STL vs OBJ
I created a fairly simple object to use as input to a sculpting program
(Sculptris). Since Sculptris only accepts OBJ files as input, I
converted from STL to OBJ in NetFabb. The result is puzzling. Much of
the surface is as one would expect, but there are visual defects
(surfaces with unexpected divots and bumps), especially at the bottom.
Not sure what to make of it. Perhaps all of the defects were introduced
by NetFabb, and this has nothing to do with OpenSCAD.
Any comments appreciated.
Jon
--
Sent from my desktop computer.
I do not receive emails while away from my desk,
nor do I receive texts on my main phone number
(which is a land line).
If you know that I am on the road, please text me.
If you know that I am home, please email me.
I was able to load both objects into Meshmixer. Neither object raised any
indication of problem with the Analysis/Inspector. I zoomed into the surface
in question and found a flat surface. When I performed a test by making an
Edit/Plane cut as closely to the bottom as possible, the resultant item was
a flat surface, no divots or bumps or problems.
I was able to load the object into Simplify3D and when running the slicing
preview, it was also flat at 0.100 layer height.
I'm inclined to agree that NetFabb is generating a problem, but not
necessarily a real one so much as a virtual one.
One curious aspect of the model is that the flat undersurface is a series of
very long thin triangles, the bases of which become the circumference of the
circle, while the other two legs are chords of the circle. I usually see
circular shapes in wireframe mode as a series of smaller triangles evenly
distributed over that surface.
If you have troubles with NetFabb in the future (and the future is NOW),
Meshmixer will export as OBJ easily enough.
--
Sent from: http://forum.openscad.org/
Thank you for your thoughtful analysis. When I take the STL and convert
it into an OBJ and then mess with it in Sculptris, the shape (which
seems fairly uniform if you look at the triangle patterns) deforms in a
variety of ways, as if the surface was inconsistent. Perhaps this is an
issue with Sculptris. I will try another sculpting program.
:)
On 3/7/2018 5:17 PM, fred_dot_u wrote:
I was able to load both objects into Meshmixer. Neither object raised any
indication of problem with the Analysis/Inspector. I zoomed into the surface
in question and found a flat surface. When I performed a test by making an
Edit/Plane cut as closely to the bottom as possible, the resultant item was
a flat surface, no divots or bumps or problems.
I was able to load the object into Simplify3D and when running the slicing
preview, it was also flat at 0.100 layer height.
I'm inclined to agree that NetFabb is generating a problem, but not
necessarily a real one so much as a virtual one.
One curious aspect of the model is that the flat undersurface is a series of
very long thin triangles, the bases of which become the circumference of the
circle, while the other two legs are chords of the circle. I usually see
circular shapes in wireframe mode as a series of smaller triangles evenly
distributed over that surface.
If you have troubles with NetFabb in the future (and the future is NOW),
Meshmixer will export as OBJ easily enough.
--
Sent from: http://forum.openscad.org/
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
--
Sent from my desktop computer.
I do not receive emails while away from my desk,
nor do I receive texts on my main phone number
(which is a land line).
If you know that I am on the road, please text me.
If you know that I am home, please email me.
Reflecting_Surface_Blank_02.obj
http://forum.openscad.org/file/t824/Reflecting_Surface_Blank_02.obj
I've attached your OBJ file, but with a slight modification. I used
Meshmixer to create face groups, then selected only the bottom surface.
Using the remesh feature, I performed two iterations and used a very small
target edge length for the parameters.
Now what used to be long slender triangles has been changed to a series of
tiny ones more uniformly scattered over the surface.
It's possible that your program will look more favorably upon such a
surface, as it lends itself better to organic modeling such as Sculptris
--
Sent from: http://forum.openscad.org/
Jon the long thin triangles are the problem. You need to remesh the
object. not sure if sculptris can do this but ZBrush can.
You need a more even distribution of less sharp triangles.
Thank you. I guess the question is "how/why did OpenSCAD produce these
triangles?"
