Yes, can confirm my fedora vbox build is now working fine! Many thanks,
Brecht! :)
On 22/11/2012 07:46, Sergey Sharybin wrote:
> Thanks, Brecht!
>
> Have been trying to make this working tonight, but something didn't work..
>
>
> On Thu, Nov 22, 2012 at 3:28 AM, Brecht Van Lommel<
> brechtvanlom.
Actually, subdivide creates n-gons in most cases (unless you use it on edge
rings or the whole mesh), so it's likely to be the problem.
Also here is a quick patch that adds a "Cursor to Center of Mass" option:
http://www.pasteall.org/37435/diff
Am Mi, 21. Nov, 2012 um 11:20 ,Jason Wilkins
schr
Thanks, Brecht!
Have been trying to make this working tonight, but something didn't work..
On Thu, Nov 22, 2012 at 3:28 AM, Brecht Van Lommel <
brechtvanlom...@pandora.be> wrote:
> Revision: 52453
>
> http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=52453
> Author:
Hi!
I'm aiming a bit lower so I prefer to attack the mechanics of the problem.
Limiting the calculation by actual volume to when it makes sense would be
nice, but detecting overlaps etc. seems overkill.
Here's later code which does it basically two ways, one is the tetrahedron
and the other is po
Every algorithm has it's limitations. Computing the center of mass for a
body that has no volume is a quite useless task and assuming
inhomogeneous density for a body that is only defined by the surface is
also questionable. For the average case a closed surface and the
assumption of homogeneit
Did you check against 'non orientable surfaces'?
like
Möbius strip
Klein Bottle
I doubt the algorithm stands with that...
As far as I remember those objects have no defined volume?
As : how would you like to make a cylinder from a Möbius strip?
You can always create a bounding object .. like in
Hi!
I finished writing some testing code (only prints out the cm though).
Planar convex/concave n-gon centroids should be correct and for closed
meshes there's the tetrahedron stuff which uses either "fake" tris from
polys or correct tris from tessfaces. Non-planar polys won't be correct
because o
Well, the calculation is used first to map the origin of the object
and this discussion was just how to extend that to moving the cursor.
As for it being because of n-gons, I do not think my model contained
n-gons. What I did was call the subdivide operator 4 times on one end
of the model, which
Oops, I forgot this was about the cursor snapping and had something else in
mind... My apologies. =)
> From: p_boel...@msn.com
> To: bf-committers@blender.org
> Date: Wed, 21 Nov 2012 20:07:04 +0100
> Subject: Re: [Bf-committers] Cursor to center of mass
>
>
> I can sort of see the reasoning b
I can sort of see the reasoning behind that, but I would also argue accuracy
can be very important. Maybe it'd be worth it to make a "high quality" operator
bool in the same vein as the "high quality normals" in the Solidify modifier?
-Patrick
> Date: Wed, 21 Nov 2012 17:36:19 +0001
> From: se
Well today I tested various builds on mac including official release build
and buildbot build, where none had numpy in it.
On windows I tested the same.
I also tried to build numpy for mac with python 3.3 and 3.2, and both were
unsuccessfull.(so on mac i didn't get it running at all...)
On w
Hey Jason,
currently there is a problem with n-gons when chalculating the center of mass.
The issue is that we don't have a good way of calculating centroids for n-gons
and just use the vertex average, so the calculated origin will drift a little
when your polygons have many vertices.
I'd prefer
On Wed, Nov 21, 2012 at 12:01:49PM +0100, Vilem Novak wrote:
> Several months ago,
> there was a discussion on this list, where integrating numpy was agreed
> with.
Not sure what the state of numpy integration with blender is, but if
you have numpy installed (for the correct version of blender'
Hi Vilem,
Are you sure it didn't happen?
I never build with numpy because I don't have it installed, but I
always get a warning from CMake saying that NumPy will not be
installed due to my system. Which makes me assume Blender itself
incorporated numpy.
Have you tried one of the official test bui
It's pretty easy to handle:
http://www.pasteall.org/37405/python
Am 21.11.2012 04:55, schrieb Brecht Van Lommel:
> Tessfaces includes both triangles and quads, it doesn't triangulate
> entirely, so you'd need to handle the quads still.
>
> Brecht.
>
> On Wed, Nov 21, 2012 at 4:30 AM, Ummi Nom w
David,
I've commited your patch. At least it can not make AAC/AC3 worse :) Also
seems the same change was needed for Vorbis with FFmpeg-1.0.
On Tue, Nov 20, 2012 at 9:23 PM, Sergey Sharybin wrote:
> Hi David,
>
> Thanks for the patch. From quick glance seems ok. Will do some more tests
> later
Several months ago,
there was a discussion on this list, where integrating numpy was agreed
with.
However, it obviously didn't happen, probably due to lack of time on the
side of platform maintainers.
I am now developing a CAM addon for blender, and numpy would really speed up
a lot of computati
Am 20.11.2012 00:43, schrieb Gaia:
> Sure.
> Any improvement of the Collada module is welcome and appreciated :)
> Please can you add the patch to the patch tracker at
> http://projects.blender.org/ ?
> Then i will take a look on it.
>
> thanks!
> Gaia
>
>
> On 20.11.2012 00:37, Daniel Jungmann wro
18 matches
Mail list logo
