FLAC v WAV
 

A well-known hi-fi magazine recently ran an article about how apparently
an uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

--
Squirrel Solutions Ltd Tel: (01453) 845735
http://www.squirrelsolutions.co.uk/ Fax: (01453) 843773

Registered in England: 08918250

FLAC v WAV
 

I understood that to play a flac the file has to be uncompressed a bit like
Zip. Thus it tends to need a lot of ram to be available. I wonder what they
were using to play the files.

What about some of the lossless formats that Apple and others use, are these
similar?

Often of course if the hardware is working hard then one might find some
issues with the extra processing needed to record and play sounds.
Brian

--
From the Sofa of Brian Gaff Reply address is active
"Glenn Richards" wrote in message
. uk...
A well-known hi-fi magazine recently ran an article about how apparently an
uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

--
Squirrel Solutions Ltd Tel: (01453) 845735
http://www.squirrelsolutions.co.uk/ Fax: (01453) 843773

Registered in England: 08918250

FLAC v WAV
 

In article , Glenn
Richards wrote:
A well-known hi-fi magazine recently ran an article about how apparently
an uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

Well, it may be that a particular device/system running particular software
gets something wrong, or struggles to run properly. That then gets blamed
on 'flac vs wave' or whatever as if that was the cause of a more general
problem.

Some years ago when doing tests using a version of audacious I found that
when I played 24 bit wave and flac files, the flac reached the dac as 24
bit, but the wave reached it as 16bit. Last byte of each value sent as a
zero.

Nothing to do with flac vs wave per se. All to do with whoever had
developed and built that version of audacity not getting something right
and not checking. Since I had a USB DAC with an spdif out and could capture
that stream I could find the difference. But I doubt the programmer could,
or would even think of it. And I doubt many hifi 'reviewers' would either,
alas.

The more general problem is when 'reviewers' say A differs from B and then
give entirely the wrong 'reason' as fact without even knowing how to check.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

Brian Gaff wrote:

Often of course if the hardware is working hard then one might find
some issues with the extra processing needed to record and play
sounds.

Apparently this was a hardware streamer connected via a digital output.
So the "processor introducing RFI into the analogue audio" hypothesis is
out.

--
Squirrel Solutions Ltd Tel: (01453) 845735
http://www.squirrelsolutions.co.uk/ Fax: (01453) 843773

Registered in England: 08918250

FLAC v WAV
 

In article ,
Bob Latham wrote:
In article ,
Glenn Richards wrote:

Apparently this was a hardware streamer connected via a digital
output. So the "processor introducing RFI into the analogue audio"
hypothesis is out.

So either this is imagination (journalism) or the streamer was poorly
engineered. Nothing to do with flac at all.

Par for the course with many Hi-Fi mags?

--
*'ome is where you 'ang your @ *

Dave Plowman London SW
To e-mail, change noise into sound.

FLAC v WAV
 

Sounds like a marketing opportunity for Russ Andrews then.
Brian

--
From the Sofa of Brian Gaff Reply address is active
"Dave Plowman (News)" wrote in message
...
In article ,
Bob Latham wrote:
In article ,
Glenn Richards wrote:

Apparently this was a hardware streamer connected via a digital
output. So the "processor introducing RFI into the analogue audio"
hypothesis is out.

So either this is imagination (journalism) or the streamer was poorly
engineered. Nothing to do with flac at all.

Par for the course with many Hi-Fi mags?

--
*'ome is where you 'ang your @ *

Dave Plowman London SW
To e-mail, change noise into sound.

