HP15c CE: 8x8 equation system

[Divasson]

I was testing the matrix capabilities of the new HP15C Collector's Edition, by trying a 8x8 equation system. This was completely impossible to do with any of the previous releases: they maxxed memory allocable to matrices at 64 - so you could only invert a 8x8 matrix, but not solving a system since you needed the independent values' vector.

I started creating the two matrices, and filling them with random data, in the hope they would not be singular. To do that, I wrote a short program using the USER feature so that it fills cell by cell of both matrices.

There were three matrices: a 8x8, b 8x1, c 8x1 (result matrix)

(used to program in other languages, it is amazing to be able to solve a problem with just 10-12 steps)

Total memory used: 64+8+8= 80 registers. You still have 19 registers left for programs and operation!

Here is the result:

8x8 Equation system solving with the HP15c CE

[J-F Garnier]

I did something similar some time ago as a simple test case for using multiple matrices beyond 64 registers and solving 8x8 systems on extended 15C platforms (but not the CE of course).
I built the test in such a way that the result can be verified.

Here are the details:

- Ensure your platform has 74 or more free ("uncommitted") registers.

- create and fill a 8x8 matrix A:

8  ENTER  DIM A
MATRIX 1
PI  STO RAN#
Set USER ON
enter and run this program:
001-42,21,11  LBL A
002-    42 36  RAN#
003-          1
004-          0
005-         20  *
006-     43 44  INT
007u     44 11  STO A
008-     22 11  GTO A

After the program terminates, the stack must be x=7, y=7, z=4, t=4.
Check that the first elements A are correct:
(USER is ON)
MATRIX 1
RCL A --> A(1,1)=8
RCL A --> A(1,2)=2
RCL A --> A(1,2)=0
...
then check A globally by calculating the norm (matrix 8):
RCL MATRIX A
MATRIX 8
--> 43.6921

- create and fill a 8x1 vector b

8  ENTER  1  DIM b
(USER is ON)
MATRIX 1
132  STO b
157  STO b
196  STO b
149  STO b
231  STO b
239  STO b
180  STO b
151  STO b

check that b is correct by calculating the norm (matrix 8):
RCL MATRIX b
MATRIX 8
--> 518.0473

- Now solve A.X=b, result in b:

RESULT b
RCL MATRIX b
RCL MATRIX A
/

- Check the solution is [ 1 2 3 4 5 6 7 8 ]

MATRIX 1
(USER is ON)
RCL b  --> b(1,1)=1
RCL b  --> b(2,1)=2
RCL b  --> b(3,1)=3
...
RCL b  --> b(8,1)=8

- Successfully tested on:

DM15L w/ 228 registers,
15C LE w/ 192 registers,
PX15C w/ 160 registers,
all with a standard partition 19 DIM (i)

J-F

[Divasson]

Check the speed with the HP15C CE: amazing!!

[J-F Garnier]

(05-12-2023 07:33 AM)Divasson Wrote:  Check the speed with the HP15C CE: amazing!!

I would like very much

The speed of the 15c LE is about the same, so yes it's very fast, much faster than the other platforms.

J-F

[jonmoore]

My favourite 15c related video is the legendary William Kahan talking about the evolution from the HP-34c to the 12c and 15c, and how he convinced senior HP management to allow him to have dedicated solve, integrate and with the 15c matrix capabilities.

https://www.youtube.com/watch?v=5mBw7tnRx1c

Not that the 15c Collectors Edition needs any pre-launch hype.

[jonmoore]

(05-12-2023 07:39 AM)J-F Garnier Wrote:  

(05-12-2023 07:33 AM)Divasson Wrote:  Check the speed with the HP15C CE: amazing!!

I would like very much

The speed of the 15c LE is about the same, so yes it's very fast, much faster than the other platforms.

J-F

Much faster than a 50g or Prime, are you sure?

[J-F Garnier]

(05-12-2023 07:51 AM)jonmoore Wrote:  

(05-12-2023 07:39 AM)J-F Garnier Wrote:  The speed of the 15c LE is about the same, so yes it's very fast, much faster than the other platforms.

Much faster than a 50g or Prime, are you sure?

Faster than the other 15C platforms (was clear from the context).

[jonmoore]

(05-12-2023 07:55 AM)J-F Garnier Wrote:  

(05-12-2023 07:51 AM)jonmoore Wrote:  Much faster than a 50g or Prime, are you sure?

Faster than the other 15C platforms (was clear from the context).

Not that I want to get into a row about such a trifling matter. But the context wasn't clear to me. Clear would have been to state something along the lines of "faster than other 15c variants".

I love my 15c's (vintage and LE); the 15c is in my top 5 HP calculators of all time. But I wouldn't consider using a 15c these days for system of equation duties. Even if i limit myself to other HP calculators, there are far, far better options available (in both user experience and performance regards).

I'm cook-a-hoop that Moravia have decided to reissue the 15c and will order one in July to add to my HP collection (when they're publicly available). But on the remote chance that they've sold out in july, due to pre-release hype, so be it. The CE appears to be broadly similar from a perfomance perspective to the LE, which is the version I use on a more regular basis, so the urge to purchase one up front is minimal (much as I don't use a 15c for matrices in 2023, I happily crank through complex arithmetic duties on a 15c).

[Steve Simpkin]

(05-12-2023 07:48 AM)jonmoore Wrote:  My favourite 15c related video is the legendary William Kahan talking about the evolution from the HP-34c to the 12c and 15c, and how he convinced senior HP management to allow him to have dedicated solve, integrate and with the 15c matrix capabilities.

https://www.youtube.com/watch?v=5mBw7tnRx1c

Not that the 15c Collectors Edition needs any pre-launch hype.

A great video excerpt from Dr Kahan

From Dr Kahan's feeling on the situation, perhaps we should thank HP's marketing department of the early 1980s for making the HP-15C more scarce that it could have been, thus driving up the used market price and helping to motivate these occasional re-issues of the HP-15C.
Thanks?

KAHAN: "The HP 12C was successful enough that they were willing to take my advice about building the 15C, but not take my advice about how many to build. They wanted a third of my figure, and Harms did half again what they wanted, and that’s what they were doing. They were producing half of my number. The marketing people had a third of my number. I shouldn’t have said a half. The marketing people had a third of my number, and Harms ended up with the production line producing half of my target number, and then these calculators were disappearing off the shelves as fast as they could be supplied. MIT, for example, for a couple of years was telling its freshmen that they should buy the calculator, and it had a special deal with HP that would get them involved in a somewhat lower price.
HAIGH: What year did the calculator appear?
KAHAN: I think it was 1982, give or take. We could probably look at the manuals and find out exactly when it appeared. Well, HP never produced an advertisement for the 15C in its own right in any Western language. It might have had advertisements that listed the 15C and the 16C and the 11C and so on, but never for the 15C in its own right except for one in Japanese. I saw an advertisement in Japanese. They were selling them by word of mouth as fast as they could produce them, and when my friends and I who had worked on this went to the marketing people and tried to persuade them, “Look, set up another production line, because you want to gather your flowers while you may,” they said, “No, if we set up another production line, we may end up with inventory after all. You know how sales go. The sales rise to a peak and then they go down,” and these guys thought that they must be hitting the peak. They weren’t hitting a peak. They were hitting a ceiling. It’s a different thing. So they never did set up another production line. In consequence, the market was starved. There were waiting lists, and that window closed, and so I never did get the calculators into the hands of sufficiently many students to change the ways in which professors would issue assignments, and that was a bitter disappointment. It colored my relations with this particular group at HP. I continued to work with them for a couple of years, but my heart just wasn’t in it anymore."

[Valentin Albillo]

.
Hi, José,

(05-12-2023 06:56 AM)Divasson Wrote:  I was testing the matrix capabilities of the new HP15C Collector's Edition, by trying a 8x8 equation system [...] Here is the result: [...]

Thank you very much for posting this, very interesting and encouraging.

Hiwi Wrote:I can comfortably recommend TCS for MoHPC members, and expect this to continue into the future.

+5.   Mr. Divasson's "The Calculator Store" is the very best, truly unique.

Best regards.
V.

[johanw]

I have not yet tried it, but I assume the DM15 with the M1B firmware should also be able to do this?

The DM15 is about the same price as the 15C CE but I'm not sure if I can justify the purchase of another calculator that does the same as one that I already have.

[Steve Simpkin]

(05-12-2023 11:50 PM)johanw Wrote:  I have not yet tried it, but I assume the DM15 with the M1B firmware should also be able to do this?

The DM15 is about the same price as the 15C CE but I'm not sure if I can justify the purchase of another calculator that does the same as one that I already have.

I think the biggest advantage of the DM15L over the HP-15C CE is the ability to connect to a computer via USB. This can be used to backup/restore the calculator and also for HP-15C program development when used in conjunction with Torsten Manz's HP-15C simulator for Windows/Mac/Linux). Having the ability to write programs with documentation and simulate/debug them on the computer before downloading them to the DM15L seems like a massive feature *if* you write a lot of HP-15C programs.
https://hp-15c.homepage.t-online.de/homepage.htm

I think the biggest advantage of the HP-15C CE will be its keyboard quality and feel.

Disclaimer: I own neither of these devices.

[johanw]

(05-13-2023 01:07 AM)Steve Simpkin Wrote:  I think the biggest advantage of the DM15L over the HP-15C CE is the ability to connect to a computer via USB. This can be used to backup/restore the calculator

Yes, that is a big advantage. Espacially since I seem unable to switch batteries fast enough to be able to maintain any stored programs so I'll have to restore the calculator each time I do that.

