prefect 695d966788 ia32_memcpy now returns the destination pointer just like the standard ...

library version does.
Fixed improper non-use of delete[] in the new model code.

This was SVN commit r3060.

2005-10-30 21:01:42 +00:00

..

graphics

profiling indicates memcpy is a bottleneck, so globally replace all of them with memcpy2 (hand-optimized). depending on transfer size, this is 10%..300% faster. also made memcpy2 into a macro instead of thunk function because vc7 was still generating a jump.

2005-10-30 16:19:20 +00:00

gui

Extension function pointers are renamed from glXyzARB to pglXyzARB

2005-10-30 01:18:15 +00:00

i18n

profiling indicates memcpy is a bottleneck, so globally replace all of them with memcpy2 (hand-optimized). depending on transfer size, this is 10%..300% faster. also made memcpy2 into a macro instead of thunk function because vc7 was still generating a jump.

2005-10-30 16:19:20 +00:00

lib

ia32_memcpy now returns the destination pointer just like the standard

2005-10-30 21:01:42 +00:00

maths

renderer: load/unload water textures (similar to alpha maps); do that in delay load and avoid leaking them.

2005-10-12 04:16:41 +00:00

ps

profiling indicates memcpy is a bottleneck, so globally replace all of them with memcpy2 (hand-optimized). depending on transfer size, this is 10%..300% faster. also made memcpy2 into a macro instead of thunk function because vc7 was still generating a jump.

2005-10-30 16:19:20 +00:00

renderer

ia32_memcpy now returns the destination pointer just like the standard

2005-10-30 21:01:42 +00:00

scripting

change requested by philip: debug_printf "tags" are now followed by '|' instead of ':' because something like "func: i am blowing up" is rather common.

2005-10-27 16:53:41 +00:00

simulation

gameview, renderer: LoadWaterTextures is now a coroutine because it takes friggin forever (3 seconds). VFS performance is down by a factor of ten! investigating is on todo.

2005-10-12 17:19:07 +00:00

sound

Lots of gcc -Wall fixes. The common ones:

2005-10-07 15:24:29 +00:00

tools

profiling indicates memcpy is a bottleneck, so globally replace all of them with memcpy2 (hand-optimized). depending on transfer size, this is 10%..300% faster. also made memcpy2 into a macro instead of thunk function because vc7 was still generating a jump.

2005-10-30 16:19:20 +00:00

main.cpp

Atlas: Bits of tool-related code.

2005-10-24 01:53:03 +00:00

ProgrammingSchedule.xls

gave it a once-over, marked 1 as complete

2005-05-09 16:30:01 +00:00