Equal-step tuning: Difference between revisions

Contribution (talk | contribs)
Split
 
(35 intermediate revisions by 2 users not shown)
Line 1: Line 1:
{{interwiki
{{interwiki
| de = Gleichstufige_Tonsysteme
| de = Gleichstufige_Tonsysteme
| en = Equal-step_tuning
| en = Equal-step tuning
| ja = 平均律
| ja = 平均律
}}
}}
Line 79: Line 79:
* [[Ed8]] (… of the 8th harmonic)
* [[Ed8]] (… of the 8th harmonic)
* [[Ed12]] (… of the 12th harmonic)
* [[Ed12]] (… of the 12th harmonic)


=== Equal multiplications ===
=== Equal multiplications ===
Line 117: Line 118:


Other edonoi contain no approximation of an octave or a compound octave (at least, not for a while), and continue generating new tones as they continue upward or downward. Such scales lack a very familiar compositional redundancy, that of [[octave equivalence]] – this might necessitate special attention.
Other edonoi contain no approximation of an octave or a compound octave (at least, not for a while), and continue generating new tones as they continue upward or downward. Such scales lack a very familiar compositional redundancy, that of [[octave equivalence]] – this might necessitate special attention.
== Alpha-beta-gamma family ==
Wendy Carlos invented in the 1980's the Alpha, Beta, and Gamma scales. These are scales that divides 3/2 with steps very near to the successive superparticular complementary pair folding in 3/2, namely 6/5 and 5/4. The happy equal divisions are 9ed3/2, 11ed3/2, and 20ed3/2. However, these scales belong to a much vaster family, where also applies the same principle of divisions of a ratio with steps very near to the successive superparticular complementary pair folding into it. Below is a table showing the first members of this family:
{| class="wikitable"
|+The Alpha-Beta-Gamma family
! colspan="4" |Tuning
! colspan="3" |Intervals
! rowspan="2" |Comment
|-
!Equal division
!Type
!Cents
per steps
!Steps
per octave
!Ratio divided
!Successive superparticular
complementary pair folding
in the ratio divided
!Approximation
in cents of these
three intervals
|-
|[[3edt|3ed3/1]]
|Alpha
|633.985
|1.893
| rowspan="3" |3/1
| rowspan="3" |3/2, 2/1
|0, -67.970, 67.970
| rowspan="2" |Too much off to be useful
|-
|[[5edt|5ed3/1]]
|Beta
|380.391
|3.155
|0, 58.827, -58.827
|-
|[[8edt|8ed3/1]]
|Gamma
|237.744
|5.047
|0, 11.278, -11.278
| rowspan="3" |Pairs are off but still recognizable
|-
|[[5edo|5ed2/1]]
|Alpha
|240
|5
| rowspan="3" |2/1
| rowspan="3" |4/3, 3/2
|0, -18.045, 18.045
|-
|[[7edo|7ed2/1]]
|Beta
|171.429
|7
|0, 16.241, -16.241
|-
|[[12edo|12ed2/1]]
|Gamma
|100
|12
|0, 1.955, -1.955
| rowspan="13" |Happy divisions musically useful
|-
|[[7ed5/3]]
|Alpha
|126.337
|9.498
| rowspan="3" |5/3
| rowspan="3" |5/4, 4/3
|0, -7.303, 7.303
|-
|[[9ed5/3]]
|Beta
|98.262
|12.212
|0, 6.735, -6.735
|-
|[[16ed5/3]]
|Gamma
|55.272
|21.711
|0, 0.593, -0.593
|-
|[[9edf|9ed3/2]]
|Alpha
|77.995
|15.386