FLAC v WAV
 

Yes, that was what i was thinking. I've noticed several sound cards sound
different for all sorts of reasons, whether it be drivers, codecs or the
actual hardware.
If I recall, most of the sound cards made by Creative a few yeas ago
sampled at 48k, b then internally converted it to whatever you were trying
to use, which some people claimed was very audible, but I could not hear any
difference between it and other cards that did it other ways.
On the other hand, lossy compression like MP3 is pretty audible even at
quite high rates due to the phase problems that seem to occur. Its
acceptable on portable gear, just like tapes were, but in my view has no
place on modern high quality systems. and for goodness sake don't use it for
old 78rpm or hissy masters, as itis crap at noise presevation!
Brian

--
From the Sofa of Brian Gaff Reply address is active
"Jim Lesurf" wrote in message
...
In article , Glenn
Richards wrote:
A well-known hi-fi magazine recently ran an article about how apparently
an uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

Well, it may be that a particular device/system running particular
software
gets something wrong, or struggles to run properly. That then gets blamed
on 'flac vs wave' or whatever as if that was the cause of a more general
problem.

Some years ago when doing tests using a version of audacious I found that
when I played 24 bit wave and flac files, the flac reached the dac as 24
bit, but the wave reached it as 16bit. Last byte of each value sent as a
zero.

Nothing to do with flac vs wave per se. All to do with whoever had
developed and built that version of audacity not getting something right
and not checking. Since I had a USB DAC with an spdif out and could
capture
that stream I could find the difference. But I doubt the programmer could,
or would even think of it. And I doubt many hifi 'reviewers' would either,
alas.

The more general problem is when 'reviewers' say A differs from B and then
give entirely the wrong 'reason' as fact without even knowing how to
check.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics
http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

In article ,
Brian Gaff wrote:
If I recall, most of the sound cards made by Creative a few yeas ago
sampled at 48k, b then internally converted it to whatever you were
trying to use, which some people claimed was very audible, but I could
not hear any difference between it and other cards that did it other
ways.

I'm currently using a Digigram VX222v2 which has balanced analogue in/out
as well as most of the digital ones. Secondhand ex BBC Bush House. I
thought I'd let them do the research. ;-)

--
*Strip mining prevents forest fires.

Dave Plowman London SW
To e-mail, change noise into sound.

FLAC v WAV
 

On Tue, 03 Jun 2014 15:16:07 +0100, Brian Gaff wrote:

snip
And for goodness sake don't use it
for old 78rpm or hissy masters, as it is crap at noise preservation!
Brian


You mean that it's great at noise preservation - at the expense of the
content that you want, Brian? ;-)

FLAC v WAV
 

On Mon, 02 Jun 2014 23:35:43 +0100, Glenn Richards wrote:

A well-known hi-fi magazine recently ran an article about how apparently
an uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

I was puzzled by this too, so I listened to a file in both formats, using
sox to produce a wav from the flac.

Initially, the wav did sound a little better - until I realised that the
volume of the wav was slightly louder. Once I'd compensated by putting
up the volume a notch on my Quad pre-amp when I played the flac, I
couldn't tell any difference.

- Richard.

FLAC v WAV
 

In article , Richard
Kimber scribeth thus
On Mon, 02 Jun 2014 23:35:43 +0100, Glenn Richards wrote:

A well-known hi-fi magazine recently ran an article about how apparently
an uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

I was puzzled by this too, so I listened to a file in both formats, using
sox to produce a wav from the flac.

Initially, the wav did sound a little better - until I realised that the
volume of the wav was slightly louder. Once I'd compensated by putting
up the volume a notch on my Quad pre-amp when I played the flac, I
couldn't tell any difference.

- Richard.



Log onto this page for the Wolfson audio card for the Raspberry Pi PC
and theres a section marked download HD audio around 550 M/bytes.


These tracks are FLAC encoded, what do you think of them?..



http://www.element14.com/community/c.../raspberry-pi-
accessories/wolfson_pi?ICID=rpispace-wolfson-sideban
--
Tony Sayer

FLAC v WAV
 

On Thu, 05 Jun 2014 15:53:24 -0500, Richard Kimber
wrote:

On Mon, 02 Jun 2014 23:35:43 +0100, Glenn Richards wrote:

A well-known hi-fi magazine recently ran an article about how apparently
an uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

I was puzzled by this too, so I listened to a file in both formats, using
sox to produce a wav from the flac.

Initially, the wav did sound a little better - until I realised that the
volume of the wav was slightly louder. Once I'd compensated by putting
up the volume a notch on my Quad pre-amp when I played the flac, I
couldn't tell any difference.

