S-expression: Difference between revisions
m →Sk/S(k + 2) (semiparticulars): convert to much more readable LaTeX |
m →Sk/S(k + 2) (semiparticulars): section, clarify |
||
| Line 889: | Line 889: | ||
== Sk/S(k + 2) (semiparticulars) == | == Sk/S(k + 2) (semiparticulars) == | ||
For differences between square-particulars of the form S | For differences between square-particulars of the form S''k''/S(''k'' + 2), the resulting comma is either [[superparticular]] or [[#Glossary|odd-particular]]. | ||
(This terminology also suggests [[#Glossary|throdd-particular]] and [[#Glossary|quodd-particular]] as generalizations.) | |||
In the below, we use S(''k'' - 1)/S(''k'' + 1) for symmetry around ''k'' to make the math visually simpler, but keep in mind it's equivalent to using an offset ''k''. | |||
Also keep in mind that ''k'' - ''a'' (for positive ''a'') is smaller than ''k'', so that ''k''/(''k'' - ''a'') > (''k'' + ''a'')/''k'' (because the former appears earlier in the harmonic series and is thus larger); this is an important and useful intuition to learn. | |||
=== Significance === | |||
Tempering S(''k'' - 1)/S(''k'' + 1) implies that (''k'' + 2)/(''k'' - 2) is divisible exactly into two halves of (''k'' + 1)/(''k'' - 1). It also implies that the intervals (''k'' + 2)/''k'' (=s) and ''k''/(''k'' - 2) (=L) are equidistant from (''k'' + 1)/(''k'' - 1) (=M) because to make them equidistant we need to temper: | Tempering S(''k'' - 1)/S(''k'' + 1) implies that (''k'' + 2)/(''k'' - 2) is divisible exactly into two halves of (''k'' + 1)/(''k'' - 1). It also implies that the intervals (''k'' + 2)/''k'' (=s) and ''k''/(''k'' - 2) (=L) are equidistant from (''k'' + 1)/(''k'' - 1) (=M) because to make them equidistant we need to temper: | ||
| Line 900: | Line 907: | ||
(''k''+2)/''k'', (''k''+1)/(''k''-1), ''k''/(''k''-2) | (''k''+2)/''k'', (''k''+1)/(''k''-1), ''k''/(''k''-2) | ||
Also note that in the above, (''k''+1)/(''k''-1) is the [[mediant]] of the adjacent two intervals, meaning that division of an interval into two via tempering a semiparticular is in some sense 'optimal' relative to the complexity. This also means that if ''k'' is a multiple of 2, this corresponds to a natural way to split the square superparticular S(''k''/2) into two parts. For example, if ''k'' = 10 then we have (10+2)/10, (10+1)/(10-1), 10/(10-2) as equidistant, which simplified is 6/5, 11/9, 5/4, with 11/9 being the mediant of 6/5 and 5/4, and therefore the corresponding superparticular S5 = (5/4)/(6/5) is split into two parts which are tempered together: (5/4)/(11/9) = 45/44 and (11/9)/(6/5) = 55/54. The semiparticular is therefore S(10-1)/S(10+1) = S9/S11 = 243/242 = (45/44)/(55/54) = ((10+2)/(10-2))/((10+1)/(10-1))<sup>2</sup>. | Also note that in the above, (''k''+1)/(''k''-1) is the [[mediant]] of the adjacent two intervals, meaning that division of an interval into two via tempering a semiparticular is in some sense 'optimal' relative to the complexity. This also means that if ''k'' is a multiple of 2 (such as in , this corresponds to a natural way to split the square superparticular S(''k''/2) into two parts. For example, if ''k'' = 10 then we have (10+2)/10, (10+1)/(10-1), 10/(10-2) as equidistant, which simplified is 6/5, 11/9, 5/4, with 11/9 being the mediant of 6/5 and 5/4, and therefore the corresponding superparticular S5 = (5/4)/(6/5) is split into two parts which are tempered together: (5/4)/(11/9) = 45/44 and (11/9)/(6/5) = 55/54. The semiparticular is therefore S(10-1)/S(10+1) = S9/S11 = 243/242 = (45/44)/(55/54) = ((10+2)/(10-2))/((10+1)/(10-1))<sup>2</sup>. | ||
This form of comma does not yet have an official name, but a proposed name is "semiparticular", because most of the time it is superparticular but less often it is odd-particular, and because when tempered they all cause an interval to be divided into two equal parts where each part is a (tempered version of a) superparticular or odd-particular, and the interval being divided in half is sometimes quodd-particular, sometimes odd-particular and sometimes superparticular. Specifically: | This form of comma does not yet have an official name, but a proposed name is "semiparticular", because most of the time it is superparticular but less often it is odd-particular, and because when tempered they all cause an interval to be divided into two equal parts where each part is a (tempered version of a) superparticular or odd-particular, and the interval being divided in half is sometimes quodd-particular, sometimes odd-particular and sometimes superparticular. Specifically: | ||
| Line 914: | Line 921: | ||
