SMSL Sanskrit 10th MK II DAC Review

[Vini darko]

Thanks for your tests. It's strange. Now I managed to force it to 768 mode abusing a resampler plugin in foobar2000. So maybe it's a windows limitation, but then why are different values in the manual?...
I don't have much experience in DACs. I have a NAD C 356BEE amplifier with integrated DAC, and I can't say the Sanskrit is better. It sounds a little different for sure, but based on the reviews I expected it to be much better.

All dacs sound the same apparently welcome to asr.

 

[csadam]

All dacs sound the same apparently welcome to asr.

This was bothering me, so I experimented a little, and I found that it sounds better on optical. (I used the NAD on optical and Sanskrit on USB).

It was strange, so I've set up an experiment:
1. play the same music on Sanscrit and record it in Audacity to different tracks. First on optical, next on USB.
2. precisely align the beginning of the tracks
3. invert one of the tracks
4. mix the tracks to a difference track.

In theory, when the two tracks are the same, the difference track will be silence.
But in practice it turned out that something is clearly wrong with my setup


When I compared the tracks, the difference was huge, much more than I expected. But not because they sound so much different, but because there is a timing issue with the USB (or with the optical) output. Even when the start of the tracks are aligned bit-precise, a few seconds later one track gets shifted, and the shifting will be more and more by the time.

In the following days, I will try to find out why this happens. I feel I'm getting too deep into this digital audio rabbithole

 

[Vini darko]

This was bothering me, so I experimented a little, and I found that it sounds better on optical. (I used the NAD on optical and Sanskrit on USB).

It was strange, so I've set up an experiment:
1. play the same music on Sanscrit and record it in Audacity to different tracks. First on optical, next on USB.
2. precisely align the beginning of the tracks
3. invert one of the tracks
4. mix the tracks to a difference track.

In theory, when the two tracks are the same, the difference track will be silence.
But in practice it turned out that something is clearly wrong with my setup
View attachment 86612View attachment 86613

When I compared the tracks, the difference was huge, much more than I expected. But not because they sound so much different, but because there is a timing issue with the USB (or with the optical) output. Even when the start of the tracks are aligned bit-precise, a few seconds later one track gets shifted, and the shifting will be more and more by the time.

In the following days, I will try to find out why this happens. I feel I'm getting too deep into this digital audio rabbithole

Nice objective test. Might be jitter wich is clocking errors. Here's the difference in the sanskrit from amirm's data.

 

[Vini darko]

The difference is significant. I think it's because usb is error checking so manages to track accurately much easier than toslink. As you spend more money on dacs (chord for example) these problems get engineered out.

 

[csadam]

The difference is significant. I think it's because usb is error checking so manages to track accurately much easier than toslink. As you spend more money on dacs (chord for example) these problems get engineered out.

I prefer to keep myself on the sane side of audio stuff Jitter or not, it simply does not seems right. This would mean I can finish an album a few seconds earlier on USB than on toslink. It's like playing a 29.97fps video at 30fps, something must be wrong somewhere. I will test different settings, USB ports, cables, (re)sampling rates, etc.

 

[Vini darko]

I prefer to keep myself on the sane side of audio stuff Jitter or not, it simply does not seems right. This would mean I can finish an album a few seconds earlier on USB than on toslink. It's like playing a 29.97fps video at 30fps, something must be wrong somewhere. I will test different settings, USB ports, cables, (re)sampling rates, etc.

Yeah I'm not keen on spending money on dacs either. Look forward to your further tests getting to the bottom of this issue.

 

[harry501]

Thanks for your tests. It's strange. Now I managed to force it to 768 mode abusing a resampler plugin in foobar2000. So maybe it's a windows limitation, but then why are different values in the manual?...
I don't have much experience in DACs. I have a NAD C 356BEE amplifier with integrated DAC, and I can't say the Sanskrit is better. It sounds a little different for sure, but based on the reviews I expected it to be much better.

If you didn't like the DAC then it could give you an out for a refund. Just show them the manual and tell them the box shouldn't have come with "V2" sticker stuck on it.

 

[AnalogSteph]

I prefer to keep myself on the sane side of audio stuff Jitter or not, it simply does not seems right. This would mean I can finish an album a few seconds earlier on USB than on toslink. It's like playing a 29.97fps video at 30fps, something must be wrong somewhere. I will test different settings, USB ports, cables, (re)sampling rates, etc.

I would suggest looking out for a software frequency counter, and having an ADC of known-decent frequency accuracy wouldn't hurt. Toslink input would have to be following the source sample rate, while in USB mode the clock is likely to be derived from an internal oscillator with a PLL. Do test different sample rates as well.

It is not necessarily the DAC that's at fault, it might be your Toslink source as well. That said, the clock deviations I've seen even in Realtek onboard audio chips have been quite small and non-concerning. (I think I was using a 10 kHz sine and letting RMAA do a high-res FFT analysis, where even fairly small deviations will show up.) Not necessarily always the case though:
http://nwavguy.blogspot.com/2011/02/sansa-clip-measured.html

 

[3dcnc]

I have just ordered. I am not clear if an external Linear power supply will solve jitter issues on toslink /optical connection? I will do listening tests myself.

