I'll try to answer your second question first. There are many ways to go about both your questions, but here's a simple way. First, let's think of the problem. You want something to happen when the players mouse hovers over something. That's basically it. Now let's think about what we already know that could help us find the solution to the problem. We know that the something that the player will hover over will be a sprite or image of some sort (as they can obviously see it). Let's assume, since you are building this is in C++, that all your "somethings" will be within a class. Let's also assume that you have a container of some sort to store all of these "somethings." SDL, im pretty sure, has a function or method that we can use to find the current coordinates of the mouse. SDL, im pretty sure, also has a function or method that can determine time for us (as like a stopwatch). So now that we have determined the problem and looked at what we knew we had to help us determine the solution its time to make our solution. Our solution is going to be inefficient (in my opinion) but simple for you to understand and work off of. Within our game loop, we need to use SDL's function for getting mouse coords, send those coords to our "something" objects, by iterating through our "something" object container. Our objects will know their current coords and based off of that will determine if the mouse coords are within their area.

obvious pseudo-code:


Hope this helps some. This solution will work, but it is very inefficient since you are basically iterating through an entire container and on top of that checking collisions every frame. This is extremely computationally heavy but you could very well split the objects up, or setup a timer to only do this every .5 seconds (or some other number), or set it up to do it only every 20 frames (or some other number). Maybe people here will have a better way, who knows :) The point to my post, besides helping you with your question, was to hopefully teach you to look at something from an engineering approach, whats the problem, what do I know, and how do I derive a solution from that. Good luck =)

#1 Get used to thinking in terms of coordinate systems -- the screen area, your map,the objects on the map, the cursors position on the screen....

#2 When doing filtering you should do something called 'lazy evaluation' where you cull out the objects NOT within an area (the screen view window) by the cheapest method available (an XY box test for example) and then subject the remainder to a more detailed (proceesing costly) test if needed.

#3 With 100 objects being tested (even in containers with their overhead) you should be able to do such comparisons hundreds or thousands of times a second and the other processing (like the rendering or collision testing) will take much more processing time.