DSD Direct vs non-DSD Direct: Myth or Turth?

[Sokel]

There have been some implementations around for Direct DSD with no DAC involved.
Rather simple ones,you can search diyaudio for those.

I know it's not the subject of the thread but if one searches for something like that...

 

[Geert]

MEASUREMENTS: DSD Testing with SoX-DSD Signals Part II (RME ADI-2 Pro FS R Black Edition, DSD Direct, Filter @ 50/150kHz) [Updated with 44.1kHz PCM --> DSD.]

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

DSD Direct versus PCM:

Multitone DSD Direct versus normal DSD:



If there's any audible difference, it's hiding very well from measurements

"Given the excellent AKM AK4493 performance for DSD playback, I certainly would not insist that "DSD Direct" must be turned on. I did not notice significant audible changes listening to the RME ADI-2 FS R BE as I played DSD64 tracks and fooled around with "DSD Direct" ON / OFF and the filter settings".

 

[Sokel]

MEASUREMENTS: DSD Testing with SoX-DSD Signals Part II (RME ADI-2 Pro FS R Black Edition, DSD Direct, Filter @ 50/150kHz) [Updated with 44.1kHz PCM --> DSD.]

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

DSD Direct versus PCM:

View attachment 359602

Multitone DSD Direct versus normal DSD:

View attachment 359603

If there's any audible difference, it's hiding very well from measurements

"Given the excellent AKM AK4493 performance for DSD playback, I certainly would not insist that "DSD Direct" must be turned on. I did not notice significant audible changes listening to the RME ADI-2 FS R BE as I played DSD64 tracks and fooled around with "DSD Direct" ON / OFF and the filter settings".

That's the reason I'm a pre-fire AKM fan-boi

It's nice 20-20Khz (not the case with some newer,specially some ESS as going lower) and in any format.

(people don't tell me about audibility,if it was about that we would only care about SNR,gain staging,clipping and maybe IMD ,it's the visuals of measurements that's nice )

 

[sunjam]

MEASUREMENTS: DSD Testing with SoX-DSD Signals Part II (RME ADI-2 Pro FS R Black Edition, DSD Direct, Filter @ 50/150kHz) [Updated with 44.1kHz PCM --> DSD.]

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

DSD Direct versus PCM:

View attachment 359602

Multitone DSD Direct versus normal DSD:

View attachment 359603

If there's any audible difference, it's hiding very well from measurements

"Given the excellent AKM AK4493 performance for DSD playback, I certainly would not insist that "DSD Direct" must be turned on. I did not notice significant audible changes listening to the RME ADI-2 FS R BE as I played DSD64 tracks and fooled around with "DSD Direct" ON / OFF and the filter settings".

Hmm... looks like it is "just some purists want their DSD music to be 'bit perfect' down to the last digital-to-analogy step (even the difference in terms of the sound produced is virutally close to none)?"

 

[sunjam]

That's the reason I'm a pre-fire AKM fan-boi

It's nice 20-20Khz (not the case with some newer,specially some ESS as going lower) and in any format.

(people don't tell me about audibility,if it was about that we would only care about SNR,gain staging,clipping and maybe IMD ,it's the visuals of measurements that's nice )

I thought the newer AKM / ESS chips (e.g. 4493S, 9039Q2M are better than the old 4493). No?

 

[Sokel]

I thought the newer AKM / ESS chips (e.g. 4493S, 9039Q2M are better than the old 4493). No?

You can see the comparison here:

AKM AK4493 vs. AKM 4493S

Sometimes AKM likes to be annoying for no reason and decides to make it hard to get one of their datasheets. So I got the one for the 4493S and made a 4493 vs. 4493S comparison in image formatfor anyone who's interested! It was released last year, and is used in the Topping DX1. It's also quite...

www.audiosciencereview.com

(I like the even older ones)

 

[sunjam]

You can see the comparison here:

AKM AK4493 vs. AKM 4493S

Sometimes AKM likes to be annoying for no reason and decides to make it hard to get one of their datasheets. So I got the one for the 4493S and made a 4493 vs. 4493S comparison in image formatfor anyone who's interested! It was released last year, and is used in the Topping DX1. It's also quite...

www.audiosciencereview.com

(I like the even older ones)

Hmm... I think I miss something. By looking at the comparison, it seems that 4993S is better than 4993. Could you highlight the difference to me that 4493 is better?

 

[Sokel]

Hmm... I think I miss something. By looking at the comparison, it seems that 4993S is better than 4993. Could you highlight the difference to me that 4493 is better?

In the reported datasheet it is,yes,by a small margin but it is.
Is more about visuals in spectrum,slopes,added "grass" in some implementations,also hints of ESS hump,etc.

Nothing wrong with either one in real world music listening of course,just measurbating

 

[drewdawg999]

Well I just replaced the DAC in my headphone rig with the SMSL D-6, which does indeed have DSD Direct. On my detailed DT770 with Dekoni Choice Leather pads, I can clearly hear a difference between the DSD Direct mode (Prd) and the DAC only mode (Pr0). For one thing it's softer (I believe 3.7V) but it's also more refined, maybe less detail in the high end but overall smoother, with a warmer and bassier presentation. I consider DSD Direct to be super analogue mode, an improvement over DAC only mode, and a killer feature of AKM chips. The great thing is that this is a cheap DAC that has this feature, it's right on the chip and is a cakewalk to implement. Downside is no EQ or volume control for DSD.

