[arnyk]

  In my experience recording and then replaying the files on a different hardware might completely change the results. At least in the direction that there might be no difference any more. The (imo) bad version will of course not get better by recording / replaying.
I was locking for someone who heard (not measured) a difference between the firmwares .. like I did.
If nobody here does comparisons like that ... sorry ... then I posted it in the wrong thread.
 
Btw. I listen with X5 + E12 and (of course) AKG K400.

 
In my experience it is far from mission impossible and obtain useful information about noise, distortion, and frequency response information in gear from well made recordings. If you can hear it there will be reliable evidence to be obtained. It is far more likely that the recording will be incompetently made than that someone on their game won't be able to obtain the necessary information from it.
 
Again, if there is an audible difference between what these firmwares do, reluiable evidence of it, both audible and measurable, will very likely to be very easy to obtain from a reasonably well made recording.
 
Why not?  Unsure of yourself?

 

[JSBachFan]

   
In my experience it is far from mission impossible and obtain useful information about noise, distortion, and frequency response information in gear from well made recordings. If you can hear it there will be reliable evidence to be obtained. It is far more likely that the recording will be incompetently made than that someone on their game won't be able to obtain the necessary information from it.
 
Again, if there is an audible difference between what these firmwares do, reluiable evidence of it, both audible and measurable, will very likely to be very easy to obtain from a reasonably well made recording.
 
Why not?  Unsure of yourself?

Actually I'm curious about the outcome of this test myself. So why not ...
But I never made a high quality recording out of the X5 playing a file.
How can I do that?

 

[Joe Bloggs]

Hello K400lover,

Basically you want to maximise signal-to-noise ratio while avoiding clipping. Load a 1kHz full-scale sine wave file on the X5, adjust the X5 to max volume, connect the X5's headphone out to your computer's line in using a 3.5mm interconnect, open a digital audio editor such as Audacity and press record, and play the sine wave file on the X5. You will see a waveform display onscreen, if the waveform fills the whole channel from top to bottom you're clipping. Adjust the recording level of the line in from the audio editor and retry until the waveform fills most but not all of the vertical space. Listen to the sine wave recorded this way to check if you still hear clipping (in which case you need to lower the volume on the X5 itself and increase the recording volume from the audio editor)

The above allow you to set the optimal volumes for the best recording your sound card allows. Oh, and you want to adjust the line in's device sample rate and bit depth and the audio editor's sample rate and bit depth to match each other at a high quality setting (such as 24bit/96kHz, if available)

Best regards,
Joe

 

[JSBachFan]

Hello K400lover,

Basically you want to maximise signal-to-noise ratio while avoiding clipping. Load a 1kHz full-scale sine wave file on the X5, adjust the X5 to max volume, connect the X5's headphone out to your computer's line in using a 3.5mm interconnect, open a digital audio editor such as Audacity and press record, and play the sine wave file on the X5. You will see a waveform display onscreen, if the waveform fills the whole channel from top to bottom you're clipping. Adjust the recording level of the line in from the audio editor and retry until the waveform fills most but not all of the vertical space. Listen to the sine wave recorded this way to check if you still hear clipping (in which case you need to lower the volume on the X5 itself and increase the recording volume from the audio editor)

The above allow you to set the optimal volumes for the best recording your sound card allows. Oh, and you want to adjust the line in's device sample rate and bit depth and the audio editor's sample rate and bit depth to match each other at a high quality setting (such as 24bit/96kHz, if available)

Best regards,
Joe

 
Hello Joe,
 
thanks for all the effort writing that answer to me, but sorry, I expected something different.
When I talk about musicality (!) getting worse after an update how can a sine wave prove me wrong?
Musicality effects (rhythm, feeling) come imo mainly from jitter. Yes jitter will also affect a sine wave, but high level digital reproduction will only reveal these tiny differences when listening to real music. Another round of recording / playback, which obviously will have another round of jitter problems, so the result (especially on a sine tone) will show something, but not what I was talking about.
 
To avoid making a fool of me I did another long AB test yesterday night after which I have to admit, that the strong tonal differenzes (clearity, sound stage) are not as big as I experienced in my first test. But the overall result clearly was the same: It is just more fun to listen to FW 2.2.
 
While I was changing the firmwares several times back and forth I had an idea how I will with 100% certainty find out for myself if I was right or not.
I'm able to do some cool stuff with Excel VBA, so what I will do is to write an Excel macro, that puts a randomly chosen FW out of given set (2.0, 2.2, 2.3, 2.4, 2.5) on my X5 and while doing that recording, which one it chose but not telling me until I want to know.
So I can listen, guess which FW it was, press a button, write down my guess, listen to another one and so on ... and then check against what the VBA program in reality did.
If in this test-setting I can 100% and blind find out which FW was playing, I have my answer. I guess this is even better than blind tests where another person does the FW changes. He might somehow influence you. A computer has a pretty good poker face
... we'll see and I will honestly share the results here ...
 
Since I think this is a very cool idea, I will publish the VBA code in case anybody else is interested in this way of testing too   (?)
 
