[geeks] Working Music vs Thinking Music

dave at cca.org dave at cca.org
Mon Jun 24 20:40:30 CDT 2002


jdboyd at cs.millersville.edu writes:

>OK, perhaps I'm just buying some peoples story, but I know that I have
>a number of classical records that despite a small amount of fuzz
>sound better than any of the classical CDs I have.  The record player
>was good, but it wasn't extremely expensive or anything, neither was
>any other part of the stereo I was doing the comparison on.  But,
>then, I was unable to find the same performance recorded on both
>mediums, but I was at least comparing the same musical pieces from the
>same company (dutch gramaphone).

>I haven't bothered trying to compare any rock I like on both vinyl and
>CD. And I never listen to those classical records because they are too
>much hassle.

You get into murky water here though, since cutting the master
for vinyl is a very tricky process, it was done by professionals
with incredibly good ears for subtleties. CDs are cut by companies
that think there's nothing to it more than "cp audio-file /dev/cd".
Lots of horrible mistakes are made in CD production that aren't inherent
to the technology.

>But from the physics perspective of how we were told sound worked, it
>made sense that increasing the bit depth would help for classical
>music were due to some parts being very loud, all the quite passages
>are limited to only 4 bits worth of precision.  Or something like
>that.  I'm not taking the time to pull up a bunch of classical CDs to
>see if that ratio actually works out that way, although it would be
>interesting to.

True. More bits will give you better dynamic range. But the question
is how much dynamic range can the human ear pick up? How many bits
do you actually need? 

If you can - generate an audio CD with a sequence of sine waves, 1 bit
high, 2 bits high, 3 bits high, etc. See which ones you can hear,
and if you can hear the difference, etc. (This might also show off
limitations or bad design in the DACs.)

As an aside - it is good to record and do studio work at higher
resolution than the final product, to avoid roundoff errors. So
24 bits in the studio makes sense.

>> And of course, at the high end, it simply *can't* be about the audio
>> any more. When people are spending more on their stereo to listen
>> to classical music than it would cost to hire a string quartet to
>> play in their living room once a week, something is wrong.

>I don't pay enough attention to the high end to really know.  But I
>must say, I was extremely impressed with the sounds coming from one
>friends multi-thousand dollar home stereo compared to what comes from
>most peoples multi $100 stereos.  That was some sweet sounding gear.

Ah yes - a few thousand dollars will get you a really nice sounding
system. That's sane. When I say "audiophile" I'm thinking of the
$100,000 and up systems.

------ David Fischer ------- dave at cca.org ------- http://www.cca.org ------
----- Being poked in the eye with a sharp stick makes baby jesus cry! -----



More information about the geeks mailing list