43S Alternative key layout --> WP43C

General discussion about calculators, Swiss Micros or otherwise
User avatar
Jaymos
Posts: 203
Joined: Sun Nov 04, 2018 6:03 pm
Location: Cape Town

Re: 43S Alternative key layout

Post by Jaymos » Sun Aug 11, 2019 1:49 am

Jaymos wrote:
Wed Jul 24, 2019 10:07 pm
Jaymos wrote:
Wed Jul 24, 2019 10:10 am
Dani R. wrote:
Wed Jul 24, 2019 5:55 am
You can set the alphabetical labels back to the original DM42 positions, i.e. K to COS, L to TAN, etc.
Good idea.
I will look at the repercussions, and do if it’s feasible.
J
The repercussions are large to change the text, as the alpha mode brings with it a large number of shifted greek letters, and also some commands. Changing this is a big deal to re-assign the greek lettering and to allow some commands and menus to still come through.

See p109 of this document (ed. ReM) to see the alpha mode layout:

The other problem is that my layout has already changed due to the shifts moving, the operators moving and the off button that moved. So I already have to analyse the effects of tis and make the alpha screen work properly.

So for now I will pend the re-arrangement of the alphas, unless I am any way forced to start from a clean page with alpha mode layout.

J
Regarding Dani's request to re-arrange the alpha keys - it is complete now.

You guys can test the alpha mode on the DM42 image now.

I removed all the commands that were interweaved with the greek letters and open buttons, as these all coincided with the letters. There is now a temporary ALPHA menu added bottom left on the screen
Clipboard01a.png
Clipboard01a.png (5.3 KiB) Viewed 176 times
which opens this menu and contains the commands previously in the way of the buttons
Clipboard02a.png
Clipboard02a.png (6.35 KiB) Viewed 176 times
For example, pressing a.INTL:
Clipboard04a.png
Clipboard04a.png (6.17 KiB) Viewed 176 times
Jaco Mostert
Elec Eng, South Africa
WP34C, HP42S, DM42 for complex math; 35S, 28C, 32Sii, WP34S, EL-506P, EL-W506, FX750P; owned FX702P & 11C; used 67 & 85. iOS: 42s (Byron), Free42, WP31S/34S, HCalc.
43S operators right. DM42 sn. 03818.

rprosperi
Posts: 479
Joined: Mon Apr 24, 2017 5:48 pm
Location: New York

Re: 43S Alternative key layout --> WP43C

Post by rprosperi » Sun Aug 11, 2019 2:06 am

Jaymos wrote:
Sun Aug 11, 2019 1:26 am
I tried to make a proper emulator layout without the confusing colours:

See http://cocoon-creations.com/download/IMG_0392.JPG
As the old adage says, a picture really is worth a thousand words. Thanks for sharing.

This image is nice and should clear up much confusion for the TL;DR folks, myself among them. :)
--bob p

DM42: β00071 & 00282

Dani R.
Posts: 130
Joined: Fri May 05, 2017 8:23 pm

Re: 43S Alternative key layout --> WP43C

Post by Dani R. » Sun Aug 11, 2019 8:00 am

Please excuse my ignorance, Jaco, I have actually not read your current contributions, so I don't know your reasons. I let the triple picture work.


Layout 1: I suspect that '#' and BASE are not the same. #' works for me as a function like RCL. So you have a value and specify the new base in a dialog. I don't know such a functionality, I had always worked with the 'Y' and 'X' registers for such functionalities. Note: Except for STO and RCL, I can't think of any function that requires parameters as further input. With RPL they have removed this exception behavior, I think. FIX, ENG etc. are not functions for me, but configuration dialogs. In my personal opinion, I would have liked to avoid functions that require parameters that are not passed via the stack. Long speech, short sense. If you have '#' you can keep '# as functionality if it already existed on the WP34s. Otherwise I would have liked to have an additional BASE menu in the number pad in layout 1. Thomas has, I think, reconstructed the functionality of the 16C in the current Free42 pretty completely. But it might also be possible that I'm not concentrated and you find functions of BASE and/or 16C in another menu.

