Denafrips ARES II USB R2R DAC Review

[ethanhallbeyer]

i have heard m400 and ares ii and comapred to my chi fi xiang sheng dac-05a dual mono burr brown balanced dac i found the former 2 both to be really good dacs.. m400 sounds great but the ares ii, to me, is the better dac based on my preferences. hard to say either is 3x as good as my $250 xiang sheng, but if you are as fixated on exacting max ROI, i think they are definitely the better dacs, as they should be for the price difference.

in terms of build quality and heft the ares ii wins hands down.

 

[BDWoody]

I wanted to know for certain which DAC offered a sound I could live with, hence the many DACs I bought

How did you go about comparing them?

 

[VintageFlanker]

i think they are definitely the better dacs, as they should be for the price difference.

That last part has been proven wrong a thousand times there....

 

[decoRyder]

I've had the opposite experience

What proved your assumption to be wrong?

There's just a lot more variables to consider than I thought possible. In the end it's really about the implementation as a whole, rather than any subset of features.

 

[decoRyder]

How did you go about comparing them?

By spending a lot of time plugging them into my reference system, and listening to the same tracks over and over again; I level-match the volume, and I think I do a halfway decent job of it.
A tedious process, but I think I learned a lot from it. The Ares II has something that my other Dacs do not have - which is that 'plasticity' for lack of a better word. I hate the term 'holographic', because I think it's overused, but that kinda describes it. Having said that I wouldn't use the Ares II on a set of 'bright' speakers, since the Ares II will add to any sibiliance inherent in the recording, or the system.

 

[BDWoody]

I level-match the volume

By voltmeter?

Was any of your testing done blind?

If not, it is almost guaranteed you (as a human being...nothing personal) will hear differences.

You were right at first... The DAC really is the least of your worries, at least in terms of SQ. There just isn't anything hiding in the signals on the way to the amp that's not getting measured to add up to audible differences (with music, in a normal room).

Without careful listening controls, people can end up chasing rainbows, and end up with 14 DAC's before they know it!

 

[decoRyder]

By voltmeter?

Was any of your testing done blind?

If not, it is almost guaranteed you (as a human being...nothing personal) will hear differences.

You were right at first... The DAC really is the least of your worries, at least in terms of SQ. There just isn't anything hiding in the signals on the way to the amp that's not getting measured to add up to audible differences (with music, in a normal room).

Without careful listening controls, people can end up chasing rainbows, and end up with 14 DAC's before they know it!

... without 14 DACs people can end up talking about facts which they have not actually experienced and verified

I use REW to adjust the volume. No blind testing. Confirmation bias does not apply, different DACs simply sound different. Some more, some less. For example, I can't tell ANY difference between the Topping D90 and the SMSL M400 using balanced outputs, but I can tell the difference between the Ares II and the Topping D90.

I think that the actual topology, particularly the output stage is often overlooked. Clearly, if DACs measure differently they can also sound different.

 

[BDWoody]

. No blind testing. Confirmation bias does not apply

I'm not sure that's how it works...or why you don't believe you would be subject to bias.

think that the actual topology, particularly the output stage is often overlooked.

What does that mean in terms of what gets measured? How is it overlooked?

Measuring differently, and being audibly distinguishable are two different things.

Anyway, glad you like your Denafrips!

 

[decoRyder]

I'm not sure that's how it works...or why you don't believe you would be subject to bias.



What does that mean in terms of what gets measured? How is it overlooked?

Measuring differently, and being audibly distinguishable are two different things.

Anyway, glad you like your Denafrips!

By nature, and profession I am trained to eliminate confirmation bias; in my field, falling prey to CB, or any prejudice will get me fired. In fact, that 'sword of damokles' hangs over my head every day; I have to analyze tons of metrics and statistical data, and if I make a wrong decision, i get fired - that simple. So in this case, I'd like to think my approach is PoE, even though it's admittedly not very scientific, or rigorous in terms of being reproducible Then again it doesn't have to - all I want is a DAC I can live with.

