I'm reading "fibonacci Heap" of Dr R.Tarjan. Basically, I know what he say. But, I don't know why he do that.
For example:
He define a potential function is:

//t(Hi) and m(Hi) is number of roots and number of marked Node, respectively at Heap Hi

1) I don't know why can think that potential function with 2 parameters t(H) and m(H). (So, in my thinking, if I can,of course, if I can, find another parameter function X, so does my potential function will be better than him ?)

2) I don't know why have number 2 before m(Hi) (It mean I don't know why he think that number).
In the prove of him, there 2 main points: "extract min" and "Decreasing key", and (maybe) I see that number 2 don't very important. I think we can change number 2-->any constant number and the result no change.
Does am I wrong, huh ;) ?

So, who know clearer than me at fibonacci Heap, please help me please.

thanks :)

Maybe noone like Fibonacci Heap or Amortized Analysis, huh ;) I think those problems are very interesting, but often rarely people care them. :(