

Underneath we have the 3800:
View attachment 241096

It's Channel 1 Digital to Analog Conversion (DAC) is very frustratingly poor with SINAD of just below 87 dB. When this gets amplified, the signal which is ready for the speaker to bring to me, the listener (> 50 yrs old, for my age considerably good hearing, still as proven in test, but of course not able to hear what I could say 20 or even 30 years ago) it is done so with 86dB for Channel 1. This is a 'whopping' 4dB better than the highly praised 8500 can do. So I'm getting kinda lost here... I feel I need to get rid of my faultless x4700PCM and sell before anyone in this world knows to stay away from Denon and Marantz PCM DAC receivers, but at the same time there is no alternative... Of course I can buy the only available 2nd offer in Europe for a 3600 (with the same AMP-impact on the signal) or spend like € 20k on Tinnov, Storm and some amps.

Am I getting this completely wrong or is at the end of the line in this example the signal from the 3800 less distorted compared to the 8500 before it hits the speaker?

That's testing amp only (analog input, bypassing the DAC).

 

[Brambo67]

That's testing amp only (analog input, bypassing the DAC).

Yes, that I understand. But ultimately isn't this the max. SINAD performance of the amp, no matter if the input is coming from DAC ór analog (external) input? So in other words, isn't the amp the bottleneck anyway? Or will the amp-performance improve when the signal is coming from the DAC?

 

[pbc]

Interesting read. I’m currently using an Anthem AVM60 with an external amp to power my pair of on wall surrounds and 4 atmos in ceiling speakers. My mains are powered Seaton Catalysts. Currently dual DIY subwoofers.

Have been thinking about moving back to an AVR to get rid of the amp and ideally either go directly to DIRAC or have an upgrade path to it. Considered the NAD T778 but the unit has numerous issues from what I’ve read and was hoping the 3800 would be “it”.

But reading this thread has me concerned … and I’m not sure if it should? Especially if I’m using the unit only to power surrounds, or would the distortion at high volumes also impact my mains because it’s the DAC that is the issue vs the amp? These numbers are confusing. At this point just thinkin mg I should stick to the Anthem AVM60.

 

[delta76]

Yes, that I understand. But ultimately isn't this the max. SINAD performance of the amp, no matter if the input is coming from DAC ór analog (external) input? So in other words, isn't the amp the bottleneck anyway? Or will the amp-performance improve when the signal is coming from the DAC?

many people would use external amp with preout. That is one selling point of x3xxx and up. X2xxx and lower do not have preout

 

[CyrusTheGreat_600BC]

So everyone that thinks noise is inaudible, here’s a post from @mike7877 from a year ago (#21 and #25) on the following thread, and he only used his ears and the volume control without any other equipment to estimate SINAD of 3700v2 to be 86db! That’s incredibly accurate and close to 3800 measurements that use the same DAC:

New Denon 3700/4700 DAC chip

So you couldn't even level match? I have a scope so could match levels, but it's overkill when the other steps are done by ear. Comparing noise, the difference between my DAC and the 3700's is notable. Playing silence through the 3700's DAC with the receiver's volume at 0db, noting the noise...

www.audiosciencereview.com

So Amir is right. Trained ears can detect the lower quality DACs.

 

[voodooless]

So everyone that thinks noise is inaudible, here’s a post from @mike7877 from a year ago (#21 and #25) on the following thread, and he only used his ears and the volume control without any other equipment to estimate SINAD of 3700v2 to be 86db! That’s incredibly accurate and close to 3800 measurements that use the same DAC:

New Denon 3700/4700 DAC chip

So you couldn't even level match? I have a scope so could match levels, but it's overkill when the other steps are done by ear. Comparing noise, the difference between my DAC and the 3700's is notable. Playing silence through the 3700's DAC with the receiver's volume at 0db, noting the noise...

www.audiosciencereview.com

So Amir is right. Trained ears can detect the lower quality DACs.

He may be right, but that "test" is no proof of that. Those numbers are pure luck basically. Only listening to the noise, without knowing the gain structures involved is not really the way to go. It's guesswork at best. And even then, you only looking at SNR, not distortion.

The 86 wasn't very hard to guess anyway. The chip does 93, and there is some volume control and also some gain circuits (output is almost 4V vs 2.1 of DAC), so let's lose about 6dB due to this and some other cheap engineering... Voila, perfect score

 

[Brambo67]

many people would use external amp with preout. That is one selling point of x3xxx and up. X2xxx and lower do not have preout

But most people don't. They just buy this AVC in order not having to buy external amps. Still you are right the theoretical signal would be less distorted in an external amp option. However probably speakers and room conditions would have way bigger impact. I guess this whole thing is about how much of an audiophile person you consider yourself to be. I guess I belong to the 'Pragmatic audiophiles'; 'we want to pay optimal and not maximal, don't like (worse) spec's to be reason for discussion but at the end can accept an AVR like the 3800.....

