Review and Measurements of Schiit Yggdrasil V2 DAC

[HiFidFan]

Ok thanks for clearing that up! I don't know why reviewers say that. One person said the Yiggy was superior in layered depth in comparison to Chord Hugo 2 which was comparatively 2 dimensional!

Sure. . .

 

[amirm]

Ok thanks for clearing that up! I don't know why reviewers say that. One person said the Yiggy was superior in layered depth in comparison to Chord Hugo 2 which was comparatively 2 dimensional!

They say that because it is fashion. And they think by commenting on it makes them a good listener and knowledgeable. The reality is the opposite of course. It is all in their imagination.

 

[Kw6]

They say that because it is fashion. And they think by commenting on it makes them a good listener and knowledgeable. The reality is the opposite of course. It is all in their imagination.

Thanks for clearing that up Amir! I kinda figured as user opinions are sometimes not all consistent and the sound they do hear are most likely due to other factors like speakers and acoustics!

 

[goldenpiggy]

I meant "Fixed EQ" as in DACs with built-in "sound signature" which you cannot change, or those so-called remastered discs that has certain frequency ranged turned up higher, or overall compressed in general. Imagine applying EQ on Roon and then you go through some DAC that adds its own sound signature. The source I meant is the source device that converts digital signal to analog signal.

Depends on implementation of the EQ, sometimes I prefer to do it in the analog domain. There are some professional processors that does the EQ in DSP without taking into account of the extra headroom needed, and caused hard clipping. Roon seems to have implemented it correctly so I do use it sometimes for my bedroom setup.

Regarding professional loudspeaker processor/DSPs, I typically apply EQ and whatever other processing (e.g. delay) on the OUTPUT side of the processor, but this has nothing to do with processor headroom or clipping. Nearly all these processors use floating point architecture or fixed point with 48 or 56 bit internal mix bus -- you're unlikely to clip it even if you max out every darn PEQ in the signal chain (before output level).

On the very rare occasion I use up all the PEQs on the output, I'll start using the PEQ on the input side.

And while on the subject of EQ, IMO there's absolutely nothing wrong with using them to get the sound you like. Just know that nothing is free: with every PEQ you use or an ever-steeper crossover slope, you're introducing phase shift or group delay. It's the nature of the beast, analog or digital.

 

[Alec Kinnear]

They say that because it is fashion. And they think by commenting on it makes them a good listener and knowledgeable. The reality is the opposite of course. It is all in their imagination.

That's actually not what you said Amir. What you said was:

Also, frequency response in 1 to 3 kHz is a large determinant of this effect. Take any headphone you have and boost this region and listen for spatial qualities. Likely you hear the effect (ignore the tonality difference -- just listen for externalization of the sound).

Hence a DAC or a player whose built-in processing slightly favours certain frequency response or slightly sweetens the sound can indeed create more or less of a soundstage effect. These differences are audible but not all listeners are reliable or have correctly set up the comparisons. As you have correctly pointed out, comparative listening must be done under identical circumstances and with near instant switchover (8 seconds is too long).

If all measurably "audibly transparent" DACs, pre-amps and amplifiers sounded the same everyone here would just have bought a $100 Chinese DAC two years ago, attached it directly to powered monoblocks and gone home. You may wish to visit the Audirvana thread. Audirvana is your player of choice. By Damien Plisson's admission and audibly, Audirvana is a euphonic player which sweetens the sound.

It's more than plausible that Schiit have subtly sweetened the sound of this Ygg. I don't know. I don't have it and would not buy it, as I prefer gear which 1. includes proper safety features (excludes most of Topping/SMSL and other such gear and older Schiit) 2. measures at least adequately well. 3. I'm based in Europe and buying Schiit is asking for trouble here and we have our own good manufacturers like RME whose ADI-2 DAC FS costs about half of what it does in the United States and includes full warranty (inverse of the Schiit situation). Any manufacturer who chooses to add euphonic qualities to his gear should do so after meeting an adequate and clear technical standard. On the other hand, this belief that the measurements done at ASR in any way fully characterise the sound of audio equipment is just myopia.

Meeting food safety standards does not a gourmet meal make. A gourmet meal (great sound) can be made in a dirty kitchen, risking making diners sick (great sounding audio gear with either safety issues or technical performance issues).

 

[goldenpiggy]

That's actually not what you said Amir. What you said was:
(excludes most of Topping/SMSL and other such gear and older Schiit).

Could you please elaborate on this? Do you mean specific regulatory approval, lack of internal fuse, etc.?

