Tunelab ver 2.

[Frank Weston]

----------
> From: John Woodrow <woodroj@syvax.email.dupont.com>
> To: pianotech@ptg.org
> Subject: Re: Tunelab ver 2.
> Date: Thursday, March 19, 1998 1:12 AM

> Frank and list,
> 
> Thanks for the great summary.  A couple of questions to clarify some
points:
> 
> 
> How do you set up your "Graphing Options" menu?  Do you (a) for clarity
and simplicity constantly select and change options depending on the
partials and octaves you are working on, or (b) just 'set and forget' the
graph options? If (b) what options do you select?

I select options depending on which partials and octaves I'm working on. 
A0 will be tuned referenced to something like the 8th partial, which will
be matched to a double or triple octave, whereas A5 will be tuned
referenced to the fundamental with a match to the 2nd partial of A4.  As a
general rule simple options and matches work above A3.  Below A3, anything
goes.  What ever selections result in a stretch curve that looks right,
also generally work well for the ear.
> 
> 
> How do you smooth the tuning curve, (a) 100% on the graph editor, or (b)
return back to the Numerical Editor tuning to create straight line or
second degree equations?  Why I ask is that I don't find the Graph editor
as good as the 'old' numerical editor for quickly smoothing a line between
2 or 3 points.  Just wanted to make sure I wasn't missing something.

I have used both the graphic editor and the numerical editor to smooth the
curve, but as I gain experience, I find I can get the job done fastest with
just the graphic editor.  I have suggested to Bob Scott that some curve
smoothing functions in the graphic editor would be helpful - maybe in the
next version.

> 
> 
> When the HT offsets are applied to the ET tuning curve, they  appear to
be applied in these same ratios across all octaves.  As the actual
intervals between notes in the octaves will be slightly different because
of the different stretch in each octave, would not each octave have an ever
so slightly different temperament?  Do you consider this a problem that has
to be fine tuned out?  I have been wondering this when using a single
octave HT offset table for any ETD, not just Tunelab.

If I understand your question, what you are asking is:  If, because stretch
over an octave is not precisely linear, will the application of stretch in
linear fashion onto the intervals within that octave  alter the desired
relationship between those intervals significantly in musical terms?   My
answer is that stretch is not necessarily applied linearly when a tuning is
developed, and in fact it is usually not.  Even if it were, the effect
would be insignificant. 

Frank Weston