Aura
Joined 31 August 2020
→Temperament(s) for Diatonicized Chromaticism?: Finished odd harmonics table (group) for 11L 2s. |
|||
(27 intermediate revisions by 3 users not shown) | |||
Line 380: | Line 380: | ||
:::::::::: I finished doing the above for 11L 2s. Note that 11/8 (the dark generator, and thereby the bright generator 16/11) remains stable throughout the entire currently posted 11L 2s table &emdash; the worst relative error is -34.8%, at 127edo, so your favorite 159edo still easily fits in. Right now I strung out the odd harmonics to high values in case I need these for composite ratios, which seem likely to be needed for 17L 2s, but seem like they may not be needed for 11L 2s (but better leave them in for now until I am sure they are not needed). [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 22:54, 9 April 2025 (UTC) | :::::::::: I finished doing the above for 11L 2s. Note that 11/8 (the dark generator, and thereby the bright generator 16/11) remains stable throughout the entire currently posted 11L 2s table &emdash; the worst relative error is -34.8%, at 127edo, so your favorite 159edo still easily fits in. Right now I strung out the odd harmonics to high values in case I need these for composite ratios, which seem likely to be needed for 17L 2s, but seem like they may not be needed for 11L 2s (but better leave them in for now until I am sure they are not needed). [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 22:54, 9 April 2025 (UTC) | ||
::::::::::: Wait, I thought that the large step of 11L 2s in 159edo was 13\159, not 11\159... --[[User:Aura|Aura]] ([[User talk:Aura|talk]]) 23:10, 9 April 2025 (UTC) | |||
:::::::::::: Good catch of typo -- fixed this. Also rechecked the rest, but didn't find any more. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 05:30, 10 April 2025 (UTC) | |||
:::::::::::: Scrolling through these table groups, I started noticing interesting things, like how even though the 11th harmonic is the only one with stable mapping all the way through 11L 2s, some of the others have stable mapping in sections, like the 3rd harmonic has stable mapping in the middle section but is all over the place in both the hard and soft ends, but the 9th harmonic actually does okay in the hard end, as does the 17th harmonic (both of these get to be all over the place in the soft end), and the 5th and 13th harmonics have stable mapping in the soft end as long as the EDO values are not too large. | |||
:::::::::::: In partial contrast, with 17L 2s, the harmonic/subharmonics of the generator have unstable mapping (because no simple ratio with a reasonable sized numerator and denominator fits into this zone), but the 3rd harmonic is nearly rock-solid (and 112b is a respectable if overly-complex quarter-comma meantone approximation), although presumably its mapping would break if I put in the rest of the right-most column of the MOS spectrum table. And there the 5th harmonic seems very much usable in the soft end of the scale tuning spectrum as long as the EDO sizes don't get too large (and even then, sometimes it is still okay), which looks to me like enabling a 2.3.5.23 meantone extension; it goes all over the place in the hard end, but there the 25th harmonic shines and is rock-solid as long as you don't go harder than 36edo (basic), and the 13th harmonic just barely misses being rock-solid in this zone (just barely breaks on 125edo, for which 125f would be not bad). Although those harmonics would also appear less solid if I included the rest of the MOS tuning spectrum. | |||
:::::::::::: [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 07:28, 10 April 2025 (UTC) | |||
In reply to your previous comment, I should mention that I'd like to know where ~3/2 is relative to the 11L 2s of 159edo. I imagine it's rather far along the generator sequence. At the same time, I'm wondering what sorts of chords you can actually get from 11L 2s in 159edo. --[[User:Aura|Aura]] ([[User talk:Aura|talk]]) 16:53, 10 April 2025 (UTC) | |||
: This is further up on my page than the tables of odd harmonics: For the fifths in the basic 11L 2s (Wyschnegradsky) diatonicized chromatic scale (the version that 24edo yields), the 11/8-span of a patent fifth is a stack of 10 intervals of 11/8, octave-reduced. I haven't yet done this calculation for 159edo — it isn't too far off from basic, but it has small enough increments that attempting to use the same 11/8-span gives the b val fifth instead of the patent fifth. 00:07, 11 April 2025 (UTC) | |||
