Review and Measurements of Topping DX3Pro DAC and Headphone Amp

[Wolven]

I just tested Roon with DX3Pro and there is no clicking at all when seeking. Ditto for switching songs, etc. So this is something to do with foobar2000.

My Foobar2000 does not do that either.
Yuno are you using 1.4 beta 6?
It is very stable.

 

[confucius_zero]

I'm sorry if this has been discussed before but
1- is the ESS hump audible? if so, what does it really do?
2- does the DX3Pro's solution to the ESS Hump make it audibly disappear?
3- since the DX3Pro is able to kill the ESS Hump but lose in other metrics, is this loss noticeable and worth it over the D10 and the D50? If noticeable, should we wait for Topping to adjust to the new DAC chip?

Thank you!

 

[amirm]

1- is the ESS hump audible? if so, what does it really do?

We don't know. There is some rise in distortion in mid levels. It is not something I have noticed in listening to a lot of DACs with this issue.

2- does the DX3Pro's solution to the ESS Hump make it audibly disappear?

Topping changed DAC chips from ESS to AKM and hence the reason the hump is no longer there. It is a new design.

3- since the DX3Pro is able to kill the ESS Hump but lose in other metrics, is this loss noticeable and worth it over the D10 and the D50? If noticeable, should we wait for Topping to adjust to the new DAC chip?

The reason to get DX3Pro is if you need a headphone output which neither the D10 or D50 have. It also has Bluetooth if you need it.

Topping is working on a high-end dac, D70, using the same chip slated to come out spring next year.

 

[confucius_zero]

Thank you very much Amirm.

Is there an article on this forum that explains how each measurements is weighted in the final recommendation verdict (say how much % is prioritized on the "SINAD" performance vs the "intermodulation distortion versus level" performance) to help us get some perspective, say on the ESS-Hump ESS devices (The topping ESS implementation) vs less-ESS-hump ESS devices (like the SMSL M8)?

 

[amirm]

Is there an article on this forum that explains how each measurements is weighted in the final recommendation verdict (say how much % is prioritized on the "SINAD" performance vs the "intermodulation distortion versus level" performance) to help us get some perspective, say on the ESS-Hump ESS devices (The topping ESS implementation) vs less-ESS-hump ESS devices (like the SMSL M8)?

Not really. It is usually my opinion of whether there are any design mistakes and overall performance.

 

[Toku]

Hi there. I played with it and figured it out.

Unplug the unit and hold the volume knob down and plug power back in. You should see F-X where X is your filter setting. Push in and hold until you see 88. Let go and it will go dark and power/down. I think this is the reset.

When I operated according to the advice, 8 - 8 was displayed and it was OK.
Amir Thank you very much.

 

[Yuno]

My Foobar2000 does not do that either.
Yuno are you using 1.4 beta 6?
It is very stable.

1.4.1. But this is not foobar specific issue, it was also happening in firefox.

I'm also getting quite a lot of clicks by using citrix. I connect remotely to work via citrix receiver and that too seems to produce clicks quite frequently. Not sure why because sampling rate is always 192kHz.

So far I noticed that each message that I get via Lync is triggering a click.

Set sound to 32bit/44.1kHz. Turned off exclusive mode in windows settings. Now the only sampling rate it can use is 44.1kHz. Still clicking... It's switching from 44.1 to 44.1.

 

[Wolven]

1.4.1. But this is not foobar specific issue, it was also happening in firefox.

I'm also getting quite a lot of clicks by using citrix. I connect remotely to work via citrix receiver and that too seems to produce clicks quite frequently. Not sure why because sampling rate is always 192kHz.

So far I noticed that each message that I get via Lync is triggering a click.

Set sound to 32bit/44.1kHz. Turned off exclusive mode in windows settings. Now the only sampling rate it can use is 44.1kHz. Still clicking... It's switching from 44.1 to 44.1.

Sorry to hear that. When I had an issue, I just turned off the computer and unplugged DX3pro and then everything was fine. Perhaps a reset could be a good idea. Contacting Topping could be an idea.

 

[Yuno]

Sorry to hear that. When I had an issue, I just turned off the computer and unplugged DX3pro and then everything was fine. Perhaps a reset could be a good idea. Contacting Topping could be an idea.

Sadly I already tried to reset the unit, and tested it on 2 different devices. If this entire constant resampling thing is not a default behavior then my unit is defective. If it is default behavior then this unit is defective by design. Either way this is not behavior I've ever seen before on dac or soundcard.
I don't think I'm even going to bother contacting topping, if it's defective then it's straight dispute via aliexpress.

 

[vafan13]

sooo, if I have no need for the amp section, do I want the D50 or the DX3Pro? I read quite extensively through stuff on the D50, but then saw this thread.

 

[Wolven]

Sadly I already tried to reset the unit, and tested it on 2 different devices. If this entire constant resampling thing is not a default behavior then my unit is defective. If it is default behavior then this unit is defective by design. Either way this is not behavior I've ever seen before on dac or soundcard.
I don't think I'm even going to bother contacting topping, if it's defective then it's straight dispute via aliexpress.

Yeah, in your situation I can see that being annoying. For my purposes it is fine. I have only had one external DAC prior to DX3pro. The Audioengine D1 would click in and out with WASAPI, and was unable to function with it. It worked with windows driver.

I'm curious about DX3pro experience of other people. Anyone?

 

[Yuno]

Ok, here is an interesting update. Somehow clicking stopped, I changed output device in foobar2k to primary sound driver and other than being stuck with 44.1kHz I can hear no clicks whatsoever. Now it's behaving how I would expect dac with static sampling rate to behave. Interesting thing is that I already tested something like this before (except I chose DS: Speakers DX3-Pro instead of primary sound driver) and it didn't do much. By the way, this is without topping driver, which I uninstalled.
I guess for now I'm happy that my dx3 pro is done clicking. It seems that in the end foobar wasapi implementation was the culprit here. I'm not really sure how exactly since I had it stopped most of the time, especially when I was working.

