WP43 Alternative key layout --> C43

[Jaymos]

1) For me, the keyboard layout developed in this discussion looks much better, than the original layout by Walter.

Welcome

Thank you. I agree.

2) Concerning the last discussion about the register view (browser) and flag view, i suggest the key labels REG.V and FLG.V.

We had a solution with names that were close enough to the original and acceptable instead of STATUS, i.e. R.BR and B.BR.

That was no problem while one was on [f][7] and the other on [g][7]. The colours and the horizontal proximity made it easy to distinguish.

The problem now, is for good reason Dani suggested to move it to the [g] cursor keys. Now both are blue, far apart vertically, and my mind (and therefore extrapolated to unsuspecting newcomers) it is ergonomically not good to have almost identical labels, in the same colour, vertically apart.

So, we are looking for names that are sufficiently different to make them stand out for themselves.

REG.V and FLG.V are also too similar.

Currently we have REGS and FLAG.V instead of the original R.BR and STATUS, and also instead of the current R.BR and B.BR. I will send and image soon after changing the emulator.

3) Please do not change the color of the key labels of the letters A-Z to red as someone suggested here some pages before. Red on an dark grey, black or dark brown is unreadable for people with red/green defect on their eyes. We need an bright/dark contrast to recognize.

Heard. Idea was discarded already. Thanx.

4) I don't understand the need of the DROP button. We can use the left-arrow button for this like it is done on the HP50G. In Walter's manual, the left-arrow button makes a CLRX, it fills the X register with zero (if i understand it right). I think, this is a nearly useless function - what do you want to do with this zero in X register?

Drop rolls down and discards X. I use it instead of CLX and I don't want to loose it.
I don't think CLX is useful. You just type over X.
Personally I like a scheme where one tap on [<--] does CLX, and another (within a time limit does a DROP). (The 42s iOS app from Byron Foster does that).

5) We have a lot of unit conversion values and constants in the calculator. We also can display an integer number as binary or hexadecimal number of different length and one's-complement or two's-complement. What i miss is the conversion (or more a display) of a real value according to IEEE754. For example, i have the value -3.1415926535E-7 in the X register and i want to have it displayed hexadecimal or bitwise as a single float (4 byte) and decomposed the sign, exponent and mantisse part (the following display is only an example):
hex: AB CD EF 01
sign: 1
exponent: 3 BC
mantisse: D EF 01
It should be possible to switch the endianess and between float/double. The extended formats could be leave out, because they are more complicated due to the lenght of mantissa and exponent are implementation dependend.

This question is valid and should be directed to Walter on the "43S News" group.
Describe it well, motivate it well, and who knows.

[Patrick]

The work you all do here is really fantastic.

Currently, I am able to compile the WP43s emulator in order to quickly track and test the changes made to it.

I was wondering if it was possible to do the same with the version you are developing and, if so, in what way.

Thanks,

Patrick

Great.

Do you use Windows or Mac or Linux?

Windows 10

[Jaymos]

The work you all do here is really fantastic.

Currently, I am able to compile the WP43s emulator in order to quickly track and test the changes made to it.

I was wondering if it was possible to do the same with the version you are developing and, if so, in what way.

Thanks,

Patrick

Great.

Do you use Windows or Mac or Linux?

Windows 10

Can the emulator that you compile on Windows run independently, on a different machine?

I want executable snapshots distributed to the contributors on this list, for further testing and feeling of the changed system. The executable on the Mac that I use is not running externally on my wife's Mac, so I have not tried to distribute anything.

It would really help if you can generate a test executable, and test it on a non-dev machine. If it works, I can give you the changed files, and you can compile and distribute alpha versions.

J

[Jaymos]

2. When you bring up the HOME menu, you can currently scroll up and down to reach a total of 5 screens. This is not fixed. This is open for discussion.

A well known concept, which should be easy to get used to. I assume it is also possible to "scroll around" the end pages?

Yes, wraparound. Up goes to the other end immediately.

6. Of course none of this is required to operate the machine. This is an additional dimension of operation. Optional to use.

Some might embrace this fully, noone should be bothered by it, but maybe a global setting to enable this behaviour anyway?

The global setting is to disable the third shift to pop up the HOME menu.
Otherwise, it sits on a [g] shift and does not come up if you don't want it. Not needed for another flag.

[Jaymos]

6. Options:
6a. REGS maybe then. NO need for BR. Fits in. Short enough. Descriptive.
6b. REGISTR for a long option. Short may be better.
6c. GFLAGS for a long name.
6d. FLSTATS or FLGST or FSTAT or FSTATS also long options.

REGS, then? It's short, descriptive, and intuitive. And hints to the first R in RBR.

