WP43 Alternative key layout --> C43

This area is for discussion about these families of custom high-end Scientific Calculator applications for SwissMicros devices.
User avatar
Jaymos
Posts: 1635
Joined: Sun Nov 04, 2018 7:03 pm
Location: Cape Town

Re: 43S Alternative key layout

Post by Jaymos »

Dani R. wrote:
Tue Jul 23, 2019 9:00 pm
Implementing a user interface is always difficult because everyone has their own opinion about it. So here are a few suggestions from me regarding a possible successor of the WP43S.
Correction, I am not making a successor for the WP43S. In fact I add no real value - the dev team does that and the math and the OS and the error checking and the structures and the clever coding. I fiddle around to hack a keyboard I like, to slap onto their code. I am not making a new calc.
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm
First suggestion. Please, please bring the alpha key unshifted, as most HP calculators with alpha input have it.
I really can't - the DM42 keys are fixed. This may be possible on the real 43S though if you convince Walter on the "43S NEWS" forum.
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm
With the introduction of the menus, topics are summarized, which also creates space on the keyboard. The WP43S for example has a menu for trigonometry, why not for logarithm. Possibly you get all functionality packed into 30 menus.
There is a menu with some logs, see EXP.
But this goes against what I want in a calculator. I have a 28C (from those days) and I don't like it because I have to dig into the menus for the most basic functions. I want a compromise calc, and that is what the 43S brings - and I compromise it further by making it 42S-like.
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm
Second suggestion. Please try to get out with only one shift key. There are several HP calculators with only one shift key.
The DM42 hardware has one only, and my goal is to not change that.
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm


Third suggestion. If it is not possible to get by with a single shift key, then follow Günter's suggestion above.
His idea of indicating the row is implemented now.
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm
Menu CNST. I can imagine that one assigns oneself more frequently used constants to the function keys F1..F5 and SHIFT-F1--SHIFT-F5. For example, F6 would be NXT to get to the next ten places, alternatively to the cursor keys. SHIFT-F6 could be used, for example, to display the constants as a vertical list at least in the CNST menu, i.e. actually a classic catalog functionality. A disadvantage is that you would only see eight entries at once. An advantage is that you can write the description in a second column in addition to the constant symbols. If a list is opened, the assignments of the function keys with functions like assign to a free position, delete a position, define a user-defined constant etc. change of course.
The menu structure and the fact that you see 6x3=18 slots at a time, and can use arrow down to access more levels is done and working in the main 43S project. I accept that and won't change it.
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm


Here is a possible keyboard layout with a lot of free space for menus.
The problem I see with this menu is that you sacrificed the arrow keys of the DM432 hardware. No go.

My effort is re-designing the 43S keyboard that Walter and the team developed to work on the DM42 hardware without changing the actual buttons.
Jaco Mostert
Elec Eng, South Africa
https://47calc.com C47 (s/n 03818 & 06199), WP43 (0015). In box: HP42S, HP32Sii, WP34S&C, HP28C, HP35s, EL-506P, EL-W506, PB700; ex: FX702P, 11C, HP67 & HP85; iOS: 42s Byron, Free42+, WP31S/34S, HCalc.
Dani R.
Posts: 349
Joined: Fri May 05, 2017 10:23 pm

Re: 43S Alternative key layout

Post by Dani R. »

Guenter wrote:
Mon Jul 22, 2019 10:04 pm
<More thoughts>. Would it be possible, feasible to give all items that call a menu the blue color, and all the items that call a function directly the yellow or gold color? This would enable the omission of the underlining . Thus reducing the clutter on the keyboard significantly...</More thoughts>

Günter
Jaymos wrote:
Tue Jul 23, 2019 10:12 pm

...
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm


Third suggestion. If it is not possible to get by with a single shift key, then follow Günter's suggestion above.
His idea of indicating the row is implemented now.

...
Dani R. wrote:
Tue Jul 23, 2019 9:00 pm


Here is a possible keyboard layout with a lot of free space for menus.
The problem I see with this menu is that you sacrificed the arrow keys of the DM432 hardware. No go.

