HP 48GX Collector’s Edition (merged) - please don't start two threads

[rprosperi]

Only on the MoHPC Forums can a fantasy wishing for a new collectible calculator devolve into a mathematical debate about the meaning of 0^0.

I'm just not yet sure if this is a good thing, or a bad thing???

[avsebastian]

(11-22-2023 01:41 AM)rprosperi Wrote:  Only on the MoHPC Forums can a fantasy wishing for a new collectible calculator devolve into a mathematical debate about the meaning of 0^0.

I'm just not yet sure if this is a good thig, or a bad thing???

jaja.

A few days ago a nephew asked me for my HP 50g calculator.

After using it, he told me that 0 ^ 0 was equal to 1.

I remembered my calculus classes (especially limits on the hopital method) that 0 ^ 0 was indeterminate. And yes, in approximate mode it gives 1, and in exact mode it gives indeterminate (strange, right?).

Check my other calculators, for the same calculation: hp 17 bII, casio 570 ES 2nd edition, hp 10bII+: undetermined. Hp 48G and hp 50g , equal to 1

The HP 48 and 50 are tremendous monsters, but one should never trust the machines 100%. ha ha ha

[Voldemar]

(11-22-2023 02:17 AM)avsebastian Wrote:  Check my other calculators

Built-in Windows calculator
0^0=1
Default Android calculator
0^0
Undefined or 1

[AMiguelTrindade]

I tried with my HP 15C CE and it gives Error 0.
My Texas TI-85 says, error DOMAIN
My HP48 gives 1.

The fact is that the limit of x^0 when x approaches 0 is 1, mas 0^0 is, in itself indetermined.

[mfleming]

(11-22-2023 01:41 AM)rprosperi Wrote:  Only on the MoHPC Forums can a fantasy wishing for a new collectible calculator devolve into a mathematical debate about the meaning of 0^0.

I'm just not yet sure if this is a good thing, or a bad thing???

Doesn't this happen about every three years? Never does get resolved...

[Gil]

This digression topic reminds me of the already treated case on the HP50G
of the undefinite integral for the special constant function `0*x'.

'0*x' INTVX
Result given by HP50G : 0 (with no dot after the 0).

And '0.*x' INTVX
Result given by HP50G : 0. (with a dot after the 0).

The (more) correct answer should be : any constant C.

Of course, the corresponding area, integrating from x=a to x=b, is 0, ie on HP50G
'S(a,b,X*0,X)' EVAL gives 0, as expected
(a rectangles of base [b-a] and height equal to zero).

[Maximilian Hohmann]

Hello!

(11-22-2023 11:05 AM)AMiguelTrindade Wrote:  The fact is that the limit of x^0 when x approaches 0 is 1, mas 0^0 is, in itself indetermined.

Only because I could not see it here does not mean that the Wikipedia article about this subject has not been mentioned yet. But just in case, this is it - a comprehensive overview of the "0^0 - Problem": https://en.wikipedia.org/wiki/Zero_to_the_power_of_zero

Regards
Max

[c3d]

(11-21-2023 02:19 AM)Liam Hays Wrote:  I can dream of an HP 51G collector's edition, with a re-created RPL OS (no Saturn emulation), a nice classic keyboard and chassis, and a SHARP memory display like the DM42 (low lag, low power).

They would never do this, but what if HP manufactured a calculator and didn't make software for it? Then, the community™️ could step in and make the software on their own.

Also, 0^0=1 is on purpose: https://www.hpcalc.org/hp48/docs/faq/48faq-5.html#ss5.2

In case you are not aware, DB48X on DM32/DM42 is quite close to your description, except that the keyboard layout is close to DM42 more than HP48:

YouTube playlist and demo

I need testers and documentation reviewers.

[c3d]

(11-22-2023 01:41 AM)rprosperi Wrote:  Only on the MoHPC Forums can a fantasy wishing for a new collectible calculator devolve into a mathematical debate about the meaning of 0^0.

I'm just not yet sure if this is a good thing, or a bad thing???

