Sorry your browser is not supported!

I'm currently working on the engine for a game and have per pixel lighting working using GLSL. I was thinking about adding deferred lighting, but it would be a lot of work considering I would have to read a bunch of papers just to understand it. However, do you think that deferred lighting is required for a modern day indie game? The graphics that I'm shooting for are like Animal Crossing, so not overly lighting intensive at all. I think that there will be, at most, 5-6 point lights on screen and there would almost always be 1 directional light. I'm still unsure about real time shadows, might just use a projector of a black circle in place of actual shadows. I'm also trying to keep the game as cross-compatible as possible. With all of that taken into account, do you think that deferred lighting is required?

I'm having trouble finding any performance comparisons of deferred lighting vs forward rendering.

Deferred shading offers 'better' performance with more lights and more options than the alternative.

I believe that with standard pixel lighting, for every light that effects the object, the object must be re-rendered, making the performance cost be equal to No.Lights*No.Objects.

Deferred shading does some wizardry with something called a g buffer to calculate all the lighting before rendering the objects, so the performance impact is equal to No.Lights + No.Objects

For a game that doesn't focus on lighting, I don't think the work required would be worth it.