HP Prime - worthy successor to HP-71B ?

[Valentin Albillo]

.
Hi, Bob,

(01-11-2024 02:04 AM)rprosperi Wrote:  

Maximilian Hohmann Wrote:No, not to everyone here.

Worry not Max! If this hypothetical machine does come out, I'll buy 4 to cover your 2.

Make that six.

Best regards.
V.

[vaklaff]

Whenever this dream of a new RPL calculator arises, I can't resist but point to Christophe de Dinechin's firmware for SwissMicross machines: https://www.hpmuseum.org/forum/thread-20157.html. It's much closer to reality than calls for Moravia to start from scratch, IMHO.

[polbit]

(01-12-2024 02:39 AM)rprosperi Wrote:  [quote='Rolief_Rechner' pid='182847' dateline='1705024216']
Actually no, not the same situation at all, for many reasons:

The 15C already had proof that a re-issue will sell out quickly: the 15C LE !

The 15C had many more years to gather fans, and it's RPN (vs. RPL) so in demand by a much larger audience.

Finally, in the case of the 15c CE, Moravia could repurpose 12C hardware since it's still in production.

For a new kind of 48GX++, with enhancements, it would mean developing the h/w from scratch and basically redeveloping the ROM, nearly from scratch. So, orders of magnitude more investment would be needed, and I don't believe anyone could justify such. Just my opinion though...

I understand it from a hardware perspective, but why would there be a need to redevelop the ROM, since it is already running on the ARM from 50g?

[rprosperi]

(01-12-2024 05:50 PM)polbit Wrote:  

(01-12-2024 02:39 AM)rprosperi Wrote:  [quote='Rolief_Rechner' pid='182847' dateline='1705024216']
Actually no, not the same situation at all, for many reasons:

The 15C already had proof that a re-issue will sell out quickly: the 15C LE !

The 15C had many more years to gather fans, and it's RPN (vs. RPL) so in demand by a much larger audience.

Finally, in the case of the 15c CE, Moravia could repurpose 12C hardware since it's still in production.

For a new kind of 48GX++, with enhancements, it would mean developing the h/w from scratch and basically redeveloping the ROM, nearly from scratch. So, orders of magnitude more investment would be needed, and I don't believe anyone could justify such. Just my opinion though...

I understand it from a hardware perspective, but why would there be a need to redevelop the ROM, since it is already running on the ARM from 50g?

The latest 50g ROM, when last updated (2.15 IIRC), had less than 10 bytes of available space, so adding any enhancements would mean major refactoring before even starting on the new features. And both Tim and Cyrille have commented that mods to the 50g ROM were extremely difficult as it's highly patched state is very frail and nearly any change led to other undesirable side-effects, etc. So, many man-years of effort by folks with System-RPL and likely both Saturn and ARM assembler experience would be needed. Not impossible, but also not cheap.

[John Keith]

(01-13-2024 02:42 AM)rprosperi Wrote:  The latest 50g ROM, when last updated (2.15 IIRC), had less than 10 bytes of available space, so adding any enhancements would mean major refactoring before even starting on the new features. And both Tim and Cyrille have commented that mods to the 50g ROM were extremely difficult as it's highly patched state is very frail and nearly any change led to other undesirable side-effects, etc. So, many man-years of effort by folks with System-RPL and likely both Saturn and ARM assembler experience would be needed. Not impossible, but also not cheap.

This is why my preference for a "modernized" HP 4X would be NewRPL or DB48X running on modern hardware, along with an emulator similar to EMU48 that can use ROM 2.15 to allow the use of "legacy" programs and libraries that contain SysRPL or assembly code.

[rprosperi]

(01-13-2024 08:15 PM)John Keith Wrote:  

(01-13-2024 02:42 AM)rprosperi Wrote:  The latest 50g ROM, when last updated (2.15 IIRC), had less than 10 bytes of available space, so adding any enhancements would mean major refactoring before even starting on the new features. And both Tim and Cyrille have commented that mods to the 50g ROM were extremely difficult as it's highly patched state is very frail and nearly any change led to other undesirable side-effects, etc. So, many man-years of effort by folks with System-RPL and likely both Saturn and ARM assembler experience would be needed. Not impossible, but also not cheap.