Well, that caused me to add a new setting to DB48X, 'ZeroPowerZeroIsUndefined' vs 'ZeroPowerZeroIsOne'. In that sense, it's a good thing.

Like all system flags on DB48X, this can either be used as a command, or as an argument to CF / SF / STO / RCL, i.e. you can use << 'ZeroPowerZeroIsUndefined' CF >> or << 'ZeroPowerZeroIsOne' FS? >>

I still have to pick the default setting for the flag... Now I'm in trouble.

[etn]

(05-04-2023 03:37 PM)Jlouis Wrote:  

(05-04-2023 03:20 PM)Thomas Radtke Wrote:  Why not the 50g? Just because of the ENTER key?

Yes, just because the enter key is enough.

I second the OP..

Cheers

JL

+1 from me. Thanks!

[Jonathan Busby]

(11-19-2023 08:44 PM)Kiwi Wrote:  Well, if a HP48GX Collector's Edition would ever be released, I would like see the 4MB bug resolved.

That's an error in the PORTnEOS table. Specifically, if it were corrected, almost all the RAM entries would change ( Starting with the RAM entry immediately following the table ( which, IIRC, is the temporary save area for the ISR, low in IRAM ) and due to the fact that they would need to be moved up by five nibbles ) rendering a version "S" HP48GX incompatible with almost all of the existing HP48 software ( With the exception, of course, of User-RPL programs represented in ASCII ). See these posts here and here for more information.

Regards,

Jonathan

[johnb]

(11-22-2023 11:05 AM)AMiguelTrindade Wrote:  I tried with my HP 15C CE and it gives Error 0.
My Texas TI-85 says, error DOMAIN
My HP48 gives 1.

In logic and formal debate circles, there's a logical fallacy called "Argumentum ab auctoritate" -- argument from authority. (Also called "Argumentum ab auctoritate" -- appeal to authority.) "So and so is smart and important and in charge of blah-dee-blah, so they cannot be wrong."

Here in MoHPC, we eschew such pedestrian errors in our discussions.

We prefer Argumentum e diversis eventibus calculatoris...

[Jonathan Busby]

(11-22-2023 07:03 PM)johnb Wrote:  We prefer Argumentum e diversis eventibus calculatoris...

E quo HP48GX optimum esse invenimus

[c3d]

For what it's worth, I was under the impression, until a couple of days ago, that SwissMicros had some interest in DB48X. Apparently, that is not (or no longer) the case.

If I understand their position correctly (but I'd rather let them speak for themselves if they want to), they are in the business of doing HP-compatible calculators, and to the extent that DB48X promises no such compatibility, they don't plan to use it.

The reason I am bringing that up here is that I may have conveyed the impression on this forum and elsewhere, based on earlier exchanges, that DB48X could one day be the software foundation for a SwissMicros RPL calculator. I am no longer so sure. I still hope to convince them, but ultimately, it's their business, and it's entirely their decision to make. In any case, I don't want to give false hope.

The good news is that it does not change the fact that DB48X does run on DM32 and DM42 today, and that is unlikely to change.

But that prompts me to ask the question in this thread: would you prefer

1) a fully-compatible HP48 clone, using e.g. Saturn emulation and HP ROMs, much like iOS or Android apps?

2) a source-compatible firmware that is simply in the spirit of the HP48, but takes full advantage of today's hardware, even if that means being unable to run anything that relies on SYSEVAL or Saturn assembly language?

3) a combo of 1 + 2, i.e. something that is modern, but still offers some backward compatibility mode or tools?

And just to share my personal preference, I'm working on option (2), not (1), though given a machine powerful enough (i.e. more in the smartphone-class in terms of CPU, Flash, RAM and connectivity than in the DM32-class), I'd probably later add emulation support and the ability to download ROMs in order to get (3).

(Meanwhile, DB48X 0.5.1 is out)

[johnb]

Option 1.

Actually option 1b. Fully compatible with the 50g.
Something that can run 48g or 50g programs and libraries.
But with better packaging than the 50.

[polbit]

I would definitely vote for 1b as well - 50g is the latest and greatest in the 48 line, sans the keyboard design.

