SMSL D-6s Balanced DAC Review

[ICIETDIYEUR]

What Amir offers us here is very useful: it allows us to know if a device is working properly and to see if what is said in the commercial links about a device is correct

That said:

would you keep a DAC only because it is very well 'classified' according to the 'SINAD references' but that it does not suit you (for some reason) in terms of sound reproduction ?

Again, the choice is yours

 

[AdamG]

Multiple off topic posts deleted and a few days off from this thread. If you can’t stay on topic this is the outcome. We would rather not but we have a responsibility to our members to keep things on topic. Otherwise every thread will turn into a macerator of you know what….

Please and thank you for your cooperation.

 

[Atgm]

Hello everyone I'm new to this topic, don't be too strict. I've been looking for a balanced dac lately and recently ordered an smsl d6s, but I came across information that it has one es9039 chip. Do I understand correctly that in order for me to have a truly balanced system, I need to look for a dual-chip DAC? Please enlighten me

 

[somebodyelse]

Hello everyone I'm new to this topic, don't be too strict. I've been looking for a balanced dac lately and recently ordered an smsl d6s, but I came across information that it has one es9039 chip. Do I understand correctly that in order for me to have a truly balanced system, I need to look for a dual-chip DAC? Please enlighten me

It depends what you mean by 'a truly balanced system', but probably not. Assuming the es9039 is similar to the es9038q2m it has differential outputs.

 

[JIW]

Hello everyone I'm new to this topic, don't be too strict. I've been looking for a balanced dac lately and recently ordered an smsl d6s, but I came across information that it has one es9039 chip. Do I understand correctly that in order for me to have a truly balanced system, I need to look for a dual-chip DAC? Please enlighten me

No, a single one is enough for two symmetrical outputs.

 

[staticV3]

Do I understand correctly that in order for me to have a truly balanced system, I need to look for a dual-chip DAC?

You do not.

A single ES9039 chip natively supports balanced output:


All that dual DAC chips achieves is reducing noise by 2-3dB.

E.g:

 

[Atgm]

You do not.

A single ES9039 chip natively supports balanced output:
View attachment 440929

All that dual DAC chips achieves is reducing noise by 2-3dB.

E.g:
View attachment 440930

Thanks a lot guys for the detailed explanation. I've got it all figured out now

 

[Atgm]

I forgot to clarify. I have a balanced Matrix Audio HPA-3B amplifier and now I use the audinst hud dx1 in conjunction with it. I'm going to connect the d6s to my amp via the xlr ports and output music via the balanced headphone output. Is this a balanced scheme? And one more question. My DAC has an es9018 and burson vivid v6 operational amplifiers. How much is the new sabre chip superior to the old one? Perhaps I shouldn't have started this upgrade?

 

[nanook]

I forgot to clarify. I have a balanced Matrix Audio HPA-3B amplifier and now I use the audinst hud dx1 in conjunction with it. I'm going to connect the d6s to my amp via the xlr ports and output music via the balanced headphone output. Is this a balanced scheme? And one more question. My DAC has an es9018 and burson vivid v6 operational amplifiers. How much is the new sabre chip superior to the old one? Perhaps I shouldn't have started this upgrade?

Whether or not this HPA indeed operates fully differentially internally is hard to tell.
(If I would design an Amp with XLR input and balanced output (and not just a connector that allows a balanced HP being plugged in - which btw. would be absolutely equally (actually better in terms of noise and distortion) good if max. power is not needed))

BUT: As @staticV3 pointed out the D6s has a DAC chip with differential outputs, the D6s has a differential I/V stage (the RCA output takes the XLR output and subtracts the signals to obtain a single ended output).

The only advantage of 2 DAC chips is lower noise just as staticV3 wrote. And these DACs are so good nowadays, that a few dB are not at all relevant when listening music (maybe if you attenuate 50dB or more in the digital domain. And even in that situation, the noise level of the listening environment will likely be dominant.)

Regarding ES9018 vs. ES9039q2m:
The ES9039q2m is a really advanced chip with an improved Sigma-Delta modulator - measurement wise clearly a progress but I would very much doubt if someone can tell a good ES9018 design and a good ES9039 design apart in a blind listening session.

I'm not familiar with the Burson OpAmp substitutes and thus I cannot tell if they eventually generate a distinguishable "sound". The OPA1612A used in the D6s is out of suspect to audibly alter the sound when implemented properly - as is the case for the D6s.

Hope this helps a bit.

 

[nanook]

I do find serious blind listening very strenuous (with my tinnitus gaining audibility), so I usually take a bit of a shortcut.
With both signal paths levelled to within 0.1dB using a multimeter and a 1 kHz sine wave (and I'm serious about the 0.1dB) I quickly switch back and forth until I don't know which source is playing.
Then I "guess" which DAC is playing, then look which one it really was and note the results (guess and fact). I do this say 8 or 10 times and when I'm close to 50% I cannot distinguish the two - for sure not when just enjoying music.
Maybe "golden ears" might be able to tell the two candidates apart - I simply cannot.
;-)

 

[Atgm]

I do find serious blind listening very strenuous (with my tinnitus gaining audibility), so I usually take a bit of a shortcut.
With both signal paths levelled to within 0.1dB using a multimeter and a 1 kHz sine wave (and I'm serious about the 0.1dB) I quickly switch back and forth until I don't know which source is playing.
Then I "guess" which DAC is playing, then look which one it really was and note the results (guess and fact). I do this say 8 or 10 times and when I'm close to 50% I cannot distinguish the two - for sure not when just enjoying music.
Maybe "golden ears" might be able to tell the two candidates apart - I simply cannot.
;-)

Thank you very much. Your opinion is very important to me.

 

[audiojack]

For a few days I use this DAC for active (monitor) speakers. With A-B testing I've compared this SMSL DAC against the internal DAC of the active speakers.
The SMSL increases the microdetail and makes the soundstage/staging bigger.

But somehow I wonder if it colors the sound? For example vocals sound more lively/more emotion on the SMSL but somewhat further away (depth) and there seems to be a certain sound characteristic that seems to be added and it's apparent especially with vocals they all seem to get the same undertone both male and female vocals, but also all the instruments. Is this normal? Is this because I'm A-B testing against the internal DAC that I'm used to? Or does my SMSL device has an issue? (I bought it used)

If I listen those same songs on the TV speakers on my Sony TV the sound tones seems more similar to the internal DAC of the active speakers, and the SMSL seems to add something texture/coloration. It's sort of like dripping caramel over all the foods you eat, the caramel taste will be everywhere.

 

[Nutul]

Are you sure you listen to the exact same volume level, down to 0.1dB...?

 

[HalSF]

For a few days I use this DAC for active (monitor) speakers. With A-B testing I've compared this SMSL DAC against the internal DAC of the active speakers.
The SMSL increases the microdetail and makes the soundstage/staging bigger.

But somehow I wonder if it colors the sound? For example vocals sound more lively/more emotion on the SMSL but somewhat further away (depth) and there seems to be a certain sound characteristic that seems to be added and it's apparent especially with vocals they all seem to get the same undertone both male and female vocals, but also all the instruments. Is this normal? Is this because I'm A-B testing against the internal DAC that I'm used to? Or does my SMSL device has an issue? (I bought it used)

If I listen those same songs on the TV speakers on my Sony TV the sound tones seems more similar to the internal DAC of the active speakers, and the SMSL seems to add something texture/coloration. It's sort of like dripping caramel over all the foods you eat, the caramel taste will be everywhere.

These mildly anxious concerns about perceiving coloration and a morphing receding soundstage and undertones and vague distortion analogous to caramel dripped on food and a vague sense of wrongness seem like exactly the kind of emotionally tinged things that subjective projections can trick your brain into feeling/hearing. It’s what people convinced that cables and interconnects have a distinctive “sound” experience all the time when swapping wire. I’d be very interested in learning what happens if you could enlist a confederate in helping you do blind switching between the SMSL and your speaker’s DAC so you have to listen without knowing which DAC you’re hearing. This feeling that something is wrong with the SMSL may disappear.

