A puzzle on Composite efficiency.

[Onishion]

Ok, so I use some complex LCs in my projects, and I was curious as to which of these two methods would be the most efficient way of dealing with them:

Code: Select all

image MyImage:
    LiveComposite(
        (100,100),     
        (0,0), ConditionSwitch(            
            "VariableA == 3 and VariableB == 1", "images/image1.png", 
            "VariableA == 3 and VariableB == 2", "images/image2.png",  
            "VariableA == 3 and VariableB == 3", "images/image3.png",            
            "VariableB == 1", "images/image4.png", 
            "VariableB == 2", "images/image5.png",  
            "VariableB == 3", "images/image6.png",
            "True", Null(),
            ),     
        )
    
image MyImage2:
    LiveComposite(
        (100,100),     
        (0,0), ConditionSwitch(    
            "VariableA != 3", Null(),        
            "VariableB == 1", "images/image1.png", 
            "VariableB == 2", "images/image2.png",  
            "VariableB == 3", "images/image3.png",
            "True", Null(),
            ),     
        (0,0), ConditionSwitch(    
            "VariableA == 3", Null(),        
            "VariableB == 1", "images/image4.png", 
            "VariableB == 2", "images/image5.png",  
            "VariableB == 3", "images/image6.png",
            "True", Null(),
            ),   
        )

I may have made a mistake in there because I threw it together kind of quickly, but if I got it right, both methods would accomplish the same effect. If Variable A is 3, then one of images 1, 2, or 3 will display, depending on what Variable B is up to. If Variable A is not three, then one of images 4, 5, or 6 will appear. Obviously in such a small case the difference might not matter at all, but what if in the first example I were checking against 3-5 different conditions to determine whether to display the image? Would it add or reduce performance to the situation to have sets long "if and and and or" strings, or would it be better/worse to instead separate them out into individual layers of the composite and have each disqualify itself if one core condition fails? Which is more of a resource hog, having one condition switch that asks a lot of complex questions before reaching the end, or a bunch of individual layers that each ask only simple questions? Or would it not matter at all no matter how complex the composite got?

[xela]

Or would it not matter at all no matter how complex the composite got?

This. First way is prolly more efficient but it depends on case per case basis. You can use time.time() to figure out which is best if you really want to know.