Testing the claim: "I can hear differences between lossless formats."

Feb 18, 2015 at 2:34 PM Post #661 of 721
I am blissfully unencumbered by offspring!
 
Feb 18, 2015 at 3:37 PM Post #662 of 721

 
Feb 18, 2015 at 3:47 PM Post #663 of 721

 
Yeahhh...this thread has gotten so far off track.
 
In somewhat related news, I got my friend to do an ABX between lossless and 320 kbps MP3. He failed. (Not really newsworthy.)
 
If only that handful of people would be willing to do the ABX between lossless formats! Not a single one has so far. =/
 
Feb 18, 2015 at 4:44 PM Post #664 of 721
I thought lossless formats keep the same quality between types ? I will try that once i get a worthy DAP between 16/44.1 vs 16/48 or so

I will first analyze the tracks between two formats, find the mark with exctinctive differences, cut it out, and ABX it.

The differences should be super slim, probably 1-2 cents, but once the differences are presented, then it mean you can hear it....but this is getting i to extreme level.

I said extreme level as ... Looking at Picasso paintings, can you tell the fake and the real ? Only a handful people can, and because they know what to look for...the same as ABX from lossless vs high-res
 
Feb 18, 2015 at 5:02 PM Post #665 of 721
I thought lossless formats keep the same quality between types ? I will try that once i get a worthy DAP between 16/44.1 vs 16/48 or so

I will first analyze the tracks between two formats, find the mark with exctinctive differences, cut it out, and ABX it.

The differences should be super slim, probably 1-2 cents, but once the differences are presented, then it mean you can hear it....but this is getting i to extreme level.

I said extreme level as ... Looking at Picasso paintings, can you tell the fake and the real ? Only a handful people can, and because they know what to look for...the same as ABX from lossless vs high-res

 
If the difference is all frequencies above 22kHz, then you probably won't be able to hear it, even in an isolated difference file.
 
Feb 18, 2015 at 5:14 PM Post #666 of 721
I thought lossless formats keep the same quality between types ? I will try that once i get a worthy DAP between 16/44.1 vs 16/48 or so

I will first analyze the tracks between two formats, find the mark with exctinctive differences, cut it out, and ABX it.

The differences should be super slim, probably 1-2 cents, but once the differences are presented, then it mean you can hear it....but this is getting i to extreme level.

I said extreme level as ... Looking at Picasso paintings, can you tell the fake and the real ? Only a handful people can, and because they know what to look for...the same as ABX from lossless vs high-res

 
At least you're willing to do an ABX. Don't expect to pass it, though!
biggrin.gif

 
Feb 18, 2015 at 5:32 PM Post #669 of 721
I thought lossless formats keep the same quality between types ?

 
Nope. Fraunhofer MP3 is the oldest codec. That is the first MP3 standard. LAME was an improvement on that. AAC is MP4, not MP3- a much better codec. Saying "I can tell lossy from lossless" doesn't mean anything unless you specify the bitrate and codec you are talking about.
 
I have a listening test that I will be happy to share with you. Ten different samples, from Fraunhofer 192 to lossless. If you would like to listen to it and rank the samples, I'll let you know which ones you picked as the best and worst. You don't need any software for this test. Just the ability to play a FLAC or ALAC file (your choice)
 
Feb 18, 2015 at 5:43 PM Post #670 of 721
  Nope. Fraunhofer MP3 is the oldest codec. That is the first MP3 standard. LAME was an improvement on that. AAC is MP4, not MP3- a much better codec. Saying "I can tell lossy from lossless" doesn't mean anything unless you specify the bitrate and codec you are talking about.
 
I have a listening test that I will be happy to share with you. Ten different samples, from Fraunhofer 192 to lossless. If you would like to listen to it and rank the samples, I'll let you know which ones you picked as the best and worst. You don't need any software for this test. Just the ability to play a FLAC or ALAC file (your choice)

 
But he never mentioned lossy. XD
 
Feb 18, 2015 at 5:59 PM Post #671 of 721
   
Nope. Fraunhofer MP3 is the oldest codec. That is the first MP3 standard. LAME was an improvement on that. AAC is MP4, not MP3- a much better codec. Saying "I can tell lossy from lossless" doesn't mean anything unless you specify the bitrate and codec you are talking about.
 
I have a listening test that I will be happy to share with you. Ten different samples, from Fraunhofer 192 to lossless. If you would like to listen to it and rank the samples, I'll let you know which ones you picked as the best and worst. You don't need any software for this test. Just the ability to play a FLAC or ALAC file (your choice)

actually this would be worth it, only if it would be done with one's music. because, if you provide the music, at first listening to it, it might be hard for one to recognize what is better and what is worse, but if you do the same thing with one's most listened music, the cahnces are that that person will recognize the lossy from lossless. 
 
Feb 18, 2015 at 6:09 PM Post #672 of 721
You will probably need to use foobar2000 and its ABX Comparator on a PC to conduct and document the ABX test.

You also need to make sure that you convert the files of higher resolution to files of lower resolution before comparing.


I will use ABX on DAP android. Any recommended program ? And yes, I will use the original highest lossless quality and down convert it to CD quality on the same program, same track, same computer
 
Feb 18, 2015 at 7:50 PM Post #673 of 721
   
At least you're willing to do an ABX. Don't expect to pass it, though!
biggrin.gif

 
You should avoid setting peoples' expectation bias.
The nature of the test is that he'll be listening for differences that are, at best, on the edge of audibility. (If they were clearly audible, we wouldn't be having this discussion.)
If he believes he won't hear a difference, he won't listen thoroughly because he fears he will hear a difference.
If he believes he will hear a difference, he'll listen thoroughly looking for a difference.
 
Feb 18, 2015 at 10:02 PM Post #674 of 721
   
But he never mentioned lossy. XD


Sorry, I misread it. Strange question. Why would there be any difference between lossless codecs? Lossless is lossless. Inaudible in inaudible. Transparent is transparent. Doing an ABX of lossless is a complete waste of time. All you might find out is that your testing procedures are inaccurate. You can't learn anything about lossless.
 
We need a new motto around here... "Sound Science! Reinventing the wheel every hour of every day!"
 
I'll patiently wait for someone who is actually interested in improving the sound quality of their sound system, rather than verifying that it doesn't sound any better with different types of files.
 
Feb 18, 2015 at 10:07 PM Post #675 of 721
  Sorry, I misread it. Strange question. Why would there be any difference between lossless codecs? Lossless is lossless. Inaudible in inaudible. Transparent is transparent. Doing an ABX of lossless is a complete waste of time. All you might find out is that your testing procedures are inaccurate. You can't learn anything about lossless.
 
We need a new motto around here... "Sound Science! Reinventing the wheel every hour of every day!"

 
hehe, well, you're right. It's more about investigating anomalies in people's systems and human perception in general. Plus, if someone perceives a difference, but then takes an ABX and realizes it was merely the power of their imagination, it can help them get a better understanding of audio.
 

Users who are viewing this thread

Back
Top