This will be a topic on NPR Science Friday (http://www.sciencefriday.com/program/archives/201202106) tomorrow at 12:30 PST. I will be one of two guests talking about this and other audio topics related to perception and measurement of sound quality. Should be fun!
Fantastic! I will be listening.
thanks for the heads up!
As always podcast downloads of the program will be available from NPR's or Science Friday's websites.
I'll try to be around for that. Sounds great. Love to be there when misinformation gets quashed.
For those who missed this, this is a link to the web page with broadcast:
https://www.npr.org/programs/talk-of-the-nation/ (https://www.npr.org/programs/talk-of-the-nation/)
Here's a direct link for the "vinyl sound" segment:
http://pd.npr.org/anon.npr-mp3/npr/totn/20...210_totn_06.mp3 (http://pd.npr.org/anon.npr-mp3/npr/totn/2012/02/20120210_totn_06.mp3)
File size is about 12 MB. Play time is about 25 minutes.
I enjoyed this very much. Well worth a listen for me.
The part about the media and listeners bias was really spot on!
Sadly, the standard dropped rapidly when it got to lossy compression - a lot of subjective and undifferentiated claims (not by Sean Olive though).
A real letdown was the kind of attention Mr. Metcalfe gave to the difference signal - does he know anything at all about psychoacoustics!?
A real letdown was the kind of attention Mr. Metcalfe gave to the difference signal - does he know anything at all about psychoacoustics!?
I've seen the phase reversal trick used on a few occasions now to show "how much is missing with mp3". What's interesting is that, especially when presented with higher bitrates, they can't actually make the case that people can hear the difference when you play them side by side, so instead they have to present you with "the differences" so that you'll understand how truly horrid mp3 encoding is.
I'd love to see someone - not necessarily Mr. Metcalfe, but just those who routinely make such claims (Fremer, for instance) - do the same with a vinyl album vs. their hallowed "master tape"...
I found the podcast and your input very interesting, Mr. Olive.
I've seen the phase reversal trick used on a few occasions now to show "how much is missing with mp3". What's interesting is that, especially when presented with higher bitrates, they can't actually make the case that people can hear the difference when you play them side by side, so instead they have to present you with "the differences" so that you'll understand how truly horrid mp3 encoding is.
The difference signal has no meaning at all regarding lossy encoding. How is MP3 "horrid" if it cannot be ABXed/differentiated with certainty? If a lossy codec is transparent to most listeners it achieved its purpose. In my opinion MP3 is horrid because of the ID3 tagging disaster and non-gapless design, the audio quality is certainly not a major issue, as has been shown in community listening tests. I, myself, have a very hard time ABXing LAME -V4.
Found the part at 20:12 fascinating, i.e. letting us hear what bits had been "left out" of the lower quality mp3.
The difference signal has no meaning at all regarding lossy encoding. How is MP3 "horrid" if it cannot be ABXed/differentiated with certainty? If a lossy codec is transparent to most listeners it achieved its purpose.
I should've been more clear - I also would've put quotes around "horrid" in my original post.
In my opinion MP3 is horrid because of the ID3 tagging disaster and non-gapless design, the audio quality is certainly not a major issue, as has been shown in community listening tests. I, myself, have a very hard time ABXing LAME -V4.
Agreed. It is for this reason that I still use Vorbis exclusively in the lossy domain for my own encodings.
I should've been more clear - I also would've put quotes around "horrid" in my original post.
I'm glad we can agree on that.