Re: [ARSCLIST] Dynamic-frequency Range

[Mike Richter <mrichter@xxxxxxx>]

Karl Miller wrote:

> I am concerned with what was available to the "average" listener.
>
> Perhaps I should explain the reasons for my questions. I am writing an
> essay about pianists who refused to make disc recordings but were willing
> to make reproducing piano roll recordings. There were problems with
> available duration for each side, in addition to the obvious limitations
> of frequency response. I find that rarely is the question of dynamic range
> mentioned. There is a quote from one pianist who was told the number of
> amplitude steps available from a particular reproducing piano...I can't
> remember how many, perhaps just 24 and he replied, well I have 25 dynamic
> steps in my playing. While shading is probably subject to one's
> ability to measure it, it seems to me that ultimately that "arguement" was
> not informed as their dynamic range on discs was signficantly limited,
> especially in the acoustic era.
>
> Further, some pianists pointed out that they did not like having to make
> adjustments to their dynamics in particular parts of the frequency range,
> adjustments that were needed to provide more balance on playback.
>
> In short, I am trying to put into slightly more quantifiable terms, the
> reasons for their objects.
>
> I am reminded of one pianist who refused to make any disc
> recordings...yet, if she had, she might have had the backing of Victor
> and, at least from my perspective, been as well known as some of the
> artists which had the benefit of the marketing of a record company.
>
> Any thoughts would be welcome.
>
> Karl

You offer much opportunity for thought and experimentation, but I see 
little chance that answers will be found otherwise. As a side note, you 
seem to assume that recording in the acoustic era was seen by the 
performers as a benefit beyond immediate payment. For some, it may have 
been, but I'm afraid that most took positions not unlike that of the 
"Maitre de musique" in the Belgian TV film (mistranslated in English as 
"The Music Teacher").

The dynamics of the various forms of piano rolls can be determined from 
the design of the player or the properties of the recording. Since the 
typical user would not have a Vorsetzer, the dynamic range was on/off. 
In fact, you are not looking at dynamic range per se but at its 
quantization, where the player piano offered a single bit. I believe you 
want to determine how many bits were available for the more elaborate 
reproducers, which as I say should be implicit in the design.

Of course, the frequency range of the reproducing piano was that of the 
piano itself as covered by the reproducer - nominally, the full range of 
the instrument.

Quantization of the acoustic recording is in practice related to the SNR 
(Signal-to-Noise Ratio), which is closely enough related to dynamic 
range to serve most purposes. Saturation, blasting and related 
distortions do come into play and I'm afraid an expert on reproduction 
of "78"-rpm discs will be needed to deal with those issues, especially 
when combined with stylus size and shape. (Many a recording from that 
era which is unlistenable on one setup will be crystalline on another.)

You may find the little experiment on Hugues Cuenod recordings at
  http://www.operas-are.us/
of interest though it deals with voice not primarily piano. If you are 
not familiar with the Nellie Melba distance tests of 1910, you will find 
them worth seeking for issues of dynamics when varying the distance from 
the source (here, a soprano) to the horn. Before the microphone was 
introduced, the size of the piano limited the acoustic available - for 
better or for worse.

Mike
--
mrichter@xxxxxxx
http://www.mrichter.com/