Slashdot Mirror

.. AU audio files, MOV video files, and AVI files using a legacy video compressor Microsoft doesn't even ship any more?

I'm sure the research is great. I'll pass until you make it presentable.

Try http://www.drdivx.com/ and http://www.mp3dev.org/ .

GNU / GPL (Copyleft) The spyware that Sony installs on the computers of music fans does not even seem to be correct in terms of copyright law.

It turns out that the rootkit contains pieces of code that are identical to LAME, an open source mp3-encoder, and thereby breach the license.

Sony complied with non of these demands, but delivered just an executable program. A computerexpert, whose name is known by the redaction, discovered that the cd "Get Right With The Man" by "Van Zant" contains strings from the library version.c of Lame. This can be conluded from the string: "http://www.mp3dev.org/", "0.90", "LAME3.95", "3.95", "3.95 ".

But the expert has more proof. For example, the executable program go.exe contains a so called array largetbl. This is a part used in the module tables.c of libmp3lame.

This discovery can have far-stretching consequences for the music giant, who claims only to protect copyrights. Previously, judges in Germany already forced various companies to release source code to the public and to deliver the goods necessary for compiling. It is also possible to demand financial compensation for damages.

Meanwhile, Other details are also becoming clear. The Electronic Frontier Foundation complains that the spyware makes the legal listening to the music on iPods impossble. The organisation is busy making a list of cds containing the hidden software and publishes this on her website.

Various calls to SonyBMG remained unanswered despite promises to call back.

"Nonetheless, the spokesmen of Sony BMG Nederland says that Xcp will be introduced in Europe, and therefore also in the Netherlands, in 2006."

The article further contains no less then 17 dubious features of the Sony software, basically the same ones as circulated the news lately. However, number 15 is interesting. Blankesteijn claims he received spam after entering his e-mail adres in the request form for the Xcp removal software:

"15. Not only is this [filling in a form] in-necessary complicated and time consuming, Beet (the magazine) immediately received spam from Sony BMG containing an advertisement for their multimedia software. It turns out somewhere along the road you could have clicked on a link to Sony MBG's privacy policy. There it says your mail address can be added to marketing lists. But is very well possible that the user will not notice this link. In any way, nowhere is asked for permission to do this, which will make this way of handling illegal in many countries."

Another interesting Dutch headline involves the discovery that the rootkit contains parts of LAME, a LGPLed mp3 encoder. A translation can be found here.

Can I use LAME in my commercial program?

*** IMPORTANT NOTE ***
The decoding functions provided in LAME use the mpglib decoding engine which is under the GPL. They may not be used by any program not released under the GPL unless you obtain such permission from the MPG123 project (www.mpg123.de).

It is unclear what LAME is used for in the kit, but according to it's about page, Lame can also be used for decoding. In that sense, this usenet post by a first4internet employee shows that the company producing the rootkit at least has an MP3 player, which of course might be part of the Sony rootkit:

"I am currently writing an MP3 player with lots of bells and whistles including a wave editor, fades, reverbs etc. What I now need is to be able to protect the files it creates. I have already written the routine to convert the MP3 into a WMA file. Does someone have some simple C++ code which can write Microsofts DRM v1 properties that the user whishes to set (i.e. 3plays 4 copies etc) over the unprotected file to make it protected. There may be some cash on offer here if its easy to use! All I need is a procedure that performs this."

And if you're thinking you're safe for these kind of kits because you're on a mac, well, it might not be for long. Of course this is pure speculation, but at least one F4I employee is asking questions about mac programming issues...

According to this article (Dutch) on the CD Get Right With The Man of Van Zant there are strings from the library version.c of Lame. The following strings are found: "http://www.mp3dev.org/", "0.90", "LAME3.95", "3.95", "3.95 ".

Also in the program go.exe their is an array called "largetbl", which is part of tables.c of libmp3lame. Can anyone confirm these findings?

LAME is licenced under the LGPL. Could this mean more trouble for Sony because of a license violation?

90% of the mp3's traded on kazaa and the others are at a lower quality than any of my 3rd generation casettes from the early 80's and are of junk quality already.

Yeah, most of the music on P2P is rubbish quality. And when people try to rip good MP3s they still often get it wrong in some way. Maybe they used LAME, but ripped with some awful piece of junkware, that just goes for speed with no error checking.

So I was chuffed when I found Übernet, a private Direct Connect network. The music traded is based around the uberstandard, basically CDs ripped with Exact Audio Copy and encoded with LAME set to APS, APX or API (though API is insane for a reason...). We also trade OGG and FLAC.

The ironic thing is, I have bought more CDs since joining Ubernet than I had in a good few years. P2P killing sales?