| rowspan="3" |3/2
| rowspan="3" |6/5, 5/4
|0, -3.661, 3.661
|-
|[[11edf|11ed3/2]]
|Beta
|63.814
|18.805
|0, 3.429, -3.429
|-
|[[20edf|20ed3/2]]
|Gamma
|35.098
|34.190
|0, 0.238, -0.238
|-
|[[11ed7/5]]
|Alpha
|52.956
|22.660
| rowspan="3" |7/5
| rowspan="3" |7/6, 6/5
|0, -2.093, 2.093
|-
|[[13ed7/5]]
|Beta
|44.809
|26.781
|0, 1.981, -1.981
|-
|[[24ed7/5]]
|Gamma
|24.271
|49.441
|0, 0.114, -0.114
|-
|[[13ed4/3]]
|Alpha
|38.311
|31.322
| rowspan="3" |4/3
| rowspan="3" |8/7, 7/6
|0, -1.307, 1.307
|-
|[[15ed4/3]]
|Beta
|33.203
|36.141
|0, 1.247, -1.247
|-
|[[28ed4/3]]
|Gamma
|17.787
|67.464
|0, 0.061, -0.061
|}
As you can see, some patterns appear:
* For each pair of superparticular ratios <math>\frac{s+1}{s}</math> and <math>\frac{s}{s-1}</math>, there exists a ratio <math>\frac{s+1}{s-1}</math> such that <math>\frac{s+1}{s}</math> and <math>\frac{s}{s-1}</math> are <math>\frac{s+1}{s-1}</math> complementary; although <math>(s+1)-(s-1)=2</math>, when simplified as a coprime ratio with <math>\frac{s+1}{s-1}=\frac{a}{b}</math>, it is observed that <math>a-b\in{1,2}</math>. In other words, for each ratio <math>\frac{a}{b}</math> where <math>a-b\in{1,2}</math>, there exists a pair of superparticular ratios <math>\frac{s+1}{s}</math> and <math>\frac{s}{s-1}</math> that are <math>\frac{a}{b}</math>-complementary.
* We can observe a converging sequence and pattern for divisions of the ratio <math>\frac{s+1}{s-1}</math> where low errors appears for <math>\frac{s+1}{s}</math> and <math>\frac{s}{s-1}</math>: 3, 5, 8; then 5, 7, 12; then 7, 9, 16; then 9, 11, 20; then 11, 13, 24; then 13, 15, 28; etc. -
{| class="wikitable"
|+Converging sequence and pattern
! rowspan="2" |<math>{s1}</math>
! rowspan="2" |<math>{s2}</math>
! rowspan="2" |<math>{s3}</math>
! rowspan="2" |a-b
! rowspan="2" |Ratio divided
! colspan="2" |SSCP
! colspan="3" |Number of divisions
|-
!Small
!Big
!Alpha
!Beta
!Gamma
|-
|3
|2
|1
|2
|3/1
|3/2
|3/1
|3
|5
|8
|-
|4
|3
|2
|1
|2/1
|4/3
|3/2
|5
|7
|12
|-
|5
|4
|3
|2
|5/3
|5/4
|4/3
|7
|9
|16
|-
|6
|5
|4
|1
|3/2
|6/5
|5/4
|9
|11
|20
|-
|7
|6
|5
|2
|7/5
|7/6
|6/5
|11
|13
|24
|-
|8
|7
|6
|1
|4/3
|8/7
|7/6
|13
|15
|28
|}
* We can describe the pattern as follow: Alpha division = <math>2\times {s2} - 1</math>; Beta division = <math>2\times {s2} + 1</math>; Gamma division = Alpha division + Beta division
* Alpha types flatten the smaller interval and sharpen the larger; Beta types do the reverse; Gamma types again flatten the smaller and sharpen the larger.
* The Alpha, Beta, and Gamma types bring their interval pairs increasingly close to just intonation.


== See also ==  
== See also ==  
Line 371: Line 132:


[[Category:Equal-step tuning| ]] <!-- main article -->
[[Category:Equal-step tuning| ]] <!-- main article -->
<!-- main article -->
[[Category:Terms]]
[[Category:Terms]]
[[Category:Acronyms]]
[[Category:Acronyms]]
[[Category:Tuning]]
[[Category:Tuning]]