Calculator Operations RPN versus DAL 2

[Ussuri]

There is a discussion ongoing on the structural engineering forum regarding the current range of HP calculators.

I was unable to comment on the discussion as I have tried a HP a few times and didn't like it. I didn't like the Reverse Polish Notation (RPN) system. I currently use a Casio CFX-9850G which uses the DAL (Direct Algebraic Logic or something similar) system. I have had it a number of years and cannot fault it. However, I seem to be alone in my department as most everyone uses a HP with RPN.

Why does RPN seem more popular? Is it a better operating method for calculators? Should I bite the bullet, ditch my Casio and move over to a HP?

---------------------------------------------------------------------------------------------------------------------


"The world keeps turning, it keeps me in my place; where I stand is only three miles from space"
Spiritualized

 

[PSE]

Use whatever you are comfortable with. The results obtained are (should be) the same. I have a now (venerable) HP11C that gets used pretty regularly. I am equally comfortable with non RPN calculators. You don't need to change just to fit in.

RPN might be a bit more popular in that, at least in my experience, HP came out with calculators that were really useful for engineering before anyone else and they utilized the RPN notation.

Regards,

 

[JAE]

Ussuri,

I too would say use what you are most comfortable with. I personally like the RPN because once you force yourself to use it a while, you begin to realize that for calcs with a few more variables in it, the RPN really helps you fly through the numbers without having to remember all the ().

For basic X times Y either method is fine.

 

[civilperson]

In speed contests, RPN has fewer keystrokes.

 

[Ashereng]

Use what you are comfortable with.

For me, RPN is easier for getting "intermediate" results:

y = mx + (b+c)

If I want to know what mx and b+c is while I am doing my work.

In my life today, my Excel spreadsheet gets more use than my HP 15C and 32S (both RPN). I still have my HPs, I just use it less often for work, and more for quick calcs.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[IRstuff]

It's not unlike the discussion of using Forth vs. any other "normal" programming language. Oddly, I use RPN, but am not a proponent of Forth, even though they're awfully similar.

When you get right down to it though, most computer language and hardware architectures are stack oriented, hence, more like RPN, even at the calculator level. The parentheses simply force a new stack entry.

Anyways, use whatever gets you the correct answer the fastest. Any other consideration is really irrelevant.

TTFN

FAQ731-376

 

[eureka]

I beleive that an old dog can learn a new trick. Learn RPN. You will have broadened you calculating ability and have two methods to pick from. It will also allow you to follow another engineer when checking calculations.

 

[65Roses]

When I attended college a Casio (fx-7000) was required for Math classes. When I started taking classes in my major an instructor strongly recommended an HP48. At first it was really hard to get use to. I kept reaching over for my Casio. What I finally had to do was put the Casio away and force myself to learn the HP. Once I did that I was glad I did.

Why is my handle 65Roses?
Please visit

http://www.cff.org/aboutCFFoundation/About65Roses/

to learn why!

 

[KENAT]

I had trouble switching to a new Casio when I lost my old one, I dread to think how long it would take on another manufacturer.

 

[zeitghost]

I acquired a D.A.L calculator recently.

Took ages to figure out why pressing 2 followed by squareroot gave "Error" all the time...

I also managed to put it in some weird mode where 22/7 == 3

 

[mendinho]

Without any kind of doubt, the HP-48 is my favourite calculator. No Casio (I had one) can be compared to an engineering HP calculator.

RPN was specially invented to use less memory and to avoid the use of parenthesis.

For example:
((((1+2)*3+4)*5+6)*7+... would be

1 2 + 3 * 4 + 5 * 6 + 7 * ....

That is the key advantage.

 

[dik]

and with my TI89Titanium... if I wanted to revise the 5 to an 8... it would require the modification of a single digit...

BTW, I use the TI and my HP48G interchangeably... I keep one where I'm working and one at home. I like both and HP's 'stack' architecture was originally chosen because of the manner in which data was entered into a computer... similar with Forth... they needed 'real time' speed.

Dik

 

[TomBarsh]

Back in the day when machine memory was limited the RPN method was useful to minimize memory requirements. But this in itself doesn't explain the decision to implement RPN or its continued popularity.

Like others, I was confused by RPN when I got my first HP calculator around 1978 but it made sense when the HP manual explained it in terms of order of execution and how the RPN method is the same method that one would use when evalulating a numerical expression by hand.

