Indeed

Did you have a friend doing randomization for you? A non-blind test isn't much of a test.

^huh?

"The bit rate of converted file is shown by IE as 314kbps." What is IE and it should be 320. . . 

Also, can you post the report(s)?

You don't have to be a musician you know. xD

On another note: 300!

Really useful as well might be samples (I think fair use allows 15 or 30 sec) of the FLAC and MP3 files used, particularly centered around an area where you feel there's signifigant difference.

Quote:

Originally Posted by JerryLove 

Really useful as well might be samples (I think fair use allows 15 or 30 sec) of the FLAC and MP3 files used, particularly centered around an area where you feel there's signifigant difference.

Why do you need a friend to do an ABX test, and why can't the difference be throughout the entire song?

Quote:

You don't; but if you know which you are listening to, it will color your experience and therefore make your results uncertain.

Differences can be through an entire song: but it should not *require* an entire song to tell the differences.

How is this blind test performed? More details? Why don't you use the ABX plugin in foobar2000? It's free and easy to use.

Since you did the test alone, how were the two bit rates sorted so they were random? How did you not know which one would be next? How did you find out which had been which once the test was finished?

There are, that I can think of immediately two basic ABX testing reasons / methods.

Some rules of good ABX testing are that the switch be instant and that the sources are level-matched. Louder sources will almost always sound better, and the brains ability to hold very specific audio detail lasts a fraction of a second. If one does not follow these rules, you are simply listening, not really ABX testing.

If we wish to arrive an a objective conclusion... to test what we think we hear, then ABX testing should be done blind. Knowing which speaker, amplifier, or encoding you are listening to will color your subjective experience.

As other posters have said, you still did not tell us how you randomize the songs. If you are going from FLAC>MP3>FLAC>MP3, you might *think* you are able to detect a difference.

Use the ABX plugin in foobar2000. After finishing the test, click "End" and it will ask you to save a log somewhere. Post the results of that log.

There are some people who can detect differences between lossless and high bit rate lossy on some tracks. There are some tracks which are not easy to convert to lossy without artifacts. I have one particular track in my library, the mp3 has added distortion which is easy to detect when you have had a bit of practice.

The question is how often is this true. If you took a range of other tracks and did the same conversion , how many would you be able to detect the difference with ?

With most tracks I cannot tell the difference between VBR 0 and lossless, with a very very small number (~1) of difficult tracks I can.

I do not have Yo Yo Ma's Bach Cello Suites but I do have several other versions and I can believe that they would be quite difficult to render to lossy without problems.

Why not try a few more conversions and see how many tracks you can tell apart. Perhaps you have better hearing than you think.

Remember that the psychoacoustic model throws away vast amounts of data that it predicts will not be missed, but models are more or less good and sometimes don't work very well. Most of the time for most people 320Kbps is perceptually transparent, for me VBR 0 does that as well (almost all the time) .

For info what version of CDEX and what verison of Lame are you using ?

For experimental purposes it is best to back convert to a single format so that when you ABX in FooBar there is the same playback operation on both tracks.

So both tracks are wav or flac , otherwise you have a separate extra conversion on one track, probably makes no difference but might be worth trying.

I'm sorry to say that, but this thread is useless if you don't upload a 10 sec snippet (cut it with something like GoldWave) of the lossless file and describe what differences you heard in that short sample.