WP43 News

This area is for discussion about these families of custom high-end Scientific Calculator applications for SwissMicros devices.
User avatar
GeisiX
Posts: 14
Joined: Thu Mar 18, 2021 11:47 pm
Location: Bavaria

Re: FIX / digits displayed

Post by GeisiX »

Walter wrote:
Mon Nov 07, 2022 10:48 pm
Natural numbers (i.e. integers) even don't receive a decimal point so you observe a feature, not an error here.
The question remains what the purpose of this feature is. It certainly does not contribute to redability when integers and reals are on the stack at the same time. I admit that this may be down to personal preference, though.
Walter wrote:
Mon Nov 07, 2022 10:48 pm
Things are slightly more complex: if you want to have the decimal point "glued" to one position, you shall know the maximum exponent you want to be displayed. On the WP43, 4-digit exponents are legal (and may be negative) - so if you allow them and "glue" the point you'd loose quite some space for smaller numbers.
This, indeed, is a valid point. However, the "lost" display space may be covered with the ever present SHOW-option.
User avatar
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: FIX / digits displayed

Post by Walter »

GeisiX wrote:
Mon Nov 07, 2022 11:13 pm
Walter wrote:
Mon Nov 07, 2022 10:48 pm
Natural numbers (i.e. integers) even don't receive a decimal point so you observe a feature, not an error here.
The question remains what the purpose of this feature is. It certainly does not contribute to redability when integers and reals are on the stack at the same time. I admit that this may be down to personal preference, though.
Depends. This way, integers and reals are clearly differentiable. YMMV (and if too many display rows confuse you, feel free to set DSTACK 2 or even 1). 8-)
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
redglyph
Posts: 177
Joined: Sat Dec 22, 2018 11:45 am

Re: FIX / digits displayed

Post by redglyph »

GeisiX wrote:
Mon Nov 07, 2022 10:27 pm
However, natural numbers (full numbers) don't receive the added decimal zeroes. (I corrected my post above and it should now be correct).
I must admit I was surprised too when I first saw this, and I share the same feeling about the displayed values. I suppose it makes the input of integer numbers simpler since you don't need any prefix. But unless a base is attached (to the right), it looks weird in FIX and ALL modes - perhaps attaching a base to all integers would make it clearer and more logical. I think it just takes some getting used to.

You can end each number with a '.' before ENTER so that it's pushed as a decimal number and not an integer.
User avatar
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: FIX / digits displayed

Post by Walter »

redglyph wrote:
Tue Nov 08, 2022 10:16 am
GeisiX wrote:
Mon Nov 07, 2022 10:27 pm
However, natural numbers (full numbers) don't receive the added decimal zeroes. (I corrected my post above and it should now be correct).
I must admit I was surprised too when I first saw this, and I share the same feeling about the displayed values. I suppose it makes the input of integer numbers simpler since you don't need any prefix. But unless a base is attached (to the right), it looks weird in FIX and ALL modes - perhaps attaching a base to all integers would make it clearer and more logical. I think it just takes some getting used to.

You can end each number with a '.' before ENTER so that it's pushed as a decimal number and not an integer.
Yes, you can in principle but we created different data types on purpose. Infinite integers (as we call them), in particular, have their own merits - please look up the OM. If you don't appreciate them, however, feel free to leave them aside; nobody is forced to use data types on the WP43 he doesn't know or need. Maybe later ...
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
User avatar
GeisiX
Posts: 14
Joined: Thu Mar 18, 2021 11:47 pm
Location: Bavaria

Re: FIX / digits displayed

Post by GeisiX »

Thank you for your explanations, Walther, and thank you redglyph for chiming in. I totally understand the reasoning for making the decimal separator the "flag" for determining a number's nature. Adding this information by adding a base / other marker would be quite cumbersome when dealing with integer-only functionalities.

However, coming back to my initial thought on the screen notation, a possible way of getting number display in a more readable order may be:

  • 1) Keep the decimal separator key as the integer / real marker during input as it is currently implemented.
  • 2) FIX (and, depending on preference, SCI and ENG) define the position of the decimal separator on screen (aka "glued to the screen").
  • 3a) Display REALs as advertised / implemented. The same goes for all other data types aside from Integers (e. g. COMPLEX etc.).
  • 3b) Display INTEGERs in the same format as implemented now, i. e. without the decimal separator displayed, but shift the number to the left by n places to match the display settings defined in FIX n.

This way, the entries on the four stack levels displayed would line up and would be more accessible at a quick glance, while maintaining the easy input and discernibility of reals and integers.

