Würschmidt family: Difference between revisions

m Units; misc. cleanup; recategorize
Switch to Sintel's badness, WE & CWE tunings
 
(2 intermediate revisions by the same user not shown)
Line 1: Line 1:
{{Technical data page}}
{{Technical data page}}
The [[5-limit]] parent comma for the '''würschmidt family''' (würschmidt is sometimes spelled '''wuerschmidt''') is [[393216/390625]], known as Würschmidt's comma, and named after José Würschmidt. The [[generator]] is a classic major third, and to get to the interval class of fifths requires eight of these. In fact, (5/4)<sup>8</sup> × 393216/390625 = 6.  
The '''würschmidt family''' (würschmidt is sometimes spelled '''wuerschmidt''') of [[regular temperament|temperaments]] [[tempering out|tempers out]] [[393216/390625]], known as Würschmidt's comma, and named after José Würschmidt.  


Similar to [[meantone]], würschmidt implies that 3/2 will be tempered flat and/or 5/4 will be tempered sharp, and therefore 6/5 will be tempered flat. Unlike meantone, it is far more accurate. Combining würschmidt with meantone gives [[31edo]] as the first practical tuning with a generator of 10\31, but increasingly good 5-limit edo generators are [[34edo|11\34]] and especially [[65edo|21\65]], which notably is the point where it is combined with [[schismic]]/[[nestoria]] and [[gravity]]/[[larry]]. Other edo tunings include [[96edo]], [[99edo]] and [[164edo]].
== Würschmidt ==
{{Main| Würschmidt }}


Another tuning solution is to sharpen the major third by 1/8 of a Würschmidt comma, which is to say by 1.43 cents, and thereby achieve pure fifths; this is the [[minimax tuning]].  
The [[generator]] of würschmidt is a [[5/4|classical major third]], and to get to the interval class of the [[3/1|3rd harmonic]] requires eight of these. In fact, (5/4)<sup>8</sup> × 393216/390625 = 6.  


[[Mos scale]]s may not be the best approach for würschmidt since they are even more extreme than those of [[magic]]. [[Proper]] scales do not appear until 28, 31 or even 34 notes, depending on the specific tuning.  
Similar to [[meantone]], würschmidt implies that [[3/2]] will be tempered flat and/or 5/4 will be tempered sharp, and therefore [[6/5]] will be tempered flat. Unlike meantone, it is far more accurate. Combining würschmidt with meantone gives [[31edo]] as the first practical tuning with a generator of 10\31, but increasingly good 5-limit edo generators are [[34edo|11\34]] and especially [[65edo|21\65]], which notably is the point where it is combined with [[schismic]] and [[gravity]]. Other edo tunings include [[96edo]], [[99edo]] and [[164edo]]. Another tuning solution is to sharpen the major third by 1/8 of a würschmidt comma, which is to say by 1.43 cents, and thereby achieve pure perfect fifths; this is the [[minimax tuning]] for the [[5-odd-limit]].  


== Würschmidt ==
[[Mos scale]]s may not be the best approach for würschmidt since they are even more extreme than those of [[magic]]. [[Rothenberg propriety|Rothenberg-proper]] scales do not appear until 28, 31 or even 34 notes, depending on the specific tuning.
{{Main| Würschmidt }}


[[Subgroup]]: 2.3.5
[[Subgroup]]: 2.3.5
Line 19: Line 19:


[[Optimal tuning]]s:  
[[Optimal tuning]]s:  
* [[CTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.734{{c}}
* [[WE]]: ~2 = 1199.6942{{c}}, ~5/4 = 387.7005{{c}}
* [[POTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.799{{c}}
: [[error map]]: {{val| -0.306 -0.045 +0.775 }}
* [[CWE]]: ~2 = 1200.0000{{c}}, ~5/4 = 387.7762{{c}}
: error map: {{val| 0.000 +0.255 +1.463 }}


