Review and Measurements of Singxer SU-1 USB to S/PDIF

[JustIntonation]

I did come across this measurement on computeraudiophile from a holo spring, plain USB vs external I²S from a singxer SU-1
lower traces are using Spring USB input, upper using SU-1

You can see here the spring does dramatically better on jitter than the SU-1. I wasn't sure whether this meant the SU-1 wasn't very good, I²S is a bad interface or the spring USB is just that good...

Uhh the label states the bottom blue-green one is SU1 + Holo Spring DSD?
Also what kind of measurement is that? It claims to be a DAC measurement but a DAC measurement that has a noisefloor of less than -162.1dB? No way.

 

[Veri]

Uhh the label states the bottom blue-green one is SU1 + Holo Spring DSD?
Also what kind of measurement is that? It claims to be a DAC measurement but a DAC measurement that has a noisefloor of less than -162.1dB? No way.

It's jitter spectrum. And yes the labels are wrong.

 

[JustIntonation]

It's jitter spectrum. And yes the labels are wrong.

Measured how? The audiophilestyle.com thread this chart came from isn't very clear to me.
And it looks like noise? Except for the bump at 14k3 which seems to be i2s specific.
I'm not at all convinced this is a fair/well done comparison or that it is actually jitter that is shown?

 

[Veri]

Measured how? The audiophilestyle.com thread this chart came from isn't very clear to me.
And it looks like noise? Except for the bump at 14k3 which seems to be i2s specific.
I'm not at all convinced this is a fair/well done comparison or that it is actually jitter that is shown?

It's just a personal j-test of someone so yeah measurement set-up and tools used is not really obvious. iirc both use DSD256 which explains the absurdly low -162dB on USB.

 

[pkane]

Uhh the label states the bottom blue-green one is SU1 + Holo Spring DSD?
Also what kind of measurement is that? It claims to be a DAC measurement but a DAC measurement that has a noisefloor of less than -162.1dB? No way.

A 1m point FFT will give you the appearance of a very low 'noise' floor due to reduced FFT bin summing. But this wasn't a relative measurement of noise floors, in fact, the plots were simply manually offset from each other so you could see them on the same chart to compare jitter. If anything, SU-1 + Spring DAC over I2S produce about a 10dB+ higher noise floor, as I recall.

The measurement was done in REW with the little Forte ADC. J-test wave file was played by HQPlayer over DSD256.

I guess I should probably re-run some of these tests, since there are so many questions remaining. I found the SU-1 not to help with jitter with any of the modern DACs I have on hand, including two with I2S input. I did find that it raised the noise floor a bit on DSD content with Spring DAC. And by the way, this is not a stock SU-1. This had some of the most-recommended mods in it: a low-noise Sparkos voltage regulator, powered by a 6v battery, and a sorted NDK oscillator with very low measured phase noise.

 

[JustIntonation]

A 1m point FFT will give you the appearance of a very low 'noise' floor due to reduced FFT bin summing. But this wasn't a relative measurement of noise floors, in fact, the plots were simply manually offset from each other so you could see them on the same chart to compare jitter. If anything, SU-1 + Spring DAC over I2S produce about a 10dB+ higher noise floor, as I recall.

The measurement was done in REW with the little Forte ADC. J-test wave file was played by HQPlayer over DSD256.

I guess I should probably re-run some of these tests, since there are so many questions remaining. I found the SU-1 not to help with jitter with any of the modern DACs I have on hand, including two with I2S input. I did find that it raised the noise floor a bit on DSD content with Spring DAC. And by the way, this is not a stock SU-1. This had some of the most-recommended mods in it: a low-noise Sparkos voltage regulator, powered by a 6v battery, and a sorted NDK oscillator with very low measured phase noise.

Ok thank you for the explanation and measurements!

 

[pkane]

Here's today's measurement:

SU1 -> i2s -> Spring DAC = black
USB-> Spring DAC = red

Notice the difference in noise floor (higher with SU1), and slightly higher jitter sidebands without SU1:

 

[JustIntonation]

Thank you!