: Did this for 159edo — the 16/11-span of 3/2 is 51. And it doesn't even work for the next EDO up or down from the last column of the tuning spectrum table of 11L 2s (135edo or 146edo, respectively) — the 3rd harmonic mapping is too unstable for EDO sizes that large in this region. Also, 51 is so many iterations of the generator that it goes well outside of the 11L 2s scale. Added: [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 05:53, 11 April 2025 (UTC) Last Modified: [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 06:52, 11 April 2025 (UTC) | |||
:: Fascinating. I guess that means we need to look into chords with 94\159 fifths, which actually approximate 128/85 rather than 3/2. These chords clearly cannot be pure 5-limit either, but that's okay. Given that you mention 2.3.5.23 meantime, I'm now wondering if we can cobble together something for 11L 2s based on the ~128/85 archagall fifth. To start on this front, what are the modes of 11L 2s that contain ~128/85 in 159edo? --[[User:Aura|Aura]] ([[User talk:Aura|talk]]) 10:40, 11 April 2025 (UTC) | |||
::: I need to do the modes for 11L 2s with respect to coverage of standard diatonic intervals anyway, but that's going to take a bit of time, so stay tuned. The 128/85 b fifth is going to stick out like a sore thumb in the midst of the much more accurate other intervals that 159edo has, though, so maybe it would be better to try to come up with a decent MODMOS derived from 11L 2s for 159edo? In the meantime, I figured out which equal temperaments this might apply to, since they get their fifth from 51 stacked and octave-reduced 16/11 generators, leading up to 159edo: 37b, 61, 98, 159 (have not yet tried to extend beyond 159edo to see how far you can get before the fifth mapping breaks again). Note that for 61edo, 51 stacked and octave-reduced 16/11 generators gives the same note as 10 stacked and octave-reduced 11/8 generators. I also included 37b to show the small endpoint of the series (and if you really want to go weird, include 24b); judging by the pattern, the next member of the series would be 257 (not yet sure of wart, if any). Graham Breed's temperament finder lists some more members of the series without warts, going up into the thousands, but doesn't give the temperament a name beyond [https://x31eq.com/pyscript/rt.html?ets=159_98&limit=2_3_11 159 & 98], and I don't see 61edo listed in the Nexus, Nexus clan, or Nexus family, although 159edo comes up several times (maybe one of those temperaments would be better, although they probably go from the fifth to the 11/8 rather than the other way around, and would not necessarily support 11L 2s). Its enormous complexity is presumably the reason it never got a name; I don't know the specifics for computing badness, but I am going to stick my neck out and guess that this temperament would have huge badness despite its high accuracy. | |||
::: As for 2.3.*.23, the 23rd harmonic mapping is pretty unstable in the zone of 11L 2s — it gets better mapping stability in 17L 2s, although this region has its generators sharp enough relative to 23/16 that larger EDO sizes often flip to the next flatter approximation, so the generator for that needs to be something between 23/16 and 13/9. | |||
::: [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 15:21, 11 April 2025 (UTC) | |||
::: The modes of 11L 2s that enable reaching 4/3 and/or 3/2 within-scale are not very many. If you want to get both, you have to use sLLLLLsLLLLLL (no other mode gets both). If you want to get 3/2 but can live without 4/3 (more common than the opposite), then you can also use LsLLLLLsLLLLL or sLLLLLLsLLLLL. If you use LLLLLLsLLLLLs or LLLLLsLLLLLLs or LLLLLsLLLLLsL you can get 4/3 but not 3/2. In EDOs with fine enough steps to distinguish 128/85(?) from 3/2, this instead gets you 128/85(?) (or 85/64(?) instead of 4/3). So if we want a MODMOS for 159edo that gets us back to 3/2 (and maybe 4/3) we need a second generator that gets tempered to be the same as 11/8 or 16/11 in the coarser EDOs supporting 11L 2s but distinguished in the finer ones. Not sure yet what that would be, and not sure yet whether 128/85 is the best slightly-inflated fifth substitute to use for this purpose. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 00:02, 12 April 2025 (UTC) | |||
