Re: [ARSCLIST] Sound card recommendations

[Jon Noring <jon@xxxxxxxxxxx>]

Steven C. Barr wrote:
> Mike Richter wrote:
>> steven c wrote:

>>> Obviously, it is in the interest of society (whether society realizes
>>> this or not is open to question) to preserve as much of the total
>>> of historic sound recordings (of any nature) as is feasible. Each
>>> historic sound recording contains information, and the more
>>> information we are able to preserve for our posterity the better
>>> off that posterity will be!

>> Let us accept as true what you assert to be obvious. I will postulate 
>> that higher quality of preservation - more bits, channels, 
>> documentation, etc. - takes more resources than less. Then it follows 
>> that one must choose a balance between quantity of material preserved 
>> and quality of preservation. (Let's discuss separately the desirability 
>> of higher quality for material deemed more valuable or more important.)
>> 
>> Given finite resources and ever-expanding source material, definitions 
>> of 'good enough' for various purposes seem appropriate. Otherwise, 
>> everyone who is not generating material for archiving will need to be an 
>> archivist.

> Well, it seems logical to me that the quality of the preservation of a
> sound recording is effectively determined by its content! Thus, there
> is no reason to preserve stereo versions of mono originals, or to use
> more bandwidth than was recorded originally (i.e. preservations of
> acoustic 78's flat to 20KHz). As well, if the recording is "spoken
> word," the frequency range could be limited without loss of
> information and/or intelligibility. This also covers the question
> of "higher quality for material deemed more...important"...keep in
> mind that a preservation can never be anymore accurate than was
> the original recording (unless computer algorithms become much
> more powerful!)

The issue we face is that indeed what seems like "overkill" may not
necessarily be overkill when it comes to the digital preservation of
older sound recordings.

For example, preserving the two channels in a stereo pickup playing a
mono recording gives the opportunity for improved sound restoration,
such as when one groove is more noisy than the other. Even a mono
phonograph record has *two* sides to the groove.

It also appears potentially beneficial to digitize the "noise" (both
hiss and impulse which have quite high frequency components) at a
very high sampling rate so the noise is accurately represented. This
aids in noise removal/reduction/masking algorithms. A quite high
sampling rate may also be beneficial for better preserving the wanted
audio signal.

And who knows? In a few recordings there just might be some strange
higher frequency stuff introduced during mastering which could have
some forensics-type value. We'd like to preserve that as well, just in
case it is there. (I wonder, has anyone looked for oddball stuff in
older electrically-recorded 78's in the 15kHz to 25kHz region which
may be buried in the high frequency noise? There might be some really
interesting stuff awaiting to be discovered. Maybe sounds from the
beyond: EVP <laugh/>)

Just my $0.02 worth.

Jon Noring