You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am having a hard time choosing the method of drawing/painting in the SDL version. Unlike GDI or QPaint, SDL does not have native methods for drawing ellipses or more complex objects. The only available shapes are points, lines and rectangles.
Now I have two choices, write them myself or using an existing library. The most tested library seems "SDL2_gfx", and I am now using it on the master branch. Unfortunately, the original repo is in SVN, so I have to migrate it from SourceForge to GitHub myself (Axure/SDL2_gfx). Moreover, the names of some functions (fortunately, private, undocumented ones) clashes with some thoses of some API functions in ACLlib, so I have to modify the names. The good news is that these are all done, and I have got a simple example working, to some extent.
But SDL can actually have OpenGL exclusively as its backend, allowing you to use pure OpenGL to draw, as demonstrated in the "sdl_opengl" branch. It seems, from the documentation however, that if you use SDL's API, the library will use OpenGL if available, and will fallback to software rendering if necessary. The reference is here:
if possible, it will use OpenGL or Direct3D behind the scenes, which means you'll get faster blits, a working Steam Overlay, and scaling for free.
This possibly makes using just SDL a better choice, instead of manually specifying OpenGL.
What makes me further entangled is the existence of sdl-gpu. The library claims to be a more efficient rendering backend of the official sdl backend.
So, what are your advices?
The text was updated successfully, but these errors were encountered:
I am having a hard time choosing the method of drawing/painting in the SDL version. Unlike GDI or QPaint, SDL does not have native methods for drawing ellipses or more complex objects. The only available shapes are points, lines and rectangles.
Now I have two choices, write them myself or using an existing library. The most tested library seems "SDL2_gfx", and I am now using it on the master branch. Unfortunately, the original repo is in SVN, so I have to migrate it from SourceForge to GitHub myself (Axure/SDL2_gfx). Moreover, the names of some functions (fortunately, private, undocumented ones) clashes with some thoses of some API functions in ACLlib, so I have to modify the names. The good news is that these are all done, and I have got a simple example working, to some extent.
But SDL can actually have OpenGL exclusively as its backend, allowing you to use pure OpenGL to draw, as demonstrated in the "sdl_opengl" branch. It seems, from the documentation however, that if you use SDL's API, the library will use OpenGL if available, and will fallback to software rendering if necessary. The reference is here:
This possibly makes using just SDL a better choice, instead of manually specifying OpenGL.
What makes me further entangled is the existence of sdl-gpu. The library claims to be a more efficient rendering backend of the official sdl backend.
So, what are your advices?
The text was updated successfully, but these errors were encountered: