Jump to content
IGNORED

Native DSD - First Impressions


Recommended Posts

I installed the latest Conductor update last night for my Aurender N10.  My Vivaldi doesn't support Native DSD (Are you listening, DCS?), but I do have an iFi iDSD Micro in the rack, normally fed by a modest laptop for serving up YouTube music and such.  That iFi does support Native DSD, so I connected it to the N10 and started some listening tests.
I started with some DSD128 tracks because that allowed me to A/B between DoP and Native DSD.  The improvement brought on with Native DSD was immediately obvious.  The sound was more relaxed, less congested, quieter, with more realistic-sounding instruments.
This all makes sense because the DAC is not working as hard receiving the DoP data and re-assembling the DSD stream.
Then I flipped to a DSD256 version of the same track.  The sound was even better.  This time the improvement was due to the increase in resolution.  And what a wonderful improvement it was.  Listening to Louis Armstrong's live album called Jack the Knife, downloaded from nativedsd.com, Armstrong's voice was more three dimensional, as was every instrument in the band.  Everything was more solid and alive.
Thank you Aurender, and thank you iFi, and what the hell - Thank you Louis Armstrong, for making my world a little nicer.
Brian.

Rockport Altairs, D'Agostino Integrated, DCS Vivaldi, Aurender N10, Transparent/HRS/Nordost.

Link to comment

So it's not just a myth, but a false myth?  Ok.
This is such a silly thing to argue about given that the point of the post was to talk about what I heard with the new Aurender update.
We don't even disagree for the most part.  Do you just object that I say 'protocol' and you say 'shell'?
DoP is a system for stuffing PCM frames with DSD data.  I think we can agree on this.  The DAC, including everything that is on the receiving end of the USB cable, must recover the DSD data from the PCM frames before processing it as DSD data.  I would guess that we even agree on this.  Amazing what we agree on, isn't it?  No 'false myths' so far.
Perhaps where we disagree is on the impact this extraction could have on the ultimate sound quality.  I guess if your point is that "I can't possibly hear the difference" or something, the conversation is over since we're now just speculating about each other's systems and ears.  All I can say is that I can hear the difference without any difficulty, as can everyone who has heard me demo the new Aurender feature.  The update makes me happy.  I'm glad iFi supports this mode.  Does this mean that the iFi will replace my Vivaldi for all future listening sessions?  Not at all.  Does it allow me to listen to my DSD256 tracks that the Vivaldi doesn't support?  You bet.
Take care - Brian.

Rockport Altairs, D'Agostino Integrated, DCS Vivaldi, Aurender N10, Transparent/HRS/Nordost.

Link to comment
49 minutes ago, firedog said:

I understand that your main point was talking about what you heard with the Aurender. I don't comment on what other people hear-I accept they hear what they claim, becuase even if the sound is based on expectation bias, that's what they are hearing. That's a general comment, not directed at you. 

But you framed your conclusion as being a result of native DSD as opposed to DoP and explained why. Your explanation, as stated originally, is based on a misundertanding of what DoP is, as it doesn't require "reassembly" of the DSD. If you didn't mean that, fine.

But it's an oft repeated (false) myth among audiophiles that DOP somehow isn't native DSD-it is. "Native" DSD originally meant DSD not converted to PCM for playback. Both what you call native and DoP fit this definition.

Manufacturers and marketers have promoted a misunderstanding, tied to redefining "native" as "not DoP" in order to create an incorrect idea that implies that DoP DSD streams are somehow altered and that the processing of them in a DAC is somehow fundamentally different from those used to process what they call "native" DSD. It's not. DoP doesn't add a "processing burden" to the DAC. They just want to make is seem that way in order to claim/imply an illusory SQ advantage for their DACs.

 

Related question: are you familiar with uncompressed (distinct from 0 level compression) flac? It a PCM file in a flac shell.  It's essentially the same size as the equivalent wav file - both are uncompressed. It's the same data. A wav file is a PCM file in a wav shell. Do you think the two  sound different because the flac shell requires "processing" that the "wav" shell doesn't? The same thing is true o the two types of DSD you are talking about. 

 

 

DoP places a burden of additional 50% on the CPU in comparison to "native" DSD. Surely it is DSD but the CPU has more work to do anyway. For DSD256 for example you need PCM768 as carrier for DoP. I am sure @BrianLS is right in hearing differences in favour of "native" DSD.

 

Matt

"I want to know why the musicians are on stage, not where". (John Farlowe)

 

Link to comment
1 hour ago, matthias said:

 

DoP places a burden of additional 50% on the CPU in comparison to "native" DSD. Surely it is DSD but the CPU has more work to do anyway. For DSD256 for example you need PCM768 as carrier for DoP. I am sure @BrianLS is right in hearing differences in favour of "native" DSD.

 

Matt

Not on my system it doesn't. I just ran DSD 64, 128, and 256 - each in both native and DoP. Same track for each rate of DSD.

Virtually no difference in CPU load reported.

Main listening (small home office):

Main setup: Surge protector +>Isol-8 Mini sub Axis Power Strip/Isolation>QuietPC Low Noise Server>Roon (Audiolense DRC)>Stack Audio Link II>Kii Control>Kii Three (on their own electric circuit) >GIK Room Treatments.

Secondary Path: Server with Audiolense RC>RPi4 or analog>Cayin iDAC6 MKII (tube mode) (XLR)>Kii Three .

Bedroom: SBTouch to Cambridge Soundworks Desktop Setup.
Living Room/Kitchen: Ropieee (RPi3b+ with touchscreen) + Schiit Modi3E to a pair of Morel Hogtalare. 

All absolute statements about audio are false :)

Link to comment
1 hour ago, firedog said:

Not on my system it doesn't. I just ran DSD 64, 128, and 256 - each in both native and DoP. Same track for each rate of DSD.

Virtually no difference in CPU load reported.

And just to pile on.  One can do those DSD sample rates in multichannel (6 mostly) without any significant processing burden on a competent player.  

Kal Rubinson

Senior Contributing Editor, Stereophile

 

Link to comment
13 hours ago, matthias said:

 

This is the proof that SQ of native DSD is superior to that of DoP.

 

Matt

This type of conclusion is incorrect.  I only use “Native DSD” because my software chain and DACs are DSD Native (when the support DSD).

 

If you, for example, find that there is a 50% CPU overhead for DOP then there is likely something wrong somewhere. The overhead should be minimal to nonexistent. A number of people who have measured this have said its nonexistent. It’s up to the DAC to take whatever formats it supports and process them.

 

Most of this is packing bits into buffers and queueing them to the USB driver — at the low data rates we are discussing this is low CPU overhead — where are you measuring a 50% CPU overhead?

 

You more likely have a configuration issue.

Custom room treatments for headphone users.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...