[Marlene]

  Most interesting.
 
It explains why I've been feeling the mp3 decoder needs a complete overhaul. As it is it isn't worthy of inclusion in a serious DAP. The thing I first noticed was the horrible quanitization noise.
 
Could you do the same measurements for the ogg decoder so we can see what is so seriously flawed in that too.
 
And while I'm whining, can FiiO please note that the mp3 decoder is most certainly not capable of gapless playback.

 
I thought that it should be audible. I´ll measure ogg too. Stay tuned.

 

[goodvibes]

X5?
 
Marlene, it was nice to have a civil conversation.

 On my Rockboxed clip, I prefer to use it with the dither on. I don't think it a great dither either because you can tell the noise floor is higher but I still feel I can hear further in. Never thought it as good as some do but it's a fab player for what it is.

 

[Marlene]

More measurements
 
Direct Decoding (a.k.a. decoded by dbPowerAmp):
 

dynamic range
 

total harmonic distortions
 

intermodulation distortions
 
As you can see, OGG generally measures pretty bad. Nevertheless, all formats are decoded to 32 bit floating point - this is the result. Apart from the original WAV file MP3 & AAC would be able to encode with high resolution (meaning: better than CD).
 
 
 
Decoded with the FiiO X3 (S/PDIF output digitally recorded):
 

dynamic range
 

total harmonic distortions
 

intermodulation distortions
 
 
Result: the FiiO X3 decodes every format (except WAV) with 16 bit precision only.
 
MP3 is much, much worse than anything else. Not pictured are phase distortions; on WAV and AAC they were absent, MP3 and OGG both showed strong phase distortions.
 
Results for 48 kHz samplingfrequencies are the same.
 
Especially the results for MP3 are worrying me. They mean, that everything you purchase online from for example Amazon.com will have additional distortions when played back by the FiiO. Perfect MP3 playback without distortions should be mandatory in this day and age. FiiO, you simply have to repair this.
 
EDIT: here are all the files used for my tests: http://www.mediafire.com/download/rk09txfrq9h44c0/FiiO+X3+TESTS.rar
 
The archive contains the testfiles and the FiiO X3-derived results and also the RMAA results.

 

[martin t]

  More measurements
 
 
 
 

 
Thanks - that was quick!
 
However, I find ogg on the X3 totally unacceptable (whilst I can put up with mp3 for casual listening) due to enormous amounts of a sort of warble type distortion on the left channel. Do you not here this at all?

 

[goodvibes]

AAC looks OK. MP3 not so much. I think you meant to say lossy formats at 16 bit could be better and all lossless are good to go. Not just wav.

 

[chengsta]

so is this fixable by converting mp3 to lossless?  Or do we have to get a new rip from the source? 

 

[Marlene]

   
Thanks - that was quick!
 
However, I find ogg on the X3 totally unacceptable (whilst I can put up with mp3 for casual listening) due to enormous amounts of a sort of warble type distortion on the left channel. Do you not here this at all?

 
I´ve not yet listened to anything encoded with OGG. I encoded my first OGG files right after you asked for measurements. And one can´t listen to measurement files 

 

[Marlene]

  AAC looks OK. MP3 not so much. I think you meant to say lossy formats at 16 bit and lossless are good to go. Not just wav.

 
Yes, AAC looks OK. MP3 wouldn´t be able to look the same but as everyone can see on the directly decoded files it would look much better if properly decoded.
 
Lossy formats derived from 16 bits will yield the same results. I expected this to be clear now; that´s also the reason why everything from for example Amazon.com won´t sound any good with the X3.
 
You see, lossy formats don´t have a defined bit depth like WAV. They´re always floating point, no matter the input. That´s why they have to be decoded to floating point, 16 or 24 are just a fake, something the MP3-decoder produces in order not to confuse people.

 

[martin t]

Cool. Would you like to give it a listen? If you do, and if you hear the problem I do, is there a measurement technique to expose this?

 

[bodewalla]

  so is this fixable by converting mp3 to lossless?  Or do we have to get a new rip from the source? 

It seems Marlene is teaching much to many experienced guys here! Cheers for you Marlene!
 
Ok Cheg, if you have the source, why bothering with MP3? It's is lossy and if you try a lossless format from a lossy one, well it will be a lossy big file, even considering it will be better played back.

 

[sbradley02]

   
  Result: the FiiO X3 decodes every format (except WAV) with 16 bit precision only.
 
MP3 is much, much worse than anything else. Not pictured are phase distortions; on WAV and AAC they were absent, MP3 and OGG both showed strong phase distortions.
 
Results for 48 kHz samplingfrequencies are the same.
 
Especially the results for MP3 are worrying me. They mean, that everything you purchase online from for example Amazon.com will have additional distortions when played back by the FiiO. Perfect MP3 playback without distortions should be mandatory in this day and age. FiiO, you simply have to repair this.
 
EDIT: here are all the files used for my tests: http://www.mediafire.com/download/rk09txfrq9h44c0/FiiO+X3+TESTS.rar
 
The archive contains the testfiles and the FiiO X3-derived results and also the RMAA results.

Interesting data. I don't hear the issue when listening to MP3s (which is almost all my portable music) but that doesn't mean it isn't there. Looking forward to hearing James' response.

 

[Marlene]

  Cool. Would you like to give it a listen? If you do, and if you hear the problem I do, is there a measurement technique to expose this?

 
I will do so. Encoding some music right now.
 

 
It seems Marlene is teaching much to many experienced guys here! Cheers for you Marlene!
 
Ok Cheg, if you have the source, why bothering with MP3? It's is lossy and if you try a lossless format from a lossy one, well it will be a lossy big file, even considering it will be better played back.
 

 
Am I? I only do this because I want the FiiO X3 to be at its best before I review for my blog. After all, I´m having roughly 150 pageviews per week for my article about the FiiO E07K. Which means, my articles might be able to skip the scale towards another manufacturer. I don´t want this to happen of course - but I also won´t hide something the FiiO is bad at. And if it can be repaired, why not do it?
 

  Interesting data. I don't hear the issue when listening to MP3s (which is almost all my portable music) but that doesn't mean it isn't there. Looking forward to hearing James' response.

 
This I know!

 It sounds a bit more aggressive than the original, imaging stability is impaired as well.

 

[Marlene]

  Cool. Would you like to give it a listen? If you do, and if you hear the problem I do, is there a measurement technique to expose this?

 
OH
MY
GOD!
 
I wouldn´t have expected decoding errors as severe as these! This sounds so horrible that I can´t find any words. The warble reminds me of lossy encoders 15 years ago... like Xing with 96 kBit/s. It´s not only in the left channel, everything in the center and on the right is affected as well. The warble is in the left channel, the rest has slight flanging artifacts. This is a major issue. 
 
It´s also very inconvenient that the FiiO X3 ignores properly tagged OGG files (including the cover).
 
Blegh... wait, have to get rid of it... my poor ears.... phew, done.
 
Though... I really wonder why this didn´t show up on the measurement files. Well, another prove to show how limited RMAA can be.

 

[mark2410]

  quick question, how do you make it "play all"  the SD card rather than onboard?

bump

 

[Septique]

Dear friends,
Could anyone send me HiRes photo of Fiio X3 board with removed (unsolder) Wm8740 DAC chip
which would have been clearly visible line contacts under the chip?

I casually damaged a few contacts under the chip
​and now it's difficult to restore them​ without a photo
I will be very grateful to you for your help!!!!
Thank you!!
my email: discostyle@gmail.com