The best DAC you never heard is now a production reality

[Ned Kelly]

New Schiit! Ragnarok and Yggdrasil - Page 306

 

Sorry BAD, MSB, EMM, Meithner, dCS, Ayre, Chord Electronics, Lampi - it would seem that all your efforts have been in vain. No, you cant order one yet, but Mike Moffat has handed his love child to the people who have to stuff seven - yep, seven - PCBs into a case smaller than my Oppo and ensure that it works after UPS have had their evil way with it. And all for 2299USD - despite a few grizzling about the price, the hypetrain has well and truly left the station and I doubt that Schitt will be able to keep up with initial demand. It's tempting to dismiss the 'Head-Fi effect', but without that forum companies like Audeze simply wouldnt have had the start they enjoyed,

 

I did start a thread here a while back asking if the R2R ladder DAC was really 'all that' and many of you replied that it absolutely can be - will be interesting to read more impressions when this thing finally appears on the product page. Forgive me if the above offends the owners of considerably more expensive gear, but my tongue remains firmly in my cheek.

[bhobba]

That's some claim.

 

No offence taken, but the proof of the pudding is in the eating.

 

How about getting one sent out my way in Brisbane/Gold Coast and we can really put it through it paces.

 

The DAC's available are:

Direct Stream

EMM Dac

Playback Designs MPD3 and 5

Phasure

Konverter

Killer

Klein DAC

Upgraded Oppo with Terra Firma clock and diamond transistor output stage

Both Tranquillity DAC's

AMR

 

Probably some others I cant recall of the top of my head.

 

And I have an upgraded Off-Ramp 4 (specially built linear supply) that will check out just how good its internal jitter is.

 

Of course tongue also in cheek, but if the maker wants to do it happy to oblige.

 

Thanks

Bill

[Ipoci]

I'm in the process to select my new DAC mainly for DSD ... I don't like Sabre based dacs so my interest is for all implementation that stay away from this well known chip ... I'm also convinced that best sound should be achieved by a good software-hardware integration so I've a couple of questions ... maybe some of you have the right answers.

 

- What kind of usb board is in use ? Is it xmos, Amanero or something different ?

- What kind of drivers will be available ? ASIO ? Linux ? OSX ?

 

Thank's in advance, have a nice day. Massimiliano

[manisandher]

I'm sure the Schiit Yggdrasil will be a great DAC for the money. But I find some of the marketing claims a little difficult to stomach. Here's one from Jason Stoppard:

 

"What we're trying to do, and what we will shortly introduce, is a DAC that is bitperfect from input to output, using the only closed-form digital filter and multibit DACs. This may be an insane goal in today's world, but, in our philosophy, "as true to the source as possible" starts with retaining the original samples. And that's what we'll do. It's up to you to decide if it's sonically meaningful, though..."

 

The Phasure NOS1 DAC (fed by a signal filtered/upsampled in XXHighEnd beforehand) has been capable of this for nearly 5 years now. In its latest NOS1"a" guise, it's seriously, seriously good.

 

I've never liked the sound of any delta-sigma DAC I've ever heard, so I do applaud Schiit's approach.

 

Mani.

[PeterSt]

is a DAC that is bitperfect from input to output, using the only closed-form digital filter and multibit DACs.

 

What is a "closed form digital filter" ?

And depending on the answer - a bit perfect filter ??

[Ned Kelly]

I'm in the process to select my new DAC mainly for DSD ... I don't like Sabre based dacs so my interest is for all implementation that stay away from this well known chip ... I'm also convinced that best sound should be achieved by a good software-hardware integration so I've a couple of questions ... maybe some of you have the right answers.

 

- What kind of usb board is in use ? Is it xmos, Amanero or something different ?

- What kind of drivers will be available ? ASIO ? Linux ? OSX ?