 

[Jimbob54]

That's actually not what you said Amir. What you said was:



Hence a DAC or a player whose built-in processing slightly favours certain frequency response or slightly sweetens the sound can indeed create more or less of a soundstage effect. These differences are audible but not all listeners are reliable or have correctly set up the comparisons. As you have correctly pointed out, comparative listening must be done under identical circumstances and with near instant switchover (8 seconds is too long).

If all measurably "audibly transparent" DACs, pre-amps and amplifiers sounded the same everyone here would just have bought a $100 Chinese DAC two years ago, attached it directly to powered monoblocks and gone home. You may wish to visit the Audirvana thread. Audirvana is your player of choice. By Damien Plisson's admission and audibly, Audirvana is a euphonic player which sweetens the sound.

It's more than plausible that Schiit have subtly sweetened the sound of this Ygg. I don't know. I don't have it and would not buy it, as I prefer gear which 1. includes proper safety features (excludes most of Topping/SMSL and other such gear and older Schiit) 2. measures at least adequately well. 3. I'm based in Europe and buying Schiit is asking for trouble here and we have our own good manufacturers like RME whose ADI-2 DAC FS costs about half of what it does in the United States and includes full warranty (inverse of the Schiit situation). Any manufacturer who chooses to add euphonic qualities to his gear should do so after meeting an adequate and clear technical standard. On the other hand, this belief that the measurements done at ASR in any way fully characterise the sound of audio equipment is just myopia.

Meeting food safety standards does not a gourmet meal make. A gourmet meal (great sound) can be made in a dirty kitchen, risking making diners sick (great sounding audio gear with either safety issues or technical performance issues).

So audirvana is both bit perfect and also not?

As I'm pretty sure I posted you a link to their site yesterday that claimed it was bit perfect.

 

[Alec Kinnear]

Could you please elaborate on this? Do you mean specific regulatory approval, lack of internal fuse, etc.?

Topping devices blow up, take headphones with them (L30). If you plug in the wrong voltage, they melt (Amir has done this to a D50 himself). Before you start shouting that the wrong voltage should never be plugged in, mature manufacturers take into account such issues and put in a safety triggers. First the device samples the current. If the voltage is outside of what it can handle the device shuts off.

Even Topping added voltage protection to the D50s after Amir's experience. Which is a good thing as both the D50s and A50s use exactly the same 2.1mm power jack, with completely different voltages (D50s is 5v, A50s is 15v) When I plugged in the wrong plug, the D50s display gives a wrong power warning and shuts down. The A50s does not have this kind of protection. If you plug in the wrong voltage, it will immediately melt.

There are other issues with the A90 and D90 which apparently have significantly more noise on the cinch/RCA/unbalanced outputs and inputs than on the balanced ones. I had been considering an A90 as a pre-amp but between the L30 meltdowns, the issues with power supplies instantly melting a device, poor warranty service outside of the Paypal 60 day refund window (service equals "mail it back to China with false customs declartion"!) and quality issues with unbalanced input/output, I decided not to.

Fortunately there are manufacturers who are very stringent about safety features. Yamaha is one of them (I've got an R-S700 which behaves very well though I don't like the sound and I'm replacing it). In the DAC world, RME who make professional gear, build very robust and safety conscious gear and that's the path I've chosen, though I suspect on just sound alone, certainly headphone amp and probably pre-amp, when it's working, I might prefer an A90.

I keep my audio equipment for decades, so reliability/robustness matters very much to me. Similar approach to cameras which makes me prefer Nikon and previously Canon over Sony. Sony offers great features and new models twice/year but build quality has never been great. Their photo gear is engineered to fail within two to five years, Sony counts on the photographer to moved on to other gear by that time.

 

[Alec Kinnear]

So audirvana is both bit perfect and also not?

As I'm pretty sure I posted you a link to their site yesterday that claimed it was bit perfect.

Read the Audirvana thread carefully please and respond there if you could. I only mentioned Audirvana here as a reference about whether and when euphonic/sweetened sound is acceptable.

 

[Jimbob54]

Read the Audirvana thread carefully please and respond there if you could.

Kindly give us a link to that audirvana support page rather than linking your pasted extract here.

 

[Alec Kinnear]

Kindly give us a link to that audirvana support page rather than linking your pasted extract here.

Jim, I'm responding to Audirvana questions in the Audible difference in players? (Audirvana, JRiver, Roon, MusicBee, etc.) thread. There's no point to duplicating the discussion.

 

[goldenpiggy]

Too late now but maybe ANSI, ISO, ETSI, etc. should have created a standard for matching external power supply plug size/geometry to voltage range (for both DC and AC output power supplies)?