For example, their compendium of software includes:

A workaround for EU volume limitation
Ripping, encoding and tagging recommendations.
A utility to mass export Outlook contacts
News and Weather syndication downloaders.
By far the best way to retrieve your MP3s (a utility that sits on your iPod itself and is executable over a network!)
The fantastic iPod Agent, which creates beautiful XML music lists as well as performing loads of useful functions

Every other area of the lounge is equally as exhaustive - from iTunes configuration (you can do amazing things with smart playlists!) to headphones and case reviews. Visit the site instead of buying a book.

(Oh, and I'm in no way involved with the Lounge other than being a fan.)

nothing the lame project cant fix or otherwise bypass.

Grip pretty much sucks in terms of speed. I dunno why since it's just a GUI wrapper over standard CLI tools, but it does.

Try something better. The combination of cdparanoia and lame is the classic. From cdparanoia you get good speed, error correction, digital ripping, and from lame you get quality encoding.

If you want a graphical ripper, I've discovered a little gem called oggre (not to be mistaken for the other OGGre which is an OGG reader written in Java.

The oggre I'm talking about is an XMMS output plugin which writes OGG to files:

1. Set your CD Audio input plugin to digital mode (analog won't work with oggre).
2. Load the playlist with tracks from the CD.
3. Select all tracks and use "get advanced info" (or whatever it's called) to load info from CDDB.
4. Set oggre as the output plugin, with quality 4.99 (recommended best quality/size ratio, read the oggre README to see why).
5. Make sure the play mode is not random or loop.
6. Press play, sit back and watch ripping work at a very nice speed.

About the only thing wrong with oggre is the fact that it uses a fixed output dir, so if you want dirs named after the CD you have to make them by hand. The files, however, do use song attributes in their names.

Oh, and the guy who wrote oggre also wrote out_lame, but that one has fewer features. I don't mind choosing OGG over MP3 so I don't care.

A friend of mine was working on the Music Analysis Toolkit (MAT) while he was still thinking of pursuing his Ph.D. in Computer Science. The toolkit does just this kind of music recognition stuff, and he was working a lot with the psychoacoustic model and using LAME to filter out inaudible stuff. It's been a while since I've talked with him about it, but it's definately interesting.

I'd be willing to bet this is based on lots of the same stuff.

Look for fatboy.wav on the LAME test samples page.

Unlike the situation under DMCA, which attempts to block reverse engineering on copyright grounds, and may be subject to overturn on the basis of fair use precedents, reverse engineering of patented techniques has always been illegal and the case law all supports that.

Actually, reverse engineering of patented techniques has always been not only legal but pretty much pointless, since the whole point of the patent law is to have the detailed explaination publicly available for people to study so everyone could be able to implement the idea as son as the patent expires.

I can't find the links, but IIRC, at least one open source program for converting between different media formats, has withdrawn support for WMA because MS threatened them with a patent infringement lawsuit. The only real defense against an existing patent is to invalidate it in court, which can be a VERY expensive undertaking.

The same is the case with MP3. See mp3/mp3PRO Patent and Software Licensing Information. From the developer FAQ:

I want to support mp3/mp3PRO in my products. Do I need a license?

Yes. ['nuff said...] As for practically any important technology (and particularly for publicly established standards), you should know that patent rights for mp3 exist. Both Fraunhofer IIS-A and Thomson have done important work to develop mp3 audio compression (before and after it became part of the ISO/IEC MPEG standards). This work has resulted in many inventions and several patents, covering the mp3 standard. Although others may also hold patents, Fraunhofer IIS-A and Thomson have an important portfolio of patents related to mp3. [...]

That's why LAME Ain't an MP3 Encoder but a GPLed patch against the dist10 ISO demonstration source, otherwise it would be just as illegal as a GPLed WMA encoder. In other words, MP3 is as evil as WMA. That's why artists should be using Ogg Vorbis.

When does it show up for Linux? xxxBSD?

Other than Wintel (where Wine could possibly run iTunes) and Mac (which comes bundled with Mac OS X and iTunes), what desktop computer hardware platform promoted for use in homes is popular?

Palm? WinCE? Symbian?

Are those devices able to decode popular lossy audio compression formats with acceptable battery performance?

With DRM, it becomes even worse, as you can't get the content from a sanctioned device

Remember that nothing but a little generation loss and possibly the MP3 patents are keeping you from burning your recordings purchased from iTunes Music Store to CD-RW in Red Book format, ripping them with digital audio extraction software, and re-encoding them with the best MP3 encoder technology available. If you're encoding for a pocket sized MP3 player that you're planning on carrying into a public place, you don't even need to encode the distraction of stereo separation. Mono MP3 actually sounds pretty good at 96 kbps.

