Very wise advice from a very wise HF'er. The only thing I'd like to point out is that MP3, while standardised in 1992-3, is as old, but primarily much newer as a 'standard' than ATRAC is. ATRAC was popular back in 1994 and gained massive use till about 2000 when MP3 started to wild with file sharing. ATRAC, of course, has never been made to sound like the source material, but been compressed in a way to make it 'sound better' than the source by compressing certain models to make the bass and high mids stand out in more exciting ways.
The others tend to be made to emulate the source as closely as possible, but of course are not perfect. In the early days, as you (did I read between the lines correctly?) seem to suggest, OGG was ahead, but today, there is no clear winner at 128 and at higher rates (which aren't tested as often), the winner seems to be harder to parse.
Here, golden-eared listeners chime in all the time about how much better lossless is than lossy, how much better OGG vorbis is than MP3, etc. and so on till they've spittle-coated their keyboards. But I've met no one here actually doing tests of the files in controlled environments with volume-matched, blind, multi-song tests.
Of course, that shouldn't matter to anyone - how you enjoy your music should be your issue only. The problem is when (and I've seen this even from the mods here) you get offended and attack other people for using codec you don't agree with or that you think are crap. Mods especially should watch out for this as it is brutal to see them attack users for using 128kbps files.
Today, I pretty much stick with LAME though I would like to see AAC officially take its intended spot as the successor to MP3, but I don't think that will ever happen.
Originally Posted by blessingx
Also keep in mind newer than MP3 codecs (Vorbis, AAC versions, ATRAC versions, etc.) are aiming at 'similar quality at smaller sizes' not 'better quality at the highest bitrates'. Now that should be that the case all the way up the ladder (the 'tuned for ~128'/'sweet spot' discussions are supposedly false), but at the top ends 256+, there's been very little tests. So though it's likely that's the case, it's probably best to be careful when saying "Ogg sounds better" unless you've compared 320 kpbs files (Vorbis obviously goes even higher, though I don't know why anyone would use 500 kbps lossy).
Also of note, is a recent HA test (and discussion) over Ogg Vorbis underperforming specifically on a selection of Classical tracks. Something to consider depending on collection and how much faith you put on the test.