HP 50g - How to log and edit RPN expression entry?

[jimbo1qaz]

I recently moved from a heavily modded TI84+ (slightly unstable) (BASIC causes brain damage) to an HP 50g (cheapest option available XD ). I've had it for a few months, but never really got familiar in practice. This is my first CAS calculator, but seems rather awkward to use.

Integration is a bit tricky: Symbolic home screen, step-by-step symbolic equation editor, numeric equation editor, what else? Also, definite integrals from the home screen require << a b INTVX PREVAL >> . Is there a way to perform fast home-screen definite integration without first INTVX-ing?

Loading the graphing screen is slower than TI-84+ Classic mode. Graphing entry forces you to name all Y-values. Cancelling a graphing operation takes an entire second. One hard issue I encountered is that the slope field only works from the <F4 2D/3D menu, not the <F1 Y= menu.

But these questions are for another time...

## EXPR HISTORY, EDITING
The biggest problem I have right now, is that when entering expressions in the RPN home screen, numeric values are not preserved when typing. This makes editing symbolic expressions suboptimal, and numeric expressions completely impossible.

On the TI-84+, I frequently performed textual equation editing directly from the home screen. Although limited (graph variables) and error-prone (parentheses), it was at least functional.

As an analogy, RPN is similar to writing a desktop program by typing code directly into a compiler. You cannot easily undo mistakes along the process, nor can you iterate on the original code after running it. RPN is actually worse, because you can't even see the full expression as it's being written (for numeric values).

## HISTORY
I discovered you can perform RPN entry in RPL tags << >> instead, duplicate the expression, then evaluate. However, this requires manual duplication, introducing a source of human error. Also, unit conversions and calculus functions behave differently in RPL (I haven't figured out). Most importantly, you no longer get intermediate values, and the wall-of-text entry is difficult to read.

You can also duplicate expressions using ENTER, then perform operations on the lower copy. Still that's error-prone, and doesn't preserve variables.

## EDITING
Is the proper solution to use algebraic (backtick) expressions in RPN mode? (Somehow, RPN single-ticks do not support typing 2x for 2*x. Inconvenient.)

Algebraic expressions can be constructed and edited in the equation editor. But I find the equation editor to be tedious. Also, they're simplified just like RPN.

This only allows editing, and does not directly enable a full history.

## LOGGING
Is there any program that collects RPN input into a RPL block, and allows you to recall RPL expressions? or something? Similar to Algebraic?

[wojtek]

Try using << >> to put your expression into a code and save it to a variable, for example:
<< 50 25 + '5 * 3 + 10' + >>
'NN'
STO
should save the expression between << and >> in variable NN
Then you can load it to the stack (with RSHIFT and NN softkey) and evaluate with EVAL or edit by pressing the DOWN key (you can also put EVAL into the code).
After evaluating your expression you can also use the UNDO command to edit the expression (and save it with LSHIFT and NN softkey).
I hope this answers all your questions.

[jimbo1qaz]

I already explained why that's suboptimal. You need to manually insert RPL tags whenever entering expressions, rather than automatically saving input in a clean fashion, as in Algebraic. Also, it doesn't behave nicely with CAS calculus.

[BruceH]

I'm struggling to understand what your problem actually is.

If you want to enter as algebraic, evaluate as algebraic and go back and edit mistakes or changes on an algebraic expression then put the 50G into algebraic mode and you're good to go.

If you want to use RPL but still enter algebraics then you need to use single tick quotes in RPL mode.

If you want to do calculations in RPL and somehow maintain a history of what was typed so that you can go back and edit it, then that is not possible on the 50G. How would the calculator know when you started a calculation so it would know how far back to let you go? Or would you want it to keep a record all the way back to when you turned it on?

If you want to do something else, then I've not understood and a specific example might makes things clearer.

[jimbo1qaz]

(01-17-2016 12:44 AM)BruceH Wrote:  If you want to do calculations in RPL and somehow maintain a history of what was typed so that you can go back and edit it, then that is not possible on the 50G. How would the calculator know when you started a calculation so it would know how far back to let you go? Or would you want it to keep a record all the way back to when you turned it on?

That's unfortunate.

[rprosperi]

To an experienced TI-84 user, the 50g certainly does feel awkward, and the opposite is equally true. I can attest to the latter personally, I tried.

It sounds like you are trying to use the 50g the same way you use the TI, and it simply will not work that way, at least not comfortably or easily, as it wasn't intended to be used that way. But I believe you will find it is very consistent and predictable, once you spend time learning to use it as intended.

In the opposite sense, as an experienced RPL user, I tried to pick and use my son's TI-84 and was not only confused, I couldn't even get it to do simple things. Nearly ready to just toss it, my son gave me some sage advice: RTFM. So, I did and suspended all prior "knowledge" of how to use a calculator intuitively (which after 40 years of RPN and RPL was extremely biased). But by just following the manual (and some online tutorials I found) I was suddenly able to use it and do meaningful stuff.

Ultimately, it was just idle curiosity, and I petty much dropped it, but the exercise was primarily to understand just how 'good' or 'bad' it really was. Obviously it turned out it was neither of those, just different, or maybe more to the point, it was not to my taste.

All that comes down to this: try to learn using the 50g the way it was intended to be used and I think you will be far more satisfied, but more importantly, will be able to effectively use some of the features that justified your switching in the first place.

HTH

[jimbo1qaz]

(01-17-2016 03:12 AM)rprosperi Wrote:  But by just following the manual (and some online tutorials I found) I was suddenly able to use it and do meaningful stuff.

I never knew the TI-84+ had a manual, and I don't like the thought of slogging through it. I figured everything out through trial and error, and Googling. Why would HP calculators need a manual to use effectively? ... Either it's more difficult to relearn how to use a calculator, than to learn it initially. Or HPs are just less intuitive or easy-to-use.

Also, the 50g is more sophisticated and complex than the 84+ (CAS, units, ARM, etc.), though unfortunately not much faster (in my experience).

What's the difference between the 50g manual, and the guide? The guide is longer.

[Massimo Gnerucci]

The RPN way. :)

[Marcio]

(01-17-2016 03:46 AM)jimbo1qaz Wrote:  ... the 50g is more sophisticated and complex than the 84+ (CAS, units, ARM, etc.), though unfortunately not much faster (in my experience).

If you are having too much trouble with the 50g and don't want to read the manuals (yes, it's necessary due to the complexity of the machine), why don't you just drop it and pick up an HP Prime? The Prime is MUCH faster and easier to use. For those who have some kind of experience with TIs and want to take the HP road for a change, the Prime offers the shortest path to satisfaction!

[rprosperi]

(01-17-2016 03:46 AM)jimbo1qaz Wrote:  I never knew the TI-84+ had a manual, and I don't like the thought of slogging through it. I figured everything out through trial and error, and Googling. Why would HP calculators need a manual to use effectively? ... Either it's more difficult to relearn how to use a calculator, than to learn it initially. Or HPs are just less intuitive or easy-to-use.

Also, the 50g is more sophisticated and complex than the 84+ (CAS, units, ARM, etc.), though unfortunately not much faster (in my experience).

What's the difference between the 50g manual, and the guide? The guide is longer.

Perhaps if you trial/error and google with the 50g as long as you did with the TI84, you'll learn as much, but expecting to effectively use a machine with literally thousands of functions that way isn't reasonable. With power comes the need to learn to use it well.

I agree with Marcio the Prime may have been a better choice (for easier transition) but since you already bought a 50g, we're trying to give you advice on how best to accomplish what you want. Both manuals are useful, but much of the material in both of them is reference (to look up how to use specific functions for specific applications), so one would not need to read them all the way through. Since you can freely download them, I suggest you get copies of both and peruse the contents, I think it will be fairly obvious which parts you will need to use.

Regarding performance, I believe you will find the 50g notably faster when running programs (vs. real-time response) however I honestly can't be sure as I've not done much programming on the TI84.

Good luck with learning the 50g, I think you will be impressed with it's capabilities if you have the patience to work through the transition.

[Bob Frazee]

My first 2 calculators were the TI SR10 and SR11, from the early 70's. An instructor in one of my engineering classes had an HP 45, which had all the bells and whistles of that era. I was hooked. Learning how to manipulate the stack took a little bit of doing for me, but once mastered, I marveled at the ease of doing manual calculations with a stack (no parenthesis), and still today I prefer RPN/RPL over algebraic for manual calculations.

