On 12/05/2013 11:52 AM, andy pugh wrote: >> But showing and moving machine and vises is a minor thing compared to >> material removal I think. Although I don't think it is trivial. > I wonder if a voxel-based approach is simpler, but it rather depends > on the required precision. > If 1mm voxels on a 100mm cube give a good enough preview then you only > need 125kB of data. > If you want to simulate 1' cubes to 0.001" then you start looking at > 200GB of data.
The voxel approach is a valid one. You can reduce the data-set size by merging voxels in a plane and volume. There are tree-algorithms to handle such cases and there is an advantage that you only need to split, never merge. However, using trees can be computationally (a lot) more work. But then, it can be used for very high resolution and can be parallelized quite well. -- Greetings Bertho (disclaimers are disclaimed) ------------------------------------------------------------------------------ Sponsored by Intel(R) XDK Develop, test and display web and hybrid apps with a single code base. Download it for free now! http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
