quote:

Originally posted by r0gue:
Reverse Polish Notation. Oh how I hated those....

I had to buy one for my MBA even though I had and use TI calculators.

I think it was for the snob appeal. It's like insisting on using WordPerfect with its command format keys rather than using any modern day document creator.

I have an HP 48G that has been affected by corroded batteries. Would HP repair it? I have to admit to having guilt feelings about the neglect.

I have a 35S, but honestly, it doesn't feel anything like my 32. It feels like very cheap and thin plastic.

I've used RPN for 30 years now. I can't use regular algebraic calculators anymore.

quote:

Originally posted by r0gue:

Reverse Polish Notation. Oh how I hated those....

Back maybe 25 or 30 years ago, I was teaching a UNIX℠ class. I mentioned Reverse Polish Notation.

At the end of the class, the clients turned in evaluation sheets with sections for the facility, the course, and the instructor.

I got slammed by one student for making an ethnic slur. Stupid snowflakes are nothing new.

quote:

Originally posted by Rey HRH:

quote:

Originally posted by r0gue:
Reverse Polish Notation. Oh how I hated those....

I had to buy one for my MBA even though I had and use TI calculators.

I think it was for the snob appeal. It's like insisting on using WordPerfect with its command format keys rather than using any modern day document creator.

RPN is absolutely harder to learn, but once you get it, it's faster, takes fewer keystrokes, and people make fewer mistakes (all largely because you don't need parenthesis).

An amusing aside on "modern day document creators," I'm a graduate student in computer science, so I both write a lot of computer code and write a lot of documents with math in them.

For actual text editing, I still use Vim quite a bit. Vim is a command-line text editor that hasn't changed a whole lot since it was released in 1991, and isn't all that different from Vi, which it was based on, which was released in 1976.

Vim has a significant learning curve, and there's certainly an element of "snob appeal," especially in the CS community, but once you really get it, text editing with Vim is IMMENSELY faster than with a normal text editor or something like MS Word.

For making the documents themselves, virtually the entire math, science, and engineering community uses LaTeX (initial release 1983), which is based on TeX (initial release 1978), because the tools for typesetting mathematics in MS Word et al are appallingly bad.

quote:

Originally posted by r0gue:
Reverse Polish Notation. Oh how I hated those....

Those who don't understand RPN are doomed to hate it

quote:

Originally posted by mcrimm:
I was just remembering times when a customer asked to borrow my 12C. After typing in a couple numbers and an operator they said "Where's the = key?"

A buddy of mine, when he was still in school, claimed that was one of the advantages of an RPN calculator: People rarely asked to borrow it more than once

quote:

Originally posted by 4x5:
I've used RPN for 30 years now. I can't use regular algebraic calculators anymore.

I can use 'em for simple expressions.

After learning RPN, and before HP came out with the 16C Computer Scientist programmer's calculator, I bought a TI calculator that could do binary, decimal, octal and hexadecimal. One day I decided to see just how bad complex expressions were in algebraic. I hated it. Remembering where the parentheses were was a real PITA.

For complex expressions RPN and the stack blows algebraic away.

quote:

Originally posted by maladat:
RPN is absolutely harder to learn, but once you get it, it's faster, takes fewer keystrokes, and people make fewer mistakes (all largely because you don't need parenthesis).

Exactly.

Interestingly: You do complex expressions on an RPN calculator exactly the way you would by hand. E.g.:

1 * 2 + 3 * 4 + 5 * 6

By hand, keeping PEMDAS in mind, you'd do the three multiplications, then add the three results, resulting in 44. On an algebraic calculator:

( 1 * 2 ) + ( 3 * 4 ) + ( 5 * 6 ) =

on an RPN calculator:

1 <enter> 2 * 3 <enter> 4 * 5 <enter> 6 * + +

or:

1 <enter> 2 * 3 <enter> 4 * + 5 <enter> 6 * +

Either way: Four less keystrokes. Ok, simple expression. Not a big deal. But, on an RPN calculator you see the results of each intermediate expression on the stack. If you bungle one, you just drop the stack by one and redo it. Bungle something in the middle of an expression on an algebraic and you have to start all over from the beginning. Sure, you can use registers, but yet more keystrokes.

quote:

Originally posted by maladat:
Vim has a significant learning curve, and there's certainly an element of "snob appeal," especially in the CS community, but once you really get it, text editing with Vim is IMMENSELY faster than with a normal text editor or something like MS Word.

