Converting MP3 to OGG

[Radiant]

Can someone suggest me a good (and free) tool to convert MP3 music files to OGG? Thanks!

[Ponch]

I use gold wave. It's an older program, but it works well.

However, you don't want to go from a compressed MP3 to a compressed OGG. It will sound like crap. Instead, convert it to a WAV file, then convert that to an OGG. That produces good results for me, anyway.

[ShiverMeSideways]

Audacity, it takes a few steps to set up, but it works with all three formats, mp3, ogg and wav.

[Snarky]

Uhmmm... when you convert an mp3 to a wav file, what you end up with is basically the raw wave data of the compressed file. In other words, exactly the same as what gets played when you play the mp3 file. You're not going to get back any of the audio data that was lost when the file was compressed in the first place.

So you should get exactly the same result, if not better, converting directly from mp3 to ogg as going via a wav file, assuming the converters are doing their job properly.

[Ponch]

you should get exactly the same result, if not better, converting directly from mp3 to ogg as going via a wav file, assuming the converters are doing their job properly.

Really? Because I've tried that and it sounds muddy or murky when I do. But I'm also using an old program, so maybe that has something to do with it? 

[David Ostman]

What happens is basically the same as what happens with a lossy JPEG that you save into different formats. Easy to see what happens with a visual demonstration:



In this case you might as well just save it directly to TIFF/PNG/TGA and skip the BMP.

[Snarky]

Well, it's not exactly the same since OGG is also a lossy format (unlike png/tiff/tga), so you're going to get additional degradation at that step. That's why direct conversion from mp3 to ogg could conceivably give better results, since a smart converter could be aware of the ways it was originally compressed, and recompress it in a way that didn't throw away even more data (as opposed to compressing a wav file, in which case you're just taking whatever audio data is there and doing the best you can with it).