 

[amirm]

@amirm was the SU-1 tested with the Singxer ASIO driver or ASIO4ALL? Latest driver version is 4.59
And do you know the firmware of the SU-1? There has been an update a while ago. Latest version is 2.22
It seems to me in the world of USB converters these things can matter?
Spec wise the SU-1 seems to have quality parts / clocks, I'm surprised it doesn't score better.
Drivers and firmware update can be downloaded here: https://download.shenzhenaudio.com/Singxer/

The owner needs to advise as to what firmware was in it (I have sent it back to him already). For drivers, I have to look at my main system. You all need to ask me these questions when it is all fresh in my mind.

I don't see how drivers make a difference in electrical performance of these devices as opposed to functionality.

BTW, the unit I tested also had a regulator hack/upgrade.

 

[Veri]

So, I just received and tested this particular Singxer SU-1 unit which amir kindly tested. Since it is 110V I am using the "5V mod kit" and simply use my laptop USB Power to feed it.

It sounds just.. fine. Good. I don't think people should worry too much over DDC devices, a unit much cheaper would have sufficed too. But I wanted AES, currently waiting for the necessary digital 110Ohm cable to arrive. Using coaxical right now. I'm not using HDMI/I2S... nor do I intend to.

Anyway. This device works. It's kinda overpriced but I bought it used. I wouldn't say you should get rid of it if you own one. It's all right..

 

[R1200CL]

@amirm
Any plans of test the upgraded SU-2 ?
I sold my SU-1, and find the SU-2 much better. And it performs even better with an external clock.

 

[Dialectic]

@amirm
Any plans of test the upgraded SU-2 ?
I sold my SU-1, and find the SU-2 much better. And it performs even better with an external clock.

If it sounds better than the SU-1, then your DAC is broken.

If it sounds even better with an external clock, then the SU-2 is broken.

 

[R1200CL]

If it sounds better than the SU-1, then your DAC is broken.

If it sounds even better with an external clock, then the SU-2 is broken.

Any reason you’re saying this ?
Are you trolling ?

 

[Veri]

I was reading the iDSD Pro thread on head-fi, apparently the external clock sync gets lost very quickly. Then on 'computeraudiophile' etc there's reviews claiming such spectacular improvements, using that iDSD device with external clock. I like to imagine it was never in sync with the clock in the first place and 100% of everything written is pure imagination Thank iFi for buggy software/firmware making these things even more hilarious, ha...

Any reason you’re saying this ?
Are you trolling ?

No. Some interesing ASR threads to read...
https://www.audiosciencereview.com/forum/index.php?threads/master-clocks-snake-oil.7253/
https://www.audiosciencereview.com/forum/index.php?threads/auralic-leo-gx-dac-clock-review.11001/

 

[Dialectic]

Any reason you’re saying this ?
Are you trolling ?

Not trolling. These devices make an audible difference only when something inside the DAC is grossly inadequate.

 

[R1200CL]

Not trolling. These devices make an audible difference only when something inside the DAC is grossly inadequate.

This goes against what most people have discovered using various of DDC, clocks etc. Anyway we just have to disagree. My original question was to Amir, and if he would like to test and measure the SU-2.

 

[Veri]

Not trolling. These devices make an audible difference only when something inside the DAC is grossly inadequate.

Or when you need many devices slaved to a single clock perhaps, in a studio environment.

My original question was to Amir, and if he would like to test and measure the SU-2.

If you start a private conversation with @amirm he'll be happy to help you out if you want to send yours in or drop ship one that's what I did with this SU-1 which amirm measured!

 

[SIY]

This goes against what most people have discovered using various of DDC, clocks etc.

Can you point me to an example where a difference was determined? I.e., an actual experiment, not an anecdote?

 

[R1200CL]

Or when you need many devices slaved to a single clock perhaps, in a studio environment.


If you start a private conversation with @amirm he'll be happy to help you out if you want to send yours in or drop ship one that's what I did with this SU-1 which amirm measured!

As I’m not living in the US, I can’t.

 

[R1200CL]

Can you point me to an example where a difference was determined? I.e., an actual experiment, not an anecdote?

I think more or less the whole audiophilestyle site is about digital audio. I’m sure you’re able to find reviews and user comments about digital products that is highly regarded. There is no need for me to start a discussion here about the subject. Wrong tread as well.