{{Optimal ET sequence|legend=1| 3, …, 28, 31, 34, 65, 99, 164, 721c, 885c, 1049cc, 1213ccc }}
{{Optimal ET sequence|legend=1| 3, …, 28, 31, 34, 65, 99, 164, 721c, 885c, 1049cc, 1213ccc }}


[[Badness]] (Smith): 0.040603
[[Badness]] (Sintel): 0.951


=== Overview to extensions ===
=== Overview to extensions ===
Line 33: Line 35:


==== Subgroup extensions ====
==== Subgroup extensions ====
Given that würschmidt naturally produces a neutral third at the interval 4 generators up, an obvious extension to prime 11 exists by equating this to [[11/9]], that is by tempering out [[5632/5625]] in addition to [[243/242]]; furthermore, like practically any 5-limit temperament with this accuracy level of [[3/2]] available, extensions to prime 19 exist by tempering out either [[513/512]] or [[1216/1215]] (which meet at 65edo and [[nestoria]]).
Given that würschmidt naturally produces a neutral third at the interval 4 generators up, an obvious extension to prime 11 exists by equating this to [[11/9]], that is by tempering out [[5632/5625]] in addition to [[243/242]].  


However, as discussed in the main article, the "free" higher prime for würschmidt outside the 5-limit is in fact 23, via tempering out S24 = [[576/575]] and S46<sup>2</sup> × S47 = [[12167/12150]]. Therefore, the below discusses the 2.3.5.23 and 2.3.5.11.23 extensions.
With this accuracy level of [[3/2]] available, extensions that add prime [[19/1|19]] exist by tempering out either [[513/512]] or [[1216/1215]] (which meet at 65edo), but they are very complex.  


=== 2.3.5.23 subgroup ===
The "free" higher prime for würschmidt outside the 5-limit is in fact [[23/1|23]], via tempering out [[576/575]] ({{S|24}}) and [[12167/12150]] ([[S-expression|S46<sup>2</sup>⋅S47]]). This is considered immediately below, and the no-11 restriction thereof is considered in [[#Other subgroup extensions]].
Extensions to harmonics [[47/1|47]] and [[49/1|49]] are also available at +11 and +5 generator steps respectively, equalizing the 45::50 [[harmonic series segment|segment]] of the [[harmonic series]]. If we derive a mapping of prime [[7/1|7]] from this mapping of 49, then we get the weak extension [[hemiwürschmidt]].


Subgroup: 2.3.5.23
=== 2.3.5.11 subgroup ===
Subgroup: 2.3.5.11


Comma list: 576/575, 12167/12150
Comma list: 243/242, 5632/5625


Subgroup-val mapping: {{mapping| 1 -1 2 0 | 0 8 1 14 }}
Subgroup-val mapping: {{mapping| 1 -1 2 -3 | 0 8 1 20 }}


Optimal tunings:  
Optimal tuning:  
* CTE: ~2 = 1200.000{{c}}, ~5/4 = 387.734{{c}}
* WE: ~2 = 1199.7508{{c}}, ~5/4 = 387.6058{{c}}
* POTE: ~2 = 1200.000{{c}}, ~5/4 = 387.805{{c}}
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.6697{{c}}


{{Optimal ET sequence|legend=0| 3, …, 28i, 31, 34, 65, 99, 164 }}
{{Optimal ET sequence|legend=0| 31, 34, 65 }}


Badness (Smith): 0.00530
Badness (Sintel): 0.477


==== 2.3.5.11.23 subgroup ====
==== 2.3.5.11.23 subgroup ====
Line 62: Line 64:


Optimal tuning:  
Optimal tuning:  
* CTE: ~2 = 1200.000{{c}}, ~5/4 = 387.652{{c}}
* WE: ~2 = 1199.8205{{c}}, ~5/4 = 387.6316{{c}}
* POTE: ~2 = 1200.000{{c}}, ~5/4 = 387.690{{c}}
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.6770{{c}}


