Commit | Date | |
---|---|---|
2006-03-28 17:55:13 | Tree | |
2006-03-23 19:32:21 | Tree | |
2006-03-10 18:49:34 | Tree | |
2006-03-08 23:18:37 | Tree | |
2006-02-28 12:39:15 | Tree | |
2006-02-27 19:55:43 | Tree | |
[r2281]
by
elias_naur
Widened VBO/PBO buffer offsets and sizes to long to better match the GLsizeiptr and GLintptr native types |
2006-02-26 11:44:34 | Tree |
[r2280]
by
elias_naur
Linux: Removed the glXIsDirect test to determine whether a context is accelerated or not. This is specifically not the case for Xgl and friends that accelerates indireect rendering. Unfortunately, I'm not sure what to replace the test with, so there is currently no way to know from LWJGL whether we're software rendered or not (other than measuring FPS). In fact, I'm not even sure that we can come up with a good test, because some (DRI) drivers use software MESA rendering for some parts of GL, while other parts are hardware accelerated |
2006-02-23 19:29:49 | Tree |
2006-02-23 19:22:00 | Tree | |
[r2278]
by
elias_naur
Win32: Made the message processing more friendly to potential foreign windows created on the same thread (and thus sharing message queue with LWJGL). Now handleMessages() in Display.c only processes messages for the current lwjgl window. |
2006-02-22 10:47:14 | Tree |