Output stage - I think that's part of the disconnect here - the part of bits being converted, we can all agree - irelevant for the actuals 'sound' such as it is - d/s, multibit, r2r - they can all do an efficient job doing that. But the signal chain doesn't stop there, does it? The decoded signal will pass through an analog output stage in the DAC, and that output stage will be dramatically different across a wide range of DACs. It's this analog ouput stage that is the part that will affect the sound, and yes - make the same decoded bits sound different from one DAC to another.

I don't 'like' the Ares II - I can listen to it without having an urge to turn it off, that's about it

 

[AnalogSteph]

The decoded signal will pass through an analog output stage in the DAC, and that output stage will be dramatically different across a wide range of DACs. It's this analog ouput stage that is the part that will affect the sound, and yes - make the same decoded bits sound different from one DAC to another.

*give measurably different performance from one DAC to another.

No more than that in a vast majority of cases, fortunately. Thankfully, human hearing is way more forgiving than measurement setups.

Those with a propensity towards problematic levels of high-frequency IM distortion should not be used with a very slow rolloff filter, perhaps. And peak levels >0 dBFS are best avoided in general.

We haven gotten whole device classes like room correction DSPs and budget active speakers with DSP crossovers because DACs are generally good enough. Like this, combating audiophilia nervosa has allowed addressing the elephant in the room, quite literally so.

 

[VintageFlanker]

Output stage - I think that's part of the disconnect here - the part of bits being converted, we can all agree - irelevant for the actuals 'sound' such as it is - d/s, multibit, r2r - they can all do an efficient job doing that. But the signal chain doesn't stop there, does it? The decoded signal will pass through an analog output stage in the DAC, and that output stage will be dramatically different across a wide range of DACs. It's this analog ouput stage that is the part that will affect the sound, and yes - make the same decoded bits sound different from one DAC to another.

Seriously, why this... again ? Amir is measuring DACs. That includes of course their output stages... you can't bypass those when measuring through analog outputs. When people say there's no audible differences between two transparent (measuring good enough) DACs, that does of course consider the performance out of the output stage.

I don't 'like' the Ares II - I can listen to it without having an urge to turn it off, that's about it

That's the best thing to hear. Enjoy your music and keep it as long as you can, then!

PS: You can't avoid blind tests just because you don't believe in your own bias. That is precisely... some bias.

 

[solderdude]

AFAIK the Ares does not have an output stage. Just a simple lowpass at a high frequency.
That's why it has an unusually high output resistance.

 

[Sukie]

By nature, and profession I am trained to eliminate confirmation bias;

I don't know which psychologists you've been reading but you cannot "eliminate" confirmation bias. You can be trained to be aware of it and that can mitigate against its impact. But that's not the same as eliminating it. It would be impossible to eliminate it as you have no base to compare against.

Unless...
you use measurable data!

Confirmation bias is excluded because measurable data exists outside of your own experience.

... without 14 DACs people can end up talking about facts which they have not actually experienced and verified .

Facts do not need to be experienced. The facts surrounding DACs (pardon the rhyme) can be verified through data/measurements.

 

[BDWoody]

I have to analyze tons of metrics and statistical data, and if I make a wrong decision, i get fired

What's that got to do with performing a blind test? You believe you are immune to bias...?

But the signal chain doesn't stop there, does it? The decoded signal will pass through an analog output stage in the DAC

Which is what get measured. Amir doesn't open the box, pry out the chip and measure it...he takes the output at the output, where anyone who uses it would.

I don't 'like' the Ares II - I can listen to it without having an urge to turn it off, that's about it

Well, then I'm sorry you don't like your DAC. Maybe some more testing comparisons and you can find one you like.

 

[Killingbeans]

There's just a lot more variables to consider than I thought possible. In the end it's really about the implementation as a whole, rather than any subset of features.

Can't really see how that answers my question

Postulating an explanation for your experience is not the same as providing proof.

 

[decoRyder]

Can't really see how that answers my question

Postulating an explanation for your experience is not the same as providing proof.

No offense mate, but I don't see how I'm obligated to provide you anything.

 

[VintageFlanker]

No offense mate, but I don't see how I'm obligated to provide you anything.

