Threaded Mode | Linear Mode

Claudio L. · 07-29-2021, 02:55 PM

(07-28-2021 10:52 PM)spiff72 Wrote: Sorry again for all the questions...

I have a few key functions I don't understand:
- What does the APPS key do (unshifted)? It seems to make the stack disappear, and pressing again makes it reappear. When the stack is gone, it won't accept entries to the stack.

- For the base rotation you described (using shift hold), do both shift buttons do this, but in opposite directions? I made this assumption for my graphic, but can't figure out how to test this on the simulator (simulate a hold).

From the list I provided earlier of right and left shifted items that are missing from the PrimeG1 layout (compared to the 50g), it looks like only the FINANCE, TIME, ARITH, and CMPLX are unaccounted for. I managed to find ARITH and CMPLX in the menu structure, but can't find TIME or FINANCE. Could those be assigned to the 1, 4, or 7 key as the LS and RS assignments, respectively? The same could be done for ARITH and CMPLX to one of the remaining of that group (if needed).

If you are missing the :-> symbol, maybe that could be the LS for the 0 key, or could replace the RS function for either 5 or 8 (since those are both <<>> currently)?

Finally, I am a bit confused by your comments about the X and Y keys being on menu keys (when outside of alpha). Are you suggesting that those SHOULD appear elsewhere? But if that is the case, wouldn't that apply to R, S, V and W as well? I agree that the unshifted C and D keys are valuable real estate, but I am not sure what would be a good candidate for those locations.

I am attaching an updated version of the graphic file with my best effort at documenting all the keys (including crude shift-hold nomenclature). I couldn't figure out how to generate the down and right arrow for carriage return, so I just used CR for that. I also wasn't sure if the "approx =" (LS Enter) was real or not, so I didn't blank that out yet.

EDIT: Note that I did NOT include any of the suggestions I made above in this graphic.

APPS switches to Form mode, in other words it runs the current Form, or App if you want to use a term these modern kids use :-)
But there's no Forms engine yet, so it just blanks the screen back and forth.

The 6 menu keys are 3 rows, 2 columns: Symb,Plot,Num, Help, View and Menu. It takes a little bit of getting used to this "vertical" menu display. The CAS key switches which menu is active with the hard keys. Unfortunately, we don't have 12 dedicated menu keys like in the 50g, so we had to settle with 6 keys, plus one key to "switch". On the other side, as soon as we get "touch" working, you'll use your preferred menu with hard keys and the other one you can use touch so you don't need to switch back and forth using CAS.

My comment was because the X and Y keys are two of the dedicated menu keys (Symb and Help), so they can only be active in Alpha mode, whereas the X key on the 50g is active all the time as X (and is also X in alpha mode, which is ideal).

spiff72 · 07-29-2021, 03:10 PM

I just reset and restored the archive to my 50g, and it DOES seem like it took the fonts with it, as well as my black text on white background setting. It even restores the contents of the stack.

I also tried a complete reset on the 39gs, and then restored my archive for that calc. It changed the inversion of the colors in the menu, but fonts stayed at default. I am 100% sure that I had those bigger, so not sure why that isn't coming through. I will try setting the fonts again and then do a new archive/restore and see if they transfer.

spiff72 · 07-29-2021, 03:12 PM

(07-29-2021 02:55 PM)Claudio L. Wrote:
(07-28-2021 10:52 PM)spiff72 Wrote: Sorry again for all the questions...

I have a few key functions I don't understand:
- What does the APPS key do (unshifted)? It seems to make the stack disappear, and pressing again makes it reappear. When the stack is gone, it won't accept entries to the stack.

- For the base rotation you described (using shift hold), do both shift buttons do this, but in opposite directions? I made this assumption for my graphic, but can't figure out how to test this on the simulator (simulate a hold).

From the list I provided earlier of right and left shifted items that are missing from the PrimeG1 layout (compared to the 50g), it looks like only the FINANCE, TIME, ARITH, and CMPLX are unaccounted for. I managed to find ARITH and CMPLX in the menu structure, but can't find TIME or FINANCE. Could those be assigned to the 1, 4, or 7 key as the LS and RS assignments, respectively? The same could be done for ARITH and CMPLX to one of the remaining of that group (if needed).

If you are missing the :-> symbol, maybe that could be the LS for the 0 key, or could replace the RS function for either 5 or 8 (since those are both <<>> currently)?

Finally, I am a bit confused by your comments about the X and Y keys being on menu keys (when outside of alpha). Are you suggesting that those SHOULD appear elsewhere? But if that is the case, wouldn't that apply to R, S, V and W as well? I agree that the unshifted C and D keys are valuable real estate, but I am not sure what would be a good candidate for those locations.

I am attaching an updated version of the graphic file with my best effort at documenting all the keys (including crude shift-hold nomenclature). I couldn't figure out how to generate the down and right arrow for carriage return, so I just used CR for that. I also wasn't sure if the "approx =" (LS Enter) was real or not, so I didn't blank that out yet.

EDIT: Note that I did NOT include any of the suggestions I made above in this graphic.

APPS switches to Form mode, in other words it runs the current Form, or App if you want to use a term these modern kids use :-)
But there's no Forms engine yet, so it just blanks the screen back and forth.

The 6 menu keys are 3 rows, 2 columns: Symb,Plot,Num, Help, View and Menu. It takes a little bit of getting used to this "vertical" menu display. The CAS key switches which menu is active with the hard keys. Unfortunately, we don't have 12 dedicated menu keys like in the 50g, so we had to settle with 6 keys, plus one key to "switch". On the other side, as soon as we get "touch" working, you'll use your preferred menu with hard keys and the other one you can use touch so you don't need to switch back and forth using CAS.

My comment was because the X and Y keys are two of the dedicated menu keys (Symb and Help), so they can only be active in Alpha mode, whereas the X key on the 50g is active all the time as X (and is also X in alpha mode, which is ideal).

Understood - thanks!

spiff72 · 07-29-2021, 04:00 PM

(07-29-2021 03:10 PM)spiff72 Wrote: I just reset and restored the archive to my 50g, and it DOES seem like it took the fonts with it, as well as my black text on white background setting. It even restores the contents of the stack.

I also tried a complete reset on the 39gs, and then restored my archive for that calc. It changed the inversion of the colors in the menu, but fonts stayed at default. I am 100% sure that I had those bigger, so not sure why that isn't coming through. I will try setting the fonts again and then do a new archive/restore and see if they transfer.

I just set up all the fonts on the 39gs to my liking and created the USBARCHIVE, Factory reset, and restored the archive, and it worked perfectly. It also reports the correct fonts when choosing the various items from the font menu (this reports the wrong values when doing it from the factory reset state).

Alejandro Paz(Germany) · 07-29-2021, 04:27 PM

Quote:The file doesn't have any extra headers. The ROM merely changes the "KINPOHP39G+IMAGE" string at the beginning of the file with "Kinposhcopyright" (total 16 bytes changed). This is checked to see if the ROM is valid, so you need to do it or the calculator won't boot.
Once you replace that string, just write the whole file to 0x4000. Make sure you DON'T erase the blocks at 0 thru 0x3fff since that's the bootloader and it's not included in the rom file.

Best of luck.

I tried again, but starting the bootloader and selectic update code/usb, I thought maybe it unlocks the flash but it still gave error downloading.

Any possibility to flash the ROM using linux ? and how ?

thanks

Claudio L. · 07-30-2021, 02:44 AM

(07-29-2021 04:27 PM)Alejandro Paz(Germany) Wrote: I tried again, but starting the bootloader and selectic update code/usb, I thought maybe it unlocks the flash but it still gave error downloading.

Any possibility to flash the ROM using linux ? and how ?

thanks

Only using a VM with Windows and the original connectivity kit that came with the calculator (or downloaded from hpcalc.org).

Alejandro Paz(Germany) · 07-30-2021, 11:57 AM

(07-30-2021 02:44 AM)Claudio L. Wrote:
(07-29-2021 04:27 PM)Alejandro Paz(Germany) Wrote: I tried again, but starting the bootloader and selectic update code/usb, I thought maybe it unlocks the flash but it still gave error downloading.

Any possibility to flash the ROM using linux ? and how ?

thanks

Only using a VM with Windows and the original connectivity kit that came with the calculator (or downloaded from hpcalc.org).

I had to install the driver from here hpcalc HP3x, 4x, 50 Drivers manually (even after it supposedly installed itself after running the executable) and then I could download the ROM.
Now I have to see how to change the fonts, it is too small, I thought about the 39gII... it is not a target (yet) Sad

, that one has nice resolution.
Well at least it is RPN but it has the 50g layout, I suppose... a calculator friendly layout would be too much to ask I suppose... the 39 is far from the 48 :/

spiff72 · (This post was last modified: 07-30-2021 12:30 PM by spiff72.)

