Virtual Models

Section comments

In general xvc2jvx handles most of the enclosed examples quite well despite of significant differences between MuPAD and JavaView concept of coordinate systems and grids. However, there are a few issues in this set of examples that are worth to look into.

1. I guess, this is difficult or perhaps even impossible to extract information about density of ticks and numbers on the axes rulers if they were done automatic. However, when user defines explicitly these parameters it would be good to transfer these parameters to JVD file. I am sure that this problem can be more difficult than it looks on the first view - users can define only some of these parameters and some other will be chosen for them automatically by VCam.
2. Even more difficult would be to establish a good match between MuPAD and JavaView grids as they use completely different concepts. Some possible improvements are shown in the enclosed examples. However, due to major differences in the concept, most of these improvements should be done by hand.
3. The most difficult is the problem of graphs of functions with very large values, for example tan(x), 1/x, etc. These examples need a special treatment - establishing a very accurate viewing box and transforming it into camera clipping in JavaView. I guess, this is not easy task when VCam chooses clipping ranges but this should be possible when user declares his own viewing box.
4. At the moment there is still a bug in converting viewing box from MuPAD to camera clipping JavaView