Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Why Ogg Vorbis?

By Idioteque in Op-Ed
Sat Jul 20, 2002 at 03:33:46 PM EST
Tags: Music (all tags)
Music

Today, the Xiph.Org Foundation released Ogg Vorbis 1.0. It was an exciting day for the Xiph.Org Foundation, the Ogg Vorbis developers and the users of Ogg Vorbis. I'm sure many users were excited to download the new encoder and try it out on some of their favorite songs, performing their own brand of listening tests, deciding what quality setting is to their liking and hopefully marveling at the great fidelity at such low bitrates. While going through these motions myself, I started thinking about why this project is important and why, for me, its carries an emotional attachment, at least more so than other open source projects.


Any song, almost immediately after it starts playing, brings about an emotional response. The responses vary from song to song and person to person, but we form an attachment to certain pieces of music either due to their content or their ties to a specific event in our lives. I'm sure we can all come up with a situation where a song played an important role, or a song reminds of that situation. Music is important to many of us.

This in turn, makes audio encoding important, at least by association. Many of us listen to music at work and at home on our computers. We listen to MPEG Audio Layer 3 (MP3) files we have downloaded and we listen to our own music, either ripped from CDs or on the CDs themselves. Many of us may have at least a gigabyte of hard drive space just devoted to music; some of us, much more.

Now, if we have spent the time ripping our own CDs and encoding them, we want them to sound good. On the other hand, we don't have an infinite amount of hard drive space, so we use lossy encoding like MP3 to maximize our space. Sometimes with lossy encoding, there may be too much quality loss. We lose some of the nuances of the music and we are instead distracted by the artifacts produced by the encoding.

Ogg Vorbis files encoded at the same bitrates as MP3 files sound much better (not as many artifacts, more of the detail in the music is retained). This may not be enough of a reason to switch to Ogg Vorbis for some.

For the most part, MP3 is owned. We must pay royalties to its patent owners. No one owns Ogg Vorbis, Xiph.Org developed it, and you are free to do with it what you wish. No one has to pay Xiph.Org any fees per file encoded or software released. It's free! Better yet, it was developed by people who love music and wanted to make something better than what existed, and they actually succeeded in their task.

I'm emotional about my music, I cherish my CD collection and enjoy listening to music both at home and work. I am also emotionally attached to Ogg Vorbis. It was developed by people who love music, for people who love music with no other motives. Yes, I and others have been critical of the Ogg Vorbis project, for lack of documentation and the amount of time it took to get to 1.0. I wish I had not been as critical, because the end result was worth the wait. I guess one of the reasons I was hard on Ogg Vorbis was because I saw its potential and had developed an emotional attachment to it, through the music I had encoded with it.

Today was an exciting day for me. It was not something I could tell the average person on the street why I was excited, or even most of my friends for that matter. There have been other successful open source projects in the past (Linux, GNU, Apache, Python, Perl, and many more) and now Ogg Vorbis is among those, but for me, this project is special, because to me, music is special.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Poll
Give Ogg Vorbis a Try?
o Sure 14%
o I've been using it 56%
o No, MP3's are fine for me 13%
o No, I really don't care 3%
o Maybe, if I have some time 10%
o Ogg why? 1%

Votes: 159
Results | Other Polls

Related Links
o Xiph.Org Foundation
o Ogg Vorbis 1.0
o MP3 is owned
o and they actually succeeded in their task.
o Also by Idioteque


Display: Sort:
Why Ogg Vorbis? | 115 comments (103 topical, 12 editorial, 0 hidden)
Depends (2.75 / 4) (#2)
by JChen on Fri Jul 19, 2002 at 11:40:25 PM EST

Ogg Vorbis is on par with MP3 in terms of quality, but because it lacks corporate muscle and mass hype with its publicity mainly limited to those who are more computer proficient than the masses, I fear it will not swing the masses in its favor. It almost feels like the battle between the Linux crowd and Microsoft, sharing many of the same arguments.

Let us do as we say.
Does not depend! (4.80 / 5) (#3)
by tftp on Fri Jul 19, 2002 at 11:52:53 PM EST

because it lacks corporate muscle and mass hype with its publicity mainly limited to those who are more computer proficient than the masses

No "corporate muscle", ever, promoted MP3.

"Masses" do not encode songs. I think, maybe 0.1% of music listeners encode songs; the rest "leeches" them (possibly legally, if they own the CDs - I know several people who did that).

Success or failure of Ogg Vorbis among "masses" is irrelevant. The codec is out there, and developers, creators of things will use it if they want it. The opinion of masses is not interesting, masses do not create anything, they consume (by definition of "masses".) Regular [ex-] Napster user probably doesn't even know what this Vorbis thing is. All s/he cares about is that WinAmp plays .ogg - and it does. So no big deal here, as long as double-clicking on the song plays it.

[ Parent ]

Just a side note... (4.40 / 10) (#4)
by seebs on Sat Jul 20, 2002 at 12:01:13 AM EST

I just redid my database as MP3.  I considered Ogg, and rejected it:

* My DVD player can play MP3 CD's.  It can't play Ogg Vorbis.

* My car stereo can play MP3.  It can't play Ogg Vorbis.

* My portable music player can play MP3.  It can't play Ogg Vorbis.

It's that simple, for me.


That's too bad. (3.00 / 1) (#7)
by Farq Q. Fenderson on Sat Jul 20, 2002 at 12:34:40 AM EST

Because Ogg sounds so much better than MP3. I just can't believe that files so small could sound so good. (Well, barring various mod formats.)

Someday, hopefully, common hardware will support Ogg Vorbis as well. In fact, I wouldn't mind buying music in Ogg format, it's just that good, IMO.

farq will not be coming back
[ Parent ]

I guess it's a question of perspectives. (3.00 / 1) (#9)
by seebs on Sat Jul 20, 2002 at 12:41:39 AM EST

There's a tradeoff from quality to accessibility.  Music on a perfect quad-speaker system with a subwoofer in a carefully designed room with lightly padded walls might give me the best possible experience - but I'll take the car stereo, any day, by preference.

I might switch to Ogg if I could get a car stereo for it; I'd miss the portable MP3 player, but not much.  The car stereo, though, is non-negotiable.


[ Parent ]

Vicious circle (3.00 / 1) (#36)
by afree87 on Sat Jul 20, 2002 at 12:47:11 PM EST

If it can't be played on a portable player, you won't burn your files to it.
If you haven't burned your files to it, they won't make a portable player that can play it.

How about you take the initiative instead of making the electronics industry do so?
--
Ha... yeah.
[ Parent ]

It doesn't work that way. (4.00 / 1) (#38)
by seebs on Sat Jul 20, 2002 at 01:05:07 PM EST

Two problems:

1.  If I "take the initiative", I lose the use of my music until such time as the electronics industry may or may not support a new format.

2.  The electronics industry is not making any decisions based on what I do; they're making decisions based on polls and focus groups, on which I have no effect.

3.  It is much easier to make a player that supports both MP3 and Ogg than it is to keep a database in both formats.

The first step is clearly theirs, and I think they'll take it, some day.  Until then, I'll stick with stuff that works for me.

[ Parent ]

Ogg is a decision factor for future buys (4.66 / 3) (#57)
by fencepost on Sat Jul 20, 2002 at 04:20:16 PM EST

When there's a portable available that supports Ogg Vorbis, I'll look very hard at it.  I'm not going to promise to buy it because it may suck or cost a fortune, but I'll definitely be considering it.  More to the point, support for Ogg Vorbis will be a major decision factor when I replace my current portable.

Currently I use MP3s in two ways - on a PC at home acting as a jukebox, and on a portable that spends most of its time in my car or with me when I travel.  I could use Ogg at home, but without the portable support it'd just mean I had to buy another hard drive and re-rip my CDs while keeping the MP3s I currently have for use in the car.  Not worth it, particularly since I rip at very high quality (they're for my use not for sharing, and I have enough disk space).
--
"nothing really says "don't hire me, I'm an idiot" quite as well as misspelling "pom-pom" on your resume." -- former Grinnellian
[ Parent ]

On that note (4.00 / 2) (#67)
by Rob Greenberg on Sat Jul 20, 2002 at 07:33:17 PM EST

The licensing issues haven't quite come into play yet but they could.  Because the MP3 format is patented, it means that hardware manufacturers could be forced to pay licenses down the road.  

