Which DACs bypass digital filtering?

[semente]

Update:

 

- Aqua DACs

- Armature Asterion (same as Holo Spring)

- Audio-GD (some models)

- Border Patrol

- Denafrips

- Esoteric (some models?)

- Holo Audio (some models)

- iFi micro (some models)

- Lampizator

- Metrum Acoustics

- MHDT (some models)

- Phasure NOS1 DAC

- RME ADI-2 DAC

- T+A DAC 8 DSD

- Teac UD-50x family

- TotalDAC (some models?)

[matthias]

3 hours ago, semente said:

- Armature Asterion (same as Holo Spring)

 

Please, can you shed some light on this?

Under this name I have seen so far only a model with the old Spring1 technique but not with the new Spring2 one.

Are both built up to the same standards by the same manufacturer or is Armature somewhat inferior to Holo?

 

Thanks

 

Matt

[semente]

15 minutes ago, matthias said:

 

Please, can you shed some light on this.

Under this name I have seen so far only a model with the old Spring1 technique but not with the new Spring2 one.

Are both built by the same manufacturer or is Armature a copy of Holo?

 

Thanks

 

Matt

 

Hi Matt,

 

The Asterion is a slightly tweaked version of the "old" Spring 1:

 

L’Armature Asterion est une version customisée du Holo Audio spring (en level 1 chez les distributeurs qui vendent des versions gonflées, avec notamment un transfo en argent au level3, ajout jugé discutable par le staff audiophonics, qui est resté avec la version standard, très homogène). Armature a fait quelques modifs, avec une entrée USB XMOS plus musclée et un afficheur différent. Il semblerait également que les connecteurs soient différents.

https://www.audiophonics.fr/en/blog-diy-audio/22-armature-asterion-dac-complete-review-by-kyosato.html

 

Translation here - > https://tinyurl.com/y2mlnlh7

 

Best,

Ricardo

 

[matthias]

2 minutes ago, semente said:

 

Hi Matt,

 

The Asterion is a slightly tweaked version of the "old" Spring 1:

 

L’Armature Asterion est une version customisée du Holo Audio spring (en level 1 chez les distributeurs qui vendent des versions gonflées, avec notamment un transfo en argent au level3, ajout jugé discutable par le staff audiophonics, qui est resté avec la version standard, très homogène). Armature a fait quelques modifs, avec une entrée USB XMOS plus musclée et un afficheur différent. Il semblerait également que les connecteurs soient différents.

https://www.audiophonics.fr/en/blog-diy-audio/22-armature-asterion-dac-complete-review-by-kyosato.html

 

Translation here - > https://tinyurl.com/y2mlnlh7

 

Best,

Ricardo

 

 

Thanks Ricardo,

do you know if they will launch a tweaked version of Spring2 as well and if yes, when?

 

Matt

[semente]

Just now, matthias said:

 

Thanks Ricardo,

do you know if they will launch a tweaked version of Spring2 as well and if yes, when?

 

Matt

 

Sorry, I have no idea if they intend to update the current version.

Maybe you can send them an email: http://armature.audio/contactez-nous/

[matthias]

6 minutes ago, semente said:

 

Sorry, I have no idea if they intend to update the current version.

Maybe you can send them an email: http://armature.audio/contactez-nous/

 

Thanks, email sent.

 

Matt

[matthias]

On 3/10/2019 at 3:26 PM, semente said:

 

Sorry, I have no idea if they intend to update the current version.

Maybe you can send them an email: http://armature.audio/contactez-nous/

 

I asked if they have plans to launch a tweaked version of the new Holo Spring2.

 

The reply:

"To be honest no, since for us the gap is at best minimal."

 

Matt

 

[Veri]

On 3/10/2019 at 11:17 AM, semente said:

Update:

 

- Aqua DACs

- Armature Asterion (same as Holo Spring)

- Audio-GD (some models)

- Border Patrol

- Denafrips

- Esoteric (some models?)

- Holo Audio (some models)

- iFi micro (some models)

- Lampizator

- Metrum Acoustics

- MHDT (some models)

- Phasure NOS1 DAC

- RME ADI-2 DAC

- T+A DAC 8 DSD

- Teac UD-50x family

- TotalDAC (some models?)

 

Apparently the RME ADI-2 is not really NOS. They only called it so,

 

it is the 'Super Slow' AKM FIR filter which shows a stepped response like NOS. Many AKM chip devices can select this filter, are they all NSO DACs? I think not.