[Valentin Albillo]

.
Hi,

(05-12-2023 11:43 AM)Steve Simpkin Wrote:  

(05-12-2023 07:48 AM)jonmoore Wrote:  My favourite 15c related video is the legendary William Kahan talking about the evolution from the HP-34c to the 12c and 15c, [...]
https://www.youtube.com/watch?v=5mBw7tnRx1c

A great video excerpt from Dr Kahan

Indeed. A little earlier in the same interview, he says (my highlights):

"Well, we could go on now with more of the Hewlett-Packard stuff because I had another ambition: the HP-15C. That was my ambition. I wanted to have a calculator that could be used by practically all engineering and most science students.

And I wanted that calculator to be able to do all the math they were going to use up to their sophomore year except for divs, grads, and curls because on a calculator, you can’t really display regions and things like that. But you could certainly do everything else."

However, on the other hand he was very dismissive of the HP-71B, despite it implementing his IEEE math and being at least two orders of magnitude more capable than the HP-15C (and only 4x more expensive). He says about the HP-71B:

"KAHAN: [...] But it only accepts Basic, or a kind of assembly language; it runs slowly; it’s got a very limited window. It’s got sort of fraction-of-a-line window, so you can’t even get a whole line of Basic in there sometimes. So it’s not a pleasant thing to use unless you have relatively simple computations. [...]

[...] also, it was expensive. You could afford it to control laboratory instruments—but if all you wanted was a handheld calculator, that was a lot of money to pay and it was easier to use a PC. [...]

HAIGH: So was that the only machine that was built to the standard, or did other people take it up, too?

KAHAN: That’s the only one I know about whose hardware was designed for the standard, and even that is a little bit stretching because, of course, this is actually a very rudimentary microprocessor microcoded from ROM to perform arithmetic in a certain way."

V.

[Steve Simpkin]

(05-13-2023 11:48 PM)Valentin Albillo Wrote:  ...
However, on the other hand he [Dr Kahan] was very dismissive of the HP-71B, despite it implementing his IEEE math and being at least two orders of magnitude more capable than the HP-15C (and only 4x more expensive). He says about the HP-71B:

"KAHAN: [...] But it only accepts Basic, or a kind of assembly language; it runs slowly; it’s got a very limited window. It’s got sort of fraction-of-a-line window, so you can’t even get a whole line of Basic in there sometimes. So it’s not a pleasant thing to use unless you have relatively simple computations. [...]

[...] also, it was expensive. You could afford it to control laboratory instruments—but if all you wanted was a handheld calculator, that was a lot of money to pay and it was easier to use a PC. [...]

HAIGH: So was that the only machine that was built to the standard, or did other people take it up, too?

KAHAN: That’s the only one I know about whose hardware was designed for the standard, and even that is a little bit stretching because, of course, this is actually a very rudimentary microprocessor microcoded from ROM to perform arithmetic in a certain way."

V.

That is rather ironic considering, as was mentioned, it was the first handheld to incorporate the IEEE floating point math standard which he contributed so much to. I suspect some of Dr Kahan's criticisms of the HP-71B were based on his vision of it it not being the best calculation tool for his target market, engineering students. In that light, his criticisms are probably more reasonable.

You, yourself mention some of the HP-71B's limitations in less than glowing terms
"unbearably high prices when released"
"rather pathetically small display, limited to a single 22-character line"
"But for every other use and purpose, it was hopelessly slow"
"built-in BASIC interpreter, which for all its power and advanced features was nevertheless excruciatingly slow when doing such extremely common operations as branching and looping"

HP-71B Modest Mater

[jonmoore]

(05-14-2023 07:33 PM)Steve Simpkin Wrote:  I suspect some of Dr Kahan's criticisms of the HP-71B were based on his vision of it it not being the best calculation tool for his target market, engineering students. In that light, his criticisms are probably more reasonable.

Agreed entirely. And in truth HP dropped the ball when it came to BASIC pocket computers. I'm fully aware of the many fans of the 71B here on the forum. But for me, Casio and Sharp were the innovators when it came to pocket computers. Sure the 71B and 75C had the Forth and Math Pac add-ons, which extended both machines considerably but the Japanese understood the market better, and iterated on their initial offerings far more successfully.

Being totally selfish, whilst also donning my fave rose tinted glasses, those HP pocket computers led to RPL,and HP certainly knew their market on that front (if we forget the 28C ever existed with its piffling 1-2kb of user program.RAM).

[rprosperi]

(05-14-2023 07:33 PM)Steve Simpkin Wrote:  That is rather ironic considering, as was mentioned, it was the first handheld to incorporate the IEEE floating point math standard which he contributed so much to. I suspect some of Dr Kahan's criticisms of the HP-71B were based on his vision of it it not being the best calculation tool for his target market, engineering students. In that light, his criticisms are probably more reasonable.