 

Thank's in advance, have a nice day. Massimiliano

 

Given Schitt's previous support for the their DACs I expect that they will ensure that it works on all 3 platforms - whether you will need a driver for non-Windows systems is a question best directed to them. As for the USB board / chipset, I dont recall Mike mentioning any other vendor in the Head-Fi thread : clearly, there are some wheels you wouldnt want to reinvent, but Schitt have gone to great lengths to stress that this was all their own work from design to production. The Head-Fi thread may be the best place to raise both questions.

[Ned Kelly]

I'm sure the Schiit Yggdrasil will be a great DAC for the money. But I find some of the marketing claims a little difficult to stomach. Here's one from Jason Stoppard:

 

"What we're trying to do, and what we will shortly introduce, is a DAC that is bitperfect from input to output, using the only closed-form digital filter and multibit DACs. This may be an insane goal in today's world, but, in our philosophy, "as true to the source as possible" starts with retaining the original samples. And that's what we'll do. It's up to you to decide if it's sonically meaningful, though..."

 

The Phasure NOS1 DAC (fed by a signal filtered/upsampled in XXHighEnd beforehand) has been capable of this for nearly 5 years now. In its latest NOS1"a" guise, it's seriously, seriously good.

 

I've never liked the sound of any delta-sigma DAC I've ever heard, so I do applaud Schiit's approach.

 

Mani.

 

No argument from me - its also expensive. Schitt are giving the rest of us a chance to get a glimpse into your world.

[bmoura]

New Schiit! Ragnarok and Yggdrasil - Page 306

 

Sorry BAD, MSB, EMM, Meithner, dCS, Ayre, Chord Electronics, Lampi - it would seem that all your efforts have been in vain.

 

$2,300 and no DSD playback? Time to go back to the design drawing board.

[4est]

The operative word in this appears to be "hypetrain"...

New Schiit! Ragnarok and Yggdrasil - Page 306

 

Sorry BAD, MSB, EMM, Meithner, dCS, Ayre, Chord Electronics, Lampi - it would seem that all your efforts have been in vain. No, you cant order one yet, but Mike Moffat has handed his love child to the people who have to stuff seven - yep, seven - PCBs into a case smaller than my Oppo and ensure that it works after UPS have had their evil way with it. And all for 2299USD - despite a few grizzling about the price, the hypetrain has well and truly left the station and I doubt that Schitt will be able to keep up with initial demand. It's tempting to dismiss the 'Head-Fi effect', but without that forum companies like Audeze simply wouldnt have had the start they enjoyed,

 

I did start a thread here a while back asking if the R2R ladder DAC was really 'all that' and many of you replied that it absolutely can be - will be interesting to read more impressions when this thing finally appears on the product page. Forgive me if the above offends the owners of considerably more expensive gear, but my tongue remains firmly in my cheek.

[Paul R]

$2,300 and no DSD playback? Time to go back to the design drawing board.

 

My first thought also, though really top end PCM DACs do sound really good.

[blueixus]

$2,300 and no DSD playback? Time to go back to the design drawing board.

 

Couldn't agree less with this. I am not a fan of DSD at all, I much prefer PCM and R2R dacs. Always have so I will be very interested in this indeed. Surely it is all about choice...or do we all have to put up with DSD?

[wisnon]

Couldn't agree less with this. I am not a fan of DSD at all, I much prefer PCM and R2R dacs. Always have so I will be very interested in this indeed. Surely it is all about choice...or do we all have to put up with DSD?

 

PUT up? What a strange choice of words.

 

What Dacs have you heard DSD on? That may better inform me as to your current position..

[Jabs1542]

"What we're trying to do, and what we will shortly introduce, is a DAC that is bitperfect from input to output, using the only closed-form digital filter and multibit DACs.

 

How is the output of a DAC bit perfect? I thought the output was analog (not being a smart ass, just trying to understand what "bit perfect from input to output" means).

 

Is this device really intended to compete with the new BADA Reference, being that they are the best at PCM without offering DSD?

[bhobba]

No argument from me - its also expensive. Schitt are giving the rest of us a chance to get a glimpse into your world.

 

It may not be to your liking.

 

I know he Phasure VERY well and have compared it to a number of DAC's in a number of comparisons.

