Sunday, November 22, 2009

Lossless: Keeping your options open

There have been a number of articles 'comparing the sound' of music encoded in a lossless way vs. a lossy way, for example, this one at Slashdot:

Slashdot News Story | Can We Really Tell Lossless From MP3?

I think this question misses a key point (though the original article does touch on it). If you have your content stored in a lossless form you can play it in a number of ways, including compressing music so it sounds as it would if it were being decoded from a lossy source, but you can't do it the other way around.

For example, if I start with a FLAC music file (at any given sample rate) I can convert it to MP3 and listen to the resulting stream with MP3 compression artefacts added (just as they do in the article), or I can listen to the music directly from the FLAC file through something like the Linn Klimax (also mentioned on the BBC, btw). It is not possible to start with an MP3 file and listen to the 'MP3 sound' and then somehow get the 'FLAC sound' from the MP3, because the MP3 has lost some information (hence lossy). The lost bits are just not available. Ever. Lost for good.

You don't need to use something as expensive as a Klimax to enjoy music encoded using FLAC. There are lots of free player programs for your computer (e.g. Play for the mac) and home music stystems (e.g. the Logitech Squeezebox). Yes, lossless formats do take up more space, but bulk storage is quite cheap today and getting cheaper, so this is no longer a really significant issue.

So, keep your valued music in FLAC format and keep your options open. You may not be able to afford a Klimax now, but who knows what the future might bring or how your ears might change.

And, of course, avoid DRM like the plague.


Andrés said...

If you use Windows, then you should consider something like Winamp (or a player that accepts Winamp plugins), because there's a Winamp plugin for basically everything. You may also be interested in Speex.

Andy said...

Also, if you are ever going to do any processing on the sound (e.g. running it through an eq, resampling, etc) then this stuff can end up amplifying noise and making it more noticable.