Abbreviated: If '#' is a menu it doesn't belong in the second row, if BASE is missing you have to make room. Possible movement; BASE on [f] [4] and ALPHA on [Sigma+] and send a red label together with the overlays, known procedure...

I had already mentioned [3], [-], [R/S] and [+] movements.


Layout 2: If this variant is a temporary test variant, you can also imagine omitting RTN in it. This means; one function to [g][TAN]. I have discussed the topic '#' above.

Leave HOME at [g] [7], currently leave ALPHA.FN at [g][4], move VIEW to [f][0], would do well there next to SHOW.






Afterwards you can actually consider to transfer EQN, ADV, timer movements to layout 1. Could work. Then the practical differences between layout 1 and 2 are only COMPLEX and ALPHA.


Dani


EDIT:
What is currently not finding which function is behind the function on the [g][STO] position in layout 1? Is this a candidate to be omitted or to be replaced with x! ?
DM42 SN:00032

Dani R.
Posts: 130
Joined: Fri May 05, 2017 8:23 pm

Re: 43S Alternative key layout

Post by Dani R. » Sun Aug 11, 2019 9:10 am

Jaymos wrote:
Sun Aug 11, 2019 1:49 am

Regarding Dani's request to re-arrange the alpha keys - it is complete now.

You guys can test the alpha mode on the DM42 image now.
;)
DM42 SN:00032

User avatar
Jaymos
Posts: 203
Joined: Sun Nov 04, 2018 6:03 pm
Location: Cape Town

Re: 43S Alternative key layout --> WP43C

Post by Jaymos » Sun Aug 11, 2019 9:53 am

Dani R. wrote:
Sun Aug 11, 2019 8:00 am
Please excuse my ignorance, Jaco, I have actually not read your current contributions, so I don't know your reasons. I let the triple picture work.
Have a look. I ask the question if Layout1 can stay.

Layout 1: I suspect that '#' and BASE are not the same. #' works for me as a function like RCL. So you have a value and specify the new base in a dialog. I don't know such a functionality, I had always worked with the 'Y' and 'X' registers for such functionalities.
# is a direct input screen like RCL.
BASE is like HP42S BASE and it brings up a menu. BASE includes a sift FN6 for #.
Neither use the stack.
Note: Except for STO and RCL, I can't think of any function that requires parameters as further input.
They use the same direct input screen called TAM, for. RCL & STO, ENG, SCI, etc. and ?????
With RPL they have removed this exception behavior, I think. FIX, ENG etc. are not functions for me, but configuration dialogs. In my personal opinion, I would have liked to avoid functions that require parameters that are not passed via the stack.
I don’t want to change this. This is inherited from WP43S.
Long speech, short sense. If you have '#' you can keep '# as functionality if it already existed on the WP34s. Otherwise I would have liked to have an additional BASE menu in the number pad in layout 1.
# is a direct input, I assume you meant WP43S.
BASE is a menu. And I cannot find a spot in the number pad for it. That is why I don’t like layout 1 anymore. No space for BASE.
Thomas has, I think, reconstructed the functionality of the 16C in the current Free42 pretty completely. But it might also be possible that I'm not concentrated and you find functions of BASE and/or 16C in another menu.
The other bit commands are in other menus. Which????. I don’t think the user interface at the moment can compete with 16C. It needs more work.
Abbreviated: If '#' is a menu it doesn't belong in the second row, if BASE is missing you have to make room.
# belonged there and I can leave # there. I did not destroy #, I added a BASE menu. Maybe there must be a FN to call BASE from #. I like that.
Possible movement; BASE on [f] [4] and ALPHA on [Sigma+] and send a red label together with the overlays, known procedure...
No :-)
I am reserving Sigma+ to have a default USER key config to call MyMenu. You can change USER sigma+ to ALPHA
I had already mentioned [3], [-], [R/S] and [+] movements.
I like them. I will make them if Layout1 stays.