 

It quite unmistakable - very clear, clean and pure - puts lesser DAC's to shame - but its not to everyones liking - some like me (and I am far from the only one) find it boring and uninteresting - while others go ga ga over it.

 

Lets not get too carried away here - wait until its gets into peoples hands and some comparisons done.

 

Like I said if someone wants to send me one I can give it a real workout.

 

Thanks

Bill

[wisnon]

How is the output of a DAC bit perfect? I thought the output was analog.

 

Is this device really intended to compete with the new BADA Reference, being that they are the best at PCM without offering DSD?

 

Hahahahaha,

 

Good question.

 

Moreover, according to S-Dad and Swenson, the PDN noise is a big SQ killer and bit perfection is irrelevant to that. According to Gordon R, many, if not most master clocks are producing more jitter than the incoming jitter from the transport (which is normally handled somewhat by async). Therefore Schitt has to come better than that. They may indeed have more up there sleeves, but the bit perfecttion argument by itself, is thin.

 

In any case, what recent discussions here and elsewhere have made clear to me is that no matter the posturing, digital audio at the highest SQ level, is STILL poorly understood and that is why trial and error/experimentation continues to produce incremental SQ gains and why the rate of improvement is still very respectable (at the very least the addition of new features, while improving or keeping constant SQ levels).

 

We know that DRR RAM moving stuff around randomly is detrimental and that SD card playback tends to be better than even SSDs. I have heard that the LauferTek Memory play is/was a great transport and that is a combination of a lot of tech that seems to not be widely recognized. I think they even espouse the concept of embedded jitter that is anathema here. However, I rule nothing out as we seem to be learning stuff month to month.

[bhobba]

In any case, what recent discussions here and elsewhere have made clear to me is that no matter the posturing, digital audio at the highest SQ level, is STILL poorly understood and that is why trial and error/experimentation continues to produce incremental SQ gains and why the rate of improvement is still very respectable (at the very least the addition of new features, while improving or keeping constant SQ levels).

 

Aren't that the truth.

 

And personal experience of being part of many comparisons shows personal preference is by far the biggest factor.

 

Thanks

Bill

[Jud]

$2,300 and no DSD playback? Time to go back to the design drawing board.

 

Brian - Mike Moffat (designer of this and the other Schiit DACs as well as the early California Audio Labs and Theta DACs - in other words, pretty much the originator of DACs as separate pieces of equipment), according to what he's written at least, hath no love for SDM. Whether this is the sum of his true feelings or whether there is a difference between what he's said publicly and his private opinion, I don't know. (Sum and difference, very small pun there.)

[Jud]

How is the output of a DAC bit perfect? I thought the output was analog (not being a smart ass, just trying to understand what "bit perfect from input to output" means).

 

 

Not only is the output analog, but as soon as you do interpolation you are no longer "bit perfect" even in the digital realm.

 

This may or may not be accurate, just my impression:

 

What I gather, amidst the marketing-speak, is that the "closed form digital filter" is meant to provably retain all the original samples during the interpolation/filtering process. That may be what "closed form" is supposed to mean - susceptible to mathematical proof that the original samples are retained.

 

Retention of all the original samples is what is being called "bit perfect."

 

[manisandher]

I know he Phasure VERY well and have compared it to a number of DAC's in a number of comparisons.

 

This is a Schiit thread, and I don't want to derail it. But just for clarification Bill, you know the old Phasure NOS1 very well, and the newer NOS1a not at all by the sound of things.

 

Mani.

[PeterSt]

Bill, you know the old Phasure NOS1 very well, and the newer NOS1a not at all by the sound of things.

 

Haha, don't bet Mani. OK, I would say the same regarding all contexts (like repeating old expressions from Bill), but IIRC 3 "a"s are in the (his) hood there.

[PeterSt]

Not only is the output analog, but as soon as you do interpolation you are no longer "bit perfect" even in the digital realm.

