* Use the GPU to render the objects - GLSL shader approach
* This way is definitely faster, but subject to the GPU's limitations on precision
*/
- virtual void RenderUsingGPU();
+ virtual void RenderUsingGPU(unsigned first_obj_id, unsigned last_obj_id);
/**
* Use the CPU to render the objects - "make a bitmap and convert it to a texture" approach
struct PixelBounds
{
int64_t x; int64_t y; int64_t w; int64_t h;
- PixelBounds(const Rect & bounds) : x(bounds.x), y(bounds.y), w(bounds.w), h(bounds.h) {}
+ PixelBounds(const Rect & bounds) : x(Double(bounds.x)), y(Double(bounds.y)), w(Double(bounds.w)), h(Double(bounds.h)) {}
};
static Rect CPURenderBounds(const Rect & bounds, const View & view, const CPURenderTarget & target);
public:
BezierRenderer() : ObjectRenderer(BEZIER, "shaders/rect_vert.glsl", "shaders/rect_frag.glsl", "shaders/bezier_texbuf_geom.glsl") {}
virtual ~BezierRenderer() {}
- virtual void RenderUsingGPU();
+ virtual void RenderUsingGPU(unsigned first_obj_id, unsigned last_obj_id);
virtual void RenderUsingCPU(const Objects & objects, const View & view, const CPURenderTarget & target);
void PrepareBezierGPUBuffer(const Objects & objects);
private: