Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.

Poll

What's your *main lossy* format of choice?

MP3
[ 497 ] (59.1%)
Ogg Vorbis
[ 120 ] (14.3%)
AAC (MP4, M4A, AAC)
[ 137 ] (16.3%)
MPC
[ 38 ] (4.5%)
WavPack lossy
[ 3 ] (0.4%)
LossyWAV + lossless
[ 8 ] (1%)
WMA Standard or PRO (lossy)
[ 4 ] (0.5%)
Atrac (any version)
[ 0 ] (0%)
other lossy format
[ 0 ] (0%)
I don't use lossy AT ALL!
[ 34 ] (4%)

Total Members Voted: 922

Topic: 2009 ripping/encoding general poll (Read 152240 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

2009 ripping/encoding general poll

Reply #75
Excuse me... where's the problem here? I don't think what you are saying is a good reason for not using AAC. First, what do you mean saying "no good open source implementation"? An encoder? Since the latest and current version of Nero produce very high quality encodings for me i don't need another and the developers are here hearing us (like Lame MP3 ones) when we find problems updating it. Anyway, most important of all, it is also not the only AAC encoder available, so, even if some day they stop to work on it there are always the others implementations like CT, Apple (free too)...

I just don't like a format that *may* be unmaintaned (encoder wise). Sure both Nero and Apple have free high quality encoders available, and it's unlikely that AAC support will be dropped in the near future, *but* if they decide to to so, or stop giving their encoder away for free you're basically stuck with whatever the last version was.
I believe people don't like codecs where no progress is made. Look at MPC, it's basically still good today and if you used it some years ago and it was good enough back then, there's basically no reason to abandon it (expect for hardware support of course), yet still after Frank Klemm stopped working on it the userbase declined dramatically.
I also think Vorbis userbase would be a lot smaller if it wasn't for ayoumi. At least here at hydrogenaudio I think people want a new version every now and then, something to play with, at least I do.

Sure open source encoders don't guarantee an encoder will me maintaned for all time, but at least there's a greater possibility.

I don't think no open source implementation is a strong point against using a format, I personally like the Nero encoder alot and use it for my movies (x264+AAC), but for my music archive, if I have the choice I will choose open over closed source.
Blubb

2009 ripping/encoding general poll

Reply #76
I find it interesting how things settle in favor of FLAC and MP3. I also feel a bit sorry for all the mpc-pioneers :-(. Will keep an eye on TAK...

Anyway... WavPack for backups, transcode to mp3 for my portable.

Am I the only one ripping with foobar??


I don't think no open source implementation is a strong point against using a format [...] but for my music archive, if I have the choice I will choose open over closed source.


Likewise. Anyway, I don't see the point in using anything but lossless codecs for archiving, even for stuff thats not "rare" or "audiophile". If you have a lossless open-source-coded backup, it really doesn't matter which lossy codec you transcode it to.

EDIT: answer to hödyr

2009 ripping/encoding general poll

Reply #77
I don't think no open source implementation is a strong point against using a format, I personally like the Nero encoder alot and use it for my movies (x264+AAC), but for my music archive, if I have the choice I will choose open over closed source.

Nero and Apple do good job around AAC.
I have feelings something like yours especially about x264. That's why I opened this poll to ask people what they think about OS AAC. http://www.hydrogenaudio.org/forums/index....64214&st=25
I think projects like Firefox, x264 and LAME wouldn't be so great without being OS. Even Google understand the importance of OS in their new browser.

2009 ripping/encoding general poll

Reply #78
On the other hand also Musepack and Ogg Vorbis are opensource and not heavily developed anymore (besides ayoumi for vorbis). Personally I don't think that it makes such a big difference if it's open source or not - at least if it's well documented (to decode). The problem just is that there aren't a lot people who are able and really want to develope an existing project any further. AFAIK also Mozilla has the problem with decreasing contributors...

2009 ripping/encoding general poll

Reply #79
On the other hand also Musepack and Ogg Vorbis are opensource and not heavily developed anymore (besides ayoumi for vorbis).

Another factor is standards. Vorbis and Musepack aren't standards while MP3 and AAC are.
In few words we can see that MP3 has a bomb mix: Standard and high quality open source encoder LAME.
AAC is standard too, but there is no high quality OS encoder.

