Denon AVR-X3800H Review

[peng]

The clipping myth just doesn’t die for preamp mode. I can’t even count how often you had to explain this in various threads. Thx.

I can also confirm, yes the SINAD is highest with ca 100dB at 1.4V and then slightly drops by ca 5dB all the way to 4Vrms. Yes my x3700 easily puts out 4Vrms in pre amp mode only (measured it myself).

Would I like the SiNAD to stay at 100 yes but the drop to ca 95dB isn’t clipping.

The only person who can put this to rest is Amir, I tried but failed. He stuck to his gun, defining "clip" as when SINAD starts to drop precipitously. He is technically not incorrect but practically speaking we should not define clipping that way because "drop precipitously" is relative. What seems like a steep slop can look flatter simply by changing the x axis scale.

Actually, in the following example, not for you but for @Beershaun who should be able to understand my point more easily using this example:

Here Amir said it was nice, "Doesn't clip until 4.2 volts":

Think about the following facts:
1) Marantz NR1510/slimline series do not have preamp mode, no HDAMs either, so why it is not "clip" at 4.2 V instead of 1.4 V?
2) See that the Red curve dropped "precipitously" at about 1.0 V to about 76 dB SINAD just like those Denon AVR did at about 2.0 V in non preamp mode, why?
3) Now take a look of the xy axis scale he use!!
4) Now look back on the Denon AVR's SINAD vs Output V graph, and notice that if Amir had done the same measurement but to 5 V like he did on the NR1510, you can see that at about 2 V, the SINAD vs V slope would become much gentler and may even level off like it did on the NR1210. There is no reason why it would behave differently, considering the parts and circuitry of D+M's AVRs basically have the same preamp/dac block diagram.

I don't have one for the 1510 but here's the one for the 1508, that's the same for other Marantz/Denon AVRs except the NR series and Denon don't have HDAMs in the signal path:

Same dac and volume chip for the preamp/dac blocks! So again, why would the NR1510 "clip at 4.2 V, while its bigger brothers "clip" at 1.4 V? Answer: It's Amir's choice, he's the boss, I can only voice my opinion.

 

[peng]

Which Anthem did you buy and which Denon did you replace and why? Well thank you.

I traded it in the AVR-X4400H for the AVM70, because, thanks to an ASR member who posted the clearance deal Anthem offered back in Nov/or Dec on the discontinued 4K version, and my nearby dealer offered very reasonable trade-in value for my 4 years old Denon. Other than the good deal, I have no other reasons.

I think that deal is off now, likely sold out.

 

[Beershaun]

The only person who can put this to rest is Amir, I tried but failed. He stuck to his gun, defining "clip" as when SINAD starts to drop precipitously. He is technically not incorrect but practically speaking we should not define clipping that way because "drop precipitously" is relative. What seems like a steep slop can look flatter simply by changing the x axis scale.

Actually, in the following example, not for you but for @Beershaun who should be able to understand my point more easily using this example:

Here Amir said it was nice, "Doesn't clip until 4.2 volts":

Think about the following facts:
1) Marantz NR1510/slimline series do not have preamp mode, no HDAMs either, so why it is not "clip" at 4.2 V instead of 1.4 V?
2) See that the Red curve dropped "precipitously" at about 1.0 V to about 76 dB SINAD just like those Denon AVR did at about 2.0 V in non preamp mode, why?
3) Now take a look of the xy axis scale he use!!
4) Now look back on the Denon AVR's SINAD vs Output V graph, and notice that if Amir had done the same measurement but to 5 V like he did on the NR1510, you can see that at about 2 V, the SINAD vs V slope would become much gentler and may even level off like it did on the NR1210. There is no reason why it would behave differently, considering the parts and circuitry of D+M's AVRs basically have the same preamp/dac block diagram.

I don't have one for the 1510 but here's the one for the 1508, that's the same for other Marantz/Denon AVRs except the NR series and Denon don't have HDAMs in the signal path:

Same dac and volume chip for the preamp/dac blocks! So again, why would the NR1510 "clip at 4.2 V, while its bigger brothers "clip" at 1.4 V? Answer: It's Amir's choice, he's the boss, I can only voice my opinion.

