Topping DX5II Balanced DAC and Headphone Amp Review

[ziggurcat]

I originally posted the following.
++++++++++++++++++++++++++++++++++++++++++++++++++
Issue 1 which is repeatable.
I use the device to replace a dbx equalizer. This means I use it primarily to equalize out a bass resonance in my Quad Esl electrostatics. Works like a charm. But of course I also exploit the extra option to use and equalize my HD 600 headphones in one go.
The strange issue occurs when I connect or disconnect the headphones while playing music over the loudspeakers (LO SE output). The second I pull the headphone plug the speaker sound drops away for a second or so. The other way around I when connect the headphones the speaker sound drops away for a second just when the headphone plug touches the metal of the 6,35 mm to 3.5 mm converter I use.
I suspect there must be something inside the DX5 II that is very sensitive to stray electric charges.

Issue 2 is not easily repeatable.
I have the Topping connected to a Maranz Streamer / Cd player via spdif. The Maranz is used a a digital selector, my record player plays over the Maranz via a minidsp pocketadc, my computer is connected via a 15 m ssnake optical spdif and when I need another input it is also connected via an ADC. The Maranz itself is a streamer / internet radio spotify endpoint etc. (This chain went from mostly analog into a digital system, only the power amp still is analog).
After this long preambule, the issue. When playing through the Topping every now and again suddenly the volume increases or drops by quite a margin for just one or two seconds. Not unlike the dropout in the headphone scenario, but it doesn't only drop out, the volume increases as well on occasions. There is no difference in connecting the Topping through coax or optical. I stress that this never happened with the amp connected to analog outputs of the Maranz itself. It's unique with the signal path through the Topping.
I suspect that both phenomena have a common origin, that might even have a connection with the usb issues
+++++++++++++++++++++++++++++++++++++++++++++++

I updated the unit to 1.93
After letting the unit play for 2 days, I can confirm that the SPDIF issues have dissappeared. No dropouts. no hick-ups anymore. Issue 2 seems fully solved.

Issue 1 however occurred once again during that period. It happened when I wanted change some settings. I picked up the unit actively playing. While I was busy doing that I I touched the 4 pin XLR socked with a finger while holding the unit in the palm of my hand and the music dropped out for about a second.
I already mentioned I consider this issue as a minor nuisance I can live with.

Thanks to all members that helped me with their thinking powers. I am sure now I am going to enjoy this little device.

Good to hear that updating the FW resolved one of your issues. I have not done anything about your other one yet, but will test as soon as I have a few moments to do that.

 

[Berwhale]

Is that something that could be added in a FW update (selectable between XLR and 4.4mm) or would that functionality require a HW revision?

I think it's likely that both the hardware and software would need to be changed to support independent activation of the XLR and Pentacon sockets. Maybe something for the Mk III

 

[Wolfram98]

Im on the latest 1.93

When switching C1 and C2 the USB drops out for a moment(can see in windows that no output is available, and Foobar2000 immediately throws tantrums), and when doing it often enough it would interfere with other devices on the USB controller, in particular my keyboard needed re-plugging after the DX5 II did its weirdness, so i just plugged it into a controller that doesn't have any other devices on it.

Is there a way to force the device to listen to the first input of the remote when the screen is off? that 1st remote input being just the "wakeup" is annoying.

The switching between EQ profiles seems very "slow" compared to other functions like switching inputs/outputs, you cant reliably and quickly scroll through them.

Using DT 880 250Ohm i would have expected the amp to be able to blow up the headphones, but even at High gain 0db volume the headphones seem to only start showing some noticable distortions.

And lastly, why did they include an IR remote in this day and age? a BT one would have been way more preferable.

 

[ziggurcat]

I think it's likely that both the hardware and software would need to be changed to support independent activation of the XLR and Pentacon sockets. Maybe something for the Mk III

At this point, even after ordering their new Mini 300 amp, the only way I'll "upgrade" my DAC/HPA is if it's a revamp of the MX5 all-in-one (and it shows itself to perform equally or better than the DX5 II/Mini 300 combo).

 

[Ziofrenko]

Using DT 880 250Ohm i would have expected the amp to be able to blow up the headphones, but even at High gain 0db volume the headphones seem to only start showing some noticable distortions.

