Schiit Yggdrasil - Best DAC available!?

[semente]

Talk to the best engineers in the industry, who are designing cutting edge products, and they'll all tell you to throw out the text book designs.

 

In this case Schiit is doing this to eliminate the DC servo and its signal bleed back into the input stage.

 

I love when companies do stuff like this. I could care less if the DAC measures terrible because of this design choice. I really care about fidelity to the music, not test tones. Fidelity and this design are not mutually exclusive.

 

A company that sticks to the text book designs so that its products measure better when played a test tone (not talking about fidelity to the musical signal) versus play music more accurately (less signal bleed back into input stage) is not cutting edge.

 

 

Please correct me if I'm wrong but I have the impression that a significant set of measurements will go a long way into characterising an equipment's performance (meaning the accuracy with which it handles the recorded signal, or in this case amplifies it, and not how pleasant it sounds).

 

 

In my opinion there's no such thing as "fidelity to the music"; audio equipment deals with sound, whatever is in the recording, and the less harm it does to that signal the more of the music one will be able to listen to.

 

R

[The Computer Audiophile]

Good quality caps can do the trick much more simply.

 

And clearly the method Schiit chose isn't 100% transparent 100% of the time. Was the cure worse than the disease? Don't listen to the Beatles " a Day in the Life" with your Schiit. So even if we find no other song we can say it is not 100% transparent 100% of the time.

 

A DC servo is there for a reason. It is faulty reasoning to say if it can be removed then removing it will make it transparent all the time. Clearly in this design that isn't so.

I meant that DC servo problems are gone when the DC servo is removed. And, all said I context of Alex's comments.

[The Computer Audiophile]

Please correct me if I'm wrong but I have the impression that a significant set of measurements will go a long way into characterising an equipment's performance (meaning the accuracy with which it handles the recorded signal, or in this case amplifies it, and not how pleasant it sounds).

 

 

In my opinion there's no such thing as "fidelity to the music"; audio equipment deals with sound, whatever is in the recording, and the less harm it does to that signal the more of the music one will be able to listen to.

 

R

Not sure what you're saying.

 

im saying the measurements of this amp mean nothing to me if the product freaks out with test tones but not music.

[semente]

Not sure what you're saying.

 

im saying the measurements of this amp mean nothing to me if the product freaks out with test tones but not music.

 

I understand now.

But why have it "freak out" with test signals in the first place?

Is there any advantage from a performance/accuracy perspective?

Why don't other high performance design act in such manner?

 

R

[esldude]

Not sure what you're saying.

 

im saying the measurements of this amp mean nothing to me if the product freaks out with test tones but not music.

 

I made a recording for some friends last year. It was in a large space and there was a fairly substantial 17 hz resonance much like a steady tone. It wasn't enough to overload anything else so I didn't worry about it too much for the first take though I later turned on a filter. I then distributed some of those first takes to the musicians. I forgot to filter it first. Two of them complained of distorted sound and woofers going crazy. It was that tone eating up too much power in their small underpowered systems. Would a recording like that trip up this Schiit amp on biasing? Seems like it might.

 

Someone already mentioned the Interstellar soundtrack. It has a 15-16 khz tone for minutes in part of the movie. So don't use your Schiit to listen to this movie or the soundtrack from it.

[sandyk]

I meant that DC servo problems are gone when the DC servo is removed. And, all said I context of Alex's comments.

 

Perhaps I should have left out the word virtually ?

The point I was trying to make is that there are far greater audible benefits to be heard by going fully DC coupled.

[The Computer Audiophile]

I understand now.

But why have it "freak out" with test signals in the first place?

Is there any advantage from a performance/accuracy perspective?

Why don't other high performance design act in such manner?

 

R

 

It freaks out because of a design decision to remove the DC servo (getting rid of noise on input stage) and use the microprocessor instead.

 

There are many ways to go about things. This one freaks out with test tones. Big deal.

