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.
Topic: The Future of FLAC (Read 60633 times) previous topic - next topic
0 Members and 2 Guests are viewing this topic.

The Future of FLAC

Reply #50
One obvious area where help may be appreciated is the website, which indeed is looking like 1997 and also is outdated content-wise.

I agree. Who do I contact regarding this?
I'm not that good at web design, but I could at least submit some news etc.


Steinberg Cubase and Adobe Audition

Both of those actually support FLAC natively.
(Information like this could be added to the FLAC website.)


Meanwhile, you can also vote for FLAC support in Windows Phone (no need to log in). Supposedly Microsoft actually reads those requests.

The Future of FLAC

Reply #51
I think Xiph should push to make FLAC an ISO-standard (or IETF-standard like Opus).
I think that would help it become the mainstream standard it deserves to be.

Does anyone know if this has been tried?
XLD // ALAC // OGG VORBIS

The Future of FLAC

Reply #52
Quote
I think Xiph should push to make FLAC an ISO-standard (or IETF-standard like Opus).
I think that would help it become the mainstream standard it deserves to be.

IETF usually involves Internet standards; stuff that would be used over the internet. Lossless audio requires a lot of bandwidth, but with today's internet speeds it might be feasible. Streaming FLAC won't work on my connection

Opus satisfies a lot of use cases for use on the Internet like low-delay, low-bandwidth, packet loss etc.

The Future of FLAC

Reply #53
Quote
I think Xiph should push to make FLAC an ISO-standard (or IETF-standard like Opus).
I think that would help it become the mainstream standard it deserves to be.

IETF usually involves Internet standards; stuff that would be used over the internet. Lossless audio requires a lot of bandwidth, but with today's internet speeds it might be feasible. Streaming FLAC won't work on my connection

Opus satisfies a lot of use cases for use on the Internet like low-delay, low-bandwidth, packet loss etc.


Yes, ISO would be the most fitting standards-body.
XLD // ALAC // OGG VORBIS

The Future of FLAC

Reply #54
FLAC needs >2G support

More or less, they reached a point of maturity. But they still need to do improvements (2gb limit for example).


This bugfixed version http://www.foobar2000.org/encoderpack does, as you can see from the news here: http://www.foobar2000.org/?page=News


AFAIK the future Xiph's format Ghost will support lossless. But it won't be ready any time soon.



I'm honestly not really interested in formats that support both lossy and lossless mode. It's purely my opinion, but I like knowing that if I have a FLAC file that I have an exact copy of the audio that was encoded. If I want audio in a form where I don't mind loss, then there are plenty of choices out there. Mixing lossy and lossless audio in the same codec does nothing but confuse the issue and introduce too much chance of mistaking a lossy version as lossless and introducing loss into a signal chain that was intended to be lossless.

Imagine if mp3 had a lossless mode and you had the hordes of noobs out there mangling their encodes because they don't understand the difference. You'd have people loading sansa clips with lossless mp3 files then complaining because they can only get 5 or 6 albums on it. Then you'd have "CD backups" loaded with 128Kbps CBR files. You know it would happen, A LOT.

Keep it simple.


You couldn't have explained better why I don't like WavPack.

Opensource projects are too easily abandoned, what does it really take to switch website like Xiph was saying? Don't want to be rude but people don't really have half an hour to do this? I'd love too see FLAC becoming the default lossless codec of every system.

The Future of FLAC

Reply #55
Oh this nonsense has been debunked before. Wavpack lossless files have a 'lossless' flag in the tag profile and most people are using lossless mode exclusive or a hybrid / lossless mix. WV Lossy files have a 'lossy' flag. You don't know the authenticity of a file unless you encode it yourself. Just because a FLAC file doesn't support lossless doesn't mean the source isn't lossy. In some cases it is on P2P networks and likewise LAME V0 / 320k files have a sharp 16khz cutoff . You don't know the a flac file hasn't been tampered with . No one will get a wavpack lossy file unless using the -b switch in the encoder explicitly.

The Future of FLAC

Reply #56
Oh this nonsense has been debunked before. Wavpack lossless files have a 'lossless' flag in the tag profile and most people are using lossless mode exclusive or a hybrid / lossless mix. WV Lossy files have a 'lossy' flag. You don't know the authenticity of a file unless you encode it yourself. Just because a FLAC file doesn't support lossless doesn't mean the source isn't lossy. In some cases it is on P2P networks and likewise LAME V0 / 320k files have a sharp 16khz cutoff . You don't know the a flac file hasn't been tampered with . No one will get a wavpack lossy file unless using the -b switch in the encoder explicitly.

Yeah ok, tell everyone to check the tag. This is how it works mostly: extension = .wv? Must be lossless, or: "it's wavpack it can't be lossy!".

Of course people convert lossy to lossless even with a lossless only codec but the chance to exchange lossy files is reduced.

The Future of FLAC

Reply #57
How exactly is a wavpack user gonna end up with a lossy / lossless confusion - unless they have seriously diminished brain function ?

