[78-L] Removal of hiss on 78 transfers

[Jeff Lichtman]

>  What I am suggesting is that the groove of a record is scanned 
> laterally or radially, however you want to describe it.  Assuming 
> the results of this scan to be shown on a computer screen, what you 
> should see is a "V" in the shape of the cutting stylus.  If
>  the actual shape of the cutting stylus is known, the scan can be 
> compared to it and only the parts of the scan which coincide with 
> the reference stylus are considered valid and other parts of the 
> scan are disregarded.

The system in question does do a scan of the playing surface. For 
lateral-cut records it does a two-dimensional scan and uses 
differences in reflectivity to isolate the sides of the groove. What 
you get is an undulating white stripe with a black background. The 
system interprets lateral displacement of the stripe as if it were 
stylus motion.

For vertical-cut records the system uses a clever means to determine 
groove depth - it uses a lens specially designed to have a large 
amount of chromatic aberration. This aberration causes different 
wavelengths to focus at different distances from the lens (it causes 
color fringes, and is usually considered a flaw in a lens). The 
system uses the different colors to construct a three-dimensional 
picture of the groove and interprets vertical displacement of the 
groove bottom as if it were stylus motion.

The problem I see with your idea is the main problem with all noise 
reduction systems: how do you distinguish between signal and noise? 
In your proposal, when the system is interpreting a noisy section of 
groove, how is it to tell which undulations to trace and which not 
to? It's not enough to say that you would use the shape of the 
original cutting stylus - when the system comes to a lump or pit on 
the groove wall, it has to decide whether to trace this as if it were 
signal or skip past it as if it were noise. Imagine, for example, 
that the system encounters a widening in the groove that *could* 
indicate impulse noise. Which side of the groove should it trace? 
There could be a flaw on either side of the groove, and if you pick 
the wrong side you're eliminating signal instead of noise.

I imagine that you could build some intelligence into the system to 
recognize noise simply by what it looks like. A noisy bit of groove 
might have an irregular surface, for example. It would be interesting 
to do research in this area, but I don't think the problem is at all easy.


                        -        Jeff Lichtman
                                 jeff at swazoo.com
                                 Check out Swazoo Koolak's Web Jukebox at
                                 http://swazoo.com/