Ehm...
First: It depends on the music and what source it comes from, if it's streaming make sure you turn off the noise leveling and listen to music that's playing loudly.
Second: There are headphones that according to the spec data can support half a watt per earpiece... From the dx5 250mW come out, and I repeat at 0dBFS.

 

[Wolfram98]

Ehm...
First: It depends on the music and what source it comes from, if it's streaming make sure you turn off the noise leveling and listen to music that's playing loudly.
Second: There are headphones that according to the spec data can support half a watt per earpiece... From the dx5 250mW come out, and I repeat at 0dBFS.

im using USB.

hmm, it feels as if my old Objective2 amp could achieve the same volume levels with these headphones.

 

[staticV3]

im using USB.

hmm, it feels as if my old Objective2 amp could achieve the same volume levels with these headphones.

With high impedance loads like the 250Ω DT880 using specifically the Differential output (4pin XLR/4.4mm Pentaconn), the DX5II can play 8dB louder than the JDS O2 Amp, which corresponds to 70% higher perceived loudness.
(17.3Vrms vs 7.0Vrms)

If you're comparing single-ended vs single-ended though, then the two are almost equal (8.7Vrms vs 7.0Vrms).
This corresponds to +2dB or +15% perceived loudness.

 

[Henreid]

Many thanks for this review, Amirm. The Topping DX5II appears to be the reasonably-priced audio component I have been hoping to find. Over the years I've amassed a ridiculously large collection of Classical CDs. I've been stuck in analysis paralysis over which headphone model to buy because I couldn't figure out a straightforward and cost-effective way to apply headphone equalization to the signal from my CD player. The DX5II should enable me to do that. There's also a very good chance that now I will finally act on your recommendation to buy the affordable Moondrop Para 2 headphones and optimize their performance with your equalization adjustments (so you won't have to resort to commandeering them due to my failure to benefit from your advice: "If you don't use EQ with the Para 2 I will personally come to your home and take it away from you!")

 

[Wolfram98]

With high impedance loads like the 250Ω DT880 using specifically the Differential output (4pin XLR/4.4mm Pentaconn), the DX5II can play 8dB louder than the JDS O2 Amp, which corresponds to 70% higher perceived loudness.
(17.3Vrms vs 7.0Vrms)

If you're comparing single-ended vs single-ended though, then the two are almost equal (8.7Vrms vs 7.0Vrms).
This corresponds to +2dB or +15% perceived loudness.

umm, the spec sheet of dx5 II says 24.4V at Gain=H and 7.2 at Gain=L for the 6.3mm jack.
or am i misinterpreting those numbers?

 

[staticV3]

umm, the spec sheet of dx5 II says 24.4V at Gain=H and 7.2 at Gain=L for the 6.3mm jack.
or am i misinterpreting those numbers?

You're not misinterpreting the numbers, you're misreading them.

The spec sheet doesn't say 24.4V, it says 24.4Vpp.

24.4Vpp = 8.6Vrms

 

[Henreid]

As we're approaching Chinese New Year (February 17, 2026) I'm wondering if China-based audio companies like Topping have ever offered discounts on their products during the holiday, similar to Black Friday sales. In previous years, has there been a dip in price on components like the DX5II?

 

[engin_ear]

Does anyone know how much latency the PEQ adds; has someone measured it? EPO is extremely efficient but also has the benefit of running on your main CPU. While I doubt PEQ running on the DAC adds much, I'm wondering if it's worth measuring (seems outside Amir's testing though) - like, is it 1ms+?

I don't know if anyone answered this, so I'll post it here. It depends on how they implemented the filter(s) in the digital domain. If IIR, the delay is basically 0, but IIR comes with some other nasties in terms of phase.
If FIR - let's assume a 101-tap filter. (Audio filters have a lot of taps to get the attenuation down to sub-100dB levels) The delay of this filter is 50 taps. At 44.1kHz sample rate (CD source assumed), the delay is 1/44.1kHz = 22.7us x 50 = 1.13ms.
So there's a few assumptions here, and you won't really know what it is exactly unless you measure it. Or you speak Chinese and have friends at the company.

They've built a fantastic unit here, especially for the price. I have one, love it.

