if you could let me know whats going on that would be grate thx
Stencyl 3.5, currently in beta, uses the new OpenFL which brings with it better shader support. Even the current tools (such as Image API) work better--as far as I've tested, anyway. So that will open new doors.
Lighting was my biggest challenge by far--that's just the nature of the beast. What ultimately helped me was to just sit down with a pen and notebook and work out all the math behind what I wanted to accomplish. Efficiency is the name of the game, which could also be a problem with any solution Stencyl provides. When the Stencyl toolset has to decide between efficiency or ease of use, it goes with ease of use. When it has to decide between efficiency or generality, it goes with generality. Because Stencyl has to provide tools that work across a broad spectrum of games, it generally won't provide anything optimized for one specific game.
Usually, you can't have efficiency and generality at the same time. In most cases that's not a problem. An unoptimized basic platformer will run 60FPS just as an optimized one. But because of the number of per-frame calculations, lighting is an exception.
To work well, and fast, lighting needs to be optimized for your specific needs. Stencyl and OpenFL do provide the tools you need. I documented all the math problems involved in designing my system.
Take a look if you think it'd be helpful--
starting around Page 13 of my dev journal. It's a 2.5/3D lighting system. 2D math would be simpler, but the same problems (shadow casting, clipping polygons, etc.) still apply. Early result: