FLAC or other uncompressed formats, which is best?

[Mxsmanic]

I'd like to compress the WAV files in my archive to something lossless but
compressed. I've read about FLAC a lot, how widely is it supported? Or is
there another format that's better still? I'm using Sound Forge.

[Arny Krueger]

"Mxsmanic" <mxsm...@gmail.com> wrote in message
news:v84oo7tj6lc3j8fsq...@4ax.com...

> I'd like to compress the WAV files in my archive to something lossless but
> compressed. I've read about FLAC a lot, how widely is it supported?

FLAC, WAV, and uncompressed AIF seem to be the most widely supported.

> Or is there another format that's better still? I'm using Sound Forge.

FLAC would be a good choice.

[Preben Friis]

"Arny Krueger" wrote in message
news:EaKdndHlurIivxHS...@giganews.com...

> FLAC would be a good choice.

Agreed .... unless you ever want to use it in a Apple software.... In that
case the best choice is Apple Lossless (ALAC).

/Preben Friis

[John Williamson]

Unless, that is, you ever intend to use it on anything other than an
Apple machine. Apple lossless is purely an Apple file format.

FLAC is open source, and should have support for quite a while. There is
even a player for FLAC files available for Macs, as well as number of
free converters.

--
Tciao for Now!

John.

[jus...@ao1.com]

For you I recommend mp3 at 48kbs. It;s lsssless and compressess very well
and removes all the hissssssssssssss.

[Don Pearce]

On Mon, 16 Apr 2012 14:32:34 +0200, Mxsmanic <mxsm...@gmail.com>
wrote:

>I'd like to compress the WAV files in my archive to something lossless but
>compressed. I've read about FLAC a lot, how widely is it supported? Or is
>there another format that's better still? I'm using Sound Forge.

I would strongly recommend you simply buy a bigger hard drive.
Actually I would be amazed if you could even get close to filling a
moderately sized drive. Do you really need to compress, or do you just
feel you should?

The less you compress, the better is the error recovery if the odd bit
goes bad.

d

[Mxsmanic]

Preben Friis writes:

> Agreed .... unless you ever want to use it in a Apple software.... In that
> case the best choice is Apple Lossless (ALAC).

I'm not likely to be using any Macs any time soon, but just out of curiosity,
who else supports this ALAC format besides Apple?

[Mxsmanic]

John Williamson writes:

> Unless, that is, you ever intend to use it on anything other than an
> Apple machine. Apple lossless is purely an Apple file format.

OK, I can scratch that off my list (if Sound Forge supports it--I'm not sure).

> FLAC is open source, and should have support for quite a while. There is
> even a player for FLAC files available for Macs, as well as number of
> free converters.

FLAC it is, then.

[Mxsmanic]

jus...@ao1.com writes:

> For you I recommend mp3 at 48kbs. It;s lsssless and compressess very well
> and removes all the hissssssssssssss.

I'm not aware of any variation of MP3 that is lossless. It wouldn't compress
very well at all if it were lossless.

[Mxsmanic]

Don Pearce writes:

> I would strongly recommend you simply buy a bigger hard drive.
> Actually I would be amazed if you could even get close to filling a
> moderately sized drive. Do you really need to compress, or do you just
> feel you should?

I use the same drive for both audio and video files, and the video files are
huge, so I was just thinking of saving some space. As long as it's a lossless
format, I'm not really sacrificing anything.

> The less you compress, the better is the error recovery if the odd bit
> goes bad.

Even with a lossless format? They shouldn't be too sensitive to dropped bits.
Then again, if I'm dropping bits on a hard disk drive, I have much more
serious problems to worry about, such as imminent failure.

I did lose a gigantic (500 MB) film scan in TIFF on a CD-R a few days ago. The
CD-R is ten years old and the TIFF file is no longer readable, although the
10-times-smaller JPEG on the same CD was readable when I put the CD in the
same drive that burned it.

[nebulax]

There's another open-source lossless format called WavPack, but I
don't see it in use as much as FLAC - <http://en.wikipedia.org/wiki/
WavPack>

-Neb

[Marc Wielage]

On Mon, 16 Apr 2012 07:15:24 -0700, John Williamson wrote
(in article <9v2nrt...@mid.individual.net>):

> Unless, that is, you ever intend to use it on anything other than an
> Apple machine. Apple lossless is purely an Apple file format.

>------------------------------<snip>------------------------------<

Not any more. Apple released Apple Lossless as open source about five months
ago. Licensing is now free, so anybody can use the source code at no charge.

--MFW

[geoff]

Technically all equal. So what is most standard and likely to 'survive' ?

Probably FLAC.

But as FLA=C only gives around 50%,and hard drives are huge/huger/cheap/etc,
why bother ? Why not straight vanilla LPCM ?

geoff

[geoff]

Mxsmanic wrote:
> Don Pearce writes:
>
>> I would strongly recommend you simply buy a bigger hard drive.
>> Actually I would be amazed if you could even get close to filling a
>> moderately sized drive. Do you really need to compress, or do you
>> just feel you should?
>
> I use the same drive for both audio and video files, and the video
> files are huge, so I was just thinking of saving some space. As long
> as it's a lossless format, I'm not really sacrificing anything.

Thgen spend another $50 for another hard drive and put just audio on it.
And another $50 just to back it all up.

geoff

[Tobiah]

Consider yourself trolled.

[Nil]

On 17 Apr 2012, "geoff" <ge...@nospampaf.co.nz> wrote in
rec.audio.pro:

> But as FLA=C only gives around 50%,and hard drives are
> huge/huger/cheap/etc, why bother ? Why not straight vanilla LPCM
> ?

For one (major) thing, FLAC supports a full compliment of information
and artwork tags.

And what's wrong with saving 50% of your disk space? If I had 50% more
room in my house, I'd be ecstatic!

[Mxsmanic]

geoff writes:

> Thgen spend another $50 for another hard drive and put just audio on it.
> And another $50 just to back it all up.

Agreed, but my budget doesn't stretch to that right now. If and when I have
enough money, I'm just going to get some more USB disk drives for precisely
that purpose. Right now I have only one archive copy of the videos and audio
files.

[Mxsmanic]

geoff writes:

> But as FLA=C only gives around 50%,and hard drives are huge/huger/cheap/etc,
> why bother ? Why not straight vanilla LPCM ?

Point taken. Just trying to leave as much space as I can free for video files,
which are very large.

[John Williamson]

Even more reason not to take a chance on file corruption, then. As has
been said here, if a few bytes in an uncompressed audio file are
corrupted, you get a short glitch, whereas the same loss on any
compressed file, even one that's losslessly compressed can irreversibly
corrupt a whole block of up to a second. Compare it with mjpeg
compressed video, where a single corrupted byte can wipe out a second or
more of video.

