SMSL DO100, Unit to unit variation?

[Wendigo79]

Hello!

I have been wonderin about unit to unit variation in SMSL dacs. I cannot escape the feeling that my DO100 sounds different when I use it trough USB or Toslink. If I use USB, the sound if "dry, hard, compressed and loud". So it sounds like having some distortion. It sounds still OK, but if I use Toslink and my wiim mini streamer, it sound more spacious, cooler, more detailed and relaxed. The dirrerence is immediatelly noticeable. I have tried it with my windows PC and RPI/moode combo. Also with volumio and so on. Every possible software. Result is always the same. Output from the computer if of course bit-perfect and all ok.

I have had the SMSL DO100 dac for about two years now and I did back then when I bought it already notice these things, but I have since used the dac with wiim mini and toslink. But now I started to play with these things again and noticed same results than back then. Amirms review of the DO100 shoved practically identical results with toslink and usb. So could my unit have some flaw in it?

I also have wondered does Qobuz use different files for PC/Linux than with streamers? I once studied it by inspectin the bitrate with Foobar2000 on windows and bubbleupnp on android, but had very similar results. So no dynamic compression on Qobuz on different platforms, I guess.

I have been looking for a firmware update for the DO100, but there has been none. For the purpose of fixing the possible USB-communication problems.

So has anyone else had similar things happening? Perhaps my DO100 has something wrong going on in it's USB input...

 

[Roland68]

Maybe there's something wrong with the USB output? Some computers have very dirty USB outputs.
For example, have you ever made a comparison with a Macbook running on battery power? Or another computer?

 

[Wendigo79]

I have used it with laptop, with and without battery. RPI and Intel NUC. I dont think that is the reason. Also it does not hum or whine etc

 

[Roland68]

I have used it with laptop, with and without battery. RPI and Intel NUC. I dont think that is the reason. Also it does not hum or whine etc

It has absolutely nothing to do with audible humming or whining. It's just about contamination of the data stream and the 5 volt line.
Remember that with USB there is no error correction, verification, etc. in this case.

 

[Roland68]

I have used it with laptop, with and without battery. RPI and Intel NUC. I dont think that is the reason. Also it does not hum or whine etc

I just saw that this DAC still has an XMOS XU208 chip.
It is significantly more sensitive to such influences than the newer generations such as the XU216 and XU316.

 

[DVDdoug]

it sound more spacious, cooler, more detailed and relaxed.

Nobody knows what those words mean. You can have distortion, noise, and frequency response errors... That's pretty much it with electronics. You could get distortion if the digital stream is isn't really "bit perfect" and the digital is boosted & clipped. Otherwise you're not going to get audible distortion from a non-broken DAC. There is always SOME noise (usually background hiss from the analog side). You could also get clicks or crackles if if there are glitches-disruptions in the data stream. Frequency response is almost always better than human hearing.

of course bit-perfect and all ok.

Level matching is usually more important than "bit perfection" when doing an A/B or ABX listening tests.

I also have wondered does Qobuz use different files for PC/Linux than with streamers?

I'm pretty sure they don't know what you've got.

So no dynamic compression on Qobuz on different platforms, I guess.

None of the popular streaming services apply dynamic compression. But you can't necessarily compare different services. They don't all use the same master, sometimes not even the same mix or the same recording. Sometimes a new re-mastered high-resolution file may be more dynamically compressed than the original master.

Most of them use volume matching, where the volume is adjusted before the track starts, but they may not all use the same volume target.

 

[mhardy6647]

Some computers have very dirty USB outputs.
For example, have you ever made a comparison with a Macbook running on battery power? Or another computer?

This (as they say)
BTDT (as they also say)

Had all kinds of issues with the laptop I use to stream music from our NAS. Sounded fine on battery power, but all kinds of bad on AC. Turns out it was the cheap aftermarket (brand X) AC power supply I was using. Bought a gen-u-ine Lenovo P/S for it and all was well.

EDIT: after I posted (sorry!), I see that @Wendigo79 claims no difference on AC vs battery. I didn't see that he tried a different 'pute, or even different USBs on the same 'pute(?).

If you wonder why the aftermarket P/S -- this is an ancient IBM (!) R90 thinkpad that I bought used and cheap -- from the I/T support "guy" (one man shop/company) I use for my consultancy. Twas he who bundled the cheapie P/S with it -- which didn't matter until I tried to use the USB -> DAC for playing music.
I resurrected the lappie a few years ago and installed Linux on it.

 

[Roland68]

This (as they say)
BTDT (as they also say)

Had all kinds of issues with the laptop I use to stream music from our NAS. Sounded fine on battery power, but all kinds of bad on AC. Turns out it was the cheap aftermarket (brand X) AC power supply I was using. Bought a gen-u-ine Lenovo P/S for it and all was well.

