Topping D90SE Review (Balanced DAC)

[Mad Bill]

I would take a shot and say it's could just interplay with your other component.

I noticed this with my rme back in the day. My cheaper dac had a better soundstage. I always guessed maybe the other is just exaggerating the soundstage. If a dac has better numbers across the board, even crosstalk, then maybe that's not the way the recording actual sounds and it's your other dac imparting a sound signature.

Looking at the other dac, if I am to believe the numbers, it's a pretty well performing dac as well (audibly transparent). Crosstalk with your nad is 123 and the d90se is 140 both at 1khz. I'm not sure if crosstalk is what would cause something to have greater soundstage, I've been trying to find out exactly it's audibility. But the d90se beats the nad in every measurable category, so it is strange.

I didn't live with the Topping for long enough to really say about the soundstage actually. And I could have been influenced by something I read-- though I think that psychological effect, if it occurred, would probably have been negated by longer-term experience (I know I'm getting into risky territory here--maybe, I'm new here--but I have reasons for thinking this).

 

[JohnYang1997]

Thanks for your reply and question. The NAD M51 is the only DAC I've had for a number of years, dating back to before my purchase of the M50.2. May I ask, is there anything about the output of the M50.2 that would be significantly different from a computer or other source? How are you looking at this question? (I don't have a high level of technical knowledge.)

There has been reported issue that some of our dacs(with cs8416 as digital receiver for AES, COAX, TOSLINK) when paired with some CD players. The mechanism is known now and D90se uses a different chip that is tested without issue. The cause was partially cs8416 can't lock on sources with large jitter. The same may happen to M50.2 here. It's possible the digital output of M50.2 is really poor so even lc89091 can't lock onto it.
I'll check on the implementation of M51 and see if there's something I can find. In the mean time, I can only recommend you to use other sources or other DAC.

 

[JohnYang1997]

Just found out it uses ak4118. Due to shortage we can't use it.
PS: Fuarrk M50.2 it's expensive....

 

[Mad Bill]

There has been reported issue that some of our dacs(with cs8416 as digital receiver for AES, COAX, TOSLINK) when paired with some CD players. The mechanism is known now and D90se uses a different chip that is tested without issue. The cause was partially cs8416 can't lock on sources with large jitter. The same may happen to M50.2 here. It's possible the digital output of M50.2 is really poor so even lc89091 can't lock onto it.
I'll check on the implementation of M51 and see if there's something I can find. In the mean time, I can only recommend you to use other sources or other DAC.

Wow, I would have never suspected that. Thanks very much.

 

[Marc v E]

My experience with the D90SE: I live in Taipei and bought from the distributor. I have been using an NAD M51, which I bought based on reviews and measurements -- I liked the 21-bit resolution -- and I've been happy with it for years. But the D90SE measures even better, so ... I was curious. But the first unit didn't work, and I exchanged it for another, which also didn't work. The sound would simply cut off for a second or two at random intervals -- sometimes after 10 seconds, sometimes after 5 minutes. My source is an NAD M50.2 server, and the problem continued with both AES and coaxial inputs, and also in both pre-amp and DAC modes. So I'm going to return the second unit today and get my money back.

My impression of the sound is that, yes, it's just a tiny, tiny bit more detailed and clean than the NAD M51. Bass weight and extension in particular is impressive. But there's not much of a soundstage; not much depth to it. The NAD M51 isn't necessarily known for that, but combined with the Buchardt Audio S300, it does pretty well. I just didn't hear the same soundstage with the Topping.

Any comments on the possible source of the above problem would be interesting.

My experience with the D90SE: I live in Taipei and bought from the distributor. I have been using an NAD M51, which I bought based on reviews and measurements -- I liked the 21-bit resolution -- and I've been happy with it for years. But the D90SE measures even better, so ... I was curious. But the first unit didn't work, and I exchanged it for another, which also didn't work. The sound would simply cut off for a second or two at random intervals -- sometimes after 10 seconds, sometimes after 5 minutes. My source is an NAD M50.2 server, and the problem continued with both AES and coaxial inputs, and also in both pre-amp and DAC modes. So I'm going to return the second unit today and get my money back.

