Slashdot Mirror

Mpeg Audio Layer-3 coding was developed by the Fraunhofer IIS institution in the late 80s, and therefore the patent is owned by them.

-Tommy

------
"I do not think much of a man who is not wiser today than he was yesterday."

Mpeg Audio Layer-3 coding was developed by the Fraunhofer IIS institution in the late 80s, and therefore the patent is owned by them.

-Tommy

------
"I do not think much of a man who is not wiser today than he was yesterday."

2*H2 + O2 = 2*H20 -> that means on a PIII laptop, you've got a river running out of your laptop. Perhaps we could use it to cool the processor? Would sound very cool: "Hydrogen-Powered laptop with watercooled PIII cpu"

Some URLs on this topic:

http://www.heise.de/newsticker/d ata/jr-06.03.98-000/ (Sorry, only german)
http://www.ise.fhg.de/Products/brennst offzelle.html
http://www.ise.fhg.de/Projects /energy99/art1.html
Yes, it seems that the Fraunhofer people already have a prototype of this.

2*H2 + O2 = 2*H20 -> that means on a PIII laptop, you've got a river running out of your laptop. Perhaps we could use it to cool the processor? Would sound very cool: "Hydrogen-Powered laptop with watercooled PIII cpu"

Some URLs on this topic:

http://www.heise.de/newsticker/d ata/jr-06.03.98-000/ (Sorry, only german)
http://www.ise.fhg.de/Products/brennst offzelle.html
http://www.ise.fhg.de/Projects /energy99/art1.html
Yes, it seems that the Fraunhofer people already have a prototype of this.

but MPEG, while being "open" isn't really free...check this out they won't even tell you what the licensing terms are...Though I suggest reading the Office Faq, There /is/ a need for a FREE standard..as opposed to a "OPEN" one...

You're forgetting that the Pro version will (hopefully) have a java vm, and it will have a more featurefull BeIDE that what's in the devtools zip file, but I tend to use Eddie for programming anyway, I still would
rather use make commands and makefiles. And you get support for it too. And hopefully Be makes Gobe keep non-point releases free for registered users just like it has in the past.

Here is the link to Gobe's website, now that they are distributing BeOS they have the info there.

But the mp3 codec is straight from THE source, Fraunhofer IIS and THOMPSON Multimedia, which means it's way better than lame and crew. And there is no need to decode a cd to a wav anymore, the cdda fs addon lets you look at your audio cd like it's full of wavs, drag and drop and boom you're done, just encode away.

To the other person who said they wanted a plugin to broadcast icecast streams.. uhh that's pretty ridiculous.

Having recently set up an icecast server, I think I know what that other guy is talking about. I think he was looking for some way to use XMMS as a more "user friendly" front end to produce an MP3 audio stream for icecast.

As it is right now, actually producing the audio stream for icecast to serve out to the world is something of a black art (IMO).


Your 3 choices are:

First option - shout: Easiest to get running, but less flexible. It only allows you to stream your MP3s at the same bitrate they were encoded at. In other words, no real-time re-encoding.

If you want to stream the same audio at several bitrates (to give people with slower connections an option of lower quality but less skipping), you need to be able to re-encode the MP3 data on the fly. You also need re-time re-encoding if your source MP3s were encoded at different bitrates and you want to give one consistent bitrate for your stream. (this would be a good idea, since some players get out of sync if you change the bitrate within the stream)


Second option - liveice: Much more flexible but a bear to set up. It allows you to re-encode real time, and manipulate the playlist, mix tracks together, and do more of the things you hear on over the air radio.

The complexity comes from the "modular" nature of the software. There are license restrictions on the MP3 encoding algorithm, so there can't legally be any GPL encoders. This means that Scott Manley can't include MP3 encoding code directly in his program. He has to rely on the encoding coming from an external source, such as a commercial encoding package like Xing or from one of the "underground" packages like LAME that are simple wrappers around the unoptimized MP3 reference encoder.

Since most people doing icecast servers want to keep things open source, they aren't very well going to pay hundreds of dollars for a commercial encoder. That leaves the "underground" packages.

There are several packages available for Linux to encode MP3, but the stability, audio quality, and program interface varies widely among them. Scott usually has a tough task getting his code and somebody else's to play nice.

I had such a difficult and unreliable experience trying to get an encoder to run, produce decent sound, and not crash with liveice that I eventually abandoned it and used shout. I manually re-encoded my MP3s at 24kbps and I just stream them statically. They say doing that usually produces better audio quality anyway.

BTW, I'm fully prepared to concede my inablility to get liveice running stems from my own incompetence and NOT from flaws in Linux or mistakes of the icecast developers. YMMV...


Third option - Winamp with DSP Plugin: Believe it or not, there are people who keep a Windoze machine around just to produce a stream to the icecast server.

For all it's shortcomings, M$ has done one nice thing in its life and that is pay for a distribution licence for the Fraunhofer MP3 codec. Fraunhofer is acknowledged to be the most efficient, stable, and best sounding MP3 codec around. M$ included it in the Windows Media Player server tools/authoring package which you can download for free.

What you do is, you get that MS package, winamp, and a Shoutcast DSP plugin that re-directs the MP3 output to an IP address/port you specify, and configure the bitrate and audio quality all from a nice user friendly dialog box. Then you load up your winamp playlists and you are "on the air", with less hassle than liveice. (unless you consider rebooting a crashing Doze box every few hours or days a hassle...)


So what I think the other guy was looking for is some way to make XMMS give similar functionality and user-friendliness to what winamp provides under the windows platform, except provide it with the stability and performace of linux, and license it under the GPL.