Unfortunately I have a feeling we will never see an RPL machine from SM, despite what seems like a lot of interest.

[DavidM]

(11-26-2023 11:36 PM)johnb Wrote:  Actually option 1b. Fully compatible with the 50g.

Honestly, I don't need much more than a 4-banger at this point. But I actually enjoy writing RPL code. I'm afraid I no longer have the energy to learn a new dialect of a largely write-only coding system, though.

My ideal "new" RPL calculator would have the colors of the 48s/x, the key layout of a 48gx, and would faithfully emulate the 50g Saturn/SysRPL/RPL platform (with the obvious key differences). I still write SysRPL and Saturn code, so those functions are important to me.

While others may not need the SysRPL/Saturn functionality from a coding standpoint, it's still important for compatibility with the extensive wealth of programs and libraries that are freely available.

Do I ever expect to see such a product from anyone? No. But it's fun to think about.

[polbit]

(11-27-2023 01:06 AM)DavidM Wrote:  

(11-26-2023 11:36 PM)johnb Wrote:  Actually option 1b. Fully compatible with the 50g.

Honestly, I don't need much more than a 4-banger at this point. But I actually enjoy writing RPL code. I'm afraid I no longer have the energy to learn a new dialect of a largely write-only coding system, though.

My ideal "new" RPL calculator would have the colors of the 48s/x, the key layout of a 48gx, and would faithfully emulate the 50g Saturn/SysRPL/RPL platform (with the obvious key differences). I still write SysRPL and Saturn code, so those functions are important to me.

While others may not need the SysRPL/Saturn functionality from a coding standpoint, it's still important for compatibility with the extensive wealth of programs and libraries that are freely available.

Do I ever expect to see such a product from anyone? No. But it's fun to think about.

My sentiment exactly!

[vaklaff]

(11-26-2023 11:18 PM)c3d Wrote:  ...
2) a source-compatible firmware that is simply in the spirit of the HP48, but takes full advantage of today's hardware, even if that means being unable to run anything that relies on SYSEVAL or Saturn assembly language?
...
And just to share my personal preference, I'm working on option (2),
...

Looks like I'll be in the minority but only the option 2 makes sense to me:

• Whoever is interested in original RPL can still get the original machines quite easily (and even cheaper than a SwissMicros flagship).
• RPL machines emulators/simulators have already been written so when the original HW is no longer available, perhaps the emulators/simulators can be ported to HW best suited at that moment.
• It's your spare time activity, enjoy it, play with new ideas. Don't write some SW just because you think you are obliged to write it.
  

Nevertheless I have to admit I haven't made the jump from RPN to RPL yet and I'm just a passive observer so far. However I do plan to learn RPL and I'd rather do it with a modern and improved version if available :-)

[john gustaf stebbins]

(11-26-2023 11:18 PM)c3d Wrote:  But that prompts me to ask the question in this thread: would you prefer

1) a fully-compatible HP48 clone, using e.g. Saturn emulation and HP ROMs, much like iOS or Android apps?

2) a source-compatible firmware that is simply in the spirit of the HP48, but takes full advantage of today's hardware, even if that means being unable to run anything that relies on SYSEVAL or Saturn assembly language?

3) a combo of 1 + 2, i.e. something that is modern, but still offers some backward compatibility mode or tools?

2 would be good for me.

I'd also push for 4) an App for the Prime, or better a reworked Prime Pro, that allows more flexibility for setting variables and intermixing Home and CAS functions. While I really am at home programing in RPL and find reverse Polish entry much more useful for back of the envelope calculations small programming tasks, the Prime programming language is quite powerful. The main reason I don't pick up the Prime when I'm doing some things is the UI which is geared towards students and teaching, which doesn't let me create variables and small programs on the fly. If I have the need and time for a bigger solution, it is worth diving into the programming environment on the Prime to write a real program. But with the 48-50 series, I could build up something incrementally and move between projects easily. For professional work I need to be able to save information, structures, and code for more than one problem or investigation at a time.