Topping D30 DAC Measurement and Review

[Jimster480]

Sure. I can send you an M8 for testing once I get it back from my buddy. Hes been borrowing it for a couple months now.
Maybe I can swap my D30 with the M8 (let him borrow the D30) and send you the M8.

 

[amirm]

That would be great! Please PM me when ready and I will provide the shipping info.

 

[Candlesticks]

Do you still have the Topping D30?

Would you consider the measurements of this device to be complete at this point?

I'd like to compile a list of completed measurements of products that can be used for purchase advice later and have comparable measurements between all devices.

Are you able to test USB input as it will be the most common input used.

The Topping D30 is good as it has USB, coaxial and optical inputs. Some people are looking for a product that is capable of more than just USB so having something good to recommend is welcome even if the price point is higher than a just USB device such as the SMSL IDEA.

 

[amirm]

Yes, I own the D30 so it is still here.

I see that its measurements are sparse relative to what I measure now. So I can add more but let me cover more of the pile of gear I have here and squeeze that in spare time.

 

[Jimster480]

Yes, I own the D30 so it is still here.

I see that its measurements are sparse relative to what I measure now. So I can add more but let me cover more of the pile of gear I have here and squeeze that in spare time.

Eventually, and re-measure it using USB with your great ENOB measurements, especially since the tests you did the other day suggested the D30 to have ~18 bits of accuracy via USB.

 

[Mike S]

eJust got in a D30 and patched it in to my monitor system, initial impression was that it sounds a bit strident. Patched it in to my studio headphones (K701 with dedicated amp) and still sounds a bit strident to my ears, almost as if it was scooped 700-1.5k and pushed 2k-6k by about 1/2 dB. I tested it with RMAA and of course its flat as a pancake... So thinking I'll socket the op amps and try some I use in other gear known to make my ears happier. My 1820m had a similar sound when stock and it warmed up fine with a change and I suspect this will respond similarly. Other than that aspect I really like it. The highs are pretty clean...cymbols and hats are always dead give aways imo and they sound good.

BTW, I bought mine out of China from an EBayer named WAITSTORE based on estimated shipping times for free shipping (plus 99.8% feedback and 60 day returns and negotiated a few $ with Make Offer). Despite their "misleading" name, it was on my doorstep 5 calendar days after clicking "buy"...pretty fantastic!

EDIT: I'm going to try an objective test to see if op-amp differences show up. Many studio softs can print spectrographs of a music clip that map frequency response as a continuous "scroll" of color representing frequencies. Since the soft allows a repeatable segment of music to be targeted, one can run a graph of the clip with one opamp and then pop in another, run it again and have an apples to apples comparison. I'll post in the next few days with results...

 

[amirm]

EDIT: I'm going to try an objective test to see if op-amp differences show up. Many studio softs can print spectrographs of a music clip that map frequency response as a continuous "scroll" of color representing frequencies. Since the soft allows a repeatable segment of music to be targeted, one can run a graph of the clip with one opamp and then pop in another, run it again and have an apples to apples comparison. I'll post in the next few days with results...

Please do. Meanwhile, some kind members are loaning me some opamps to swap in Topping D10 and I will report with measurements of those when I get to it.

 

[Jimster480]

eJust got in a D30 and patched it in to my monitor system, initial impression was that it sounds a bit strident. Patched it in to my studio headphones (K701 with dedicated amp) and still sounds a bit strident to my ears, almost as if it was scooped 700-1.5k and pushed 2k-6k by about 1/2 dB. I tested it with RMAA and of course its flat as a pancake... So thinking I'll socket the op amps and try some I use in other gear known to make my ears happier. My 1820m had a similar sound when stock and it warmed up fine with a change and I suspect this will respond similarly. Other than that aspect I really like it. The highs are pretty clean...cymbols and hats are always dead give aways imo and they sound good.

BTW, I bought mine out of China from an EBayer named WAITSTORE based on estimated shipping times for free shipping (plus 99.8% feedback and 60 day returns and negotiated a few $ with Make Offer). Despite their "misleading" name, it was on my doorstep 5 calendar days after clicking "buy"...pretty fantastic!

EDIT: I'm going to try an objective test to see if op-amp differences show up. Many studio softs can print spectrographs of a music clip that map frequency response as a continuous "scroll" of color representing frequencies. Since the soft allows a repeatable segment of music to be targeted, one can run a graph of the clip with one opamp and then pop in another, run it again and have an apples to apples comparison. I'll post in the next few days with results...

I'd be interested in reading that.

 

[Mike S]

Another thing everyone in the studio world does is "null testing". The idea is to take a single clip of material played/recorded thru two different target devices or software plugins, and then play both recordings back at the same time with one of them having the phase flipped 180 degrees. The two signals cancel each other out *except* for the differences between them. It works great for identifying subtle differences and if they are identical it will be dead silent...so I'll try it with the two opamps (stock and a 4562 replacement) and see what we get. I use a German software called Samplitude and its kind of expensive, but for those interested I believe Audacity is still freeware and has enough functionality to do a null test. And if you have a Behringer 204 it may have come with a studio software... Bottom line is null testing (and spectrographing) the D30 with actual music to compare different opamps should provide objective data that accounts for what an opamp does when its hit with complex music signals as opposed to a couple of sine waves. Of course it isn't going to tell you which is better, lol. But at least you'll know your preference isn't a placebo.

