Large transparencys and C2

[GregHart1965]

One thing I have noticed is that C2 doesn't like very large areas of transparency. I am building a cable car, complete with cables, and an antenna with cable stays. The cables for the cable car are over 1km long. In C1 it was better to create the cables on a flattened transparent cube in order to keep the number of vertices low. In C2 this places a huge load on the graphics engine as it tries to figure out the shadow and then it casts a shadow for most (all?) of the transparent sheet. This load may not be apparent on computers with a strong graphics card but on my small laptop (WAAAAY below minimum spec - yet C2 still works!) it is painfully obvious with frame rates dropping below usable levels.

Replacing the cables with a solid piece (three sided cylinder made very skinny and long) removes the load on the graphics engine and makes clear accurate shadows. It helped a lot on the antenna so I am going to do the same on the cable car. (Will write here the results).

Anyway, this is just one difference Ive noted that implies building things a bit differently in C2. The smaller areas of transparency make little difference to FPS so for example leaving the tower of the antenna as four flattened cubes (rather than building the entire complex structure) is fine but not with something as huge as the cables - those are better built as solid units for C2.

PS: I built the transparent sheets without making the edges hard - this is incorrect. It is always better to have hard edges.

Shot54.jpg

[GregHart1965]

I converted my cable car cables to solid objects (mission!). The results are very nice, the cables are easy to see, the strange shadows of the transparency are gone BUT the problem of very large objects persists in that they use up a lot of resources. I had to remove a lot of other objects to make the FPS workable, flying near the cable car, but it works in the end.

Shot59.jpg