View attachment 261976

Thanks for the time to explain this to me and help me better understand what is going on. I now understand you are correct that it's not clipping and the full operation range is 0-4v. In that case, my question then becomes is that significant drop off after 1.4v acceptable or not? I know your answer is, 75db is still "good enough " but this discontinuity in the curve indicates a significant change that people using this in mixed mode should be aware of and work around. Meaning ensuring they have a high gain amp and are staying below that 1.4v in their normal operating range. Also, there is a separate discussion here about what should be "good enough" for a hifi DAC/preout sinad and at what voltage. Amirs benchmark is 2v for his tests. this two step curve doesn't look like it should meet our quality bar. People should rule out using our favorite high measuring low gain amps like the Benchmark and Ncx500 for example.

Comparing the Denon and marantz amps above shows that they are using the same design choice across their products. Our Anthems, for example, do not exhibit this behavior and are much more linear in their output range.

So I'm on board with you that these can output the full voltage range, and these are good options for full preamp mode. I think people should not use this in a mixed mode with say a 3ch external amps for example, unless they are comfortable figuring out how to set up their volume levels to keep the output below 1.4v.

Regarding my original contention about subwoofers:

It could be Audessey vs ARC, but then however Denon implemented Audessey is getting it wrong and my Denon still does not drive my subwoofers as well as my Anthem.

 

[Beershaun]

It's worth noting that 1.4V (RMS) is the Pro "standard" - so for example, the Crown XLS power amps designed for the pro band and PA markets, are designed to achieve their max rated output with 1.4V.

And consumer/domestic use equipment has traditionally required no more than 1V for max rated output (with many components requiring only 0.7V... and higher gain amps eg:Quad requiring only 0.5V!! )

So yes.... 1.4V can be a limitation.... but only for a specific subset of the market.

I have a number of Quad 606 amps which require only 0.7V...
I have a pair of first generation Crown XLS2500 amps, which require 1.4V.
(and I did check before purchasing my current AVR, to make sure it could put out 1.4V)

Why would this be a problem !?

It's worth noting that 1.4V (RMS) is the Pro "standard" - so for example, the Crown XLS power amps designed for the pro band and PA markets, are designed to achieve their max rated output with 1.4V.

And consumer/domestic use equipment has traditionally required no more than 1V for max rated output (with many components requiring only 0.7V... and higher gain amps eg:Quad requiring only 0.5V!! )

So yes.... 1.4V can be a limitation.... but only for a specific subset of the market.

I have a number of Quad 606 amps which require only 0.7V...
I have a pair of first generation Crown XLS2500 amps, which require 1.4V.
(and I did check before purchasing my current AVR, to make sure it could put out 1.4V)

Why would this be a problem !?!

This would be an issue for lower gain amps like the Benchmark and NCx500 no?

 

[peng]

This would be an issue for lower gain amps like the Benchmark and NCx500 no?

Yes and no, not always. The fact is, whether 1.4 V, or even 1.0 V is an issue, depends on not just the gain of the power amp you use, but also the rated output.

For example, the Benchmark AHB2 you mentioned, has the following gain/input sensitivity specs:

INPUT SENSITIVITY​

• Low-Gain = 22 dBu (9.8 Vrms), Gain = 9.2 dB
• Mid-Gain = 14.2 dBu (4 Vrms), Gain = 17.0 dB
• High-Gain = 8.2 dBu (2 Vrms), Gain = 23 dB
• Use Mid-Gain or High-Gain settings for unbalanced inputs
• Unbalanced inputs require RCA to XLRM adapter cables

So if you use the High-Gain setting, it is still quite low, at 23 dB, but it only need 2 Vrms to drive it to its rated output, that's because its rated output is only 100 W. If you use it with an amp that also has 23 dB gain, but is rated 400 W, then you will need 4 Vrms to drive it to its rated 400 W output. But you still only need 2 Vrms if you only drive the amp to output 100 W. You can use the formula to calculate the V vs output for any given gain.
The site linked below has all the formula you need:

Gain factor dB conversion to factor ratio gain conversion to level in decibels dB amplitude voltage gain field quantity dBu dBm dBW dBV dBmV dBµV dBu energy size power voltage damping factor attenuation amplitude amplification acoustic intensity soun

gain factor ratio gain conversion to level in decibels dB amplitude voltage gain conversion to factor field quantity dBu dBm dBW dBV dBmV dBµV dBu energy size power voltage damping attenuation amplitude amplification acoustic intensity sound pressure factor ratio voltage gain powerloss level...

