Audiophile… or not?

rs2000

I'd agree that high bitrate mp3s encoded with a modern encoder are usually indistinguishable as long as we're talking about common full mixes and not one shot samples.
What Beato does here is kind of a subjective over-simplification though - I prefer a more neutral analysis like this one, done with different listeners and different music styles rated separately:
https://hydrogenaud.io/index.php?topic=113324.0

mp3 encoding is not about high frequency limits. It's much more about spectral artifacts coming from sub band quantization changing from mp3 frame to mp3 frame. This is something no oscilloscope or spectrum analyzer will be able to visualize.
A good experiment is to take one of the spectral denoisers (Adobe Audition noise reduction based on noise profiling or iZotope Denoiser) and turn down the noise to extremely low levels until artifacts become audible. That sound is somewhat similar (if highly exaggerated of course) to what mp3 compression does in the worst case.

Another nice chart from the great hydrogenaudio forums:

Source: https://hydrogenaud.io/index.php?topic=120166.0

Side note: I've switched from WAV to mostly AAC 160kbps for my samples on iOS and I find the iTunes encoder sounding "perfect" (i.e. insistinguishable for me) with much better transient behavior than mp3, which was apparent in short percussive synth sounds and samples with sharp transients like snare drums.

LinearLineman

@rs2000 said:
I'd agree that high bitrate mp3s encoded with a modern encoder are usually indistinguishable as long as we're talking about common full mixes and not one shot samples.
What Beato does here is kind of a subjective over-simplification though - I prefer a more neutral analysis like this one, done with different listeners and different music styles rated separately:
https://hydrogenaud.io/index.php?topic=113324.0

mp3 encoding is not about high frequency limits. It's much more about spectral artifacts coming from sub band quantization changing from mp3 frame to mp3 frame. This is something no oscilloscope or spectrum analyzer will be able to visualize.
A good experiment is to take one of the spectral denoisers (Adobe Audition noise reduction based on noise profiling or iZotope Denoiser) and turn down the noise to extremely low levels until artifacts become audible. That sound is somewhat similar (if highly exaggerated of course) to what mp3 compression does in the worst case.

Another nice chart from the great hydrogenaudio forums:

Source: https://hydrogenaud.io/index.php?topic=120166.0

Side note: I've switched from WAV to mostly AAC 160kbps for my samples on iOS and I find the iTunes encoder sounding "perfect" (i.e. insistinguishable for me) with much better transient behavior than mp3, which was apparent in short percussive synth sounds and samples with sharp transients like snare drums.

You don't say!