 

[Wolven]

Ok, here is an interesting update. Somehow clicking stopped, I changed output device in foobar2k to primary sound driver and other than being stuck with 44.1kHz I can hear no clicks whatsoever. Now it's behaving how I would expect dac with static sampling rate to behave. Interesting thing is that I already tested something like this before (except I chose DS: Speakers DX3-Pro instead of primary sound driver) and it didn't do much. By the way, this is without topping driver, which I uninstalled.
I guess for now I'm happy that my dx3 pro is done clicking. It seems that in the end foobar wasapi implementation was the culprit here. I'm not really sure how exactly since I had it stopped most of the time, especially when I was working.

Did you try going to properties for sound in windows, shared mode can be switched to different setting, and should go up to 24/192.
Personally, I don't like this solution as everything gets upsampled. I'm not clear what is processing the sound in this setup, is it DX3pro or is it the computer chip?

 

[Yuno]

Did you try going to properties for sound in windows, shared mode can be switched to different setting, and should go up to 24/192.
Personally, I don't like this solution as everything gets upsampled. I'm not clear what is processing the sound in this setup, is it DX3pro or is it the computer chip?

Yeah, I prefer to downsample to 44.1 rather than upsample. Majority of my music is 44.1 to begin with, and benefits of so called hi res music are basically none for human beings, and my pets don't have much interest in audio to begin with.
Bigger benefit of WASAPI is complete lack of any processing on PC side, everything is handled DAC side. Without exclusive mode sound is first processed by Windows, then by DAC. In theory Windows shouldn't change the sound unless you explicitly pick some filter or such, but who knows.
It's one of the things I would love to see measured honestly, WASAPI vs no-WASAPI.

 

[bleachershane]

First time poster, have been considering a USB DAC for an age (always eyeing up the Topping D30) but also needed a headphone amplifier. Noticed Shenzen Audio were selling through Amazon UK and that some of their items were fulfilled by Amazon (including the D50).
Stumbled on the DX3 Pro and it ticked every box for me, USB DAC, seemingly great headphone amplifier and as a bonus can be used as a preamplifier (I'm considering converting one of my old Cyrus One amplifiers into a pure power amplifier and I think I could get away with all digital sources through the DX3 Pro). Sadly the DX3 Pro isn't fulfilled by Amazon yet so have to wait for shipping from China, £144.99 (about $185 in today's conversion rates). No idea what will happen with customs and taxes, fingers crossed!

Really looking forward to this as an upgrade from my Focusrite Solo being my PC based listening system. Hope it's an audible upgrade!

 

[bona998]

HA! I have just grabbed last unit on audiophonics. They sold out all units in 1 day ! Let's see now what is this unit capable of by using own ears :> (couldn't find brand new aune S16 and RME seems to be too expensive for me). It is a bit pricey here but with official audiophonics 2 years warranty and without a lot of troubles when sending back to China or for repair.

Are there some drawbacks of this unit ? Like dying D50 chips / regulators or some radiators/thermal grease that are mounted wrong etc ? Has anyone had some issues with dx3pro yet ?

 

[amirm]

sooo, if I have no need for the amp section, do I want the D50 or the DX3Pro? I read quite extensively through stuff on the D50, but then saw this thread.

There is no clear distinction between them. The D50 is superior in low level detail. The DX3 Pro has less intermodulation distortion. I like the fact that DX3 Pro has a remote.

 

[bona998]

Damn ;/ TOSLINK performance regarding jitter suppression is not too good - checked it out too late ; / I was looking at coaxial performance earlier as SPDIF. Still worth a try, I can always send it back if I don't like it. For great jitter reduction and all in one combo for headphones when using TOSLINK, there will be only RME or Benchmark I guess and these cost much more.

Or maybe I will just use noisy USB connection, hoping that they put a bit of effort to filter/isolate things out there.

 

[bellyofthebizd]

Damn ;/ TOSLINK performance regarding jitter suppression is not too good - checked it out too late ; / I was looking at coaxial performance earlier as SPDIF. Still worth a try, I can always send it back if I don't like it. For great jitter reduction and all in one combo for headphones when using TOSLINK, there will be only RME or Benchmark I guess and these cost much more.

Or maybe I will just use noisy USB connection, hoping that they put a bit of effort to filter/isolate things out there.

Toslink is a 35 years old tech - always choose coax

 

[bona998]

Toslink is a 35 years old tech

Hahaha, so what ? It is still used in all audio equipment and should be handled well.

Only USB has nice results here. For both - toslink & coax since there is poor jitter rejection (coax a bit better but still far away from usb performance). I would have to use some third party usb->spdif asynchronous converter which doesn't make sense at all price-wise overall. I want to be fully separated from PC (galvanic separation). Don't care how old is toslink, only light (opto-transistors + light) can provide full separation although new generation transformers coupling do the job too on COAX but ... I just don't want it, don't want any galvanic connection that can cause any degradation of performance like ground issues/loops, interfering with HF noise from PC etc It can degrade total performance of the unit significantly depending on the audio chain & environment).

Have a look at jitter rejection in other DACs tests. If this is nicely implemented, there is perfect suppression regardless of digital input being used.

The question is if it's audible or maybe incoming jitter from toslink output of high-end mobo powered by seasonic won't be so polluted. I guess amir's test is the worst case scenario signal when testing jitter.

Let's just wait for the audition of the unit, in this price range I am curious how it competes with aune x1s.