Layout 2: If this variant is a temporary test variant, you can also imagine omitting RTN in it. This means; one function to [g][TAN].
I see Layout2 being very useful and not only temporary.
Do we need Layout1 still?
I have discussed the topic '#' above.
BASE works perfectly here.
Leave HOME at [g] [7], currently leave ALPHA.FN at [g][4], move VIEW to [f][0], would do well there next to SHOW.
To confirm: On Layout2,
move HOME to g[7],
move a.FN to g[4],
move VIEW to g[.],
move TEST to f[0].

That looks very good and puts HOME where I like it. I will do it.

Afterwards you can actually consider to transfer EQN, ADV, timer movements to layout 1. Could work. Then the practical differences between layout 1 and 2 are only COMPLEX and ALPHA.
This brings me to the same question, ie does Layout 1 live or not.

EDIT:
What is currently not finding which function is behind the function on the [g][STO] position in layout 1? Is this a candidate to be omitted or to be replaced with x! ?
Layout1 g[STO] is the complex angle function. Very important to have on direct key.
Jaco Mostert
Elec Eng, South Africa
WP34C, HP42S, DM42 for complex math; 35S, 28C, 32Sii, WP34S, EL-506P, EL-W506, FX750P; owned FX702P & 11C; used 67 & 85. iOS: 42s (Byron), Free42, WP31S/34S, HCalc.
43S operators right. DM42 sn. 03818.

User avatar
Jaymos
Posts: 203
Joined: Sun Nov 04, 2018 6:03 pm
Location: Cape Town

Re: 43S Alternative key layout --> WP43C

Post by Jaymos » Sun Aug 11, 2019 12:04 pm

Jaymos wrote:
Sun Aug 11, 2019 9:53 am
Dani R. wrote:
Sun Aug 11, 2019 8:00 am
Layout 1: I suspect that '#' and BASE are not the same. #' works for me as a function like RCL. So you have a value and specify the new base in a dialog. I don't know such a functionality, I had always worked with the 'Y' and 'X' registers for such functionalities.
# is a direct input screen like RCL.
BASE is like HP42S BASE and it brings up a menu. BASE includes a sift FN6 for #.
Neither use the stack.
Note: Except for STO and RCL, I can't think of any function that requires parameters as further input.
They use the same direct input screen called TAM, for. RCL & STO, ENG, SCI, etc. and ?????
All the following functions use the same direct input method like RCL. The confusing part is that the same menu appears when you press any of these functions. Better will be relevant soften point per function, or no soft menu functions because it is confusing.

RCL input screen:
Clipboard01.png
Clipboard01.png (6.33 KiB) Viewed 140 times


All other input screens:
Clipboard02.png
Clipboard02.png (2.13 KiB) Viewed 140 times

RCL, STO
FIX, SCI, ENG, ALL,
DISP/Dn/GAP (grouping of numbers (i.e. groups 3, 4 ... [100 000 000, or 1 0000 0000])
Base change (2-16)
DISP/Dn/DSTACK (display number of stack items 1-4)
Word size input (bit length 2-64)

ITM_DSP I am not sure what this does. I think deleted in main project. See OM v0.12 p312.

Long speech, short sense. If you have '#' you can keep '# as functionality if it already existed on the WP34s. Otherwise I would have liked to have an additional BASE menu in the number pad in layout 1.
# is a direct input, I assume you meant WP43S.
BASE is a menu. And I cannot find a spot in the number pad for it. That is why I don’t like layout 1 anymore. No space for BASE.
Thomas has, I think, reconstructed the functionality of the 16C in the current Free42 pretty completely. But it might also be possible that I'm not concentrated and you find functions of BASE and/or 16C in another menu.
The other bit commands are in other menus. Which????. I don’t think the user interface at the moment can compete with 16C. It needs more work.
The other bit commands are in BITS, in three menu screens.Main screen below, shifts in next screen.
Clipboard07.png
Clipboard07.png (2.77 KiB) Viewed 140 times
Jaco Mostert
Elec Eng, South Africa
WP34C, HP42S, DM42 for complex math; 35S, 28C, 32Sii, WP34S, EL-506P, EL-W506, FX750P; owned FX702P & 11C; used 67 & 85. iOS: 42s (Byron), Free42, WP31S/34S, HCalc.
43S operators right. DM42 sn. 03818.

