(Fix) Rendering intermediate levels for very big images no longer sometimes fails. (We are not sure this ever happened in production on any system, we could only see the failure on a synthetic proof-of-concept test that we created after discovering a defect in our code.)
(Fix) Projecting an image no longer fails if the target image has tiles in more than one XY quadrant. (The majority of operations automatically adjust the coordinate system of the produced image so that the image is entirely in the positive XY area, which is why this bug was occurring rarely.)
(Fix) Computing statistics for a big image no longer sometimes crashes.
The Style pane used with a big image no longer tries to compute statistics for the entire image and instead computes statistics from intermediate levels. (This greatly improves the usability of the Style pane with big images without losing much in terms of accuracy. We will likely add separate means to compute statistics for the entire image from the UI in the future.)
(Fix) Reading DWG R2010 and later correctly processes splines.
New transform templates for interpolating vector data:
- Interpolate, Gravity - uses TileInterpolateGravity,
- Interpolate, Kriging - uses TileInterpolateKriging,
- Interpolate, Kriging with Median Polish - uses TileInterpolateKrigingMedianPolish,
- Interpolate, Triangulation - uses TileInterpolateTriangulation,
- Interpolate, Triangulation with Segments - uses TileInterpolateTriangulationSeg.
All added templates allow specifying margin in drawing coordinate system units (by default set to 0), and resolution as a multiplier of drawing coordinate system units (by default set to 1). (We will extend the templates to both display the coordinate system unit in use and allow changing it in the future.)
All added templates allow using multiple threads both for preparing interpolation data (except for gravity which does not benefit from multiple threads here) and for performing actual interpolation.
TileInterpolateGravity and TileInterpolateKrigingXxx query functions interpret provided (positive) radius and auto-computed (zero or negative) number of neighbors as "use all neighbors within specified radius". (This covers all possible combinations of radius and the number of neighbors being provided specifically or specified to be auto-computed, the other combinations are: both radius and neighbors auto-computed = Voronoi neighbors, specific number of neighbors but radius auto-computed = radius set by the system according to global interpolation picture, specific number of neighbors and specific radius = the system uses specified values.)
End of list.