{{Optimal ET sequence|legend=0| 31, 34, 65 }}
{{Optimal ET sequence|legend=0| 31, 34, 65 }}


Badness (Smith): 0.00660
Badness (Sintel): 0.300


== Septimal würschmidt ==
== Septimal würschmidt ==
Würschmidt, aside from the commas listed above, also tempers out [[225/224]]. [[31edo]] or [[127edo]] can be used as tunings. It extends naturally to an 11-limit version which also tempers out [[99/98]], [[176/175]] and 243/242. 127edo is again an excellent tuning for 11-limit würschmidt, as well as for [[minerva]], the 11-limit rank-3 temperament tempering out 99/98 and 176/175.
Septimal würschmidt, aside from the commas listed above, also tempers out [[225/224]]. [[31edo]] or [[127edo]] can be used as tunings. It extends naturally to an 11-limit version which also tempers out [[99/98]], [[176/175]] and 243/242. 127edo is again an excellent tuning for 11-limit würschmidt, as well as for [[minerva]], the 11-limit rank-3 temperament tempering out 99/98 and 176/175.
 
2-würschmidt, the temperament with all the same commas as würschmidt but a generator of twice the size, is equivalent to [[skwares]] as a 2.3.7.11 subgroup temperament.


The S-expression-based comma list of the 11-limit würschmidt discussed here is {[[176/175|S8/S10]], [[243/242|S9/S11]], [[225/224|S15]]}. Tempering out [[81/80|S9]] or [[121/120|S11]] results in [[31edo]], and in complementary fashion, tempering out [[64/63|S8]] or [[100/99|S10]] results in [[34edo]], but specifically, the 34d [[val]] where we accept 17edo's mapping of ~7. Their val sum, 31 + 34d = 65d, thus observes all of these [[square superparticular]]s by equating them as S8 = S9 = S10 = S11, hence its S-expression-based comma list is {{nowrap| {[[5120/5103|S8/S9]], [[8019/8000|S9/S10]], [[4000/3993|S10/S11]]} }}, which may be expressed in shortened form as {{nowrap| {S8/9/10/11} }}*. As a result, [[65edo]] is especially structurally natural for this temperament, though high damage on the 7 no matter what mapping you use (with the sharp 7 being used for this temperament); even so, it's fairly close to the optimal tuning already if you are fine with a significantly flat ~9/7, which has the advantage of ~14/11 more in tune. However, as 31edo is relatively in-tune already, 65d + 31 = [[96edo]] is also a reasonable choice, as it has the advantage of being [[patent val]] in the 11-limit, though it uses a different (more accurate) mapping for 13.
2-würschmidt, the temperament with all the same commas as würschmidt but a generator of twice the size, is equivalent to [[skwares]] as a [[2.3.7.11-subgroup|2.3.7.11 subgroup]] temperament.


(<nowiki>*</nowiki> The advantage of this form is we can easily see that all of the [[semiparticular]] commas expected are implied as well as any other commas expressible as the difference between two square superparticular commas by reading them off as ratios like 8/10 (S8/S10) and 9/11 (S9/S11).)
The S-expression-based comma list of the 11-limit würschmidt discussed here is {[[176/175|S8/S10]], [[243/242|S9/S11]], [[225/224|S15]]}. Tempering out [[81/80|S9]] or [[121/120|S11]] results in [[31edo]], and in complementary fashion, tempering out [[64/63|S8]] or [[100/99|S10]] results in [[34edo]] in the 34d [[val]], where we accept [[17edo]]'s mapping of prime 7. Their val sum, 31 + 34d = 65d, thus observes all of these [[square superparticular]]s by tempering them together. As a result, [[65edo]] is especially structurally natural for this temperament, though high damage on the 7; even so, it is fairly close to the optimal tuning already if you are fine with a significantly flat ~9/7, which has the advantage of ~14/11 more in tune. However, as 31edo is relatively in-tune already, [[96edo]] (= 65d + 31) is also a reasonable choice, as it has the advantage of being a [[patent val]] in the 11-limit, though it uses a different (more accurate) mapping for 13.


