I did some experiments with drawing hexagons and decided a ratio of 0.9 worked best: https://code.jsoftware.com/wiki/NYCJUG/HexagonAspects .
On Mon, Mar 21, 2022 at 10:48 AM Raul Miller <[email protected]> wrote: > On Mon, Mar 21, 2022 at 7:18 AM Martin Kreuzer <[email protected]> > wrote: > > The option 'aspect' which was suggested as a remedy (using a value of > > 0.866, which -for reasons unknown to me- is suspiciously close to > > -:%:3) has probably more to do with the aspect of the plotted grid > > (and if so, in a weird way, at least one I do not fully understand). > > 1 2 o./ 2p1 * 0 1 2 % 6 > 0 0.866025 0.866025 > 1 0.5 _0.5 > > Or, another approach: https://en.wikipedia.org/wiki/Eisenstein_integer > > Basically, if you are working with hexagons aligned on a square grid, > you are working with multiples of the square root of 0.75 along one > axis and multiples of 1 along the other axis. > > Anyways, personally, I think using an aspect of 1 and putting that > %:0.75 ratio into the data would be the right approach. My reading of > https://code.jsoftware.com/wiki/Plot/Options (and > https://code.jsoftware.com/wiki/Plot/Function) suggests to me that > using an aspect of %:0.75 should mean that the y axis is slightly > shorter than the x axis. This would let me use 1 in the y direction to > visually represent a %:0.75 distance in the y direction, which would > work, but is a bit confusing to talk about. Still, > > 'aspect 0.866' plot 1 0.5j1 _0.5j1 _1 _0.5j_1 0.5j_1 1 > > does give me something that looks like > > plot j./2 1 o./2p1*(i.7)%6 > > I hope this helps, > > -- > Raul > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > -- Devon McCormick, CFA Quantitative Consultant ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