The beauty of Ogg Vorbis being patent and copyright free is truly remarkable.  This means that as hardware costs decrease licensing becomes a larger and larger percentage of the cost of production.  In the far east, Chinese and Taiwanese manufacturers are trying develop new DVD type formats so that they can avoid the licensing issues with DVD.  

The less the cost, the more likely the use.
I feel that Ogg Vorbis hitting the 1.0 mark is a great step forward for all of us because of the hardware cost issues that will become more apparent in the future.

-Rob

[ Parent ]

Ogg vs. MP3 (4.20 / 10) (#6)
by SMN on Sat Jul 20, 2002 at 12:23:42 AM EST

I've seen Ogg bragging about its patent-free status being a huge benefit over MP3 since its inception, and while I don't argue that it's nice to be entirely patent-free, I don't see what benefit it has for me, the consumer. There are plenty of free MP3 encoders around, and there are plenty of free applications capable of playing MP3s. There are now plenty of MP3 CD players. If there's some sort of premium/licensing fee on MP3-related products (I believe there is?), I haven't seen that hurt MP3's proliferation.

So what I'd like to see someone explain is, in practical terms, why is Ogg Vorbis' patent-free status so great? How does this impact consumers directly? Are the MP3 licensing terms somehow limiting MP3 technology? I don't see too many companies adding Ogg playback to MP3 devices, even though that feature would be "free" to implement (although I've heard this is because current Ogg Vorbis decoding requires floating point operations, while MP3/WMA do not).

When there's a reason to use free/open software instead of proprietary products, I'll happily do so. I was using Mozilla since something like Milestone 17, long before it was actually a better/faster/more stable browser than IE and other closed-source apps, simply because it was free, open, and standards compliant. But Ogg seems to be about offering a free and open alternative because they can, not because they should. I don't see the practical reasons why free and open are relevant here.

I see people saying that Ogg is better sounding than MP3 at smaller bitrates, but I really don't see that as too great an incentive. I'm the type of person who doesn't really notice the difference between your 64k MP3s and your 128k MP3s (although I usually rip mine at 128), and I happily listen to the radio all the time without any qualms about the sound quality. My car's speakers are a bit old and the bass rattles a bit; my friends can't stand it, but it's not bad enough for me to get new speakers. My MP3s are small "enough" that filesize isn't a concern.

So, for my needs, MP3 and Ogg would both be sufficient. But MP3s are common and can be played by many devices, which is a major advantage to me. On the other hand, I can't find any major advantage or killer feature in Ogg.

I'm certainly open to persuasion. Am I missing something big here, or is this a case of "free because we can, not because we should"?

-- Imagine how much more advanced our technology would be if we had eight fingers per hand.

Well, here's the thing. (4.80 / 5) (#12)
by regeya on Sat Jul 20, 2002 at 12:48:05 AM EST

Apparently Fraunhofer (or however it's spelled) just sorta tolerates free encoders. Being tolerated and being legal aren't entirely the same. I'm probably wrong on this, and will probably get a bunch of the anti-Free zealots worked into a lather. My apologies. ;-D

Also, a number of commercial projects would like to see Ogg grow and prosper, as they can't exactly get out of paying licensing fees for MPEG. Having a free equivalent to MP3 would help keep the costs of development down (which will hopefully be passed on to consumers.)

It's not a totally killer app, and it's hard to convince kids that they shouldn't use LAME unlicensed because it's illegal if all they're doing is trading MP3s via Limewire or some better client. But for legit operations, Ogg Vorbis is compelling.

My only reasons for using Ogg Voris are that, well, on this issue I've decided to be a GNUbie and do things nice and legal. While ripping CDs and taking the results to work might not be entirely legal, I try to keep it as kosher as possible by using Oggs rather than MP3s, 'coz my other choice on my platform of choice would be LAME, an unlicensed MP3 encoder. It may seem a little odd for a longtime KDE user to take such a stand ;-D but that's how I feel. If that were the only reason, I wouldn't use it anyway and just use LAME. However, I feel that Vorbis does indeed seem to produce better quality for the bitrate.

Heck, I can remember being told by several people that MP3 would never catch on. I'm hoping the naysayers are wrong on this one.


[ yokelpunk | kuro5hin diary ]
[ Parent ]

OGG distributed online (3.50 / 2) (#23)
by fluffy grue on Sat Jul 20, 2002 at 09:17:46 AM EST

I already have a hard enough time getting people to listen to my music in mp3 format. A few months ago on somesongs, someone had posted a few songs in ogg format, and pretty much everyone there said, "I'm not installing a new plugin just to listen to your music!" So it hasn't reached a critical mass just yet, and until it does, I'm still going to distribute illegally-encoded mp3s from my site.
--
"Is a sentence fragment" is a sentence fragment.
"Is not a quine" is not a quine.

[ Hug Your Trikuare ]
[ Parent ]

mostly supported already (5.00 / 1) (#45)
by uhoreg on Sat Jul 20, 2002 at 02:40:25 PM EST

FYI, the latest WinAMP supports vorbis out of the box. Sonique has supported it for a while. The only major (software) audio player left is iTunes.

[ Parent ]
Unfortunately... (none / 0) (#52)
by fluffy grue on Sat Jul 20, 2002 at 03:39:45 PM EST

I still use iTunes on my iBook (actually, I've downgraded to 1.0; lots of the "features" of 2.0 and especially 3.0 bug the hell out of me, especially with how iTunes assumes that it's the ONLY thing you play music through, when my entire collection won't even fit on my iBook - I have my own song collection synchronization thing so I can treat my iBook like an iPod), and I recently bought a perfectly-good car mp3cd player...
--
"Is a sentence fragment" is a sentence fragment.
"Is not a quine" is not a quine.

[ Hug Your Trikuare ]
[ Parent ]

you are not "the masses" (none / 0) (#80)
by codemonkey_uk on Sun Jul 21, 2002 at 10:58:31 AM EST

Apple is a niche market. The majority of online music consumers are using Windows, and either media player, or WinAmp.

Of course, that doesn't invalidate "I'm happy with what I've already got" argument for not switching - but it does make you reply a poor retort to the comment it was posted in reply to, which was suggesting that for most people playing ogg is already zero effort.
---
Thad
"The most savage controversies are those about matters as to which there is no good evidence either way." - Bertrand Russell
[ Parent ]

Except it's not (none / 0) (#83)
by fluffy grue on Sun Jul 21, 2002 at 01:35:33 PM EST

I'm sure that most people out there are happy with some old version of Winamp or Media Player, one which doesn't support OGG. At least, none of the (largely non-computer-junkie type) users at somesongs couldn't listen to a few songs which were posted in OGG format, and they were unwilling to go and even download a single plugin to do it.
--
"Is a sentence fragment" is a sentence fragment.
"Is not a quine" is not a quine.

[ Hug Your Trikuare ]
[ Parent ]

iTunes and Ogg (5.00 / 1) (#87)
by sjl on Mon Jul 22, 2002 at 01:00:26 AM EST

It Can Be Done. And has been done. I've actually managed it. Unfortunately, it needs a bit of fiddling... here's the details.

In order:

  1. Download the Quicktime Ogg components: at http://qtcomponents.sourceforge.net/.
  2. Install, as the documentation directs.
  3. Rename all your .ogg files to .mov (eg: .ogg.mov is probably the best bet.) iTunes will only pass a file to the Quicktime subsystem if it has a .mov file extension (what the HELL were these guys thinking? Oh, they weren't. Nevermind.)
  4. Import, and play away.
I seem to recall that you need to run some command to set the various bits of meta data for Quicktime to recognise it. Unfortunately, I can't remember what that command is. :-( I've sent a message to a friend that I gave all this stuff to a while back; hopefully, he can jog my memory (I'll post the details if/when I find them.)

Problems/drawbacks:

  • The component won't work with QT 6. QT has problems with variable bit encoding, according to the guys that did the component.
  • iTunes still won't encode the files for you -- you'll have to rip to .WAV, and then encode with the command line tool (unless there's some GUI tool for Mac OS X -- I don't know.)
  • iTunes won't import all the tags from the file (with iTunes v2 -- fixed in v3, according to the same guys.)
I've tested this with QT 5 and iTunes 2. Good luck. (Obvious comment: for various reasons, I haven't bothered re-ripping my collection. One of these days, in my Copious Spare Time(tm)...)

