Review and Measurements of Schiit Yggdrasil V2 DAC

[Krunok]

Please, the argument I'm making about bias is a basic tenet of psychological research that is accepted by all psychologists and scientists who work with subjective data. It is 100% uncontroversial.

Heh.. I can't believe there are still folks willing to dispute this..

 

[zalive]

I don't agree, to me it sounds pretty much the same. If your claim is to be valid it should be tested with the same recording in the controlled environment, otherwise it is just subjectve impression.

Kruno we seem to be from the same town, I'm in Zagreb as well
Which equipment do you use? If it's a very good system (for example a DAC as good as BADA Alpha plus amplification of a similar level of performance), I can understand closeness to realism.

 

[andreasmaaan]

Well it will take a while with 170 pages. I'm not sure it is the one I was thinking of either. I believe the one I had in mind was only about op amps and not capacitors.

Thanks. No pressure, yeh it's a fucking long read

 

[zalive]

In the example you give, my subjective opinion is not good enough (it is not fully reliable; I may be mistaken).

Rather, it is merely the best information available on which to form a judgement. Relying on the information is my only option.

In audio, on the other hand, there is better information available than my own unavoidably biased perceptions obtained through sighted tests.

Please, the argument I'm making about bias is a basic tenet of psychological research that is accepted by all psychologists and scientists who work with subjective data. It is 100% uncontroversial.

I agree with you, but you still need to rely on your vision in a given situation. And if you felt clear reasons why you cannot trust your eyes sufficiently for the purpose, you'd probably abandon driving the car.

In audio, enjoyment comes through ears, not through looking to graphs. That might give me some enjoyment too, but it's a different kind of enjoyment

 

[andreasmaaan]

I agree with you, but you still need to rely on your vision in a given situation. And if you felt clear reasons why you cannot trust your eyes sufficiently for the purpose, you'd probably abandon driving the car.

In audio, enjoyment comes through ears, not through looking to graphs. That might give me some enjoyment too, but it's a different kind of enjoyment

The data are ofc no substitute for the experience of listening (the two are completely different things). And, while reading and understanding the measurements and studies might be enjoyable for some, that is not their purpose.

Their purposes are:

• for designers, to know how their measurements are likely to translate into subjective experiences of their devices
• for recording and mastering engineers, to know how their decisions are likely to translate into subjective experiences of their recordings
• for consumers, to know how published measurements are likely to translate into their experiences of their systems, and to efficiently and reliably identify problems or limitations in their systems that they might be hearing

Among other things...

 

[andreasmaaan]

And I should add @zalive: for consumers, another purpose of science and measurement is to help them understand when their subjective experiences are likely to be the product of something material occurring at an electronic or acoustic level in their systems, or likely to be the product of their own psychology

 

[Krunok]

Kruno we seem to be from the same town, I'm in Zagreb as well
Which equipment do you use? If it's a very good system (for example a DAC as good as BADA Alpha plus amplification of a similar level of performance), I can understand closeness to realism.

Hi!

I'm using Topping D10 DAC. Although it is a cheap device in my opinion it is completely transparent hence no need for more expensive device. I'm using Volumio running on RPI to play music from NAS server. D10 is connected to Rotel RMB-1075 amp and Audio Aero Capitol amp (Class A 50W tube amp). I also have Audio Aero DAC/preamp which I feed via coax SPDIF from D10. My speakers are Castle Harlech S2.

 

[zalive]

• for consumers, to know how published measurements are likely to translate into their experiences of their systems, and to efficiently and reliably identify problems or limitations in their systems that they might be hearing

Well it's not established at all whether measurements will translate well into the listening experience.
This is particularly what I put in question, and what I would like to confirm with double blind listening tests. How the objective measuring as done today translates to listening experience.

 

[Krunok]

Well it's not established at all whether measurements will translate well into the listening experience.
This is particularly what I put in question, and what I would like to confirm with double blind listening tests. How the objective measuring as done today translates to listening experience.

For electronics it traslates extremely well. Loudspeakers may be a little tricky..

 

[andreasmaaan]

Well it's not established at all whether measurements will translate well into the listening experience.
This is particularly what I put in question, and what I would like to confirm with double blind listening tests. How the objective measuring as done today translates to listening experience.