www.sengpielaudio.com

For the Hypex amps that typically have 25 to 26 dB gain, say 26 dB:

Preout voltage required.......................................Output power into 8 ohm, gain = 26 dB....................... Gain = 23 dB

0.5V.............................................................................. 12.44 W....................................................................................... 6.24 W
1.0 V............................................................................. 49.76 W....................................................................................... 24.94 W
1.5 V............................................................................. 112 W........................................................................................... 56.12 W
2.0 V............................................................................. 199 W .......................................................................................... 99.76 W

Double the voltage will get you 4X the output power.

And you are right, if one wants to have >76 dB SINAD pre out to drive the Benchmark amp, don't use the Denon AVR in non preamp mode.

However, that is also untrue because the Denon AVR-X3600H and higher models have the amp assign mode that allows you to disconnect the front left and right channels. That's how the AVR-X3600H, that has no preamp mode, was able to achieve almost 100 dB SINAD (as good as any Anthem AVRs/AVPs) in Amir's review. So those Denon AVRs can drive the Benchmark amps, if you are happy with the preout SINAD >90 dB.

 

[peng]

Regarding my original contention about subwoofers:

It could be Audessey vs ARC, but then however Denon implemented Audessey is getting it wrong and my Denon still does not drive my subwoofers as well as my Anthem.

As I said before, it might have been a setting issue on your AVR, which Denon AVR you had before? I suppose like me, you no longer have the Denon, otherwise we can troubleshoot some more. As it is now, I don't know you have that trouble, that as far as I know you may be among the very few that had experience such issue.

Obviously, something was not right, but it had nothing to do with the pre out being too low. If it was, there would have been numerous Denon and Marantz owners complaining.

Edit: just remember one other thing, Anthem's volume range is -90 to +10, Denon's -81 to +18 (going by memory), so at the same volume setting, your Anthem will sound louder, the subout to your subwoofer input will have a higher voltage. That may be the difference you were experiencing.

 

[HarmonicTHD]

The only person who can put this to rest is Amir, I tried but failed. He stuck to his gun, defining "clip" as when SINAD starts to drop precipitously. He is technically not incorrect but practically speaking we should not define clipping that way because "drop precipitously" is relative. What seems like a steep slop can look flatter simply by changing the x axis scale.

Actually, in the following example, not for you but for @Beershaun who should be able to understand my point more easily using this example:

Here Amir said it was nice, "Doesn't clip until 4.2 volts":

Think about the following facts:
1) Marantz NR1510/slimline series do not have preamp mode, no HDAMs either, so why it is not "clip" at 4.2 V instead of 1.4 V?
2) See that the Red curve dropped "precipitously" at about 1.0 V to about 76 dB SINAD just like those Denon AVR did at about 2.0 V in non preamp mode, why?
3) Now take a look of the xy axis scale he use!!
4) Now look back on the Denon AVR's SINAD vs Output V graph, and notice that if Amir had done the same measurement but to 5 V like he did on the NR1510, you can see that at about 2 V, the SINAD vs V slope would become much gentler and may even level off like it did on the NR1210. There is no reason why it would behave differently, considering the parts and circuitry of D+M's AVRs basically have the same preamp/dac block diagram.

I don't have one for the 1510 but here's the one for the 1508, that's the same for other Marantz/Denon AVRs except the NR series and Denon don't have HDAMs in the signal path:

Same dac and volume chip for the preamp/dac blocks! So again, why would the NR1510 "clip at 4.2 V, while its bigger brothers "clip" at 1.4 V? Answer: It's Amir's choice, he's the boss, I can only voice my opinion.

View attachment 261976

Oh I am converted and fully with you.

I also would not call a drop of ca 5dB from ca 100 to ca 95ish “precipitously”

The logic is inconsistent and I would rather trust Don here:

Clipping 101

Amplifier clipping has been cited as the cause of audible distortion, destruction of speakers (especially tweeters), and total annihilation of the known universe. More or less. In fact, clipping does cause distortion that adds significant high-frequency content not in the original signal, and...

www.audiosciencereview.com

 

[peng]

Oh I am converted and fully with you.

