How to match these two surfaces smooth?

[eg1]

Where does the edge in the zebra come from? And why is the curvature map soo disgusting?
Do I need a higher degree srf? Shouldnt it be possible to match a srf to “the half” of another srf?

rhino srf question.3dm (2.8 MB)

[Tom_P]

zebra is rendered on the so called analysis mesh.
in the zebra dialog there is a “adjust mesh”
go to detailed control and use some where fine settings.

this is already very dense / fine:

@eg1

Where does the edge in the zebra come from?

… from the default analysis mesh

[eg1]

Ok thx that explains that! Why is the curvature analysis soo disgusting though? Is this normal for what Im trying to do?

[Tom_P]

@eg1

curvature analysis soo disgusting though

you are showing “mean” curvature ?
check the help - explaining the styles

Curvature | Rhino 3-D modeling

and / or read some background info in wikipedia:

Mean curvature

min radius will show a nice distribution:

see this topic to understand saddle like (“inner part of torus”) - your surface - max Radius:

Command Error: CurvatureAnalysis - Max Radius

[image] as i came across this issue again with my students - see an illustration for @mikko 's explanation. The grey saddle surfaces is evaluated at the center. The cyan arcs show the 2 principal curvatures. Somewhere between convex and concave there is zero curvature or infinity radius. (the red lines - arcs with infinit radius and zero curvature). hope this clarifies / visualises it better. maybe a bit more info in the help would be nice ? kind regards -tom

mean curvature is the average of the 2 principal curvatures.
from the help:

Mean
Displays the absolute value of the mean curvature. It is useful for finding areas of abrupt change in the surface curvature.

because of its “averaging” math - it is not the best choice to say perfectly distributed curvature / perfectly smooth…

note the curvatureGraph can also be applied to surfaces.
some people like _emap analysis a lot.

what are the surfaces for ? for most use cases i would claim they look good.
For a chapter in a nurbs-modelling school book i would criticise the controlpolygon / CV-distribution.

happy modelling.

[Lagom]

Another good explanatoral for beginners is this one here.