I have been wanting to test this little guy out for a long while and finally have a chance at a short loaner as one of my friends got it as well. The X1 uses a PCM5142 which is the same DAC that is used in Bottlehead's. First impression wise, I thought the X1 fits perfectly in with a good enough price point. It does not strike me as a giant killer or whatever but does well at whatever it does at its price point, DAC wise.
UI is pretty decent and with my needs, very useable considering how shitty other GUI's like the DX50 is. There are complains of compatibility with playlist and the way titles are parsed but these are pretty minor. The scroll works well enough though there is some backtrack. Interface is not very laggy as well.
Comparing with the GEEK IEM, its grayish background and a weirdish soundstage. Transients are not very well defined and come across to me as slightly rounded off. Its not a hard hitting DAC either. Might be somewhat grainy sounding.
Comparing with the UD120 a 5102a based dac, I think the Fiio gets the tonality thing down slightly better though the UD120 is more refined, dynamic and less rough sounding.
Comparing with the DX50, the X1 is welcome in the DAP world, the DX50 was unusable and sounds very lo-fi
Compared with my Ipod 4g, I thought the X1 provided decent soundstaging compared to the 4g which has a collapsed soundstage.
Using it as a headphone amp, low bass extension is something that needs to be addressed together with tonality of sorts.
Rig:
Fiio X1(Headphone amp mode) -> Short 3.5mm to RCA cable-> EMU0404 via RCA to XLR adapters-> Fujitsu Laptop
Software: WaveGene, WaveSpectra, ARTA, Audition CS5.5
DISCLAIMERS:
Not science, YMMV. Also do take note of the rising noise with wide bandwidth measurements of the 0404.
Impulse response shows a linear phase IIR symmetrical ringing filter. Somehow the polarity is inverted for whatever reason.
The X1 uses a 4X normal interpolation filter with a sharp rolloff, its not high attenuating though so you can see the stopband ripple. There is some aliasing distortion that can be seen at 25khz. This seems common with the TI chips as the UD120 also has similar levels of aliasing. There is less measured distortion products but the noise floor on the X1 is notably higher than the UD120
I wanted to measure the even harmonic products from the square wave but was unable to do so due to glitching involved with the X1, it seems to be constant no matter the level of the square wave. Using shorter FFT sampling allowed me to see that there is really glitches(in the form of flashing of the spectrum window) involved when playing the square wave. I wonder if this has an effect on complex signals?
As a DAC, the X1 does pretty okay compared to the UD120 and the GEEK IEM in TCM mode. It shows the sonically preferable decaying harmonics though 4th and above order harmonic distortions are notably more than the UD120.
In the real world test, an 18ohm dynamic IEM is used as a load. Distortion is increased though its notably much better than the Iphone 5s in driving headphone loads(0.05% THD). Of course, it does better than the Sansa Zip(given the price)
The X1 does okay with low level resolution test despite the slightly raised noise floor, the result is a defined but slightly noisy sinewaves. It resolves the 1 6bit quantisation steps with ease.