 

[dess]

I recently acquired the MKII. After 20 hours of break in, I tested the unit for a couple hours but unfortunately my preamp went off (and I’m waiting to be fixed).
For the few A/B comparisons I did with 16/44 cd (A from a dvd player > analog rca > preamp and B the same dvd player as transport > coaxial digital > Sanskrit MKII > analog rca > preamp), I heard no significant difference with 16/44 cd material.
Anyone tested Sanskrit MKII as a transport dac with 16/44?

Edit: Anyone tested Sanskrit MKII with a transport and 16/44?
Sorry, it didn't make sense..

 

[vert]

All dacs sound the same apparently welcome to asr.

Then at least your integrated's DAC is competent, it isn't always the case! In my case early on I noticed music sounded a lot cleaner with an external DAC on my Marantz AVR. Amir's reviews of Marantz products have confirmed their average performance since then. I have the Sanskrit as a second DAC for my headphones system by the way, to me it's a perfect product.

 

[tvrgeek]

Well, based on this review, I bought one. Sending it back .
USB input. USB powered.
First of all it has residual artifacts at 538 and 1076 of roughly -88 dB. I do not get these at all from my old MUSE DAC or the Focusrite Scarlett 2i2. (also USB powered) Zip. Not even from my Woofer Tester used as a DAC.
From 50 Hz to 6K, predominantly second order, -70 dB. Third @ -95
Linearity is fine.
Noise floor is below my bench. Using a Focusrite 2i2. My loopback noise floor runs from about -120 @ 10 Hz to about -95@ 24K. Distortion runs in below the floor from about 500 up.

This could possibly be USB power line induced. If so, it is still going back as it is no where near the .0002 spec.

 

[3dcnc]

Have not tried USB imput- arrived today. Optical sounds fantastic! better than co-axial (more tests needed) . Using Ifi linear power supply. best DAC I own - I think!. Also own - modi 3 , Pre Box digital S2. Runing off a Douk U2.

 

[3dcnc]

Well, based on this review, I bought one. Sending it back .
USB input. USB powered.
First of all it has residual artifacts at 538 and 1076 of roughly -88 dB. I do not get these at all from my old MUSE DAC or the Focusrite Scarlett 2i2. (also USB powered) Zip. Not even from my Woofer Tester used as a DAC.
From 50 Hz to 6K, predominantly second order, -70 dB. Third @ -95
Linearity is fine.
Noise floor is below my bench. Using a Focusrite 2i2. My loopback noise floor runs from about -120 @ 10 Hz to about -95@ 24K. Distortion runs in below the floor from about 500 up.

This could possibly be USB power line induced. If so, it is still going back as it is no where near the .0002 spec.

I assume you are not using the supplied USB cable. There seems to be complaints related to this. - Based on your measurements , I think your knowledge is 1000 x more than mine. So i might be wrong.

 

[Eddie_Vanjovi]

Some people have been posting here about lag? Will that hold true if used for gaming? How does one fix? (appears people swapping power supplies does the trick?)

 

[3dcnc]

Google took me to the wrong website looking for drivers. I eventually found a the correct link

https://www.smsl-audio.com/portal/product/downlist/id/11.html

 

[tvrgeek]

And what does this driver do? There was nothing in the instructions ( 2 point font if I could read it) and Windows had no difficulty detecting it, detecting format etc. Worked as expected after the "cal" function. Except failing to meet spec or even expectations which were lower.

A USB cable will not magically generate artifacts on the output.
Seems unlikely a power supply would effect lag as that is a architecture result.
It is advertised and sold to work without external power unless used on a portable device that does not meet USB power spec, such as a phone.

 

[3dcnc]

The driver gets the Xmos chip to work with your PC. I first could not get the official driver to work. You need to put the dac in a "usb line in" mode. I think it is line 1.

 

[3dcnc]

And what does this driver do? There was nothing in the instructions ( 2 point font if I could read it) and Windows had no difficulty detecting it, detecting format etc. Worked as expected after the "cal" function. Except failing to meet spec or even expectations which were lower.

A USB cable will not magically generate artifacts on the output.
Seems unlikely a power supply would effect lag as that is a architecture result.
It is advertised and sold to work without external power unless used on a portable device that does not meet USB power spec, such as a phone.

I am not electrically proficient. My limited experience, is that linear power does make a small difference.The xmos chip with the driver should give you clean usb signal. On my setup with Dali speakers , which is very transparent the DAC is very clear on USB. I think this review on youtube confirms the info I gave.

 

[3dcnc]

I was wrong . you need to get the official driver to work with the unit. The way i got it to work is plug the DAC in the PC get windows to recognise it. remember to have the DAC in USB in mode. then install driver .Xmose Control panel shows device. The sound is a little more clearer with the driver installed. driver attached