 

[amirm]

Hence a DAC or a player whose built-in processing slightly favours certain frequency response or slightly sweetens the sound can indeed create more or less of a soundstage effect.

I have never, ever seen a DAC create a dip in 1 to 3 kHz region. My comment there was regarding speakers and headphones where frequency response is variable. With DACs, that explanation is not relevant due to their rule flat frequency response.

 

[Alec Kinnear]

I have never, ever seen a DAC create a dip in 1 to 3 kHz region. My comment there was regarding speakers and headphones where frequency response is variable. With DACs, that explanation is not relevant due to their rule flat frequency response.

Thanks for the clarification. My own tests show that good modern DACs measure almost identical but still show significant differences when listening. While some of these heard differences have their origins in psychoacoustics, some are real (otherwise we'd all be listening to Topping's excellent budget D30 DAC with inexpensive monoblocks). We need better and more revealing tests to account for these perceptual differences.

In the absence of such tests, this Ygg may genuinely sound better to most humans than better measuring DACs (solution proposed at the end of this post).

I've recently run through a whole pile of amplifiers which tested very well in your tests (amp section of Yamaha R-S700, NAD C320BEE, ones which tested less well Behringer A500, Marantz SR7001). There are huge differences in sound quality, with the Behringer doing much, much better in subjective tests than it possibly should in the field. Yes, there is hiss and crackle on near silent passages at high volume, but when music with a strong signal is playing, the A500 often sounds better than the rest (probably thanks to its enormous power reserves). The pre-amp tests which also included an NAD C372 (amp section long dead, preamp still works). As well as the variable output of the DACs (not recommended after testing, proper pre-amping with extra power does make a difference), there was a Topping A50s (excellent balanced headphone sound, excellent pre-amp, relatively better than its stablemate D50s), a NAD C165BEE (very, very good pre-amp but takes up a lot of space, very similar to the pre-amp in the C372) and a Pro-Ject Pre Box Analogue DS2 (great for rock, didn't like at all for classical/vocal) and now an RME ADI-2 DAC FS (compact, flexible, configurable, but a small step below the C165BEE and Pre Box as a pre-amp on congested passages). Worst pre-amp of all was the Yamaha R-S700, which sterilises and deadens sound while still measuring adequately. I was able to do these tests simultaneously and/or back to back as all the gear was here at the same time. I won't be able to do this again as almost all of it will be shipping out.

You have enormous stamina to go through so much gear every week. When people ask for you to put gear back in to run just one more test, I shudder. Thankfully you often refuse these days.

I correlated your measured findings with listening tests on as much of this gear as possible. The measurements provide a rough guide to performance but there were many cases where relatively poor measurers sonically outperformed measurement champs. To bring audio down to four or five measurements is a horrific simplification of sound, music and the capabilities of our ears. We need better and more revealing tests.

With DACs, a simple and significant test would be to diff output. Even subtle differences in sound cannot easily hide from a well-run diff as it's not a partial measurement but a full measurement. It's all the information which we hear. Diff files could work for pre-amps as well. What to do to better measure amplifiers, I have no idea.

 

[amirm]

In the absence of such tests, this Ygg may genuinely sound better to most humans than better measuring DACs (solution proposed at the end of this post).

I performed controlled testing of the Yggy against Topping DAC which the last owner said he had compared. He even gave me the exact pieces of music that replicated his experience. I tested that and there was no difference at all in controlled AB testing. Everything we know about audio research backs my experience. And and predicts that someone would think the Yggy sounds better even though in reality it does not.

With DACs, a simple and significant test would be to diff output. Even subtle differences in sound cannot easily hide from a well-run diff as it's not a partial measurement but a full measurement.

Such tests are difficult to do since due to noise and clock drift, the output differs from run to run on the same DAC let alone different ones. We have tools to compensate for this but they don't null out to infinity.

Also, difference signals that are audible may not at all be in context of the music itself due to masking.

 

[rkbates]

Thanks for the clarification. My own tests show that good modern DACs measure almost identical but still show significant differences when listening. While some of these heard differences have their origins in psychoacoustics, some are real (otherwise we'd all be listening to Topping's excellent budget D30 DAC with inexpensive monoblocks). We need better and more revealing tests to account for these perceptual differences.

In the absence of such tests, this Ygg may genuinely sound better to most humans than better measuring DACs (solution proposed at the end of this post).

