This is the model I am working on, in shaded and ghosted display.
After I do a Boolean differfence (I am subtracting the red objects from the large object), this is what my display looks like.
As you can see, the model seems completely transparent, I can see the grid through it. But isocurves are still present. However something seem wrong in this display. No matter whether I choose shaded or rendered display, the model stays transparent.
However when I execute the command SelOpenPolysrf nothing shows, and when I do Analyze > Edge Tools > Show Edges, no naked edges are shown.
When I export the model as an STL file, and then open it again, everything seems fine.
Is this normal or is something wrong? Or, more likely, am I doing something that causes this to happen?
rhino for mac has been doing this for as long as i can remember… on anything but the simplest of trims or booleans, the display mode switches to wireframe (though yours seems to show some surfaces so that’s new to me)…
did you try switching back to shaded or ghosted after the boolean?
Hi Zews, my guess is that the intersection of the two objects has one or more loops in the curve someplace - the result is that the inside and outside of the object are not clearly defined - can you post the ‘pre-booleaned’ objects?
Here is the file with the Pre Boolean objects. Yes, I did try to switch back to shaded, ghosted and rendered after the boolean. That did not change anything.
What does “has one or more loops in the curve” mean?
Jeff, I think what you are showing in the file you uploaded is something very different. What seems to happen there is that instead of trimming all the cylinders at once, you have to pick them one by one, even though all the cylinders are in a group. But when you split, the group stays intact. I do see that the display changes from shaded to wireframe. I have noticed that also on several occasions doing different things. I never mentioned it here, because it did not bother me much.
This is different though, because when I assign a material to the model after the boolean difference, the top and bottom surfaces also do not get a material applied to it. And my rendering then looks like this. Notice how the area that has been removed with boolean difference looks fine, only to top and bottom have disappeared.
Jeff, I think what you are showing int the file you uploaded is something very different. What seems to happen there is that instead of trimming all the cylinders at once, you have to pick them one by one, even though all the cylinders are in a group.[/quote]
Split them then instead of trim (or window select if Trimming)… the display mode will switch to wireframe… it won’t switch with simple trims such as one plane / one cylinder… doing all the cylinders at once will show the issue.
they should be open… using BooleanSplit instead of split will close them.
Pascal, thank you! That is interesting. How did you know where to look?
I made this bad polysurface as an extrusion from a polycurve. The polycurve was created from a number of curves after having done a number of trims on the original curves. It turns out, that the original curves were not intersecting accurately.
When after all the trims I did “Join”, command history in the lower left corner said “20 curves joined into one open curve”, so I knew I would not be able to extrude this curve into a solid without naked edges. When I could not find where the opening(s) was/were (it turns out there were four), I did “Connect”, thinking that this solved the issue. I never checked the object details.
When I then did an extrude, I got the following message window.
I clicked “Yes”. But because I had gotten this message I did SelOpenPolysrf and Analyze > Edge Tools > Show Edges to make sure everything was OK. Both commands did not yield anything bad, i.e. no naked edges, so I thought everything was fine.
When I then did the boolean difference, the result was that the top and bottom surfaces disappeared.
It turns out that I should have been more careful/precise when I added those two circles at both ends of the object. I redid it, and now everything is fine. Open curve.3dm (56.1 KB)
Since the thing was closed and the normals appeared to be correct, I used the ‘bigger hammer’ approach - I kept cutting the object in half and deleting the part that then shaded correctly until I had just a bit that I could inspect more carefully.
<img