Well first of all it's only one render, using multiple cameras, all together smaller than one full viewport. Additionally, logically the number of cameras can be reduced to 3, because you never see more sides of a cube than 3 at a time.
Finally, if you make a cubemap, you may use it for all reflecting things, so your level may contain water, marble, chrome etc. all done with one cubemap. I think it may be worth to do that for some kind of game fx.
But of course I agree, realtime cubemapping should only be used when it makes sense. Eg. in a game like Reds Sub Game.
Finally, if you make a cubemap, you may use it for all reflecting things, so your level may contain water, marble, chrome etc. all done with one cubemap. I think it may be worth to do that for some kind of game fx.
But of course I agree, realtime cubemapping should only be used when it makes sense. Eg. in a game like Reds Sub Game.