I don't understand the HP-50g, what am I missing?

[cdmackay]

One further point regarding MENU.

Hold down left-shift and press the MODE key, release left-shift, then press MENU, then MENU again.

Whether that's easier than just typing MENU manually is another matter

[Voldemar]

(04-11-2024 05:07 PM)cdmackay Wrote:  Hold down left-shift and press the MODE key, release left-shift, then press MENU, then MENU again.

Thank You. Useful.

[carey]

It’s interesting that this is the most helpful discussion of the many peculiarities of the HP50g I’ve seen and it comes 18 years after the HP50g was introduced and 9 years after it was discontinued. No point, just interesting.

[drbarnack]

(04-08-2024 08:34 PM)bxparks Wrote:  5) How do I Invoke the Equation Writer on a Previously Entered Equation?

One of the first examples of using the Equation Writer (RightShift+EQW) is on page 1-13, which has the following example, sqrt(3 * (5-1/(3*3))/(23^3) + e^2.5), which looks like this in MathJax/Latex mode:

\[
\sqrt{\frac{3\cdot(5 - \frac{1}{3 \cdot 3})}{23^3}+e^{2.5}}
\]

I have tried to enter this equation into EQW about 4-5 times over 3 months, and I have never been able to enter it correctly. I cannot seem to be able figure out how to use the arrow keys to get to the correct location to remove or add the correct parenthesis or term. (Amusingly, I was able to enter this expression correctly in RPN mode on the first try.)

So I enter the incorrect equation, and it shows up on the RPL stack, and I can see that it is wrong. How do I edit that equation?? I have tried everything:

• I scroll up to that term with the arrow keys, hit RightShift+EQW, and my expectation is that it should copy that back into the EquationWriter. Nope.
• I scroll up, hit RightShift+COPY, then go into RightShift+EQW, then hit RightShift+PASTE. Nothing
  

I have searched the User's Guide for information on how to edit a previously entered equation. It must be in there somewhere in the 887 pages, but I cannot find it.

This had been bothering me too, so I think I tracked it down. On page G-2 of the AUR it says you can hold down left shift and down arrow ("LS & DA") to invoke VISITB ("VISIT Best?") for variables, EDITB for everything else.

And then looking up VISITB/EDITB in the AUR says that for a specified variable, opens the contents in the most suitable editor for the object type. For example, if the variable holds an equation, open in the Equation Editor.

(FWIW I'm coming from both a HP 48 GX and a HP Prime. I wasn't sure if I was going to like the keyboard layout but it's fine, the SD card slot makes getting stuff back and forth from the PC much easier, and I feel like the proverbial snail riding on the back of a turtle -- I was blown away by Fast3D plot.)

[bxparks]

I'm glad to hear that some people are finding this thread useful. I do want to recognize the tremendous amount of work that the developers must have spent creating this product. I am not trying to diminish their accomplishments or be unfairly critical. The 50g looks and feels like a great product. I want to like this calculator. I am just having difficulty figuring out how to use it.

In this post, I want to follow up on the topic referenced by 2-3 people, the 50g menu system.

(04-10-2024 08:28 PM)Voldemar Wrote:  There is no soft menu «button» TOOL, there is a physical button TOOL.

Are you sure that this matters? There shouldn't be any difference whether a menu gets triggered by an unshifted key (like TOOL) or a shifted key (like LeftShift+MTH or RightShift+CMPLX).

(04-11-2024 11:56 AM)David Hayden Wrote:  [...]
Menus are not hierarchical. There is no "parent menu," only the previous menu that you were using. Unfortunately, I don't think the 50g remembers more than one previous menu.
You've mentioned parent and child "folders" a few times.
[...]

You are correct that the 50g User's Guide never uses the word "folder" in the context of menus. It only uses "folder" to explain the concept of "directories" which are used to organize storage variables.