REG.B
FLG.V or FLG.S or GFS.V

and

P.FCN and X.FCN and ALPHA.FCN

would restore symmetry?

Edit

I see no benefit in FCN over FN. If no benefit to change, rather stay with WP43S.

[Jaymos]

... I added [COMPLX] to the 4th HOME screen for shiftless operation...

.. maybe some more complex math functions on the 4th HOME screen might involve the soft keys more in complex calculations and defend COMPLX's assignment there?

Edit: I should perhaps mention that I use my left thumb to press [SHIFT] and [ENTER] when I can. For single-handed operation, I'd probably use my right hand, and shiftless operation might be more appealing.

I would like suggestions of what to group in a practical grouping with COMPLX.

The benefit of the HOME menu over the normal menus, is that the HOME menu can easily group functions from other groups together, on primary keys if needed.

[Jaymos]

…
202. Fix Alpha key layout and menus that is broken due to layout change
…

I think you can change the letter assignment in 'assign.c' to the new old needs (primaryAim, fShiftedAim, gShiftedAim).

This is precisely where it has to change in the source.

But the problems are more specifically,

1. I need to first understand which shifted commands must be accessible when the Aim is active, and also the TAM input screen too.
2. When that is figured out, I need to see how that can be allowed with the changed main keyboard, because the required functions must be on the same places in both keyboards (I tried, it does not work if you try misalign labels on the screens).
3. Then I must see how those required positions influence the greek characters, and shifted alpha characters.
4. I do not want to change the greek layout - it already had to follow the alpha keys in a way. (I want the manuals to still make sense in a way).
5. So, my solution will probably be to make a custom MENU for the apha screen, so that all commands are accessed via the FN keys, not via the actual buttons which are conflicted with the Greek lettering,

[Patrick]

Great.

Do you use Windows or Mac or Linux?

Windows 10

Can the emulator that you compile on Windows run independently, on a different machine?

I want executable snapshots distributed to the contributors on this list, for further testing and feeling of the changed system. The executable on the Mac that I use is not running externally on my wife's Mac, so I have not tried to distribute anything.

It would really help if you can generate a test executable, and test it on a non-dev machine. If it works, I can give you the changed files, and you can compile and distribute alpha versions.

J

I think I could compile the files, I'm willing to try it. Currently, I use, as Martin suggested, MSYS2 to clone the Git of the original project on my computer and to compile the emulator.

If I replace the files that differ, I hope everything will work.

I am confident that the emulator could be distributed without any problems.

[Jaymos]

Windows 10

Can the emulator that you compile on Windows run independently, on a different machine?

I want executable snapshots distributed to the contributors on this list, for further testing and feeling of the changed system. The executable on the Mac that I use is not running externally on my wife's Mac, so I have not tried to distribute anything.

It would really help if you can generate a test executable, and test it on a non-dev machine. If it works, I can give you the changed files, and you can compile and distribute alpha versions.

J

I think I could compile the files, I'm willing to try it. Currently, I use, as Martin suggested, MSYS2 to clone the Git of the original project on my computer and to compile the emulator.

If I replace the files that differ, I hope everything will work.

I am confident that the emulator could be distributed without any problems.

Thank Patrick, please try it.

Maybe test it on a clean machine to gauge whether it would work on other testers' devices. I also have a Windows 10 machine here, but I have a few dev things on there, so I can also test it.

In the mean time I would do some revamping of the source files prior to sending to you. Let me know.

The files I have modified are all in the src folder, and are .c and .h files. There are also 2 files in the root folder to replace, i.e. a CSS file and PNG.

[Patrick]

Can the emulator that you compile on Windows run independently, on a different machine?

I want executable snapshots distributed to the contributors on this list, for further testing and feeling of the changed system. The executable on the Mac that I use is not running externally on my wife's Mac, so I have not tried to distribute anything.

It would really help if you can generate a test executable, and test it on a non-dev machine. If it works, I can give you the changed files, and you can compile and distribute alpha versions.

J

I think I could compile the files, I'm willing to try it. Currently, I use, as Martin suggested, MSYS2 to clone the Git of the original project on my computer and to compile the emulator.

If I replace the files that differ, I hope everything will work.

I am confident that the emulator could be distributed without any problems.

Thank Patrick, please try it.

Maybe test it on a clean machine to gauge whether it would work on other testers' devices. I also have a Windows 10 machine here, but I have a few dev things on there, so I can also test it.

In the mean time I would do some revamping of the source files prior to sending to you. Let me know.

The files I have modified are all in the src folder, and are .c and .h files. There are also 2 files in the root folder to replace, i.e. a CSS file and PNG.

There is only one way to know if it will work, and that's to try it.

Send me the files as soon as possible, I'll try to compile them.