• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Review and Measurements of the Topping D70 DAC

Joined
Jun 19, 2019
Messages
21
Likes
2
This measurement is what brought up the discussion. Whether the distortion above -12 dBFS is audible is another topic, though. Do you have particularly acute hearing?

View attachment 31197
You're kidding. I mean, the best sound is around -12dB, at -30db, -40db is not good and as described Leepc:
I also regret buying the D70.
I bought D70 because Grace m9xx(also $499) doesn't have XLR output. I thought D70 with XLR will drive Sennheiser HDVA 600 and Gustard H10 better than m9xx.
Unfortunately, I was wrong. Comparing to m9xx, D70 is lacking dynamic and definition at bass/mid.
D70 struggles to reproduce foot tapping and clapping on Eric Clapton Unplugged album on both my T1 and LCD2 headphones. It's not a volume matching issue. Its just D70 is lacking details and power comparing to m9xx.
I have a very sensitive amplifier for changes in the audio track and I ask. What impact does the sound have and why does the volume control have an effect?
 

maxxevv

Addicted to Fun and Learning
Joined
Apr 12, 2018
Messages
844
Likes
631
You're kidding. I mean, the best sound is around -12dB, at -30db, -40db is not good and as described Leepc:


I have a very sensitive amplifier for changes in the audio track and I ask. What impact does the sound have and why does the volume control have an effect?
I wouldn't be too quick to say that -30dB and -40dB isn't great. Just that the the lowest levels of distortion measured seems to max out at -12dB. Whether its obvious or not, I certainly am in no position to say. But setting the output at -12dB based on its measurements does make some sense.

As for volume, usually even if its just 0.5 ~2 dB difference in volume (depending on how sensitive the individual is), there is the perceived enhancement in sound of the source that is louder. So you can have 2 identical DACs playing that wee bit different in volume. For a blind test, there's a over 90% confidence that the louder one will be perceived to "sound better".

Hence, its pretty important to level match as close as possible.
 
Joined
Jun 19, 2019
Messages
21
Likes
2
I wouldn't be too quick to say that -30dB and -40dB isn't great. Just that the the lowest levels of distortion measured seems to max out at -12dB. Whether its obvious or not, I certainly am in no position to say. But setting the output at -12dB based on its measurements does make some sense.

As for volume, usually even if its just 0.5 ~2 dB difference in volume (depending on how sensitive the individual is), there is the perceived enhancement in sound of the source that is louder. So you can have 2 identical DACs playing that wee bit different in volume. For a blind test, there's a over 90% confidence that the louder one will be perceived to "sound better".

Hence, its pretty important to level match as close as possible.
I understand and make corrections, but why the color and tone change so much. At -12db is a very good DAC for $ 450.
 
Joined
Apr 4, 2019
Messages
35
Likes
50
But setting the output at -12dB based on its measurements does make some sense.
.
If I remember correctly, @amirm said that he is planning a project to test this? I'd imagine that, considering the chart I posted last page in post #733, the D70's SINAD jumps to elite levels and the dynamic range is still very large. If so, for those that plan on leaving the volume set at a single setting (like if connected to a amp with volume control), choosing ~-12 vol is a pretty easy little tweak to nudge the D70 from "great" to "among the best of the best."

This is the part of the review that led to my thinking about this:
We can see why SINAD suffers a bit because once the levels get close to maximum, more distortion sets in than best of the best DACs. If you operate the DAC using its volume control you would avoid this and get all the benefits of the D70.
 
Last edited:
Joined
Jun 19, 2019
Messages
21
Likes
2
If I remember correctly, @amirm said that he is planning a project to test this? I'd imagine that, considering the chart I posted last page in post #733, the D70's SINAD jumps to elite levels and the dynamic range is still very large. If so, for those that plan on leaving the volume set at a single setting (like if connected to a amp with volume control), choosing ~-12 vol is a pretty easy little tweak to nudge the D70 from "great" to "among the best of the best."
;)
 
Joined
Aug 13, 2019
Messages
3
Likes
1
That would seem highly odd that any AK4490 implementation would sound so drastically different from a AK4497 implementation and that you prefer the first over the supposedly more transparent one. Very odd indeed.
it is not odd. It's not personal preference.
It's a fact that m9xx is able to interpret the record more accurately compared to D70( both are about $500)

I know some people here only trust measurements and don't believe a human can hear the difference between DAC.
So If someone also has m9xx and D70, please listen to Eric Clapton's Walking Blues.
If it's hard to tell difference, Just focus on the simple foot-tapping.
The foot-tapping from m9xx sounds just like foot-tapping. D70 makes it sound like a drumbeat or a basketball hitting floor. Increasing volume for D70 can't help.