[ Parent ]
Found it. (none / 0) (#89)
by sjl on Mon Jul 22, 2002 at 01:27:46 AM EST

From the xiph archives (and also this one:

  • /Developer/Tools/SetFile -c hook -t OggS
  • open -a /Applications/iTunes.app foo.ogg.mov
(the first to set the meta data, the second to import the file into iTunes.)

Hope this helps,

Stuart.

[ Parent ]

I hope to not sound alarmist, but... (3.66 / 6) (#21)
by martingale on Sat Jul 20, 2002 at 01:57:38 AM EST

Are you a Windows XP user, by any chance? The license you agreed to states that Microsoft reserves the right to modify the contents of your computer, including deleting any MP3 files you might have, if it can be justified under Digital Rights Management.

Most (all?) of the MP3 format songs you have on your hard drive, if they were encoded by a free encoder which doesn't pay Fraunhofer's license fee for each encoded song, can be legally deleted by a Microsoft Media Player Update script at any time. *Poof*, no more music.

Of course, this supposes that properly licensed MP3 files can be distinguished from unlicensed ones. That's certainly possible by having for example the encoder add some kind of digital watermark identifying the encoder, in each properly licensed, newly created MP3 file. Clearly, the free encoder you use now, or even those which have been used in the past to create your collection, is/are unlikely to write such a watermark. So most likely, 99% of your current collection is clearly identifiable as unlicensed by an automated script. I hope you have made backups.

[ Parent ]

eggs milk toilet paper coffee filters bran (4.00 / 1) (#107)
by guergle on Wed Jul 24, 2002 at 02:20:57 AM EST

Your comment scared me a little, so I went and read the EULA and I can't find anywhere where it says anything about agreeing to have files deleted. Where'd you get that?

There was an interesting part though:

Except as otherwise permitted by the NetMeeting, Remote Assistance, and Remote Desktop features described below, you may not use the Product to permit any Device to use, access, display or run other executable software residing on the Workstation Computer, nor may you permit any Device to use, access, display, or run the Product or Product's user interface, unless the Device has a separate license for the Product.

I think that means you are not allowed to run X or VNC on an XP box, unless both client and server are running on licensed XP machines...

[ Parent ]

link to eula, or lack thereof (4.00 / 1) (#108)
by martingale on Wed Jul 24, 2002 at 04:21:57 AM EST

I should have linked to the EULA at the time, but I was lazy ;-). Actually, I tried linking to it on the web just now (since I don't run Windows myself) and I can't find a direct copy (ie I found form that Microsoft wants me to fill out and they'll send it to me in the mail).

However, the offending paragraph was widely discussed on the net at the time it was published. Here's a link to The Register, but I'm sure you can google for others.

Oh shucks, here's the paragraph anyway.

"You agree that in order to protect the integrity of content and software protected by digital rights management ('Secure Content'), Microsoft may provide security related updates to the OS Components that will be automatically downloaded onto your computer. These security related updates may disable your ability to copy and/or play Secure Content and use other software on your computer. If we provide such a security update, we will use reasonable efforts to post notices on a web site explaining the update."
You'll notice that "disabling the ability to use other software on your computer" is a very broad right, which is certainly compatible with the scenario I outlined in the previous post. I'm sure you can imagine other scenarios yourself.

[ Parent ]
ziplock bags, pickled walnuts, romaine lettuce (none / 0) (#109)
by guergle on Wed Jul 24, 2002 at 04:10:31 PM EST

My eula has slightly different wording:

"Owners of such Secure Content ("Secure Content Owners") may, from time to time, request Microsoft to provide security related updates to the Microsoft DRM components of the Product ("Security Updates") that may affect your ability to copy, display and/or play Secure Content through Microsoft software or third party applications that utilize Microsoft DRM."

I think that means "only at the request of the secure content owner" and "only microsoft software".

[ Parent ]

fair comment, but here's another interpretation (none / 0) (#110)
by martingale on Wed Jul 24, 2002 at 08:46:33 PM EST

Good to see yours isn't as draconian as the other one. I'm not convinced this is entirely positive for the user however.

For one, the secure content owner would be the original producer, typically a member of the RIAA/MPAA. So this can be interpreted as "whenever Sony feels like it" (- possibly replacing Sony by your favourite media company).

Another point is that finding programs which don't use the Microsoft DRM in future updated Windows OSes might be difficult. As I understand it, the technology is supposed to compartmentalize and encrypt sections of your computer's memory and hardware services, so for example to display an image on screen, your secure authenticated program decrypts the file into a secure area of memory, and the graphics subsystem will regulate display authorizations to ensure the contents of this memory area are displayed by the authorized program and the displayed image cannot be read by some unauthorized third party program.

In a scenario like the above, I'd expect that half the Windows system call API would contain DRM component code, which would in effect imply that any program compiled for Windows makes use of DRM code, whether they want to or not.

So under my scenario above I would expect that your "only Microsoft software" interpretation is too narrow. Instead, it would be "only software which makes Windows system calls related to memory/audio/display".

[ Parent ]

Freedom to choose your software. (4.00 / 1) (#34)
by Tezcatlipoca on Sat Jul 20, 2002 at 12:19:53 PM EST

With an open format you are absolutely sure you can access your music in the future.

With closed, propietary formats, your right to access your music is in the hands of others (Normally powerful companies).

You choose your poison.
---
"Every duck should aspire to be crispy and aromatic." sleepyhel

[ Parent ]

while is fine and good for .doc (none / 0) (#41)
by mikpos on Sat Jul 20, 2002 at 01:45:00 PM EST

But has no relevance for .mp3. There are already a fair number of free MP3 players. Unless the source code for all of them somehow disappears, you will always be able to listen to your MP3s.

[ Parent ]
Fear the worst. (none / 0) (#70)
by Tezcatlipoca on Sat Jul 20, 2002 at 10:20:23 PM EST

And prepare for it.

Why to risk be dependent on unsuported software in the future (that may be even illegal then) if one can promote the use of Open alternatives that benefit everybody?

I know that MP3 will not disappear anytime soon, but using Open formats is a matter of principle for me.
---
"Every duck should aspire to be crispy and aromatic." sleepyhel

[ Parent ]

Have you checked (none / 0) (#92)
by CaptainZapp on Mon Jul 22, 2002 at 05:57:48 AM EST

Your Microsoft license agreements recently?

[ Parent ]
have you assumed (none / 0) (#96)
by mikpos on Mon Jul 22, 2002 at 09:05:27 AM EST

That I have a Micrisoft licence agreement? Why would I use Microsoft software?

[ Parent ]
The big 1.0 (none / 0) (#60)
by swr on Sat Jul 20, 2002 at 05:11:41 PM EST

There are plenty of free MP3 encoders around, and there are plenty of free applications capable of playing MP3s. There are now plenty of MP3 CD players. If there's some sort of premium/licensing fee on MP3-related products (I believe there is?), I haven't seen that hurt MP3's proliferation.

There hasn't really been an alternative. Ogg Vorbis has been around, but just now hit 1.0. A lot of people have reservations about pre-1.0 things; often for good reason.

Not many hardware devices support .ogg files. Hopefully that will change now that Ogg Vorbis is "officially" released. Then you will be able to fit more tunes with better quality into one device.



[ Parent ]
Every few months... (2.31 / 22) (#10)
by Talez on Sat Jul 20, 2002 at 12:43:55 AM EST

Some GNU hippie comes out of the woodwork preaching the GPL goodness and the so called "patent free" status of Ogg Vorbis.

Ogg will never be 100% sure "patent free" because there might be patents on the some of the methods that ogg uses in its compression. While this might not be immediately apparent (the person who owns the patent might not realise it) and ogg has no specific patents on the Ogg Vorbis format, the fact that Ogg Vorbis can come out and declare itself "patent free" is laughable at best.

Secondly, MP3 is owned and there's nothing bad about that. I'm sure The Fraunhoffer(sp?) Institute spend a great deal of money in R&D to work out the voodoo magic that's performed during MP3 encoding. Should they be allowed to get money/glory/recognition for this fact without feeling guilty? Do you know how hard it was to make MP3? I'll tell you now it involved thousands of man hours of listening tests to get as good as it is. Get off their backs already.

