A good 320 MP3 is essentially indistinguishable from WAV#
If you encode a 320 kbps MP3 properly, constant bit rate 320 with the LAME encoder at its default settings, then under normal listening conditions you really can't hear the difference from the uncompressed WAV.
It is not a night and day difference that you hear once you have "trained your ears".
A blind test is only as good as its files#
A blind test built on badly encoded files is testing the encoder, not the format.
YouTube's codec does not obliterate differences#
Every time I post an A/B, someone says you can't compare anything on YouTube because the compression completely destroys the audio. So I tested it. I made two subtle EQ moves, a 1.5 dB boost at 400 Hz on a ukulele recording and a 1.5 dB boost at 10 kHz on a drum loop, and passed an ABX test on both in my DAW. Then I uploaded them, recorded the playback with a screen recorder that turned the level down by 20 dB and added its own lossy codec, and did the ABX again. I could still reliably identify both.
YouTube does change the audio. There is a high-frequency roll-off around 17 to 18.5 kHz that most adults can't hear, and a tiny bit of drift, smaller than a turntable's. But considering how far it squeezes the files down, it does a remarkably good job, and critical listening on YouTube is not a nonsensical activity.
This website uses no cookies directly. However, embedded YouTube videos may set cookies and process personal data once played. By interacting with the video, you consent to YouTube’s data processing. See YouTube’s privacy policy for details.