(locked because this probably wont be possible.)

[Rally]

one thing that affects performance of static shapes is how big it is. having many big pieces may actually be okay on the engine as long as no single piece is too big. this is worth looking into, since knowing the difference in performance of one big shape vs multiple small ones could mean porting maps like this could work.

do you think it would be possible/practical to split the map down into a bunch of sections and then scope them to clients individually with Selective Ghosting as they're needed?

[Ad Bot]

[Fire Vine]

do you think it would be possible/practical to split the map down into a bunch of sections and then scope them to clients individually with Selective Ghosting as they're needed?

that could work

[Smallguy]

one thing that affects performance of static shapes is how big it is. having many big pieces may actually be okay on the engine as long as no single piece is too big. this is worth looking into, since knowing the difference in performance of one big shape vs multiple small ones could mean porting maps like this could work.

I remember running into that issue this summer. I'd love to do more research into what BL can actually handle in different scenarios.

EDIT: OP, another issue with porting complex geometry to blockland is dealing with collision since only convex geometry is supported. For complex models I would just separate each face into its own collision plane but with too many faces blender will just crash when attempting to split the model. I'll get good performance on my Altar of Storms map which has 2000 collision faces in a tight area but bad performance with a large terrain map of 800 collision faces. Keep in mind the loading time is terrible in either scenario

[Mouse droidz 21]

forget about GTA SA that's just not happening, the map is too big. As for vice city and gta 3 something could be done