Talk:Free Lossless Audio Codec

From Wikipedia, the free encyclopedia

This is the talk page for discussing improvements to the Free Lossless Audio Codec article.
Article policies
Free software Portal
This article is part of WikiProject Free Software, an effort to create, expand, organize, and improve free software-related articles.
B rated as B-Class on the assessment scale
Mid rated as Mid-importance on the assessment scale

Contents

[edit] Change "m4a" to "aac"?

In the intro text should 'm4a' be changed to 'aac' since 'm4a' isn't a compression codec, but rather a container that may contain lossless audio?

[edit] Quality setting dependent on processor power?

"...a fast decode time that is independent of compression level." I was under the impression that the Quality field was for exactly that, to compress better or worse depending on the amount of processor power required to decode it.

I may be wrong here (not having looked at the code), but no, the "quality" of a FLAC does not adjust the decode time. It adjusts the encode time. It takes 4-5x longer for a --best compression to happen. (Of course, since it's lossless, you're generally better off just letting it take more time.) But they all decode with pretty much the same CPU usage. Phil Bordelon 14:00, 2005 Jun 25 (UTC)
Files encoded at a higher compression level require slightly more CPU power to decode, I don't have a citation but you could easily fine one in the developer mailing list or hydrogenaudio or something. Maybe it's just semantics, but using the term "quality" here is misleading and inaccurate seeing as lossless is lossless. - 85.210.50.158 22:48, 5 May 2007 (UTC)

[edit] Couple minor wonders.

  1. Should "Xiphophorus (now called the Xiph.Org Foundation)" become "Xiph.Org Foundation (formerly the Xiphophorus Foundation)"?
  2. Does Xiphophorus even bear mention? It seems to be real, but I can't find enough reference to it to even feel comfortably making the changed posited in $1 above. Didn't look a lot, but didn't readily find reference to the name change, at least.
  3. Xiphophorus does not have a disambiguation page. It just goes straight to the fish, which could lend weight to not bearing mention, but could also of course just mean there's need for a disambiguation page.

Bits worth of wonder. -Ozzyslovechild 7 July 2005 04:14 (UTC)

xiph originally chose the name after the fish; from their website:
Xiphophorus helleri is a small aquarium fish (the common Swordtail). What's special about it? Not much, really. The Xiph.Org logo doesn't even look anything like a real swordtail, but it's a logo that's been in use a long time.
What the name *does* have is the minimum requirement of one letter 'X' for a technology-related organization. That fact that it's impossible to spell is an added bonus." --Morgajel 15:15, July 27, 2005 (UTC)

[edit] FLAC vs. TTA

How does FLAC compare to TTA, and vice versa? Guaka 17:04, 11 August 2005 (UTC)

There is no need to do that as it is already well done at the HydrogenAudio Wiki linked in the comparisons section: [1] (section 10) VodkaJazz 03:03, 6 September 2005 (UTC)

[edit] Good idea

There should be a FLAC portable audio player. iRiver's next IFP line should support it. That would rock.

Look up JetAudio, ipod alternative, iaudio plays flac, ogg, mp3, wav, wma - iAUIDO X5L My sister and I have one, and it works quite well.

[edit] Lossless AAC?

Doesn't AAC have a lossless version?

There is the recently-ratified MPEG-4 ALS codec. Information on it can be found here: [2] Aottley 18:26, 16 February 2006 (UTC)

[edit] First paragraph redundancy

Is it wrong that the first two sentences of the article are somewhat redundant? Would it be better to just link the expanded acronym, i.e. "FLAC is an acronym for Free Lossless Audio Codec?

Done.--Frenchman113 21:05, 11 April 2006 (UTC)

[edit] Add trivia section?

I was thinking of adding a trivia section including the --super-secret-totally-impractical-compression-level option for flac. However it seems silly to add a trivia section with only one item. Perhaps someone knows of more trivia, or can think of a place to integrate this into the article?