In other words, when evaluating by hand you don't simply start at the left and work towards the right (unless you can stack up those intermediate results in your brain, I can't), you start somewhere "in the middle", evaluating expressions within parentheses first, then by other standard procedures.

For example, in my original "Solving Problems With Your HP Calculator" (complete with helpful Charlie Chaplin cartoon manipulating the stack register) from 1978 they give the example of:

5*((3/4)-(5/2)+(4*3))/(3*0.213)

Evaluating with an RPN calculator proceeds just as though you were solving by hand, start with the innermost parentheses, then work outward (there may be more than one path taken). The intermediate results are stored in the stack, then brought down as they are needed. Solving with an algebraic calculator, well, I guess the modern ones you can enter the whole expression and then even edit it, then evaluate it. That's pretty handy too. But by the time those were available I was an HP convert. Nowadays, I still use my trusty old HP-15C but for anything fancy I will make an Excel spreadsheet (and use it over and over again).

Some Polish mathematician found that the minimum requirement for a stack is 4, anything less and values would pop off the top and be lost. But some later HP calculators have very large stacks (100?). I have an HP 48GX stuck somewhere in my desk, I don't have much use for it.

By the way, the old HP-15C "Advanced Functions Handbook" (separate purchase) had an incredible Appendix on "Accuracy of Numerical Calculations". I still refer to that to this day. It was written by some very brilliant person(s) but in a form that even I can understand.

 

[ScottyUK]

I'm curious as to the age profile of the RPN advocates compared with those who prefer the DAL system. I suspect a bias toward younger users prefering DAL but I might be completely wrong (it won't be the first time this week!).

Personally I like DAL, although I can use RPN at a push. I don't use RPN often enough to develop proficiency in it, so it is determined effort rather than second nature. Practice would change that, but since I can have a result from my Casio FX-115 much quicker at present than I would with a RPN calculator I have little reason to change for the sake of saving a few keystrokes some time in the future.

For the sake of the age profile, I'm 35. By the time I was using a scientific calculator at school the RPN types were an expensive minority and Casio had cornered the bulk of the market.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[dik]

I'm pushing 60 and I like both... I liked HP because of the reliability, although I'm a little more reserved about this in the last 5 years or so... It replaced a model (I can't recall the model) that was, IMHO, the best calculator HP ever shipped... When my wife was hospitalized, I replaced it with the TI because the TI had a Motorola 68000 processor in it (funny reason for choosing it) and it was purchased simply to learn a new calculator... both work well, and are used almost equally. Only problem with the TI, is that in the two years I've had it, it has 'locked up' a couple of times... complete freeze; some kind of programming error...

Dik

 

[berkshire]

ScottyUK
I am now 65. I started with a national semiconductor RPN calculator, then went to a TI59, until somebody pinched it.
I then went to a Sharp PC1600 which was a strange beast, you entered the equation as you wrote it then hit enter. It was also programmable in basic but would only hold 900 lines of code, I now use a Casio CFX-9750G.The small brother to the one Ussuri uses, the large screen allows you to enter the whole equation and see what you have entered before you hit the execute key. I have no problem switching between DAL and RPN.except when I am entering strings of numbers on a DAL and get to the last one.
B.E.

 

[ScottyUK]

Those Casio CFX series look nice - a lot more capabilities than my simple non-graphical one. Do you find you use the graphing function much, or is it the big screen which appeals? I had one of the early graphing calculators about 17 years ago and found it invaluable in tracing the roots of equations where you might otherwise use an iterative process, but used the function little other than that.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[GregLocock]

I think for kids a graphing calculator is brilliant - to me maths is most easily understood via visuals.

I think the RPN vs normal logic is a religous argument - RPN is to some extent more closely related to slide rules... but I'm not really sure that translating a normal equation into RPN is a robust solution.

I don't actually care much either way - a Toshiba Libretto running an early version of Mathcad was as portable as a proper calculator, and 200 times as useful. But some bastard stole mine to fund his crack habit.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[berkshire]

ScottyUK (Electrical)
I really do not use the graphing function that much, the large screen and the fact that I can program the calculator appeal to me the most, I have a large number of trig functions programmed into mine.
Greg,
Was that Mathcad 3.1 or 3.0 ?
B.E.

 

[GregLocock]

v 4.0

I preferred 3.1 to be honest



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.