My effort is re-designing the 43S keyboard that Walter and the team developed to work on the DM42 hardware without changing the actual buttons.
Hello Jaco

I've abused your thread to keep my thoughts somewhere. I am aware that practically none of the ideas are taken up immediately. But hey, maybe someone who wants to create a completely new design is reading it towards the HP-48.

Your efforts to make the layout compatible with the DM42 are very welcome. I appreciate that. Even though it's hard work, Günter's contribution seems worth considering what's under (f) and what's under (g). Putting all functions under (f) and all menus under (g), or vice versa. This additional effort to clean up the layout would, in my opinion, greatly improve usability.

Regards Dani
C47(DM42) SN:00032 WP43 SN:00016
https://47calc.com
User avatar
Jaymos
Posts: 1635
Joined: Sun Nov 04, 2018 7:03 pm
Location: Cape Town

Re: 43S Alternative key layout

Post by Jaymos »

Dani R. wrote:
Tue Jul 23, 2019 10:45 pm

...Günter's contribution seems worth considering what's under (f) and what's under (g). Putting all functions under (f) and all menus under (g), or vice versa. This additional effort to clean up the layout would, in my opinion, greatly improve usability.
I reshuffled the listed keys that I mentioned in the first post of this thread, and I considered this comment. There are of course exceptions, but I succeeded now to make the first 5 rows (including F keys) such that ALL yellow's are non-menus, and 7 blue labels are menus. Rows 6-8 have 10/14 yellow labels being menus, and ALL blue labels menus.

J
Jaco Mostert
Elec Eng, South Africa
https://47calc.com C47 (s/n 03818 & 06199), WP43 (0015). In box: HP42S, HP32Sii, WP34S&C, HP28C, HP35s, EL-506P, EL-W506, PB700; ex: FX702P, 11C, HP67 & HP85; iOS: 42s Byron, Free42+, WP31S/34S, HCalc.
User avatar
Jaymos
Posts: 1635
Joined: Sun Nov 04, 2018 7:03 pm
Location: Cape Town

Re: 43S Alternative key layout

Post by Jaymos »

H2X wrote:
Mon Jul 22, 2019 12:58 pm
I just wanted to let you know that if ASN stands for "assign", I think it sits well on [STO] for the same reason as SAVE.
I considered it, but found that even though ASN is saving a key assignment, it has little to do with math, is much lower down on the 42S (where the label space is already FULL of menu items), and I needed the real estate for math, so I moved it a bit lower to CHS.
H2X wrote:
Mon Jul 22, 2019 12:25 pm
Regarding the [STO] key, I do feel that this is the prime candidate for the SAVE, but mostly because of the semantic connection between "storing" and "saving" something. It just feels like the natural place to start looking for SAVE, and it would feel good to find it there.
I moved EXP out to [g] STO, because I wanted the 42S compatible % on RCL. I moved SAVE to [f] STO as you rightly said it sort of belongs there. I also shuffled FIN to CHS to make space for VIEW next to DISP on EEX.
H2X wrote:
Mon Jul 22, 2019 12:25 pm
I also like the pattern which is beginning to emerge that the (f) shifted labels are yellow (like on the DM42 and almost like the HP42S). Finding (f) shifted labels at the familiar places and in the familiar colour must be a good thing.
I swapped the [f] and [g] keys of [5] and [8], and replaced [9] with my favourite |x| and angle.

So, now there is a bit more structure, i.e. the first 5 rows are clearly [f] are commands, and most [g] are menus. And the lower 3 rows are a mix match of menu stuff. Also, the upper part is for direct access math and close to HP42S, middle area is a bit hybrid, lower area is full with new menus.
H2X wrote:
Mon Jul 22, 2019 12:25 pm
Coincidentally, the (f) shifted options are also more accessible in terms of key, suggesting that (g) shifted options be reserved for lesser used functions - unless there is strong affinity between (f) and (g) shifted options (such as % and d%, R< and >P, etc.) which makes it logical to find them om the same key.
I tried to improve this even more.

