This ProtocolZero query uses the graphical abilities of Quake 3 to render the world as seen from the bot's perspective, much like vc. Unlike vc, however, it returns the contents of the OpenGL depth component buffer instead of the color buffer. Larger values are further away, while smaller values are closer to the camera plane.

See vc for some additional notes on vc and vd.

Parameters: None.

Responds:

• w h DATA where w is the width of the image, h is the height of the image, and DATA is w*h*4 bytes of depth information, formatting according to GL_INT. I believe this typically means that they are little-endian, but you're best off determining it for yourself. The first 4 bytes belong to the lower-left corner of the image, and they proceed to the right (and then up when the row is filled).

Reports:

• done when the task completes.
• popped if popped by po.
• forgotten if deleted by fa or fm.
• replaced if replaced by a command scheduled with 'r'.

Example:

• User sends: "n vd 512"
• Bot replies "rs vd 512 640 480 5q'5hn?o1qin716[32\-..."
• Bot replies "cp vd 512 done"