- Richard.

Assuming you can playout the flac to a wav file, you can compare the
original wav file against the flac encoded one using an audio
processing utility such as audacity or CoolEdit Pro by lining them up
exactly, normalising and then subtract one from the other. If flac
fails to losslessly do its magic, you'll hear something as opposed to
the expected dead silence.

IOW, it should be easy enough to verify the "Bollicks Factor" in that
article. :-)
--
J B Good

FLAC v WAV
 

In article , Richard
Kimber
wrote:
On Mon, 02 Jun 2014 23:35:43 +0100, Glenn Richards wrote:

A well-known hi-fi magazine recently ran an article about how
apparently an uncompressed WAV file sounds better than FLAC.

Ummmm...

*facepalm*

I was puzzled by this too, so I listened to a file in both formats,
using sox to produce a wav from the flac.

Initially, the wav did sound a little better - until I realised that the
volume of the wav was slightly louder.

What's curious about that is that sox should convert with no changes in the
values when 'reconstituted' back into LPCM. So either you'd added a scaling
or the player was up to something akin to the problem I reported.

Did you just use something as simple as

sox in.wav out.flac

to do the conversion?

And what playing software did you use?

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

On Fri, 06 Jun 2014 12:11:02 +0100, Jim Lesurf wrote:

What's curious about that is that sox should convert with no changes in
the values when 'reconstituted' back into LPCM. So either you'd added a
scaling or the player was up to something akin to the problem I
reported.

Did you just use something as simple as

sox in.wav out.flac

to do the conversion?

It was quite a long time ago (when I first saw reference to this in HFN),
but as far as I can remember, yes that's what I did.

And what playing software did you use?

I streamed the files to a Pioneer N-50 using minidlna using a wired
connection.

- Richard.

FLAC v WAV
 

In article , Richard
Kimber
wrote:
On Fri, 06 Jun 2014 12:11:02 +0100, Jim Lesurf wrote:

What's curious about that is that sox should convert with no changes
in the values when 'reconstituted' back into LPCM. So either you'd
added a scaling or the player was up to something akin to the problem
I reported.

Did you just use something as simple as

sox in.wav out.flac

to do the conversion?

It was quite a long time ago (when I first saw reference to this in
HFN), but as far as I can remember, yes that's what I did.

And what playing software did you use?

I streamed the files to a Pioneer N-50 using minidlna using a wired
connection.

I don't have any experience with 'streamers' of that kind. I just play or
record on the basis that the player or recorder read or write the file.
Given that I *assume* the flac file is read by the N-50 and that's what
turns it into LPCM or whatever to output the eventual analogue signals. On
that basis it looks like the N-50 is scaling flac and wave differently.

This may be quite understandable as the device doing the conversion may
simply scale values in some way for some reason - e.g. shy of clipping
during the format conversion processing - which then means the outcome
differs from the LPCM that the flac file data represented. As I guess
you'll know, *correctly done*, wave - flac - should give a result
identical to the original. Alas, 'should' may not mean 'does' in all cases
with real devices, programs, etc. :-/

So it does look like an example of where a setup might cause an 'innocent'
reviewer would say "wave is better than flac" as a general conclusion
rather than "slightly louder sounded better" because they didn't check.

I've also lost count of reviewers saying 'DSD sounds better than LPCM' even
for the same source, and when many ADCs and DACs may use low bit
(essentially DSD) anyway. Such confusions between container and contained
are rife, alas. No help to the poor readers.

Jim


Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

On Fri, 06 Jun 2014 15:06:06 +0100, Jim Lesurf wrote:

I don't have any experience with 'streamers' of that kind. I just play
or record on the basis that the player or recorder read or write the
file. Given that I *assume* the flac file is read by the N-50 and that's
what turns it into LPCM or whatever to output the eventual analogue
signals. On that basis it looks like the N-50 is scaling flac and wave
differently.

