Been working quite busily today. Here’s my list:
- Noticed, verified, and reported a bug in cgdb under OS X 10.8/Xcode 4.4. (I’m the proud owner of the very first bug report on its new GitHub page). (cgdb is a text-console-based debugger, and is how I usually prefer to debug my programs)
- Further work on SDL to support retina displays. Included some discussions on mailing lists and development forums. I’ve got it almost fully working. (The bit still broken involves being on a retina-capable machine with a non-retina-capable program, SDL reports some illegal screen resolutions as being available. Still working on finding a workaround for that)
- Noticed that there are issues in the current trunk version of the VectorStorm Testbed repository on GitHub, of the “doesn’t even build via Xcode” variety, so I spent some time ironing all of that out, and updating it to use the latest revision of the VectorStorm library.
- Noticed that since adding support for arbitrary render targets, changing resolution doesn’t work any more; it leads to a crash. I’ve fixed the crash, but that now causes rendering to stop after the resolution change. Based upon the behaviour that I’ve seen, I suspect that it’s probably trying to render to the old (pre-resolution-change) offscreen textures, but I haven’t yet actually tracked down why it would be doing that. So my hunch may be completely wrong.
- Noticed that for some reason, SDL_image is failing to load a particular png file from the “CameraTest” sample. Opening and exporting that png file again seems to have sorted out the problem. No clue what the real issue was.
- Noticed that the VectorPhysics sample game in the Testbeds suite is broken again. Completely broken. HUD not visible, interface doesn’t work, and Box2D appearing to cause odd crashes. Not sure whether to bother dealing with that right now; I don’t really need Box2D for anything at the moment, so maybe I’ll just ignore the problem and hope it goes away, and get back to MT2 development sooner.