I use vi/vim almost exclusively. I don't think it's "snob appeal." It's simply a million times faster than point-and-drool text editor interfaces. Or even Emacs, with its arcane control-shift-right-foot-pedal-left-cokebottle keystrokes.

I'm so wired-in with vi that, when I was using a Commodore Amiga I got my hands on the source code for STEVIE (ST Editor for VI Enthusiasts), ported it to the Amiga OS, then set to making it 99.9% vi-compatible--warts and all. By the time I was finished it was so true to vi that I, a vi power user, could hardly tell the difference.

Every time I fire up a new account on a *nix box, the first thing I do is change the command line editor to vi mode.

quote:

Originally posted by maladat:
For making the documents themselves, virtually the entire math, science, and engineering community uses LaTeX (initial release 1983), which is based on TeX (initial release 1978), because the tools for typesetting mathematics in MS Word et al are appallingly bad.

Never got the hang of LaTeX. Never really needed to, though.

quote:

Originally posted by KenS:
<snip>
I have both an HP35S on my desk and the free42 app on my phone. Both highly recommended.

Ken

For a smartphone calculator I’ve been using PCalc for several years. There’s a free Lite version, but I have the full feature version – a few bucks, well worth it. A plethora of selectable keyboard layouts, from basic to complex. Algebraic or RPN (selectable stack depth). RPN for me. For most keyboards you get additional function keys by rotating the phone to landscape orientation.

Its unit conversion feature “A>B”, is wonderfully extensive and convenient. And many built-in constants – Pi, etc.

Scientific or engineering, and bin, oct or hex integers.

PCalc is actively maintained – additions of new features and occasional bug fixes.

iPhone and (I think) Android.

quote:

Originally posted by Pipe Smoker:

quote:

Originally posted by KenS:
<snip>
I have both an HP35S on my desk and the free42 app on my phone. Both highly recommended.

Ken

For a smartphone calculator I’ve been using PCalc for several years. There’s a free Lite version, but I have the full feature version – a few bucks, well worth it. A plethora of selectable keyboard layouts, from basic to complex. Algebraic or RPN (selectable stack depth). RPN for me. For most keyboards you get additional function keys by rotating the phone to landscape orientation.

Its unit conversion feature “A>B”, is wonderfully extensive and convenient. And many built-in constants – Pi, etc.

Scientific or engineering, and bin, oct or hex integers.

PCalc is actively maintained – additions of new features and occasional bug fixes.

iPhone and (I think) Android.

This. I've had this iPhone app since 2010.

I have only used RPN since 1985! You can even change the layout of the keys. I fashioned mine after my HP41CV (Which I still have and still works).

quote:

Originally posted by maladat:

quote:

Originally posted by Rey HRH:

I had to buy one for my MBA even though I had and use TI calculators.

I think it was for the snob appeal. It's like insisting on using WordPerfect with its command format keys rather than using any modern day document creator.

RPN is absolutely harder to learn, but once you get it, it's faster, takes fewer keystrokes, and people make fewer mistakes (all largely because you don't need parenthesis).

An amusing aside on "modern day document creators," I'm a graduate student in computer science, so I both write a lot of computer code and write a lot of documents with math in them.

For actual text editing, I still use Vim quite a bit. Vim is a command-line text editor that hasn't changed a whole lot since it was released in 1991, and isn't all that different from Vi, which it was based on, which was released in 1976.

Vim has a significant learning curve, and there's certainly an element of "snob appeal," especially in the CS community, but once you really get it, text editing with Vim is IMMENSELY faster than with a normal text editor or something like MS Word.

For making the documents themselves, virtually the entire math, science, and engineering community uses LaTeX (initial release 1983), which is based on TeX (initial release 1978), because the tools for typesetting mathematics in MS Word et al are appallingly bad.

The inference from my modern day document program versus WordPerfect is that WYSIWYG feature of modern doc programs versus typing the embedded formatting codes of WordPerfect is no contest. The issue isn’t about the latest but the best solution for the task. You use Vim because it’s a good match for you. When I have to write equations, Word’s equation editor is good enough for me. There’s no payoff for me to learn Vim.

With RPN, I really doubt it’s faster and make fewer mistakes. You have to first map in your head the sequence of inputting the problem, right? You’ll find that any advantage you have with RPN will be erased against a TI user if presented with a super long sequence of multi-nested, series of random operations including scientific notation and angular measurements. The TI user can just start entering carefully while the RPN user will have to first understand the equation he is entering.

quote:

Originally posted by Rey HRH:
With RPN, I really doubt it’s faster and make fewer mistakes.

