• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Denon Replaces AKM AK4458 DAC IC in X4700H and X6700H

The 3700 or 3xxx range not sold in Japan. Drop to 2700 or jump to 4700, 4500. Disappointing since 3700 offers MultEq xt32 as well as other features in 4700 but not quite as expensive. Guess Denon wants folks to choose up or down for margins maybe leave mid tier to Marantz ? Anyway sucks for Japan consumers.
 
Keep in mind not everyone would actually crank the volume high enough to exceed the sweet spot of around 1.4 to 1.5 V preamp output.
At below the 1.4-1.5 V range, SINAD will still be above 95 dB without disconnecting the signal line of the preamp connections to the internal power amps.

You don't know that. The issues are numerous in an AVR, especially one with a single transformer. Even though the digital supplies may be off separate windings with their own local regulators, the power stage rails can easily cross modulate and/or cause noise on the digital supply rails.

AVRs are a disaster for high performance 2 channel. They always have been.
 
You don't know that. The issues are numerous in an AVR, especially one with a single transformer. Even though the digital supplies may be off separate windings with their own local regulators, the power stage rails can easily cross modulate and/or cause noise on the digital supply rails.

AVRs are a disaster for high performance 2 channel. They always have been.
That's a valid assumption, but in this specific case, measurements seem to indicate otherwise.

index.php
 
Last edited:
You don't know that. The issues are numerous in an AVR, especially one with a single transformer. Even though the digital supplies may be off separate windings with their own local regulators, the power stage rails can easily cross modulate and/or cause noise on the digital supply rails.

AVRs are a disaster for high performance 2 channel. They always have been.

I do know that based on ASR's measurements.:)
 
That's a valid assumption, but in this specific case, measurements seem to indicate otherwise.

index.php


Some people just don't like AVRs, sorry if I am stating the obvious.:D By the way, I think Amir's "clip" is a little ambiguous (not wrong just ambiguous, subject to interpretation imo. I believe what actually happen is, as Gene explained in one of his video (on the X3600H iirc) that it is actually the power amp that would begin to clip when the pre out that drives it reaches above 1.2 V, so by the time it reaches 1.4 V, the internal power amp would be well into clipping and the distortions/noise got fed back to the input that would obviously then affect the pre out signal because they are connected.

So again, if you break that connection point using amp assign to the front left/right channels, or in preamp mode then pre out will remain cleaner until it reaches the clipping point, and that would be closer to 4 V.
 
Last edited:
I do know that based on ASR's measurements.

OK. Let's throw this out there:

1 channel driven, 1W@8R and measure the peformance on the line stage (the D/A converter through to RCA/line out)
2 channels driven
4 channels driven
etc.

And then, the same tests at rated power. Amir uses 5 watts. Another plucked out of the air number that means absolutely nothing.

The "tests" are meaningless as they stand.

As for the "clipping" point, I don't know how many times we all have to bring it up. Pick a number. It's really simple. Draw a line in the sand. Make it 0.1%, 1% whatever, but for FFS, stop trying to put crosshairs on an arbitrary point on an FFT that varies from product to product. It just makes everyone look stupid.
 
Well, at least this will stop my lust after a DENON x3700 or x4700 upgrade, I will stick with my AK4458 based Sony receiver for the foreseeable future. :)
 
OK. Let's throw this out there:

1 channel driven, 1W@8R and measure the peformance on the line stage (the D/A converter through to RCA/line out)
2 channels driven
4 channels driven
etc.

And then, the same tests at rated power. Amir uses 5 watts. Another plucked out of the air number that means absolutely nothing.

The "tests" are meaningless as they stand.

As for the "clipping" point, I don't know how many times we all have to bring it up. Pick a number. It's really simple. Draw a line in the sand. Make it 0.1%, 1% whatever, but for FFS, stop trying to put crosshairs on an arbitrary point on an FFT that varies from product to product. It just makes everyone look stupid.

I am not sure what you are talking about, power amps? No arguments there, but my post was about pre out voltage vs SINAD. Macfox knows, he posted the graph in post#43 As for the "clipping" point, I agree with you but you should talk to Amir about it.:)

By the way, some of those AVRs pre out measured better in SINAD at 1 kHz, (better in some other criteria too) than some integrated amps and AV preamp processors. That's factual whether you like it or not.
 
So the 3700 4700 gets the pcm5102 and the 8500 the es9010
 
Just opened the cover again of my new X4700.

I removed the first board to be able to see the DACs.

Any help how to find them? The rear side of the first board shows no DACs, also the upper side of the 2nd board (which is brown colour and connected with the analog video rca) contains nearly nothing ( just 7 caps and a lot of junctions).

Are they on the 3rd board (with all analogue inputs and pre-outs)?

Don't want to keep the AVR open for days :)
 
Just opened the cover again of my new X4700.

I removed the first board to be able to see the DACs.

Any help how to find them? The rear side of the first board shows no DACs, also the upper side of the 2nd board (which is brown colour and connected with the analog video rca) contains nearly nothing ( just 7 caps and a lot of junctions).

Are they on the 3rd board (with all analogue inputs and pre-outs)?

