HP & RPN

[John Garza (3665)]

I've seen lots of books & articles extolling the virtues of RPN. But I've never seen any information on how & why HP chose RPN for their calculators. From a business standpoint, I'm sure there must have been some analysis done to justify the use of this obscure system. After all, how many people would know of Polish Notation and that Jan fellow with-the-funny-last-name-I-can't-spell, if it weren't for HP calculators?

[Raymond Del Tondo]

It seems that engineers had much more influence on design matters back then.

One of the sentences in their owner's manuals was
"Made for engineers by engineers"
(maybe the 2nd and 4th word exchanged)

It seems that engineers wanted to have highly efficient, high quality tools. So some kind of RPN was obligate,
and it was even cheaper to produce an RPN machine,
because they needed less storage registers,
and RAM was very expensive those days.

Nowadays it's vice versa: The 'marketeers' hold the strings,
and we all see the result, that new HP calculators are kinda crap (IMHO).

Raymond

[Vieira, Luiz C. (Brazil)]

Hi;

as a matter of fact, let's face one important point: RPN is related to the direct CPU operation, in a way operands must be placed before the CPU can use them. As it happens with RPN. This means an operating system that does not offer an AOS interface to the user. It's known that algebraic calculators store the keystroke sequences till an ")" or "=" initiates a LIFO-style (Last In, First Out) execution, that resambles RPN.

There is more about it, and I'm in rush for now.

Cheers.

[John Garza (3665)]

Good point about the engineers vs. the marketeers. Very true. And I know a postfix parser is easier to code than infix. Been there done that :) But what I'm getting at is that (knowing HP), there must have been some document produced by HP along the way that founded the basis for their decision to use RPN. After all, it was not a minor design decision. It's very un-HP like to make arbitrary decisions like that. They are almost always well considered.

-John

[Dia C. Tran]

The first hand held calculator the HP35 was supposed to be for HP engineers use only not for sale