This is why all my backups are stored on uncompressed drives, with only
working copies being on the compressed NTFS HD in the laptop.

[geoff]

I wouldn't trust a USB memory stick for archiving things.

geoff

[geoff]

Actually there *is* a lossless MP3 format. MP3HD. Never caught on.

geoff

[geoff]

Your house is too small for a HDD, that fits inside an existing space anyway
?!!

;-)

geoff

[Mxsmanic]

geoff writes:

> I wouldn't trust a USB memory stick for archiving things.

I was talking about external hard disk drives with a USB interface.

[Nil]

You haven't seen my house. Oy vey.

[Trevor]

"geoff" <ge...@nospampaf.co.nz> wrote in message
news:dLCdnXV1ZbKIohDS...@giganews.com...

> Technically all equal. So what is most standard and likely to 'survive' ?
> Probably FLAC.

> But as FLAC only gives around 50%,and hard drives are

> huge/huger/cheap/etc, why bother ? Why not straight vanilla LPCM ?

Yep, I use wave rather than waste my time compressing/uncompressing files to
save a few dollars.

Trevor.

[Nil]

On 17 Apr 2012, "Trevor" <tre...@home.net> wrote in rec.audio.pro:

> Yep, I use wave rather than waste my time
> compressing/uncompressing files to save a few dollars.

Why are you spending so much time compressing/uncompressing? It might
make sense to keep them in WAV format while they're being edited or
processed, but the final product can be compressed and left that way.
Most digital players can cope with them just as easily as an
uncompressed version. Plus FLAC files can contain lots of embedded
information like title, author, composer, date, cover art, lyrics, and
much more.

[Trevor]

"Nil" <redn...@REMOVETHIScomcast.net> wrote in message
news:XnsA038D316...@wheedledeedle.moc...

> On 17 Apr 2012, "Trevor" <tre...@home.net> wrote in rec.audio.pro:
>> Yep, I use wave rather than waste my time
>> compressing/uncompressing files to save a few dollars.
>
> Why are you spending so much time compressing/uncompressing? It might
> make sense to keep them in WAV format while they're being edited or
> processed, but the final product can be compressed

Without "spending time compressing"?

>and left that way. Most digital players can cope with them just as easily
>as an
> uncompressed version.

Mine can't, but I don't use wave files for a portable player in any case. I
use wave for storage and home use, MP3 for portables and car. FLAC would
just add one more layer of time and effort. A 1TB disk holds quite a few
songs in wave format already :-)

Trevor.

[Nil]

On 17 Apr 2012, "Trevor" <tre...@home.net> wrote in rec.audio.pro:

> Without "spending time compressing"?

Of course not. But you don't have to wait for decompression. Besides,
it doesn't take very long to compress them, and the process can be
automated. It's not like you have to squeeze them into a little can by
hand.

>>and left that way. Most digital players can cope with them just as
>>easily as an uncompressed version.
>
> Mine can't, but I don't use wave files for a portable player in
> any case.

Me, either. I use mp3 for the portable and FLAC for home use. The info
tags are invaluable.

Too bad about your player. I rely on the embedded info tags in FLAC and
MP3 files for organization, and I wouldn't be interested in a player
that can't use them.

> FLAC would just add one more layer of time and effort. A
> 1TB disk holds quite a few songs in wave format already :-)

Really the time is minimal, and I like the idea of the disk holding
quite-a-few x 2 songs.

[Trevor]

"Nil" <redn...@REMOVETHIScomcast.net> wrote in message

news:XnsA038D7DA...@wheedledeedle.moc...

> Too bad about your player. I rely on the embedded info tags in FLAC and
> MP3 files for organization, and I wouldn't be interested in a player
> that can't use them.

Mine handles the tags in the MP3 files just fine, and I have NO need for
FLAC in a portable player when 256 or 320kbs is more than enough for the
purpose, and when storage space is more limited than my desktop.

>> FLAC would just add one more layer of time and effort. A
>> 1TB disk holds quite a few songs in wave format already :-)
>
> Really the time is minimal, and I like the idea of the disk holding
> quite-a-few x 2 songs.

Your choice, whatever works for your needs. It's simply a trade off between
time/convenience/expense/security/future needs etc.
And good luck with your 50,000 - 100,000 songs per 1TB HD in FLAC format.
That's an awful lot of ripping/encoding time in any case! :-)

Trevor.

[Mxsmanic]

Nil writes:

> Why are you spending so much time compressing/uncompressing? It might
> make sense to keep them in WAV format while they're being edited or
> processed, but the final product can be compressed and left that way.

I keep mine as WAV files if I'm still doing something with them, but once they
are stable and ready to be archived, I figure saving 50% on space--with no
loss in quality--isn't a bad idea. It can eventually add up.

It's true that the disk isn't nearly full yet, but video files are chewing
through the available space rapidly.

[anahata]

On Wed, 18 Apr 2012 08:47:44 +1200, geoff wrote:

> Tobiah wrote:
>> On 04/16/2012 08:36 PM, Mxsmanic wrote:
>>> jus...@ao1.com writes:
>>>> For you I recommend mp3 at 48kbs. It;s lsssless

>>> I'm not aware of any variation of MP3 that is lossless.

>> Consider yourself trolled.

> Actually there *is* a lossless MP3 format. MP3HD. Never caught on.

Not at 48kbs ;-)

--
Anahata
ana...@treewind.co.uk --/-- http://www.treewind.co.uk
+44 (0)1638 720444

[Low...@ao1.com]

Nil <redn...@REMOVETHIScomcast.net> wrote:
> FLAC files can contain lots of embedded
>information like title, author, composer, date, cover art, lyrics, and
>much more.

You've said that twice already. But flac compression doesn't preserve index
markers. There is some alternate flac program that will keep the markers
but in my experience, the standard flac programs won't save the markers.

That kills it for me. The SHN compression format, on the other hand, does
indeed save the markers.

[Nil]

On 19 Apr 2012, Low...@ao1.com wrote in rec.audio.pro:

> You've said that twice already.

I only repeated it because people don't seem to have noticed it. That's
a very important aspect to me. Raw audio files with no ability to hold
more information than just the file name seem primitive and useless to
me. Most digital audio players can use those tags to organize, sort,
and search for music. If you've got a lot of digital music, I don't
know how you could find and listen to the stuff without the use of good
tags. It's the key for me.

> But flac compression doesn't
> preserve index markers. There is some alternate flac program that
> will keep the markers but in my experience, the standard flac
> programs won't save the markers.

Well, I'm not even sure what "index markers" are and I've never used
them, so it's never been a concern to me.