I think the functionality can certainly be written by somebody, we have a lot of real smart developers in the open source world and XMMS/icecast in particular. The tough nut is going to be a high quality, free MP3 encoder.

The best option would be a freely distributable version of the Fraunhofer encoder for Linux. This could actaully be done, the only barrier is money. It costs MEGABUCKS (several thousands) for a distribution license from Fraunhofer, and the result would not be possible to put under GPL since the algorithm iself has license and NDA restrictions. So the best we could do is get a binary only encoder that would only run on whatever distribution and platform it was compiled for, and we'd have to find a charitible sugar daddy to pay for the license. And we would still have to hold our noses and accept a non-GPL solution.


BTW, if anybody cares, my icecast server is online at http://24.5.234.110:8000. I have several hours worth of Dave Matthews Band live show bootlegs up there. If you're a DMB fan, giv it a listen sometime. [/me hunkers down for the /. effect...]


-James

Ok, i'll never do that again, i promise. My conversion to geekdom is complete now, i suppose. And as retribution for being a "first post loser", i'd like to contribute this comment, completely on-topic, and relavent. :)


Nuclear power is an amazing tool. Sadly, it is a tool that we simply do not have the technological prowess to to control. Even sadder, this hasn't stopped us.

There is no way to assure safety when dealing with nuclear power, or nuclear concepts as a whole. We're still a long way from "cold fusion", and even then the danger will never be eliminated. We're dealing with the very basic building blocks of reality as we know it - it cannot possibly be safe.

Many countries turn to nuclear power a (relatively) cheap and efficient way to power their land. Many do not have a choice. For countries who do not yet have nuclear power, it has become a sort of "watermark" for the economic evolution of their country. This *has to stop*.

Nuclear power is *not* safe, and it never will be. There are lots of options - the most popular and well-known include Solar and Wind power units. Organizations like the EPSEA and Fraunhoer ISE have been pushing and making immense progress in the solar power field - proving it as a viable source of power. Wind power has been used since the dawn of civilization, and is so prevalent and useful now that companies like Wintec are popping up everywhere.

It's time we realize that nuclear power isn't a safe option. Not for us, not for our children, not for earth. It's that simple.


.------------ - - -
| big bad mr. frosty
`------------ - - -

hehe nope :)
from mp3 faq

Q: Talking about MPEG audio, I always hear "Layer 1, 2 and 3". What does it mean?
A: MPEG describes the compression of audio signals using high performance perceptual coding schemes. It specifies a family of three audio coding schemes, simply called Layer-1, Layer-2, and Layer-3. From Layer-1 to Layer-3, encoder complexity and performance (sound quality per bitrate) are increasing. The three codecs are compatible in a hierarchical way, i.e. a Layer-N decoder may be able to decode bitstream data encoded in Layer-N and all Layers below N (e.g., a Layer-3 decoder may accept Layer-1,-2,-3, whereas a Layer-2 decoder may accept only Layer-1 and -2.)



Basically they are all audio formats (the video "layers" arn't called "layers") and the higher layers are most complex than the lower layers. Layer-1 is really not that much far from .wav just a slight bit of compression.

FIPS has supported FAT32 for over a year. Look at the FIPS home page for the details.

For MPEG-4, the licensing authority details are not yet set, but there is general intent to provide a similar one-stop mechanism. I would expect the legal machinery to be in place within a year. There are a lot more patents that apply to MPEG-4 than to MP3, maybe as many as 25 or more (but see below).

Purely in my own opinion, I think that ISO is not nearly strict enough about managing patent claims on the MPEG standards. The basic argument, which is a good one, is that all of the relevant technology is held by patents, and good standard codecs could not be made if patented technology was not admitted. But IMHO the patent-holders are today allowed to abuse their privileges under this system.

Of course there is a point to standards that are protected by patents -- many companies make a lot of money from them in spite of the licenses they have to pay (MPEG-2 video decoders, for example). But these patents also have the effect of making it difficult for hobbyists to develop conforming freeware and open-source products. This may, in fact, be the goal, but it is wrong IMHO.

Not all of MPEG-4 audio technology is protected by patent. The audio synthesis codec that I developed at MIT and contributed to MPEG has been released into the public domain at my request.

-- Eric Scheirer
Editor, ISO 14496-3 (MPEG-4 Audio standard)

http://www.iis.fhg.de/amm/legal/index.html

Now I hope you appreciate what Apple had to invest in order to incorporate MP3 technology into Quicktime.

Fraunhofer discusses that they have redused the required number of DSP from 5 down to one, and that you can get their system for broadcasting MP3 streaming content. The problem is in the Real-time encoding. Fraunhofer is being very protective of their intellectual property, and don't want anyone else to create high quality, low bit rate audio without giving them money. A real-time encoder is very expensive. From a bussiness perspective, this makes perfect sense. As anyone who has encoded their own MP3's knows, encoding CD quality audio is time consuming, and even a DSP hardware implementation is usually two slow for a real-time application.


In the mean time, HW decoding soluions are just now becoming a viable option. Several companies are now creating IC's with MPEG 1-Layer 3 (and Layer 2.5) decoding, but these are recent developments. Prior to the release of these chips, only DSP implementations were posible. MP3 decoding is avalible from some sources in the from of a PCI card, my roomate is also trying a different approach using these chips for a school project. Decoding is a reality, but encoding is a few more years down the road for the average user. This means that a real-time MP3 CODAC is not a viable concept, yet.


Baggio
Time flies like an arrow;