Stack calculations require a different mind set than algebraic. You Must understand how the stack operates, or you'll just frustrate yourself. I visualize the stack as just a stack of numbers, one on top of the other. RPN generally implies a 4 level stack, RPL generally implies an infinite stack (at least until you run out of memory). You have more instructions available to manipulate the RPL stack, but for both, you're generally working with the bottom 2 registers, RPN calls them X and Y, RPL calls them Level 1 and Level 2. There are tutorials on how the stack operates on this website. Please check them out.

For performance of user written programs (I program in User RPL on the HP50g, and TI Basic on the TI89 Titanium). I have found that the programs that I run (mostly engineering types), will generally get an answer within 2 or 3 seconds of each other, which is close enough for me. If you want to check one out, there is a program called FluidTI for the TI84+ (located at ticalc.org), and the same program called FluidHP for the HP50g (located at hpcalc.org). That should give you some idea of program performance.

Perservere in learning how the stack operates, and I don't think you'll regret your decision for purchasing the HP50g.

[smnsparty]

Sorry to be bumping an old thread, but I think this is actually a very helpful discussion (full disclosure: I've been a TI-89 user since 1999, and have owned an HP-50g since 2013).

The 50g has remarkable computational and programming potential for the price (while it was still in production), but unfortunately, unlike TI products the stock HP software/firmware has a very Jekyll-and-Hyde nature, which can be quite jarring for the uninitiated.

The OP's observations are also my main concerns from the standpoint of having easy and convenient access to a sophisticated CAS in a handheld device, but I believe the following are also worth adding to the list (since I'm relatively inexperienced with the device despite having owned it so long, please feel free to point out any system flags or scripts/workarounds which might easily address the items below).

1. Textbook mode (aka Pretty Print in the TI/Python universe) for the stack seems not to have been fully implemented; the moment I hit UA to browse the stack, the display regresses into looking like a TI-83 of yore, complete with nested-parenthetical one-liner infix notation entries forming a veritable wall of ASCII -- in fact, I swear I actually end up staring at algebraic mode (as opposed to EQW or textbook mode) expressions at least 3x longer on the 50g than I ever do on the TI-89!
2. RPN mode seems to rely a fair amount upon UNDO to fix mistaken entries or operations, but it inexplicably offers only a single level of undo, and to add to the confusion, whilst 'redo' is technically feasible, it's hidden inside the cryptically named CMD feature -- to compound the confusion even further, despite the name, CMD only captures the 4 most recent operands pushed onto the stack, and instead entirely leaves out the operators and functions that had been applied!
3. The 50g strongly favours all-caps, even in CAS mode -- if one wishes to use lowercase letters, any keystroke advantage provided by RPN or the EQW relative to other devices instantly vanishes due to the repeated Alpha-LShift nonsense (and locking lowercase alpha isn't much better, as it overrides critical items like the inverse, exponential, trig functions, etc.)
4. The system font is horrendously ugly, and the smaller fonts are often illegible (especially when using a lot of lowercase letters) -- rather surprising from a company with a long history of solid PostScript support for its printers
5. The only way COPY/CUT/PASTE seem to work in RPN mode is if they're accompanied by the BEGIN + END markers -- these seem like unnecessary additional keystrokes to me, and this might've been a missed opportunity to instead, for example, allow holding LShift whilst using the cursor keys to highlight the desired items, just like we're used to doing on computers or calculators from other manufacturers
   

I would be very interested to know if there are easy fixes or workarounds to the above points.

Last but not least, I haven't been very impressed with the calculator's documentation, and have instead lurked on various fora over the years and watched YouTube videos to find out what I currently know about the 50g (which admittedly isn't much -- the above drawbacks have held me back from delving deeper into its feature set). To be fair, the manuals for the TI-89 Titanium aren't great either -- in stark contrast to the original TI-89 Guidebook -- now that document is an absolute masterpiece of accessible, hands-on technical writing with beautiful illustrations and overall typesetting, and a very large reason why I know the 89 like the back of my hand.