But, you see, OSS for Windows is catching on! Some of the most popular programs on Sourceforge are win32. Everything that you need for spyware-free commercial-free RIAA-free music and video sharing is available there, on sf.net. Here's a sampling:

CDEX - a great MP3 ripper. Use with LAME for great, free rips.

eMule and DC++ - very popular P2P clients

BitTorrent - For large file sharing (movies, etc)

VirtualDub - for video format conversion (DiVX, VCD, etc)

Audacity - multi-track audio editor

I could go on and on. Look at this list and all the win32 apps there.

I have only used bladeenc. Is there much of a difference?

Well I wouldn't choose this one.
Let's say a good ExactAudioCopy v0.9 extractor (reads up to 80 times the same part to avoid errors) coupled with a nice LAME 3.92 win32 DLL would do the job !
LAME is among the best MP3 compressor so far.
They are both freeware and don't need install

C o A X

Do also point out that LAME, the MP3 encoder that CDex includes, is not only open source but is also the best MP3 encoder there is, period.

Use Exact Audio Copy with LAME and the --r3mix command line, and never worry about quality again.

Perhaps it would be possible to redo (for example) mpg321 in a similar way?

What about LAME? They've managed to create an MP3 encoder without, from what I understand, infringing upon FhG's patents. Couldn't they create an encoder to do the same? I'm currently in the middle of reading an FAQ off their site about the MPEG patents. It is a good read: http://web.media.mit.edu/~eds/mpeg-patents-faq.

LAME actually is an MP3 encoder now.

"Following the great history of GNU naming, LAME originally stood for LAME Ain't an Mp3 Encoder. LAME started life as a GPL'd patch against the dist10 ISO demonstration source, and thus was incapable of producing an mp3 stream or even being compiled by itself. But in May 2000, the last remnants of the ISO source code were replaced, and now LAME is the source code for a fully LGPL'd MP3 encoder, with speed and quality to rival all commercial competitors."

http://www.mp3dev.org/mp3/ - the LAME project.

Even better is EAC, or Exact Audio Copy. Alongside LAME, OGG Vorbis, or, for you lossless zealots, Monkey's Audio, you've got the most accurate copy of the CD that you can get.

Of course, if it's N'Sync, it won't really make a difference -- it's crap no matter which CD ripper you use.

I guess this could be done using MPlayer and LAME.. Mplayer supports mms:// urls and about all codecs there is.. The proprietary extensions to rtsp:// (streaming realaudio/video) is also about to be supported.. It also supports writing audio to raw pcm or wav, which can be encoded to mp3 using LAME.

As far as try-before-you-buy goes, how come no one ever complains that we don't get to preview entire movies before we decide to shell out for the tickets?

A movie is one coherent audiovisual work that tells a single story. Reviews dissecting every part of a movie are available in almost every imaginable news medium. On the other hand, a typical album is a recording comprising twelve musical works, unrelated except for having been recorded (and possibly written) by a single team of performers called a "band". (Themed albums are the exception to this rule, but they are also the exception in the pop marketplace.) It's hard to judge whether or not critics like a whole album because 1. the music reviews don't get as much publicity as the movie reviews, and 2. music listeners have much more diverse tastes than movie viewers.

It's about time media companies realized that if they want customers' money, they must work with their customers, not against them. To let listeners preview a whole album, I'd suggest that the label publicly release an MP3 file containing a representative 20 second snippet of each song for free promotional redistribution.

In addition, if the RIAA labels put up a site where I could download high-quality singles (MP3 encoded with LAME 3.92, preset r3mix) for $1.00 each, and the site showed exactly how much of my buck went to the songwriters and performers, I would sign up in a heartbeat. The most popular legit major-label MP3 site (eMusic, $15 per month for unmetered downloads) offers only 128 kbps MP3, and 128 kbps MP3 sounds like crap on my speakers because it throws away so much information.

Listening to that on one of those teenie Mac speakers, of COURSE it's not going to sound any different. ;) While I can't entirely vouch for the quality of AAC, from what I've heard, it's really fitting to be alongside MPEG-4 video -- MPEG-4 video was designed for Internet viewing, and IIRC, AAC was designed for the same purpose.

Really, it comes down to this:

In the beginning, there was MPEG-1 and MPEG-1 Layer 1,2,3 audio. Big deal, koz they were all more-or-less firsts in semi-quality video compression.
Then came (in no particular order) MPEG-2 and MPEG-4. MPEG-2 was a format designed for DVDs and other high-resolution media, and AC3, its companion, designed for high-quality audio. MPEG-4 was designed for Internet viewing, and, as I said, AAC was (is) its companion.