My impression of the sound is that, yes, it's just a tiny, tiny bit more detailed and clean than the NAD M51. Bass weight and extension in particular is impressive. But there's not much of a soundstage; not much depth to it. The NAD M51 isn't necessarily known for that, but combined with the Buchardt Audio S300, it does pretty well. I just didn't hear the same soundstage with the Topping.

Any comments on the possible source of the above problem would be interesting.

I have a nad dac/pre C510. The innards of this unit are, as far as I can tell, exactly the same as the M51. Only the casing differs.

I never experienced a dropout over spdif with this unit but neither did I with the Topping dx7pro. I do know that the nad is very forgiving of poor sources, in fact the best I ever experienced.

Here is what I wrote down earlier after comparing the two. I can only explain our different experience by the effects of subjective reviewing. Https://www.audiosciencereview.com/...adphone-amp-reviewed.9446/page-71#post-789694

 

[gvl]

The sound would simply cut off for a second or two at random intervals -- sometimes after 10 seconds, sometimes after 5 minutes. My source is an NAD M50.2 server, and the problem continued with both AES and coaxial inputs, and also in both pre-amp and DAC modes. So I'm going to return the second unit today and get my money back.

It's a waste but you could try a jitter killer such as iFi SPDIF iPurifier in front of the DAC.

 

[DHT 845]

My experience with the D90SE: I live in Taipei and bought from the distributor. I have been using an NAD M51, which I bought based on reviews and measurements -- I liked the 21-bit resolution -- and I've been happy with it for years. But the D90SE measures even better, so ... I was curious. But the first unit didn't work, and I exchanged it for another, which also didn't work. The sound would simply cut off for a second or two at random intervals -- sometimes after 10 seconds, sometimes after 5 minutes. My source is an NAD M50.2 server, and the problem continued with both AES and coaxial inputs, and also in both pre-amp and DAC modes. So I'm going to return the second unit today and get my money back. My impression of the sound is that, yes, it's just a tiny, tiny bit more detailed and clean than the NAD M51. Bass weight and extension in particular is impressive. But there's not much of a soundstage; not much depth to it. The NAD M51 isn't necessarily known for that, but combined with the Buchardt Audio S300, it does pretty well. I just didn't hear the same soundstage with the Topping.

I owned NAD M51 for a few years and know its sound very well (a bit hazy and grey, bloated bass, wide soundstage, not so good depth, very nice musicality and liquidity). I think you should definately try Topping D70s. I am not sure what was sync issue cause in your case however but I suspect NAD M50.2 as culprit (high jitter and some chip incompability?)

 

[Marc v E]

I owned NAD M51 for a few years and know its sound very well (a bit hazy and grey, bloated bass, wide soundstage, not so good depth, very nice musicality and liquidity). I think you should definately try Topping D70s. I am not sure what was sync issue cause in your case however...

That comes very close to how I would describe the sound of my current nad dac c510...

 

[Marc v E]

Can anyone explain why the results in the Imd graph start at 60db while in the multitone test I see only noise under 120db or thereabouts? I'm probably missing something but I would like to know and learn.

 

[JohnYang1997]

Can anyone explain why the results in the Imd graph start at 60db while in the multitone test I see only noise under 120db or thereabouts? I'm probably missing something but I would like to know and learn.

-69dB is where the IMD graph starts at and it's at -60dB attenuation. If you add these two numbers together you get 129dB which is precisely the SNR at full level measured here. Every 1dB of attenuation deducts 1dB of SNR.

In multitone, the noise floor is half determined by the component itself and AP's own noise+distortion, half determined by FFT length. You gain more resolution in the graph(lower "noise floor") by increasing the FFT length and averaging in order to show remaining distortion.

This test is limited by AP's ADC which in itself is not powerful enough. The low THD+N is measured by notching out the fundamental frequency. This is the technique AP having used for long time, which gives around 30dB boost in performance.