[[Subgroup]]: 2.3.5.7
[[Subgroup]]: 2.3.5.7
Line 85: Line 85:


[[Optimal tuning]]s:  
[[Optimal tuning]]s:  
* [[CTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.379{{c}}
* [[WE]]: ~2 = 1199.9741{{c}}, ~5/4 = 387.3742{{c}}
* [[POTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.383{{c}}
: [[error map]]: {{val| -0.026 -2.936 +1.009 +3.987 }}
* [[CWE]]: ~2 = 1200.0000{{c}}, ~5/4 = 387.3809{{c}}
: error map: {{val| 0.000 -2.907 +1.067 +4.031 }}


{{Optimal ET sequence|legend=1| 31, 96, 127 }}
{{Optimal ET sequence|legend=1| 31, 96, 127 }}


[[Badness]] (Smith): 0.050776
[[Badness]] (Sintel): 1.28


=== 11-limit ===
=== 11-limit ===
Line 100: Line 102:


Optimal tunings:  
Optimal tunings:  
* CTE: ~2 = 1200.000{{c}}, ~5/4 = 387.441{{c}}
* WE: ~2 = 1199.9618{{c}}, ~5/4 = 387.4347{{c}}
* POTE: ~2 = 1200.000{{c}}, ~5/4 = 387.447{{c}}
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.4446{{c}}


{{Optimal ET sequence|legend=0| 31, 65d, 96, 127 }}
{{Optimal ET sequence|legend=0| 31, 65d, 96, 127 }}


Badness (Smith): 0.024413
Badness (Sintel): 0.807


==== 13-limit ====
==== 13-limit ====
Line 115: Line 117:


Optimal tunings:  
Optimal tunings:  
* CTE: ~2 = 1200.000{{c}}, ~5/4 = 387.469{{c}}
* WE: ~2 = 1199.0325{{c}}, ~5/4 = 387.3137{{c}}
* POTE: ~2 = 1200.000{{c}}, ~5/4 = 387.626{{c}}
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.5893{{c}}


{{Optimal ET sequence|legend=0| 31, 65d }}
{{Optimal ET sequence|legend=0| 31, 65d }}


Badness (Smith): 0.023593
Badness (Sintel): 0.975


==== Worseschmidt ====
==== Worseschmidt ====
Line 130: Line 132:


Optimal tunings:  
Optimal tunings:  
* CTE: ~2 = 1200.000{{c}}, ~5/4 = 387.179{{c}}
* WE: ~2 = 1200.4712{{c}}, ~5/4 = 387.2511{{c}}
* POTE: ~2 = 1200.000{{c}}, ~5/4 = 387.099{{c}}
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.1252{{c}}


{{Optimal ET sequence|legend=0| 3def, 28def, 31 }}
{{Optimal ET sequence|legend=0| 3def, 28def, 31 }}


Badness (Smith): 0.034382
Badness (Sintel): 1.42


== Worschmidt ==
== Worschmidt ==
Worschmidt tempers out 126/125 rather than 225/224, and can use [[31edo]], [[34edo]], or [[127edo]] as a tuning. If 127 is used, note that the val is {{val| 127 201 295 '''356''' }} (127d) and not {{val| 127 201 295 '''357''' }} as with würschmidt. In practice, of course, both mappings could be used ambiguously, which might be an interesting avenue for someone to explore.
Worschmidt tempers out [[126/125]] rather than [[225/224]], and can use [[31edo]], [[34edo]], or [[127edo]] as a tuning. If 127 is used, note that the val is {{val| 127 201 295 '''356''' }} (127d) and not {{val| 127 201 295 '''357''' }} as with würschmidt. In practice, of course, both mappings could be used ambiguously, which might be an interesting avenue for someone to explore.


