SUNDIAL - hours distortion

Hi there,

I’m currently working on a sundial project and I’m having problems adjusting the numbers for the hours. The closest solution I have found is “cage edit” in rhino though it’s far from perfect, even the width of the numbers is distorted… I can’t think of a similar solution in grasshopper to smootlhy parametrize my numbers…

I’d like to achieve a similar effect to the example on the sundial above. So far I could define that the numbers in the example follow two rules: they’re parallel to the angle of the hour line and their base and top follow two different curvatures of the sundial face.

SUNDIAL_FACE_hour_problem.3dm (9.4 MB) (16.9 KB)

Hello Bed, fine design you’ve got started. Curious why you used IIII instead of IV for four? I’ll be tagging along to see how it works, because I’m trying to learn Blender; and for me everything is so much easier to draw/model in Rhino.

Hi Rob, I didn’t even notice :), but that’s in the example i’m trying to emulate.

I love it when a question make me look it up:
""The numerical notation of 4 is IV in Roman numerals. You probably think so, too.
However, there are many cases where IIII is used at the 4:00 position on the dial plates of clocks that use Roman numerals. “”
The Mystery of Numerical Notation on the Dial Plate - 4 is Expressed as IIII, not IV | THE SEIKO MUSEUM GINZA.


all the roman numerals that might fit in a clock can be drawn a single strokes or with a combination of single strokes that starts from the top and reaches a point in the bottom (or the opposite)

this leads me to think that as long as you move the top points of the cage together, everything should be fine :slight_smile:

thinking about it a bit more, maybe you don’t even need cage edit (for the reason that all the strokes start from top_line and end on bottom_line) and could directly do something like an CurveClosestPoint with always Reparametrized reference and destination curves?

for instance here is a simple experiment with some arcs drawn in Rhino, just to understand if it might work (I have used arcs because it looks like the numerals on the reference image have a round bottom-line and top-line) (13.6 KB)