User avatar
Jaymos
Posts: 203
Joined: Sun Nov 04, 2018 6:03 pm
Location: Cape Town

Re: 43S Alternative key layout --> WP43C

Post by Jaymos » Sun Aug 11, 2019 12:14 pm

Jaymos wrote:
Sun Aug 11, 2019 9:53 am
Dani R. wrote:
Sun Aug 11, 2019 8:00 am
Leave HOME at [g] [7], currently leave ALPHA.FN at [g][4], move VIEW to [f][0], would do well there next to SHOW.
To confirm: On Layout2,
move HOME to g[7],
move a.FN to g[4],
move VIEW to g[.],
move TEST to f[0].

That looks very good and puts HOME where I like it. I will do it.
Dani,

I looked at this.

How about the following :

move HOME to g[7], as you suggested,
move a.FN to g[4], as you suggested,
move VIEW to g[.], as you suggested,

but

move TEST to g[-]
move TIMER to g[0].

If we do this I think that the Layout2 is becoming good enough to replace Layout1.

What do you think?
Jaco Mostert
Elec Eng, South Africa
WP34C, HP42S, DM42 for complex math; 35S, 28C, 32Sii, WP34S, EL-506P, EL-W506, FX750P; owned FX702P & 11C; used 67 & 85. iOS: 42s (Byron), Free42, WP31S/34S, HCalc.
43S operators right. DM42 sn. 03818.

User avatar
Jaymos
Posts: 203
Joined: Sun Nov 04, 2018 6:03 pm
Location: Cape Town

Re: 43S Alternative key layout --> WP43C

Post by Jaymos » Sun Aug 11, 2019 2:18 pm

Jaymos wrote:
Sun Aug 11, 2019 12:14 pm
Jaymos wrote:
Sun Aug 11, 2019 9:53 am
Dani R. wrote:
Sun Aug 11, 2019 8:00 am
Leave HOME at [g] [7], currently leave ALPHA.FN at [g][4], move VIEW to [f][0], would do well there next to SHOW.
To confirm: On Layout2,
move HOME to g[7],
move a.FN to g[4],
move VIEW to g[.],
move TEST to f[0].

That looks very good and puts HOME where I like it. I will do it.
Dani,

I looked at this.

How about the following :

move HOME to g[7], as you suggested,
move a.FN to g[4], as you suggested,
move VIEW to g[.], as you suggested,

but

move TEST to g[-]
move TIMER to g[0].

If we do this I think that the Layout2 is becoming good enough to replace Layout1.

What do you think?

I made the above changes,
IMG_0394c.png
IMG_0394c.png (53.85 KiB) Viewed 130 times

The only problem now is that HOME is not visible to the DM42 user with NO blue labels.
The default is triple shift brings HOME, but some people might not know about this shortcut.

so, this solves that problem:

HOME to f[0]
TIMER to g[0]
INFO to g[7]

Ideas?
Jaco Mostert
Elec Eng, South Africa
WP34C, HP42S, DM42 for complex math; 35S, 28C, 32Sii, WP34S, EL-506P, EL-W506, FX750P; owned FX702P & 11C; used 67 & 85. iOS: 42s (Byron), Free42, WP31S/34S, HCalc.
43S operators right. DM42 sn. 03818.

Dani R.
Posts: 130
Joined: Fri May 05, 2017 8:23 pm

Re: 43S Alternative key layout --> WP43C

Post by Dani R. » Sun Aug 11, 2019 7:25 pm

I have understood, in the long run you probably won't want to maintain two layouts, not even with #ifdef. I really appreciate that layout 1 and layout 2 are still maintained at the moment.