:::: Well, the ~128/85 fifth of 159edo is indeed useful in its own right as an imitation of the kind of thing you see in 22edo. Perhaps we can play to the strengths of that type of system where two ~128/85 intervals octave-reduced add up to ~17/15, and two instances of ~17/15 add up to ~9/7- yes, that is how 159edo works since both 22edo and 159edo support archagall temperament. Not only that but the ~128/85 interval is 159edo's best approximation of 17edo's fifth, so we could also shoot for 11L 2s harmonies that evoke the kind of stuff found in 17edo and it's multiples. The point is that if the ~11/8 and ~16/11 intervals count as ambisonances (basically, they're both consonances and dissonances at the same time) for our intents and purposes, as I know they do for mine, then we can afford to play with some of the less accurate intervals in our 11L 2s harmonies- assuming we play our cards right. --[[User:Aura|Aura]] ([[User talk:Aura|talk]]) 04:28, 12 April 2025 (UTC) | |||
::::: [[24576/24565#Mavka_a.k.a._archagallismic|Mavka a.k.a. archagallismic]] looks sort of like what we want, although that seems like more dimensions than we need, and I don't know how you would map that many dimensions onto any kind of keyboard, even before considering the size. (But maybe that could be trimmed down to a rank-3 temperament that is different from Archgallic, so as to get the 11/8 in there?) I see that its multiple generators include ~3, and it includes 150edo as a patent val, so you can get both 128/85 and 3/2. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 08:25, 12 April 2025 (UTC) | |||
:::::: If I can map 159edo onto an isomorphic keyboard layout using [[tertiaschis]] temperament, albeit by squashing the hexagons, then we can work together to map a lower-dimensional temperament that relates to archagallic temperament. --[[User:Aura|Aura]] ([[User talk:Aura|talk]]) 09:39, 12 April 2025 (UTC) | |||
::::::: Tertiaschis looks interesting. Meanwhile, I determined that the Rastma (243/242) is NOT the right interval for differentiating 159edo from 61edo (lower on the series I noted above) — 159edo maps it correctly to 1 increment, but 61edo not only doesn't temper it out, but inflates it to 2 increments, while 98edo inverts it. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 18:16, 12 April 2025 (UTC) | |||
::::::: Should have tried the Char comma/chroma ([[256/255]]) first. This is tempered out in 24edo, 37edo, 61edo, and 98edo, but correctly maps to 1 increment of 159edo; it is exactly the amount by which an archagall fifth exceeds a normal fifth. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 10:13, 13 April 2025 (UTC) | |||
:::::::: Nice catch. Yeah, the char subchroma will prove useful to us. Come to think of it, however, the way that this interval plays with the intervals of 11L 2s reminds me of how the syntonic comma plays with the intervals of 5L 2s... --[[User:Aura|Aura]] ([[User talk:Aura|talk]]) 10:30, 13 April 2025 (UTC) | |||
::::::::: That's the idea. Although other parts of the 11L 2s tuning spectrum may need alternate syntonic comma/chroma analogs, just like some parts of the 5L 2s tuning spectrum are best served by 64/63 instead of 81/80. (And sorry about the typo in the [[256/255]] page.) [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 21:12, 13 April 2025 (UTC) | |||
::::::::: Checking in to let you know I didn't forget about this. In addition, I've been coming to the conclusion that while 17L 2s seems to fit with temperaments that proceed along at least part of the vertical axis of the corresponding tuning spectrum table, 11L 2s requires temperaments that proceed along the horizontal axis of the corresponding tuning spectrum table. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 01:49, 21 April 2025 (UTC) | |||