Wikipedia discourages the inclusion of trivia sections. 83.104.249.240 (talk) 12:39, 18 November 2007 (UTC)

[edit] cue and track shortcuts

Is FLAC capable of using a cue file to make shortcuts to individual tracks on a big .flac file, like with Monkey's Audio? Shawnc 13:56, 21 September 2006 (UTC)

[edit] Comparison to other Lossless formats

I'd like to see some info about how FLAC compares to other loseless formats in regard to compression speed, decompression speed, gapless playback, and random access. --24.249.108.133 20:43, 28 September 2006 (UTC)

What's wrong with the links that are there? --Mcoder 00:10, 29 September 2006 (UTC)

[edit] Add Link Back?

I would like to see if anyone would object to me adding the link back to our site, The Lossless Audio Blog? Our site tries to bridge the gap between the forums and the various EAC Guides by providing information on getting started with lossless audio formats as well as current news and information. Because the Wiki pages for lossless audio formats are such a great place for those learning about the various formats I feel that our site compliments this and have heard from a lot of users voicing the same opinion.

We are also in the process of moving to our own domain, links are below. Our site was linked here for a few months but it was taken down sometime ago and I did not notice until a user notified me. Thanks for the consideration! Windmiller 11:24, 6 October 2006 (UTC)

Old Site The Lossless Audio Blog New Site The Lossless Audio Blog

Link added back. This was discussed on the talk page. thanksWindmiller 17:38, 13 October 2006 (UTC)

[edit] Check the ALAC article

Please check the talk page of the ALAC article. Some of the claims made by the article are in dire need of sources. For example I would assume ALAC has a faster decode time cf FLAC if I simply read the ALAC article. Looking at hydrogen audio wiki I don't think this is true but it would be a resonable assumption Nil Einne 16:54, 27 October 2006 (UTC)

[edit] Command-line quibbles/questions

The example commands for flac include things like '-r 0,8' which seems to be unnecessary, as 0 is default if -r is specified and MIN isn't. '-r 8' should achieve the same thing. It also ignores the -{0..8} flags and --fast|--best options, which might be better for the objective. Not really knowing anything about flac, I dunno, but I note those discrepancies, FWIW.

[edit] pistonsoft.com

Hello! I want to add link to the multi format audio converter.
Direct Audio Converter and CD Ripper - is a very useful tool that allows you to convert audio files between various audio formats and rip CD audio tracks directly to MP3, WMA, WAV, OGG, FLAC, Monkey's Audio APE or MusePack MPC.
Please tell me how can I do this?
Thank you.

--AudioMaster 15:19, 19 January 2007 (UTC)

You can't. Wikipedia is not a place for product promotion. Femto 15:27, 19 January 2007 (UTC)

[edit] Decoding, Encoding, Ripping sections clean-up?

Hi there,

I'm concerned about the fact that most tools for decoding, encoding and ripping listed under Linux (ogg123, mpd, CDDa2wav, CDParanoia, Mencoder) are in fact POSIX-like command shells which are compiled for most UNIX-like OSes (Including BSD-Based OSes, Linux, Mac OS X, etc.). iTunes is listed under Windows and not Mac OS X, and, quite generally, this section seems quite incomplete. I'll do what I can with what I know, but I don't think I'll be able to straighten it up all by myself... Anyways, keep up the good work!

Hara-Tiri 20:57, 8 February 2007 (UTC)

[edit] WMA can be lossless

Deleted the false statement that said WMA is lossless: WMA CAN be configured to be lossless, as its own Wikipedia article says. F15x28 07:19, 22 February 2007 (UTC)

This article did that wrong... but I have a comment... why is this the only article that says a 40-50% rate? - 68.228.56.158 02:19, 1 June 2007 (UTC)

[edit] Software header reorganization

I find the Encoding and the Ripping section to be confusing. From a FLAC perspective isn't ripping the same as encoding? I also notice that some pieces of software are listed multiple times under different headers, this makes it messier than is needed. I propose merging the Encoding and Ripping sections. I also propose creating a combined Decoding / Encoding section to separate the duplicate entries. What do you think? (Requestion 16:47, 27 February 2007 (UTC))

I have a small question that I haven't seen related yet, can the flac implementation encode in realtime?

[edit] Technical details discrepancy?