This is why my preference for a "modernized" HP 4X would be NewRPL or DB48X running on modern hardware, along with an emulator similar to EMU48 that can use ROM 2.15 to allow the use of "legacy" programs and libraries that contain SysRPL or assembly code.

I'd prefer a complete re-write for a new product too, but I won't expect one...

And I'm likely in the minority, but I believe any theoretical successor should be fully backward compatible with 50g RPL out of the box, possibly allowing some very specific behavior changes via flags.

Note that I'm not looking for a religious discussion here about why this view is so evil and indefensible that it defies all human logic, it's simply what I think makes the most sense, so use of it can leverage the huge base of existing apps and code, rather than enable an infinite number of discussions about why the new this or that was changed when the old way worked just fine (see HP Prime for an example). But it won't happen, so I suppose there is nothing to fear....

[Rolief_Rechner]

(01-12-2024 02:39 AM)rprosperi Wrote:  

(01-12-2024 01:50 AM)Rolief_Rechner Wrote:  You realize the same argument could have been made for not producing the 15CE? Can you hear the marketing types? - 'Who on earth would want a 40-year old machine with a strange design, and a backwards calculation system? Only those old timers in the newsgroup.'

Actually no, not the same situation at all, for many reasons:

The 15C already had proof that a re-issue will sell out quickly: the 15C LE !

The 15C had many more years to gather fans, and it's RPN (vs. RPL) so in demand by a much larger audience.

Finally, in the case of the 15c CE, Moravia could repurpose 12C hardware since it's still in production.

For a new kind of 48GX++, with enhancements, it would mean developing the h/w from scratch and basically redeveloping the ROM, nearly from scratch. So, orders of magnitude more investment would be needed, and I don't believe anyone could justify such. Just my opinion though...

All your points are good. What is discouraging is the apparent lack of enthusiasm, even from those might enjoy such a machine. Where there's a will there's a way, but the will is missing on this. Which means the probability of it ever happening is essentially zero. Oh well...

[jthole]

(01-14-2024 05:45 AM)Rolief_Rechner Wrote:  All your points are good. What is discouraging is the apparent lack of enthusiasm, even from those might enjoy such a machine.

It's called realism. If they can't reuse an existing hardware platform and most of the software, it won't happen. Too expensive and too much risk.

The best you can hope for (and even that is questionable) is something like a SwissMicros DM48 (maybe running C3D's DB48X program when that's mature enough).

[Rolief_Rechner]

(01-14-2024 06:19 AM)jthole Wrote:  

(01-14-2024 05:45 AM)Rolief_Rechner Wrote:  All your points are good. What is discouraging is the apparent lack of enthusiasm, even from those might enjoy such a machine.

It's called realism. If they can't reuse an existing hardware platform and most of the software, it won't happen. Too expensive and too much risk.

The best you can hope for (and even that is questionable) is something like a SwissMicros DM48 (maybe running C3D's DB48X program when that's mature enough).

No. It's lack of desire for the product. And thus the will to solve problems that go with such a project is nonexistent. The market wants a calcphone now. Or would it be called a phonecalc? As long as users can slide their fingers all over the screen, they're happy.

As for SM, what they've done with so little is phenomenal. But calling it like it is - none of their products are the original. For example, just happen to have a DM15L right here. The keys are stiff as a frozen tree. You have to bang those keys not to miss keystrokes. The user experience is not the same.

To end, guess I'm lucky. When the abomination that was the HP49 came out over ~20 years ago, a local university bookstore put the 48GX on sale for $79 US. I bought the last 3 they had. One of the smartest buys I've ever made.

[Peet]

(09-25-2017 03:57 PM)Vincent Weber Wrote:  Would you say "Long live the HP Prime ?"

I like PPL much more than Basic and Python is a complete new level. The Prime's rudimentary RPN even makes him a more "real" HP than the 71b.

