View Single Post
Old 16th Apr 2018, 4:59 pm   #7
bluepilot
Heptode
 
Join Date: Feb 2010
Location: Duffort, Gers, France
Posts: 714
Default Re: Big Band records - plummy sound

Quote:
Originally Posted by mark_in_manc View Post
That's a neat trick, noise being noise. I wonder how deterministic surface 'noise' is - that is, whether with multiple plays of one pressing one could average out the random content. Otherwise, one might need to collect large numbers of pressings of the same recording...or possess a clean version of the signal with which one could train an adaptive filter, which rather defeats the object
I think you'll find that the noise is "recorded" on the record in the granular nature of the record material and wear produced by steel needles. If you play the same record several times you'll get the same noise. I suppose you could superimpose several copies of the same record but in practice synchronizing them would be virtually impossible.

Clickrepair (well actually Denoise) works by taking a sample of the noise somewhere where there is no music. Then it determines the spectrum and removes that spectrum of signal from the rest of the record. I assume the other processors do something similar.

I find that often the noise changes throughout the record as the steel needles used to play them originally wore down during playing. If you want to do things really properly you need to sample the noise in the run in and run out. Then process the record twice, once with each noise spectrum. Then splice the start of one to the end of the other.
__________________
Stuart

The golden age is always yesterday - Asa Briggs
bluepilot is offline