And as the comments above notes, level matching is a crucial consideration.

 

[audiojack]

Are you sure you listen to the exact same volume level, down to 0.1dB...?

I matched the volume by ear by A-B switching between them and by focusing on bass and treble volume, because I don't have a measurement mic nor a dB meter. Besides that the frequency profile of the SMSL DAC and the active speakers DAC is noticeably different, and since the loudness of different frequencies are not perceived equal by the human ear I wonder if dB matching would help. Yes the loudness of all the frequencies combined can be matched but if one of the DACs has way more frequencies in the range that the human ear is most sensitive to this one would be perceived as louder hence "better"? Or is my thinking flawed?

 

[audiojack]

These mildly anxious concerns about perceiving coloration and a morphing receding soundstage and undertones and vague distortion analogous to caramel dripped on food and a vague sense of wrongness seem like exactly the kind of emotionally tinged things that subjective projections can trick your brain into feeling/hearing. It’s what people convinced that cables and interconnects have a distinctive “sound” experience all the time when swapping wire. I’d be very interested in learning what happens if you could enlist a confederate in helping you do blind switching between the SMSL and your speaker’s DAC so you have to listen without knowing which DAC you’re hearing. This feeling that something is wrong with the SMSL may disappear.

And as the comments above notes, level matching is a crucial consideration.

Like I said I can hear more microdetail with the SMSL so it's not that I think it's a flawed but just wonder if the perceived coloration can be explained otherwise, yes of course it can be explained by brain perception fair. Maybe my concern is also because I bought it used (returned product) and think something could be wrong with it. Maybe it's just that the active speaker DAC sounds too thin in a certain frequency range which adds less texture/richness (caramel) to the sounds, and too bright in others which brings vocals more forward than the SMSL.

 

[Nutul]

You play a sine tone at 1, or 2, or 4 kHz, then with a cheap multimeter measure the output at the speaker's terminal. It's enough for testing purposes. No need for fancy tech.

 

[nanook]

@audiojack: Which active speakers are you talking about?
I cannot rule out, there might be some kind of equalization in the signal path when using the internal DAC - but it's very, very unlikely they do. An active speaker has to sound the same, no matter which input is used.

As @Nutul proposed, the definite proof is to measure the AC voltage present on the terminals of the chassis. 1kHz should be woofer (or midrange in case of 3-way), everything above 4 kHz is tweeter.
But: most multimeters cannot really measure beyond ca. 400 Hz. For comparing the levels using internal DAC vs. external DAC you can as well use e.g. 100 Hz. I really don't expect either of the DACs to have a frequency response with more than 0.x dB in the audio band.

In case you have a really good multimeter available or you have access to an oscilloscope, you can check if the D6s has a constant output level across the audio band.
Source could be the freeware tool REW. It has an easy to use sine generator.

Do NOT play frequencies above "woofer" (a few 100 Hz) via your speakers! With unintended high levels the tweeter or midrange may die.
I just meant checking the D6s with nothing connected other than the oscilloscope or DMM.

It's as well extremely unlikely that the D6s shows a response deviation in the audio band.
I rather think that your brain is fooling you, just like @HalSF wrote.

 

[Esprit]

because I don't have a measurement mic nor a dB meter

Do you have a cell phone?
It's full of apps.
They may not be accurate if you want to know the exact dB, but now you only need to measure two identical volumes.

 

[nanook]

Do you have a cell phone?
It's full of apps.
They may not be accurate if you want to know the exact dB, but now you only need to measure two identical volumes.

And the position of "everything" (including you) must be identical.
I'd measure "near field" i.e. have the mic of the phone close to the membrane (several cm, less than 20cm), such that reflections are not so relevant.