philip.atha Posted April 30, 2014 Share Posted April 30, 2014 I'm trying to understand the meaning of terrain step. Evidently I'm missing the core concept. I'm considering the units system to be meters for my purposes. If I have terrain source data that's 4096px x 4096px and 1px = 0.854m (real world meters) Does that therefore mean that my 'step' is 0.854? If not, what then is 'step' exactly? The documentation says: Step The size of a grid cell in units. Step controls how big the created terrain will be. If I create a terrain using in this context, my terrain appears relatively smaller than where the rest of my objects should be. Link to comment
ulf.schroeter Posted April 30, 2014 Share Posted April 30, 2014 If I have terrain source data that's 4096px x 4096px and 1px = 0.854m (real world meters) Does that therefore mean that my 'step' is 0.854? Yes. Therefore your terrain total world size (ObjectTerrain bounding box size) should be approx. 3498m x 3498m. You can check the size by selecting your ObjectTerrain instance in Unigine editor (dynamic node info enabed on tools dialog, see screenshot). Radius should then be approx. sqrt( 3498m ^ 2 + 3498m ^ 2 ) = 4947 m. If this is not the case, than there might be another issue (e.g. node scaling). Link to comment
ulf.schroeter Posted May 1, 2014 Share Posted May 1, 2014 @UNIGINE: hm, maybe dumb question, but just by thinking about the above screenshot for ObjectTerrain: the ObjectTerrain centre (= bounding sphere centre ?) is at x=0,y=0 as indicated by the coordinate frame, but the terrain itself only extends in positive x/y direction. Expecting that the depicted radius (= bounding sphere radius ?) encloses the complete terrain, than this bounding sphere would be far too large (inefficient frustum culling) ? Why is the ObjectTerrain bounding sphere centre not at (worldsize x)/2 , (worldsize y)/2 as depicted in the right image ? Link to comment
Recommended Posts