This is different from my previous DAC, the SMSL DO100 which uses an ESS chip. On that DAC the DSD gets all processed through all the same modulators that PCM gets so there's volume control. There is no difference on that DAC between DSD and PCM, and in addition the native DSD decoding can be glitchy, so I just converted DSD to PCM beforehand through Foobar to 24/176, and then applied EQ on all files. I stuck the DO100 back in the big stereo where it's on CD duty and Bluetooth for my phone. It remains superb for redbook and PCM, but so is the D-6 despite having the entry level Velvet chip. DSD Direct puts it over the top and has made the D-6 my main DAC.

 

[ppataki]

My D-6 shall arrive next week, I will also share my experience + some measurements

 

[antcollinet]

I will just use my UMIK-1 microphone in the main listening position and will perform sweep measurements in REW
That way not only the frequency response but everything else will be visible and comparable

There is no way that measuring in the audio domain will be accurate enough to detect any difference. By several orders of magnitude.

For a start any differences between PCM and DSD will be many 10s of dB below your room noise floor - even if the mic were sensitive enough to detect them (which it is not).

 

[antcollinet]

Well I just replaced the DAC in my headphone rig with the SMSL D-6, which does indeed have DSD Direct. On my detailed DT770 with Dekoni Choice Leather pads, I can clearly hear a difference between the DSD Direct mode (Prd) and the DAC only mode (Pr0). For one thing it's softer (I believe 3.7V) but it's also more refined, maybe less detail in the high end but overall smoother, with a warmer and bassier presentation. I consider DSD Direct to be super analogue mode, an improvement over DAC only mode, and a killer feature of AKM chips. The great thing is that this is a cheap DAC that has this feature, it's right on the chip and is a cakewalk to implement. Downside is no EQ or volume control for DSD.

This is different from my previous DAC, the SMSL DO100 which uses an ESS chip. On that DAC the DSD gets all processed through all the same modulators that PCM gets so there's volume control. There is no difference on that DAC between DSD and PCM, and in addition the native DSD decoding can be glitchy, so I just converted DSD to PCM beforehand through Foobar to 24/176, and then applied EQ on all files. I stuck the DO100 back in the big stereo where it's on CD duty and Bluetooth for my phone. It remains superb for redbook and PCM, but so is the D-6 despite having the entry level Velvet chip. DSD Direct puts it over the top and has made the D-6 my main DAC.

Come back when you've done the comparison blind and properly controlled.

Then there might be something worth discussing.

 

[Scytales]

Come back when you've done the comparison blind and properly controlled.

And level matched ! He wrote that the output voltage in "DSD direct" is lower !

By the way, even in "DSD Direct", the AKM chip is a DAC (Digital to Analogue Converter) ! The actual DAC stage, inside the AKM chip, is the switched capacitor filter (SCF) in wich DSD is directly inputted without having been previously digitally processed as in "non-DSD direct" mode. It is that very SCF which transforms the representation of the 1s and 0s of the DSD stream in actual usable voltages in the continuous (analogue) domain.

 

[antcollinet]

And level matched

That is part of the proper controls.

 

[drewdawg999]

Come back when you've done the comparison blind and properly controlled.

Then there might be something worth discussing.

But isn't it something that I prefer the softer signal? Don't people almost always gravitate towards the louder signal? Anyways I had to switch cables last night, the USB-C to C was producing static on 24-bit PCM. Now using a USB-C to A and that fixed it. But I tried the comparison again and it was far more subtle, could barely tell the volume difference. Still think it's a touch smoother in DSD Direct but am no longer confident I could pass a DBT. My mind could be playing tricks on me, but DSD somehow brings the happy super strong, perhaps on a subliminal level. But I for one am curious to see some mic measurements, hoping that a difference shows up in the waveforms.

 

[Scytales]

But isn't it something that I prefer the softer signal?

No. Everyone's taste is different.

But that's precisely the point of doing a level matched comparison. Since you prefer a softer signal, it is necessary to match the output level of the "non-DSD direct' signal processing to the softer level of the DSD direct signal processing. That way, you will hear both at the same softer level you prefer and, by eliminating the level difference, you will eventually hear or not hear an actual sound difference (if it exists) between the two modes of operation irrespective of levels.

Otherwise, the only conclusion you can draw is that you personnaly prefer to turn down the volume control just a little bit.

 

[antcollinet]

But isn't it something that I prefer the softer signal? Don't people almost always gravitate towards the louder signal? Anyways I had to switch cables last night, the USB-C to C was producing static on 24-bit PCM. Now using a USB-C to A and that fixed it. But I tried the comparison again and it was far more subtle, could barely tell the volume difference. Still think it's a touch smoother in DSD Direct but am no longer confident I could pass a DBT. My mind could be playing tricks on me, but DSD somehow brings the happy super strong, perhaps on a subliminal level. But I for one am curious to see some mic measurements, hoping that a difference shows up in the waveforms.

As I said upthread, microphonic measurements are orders of magnitude below the sensitivity needed to detect the microscopic differences you can get between different audio formats being discussed here. They cannot show anything.