Image

The latest DM42 compatible layout of today.
Note the dots both sides of the menu "f"-line. I could not use an arrow or blocks or lines, as the area is very restricted there.
Also note the current EXP menus on the display.

Questions for comment:

There are cases to be made for the following:
1. Should COMPLEX move to [f] STO, for another key more compatible with HP42S? Bear in mind the COMPLEX functionality offered by the main 43S project is that of "CC"and a little different than HP42S "COMPLEX" even though it achieves the same thing.

2. If COMPLEX moves to STO, should ALPHA (currently [f] [5]) move to [f] ENTER? Advantage to be a bit more like HP42S, and disadvantage is that that ALPHA function is now removed from the ALPHA menu on [g] [5] (I am not negotiable on ROLLdn on [g] ENTER). {edit: I meant DROP, not ROLLdn}.

3. If both (1) and (2), then SAVE goes down to [f] [5] where it is far from STO and a bit in the cold.

4. A different question: Should the shifted labels on [8] and [9] swap with [CHS] and [EEX}? More math to the top and admin to the middle rows.


I would like to add the colour arrows to the emulator image, above the shift key ↰f g↱, but that would need some research. I also want to indicate "lx"above "L" in white, to indicate RCL L is last x which also needs research.

Comments or suggestions?
J
Last edited by Jaymos on Wed Jul 24, 2019 8:17 am, edited 2 times in total.
Jaco Mostert
Elec Eng, South Africa
https://47calc.com C47 (s/n 03818 & 06199), WP43 (0015). In box: HP42S, HP32Sii, WP34S&C, HP28C, HP35s, EL-506P, EL-W506, PB700; ex: FX702P, 11C, HP67 & HP85; iOS: 42s Byron, Free42+, WP31S/34S, HCalc.
Dani R.
Posts: 349
Joined: Fri May 05, 2017 10:23 pm

Re: 43S Alternative key layout

Post by Dani R. »

You could put Alpha on Sigma+. I think Sigma+ can also be found in STAT. This would be the first violation compared to the original DM42 assignment.
My XEQ ALPHA problem could possibly be solved by pressing XEQ twice.
I like to have COMPLEX on ENTER, although I practically don't need complex numbers anymore.

You can set the alphabetical labels back to the original DM42 positions, i.e. K to COS, L to TAN, etc.
C47(DM42) SN:00032 WP43 SN:00016
https://47calc.com
H2X
Posts: 885
Joined: Tue Apr 25, 2017 8:00 am
Location: Norðvegr
Contact:

Re: 43S Alternative key layout

Post by H2X »

Jaymos wrote:
Wed Jul 24, 2019 12:17 am
Questions for comment:

There are cases to be made for the following:
1. Should COMPLEX move to [f] STO, for another key more compatible with HP42S? Bear in mind the COMPLEX functionality offered by the main 43S project is that of "CC"and a little different than HP42S "COMPLEX" even though it achieves the same thing.
2. If COMPLEX moves to STO, should ALPHA (currently [f] [5]) move to [f] ENTER? Advantage to be a bit more like HP42S, and disadvantage is that that ALPHA function is now removed from the ALPHA menu on [g] [5] (I am not negotiable on ROLLdn on [g] ENTER). {edit: I meant DROP, not ROLLdn}.
I think not, and I'd argue along the lines that COMPLEX / CC is about number entry, and thus belongs on the [ENTER] key.

One could counter that entering text also has a semantic connection with [ENTER], and one could debate which connection is stronger - I think the semantic is also different, and that ALPHA is more of a "mode" thing than an "entry" thing.

I also think that ergonomics would be better with COMPLEX on the [ENTER] key. Number entry and text entry are the essential user interactions with the calculator, and of these I think number entry benefits most by having its shifted options, i.e. COMPLEX / CC, on the [ENTER] key.

Actually, I wouldn't mind if the COMPLEX and ALPHA labels were swapped on the HP42S. If I were on that team, I think I'd be having the same discussion with them. I imagine that a significant reason for them to put ALPHA on the [ENTER] key is to emphasize or promote that capability. Today, it doesn't need promoting.