Don't want to keep the AVR open for days :)

Without the help of a good service manual, even if the board is exposed, the DAC ICs (should be 2 of them) could be on the side that is not visible to you. In that case the picture of the board might at least help you identify it, and go from there.

New or old, the boards look the same (that's assuming the X4500H has the same one as the original X4700H, both are based on the AK4458). You cannot see the DAC's ID in the pictures but you can clearly see the layouts are quite different.

Both pictures below are from Denon.jp:
AVR-X4700H | 9.2chプレミアムAVサラウンドレシーバー | Denon公式
AVR-X4500H | Denon公式

NewDAC.jpg



OldDAC.jpg
 
Ok, found the DACs.

They are not on the 2nd board underneath the 1st one. There is a vertically fixed board between all horizontal boards and the power-amp section.
After removing the 1st horizontal board you can clearly see the DACs on vertical one.

At least for my 4700, they are still AK4458 (2 pieces).

ak4458-2.png
 
Ok, found the DACs.

They are not on the 2nd board underneath the 1st one. There is a vertically fixed board between all horizontal boards and the power-amp section.
After removing the 1st horizontal board you can clearly see the DACs on vertical one.

At least for my 4700, they are still AK4458 (2 pieces).

View attachment 128496
How deep did you dig to find it?
 
Ok, found the DACs.

They are not on the 2nd board underneath the 1st one. There is a vertically fixed board between all horizontal boards and the power-amp section.
After removing the 1st horizontal board you can clearly see the DACs on vertical one.

At least for my 4700, they are still AK4458 (2 pieces).

View attachment 128496

That's great! The AKM IC is a known quantity.
 
From the photos of the new board, I can barely see the pin layout of the two main DAC ICs. Based on that, I would say the ES9006 and the AK4468VN are the likely candidates. Somewhere the AKM factory fire was mentioned, so it is possible that inventory of the old and very popular Ak4458VN might have been depleted to the point D+M would be wised to switch to the AK4468 that is a successor of the 4458 so production should be at least ongoing, whereas AKM may not bother produce the old IC. Just guessing.., but seems logical to me.:D

If it is the AK4468, then I am not concerned and may still consider D+M if and when I am ready.., and would still want to see one or two Denon that has the new ICs measured first.
 
I just had a look at the service manual of my Marantz SR-6010. It actually already contains two versions of the DAC boards, one with AK4458, the other one with PCM1690. Looks like the PCM is for the SR-6010, and AK for SR7010. SINAD difference is quite big between the two. It's probably not the only series of models to have this kind of setup and Denon will be similar in this regard as well, so it should be easy for them to just exchange the DAC board(s) for another. Most likely it was already designed long ago.
 
I just had a look at the service manual of my Marantz SR-6010. It actually already contains two versions of the DAC boards, one with AK4458, the other one with PCM1690. Looks like the PCM is for the SR-6010, and AK for SR7010. SINAD difference is quite big between the two. It's probably not the only series of models to have this kind of setup and Denon will be similar in this regard as well, so it should be easy for them to just exchange the DAC board(s) for another. Most likely it was already designed long ago.

As you alluded to, the SR6010 has the PCM1690, even the SR6011 has the PCM1690 according to the service manual.
It looked like Marantz did not standardize on the AK4458 until the model year SR6012 and SR7010, AV7702A (the 7702 has the PCM1690 too). Same on the Denon side in terms of model year vs DAC IC used.

It would appear that D+M and Yamaha would let people know (most of the time I think) when they launched new models in their marketing material. Others like Anthem seemed lacking transparency and you would hard time tracking down any service manuals. D+M and Yamaha's owner's manual, product info sheets also typically provide much more details than their more expensive "boutique" brands competitors. I may be wrong, that's just my impression, from reading many product info sheets and manuals of gear I have owned, and not owned, yet..:)
 
Last edited:
You don't know that. The issues are numerous in an AVR, especially one with a single transformer. Even though the digital supplies may be off separate windings with their own local regulators, the power stage rails can easily cross modulate and/or cause noise on the digital supply rails.

AVRs are a disaster for high performance 2 channel. They always have been.

What do You mean by that?
I am figuring out if I Denon 3700h will be audible worse than Purifi amp. Data and other people indicates it will be not. Can You explain Your point of view?
 
You don't know that. The issues are numerous in an AVR, especially one with a single transformer. Even though the digital supplies may be off separate windings with their own local regulators, the power stage rails can easily cross modulate and/or cause noise on the digital supply rails.

AVRs are a disaster for high performance 2 channel. They always have been.

Major generalisation without any hard evidence, based on assumptions about the effects of theoretical problems that don't always lead to measurable and/or audible problems.

What do You mean by that?
I am figuring out if I Denon 3700h will be audible worse than Purifi amp. Data and other people indicates it will be not. Can You explain Your point of view?
Do you prefer measurements or assumptions? Interpretation of the first tells you the answer is 'probably no'. The post you quoted seems to go by the second. Based on the second, lots of people will tell you there must be a difference, because of theoretical differences. One can buy an expensive amplifier based on those motivations and be perfectly happy with it. Nothing wrong with that. Just don't test these assumptions in a double blind setup or prepare to be disappointed.
 
Back
Top Bottom