K400Lover

 

[Joe Bloggs]

Hello K400lover,


Basically you want to maximise signal-to-noise ratio while avoiding clipping. Load a 1kHz full-scale sine wave file on the X5, adjust the X5 to max volume, connect the X5's headphone out to your computer's line in using a 3.5mm interconnect, open a digital audio editor such as Audacity and press record, and play the sine wave file on the X5. You will see a waveform display onscreen, if the waveform fills the whole channel from top to bottom you're clipping. Adjust the recording level of the line in from the audio editor and retry until the waveform fills most but not all of the vertical space. Listen to the sine wave recorded this way to check if you still hear clipping (in which case you need to lower the volume on the X5 itself and increase the recording volume from the audio editor)


The above allow you to set the optimal volumes for the best recording your sound card allows. Oh, and you want to adjust the line in's device sample rate and bit depth and the audio editor's sample rate and bit depth to match each other at a high quality setting (such as 24bit/96kHz, if available)


Best regards,

Joe

Hello Joe,

thanks for all the effort writing that answer to me, but sorry, I expected something different.
When I talk about musicality (!) getting worse after an update how can a sine wave prove me wrong?
Musicality effects (rhythm, feeling) come imo mainly from jitter. Yes jitter will also affect a sine wave, but high level digital reproduction will only reveal these tiny differences when listening to real music. Another round of recording / playback, which obviously will have another round of jitter problems, so the result (especially on a sine tone) will show something, but not what I was talking about.

To avoid making a fool of me I did another long AB test yesterday night after which I have to admit, that the strong tonal differenzes (clearity, sound stage) are not as big as I experienced in my first test. But the overall result clearly was the same: It is just more fun to listen to FW 2.2.

While I was changing the firmwares several times back and forth I had an idea how I will with 100% certainty find out for myself if I was right or not.
I'm able to do some cool stuff with Excel VBA, so what I will do is to write an Excel macro, that puts a randomly chosen FW out of given set (2.0, 2.2, 2.3, 2.4, 2.5) on my X5 and while doing that recording, which one it chose but not telling me until I want to know.
So I can listen, guess which FW it was, press a button, write down my guess, listen to another one and so on ... and then check against what the VBA program in reality did.
If in this test-setting I can 100% and blind find out which FW was playing, I have my answer. I guess this is even better than blind tests where another person does the FW changes. He might somehow influence you. A computer has a pretty good poker face
... we'll see and I will honestly share the results here ...

Since I think this is a very cool idea, I will publish the VBA code in case anybody else is interested in this way of testing too   (?)

K400Lover

Hey K400, what I wrote was as preparation for actual recordings of music from different firmwares. Once you have the correct levels set the next step is obviously to record the actual music with the two different firmwares for ABX testing.

 

[JSBachFan]

Hey K400, what I wrote was as preparation for actual recordings of music from different firmwares. Once you have the correct levels set the next step is obviously to record the actual music with the two different firmwares for ABX testing.

ok - missunderstanding ... but I will do my new test setting first.
 
If I manage to determine the computer chosen FW version blind, I might do the technical test too.
If not, the whole technical comparison is pretty obsolete. At least for me.

 

[JSBachFan]

Just to give a final statement to this (at least from me): I was wrong about 2.5 being worse. However that is possible it seems that even firmware updates have to "burn in".
I can here no differences any more when switching between two firmware versions, that have been used both for a while.
... but still confident that I heard differences, when 2.5 was newly installed ...

 

[issachar1943]

Interesting that this thread is still going on.
 
However, from other forum, I'm also aware that people are complaining the sound quality is performed differently with different firmware.
 
While I agree our ears and brain might be bias from varies reasons, I still tend to believe firmware affect sound quality of the player.

 

[castleofargh]

  Interesting that this thread is still going on.
 
However, from other forum, I'm also aware that people are complaining the sound quality is performed differently with different firmware.
 
While I agree our ears and brain might be bias from varies reasons, I still tend to believe firmware affect sound quality of the player.

there is a difference between "can a FW affect the signal output? " to which you can usually answer yes it can, and "does this given FW affect the signal output?".
but people who really care should just try and record both on a quality input and then ABX if it's music, or analyze if it's a sweep, an impulse, or other test signals.
 
if I listened to a DAP, put it away and had someone tell me he changes something on it, I'm pretty sure I would then hear a difference. real or not that's how a human is, if we look for a difference we'll find one even if it's a made up difference. so it's important to be careful with gut feelings over such possible differences, and look for objective data before drawing conclusions. which is exactly what ClieOS did in the first post. everything else will leave us all with lingering doubts.

 

[issachar1943]

 
there is a difference between "can a FW affect the signal output? " to which you can usually answer yes it can, and "does this given FW affect the signal output?".
but people who really care should just try and record both on a quality input and then ABX if it's music, or analyze if it's a sweep, an impulse, or other test signals.
 
if I listened to a DAP, put it away and had someone tell me he changes something on it, I'm pretty sure I would then hear a difference. real or not that's how a human is, if we look for a difference we'll find one even if it's a made up difference. so it's important to be careful with gut feelings over such possible differences, and look for objective data before drawing conclusions. which is exactly what ClieOS did in the first post. everything else will leave us all with lingering doubts.

 
Yes, I completely agree with your view here.
 
And come to think twice, of coz software could affect how hardware perform, it's just a very 101 Q&A.

 

[Rearwing]

We can over think when it comes to sq, if you like something thats great, if you don't try something else there are a myriad of toys, settings and headphones to try; thats the joy of this hobby. However, if you want definitives........you possibly need a new interest, or you just love a good puzzle/argument/brain twister!
 
After a tortuous journey I like my kit, but I am not sure that I can tell you why in any scientific sense, the interplay between kit, content and listener is too complex for me to either understand or explain, but I know I like listening to music, and thats what keeps me in the game.