Ticket #557 (closed defect: fixed)

Opened 3 years ago

Last modified 3 years ago

Reworked lighting model

Reported by: kozmo Owned by: kozmo
Priority: normal Milestone: 0.3.3
Component: engine Version: trunk
Severity: normal Keywords:
Cc: Blocked By:
Blocking: Time planned:
Time remaining: Time spent:

Description

The current lighting model introduce confusion. We have two different lighting settings, yet sometimes giving the same results. There's few settings that could be settable but they aren't exposed by the interface. Some of them shouldn't be touched by the end user since it's too low level.

Change History

comment:1 Changed 3 years ago by kozmo

(In [3721]) * Removed unnecessary alpha value for lighting as well as alpha reference for alpha test

  • Removed appropriate fields in light editor
  • Reworked light model to be compatible with new render backend
  • Changed layer z offset computation for new render backend to use more precisely part of depth buffer as it's not linear
  • Added the ability for guichan to create atlas from its images on runtime - this is only applicable when using its image loader (see engine/python/fife/extensions/pychan/compat.py) refs #557

comment:2 Changed 3 years ago by kozmo

(In [3722]) * More work on lighting model - more uniform interface refs #557

comment:3 Changed 3 years ago by kozmo

(In [3729]) * Demo rio_de_hola now uses atlases

  • Modified addFrame method animation to reflect recent changes for lazy initialisation
  • Modified AnimPacker? tool to support mixed objects (with static images and actions) and to write correct xml header
  • Added a statusbar to Atlas Creator indicating current atlas size
  • Fixed a bug with making directories not correctly (parent directories) in case of disassembling atlas
  • Further changes to lighting model (settings from python extensions) refs #557

comment:4 Changed 3 years ago by prock

  • Status changed from new to closed
  • Resolution set to fixed

closed

Note: See TracTickets for help on using tickets.