Let's focus on layout 2 and face the facts. For layout 2 to work, lines two and three must work again.
Jaymos wrote:
Sun Aug 11, 2019 9:53 am
Dani R. wrote:
Sun Aug 11, 2019 8:00 am
EDIT:
What is currently not finding which function is behind the function on the [g][STO] position in layout 1? Is this a candidate to be omitted or to be replaced with x! ?
Layout1 g[STO] is the complex angle function. Very important to have on direct key.
This leads me to believe that this function should be left on [g][STO], right next to [COMPLEX].

I also think that you should leave '#' at [g][1/x]. Maybe you'll notice later that one of the superimposed functionalities of '#' is worth having on a direct function key and you're glad to have kept it there.
Jaymos wrote:
Sun Aug 11, 2019 12:04 pm
RCL, STO
FIX, SCI, ENG, ALL,
DISP/Dn/GAP (grouping of numbers (i.e. groups 3, 4 ... [100 000 000, or 1 0000 0000])
Base change (2-16)
DISP/Dn/DSTACK (display number of stack items 1-4)
Word size input (bit length 2-64)
As assumed, except STO and RCL, these are all configuration functions, if I am not mistaken, which have no influence on the content of the stack, but at most on the display. But I won't stay here any longer, if Walter decided to use the TAM in the '#'-function and this can be in the tradition of HP, this is not a real issue for me. Personally I still can't like it.

So we lost the position for ABS. ABS is an extremely important function. ABS belongs to a prominent position. It is also used, for example, for the matrix. I actually only see [g][RCL] or [g][RollDown] as possible positions. I will never use DELTA% and you will surely find in the FIN menu. RollUP requires [SHIFT][SHIFT][RollDown]. I had the problem solved with [RollDown] [RollDown] [RollDown]. They are the same number of keystrokes. Well, if you have switched to the eighth stack, you need [STK].

Yes, a good solution for ABS might possibly make layout 1 obsolete in the medium term.
Jaymos wrote:
Sun Aug 11, 2019 9:53 am
Dani R. wrote:
Sun Aug 11, 2019 8:00 am
Thomas has, I think, reconstructed the functionality of the 16C in the current Free42 pretty completely. But it might also be possible that I'm not concentrated and you find functions of BASE and/or 16C in another menu.
The other bit commands are in other menus. Which????. I don’t think the user interface at the moment can compete with 16C. It needs more work.
I also think that it's worth having the BASE menu and I suspect that there's still some work to be done here, that you'll find the functionality of the 16C united here. But it's possible that most of the functionality is there. Simply not in the new BASE menu.
Jaymos wrote:
Sun Aug 11, 2019 12:14 pm
How about the following :

move HOME to g[7], as you suggested,
move a.FN to g[4], as you suggested,
move VIEW to g[.], as you suggested,

but

move TEST to g[-]
move TIMER to g[0].
I had suggested taking VIEW on [f][0], but on [g][.] is also interesting.
Jaymos wrote:
Sun Aug 11, 2019 2:18 pm
so, this solves that problem:

HOME to f[0]
TIMER to g[0]
INFO to g[7]
I just don't like that there except for [f]CLR], [g]UNDO on [CLR] another exception exists with menu/function. But since HOME is more of a super function and the whole thing happens in the last line, I can probably turn a blind eye.

Yes, it's worth another mock up.

Maybe you could swap ALPHA.FN and CLK, then CLK would be closer to TIMER again. That's still what I don't like so much that TIMER is far away from CLK. But this is not a real issue.
DM42 SN:00032

User avatar
Jaymos
Posts: 203
Joined: Sun Nov 04, 2018 6:03 pm
Location: Cape Town

Re: 43S Alternative key layout --> WP43C

Post by Jaymos » Mon Aug 12, 2019 12:30 am

Dani R. wrote:
Sun Aug 11, 2019 7:25 pm
I have understood, in the long run you probably won't want to maintain two layouts, not even with #ifdef. I really appreciate that layout 1 and layout 2 are still maintained at the moment.
It is possible to keep two alive. But it makes it difficult especially with unknown changes coming from the main project. And the main question is why two. If we can make ONE compromise solution, then it will be better.