[The Computer Audiophile]

I made a recording for some friends last year. It was in a large space and there was a fairly substantial 17 hz resonance much like a steady tone. It wasn't enough to overload anything else so I didn't worry about it too much for the first take though I later turned on a filter. I then distributed some of those first takes to the musicians. I forgot to filter it first. Two of them complained of distorted sound and woofers going crazy. It was that tone eating up too much power in their small underpowered systems. Would a recording like that trip up this Schiit amp on biasing? Seems like it might.

 

Someone already mentioned the Interstellar soundtrack. It has a 15-16 khz tone for minutes in part of the movie. So don't use your Schiit to listen to this movie or the soundtrack from it.

 

I think someone should test these recording on the Schiit before speculating :~)

[Jud]

I made a recording for some friends last year. It was in a large space and there was a fairly substantial 17 hz resonance much like a steady tone. It wasn't enough to overload anything else so I didn't worry about it too much for the first take though I later turned on a filter. I then distributed some of those first takes to the musicians. I forgot to filter it first. Two of them complained of distorted sound and woofers going crazy. It was that tone eating up too much power in their small underpowered systems. Would a recording like that trip up this Schiit amp on biasing? Seems like it might.

 

Someone already mentioned the Interstellar soundtrack. It has a 15-16 khz tone for minutes in part of the movie. So don't use your Schiit to listen to this movie or the soundtrack from it.

 

What JA did was run a single low level continuous test tone as the only signal into the Ragnarok for five minutes. This caused the bias to be adjusted in such a way as to result in very high THD. Nothing you've described is a situation where a single continuous low level pure sine wave tone is played with no other signal for five minutes.

[mansr]

Not sure what you're saying.

 

im saying the measurements of this amp mean nothing to me if the product freaks out with test tones but not music.

If it freaks out on a test signal, what guarantee is there that it doesn't freak at least some of the way out on some non-test signal? I'd rather not have some engineer at Schiit be the arbiter of what constitutes real music.

[Jud]

If it freaks out on a test signal, what guarantee is there that it doesn't freak at least some of the way out on some non-test signal? I'd rather not have some engineer at Schiit be the arbiter of what constitutes real music.

 

As soon as someone stops talking in generalities and gives me an example of music that has five minutes of a low level pure sine wave tone as the only signal, I'll be more concerned than I am currently.

[mansr]

As soon as someone stops talking in generalities and gives me an example of music that has five minutes of a low level pure sine wave tone as the only signal, I'll be more concerned than I am currently.

Who said that's the only signal it can't handle?

[Jud]

Who said that's the only signal it can't handle?

 

As soon as you show me measurements, as JA did, I'll believe it.

[mansr]

As soon as you show me measurements, as JA did, I'll believe it.

I don't have one of those amps, and given what I've read, I never will.

 

The problem I see is that with a design like that, the amp is impossible to characterise. It is unknowable how it will perform on a signal you haven't explicitly tested. That's not an amp I'll be buying.

[Jud]

I don't have one of those amps, and given what I've read, I never will.

 

The problem I see is that with a design like that, the amp is impossible to characterise. It is unknowable how it will perform on a signal you haven't explicitly tested. That's not an amp I'll be buying.

 

As I said earlier in the thread, I favor equipment that naturally tends toward stability at operating characteristics that sound good. (Wonder if the latest much-admired Class D amps meet that description?) But I saw the thread descending toward unsubstantiated claims that any music with sustained tones could not be played by this amp, and felt more sober consideration was called for.

[esldude]

As soon as someone stops talking in generalities and gives me an example of music that has five minutes of a low level pure sine wave tone as the only signal, I'll be more concerned than I am currently.

 

We don't know what the amp would do. JA didn't test for a mix of some considerable tone with music. The fact it could happen represents a poor design choice in my opinion. JA did play music and switch to a tone and noted the THD immediately started to rise. Would it have risen some with the Interstellar high pitch tone which is fairly high in level mixed with music? Would it rise with the 17 hz I described mixed with music? It might, it might rise a little or a lot. JA was switching from moderately loud music to equally moderate tones of 14w or 16 watts. This from a not exactly super powerful amplifier.

