Ogg Vorbis And Xiphophorus

mental666 writes: "Ogg Vorbis is a fully open, non-proprietary, patent-and-royalty-free general purpose compressed audio format for high quality (44.1-48.0kHz, 16+ bit, polyphonic) audio and music at fixed and variable bitrates from 16 to 128 kbps/channel. This places Vorbis in the same class as audio representations including MPEG-1 audio layer 3, MPEG-4 audio (AAC and TwinVQ), and PAC. It sounds great; I've tried it and while the file size is slightly larger than mp3, the quality is there. There's also an xmms plugin in the cvs module. For details check out The Xiph Web site."

Comments

Wow.

[Chris Johnson]

I _really_ like this guy's attitude. My initial reaction was "But I have five CDs worth of songs released as mp3!", with concern about whether this could be used as a weapon to hurt mp3 and damage the viability of _any_ unrestricted digital music format (I specifically mean 'unrestricted for people to download', i.e. no 'security features', I don't mean unrestricted for the programmer)

Then, reading the inventor's interview, I was blown away by how good his attitude was. He GPLs: I GPL. He _specifically_ and eloquently objects to the notion of 'security features'. He's clearly coming from an audiophile sort of place- I'd given up hope of finding that in compressed digital music at _all_- I currently like mp3s because they distort pretty smoothly and unaggressively, the attempt to deliver 'bright clear spectacular highs' will inevitably produce horrid distortions and breakup, just as it does in hardware and analog audio, only much worse. Now here's somebody with an actual clue concerning himself with issues that mean a great deal to me! Wow. I kinda like that.

My take on the matter as an active and increasingly prolific net musician is this: I'm happily using mp3 128bit from BladeEnc, partly because I dislike the notion that I have to shell out big bucks to Xing or somebody (no laughing in the back there! I hear Xing is actually pretty unmusical, 'flashy' sonically) just in order to produce my work- also I know mp3 is well established, and again it degrades kind of euphonically compared to some stuff I've heard.

CD audio is _already_ severely severely compromised. It is. I can't get over how people get into these arguments over what sounds better or worse and miss how degraded CD sound already is... I would say with my recent work, it is about the same degradation in quality going to 44.1/16 as it is going to 128k BladeEnc. I'm not fooling- this is due to my use of a customised 48/20 ADAT with custom coupling caps throughout the analog board, combined with the removing of all the (wave soldered!) anal hiss control caps which are only there to change the op-amp noise levels from -104db to -73252346 db for crazed silence freaks ;) oh, and it feeds a custom analog board through custom cabling into the computer's A/D converters through more custom cabling. (To hear all this working, listen to songs off "anima", which uses the whole setup- expect the beginnings of a techno album soon. The album "Hard Vacuum", which is pure Noise, is actually even harder on an encoder, recorded direct to 44.1/16 and I kept the masters...)

Which is to say, I _know_ mp3 is quite flawed. I can mix and master to make this as unintrusive as possible, but it's still on a level with very good cassette tape (say, Nakamichi Dragon decks) at its best. That can be quite listenable. It can also be useful- when you start dealing with very high transparency equipment like I use, the performances have to be at a higher level, and it's easy to fake yourself out by mixing so that the slightest mistakes become obvious. If you record pretty dry it can be very unforgiving, though real punchy and involving- there's no room for error at all. I do this and then mix to 44.1/16 and then mp3 and by the time that's done, it's much harder to find flaws because the transparency just isn't there anymore :)

However, I will still be among the first to go with Nanny Ogg (gotta love the pterry references! Didn't even clue to this right away) and will even do special high-resolution mixes tailored to stress the new format to the limit- given two things.

• tools that run on a PowerMac, or which can be easily built in MPW, which I'd be willing to struggle with for the cause- I'm _no_ C programmer. I'm a drummer ;) (well, and bassist and guitar player etc etc etc)
• somewhere to PUT the damned music. I have 177.1 megs of _compressed_ music on the Web at mp3.com, _all_ freely downloadable. That is orders of magnitude beyond what I can get for my own website, which I'm quite happy with for what it is, but it charges $1 a meg, roughly, and I'm paying $45 a month all things considered.

I like to think I am the poster child for audiophile geekness ;) God knows I am ruthless enough with my equipment hacking, and I am quite capable of producing jaw-dropping sonics, and even have credentials including articles in internationally published audio journals. Maybe that makes me more prone to embrace a new format that is genuinely more capable of delivering sonics in line with what I'm after. I can only say that I'm not getting them now, am reasonably comfortable with mp3 in general terms in spite of this, am typically hostile to things like flashy Microsoft formats tailored to outmarket mp3- and from reading the views of this format's inventor (especially about 'secure' formats) I find I would really like to support Ogg- and doubt there are many other musician audiogeeks who could outdo me at producing codec-show-off materials in a musical sense. Hell, all I'd have to do is put things more up-front and finish up the multiband compression for mastering, and I could get a sound that's airy as hell and still musical. mp3 won't handle that and I know it- CD-audio barely copes with it already, mp3 would go all wavey and so I mix for soundstage. It would be _so_ _easy_ to make mixes that went a quantum leap beyond mp3. And the next album I'm doing will kind of _want_ that sound, I already know- and mp3 will be a compromise that I accept because of where I'm publishing- and I'll have all the masters handy for anytime I want to go beyond that.