 

[amirm]

It is hard to do a null testing because the clock in the DAC/ADC can drift. Audiodiffmaker can compensate for it but I have had no luck using it. It almost always crashes on me. I would be interested if you are able to get it done.

 

[Mike S]

It is hard to do a null testing because the clock in the DAC/ADC can drift. Audiodiffmaker can compensate for it but I have had no luck using it. It almost always crashes on me. I would be interested if you are able to get it done.

Been doing it for years, and its very easy to get a perfect null. Maybe the 1820m was just a really good piece? One of its selling bullets was ultra-low jitter. Anyway, I'll try to post something in a day or so....

 

[bhigh]

The D30 is available on Massdrop again for $110.

 

[Old Listener]

Been doing it for years, and its very easy to get a perfect null. Maybe the 1820m was just a really good piece? One of its selling bullets was ultra-low jitter. Anyway, I'll try to post something in a day or so....

If you are using the 1820m for both the DAC and ADC functions in the null test, it's easy. Of course, you aren't fully testing the performance of the DAC in creating an analog signal.

For a full test, you need a separate device for the ADC function (with its own clock.) That second clock will not have the same characteristics as the one in the DAC device and the analog waveform won't be sampled at the exact time the DAC produced the waveform. You'll need a more sophisticated analysis for the comparison and you won't get a perfect null.

 

[Mike S]

Of course, you aren't fully testing the performance of the DAC in creating an analog signal.

How's that? The target DAC just chases the clock in the 1820m is all. 1820 digital out via toslink to target DAC then target's analog out to 1820m ADC completes the loop. The 1820m DAC isn't in the loop. I suppose the target chasing via toslink alone could very well cause a null test to fail, but I'm not following you on the analog signal part. I'll try it though and see what I get, and I promise to be 100% honest ;-) !

All that said, I found an opamp I really like in the D30. First up I tried my usual go-to the LME49720 and it really didn't sound much different. If you're familiar with classical, baroque played on period instruments has very identifiable sound in the violins and violas thru the 1820m. Thru the D30 they lose that signature somewhat. Not bad mind you, just something subtle missing to me anyway. So back to the opamp...I checked my stash and found a few oft maligned on the net pieces I've retrofitted in a bunch of gear over the years and people who heard those pieces always liked the change. So (drum roll, and try not to laugh) the lowly LM 833 got the gig in the D30. It immediately satisfied my ears. And it even specs essentially the same, limited as my setup may be:

As with all the subjective aspects, YMMV, JMO, etc. etc. But for me anyway, it shows that a good sounding piece can get lost in the spec race if one gets too caught up with numbers and doesn't listen to ones ears.

 

[Mike S]

Meanwhile, some kind members are loaning me some opamps....

What do you have coming? I'd be happy to send you a few out of my stash if you want them...I have some PDIP duals in 49720 (a slightly tweeked 4562 by most accounts), 833, 2227, and of course 2134 (but you already have that for sure), I may have some others I used to use... Shoot me a PM if you want some.

 

[Mike S]

Update: I ended up using an OPA2227 in this box. Similar mids to the LM833 but better highs. Also was able to verify on oscilloscope that the circuit surrounding the op amp allows use of uncompensated duals like the OPA2228 (which although nothing more than an uncompensated 2227, has a different sound).

 

[cjfrbw]

https://www.amazon.com/gp/product/B01MXRHW9M/ref=pe_153050_276939440_em_1p_0_im

Looks like a nice package. Any clues on the topology of the amp portion?

Oops, got the answer: TI HIFI chip OPA1611 and TPA6120

 

[Old Listener]

How's that? The target DAC just chases the clock in the 1820m is all. 1820 digital out via toslink to target DAC then target's analog out to 1820m ADC completes the loop. The 1820m DAC isn't in the loop. I suppose the target chasing via toslink alone could very well cause a null test to fail, but I'm not following you on the analog signal part. I'll try it though and see what I get, and I promise to be 100% honest ;-) !

The 1820m is generating the clock on the Toslink data stream. If that clock is used by the target DAC (with some jitter reduction perhaps) to drive the D to A conversion, the effect is about the same as if the D to A process was done in the 1820m. If the target DAC resamples the Toslink input, the D to A process will be using a different clock than the A to D process.

 

[sonci]

I also got this dac from shenzen audio, I needed something for stereo out of my blu ray.
Sound quality is very good, even though not quiet as good as my cd players.
Initially I liked the usb input more than the spidf, but after some days of use, the sound from the laptop has become a bit darker, less airy and spidf in seems better now.
What output are you using with windows? foo dsd like recommended in the manual or just wasapi?
I opened it up, and unfortunately they're using cheaper components on later revisions, some Capxons are in place of the Elnas.

LM317 and LM 337

Different clocks around XMOS chip..

 

[amirm]

What output are you using with windows? foo dsd like recommended in the manual or just wasapi?

I only use WASAPI.

Bummer on switch to cheaper caps.