Measurements and Review of SMSL iDEA and AudioQuest Dragonfly Black DACs

[sejarzo]

I did a few casual tests against the K1 and sadly, its measured performance is worse than SMSL iDEA. Will need to repeat the tests again to be sure but thought I share that with you.

Does the performance degrade terribly at a 0 dBFS level versus something less as some have claimed is a consistent problem with PCM5102-based devices?

Also, what load impedance did you use?

I ask because, given that it the K1 has no volume control, I use the volume slider in Foobar with my 36Ω RE400 IEMs, and it shows that I normally run ~ -20 dB attenuation on LOUDLY mastered pop/rock and rarely louder than ~ -14 dB on older classical material that's mastered "classically."

 

[amirm]

The current distortion tests are essentially open load with a 100 K impedance. I plan to extend testing with actual headphone loads and various dummy load one. The amount of work explodes unfortunately.

 

[Zilfallion]

Amir did some measurements on my E07K recently. As I recall it looked OK but had a DC offset through the L7 line-out dock. (I don't think we looked at the headphone out.) The offset was slightly negative when running off of USB power, and slightly positive when running off of battery.

I'd be interested in seeing the results of that. It would at least give a better idea of FiiO products. We saw the D03k do decently at 44.1kHz(and poorly at 48), the E10k did fairly well. I'd be curious to have a look at their mobile DAC/Amps.

 

[Jimster480]

FiiO K1 measurements!! I think these will surprise many because this device sounds pretty damn good atleast to my ears. And it seems to be capable of powering my Aeon!

 

[gepardcv]

I tried a black DragonFly a while back, and found it appalling enough to return. With IEMs, it (1) played much too loud even at minimum volume setting in macOS, and (2) hissed loudly enough to hear over music. All-in-all, it worked significantly worse than the headphone jack of anything I tried plugging it into.

Which measurements would reflect this poor performance?

 

[Jimster480]

I tried a black DragonFly a while back, and found it appalling enough to return. With IEMs, it (1) played much too loud even at minimum volume setting in macOS, and (2) hissed loudly enough to hear over music. All-in-all, it worked significantly worse than the headphone jack of anything I tried plugging it into.

Which measurements would reflect this poor performance?

Since he isn't measuring gain... but there is electrical interference as shown in the sine wave and the THD+N being up under -100db

 

[gepardcv]

Meaning the 3kHz peak hitting -70dB? The hiss I heard is way, way louder than that. In fact, the room's noise floor was probably above -70dBFS, and I could still hear a clear hiss above the playing.

 

[amirm]

It is possible that it doesn't have good USB isolation and your USB port is a lot noisier than mine. And it was that noise that was travelling through.

 

[Jimster480]

Meaning the 3kHz peak hitting -70dB? The hiss I heard is way, way louder than that. In fact, the room's noise floor was probably above -70dBFS, and I could still hear a clear hiss above the playing.

Remember that with headphone loads, these numbers may change.
They could be worse or as Amir said it could easily be your noise from your USB port and the isolation could be subpar.

 

[SchwarzeWolke]

I also have the DF Black and I have also the problem with hiss when I‘m using sensitive IEMs. I would also say that this is related to a noisy USB I have with my desktop setup. I don‘t have the hiss problem when running my laptop from the internal battery and the hiss reacts to the volume control.

 

[duo8]

This device has volume controls on it. According to that ESS slide digital volume control can cause loss of resolution.
Can anyone confirm it is the case with the iDEA?

 

[sejarzo]

This device has volume controls on it. According to that ESS slide digital volume control can cause loss of resolution.
Can anyone confirm it is the case with the iDEA?

Something I have also wondered about, too.

I have always thought that since virtually all of my library is 16 bit, and that's padded up to 24 bit by Foobar before it is sent to the DAC, that leaves at least 8 bits/48 dB of attenuation available before there would be a definite loss of resolution, assuming that the DAC is capable of 16 bits of resolution overall and the source material does indeed have the maximum dynamic range of 96 dB (doubtful.)

 

[amirm]

This device has volume controls on it. According to that ESS slide digital volume control can cause loss of resolution.
Can anyone confirm it is the case with the iDEA?

I would have to devise a test for it. For now, can you point me to where they say this?

 

[duo8]

I would have to devise a test for it. For now, can you point me to where they say this?

ESS presentation: http://www.esstech.com/files/3014/4095/4308/digital-vs-analog-volume-control.pdf

 

[amirm]

Thanks. But as they explain if the word length is much larger than that of the DAC output, then digital performance is excellent:

The problem was mainly there when they used 16 bit arithmetic for the volume control for most of the presentation.

 

[Blumlein 88]

This below is an excerpt from the jriver wiki about volume control in their media center.



The precision offered by Media Center's 64bit audio engine is billions of times greater than the best hardware can utilize. In other words, it is bit-perfect on all known hardware.


To demonstrate the incredible precision of 64bit audio, imagine applying 100 million random volume changes (huge changes from -100 to 100 dB), and then applying those same 100 million volume changes again in the opposite direction.


Amazingly, you will have the exact same signal at 32bit after 200 million huge volume changes as when you started.


In other words, this incredible number of changes results in a bit-perfect output at 32bit, which is the highest hardware output bitdepth (most high-end hardware is 24bit).


This also means one volume change or a series of 100 million volume changes that add up to the same net result is bit-identical.

 

[tili]

I did a few casual tests against the K1 and sadly, its measured performance is worse than SMSL iDEA. Will need to repeat the tests again to be sure but thought I share that with you.

thank you for the warning! now I gotta find the cheapest-you-can-getxaway-with DAC after crossing K1 out

 

[amirm]

Here is a quick jitter and noise test against Fiio K1:

As we see it has lower dynamic range (in green). Noise floor is higher yet the output is lower.

 

[Zilfallion]

Here is a quick jitter and noise test against Fiio K1:
As we see it has lower dynamic range (in green). Noise floor is higher yet the output is lower.

That's really not that bad for a 40 dollar DAC/Amp combo though. Higher noise floor, weaker output, but no bad spikes seen.

 

[Jimster480]

Here is a quick jitter and noise test against Fiio K1:

View attachment 11008

As we see it has lower dynamic range (in green). Noise floor is higher yet the output is lower.

Output actually looks pretty clean overall. Despite the higher noise floor its still mostly under -120db. So it shouldn't be audible?