Also, the interval in the denominator of an expression of a semiparticular of the form (a/b)/(c/d)<sup>2</sup> is significant in that it has a special relationship: specifically, consider tempering (a/b)/(c/d)<sup>2</sup> so therefore the interval c/d is equal to the interval (a/b)/(c/d). This is significant because it allows the intuitive replacement of the two superparticulars composing a superparticular or odd-particular with the two superparticulars directly adjacent to them. For example, as 9/8 = 18/17 * 17/16 we can replace 18/17 with 19/18 and 17/16 with 16/15 by tempering S16/S18 = (19/15)/(9/8)<sup>2</sup> because we can multiply 9/8 by the tempered comma (19/15)/(9/8)<sup>2</sup> to get (19/15)/(9/8) = (19/18)(16/15) (because 9/8 = 18/16), or as 13/11 = 13/12 * 12/11 we can replace 13/12 with 14/13 and 12/11 with 11/10 by tempering S11/S13 = (7/5)/(13/11)<sup>2</sup> because we can multiply 13/11 by the tempered comma (7/5)/(13/11)<sup>2</sup> to get (7/5)/(13/11) = (14/13)(11/10) (because 7/5 = 14/10). Note we have to replace ''both'' intervals ''simultaneously'' as this is lower error, and note that if we want to be able to replace them individually we must pick the higher error route of tempering S16 and S18 or S11 and S13 individually (for which tempering the semiparticular is then an implied consequence). (The broader lesson is that you can rewrite exact JI equivalences with the commas you are tempering to find new interesting consequences of those commas.) | Also, the interval in the denominator of an expression of a semiparticular of the form (a/b)/(c/d)<sup>2</sup> is significant in that it has a special relationship: specifically, consider tempering (a/b)/(c/d)<sup>2</sup> so therefore the interval c/d is equal to the interval (a/b)/(c/d). This is significant because it allows the intuitive replacement of the two superparticulars composing a superparticular or odd-particular with the two superparticulars directly adjacent to them. For example, as 9/8 = 18/17 * 17/16 we can replace 18/17 with 19/18 and 17/16 with 16/15 by tempering S16/S18 = (19/15)/(9/8)<sup>2</sup> because we can multiply 9/8 by the tempered comma (19/15)/(9/8)<sup>2</sup> to get (19/15)/(9/8) = (19/18)(16/15) (because 9/8 = 18/16), or as 13/11 = 13/12 * 12/11 we can replace 13/12 with 14/13 and 12/11 with 11/10 by tempering S11/S13 = (7/5)/(13/11)<sup>2</sup> because we can multiply 13/11 by the tempered comma (7/5)/(13/11)<sup>2</sup> to get (7/5)/(13/11) = (14/13)(11/10) (because 7/5 = 14/10). Note we have to replace ''both'' intervals ''simultaneously'' as this is lower error, and note that if we want to be able to replace them individually we must pick the higher error route of tempering S16 and S18 or S11 and S13 individually (for which tempering the semiparticular is then an implied consequence). (The broader lesson is that you can rewrite exact JI equivalences with the commas you are tempering to find new interesting consequences of those commas.) | ||
=== Table of semiparticulars === | |||
Here follows a table of [[23-limit]] semiparticulars corresponding to square-particulars S''k'' for ''k'' < 96, plus all semiparticulars up to [[9801/9800|S33/S35 = S99]], an exceptional [[11-limit]] comma, plus all semiparticulars dividing [[superparticular interval]]s up to [[13/12]] (corresponding to the [[17-limit]] semiparticular [[31213/31212|S49/S51]]) for completeness. This table also shows all semiparticulars corresponding to splitting an [[#Glossary|odd-particular]] in two up to [[17/15]] (although a common strategy is to temper the square-particular that is the difference between the two superparticular intervals the odd-particular is composed of instead). The bound ''k'' < 96 was chosen as it corresponds to another remarkable semiparticular [[123201/123200|S78/S80 = S351]]. Perhaps many of the patterns will become clearer if you examine this table: | Here follows a table of [[23-limit]] semiparticulars corresponding to square-particulars S''k'' for ''k'' < 96, plus all semiparticulars up to [[9801/9800|S33/S35 = S99]], an exceptional [[11-limit]] comma, plus all semiparticulars dividing [[superparticular interval]]s up to [[13/12]] (corresponding to the [[17-limit]] semiparticular [[31213/31212|S49/S51]]) for completeness. This table also shows all semiparticulars corresponding to splitting an [[#Glossary|odd-particular]] in two up to [[17/15]] (although a common strategy is to temper the square-particular that is the difference between the two superparticular intervals the odd-particular is composed of instead). The bound ''k'' < 96 was chosen as it corresponds to another remarkable semiparticular [[123201/123200|S78/S80 = S351]]. Perhaps many of the patterns will become clearer if you examine this table: | ||