There is only one "standard FLAC program", as far as I know. There are
many GUI interfaces for that program, and maybe some can access more
features of the core codec than others. I do notice that WAV files can
sometimes contain metadata, and the FLAC codec can strip out that
metadata or not, depending on your instructions. Maybe those indexes
are of that type.

> That kills it for me. The SHN compression format, on the other
> hand, does indeed save the markers.

SHN is decades outdated, compresses far worse, and doesn't support
tags. It's dead to me.

[Jason]

On Thu, 19 Apr 2012 21:01:13 -0400 "Nil" <redn...@REMOVETHIScomcast.net>
wrote in article <XnsA03AD5D4...@wheedledeedle.moc>

fwiw, the upcoming new edition of Audition, CS6, supports FLAC and APE
formats directly (as well as others that are new to Auditon).

[Nil]

On 19 Apr 2012, Jason <jason_...@ieee.org> wrote in
rec.audio.pro:

> fwiw, the upcoming new edition of Audition, CS6, supports FLAC and
> APE formats directly (as well as others that are new to Auditon).

That's good to know. My old stalwart Audition 1.5 can open and save
FLAC files (courtesy of some add-on I added a long time ago.)

APE format seemed to have most of the advantages of FLAC, and
compressed a little tighter. As I recall, it was originally a
proprietary technology but later became open source. Still, I haven't
seen it used much in the past few years. FLAC seems to be the most
common and popular lossless codec these days.

[Scott Dorsey]

Nil <redn...@REMOVETHIScomcast.net> wrote:
>On 19 Apr 2012, Low...@ao1.com wrote in rec.audio.pro:
>
>> You've said that twice already.
>
>I only repeated it because people don't seem to have noticed it. That's
>a very important aspect to me. Raw audio files with no ability to hold
>more information than just the file name seem primitive and useless to
>me. Most digital audio players can use those tags to organize, sort,
>and search for music. If you've got a lot of digital music, I don't
>know how you could find and listen to the stuff without the use of good
>tags. It's the key for me.

Well, then use Broadcast WAV. It has all those tags in it, and it's
uncompressed, and has been an industry standard for years.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Nil]

On 20 Apr 2012, klu...@panix.com (Scott Dorsey) wrote in
rec.audio.pro:

> Well, then use Broadcast WAV. It has all those tags in it, and
> it's uncompressed, and has been an industry standard for years.

But it's twice the size of FLAC, and I don't think any of the media
players I use will read Broadcast Wav tags. No, FLAC is the best fit
for me.

[Marc Wielage]

On Fri, 20 Apr 2012 06:30:55 -0700, Scott Dorsey wrote
(in article <jmroef$2cg$1...@panix2.panix.com>):

> Well, then use Broadcast WAV. It has all those tags in it, and it's
> uncompressed, and has been an industry standard for years.

>------------------------------<snip>------------------------------<

The tags aren't readable in any consumer players, cars, or home music
servers, plus there's not a lot of agreement on where and how BWF tags are
implemented.

I think the tags for Apple Lossless (or FLAC) make a lot more sense for
consumer use. Also lossless, very reliable, and both open-source, plus the
tag support is beyond compare.

--MFW

[geoff]

But on the other hand nobody has ever heard of it, so unlikely to be
supported for long.

geoff

[Preben Friis]

"Marc Wielage" wrote in message
news:0001HW.CBB728F3...@news.giganews.com...

On Fri, 20 Apr 2012 06:30:55 -0700, Scott Dorsey wrote
(in article <jmroef$2cg$1...@panix2.panix.com>):

> Well, then use Broadcast WAV. It has all those tags in it, and it's
> uncompressed, and has been an industry standard for years.
>------------------------------<snip>------------------------------<

> ... plus there's not a lot of agreement on where and how BWF tags are
> implemented.

Can you elaborate on this? The majority of recorders seem to agree on using
cue points with a chunk ID of 'cue ' according to EBU specifications. At
least this goes for Tascam, Zoom, Edirol, Sound Devices etc. Only Marantz
and Olympus uses their own self-invented formats.

/Preben Friis

[Mike Rivers]

On 4/20/2012 10:08 AM, Nil wrote:

>> Well, then use Broadcast WAV.

> But it's twice the size of FLAC

I didn't think that size mattered these days.



--
"Today's production equipment is IT based and cannot be operated without
a passing knowledge of computing, although it seems that it can be
operated without a passing knowledge of audio" - John Watkinson

Drop by http://mikeriversaudio.wordpress.com now and then

[geoff]

Mike Rivers wrote:
> On 4/20/2012 10:08 AM, Nil wrote:
>
>>> Well, then use Broadcast WAV.
>
>> But it's twice the size of FLAC
>
> I didn't think that size mattered these days.

It shouldn't. And every 6 months about 50% less significant.

geoff

[Ron Capik]

On 4/21/2012 10:16 PM, geoff wrote:

> Mike Rivers wrote:
>>
>> I didn't think that size mattered these days.
>
> It shouldn't. And every 6 months about 50% less significant.
>
> geoff
>
>

I think you guys need to have a talk
with the women in your lives. 8-)
==

Later....
Ron Capik
--
[Sorry, the devil made me do it.]

[Marc Wielage]

On Sat, 21 Apr 2012 08:29:33 -0700, Preben Friis wrote
(in article <jmujou$vgm$1...@speranza.aioe.org>):

> Can you elaborate on this? The majority of recorders seem to agree on using
> cue points with a chunk ID of 'cue ' according to EBU specifications.

>------------------------------<snip>------------------------------<

I'm speaking strictly of song title, artist, JPEG artwork, and other
consumer-type information. For normal timecode, scene, take, notes, and
related information, I think all the major players (Fostex, Marantz, Nagra,
Sound Devices, Tascam, Zaxcom, etc.) are compatible on BWF. But BWF is not
ideal as a music format for home use.

After months of my own tests, I came to the conclusion that lossless is
lossless, and AIFF = WAV = FLAC = ALAC and every other lossless format. The
key to me is how the CDs are ripped and where the metadata comes from, which
is a different problem than what the o.p. asked.

--MFW

[geoff]

Ron Capik wrote:
> On 4/21/2012 10:16 PM, geoff wrote:
>> Mike Rivers wrote:
>>>
>>> I didn't think that size mattered these days.
>>
>> It shouldn't. And every 6 months about 50% less significant.
>>
>> geoff
>>
>>
> I think you guys need to have a talk
> with the women in your lives. 8-)
> ==

No, that side of things is sorted - she's had 4 kids, and size doesn't seem
to be an issue for either, in fact 'complimentary' ;-)

geoff

[geoff]

Each of our houses has enough room for all the kids , if you were wondering
!

geoff

[Low...@ao1.com]

>> The SHN compression format, on the other hand,
>> does indeed save the markers.


