[xbmcreallyrocks]

How about 0=high, 2=mid, 1=low ? If that's still wrong then I bow out. I do wish there were some vocals in your samples but this is fun anyway.

I will say that the difference between great and not-as-great sound is frequently how fatiguing it is during a long listening session.

[fritsch]

Here is the solution:

0 = low
1 = mid
2 = high

- so now you even could not distinguish the far worst quality and the very best quality, hehe - so much about "the greatest resampler". But Let's do another test. Can you post a piece of music that you know - some 96 / 192 khz flac - I will resample it and post the results here - so you can try again?

Edit: Btw. please post an xbmc.log in debug mode when playing those posted three samples - I want to see that you are really running the ALSA sink and not e.g. pulseaudio which could resample internally again.

[fritsch]

Test 1
https://dl.dropboxusercontent.com/u/5572...er2-0.flac
https://dl.dropboxusercontent.com/u/5572...er2-1.flac
https://dl.dropboxusercontent.com/u/5572...er2-2.flac

The original file you can be found here: http://www.2l.no/hires/ - the tilte is called: Finzi: Come Away, Death Marianne Beate Kielland, mezzo soprano Sergej Osadchuk, piano 192 khz / 24 bit

Test 2
https://dl.dropboxusercontent.com/u/5572..._orig.flac (Original)
https://dl.dropboxusercontent.com/u/5572...ham_0.flac
https://dl.dropboxusercontent.com/u/5572...ham_1.flac
https://dl.dropboxusercontent.com/u/5572...ham_2.flac

What you need to do:
Download the files and listen to them. I for my part can clearly hear the difference on my B&W speakers - so, the big question is again: Which number corresponds to low, which to middle and which to high? And as a sidenote which one compares best to the original?

[fritsch]

Here is an image comparing the frequencies. As audiophile and resampler specialist - you are familiar with fourier analysis, right? So I don't need to explain the two plots (besides you see what the nyquist parameter does):

[fritsch]

Nobody? :-(

Come on Audiophiles.

[fritsch]

Audiophiles bump again.

[fritsch]

Another audiophile bump ...

[MrNice]

I try:
Config;
Raspberry Pi < HDMI > Denon AVR-2313 > Monitor Audio RX6 5.0 Speakers

sample_gotham_x.flac files:
Best is 2
Then 1
Worse is 0

singer2-x.flac files:
Best is 2
Then 1
Worse is 0

Let me know the result

[fritsch]

Also answer the second part. What difference do you hear between "Best" and the original file?
Assuming your answer would be correct (:-)) - how far is 1 worse than 2 and how far is 0 worse than 1 or 2?

[MrNice]

I was waiting for your result but I have a secret to say:
I can't listen any difference, I cheated using Spek to find the difference visually.
I think my hardware is able to play only a tiny difference but my ears can't do any difference, too old at 60 years.

Please have a look your files with Spek. It's free and multi OS.

sample_gotham_x.flac files are not so good for this kind of comparaison because the orchester plays in bass register, just a bit of high-pitched at the end of the extract.

And I can't really understand the voice in the other file.
With Spek, the first few seconds we can see the piano register; no high harmonic, low intensity.
Then the voice. How the femal voice can have so high harmonics with strong intensity? I'd say there is a manipulation in this record.

I think I could try to compare records with high-pitched sound instrument like metal drums, high register solo violon or organ but obviousely not with piano and voice.

BTW did you read this article on why not to use 192 kHz sampling?

[fritsch]

http://xiph.org/~xiphmont/demo/neil-young.html <- yeah - exactly.

This whole thread here is there to show audiophilists, that they should more use their ears and not what "light" of the DAC is showing.
Btw. the "high quality" file uses 8 times more processing power than the "mid quality" which uses double the processing power.

Edit: And until now I only found exactly one person that could hear the difference.

[MrNice]

Could you have a look at the sing2L-064_stereo192kHz_01_08.flac file with Spek.
What is the cloud above 70 kHz? Why this noise?

[fritsch]

No idea - really. Seems like an somehow unwanted effect by the 2L people.

[MrNice]

But a lot of music files has this kind of noise.

If you have some time, you should analyse your music files with Spek, you will be surprised, even with 16/44.1 CD files.

BTW, do you know this article about SACD?

[fritsch]

Hehe :-)

Last time - I upsampled a 44.1 mp3 to 192 khz and gave it (as flac) to some audiophiles - and all liked the crappy upsampled 192 khz best, cause their "DAC" had a green light when 192 khz is coming through.
Basically I want that people judge with their ears.

Always funny.