Review and Measurements of SMSL SU-8 DAC

[SIY]

Since the deviation is negative, it is not noise. Here is the linearity using very narrowband FFT method:

Each peak needs to hit the corresponding horizontal line. We see deviations starting around -100 dB and it gets worse, always showing lesser amount. Not that I measure it past -120 but you can see it is all pretty wrong in that region.

Noise will not always show positive in a single scan. But the single tone stuff nails it. Many thanks.

 

[frogmeat69]

Thanks Amir for doing the review, I have been enjoying this DAC when I can for about a week, so far no problems with drivers or any glitches, and has been pleasant to listen to through both headphones and (not the greatest, lol) speakers.

 

[amirm]

I am glad you are enjoying it. It is always good to hear about people's experiences with what I review.

 

[t129]

@amirm Did you have time/interest to test the different sound colour settings available? I suppose SU-8 is one of the few (only one?) to implement the on chip dsp options of the Sabre for the customer to mess with but gives no information of the actual effects.

I think the previous question about them got lost in the noise. (Well, it was distortion conversation to be precise but that doesn't work as a pun.)

 

[amirm]

What measurement would you like to see there. Frequency Response?

 

[t129]

Massdrop description talks about adjusting the levels of distortion and making your music sound different. Which seems a bit far fetched to be honest (on making the music sound different part), but perhaps doing 1 kHz thd plots with couple of different sound "colours" could show what is what.

I don't think it makes sense to do a whole lot of measuring if the filters just make things "audibly" worse.

 

[pos]

Maybe these settings will have an effect on the IMD vs level hump that seem to show with all recent devices equipped with mobile ES9028/ES9038?

 

[pos]

Also Amir, I had to remind you about measuring one leg of the XLR to see if the unit might have common mode noise/DC problems.

 

[amirm]

@amirm Did you have time/interest to test the different sound colour settings available? I suppose SU-8 is one of the few (only one?) to implement the on chip dsp options of the Sabre for the customer to mess with but gives no information of the actual effects.

Doesn't look like it. Here is the manual:

As you see, they are the standard filter responses and nothing else.

 

[amirm]

Also Amir, I had to remind you about measuring one leg of the XLR to see if the unit might have common mode noise/DC problems.

I am sorry but I don't remember the context for this. What am I measuring and why?

 

[amirm]

Here are filters 1 through 7 response to white noise:

 

[t129]

As you see, they are the standard filter responses and nothing else.

Eh, not surprised but still slightly disappointed. Weird that 7 reconstruction filters have changed to 9 "sound colour settings". Marketing I suppose...

Thank you for looking into this anyway.

 

[Ruben G. Remus]

The "sound colour settings" are distinct from the filter settings. They are Standard (no "colour"), Rich 1, Rich 2, Rich 3, Tube 1, Tube 2, Tube 3, Crystal 1, Crystal 2, and Crystal 3. I don't have the unit with me, but if you repeatedly hit the round button at the center of the UP/DOWN/RIGHT/LEFT cluster in the remote you will get to the colour settings, then you can select the desired setting using the RIGHT/LEFT buttons. I will attach a picture from the D1 manual (the SU-8 appears to be a "poor man's" D1).

 

[pos]

I am sorry but I don't remember the context for this. What am I measuring and why?

Here is the context: https://audiosciencereview.com/foru...ing-audio-measurements.2351/page-2#post-89480

 

[yummy]

Maybe these settings will have an effect on the IMD vs level hump that seem to show with all recent devices equipped with mobile ES9028/ES9038?

bump in THD+n or IMD is the ESS "landmark", read more measurements, I know you are king guy in the JBL speakers forums, welcome to the new world, buddy, bravo, hats off

 

[Ruben G. Remus]

From the SMSL page for the D1 (http://www.smsl-audio.com/productsh2) : "
... DSP is programmable! D1 used this programmability and creatively designs a tone adjustment function! ... D1's tone adjustment is to adjust the harmonic content of the audio, unlike EQ, it will not affect the whole frequency response curve! The D1 is built in total of 10 modes! Among them, the Standard is straight through, unprocessed, and the others are Rich, Crystal and Tube. There are three of each type! Let users find the listening type that suits them as much as possible! Rich stands for rich harmonics, Crystal stands for clear voices, and Tube stands for unique tone of tube!"

Well, they sure like like their exclamation signs!

I gather the "colours" are adding euphonic (hopefully) distortion to the output. If you are going to do this, I guess this is the right way to do it (allow the user to turn it off).

To my 64-year-old-low-level-tinnitus-impaired ears, the effect is noticeable, but small (I only tried the "Tube" settings, and not for very long).

 

[amirm]

The "sound colour settings" are distinct from the filter settings. They are Standard (no "colour"), Rich 1, Rich 2, Rich 3, Tube 1, Tube 2, Tube 3, Crystal 1, Crystal 2, and Crystal 3. I don't have the unit with me, but if you repeatedly hit the round button at the center of the UP/DOWN/RIGHT/LEFT cluster in the remote you will get to the colour settings, then you can select the desired setting using the RIGHT/LEFT buttons. I will attach a picture from the D1 manual (the SU-8 appears to be a "poor man's" D1).

Thanks for the correction. I will go back and see if I can test those.

 

[amirm]

OK, I cycled through the settings in "color" and the frequency response does not change at all!

Looking at the distortion figures, all settings but standard lower SINAD by some 20 to 30 dB! In other words, it is working by changing the distortion profile by a significant amount.

Here is a 1 kHz tone response in standard/bypass mode:

And here is Rich1:

Notice the much higher 2nd and 3rd harmonic.

Here is Tube 1:

Notice how the second harmonic is accentuned a lot but not the 3rd as much.

Unfortunately it mutes for a couple of seconds between changes so it is hard to do an AB comparison.

Net summary, it is changing the transfer function of the DAC to mimic different distortion profiles. In that sense it is different than just an EQ.

 

[andreasmaaan]

@amirm I would be really interested to see if you can hear a difference between the different colour settings, and if so, how they sound different to you!

 

[Krunok]

OK, I cycled through the settings in "color" and the frequency response does not change at all!

View attachment 14200

Looking at the distortion figures, all settings but standard lower SINAD by some 20 to 30 dB! In other words, it is working by changing the distortion profile by a significant amount.

Here is a 1 kHz tone response in standard/bypass mode:
View attachment 14201

And here is Rich1:

View attachment 14202

Notice the much higher 2nd and 3rd harmonic.

Here is Tube 1:
View attachment 14203

Notice how the second harmonic is accentuned a lot but not the 3rd as much.

Unfortunately it mutes for a couple of seconds between changes so it is hard to do an AB comparison.

Net summary, it is changing the transfer function of the DAC to mimic different distortion profiles. In that sense it is different than just an EQ.

That seems as they are trying to make the sound "tube alike" using DSP!