Topping E30 DAC Review

[solderdude]

It's all about learning.
Obviously aside from the 2 mentioned DACs @Pdxwayne also had an UCA 202.
It was clear from the start it could not be the E30 but he wanted to learn how to confirm this.

 

[Pdxwayne]

It's all about learning.
Obviously aside from the 2 mentioned DACs @Pdxwayne also had an UCA 202.
It was clear from the start it could not be the E30 but he wanted to learn how to confirm this.

Next is to measure actual dB at 100hz and 2000hz for each DAC.

If I use Audacity to create sine of 100hz and 2000hz, play those tones to DAC, then capture those outputs, would the frequency analysis dB analysis for those freq good accurate enough?

Thanks!

 

[companyja]

I tested just for fun and my soundcard's line-in has correct polarity, my onboard realtek's line-in has reverse polarity.

I also have this really interesting thing happening with that tone generator @solderdude mentioned - on the website itself, it records fine. But when I click save and load it in a player, it will record a weird sort of sawtooth where it will modulate the volume up and down and up and down (but the actual waveform of the original file downloaded from the website is fine)...I first thought foobar was doing something weird, so I tried MPC-HC and got the same thing. I also put the saved file on my phone and in poweramp it also does that weird volume variation, while the tone on the website is rock stable.

For added weirdness, it will NOT do this weird modulation on the saved file if I do a loopback of Creative X-Fi stereo out > Creative X-Fi line-out. Any other input and the file varies in volume. For extra added weirdness, if I record the website output manually and then play that file back in a player, it will record regularly and not modulate. I'm still stumped as to why this would be happening. I was at first really scared that my X3 III was broken beyond belief or something

 

[voodooless]

Next is to measure actual dB at 100hz and 2000hz for each DAC.

What would be the point? If you want to measure the frequency response, download REW and use that. Should be plenty of guides out there to help you.

 

[Pdxwayne]

What would be the point? If you want to measure the frequency response, download REW and use that. Should be plenty of guides out there to help you.

All these started because I heard something with E30 in the highs that I didn't hear with x16. Since this is ASR, people will not accept subjective observations.

I have eliminate one possible reason (reverse phase). Now I move on to checking difference in db instead of voltage.

I did voltages measurements for multiple freq already. The odd thing is around 2Khz. X16 a little lower in voltage there as compared to E30. Otherwise, it is similar in voltage to E30 up to 5Khz. But since my multimeter is not accurate, I don't know if the 2khz difference translate to something audible or not.

So, the point is to further check if what I heard is something different in term of loudness around 2khz.

 

[voodooless]

Why not not use Occam’s Rasor and eliminate the most logical thing that could be the cause of all this first: your own perception bias?

 

[Pdxwayne]

Why not not use Occam’s Rasor and eliminate the most logical thing that could be the cause of all this first: your own perception bias?

Aren't you the one who said due to perception bias, people say new toy sounds better? In this case, my old toy (E30) sounds more magical in the highs than my new toy (X16) when played back in my stereo system. Thus I would like to find out why.

 

[voodooless]

Aren't you the one who said due to perception bias, people say new toy sounds better? In this case, my old toy (E30) sounds more magical in the highs than my new toy (X16) when played back in my stereo system. Thus I would like to find out why.

That can still be perception bias. Just look at the people that started doubting the E30 just because you thought there might be phase issue (and don’t get me wrong, I’m not blaming you for that). Basically any little thing influences your (and anyone else’s bias). The mind is a strange thing.. who knows why it makes you find the E30 highs better than the more expensive X16.. AKM calls it “velvet sound”.. that could explain it already. Marketing at it’s best. It’s just unlikely that it has anything to do with the objective performance of the two devices.

I’m all for doing some measurements. It’s obviously part of a discovery journey you’re going though. You’ll always be wiser afterwards.

