Since the 80's when arcade video games has been first produced you could notice that in many games there is the graphical object itself along it's virtual borders. Every object had it's detection box, Sometime an object could be much bigger than the box itself and that could let you for example to move a big vehicle or a an attacking giant walking in a maze in a way that it will collide with other objects only in it's detection box borders.
Up to this point, in order for a user to make an objects with a smaller borders than the object itself you had to create a small dummy object and set the condition of set X and Y of the graphical object to dummy. That way you could make a 64x64 car for example move in a maze that has a path of 32x32 as the dummy object was a 32x32 size and over it your could connect by that specific condition the graphic part of the vehicle, or the giant or any other unit you would consider inventing.
I suggest the option to include a box size for every active object that isn't related to the size of the object itself and can be smaller than the graphic part. Right now when I make a car of a size of 64X64 I cannot set it's border to be less than that, said like, 32X32 and having an in object virtual box that can detect collision could really aid to simple process and can save the use of actual conditions. Up to now the best I might decide to do is create another active object instead of background object to handle collision in a manner of, like If "Active Car" position X is lower than "Active Background" + 4 and If "Active Car" position X is greater than "Active Background" - 4, stop object or bounce object into specific direction, plus others conditions like the direction an object came from, I could have it pre-set in the object's parameters themselves. It would save time setting condition and could be applied a handle with background objects and I could use a single object without connecting it to a dummy object as well as in case of multiple object, having to worry to connect each dummy object to it's respected graphics by let's say, spreading numbers...
What I can sum up is that you could have an option to crate an object in a specific size, like 32x32, 64x56, or any other desired size but set it's virtual size for collision to be smaller, as well as deciding where the area of the box is aligned compared to the object itself. As for example, a human which will collide with a wall only when the part of it's shoulders that are in the virtual box area collides with the background instead of starting of the head area... Or like, a 32x64 object but it's virtual box made for collision would start at 0x20 of said object and it's size could be for example 32x32, or any position and size I would like to set for it.
Also, You could always make more than what I suggested by letting an object have multiple virtual boxes by using conditions. I guess that there can always be a special object for that which I do not know, but it is time to implement it as an optional possibility for the active object itself!
I thank you in advance. If these who working on programming new features for future updates could consider it I can assure it will be useful for me and others as well!
