36edo: Difference between revisions
| Line 31: | Line 31: | ||
36edo has almost 50% relative error on harmonics 5/1 and 11/1. This means that whether one [[Octave stretch|stretches]] or [[Octave shrinking|compresses]] the octave, either way it will improve 36edo’s approximations of [[JI]], but in opposite directions (as long as it is done by the right amount). | 36edo has almost 50% relative error on harmonics 5/1 and 11/1. This means that whether one [[Octave stretch|stretches]] or [[Octave shrinking|compresses]] the octave, either way it will improve 36edo’s approximations of [[JI]], but in opposite directions (as long as it is done by the right amount). | ||
What follows is a comparison of | What follows is a comparison of compressed- and stretched-octave 36edo tunings. | ||
Of the compressed-octave tunings listed, ''36ed513/256'', with a step size about 33.25 cents, performs best in this comparison, approximating all 11-limit primes with less than 36% relative error (<12 cents error). | Of the compressed-octave tunings listed, ''36ed513/256'', with a step size about 33.25 cents, performs best in this comparison, approximating all 11-limit primes with less than 36% relative error (<12 cents error). | ||