2009 ripping/encoding general poll

Reply #80
I think having an OSS encoder/decoder is awfully important. I want to know that I (as a programmer) can, in the future, rely on published specifications to somehow get my music back. Software rots, and computers wind up in landfills.

That said, I don't know much about WavPack. Why would I use that rather than FLAC?

2009 ripping/encoding general poll

Reply #81
Slightly better compression, slightly better encoding efficiency (EDIT: well maybe not, kind of hard to say), and the ability to produce hybrid files (though there's now lossywav).

I don't buy into all this fear people are expressing about closed-source codecs; never have, probably never will.  Funny how every year a new poll comes up, we get the same old arguments.  Sort of like every time Thomas announces something about TAK he gets trolled for it not being open-source.

2009 ripping/encoding general poll

Reply #82
I don't buy into all this fear people are expressing about closed-source codecs; never have, probably never will.  Funny how every year a new poll comes up, we get the same old arguments.  Sort of like every time Thomas announces something about TAK he gets trolled for it not being open-source.

Totally agree. I've got loads of TAK and LossyTAK files and if I ever need to it's no big deal to re-encode them to FLAC or some other codec. There simply isn't a problem.

C.
PC = TAK + LossyWAV  ::  Portable = Opus (130)

2009 ripping/encoding general poll

Reply #83
I don't buy into all this fear people are expressing about closed-source codecs; never have, probably never will.  Funny how every year a new poll comes up, we get the same old arguments.  Sort of like every time Thomas announces something about TAK he gets trolled for it not being open-source.

I completely understand the annoyance at people whining every time a new version of TAK is released.  It's Thomas's code and he's free to do whatever he likes with it.

However, the "fear" at closed-source is, in my opinion, justifiable; at least for some of us.  If you run Windows, you're probably fine.  However, I don't, and as such I'd be at the mercy of wine to get windows-only programs running.  That's to say nothing of getting them integrated into my media player.  Because programs I use are open source, they are going to have better support for open formats.  It's due to this that I avoid closed formats.

It would be the height of arrogance for me to assume that authors have an obligation to release their code under a license I approve of, and of course I do not feel this way.  But for those of us whose programs are open, it's really not feasible to use closed components.  Not because of ideological reasons (at least, not in my case), but because of the simple fact that I can't use them.

I think one of the reasons that people have a reaction to the "open-source" idea, at least here on HA, is because so many people demand that all code be open.  Such attitudes do nothing to promote open code, and are probably harmful.  But not all of us who prefer open source are of the opinion that everybody needs to provide it.  I suppose we're just not as vocal as those who demand.

2009 ripping/encoding general poll

Reply #84
Most of HA's members know the reason why they prefer some particular formats. But in real life people even don't know or don't care about alternative formats to MP3.
It wouldn't be an exageration if I think that MP3 userbase is higher than 90% (outside of HA)

Many friends of mine who have iPods and iPhones (even some electonic lovers and geeks, students and already ingeneers of electronic and/or sound ) don't know about AAC and less about its efficiency.


I think there's a wide variation in how much people care or know.  I know people who use Itunes/Ipod and don't even know that most of their songs aren't mp3.

On the Sansa forums there's a lot of buzz about vorbis.

2009 ripping/encoding general poll

Reply #85
Hi,

MP3 3.98 V3 for DAP and mobile phone
FLAC for archiving and listening on PC
[Ripping with EAC to single file with cuesheet]

Happy new year!
Harald

2009 ripping/encoding general poll

Reply #86
Monkey's Audio, one file per CD with cuesheet
MP3, one file per track.
FB2K,APE&LAME

2009 ripping/encoding general poll

Reply #87
MP3 (LAME 3.98.2 -V2) for listening on PC. One file per track.
...
MP3 will be my choice for "almost transparent" encodings until other codec reach the same compatibility than it. For "medium" [-q0.35] or "streaming" [-q0.2] encodings MP4 without doubt.
Ogg is good but anyway at any quality below q6 it sounds worse than MP3 e MP4 to me, but love how it sounds with "classical" music (that is only 2% of all my music!). Musepack no way, only sounds transparent to me at high bitrates (256~320 kbps).


Interesting... What kind of music do you prefer?
I've never supposed that there will be a real music sample that will sound "almost transparent" with Lame V2 (about 190 kbps), worse with Ogg Vorbis below q6 (q5 ~ 160 kbps) and worst (not transparent) with Musepack < 256 kbps ?
Would you upload some 30 sec. sample so I could perform some ABX?
Thanks in advance.