 

[elvisizer]

Getting dropouts over toslink from my lg oled tv- never had that before with that device and toslink cable feeding other dacs

 

[gvl]

Dang, and this is after all those D30 Pro complaints. Seems Topping knows its opamps but not SPDIF receivers. There should be a way to relax the PLL at the cost of higher jitter which is a moot point given ASRC is always on, so perhaps provide a user setting that allows to run PLL at different levels to improve compatibility with bad sources in your next DAC @JohnYang1997. Will it be called D90SE Pro?

 

[JohnYang1997]

Dang, and this is after all those D30 Pro complaints. Seems Topping knows its opamps but not SPDIF receivers. There should be a way to relax the PLL at the cost of higher jitter which is a moot point given ASRC is always on, so perhaps provide a user setting that allows to run PLL at different levels to improve compatibility with bad sources in your next DAC @JohnYang1997. Will it be called D90SE Pro?

I have already tried on cs8416. We have relaxed the PLL BW of cs8416 by more than 10 times. No lock with the CD player that's reported to have issue before.
Lc89091 is tested with a CD player that's previously not lockable with D30pro but very stable with Lc89091.
The issue has nothing to do with the so called PLL that will introduce higher jitter. The DAC's internal DPLL will always be locked.
What's making me a little uncomfortable is that you really don't have any idea of this matter. Why do you talk like this??

 

[JohnYang1997]

Getting dropouts over toslink from my lg oled tv- never had that before with that device and toslink cable feeding other dacs

May I ask what other DACs do you have?

 

[gvl]

I have already tried on cs8416. We have relaxed the PLL BW of cs8416 by more than 10 times. No lock with the CD player that's reported to have issue before.
Lc89091 is tested with a CD player that's previously not lockable with D30pro but very stable with Lc89091.
The issue has nothing to do with the so called PLL that will introduce higher jitter. The DAC's internal DPLL will always be locked.
What's making me a little uncomfortable is that you really don't have any idea of this matter. Why do you talk like this??

CS8416 has been around for ages, there must be a solution thar supports every marginal source out there. Yes, I may not be technically deep into these things, but having multiple reports about this repeating issue in your flagship DAC is a turn off.

 

[JohnYang1997]

CS8416 has been around for ages, there must be a solution thar supports every marginal source out there. Yes, I may not be technically deep into these things, but having multiple reports about this repeating issue in your flagship DAC is a turn off.

Countless devices also have this issue. X16, fiio k5 etc.
Sadly there's simply no way to making cs8416 work for these cd players. And that's exactly why we changed the receiver. Part of the issue here is we can't find the worst spdif sources to test. If I don't see reports of some more affordable SPDIF sources that have the issue we may purchase the NAD. We have purchased over 10 different cheap SPDIF converters or CD players that they all work flawlessly with CS8416. We only found one device that has the issue on CS8416 and it's working very stable with LC89091.

 

[JohnYang1997]

Also if you think the PLL bandwidth in ESS dac has anything to do with this we can simply release a firmware that allows you to change the BW. We have managed to make firmware upgrade for every components in the DAC work. No need for next model. But reality is it doesn't matter.

 

[gvl]

Countless devices also have this issue. X16, fiio k5 etc.
Sadly there's simply no way to making cs8416 work for these cd players. And that's exactly why we changed the receiver. Part of the issue here is we can't find the worst spdif sources to test. If I don't see reports of some more affordable SPDIF sources that have the issue we may purchase the NAD. We have purchased over 10 different cheap SPDIF converters or CD players that they all work flawlessly with CS8416. We only found one device that has the issue on CS8416 and it's working very stable with LC89091.

Can't you guys use your AP to simulate incoming jitter on those inputs?

 

[JohnYang1997]

Can't you guys use your AP to simulate incoming jitter on those inputs?

Yes. Cs8416 works with 20ns of square wave jitter without issue. This is a lot of jitter.

 

[gvl]

Yes. Cs8416 works with 20ns of square wave jitter without issue. This is a lot of jitter.

Perhaps it's not a jitter issue but simply the optical receiver is not sensitive enough?