MP3 is already mature. People have gigs of stuff encoded in MP3. It's a defacto standard and defacto standards are EXTREMELY hard to change. Witness the domination of Flash and PDF around the net.

Fourth, why bother? MP3 does the job. It's good enough. Ogg Vorbis achieves too little, too late. Until they can get up to MPEG-2 AAC standard type audio I really don't want to hear about it.

No prize for second I'm afraid. Idealism isn't a feature we care about. The world isn't going to convert to Ogg Vorbis because its "free". MP3 is free enough for all of us. Get over it already.


Si in Googlis non est, ergo non est

Not the angle I wanted (4.00 / 2) (#13)
by Idioteque on Sat Jul 20, 2002 at 12:53:35 AM EST

I am by no means a free software zealot. In fact, I am typing this on a Windows box, although using Mozilla. My firewall/mail server runs Linux, but I don't consider myself a big preacher/promoter or open source. I, as a musician, music lover, and audio software developer, like Ogg Vorbis, and just felt like telling others why I was excited about it, not forcing it down their throats...I hope I didn't come across that way.


I have seen too much; I haven't seen enough - Radiohead
[ Parent ]
I admire your spirit (1.50 / 2) (#17)
by Talez on Sat Jul 20, 2002 at 01:04:24 AM EST

But we get far too many people trying to force ogg down our throats when its obvious that nobody besides the hardcore really care.

Si in Googlis non est, ergo non est
[ Parent ]
See your point (3.00 / 1) (#19)
by Idioteque on Sat Jul 20, 2002 at 01:19:34 AM EST

I changed the last paragraph to make it less down your throat and more, just there for you to think about. Let me know what you think.


I have seen too much; I haven't seen enough - Radiohead
[ Parent ]
Forcing down your throat? (none / 0) (#111)
by phliar on Thu Jul 25, 2002 at 11:41:45 PM EST

Excuse me, I need to clear my throat so I can yell
BULLSHIT!
Did idioteque come over to your house, tie you down, set up his computer, and make you listen to Ogg encoded music? No; his/her article is nothing more or less than singing the praises of (call it evangelising or proselytising if you want) something that he/she thinks is cool in an area that at least 95 other people found interesting.

Not just that, but the article argues in rather persuasive terms why Ogg is better than MP3, and that its availability and popularity (the admitted ways in which MP3 is superior) will grow. A case is made that the 1.0 milestone is a sign that finally it's not just what you in fact say is the case: that Ogg and other Gnu utilities are just self-important chest-thumping rants from zealots, er, I mean "Gnu hippies." In short, it's a well written opinion piece (if you'd been paying attention, you'd have noticed that in fact it's in the Op-Ed section) that even has links to other sites that buttress its thesis; something the whole world needs more of, let alone K5.

In other words: grow up, get a life, whatever; but please don't post drivel where others can read it.


Faster, faster, until the thrill of...
[ Parent ]

In Talez's defense... (none / 0) (#114)
by Idioteque on Mon Jul 29, 2002 at 10:52:11 AM EST

Talez's comment was posted when the story was in the edit queue. There was a bit of a different tone to the piece, which I changed and I think made it a better piece. Although his/her point could have been made more politely, it did help me refine the article.

Thanks for the nice comment though. Glad you liked it.


I have seen too much; I haven't seen enough - Radiohead
[ Parent ]
I disagree. (5.00 / 4) (#14)
by Cluster on Sat Jul 20, 2002 at 12:54:08 AM EST

While it's true that Ogg Vorbis would most likely not be accepted if the public had to choose MP3 or Ogg Vorbis, the two formats can peacefully coexist.

There are already gadgets that offer Vorbis functionality -- iPaq and Zaurus.  Any hardware manufacturer has to pay Fraunhofer something like $20,000 for the first 15,000 devices, then something like $.75 for each one thereafter.  The same manufacturer does not have to pay ANYTHING for inclusion of Ogg Vorbis.

When most devices out there accept both MP3 and Ogg Vorbis, people can't help but compare the two -- and there is no doubt that OV exceeds (or at least meets, for the public that doesn't care much) MP3 quality while resulting in smaller filesizes.  As ignorant as the public can be, there's just no denying that given choice A: 100 "CD-quality" songs (MP3 160* Kb/s), or choice B: 200 "CD-quality" songs (OV -q 3*) on your Walkman/iPaq/PDA/whatever, anyone will choose B.

-----
* I do not plan on starting a war about exactly what quality/bitrate is "CD quality" -- that's purely subjective.

[ Parent ]

Good point... (2.00 / 3) (#15)
by Talez on Sat Jul 20, 2002 at 01:04:17 AM EST

They can coexist and I'm not bothered by people that use .ogg because I can open the format with Winamp. What I do have a problem with is GNU hippies that keep putting ogg in my face when it's obvious that nobody outside of highly techie circles care.

It's like religious people. So long as they don't bother me about <insert religion here> I don't care. If they start knocking on my door I tend to get nasty and do things like slam doors in faces.

As for royalties, $1 a device is hardly strangling for the functionality that MP3 provides. Besides, if Fraunhofer charged too much to use their tech everyone would drop MP3 like a hot rock anyway.

And yes, we don't need to go into a quality war since I know and acknowledge that ogg is better quality audio in general than MP3 and almost every other person on the planet will agree with that.

Si in Googlis non est, ergo non est
[ Parent ]

What? (4.20 / 5) (#20)
by carbon on Sat Jul 20, 2002 at 01:56:22 AM EST

This is a public forum site. If you don't like certain stories, vote Abstain and ignore them from then on. But nobody's forcing you to read anything, and nobody is shoving anything in your face.

In fact, the same could be said about the rest of the Web, which is effectively the only place you're likely to hear about Ogg Vorbis. No "GNU hippies" are knocking on your door; you're just overhearing them in the park, presumably as you head somewhere else.

One wonders why, if you're so completely unconcerned about the technical and political future of Ogg Vorbis, you've written several comments about your opinion of those very things.


Wasn't Dr. Claus the bad guy on Inspector Gadget? - dirvish
[ Parent ]
Licensing costs (4.50 / 4) (#48)
by tftp on Sat Jul 20, 2002 at 02:56:36 PM EST

As for royalties, $1 a device is hardly strangling for the functionality that MP3 provides.

You are forgetting that all gadgets are low-margin items. Manufacturers fight for pennies. A whole dollar wasted on a "right" is a visible loss. It would not be unreasonable to assume that a $100 player brings only $2-3 of profit to the OEM, the rest being spent on parts, assembly, QA, marketing, sales, customer support, administrative etc. If you look at it this way, the $1 licensing fee can easily be from 10% to 30% of your profits!

[ Parent ]

profit margin (3.00 / 1) (#69)
by ReverendX on Sat Jul 20, 2002 at 09:16:13 PM EST

2 or 3%? I'm sure they are making way more than that. Average profit margins are on the order of 15% I believe.

Being able to piss in an allyway is however, a very poor substitute for a warm bed and a hot cup of super-premium coffee. - homelessweek.com
[ Parent ]

$100 player - who gets the cash (4.00 / 2) (#72)
by bigbird on Sun Jul 21, 2002 at 12:24:24 AM EST

The store sells it for $100.

The distributor sold it to the store for $50.

The manufacturer sold it to the distributor for $30.

The manufacturer's 10-15% is thus $3-5, against which $1 licensing fees suddenly look a lot more significant. The actual numbers may differ (perhaps the manufacturer is getting $60, and the distributor $80), but most of the $100 goes into marketing and distribution.



[ Parent ]
All the same... (4.00 / 1) (#74)
by magney on Sun Jul 21, 2002 at 01:59:28 AM EST

As the other guy said, MP3 is a mature format, in wide use, and no media-player company in their right mind is going to drop MP3 support anytime soon, not unless Ogg Vorbis catches on in a mega-huge way.

Still, the fact that they won't have to pay that fee again to take on Ogg Vorbis support (beyond any relevant development costs) will make it easier for them to tolerate adding the feature.

Do I look like I speak for my employer?
[ Parent ]

you should care (4.00 / 2) (#55)
by klash on Sat Jul 20, 2002 at 04:00:46 PM EST

Besides, if Fraunhofer charged too much to use their tech everyone would drop MP3 like a hot rock anyway.

Just like the way everyone dropped GIF like a rock once Unisys started demanding royalties?

Oh wait, they couldn't, because nothing else had anywhere near the application support. So along came PNG.

