As the title mentions, my question is about performance and optimization. I was using panel sprites, with objects on the screen stretched to occupy large areas. However, I still don’t have the workflow defined for building the scenes and I need to resolve these questions before doing so.
Is it better to use panel sprites than to have different images (center and edges) of the object I want to build and put it together as a puzzle in the editor? or is it the opposite?
On the other hand, I have seen that you can import sprite sheets. Does it improve game performance to have objects and animations on sprite sheets rather than importing or creating the images separately in the animation editor?
Nothing out of the box in the engine uses spritesheets natively. Piskel just slices the sprite sheet into separate sprites for a single frame of animation.
There is a sprite sheet extension but it has to use a tiled sprite to work and they are less performant than normal sprites or tilemaps in many cases.
Panel sprites are generally for UI and are fine at that, using them for things like terrain can be taxing because of the amount of transformational math that happens every frame.
1 Like
Thanks for the clarification! I have one more question regarding rendering performance. If, for example, I had to make a grass background, which option consumes less rendering resources? Use a colored pixel and stretch it to the size you want that grass to be or make larger squares (not stretched) and place them one with another to make up the entire grass? I assumed that the first option was better performing, but when I did the tests I didn’t notice much difference.
If there is another option with better performance in terms of calculations and rendering, I’m all ears. 