i just read some reviews and stuff, and decided test it myself, there are 160 and 192 kbps tests:
nothing new so far … ogg is superior. only problem that mp3 is around for a longer period of time and is the “standard”. but it sucks definately.
well, i made the test coz i read some bullshit review where says “ogg sux at 160 kbps” or something…
good that we already decided to use ogg instead of mp3 in renoise
I will agree that Fraunhofer and Xing are mediocre perceptual encoders when it comes to optimum quality. Both companies have focused primarily on low bitrate streaming applications and have left quality striving audiophiles in the dust. LAME, on the other hand, is a different story. Not only does it produce files that are playable in all .MP3 players, but it has vastly superior psycho-acoustic and noise shaping models for audio compression. Ogg is very good, but LAME is king.
I was expecting that atleast one person would grab this oppertunity and say: “Zed’s dead bayby, Zed’s dead.”
Oh how fun that would have been.
On the other hand it might be that all of you saw this but just thought it was a lame joke… hehe
Zed: Could you include some charts on the lame encoder performance?
It would be interesting to see them in comparision (is that a gramaticly correct form of compare? - I’m a swede).
I would do it myself if I only had a computer… but I don’t. “So pretty please, with sugar on the top. Make that f-ing chart”
(For those of you who haven’t seen Pulp Fiction… I’m not being rude. I’m just trying to be funny… I will stop that… right now )
lame is not king, it sucks as much as mp3.
and please just make a listening test to lame, mp3 and ogg at very low bitrates (like 24, 32 and 64 kbps). And dont forget to close your mouth after.
why are 21 and 55 world’s most important numbers?
heh, yeh, i’ve seen pulp fiction, but i don’t like that zed actually, i took “zed” nickname many years ago because i love my Spectrum and there is “z-80” microprocessor, and i love to code for it, i just took “z” for nick name.
about encoders, yeh, i could, just need some time for that, i also want to encode different music style (the problem is i don’t have that much cd’s with such music). we’ll see.
Why don’t you post which styles you need, and I (and others) can send you what I have. Somebody probably have it.
man, i’ve got friends i guess, i’ll get some cds soon i’m just into electronic stuff a lot, and not used to ask for some pop cds or something…
btw, i’ve just done test of lame encoder, and i should say it gave VERY similar results to WavLab’s encoding… check the same url…
its not 21 and 55 which are the most important number, but 21 divided with 55. this leads roughly to 0.39 or 39% and this is the so called “golden number” which represents a psychological value.
if something good happens 39% of the time you wait for it, then humans call the whole thing “positive”. for example if you try something alot of times, and you are successful more than 39% of the all tries, you have the feeling that you made it.
so basically it all goes down to : make your life “good” for atleast 39% of it and you have a “good” life.
I have no interest in low bitrate encoding. If you read my post, I mentioned “quality striving audiophiles”. Low bit rate encoding, regardless of codec, sounds mediocre in my opinion. Of course, everyone’s quality tolerance is different, but I demand true CD quality from my MP3s. I encode MP3s for archival purposes and want the optimum quality possible. File size is not nearly as important to me. Try encoding something with high levels of dynamics and rich overtones like acoustic piano or chamber music and you’ll soon see the limitations of most encoders. LAME with the --r3mix preset is unmatched to my ears at this time.
well, but I guess an encoder which is able to produce superior results even with low bitrates should be capable of much more if using high bitrates …
not true actually with high bitrates case.
With high bitrates LAME is really better than Fraunhofer based encoder while Fraunhofer is better in lower bitrates.
The difference is because Lame simply has much better psychoacustic model when working with higher bitrates. There are tons of tests on the net here is first I found which compares Lame, Xing and Fraunhofer.
So when encoding to 192kbps (something like CD quality) or more you should use Lame. For 128kbps (radio quality) use Fraunhofer.
these tests looks very fishily to me, really, i’ve done test with lame 3.92, and i’ve got very different results to these tests, well, at least, i’ve done my tests honestly, and i’m not against mp3, i just choose the best for me, and now i choose ogg.
i’m sorry, but your test has absolutely no relevance.
of course, i’m a total loser, i wanted to demonstrate to all the world that ogg better than mp3.
all you did was a
frequency analysys on decoded files, and that has nothing to do with
how the files actually sound.
of course, sound has nothing to do with frequency, sound is about radiation.
i don’t even know if you made double-blind listening tests (probably not)
of course not, and i hate to listen to the music. music sux, silence rocks.
and what audio equipment you have, but let me assure you that ogg
still trails mp3 in sound quality by a slight margin (confirmed by DBT’s on
now i’ve got $8000 equipment (Vorbis payed me for my tests)
its biggest advantage is being royalty-free. well, that,
and low-bitrate modes which could come in handy for mp3/ogg walkmans
that’s right, i’ve bought a walkman for these $8000, and the sound is great.
Only reason I use mp3 in stead of ogg is cos my mp3-discman. It’d be cool to see some discman with ogg-support… Maybe even mod-support! Why, oh why can’t somebody make some portable player you can install plugins into? Winamp should do something like this! =)
Ah well, I’m going ogg cos it’s royalty free, not cos of quality. Can’t hear much difference anyway… There should be better information-fields, like the id3v2 in mp3, now that’s good!
ogg support any info tags, it’s totally customable, you can even attach a picture to the ogg’s tag.