>But on the other hand nobody has ever heard of it, so unlikely to be
>supported for long.

I believe SHN precedes FLAC by a number of years.

They are both based on an ingenous compression algorithim which was then
optimized for audio probabilities.

[Randy Yates]

Mxsmanic <mxsm...@gmail.com> writes:

> I'd like to compress the WAV files in my archive to something lossless but
> compressed. I've read about FLAC a lot, how widely is it supported? Or is
> there another format that's better still? I'm using Sound Forge.

A nit, but FLAC *is* compressed - it's just losslessly compressed, as
opposed to lossy compression. I think you know that - just were a bit
sloppy in the title.
--
Randy Yates
Digital Signal Labs
http://www.digitalsignallabs.com

[Mxsmanic]

Randy Yates writes:

> A nit, but FLAC *is* compressed - it's just losslessly compressed, as
> opposed to lossy compression. I think you know that - just were a bit
> sloppy in the title.

Yes, I was very sloppy in the title. Sorry about that.

[Randy Yates]

Cool. And I also recommend FLAC - been using it for several years.
Seeing the evolution of hard drive space to the TB range, and knowing
that quality matters to me, I can say in retrospect it was a good
choice.

[Mxsmanic]

Randy Yates writes:

> Cool. And I also recommend FLAC - been using it for several years.
> Seeing the evolution of hard drive space to the TB range, and knowing
> that quality matters to me, I can say in retrospect it was a good
> choice.

Thanks, that's reassuring. I was just looking for a lossless format that is
likely to be around as long as WAV. It sounds like FLAC is widely used enough
to have some sort of perennity.

[geoff]

asLTHOUGH THERE ARE OTHERS, i SUSPECT flac WILL HAVE THE LONGEVITY.

GEOFF

Oooops capslock.

[Mike Rivers]

On 5/15/2012 10:27 PM, Mxsmanic wrote:

> Thanks, that's reassuring. I was just looking for a lossless format that is
> likely to be around as long as WAV. It sounds like FLAC is widely used enough
> to have some sort of perennity.

And just how long do you expect WAV to be around? If you
want permanence, get it in writing. ;)

"Sort of perennity" is a good way to look at it. You'll have
it for as long as you care about it, but you may realize
that eventually, when you don't have it any more (like you
no longer have anything that will play WAV or FLAC or a SATA
hard drive) you no longer care about it.

--
"Today's production equipment is IT based and cannot be
operated without a passing knowledge of computing, although
it seems that it can be operated without a passing knowledge

of audio." - John Watkinson

http://mikeriversaudio.wordpress.com - useful and
interesting audio stuff

[Trevor]

"Mike Rivers" <mri...@d-and-d.com> wrote in message
news:jp021b$oo6$1...@dont-email.me...

> And just how long do you expect WAV to be around?

I bet I'll be able to convert .wav files into anything else that takes it's
place for as long as I'm alive, and probably anybody else living today for
that matter. That's the great thing about uncompressed digital, easy peasy
to write a conversion program for a new data format/file container, without
loss.

> "Sort of perennity" is a good way to look at it. You'll have it for as
> long as you care about it, but you may realize that eventually, when you
> don't have it any more (like you no longer have anything that will play
> WAV or FLAC or a SATA hard drive) you no longer care about it.

Well obviously you need to copy all data to new media before your hard
drives are no longer readable for whatever reason.
If you don't, then as you say, you no longer care about it.

Trevor.

[William Sommerwerck]

>> And just how long do you expect WAV to be around?

> I bet I'll be able to convert .wav files into anything else that

> takes its place for as long as I'm alive, and probably anybody
> else living today, for that matter.

How about the next 10,000 years, give or take a couple of hundred thousand?


> That's the great thing about uncompressed digital, easy to

> write a conversion program for a new data format/file container,
> without loss.

Or lossy-compressed, too. (Additional loss might occur when converting to
another lossy-compression scheme.)

Digital data is highly fungible. The idea that, even a couple of thousand
years from now, the contents of WAV files (or //any//other kind of file)
won't be recoverable, is absurd.

[Randy Yates]

In celebration of FLAC, here's a good oldie and an old goodie:

http://www.digitalsignallabs.com/flac.flac

[Mxsmanic]

Mike Rivers writes:

> And just how long do you expect WAV to be around?

About as long as .TXT or .JPG files, which is to say, a very long time.

> "Sort of perennity" is a good way to look at it. You'll have
> it for as long as you care about it, but you may realize
> that eventually, when you don't have it any more (like you
> no longer have anything that will play WAV or FLAC or a SATA
> hard drive) you no longer care about it.

I haven't looked at the FLAC file format, but the WAV format is so simple that
you can write something fairly quickly that will play it.

[Mxsmanic]

William Sommerwerck writes:

> How about the next 10,000 years, give or take a couple of hundred thousand?

A WAV file is essentially just a string of digital samples, so it will be
readable even after the specs are gone, for anyone who cares to write a small
program to read it.

> Or lossy-compressed, too.

Compressed files are much, much harder to figure out. Lossy compression is
even worse. Compression is akin to encryption in terms of information theory.

[geoff]

"Mike Rivers" <mri...@d-and-d.com> wrote in message
news:jp021b$oo6$1...@dont-email.me...

> On 5/15/2012 10:27 PM, Mxsmanic wrote:
>
>> Thanks, that's reassuring. I was just looking for a lossless format that
>> is
>> likely to be around as long as WAV. It sounds like FLAC is widely used
>> enough
>> to have some sort of perennity.
>
> And just how long do you expect WAV to be around? If you want permanence,
> get it in writing. ;)
>
> "Sort of perennity" is a good way to look at it. You'll have it for as
> long as you care about it, but you may realize that eventually, when you
> don't have it any more (like you no longer have anything that will play
> WAV or FLAC or a SATA hard drive) you no longer care about it.

WAV ( aka Linear PCM) though a file format, is also pretty much a
fundamental level of physics wrt digital technology.

A technological equivalent of suggesting that 'speech' will go out of date.

geoff

[geoff]

"Mxsmanic" <mxsm...@gmail.com> wrote in message
news:b5u7r79c5j7g30c7p...@4ax.com...

Unless I'm wrong, I think FLAC possibly evolved from the WavZip, a Zip
algorithmn tuned to the audio.

geoff

[Mike Rivers]

On 5/16/2012 9:11 AM, William Sommerwerck wrote:

> Digital data is highly fungible. The idea that, even a couple of thousand
> years from now, the contents of WAV files (or //any//other kind of file)
> won't be recoverable, is absurd.

I don't think it's absurd at all. I can't imagine that the
documentation for the WAV format, or a data reduction
algorithm, will be around for a thousand years. Unless we
get really stupid (depends on how long Facebook lasts) in
the next thousand years, I suppos it would be possible with
enough time and trials, to reverse engineer the process from
the data, but people like me aren't going to do that. The
material would have to be really important to spend much on
recovering it.