Well, you could be right about that. FWIW, I have to set *very*
different volume levels when I'm playing internet radio via the N-50
compared to streaming from the hard disk via minidlna.

On a completely different matter, I note that you put streamers in
quotes. I've always thought that using a dlna server was 'streaming' and
that doing anything else was 'playing'. Am I right that there's been a
takeover of the term 'streaming' by the wider community that does not use
dlna? IMHO Software players 'play'. But, then, nowadays, nothing seems
to matter :-)

- Richard.

FLAC v WAV
 

In article , Richard
Kimber
wrote:
On Fri, 06 Jun 2014 15:06:06 +0100, Jim Lesurf wrote:


On a completely different matter, I note that you put streamers in
quotes. I've always thought that using a dlna server was 'streaming'
and that doing anything else was 'playing'. Am I right that there's
been a takeover of the term 'streaming' by the wider community that
does not use dlna? IMHO Software players 'play'. But, then, nowadays,
nothing seems to matter :-)

I confess I've become quite confused by the ways people have started to use
the terms 'streamer' and 'streaming'. I have the feeling that hifi mags,
etc have started using the terms for more than one thing without making a
distinction.

I use the old fashioned approach that I have something like a set of wave
or flac files stored on a device, and then a program 'plays' a file. I
don't think of that as 'streaming' even if the data is being fetched from
one machine as a program on another is playing it. Or if it is being played
from a NAS.

I *suspect* the key distinction is that I'm using a computer to run a
program to play a file and then send the output to a DAC. Whereas
'streamer' may be a device that isn't a general purpose computer. But a
device that provides its own user interface, and reads files for itself,
playing them out via its own DAC.

For me the advantage of the 'run you chosen software on your chosen
computer' is flexibility. e,g, On the PandaBoard I'm writing this email
using, I wrote my own simple wave file player to play wave files out to a
USB DAC. Yesterday I was playing a wave file I have on a NAS that way,
using SunFish and NFS. Is this 'streaming'? At present I wouldn't say it
was. But I may be conditioned by the *nix 'everything is a file' mindset.
:-)

But I'd be interested to see what definitions/distinctions others give of
this. TBH I'm not even clear what the details of things like dlna *are*
except beyond assuming they require some standards so the device can find
and identify what it can play.

The above has files in mind. Whereas 'streaming' could be applied to
situations where the source is 'stream' in the sense that there is no
predefined 'file' of a set size, name, etc. But a sequence of data from a
chosen source address.

Comments?

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

On Sat, 07 Jun 2014 09:39:32 +0100, Jim Lesurf
wrote:

In article , Richard
Kimber
wrote:
On Fri, 06 Jun 2014 15:06:06 +0100, Jim Lesurf wrote:


On a completely different matter, I note that you put streamers in
quotes. I've always thought that using a dlna server was 'streaming'
and that doing anything else was 'playing'. Am I right that there's
been a takeover of the term 'streaming' by the wider community that
does not use dlna? IMHO Software players 'play'. But, then, nowadays,
nothing seems to matter :-)

I confess I've become quite confused by the ways people have started to use
the terms 'streamer' and 'streaming'. I have the feeling that hifi mags,
etc have started using the terms for more than one thing without making a
distinction.

I use the old fashioned approach that I have something like a set of wave
or flac files stored on a device, and then a program 'plays' a file. I
don't think of that as 'streaming' even if the data is being fetched from
one machine as a program on another is playing it. Or if it is being played
from a NAS.

I *suspect* the key distinction is that I'm using a computer to run a
program to play a file and then send the output to a DAC. Whereas
'streamer' may be a device that isn't a general purpose computer. But a
device that provides its own user interface, and reads files for itself,
playing them out via its own DAC.

For me the advantage of the 'run you chosen software on your chosen
computer' is flexibility. e,g, On the PandaBoard I'm writing this email
using, I wrote my own simple wave file player to play wave files out to a
USB DAC. Yesterday I was playing a wave file I have on a NAS that way,
using SunFish and NFS. Is this 'streaming'? At present I wouldn't say it
was. But I may be conditioned by the *nix 'everything is a file' mindset.
:-)

But I'd be interested to see what definitions/distinctions others give of
this. TBH I'm not even clear what the details of things like dlna *are*
except beyond assuming they require some standards so the device can find
and identify what it can play.