[[Subgroup]]: 2.3.5.7
[[Subgroup]]: 2.3.5.7
Line 147: Line 149:


[[Optimal tuning]]s:  
[[Optimal tuning]]s:  
* [[CTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.406{{c}}
* [[WE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.406{{c}}
* [[POTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.392{{c}}
: [[error map]]: {{val| +0.763 -1.610 +2.851 -2.786 }}
* [[CWE]]: ~2 = 1200.0000{{c}}, ~5/4 = 387.3950{{c}}
: error map: {{val| 0.000 -2.795 +1.081 -4.960 }}


{{Optimal ET sequence|legend=1| 31, 96d, 127d }}
{{Optimal ET sequence|legend=1| 31, 96d, 127d }}


[[Badness]] (Smith): 0.064614
[[Badness]] (Sintel): 1.64


=== 11-limit ===
=== 11-limit ===
Line 162: Line 166:


Optimal tunings:  
Optimal tunings:  
* CTE: ~2 = 1200.000{{c}}, ~5/4 = 387.472{{c}}
* WE: ~2 = 1200.7911{{c}}, ~5/4 = 387.6624{{c}}
* POTE: ~2 = 1200.000{{c}}, ~5/4 = 387.407{{c}}
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.4189{{c}}


{{Optimal ET sequence|legend=0| 31, 65, 96d, 127d }}
{{Optimal ET sequence|legend=0| 31, 65, 96d, 127d }}


Badness (Smith): 0.033436
Badness (Sintel): 1.11


== Whirrschmidt ==
== Whirrschmidt ==
Line 179: Line 183:


[[Optimal tuning]]s:  
[[Optimal tuning]]s:  
* [[CTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.853
* [[WE]]: ~2 = 1199.5703{{c}}, ~5/4 = 387.7422{{c}}
* [[POTE]]: ~2 = 1200.000{{c}}, ~5/4 = 387.881
: [[error map]]: {{val| -0.430 +0.412 +0.569 -0.216 }}
* [[CWE]]: ~2 = 1200.0000{{c}}, ~5/4 = 387.8729{{c}}
: error map: {{val| 0.000 +1.029 +1.559 +0.567 }}


{{Optimal ET sequence|legend=1| 34d, 65, 99 }}
{{Optimal ET sequence|legend=1| 34d, 65, 99 }}


[[Badness]] (Smith): 0.086334
[[Badness]] (Sintel): 2.18


=== 11-limit ===
=== 11-limit ===
Line 194: Line 200:


Optimal tunings:  
Optimal tunings:  
* CTE: ~2 = 1200.000{{c}}, ~5/4 = 387.829
* WE: ~2 = 1199.2839{{c}}, ~5/4 = 387.6507{{c}}
* POTE: ~2 = 1200.000{{c}}, ~5/4 = 387.882
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.8682{{c}}


{{Optimal ET sequence|legend=0| 34d, 65, 99e }}
{{Optimal ET sequence|legend=0| 34d, 65, 99e }}


Badness (Smith): 0.058325
Badness (Sintel): 1.93
 
== Other subgroup extensions ==
=== Würschmidt (2.3.5.23) ===
Extensions to harmonics [[47/1|47]] and [[49/1|49]] are also available at +11 and +5 generator steps respectively, equalizing the 45::50 [[harmonic series segment|segment]] of the [[harmonic series]]. If we derive a mapping of prime [[7/1|7]] from this mapping of 49, then we get the weak extension [[hemiwürschmidt]].
 
Subgroup: 2.3.5.23
 
Comma list: 576/575, 12167/12150
 
Subgroup-val mapping: {{mapping| 1 -1 2 0 | 0 8 1 14 }}
 
Optimal tunings:
* WE: ~2 = 1199.7075{{c}}, ~5/4 = 387.7106{{c}}
* CWE: ~2 = 1200.0000{{c}}, ~5/4 = 387.7807{{c}}
 
{{Optimal ET sequence|legend=0| 31, 34, 65, 99, 164 }}
 
Badness (Sintel): 0.216


[[Category:Temperament families]]
[[Category:Temperament families]]