Kite's thoughts on pergens: Difference between revisions

Wikispaces>TallKite
**Imported revision 625206059 - Original comment: **
Wikispaces>TallKite
**Imported revision 625221961 - Original comment: **
Line 1: Line 1:
<h2>IMPORTED REVISION FROM WIKISPACES</h2>
<h2>IMPORTED REVISION FROM WIKISPACES</h2>
This is an imported revision from Wikispaces. The revision metadata is included below for reference:<br>
This is an imported revision from Wikispaces. The revision metadata is included below for reference:<br>
: This revision was by author [[User:TallKite|TallKite]] and made on <tt>2018-01-22 13:54:36 UTC</tt>.<br>
: This revision was by author [[User:TallKite|TallKite]] and made on <tt>2018-01-22 20:23:25 UTC</tt>.<br>
: The original revision id was <tt>625206059</tt>.<br>
: The original revision id was <tt>625221961</tt>.<br>
: The revision comment was: <tt></tt><br>
: The revision comment was: <tt></tt><br>
The revision contents are below, presented both in the original Wikispaces Wikitext format, and in HTML exactly as Wikispaces rendered it.<br>
The revision contents are below, presented both in the original Wikispaces Wikitext format, and in HTML exactly as Wikispaces rendered it.<br>
Line 568: Line 568:
How many edos support a given pergen? In general, an infinite number. For (P8/m, M/n) to be supported by N-edo, N must be divisible by m, and k by n, where k is the multigen's N-edo keyspan. To be fully supported, N/m and k/n must be coprime.
How many edos support a given pergen? In general, an infinite number. For (P8/m, M/n) to be supported by N-edo, N must be divisible by m, and k by n, where k is the multigen's N-edo keyspan. To be fully supported, N/m and k/n must be coprime.


Given an edo, a period, and a generator, what is the pergen? There is more than one right answer. For 12edo, if P = 12\12 and G = 1\12 (one semitone), it could be either (P8, P4/5) or (P8, P5/7).  
Given an edo, a period, and a generator, what is the pergen? There is more than one right answer. For 12edo, if P = 12\12 and G = 1\12 (one semitone), it could be either (P8, P4/5) or (P8, P5/7).


For N-edo, if P = p\N and G = g\N, ...
For N-edo, if P = p\N and G = g\N, ...
Line 698: Line 698:
Therefore assume m &gt; |b| and unreduce
Therefore assume m &gt; |b| and unreduce
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn)
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn)
Simplify by dividing by b to get (P8/m, (n/b - a(m/b), -m) / m(n/b)) = (P8/m, (a',b')/n')
Let p = m/b and q = n/b, p and q are coprime, |p| &gt; 1 and |q| &gt; 1
Simplify by dividing by b to get (P8/m, (q - ap, -m) / qm) = (P8/m, (a',b')/n')
Can b' be reduced by simplifying further?
No, because GCD (a', b') = GCD (q - ap, -bp) = r
Since a and b are coprime, GCD (-ap, -bp) = p
Since GCD (p, q) = 1
b' = -m, therefore m = |b'|, and the unreduced pergen is explicitly false
b' = -m, therefore m = |b'|, and the unreduced pergen is explicitly false
Therefore the original pergen is a false double
Therefore the original pergen is a false double
Line 707: Line 712:
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn) = (P8/m, (a',b')/n')
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn) = (P8/m, (a',b')/n')
Simplify using b = GCD (m,n): a' = (n-am)/b, b' = -m, and n' = mn/b
Simplify using b = GCD (m,n): a' = (n-am)/b, b' = -m, and n' = mn/b
Can b' be reduced by simplifying further?
No, because GCD (a', b', n') = GCD (n/b - am/b, -m, mn/b)
GCD (b', n') = m
GCD (n/b, m) = 1
GCD (
|b'| = m, so the unreduced pergen is explicitly false, and the test works
|b'| = m, so the unreduced pergen is explicitly false, and the test works


Line 3,584: Line 3,584:
How many edos support a given pergen? In general, an infinite number. For (P8/m, M/n) to be supported by N-edo, N must be divisible by m, and k by n, where k is the multigen's N-edo keyspan. To be fully supported, N/m and k/n must be coprime.&lt;br /&gt;
How many edos support a given pergen? In general, an infinite number. For (P8/m, M/n) to be supported by N-edo, N must be divisible by m, and k by n, where k is the multigen's N-edo keyspan. To be fully supported, N/m and k/n must be coprime.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
Given an edo, a period, and a generator, what is the pergen? There is more than one right answer. For 12edo, if P = 12\12 and G = 1\12 (one semitone), it could be either (P8, P4/5) or (P8, P5/7). &lt;br /&gt;
Given an edo, a period, and a generator, what is the pergen? There is more than one right answer. For 12edo, if P = 12\12 and G = 1\12 (one semitone), it could be either (P8, P4/5) or (P8, P5/7).&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
For N-edo, if P = p\N and G = g\N, ...&lt;br /&gt;
For N-edo, if P = p\N and G = g\N, ...&lt;br /&gt;
Line 3,958: Line 3,958:
Therefore assume m &amp;gt; |b| and unreduce&lt;br /&gt;
Therefore assume m &amp;gt; |b| and unreduce&lt;br /&gt;
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn)&lt;br /&gt;
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn)&lt;br /&gt;
Simplify by dividing by b to get (P8/m, (n/b - a(m/b), -m) / m(n/b)) = (P8/m, (a',b')/n')&lt;br /&gt;
Let p = m/b and q = n/b, p and q are coprime, |p| &amp;gt; 1 and |q| &amp;gt; 1&lt;br /&gt;
Simplify by dividing by b to get (P8/m, (q - ap, -m) / qm) = (P8/m, (a',b')/n')&lt;br /&gt;
Can b' be reduced by simplifying further?&lt;br /&gt;
No, because GCD (a', b') = GCD (q - ap, -bp) = r&lt;br /&gt;
Since a and b are coprime, GCD (-ap, -bp) = p&lt;br /&gt;
Since GCD (p, q) = 1&lt;br /&gt;
b' = -m, therefore m = |b'|, and the unreduced pergen is explicitly false&lt;br /&gt;
b' = -m, therefore m = |b'|, and the unreduced pergen is explicitly false&lt;br /&gt;
Therefore the original pergen is a false double&lt;br /&gt;
Therefore the original pergen is a false double&lt;br /&gt;
Line 3,967: Line 3,972:
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn) = (P8/m, (a',b')/n')&lt;br /&gt;
(P8/m, (a,b)/n) unreduced is (P8/m, (n-am, -bm) / mn) = (P8/m, (a',b')/n')&lt;br /&gt;
Simplify using b = GCD (m,n): a' = (n-am)/b, b' = -m, and n' = mn/b&lt;br /&gt;
Simplify using b = GCD (m,n): a' = (n-am)/b, b' = -m, and n' = mn/b&lt;br /&gt;
Can b' be reduced by simplifying further?&lt;br /&gt;
No, because GCD (a', b', n') = GCD (n/b - am/b, -m, mn/b)&lt;br /&gt;
GCD (b', n') = m&lt;br /&gt;
GCD (n/b, m) = 1&lt;br /&gt;
GCD (&lt;br /&gt;
|b'| = m, so the unreduced pergen is explicitly false, and the test works&lt;br /&gt;
|b'| = m, so the unreduced pergen is explicitly false, and the test works&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;