2009 ripping/encoding general poll

Reply #88
For 2008 I answered FLAC + Vorbis.

I really love Vorbis and want to use it but this year I've picked up a PS3, iPod, iPhone, and an MP3 CD player in the car. So I've switched to MP3 just to keep my sanity. I could have went to AAC but it looks like tagging can be an issue with the PS3 unless it is in the right container/etc and I'm not sure what plays AAC+/HE/whatever etc. It's so much easier to just encode to MP3 and know it will work everywhere without issues.

As the PS3 is my main listening device and set up for quality, I encode with LAME at -V2. It's a bummer wasting that many bits on a lossy format when there are better options like AAC and Vorbis out there, especially when I want to cram music on the iPods or MP3 CDs, but it is the trade off I am making to keep life simple. I sure wish the PS3 could play FLACs natively (I know I can transcode with TVersity, but...), and I really wish its wireless connection wasn't so fragile as I can't play off a media server without it repeatedly dropping and stopping.

2009 ripping/encoding general poll

Reply #89
WavPack for lossless....Yeah, baby!

LAME V2-V0 for iPod
Vorbis q2-q4 for Rockboxed Sansa 4 GB

Sometimes Musepack because it holds a special place in my heart.  But only on the Sansa or PC, of course...


2009 ripping/encoding general poll

Reply #90
FLAC -8 for main archive
LAME -V5 for Rio/PS3
AAC-HE for mobile phone

Using dbPowerAMP Music Converter to convert from my FLAC archive to the lossy formats. Using the Rio less and less, mobile phone has taken over for out-of-home music and the PS3 takes over more and more for music at home... will probably switch to LAME -V2 as the size issue is of less and less importance.

2009 ripping/encoding general poll

Reply #91
It is such a good idea to do this yearly (but please move it to the Polls section, with the others)

In the last year I moved to a practical approach, I went mostly with LAME -V 2to1 and FLAC for lossless.
Next to that I also use lossyWav with FLAC as the great compromise, total lossless fills up my diskspace rather quickly. Musepack, WavPack and TAK, however much I like them, are on the backburner now.
In theory, there is no difference between theory and practice. In practice there is.

2009 ripping/encoding general poll

Reply #92
My lossy library consists mostly of MPC, but for new encodes I would likely choose Ogg Vorbis today.  Lossless is all FLAC.

2009 ripping/encoding general poll

Reply #93
I've always had a mix of multiple formats, but with the growing number of devices I own I have switched back to MP3 for lossy encoding. I've always wished that Vorbis support would take of on portable devices, but it remains just that, a wish.

For lossless I mainly use FLAC, but since Wavpack has been moved to gstreamer-plugins-base I am becoming a lot more interested in the format. Especially the hybrid mode. I guess I'll have to wait for the 2010 poll to see where I am with that.
--
Eric

2009 ripping/encoding general poll

Reply #94
I answered, Wavpack, MP3, disc image + CUE sheet.

Though I use lossyWAV frequently, I consider it more as near-lossless and voted MP3 as my lossy choice. If I'm typical of lossyWAV users, it would skew the poll against lossyWAV, though I suspect its user base is small anyway.

In fact, my lossless WV is mainly existing rips, not new ones, mostly image+CUE with Replay Gain data. I chose Wavpack for fast decode speed and good compression. FLAC and TAK are similar in these aspects, and FLAC is well supported and would be easy to switch to in an overnight lossless conversion session. I have no particular loyalty, so long as it can also benefit from lossyWAV.

I barely use true lossless for current CD rips - just for some precious work, especially where I've restored the audio in some way.

PC archive and transcoding source: My new disc image archives are now mostly lossyWV (lossyWAV --standard | wavpack). I use these for PC playback and to transcode for MP3 devices (via fb2k) and consider them practically as good a source as either lossless or wavgain | lossless (with dither but no noise shaping) that was a partial solution to lossless bitrate bloat in over-loud CDs before lossyWAV came along. I'm tending to embed CUE sheets now.

