flac dev - Mar 2012 - uncompressed FLAC

Declan Kelly <flac-dev at groov.ie> wrote:
> On Thu, Mar 08, 2012 at 09:50:20AM -0800, giles at thaumas.net wrote:
>> I wouldn't worry about it though. It's unfortunate the
dbPowerAmp
>> developers want to take advantage of the subset of customer who
don't
>> understand what 'lossless' means.
>
> I read some of the articles on www.audiostream.com earlier, and some of
> those linked from it. I can't understand the "uncompressed is
better
> than lossless" notion either. At least with CD playback, a CD-R copy
is
> going to be more prone to jitter (and added gaps from poorly configured
> rip/burn software) which does affect the sound, but comparing FLAC and
> WAV on exactly the same hardware should yield no difference in audio.
>
> On systems where the I/O is the bottleneck (for example, a smartphone
> app or hardware player with slow storage) there can be a higher risk
> of buffer underruns with uncompressed source material. And with a low
> powered CPU, there should be more of a risk of underruns with tighter
> compression ratios. This could explain why some audiophiles have heard
> better results with looser FLAC compression on the same device. But on
> modern (and reasonably powered) hardware, there should be no practical
> difference at all.
What I read was that most contributors accept
that FLAC is bit perfect.  It is easy to propose
mechanisms to explain why people hear
differences.  (The most obvious one is that
there are no differences to hear.  Another is
that the extra computation load on the
processor places noise onto the power rails,
affecting the D/A convertors.)  However,
without proper double blind testing, there is
little point.

Regards,
Martin
-- 
Martin J Leese
E-mail: martin.leese  stanfordalumni.org
Web: http://members.tripod.com/martin_leese/

Martin Leese wrote:
> What I read was that most contributors accept
> that FLAC is bit perfect.  It is easy to propose
> mechanisms to explain why people hear
> differences.  (The most obvious one is that
> there are no differences to hear.  Another is
> that the extra computation load on the
> processor places noise onto the power rails,
> affecting the D/A convertors.)  However,
> without proper double blind testing, there is
> little point.
Until there are a couple of reliable ABX tests that confirm
this its simply not worth discussing.

Erik
-- 
----------------------------------------------------------------------
Erik de Castro Lopo
http://www.mega-nerd.com/

> Martin Leese wrote:
>
>> What I read was that most contributors accept
>> that FLAC is bit perfect.  It is easy to propose
>> mechanisms to explain why people hear
>> differences.  (The most obvious one is that
>> there are no differences to hear.  Another is
>> that the extra computation load on the
>> processor places noise onto the power rails,
>> affecting the D/A convertors.)  However,
>> without proper double blind testing, there is
>> little point.
>
> Until there are a couple of reliable ABX tests that confirm
> this its simply not worth discussing.
>
> Erik
I remember Josh Coalson one time pointing out that, for some devices,
decoding FLAC is actually faster and less resource intensive than WAV
because the reduced I/O cost outweighs the CPU cost.  This would be
espcially true in a fast laptop with a 5400 RPM drive, or a mobile phone
with a 1.2GHz ARM processor and a class-2 SD card.

-Ben Allison