EDIT: after I posted (sorry!), I see that @Wendigo79 claims no difference on AC vs battery. I didn't see that he tried a different 'pute, or even different USBs on the same 'pute(?).

If you wonder why the aftermarket P/S -- this is an ancient IBM (!) R90 thinkpad that I bought used and cheap -- from the I/T support "guy" (one man shop/company) I use for my consultancy. Twas he who bundled the cheapie P/S with it -- which didn't matter until I tried to use the USB -> DAC for playing music.
I resurrected the lappie a few years ago and installed Linux on it.

In terms of hardware, the old Lenovos were in the same league as HP Business Books and MacBooks, but there's nothing you can do about a modest power supply other than replacing it

 

[mhardy6647]

In terms of hardware, the old Lenovos were in the same league as HP Business Books and MacBooks, but there's nothing you can do about a modest power supply other than replacing it

Oh, I love 'em (the Thinkpads, that is). I've been using them, purchased used, as work computers for years.
The funny thing about the R90 is -- I am not really quite sure of its provenance! I think I bought it for my wife to use on an interim basis when her workhorse lappie at the time (a Dell) had developed some sort of issue (which may have been a broken hinge that took it to the precipice of rapid unscheduled disassembly).
I found the R90 in a closet looking for something else one day a few years back. It is the only IBM/Lenovo lappie here (and there are... several) with an aftermarket P/S, as it happens.



EDIT: Sorry for the derailment -- now back to the actual thread topic!

 

[Dunring]

Test your USB ports with this if you have Android. It's an EMF detector and will tell how much is going on moving the phone near each one. I only use the Samsung ETA0U60JBE chargers for DACs that can use it, measures as clean as the Apple chargers. I was surprised on our systems how much more noise was on some ports than others.

 

[somebodyelse]

I just saw that this DAC still has an XMOS XU208 chip.
It is significantly more sensitive to such influences than the newer generations such as the XU216 and XU316.

What's your evidence for that? The XU208 and XU216 are the same generation and very similar, the 216 having more cores than the 208 for applications that need the extra processing power like decoding MQA.

 

[Roland68]

What's your evidence for that? The XU208 and XU216 are the same generation and very similar, the 216 having more cores than the 208 for applications that need the extra processing power like decoding MQA.

It is a finding from several extensive tests on the impact of USB cables, which we verified with a USB analyzer.
While the influence of heavy contamination on the USB interface could also be audibly affected on the XU208, this effect could no longer be clearly verified on the XU216/316. The result was the same with different DAC and DDC.
But it was just a coincidence that it was noticed and of course it could also be due to other factors.

It is up to each person what they do with this information. Just ignore it if you don't like it.

 

[Wendigo79]

Nobody knows what those words mean. You can have distortion, noise, and frequency response errors... That's pretty much it with electronics. You could get distortion if the digital stream is isn't really "bit perfect" and the digital is boosted & clipped. Otherwise you're not going to get audible distortion from a non-broken DAC. There is always SOME noise (usually background hiss from the analog side). You could also get clicks or crackles if if there are glitches-disruptions in the data stream. Frequency response is almost always better than human hearing.


Level matching is usually more important than "bit perfection" when doing an A/B or ABX listening tests.

I'm pretty sure they don't know what you've got.


None of the popular streaming services apply dynamic compression. But you can't necessarily compare different services. They don't all use the same master, sometimes not even the same mix or the same recording. Sometimes a new re-mastered high-resolution file may be more dynamically compressed than the original master.

Most of them use volume matching, where the volume is adjusted before the track starts, but they may not all use the same volume target.

I tried to describe "compressed" sound quality. For me the difference was strong and I think Im able to hear it genuinely. They use different files for example Chromecast on TV. Instead of flac they send i guess 256/AAC instead. Just like Spotify does. It uses .ogg for android and pc,but AAC for chromecast. What suspect is that inside of the XMOS-usb thing something happens and sound gets somehow "messed up".

But all this is history now as my DO100 died today. I just moment ago was fiddling with it again. Connected to RPI3/moode and was again comparing things (I know... ). Then the dac started to turn off by itself. Just shut down. I disconnected all cables and the power cord also. Waited for a while and reconnected. Same thing. Turned itself off. With or without any other cables than the power cord Then I found that the coil whine it always had, had also disappeared. So i guess the PSU failed. Well it lasted almost two years So i had to dig my topping E30 back to action. Searching now for a new balanced dac. E30 has only rca

 

[Wendigo79]

It has absolutely nothing to do with audible humming or whining. It's just about contamination of the data stream and the 5 volt line.
Remember that with USB there is no error correction, verification, etc. in this case.

I think that the DO100 dac would drop the signal completely if it would be interfered somehow. When I use toslink with tv it sometimes drops the signal because of too much jitter or something. The DPLL setting has to be above 8 and it the works. So I think that if the USB-signal would be messed up the dac would drop it.