So to keep it short do I understand it right that it is completely up to me what scale the models should be yet it is best to interpret a vector like (1 , 0, 0) as beeing exactly 1 meter long if I want accurate physics and becouse the “PhysicsSpace” in JBullet works with that scale by default?
I use DeleD 2.45CE and Eclipse for my first steps which seems to work quite good. Should I make my models from the beginning in the right scale which means for example that the dimesion of a car should be let’s say randomly (4.2, 1.3, 2.1)? Or is it better to work with large numbers like (420, 130, 210) and then scale the model down when exporting to ogre mesh?
In DeleD I’m wondering why the default grid sizes are power of 2 e.g. 64 should I set this to small size like 0.01 for centimeter accuracy (this question applys to other modeling tools too)?
Sorry for my english and thanks for helping me in andvance.
normally, if you only have large objects and try to simulate a solarsystem you should probably try a different sclae due to float problems.
whatever you and your art pipeline prefer/capable of
because you often have large models and start with low detail then go on to finer, jsut set it to 0,01 and your fine tho (dont set the size too high as it eats quite some performance else)