[Mike Rivers]

On 5/16/2012 3:00 PM, Mxsmanic wrote:

> A WAV file is essentially just a string of digital samples, so it will be
> readable even after the specs are gone, for anyone who cares to write a small
> program to read it.

But suppose you had no idea what it was. And even if you
managed to decode it into a string of values, would you know
how to build a D/A converter?

I pose a similar question about analog magnetic tape. The
difference there is that if you know what it is, it's very
easy to convert what might be left of the magnetic domains
into audio.

[Mike Rivers]

On 5/16/2012 2:58 PM, Mxsmanic wrote:

>> And just how long do you expect WAV to be around?
> About as long as .TXT or .JPG files, which is to say, a very long time.

How can you justify that? Neither of them have been around
half the time of a phonograph record or analog recording
tape yet. We can still play those because it's easy. But if
I were to hand you an 8" floppy disk, what would you have to
do in order to read it? Sure, there are probably still some
drives in computer museums (and in my friend Don's storage
shed) but you'd pretty much have to build a computer around
it, which means figuring out the hardware interface, for
starters. Then figure out how the data is written, what bits
are the data you want to recover and what bits are checksums
and parity bits. It's not a trivial task, and most people
aren't going to take that much trouble to save what might be
a musical recording that hasn't been preserved in some other
medium.

> I haven't looked at the FLAC file format, but the WAV format is so simple that
> you can write something fairly quickly that will play it.

Your presumption is that you'll know what a WAV file is. I
predict that documentation, and perhaps even history, will
be lost in a couple of hundred years. That's the way things
have been going in the computer age.

[Frank]

On Thu, 17 May 2012 08:56:22 +1200, in 'rec.audio.pro',
in article <Re: FLAC or other uncompressed formats, which is best?>,

"geoff" <ge...@nospampaf.co.nz> wrote:

>WAV ( aka Linear PCM) though a file format, is also pretty much a
>fundamental level of physics wrt digital technology.

Okay, but remember that a .wav file can quite validly contain other
than LPCM audio data. I have perfectly valid .wav files with MPEG-1
Layer III audio data, Dolby Digital AC-3 audio data, Sony ATRAC audio
data, etc.

To reverse engineer it, you would need to know the values of the
format tags, such as 0x0055 for MP3 and 0x2000 for AC-3, so as to know
which codec to use in order to decode the audio data.

And of course, you would have to know where the RIFF header ended and
the audio data began.

It would also be nice to how many channels of audio data were
contained within the file. I have .wav files, as I'm sure many people
here do, with a single channel, two channels, six channels (5.1
surround), etc.

And don't forget about BWF (EBU Broadcast Wave Format) .wav files that
contain time code.

--
Frank, Independent Consultant, New York, NY
[Please remove 'nojunkmail.' from address to reply via e-mail.]
Read Frank's thoughts on HDV at http://www.humanvalues.net/hdv/
[also covers AVCHD (including AVCCAM & NXCAM) and XDCAM EX].

[Les Cargill]

Mike Rivers wrote:
> On 5/16/2012 9:11 AM, William Sommerwerck wrote:
>
>> Digital data is highly fungible. The idea that, even a couple of thousand
>> years from now, the contents of WAV files (or //any//other kind of file)
>> won't be recoverable, is absurd.
>
>
> I don't think it's absurd at all. I can't imagine that the documentation
> for the WAV format, or a data reduction algorithm, will be around for a
> thousand years.

Why not? Unless it just goes completely obsolete, it'll be used.

> Unless we get really stupid (depends on how long
> Facebook lasts) in the next thousand years, I suppos it would be
> possible with enough time and trials, to reverse engineer the process
> from the data, but people like me aren't going to do that. The material
> would have to be really important to spend much on recovering it.


--

Les Cargill

[Trevor]

"William Sommerwerck" <grizzle...@comcast.net> wrote in message
news:jp08vt$jso$1...@dont-email.me...

>>> And just how long do you expect WAV to be around?
>
>> I bet I'll be able to convert .wav files into anything else that
>> takes its place for as long as I'm alive, and probably anybody
>> else living today, for that matter.
>
> How about the next 10,000 years, give or take a couple of hundred
> thousand?

You think anybody will care in 10,000 years?
(I doubt there will even be anybody left to care)

>> That's the great thing about uncompressed digital, easy to
>> write a conversion program for a new data format/file container,
>> without loss.
>
> Or lossy-compressed, too. (Additional loss might occur when converting to
> another lossy-compression scheme.)

Exactly, do YOU want that?

> Digital data is highly fungible. The idea that, even a couple of thousand
> years from now, the contents of WAV files (or //any//other kind of file)
> won't be recoverable, is absurd.

Not at all. Assuming anybody cared, (and is left to care) they will simply
copy to some really archival storage when one is invented. Actually we have
some now, but not common enough or cheap enough to be useful.

Trevor.

[Trevor]

"Mike Rivers" <mri...@d-and-d.com> wrote in message

news:jp1gbc$tn4$2...@dont-email.me...

>> A WAV file is essentially just a string of digital samples, so it will be
>> readable even after the specs are gone, for anyone who cares to write a
>> small
>> program to read it.
>
> But suppose you had no idea what it was. And even if you managed to decode
> it into a string of values, would you know how to build a D/A converter?
>
> I pose a similar question about analog magnetic tape. The difference there
> is that if you know what it is, it's very easy to convert what might be
> left of the magnetic domains into audio.

And why would you suppose they know what one is and not the other, and why
would they know the more obsolete one rather than the newer? The way such
knowledge is being archived these days, the only reason they would not know
is when there's nobody left to care anyway.

Trevor.

[Trevor]

"Mike Rivers" <mri...@d-and-d.com> wrote in message

news:jp1h3p$30p$1...@dont-email.me...

>>> And just how long do you expect WAV to be around?
>> About as long as .TXT or .JPG files, which is to say, a very long time.
>
> How can you justify that? Neither of them have been around half the time
> of a phonograph record or analog recording tape yet. We can still play
> those because it's easy. But if I were to hand you an 8" floppy disk, what
> would you have to do in order to read it? Sure, there are probably still
> some drives in computer museums (and in my friend Don's storage shed) but
> you'd pretty much have to build a computer around it, which means figuring
> out the hardware interface, for starters. Then figure out how the data is
> written, what bits are the data you want to recover and what bits are
> checksums and parity bits. It's not a trivial task, and most people aren't
> going to take that much trouble to save what might be a musical recording
> that hasn't been preserved in some other medium.

Now you are simply changing the whole argument from one of file formats,
back to the perennial argument of physical file storage.