More than anything else, I think Kahan's vastly differing views on the 15C and 71B reflect the dramatic changes that took place in personal computing between 1978 and 1984. In 78, when working in the 34C, there were effectively no personal computers in widespread use by college students, so the 15C could easily be considered the best tool available for Engineering students, while by 84-85, PCs had already become relatively commonplace, if not individually owned by each student, at least available to them, along with ubiquitous minicomputer terminals everywhere.

It has been suggested a reason for the difference is because Kahan was involved in the development of the 15C and not the 71B, but that's actually not correct, he advised HP on both the IEEE implementation and the PROOT solver function.

[Valentin Albillo]

      
Hi, Steve Simpkin,

(05-14-2023 07:33 PM)Steve Simpkin Wrote:  You, yourself mention some of the HP-71B's limitations in less than glowing terms

Correct. And I can add some more to the ones you quoted, for instance:

• CALC mode is an utterly useless abomination that wastes at least 2.5 Kb of precious System ROM space which could've been put to much better use, e.g. including sorely needed additional string functions (RPT$, TRIM%, REV$, REPLACE$, NOC, etc.) or structured statements (REPEAT..UNTIL, WHILE..END WHILE, LOOP..END LOOP, SELECT..CASE..END SELECT, EXIT FOR, etc.)
        
• USING / IMAGE processing is incredibly slow, the whole shebang having an unnecessarily complicated syntax while also having glaring limitations. Probably J-F Garnier could find out exactly how much System ROM space is wasted implementing this overbloated feature but it surely exceeds even CALC's wastefulness.
        
• INTEGER operations being slower than REAL ones is counterintuitive and frankly appalling, to say the least.
        
• User-defined functions have large overheads, which significantly slows down their execution and uses lots of RAM to define and invoke, so much in fact that it frequently pays off to use subroutines instead. They also lack LOCAL variables, passing arguments by reference, accepting array/matrix arguments, etc., which are all glaring limitations (among others) which greatly detract from their usefulness.
        
• etc., etc. (variables implementation and handling, ...)

The difference between me and Kahan is that he's quite dismissive of the HP-71B but enthralled and deeply in love with the HP-15C, while I'm fully aware of the HP-71B's shortcomings but I do dearly love both the HP-15C and the HP-71B, warts and all.

Regards.
V.

[Steve Simpkin]

(05-14-2023 09:26 PM)jonmoore Wrote:  …
Being totally selfish, whilst also donning my fave rose tinted glasses, those HP pocket computers led to RPL,and HP certainly knew their market on that front (if we forget the 28C ever existed with its piffling 1-2kb of user program.RAM).

Regarding the HP-28C. Around 1986 Bill Wickes proposed designing a new advanced graphing scientific calculator with algebra capabilities. At the time there was no budget available to develop an entirely new calculator like Bill wanted to do. Bill's choice was to either use the existing hardware of the HP-18C with its limited 2K of RAM (along with more ROM) or gamble and wait sometime in the future to see if management would approve a development budget for the hardware he really wanted to develop (basically the HP-48SX). No one at HP was certain how much of a market there was for an advanced scientific graphing calculator with algebra capabilities. HP allowed the team to test the marketplace using the already developed HP-18C hardware. When the HP-28C turned out to be a huge hit, they approved the budget needed for the HP-28S and then for the HP-48SX.

I am sure some people found the 1.6 KB of user RAM quite limiting, particularly when using its advanced algebra/calculus applications. Coming from a HP-11C, it was a huge step up for me. Even with its meager memory, I was able to write dozens of small utility programs that I used when designing new analog/digital circuit boards. I also used the Solver a lot.

In any event, it wasn't that HP deliberately decided to limit the amount of memory available in the HP-28C, it was more that they decided to build it at all. The HP-28C became a pivotal stepping stone for the models that came after it.

[OlidaBel]

(05-13-2023 01:07 AM)Steve Simpkin Wrote:  I think the biggest advantage of the DM15L over the HP-15C CE is the ability to connect to a computer via USB. This can be used to backup/restore the calculator and also for HP-15C program development when used in conjunction with Torsten Manz's HP-15C simulator for Windows/Mac/Linux). Having the ability to write programs with documentation and simulate/debug them on the computer before downloading them to the DM15L seems like a massive feature *if* you write a lot of HP-15C programs.
https://hp-15c.homepage.t-online.de/homepage.htm

I think the biggest advantage of the HP-15C CE will be its keyboard quality and feel.

Disclaimer: I own neither of these devices.

I totally approve. The connection to this superb flexible simulator with the DM15L and the beautiful object with the 15C CE (I hope !!!)