eminently practical!

> From: PERLICH@...
> Subject: Re: You too can calculate harmonic entropy!
>
> Or if you don't want to, I can just send you the values! Or put them in a
> file, like Manuel has already done.
>

Well, that would get it into my program in a jiffy. If you had it
in excel or a text file, it would be a cinch.

If its a big effort, then don't make it for my sake. I have found
how to control the "accuracy" of the "assumed listener" in my
scale-minimal-complexity-searching-program. The main thing
that would be of interest is if there are major differences
between the models.

Of course, it would probably be nice to have a table of the
values on Yahoo for the occasional wanderer who wants to use
the results without generating them.

thanks,

Bob Valentine

Hi Robert,

Since you include local mimima for rather complex ratios, and you're looking
at scales in an octave-equivalent sense, I decided to give you the data for
this graph:

/harmonic_entropy/files/dyadic/partch.bmp

The data is here:

/harmonic_entropy/files/dyadic/partch.txt

The first column is cents, the second column is entropy.

It only goes up to 600 cents, as thereafter you only get the same interval
classes in reverse, under octave-equivalence.

Let me know if you want the data for a different graph.

>
> 1. RE: eminently practical!
> From: "Paul H. Erlich" <PERLICH@...>
> Hi Robert,
>
> Since you include local mimima for rather complex ratios, and you're looking
> at scales in an octave-equivalent sense, I decided to give you the data for
> this graph:

Excellent! I've got a long weekend coming up and will plug in this data.

>
> /harmonic_entropy/files/dyadic/partch.bmp
>
> The data is here:
>
> /harmonic_entropy/files/dyadic/partch.txt
>
> The first column is cents, the second column is entropy.
>
> It only goes up to 600 cents, as thereafter you only get the same interval
> classes in reverse, under octave-equivalence.
>

But a complexity formula like a simple product does differentiate between
6/5 and 5/3, etc... Does that distinction go away due to summing or is
the complexity formula completely different?

thanks,

Bob

>But a complexity formula like a simple product does differentiate between
>6/5 and 5/3, etc...

So does regular, non-octave-equivalent harmonic entropy.

Does that distinction go away due to summing or is
>the complexity formula completely different?

Hi Robert,

I gave you the octave-equivalent results because I understand you are
looking at scales in an octave-equivalent sense; i.e., the interval between
D and F is just as much 6/5 as 5/3, 12/5, 10/3, etc. In a sense, the
octave-equivalent formulation takes _all_ of these intervals into account,
and the ratios in their neighborhoods, when computing a HE value for 316
cents.