Quote:
My point is that all of the CoD screenshots have felt slightly 'lightweight'. The reason for this is that they contain very little true black, and the eye sees a lot of black in shadows in RL.
The effect you end up with is a slightly super-real look where you can see details in shadows that you wouldn't be able to in RL.
|
The human brain is remarkably capable of resolving objects in shadow as well as in light, even at the same time. (ex, sky behind some trees).
Cameras on the other hand, don't and make shadows which are totally black, or light which is totally white and has no detail. A photograph would show a black silhouette of trees on a blue sky or show some nice green trees on a white sky. The only way to mitigate this problem is to take the same photo two or three times with different exposures and composite them for the best results
I think Oleg's code is simplifying what our brain does automatically without our perceiving it, so the result is low contrast overall. Which is fine, I think. I imagine it would be difficult for a computer to know whether your eyes are glancing to the instrument panel or outside at any given moment (like in the screenshot where both are visible at once). Currently, the mouse just moves our "head", which would lead to some awkward situations I think, including having to move the mouse around much more to "see" everything.
@SpecJockey - I agree with your points, and with Oleg's philosophy on the matter. I prefer a realistic shot than "filtered" ones, but I imagine the possibility to fiddle with effects is already implemented or in the works (I recall Oleg wanted movie makers to be able to the engine for truly cinematic sequences).