However, the 50g menu system definitely seems hierarchical to me, just like the HP-42S. A good example is the LeftShift+MTH menu, which provides at least 3 levels of hierarchy. The top-level 'MTH' menufolder has many sub menufolders, one of them is 'MTH > MATRIX' on the second level. The 'MTH > MATRIX' menufolder in turn has a 'MTH > MATRIX > MAKE' subfolder on the third level. Each sub-menufolder has a back-link to its parent menufolder.

The 50g menu system seems to have 2 types of nodes in the hierarchical tree:

- a MenuFolder, that holds one or more nodes (MenuFolder or MenuItem), and
- a MenuItem, which invokes a function, command, or operation.

I know that the 50g distinguishes between these 2 node types, because it uses 2 different icons for them: A MenuFolder has a small dash above the menu box, a MenuItem does not. That little dash turns a menu box into a file folder icon. (It was only a few weeks ago that I realized that this was a folder icon, after 10-20 years of wondering what those little lines were in screenshots of the 48-series calculators.)

One complication of the 50g menu system is that it seems to be a set of multiplely-rooted trees, not a simple singly-rooted tree. Each of the various menu buttons (e.g. TOOL, LeftShift+MTH, RightShift+CMPLX) seems to activate a separate hierarchy of menus. (Of course, internally, the 50g menu system could be implemented as a singly-rooted tree, with the top most "root" MenuFolder invisible to the user.)

The second complication is that the 50g menu system actually seems to be a directed acyclic graph (DAG), which allows the TOOL menu folder to contain a link to the STACK menufolder, which also appears in the PRG menufolder. So the STACK menufolder has 2 parents. If the menu system was strictly hierarchical, every node would have only a single parent. But I think it is convenient to continue to think of the menu system as mostly a set of singly-rooted trees.

Anyway, that's my mental model of the 50g menu system. If I think of it as an arbitrary directed graph (i.e. like a set of HTML hyperlinks), I get lost in a sea of proliferating menu items. A hierarchical organization is easier to remember and navigate. I think my biggest complaint about the 50g menu system is that it is difficult to navigate for a new user who is exploring the features of the calculator.

Brian

[cdmackay]

I find it easier to think of each menu as just an arbitrary collection of items, i.e. a layout, rather than part of a strict hierarchical design.

Menus (layouts) can contain pointers to other menus, variables, commands, etc.

They are just convenient groupings of these things. i.e. a command may appear in multiple menus.

There are often many different ways to get to the same underlying object.

Some menus are triggered by a key press. That's just one way to display a menu.

Each menu has a number. That can be used with the MENU command, to display any menu (e.g. within a program). Or manually, e.g. <<96 MENU>> shows the STAT menu from the HP48.

The 50g also knows about many of the menu layouts from the HP48G-series, and they can be accessed via their menu numbers, using the MENU command.

e.g.

0 MENU # last menu, also ↰+PREV
1 # CST custom menu
2 # VAR
30 # HP48 "white menu" solver
63 # MODES
74 # HP48 solver, also ↱+NUM.SLV
81 # PLOT
93 # HP48 SYMB
94 # HP48 TIME, also ↱+TIME
96 # HP48 STAT
104 # HP48 I/O
110 # HP48 LIBRARY
2269 # lots of goodies library 221 Meta Kernel commands (2269 == 221+2048)

libraries < 256 are accessed via: 2048+#lib MENU
libraries ≥ 256 are accessed via: #lib MENU



So stop trying to make logical sense of the existing menu hierarchies, and just accept them as a partly-historical backwards-compatible mismash of lots of different ways to get to various related collections of underlying objects, e.g. vars, commands, etc.


What works for me is to make notes of the underlying objects (e.g. commands) that I use often, along with one (or more) ways to get to them, e.g. one of several menus. The AUR gives a menu path for every command, often the shortest menu path.

e.g. from my notes:

RCLKEYS ↰+MODE KEYS shows current key assignments


===

Note, in the above, I use + to indicate holding down. e.g. ↰+MODE means hold-down left shift and press MODE, then release both.