Besides, I like that you have placed &alpha; and &alpha;.FN on the same key,
Jaymos wrote:
Wed Jul 24, 2019 12:17 am
3. If both (1) and (2), then SAVE goes down to [f] [5] where it is far from STO and a bit in the cold.
Neither (1) nor (2), besides I personally prefer SAVE in the warm. :-)
Jaymos wrote:
Wed Jul 24, 2019 12:17 am
4. A different question: Should the shifted labels on [8] and [9] swap with [CHS] and [EEX}? More math to the top and admin to the middle rows.
Fine with me, and I agree with your reasoning. Besides, LAST x and MODES are gone from the [X<>Y] and [+/-] keys (on the HP42S), so there is already a case for DISP not having to be on the [EEX] key.
Jaymos wrote:
Wed Jul 24, 2019 12:17 am
I would like to add the colour arrows to the emulator image, above the shift key ↰f g↱, but that would need some research. I also want to indicate "lx"above "L" in white, to indicate RCL L is last x which also needs research.

Comments or suggestions?
J
I like the visual indication of which soft menu row is active! As an alternative to dots, how would a vertical bar of a few pixels look?
What is the metric tensor in imperial units?
H2X
Posts: 885
Joined: Tue Apr 25, 2017 8:00 am
Location: Norðvegr
Contact:

Re: 43S Alternative key layout

Post by H2X »

@Jaco,

are you planning to implement the screenshot feature of the DM42 (hold SHIFT and press DISP)?

https://www.swissmicros.com/dm42/doc/dm ... creenshots

This doesn't have to be a problem vs. the cycling shift key behaviour you have adopted - holding (long pressing) the shift key need not advance the cycle, allowing it to both be "terminated" by a key press, and aborted by just releasing the shift key - as long as it has been pressed long enough.
What is the metric tensor in imperial units?
User avatar
Jaymos
Posts: 1635
Joined: Sun Nov 04, 2018 7:03 pm
Location: Cape Town

Re: 43S Alternative key layout

Post by Jaymos »

H2X wrote:
Wed Jul 24, 2019 9:59 am
@Jaco,

are you planning to implement the screenshot feature of the DM42 (hold SHIFT and press DISP)?

https://www.swissmicros.com/dm42/doc/dm ... creenshots

This doesn't have to be a problem vs. the cycling shift key behaviour you have adopted - holding (long pressing) the shift key need not advance the cycle, allowing it to both be "terminated" by a key press, and aborted by just releasing the shift key - as long as it has been pressed long enough.
No. At this moment that needs too much OS functionality and depends on software modules I don’t want to touch.
Jaco Mostert
Elec Eng, South Africa
https://47calc.com C47 (s/n 03818 & 06199), WP43 (0015). In box: HP42S, HP32Sii, WP34S&C, HP28C, HP35s, EL-506P, EL-W506, PB700; ex: FX702P, 11C, HP67 & HP85; iOS: 42s Byron, Free42+, WP31S/34S, HCalc.
User avatar
Jaymos
Posts: 1635
Joined: Sun Nov 04, 2018 7:03 pm
Location: Cape Town

Re: 43S Alternative key layout

Post by Jaymos »

H2X wrote:
Wed Jul 24, 2019 9:44 am
Jaymos wrote:
Wed Jul 24, 2019 12:17 am
Questions for comment:

There are cases to be made for the following:
1. Should COMPLEX move to [f] STO, for another key more compatible with HP42S? Bear in mind the COMPLEX functionality offered by the main 43S project is that of "CC"and a little different than HP42S "COMPLEX" even though it achieves the same thing.
2. If COMPLEX moves to STO, should ALPHA (currently [f] [5]) move to [f] ENTER? Advantage to be a bit more like HP42S, and disadvantage is that that ALPHA function is now removed from the ALPHA menu on [g] [5] (I am not negotiable on ROLLdn on [g] ENTER). {edit: I meant DROP, not ROLLdn}.
I think not, and I'd argue along the lines that COMPLEX / CC is about number entry, and thus belongs on the [ENTER] key.

