Review and Measurements of Topping D10 DAC

[Krunok]

Yep it's Austria, near the Slovenian border, so a 3 hour drive. Feel free to drop me a note when you are around.

Will do!

 

[Krunok]

Btw, as GAS is also catching on me I'm looking at this little guy: http://matrix-digi.com/en/products/122/index.html

Take a look at these impressive specs! http://matrix-digi.com/en/specifications/115/index.html

 

[Michael Guntsche]

If you really want to satisfy your GAS go with a RME Adi-2 Dac.

 

[Krunok]

If you really want to satisfy your GAS go with a RME Adi-2 Dac.

Impressive device! But, as is the case with all professional devices, it's WAF falls somewhere between "no way" and "only over your dead body"!

WAF = woman acceptance factor

 

[Guermantes]

As I said, I cannot think of a single technical reason why would anyone prefer DSD format instead of PCM.

If you look at the history of DSD, you can perhaps understand why it was a good idea at the time.

In the mid-1990s Sony archival audio engineers decided that it was a better idea to tap the 1-bit stream of the A-D converters they were using (Delta Sigma designs) before the decimation stage (conversion to LPCM) and record this as the archival master. This decision was based on the A-D design and potentially meant that the new master was "purer" than the LPCM output of the same A-D converter. Conversion to other formats such as CD, Hi-Res PCM, etc., could be done at a later stage. Also DAC technologies at the time were a mirror of the ADC, so it was possible to keep this "pure" 1-bit path all the way back to the analogue domain.

Sony took the idea of the "purity" of the 1-bit stream and with Philips developed the SACD format as a "better-than-CD" alternative. However a couple of problems developed: editing and mixing in native DSD was difficult so it was really only suited to archival use or as a final master; ADC and DAC designs changed and the "pure" 1-bit path became compromised by various (superior) multi-bit Delta Sigma alternatives.

So DSD was based on '90s ADC technology but today things have moved on. It is better and "purer" to use modern LPCM technologies and formats. In my mind, Blu-Ray Audio is the optical format of choice, especially for multi-channel recordings, but it, too, is sadly neglected in favour of streaming-ready formats.

I have a lot of SACDs -- a release on the format is usually an indicator that there has been attention to detail in recording and mastering. The Living Stereo SACDs are a point in case, as are releases by Alia Vox (Jordi Savall), Channel Classics, Chandos, and some others. Some of them are gorgeous in multi-channel so I hope that multi-channel flac or streaming equivalents will become more available.

 

[amirm]

So DoP only up to 128

This is correct. Using Roon, I can play DSD256 using WASAPI/DoP with Topping D50. With Topping D10, Roon automatically downsamples DSD256 to DSD128 for playback.

 

[Michael Guntsche]

......

I have a lot of SACDs -- a release on the format is usually an indicator that there has been attention to detail in recording and mastering. The Living Stereo SACDs are a point in case, as are releases by Alia Vox (Jordi Savall), Channel Classics, Chandos, and some others. Some of them are gorgeous in multi-channel so I hope that multi-channel flac or streaming equivalents will become more available.

For me the main problem was availability. I had a SACD compatible player back then and I bought some SACDs, but the music I was interested in was not available as SACD. As a result I bought some albums would normally not listen to just because they were available which ultimately ment that I did not listen to them at all. Yes multi-channel was very nice to listen to and I liked the mix but I did not like the music.

 

[Krunok]

If you look at the history of DSD, you can perhaps understand why it was a good idea at the time.

In the mid-1990s Sony archival audio engineers decided that it was a better idea to tap the 1-bit stream of the A-D converters they were using (Delta Sigma designs) before the decimation stage (conversion to LPCM) and record this as the archival master. This decision was based on the A-D design and potentially meant that the new master was "purer" than the LPCM output of the same A-D converter. Conversion to other formats such as CD, Hi-Res PCM, etc., could be done at a later stage. Also DAC technologies at the time were a mirror of the ADC, so it was possible to keep this "pure" 1-bit path all the way back to the analogue domain.

Sony took the idea of the "purity" of the 1-bit stream and with Philips developed the SACD format as a "better-than-CD" alternative. However a couple of problems developed: editing and mixing in native DSD was difficult so it was really only suited to archival use or as a final master; ADC and DAC designs changed and the "pure" 1-bit path became compromised by various (superior) multi-bit Delta Sigma alternatives.

So DSD was based on '90s ADC technology but today things have moved on. It is better and "purer" to use modern LPCM technologies and formats. In my mind, Blu-Ray Audio is the optical format of choice, especially for multi-channel recordings, but it, too, is sadly neglected in favour of streaming-ready formats.

I have a lot of SACDs -- a release on the format is usually an indicator that there has been attention to detail in recording and mastering. The Living Stereo SACDs are a point in case, as are releases by Alia Vox (Jordi Savall), Channel Classics, Chandos, and some others. Some of them are gorgeous in multi-channel so I hope that multi-channel flac or streaming equivalents will become more available.

