Seems simple enough to comprehend. It appears to be simplistic and straightforward. I'm sure there will be some tricks when it comes to more advanced functions, though.
I think you'l find that there are *no* tricks, and that is part of RPN's charm. When using in-fix, you have to remember a potentially complex and subtle order of operations, or else use lots of parentheses. Most of us might remember "PEMDAS" (parentheses, exponents, multiplication, division, addition, subtraction) but in a programming language with lots of operators there can be subtle variations that can bite and annoy you. If you've ever written a grammar and accompanying parser for an in-fix language... you'll know it's work.
Enter RPN. There is no order of operations. Every operation pops its operands directly from the top of the s stack and pushes its results back onto the stack. There's nothing to parse. There may be some special functions to 'rotate' and 'swap' things' position on the stack (so that the next function can access them), but even those can be implemented as a bunch of pops, rearrange and then pushes. Honestly, once you get used to it, you'll think it is _easier_ than infix.
Well, if you get the bug, anyway.