Whereas ↰MODE would mean press left-shift, release, then press MODE.

e.g:

⬇️ EDITB
↰⬇️ EDIT obj, VISIT 'name'
↰+⬇️ EDITB obj, VISITB 'name'

[carey]

(04-10-2024 07:53 PM)Eric Rechlin Wrote:  A lot of the inconsistencies are in optional features that you can ignore for simplicity and only learn if you are looking for extra efficiency as a power user. Once you get over a few hurdles, things will make sense and you'll be able to just treat it as a super-48.

As I’m fond of the 48GX UI, I like your perspective on the 50g as a “super-48” and would like configure a 50g to behave, as much as possible, like a “super-48.”

I think I can figure out many of the flag settings and key assignments that would move the 50g UI a little in the direction of the 48GX UI, but wonder if someone has already tried this or compiled a list of changes to make 50g behavior, as much as possible within the limitations of different key layout and menu options, more closely mimic the 48GX. I didn’t find an example in the forum archives, but may have missed it. If not, I’ll give it a try and report on the results. Or do you think a keyboard overlay would be needed to really make much difference?

Thanks!

[Steve Simpkin]

(04-19-2024 08:39 PM)carey Wrote:  

(04-10-2024 07:53 PM)Eric Rechlin Wrote:  A lot of the inconsistencies are in optional features that you can ignore for simplicity and only learn if you are looking for extra efficiency as a power user. Once you get over a few hurdles, things will make sense and you'll be able to just treat it as a super-48.

As I’m fond of the 48GX UI, I like your perspective on the 50g as a “super-48” and would like configure a 50g to, as much as possible, behave like a “super-48.”

I think I can figure out many of the flag settings and key assignments that would move the 50g UI a little in the direction of the 48GX UI, but wonder if someone has already tried this or compiled a list of changes to make 50g behavior, as much as possible within the limitations of different key layout and menu options, more closely mimic the 48GX. I didn’t find an example in the forum archives, but may have missed it. If not, I’ll give it a try and report on the results. Or do you think a keyboard overlay would be needed to really make much difference?

Thanks!

I would like to see a collection of tips as well. I can find bits and pieces of tips through searching but have not found a collection of them yet, particular for HP-48 users.

This site features comparison tables and photo galleries for various HP RPN LCD models. Here are some links to information on the HP 50g. You may find some of them helpful.
https://sites.google.com/view/hp-plus-ca...2zeobzgr2l

[carey]

(04-19-2024 09:33 PM)Steve Simpkin Wrote:  This site features comparison tables and photo galleries for various HP RPN LCD models. Here are some links to information on the HP 50g. You may find some of them helpful.
https://sites.google.com/view/hp-plus-ca...2zeobzgr2l

Thank you Steve! The site will be helpful.

[rprosperi]

(04-19-2024 09:33 PM)Steve Simpkin Wrote:  I would like to see a collection of tips as well. I can find bits and pieces of tips through searching but have not found a collection of them yet, particular for HP-48 users.

Not exactly a list of tips, buy if you've not seen it before, "One Minute Marvels" is an excellent collection of small, focused 48 programs written by Richard Nelson and Wlodek Mier-Jedrzejowicz. I've read it numerous times, but upon each re-reading I still learn new RPL things.

https://www.hpcalc.org/details/1691

[Klaus]

Although it refers to the 48, the principles are similar: https://literature.hpcalc.org/items/1571
Greetings!

[John Keith]

(04-20-2024 01:38 PM)Klaus Wrote:  Although it refers to the 48, the principles are similar: https://literature.hpcalc.org/items/1571
Greetings!

The 48-GX version here may be more relevant, it covers the advanced list processing and matrix functions of the GX. They are very good references- Bill Wickes is the chief developer of the 28 and the 48 series.