Have you read the body of research I referred to in my first reply to your post? If not, I'm sorry to say that I don't think you're qualified to say "it's not established at all"

 

[zalive]

Have you read the body of research I referred to in my first reply to your post? If not, I'm sorry to say that I don't think you're qualified to say "it's not established at all"

But you still don't disagree that it would be a good thing to do double listening blind tests to approve or disprove the objective measurement evaluation of a sound quality?
The way I see it, it's objectivists who have more to lose if such tests should disprove their interpretation of objective results

 

[andreasmaaan]

But you still don't disagree that it would be a good thing to do double listening blind tests to approve or disprove the objective measurement evaluation of a sound quality?
The way I see it, it's objectivists who have more to lose if such tests should disprove their interpretation of objective results

In this case, yes, because I think the Schiit product's measurements suggest that it is about on the cusp of audibility thresholds, whereas the Topping D50 (for example) is all but guaranteed to be transparent. In other words there is some chance the test could go either way in this case, which makes it interesting.

But I don't see how this leaves an objectivist with anything to lose?

 

[andreasmaaan]

Lol, sorry

Objectivists would have a lot more to lose from a double blind test of two devices that should fall clearly in the green circle on my diagram.

Although note that an unexpected result in such a test would not refute objectivism itself; it would merely cast doubt on the thresholds that the research had established, and/or point to the possibility that the measurements that led us to place the devices in the green circle in the first place were not comprehensive enough.

 

[Krunok]

Although note that an unexpected result in such a test would not refute objectivism itself; it would merely cast doubt on the thresholds that the research had established, and/or point to the possibility that the measurements that led us to place the devices in the green circle in the first place were not comprehensive enough.

IMO the thresholds, or better to say huge lack of them, are the biggest problem of the objectivists, because if they exist then we won't be calling certain measurement figures "issues" as they would be so far below thresholds and thus not worth mention of.

 

[andreasmaaan]

IMO the thresholds, or better to say huge lack of them, are the biggest problem of the objectivists, because if they exist then we won't be calling certain measurement figures "issues" as they would be so far below thresholds and thus not worth mention of.

I wouldn't say there is a lack of thresholds as such, just that the precise lines are not very clear (hence the grey area in my ridiculous Venn diagram).

 

[Krunok]

I wouldn't say there is a lack of thresholds as such, just that the precise lines are not very clear (hence the grey area in my ridiculous Venn diagram).

Sure. But If we agreed that practically no one can hear THD of less than 0.1% of the 1kHz single tone (not to mention that such distortion in some music would be masked, so it would be much harder to hear) why are we still discussing some measured artifacts that are happening in the range of 0.00X% or even less?

 

[andreasmaaan]

Sure. But If we agreed that practically no one can hear THD of less than 0.1% of the 1kHz single tone (not to mention that such distortion in some music would be masked, so it would be much harder to hear) why are we still discussing some measured artifacts that are happening in the range of 0.00X% or even less?

I agree. I think in light of a few studies (e.g. the one I posted in response to Blumlein's question before), there is evidence to say that slightly less than 0.01% might be audible with some material. The main problem in this particular case is that THD is a crap metric in the first place, with low correlation to listener experience. So I suggest placing the benchmark a little lower (0.001% for example would be extremely safe).

I suspect that in circumstances where distortion in the range of 0.1% has been shown to be audible, the type of distortion was of great significance, e.g. something resembling crossover distortion with lots of high order content. So probably not the type of distortion produced by a DAC. But this is just speculation...

 

[andreasmaaan]

There have also of course been cases where particular kinds of THD in the range of 10%+ have not been audible. Hence my saying that THD itself is a crap metric.

 

[Krunok]

I agree. I think in light of a few studies (e.g. the one I posted in response to Blumlein's question before), there is evidence to say that slightly less than 0.01% might be audible with some material. The main problem in this particular case is that THD is a crap metric in the first place, with low correlation to listener experience. So I suggest placing the benchmark a little lower (0.001% for example would be extremely safe).

If that is the case wouldn't the loudspeakers, and even the best of them are having distortion in the range of 0.0X% to 0.X%, mask the distortion of the DAC and amp?

 

[andreasmaaan]

If that is the case wouldn't the loudspeakers, and even the best of them are having distortion in the range of 0.0X% to 0.X%, mask the distortion of the DAC and amp?

Certainly to an extent, but if a listener can discern between e.g. 0.1% of a particular kind of electronically added distortion and 0.001% of electronically added distortion in a properly controlled test, then we have to conclude that whetever distortion the speakers/headphones introduced was not sufficient to fully mask the electronically added 0.1% distortion.

In my estimation, the most likely reason this might be the case is that the electronically added distortion is similar to crossover distortion, i.e. it affects low level components of the material more than high level components, meaning it does the opposite to what the speakers are doing, or that the electronically added distortion contains significant high-order components (speakers tend to produce predominantly low order harmonic distortion).