That said, there is a lot of crossing over within the formats. Personally, I encode DivX video with MPEG-1 Layer 3 audio (VBR, which is against the AVI standard, bad me, but oh well).

I don't mean to get off on a rant here, but the largest problem I see with MP3 is that people aren't using VBR. It's an EXCELLENT thing: take bits from places where they aren't needed and put them where more bandwidth is needed. You can have a "CD-quality" MP3 file using VBR -- at lower filesizes -- that you would need 320Kbps CBR to achieve.
Well, that, and people are still using Xing. stop it! Use LAME + EAC to rip and encode your CDs. Honestly.

And now, the obligatory plug. For more information on MP3 encoding, visit r3mix.net. Of course, these are facts, not opinions; I couldn't be wrong.

Listening to that on one of those teenie Mac speakers, of COURSE it's not going to sound any different. ;) While I can't entirely vouch for the quality of AAC, from what I've heard, it's really fitting to be alongside MPEG-4 video -- MPEG-4 video was designed for Internet viewing, and IIRC, AAC was designed for the same purpose.

Really, it comes down to this:

In the beginning, there was MPEG-1 and MPEG-1 Layer 1,2,3 audio. Big deal, koz they were all more-or-less firsts in semi-quality video compression.
Then came (in no particular order) MPEG-2 and MPEG-4. MPEG-2 was a format designed for DVDs and other high-resolution media, and AC3, its companion, designed for high-quality audio. MPEG-4 was designed for Internet viewing, and, as I said, AAC was (is) its companion.

That said, there is a lot of crossing over within the formats. Personally, I encode DivX video with MPEG-1 Layer 3 audio (VBR, which is against the AVI standard, bad me, but oh well).

I don't mean to get off on a rant here, but the largest problem I see with MP3 is that people aren't using VBR. It's an EXCELLENT thing: take bits from places where they aren't needed and put them where more bandwidth is needed. You can have a "CD-quality" MP3 file using VBR -- at lower filesizes -- that you would need 320Kbps CBR to achieve.
Well, that, and people are still using Xing. stop it! Use LAME + EAC to rip and encode your CDs. Honestly.

And now, the obligatory plug. For more information on MP3 encoding, visit r3mix.net. Of course, these are facts, not opinions; I couldn't be wrong.

Well since LAME Ain't an MP3 Encoder, they don't have to.

It's just C code. It, uh, could be an encoder, it could be C source to the Space Shuttle's navigation system, or C source to my wired toilet bowl! ;-)

Actually from visiting the page they don't say that anymore, but there is nothing that indicates that they have to pay $50,000!!!

Actually, it's just MPEG layer 3. Most MP3s are MPEG-1, but if you're using a good encoder, you can also do MPEG-2[.5] layer 3. It's quite handy if you want to encode something that's not (or shouldn't be (*cough128kbps/44.1kHzDubyaSpeechescough*)) sampled at 44.1kHz, because you can get roughly the same quality you'd get by upsampling and encoding, but at around half to 1/4 the bitrate.

If you are going to use Windows software from untrusted (i.e. most everyone, especially M$) sources you must take steps to protect yourself. First, trust your gut. Does the developer "smell funny"? Is the software from a startup company with no visible means of revenue? I tend to trust programs created by individuals or small teams that demonstrate some passion for what they do (EAC, or LAME for example)

Then, get Technological on their ass. Start with a personal firewall that monitors all outgoing traffic. Zone Alarm is the one I trust - gut feelings, and I've read some negative things about Black ICE. Amaze and astound your friends as you block requests from RealPlayer, Windows Update, and other "legitimate" programs that like to access the net without asking permission.

Then get Ad Aware and get that sinking feeling as you see the total number of unauthorized programs, components, and services on your system.

Finally, install Proxomitron to make make your browser behave a bit more politely by re-writing the html it sees before it sees it (and find yet another reason to love Shonen Knife. They're way kawaii!)

Forewarned and fore-armed (hairy ones, even), you stand a much better chance of maintaining control of your system.

Fifth post for the Happy Dayz Krew. Propz to all dead Joanie.

Dear Mr Trolling For Free Software -- you might want to check the licences on OpenDivX.

Not that licencing problems would hurt the popularity of a video codec or affect distribution in the real IP-is-optional world..

128 kbps is not CD quality. Please refer to this site to learn a bit about this stuff -- it is very informative. They even got a --r3mix option added to LAME, which encodes with their recommendations.

No no, Lame ain't an MP3 encoder!

