Live data from Hacker News

The Ghost in the MP3 (2014)

theghostinthemp3.com

71–72 of 72 posts

Re: The Ghost in the MP3 (2014)

#71
post #29
post #8

A few years ago I did a much more detailed comparison of lossy audio codecs: https://andrewrondeau.herokuapp.com/objectively_comparing_au... TLDR: If you need to go lossy, use Opus. At higher bitrates, it's the only lossy codec that has a flat frequency response.

You cannot objectively analyze the performance of perceptual audio codecs. They exploit subjective perception of audio signals... I haven't dug into your analysis but on the surface it appears quite biased against MPEG codecs that exploit masking to determine quantization levels (one of the major advantages being that humans can't perceive quantization error in the presence of loud/noisy content, which can be exploit…

I demonstrate the equalization differences.

That's about as objective as you can get.

Post reply on HN