[Bf-committers] Proportional editing distance calculation ?

Bradley Smith bradley at baysmith.com
Fri Apr 12 18:17:08 CEST 2013

Proportional editing connected mode is using the graph distance (the
distance calculated when traversing the mesh to find connected
vertices) rather than the Euclidean distance. The behavior seems to
have been altered on Feb 1, 2010, with commit 26537 "Proportional Edit
Connected: Use connectivity distance for falloff (connectivity
distance now also calculated across inner face edges)."

Reversing the behavior can be easily accomplished by deleting three
lines of code (lines 1724-1726 in transform_generics.c). However, I
don't know what effect this would have beyond fixing the behavior of
mesh editing with proportional editing connected mode.

Does someone know a reason for using the graph distance rather than
the Euclidean distance?


On Sat, Apr 6, 2013 at 2:13 PM, Hadrien Brissaud <hadriscus at gmail.com> wrote:
> Hello,
> I have come across an annoying behaviour when using proportional editing -
> connected mode. This mode is very useful for situations where we don't want
> nearby, unconnected areas to be affected by transformations made with
> proportional editing enabled.
> This happens with "*proportional edit"* on : http://i.imgur.com/FE6uPh1.png
> Now, this happens with "*proportional editing connected"* on :
> http://i.imgur.com/U5TZW3i.png
> I understand this is likely due to the way Blender calculates distance to
> nearby vertices - for instance, diagonals may be calculated as twice
> farther than directly-connected vertices. This is verified by attempting
> the same deformation on a radially-subdivided mesh :
> http://i.imgur.com/O6PPKVf.png (it works "as expected")
> As reference, Maya handles both situations the same way, and this is the
> result : http://imgur.com/B6bQ0sL
> How easily could this be fixed ?
> Thank you,
> Hadrien
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers

More information about the Bf-committers mailing list