RE: [MLUG - DISCUSSION] Audio file quality tests

["Hargus, Diana" <EMAIL:PROTECTED>]

> -----Original Message-----
> From: EMAIL:PROTECTED [mailto:discussion- 
> EMAIL:PROTECTED] On Behalf Of Mike Miller
> Sent: Tuesday, January 08, 2008 12:41 PM
> To: MLUG Off-Topic Discussion
> Subject: RE: [MLUG - DISCUSSION] Audio file quality tests
> 
> On Tue, 8 Jan 2008, Hargus, Diana wrote:
> 
> > Would it be a feasible test to have a band perform and then compare
> an
> > analog recording to various bitrate samples?
> 
> Go for it.  It was a big issue back in the '80s -- was digital sound 
> as good as analog?  Some people would claim that they could tell the 
> difference and the digitized version was worse.  I know of no test of 
> their claims.  I can tell the difference between a CD and a vinyl 
> recording because the CD has no noise.  So to blind appropriately, one

> would have needed to digitize the noisy analog signal.

Would a reel-to-reel have as much signal noise? And why not just the
output source in a blank room with no way of seeing the originating
medium source of the sound?

 
> 
> > Just curious, as I like "live" concerts" and have noticed that the
> cd's
> > never catch the sound quality of the show (Moody Blues -Live at red
> rock
> > is the only example I can give, sorry).
> 
> I'm not sure of what you are getting at here.  A live concert is 
> played using a massive PA system and all kinds of speakers all over 
> the place in some giant room.  Later you use a little stereo system in

> a little room.
> Of course they sound different.

Which would invalidate any blind test unless you used an amphitheatre or
concert hall with said same speaker arrangement. As you said, of course
they will sound different.

> 
> 
> > As to the argument that this bit rate is better than that bit rate, 
> > I can only say 192 is fine for my car, as the road noise is working 
> > against the sound of the mp3 file, and the same goes for any 
> > portable listening device; that said, high bit rates would be a must

> > for archiving purposes.
> 
> Again, the question is "how do you know?"  You say that 192 is fine, 
> but what does that mean?  You haven't compared it to other bit rates 
> in any sort of controlled test.  Of course your main point is that the

> listening experience in the car is rather poor so that low bitrate 
> will be less noticable, and that has to be true -- I'm confident 
> enough that I wouldn't want to study it.  I'm mostly interested in 
> optimal listening conditions, but I'd also be interested in, say, what

> is bitrate is detectable as imperfect on an iPod.

But by limiting it to one device, you have introduced a bias into the
study, correct?  

As far as saying that 192 is fine, I am commenting that to my limited
perceptions and abilities, this is an adequate bitrate as compared to
the quality of a standard cd (WAV?) output.  The same would go for most
portable listening devices, as the output is limited by the output
device (speakers, headphones, etc) and to the environment that you are
listening in, be it the while you are jogging, biking, or driving.  Any
of these activities that utilize a portable device will invariably
introduce environmental noise in the background to degrade any music you
are listening to.  

But, now that we have defined the study, we can sum it up as this:

In a soundproofed room, what bitrate is optimal, as defined by an
overall majority of listeners, reproduced on a specific portable device,
that device being the Apple IPod using the stock headset that came with
the portable device.

Or this:

In a soundproofed room, what bitrate is detectable as imperfect, as
defined by an overall majority of listeners, reproduced on a specific
portable device, that device being the Apple IPod using the stock
headset that came with the portable device.



Is this correct?  I want to be sure of the parameters of this debate so
I will not be asked " what does that mean?" or accused of not following
the proper forms of the debate.  Thanks for the lively conversation  :)

Diana
 
ps I just love meetings...

> Mike
> 
> _______________________________________________
> discussion mailing list
> EMAIL:PROTECTED
> http://mlug.missouri.edu/mailman/listinfo/discussion

_______________________________________________
discussion mailing list
EMAIL:PROTECTED
http://mlug.missouri.edu/mailman/listinfo/discussion