$fn = 200;
module torus()
translate([0, 0, 20])
rotate_extrude(convexity = 10)
translate([40, 0, 0])
circle(d = 20);
module Main() {
hull() {
cylinder(h = 1, d = 80);
torus();
}
hull() {
torus();
translate([0, 0, 120]) sphere(30);
}
}
difference() {
Main();
for (a = [0:90:359]) {
rotate([0, 0, a ]) translate([135, 0, 90]) sphere(100);
rotate([0, 0, a ]) translate([130, 0, 130]) sphere(100);
rotate([0, 0, a + 10]) translate([145, 0, 45]) sphere(100);
rotate([0, 0, a + 50]) translate([132, 0, 110]) sphere(100);
rotate([0, 0, a + 35]) translate([141, 0, 70]) sphere(100);
rotate([0, 0, a + 63]) translate([143, 0, 60]) sphere(100);
}
}
On 3/8/2018 3:33 AM, Mark Schafer wrote:
Jon the long thin triangles are the problem. You need to remesh the
object. not sure if sculptris can do this but ZBrush can.
You need a more even distribution of less sharp triangles.
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
--
Sent from my desktop computer.
I do not receive emails while away from my desk,
nor do I receive texts on my main phone number
(which is a land line).
If you know that I am on the road, please text me.
If you know that I am home, please email me.
Thank you. It is interesting that Sculptris shows your repaired object
as having inconsistent (not pretty) surfaces.
On 3/7/2018 7:58 PM, fred_dot_u wrote:
Reflecting_Surface_Blank_02.obj
http://forum.openscad.org/file/t824/Reflecting_Surface_Blank_02.obj
I've attached your OBJ file, but with a slight modification. I used
Meshmixer to create face groups, then selected only the bottom surface.
Using the remesh feature, I performed two iterations and used a very small
target edge length for the parameters.
Now what used to be long slender triangles has been changed to a series of
tiny ones more uniformly scattered over the surface.
It's possible that your program will look more favorably upon such a
surface, as it lends itself better to organic modeling such as Sculptris
--
Sent from: http://forum.openscad.org/
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
--
Sent from my desktop computer.
I do not receive emails while away from my desk,
nor do I receive texts on my main phone number
(which is a land line).
If you know that I am on the road, please text me.
If you know that I am home, please email me.
Interesting that a planar object can appear differently depending on how
the triangles are arranged on that planar surface. It is still a plane,
after all...
On 3/8/2018 3:33 AM, Mark Schafer wrote:
Jon the long thin triangles are the problem. You need to remesh the
object. not sure if sculptris can do this but ZBrush can.
You need a more even distribution of less sharp triangles.
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
--
Sent from my desktop computer.
I do not receive emails while away from my desk,
nor do I receive texts on my main phone number
(which is a land line).
If you know that I am on the road, please text me.
If you know that I am home, please email me.
I downloaded and installed Sculptris to see how things developed. The first
item of note is that the import of the OBJ file is not accurate. The
perimeter of the base of the model shows in wireframe in Meshmixer as a
clean circle. In Sculptris, that feature is destroyed. The triangles are
arrayed almost randomly at the interface between the circular base and the
curved portion moving upward.
I also noted that some of the sides are irregular as well.
I think the blame falls on the internal works of Sculptris and not within
any particular slicer or other modeling program, especially not within
OpenSCAD.
--
Sent from: http://forum.openscad.org/
I think I agree. So much for a 4 year old free program!
Thank you all!
On 3/8/2018 9:42 AM, fred_dot_u wrote:
I downloaded and installed Sculptris to see how things developed. The first
item of note is that the import of the OBJ file is not accurate. The
perimeter of the base of the model shows in wireframe in Meshmixer as a
clean circle. In Sculptris, that feature is destroyed. The triangles are
arrayed almost randomly at the interface between the circular base and the
curved portion moving upward.
I also noted that some of the sides are irregular as well.
I think the blame falls on the internal works of Sculptris and not within
any particular slicer or other modeling program, especially not within
OpenSCAD.
--
Sent from: http://forum.openscad.org/
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
--
Sent from my desktop computer.
I do not receive emails while away from my desk,
nor do I receive texts on my main phone number
(which is a land line).
If you know that I am on the road, please text me.
If you know that I am home, please email me.