Tweet
Did anybody test extracted audio from cd using flac vs mp3? Appreciate if you can share album title and I plan to do testing on any audio quality difference as well in my home theater.
-
-
Re: flac vs mp3
It would depend on bitrate used for mp3, it would potentially be easy to hear the difference between 128kbps mp3 and flac, yet very difficult 320kbps mp3 vs flac. -
Re: flac vs mp3
Lots of work has been done on testing the transparency of mp3 vs lossless (flac etc). As Spoon notes, very unlikely that you could detect lossless vs lossy at high enough bitrate mp3 for virtually all normal music (i.e., not "problem" samples of music). See this as one example on this sort of testing:
Also, install music player foobar2000 and install ABX component and test for yourself. You'll be surprised at the level that mp3 becomes transparent (indistinguishable from lossless).Comment
-
Re: flac vs mp3
Thanks. I went through article and it is good to know that mp3s with 320kb are not anyway inferior. i will keep my flac for future and try to use mp3sLots of work has been done on testing the transparency of mp3 vs lossless (flac etc). As Spoon notes, very unlikely that you could detect lossless vs lossy at high enough bitrate mp3 for virtually all normal music (i.e., not "problem" samples of music). See this as one example on this sort of testing:
Also, install music player foobar2000 and install ABX component and test for yourself. You'll be surprised at the level that mp3 becomes transparent (indistinguishable from lossless).Comment
-
Re: flac vs mp3
I know lots of people use ABX to test between 320Kbps and FLAC but that is heavily flawed.
In short durations, things like loudness and EQ plays more of an effect on the perception of quality than anything else. There is a reason for the loudness war at the detriment of sound dynamics. So, if you AB flac and mp3, you're testing the loudness and EQ of the samples rather than the quality of the encodings. Unless you know what you're looking for, if the loudness and EQ are matched you'll have a hard time telling the difference.
This is similar to the "Pepsi challenge". You take a sip or two drinks and see which one you prefer. It's always Pepsi and the reason is that the sweeter drink will always win in the sip test. Doesn't mean you'll enjoy the entire can of the sweeter drink more.
I think FLAC has better sound-stage and sharper attacks. In MP3s, instruments meld into one another and move around and has boinging and smoothening effect. Cymbals and hi-hats are obviously affected in mp3 and the snare clatter is gone in MP3. Sharp guitar fuzz and synths buzz are gone in MP3s as well. When a lot of stuff happens very quickly, you can hear MP3 butcher the cymbals. I almost brace for that noise like high end when the whole band comes in, even when I'm listening to FLAC - but its probably due to clipping from high loudness as well.
But, my point is that don't rely on ABX. It doesn't work on high birate mp3 and FLAC. However, you should listen to the mp3 version and FLAC all the way through a few times and see which you prefer. I've found that I like FLAC better and I find mp3s in high volume annoying but not every player supports FLAC. It could all be psychological as well.
I think Mp3 is the only codec which doesn't mess the EQ of the original source. I can easily tell the difference in some of the other audio codec just by ABing them. WMAs boost the highs, AAC cuts the mids and OGG has its own weird thing with a weird high end. The codecs are always changing so the newer versions are probably different - I haven't done a recent test.Comment
-
Re: flac vs mp3
A couple of misconceptions in your statement. There is nothing about an ABX test that requires "short duration" listening. And a proper ABX test will carefully adjust the loudness of the two samples to match perfectly (for example, foobar2000 with the ABX component). This is because you are correct that even an imperceptible slightly louder track will often be considered "better" sounding. You might find this large scale public test of mp3 vs FLAC interesting reading:
Comment
-
Re: flac vs mp3
ABing also has its own set of psychological problems.
If you match loudness and EQ in both samples, it becomes hard to make a decision but you are required to make a decision. So, people just base their decision on something small, like how the highs sound or how the lows sounds, instead of the general experience. People want to find that one weird trick that will make their decision super easy.
With ABX I can tell the difference most of the time between mp3 and flac. The short gap in playback when switching between A and B in the foobar plugin makes everything harder though (at least for me). It's not 100% for me between mp3 and flac but probably 70%-80% (after a few tests, I get tired so I haven't done a long test and I get better results if I listen longer).
Anyway, I feel FLACs have a more pleasurable listening experience on the computer (I have a discrete PCI 24/192 card, a 100W amp and reference speakers, cheap sub - not super high end). The iPhone's sound is not good enough to tell the difference - there is a blanket over speaker thing on it. I think with FLAC I find myself admiring the instruments and recording lot more than with mp3. With mp3, I listen to the voice mostly and I feel the instruments like to hide. Could be psychological though as well.Comment
Comment