 

Of course it doesn't matter. Why not fix what isn't broken and call it innovation. The only thing that matters is I won't be buying one. If plenty of others do then they'll have a hit on their hands and be encouraged to design more things the same way.

 

You can read in the middle of this page the develop of this amp by Jason Stoddard.

 

http://www.head-fi.org/t/701900/schiit-happened-the-story-of-the-worlds-most-improbable-start-up/2205#post_10812955

[sandyk]

As soon as you show me measurements, as JA did, I'll believe it.

 

Hi Jud

I thought that perhaps you would be one of the last people to rely on measurements which are not always the appropriate type, and don't always tell the full story.

Regards

Alex

[Jud]

We don't know what the amp would do. JA didn't test for a mix of some considerable tone with music. The fact it could happen represents a poor design choice in my opinion. JA did play music and switch to a tone and noted the THD immediately started to rise. Would it have risen some with the Interstellar high pitch tone which is fairly high in level mixed with music? Would it rise with the 17 hz I described mixed with music? It might, it might rise a little or a lot. JA was switching from moderately loud music to equally moderate tones of 14w or 16 watts. This from a not exactly super powerful amplifier.

 

A particular problematic scenario was pointed out. You are of course correct that we can't tell from this whether the amp is held together with spit, baling wire, and microprocessors, riding on a knife edge until someone gets too enthusiastic with pedal sustain in a piano concerto.

 

I suppose I am fairly eclectic in my skepticism, choosing to exercise it on both quantum cables and on the idea this amp is a THD grenade with a loose pin.

[esldude]

Well the amp is on back order for a few weeks. So a successful review by Stereophile.

[Jud]

Hi Jud

I thought that perhaps you would be one of the last people to rely on measurements which are not always the appropriate type, and don't always tell the full story.

Regards

Alex

 

Hi Alex. So you have something else besides JA's measurements showing a scenario where this amp operates at outlandish distortion levels?

[Jud]

Well the amp is on back order for a few weeks. So a successful review by Stereophile.

 

It's been on backorder a substantial amount of the time it's been for sale prior to the Stereophile review. Of course the Stereophile review won't hurt with those who still pay attention to magazine reviews, of which I'm not one. I quickly lose interest when I can write the ending without having to read the piece.

[mansr]

I quickly lose interest when I can write the ending without having to read the piece.

 

Good one.

[jacquesr]

I have DACs on the brain. My mistake. Thanks for the correction.

 

Hello Chris,

Great to see you in this thread...

Can you chime in on the possibility of the so called Yggy "glitch"? Just be sure, it is a DAC

 

You review of the Yggy was very clear. But the recent events in this thread being what they are, your input (USB, of course) would be much appreciated.

 

Also, I was wondering, does USB mean Unequivocally Supporting Baldur" ?

[mansr]

Also, I was wondering, does USB mean Unequivocally Supporting Baldur" ?

 

No, it's Unlimited Source of Bugs.

[jabbr]

Talk to the best engineers in the industry, who are designing cutting edge products, and they'll all tell you to throw out the text book designs.

 

In this case Schiit is doing this to eliminate the DC servo and its signal bleed back into the input stage.

 

I love when companies do stuff like this. I could care less if the DAC measures terrible because of this design choice. I really care about fidelity to the music, not test tones. Fidelity and this design are not mutually exclusive.

 

A company that sticks to the text book designs so that its products measure better when played a test tone (not talking about fidelity to the musical signal) versus play music more accurately (less signal bleed back into input stage) is not cutting edge.

 

Agreed. Case in point would be the new Nelson Pass/First Watt F7 which I haven't heard yet but has gotten rave reviews. Its claim to fame is the use of positive current feedback. Consequently not tuned for lowest THD! Similarly the V-FET/SIT designs, in which case the transistors are designed to act more like triodes. Again, the design is tailored to simplicity and improved sound, not blindly reduced THD.

 

http://www.6moons.com/audioreviews2/firstwatt2/4.html

http://www.diyaudio.com/forums/pass-labs/275921-first-watt-f7-review-44.html