I'm just trying to understand the current system, to see if it could be used to make something that doesn't need attachements. My best idea so far is to simply have the box with all possible single-launcher tube counts bezels only occupying the bottom third, and then having the secondary and third launchers occupy a bezel within that box that cover the middle and top third respectively. Something like this:
There would be a red model with the box and a bezel for 1, 2, 4, 5, 6, 10, 15 and 20 tubes, and green and blue models that are only the bezels with 0, 1, 2, 4, 5, 6, 10 tubes (+ one with 15 for the green). Maybe the model count could be brought down if some combinations were merged, e.g. 2+2 would only show the red box with 4 tubes and green and blue with 0, same as 1x4. Not sure if the current system can do this, but the lack of 4 and 5 tube boxes makes me believe it can.
Simply having one bezel for all possible tube counts would be too much. Of course, this could be done with texture animation instead of requiring a model for each tube count, but this would probably mean code changes. The solution above could work with the current system, look reasonably good and require about 23 models, 5 more than the current solution, if no merging could be done, otherwise even less.
Edited by zagibu, 31 March 2014 - 03:54 PM.