Hello, I'm a student in a hight school of art, and I have started few month ago to learn to work with Processing.

So, in  the framework of my studies and to lead my own project I have realized a small program which read the value of a pixel according to the mouse X and Y position. The problem is that the values I get back are negatives numbers like
"-14212709" for example. I know what this value can represent and why it's negative BUT, I would like to get back the Red, the Green and the Blue values instead of these negatives numbers.

I don't know if it's possible but I hope so.
If someone had a part of the answer it would be a great help for me.

thanks for your attention.

As memo said, although I am not sure that the second solution is more optimized. I prefer the clearer one, for sure...

The issue and solution are explained in the color page, the reason being that by default, colors are opaque, ie. have a 0xFF alpha value which is in the high bits of the word.

If you will only need to do this once or twice per update frame (e.g. getting the color components of the one pixel under the mouse cursor every frame), then the performance loss from using the red(), green(), blue() functions will be negligable. However if you want to loop through every pixel in a 1024x768 image - i.e. 2.35 MILLION times a frame, then I advise you use the second method! (In that case i also advise that the int r, int g, int b variable declations be done OUTSIDE the for loop as well!)

Any time someone says a method is ever-so-slightly faster than another method, generally you don't need to use the ever-so-slightly faster method if you will only be doing it a few times every frame. Its when you have it in a massive for-loop or something that you need to start saving nano-seconds! (e.g. iterating pixels in an image)

To be honest Philho, I haven't measured the difference myself!
Its just a habit I've had for many many years, and an optimization tip which all programmers will recommend whether its for C, C++, C#, Java, ActionScript etc. It is true that some compiler/interpreter options can make it irrelevant by creating the variable only once outside the for-loop as if thats what you coded, but I think it still is good practice  to bear in mind when performance really is an issue...

Hey Philho, yea like I said, I haven't measured the performance difference myself so can't comment on the nessecity of keeping the vars outside. But I do copy/paste code between C++/AS3/Java quite often so like to keep it as compatible as possible, so will probably carry on keeping it outside. As long as its in a function or other block the scope is still limited - I am quite intrigued that the bytecode is identical, I gotta look into that.

Though I must say, every Java optimization page I've read gives the specific example that for (i = 0; i < someString.length(); i++) is bad because someString.length() is executed every frame. In fact they generally go as far as to say even not to use for (i = 0; i < myObject.customProperty; i++) (in this case customProperty is a property NOT a method), and to cache object properties as well. Memory management may be quite different in C and Java (and AS3), but thats a bit irrelevant when it comes to compiling such code. A function call, or accessing an object property, or multi-dimensional array etc. will always have overheads compared to a local or static variable...

For others reading this thread confused as to what to do - like I said in my first post, all of these are negligable if they will be done a few times a frame. Its only when you are doing them millions of times per frame does it start to make a difference.