Changeset 1918
- Timestamp:
- Oct 14, 2005, 11:24:54 AM (20 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
inundation/wiki/least_squares_refactoring.txt
r1915 r1918 66 66 situation) 67 67 68 Fitting 2million points to a one million triangles mesh will not cause a68 Fitting 5 million points to a one million triangles mesh will not cause a 69 69 memory error. 70 70 … … 131 131 132 132 *To reduce the time spent building matrix A. 133 133 134 Change the expansion algorithm, when determining what triangle a point 134 is in. It should look in the cell of the point and then all adjacent 135 is in, when none of the vertices are in the cell the point is in. 136 137 Options (still thinking about this one) 138 * It should look in the cell of the point and then all adjacent 135 139 cells. If it isn't there then ... we have problems. My feeling is it 136 140 has to be there. 141 - Building the structure to keep this info could be prohibative. 142 137 143 138 144 Another speed up option is to write the bottlenecks in C. Do it if necessary.
Note: See TracChangeset
for help on using the changeset viewer.