One could counter that entering text also has a semantic connection with [ENTER], and one could debate which connection is stronger - I think the semantic is also different, and that ALPHA is more of a "mode" thing than an "entry" thing.

I also think that ergonomics would be better with COMPLEX on the [ENTER] key. Number entry and text entry are the essential user interactions with the calculator, and of these I think number entry benefits most by having its shifted options, i.e. COMPLEX / CC, on the [ENTER] key.

Actually, I wouldn't mind if the COMPLEX and ALPHA labels were swapped on the HP42S. If I were on that team, I think I'd be having the same discussion with them. I imagine that a significant reason for them to put ALPHA on the [ENTER] key is to emphasize or promote that capability. Today, it doesn't need promoting.

Besides, I like that you have placed &alpha; and &alpha;.FN on the same key,
Jaymos wrote:
Wed Jul 24, 2019 12:17 am
3. If both (1) and (2), then SAVE goes down to [f] [5] where it is far from STO and a bit in the cold.
Neither (1) nor (2), besides I personally prefer SAVE in the warm. :-)
Jaymos wrote:
Wed Jul 24, 2019 12:17 am
4. A different question: Should the shifted labels on [8] and [9] swap with [CHS] and [EEX}? More math to the top and admin to the middle rows.
Fine with me, and I agree with your reasoning. Besides, LAST x and MODES are gone from the [X<>Y] and [+/-] keys (on the HP42S), so there is already a case for DISP not having to be on the [EEX] key.
Jaymos wrote:
Wed Jul 24, 2019 12:17 am
I would like to add the colour arrows to the emulator image, above the shift key ↰f g↱, but that would need some research. I also want to indicate "lx"above "L" in white, to indicate RCL L is last x which also needs research.

Comments or suggestions?
J
I like the visual indication of which soft menu row is active! As an alternative to dots, how would a vertical bar of a few pixels look?

All valuable inputs above, thanx.

I also like the improvement over the HP42S and reasoning re ALPHA. COMPLEX stays on [f] ENTER.

Dots: I cannot make pixel shapes as I can only use characters defined already, not lines, and the dot takes up least space, as it overwrites the actual menu text and lines.
Jaco Mostert
Elec Eng, South Africa
https://47calc.com C47 (s/n 03818 & 06199), WP43 (0015). In box: HP42S, HP32Sii, WP34S&C, HP28C, HP35s, EL-506P, EL-W506, PB700; ex: FX702P, 11C, HP67 & HP85; iOS: 42s Byron, Free42+, WP31S/34S, HCalc.
User avatar
Jaymos
Posts: 1635
Joined: Sun Nov 04, 2018 7:03 pm
Location: Cape Town

Re: 43S Alternative key layout

Post by Jaymos »

Dani R. wrote:
Wed Jul 24, 2019 7:55 am
You could put Alpha on Sigma+. I think Sigma+ can also be found in STAT. This would be the first violation compared to the original DM42 assignment.
My XEQ ALPHA problem could possibly be solved by pressing XEQ twice.
I like to have COMPLEX on ENTER, although I practically don't need complex numbers anymore.

You can set the alphabetical labels back to the original DM42 positions, i.e. K to COS, L to TAN, etc.

XEQ is not yet functional in the main project and therefore not in the emulator, so we cannot determine how the alpha exactly would work. Let’s wait until it can be tested.

I quite like the XEQ XEQ.

I do dislike sigma+ top left, so maybe...
either way it cannot get another label on the DM42 hardware.
Jaco Mostert
Elec Eng, South Africa
https://47calc.com C47 (s/n 03818 & 06199), WP43 (0015). In box: HP42S, HP32Sii, WP34S&C, HP28C, HP35s, EL-506P, EL-W506, PB700; ex: FX702P, 11C, HP67 & HP85; iOS: 42s Byron, Free42+, WP31S/34S, HCalc.
Post Reply