>> I haven't looked at the FLAC file format, but the WAV format is so simple
>> that
>> you can write something fairly quickly that will play it.
>
> Your presumption is that you'll know what a WAV file is. I predict that
> documentation, and perhaps even history, will be lost in a couple of
> hundred years. That's the way things have been going in the computer age.

Actually the way things are going these days is that nearly all human
knowledge is stored on current servers somewhere, which are regularly
replaced and backed up as necessary. There is no reason short of nuclear
holocost that such knowledge will simply be erased. Especially given the
continual reduction in data storage costs.

Trevor.

[Don Pearce]

On Wed, 16 May 2012 21:00:40 +0200, Mxsmanic <mxsm...@gmail.com>
wrote:

A Wav file can actually be almost anything. Your programme has no idea
what it might have to decode until it reads the RIFF.

d

[Mike Rivers]

On 5/16/2012 11:41 PM, Trevor wrote:

> And why would you suppose they know what one is and not the other, and why
> would they know the more obsolete one rather than the newer?

I don't expect the history or technical principles of
magnetic tape or a phonograph record will necessarily
survive longer than that of anything digital, but if you
have some idea of what it is, it's pretty easy to figure out
how to play it. There isn't very much that you need to try
before you can get SOME sound out of it - and from that
point your ears will tell you how to refine it. Assuming
that anyone 1000 years from now knows what today's music
sounds like.

> The way such
> knowledge is being archived these days, the only reason they would not know
> is when there's nobody left to care anyway.

That's a real problem. At some point nobody will remember
the principle of sampling, but I suspect that how magnetic
fields work will continue to be studied in whatever the
equivalent of 8th grade science class is in the future.
Today there is some interest in hearing what Enrico Caruso
or Robert Johnson or Hank Williams sounded like. But will
they in Y3K? And if they don't care what those guys sound
like, will they care what Michael Jackson sounds like?

We can speculate about either aspect of retrieving archives,
but without both, they'll remain unheard.



"Impossible" is a strong word, and I'm not saying that
nobody will ever figure out a CD 1000 years from now, but
there indeed may not be anyone who cares enough and is well
enough funded.

[Randy Yates]

typedef enum
{
PCM = 0x0001,
ADPCM = 0x0002,
IBM_CVSD = 0x0005,
ALAW_G711 = 0x0006,
MULAW_G711 = 0x0007,
ADPCM_OKI = 0x0010,
ADPCM_IMA = 0x0011,
ADPCM_SIERRA = 0x0013,
ADPCM_G723 = 0x0014,
DIGISTD = 0x0015,
DIGIFIX = 0x0016,
SONARC = 0x0021,
ADPCM_YAMAHA = 0x0020,
DSPGROUP_TRUESPEECH = 0x0022,
ECHOSC1 = 0x0023,
AUDIOFILE_AF36 = 0x0024,
APTX = 0x0025,
AUDIOFILE_AF10 = 0x0026,
DOLBY_AC2 = 0x0030,
GSM610 = 0x0031,
ADPCME_ANTEX = 0x0033,
VQPLC_CONTROL_RES = 0x0034,
DIGIREAL = 0x0035,
ADPCM_DIGI = 0x0036,
CONTROL_RES_CR10 = 0x0037,
VBXADPCM_NMS = 0x0038,
ADPCM_G721 = 0x0040,
MPEG = 0x0050,
ADPCM_CREATIVE = 0x0200,
FM_TOWNS_SND = 0x0300,
GSM_OLI = 0x1000,
ADPCM_OLI = 0x1001,
CELP_OLI = 0x1002,
SBC_OLI = 0x1003,
OPR_OLI = 0x1004,
} SAMPLE_T;

[Mike Rivers]

On 5/16/2012 11:48 PM, Trevor wrote:

> Now you are simply changing the whole argument from one of file formats,
> back to the perennial argument of physical file storage.

The two really can't be separated. There are already tales
of digital magnetic tapes that can't be played now, though
this seems like it's more along the line that nobody cares
enough.

> Actually the way things are going these days is that nearly all human
> knowledge is stored on current servers somewhere, which are regularly
> replaced and backed up as necessary.

But some of them aren't. rec.music.makers.synth has been
around longer than Google. Can you find my posts about the
first NAMM show I attended, 1988, I believe. I can't. I
Nobody archived that, unless it's in someone's private
collection. I didn't archive it myself because I figured
that the newsgroups and their content would be around for a
very long time. And if I did save the text files, they'd
probably be on an ST506 disk drive, backed up on a 5-1/4"
floppy. I still have a working computer with 5-1/4" and
3-1/2" floppy drives, at least I think it still works. But
I'm not planning to preserve it forever.

> There is no reason short of nuclear
> holocost that such knowledge will simply be erased. Especially given the
> continual reduction in data storage costs.

While data storage cost for media may have dropped, we're
putting more data in smaller containers, which means that
the risk of greater loss with smaller mishaps is greater. So
refreshing the archive must be done more frequently. Even
though this can be automated, it still involves labor, and
that's getting more expensive.

And what makes you think there won't be a nuclear holocaust?
Or a great electromagnetic pulse? The LIbrary of Congress
has a pretty good holocaust-resistant storage facility for
film, video, and recorded sound, but they don't have
everything, and they have to be selective as to what they
add to that archive. .

[Mike Rivers]

On 5/16/2012 9:42 PM, Frank wrote:

> Okay, but remember that a .wav file can quite validly contain other
> than LPCM audio data. I have perfectly valid .wav files with MPEG-1
> Layer III audio data, Dolby Digital AC-3 audio data, Sony ATRAC audio
> data, etc.

There's a common example of this in the DAW world. Questions
about why a WAV file recorded in one program (or on one
recorder) won't import into this or that DAW? The usual
answer is "open it in Audacity and export it as a WAV" with
no understanding of why this works (though, it often does).

You may think you're archiving a file, but you may have
inadvertently done it in a format that will become
unsupported sooner than others. Professional archivists have
their club and secret handshake and try to limit the chaos.

> And don't forget about BWF (EBU Broadcast Wave Format) .wav files that
> contain time code.

Wouldn't it be nice to know exactly when something was
recorded even if you can't find the track sheet?

[Mike Rivers]

On 5/16/2012 10:07 PM, Les Cargill wrote:

> Mike Rivers wrote:

>> I can't imagine that the
>> documentation
>> for the WAV format, or a data reduction algorithm, will be
>> around for a
>> thousand years.
>
> Why not? Unless it just goes completely obsolete, it'll be
> used.

And that's exactly my point. I believe that it will be
completely obsolete in a thousand years. Don't ask me to
justify that but I've seen a lot of things become obsolete
in my lifetime.

