FiiO X5 firmware 2.20 download and discussion
Sep 6, 2014 at 6:10 AM Post #106 of 266
   
How about you prove - properly blind tested and volume matched - multiple tests - that the perceived minor distortion changes you measured - are actually audible.
 
Seeing how you have the gear - maybe record the output under both FW versions (playing real music), and post examples of the clips for other people to abx as well ........

evidence on the face of interference is more than 2 times of 2.2, 2.2 vs 2.07 used screening tracks 5, 10, 25, 50, 100, 500, 1000, 21000, 22000 Hz, Flac 24bit/196khz. I hear it, otherwise I would not have done the test. Tests have shown a considerable increase of noise. If you do not hear you, then you have bad headphones or hearing problems. Since firmware 2.2 records do not play is not clean, blurry.
 
Sep 6, 2014 at 6:20 AM Post #107 of 266
  evidence on the face of interference is more than 2 times of 2.2, 2.2 vs 2.07 used screening tracks 5, 10, 25, 50, 100, 500, 1000, 21000, 22000 Hz, Flac 24bit/196khz. I hear it, otherwise I would not have done the test. Tests have shown a considerable increase of noise. If you do not hear you, then you have bad headphones or hearing problems. Since firmware 2.2 records do not play is not clean, blurry.

 
So prove it - post the tracks so everyone can abx.
 
And BTW - you know nothing about me.  My choice of gear is in my signature.  My main headphones are Beyer T1s.  Guess that just means I have tin ears eh 
wink.gif
 
 
Just noticed you mentioned 21 kHz and 22 kHz - you can actually hear that distortion ....... ?
 
Sep 6, 2014 at 6:27 AM Post #108 of 266
   
So prove it - post the tracks so everyone can abx.
 
And BTW - you know nothing about me.  My choice of gear is in my signature.  My main headphones are Beyer T1s.  Guess that just means I have tin ears eh 
wink.gif
 
 
Just noticed you mentioned 21 kHz and 22 kHz - you can actually hear that distortion ....... ?

21 kHz and 22 kHz
The test is made because of the interest, the player plays these frequencies or not ...
 
Sep 6, 2014 at 9:21 AM Post #109 of 266
Sep 6, 2014 at 9:34 AM Post #110 of 266
Where are the test results?

Equipment. This equipment has been tested and calibrated state.
http://imgur.com/a/41k2P#0
2.07
http://imgur.com/a/RgI7b
2.20
http://imgur.com/a/eQiwN
 

evidence on the face of interference is more than 2 times of 2.2, 2.2 vs 2.07 used screening tracks 5, 10, 25, 50, 100, 500, 1000, 21000, 22000 Hz, Flac 24bit/196khz. I hear it, otherwise I would not have done the test. Tests have shown a considerable increase of noise. If you do not hear you, then you have bad headphones or hearing problems. Since firmware 2.2 records do not play is not clean, blurry.
 
Sep 7, 2014 at 5:49 AM Post #111 of 266
  Equipment. This equipment has been tested and calibrated state.
http://imgur.com/a/41k2P#0
2.07
http://imgur.com/a/RgI7b
2.20
http://imgur.com/a/eQiwN
 

evidence on the face of interference is more than 2 times of 2.2, 2.2 vs 2.07 used screening tracks 5, 10, 25, 50, 100, 500, 1000, 21000, 22000 Hz, Flac 24bit/196khz. I hear it, otherwise I would not have done the test. Tests have shown a considerable increase of noise. If you do not hear you, then you have bad headphones or hearing problems. Since firmware 2.2 records do not play is not clean, blurry.

 
This is interesting. Can you tell us more about the method of testing? I.e. files on card / from computer, battery fully charged / partially discharged, etc?
 
Sep 7, 2014 at 8:00 AM Post #112 of 266
As a means of measurements performed state standards: oscilloscope and resistor. Manufacturer and type, you can see from the photos. Assembly diagram is simple: connect the cable mini-plug into the player, on the other hand by using the "crocodiles remove" signal from one of the channels, see the reproduced signal on the oscilloscope screen. oscilloscope as a measuring instrument is able to determine the frequency ranging from 20 Hz, and shows the signal amplitude voltage.Workplace, as well as the standards are shown in the photos. were selected signals: sinusoidal frequencies of 5, 10, 25, 50, 100, 250, 500, 1000, 2500, 5000, 7500, 10000, 12500, 15000, 17500, 20000, 21000, 22000, 23000 Hz; triangular shapes frequencies 1000, 5000, 10,000, 15,000 Hz; rectangular wave frequencies 1000, 5000, 10000, 15000 Hz. Signal is removed from the line-out with a load of 32 ohms.It was interesting as it reproduces signals triangular shape with the frequencies 1000, 5000, 10,000, 15,000 Hz; rectangular wave frequencies 1000, 5000, 10000, 15000 Hz. But they will not show, because there is quite bad. Player to use as a frequency generator is impossible. On the other hand there is an idea, why not make a player that would reproduce the signal with absolute certainty? because then we would have heard and all that is recorded in the studio with sterile cleanliness. Maybe expensive players do exactly that?
thank you
 
Sep 7, 2014 at 5:40 PM Post #113 of 266
If you're measuring from the line-out, you should probably be using a load more like 10k ohm because it's designed to feed an amplifier, not a headphone load. That could account for extra noise which is apparent in both your measurements (2.07 & 2.2)
 
Also, if you're feeding the signal via the micro-USB (using the X5 like a DAC) then tat will introduce noise from the USB of the transport. To fairly measure the X5 you need to load the test tones onto a micro SD card and play directly from the device to isolate the device from external sources of noise.
 