The above has files in mind. Whereas 'streaming' could be applied to
situations where the source is 'stream' in the sense that there is no
predefined 'file' of a set size, name, etc. But a sequence of data from a
chosen source address.

Comments?

A "Streaming server" is one that provides a download at a rate that
matches consumption for 'playback' purposes (with a modicum of
buffering to guard against congestion/contention induced 'dropouts').

At least, that's _my_ understanding of the distinction between
dowloading / torrenting a 1 hour movie in a matter of minutes and
watching the 'stream' direct. AFAICS, other than that distinction,
they're both downloading processes.

The big advantage of streaming a media file over downloading one for
later consumption is that you can terminate (or even pause) the stream
at will, saving bandwidth consumption.

If, for instance, it turned out that you got bored with the streamed
movie or realised it was one you'd already seen before after just a
few minutes of watching it, you could terminate the stream and save
the server and the network links the burden of sending the whole movie
file.

As an example of this difference, I aquired some 54 episodes of the
Goon Show several years ago by redirecting winamp's output to a large
wav file over a 24 hour period from an internet radio station. The
stream was 64Kbps mono mp3 and the wav file grew to in excess of 7GB
(16 bit 44.1Ksps stereo).

After splitting the file into 2 sub 4GB parts (CoolEdit Pro wouldn't
handle bigger files), I was able to slice and dice the giant wav files
into seperate episodes, discovering that the 24 hours had actually
been enough to capture the full 54 episodes being looped by the
internet radio station.

Once I had my episodes, I converted them back into 64Kbps mono mp3
files creating a collection that occupied some 662MB of disk space. At
that time, I had a 10Mbps cable downlink service from NTL/Telewest
which, in theory, would have allowed me to collect the whole lot in a
matter of 11 or so minutes downloading time[1] as opposed to the 1440
minutes or so streaming time it actually took.

What that meant was I consumed less than 1% of the bandwidth I had
available to listen to the 'stream' live versus using up a maximum of
10Mbps if I could have found a server obliging enough[2] to match my
D/L speed with the exact same content.

[1] I swiftly discovered, after upgrading from the 4Mbps service, how
few servers on the internet could match my D/L speed. It was very rare
indeed to achieve a CD's worth in less than 12 minutes. Only a couple
of notable exceptions stick in my mind, Microsoft and Nvidia's driver
downloads sites.

[2] As noted, a very rare event. However, with VM's never ending 'free
speed upgrades', it seems I'm still ahead of the curve in overall
speed of the internet in general despite being on the slowest offering
short of falling back to a 'retentions' service.

Back in the days of 150Kbps - 4Mbps, when I first tried P2P file
sharing networks (notably torrents), I wasn't too impressed with the
download speeds compared to a direct download from a decent server.

Perhaps it was the quality of the client software I was using
(probably compounded by my lack of experience in setting such software
up) but the situation is somewhat reversed today. A lot of my torrent
downloads will match my self imposed 2MB/s (that's ~20Mbps) limit,
often exceeding direct downloads by a considerable margin.

I only discovered "The Joys of Torrenting" after a few years break in
trying to use torrent clients on the desktop PC when I decided to make
use of the Transmission Client service on the FreeNAS/NAS4Free server
(a no-brainer place to run such software) two or three years ago.

I'm not one of those folk who keep the torrent client active
regardless of use so I usually have the service disabled most of the
time, only starting it up when I get a hankering to collect another
bunch of 'large files'. :-)
--
J B Good

FLAC v WAV
 

On 07/06/2014 09:39, Jim Lesurf wrote:
In article , Richard
Kimber
wrote:
On Fri, 06 Jun 2014 15:06:06 +0100, Jim Lesurf wrote:


On a completely different matter, I note that you put streamers in
quotes. I've always thought that using a dlna server was 'streaming'
and that doing anything else was 'playing'. Am I right that there's
been a takeover of the term 'streaming' by the wider community that
does not use dlna? IMHO Software players 'play'. But, then, nowadays,
nothing seems to matter :-)


I'd have thought Youtube (etc) is an example of streaming, so DLNA not
required.