I tried both D70's RCA and XLR to drive HDVA 600 amp, also directly to JBL 306 speakers. All the same results.
Foot-Tapping still sounds more like drum beats on D70.

I just hope people don't make the same costly mistake I did. If you have m9xx, don't expect the newer chip and XLR on D70 to get you a better sound. It is not. Just enjoy your m9xx and save that $499 for feeding your family.
 
Joined
Aug 4, 2019
Messages
5
Likes
3
D70 makes it sound like a drumbeat or a basketball hitting floor. Increasing volume for D70 can't help.
I haven't m9xx to compare, but I can give some hints. Try to check AC polarity of power cord connection. D70 is sensetive to polarity, in wrong connection sound is a bit smoothed/blured (especially highs). Also make sure that the PCM filter is not 2, 4 or 5, they're affects bass presentation pretty much. Also if you're using foobar, try 32-bit ASIO mode, it reproduces low frequencies more percise. These are my conclusions after many experiments.
Listened to foot-tapping on ES9038 smartphone, there is more punch, bass is less fuller, but it really is recognized more like foot-tapping. I understand what you are talking about. Closest result on D70 is with PCM filter 6 applied (in DAC mode, not in PRE -12db), it adds density to sounds a bit in comparison with filter 3.
ADD: most dynamic and crisp sound - KS output (Kernel Streaming). Check it out.
 
Last edited:
Joined
May 31, 2019
Messages
24
Likes
11
Location
San Francisco Bay Area
I haven't m9xx to compare, but I can give some hints. Try to check AC polarity of power cord connection. D70 is sensetive to polarity, in wrong connection sound is a bit smoothed/blured (especially highs). Also make sure that the PCM filter is not 2, 4 or 5, they're affects bass presentation pretty much. Also if you're using foobar, try 32-bit ASIO mode, it reproduces low frequencies more percise. These are my conclusions after many experiments.
Listened to foot-tapping on ES9038 smartphone, there is more punch, bass is less fuller, but it really is recognized more like foot-tapping. I understand what you are talking about. Closest result on D70 is with PCM filter 6 applied (in DAC mode, not in PRE -12db), it adds density to sounds a bit in comparison with filter 3.
how can polarity go wrong given all plugs are three-pronged?
 
Joined
Aug 4, 2019
Messages
5
Likes
3
how can polarity go wrong given all plugs are three-pronged?
Ah sorry, I'm speaking about EU plugs. And it happens there is default computer cables with different pinouts. I'm using tester to be sure. This is wrong polarity on photo:
 

Attachments

Last edited:
Joined
Apr 24, 2019
Messages
24
Likes
8
So, as far as I could tell the following was never resolved:
Was the filter measurements done with a 48 kHz samplerate, or was it done with a 44.1 kHz samplerate yet still not reaching full attenuation until 24 kHz?
If it was the latter the Topping D50S would have better filters then.
 

3434

New Member
Joined
Jul 4, 2019
Messages
2
Likes
2
So, as far as I could tell the following was never resolved:
Was the filter measurements done with a 48 kHz samplerate, or was it done with a 44.1 kHz samplerate yet still not reaching full attenuation until 24 kHz?
If it was the latter the Topping D50S would have better filters then.
Measurement sample rate (not the generation sample rate) aside, after I asked that question earlier in this thread, I looked at more "linear fast" filters for ESS and AKM DACs and the spec sheet for AK4970, I am more convinced that it is typical for linear or minimum phase "fast" filters to attenuate over the nyquist frequency for 44.1 KHz these days. Seems unintuitive but I guess it is normal.
 
Top Bottom