 

[delta76]

But most people don't. They just buy this AVC in order not having to buy external amps. Still you are right the theoretical signal would be less distorted in an external amp option. However probably speakers and room conditions would have way bigger impact. I guess this whole thing is about how much of an audiophile person you consider yourself to be. I guess I belong to the 'Pragmatic audiophiles'; 'we want to pay optimal and not maximal, don't like (worse) spec's to be reason for discussion but at the end can accept an AVR like the 3800.....

uhm no. people who spent $5000 buying x8500h/a usually spent money on a lot of amps as well. especially in home theatre setup.

people powering their LR with external amp is even more common. Having a good dac makes sense.

 

[oivavoi]

So everyone that thinks noise is inaudible, here’s a post from @mike7877 from a year ago (#21 and #25) on the following thread, and he only used his ears and the volume control without any other equipment to estimate SINAD of 3700v2 to be 86db! That’s incredibly accurate and close to 3800 measurements that use the same DAC:

New Denon 3700/4700 DAC chip

So you couldn't even level match? I have a scope so could match levels, but it's overkill when the other steps are done by ear. Comparing noise, the difference between my DAC and the 3700's is notable. Playing silence through the 3700's DAC with the receiver's volume at 0db, noting the noise...

www.audiosciencereview.com

So Amir is right. Trained ears can detect the lower quality DACs.

I don't think anybody has said that noise is inaudible? The whole point (at least for me) is that some forms of distortion or noise are close to inaudible, whereas others are audible indeed. Lower-order harmonic distortion is very hard to detect once it gets down to 50 or 60 db below the signal. Heck, even 20, 30 or 40 db can be hard to hear with music. Some forms of noise, on the other hand, is fairly easy to hear. Higher-order harmonic distortion or IMD can also be fairly easy to pick out. So one device with a benign SINAD profile of 60 db can be completely transparent to the human ear, whereas a device with a non-benign SINAD profile of 80 db can be non-transparent. Now if SINAD numbers get high enough, things stop being audible altogether, of course.

That said, proper gain staging is usually a smart thing to do in order to get noise to inaudible levels. That often means to have much gain early in the chain, and to lower the gain as much as possible in the final amplifier section. If one does the opposite - i.e. lowers the gain from the dac and increases the gain from the final power amp - it's not difficult to end up with audible hiss.

 

[Bren Derlin]

Allowed by whom? They say "jump", you say "how high"? That it?
Feel free to be their victim. Or their pray...

You should probably never buy anything from an electronics company again, so you're not a victim falling to their prey... Or go ahead and spend fists-full of dollars to ensure your getting the absolute best quality available (maybe)...

Edit: ...and you're getting all bent out of shape over a measured product that was never actually used in a real world environment. ...That's like missing HALF the science equation. ...then there's "peer reviewed and tested" to verify the findings. So far, none of that has actually taken place. For that, I absolutely LOVE Amir's speaker tests. He sits and listens to the speakers to verify the measurements, and adds the subjectivity comments to see if matches up with the objectivity measurements.

 

[birdog1960]

Thanks for the info. One more thing I did not mention is these surround speakers are pretty efficient at 91db if that makes any difference. I do have 2 subs but they are also martin logan and have ARC room correction so I can adjust them individually. I also had an Onkyo TXNR 5010 that was great just did not pas 4k so I sold it for the Marantz sr8015. But the Onkyo sounded just as good as the 8015

Not sure how muck you are gaining in sound using those high end, power eating speakers as surrounds. You could likely sell them and get some easier to push but perfectly adequate small surrounds and have money left over with no perceptible drop in movie performance. If you are playing music in all channel surround it might be an issue. fwiw, I'm using a 3500 to push 4 ohm BMR's as fronts with Definitive tech 8 ohm surrounds with more than enough power to fill a huge room and minimal (to my ears) distortion/noise at the highest volumes I need.

 

[peng]

Again hope this is not a dumb question but. I will be running the FCR with the ATI 200 wpc amp so would the Cinama 50 at 110 wpc be ok for the 4 surround speakers? Or would it make more sence to wait for a 4800 at 125 wpc to run the 4 surround speakers? Thanks

110 W vs 125 W makes little difference, that's about 0.5 dB difference that people can barely notice.

Whether it would be ok for 4 surround speakers will depend on the distance, speaker sensitivity and the spl you desired. In order to know for sure you should use an online calculator to figure it out. Keep in mind that even the 200 W ATI amp can only give you just under 3 dB more spl, all else being equal, though taking loads off the AVR is always a good thing, and external power amps typically can handle speakers with low impedance and/or high phase angles better, to much better..

 

[peng]

FYI I have already shipped this unit to its owner so no more tests.

Keep that in mind though (up to 3 V in non preamp mode, and please try again measure with an external amp or simulated load connected, when you measure the next one, hopefully the X4800H or the Cinema 50.