Well this time creating of an alternative is preempting the showdown between the patent-holder and the masses. If you don't see the significance in that, then I invite you to throw a few cents in a jar for every gif that loads in your web browser, and monthly mail the jar to me.

[ Parent ]

Some companies already use Ogg (4.00 / 2) (#25)
by Echo5ive on Sat Jul 20, 2002 at 09:26:05 AM EST

Croteam used Ogg for the music in Serious Sam 2. I even managed to find the announcement:

We did a major change in the sound engine between FE and SE. And its name is Ogg Vorbis. Yeah, that's right, we're using ogg for music playing. In case someone hasn't heard of it yet, Ogg Vorbis (http://www.vorbis.com) is a patent-free, open source audio codec project. Or in english: a music compressor that plainly rocks. Make sure you check it out. We've tried encoding all the music for SE with Oggdrop at 64kbps and the quality was perfect even at such low bitrate. In the final version, since we won't need the extra space, we'll be shipping with 128kbps music tracks, for even higher fidelity. The guys there are really helpful and supportive and the whole project is surprisingly functional already. There are plugins for all major music players and other music programs.


--
Frozen Skies: mental masturbation.

[ Parent ]
Also... (3.50 / 2) (#79)
by Talez on Sun Jul 21, 2002 at 09:59:06 AM EST

UT2003 uses the .ogg format... I've noticed some little niggles but so far it sounds good.

Si in Googlis non est, ergo non est
[ Parent ]
No need to change (3.50 / 2) (#28)
by Nickus on Sat Jul 20, 2002 at 11:15:22 AM EST

Does it matter that there is terabytes of music out there already encoded in the mp3 format? For me it doesn't matter. When I encode my albums nowadays I use Ogg Vorbis and I keep my old mp3s. My player doesn't care which format it uses.



Due to budget cuts, light at end of tunnel will be out. --Unknown
[ Parent ]
Nope (2.50 / 4) (#33)
by Bob Dog on Sat Jul 20, 2002 at 11:52:21 AM EST

The krauts shouldn't be allowed to make money of patenting mathematics.  No one should.


[ Parent ]
Don't re-encode mp3 -> ogg! (4.72 / 18) (#22)
by Echo5ive on Sat Jul 20, 2002 at 09:11:41 AM EST

If so, it will sound like crap and you'll most likely think "Hey, this here ogg thingie stinks!"

In other words, re-rip your CDs directly to ogg -- converting mp3s to ogg is just plain stupid. It will sound bad.

You do have an original CD you can rip, right?



--
Frozen Skies: mental masturbation.

Portable ogg players? (4.00 / 2) (#24)
by Echo5ive on Sat Jul 20, 2002 at 09:21:23 AM EST

So, are there any portable players that support ogg yet?

Some googling resulted in the Nomad players -- they don't have ogg support yet, but the Jukebox model "supports standard MP3, WMA, and WAV formats and is geared for access to future audio formats, too. Upgradeable technology allows new features and enhancements to be downloaded to it when available."

Sounds promising, at least. Though with the current DRM craze I wouldn't be surprised if they refuse to support Ogg officially...



--
Frozen Skies: mental masturbation.

need FPU (4.00 / 1) (#29)
by mikpos on Sat Jul 20, 2002 at 11:40:38 AM EST

Most portable players do not have FPUs. Unfortunately, the Ogg Vorbis spec and reference implementation are written using floating-point operations. I believe you can get an integer-only implementation, but it costs a king's ransom and half an arm to get it from the Xiph people (this is how they make their money, I assume).

[ Parent ]
Yeah, right... (3.00 / 1) (#42)
by gordonjcp on Sat Jul 20, 2002 at 02:12:10 PM EST

... and you can pick up the integer MP3 reference decoder right there alongside the Daily Mirror, it's so cheap.

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


[ Parent ]
Yes, that's about right (4.00 / 1) (#63)
by damiam on Sat Jul 20, 2002 at 06:20:13 PM EST

Open-source integer MP3 decoder.

[ Parent ]
*REFERENCE* decoder? (3.50 / 2) (#65)
by gordonjcp on Sat Jul 20, 2002 at 06:40:09 PM EST

I mean, there's a pretty usable ARM Linux Ogg decoder out there, that's purely integer...

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


[ Parent ]
Pure integer ARM/Linux Ogg decoder (5.00 / 2) (#78)
by codemonkey_uk on Sun Jul 21, 2002 at 08:16:07 AM EST

Jan 09, 2002, Russell King wrote:
" I was [working on a fixed point decoder on ARM for .ogg], but I haven't had the time recently. It's not nice code, and currently there's no chance of merging it into Xiaph's CVS since they're in a release freeze." - arm.linux.org.uk mail archive

And that, it seems, was the last of that...
---
Thad
"The most savage controversies are those about matters as to which there is no good evidence either way." - Bertrand Russell
[ Parent ]

Um, no, that wasn't the last of that... (4.50 / 4) (#91)
by vmlinuz on Mon Jul 22, 2002 at 05:33:10 AM EST

nico, one of the other ARM-Linux guys, successfully produced a fixed-point decoder a few months ago. I 'ported' and packaged (ipkg for Familiar Linux on the iPAQ) the resulting libraries and ogg123 a few weeks later - and I'm working on an audio player based on famplayer now. I also know there is at least one other integerised version of the libraries under development, with the author hoping for a first release this week...

[ Parent ]
integer implementation probably on the way (4.00 / 2) (#53)
by klash on Sat Jul 20, 2002 at 03:47:58 PM EST

Yesterday I asked Rob Leslie (author of MAD and this kuro5hin article) if he was planning to write an integer vorbis implementation now that a vorbis spec has been released.

He said he probably would.

[ Parent ]

the imp-350 doesn't have it yet (4.00 / 2) (#43)
by nodsmasher on Sat Jul 20, 2002 at 02:37:08 PM EST

but its promised as a firmware update
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Most people don't realise just how funny cannibalism can actually be.
-Tatarigami
[ Parent ]
Zaurus (3.00 / 1) (#64)
by Sloppy on Sat Jul 20, 2002 at 06:36:48 PM EST

About the closest thing right now, is the Sharp Zaurus, which is really a general-purpose portable, not just a player. Better than nothing, but still not optimal.
"RSA, 2048, seeks sexy young entropic lover, for several clock cycles of prime passion..."
[ Parent ]
Idiotheque?! (2.50 / 2) (#27)
by europeanson on Sat Jul 20, 2002 at 10:16:06 AM EST

Didn't I just leech some of your work from vorbis.org ?
Due to the fact that I did, ogg/vorbis is doing it's job as a free, legal format
and due to the facts mentionened, you obviously know what you are writing about.
+1FP

-- my other signature is even funnier
Huh? (none / 0) (#31)
by Idioteque on Sat Jul 20, 2002 at 11:51:12 AM EST

I'm not sure what you are talking about.


I have seen too much; I haven't seen enough - Radiohead
[ Parent ]
um (none / 0) (#37)
by johann public on Sat Jul 20, 2002 at 12:52:09 PM EST

he downloaded a track by a group called Idioteque, in the ogg format, and he thought it was yours.
i'm guessing its not, of course, by your reaction...

[ Parent ]
Strange coincidence [nt] (1.00 / 1) (#40)
by europeanson on Sat Jul 20, 2002 at 01:05:43 PM EST


-- my other signature is even funnier
[ Parent ]
betamax vs VHS (3.00 / 8) (#32)
by jgk on Sat Jul 20, 2002 at 11:51:55 AM EST

I remember visiting an Ogg Vorbis site a while back and it said that Ogg would beat mp3s just as VHS beat Betamax because Ogg and VHS are superior to their rivals.

This was the wrong way round. Betamax was better than VHS and it died because it wasn't as popular. I love my now busted Betamax VCR but Ogg will go the same way because mp3 is MUCH MUCH more popular and being open source adds VERY little to Ogg's charm.

Gore Vidal is cool.
Gonna have to agree here ... (5.00 / 1) (#46)
by omegadan on Sat Jul 20, 2002 at 02:44:14 PM EST

I've developed with OGG and its really great (sfx/music engine for games).  However, I own *4* mp3 players (archos jukebox, samsung yepp, neo carjukebox, and an RFX9400) none of which play oggs :) Thus my personal music collection remains in mp3, and if I did have OGG players today, I dont know what could convince me to rip 250 cds again (such a painfull process).