No, you're not, actually.

But don't claim anything barely scientific, then.

 

[decoRyder]

No, you're not, actually. But don't claim anything, then.

I didn't make any claims beyond that it a : ) measures well, and b : ) I prefer this DAC over my other Dacs.

 

[Killingbeans]

No offense mate, but I don't see how I'm obligated to provide you anything.

Non taken. And you're right.

But you did mention having proven you own assumption wrong, and I was just curious as to what the proof was.

I think you have already explained your testing method in previous posts, and I'll just take the conclusions you've made from them with a serious grain of salt. Best thing we can do at this point is probably to agree to disagree

I didn't make any claims beyond that it a : ) measures well, and b : ) I prefer this DAC to my other Dacs.

I think the claim VintageFlanker is referring to is this one:

It took me too long to figure out that the DAC is probably the most significant component in the signal chain when playing digital audio. I used to think any halfway decent DAC would be sufficient - that assumption turned out to be very wrong : )

 

[decoRyder]

Non taken. And you're right.

But you did mention having proven you own assumption wrong, and I was just curious as to what the proof was.

I think you have already explained your testing method in previous posts, and I'll just take the conclusions you've made from them with a serious grain of salt. Best thing we can do at this point is probably to agree to disagree

I think the claim VintageFlanker is referring to is this one:

Appreciate the reply Ok, I do not make any claim that what I'm about to say has any validity outside my own personal experience and preferences. Before I decided to try a wide range of DACs, I ran the Topping D30 for about 2 years. The D30 measures well, and for the money it's an absolute bargain, like most Topping offerings. So then, around xmas last year, I got the Wharfedale Diamond 11.5's. Great speakers, but I could not listen to them for extended periods of time, most recordings sounded outright shrill, and harsh. Before the 11.5's I was running a pair of Zu Audio Omen DW II's which due to their eccentric design do have the advantage of being able to play a wide variety of recordings without my ears falling off. Nothing 'wrong' with the Wharfedales in comparison to the Zu's, but the Wharfdales are definitely more resolving, and will do a much better job at producing the recording, without any obvious coloration, or emphasis of any particular frequency range. The Zu's just kind of 'mask' harsh treble, particularly on poor recordings.

Ok, so I have a decent range of amps of all classes - Class A, A/B, Class D - some DYI, and in terms of brands I own amps by Rotel, Oddyssey Audio, Schiit, Emotiva and others. Tried them all, still couldn't get to the point where I could listen to the Wharfedales and find the sound 'enjoyable'. I even tried tone controls, using a Schiit Loki, but there was still this 'harsh' edge that I just couldn't eliminate from the sound. Next I ordered a pair of JBL Studio 590's, thinking that for whatever reason, I just couldn't warm up to the Wharfedales. Well - I'm really annoyed at this point - the same thing happened - it still sounded harsh, shrill and 'etched' - no difference between the two sets of speakers in that regard. I really didn't expect that the DAC would make any difference, the D30 measures well, and DACs aren't supposed have a 'sound' - but at some point I decided to try a different DAC - some old no-name 'Ambery' brand DAC I had flying about in the garage, with some old Burr-Brown Chip, I don't remember which one. Well, imagine my surprise when this ancient, but well-built DAC sounded considerably better than the D30. Oh yes, some of the detail was lost, but what was there sounded considerably 'warmer' and fuller in the midrange, while making the treble bearable. So, while the D30 measures well, I don't like the way it sounds on a set of 'revealing' speakers.

I have funds, what I don't have is much time, for anything beyond work. That's why rather than selecting just one or two DACs I ordered a whole bunch of them, and tested them over a period of several months. I like the Ares II, the SMSL M400, the Topping D90, the Audio GD R1, and one DYI DAC based on dual AD1865R NOS chips - in that order. I particularly don't like the Topping D30, the Topping D10, and the Topping D50s - on anything other than the ZU Omen DW II's. I'm still missing the RME ADI-2 DAC FS, which I'll get soon.

... and that's it really - using 'revealing' speakers - some DACs enable me to be able to listen without having an urge to turn the music off after a short period of time : )