Path-based goodness: Difference between revisions

Line 1:

Path-based goodness is a measure of your ability as a composer to create music recognizable as tempered JI. we assume a composer wants to return to previously written pitches; otherwise, they would just use JI. so their music contains multiple paths from a tone to the same tone. They are likelier to use a path if the intervals within it are simple and close to just. So path-based goodness is a measure of how much total ''option'' (which I use as a mass noun here because there is an infinite number of options and some paths are more of a compositional option for them than others) a composer has when writing music. ~~brb going to put that on the path-based goodness page~~

Path-based goodness is a measure of your ability as a composer to create music recognizable as tempered JI. we assume a composer wants to return to previously written pitches; otherwise, they would just use JI. so their music contains multiple paths from a tone to the same tone. They are likelier to use a path if the intervals within it are simple and close to just. So path-based goodness is a measure of how much total ''option'' (which I use as a mass noun here because there is an infinite number of options and some paths are more of a compositional option for them than others) a composer has when writing music.

The parameters are calibrated according to what best reflects the preferences of composers (how much they use simple intervals, and how much error they're willing to tolerate). If a composer works under different aesthetic criteria, they may want to use different parameters. For example, if one uses timbres less sensitive to mistuning, one may want to increase <code>inaccuracy_fondness</code>. If one is interested in essentially tempered chords more than long comma pumps, one may want to lower <code>target_score</code>. hkm has found that changing <code>target_score</code> by an order of magnitude affects rankings minimally; changing the other parameters changes the rankings in a predictable and small way. (For example, if we increase <code>inaccuracy_fondness</code>, all scores go up, but scores for less accurate systems increase more.) We do not treat some JI intervals as having more room for error than others, both because it is not entirely agreed upon whether they do, and because it should have little change in the results (since simple intervals can be stacked upon other intervals to become complex intervals).

Revision as of 20:17, 17 August 2025 view source Hkm (talk \| contribs) autopatrolled 996 edits No edit summary Tag: Visual edit: Switched ← Older edit		Revision as of 00:29, 26 August 2025 view source Hkm (talk \| contribs) autopatrolled 996 edits No edit summary Tag: Visual edit Newer edit →
Line 1:		Line 1:
	Path-based goodness is a measure of your ability as a composer to create music recognizable as tempered JI. we assume a composer wants to return to previously written pitches; otherwise, they would just use JI. so their music contains multiple paths from a tone to the same tone. They are likelier to use a path if the intervals within it are simple and close to just. So path-based goodness is a measure of how much total ''option'' (which I use as a mass noun here because there is an infinite number of options and some paths are more of a compositional option for them than others) a composer has when writing music. ~~brb going to put that on the path-based goodness page~~		Path-based goodness is a measure of your ability as a composer to create music recognizable as tempered JI. we assume a composer wants to return to previously written pitches; otherwise, they would just use JI. so their music contains multiple paths from a tone to the same tone. They are likelier to use a path if the intervals within it are simple and close to just. So path-based goodness is a measure of how much total ''option'' (which I use as a mass noun here because there is an infinite number of options and some paths are more of a compositional option for them than others) a composer has when writing music.

	The parameters are calibrated according to what best reflects the preferences of composers (how much they use simple intervals, and how much error they're willing to tolerate). If a composer works under different aesthetic criteria, they may want to use different parameters. For example, if one uses timbres less sensitive to mistuning, one may want to increase <code>inaccuracy_fondness</code>. If one is interested in essentially tempered chords more than long comma pumps, one may want to lower <code>target_score</code>. hkm has found that changing <code>target_score</code> by an order of magnitude affects rankings minimally; changing the other parameters changes the rankings in a predictable and small way. (For example, if we increase <code>inaccuracy_fondness</code>, all scores go up, but scores for less accurate systems increase more.) We do not treat some JI intervals as having more room for error than others, both because it is not entirely agreed upon whether they do, and because it should have little change in the results (since simple intervals can be stacked upon other intervals to become complex intervals).		The parameters are calibrated according to what best reflects the preferences of composers (how much they use simple intervals, and how much error they're willing to tolerate). If a composer works under different aesthetic criteria, they may want to use different parameters. For example, if one uses timbres less sensitive to mistuning, one may want to increase <code>inaccuracy_fondness</code>. If one is interested in essentially tempered chords more than long comma pumps, one may want to lower <code>target_score</code>. hkm has found that changing <code>target_score</code> by an order of magnitude affects rankings minimally; changing the other parameters changes the rankings in a predictable and small way. (For example, if we increase <code>inaccuracy_fondness</code>, all scores go up, but scores for less accurate systems increase more.) We do not treat some JI intervals as having more room for error than others, both because it is not entirely agreed upon whether they do, and because it should have little change in the results (since simple intervals can be stacked upon other intervals to become complex intervals).