Ogg took *WAY* too long to come to anything (2 or 3 years?), and their failure to get into *any* hardware devices has already sealed their fate.

On another note, OGG is widely used in videogames.

Religion is a gateway psychosis. - Dave Foley
[ Parent ]

New CDs (4.50 / 2) (#50)
by uhoreg on Sat Jul 20, 2002 at 03:03:10 PM EST

But if you had Ogg players today, would you rather rip new CDs into (higher quality/smaller size) Ogg files, or MP3 files? The difference between Betamax vs VHS and Ogg vs MP3 is that with the former, you would need two different devices, whereas with the latter, the same device could work for both, making the transition easier.

[ Parent ]
you've got it backwards (4.75 / 4) (#56)
by klash on Sat Jul 20, 2002 at 04:15:07 PM EST

Betamax is useless today since no one makes betamax players or movies. Any kind of investment into it was wasted, due to how things played out.

Today, I can tell you for a fact that Ogg Vorbis encoders and decoders will be available as free software in ten years. It's open-source and patent-free (and despite the boogie-man FUD, I believe Monty and friends have done their homework here).

What I can't tell you for a fact is that mp3 encoders and decoders will be around in ten years. It is extremely likely that they will, since Thompson now knows that the masses have an alternative they can switch to now that ogg is available, but they have the legal ability to shut down winamp, xmms, and US use of LAME tomorrow if they want.

This is why the safe bet is to go with vorbis, a technology in the hands of the people.

[ Parent ]

i'm not so optimistic (3.00 / 2) (#100)
by tps12 on Mon Jul 22, 2002 at 11:45:26 AM EST

This is why the safe bet is to go with vorbis, a technology in the hands of the people.
Based on my experience, I'd feel more confident about something that is in the hands of a business with a stake in its well-being than one in the hands of the fickle public.

[ Parent ]
what about vorbis requires optimism? (4.00 / 1) (#104)
by klash on Mon Jul 22, 2002 at 11:45:31 PM EST

Based on my experience

What sort of experience might this be?

I'd feel more confident about something that is in the hands of a business with a stake in its well-being than one in the hands of the fickle public.

In a business's view, well-being of a technology means they are able to extract a maximum amount of money from the public with a minimum amount of effort. This is why, especially in the case of a non-scarce resource like codecs (and software in general), the goals of a business are directly at odds with the goals of the public.

And what is there about trusting Vorbis that requires optimism? What kind of scenerio can the pessimist construct that can tarnish the usefulness of Vorbis? Even if xiph fell off the face of the planet today, vorbis is usable, free, and guarenteed to remain available because of its open nature.

[ Parent ]

stagnation (none / 0) (#106)
by tps12 on Tue Jul 23, 2002 at 07:10:48 AM EST

I was speaking in more general terms, but it may apply here. In essence, a business would try to develop Ogg and sign deals getting people to use it, improving it constantly. They do this because, yes, it makes them the most money with the minimum effort. A volunteer team will only take it as far as they have interest. Without the possibility of monetary reward, they are more likely to become bored and move on to something else.

You do have a point about what happens when a business fails or an open source team dissolves. In that case, clearly we are better off with their products in public possession.

And also, Ogg is not the best example for this argument.

[ Parent ]

beta vs. vhs (4.85 / 7) (#62)
by forii on Sat Jul 20, 2002 at 05:41:34 PM EST

Betamax was better than VHS and it died because it wasn't as popular.

Your comparison with beta vs. vhs isn't really correct. While Beta had better picture quality, it suffered from the fact that the tapes could not be as long of duration as VHS tapes, and more importantly, Beta was a proprietary format, controlled by Sony. A lot of manufacturers were not thrilled with the prospect of having to pay royalties to Sony, who many of them were direct competitors of, so instead they went with VHS, which was an open format.


Proud member of the ACLU, the NRA, and the EFF.
[ Parent ]

Beta isn't dead... (4.00 / 4) (#85)
by SvnLyrBrto on Sun Jul 21, 2002 at 05:17:01 PM EST

A minor nitpick, but I think it's worthy of note.

While the unwashed masses may have ignorantly settled for the inferior quality of VHS; people who know what they're doing, and rely on videotape PROFESSIONALLY, still use the superior format: Beta.

Visit any television studio.  They "film" and edit in Beta.  The cameras that the newsstations send in to the field with their reporters?  Beta.

This MAY be changing these days, as digital video cameras now offer superior quality to even Beta.  But that doesn't change the LONG history of people who know what they were doing choosing Beta over VHS.  And even WITH high quality DV cameras available, those professional-level Beta cameras were expensive, and the studios are NOT going to toss them before they wear out anyway.

In any event, amongst people who know better, Beta is alive and well; and probably WILL be for quite some time still.

cya,
john

Imagine all the people...
[ Parent ]

I did a simple test. (4.71 / 7) (#44)
by gordonjcp on Sat Jul 20, 2002 at 02:38:50 PM EST

I took a CD track that I expected would give the encoder a workout, and tried it with Lame, Blade and Ogg.  I deliberately tried to get the lowest bitrate I could.
I chose The Cardigans/Hey! Get Out of My Way, track 11 off Life (Stockholm Records 523 556-2).  This track has a lot of splashy cymbals and crisp hihats in the drums, bright vocals, and generally a lot of definition on the original CD.  I ripped the track using CDparanoia, and played back with both play and xmms, neither of which revealed any nasty noises.  Onto the encoders, then.  The wav file was 36M long!
First I encoded with oggenc on -1.  The resulting track was 1.2M and encoded in about 1:30.  Blade doesn't do VBR, so I set it to 56kbps - about the average that oggenc produced.  It came up with a file 1.4M long.  Finally, I tried lame's VBR, but it would only go down to an average of 112kbps - not really a fair fight.  So, I got that going at 56kbps as well.  It took about a minute, and produced a file 1.5M long.

On playback, the ogg file seemed to sit around 47-50kbps with occasional peaks to 60kps.  It sounded very clean - a little smearing of sharp transients (hihats got a bit buried in busy passages), but I've actually heard worse from 128k MP3s!
Blade was bloody awful.  It sounded like it was underwater.  Horrible.  Never liked blade much anyway, but this put me right off it.
Lame was, for a long time, my "standard" encoder.  In this test it gave pretty good results, but had nowhere near the top and clarity of the ogg file.  It was a lot duller, with quite severe artifacts on sharp transients.

Anyway, this was just a quick test I did.  It took me about 45 minutes, including getting and compiling all the sources, choosing a track, ripping, encoding and all the rest of it.  Speakers were a fairly reasonable pair of Panasonic Dolby Prologic speakers, amp was a massive old Sony hi-fi amp.  I used an SBLive! soundcard, with ALSA drivers (and OSS emulation).

You can try it yourself.

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


Blade (4.25 / 4) (#54)
by Betcour on Sat Jul 20, 2002 at 03:55:42 PM EST

Actually Bladeenc is specifically targetted at high bitrate. The original author admit that below 160 kbit, other MP3 encoders are much better because they use additionnal tricks to cut the bitrate down that would not be acceptable at high bitrate.

[ Parent ]
Classical test (5.00 / 6) (#58)
by thebrix on Sat Jul 20, 2002 at 04:21:52 PM EST

As a classical musician I tried two very different pieces with similar rips to WAV then encodings using the KDE 3 default settings (MP3 160KB constant bitrate with stereo; OGG with [min,average,max] bitrates [40,160,350]KB).

Messiaen: Turangalîla Symphony, fifth movement (enormous orchestra with nearly 140 musicians, fast music, huge range of colours)

J S Bach; Sarabande from Suite no.2 for solo cello (slow and melancholic music, an unaccompanied string instrument).

The difference, in both cases, is phenomenal. With the OGG I can hear individual instruments in the Messiaen and the rawness of the open string tone in the Bach: with MP3 the first frequently blurs out (to a general brouhaha) in the Messiaen and the second is not audible in the Bach.

Doubtless I could play around with bitrates, but I think these differences are intrinsic to the encoding.

[ Parent ]

I haven't tried orchestral music yet... (none / 0) (#61)
by gordonjcp on Sat Jul 20, 2002 at 05:26:56 PM EST

... but it's got a date with Berlioz - Harold in Italy and Saint-Saens - La Carnival des Animaux.