I confess I've become quite confused by the ways people have started to use
the terms 'streamer' and 'streaming'. I have the feeling that hifi mags,
etc have started using the terms for more than one thing without making a
distinction.


FWIW, I've always taken 'streaming' as the process of delivery (of data)
from one place to another, delivered by the owner (etc) and accessible
to authorised recipients.

Confusion can build because a lot of hardware can become involved, perhaps?

Don't think it's much more than that!

--
Cheers, Rob

FLAC v WAV
 

In article , Bob Latham
wrote:
In article , Jim Lesurf
wrote:
[snip]

Oh this is interesting.

I store flac files on a Synology NAS and then use a Sonos box (under
iPad control) to pull the data across the network and play it with spdif
into my pre-amp.

Now my definition of streaming which I assume is wrong, is that chunks
of the file is pulled or pushed just before it is needed with the
minimum amount of data stored at the player, just enough to overcome
blips on the data stream.

So I do consider what I do to be streaming.

From my POV I have a program I've written myself as an example I can use to
illusrate.

The program reads a wave file and shovels data from it to a USB DAC. First
it reads the header of the file so it finds the sample rate, etc. It then
sends to the DAC the value of the sample rate at which to operate.

Then it proceeds by grabbing successive 1-sec 'blocks' of data from the
file and dumping them into an output buffer.

The DAC reads samples from this buffer at a rate the *DAC* now controls. It
reads at regular intervals, each time taking a few samples to allow it to
play at the rate required. Once this buffer-full has been read, this is
detected and the program - which has loaded another buffer-worth ready -
gives that via the buffer.

In effect the software in the computer is just shuffling blocks of data,
and doing each new block when the DAC says it wants it.

This works either from a local hard disc, or something like a NAS or device
like a USB memory stick, say. In each case a filing system makes the file
available to be read.

I *guess* that a 'streamer' is an all-in-one chunk of hardware that
provides its own ability to access files, let the user choose one, and play
it via its own DAC. Thus saving the user the bother of having a computer
run a program to do the job. But I've only deduced this by reading what
people say.

What I don't understand and would love to is, what the blazes is the
point of dlna for audio streaming? What I do does not use dlna at all,
the Sonos just opens the file and pulls data from the NAS using good old
SMB, no dlna server needed or running. It works just fine. What does
dlna do that this doesn't?

I'm also curious about the role and point of dlna.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

On Sun, 08 Jun 2014 09:44:44 +0100, Jim Lesurf
wrote:

In article , Bob Latham
wrote:
In article , Jim Lesurf
wrote:
[snip]

Oh this is interesting.

I store flac files on a Synology NAS and then use a Sonos box (under
iPad control) to pull the data across the network and play it with spdif
into my pre-amp.

Now my definition of streaming which I assume is wrong, is that chunks
of the file is pulled or pushed just before it is needed with the
minimum amount of data stored at the player, just enough to overcome
blips on the data stream.

So I do consider what I do to be streaming.

From my POV I have a program I've written myself as an example I can use to
illusrate.

The program reads a wave file and shovels data from it to a USB DAC. First
it reads the header of the file so it finds the sample rate, etc. It then
sends to the DAC the value of the sample rate at which to operate.

Then it proceeds by grabbing successive 1-sec 'blocks' of data from the
file and dumping them into an output buffer.

The DAC reads samples from this buffer at a rate the *DAC* now controls. It
reads at regular intervals, each time taking a few samples to allow it to
play at the rate required. Once this buffer-full has been read, this is
detected and the program - which has loaded another buffer-worth ready -
gives that via the buffer.

In effect the software in the computer is just shuffling blocks of data,
and doing each new block when the DAC says it wants it.