 

[Wolfram98]

I have a weird "bug" with the 1.93 fw.
the first sound never goes through the headphones, It is noticable if you have very short message beeps, they simply dont make a sound if the system is completely silent for the last ~5s. And i can see that it receives a sound from the computer via USB , FFT analyzer ,VU, and Volume meter(On the DX5 screen itself) all show that there is sound, but the headphones are silent and i can hear the very silent end of the beep, as if they have some sort auto mute function.
It is both present on HPA and LO outputs separately.

 

[Mutex]

I have a weird "bug" with the 1.93 fw.
the first sound never goes through the headphones, It is noticable if you have very short message beeps, they simply dont make a sound if the system is completely silent for the last ~5s.

What you're experiencing is probably this: ZH3 cuts off first bit of audio (Less than 1 sec)

Discussion is from another unit, but the concept is the same. They probably mute audio for the first 200ms to avoid pops during wake-up. You can prevent that by keeping your unit awake at all times. On Linux there is a configuration you can do to achieve that. On Windows I believe someone mentioned a setting for that as well in here. On MacOS, you'll need to run an App that constantly sends zero streams to the unit to keep it awake.

 

[Wolfram98]

What you're experiencing is probably this: ZH3 cuts off first bit of audio (Less than 1 sec)

Discussion is from another unit, but the concept is the same. They probably mute audio for the first 200ms to avoid pops during wake-up. You can prevent that by keeping your unit awake at all times. On Linux there is a configuration you can do to achieve that. On Windows I believe someone mentioned a setting for that as well in here. On MacOS, you'll need to run an App that constantly sends zero streams to the unit to keep it awake.

Tried out Sound Keeper, seems to work in default config.
Thanks! i was already starting the refund process with audiophonics for this "quality" product.

Quick edit:
Updated the windows driver to 5.74 manually and it seems it doesnt do the automute anymore, without the help of soundkeeper.
Cant check if this setting was available from the getgo, but now it is defaulted and works flawlessly.

 

[ziggurcat]

Tried out Sound Keeper, seems to work in default config.
Thanks! i was already starting the refund process with audiophonics for this "quality" product.

Quick edit:
Updated the windows driver to 5.74 manually and it seems it doesnt do the automute anymore, without the help of soundkeeper.
Cant check if this setting was available from the getgo, but now it is defaulted and works flawlessly.
View attachment 503911

For the record, the data streaming popping isn't specific to Topping or the DX5 II, it is an issue with several DACs from several different companies. For Windows users, turning the DAC streaming to be always on (shown in the image you've attached) is the way around the problem. For Mac users -- since there are no drivers -- we have to find other solutions like running Qobuz in the background or the Keep My Mac Alive app from user @CuriousAli or putting a USB DDC in between the computer and DX5 II so that either SPDIF -- TOSLINK or coax -- inputs are used instead of USB.

Again, this is not a Topping-specific issue.

 

[Wolfram98]

For the record, the data streaming popping isn't specific to Topping or the DX5 II, it is an issue with several DACs from several different companies. For Windows users, turning the DAC streaming to be always on (shown in the image you've attached) is the way around the problem. For Mac users -- since there are no drivers -- we have to find other solutions like running Qobuz in the background or the Keep My Mac Alive app from user @CuriousAli or putting a USB DDC in between the computer and DX5 II so that either SPDIF -- TOSLINK or coax -- inputs are used instead of USB.

Again, this is not a Topping-specific issue.

i am coming from an O2 ODAC which worked flawlessly till this day, with no automute shenanigans.

what was particularly weird in topping instance is that the FFT display on the dac/amp itself showed the sound, but nothing came out of the outputs.

Need to check further, but also when turning off the DX5 II device(through the remote) the usb connection will still be kept alive, but after a while will disconnect from the PC, and Foobar2000 will be popping error windows due to no outputs available.

 

[MarioBoi]

Have gone through 2 units of this combo. Has anyone else experienced clipping during sub-bass heavy tracks when adjusting the volume? Both of my units have exhibited this behavior which is quite annoying

 

[Viper780]

Not yet.
Which tracks?
What headphones you are using?
what settings and how loud?

 

[MarioBoi]

Limit to your love by James Blake (0:55 when bass starts), HD6xx and normal listening volume (low gain and -15db)