[semente]

10 hours ago, Veri said:

 

Apparently the RME ADI-2 is not really NOS. They only called it so,

 

it is the 'Super Slow' AKM FIR filter which shows a stepped response like NOS. Many AKM chip devices can select this filter, are they all NSO DACs? I think not.

If I'm not mistaken if you feed it a stream at the highest possible sample rate it will not oversample so it is NOS, but fron what you say it is not filterless.

My TEAC also has a FIR filter for DSD but I think that it's an analogue filter. And crossing quite high.

[Miska]

12 hours ago, Veri said:

Apparently the RME ADI-2 is not really NOS. They only called it so,

 

it is the 'Super Slow' AKM FIR filter which shows a stepped response like NOS. Many AKM chip devices can select this filter, are they all NSO DACs? I think not.

 

If you run it at >= 352.8 it bypasses the digital filter stage and runs only S/H, like many other SDM DAC chips. It is still oversampling, but not through digital filter anymore, so it just copies same sample N times. You cannot really run an SDM DAC at PCM rates because the rate would be too low for the modulator to do it's job. So "NOS" in context of SDM DAC with PCM input usually means "no digital filter".

 

Then it has switched capacitor analog filter in the output as a D/A conversion stage to produce voltage output. In this respect is is similar to Cirrus Logic chips.

 

In DSD Direct mode, it sends DSD inputs straight to the switched capacitor filter, which can be configured for two different cut-off points. If it is not in DSD Direct mode, then there is quite a bit of extra DSP going on. TEAC runs the chip in DSD Direct mode always since they have analog volume control. RME has configuration option to switch one of the two DAC chips into DSD Direct mode and consequently volume control is gone (and corresponding headphone output is muted). The other DAC chip is never in DSD Direct mode and thus the other headphone output is active and has volume control.

 

[Superdad]

4 hours ago, Miska said:

You cannot really run an SDM DAC at PCM rates because the rate would be too low for the modulator to do it's job.

 

Great post Jussi.  People should study and remember that important information. 

[asdf1000]

On 4/16/2019 at 7:57 AM, Miska said:

If it is not in DSD Direct mode, then there is quite a bit of extra DSP going on.

 

But it is converted to PCM for this extra DSP?

 

[asdf1000]

On 5/29/2017 at 11:49 PM, mansr said:

DSD handling is quite different from the TI chip. In the default mode of operation, DSD input is low-pass filtered before going through the same digital volume control and sigma-delta modulator as PCM. As far as I can tell, the modulator is operated at the same rate as the DSD input, i.e. no resampling is performed.

 

Quoting an old post here but with the AKM chips, would DSD256 input (for example) be converted to PCM for digital volume control?

 

For example, the AK4493EQ.pdf

 

I'm trying to understand, if I upsample my PCM music to DSD256 (1 bit 11.2896 MHz) and use digital volume control, is my DSD256 being converted to PCM705kHz (or something else) or does it remain at 11.2896 MHz with some extra bits for the digital volume control?

 

Or something else?

 

 

[mansr]

2 hours ago, Em2016 said:

Quoting an old post here but with the AKM chips, would DSD256 input (for example) be converted to PCM for digital volume control?

 

For example, the AK4493EQ.pdf

 

I'm trying to understand, if I upsample my PCM music to DSD256 (1 bit 11.2896 MHz) and use digital volume control, is my DSD256 being converted to PCM705kHz (or something else) or does it remain at 11.2896 MHz with some extra bits for the digital volume control?

 

Or something else?

My guess is that it uses a multi-bit format at the same rate as the DSD input. Only AKM themselves know for sure how it works.

[firedog]

2 hours ago, Em2016 said:

I'm trying to understand, if I upsample my PCM music to DSD256 (1 bit 11.2896 MHz) and use digital volume control, is my DSD256 being converted to PCM705kHz (or something else) or does it remain at 11.2896 MHz with some extra bits for the digital volume control?

You don't have to convert it to PCM to do the volume control. Roon performs all sorts of DSP on DSD files without conversion to PCM. Multibit, but not PCM. They have an explanation in either the forum or the wiki of how they manipulate DSD without conversion to PCM.
A chip can be setup to do the same. How it actually works in that chip - don't have a clue. 