The article states this in one paragraph:

"It can handle any PCM bit resolution from 4 to 32 bits per sample, any sampling rate from 1 Hz to 1,048,570 Hz"

and this in another:

"FLAC allows for ... a wide range of sampling rates up to 192 kHz, in various bits-per-sample width."


Which is it?

MGlosenger 01:52, 15 June 2007 (UTC)

[edit] Several errors need correcting

There are a few errors in this article. As the creator of flac I don't want to edit these myself to avoid any problems with bias, so I'll just list some things that are wrong (in order of appearance) and hopefully someone will investigate:

  • the reference FLAC project itself contains only plugins for xmms and winamp (there are many other developed independently)
  • our winamp plugin is LGPL
  • the reason given for not supporting floating point is incorrect. it is possible for float data to be losslessly coded; FLAC doesn't support this due to little demand
  • the max sample rate is 655350Hz. however this limitation can be worked around since the sample rate is not considered in the algorithm; FLAC simply stores it for use in playback at the right rate
  • someone has a burr under their saddle about the api changes. they did not "break the compilation of all existing players" or make the software support list "inaccurate because of the changes" (driven home in 3 places in the text!). in fact I submitted patches to dozens of projects (all I could find) for the new api and published an exhaustive porting guide (http://flac.sourceforge.net/api/group__porting.html).
  • the stream/seekable/file api layers have been collapsed. I'm not even sure describing the api organization is all that useful

Josh Coalson 18:25, 18 June 2007 (UTC)

[edit] Compression level?

The article mentions up front that it compresses 40-50% -- is there any statistic to back this up? Most of my FLAC rips (highest compression) don't hit 50% compression or even close to it.

[edit] Ability to stream

Can this format be streamed from a streaming server? Might be useful to add a comment about streaming to the page? --Tomhannen 21:55, 27 August 2007 (UTC)

[edit] compression vs mp3

Do you think it should be mentioned somewhere in the article that CDs encoded to FLAC make files that are about 8 times larger than equivalent mp3s? At least, that's what I've found from my own experience. I just think it would be useful to know (it would have saved me a lot of time if I'd known this before I downloaded software to rip CDs to FLAC). Of course, it would be difficult to verify, but whoever wrote this article seems not to have put much importance on citing their sources anyway.

TheGreekMind (talk) 06:45, 29 November 2007 (UTC)

Depends on the mp3 bitrate but my files are like 3 times larger than my mp3s (LAME VBR V0) l33tb0b (talk) 16:32, 4 March 2008 (UTC)
Audio sources encoded to FLAC are typically reduced in size 40 to 50 percent (47% according to their own comparison) - this is pulled directly from the opening paragraph. Obviously, this level of compression is nowhere near MP3.
MP3, and other lossy formats, may compress a file to ~70% of its original size. (from a lossless source, such as a CD recording) Since FLAC is a lossless compression method, rates are much lower in order to preserve a 1:1 copy of the original source. To achieve a level of compression similar to that of an MP3, the data would have to be corrupted. (made lossy) MP3 encoding destroys the original recording's integrity in order to achieve these levels of compression, which can cause arificating and similar degredation in quality. So, FLAC = exact copy of original, but very limited compression; MP3 = bad quality rendition of original, but greater compression. That's the jist of FLAC versus MP3 for you. 68.209.235.149 (talk) 01:16, 22 March 2008 (UTC)

[edit] VLC info wrong?

http://wiki.videolan.org/FLAC (the wiki of the developers), says: VLC can only decode [a FLAC] container, and cannot encode it. (Sometimes it can't do that either; I'm getting "flacdec: This stream uses too many audio channels" with 0.8.6c on Tiger) So it should be moved, yes? —Preceding unsigned comment added by 76.21.22.240 (talk) 01:16, 29 December 2007 (UTC)

[edit] A format or a codec?

Suggestions that it's a codec:

Suggestions that it's a format:

  • stated in the intro
  • the paragraph on libFLAC. You don't get implementations of a codec - a codec is an implementation.

Can anybody explain the discrepancy? Was it named FLAC by somebody who doesn't know the meaning of the word, or was the name transferred from a codec to the format as a whole at some time in history? -- Smjg (talk) 16:26, 15 April 2008 (UTC)