[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ARSCLIST] historical stylus rake angles (SRA)
Pete Tinker wrote:
> My (admittedly limited) experience has convinced me that no algorithms
> exist that will eliminate the noise. In this instance, noise has
> usually replaced the signal (the recorded sound), and any computational
> approach to restoring the signal would have to create "new" signal by
> trying to infer the original signal from existing signal surrounding the
> noise. Algorithms are typically only useful when you have to extract
> signal from a combination of signal+noise.
Noise is caused by geometric imperfections in the 3-D groove of the
record. These imperfections are from various sources: 1) noise in the
original signal fed to the cutter, 2) noise introduced by the whole
mastering process, 3) granularities in the substrate formulation of
the record (like the abrasive added to shellac, which leaves bumps and
pits), and 4) damage to the groove from playback. (I think 3 and 4 are
the biggies for cylinder and shellac records.)
The problem with present playback systems is that they drag something
through the groove, usually a stylus, and this "integrates" whatever
is encountered in the groove. A lot of fine 3-D detail is lost by this
It's a lot like taking a picture with a low resolution digital camera
-- each pixel may "see" detail, but that detail must be averaged out
to a single color. Once averaged out, that detailed information is
lost forever and cannot be recovered.
(For those with a thermodynamics background, or a signal processing
background, we are talking about an increase in entropy.)
Separating out the "noise", once it is integrated into the signal, is
much more difficult, and there are theoretical limits as to how much
noise can be removed. (Many noise reduction systems simply mask the
noise based on psychoacoustic principles.)
However, if we are able to capture the full 3-D information of the
groove (ultra-microtopography), it is theoretically possible to use a
computer algorithm to discern the various anomalies in the groove
geometry: "potholes" can be filled in, "boulders" sanded down, etc.
This will result in determining what the groove shape "should be",
and thus do a much better job at restoration.
This is essentially what Dr. Carl Haber, et al, at Lawrence Berkeley
National Laboratory are researching (see: http://www-cdf.lbl.gov/~av/ )
The current problem is that gathering microtopographic data of
sufficiently fine resolution is *very* time-consuming, and the whole
process is still quite experimental. However, I believe that with
improvements in technology, and with a few clever ideas along the way,
it will eventually be possible to do high-quality restoration using
microtopographic techniques which will far surpass what is possible
today with dragging a stylus through a groove. Haber is making
progress, but it will take time.
So don't throw away those broken records!