[grc] Fwd: audacity-devel Digest, Vol 118, Issue 70

[Frieda Werden]

This is an interesting but perhaps not a definitive test of audio sampling
preferences.  One question I would have is what speakers or headphones they
were using to run the test.

On a related topic, does anyone on this list have programming experience
and an interest in working on features for Audacity?  After many years,
this open source audio editing program has made great strides - I use it
all the time now.  But I still dearly want a feature that allows a tag to
stay with the same spot in the audio, regardless of whether the time
position of that bit of sound changes.

---------- Forwarded message ----------
From: <audacity-devel-request at lists.sourceforge.net>
Date: Wed, Feb 24, 2016 at 3:14 PM
Subject: audacity-devel Digest, Vol 118, Issue 70
To: audacity-devel at lists.sourceforge.net

...


Message: 1
Date: Wed, 24 Feb 2016 16:32:46 +0200
From: Jari Aalto <jari.aalto at cante.net>
Subject: Re: [Audacity-devel] MP3 format and music
To: audacity-devel at lists.sourceforge.net
Message-ID: <87a8mqtbzl.fsf at cante.net>
Content-Type: text/plain

2016-02-18 01:56 Martyn Shaw <martynshaw99 at gmail.com>:
| On 17/02/2016 04:02, M C Sharma wrote:
| > Dear All,
| > I appreciate that the mp3 format does a lot of space saving and is
| > useful at many places. But for me (at least in some cases) steals away
| > the soul of music. I find every thing sounding hollow.
|
| Monty has the last word on this, I believe
|
| https://xiph.org/~xiphmont/demo/neil-young.html
|
| Martyn

This articale is indeed a gem, thanks to the link[0].

Some audiophiles are even afraid of ABXing[1].

I read that in 2000, a German CT (computer technik) magazine
set up a double blind test against MP3 128 kbps vs 256 kbps
vs CD quality (Note: codes have improved since then). They
used profesional audio room: dampened walls, reflection and
resonance free suspended ceiling, professional audio studio
room with a B&W Nautilus 803 speakers (about $5000). The 12
test persons where professionell musicians, soundtechs and
so on. They could hardly ever distinguish 128 kbps from 256
(only by 5% statistical margin) but they couldn't tell 256
kbps from the original CD. And even with some sound samples
the testers surprisingly voted 128 kbps to be the best.
Summary: anything above 128 is good enough for critical
listening[2]

In another critical audiophile forum, after ABX testing,
noone could hear any difference between FLAC and 320 kbps
MP3[3]

Jari

[0]  https://xiph.org/~xiphmont/demo/neil-young.html
     ([75%]...) Double-blind listening tests are the gold
     standard; in these tests neither the test administrator
     nor the testee have any knowledge of the test contents
     or ongoing results. Computer-run ABX tests are the most
     famous example, and there are freely available tools for
     performing ABX tests on your own computer. ABX is
     considered a minimum bar for a listening test to be
     meaningful; reputable audio forums such as Hydrogen
     Audio often do not even allow discussion of listening
     results unless they meet this minimum objectivity
     requirement (...) Science is science, no slacking.

[1]  The ABX is a rigorous statistical test where anything
     below 9/10 is pure guessing; provided that you've set
     test sample count to 10; It is NOT a reding of
     "Percetage right" or "success rate".

     NOTE: forget listening or making decisions based on
     anything you hear from Youtube: it has already
     compressed the audio in all uploaded videos. Do ABX at
     home.

     https://en.wikipedia.org/wiki/ABX_test
     https://www.youtube.com/watch?v=jt7GyFW4hOI ABX tutorial
     https://www.youtube.com/watch?v=UrfX-g8auc8 ABX test

[2]  http://www.heise.de/ct/artikel/Kreuzverhoertest-287592.html

[3]  http://www.head-fi.org/t/431522/abx-test-of-320kbps-vs-flac-results




-- 
Frieda Werden, Series Producer
WINGS: Women's International News Gathering Service www.wings.org