This works either from a local hard disc, or something like a NAS or device
like a USB memory stick, say. In each case a filing system makes the file
available to be read.

I *guess* that a 'streamer' is an all-in-one chunk of hardware that
provides its own ability to access files, let the user choose one, and play
it via its own DAC. Thus saving the user the bother of having a computer
run a program to do the job. But I've only deduced this by reading what
people say.

What I don't understand and would love to is, what the blazes is the
point of dlna for audio streaming? What I do does not use dlna at all,
the Sonos just opens the file and pulls data from the NAS using good old
SMB, no dlna server needed or running. It works just fine. What does
dlna do that this doesn't?

I'm also curious about the role and point of dlna.


The point, afaik, is that it provides a searchable database of the
media content placed in its charge on the server for use by a decent
streaming client and can be configured to transcode certain media file
formats to a form that _can_ be recognised and played by the dnla
client being used.

This transcoding feature can take its toll[1] on the server's CPU so
it's best avoided where possible, especially when the server is "cpu
cycles challenged". Unfortunately, this requires that you choose a
client (media streaming box) capable of playing your files in their
native formats.

A few years ago I "Hired" a Medion streaming player from Aldi, set up
the dnla service on the FreeNAS box, drilled the necesary hole through
the basement wall into the kitchen/diner to feed a network cable to
the tiny Medion server box slung under the wall mounted TV stand on
the brackets designed just for such STB support and... was rather
disappointed with the whole experience.

At least two problems became apparent. One, it offered a very klunky
and, as seems to be the way with "TV interfaces", cumbersome/ponderous
explorer style list of the content placed in the charge of the dnla
service on the server box, and two, it couldn't distinguish between
the TV folders stored on the 4 seperate disk volumes under the control
of dnla. I landed up having to rename those folders from "TV Programs"
on each volume to "TV Programs 1" on the first disk through to "TV
Programs 4" on the fourth disk before the Medion stream player would
allow me to access the whole database.

At that point, I did rather wonder what the feckin' point of the 50
quid streaming player box was. I could have just pressed a cheap 2nd
hand laptop into mediaplayer service _without_ the faffing about with
dnla and had a much slicker PC style interface to boot. Needless to
say, I took the gadget back to Aldi for a full refund on the basis of
it not being fit for purpose.

[1] The transcoding won't always be a 'heavy duty' process, sometimes
simply a matter of transcoding the audio stream portion or stripping
out FEC crap from a TS or vice versa for a PS.
--
J B Good

FLAC v WAV
 

In article , Johny B Good
wrote:
On Sun, 08 Jun 2014 09:44:44 +0100, Jim Lesurf
wrote:
[big snip]

I'm also curious about the role and point of dlna.


The point, afaik, is that it provides a searchable database of the
media content placed in its charge on the server for use by a decent
streaming client

Ah. I can understand the logic of that on platforms that don't already have
decent filing systems, etc.

and can be configured to transcode certain media file
formats to a form that _can_ be recognised and played by the dnla client
being used.

Ah. That sounds a bit like some of the 'streaming' re-encoders, etc, that
I've come across on Linux. And which I always avoid like a plague as they
so often foul up playback by applying presumptions I don't share with
whoever set them up. :-)

All too often I've found that they not only convert the format in terms
like mp3 to lpcm but resample down in fairly clumsy ways. Hence giving you
44.1k/16 output from a 96k/24 file, having also 'improved' the gain level
without bothering to use dither, etc. PITA. Another example of how
processes neatly hidden from the user can lead to conclusions like "files
of type A sound better than files of type B" in magazines, etc. Ditto for
claims about always playing from ram, etc, etc.

The more options and clever automations are involed without the user
knowing or understanding, the more 'failure modes' there are, and reasons
for 'changes' which end up being blamed on something they can see.