I tried recording some acoustic, electric and electronic instruments.  I tried a steel-strung acoustic (with integral piezo pickup - tried that DI'd too), an electric bass, a cello, a Yamaha DX21 (digital FM synth), Polysix (analogue polysynth - very distinctive analogue chorus effect) and a Transcendent 2000 monosynth.

Generally, the ogg codec gave a more open and honest sound than the mp3 codec.  You must remember, though, that perhaps the mp3 *player* is doing stuff to the sound too - some players seem brighter and less middly than others.

Guitar (miked and DI'd) and bass were clear, good definition and you could hear the pick on the strings even at 56k.  The MP3 codec wasn't bad but a little duller sounding.  The ogg codec had a little better definition on the lower bass strings, too.

The cello came out much as you've described above, despite only using a fairly old AKG D190E dynamic mike just above the bridge.

The synths didn't seem to show up much differences.  However, I rarely record them DI'd straight into the computer, preferring to run them through some outboard stuff first.  The DX21 was a bit too bright and glassy with ogg, but would be OK in a mix.  The Polysix's effects unit sounded pretty good compressed with ogg, whereas the subtle tone was lost a bit with mp3.  The Transcendent 2000 lost none of its woody tone through ogg, but seemed a little flatter through MP3 - the *very* sharp attack from its envelope section was smeared quite badly even at 128kbps.

Any thoughts on using it as a codec for a hard disk recorder?  I know that gear like the Roland VS recorders use some kind of compression.

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


[ Parent ]
Bitrate comparisons (4.00 / 1) (#66)
by smallstepforman on Sat Jul 20, 2002 at 07:14:47 PM EST

Your test is flawed in a number of ways, but the 2 biggest issues are that of Encoders and Bitrates.  There are competing MP3 encoders which each encode the original PCM file differently.  The BLADE encoder is different to LAME (or GUGU), which are different to the Frahencantspelloum.  The differences can be attributed to many design goals, one of which was target bitrate.  You used BLADE in your tests, but BLADE wasn't intended for low bit rates (56kbps), and any conclusion based on such a flawed test is meaningless.  Well, to be more precise, OGG preforms better at lower bitrates than BLADE, but seriously, who encodes music at 56kps?   A majority of my collection is 192kbps.

Also, the argument of FREE becomes moot when you realise that mose players/encoders/OS's (like BeOS) have licenced the codec, hence users can legally use the format.  Seeing how the JPEG issue is panning out these days, I'd say its safer to use MP3 over OGG, simply because we dont know who owns a patent OGG is unknowingly infringing on, and since we licenced MP3, let Frauhen... sort out the patent issues.

[ Parent ]

no, Vorbis was specifically designed for no patent (5.00 / 3) (#75)
by jbridge21 on Sun Jul 21, 2002 at 04:15:52 AM EST

Both a patent lawyer hired by Xiph.org AND the swarm of lawyers belonging to AOLTW looked over the Vorbis codec and gave it a clean bill of health in terms of patents. (AOLTW did this before they allowed it to be included in Winamp.) So the likelihood is very low that there are any patents to worry about, plus you've got the bulk of AOLTW somewhat on your side. On the other hand, 8hz and many other mp3 encoder projects have been shut down in the past due to patent enforcement.

[ Parent ]
You've missed my point a bit... (4.50 / 2) (#76)
by gordonjcp on Sun Jul 21, 2002 at 05:42:11 AM EST

So many people were asking "Why should I switch to Ogg?" or "Why should I try to convince the people who made my hardware MP3 player to switch?" without being given any sensible answers.  So, let me try to explain why I did this and wrote it up.

There didn't seem to be even any anecdotal evidence as to how much better or worse the Ogg codec was compared to MP3.  I selected a pathological case (hard to compress cleanly, very low bitrate, and material I'm familiar with - this is important in listening tests) and tried the two main MP3 encoders, along with the Ogg encoder.  In the final result, the Ogg codec produced better sound quality than most higher bitrate MP3s at its *lowest* quality setting.  It also produced the smallest file, although it took a bit longer to do it.

If you had a portable player that supported Ogg, you could get approximately four times as much music on, as you can with MP3, for more-or-less the same sound quality.  Now, are you going to buy the player that only does MP3 and gives you an hour of music, or the player that gives you Ogg as well and four hours?

Finally, I'm not bothered about licensing.  As another poster has commented, Xiph has had a large, clueful legal team check everything over carefully (presumably at great expense).  Xiph say that their legal team say that it infringes no patents.  That's as close as I'll ever get to finding out about it, so I'll have to take them at their word.

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


[ Parent ]
Don't forget "drop packets" == "dro (4.50 / 2) (#88)
by ramses0 on Mon Jul 22, 2002 at 01:01:19 AM EST

One of the neatest things that I haven't seen OGG do yet is that they designed it so you can "scale down" the bitrate of an ogg file without re-encoding the file.

With MP3, you keep the $huge .WAV source file and encode to 256kb, 128kb and 48kb for different bandwidth levels.  With the OGG, you ditch the original .WAV and only encode the 256kb version.  Supposedly you can then stream out "every other bit" and you'll automagically get a 128kb file.  No need to re-encode.  I'm not familiar with how this works, but I remember reading that it was a design goal of the codec.  All in all, I'm very glad that I've chosen OGG as my codec of choice for ripping my CD collection.

Big thanks to any xiph.org people reading here.

--Robert
[ rate all comments , for great ju
[
Parent ]

Hadn't heard of that... (none / 0) (#90)
by gordonjcp on Mon Jul 22, 2002 at 05:12:08 AM EST

But it sounds great for streaming media, especially live streams.

At the moment you've either got to run (say) three encoders, for 128kbps, 64kbps and 32kbps, or stream at 128kbps and then at the server play the stream back and recode at a lower bitrate.  This would make reducing the bitrate for lower speed streams a really simple job.

Must investigate this...

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


[ Parent ]
A bit unfair (none / 0) (#77)
by Cameleon on Sun Jul 21, 2002 at 08:01:20 AM EST

While I believe that Ogg is better than MP3 at the same bitrate, this test is a bit unfair. At the same bitrate, VBR will always be superior to CBR. So Ogg had a huge advantage over MP3 here. Try Ogg with a target bitrate instead of a quality setting, that would be a fairer test.

[ Parent ]
Fair comment... (none / 0) (#82)
by gordonjcp on Sun Jul 21, 2002 at 12:31:23 PM EST

but I did set the CBR stuff to 56k as opposed to Ogg's average of 48k.  I'll try it again.

The resulting Ogg file doesn't go much above 56k, and never goes above 60k.

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


[ Parent ]
Inaccuracy? (none / 0) (#93)
by Jim Dabell on Mon Jul 22, 2002 at 07:58:39 AM EST

Why did you get different size output files for the same source file, using the same bitrate, when encoding the mp3s with two different encoders? Doesn't that suggest that one of the encoders is bending the rules?

[ Parent ]
Probably inaccuracies, yes. (none / 0) (#97)
by gordonjcp on Mon Jul 22, 2002 at 09:16:17 AM EST

Although, bear in mind that the encoders do use very different algorithms, and the actual sound is different.

Apparently, Blade uses tweaks to make it sound better and work faster at high bitrates, but I always have found it to be slow and artifact-y.  Lame still seems to be the best MP3 encoder.

Give a man a fish, and he'll eat for a day. Teach a man to fish, and he'll bore you rigid with fishing stories for the rest of your life.


[ Parent ]
Most interesting (3.50 / 2) (#47)
by spacejack on Sat Jul 20, 2002 at 02:49:09 PM EST

to me is the ability to use the en/decoders all nice and legal-like in an application. For something like this, it doesn't really matter how much support there is out there in players or devices. I would expect to see more & more game/sound app. developers picking up on this soon.

Some game developers have already picked up on it (5.00 / 2) (#71)
by Majromax on Sat Jul 20, 2002 at 10:50:03 PM EST

Offhand, I know of at least one game that apparantly _does_ use Vorbis for sound -- America's Army

Propaganda arguments aside, looking over an installation I see "vorbis.dll", "vorbisfile.dll", and "ogg.dll" in the system directory. I could concievably account for one of the DLL's through filename conincidence, but definitely not all 3.