I agree with you about the history reasons. Still, I was not able to find more than a few SACDs with the music I like. I'm afraid the same destiny that happened to DVD Audio awaits Blu-Ray Audio as well. But frankly, I don't mind much if that happens as new recordings sound better and better even in CD quality and there are a lot of remasters been done lately which greatly improve the sound quality of older recordings, so things are actually sounding very good these days.

 

[Guermantes]

But frankly, I don't mind much if that happens as new recordings sound better and better even in CD quality and there are a lot of remasters been done lately which greatly improve the sound quality of older recordings, so things are actually sounding very good these days.

I wish I could share your optimism! I think we have the technical potential to make great recordings and produce impressive remasters but much of that potential is sadly squandered . . .

 

[Krunok]

Amir should now have the OPA1612 I sent him, mounted in such a manner. However, as bhigh suggested:



It could be for naught. Perhaps Amir can give us some feedback on this in the future, before you spend your money on something that may be pointless.

As stated here TI is actually suggesting to replace OPA2134 with OPA1642 as both of them are FET designs, while OPA1612 is bipolar, so some compatibility issues may arise.

https://e2e.ti.com/support/amplifie...gn_notes/upgrading-op-amps-in-audio-equipment

 

[Krunok]

Let's look at harmonic distortion where the Behringer has been an excellent champion at:

View attachment 11521

Ah, we discover something interesting. The Behringer doesn't show a lot of harmonic distortion because its noise floor is much higher. The Behringer beats it by as much as 25 db. By doing so, its harmonic distortion is visible but they are all lower than what the Behringer produces (note: this noise is the combination of the DAC and my analyzer). Net, net the Topping D10 has far better performance even though its graph looks more "busy."

@amirm - Can you please check if I'm reading this graph correctly? Is this graph showing 3rd harmonic distortion at app -107db which corresponds to app 0.00045%?

 

[amirm]

@amirm - Can you please check if I'm reading this graph correctly? Is this graph showing 3rd harmonic distortion at app -107db which corresponds to app 0.00045%?

Correct.

 

[Krunok]

Correct.

I find that truly fascinating, especially when taking into account that DAC that managed to show such performance costs less than $100.

 

[Sythrix]

As stated here TI is actually suggesting to replace OPA2134 with OPA1642 as both of them are FET designs, while OPA1612 is bipolar, so some compatibility issues may arise.

https://e2e.ti.com/support/amplifie...gn_notes/upgrading-op-amps-in-audio-equipment

Hmm. The Muses 8920 is the only FET op-amp I sent. I have replaced an opa2134 with an lme49720 before and suffered no ill effects. I don't believe it will cause damage... maybe oscillations in a bad scenario, but that is up to Amir to decide on for risks.

I would be interested in whether or not it caused a noticeable difference, given that it's being fed by another opa2134 anyways.

 

[PyramidElectric]

Finally got one of these, very impressive for the money, thanks for doing the review!
I was able to compare with a Chord Mojo, using the D10's optical out to feed the Mojo, and the D10's line outs. Once level matched I couldn't hear any difference whatsoever, maybe the rest of my system isn't transparent/revealing enough (Naim Nait5si with B&W DM602 speakers) but I don't really care, it all sounds good!

 

[minh]

Are you able to play DSD256 in DoP mode or in Native?

hi I was outa town for work. I play DSD256 in DoP mode

 

[Krunok]

Hmm. The Muses 8920 is the only FET op-amp I sent. I have replaced an opa2134 with an lme49720 before and suffered no ill effects. I don't believe it will cause damage... maybe oscillations in a bad scenario, but that is up to Amir to decide on for risks.

I would be interested in whether or not it caused a noticeable difference, given that it's being fed by another opa2134 anyways.

No, it won't cause damage, in worst case you will have noisy and/or distorted output.

 

[Krunok]

I play DSD256 in DoP mode

Are you sure?

 

[Krunok]

Anybody tried installing ASIO driver? It would be nice to know how that works..

 

[DataX]

I've installed the driver on Windows 10 1703 and ASIO works fine in foobar2000 with PCM and both DoP / Native DSD (using DSD Transcoder plugin). I only have a couple of DSD64 and DSD128 samples, so I haven't tested DSD256 DoP vs Native.

I was also pleasantly surprised with the driver. Typically I dislike installing any type of drivers and prefer to stick to UAC/UAC2 audio, however Topping's is very small (~2 MB) and installs with no fuss. The only thing it leaves behind is a simple control panel (which you can optionally keep in the system tray) to change some settings like 24/32-bit output, USB streaming latency, ASIO buffer size and internal output volume. I've attached a screenshot.

--

On another note though, I've noticed an odd quirk with my unit which might be related to issue that Michael reported, which is that the D10 appears to be extremely sensitive to ESD. I had caught it resetting a couple times while moving stuff around and figured I had bumped a cable or something, however I noticed this also appears to happen with even the slightest amount of ESD when touching the case.

When this happens, I can hear a relay click and the display lights up (like when changing sample rates). Presumably the relay is on the RCA output as audio playback cuts out for a split second and then resumes. The USB device does not disconnect though and playback in foobar2000 never stops.

I've been able to reproduce this a couple times, but not every time.