I've recently run through a whole pile of amplifiers which tested very well in your tests (amp section of Yamaha R-S700, NAD C320BEE, ones which tested less well Behringer A500, Marantz SR7001). There are huge differences in sound quality, with the Behringer doing much, much better in subjective tests than it possibly should in the field. Yes, there is hiss and crackle on near silent passages at high volume, but when music with a strong signal is playing, the A500 often sounds better than the rest (probably thanks to its enormous power reserves). The pre-amp tests which also included an NAD C372 (amp section long dead, preamp still works). As well as the variable output of the DACs (not recommended after testing, proper pre-amping with extra power does make a difference), there was a Topping A50s (excellent balanced headphone sound, excellent pre-amp, relatively better than its stablemate D50s), a NAD C165BEE (very, very good pre-amp but takes up a lot of space, very similar to the pre-amp in the C372) and a Pro-Ject Pre Box Analogue DS2 (great for rock, didn't like at all for classical/vocal) and now an RME ADI-2 DAC FS (compact, flexible, configurable, but a small step below the C165BEE and Pre Box as a pre-amp on congested passages). Worst pre-amp of all was the Yamaha R-S700, which sterilises and deadens sound while still measuring adequately. I was able to do these tests simultaneously and/or back to back as all the gear was here at the same time. I won't be able to do this again as almost all of it will be shipping out.

You have enormous stamina to go through so much gear every week. When people ask for you to put gear back in to run just one more test, I shudder. Thankfully you often refuse these days.

I correlated your measured findings with listening tests on as much of this gear as possible. The measurements provide a rough guide to performance but there were many cases where relatively poor measurers sonically outperformed measurement champs. To bring audio down to four or five measurements is a horrific simplification of sound, music and the capabilities of our ears. We need better and more revealing tests.

With DACs, a simple and significant test would be to diff output. Even subtle differences in sound cannot easily hide from a well-run diff as it's not a partial measurement but a full measurement. It's all the information which we hear. Diff files could work for pre-amps as well. What to do to better measure amplifiers, I have no idea.

Whilst I expect we have vastly different view on measurements, I believe running a music file through different devices and comparing their digitized output would be an interesting thing to do (probably been done a thousand times on this forum). Does anyone have some high quality musically challenging/revealing files they can share with me so I can do some playing around?

Edit - maybe in light of Amirm'sresponse I'm just wasting my time

 

[Alec Kinnear]

I'm sure that RME is tip top but I hear it's like figuring out Vulcan chess and dazzling light displays keep us from the inner sanctum where we truly hear what's going on. My main listening room is pretty strict, relatively speaking

The RME ADI-2 DAC FS is a great choice in Europe. It's affordable, sounds good, can take the job of DAC, headphone amp, pre-amp, DSP. There are better DACs, better headphone amps, better pre-amps (much bigger though) and more sophisticated DSP. But the ADI-2 FS does all of those things very well (top 20%) and costs roughly as much as any single device which would beat it in any single category.

On top of the features, the engineering is first rate. I don't get any electricity tingles when touching the case, it can handle multiple difference voltages, its feet are not just rubber but soft padded rubber. Oh and it comes with a high quality power supply which they RME engineers recommend you keep (instead of adding on a power supply which costs as much as the device, like high end and even low end manufacturers are doing whether Topping, Musical Fidelity, Meridian, Pro-Ject, PS Audio). From the manual:

Still the unit includes a high-quality switching power supply, 12 V / 2 A, which not only accepts any mains voltage between 100 V and 240 V (usable world-wide), but is also fully regulated against voltage fluctuations and suppresses line noise. Additionally it only weights 150 g in spite of its high power of 24 Watts.

Complexity: what's great about the ADI-2 FS is that it comes with a fat, well-written manual which not only helps you learn to configure the device but also teaches you about audio. Once the ADI-2 FS is configured as a DAC in your environment (balance, EQ, loudness starting point), you won't have to look at the advanced features again.

Oh and the latest version comes with a killer remote which looks good, feels good in the hand and covers all of the features you'd ever want to change or use when the ADI-2 is already configured.

There's a great feature where the fancy screen goes dark after twenty seconds. This feature was specifically added to make the ADI-2 better for use in a hifi as opposed to studio environment.

If you're in the States (sounds like you are) it's a little more difficult decision as RME products are a little more expensive and warranty service is further away. But the ADI-2 FS is the only sensibly designed and high performing all-in-one I can think of, and I've been looking.

 

[Alec Kinnear]

So let me get this straight, even after attempting to find some kind of explanation for what you seem to hear (good on you) and failing to do so, you still think there are changes "beyond" measuring, not seen in the FR?