(07-30-2021 11:57 AM)Alejandro Paz(Germany) Wrote:
(07-30-2021 02:44 AM)Claudio L. Wrote: Only using a VM with Windows and the original connectivity kit that came with the calculator (or downloaded from hpcalc.org).

I had to install the driver from here hpcalc HP3x, 4x, 50 Drivers manually (even after it supposedly installed itself after running the executable) and then I could download the ROM.
Now I have to see how to change the fonts, it is too small, I thought about the 39gII... it is not a target (yet) , that one has nice resolution.
Well at least it is RPN but it has the 50g layout, I suppose... a calculator friendly layout would be too much to ask I suppose... the 39 is far from the 48 :/

I ended up settling on the following fonts on my 39gs (I may have the actual names of some of these incorrect, since the last characters can't be seen in the menu structure):
FNTSTK: Font7A
FNT1STK: Font8A
FNTMENU: Font6A
FNTFORMS: Font8A
FNTCMD: Font8A
FNTSTATUS: Font6A

Just go to the menu: System - Fonts

From there, FNTSTK menu item will return the current value of this font setting (but it may not be correct due to a bug). Edit that value by pressing the down arrow to invoke the editor, change the name of the font, press enter to put it on the stack, then press the ->FNTSTK menu key to set that font. Repeat for each of the font settings as listed above. (My list above is in order of appearance in the Fonts menu). Each font will show up (you have to press the NXT button to get to the others), and they always appear with the FNTxxx item, followed by ->FNTxxx. I find it easier to recall the font with the FNTxxx item, edit it, and then set it with the ->FNTxxx menu item (rather than typing in the full font name each time).

Font8C is slightly wider than Font8A if you would prefer that.

Claudio L. · 07-30-2021, 04:12 PM

(07-30-2021 11:57 AM)Alejandro Paz(Germany) Wrote:
(07-30-2021 02:44 AM)Claudio L. Wrote: Only using a VM with Windows and the original connectivity kit that came with the calculator (or downloaded from hpcalc.org).

I had to install the driver from here hpcalc HP3x, 4x, 50 Drivers manually (even after it supposedly installed itself after running the executable) and then I could download the ROM.
Now I have to see how to change the fonts, it is too small, I thought about the 39gII... it is not a target (yet) , that one has nice resolution.
Well at least it is RPN but it has the 50g layout, I suppose... a calculator friendly layout would be too much to ask I suppose... the 39 is far from the 48 :/

There's another thread (or was it somewhere in this thread?) with a nice template to create overlays for the 39gs keyboard. I have one and it looks VERY nice.
Changing the fonts is straightforward, just type the name of a system (or user) font and use ->FNTxxx to use that font on a particular area of the screen (see the wiki, Chapter 6 and open the Fonts commands to see what I mean.

Besides creating your own custom fonts, which isn't all that hard to do, there's quite a few system fonts you can use simply by typing the name as a variable name 'Font5A' will use the 5 pixel high, Style A font.

You can see all available system fonts directly in the source code, if you click on the bitmaps you can even preview them in your browser before you choose.

The name on the calculator is always "Font" (case sensitive), the height in pixels and the style, as in 'Font10A'.

compsystems · 08-02-2021, 12:47 AM

Hello
With the latest release of NewRPL, I am motivated to use the simulation version.

In the PC version, the following key combinations are not working

Quote:While the Vars menu allows the user to enter into sub-directories, it does not provide any means to exit back into the parent directory. This is achieved with the UPDIR command (always accessible through the keyboard shortcut LeftShift-UP).

To go directly to the home directory, the HOME command (also always accessible at LeftShiftHold-UP) is used.

Petition:
: It is necessary that when pressing the virtual key some effect of pulsation is carried out, like for example inverting the color of the area of pulsation.

: To hold down the keys, without using a keyboard, an alternative may be to right-click the button

Claudio L. · 08-02-2021, 04:34 PM

(08-02-2021 12:47 AM)compsystems Wrote: Hello
With the latest release of NewRPL, I am motivated to use the simulation version.

In the PC version, the following key combinations are not working

Quote:While the Vars menu allows the user to enter into sub-directories, it does not provide any means to exit back into the parent directory. This is achieved with the UPDIR command (always accessible through the keyboard shortcut LeftShift-UP).

To go directly to the home directory, the HOME command (also always accessible at LeftShiftHold-UP) is used.

Petition:
: It is necessary that when pressing the virtual key some effect of pulsation is carried out, like for example inverting the color of the area of pulsation.

: To hold down the keys, without using a keyboard, an alternative may be to right-click the button

The UPDIR and HOME key combinations work just fine in the simulator. Both using the mouse and the keyboard.
Left shift is the Caps Lock key on the keyboard, right shift is Ctrl and Alpha is Tab.

I agree it would be nice to show some feedback, but it's very low priority right now.

Alejandro Paz(Germany) · (This post was last modified: 08-05-2021 08:56 AM by Alejandro Paz(Germany).)

Thanks for the info, I changed the fonts and now it is better. The mentioned overlays look very good. I think one get for like 35 € ? some vinyl overlays, that would be good. I have to search for the company again, as a matter of fact, there are several offers.

lmamakos · 08-06-2021, 05:11 AM

(08-05-2021 08:43 AM)Alejandro Paz(Germany) Wrote: Thanks for the info, I changed the fonts and now it is better. The mentioned overlays look very good. I think one get for like 35 € ? some vinyl overlays, that would be good. I have to search for the company again, as a matter of fact, there are several offers.

I've never sold overlays for newRPL that I did for the HP 39gs/40gs/50g. Did someone else do that previously? If so I missed it, and it would have saved me a bunch of time writing OpenSCAD scripts and fiddling around in Inkscape. Oh, and convincing my wife she needed a cutting machine :-)

I recall that I had read of some overlays for sale for WP-34S conversions, but that's a different calculator body.

I'm not intending to go into the overlay business, but if someone would like to adapt the files that I produced, they're welcome to do so.

Wes Loewer · 08-07-2021, 01:13 PM

One of my to-do tasks for the summer was to install newRPL. Yesterday I finally got around to putting in on a 50g and Windows. This is really quite impressive. I'm still finding my way around, but here are my first day comments:

1. I noticed a bug on the ATAN function when used with complex arguments. For instance, « (3,2) ATAN » should return (1.338972522294.,0.146946666226.) but instead returns (1.338972522294.,0.402359478109.)

2. I noticed was there was not a LASTARG command. Was this intentional or are there plans to implement it later? Or did I just miss it?

3. Three lines on the screen seemed like a lot of screen space to use for menus and indicators, but I'm getting used to it.

4. I love how the calculator responds so quickly that it turns on before I finished depressing the ON key.

I'm looking forward to exploring more. Great job!
-wes

Wes Loewer · 08-07-2021, 02:20 PM

(08-07-2021 01:13 PM)Wes Loewer Wrote: 1. I noticed a bug on the ATAN function when used with complex arguments. For instance, « (3,2) ATAN » should return (1.338972522294.,0.146946666226.) but instead returns (1.338972522294.,0.402359478109.)

I could be mistaken, but I think the problem can be traced back to an algebraic error which fortunately is shown clearly in the comments above the code.
See lib-66-transcendentals.c, lines 1444-1445

Code:

// Denominator = (1-im^2) - (i*re)^2

// = 1-im^2+re^2 (real number)

The code that follows seems consistent with this comment.

However, if I'm not mistaken, I think that should have been

Code:

// Denominator = (1-im)^2 - (i*re)^2

// = (1-im)^2+re^2 (real number)

which changes the final result.

compsystems · 08-07-2021, 05:41 PM

the image could be used as an alternative mask to the current one.
The submenu is best displayed in dark gray instead of black.

$[Image: hp50g_newRPL_skin03.png]$

Claudio L. · (This post was last modified: 08-07-2021 09:41 PM by Claudio L..)

(08-07-2021 01:13 PM)Wes Loewer Wrote: One of my to-do tasks for the summer was to install newRPL. Yesterday I finally got around to putting in on a 50g and Windows. This is really quite impressive. I'm still finding my way around, but here are my first day comments:

1. I noticed a bug on the ATAN function when used with complex arguments. For instance, « (3,2) ATAN » should return (1.338972522294.,0.146946666226.) but instead returns (1.338972522294.,0.402359478109.)

Aargh... what an embarrassment. Dumb bug, I fixed it, will come out on the next unofficial release (probably next week).Thanks for the report! We actually need testers, lots of them.

(08-07-2021 01:13 PM)Wes Loewer Wrote: 2. I noticed was there was not a LASTARG command. Was this intentional or are there plans to implement it later? Or did I just miss it?