You doubt, but you don't know. I know. I've used them both. RPN is definitely faster and less mistake-prone. Everybody I've encountered who's used both has said the same. Without exception.

quote:

Originally posted by Rey HRH:
You have to first map in your head the sequence of inputting the problem, right?

PEMDAS.

quote:

Originally posted by Rey HRH:
The TI user can just start entering carefully while the RPN user will have to first understand the equation he is entering.

And if the TI user makes the tiniest mistake somewhere along the line they may never notice it if the result comes out looking feasible. If they do notice they have to start all over from the beginning. Neither of those problems obtain with RPN, because you get to see intermediate results on the stack as you go along--which means you can correct them, if necessary, without having to redo from the start.

This message has been edited. Last edited by: ensigmatic,

I'll be damned. You nerds blow my mind about once a quarter, talking about something I've never heard or thought of before.

And I'll admit, after a 5 minute googling of that polish stuff you're talking about, I was able to figure it out and I think I like it better than the algebraic variety.

But then again, I flunked calc twice. So there's that...

The most calculator computing I did was in college. I have a degree in engineering and I went from a TI-66 to a CV41CV in my sophomore year. That was a game changer.

Not every engineering major got a RPN calculator but the vast majority did. The equations were just too long and too complex to input dozens of ()'s.

Trust me, for heavy weight calculator computations, a RPN calculator is the only way to go.

quote:

Originally posted by Rey HRH:
The inference from my modern day document program versus WordPerfect is that WYSIWYG feature of modern doc programs versus typing the embedded formatting codes of WordPerfect is no contest. The issue isn’t about the latest but the best solution for the task. You use Vim because it’s a good match for you. When I have to write equations, Word’s equation editor is good enough for me. There’s no payoff for me to learn Vim.

Vim is a text editor - you can think of it as something like Notepad with a set of extraordinarily powerful keyboard shortcuts for navigating and editing text. LaTeX is the typesetting software.

Regardless, you're right that if you're just inserting a couple of equations into a document, the Word equation editor is OK.

When you're writing a 10+ page document with hundreds of bits of mathematical notation interspersed in the text along with dozens of complex equations, entering all the math in the Word document will take multiple times as long as doing it in LaTeX. That doesn't even consider the fact that a Word document like that will have an INSANE file size and will pretty much cause MS Word to grind to a smoking halt.

Of course, that doesn't matter for how most people use Word.

quote:

Originally posted by Rey HRH:
With RPN, I really doubt it’s faster and make fewer mistakes. You have to first map in your head the sequence of inputting the problem, right? You’ll find that any advantage you have with RPN will be erased against a TI user if presented with a super long sequence of multi-nested, series of random operations including scientific notation and angular measurements. The TI user can just start entering carefully while the RPN user will have to first understand the equation he is entering.

This is an actual scientific study done to evaluate RPN vs. algebraic notation:

https://www.sciencedirect.com/...094900485?via%3Dihub

I am able to access the full article through my university as a grad student. The short summary is that they wrote a bunch of equations and had people compute them using both RPN and algebraic notation calculators.

They found that both computation time and probability of making a mistake pretty much only depended on number of operations, not what sort of calculator was used - but that RPN calculators required a lot fewer operations. The end result is that for equations at the complex end of their test set, RPN calculators were more than 30% faster. The overall error rates (per complete computation) across the entire test set were 27% with algebraic notation calculators and 18% with RPN calculators.

My HP41 gave up the ghost last year. I got a TI30xa but I wish I had looked around to find one to replace the HP41 with same keyboard layout and functions.

I remember when the HP35 came out while working at GE. They were a status symbol for engineers with them wearing them in a case attached to their belt.

There was a time when I lusted for this Curta hand held mechanical calculator. A neat design, and would be quite a show-and-tell piece.

https://en.m.wikipedia.org/wiki/Curta

just looked for one

used...

$385

no modules

quote:

Originally posted by ensigmatic:

quote:

Originally posted by Rey HRH:
With RPN, I really doubt it’s faster and make fewer mistakes.

You doubt, but you don't know. I know. I've used them both. RPN is definitely faster and less mistake-prone. Everybody I've encountered who's used both has said the same. Without exception.

quote:

Originally posted by Rey HRH:
You have to first map in your head the sequence of inputting the problem, right?

PEMDAS.

quote:

Originally posted by Rey HRH:
The TI user can just start entering carefully while the RPN user will have to first understand the equation he is entering.

And if the TI user makes the tiniest mistake somewhere along the line they may never notice it if the result comes out looking feasible. If they do notice they have to start all over from the beginning. Neither of those problems obtain with RPN, because you get to see intermediate results on the stack as you go along--which means you can correct them, if necessary, without having to redo from the start.