I also would not call a drop of ca 5dB from ca 100 to ca 95ish “precipitously”

The logic is inconsistent and I would rather trust Don here:

Clipping 101

Amplifier clipping has been cited as the cause of audible distortion, destruction of speakers (especially tweeters), and total annihilation of the known universe. More or less. In fact, clipping does cause distortion that adds significant high-frequency content not in the original signal, and...

www.audiosciencereview.com

It's not a drop to 95 dB though, but to 75 dB at 2 V, in non preamp mode. Again, 75 dB at 2 V is inline with the Marantz AV7705, and avrs such as the SR7015 that did 79 dB at 2 V even in preamp mode.

I am with Don on that too.

 

[HarmonicTHD]

It's not a drop to 95 dB though, but to 75 dB at 2 V, in non preamp mode. Again, 75 dB at 2 V is inline with the Marantz AV7705, and avrs such as the SR7015 that did 79 dB at 2 V even in preamp mode.

I am with Don on that too.

My initial answer always referred to Denon x3700 preamp only (as this is what I measured at home). Ca 96 at 2V. Ok, maybe as Amir didn’t go to 4V, it is lower than 95 at 4.
Still I would not call a drop from 101 to 96 precipitously either - but I think we agree here anyhow.

 

[peng]

My initial answer always referred to Denon x3700 preamp only. Like here. Ca 96 at 2V. Ok, maybe as Amir didn’t go to 4V, it is lower than 95 at 4.

View attachment 262053

That's preamp mode. Amir's 1.4V clip referred to non preamp mode, so it is the green curve. About 76 dB in this case, at 2 V and you can that the drop started to level off at about 1.7 V, that's clearly not "clipping" would behave.

 

[HarmonicTHD]

That's preamp mode. Amir's 1.4V clip referred to non preamp mode.

But didn’t the OP refer to preamp mode too?
At least that’s what I read from this, see quote. But on second thought it also could have been understood differently… sorry if that was the case here.
“So if Denons preamp stage is designed to drive it's amplifiers with an output range of 0 to 1.4v”

 

[peng]

But didn’t the OP refer to preamp mode too?

I thought he was talking about non preamp mode He mentioned mixed use case, meaning when some of the internal amps will be use, so preamp mode will not work in those cases.

 

[HarmonicTHD]

I thought he was talking about non preamp mode He mentioned mixed use case, meaning when some of the internal amps will be use, so preamp mode will not work in those cases.

We both got confused probably. See my edit above. Oh well. Never mind .

As always, I appreciate your thoughts and input. Thx and have great weekend.

 

[jbonn5000]

Its been up on the US site for a while now:
https://www.denon.com/en-us/product/av-receivers/avr-x4800h

Unless I'm getting confused, that's a different product than the AVC (monolith amp modules, made in Japan etc)

ETA: they are probably the same product.

 

[dlaloum]

This would be an issue for lower gain amps like the Benchmark and NCx500 no?

I would suggest that the "odd ones out" are the lower gain amps - audio preamps have for decades been relatively standardised around the 0.7V domestic and 1.4V pro/balanced levels - so why are these newcomers imposing higher V requirements for their input?

Basically they need an input buffer board to raise their gain... this imposes noise and distortion overheads, and probably makes their specs less attractive - bringing them back into line with their many competitors.

 

[rvsixer]

Unless I'm getting confused, that's a different product than the AVC (monolith amp modules, made in Japan etc)

ETA: they are probably the same product.

AVR = AVC + tuner

 

[rvsixer]

Refurb x3800 (and x4800) at Denon USA site. Kind of a bargain imo.

 

[peng]

I have to wonder why the X4800H just came out and they already have refurbished ones.

 

[rvsixer]

I have to wonder why the X4800H just came out and they already have refurbished ones.

Happens with all new products...refurb 3800's were available about 2 weeks after US release. My last two AVR's were both refurbs with no issues, so that's pretty much all I buy.

 

[peng]

Happens with all new products...refurb 3800's were available about 2 weeks after US release. My last two AVR's were both refurbs with no issues, so that's pretty much all I buy.

Probably are just returns from the buyer's remorse thing, as opposed to repaired, hopefully.