There's no LASTARG. However...user input gets automatically pushed to the UNDO stack when you run a command from the keyboard. You can get any value from that last user input using STKPICK.

(08-07-2021 01:13 PM)Wes Loewer Wrote: 3. Three lines on the screen seemed like a lot of screen space to use for menus and indicators, but I'm getting used to it.

Press and hold VAR to switch back to 1 menu mode if you like, but having your variables menu always available is a time saver.

**EDIT**: On+VAR, not just VAR. Long press.

(08-07-2021 01:13 PM)Wes Loewer Wrote: 4. I love how the calculator responds so quickly that it turns on before I finished depressing the ON key.

I'm looking forward to exploring more. Great job!
-wes

Thanks for giving it a try, keep the bug reports coming!

Wes Loewer · 08-08-2021, 06:08 PM

(08-07-2021 09:28 PM)Claudio L. Wrote: Thanks for giving it a try, keep the bug reports coming!

Is there a preferred venue to report bugs and such?

While on the topic of complex numbers, I noticed that most operations return a complex value in the same cartesian or polar format as the original argument. However, EXP seems to always return a value in polar format, and LN always returns a value in cartesian format, regardless of the format of the argument. Was this intentional?

For consistency, it seems like it would be better to always return the value in the same format like the other commands.

Claudio L. · 08-09-2021, 01:35 PM

(08-08-2021 06:08 PM)Wes Loewer Wrote: Is there a preferred venue to report bugs and such?

Yes, I can give you access to our bug tracker. PM me with a valid email address and I'll set it up if you want.

(08-08-2021 06:08 PM)Wes Loewer Wrote: While on the topic of complex numbers, I noticed that most operations return a complex value in the same cartesian or polar format as the original argument. However, EXP seems to always return a value in polar format, and LN always returns a value in cartesian format, regardless of the format of the argument. Was this intentional?

For consistency, it seems like it would be better to always return the value in the same format like the other commands.

The general rule is all results are in the same format as the first argument. This is valid for all commands, when you operate with arguments in different formats (if you add a hex number to a decimal number, result is hex, if you add a unit 1_km + 1_m the result is in _km, etc.
LN and EXP are an exception because of:
a) Usefulness: One of the more important side effects of LN and EXP is the sort-of conversion from polar to cartesian.
b) Performance and precision loss: The conversion involves folding the number to the first quadrant, so it wouldn't be reversible. It also adds a little noise on the last few digits due to the trig functions. Right now EXP and LN are fully reversible (don't worry, I just saw 2 more bugs in LN and fixed them, will send updated firmware in a couple of days, I'm surprised nobody caught it before).

Wes Loewer · 08-09-2021, 07:25 PM

(08-09-2021 01:35 PM)Claudio L. Wrote: Yes, I can give you access to our bug tracker. PM me with a valid email address and I'll set it up if you want.

Okay, will do. The following aren't bugs but rather more design decision questions.

I was relieved to see that ADD and + are reversed for list processing. This was a longstanding shortcoming in the 50g and its ancestors. (For those unfamiliar with this issue, see https://groups.google.com/g/comp.sys.hp4...rvP3IHIdEJ from 10+ years ago.)
I wonder though if using a word like CONCAT might be better than ADD as it is more descriptive. Perhaps you could make CONCAT a synonym but keep ADD for historical reasons.

Another quirk of the 50g that I found annoying when programming was the behavior of the FOR and START loops in that they alway execute at least once, even when the condition would indicated that the loop should not be executed at all.

In other languages that I am familiar with (pascal, c/c++, java, hp ppl), something equivalent to

Code:

 FOR I FROM 5 TO 1 DO PRINT(I) END

would not print anything at all since the exit condition is already met before the loop begins. However, on the 50g, FOR loops are always executed at least once. So instead of something like

Code:

A B FOR I ... NEXT

you'd have to nest it within an IF statement.

Code:

IF A B <= THEN A B FOR I ... NEXT END

which always struck me as a bit of a kludge.

I noticed that newRPL reproduces this behavior. I can understand the pressure to do this to maintain compatibility, but I wondered since you were willing to break compatibility with ADD/+ for a good cause, then maybe you'd be willing to do the same with the FOR and START loops in order to be more logical and more consistent with other languages.

(Are there other languages that execute FOR loops at least once like this?)

Thoughts?