[William Sommerwerck]

Mike has a point about people "forgetting" how particular technologies
work -- or even what they were for -- but no one is going to forget the
sampling theorem, or any of hundreds and scientific or technical principles.

A couple of millenia from now, an engineer or scientist would have no
trouble figuring out what a phonograph record was, or how it worked, in two
or three seconds -- even if he or she had never heard of such a thing. Any
more than you or I would be the least confused about what cuneiform tablet
was for.

[William Sommerwerck]

Is "holocost" what you're charged to use the holodeck?

[Arny Krueger]

"William Sommerwerck" <grizzle...@comcast.net> wrote in message

news:jp2mao$s62$1...@dont-email.me...

> A couple of millenia from now, an engineer or scientist would have no
> trouble figuring out what a phonograph record was, or how it worked, in
> two
> or three seconds -- even if he or she had never heard of such a thing. Any
> more than you or I would be the least confused about what cuneiform tablet
> was for.

Really? We are currently in the midst of a tremendous shift from physical
mechanisms to software that could not have been imagined in say, 1930. That
was 80 years ago. 200 years from now, can we even imagine what the ongoing
shift will be from/to?

[Mxsmanic]

Mike Rivers writes:

> But suppose you had no idea what it was.

The format is so simple that I should think it wouldn't be hard to figure out
what it is, especially when the file extension is WAV.

> And even if you managed to decode it into a string of values, would you know
> how to build a D/A converter?

Sure. It's very basic stuff.

> I pose a similar question about analog magnetic tape. The
> difference there is that if you know what it is, it's very
> easy to convert what might be left of the magnetic domains
> into audio.

One of the advantages of analog. But in the future, converting from digital to
analog won't be significantly harder.

The problematic formats are the ones that are more complex: those that use
compression, for example. Compression greatly reduces the redundancy of data
(by design), which makes it a lot harder to guess the nature of the data that
has been compressed. Just looking at the raw data from a WAV will suggest
digitization of a waveform, but looking at raw data from a compressed file may
not immediately suggest anything.

[Mxsmanic]

Mike Rivers writes:

> At some point nobody will remember the principle of sampling ...

That would surprise me. It's so elementary that it's self-evident. And what
would replace it?

[Mxsmanic]

Arny Krueger writes:

> Really? We are currently in the midst of a tremendous shift from physical
> mechanisms to software that could not have been imagined in say, 1930.

Software had already been imagined 100 years before 1930. Babbage incorporated
the concept into his Analytical Engine. Vannevar Bush even foresaw the Web in
a 1945 paper. There's really nothing new under the sun.

> That was 80 years ago. 200 years from now, can we even imagine what the ongoing
> shift will be from/to?

Yes. The main problem with predicting the future, though, is that we tend to
assume that past trends always reflect future trends, so that we predict
change in the wrong places.

However, a technical nerd transported to the present day from 1930 would grasp
practically all modern technology very quickly indeed. It's not magic.

[Mxsmanic]

Don Pearce writes:

> A Wav file can actually be almost anything. Your programme has no idea
> what it might have to decode until it reads the RIFF.

The pattern of the raw data in the file, as well as its extension, very
strongly suggests samples of a waveform.

[Mxsmanic]

Mike Rivers writes:

> I can't imagine that the
> documentation for the WAV format, or a data reduction
> algorithm, will be around for a thousand years. Unless we
> get really stupid (depends on how long Facebook lasts) in
> the next thousand years, I suppos it would be possible with
> enough time and trials, to reverse engineer the process from
> the data, but people like me aren't going to do that. The
> material would have to be really important to spend much on
> recovering it.

WAV file format is trivially simple. It will be very easy to figure out in the
future, even for someone with no documentation.

[Mxsmanic]

Trevor writes:

> You think anybody will care in 10,000 years?

People today are still looking at things that are 10,000 years old. And 10,000
from now, there will be a lot more to look at.

I'm sure there will be people around who will want to know how people talked
today. It's unfortunate that we don't have digital sound recordings from
10,000 years ago. But we do have digital text recordings from thousands of
years ago, and those have been extremely instructive.

[Mxsmanic]

Mike Rivers writes:

> How can you justify that?

Well, text has been around for thousands of years, and putting it into a
computer file instead of engraving it on a clay tablet is not a significant
transformation.

> Neither of them have been around
> half the time of a phonograph record or analog recording
> tape yet. We can still play those because it's easy.

We can play sheet music from centuries ago; in fact, we can play sheet music
from as long ago as it has existed. We can read the words written by people
even further back into the past.

> But if I were to hand you an 8" floppy disk, what would you have to
> do in order to read it? Sure, there are probably still some
> drives in computer museums (and in my friend Don's storage
> shed) but you'd pretty much have to build a computer around
> it, which means figuring out the hardware interface, for
> starters.

There is still abundant documentation for all of this, and old 8-inch drives
are still easy to find if you really need one. It's possible to build one if
necessary.

> Then figure out how the data is written, what bits
> are the data you want to recover and what bits are checksums
> and parity bits. It's not a trivial task, and most people
> aren't going to take that much trouble to save what might be
> a musical recording that hasn't been preserved in some other
> medium.

It is indeed a trivial task. Keep in mind that cryptologists accomplish a lot
more than this with a lot less to go on.

> Your presumption is that you'll know what a WAV file is.

The raw data and the extension will make it obvious.

I'm not worried about WAV files.

[Mxsmanic]

Mike Rivers writes:

> The two really can't be separated. There are already tales
> of digital magnetic tapes that can't be played now, though
> this seems like it's more along the line that nobody cares
> enough.

Yes, nobody cares enough. And reporting tapes that _can_ be read isn't
particularly newsworthy.

> But some of them aren't. rec.music.makers.synth has been
> around longer than Google. Can you find my posts about the
> first NAMM show I attended, 1988, I believe. I can't.

I can still find my Usenet posts from much earlier than that (which is not
necessarily a good thing in my case, but that's the way it is).

> While data storage cost for media may have dropped, we're
> putting more data in smaller containers, which means that
> the risk of greater loss with smaller mishaps is greater. So
> refreshing the archive must be done more frequently. Even
> though this can be automated, it still involves labor, and
> that's getting more expensive.

You can put all the world's literature on a single drive today. With millions
of such drives scattered around the world, chances are that there will always
be some that are still readable.

> And what makes you think there won't be a nuclear holocaust?
> Or a great electromagnetic pulse?

Anything extreme enough to wipe out every disk drive everywhere will be
extreme enough to make the recovery of those drives unimportant.

[Mxsmanic]

Frank writes:

> Okay, but remember that a .wav file can quite validly contain other
> than LPCM audio data.

Yes, but that won't make the files that do contain LPCM any harder to read.