I can sympathise with all this. I've been fighting to understand how a new
NAS is behaving for the last 2-3 days. Slooowly making progress, but not
exactly simple when encountered with no prior experience. The manuals are
goldmine of clear information... not. :-/

[snip interesting example of a 'too clever' such system]

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

On Sat, 07 Jun 2014 19:16:11 +0100, Bob Latham wrote:

What I don't understand and would love to is, what the blazes is the
point of dlna for audio streaming? What I do does not use dlna at all,
the Sonos just opens the file and pulls data from the NAS using good old
SMB, no dlna server needed or running. It works just fine. What does
dlna do that this doesn't?

That's OK if you've got a Sonos.

- Richard.

FLAC v WAV
 

On Thu, 05 Jun 2014 22:52:44 +0100, tony sayer wrote:

Log onto this page for the Wolfson audio card for the Raspberry Pi PC
and theres a section marked download HD audio around 550 M/bytes.


These tracks are FLAC encoded, what do you think of them?..



http://www.element14.com/community/c.../raspberry-pi-
accessories/wolfson_pi?ICID=rpispace-wolfson-sideban

Unless I've been looking at the wrong files, I think most of these are
binaural, aren't they?

- Richard.

FLAC v WAV
 

In article , Bob Latham
wrote:
In article , Richard
Kimber wrote:
On Sat, 07 Jun 2014 19:16:11 +0100, Bob Latham wrote:

What I don't understand and would love to is, what the blazes is the
point of dlna for audio streaming? What I do does not use dlna at
all, the Sonos just opens the file and pulls data from the NAS using
good old SMB, no dlna server needed or running. It works just fine.
What does dlna do that this doesn't?

That's OK if you've got a Sonos.

True I suppose but that rather misses the point.

A couple of years back when I was last looking at streamers I asked
manufacturers about some facilities I wanted and their product didn't
appear to have. I was told that DLNA doesn't support xxxyyyx and we will
have to do some hacking to make that work. At the time Sonos could do it
using Samba/SMB and so I was mystified as to why the use of DLNA seemed
so compulsory for many manufacturers especially if it was limiting as
they implied.

My impression from what people have said is that dlna works on the basis of
some specific assumptions / requirements. Possibly involving it insisting
on forms of metadata its creators have adopted, and formats they can
process. So limiting you to what they've decided, I suspect.

Personally I dislike on-the-fly convertors (gstreamer, you know who I
mean!) because you may not know if they're fiddling with the data in ways
you'd not want. As discussed earlier for player software.

FWIW All I do is give files and directories names that tell me what I want
to know - composer, etc, for classical, for example. And for something like
an LP I've made a digital file from, I add scans of the front and rear
covers of the LP sleeve, and any notes. Keep these as jpegs to display if I
choose. I've noticed that Audacity will find and display a thumbnail of
these when I give it a directory of files or a file from a directory with
such a bitmap. But to see it full-screen-size I just use a suitable bipmap
viewer. Normally, I don't bother as I'm listening to the music! :-)

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

FLAC v WAV
 

In article , Richard
Kimber scribeth thus
On Thu, 05 Jun 2014 22:52:44 +0100, tony sayer wrote:

Log onto this page for the Wolfson audio card for the Raspberry Pi PC
and theres a section marked download HD audio around 550 M/bytes.


These tracks are FLAC encoded, what do you think of them?..



http://www.element14.com/community/c.../raspberry-pi-
accessories/wolfson_pi?ICID=rpispace-wolfson-sideban

Unless I've been looking at the wrong files, I think most of these are
binaural, aren't they?

- Richard.

Yes they are!..

Do sound good on headphones:)...
--
Tony Sayer

FLAC v WAV
 

On Wed, 11 Jun 2014 17:01:07 +0100, Bob Latham wrote:

That's OK if you've got a Sonos.

True I suppose but that rather misses the point.

A couple of years back when I was last looking at streamers I asked
manufacturers about some facilities I wanted and their product didn't
appear to have. I was told that DLNA doesn't support xxxyyyx and we will
have to do some hacking to make that work. At the time Sonos could do it
using Samba/SMB and so I was mystified as to why the use of DLNA seemed
so compulsory for many manufacturers especially if it was limiting as
they implied.

I don't have any xxxyyyx, so it's not a problem for me ;-)

- Richard.