 

Allow me (and discuss if you like) :

As per your own guestimate, Jud, I personally would call that bit perfect just the same. So say the filtering is lossless, meaning that the original can be restored from the filtered data, I think it is allowed to call that bit perfect because it is about the consistency. Touch that and good sound is lost. This, for example, also means that any losslessly attenuated data is also as bit perfect as it was; only the volume is lower.

From the other angle I would not call this "closest form" because then my means (Arc Prediction) is as close as close can be (because lossless indeed). The sheer fact that someone tries to approach it is already a good thing though.

 

For fun we can also think like this :

If I'd make a filter which adds a pile of ringing then it still could be a lossless filter (normally it will not, but it can be). But 1:1 it is never any more. That we have a penalty on the frequency domain when the ringing is not added is something else (think 16/44.1) but still it is that (high THD) which actually is in the data. So I tend to call that 1:1. The big trick of course is to sustain the 1:1 while not hurting the frequency domain too much.

OK, this has been said a 1000 times by now.

[blueixus]

PUT up? What a strange choice of words.

 

What Dacs have you heard DSD on? That may better inform me as to your current position..

 

Currently i am running a Hugo which is good but i find DSD consistently too polite and rather dull, the reason for my choice of words

"put up with DSD Dacs" is that to build a DAC that manages both DSD & PCM will reduce the PCM capability. For me PCM should be managed by R2R dacs but these are increasingly rare because they are expensive to produce, not because they are inferior.

[bmoura]

Brian - Mike Moffat (designer of this and the other Schiit DACs as well as the early California Audio Labs and Theta DACs - in other words, pretty much the originator of DACs as separate pieces of equipment), according to what he's written at least, hath no love for SDM. Whether this is the sum of his true feelings or whether there is a difference between what he's said publicly and his private opinion, I don't know. (Sum and difference, very small pun there.)

 

I remember Moffat from the days he designed and marketed the Angstrom Home Theater processor as the first DTS decoder on the market. A colorful guy.

 

It also reminds me of the days that software companies like Ashton Tate, Lotus and Word Perfect attended an IBM session at Comdex years ago and proudly announced that "We don't like Microsoft Windows" and we're not going to put our best selling products (1-2-3, dBase and Word Perfect) out on Windows. We will issue them only on OS/2 from IBM. Everyone knows what happened to those companies after that one!

 

Companies need to remember that what is important when selling products is offering what consumers are looking to buy - not what companies want to sell to consumers.

 

Oh well, we'll see how the "hypetrain" goes on this one. Will music fans buy a $2,300 DAC that omits a key feature for many? When they can buy a DAC from many other companies that has that feature - at or below $2,300? Stay tuned....

[Jud]

From the other angle I would not call this "closest form" because then my means (Arc Prediction) is as close as close can be (because lossless indeed). The sheer fact that someone tries to approach it is already a good thing though.

 

 

Ah, but this I think is an "English" (language) thing. Not "closest" (in contrast to "farthest") form but "closed" (in contrast to "open") form. Here I take the use of "closed" to mean something like "able to be exactly reconstructed mathematically," thus provably retaining all original samples.

[Jud]

I remember Moffat from the days he designed and marketed the Angstrom Home Theater processor as the first DTS decoder on the market. A colorful guy.

 

It also reminds me of the days that software companies like Ashton Tate, Lotus and Word Perfect attended an IBM session at Comdex years ago and proudly announced that "We don't like Microsoft Windows" and we're not going to put our best selling products (1-2-3, dBase and Word Perfect) out on Windows. We will issue them only on OS/2 from IBM. Everyone knows what happened to those companies after that one!

 

Companies need to remember that what is important when selling products is offering what consumers are looking to buy - not what companies want to sell to consumers.

 

Oh well, we'll see how the "hypetrain" goes on this one. Will music fans buy a $2,300 DAC that omits a key feature for many? When they can buy a DAC from many other companies that has that feature - at or below $2,300? Stay tuned....

 

Their worst selling product is a *very* inexpensive, decent performing DSD DAC. You can attribute this to knowing their market, or trying to market something without true conviction behind it. Either way, it adds up to not convincing the Schiit principals terribly easily that foregoing DSD is a bad marketing choice.