[mansr]

1 minute ago, firedog said:

You don't have to convert it to PCM to do the volume control. Roon performs all sorts of DSP on DSD files without conversion to PCM. Multibit, but not PCM. They have an explanation in either the forum or the wiki of how they manipulate DSD without conversion to PCM.

PCM at a high rate is still PCM.

[firedog]

20 minutes ago, mansr said:

PCM at a high rate is still PCM.

Roon can do some DSD processing, including Parametric EQ,  by converting it to DSD wide (64 bit multibit) and then back to one bit DSD in the Delta Sigma Modulator after the 64 bit DSD is processed. 

 

I'm not going to have the argument about whether that is still DSD or not. Roon and others consider that it is. 

[mansr]

45 minutes ago, firedog said:

Roon can do some DSD processing, including Parametric EQ,  by converting it to DSD wide (64 bit multibit) and then back to one bit DSD in the Delta Sigma Modulator after the 64 bit DSD is processed. 

 

I'm not going to have the argument about whether that is still DSD or not. Roon and others consider that it is. 

It doesn't matter what anyone calls it, only what it actually is. The important thing here is that if DSP is performed on DSD, there is necessarily a sigma-delta modulator at the end of it. Otherwise you wouldn't have 1-bit output.

[Ralf11]

1 hour ago, mansr said:

It doesn't matter what anyone calls it, only what it actually is. ...

 

No no no - you are thinking like  a Modernist.

 

In the post-modern weltmockerschnaug of our culturally dependent uni-multi-verse, signification is central to the deconstrucivist unfolding of deconvoluted reality.  Therefore, the naming is crucial to our conceptual continuity.

[mansr]

Alice could only look puzzled: she was thinking of the pudding.

‘You are sad,’ the Knight said in an anxious tone: ‘let me sing you a song to comfort you.’

‘Is it very long?’ Alice asked, for she had heard a good deal of poetry that day.

‘It’s long,’ said the Knight, ‘but very, very beautiful. Everybody that hears me sing it—either it brings the tears into their eyes, or else—’

‘Or else what?’ said Alice, for the Knight had made a sudden pause.

‘Or else it doesn’t, you know. The name of the song is called “Haddocks’ Eyes.”’

‘Oh, that’s the name of the song, is it?’ Alice said, trying to feel interested.

‘No, you don’t understand,’ the Knight said, looking a little vexed. ‘That’s what the name is called. The name really is “The Aged Aged Man.”’

‘Then I ought to have said “That’s what the song is called”?’ Alice corrected herself.

‘No, you oughtn’t: that’s quite another thing! The song is called “Ways and Means”: but that’s only what it’s called, you know!’

‘Well, what is the song, then?’ said Alice, who was by this time completely bewildered.

‘I was coming to that,’ the Knight said. ‘The song really is “A-sitting On A Gate”: and the tune’s my own invention.’

 

-- Lewis Carroll, Through the Looking-glass

[Ralf11]

always weird when a logician writes a children's book...

 

then there is twillig with slithy goves byring and wimbling about....

[asdf1000]

5 hours ago, firedog said:

You don't have to convert it to PCM to do the volume control. 

 

I know, that’s the reason I asked “or does it remain at 11.2896 MHz with some extra bits for the digital volume control?“

 

I was asking about this chip specifically...but it still appears to be a black box of sorts, like ESS Sabre chips.

[Miska]

9 hours ago, Em2016 said:

I'm trying to understand, if I upsample my PCM music to DSD256 (1 bit 11.2896 MHz) and use digital volume control, is my DSD256 being converted to PCM705kHz (or something else) or does it remain at 11.2896 MHz with some extra bits for the digital volume control?

 

Don't do it, do the digital volume control in PCM before it ends up being DSD256. And keep the AKM chip in DSD Direct mode. What comes out with/without DSD Direct is vastly different.

 

[asdf1000]

2 hours ago, Miska said:

 

Don't do it, do the digital volume control in PCM before it ends up being DSD256. And keep the AKM chip in DSD Direct mode. What comes out with/without DSD Direct is vastly different.

 

 

The Topping AKM DACs don’t have “DSD Direct” mode though, like RME and Teac do.

 

And I need to use the built in digital volume control for built in headphone output. 

 

So should I upsample my PCM content to highest PCM rate instead , with AK4493EQ? Avoid PCM to SDM (DSD) upsampling?

 

[luisma]

5 hours ago, mansr said:

5 hours ago, mansr said:

-- Lewis Carroll, Through the Looking-glass

Beatiful but completely OT, off with you head