Conventional low-bitrate lossy: MP3 for compatibility. LAME -V5 frequently, sometimes -V3 or -V2 from lossyWV source with Album Gain applied by fb2k before encoding. LAME -V6 for speech. Usually file-per-track for normal listening & shuffle play unless I have a gapless album and prefer to encode as a single file to work around consumer MP3 players' limitations. I'd consider Musepack at -q4 or -q5 for low processor load if I had a phone/PDA with compatible software player or a Rockboxed DAP.

Quiet background music MP3s: Use foobar2000 to transcode to LAME -V5 as single file for a groups of 20-30 tracks from my playlist. For dumb MP3 player/FM/DABradio with SD card slot, I apply Album Gain when encoding, use DSPs to Resample to 44100 Hz (only activates if needed), foo_dsp_vlevel for very necessary Dynamic Range Compression, Advanced Limiter (should do nothing!) and could use Crossfader. Once encoded, typical Album Gain with my settings is near to -6 dB, and I use mp3gain to apply -9.0 dB Constant Gain. This gives finer steps on the dumb SD-card MP3 player, whose volume control is coarse at low levels. I ignore transcoding warnings even where my source files include MP3, AAC, MPC etc., though I choose lossless or lossyWAV source where possible.  I also have a couple of very old MP2 files I can play there or in my car by renaming to MP3 extension without transcoding. Can typically fit about 26 hrs of music on 2GB SD card, and modest bitrate has solved sporadic glitch problem with my old -V2 or equivalent files on that player. Had considered trying MP2 if glitch was related to decoder processing complexity, and in any case I wanted VLevel or other DRC to avoid apparent silences.
Dynamic – the artist formerly known as DickD

2009 ripping/encoding general poll

Reply #95
PC archive and transcoding source: My new disc image archives are now mostly lossyWV (lossyWAV --standard | wavpack).


I am curious. What advantage does lossyWAV + Wavpack have over Wavpack in hybrid mode?
--
Eric

2009 ripping/encoding general poll

Reply #96
The simplest advantage is that lossyWAV output can be efficiently losslessly encoded using several different lossless encoders (FLAC, WavPack, Tak, WMA Lossless).

Additionally, once processed with lossyWAV, there is no need to re-process when re-coding from one lossless codec to another.
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

2009 ripping/encoding general poll

Reply #97
...Additionally, once processed with lossyWAV, there is no need to re-process when re-coding from one lossless codec to another.

To me this is the major advantage in a practical sense (next to the even more important advantage of being able to use a lossy variant of FLAC which is supported n several mobile DAPs): being safe for the future because the lossyWAV + lossless codec result can be losslessly transcoded to any lossless codec, while the storage saving feature remains as long as the new lossless codec is lossyWAV-friendly which is true for several codecs right now).

From a more theoretical point of view lossyWAV --standard or better IMO has the better audio-theoretic foundation as the result is controlled by keeping added noise at or below the signal's noise level.
There's no such signal related quality control with wavPack lossy AFAIK. wavPack lossy basically controls the predictor error which is not exactly related to absolute error.
In practice this doesn't mean a lot because wavPack lossy yields great results as well, but it may give a better feeling of safety.
lame3995o -Q1.7 --lowpass 17

2009 ripping/encoding general poll

Reply #98
I am curious. What advantage does lossyWAV + Wavpack have over Wavpack in hybrid mode?


I'm personally keen on both reasons. Convertability to other lossless codecs could be great if I get a DAP with native FLAC support. Also the safety and theoretical justification of following the measured noise floor (constant quality or VBR) rather than following a slightly-arbitrary bit-rate/predictor coding length (variable quality ABR or CBR) that one tends to do with wavpack's built-in lossy mode.

In practice both wavpack lossy and lossyWAV + Wavpack provide excellent quality at comparable bitrates, and the bitrate overlap between methods is where wavpack lossy is generally considered transparent. Also, lossyWAV has been pretty well tested for transparency against wavpack lossy problem samples and has no known non-transparencies at standard or portable presets so far. In fact, even lossyWAV portable seems to be a decent source for transcoding to MP3 (someone tested it), so I'm particularly confident in standard mode.

At the moment, I don't have access to decent listening equipment, but I'm confident that lossyWAV standard will meet my needs in the future and will not require re-ripping from the source CDs.
Dynamic – the artist formerly known as DickD

2009 ripping/encoding general poll

Reply #99
lame -V5 for portable listening.
flac -8 -Ax2 for archiving.
No cuesheets.
She is waiting in the air