[ Parent ]

-1: (2.80 / 5) (#49)
by athagon on Sat Jul 20, 2002 at 03:00:08 PM EST

I would have preferred more technical reasons for your thought that Ogg > MP3, rather than the general I-Love-Music-And-I-Love-Ogg-Too feeling this article had. Just my 2c.

Ogg, nah. MP3, nah. APE, yes! (4.20 / 5) (#68)
by MmmmJoel on Sat Jul 20, 2002 at 08:31:28 PM EST

Instead of deciding between Ogg and MP3, many of us audio collectors compress our CDs into a lossless format. Then we do with the lossless files what we wish.

That allows us to convert, from source, to any lossy compression we want. MP3 for your portable MP3 player, Ogg for you buddies computer, WAV to burn CDs, XXX for whatever future format gains ground. Reencoding is easy, and preferably to losing quality the next time another compression algorithm.

HD space is cheap. Lossless compressors are getting better everyday. Play with Monkey's Audio and see how impressive this technology is. I prefer it to the Shorten (used mostly by live show trading), FLAC (used mostly for cross-platform compatibility, though Monkey's Audio just went open-source and a Linux port is available), and all the others.

Comparison (none / 0) (#73)
by rehan on Sun Jul 21, 2002 at 01:54:18 AM EST

Can you give a comparison of the 3 lossless formats that you mention? I've used FLAC and I didn't think much about it because I assumed all lossless compression is pretty much the same. Which one compresses more? Do any of them have any problems streaming? Are there any standouts in terms of CPU and memory usage?

Stay Frosty and Alert


[ Parent ]
There are several. (5.00 / 1) (#81)
by MmmmJoel on Sun Jul 21, 2002 at 12:05:57 PM EST

Monkey has his own comparison that is a bit dated, but still useful. There are a few other reviews out there, all of which often show the same results. Monkey's Audio compresses the best and does it very fast.

There are also Winamp and Media Jukebox (many APE enthusiasts player of choice) plugins that me listen to and tag files just as if they were MP3s.

[ Parent ]

Ogg Flac (none / 0) (#99)
by Inhibit on Mon Jul 22, 2002 at 10:33:21 AM EST

Why not use the Flac lossless audio format, which is also legally free from patents, unlike monkeyaudio from what I hear?
-- Inhibit, PCBurn Linux hardware/software reviewer
[ Parent ]
Monkey's Audio is patentless (none / 0) (#102)
by MmmmJoel on Mon Jul 22, 2002 at 03:21:40 PM EST

I'm not sure what gave you the idea that MA wasn't patentless, but I severely doubt it infringes on any existing patents. I use MA over FLAC because of the improved compression, better speed, and the smooth integration with Media Jukebox.

[ Parent ]
Well (none / 0) (#103)
by EriKZ on Mon Jul 22, 2002 at 11:29:09 PM EST


I checked out MA FAQ, and it doesn't have any questions.

I'm surprised no one has asked "What kind of compression can one expect from MA?"

[ Parent ]

Patents (1.50 / 2) (#84)
by chbm on Sun Jul 21, 2002 at 03:03:09 PM EST

For the most part, MP3 is owned. We must pay royalties to its patent owners. No one owns Ogg Vorbis
Go read some of the patents on your link, then read some Ogg docs/code.

Anyway, this was all discussed a couple of months ago.

-- if you don't agree reply don't moderate --

Alright, which ones (none / 0) (#86)
by Idioteque on Sun Jul 21, 2002 at 06:22:11 PM EST

Tell me which patents, with links, you believe the Ogg Vorbis code is violating and why. Just saying that you've read them without anything to support your answer doesn't fly with me.


I have seen too much; I haven't seen enough - Radiohead
[ Parent ]
Ogg for broadcasting radio? (none / 0) (#94)
by Echo5ive on Mon Jul 22, 2002 at 08:46:56 AM EST

Ogg encoded with "-q 1" rocks the socks off of everything else I've tried. If used for radio, this means higher quality using lower bandwidth.



--
Frozen Skies: mental masturbation.

Clarification (none / 0) (#95)
by Echo5ive on Mon Jul 22, 2002 at 08:47:48 AM EST

...using the same bitrate, of course.



--
Frozen Skies: mental masturbation.

[ Parent ]
Ogg is owned (4.00 / 3) (#98)
by enry on Mon Jul 22, 2002 at 09:29:05 AM EST

No one owns Ogg Vorbis,

Just because something is under a GPL or BSD license doesn't remove ownership. Xiph.org owns Ogg and can do with it as they like (for example, relicense it from GPL to BSD, as they did a few months ago).

Software vs. Standard (5.00 / 3) (#101)
by thedward on Mon Jul 22, 2002 at 11:58:40 AM EST

They own the software, not the standard. The standard is open, and anyone is free to reimplement it. In order to reimplement the mp3 standard legally, you must pay royalties.

[ Parent ]
Re-licensing doesn't take away your rights (none / 0) (#112)
by phliar on Fri Jul 26, 2002 at 12:05:17 AM EST

Xiph.org owns Ogg and can do with it as they like (for example, relicense it from GPL to BSD, as they did a few months ago).
Once something is released under a license -- say GPL -- the only way you can lose the terms granted under that license is if you violate some of its terms. For instance, if Person A uses GPL'ed code in a proprietary project that is distributed, but does not also offer source code, then the person loses the right to use GPL'ed code.

When Sun bought Star Office and released the source, but later decided to go back to a closed-source version, what happened? OpenOffice. Features Sun adds in the future to Star Office will not be free, but all the things Star Office had in the time that it was free, those will always be available to everyone.

Nitpick: Xiph only put the codec libraries under a BSD like license; the other tools are still under GPL.


Faster, faster, until the thrill of...
[ Parent ]

To late... To late... (2.00 / 2) (#105)
by parasite on Tue Jul 23, 2002 at 04:33:45 AM EST


I dunno about you'all, but I see little reason to re-encode my entire CD collection of 250+ CD's when in only ~2 more years I'll be able to afford a harddrive that will contain them all in raw and pure PCM or WAV format -- and still huffing away on my beloved aged dual Celeron-333 Linux box, I'll be able to enjoy the perfect quality of the original recordings through my SB Live digital-optical daughterboard connection to my stereo.

You're right... (none / 0) (#115)
by Idioteque on Mon Jul 29, 2002 at 10:56:16 AM EST

There is no reason to re-encode, but why not start using it with the new CDs you get, and maybe a couple of favorites you listen to a lot. I used MP3 for a long time too. Most of your favorite software players support both formats, so it no big deal.


I have seen too much; I haven't seen enough - Radiohead
[ Parent ]
Another reason Ogg Vorbis is better (4.00 / 1) (#113)
by phliar on Fri Jul 26, 2002 at 05:49:37 PM EST

There is another reason Ogg Vorbis is better than MP3. Here is a section from the LAME technical FAQ:
ISO based encoders (BladeEnc, 8hz-mp3, etc) use a MDCT/filterbank routine similar to the one used in decoding, and thus also introduce their own 528 sample delay. A .wav file encoded & decoded will have a 1056 sample delay (1056 samples will be appended to the beginning).

The FhG encoder (at highest quality) introduces a 1160 sample delay, for a total encoding/decoding delay of 1688 samples. I haven't tested Xing.

LAME 3.55 has reduced the delay to 48 samples. While technically there is no reason the delay couldn't be reduced further, there are problems with very short delays: as the page says,
there are a few problems with using such a short delay:
  1. The 96 samples of the first frame are attenuated by the MDCT window. If the encoder delay is greater than 96, this window will have no effect since the first 96 samples are all padding. With a 48 sample encoder delay, the first 48 samples will be improperly attenuated. (.001 seconds worth of data at 44.1kHz).
  2. Filterbank contamination problems. This it rather technical, see below.
  3. In LAME, psycho-acoustics for the first 576 granule are not correct. This could be fixed, but at the expense of adding more buffering and code complexity.
This is not a problem if you're just listening to songs etc.; however if you're looping or splicing, this is an issue. Similarly, there will be problems when splicing, since data for frame N is not only stored in that frame but can be distributed over a region; this can be mitigated with VBR and using certain options. (See that page for detauls.) Ogg Vorbis does not suffer these boundary problems; see this page on Acidfanatic for electronic musicians' points of view.


Faster, faster, until the thrill of...

Why Ogg Vorbis? | 115 comments (103 topical, 12 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!