Closed GoogleCodeExporter closed 8 years ago
This should have solid representations of the tools and the holders
Original comment by danhe...@gmail.com
on 16 Jul 2009 at 9:16
Dan,
I don't know if this would be applicable to this issue, but perhaps something
could
be learned from the code... Blender (www.blender.com) incorporates a fairly
sophisticated sculpting tool
(http://wiki.blender.org/index.php/Doc:Manual/Modelling/Meshes/Sculpting) for
use
with its mesh geometry. This is basically an additive/subtractive geometry
algorithm
driven by user mouse/stylus input. If this input were changed to be fed by
G-Code
and the sculpting shape/mode adjusted to reflect the machine tool, could this
be
used to simulate subtractive machining?
Sorry if this is totally unrelated or unusable... I don't know much about
coding or
the inner workings of what you're considering. I'd love to see something like
this
incorporated, tho.
Cheers,
-Kelly (mzungu)
Original comment by kelly.la...@gmail.com
on 28 Jul 2009 at 8:43
Kelly,
The documentation says
"Add causes the brush to pull an area of the model in the positive direction,
Sub in
the negative direction."
This just distorts the mesh, I think. What we need is to be able to subtract any
solid from any other solid, where the solids are made from triangles. Is there a
function in Blender to cut one solid from another?
Original comment by danhe...@gmail.com
on 29 Jul 2009 at 7:37
[deleted comment]
Dan,
There are pretty good boolean functions for meshes in blender:
http://wiki.blender.org/index.php/Doc:Manual/Modifiers/Mesh/Booleans -
implemented
both as a mesh modifier and a separate tool, as I understand it.
I would guess that these would have to be quite optimized in order to be
efficient
for an animated feedback. The thing that pops into my mind is to do the
subtractions
by g-code block, rather than on a time-based algorithm.
Utilizing the modifier stack to do the boolean subtractions would be very
interesting, as it would allow the "backing up" in time, & etc. But, knowing
that
the stack is not designed for this, it may not have the capacity to pull it off.
Again, this may only be useful to you to take a peek at their source code, then
apply that knowledge to your situation here.
Hope this helps!
-Kelly
Original comment by kelly.la...@gmail.com
on 29 Jul 2009 at 2:34
Dan,
I just recalled another resource that may be of benefit to this project (and
the
HeeksCAD one, obviously) in general and to your interest in booleans (CSG) in
particular: AYAM (http://ayam.sourceforge.net/) is open source under the BSD
license. There may be some code in there to glimpse.
It also incorporates NURBS functionality, which would totally rock to have in
HeeksCAD. (For a great implementation of NURBS, have you tried out MoI
(http://moi3d.com/)? Its written by a former Rhino coder, is cheap, and has
access
to the latest beta for download & try, I believe.)
Meh, perhaps for future dev...
Cheers,
-Kelly
Original comment by kelly.la...@gmail.com
on 29 Jul 2009 at 3:22
I stumbled across Ayam the other day when looking at one of the libraries it
uses.
Ayam uses the OpenCSG library to quickly render the CSG operations.
http://www.opencsg.org/
Perhaps that help, perhaps it dosen't.
Original comment by geo01...@gmail.com
on 29 Jul 2009 at 4:06
Oh, I suppose would would also need something like the GTS library to actually
do the
CSG operations in the first place.
Original comment by geo01...@gmail.com
on 29 Jul 2009 at 4:08
I think the first thing to look at would be GTS, as it is likely to be the most
simple and so I can hope that it might be fast.
It looks like Ayam is NURBS based, so it won't be optimized for fast booleans
with
triangle models.
This conversation should be in issue 132 ( Simulation - Solid )
http://code.google.com/p/heekscnc/issues/detail?id=132
Original comment by danhe...@gmail.com
on 29 Jul 2009 at 6:59
no new features, just bug reports in this issue list
Original comment by danhe...@gmail.com
on 24 Mar 2014 at 3:35
Original issue reported on code.google.com by
danhe...@gmail.com
on 16 Jul 2009 at 8:50