The lossy mode is disabled by default.  I believe shorten also has a lossy mode.

Who is publishing wavpack files these days ? I am not aware of any commercial vendors but I have seen stuff P2P and was always labeled 'wavpack lossless - EAC'.

The Future of FLAC

Reply #58
You couldn't have explained better why I don't like WavPack.

Opensource projects are too easily abandoned...

so, when FLAC was abandoned, how did someone pick it up again?  because it was open source.  a closed source project can be abandoned just as easily, but then no one can continue the work without the source.  in this case, and most others, open source is better.

and even if it were to happen that anyone were to distribute a lossy WavPack file as lossless, the recipients would become aware of this probably as soon as:
- looking at the bitrate.
- looking at the tags.
- looking at the file size.
as said, you cannot encode a lossy WavPack file without making an educated effort at doing so.

The Future of FLAC

Reply #59
I don't get it.
Container formats like WAV, AIFC, CAF, AVI, MP4, MKV, MOV can actually contain many different codecs under the same container/file extension, and it might indeed be confusing to a noob.
On the other hand, I don't think Wavpack is so much confusing... You are not required to get codec packs or something from somewhere. If a player supports Wavpack, it will play it. At least much simpler than MP4 codecs. Isn't it?

BTW you can still use lossyWAV + FLAC, and it actually works fine.
And if it is so much confusing, one can just name a file like .hybrid.wv or something, I suppose.

The Future of FLAC

Reply #60
How exactly is a wavpack user gonna end up with a lossy / lossless confusion - unless they have seriously diminished brain function ?


By converting all their .wv files to .flac, without thinking that some of their .wv's were lossy? Which could of course happen if the user retrieves .wv from TOS9-uncompliant sources. Ignorance is bliss, thus so is transpacence.

Or by messing up folder structure? Myself I apply one-folder-per-disc, but even if the folder has artist + year + albumtitle, so does the filename. Will easier detect it if I manage to drag+drop by mistake. Also, those pesky pre-emph'ed CDs are not only tagged as such, they are stored in a different lossless format (WavPack, actually – my codec of first choice is FLAC), in case I manage to mess up a tag.

The Future of FLAC

Reply #61
Some of this is pure naivety. lossyFLAC exists  The trivial ability to transcode your entire playlist (WAV, mp3, WV, etc etc) into FLAC exists. Bad rippers exist.

It's up to you what level of paranoia you want to adopt.

Cheers,
David.

The Future of FLAC

Reply #62
FLAC is a great format.

But given that the market is moving to thinner and lighter devices a storage size will remain small compared to one of a desktop devices.
Lossy formats will remain more attractive alternative for a long time.


It's very uncommon that somebody uses lossless on his/her smartphone or tablet with 16-32 GB or so.  Oh, and time delay to transfer big files to mobile devices quite uncomfortable. Too many limitations to my taste.

The Future of FLAC

Reply #63
Class 10 64GB µSDHC card in phone, FLAC to phone in not too much time at all.
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)

The Future of FLAC

Reply #64
I keep a collection of 8 GB cards ($4 each) and load each with a bunch of FLAC files. That way I don't worry about how long it takes to reload one. Each card then has about 20 hours of music, which is longer than the battery in the clip+ lasts.

The Future of FLAC

Reply #65
I use FLAC as my playable local network archive. If I'm moving music to a mobile device then I transcode it on the fly from those FLAC files. My machine can transcode the files faster than any of my players or thumb drives can write, so speed is not an issue.

The Future of FLAC

Reply #66
I find this thread to be a fascinating look into a codec that I'm just starting to use (yes, I'm a latecomer).

So here's a simpleton's question: am I correct that Sourceforge still contains the latest FLAC binaries?  And that there are no "variant" versions available (akin to halb27's -V0+ variant of LAME)?

The Future of FLAC

Reply #67
So here's a simpleton's question: am I correct that Sourceforge still contains the latest FLAC binaries?  And that there are no "variant" versions available (akin to halb27's -V0+ variant of LAME)?

Yes, the official reference binaries are still hosted on SourceForge.  But there are different FLAC encoders out there using different approaches to encoding files - typically for better encoding speed, better compression, etc.  However, since they're all lossless, the actual data stored is bit-for-bit identical no matter what approach is used.  So the encoder doesn't really matter.

The Future of FLAC

Reply #68
I find this thread to be a fascinating look into a codec that I'm just starting to use (yes, I'm a latecomer).

So here's a simpleton's question: am I correct that Sourceforge still contains the latest FLAC binaries?  And that there are no "variant" versions available (akin to halb27's -V0+ variant of LAME)?

There is also a bugfixed version allowing >2GB files. You can get the .exe from inside this archive: http://www.foobar2000.org/encoderpack

The Future of FLAC

Reply #69
However, since they're all lossless, the actual data stored is bit-for-bit identical no matter what approach is used.  So the encoder doesn't really matter.