I very much suspect that users working extensively with integers would not use display modes with large n (i. e. many decimal places) anyways, so the loss of space when leaving decimal places blank might be acceptable.

Nevertheless, judging from the rather sparse reactions from other fellow members of this forum, I suspect, mildly put, that this issue is not on the top of the priority list for many (and so it surely shouldn't be for you either, Walther, and the great team behind the project). Nothing one cannot get used to.
redglyph
Posts: 177
Joined: Sat Dec 22, 2018 11:45 am

Re: FIX / digits displayed

Post by redglyph »

Walter wrote:
Tue Nov 08, 2022 10:42 am
Yes, you can in principle but we created different data types on purpose. Infinite integers (as we call them), in particular, have their own merits - please look up the OM. If you don't appreciate them, however, feel free to leave them aside; nobody is forced to use data types on the WP43 he doesn't know or need. Maybe later ...
I know, and it's well done. I understand it's mainly the presentation that disturbed GeisiX, and me (at least in the beginning).
GeisiX wrote:
Wed Nov 09, 2022 11:23 am
Adding this information by adding a base / other marker would be quite cumbersome when dealing with integer-only functionalities.
Maybe. I suggested that because it's how the integers are displayed when the base is explicitly given (see screenshot). But at that point they are not infinite anymore I think, so it could also be misleading. I don't have the documentation right now, I may be wrong.
20221109-12591200.png
20221109-12591200.png (1.35 KiB) Viewed 1588 times
GeisiX wrote:
Wed Nov 09, 2022 11:23 am
Nevertheless, judging from the rather sparse reactions from other fellow members of this forum, I suspect, mildly put, that this issue is not on the top of the priority list for many (and so it surely shouldn't be for you either, Walther, and the great team behind the project). Nothing one cannot get used to.
Maybe they are more interested by the latest news about the new platform for the DM32 and WP43 at the moment. ;)
User avatar
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: 43S News

Post by Walter »

In a nutshell:
  • What shows a decimal point or comma is a real number at least.
  • What is trailed by an index is a finite integer.
  • What shows neither a point nor an index is an infinite integer.
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
User avatar
GeisiX
Posts: 14
Joined: Thu Mar 18, 2021 11:47 pm
Location: Bavaria

Re: 43S News

Post by GeisiX »

Thank you for the roundup, Walter. This is clear. As redglyph said, my request is all about the on-screen organisation, not wanting to touch any datatypes or their functionalities (which, I think, have been chosen wisely as implemented).

Something completely different:
Are the function parameters of the data fit functions stored in / accessible from some variable? (e. g. a0 and a1 in LogF: R(x) = a0 + a1*ln(x) )
User avatar
Jaymos
Posts: 1635
Joined: Sun Nov 04, 2018 7:03 pm
Location: Cape Town

Re: 43S News

Post by Jaymos »

GeisiX wrote:
Fri Nov 11, 2022 3:21 pm
Something completely different:
Are the function parameters of the data fit functions stored in / accessible from some variable? (e. g. a0 and a1 in LogF: R(x) = a0 + a1*ln(x) )
They are stored in registers X & Y in your example and if otherwise applicable in register Z. If you want to keep it, you need to store it from those registers.
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
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: 43S News

Post by Walter »

Eric Rechlin published the conference videos of the Allschwil Meeting 2022 - 8 hours, 58 minutes, and 19 seconds:
Eric Rechlin wrote: HP-92 Bobcat Prototype (Chuck McCord): 06:47
HP-75 Software Development System Prototype (Chuck McCord): 07:52
Inside HP's Pioneer Calculators: The Sacajawea IC (Chuck McCord): 34:17
Inside the HP-19C (Bernhard Emese): 51:57
Advanced HP-IL Communication (Sylvain Côté): 1:09:19
V41 The Windows HP-41 Emulator, Part 2: A Dream Became Truth (Christoph Giesselink): 21:36
PIL Storage Drive (Pascal Dagornet): 36:24
Towards HP-41 Visual FOCAL++ (Angel Martin): 1:04:20
YPRT: Developing / patching the Printer ROM (Meindert Kuipers): 34:13
Voyager Single Chip Series 10 (SST) (Chuck McCord): 31:42
What if? We could have another model in the Voyager Series (Chuck McCord): 13:54
HP-41 MAXX Module (Sylvain Côté): 16:03
Vapourware Condensed [WP43 Update] (Walter Bonin): 50:58
C43: The Classic 43 [WP43 Variant] (Jaco Mostert): 1:02:12
Repair of a 1971 Japanese Calculator (Jef Ongena): 36:45
Enjoy watching!
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
Post Reply