ST-CS-10-339-75
March 2002

2.  Consequences of the BSP File Format  (U)

The crucial issue determining projectile penetration behaviour is the number of BSP cubes a projectile passes through when inside an object. The actual length of the object is totally and completely irrelevent.

An AWP, the Scout and H&K G3 can penetrate two BSP cubes. Assult rifles, the M249 machine gun, the SSG-550 and the the Desert Eagle pistol can penetrate one BSP cube.

It must be emphasised that the actual visible length or size of an object is totally and completely irrelevent.

Because the map creation tool tries to kept roughly the same number of objects in each cube when making a split, adding a single new object anywhere in the map can alter the location and sizes and the BSP cubes throughout the entire map.

Imagine the very first split during the BSP creation process. If adding a single new object on the map makes the BSP creation tool choose to perform a fourty-sixty split rather than a fifty-fifty split, then the size and location of all the BSP cubes in the map will change.

This phenomenon is demonstrated in the reproduced footage.

The first map, te_pen, has a single extremely long block with a hostage at one end and a firing position at the other end. It is possible to shoot through this entire block and eliminate the hostage.

The second map, te_nonpen, is totally and completely identical to te_pen except that one additional block (as seen in the footage, beside the long block at the hostage's end) has been added to the map. This block is totally seperate from the block which was penetrated in te_pen; it is not in the line of fire of the projectile fired from the AWP. As can be seen, it is no longer possible to penetrate the block. This is because the change in BSP cube distribution has increased the number of BSP cubes inside the long block.

Both maps are available for download in appendix 1, in BSP and RMF format.

In summary, the actual length of an object has absolutely no relationship to its ability to sustain projectile penetration. As such, estimating the length or thickness of an object with a view to guessing whether penetration will occur is completely false and misleading. It is impossible to know from visual inspection whether or not an object can be penetrated.