Sep 7, 2014 at 10:19 PM Post #114 of 266
  As a means of measurements performed state standards: oscilloscope and resistor. Manufacturer and type, you can see from the photos. Assembly diagram is simple: connect the cable mini-plug into the player, on the other hand by using the "crocodiles remove" signal from one of the channels, see the reproduced signal on the oscilloscope screen. oscilloscope as a measuring instrument is able to determine the frequency ranging from 20 Hz, and shows the signal amplitude voltage.Workplace, as well as the standards are shown in the photos. were selected signals: sinusoidal frequencies of 5, 10, 25, 50, 100, 250, 500, 1000, 2500, 5000, 7500, 10000, 12500, 15000, 17500, 20000, 21000, 22000, 23000 Hz; triangular shapes frequencies 1000, 5000, 10,000, 15,000 Hz; rectangular wave frequencies 1000, 5000, 10000, 15000 Hz. Signal is removed from the line-out with a load of 32 ohms.It was interesting as it reproduces signals triangular shape with the frequencies 1000, 5000, 10,000, 15,000 Hz; rectangular wave frequencies 1000, 5000, 10000, 15000 Hz. But they will not show, because there is quite bad. Player to use as a frequency generator is impossible. On the other hand there is an idea, why not make a player that would reproduce the signal with absolute certainty? because then we would have heard and all that is recorded in the studio with sterile cleanliness. Maybe expensive players do exactly that?
thank you

 
Purpose? i.e. measuring Noise? or Distortion?
Conclusions?
You're showing different frequencies on the two image sets.
You have to compare 5.0 kHz with 5.0 kHz, for example.
 
Sep 9, 2014 at 10:44 AM Post #115 of 266
   
Purpose? i.e. measuring Noise? or Distortion?
Conclusions?
You're showing different frequencies on the two image sets.
You have to compare 5.0 kHz with 5.0 kHz, for example.

Purpose: Show that the change in sound is, against the fact that the developers claim that the sound did not change. Noise? or Distortion?: Look at the test results that you see in the sine wave? that it shows?
  If you're measuring from the line-out, you should probably be using a load more like 10k ohm because it's designed to feed an amplifier, not a headphone load. That could account for extra noise which is apparent in both your measurements (2.07 & 2.2)
 
Also, if you're feeding the signal via the micro-USB (using the X5 like a DAC) then tat will introduce noise from the USB of the transport. To fairly measure the X5 you need to load the test tones onto a micro SD card and play directly from the device to isolate the device from external sources of noise.

Read carefully, the load 32 ohms. This load was chosen as most headphones are such a drag.
 
Sep 9, 2014 at 10:52 AM Post #116 of 266
  If you're measuring from the line-out, you should probably be using a load more like 10k ohm because it's designed to feed an amplifier, not a headphone load. That could account for extra noise which is apparent in both your measurements (2.07 & 2.2)
 
Also, if you're feeding the signal via the micro-USB (using the X5 like a DAC) then tat will introduce noise from the USB of the transport. To fairly measure the X5 you need to load the test tones onto a micro SD card and play directly from the device to isolate the device from external sources of noise.

all played as in the picture. Have you watched it? we'll see that the files were written, not by the player, and playing it, not as a DAC
 
Sep 9, 2014 at 4:13 PM Post #118 of 266
I don't understand why so many albums show up without art even though iTunes embeds the art in every track when available. The X5 shows about 50% of my files without the art. When using a tag editor like Kid3 I see no difference between the tracks that show art on the X5 and those that don't.
 
Sep 9, 2014 at 4:25 PM Post #119 of 266
  I don't understand why so many albums show up without art even though iTunes embeds the art in every track when available. The X5 shows about 50% of my files without the art. When using a tag editor like Kid3 I see no difference between the tracks that show art on the X5 and those that don't.

 
I'd suggest it's maybe the other extraneous tagging that iTunes also adds.  On my X5, I have around 5200 Flac tracks - all personally tagged by me - Artist, Album Artist, Album, Title, Date, Genre, Track #, and artwork (all jpg 600x600).  Artwork shows up consistently on every track. 
 
Sep 9, 2014 at 6:02 PM Post #120 of 266
  Read carefully, the load 32 ohms. This load was chosen as most headphones are such a drag.

 
I did read carefully and you referred to "line out" which is for amplifiers, not headphones. Did you actually take the measurements from the headphone out or the line out? The pictures suggest you used the line-out in which case a 32 ohm load is way too low and putting far too much strain on the circuit. You need to conduct these measurements via the headphone out or raise your measuring load to around 10k ohm
 

Users who are viewing this thread

Back
Top