The prime is decades ahead of the 71b in almost all respects but is missing an essential function: BEEP.

[Garth Wilson]

(02-04-2024 06:58 PM)Peet Wrote:  The prime is decades ahead of the 71b in almost all respects but is missing an essential function: BEEP.

I suppose that was supposed to be a joke; but as Keith wrote, " As was already said, the main advantage for the HP71 is I/O: the HPIL/GPIB/GPIO interface can let you control just about anything on the planet - and with an RPi ethernet bridge, you can control the rest. 8^) "  To expand on that, the 71 can interface to dozens, even hundreds, of devices at once.  My attraction to it was for lab instrumentation, not just the usual things people think of like data storage and printers.  The Prime cannot compete, even with its wireless connectivity kit and USB OTG.  I don't know why this keeps getting overlooked.  [Edit:  Oh, and about the beep: I've played music on mine as well, and you can link 71's together to play more voices together.]

[Valentin Albillo]

(02-04-2024 06:58 PM)Peet Wrote:  The Prime's rudimentary RPN even makes him a more "real" HP than the 71b.

Oh, really ? Are you sure ? Who'd have thought ...

Quote:The prime is decades ahead of the 71b in almost all respects [...]

Sure. Documentation would be a case in point, right ?

Where's the buggy toy aka Prime equivalent of the 568-page (small font) HP-71B Software Internal Design Specification Volume I, which among other things includes the whole assembly-language system listed and fully commented line by line ? And there's also Vol. II, etc.

I could go on and on but sorry, it would be wasting my time, I can't take you seriously when you dare to belittlingly compare a buggy toy vs. a professional, reliable, seasoned real-world device as the HP-71B.

V.

[rprosperi]

(02-04-2024 06:58 PM)Peet Wrote:  The Prime's rudimentary RPN even makes him a more "real" HP than the 71b.

The prime is decades ahead of the 71b in almost all respects but is missing an essential function: BEEP.

I've been using RPN on my 71b since late 1984.

Lots of HP machines are smaller or newer, many faster, a few more accurate and a handful even came in multiple colors, but for me, none since are better. I can use Algebraic equations, RPN, Forth, FOCAL or Assembler, and have used some of each in the last year alone. Oh, and you can BEEP in all of them (Algebraic BEEP? Hmmm...)

It's OK to like the Prime; despite my instincts, I still think that's fine!

If it floats your boat, like and use it, and even write about it, but there's no cause to do so at the expense of other loved machines; especially my most loved machine.

[KeithB]

Lets be fair here:
The Prime does have some rudimentary DAQ capabilities with the smart stream 410 - which I guess is still around.

I don't see why they couldn't have created a GPIB bridge.

The problem is that it was focused on the education market. There is no current market for calculator based instrument control. Except, maybe, for Garth. 8^)

[StephenG1CMZ]

(02-04-2024 11:01 PM)Valentin Albillo Wrote:  

(02-04-2024 06:58 PM)Peet Wrote:  The Prime's rudimentary RPN even makes him a more "real" HP than the 71b.

Oh, really ? Are you sure ? Who'd have thought ...

Quote:The prime is decades ahead of the 71b in almost all respects [...]

Sure. Documentation would be a case in point, right ?

Where's the buggy toy aka Prime equivalent of the 568-page (small font) HP-71B Software Internal Design Specification Volume I, which among other things includes the whole assembly-language system listed and fully commented line by line ? And there's also Vol. II, etc.

I could go on and on but sorry, it would be wasting my time, I can't take you seriously when you dare to belittlingly compare a buggy toy vs. a professional, reliable, seasoned real-world device as the HP-71B.

V.

Impressive documentation.
From a very quick look at it, I'm puzzled by several references to mainframe - which would normally have referred to large computers, whereas the HP-71 is a calculator.
Does mainframe have a special meaning in the context of the HP-71?

[KeithB]

HP-71 mainframe:
The calculator without considering any plug-in modules or accessories.
It is so expandable, you need to make this distinction.