Let's focus on layout 2 and face the facts. For layout 2 to work, lines two and three must work again.
This leads me to believe that this function (complex angle) should be left on [g][STO], right next to [COMPLEX].
Placing it by pi makes sense in that both represent angles.
When dealing with complex numbers, both ABS value and angle must be there. Both are important to be in the top section. In my opinion the complex number is given as ABS.angle, therefore ABS must be on g[ENTER] and angle on its right, not the other way round.

I also think that you should leave '#' at [g][1/x]. Maybe you'll notice later that one of the superimposed functionalities of '#' is worth having on a direct function key and you're glad to have kept it there.
For my wishes, |x| and angle are far more important on the top line than #. But # must be there.
BASE on [f][4] is optional and nice and can be used for conversions only. Number entry is on a key, i.e.

Entering 3A00 Hex using # in top line:
[f][f][3] 3A00 [f][f][1/x] [RCL] [ENTER]


So we lost the position for ABS. ABS is an extremely important function. ABS belongs to a prominent position. It is also used, for example, for the matrix. I actually only see [g][RCL] or [g][RollDown] as possible positions. I will never use DELTA% and you will surely find in the FIN menu. RollUP requires [SHIFT][SHIFT][RollDown]. I had the problem solved with [RollDown] [RollDown] [RollDown]. They are the same number of keystrokes. Well, if you have switched to the eighth stack, you need [STK].
I like your questioning of the need for RollUp.

RollUp can be useful to push Y and Z out of the way, but so could ENTER do, if just for pushing up. RollUp will also rotate the stack, top down to X. For a 4 level stack, 3 x RollDown is easier than [f][f][Rdn]. I do not think for an 8 element stack somebody wants to roll around 8 elements? Does one really keep track of the 8th element you want in X? I think roll around from D to X is not needed on the key but a menu is ok.

So in summary I agree Rollup can be placed in the STK menu for programming. It was not in the STK menu, but I added it there any way.

I changed it in the mockup for now. Please consider the effect.
I also think that it's worth having the BASE menu and I suspect that there's still some work to be done here, that you'll find the functionality of the 16C united here. But it's possible that most of the functionality is there. Simply not in the new BASE menu.
I placed the # back in the top line, and kept the BASE in the mockup. I added BITS and INTS submenus in the BASE menu.
We can decide if we really want BASE or not.
Jaymos wrote:
Sun Aug 11, 2019 12:14 pm
move HOME to g[7], as you suggested,
move a.FN to g[4], as you suggested,
move VIEW to g[.], as you suggested, (sorry misunderstood, but ok)
move TEST to g[-]
move TIMER to g[0].
HOME to f[0]
TIMER to g[0]
INFO to g[7]
I just don't like that there except for [f]CLR], [g]UNDO on [CLR] another exception exists with menu/function. But since HOME is more of a super function and the whole thing happens in the last line, I can probably turn a blind eye.
In the bottom and left area there are a few blue items which are not menus. That is no problem. Starting from DROP, REGS, FLAG.V, SAVE, TIMER, VIEW, PRINTX. I suspect you mean the yellow ones that are not.

Maybe you could swap ALPHA.FN and CLK, then CLK would be closer to TIMER again. That's still what I don't like so much that TIMER is far away from CLK. But this is not a real issue.
a.FN and CLK both are in the lower half and the exact position does not matter. So I change it in the mockup. I don't mind either way. CLK has no real connection to TIMER anyway except they are working with time functions. CLK has date conversions, and on the next page clock setup. TIMER works in seconds and does timing.

Image
Jaco Mostert
Elec Eng, South Africa
WP34C, HP42S, DM42 for complex math; 35S, 28C, 32Sii, WP34S, EL-506P, EL-W506, FX750P; owned FX702P & 11C; used 67 & 85. iOS: 42s (Byron), Free42, WP31S/34S, HCalc.
43S operators right. DM42 sn. 03818.

Post Reply