I actually do know because I had to buy one for one of my MBA classes and had to use it. I learned enough to use it. That’s why I remember I had to actually understand the problem first in order to determine the sequence of entries that I had to make. Went back to TI when I didn’t have to use RPN. In my experience, I determined no benefit for me expending the time and effort to be proficient in RPN. I’m a snobby kind of guy, I use a montblanc fountain pen at work; I would have loved to take on the snob appeal of RPN if it was worth it to me.

quote:

if the TI user makes the tiniest mistake

Math is funny that way that it requires 100% accurate input to get the right output.

quote:

Originally posted by maladat:


Vim is a text editor - you can think of it as something like Notepad with a set of extraordinarily powerful keyboard shortcuts for navigating and editing text. LaTeX is the typesetting software.

Regardless, you're right that if you're just inserting a couple of equations into a document, the Word equation editor is OK.

When you're writing a 10+ page document with hundreds of bits of mathematical notation interspersed in the text along with dozens of complex equations, entering all the math in the Word document will take multiple times as long as doing it in LaTeX. That doesn't even consider the fact that a Word document like that will have an INSANE file size and will pretty much cause MS Word to grind to a smoking halt.

Of course, that doesn't matter for how most people use Word.

quote:

Originally posted by Rey HRH:
With RPN, I really doubt it’s faster and make fewer mistakes. You have to first map in your head the sequence of inputting the problem, right? You’ll find that any advantage you have with RPN will be erased against a TI user if presented with a super long sequence of multi-nested, series of random operations including scientific notation and angular measurements. The TI user can just start entering carefully while the RPN user will have to first understand the equation he is entering.

This is an actual scientific study done to evaluate RPN vs. algebraic notation:

https://www.sciencedirect.com/...094900485?via%3Dihub

I am able to access the full article through my university as a grad student. The short summary is that they wrote a bunch of equations and had people compute them using both RPN and algebraic notation calculators.

They found that both computation time and probability of making a mistake pretty much only depended on number of operations, not what sort of calculator was used - but that RPN calculators required a lot fewer operations. The end result is that for equations at the complex end of their test set, RPN calculators were more than 30% faster. The overall error rates (per complete computation) across the entire test set were 27% with algebraic notation calculators and 18% with RPN calculators.

That’s an interesting summary. Was each person timed using both types?

I think it matters how people think along with perceived cost/benefit of learning a new system. It may very well be that RPN has advantages over TI but will the time spent learning RPN proficiency be worth it is the question for the individual.

Which reminds me, i’ve Been an Excel super user and VBA programmer. But up until maybe two years ago if I needed to validate a formula, I would reach for my calculator or the calculator app. Then I realized a new Excel sheet would do just as well and I can check the numbers in the formula.

quote:

Originally posted by Rey HRH:
That’s an interesting summary. Was each person timed using both types?

The study used 20 engineering students familiar with both types of calculator and had each of them perform each calculation with each type of calculator.

Between the two types of calculator, the keystrokes per second and average keystrokes between mistakes were effectively identical, but the RPN calculators require a lot fewer keystrokes.

quote:

Originally posted by Rey HRH:
I think it matters how people think along with perceived cost/benefit of learning a new system. It may very well be that RPN has advantages over TI but will the time spent learning RPN proficiency be worth it is the question for the individual.

For sure. This was actually a topic for an XKCD a while ago.



https://xkcd.com/1205/

quote:

Originally posted by ensigmatic:

For complex expressions RPN and the stack blows algebraic away.

Exactly.

Interestingly: You do complex expressions on an RPN calculator exactly the way you would by hand. E.g.:

1 * 2 + 3 * 4 + 5 * 6

By hand, keeping PEMDAS in mind, you'd do the three multiplications, then add the three results, resulting in 44. On an algebraic calculator:

( 1 * 2 ) + ( 3 * 4 ) + ( 5 * 6 ) =

on an RPN calculator:

1 <enter> 2 * 3 <enter> 4 * 5 <enter> 6 * + +

or:

1 <enter> 2 * 3 <enter> 4 * + 5 <enter> 6 * +

Either way: Four less keystrokes.

You do realize that you can get algebraic calculators that understand the order of operations. So in your example you would enter:

1 * 2 + 3 * 4 + 5 * 6 =

This is 12 keys, 2 keys less that the 14 required for RPN. It also has the advantage of being able to enter the equation exactly as you would typically find it written, so it saves you the time required to reformat the equation in your mind.