[Peet]

(02-04-2024 11:01 PM)Valentin Albillo Wrote:  Where's the ... Prime equivalent of the 568-page (small font) HP-71B Software Internal Design Specification ...

Yes, times were simple back then and you could still print technical instructions for a pocket computer as a book. Today a Calculator is 1000 times more complex and not so easy to describe.

[Sylvain Cote]

(02-05-2024 02:36 PM)KeithB Wrote:  There is no current market for calculator based instrument control.

Of all the handheld calculators/computers that HP has produced, the 41, 71 and 75 are in a separate class.
While the 41 could be called a hybrid/transitional handheld calculator/computer, the 71 and 75 are clearly handheld computers.
They are highly customizable, have advanced IO capabilities (HP-IL) & plenty of interfaces available.
The problem is that they are slow and have limited memory when compared to today computers.

The fact that manufacturing lines are still using NI TestStand & NI LabVIEW products today, means that there was/is a market for this type of device.
Several National Instruments products are very expensive, complex and not always reliable. I am pretty sure that lots of peoples would be interested in a reasonably priced Keysight modern instant-on programmable handheld computer with advanced IO (USB-C, Ethernet, WiFi, GPIB, etc.).

(02-05-2024 02:36 PM)KeithB Wrote:  Except, maybe, for Garth. 8^)

... and me.

Edit: HP Electronic Instruments Division spin-offs
Hewlett-Packard → Agilent Technologies → Keysight Technologies

[KeithB]

I stay well away from NI software - I particularly hate TestStand with a passion.

I do all my programming in C#.

Actually, you could probably make a pretty good executable device out of an windows tablet and a GPIB interface.

[Garth Wilson]

(02-05-2024 06:52 PM)Sylvain Cote Wrote:  

(02-05-2024 02:36 PM)KeithB Wrote:  There is no current market for calculator based instrument control.

Of all the handheld calculators/computers that HP has produced, the 41, 71 and 75 are in a separate class.
While the 41 could be called a hybrid/transitional handheld calculator/computer, the 71 and 75 are clearly handheld computers.
They are highly customizable, have advanced IO capabilities (HP-IL) & plenty of interfaces available.
The problem is that they are slow and have limited memory when compared to today computers.

The fact that manufacturing lines are still using NI TestStand & NI LabVIEW products today, means that there was/is a market for this type of device.
Several National Instruments products are very expensive, complex and not always reliable. I am pretty sure that lots of peoples would be interested in a reasonably priced Keysight modern instant-on programmable handheld computer with advanced IO (USB-C, Ethernet, WiFi, GPIB, etc.).

(02-05-2024 02:36 PM)KeithB Wrote:  Except, maybe, for Garth. 8^)

... and me.

Admittedly, I very seldom use my 41's or 71's for instrument control anymore.  I still want to keep the option though.  The 41 (which I started with, before the 71) came along at the right time.  Laptops didn't really exist yet (and besides, they take up too much room on my messy workbench).  It was nice to be able to quickly unplug the 41 and take it from the workbench to my desk, and back, and put it in the corner of my attache case and take it home at the end of the day.  I have enjoyed working for small, low-budget companies, as that suits my personality.  The first ATE (automated test equipment) setup I did, in about 1986, used the 41 and a ten-page program.  I didn't realize how quickly the usage would escalate, and soon production people were using it all day and I hardly got to use it anymore, so I stepped up to the 71 for my own use.  At some point, the ATE control "graduated" to an HP 68000-based controller.  I went to a seminar on it, and one of the attendees asked why they didn't put something more powerful in it, like the new 68030, and the presenter answered that there'd be no benefit, because so much of the time is spent on things like waiting for readings to come back from the equipment.  And that's what I found, too.  Although the 41 was very slow, stepping up to the 68000-based computer, the speed less than doubled.  (And after that, we went to STD bus.)  The 41 was easier to program though, and in fact the HP Rocky Mountain BASIC 5.1 was quite a disappointment after I had become fluent with the 71 with the various modules and the LEX files from the Paris users' group and the 160KB additional RAM I had.