I _like_ hearing about this stuff. I hope this works out. I hope this sees Slashdot again and somebody comes up with a Mac version of the programs, no matter how klugey, so I can work with the format and see what I can get out of it.

Count me in- just so happens that the contract with mp3.com is NONEXCLUSIVE rights to my recordings. If anyone sets up an Ogg Music site, I can right now throw five albums of original material on it, producing custom mixes (which I needed to do anyway for a private 'audiophile CD' project I was working on) and at least some material (Extended Play, Hard Vacuum, and anima) of very demanding audiophile quality to show off the format. All I need is the web space and the software...

(*g* dare I say it, wouldn't it be cool if I was some kinda bigshot on mp3.com by the time this happens? I'd _loooove_ to be on some sort of Successful MP3.com Musician soapbox when I start saying these things like 'use GPLed codecs' and 'no secure format bullshit' and 'there's this format called Ogg' ;) Therefore, I would be thrilled if people went and downloaded music from http://www.mp3.com/ChrisJ which is all free as in beer and speech (meaning, if you want a sample from it I'll actually let you sample individual instruments off my master tapes with my total blessing and cooperation, instead of being a dick and forbidding it), and which is fine music and great for showing off stereos and much of it makes for a nifty Bass Test, especially "Koala" which has huge warm fretless bass and a nearly subsonic string pedal tone to freak people out with... end parenthesis)

:)

Re:Isn't the real issue stream control, not codec

[Ed Avis]

I think you left out MP3 from your list of successful formats...

'Stream control' is a red herring. If you can
listen to it, you can record it. There are sound
drivers for Windows and for Linux that let you
save the output stream to disk, so you can make
perfect digital recordings of anything coming
from RealPlayer and other 'secure' players.

The Corollary Slashdot Effect

[Dr.Evil]

The fatalist attitude of so many on Slashdot amazes me sometimes. Here we are, a community numbering in the six figures (maybe even seven by now), and we sit here and go, "Too bad it won't make a difference." Hundreds of thousands of users could make a huge difference, if we just agreed to do something together! Heck, just going and looking at the web sites in most stories on Slashdot is enough to bring mortal servers to their knees!

We don't have to overwhelm MP3 to make Vorbis work - we just have to create a niche market. Especially if clever programmers write converters (tricky without compunding artifacts), Vorbis can stand on its own.

If we're serious about supporting Open Source, then we should support it right down the line - no patent-encumbered formats like MP3 and GIF. If we all live by that, then new standards like Vorbis and PNG will win in the end.

Some questions answered

[xiphmont]

Hi folks,

I'm taking time to answer a few questions as it seems a number of vocal folks have started posting without looking at the Vorbis web pages first. Practically every question and musing here is addressed there...

First off, I wasn't ready for this. Vorbis is not at release, although I hope that will be soon. I'm not releasing before it's ready.... and it isn't quite ready. Most of the fun stuff has been going on on a CVS branch; the mainline is only a functional, stable, dull, unimpressive version for starting application work. That way when Vorbis *is* released, all the Sonique, XMMS, Winamp, kmpg and Freeamp folks will have to do is recompile.

Second, the name 'Xiphophorus'; the organization is a democracy and I was outvoted. I personally like to emphasize 'xiph.org'. I rather Like 'Ogg' and 'Vorbis' though. Oh, and it's not Jesus spanking a snake. It's Thor, Mjollnir and Jörmungandr. I have a page about the names/logos; go read it.

Lastly, distributing hacked up encoders right now is fine, but keep in mind, that as soon as the new psychoacoustic engine is merged into the CVS main branch the bitstream format will change. The change is minor, but it will break existsing streams. That will happen this week, so you don't need to contain your enthusiasm too long :-)

Monty
xiph.org

The name is 'Ogg'. That's harder than mp3?

[xiphmont]

Hee hee. Let's ignore everything even remotely having to do with the format and it's implications and flame about names. This is more fun :-)

The name of the format is 'Ogg'. Just 'Ogg'. It has less unique characters than 'mp3' and can be pronounced in one syllable. Where I come from, that counts as pretty easy. Try it.

• "Fish"
• "Cat"
• "Ogg"

Vorbis is the name of a CODEC that Ogg uses. Just like 'Sorensen' is one of 'Quicktime's' video codecs. 'Ogg' is much easier than 'Quicktime' too.

...and the file extention is *also* .ogg. In the future when you get an ogg file, you know an .ogg player will handle it. How could you get simpler?

But wait! There's More!

...well, no, actually, there isn't. Anything more would make this harder, and we're all about 'easy'!

Monty xiph.org

Why Ogg Vorbis is better

[aat]

Here's a link to an interview with the author, with his explanation of why vorbis is better than mp3.

http://www.advogato.org/article/56.html

Arun