::::::::: Also, 256/255 currently has an edit war going on about whether its names should include "char comma", from our point of view, "char comma" would have been nice for enabling "char chroma" in tuning systems that use it as an interval instead of tempering it out. I thought about putting this in the associated discussion page, but the edit war is going on with apparently no inclination to put any further discussion there (although plenty of old discussion about naming exists there). | |||
::::::::: Separately, work on 17L 2s is looking like it might give me a decent 2.3.5.13.23 meantone extension (but still need to check 13th harmonic mapping stability to be sure). | |||
::::::::: [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 10:20, 21 April 2025 (UTC) | |||
::::::::: For 17L 2s, had to go for a 2.3.5.23.53 meantone extension to get the soft half of the spectrum — the 13th harmonic mapping just wasn't stable enough. Now that I've got that done, I wonder if giving the very high harmonics another look might turn up something similar for 11L 2s? Bonus points if it works accurately enough to include an EDO as large as 159edo (for the 17L 2 meantone extension, 146edo was too big to fit without an accuracy-degrading 'c' wart as well as an 'i' wart). [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 10:13, 22 April 2025 (UTC) | |||
::::::::: On second thought, what turned out to be necessary for (at least the soft half of) 17L 2s might not help for 11L 2s, because the former didn't have a stable generator fraction within range, whereas the latter does &mbdash; hence the same extraordinary measures might not do much for it. But it can't hurt to check. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 11:28, 22 April 2025 (UTC) | |||
:::::::::: Sorry about not responding for a while, I had stuff to do, but I was reading what you were saying, and I've been checking in on harmonies that you can get from instances of ~85/64 and ~128/85. It seems that in 159edo, starting on your ~17/16, you can build an approximation of a 1/1-25/22-128/85-320/187 suspension easily, and you can use this as an unexpected option for something resembling a Neapolitan chord. Not sure how this chord fits with 11L 2s however- it probably doesn't. --[[User:Aura|Aura]] ([[User talk:Aura|talk]]) 16:24, 24 April 2025 (UTC) | |||
::::::::::: 11L 2s won't include 25/22 (29\159, which is 3\159 too high) or 320/187 (123\159, which is also 3\159 too high — I just checked to make sure it maps correctly, although 200/117 is a lot more accurate). I was thinking maybe a MODMOS derived from 11L 2s would do it, but for that to work, one would have to be too high and the other too low. So it would have to be made with accidentals. Of course, 159edo has a load of other scales, so maybe one of these might do the job without accidentals. [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 18:10, 24 April 2025 (UTC) | |||
:::::::::::: I don't know if this is a good temperament for 11L 2s, but I stumbled upon [[Alphatricot_family#Tritricot|Tritricot]] while looking up stuff for temperamental generator of temperaments supporting 17L 2s: All listed *-limits of Tritricot list 159edo as their first optimal tuning. (And even if Tritricot isn't optimal for 11L 2s, it WILL work by hook or by crook, since all EDOs listed in the optimal tuning sequence are ≥159edo, and 159edo is already >143edo, which is the largest EDO that DOESN'T support 11L 2s, being to 11L 2s what 35edo is to 5L 2s. But I have yet to check whether you have to use some awful number of generators to get 11/8 or 16/11. Although at that high an EDO size, a MODmos based upon 11L 2s is likely to be better anyway, and for that you're going to want at least a secondary generator.) [[User:Lucius Chiaraviglio|Lucius Chiaraviglio]] ([[User talk:Lucius Chiaraviglio|talk]]) 05:05, 28 April 2025 (UTC) | |||
== Notability guidelines review == | |||
I drafted up a new proposal for the notability guidelines over at [[User:Sintel/Notability guidelines]]. | |||
You can use the talk page there to let us know what you think about them, especially in context of the discussion on the previous proposal wrt some of the pages you created. | |||
Thank you! – [[User:Sintel|Sintel🎏]] ([[User_talk:Sintel|talk]]) 23:46, 7 May 2025 (UTC) | |||
: Note for future readers: this conversation moved to [[User talk:Sintel/Notability guidelines]] from this point. --[[User:BudjarnLambeth|BudjarnLambeth]] ([[User talk:BudjarnLambeth|talk]]) 02:41, 8 May 2025 (UTC) |