Problem68.3dm (14.3 MB)
I am stepping back from my previous question. In the attached file I have the side of a hull. I would love to build it all in one piece but I have had to break it up into 8 due to the shape of the outlines.
There are two issue that I see.
- The surfaces need to to be simplified.
- The surfaces need to be joined and smoothed at the seams.
I am wondering what the best approach would be to doing this.
The problem I find is that, if I do not use a lot of frames, I get mismatches at the seams. As it is, the two bow surfaces and the lowest keel surface, have visible discontinuities.
First step - change the tolerance so something reasonable. 0.01 is reasonable for an object 887 units long. (Side question, why did you use a tolerance of 0.00001?)
Second step - use FitSrf to rebuild the surfaces with reasonable numbers of control points. FitSrf with a tolerance of 0.1 results in much simpler surfaces that are will within 0.1 units of your original surfaces.
Third step - Clean up the surfaces with visible discontinuities.
Fourth step - Use MatchSrf to match the surfaces at the edges. Some experiments with different orders of matching may be needed.
1 Like
The precision setting was just was was in the file I copied the surfaces to.
Thanks, I learned a couple of new command
Problem I have with the discontinuities (e.g., the end of the keel) comes from doing a networksrf.
I have waterlines, frames, and outlines for that panel but I cannot use the waterlines for networksrf because it would create a non-grid pattern.
In other places I have been able to keep adding more frames or waterlines until visible discontinuities disappeared.
I am puzzled why they show up at the bow because it uses continuous waterlines.