Review and Measurements of Topping DX3Pro DAC and Headphone Amp

[Veri]

Haha, thanks. I have been pretty excited for this review and completely missed it

It was posted 2 hours ago..

 

[Pavano]

this seems to be a better overall package than the D50. I even somewhat regret now that I got one a few weeks ago

 

[lucadoc]

I've already the metal remote for the Dx7s (that finally I didn't buy).

Topping as confirmed to me that it can replace the plastic one given with the DX3 Pro.... I think I've found my Dac

 

[derp1n]

In addition, the output impedance of it is lower than 0.1 ohm by design. The retail version may have an even lower value.

Or maybe not.

 

[Sythrix]

It was posted 2 hours ago..

He was quoting my reponse to his price inquiry, which was a quote-only post. It told me in the notifications, but for whatever reason, didn't show up above his post.

 

[Soniclife]

Yes, I forgot to mention that this also has Bluetooth input. Will see if I can test it a bit.

That will be very interesting, this has such good basic and predictable normal performance that any badness introduced will be obvious.

When you test Bluetooth are you able to point asio4all at the laptops Bluetooth output and run the full suite easily?

 

[rajapruk]

Topping seems to be an very attentive and agile company, being able to switch to AKM-chips this fast and still deliver a top notch product.
Do you think they will switch to AKM-chips in their next generation top-shelf dacs? I sure hope so (if not ESS solves the problem).

 

[derp1n]

@amirm you be measuring a production unit once they are available? This could be a hand picked review unit for all we know.

 

[Deleted member 3566]

Maximum output power at 32ohm was what, just over 0.5W? That's half of what they promised?

 

[mindbomb]

Maximum output power at 32ohm was what, just over 0.5W? That's half of what they promised?

Many times, companies will cite the amount of power at 1% thd. With the full graph though, you can see exactly when the amp starts to clip, and pointing out that value is a superior way of describing power imo.

 

[andreasmaaan]

That smooth exponential rise is just the noise floor of the DAC dominating. So it is not an offset error as we see many times (with jagged output). All else being equal, I rather seen noise here than non-linearities. I think this is a limitation of the AKM DACs relative to ESS parts.

I'm not sure the chip is to blame here. We saw 20-bit linearity with the RME Adi-2 DAC, which uses the lower spec'd AK4490.

Shouldn't the AK4493 used here be capable of at least equal performance when optimally implemented?

(I don't believe this is an audible concern in any case, btw.)

 

[mindbomb]

With the 117db dynamic range in 22khz bandwidth (I think the highest we have seen at this price point), is it ultrasonic noise that is causing the linearity issues?

 

[Veri]

I'm not sure the chip is to blame here. We saw 20-bit linearity with the RME Adi-2 DAC, which uses the lower spec'd AK4490.

Shouldn't the AK4493 used here be capable of at least equal performance when optimally implemented?

(I don't believe this is an audible concern in any case, btw.)

The RME is a $1k device and can allocate a bigger chunk into power filtering/regulation though. The linearity is bad because of noise, not because of inherent non-linearity issues. If the topping was $400-$500 we could have a way better power stage / power adapter.

As it stands, for its budget it is killer, I think. The EL Dac had better linearity but it has no amp stage, no bluetooth, no remote...

 

[andreasmaaan]

The RME is a $1k device and can allocate a bigger chunk into power filtering/regulation though. The linearity is bad because of noise, not because of inherent non-linearity issues. If this device was $400-$500 we could have a way better power stage / power adapter.

As it stands, for its budget it is killer, I think. The EL Dac had better linearity but it has no amp stage, no bluetooth, no remote...

Yeh, I totally agree. I just thought it was worth suggesting that probably the chip itself was not the culprit here.

 

[zephyros]

The D50 did fine with a phone power adapter, so I doubt a linear PSU would improve things much. That being said this one has more possible source of noise than a standalone dac given its feature set (amps, bluetooth), and much cheaper too. Makes me wonder if the bluetooth is always on or not, and if turning it off would change anything.

 

[rajapruk]

I think price is getting a little bit too much mixed in to the reviews, coloring the whole review.
Just objective performance is what I am after reading, then I can judge myself if the price is good or bad. Also price changes sometimes after hand, then the review gets kind of obsolete if the price is too intertwined in it.

The sudden unimportance of resolution (linearity in bits) in this review was a bit surprising for me. This was kind of a big deal in previous reviews, as I have interpreted it.

(I love the reviews, just a small note this)

 

[rajapruk]

Maybe price vs. performance ratio could be an own little segment of the review, in the end (in the conclusion). But not setting the tone for a whole review, if you get what I mean.
(NB, I am not talking about this review specifically. This comes more from my total view after reading a lot of reviews here)

 

[bennetng]

I'm not sure the chip is to blame here. We saw 20-bit linearity with the RME Adi-2 DAC, which uses the lower spec'd AK4490.

Shouldn't the AK4493 used here be capable of at least equal performance when optimally implemented?

(I don't believe this is an audible concern in any case, btw.)

https://www.audiosciencereview.com/...w-and-measurements-of-schiit-modi-3-dac.4742/
Take a look at Modi3's SPDIF linearity, it uses AK4490 as well.

AK4493 should be better, it reveals the whole face in full color.

 

[Raimei Templar]

I think price is getting a little bit too much mixed in to the reviews, coloring the whole review.
Just objective performance is what I am after reading, then I can judge myself if the price is good or bad. Also price changes sometimes after hand, then the review gets kind of obsolete if the price is too intertwined in it.

The sudden unimportance of resolution (linearity in bits) in this review was a bit surprising for me. This was kind of a big deal in previous reviews, as I have interpreted it.

(I love the reviews, just a small note this)

Well he said he had less of a problem with linearity here due to it being a result of noise rather than outright errors, that is why is has that smooth rise rather the crazy jaggies some others have had (eg Yggdrasil) and that is less offensive to him. Think Amir really hates jaggy lines .

 

[bunkbail]

AK4493 should be better, it reveals the whole face in full color.

I know you're joking but AKM has an image on its product page that shows which chip has better sound quality.

You see, the AK4493 bubble is twice as large as the AK4452, which should mean the DX3Pro should sound twice as better than the Grace SDAC, no?