 

[voodooless]

Is that a pure 2Khz tone? If so, looks like the ADC is clipping badly. In that case, quite useless.

 

[Pdxwayne]

Is that a pure 2Khz tone? If so, looks like the ADC is clipping badly. In that case, quite useless.

Yes, pure sine tone created by Audacity at 0.8 out of 1.0 intensity. Both E30 and X16 was at 0db. OK, I will do -10db for both E30 and X16 and measure again.

 

[voodooless]

Just try REW. It will give you a proper way to first get the levels correct, and also had more tools to discover.

 

[Pdxwayne]

Just try REW. It will give you a proper way to first get the levels correct, and also had more tools to discover.

I do have REW. My previous voltages measurements were using REW.

And the measurements I just provided, they both were within 0.1db of each other at 2khz, at least according to audacity.

 

[Jimbob54]

Actually, the magical "sparkle" of E30 could be just harmonic distortions. : (

I just used Audacity to captured 2Khz RCA playback from both Gustard X16 and Topping E30.

It would appears that E30's 6Khz harmonic is ~2.5db higher than X16. It could that additional harmonic distortions for all higher freq is making it having more "sparkles" than x16.

The top one is Gustard. Bottom is E30. As you can see, Gustard X16 6khz db is -14.7db. E30 6Khz db is -12.1. Gustard is definitely cleaner looking for the rest of higher freq.

I think I have answered my own question about why E30 having more "sparkles"....

View attachment 112357

How do you reconcile your results here to @amirm or anyone else's results of the same devices, or indeed any modern DAC? Neither chart looks like a device working properly. So either you have 2 broken devices, your whole set up is somehow flawed, or your measuring kit / methodology is flawed.

 

[Pdxwayne]

How do you reconcile your results here to @amirm or anyone else's results of the same devices, or indeed any modern DAC? Neither chart looks like a device working properly. So either you have 2 broken devices, your whole set up is somehow flawed, or your measuring kit / methodology is flawed.

Likely my adc clipped like @voodooless said.

 

[Pdxwayne]

Is that a pure 2Khz tone? If so, looks like the ADC is clipping badly. In that case, quite useless.

You are correct. Lowering both DAC to -10 db and I got much cleaner chart.

Also, don't really see freq db difference between DAC too, assuming my UCA202's ADC is accurate and audacity is good enough to do such db comparison.

Left is x16. Right is E30. E30 is ~0.5db louder than X16, when both are set to -10db and received same signals.

Looking at the "Peak" db for both, I don't see much of a different between E30 and X16 for 2Khz and 10Khz.

So far I have found out the difference I heard is not difference of phase. Also, it is not likely db different in freq (at least those freq I checked).

Like you said, next is to set both DAC to match db as close as possible when getting same signals, then do listening tests.

 

[vrobec]

Had someone ordered E30 from french online store Masters of Audio? Are they reliable?

 

[Pdxwayne]

Hmm, what do you say about this comparison?
DAC volume matched at 2khz

Same song, same first bass beat. Zoomed way in to select start of first beat to the start of next beat. Same amount of time selected to create frequency analysis.

It would seems they don't behave the same in multiple area in term of energy level for many frequencies. ~440 HZ is one area of interest. Top of each picture is Gustard. Bottom of each picture is Topping.

Or is this all for naught again? Bad ADC that caused the differences?

 

[AnalogSteph]

Your selection seems way too inaccurate, it's not even the same length. You'd have to select the exact same samples are far as possible.

 

[Pdxwayne]

Your selection seems way too inaccurate, it's not even the same length. You'd have to select the exact same samples are far as possible.

You might get confused by the time start different. This is because recording started slightly off for each DAC. If you deduct the end time from the first time selected for the freq analysis for each DAC, you should come to the same length: 461ms.

 

[voodooless]

Or is this all for naught again? Bad ADC that caused the differences?

View attachment 112426

View attachment 112427

Can you share the files? You might also try some
White noise.