Lighting rendering quirk

srmalloy

At the SE side of the Russalfield Wayshrine structure, at 44.70x47.12 (summerset/summerset_base), there is an oddity with the lighting. That position is atop the stonework platform of the wayshrine. Below it, just off the path, is a lantern illuminating the path. Oddly, though, despite the pavement of the wayshrine platform being facing away from the lantern, as well as being blocked by the side of the platform itself, the platform surface is rendered with the light from the lantern shining on it.

Because this is a purely graphical issue, and has zero effect on gameplay, I have no expectation that this will ever get fixed; it's just intellectual curiosity that makes me wonder whether this is a bug in the game's rendering engine or an artifact of the zone design, where the devs put the origin point for the illumination source of the lantern high enough that it is properly reflecting off the pavement.

Mystrius_Archaion

Lights in many games, definitely this one, work like balls of paint. They show on both sides of objects that are within the range of the lighting effect.
This is because it is not actually using ray tracing and is instead using a "ven diagram" sort of method of determining what gets lit. It is probably something that can be coded to check the direction, but maybe that isn't possible in this game.

It definitely is jarring and a little annoying.

srmalloy

Mystrius_Archaion wrote: »

This is because it is not actually using ray tracing and is instead using a "ven diagram" sort of method of determining what gets lit. It is probably something that can be coded to check the direction, but maybe that isn't possible in this game.

The cross product of the face normal for a polygon (needed to tell which way a poly is facing) and the vector from the light source to the poly would tell you whether the polygon is illuminated by the light source. That should be a primitive function in the hardware. It's just curious that it wasn't implemented; you have the same computations going on when you have the player and a light source with the corner of a wall between them; the wall facing the light source will be illuminated by the light source and the wall facing the player won't. It may be an oversight in the design of the game engine, though, in which case, it would be painful to fix at this point.

Mystrius_Archaion

srmalloy wrote: »

Mystrius_Archaion wrote: »

This is because it is not actually using ray tracing and is instead using a "ven diagram" sort of method of determining what gets lit. It is probably something that can be coded to check the direction, but maybe that isn't possible in this game.

The cross product of the face normal for a polygon (needed to tell which way a poly is facing) and the vector from the light source to the poly would tell you whether the polygon is illuminated by the light source. That should be a primitive function in the hardware. It's just curious that it wasn't implemented; you have the same computations going on when you have the player and a light source with the corner of a wall between them; the wall facing the light source will be illuminated by the light source and the wall facing the player won't. It may be an oversight in the design of the game engine, though, in which case, it would be painful to fix at this point.

Sounds about right. I'm betting it's just not something they can easily do and money is involved due to time also.