[Mxsmanic]

Mike Rivers writes:

> There's a common example of this in the DAW world. Questions
> about why a WAV file recorded in one program (or on one
> recorder) won't import into this or that DAW? The usual
> answer is "open it in Audacity and export it as a WAV" with
> no understanding of why this works (though, it often does).
>
> You may think you're archiving a file, but you may have
> inadvertently done it in a format that will become
> unsupported sooner than others. Professional archivists have
> their club and secret handshake and try to limit the chaos.

You can write something in half an hour that will read PCM audio from a WAV
file. It's just not that hard.

[Randy Yates]

If it's linear PCM, that's true. As I posted earlier today, .wav can
contain a lot more formats than linear PCM, however.

[William Sommerwerck]

"Arny Krueger" <ar...@cocmast.net> wrote in message
news:0N6dnYya-onTfinS...@giganews.com...

I'm looking at it in terms of fundamental principles. These don't change,
and any reasonably perceptive person can apply them.

[William Sommerwerck]

>> At some point nobody will remember the principle
>> of sampling ...

> That would surprise me. It's so elementary that it's
> self-evident.

It's not self-evident, not at all. (This is why many people still believe
that a higher samping rate, per se, produces better sound.) But it is a
basic mathematical principle, and the textbooks of 10,000 years from will
still cover it.)

> And what would replace it?

Exactly. It is fundamental, and can't be "replaced".

[Don Pearce]

On Thu, 17 May 2012 14:35:18 +0200, Mxsmanic <mxsm...@gmail.com>
wrote:

The extension suggests nothing more than that the file contains sound.
The pattern of raw data may well suggest LPCM encoding.

d

[Frank]

On Thu, 17 May 2012 06:32:21 -0400, in 'rec.audio.pro',
in article <Re: FLAC or other uncompressed formats, which is best?>,

And that list just includes formats defined through early 1994.

OT: Just heard that Donna Summer, born 1948, has passed.

--
Frank, Independent Consultant, New York, NY
[Please remove 'nojunkmail.' from address to reply via e-mail.]
Read Frank's thoughts on HDV at http://www.humanvalues.net/hdv/
[also covers AVCHD (including AVCCAM & NXCAM) and XDCAM EX].

[Randy Yates]

Frank,

I knew the list was dated, but where did you get the specific year
1994 from?

[Les Cargill]

Mike Rivers wrote:
> On 5/16/2012 10:07 PM, Les Cargill wrote:
>> Mike Rivers wrote:
>
>>> I can't imagine that the
>>> documentation
>>> for the WAV format, or a data reduction algorithm, will be
>>> around for a
>>> thousand years.
>>
>> Why not? Unless it just goes completely obsolete, it'll be
>> used.
>
> And that's exactly my point. I believe that it will be completely
> obsolete in a thousand years. Don't ask me to justify that but I've seen
> a lot of things become obsolete in my lifetime.

I've seen more things *not* go obsolete than *go*
obsolete. If .wav goes obsolete, it doesn't matter.

Unlike old movies, .wav is purely an abstract data format.
It could theoretically last forever.

--
Les Cargill

[Les Cargill]

Mike Rivers wrote:
> On 5/16/2012 11:48 PM, Trevor wrote:
>

<snip>

>
> And what makes you think there won't be a nuclear holocaust? Or a great
> electromagnetic pulse? The LIbrary of Congress has a pretty good
> holocaust-resistant storage facility for film, video, and recorded
> sound, but they don't have everything, and they have to be selective as
> to what they add to that archive. .
>

in that case, whoever is left will be back to stories
around a campfire.

http://en.wikipedia.org/wiki/Time_Enough_at_Last


>
>

--
Les Cargill

[Les Cargill]

William Sommerwerck wrote:
> Is "holocost" what you're charged to use the holodeck?
>
>

Not sure. They never had money on Star Trek.

I think "hollowcost" is the price difference between an
ES335 and a Les Paul...


--
Les Cargill

[Les Cargill]

Mike Rivers wrote:
> On 5/16/2012 9:42 PM, Frank wrote:
>
>> Okay, but remember that a .wav file can quite validly contain other
>> than LPCM audio data. I have perfectly valid .wav files with MPEG-1
>> Layer III audio data, Dolby Digital AC-3 audio data, Sony ATRAC audio
>> data, etc.
>
> There's a common example of this in the DAW world. Questions about why a
> WAV file recorded in one program (or on one recorder) won't import into
> this or that DAW? The usual answer is "open it in Audacity and export it
> as a WAV" with no understanding of why this works (though, it often does).
>

That's because the people who develop those recorders aren't very nice
to their customers.

*Some* of us actually wrote code to read things like the Fostex VF16
file format. And we converted 'em to .wav files.

(actually, I borrowed source code from the FDMS3 FUSE FS as a template,
since doing FUSE on Windows was impossible ).

> You may think you're archiving a file, but you may have inadvertently
> done it in a format that will become unsupported sooner than others.

Doctor, doctor, it hurts when I do that...

> Professional archivists have their club and secret handshake and try to
> limit the chaos.
>

Since they're "professional", hopefully they develop a sense of
humo(u)r and treat it as an opportunity to charge more...

>> And don't forget about BWF (EBU Broadcast Wave Format) .wav files that
>> contain time code.
>
> Wouldn't it be nice to know exactly when something was recorded even if
> you can't find the track sheet?
>

--

Les Cargill

[Frank]

On Thu, 17 May 2012 13:21:41 -0400, in 'rec.audio.pro',

Randy, go to my documentation index at
http://www.humanvalues.net/hdv/pdf/ and download the file with the
following name:
Microsoft_Multimedia_Standards_Update_-_Revision_30_-_1994_April_15.pdf

And if you should see anything else there that's of interest, please
help yourself.

Regards,

[Mxsmanic]

William Sommerwerck writes:

> It's not self-evident, not at all.

It is to the sort of people who build these things. Digitization and digital
sampling have been known since long before they were actually implemented.
They are very simple concepts. The potential user might not stumble upon the
idea spontaneously, but engineers would.

A great deal of the technology we see today was thought of a very long time
ago, and the only reason it took so long to appear was that it was difficult
to engineer and build in a practical, affordable way. Engineering tends to
advance slowly and incrementally, and often takes quite a while to catch up to
theory.

I remember reading about CD audio in science magazines more than a decade
before it actually appeared. Everyone understood the concept, it was just a
matter of finding a way to actually make CDs from an engineering and
manufacturing standpoint.

[Mxsmanic]

Don Pearce writes:

> The extension suggests nothing more than that the file contains sound.

It suggests that the file models a waveform. WAVeform. Of course sound is a
likely candidate waveform.

> The pattern of raw data may well suggest LPCM encoding.

Yes, and that's the first time of encoding that would spring to mind as well.