Should be pointed out that not all FLAC files are streamable, though. The safe option is to use the reference encoder (and avoid the --lax command-line option ... I guess beginners will be more than satisfied with -8).

The Future of FLAC

Reply #70
Class 10 64GB µSDHC card in phone, FLAC to phone in not too much time at all.

An increment happens for all aspects. A storage, resolution, performance.
Phones are getting HD displays. Now people want to store a higher resolution photos, HD videos and bigger applications. 64 GB isn't exclusively for  audio files.

The Future of FLAC

Reply #71
Should be pointed out that not all FLAC files are streamable, though. The safe option is to use the reference encoder (and avoid the --lax command-line option ... I guess beginners will be more than satisfied with -8).

Thanks for the tip, as I was about to ask where I could find some of the alternate versions which allow for compression beyond -8.

But yes, I've found -8 -p to be more than satisfactory in most cases.  That said, I have been surprised at the apparent complexity of some of my favorite tracks, as FLAC could only go to a .800 or so ratio on some of them, even with these settings.


The Future of FLAC

Reply #73
But yes, I've found -8 -p to be more than satisfactory in most cases.  That said, I have been surprised at the apparent complexity of some of my favorite tracks, as FLAC could only go to a .800 or so ratio on some of them, even with these settings.


If you want to be completely and uncompromisingly lossless, that's something you'll have to accept if you like any pop or rock music made after the late 1990s.

It's not uncommon to find lossless bitrates of around 1000 kbps in albums from the last few years which are 'competitive with their peers' in terms of loudness on shuffle play - i.e. they are victims of the Loudness War. If the volume is very high, distortion is introduced to make it so high, which is less predictable, causing lossless encoders to use more bits encoding the error between their prediction and the actual values. Also the bottom bits are full of much more random noise floor below the non-random music (what's not predictable is called the residual, which takes up most of the space of a lossless file) and a lossless encoder dutifully encodes every last bit of the noise floor too, which might be thought of as 10 to 13 bits per channel of essentially random data in some extreme Loudness War victims!

As greynol says, TAK will probably perform a little better than FLAC on these, though don't expect miracles!

There are some ways to overcome this needless bloat, which while not 100% lossless compared to what's on the CD, are effectively lossless according to the question "If this sound were mastered at a reasonable level that's still transparent, what would lossless look like?". In other words it's as though you re-mastered to a normal volume (but kept the distortion) then encoded to lossless.

  • Remaster at a sensible volume, e.g. 83 to 89 dB SPL like a classical or early 1990s CD
    Bitrate: 650-800kbps perhaps - Use ReplayGain Apply Album Gain (with clipping prevention) and dither e.g. in foobar2000 Converter. Changes original volume, but saves reaching for volume control in shuffle.
  • Use lossyWAV to make filename.lossy.flac
    Bitrate: 440-550 kbps (standard, high or extreme settings) - never shown non-transparent at standard or above. Becomes fully lossless in quiet tracks or deep fades. No low pass filtering. CUEripper / CUETools includes early version (with no adaptive noise-shaping so slightly higher bitrates) but very easy to use when ripping or converting.
  • WavPack Lossy mode
    Bitrate: 384, 448 or 512 kbps typically. Doesn't analyse noise floor like lossyWAV but very robust around 512. Becomes fully lossless in deep fades or very quiet tracks, no low pass filtering.

Dynamic – the artist formerly known as DickD

 

The Future of FLAC

Reply #74
I have a new iPhone 5, my first Apple device. My audio collection is all FLAC. In the app store, I have my choice of about 10 players that handle FLAC, ranging from free to $10 (I like HD Player and Capriccio, both of which support ReplayGain), so the lack of native Apple support doesn't bother me (I don't use iTunes).

However, I only have the 16GB iPhone, so I don't store much of my FLAC stuff on the phone. I've used dbPoweramp to convert a few FLAC albums to Nero AAC using high bitrate VBR, and frankly I can't hear much of a difference even with a good pair of earbuds. Until Apple starts supporting hi-res audio (which I think they will do simply to charge more for the music), they're not going to significantly upgrade the DACs in their portable products. I'm very impressed that they managed to fit any DAC in such a small, thin device as the iPhone 5. With the faster dual-core CPU, they could have gone DSP-only.

IMO, the current Cirrus Logic DACs aren't good enough to warrant the extra file size of FLACs for portable use. Interestingly, Apple has included a Wolfson DAC in the Lightning-to-30-pin adapter, so they're still interested in pushing quality sound out to external playback devices. And Lightning can output both digital and analog signals to external devices.

I haven't tried Opus yet, but if it's as good as the reports indicate the future for portable lossless files of any kind may be limited. The masses are all streaming now, they don't want to mess with managing a collection (from what I've seen of the new iTunes, all the improvements have been made to the store experience, not collection management). I have a feeling that FLAC is going to be relegated to the music server/HTPC segment, and its days as a portable format are limited, even as storage gets cheaper.