The difference between the D50s and the V90/E30 with vocals vs instrumentals was not a vague maybe it's here, maybe it's not difference. It was clear that the either the V90 or E30 could be tuned (via pre-amp, or at the digital source) to give me endless pleasure as a listener. The recessed vocals in the D50s could not be fixed. My listening notes compared well with other reviewers who as collectively praised the D50s but came to the conclusion there is something missing and they would not want a D50s to remain in their system.

Anyone who thinks these multimeter tests adequately express sound and musical performance has a very poor understanding of the complexity of the universe and the sophistication of human senses. Yes, these tests are very useful to unearth flaws or performance limitations, but they do not capture tonality/colour. Sitting in a group chanting "APx555 is our one true god, APx555 betters our ears, APx555 better than human" will not make it so, guys.

Such primitive behaviour is just as ludicrous as the hard core subjectivists who refuse to measure anything or consider measurements while reviewing audio equipment in isolation without a reference. This website is called Audio Science Review not Church of Audioscientology.

Scientists are still searching for the nature of life. We have fragmentary explanations which cover many of the common processes, but others, such as ageing for instance, remain a near total mystery. Good scientists know their limitations.

The most beautiful thing we can experience is the mysterious. It is the source of all true art and science.

Do I have any takers for "Albert Einstein is not a scientist, Albert Einstein is a stupid subjectivist." Come on, @VintageFlanker this one has your name on it. With a double dose of threatening to block or ignore such quotations. The top of the website calls for:

Happy members who like to discuss audio....Come here to have fun.

not:

Angry members who like to browbeat others...Come here to be abused and abuse others.

Where is the fun? Where is the good nature? Where is the comparing notes? Where is the quest for better sound and more joy in music? I have a serious question for you, all of you:

What differences in sound do you hear between your Topping D50s and the Topping E30 (or D90 or Schiit Ygg or Modi)? If your answer is "none, as I haven't listened to either and the measurements show they are the same," you are no scientist. A scientist gets out in the field and actually tests his or her hypothesis and doesn't miserably lurk at home measurebating through someone else's work (Amir's in this case). Go out and listen!

Also any proof to your Audirvana claims?

Audible difference in players? (Audirvana, JRiver, Roon, MusicBee, etc.) thread please.

 

[Alec Kinnear]

Such tests are difficult to do since due to noise and clock drift, the output differs from run to run on the same DAC let alone different ones. We have tools to compensate for this but they don't null out to infinity.

Also, difference signals that are audible may not at all be in context of the music itself due to masking.

Amir, I understand such tests are not easy. But as most good DACs all measure more or less identical these days, to be able to advance the art, new tests are required. Such tests have a huge benefit to the science portion. If DIFF files show no difference, then it's time for the subjectivists and even the agnostic like myself to go home. Such a DIFF test would effectively cover both digital to analogue conversion and amplification

If rerecording and DIFF files are not really your cup of tea, ASR seems a bit enough operation with enough traffic and enough commercial partners to allow itself the luxury of engaging a young engineer to work on creating some tests for you and then processing them. You are the one with direct access to all the equipment so unless you find someone locally, there is some participation but you could limit your own participation to a minimum.

It's depressing to hear the objectivists simply claim these multimeter/APx555 measurements are enough to explain hearing and sound while the subjectivists claim measurements are irrelevant. Does no one want to break this stalemate? Is not the goal of science is avoid becoming stuck in dogmatic beliefs but to push knowledge forward.

The work you've done in debunking the nearly criminal claims of many high end audio manufacturers and much of the urban legend ($5000 connection cables? power plugs?) was and is enormously important. But if we don't find better measurements, it will be difficult for manufacturers of goodwill to advance the art. While audio equipment has improved in performance and price (it seems there was a very good period from the seventies through the nineties: the Japanese) and a very bad period (late nineties until about five years ago) where the charlatans reined and everything hifidelity was more and more expensive and chrome plated and we have now entered another productive period.

 

[Racheski]

What differences in sound do you hear between your Topping D50s and the Topping E30 (or D90 or Schiit Ygg or Modi)? If your answer is "none, as I haven't listened to either and the measurements show they are the same," you are no scientist. A scientist gets out in the field and actually tests his or her hypothesis and doesn't miserably lurk at home measurebating through someone else's work (Amir's in this case). Go out and listen!

Fair enough, but let's get one thing straight...whatever it is you are doing for your "listening tests" has nothing whatsoever to do with science. DACing-off to convince yourself that differences do exist among DACs does not make your conclusions "real" with respect to scientific validity, and does not justify the need for "better and more revealing tests to account for these perceptual differences." Your tests need to be conducted with far more rigor to be taken seriously.