I think that if you're using any recent Debian distribution (like testing), then compiling MPlaying from sources (including getting the dependencies right) is quite easy, since the MPlayer developers include a debian directory containing scripts for generating a Debian package.

To compile MPlayer from sources (and as a result, generate a binary package suitable for your system), just grab the sources, unpack them, get into the root directory of MPlayer and issue:

dpkg-buildpackage -us -uc -rfakeroot

You'll have to have debmake and fakeroot installed, as well as anything needed to make MPlayer (which is not much). Some (all?) of the packages needed for the build process may be listed in the debian/control file (I don't remember).

The only problem so far that I've had with Mplayer is that it doesn't compile on a PowerPC computer that I have here (due to its usage of x86 Assembly).

The same comments about compiling MPlayer from source in a Debian system also apply to compiling Lame from source.

Sure, if you want to sit down and go crazy on a 100-CD ripping binge, it's going to take forever, but "time consuming" != "impossible."

Ripping (win32) 15-20 CDs per hour isn't difficult if you have lots of hard drive space.

(linux/win32) The time-consuming encoding can easily be automated with LAME or a front end for it like (win32) Razorlame

miniDisc is 100% proprietary, down to not being able to make more than 1 copy of anything in miniDisc format and having to wait real-time to make those copies (Thanks Sony!)

MP3 is proprietary if you choose to use or buy the licenses to proprietary software and codecs. You do not have to do this with the LAME MP3 encoder.

That being said, use Ogg Vorbis.

If you're running Windows, you can get ABX from http://www.pcabx.com/. On UNIX systems, the LAME source code comes with an ABX program (in the misc/ directory, I think).

Here is an example of a test that took place using a slightly different testing methodology, more akin to MUSHRA (which is used to evaluate lots of encoders at the same time): http://www.ff123.net/128tests.html.

LAME is the best MP3 encoder available. It's free, open source, and multiplatform.

On Windows, the best ripper/encoder is CDex. On Linux, look for packages for Lame, CDParanoia and Grip for your distribution.

For discussion of the best MP3 encoders, take at look at R3Mix (and in particular at the forums).

BladeEnc is just a mildly tweaked version of the ISO sample code. If you want to stay with MP3, get a recent version of Lame and you'll be amazed how much better the music will sound at the same bitrate.

I tinkered around with Vorbis, but "ogg123" (clone of mpg123) locked my FreeBSD system up solid when I tried to play a tune--so I scrapped it.

Looking through the Vorbis development archives, I see some reports of (fixed) problems with OpenBSD, but nothing about FreeBSD. Download it again and try it - and if there are still problems, email them a bug report.

There is also a lot of information coming from the Lame Project, particular of your interest may probably be reading the mailing list which probably answers all your questions.

In your particular case, in my opinion you should try experimenting with the --r3mix switch using the latest Lame3.89beta or maybe even a Lame3.90alpha, which should give you VBR files averaging 170~220kbps. In any case reading the forums there is the best way to keep up to date with each Lame development.

In my opinion, Lame exceeds by far Fraunhofer's quality, but only if allowed to use the higher bitrates. On the other hand, Fraunhofer can't do better than average even with higher bitrates.

Ogg is on its way, but it is *still* behind the current state of mp3 encoding quality. Due to its design, it seems it has all what is needed to improve in both quality and size constraints; but its not there yet, needs a lot of testing and tweaking, a process very time consuming that needs a lot of people. I don't think its even fair to test it before the first non beta release is made, much less compare it.

--

There is also a lot of information coming from the Lame Project, particular of your interest may probably be reading the mailing list which probably answers all your questions.

In your particular case, in my opinion you should try experimenting with the --r3mix switch using the latest Lame3.89beta or maybe even a Lame3.90alpha, which should give you VBR files averaging 170~220kbps. In any case reading the forums there is the best way to keep up to date with each Lame development.

In my opinion, Lame exceeds by far Fraunhofer's quality, but only if allowed to use the higher bitrates. On the other hand, Fraunhofer can't do better than average even with higher bitrates.

Ogg is on its way, but it is *still* behind the current state of mp3 encoding quality. Due to its design, it seems it has all what is needed to improve in both quality and size constraints; but its not there yet, needs a lot of testing and tweaking, a process very time consuming that needs a lot of people. I don't think its even fair to test it before the first non beta release is made, much less compare it.

--

I'm conflicted about this: on the one hand, I am concerned that companies will glom on to Vorbis, make proprietary extensions, and not release them back into the free software pool. Not good.

They can embrace and extend all they want, but the free stuff will still be there.

Ever heard of L.A.M.E.? And doesn't Fraunhofer hold a patent on MP3 coding only in Germany and U.S.?