comment out APPLE_client_storage extension usage, I think this was messing some stuff up on OS X and causing memory errors
fix handling of cubemap data so that they can be read properly (can't believe it took so long to notice to an obvious and stupid bug)
be sure to un-bind textures during preload to avoid freaky little errors later
ifdef an error message from the framebuffer error code, Apple doesn't have this one define in their headers for some reason
restore lighting falloff capability from retail and non-HTL modes (only used for asteroids as far as I know)
various cleanups and speedups for dealing with lights
handle window/fullscreen/minimize changes better, fixes cursor handling mostly (Mantis bug #1146)
don't flush on gamma change, we don't really need to any more
remove two old/unused MR flags (MR_ALWAYS_REDRAW, used for caching that doesn't work; MR_SHOW_DAMAGE, didn't do anything)
add MR_FULL_DETAIL to render an object regardless of render/detail box setting
change "model_current_LOD" to a global "Interp_detail_level" and the static "Interp_detail_level" to "Interp_detail_level_locked", a bit more descriptive
minor bits of cleanup
change a couple of vm_vec_scale_add2() calls to just vm_vec_add2() calls in ship.cpp, since that was the final result anyway
deal with a index_buffer memory leak that Valgrind has always complained about
make HTL model buffers dynamic (get rid of MAX_BUFFERS_PER_SUBMODEL)
get rid of MAX_BUFFERS
make D3D vertex buffers dynamic, like OGL has already done
- add some basic bmpman functionality to grstub, since it needs to do something at least
- add missing gr_* function ptrs to grstrub
- (re-)enable radar and hud setup functions that used to crash (problems are fixed now)
- deal with default pilot file properly (also caused a bmpman headache)
- don't bother with Multi_common_icons[] in standalone mode (they don't load, so don't let them unload either)
clean up view/proj matrix fubar that made us need far more matrix levels that actually needed (partial fix for Mantis #563)
add debug safety check to make sure that we don't use more than 2 proj matrices (all that GL is required to support)
set up a texture matrix for the env map to that it doesn't move/look funky
move color and shader functions to 2d.cpp since they are exactly the same everywhere
don't bother with the function pointer for gr_set_font(), it's the same everywhere anyway
various lighting code cleanups
- try to always make sure beam origin lights occur outside of model
- make Static_lights[] dynamic
- be sure to reset to first 8 lights when moving on to render spec related texture passes
- add ambient color to point lights (helps warp effects)
- sort lights to try and get more important and/or visible lights to always happen in initial render pass
- disable depth buffer for RTT, may be causing the ATI issues, should work ok without it for most things
- disable mipmap generation for RTT, something strange is going on there, it's sometimes making black mipmap levels (even for the same texture)
- go back to using minimize mode which non-active, but doin't minimize when Fred_running
- remove temporary cmdline options (-spec_scale, -env_scale, -alpha_alpha_blend)
- change FBO renderbuffer link around a little to maybe avoid freaky drivers (or freaky code)
taylor