 | | Math Help Needed PLEASE
|
|
Hy, I'm running into math problems again, after I recogniced whats the problem with the movements in MLCad. Now here is what I would like to do: There is a virtual grid which is bound to a part. The distance of the grid lines is given by a vector (...) (24 years ago, 19-Jun-00, to lugnet.cad.dev)
|
| |
| | Re: Math Help Needed PLEASE
|
|
(...) What do you mean by "translated into model space" ? Back from World coordinates? By applying inv(VM)? How can you find a z value for the screen point. There are many possible values. (How do you go from 2D to 3D?) Assuming that you could find (...) (24 years ago, 19-Jun-00, to lugnet.cad.dev)
|
| |
| | Re: Math Help Needed PLEASE
|
|
(...) Transform PS from model space to unrotated part space. Round the x, y, z values to the nearest grid points. (add half a grid size, divide by grid size, multiply by grid size) Transform PS back to model space. I think this algorithm may be (...) (24 years ago, 19-Jun-00, to lugnet.cad.dev)
|
| |
| | Re: Math Help Needed PLEASE
|
|
How embarrassing. Please ignore my previous post. Apparently Lars beat me to it and even brought up a point I missed. Now I have to go figure out how to retract a message. Don (24 years ago, 19-Jun-00, to lugnet.cad.dev)
|
| |
| | Re: Math Help Needed PLEASE
|
|
(...) Ok I got it partly somehow running, but currently I'm fighting with matrixes which do not have an inversion-matrix, I got to algorythms on how to get the inversion-matrix and in some rare cases they return different values, and tell me that (...) (24 years ago, 19-Jun-00, to lugnet.cad.dev)
|
| |
| | Re: Math Help Needed PLEASE
|
|
Michael Lachmann <michael.lachmann@maxmobil.at> wrote in message news:FwEsqI.Mop@lugnet.com... (...) matrixes (...) the (...) tell (...) one (...) also by (...) I think either your inversion routine is broken, or your rotation routine is broken, (...) (24 years ago, 20-Jun-00, to lugnet.cad.dev)
|
| |
| | Re: Math Help Needed PLEASE
|
|
:-) Much better ... the compution of the determinant of a matrix was wrong ... this explains also some strange effects with bfc ... Thanks anyway! Michael (...) (24 years ago, 20-Jun-00, to lugnet.cad.dev)
|
