HP33S simulator/emulator



#27

Hi all,

I'm thinking of porting a number of my programs of old for such classics as the HP-67/97, HP-34C, and such to the new HP33S, with the idea of making them available in the new, easily affordable HP33S for people who might need them. I'm thinking in such programs as linear systems of up to 7 linear equations in 7 unknowns, tridiagonal systems up to 12x12, matrix operations, complex operations, non-linear systems, etc, etc, in the fields of applied mathematics and engineering, perhaps even including the odd game or two.

This idea crossed my mind after reading the HP33S manual (PDF format) and having a look at HP's pathetic 3x3 system solving example. Imagine, including such a *puny* program for a 32Kb machine, when you could do a 7x7 system in a 224-byte HP-67/97 ! I know it's intended to be an example, but even so ...

Then I thought that it could probably help a number of people interested in those fields if I were to make versions of said programs of mine for the HP33S, which people can take to exams, and while I'm at it, I could also write some articles for publication in Datafile or posting here, maybe even a 'challenge' or two.

The problem is, I do not have an HP33S in my HP collection nor do I intend to get one in the near future. But if any of you would be kind enough to provide me with an HP33S simulator/emulator, that would do, as long as it *faithfully* emulates/simulates the workings of an HP33S, so I could test my programs on it and rest assured that they'd also run in an actual, physical HP33S. I wouldn't want to publish/post untested software, that would be utterly unacceptable to me.

Can anyone help ? :-)

Best regards from V.


#28

I can send you a 33S in the flesh. Emulating it is tricky--there are bugs to be emulated!


#29

Hi, Bill:

Bill Posted:
"I can send you a 33S in the flesh. Emulating it is tricky--there are bugs to be emulated! "

Why, thank you, you're extremely kind. But it's not a matter of money, though the HP33S is more than twice as costly here in Spain than in the US, say.

It's more that I completely disapprove of HP's decisions regarding this product, from its manufacture (KinHPo) to its keyboard to its size to its style to its LCD to its memory management to ... to everything, actually.

So, while I'm perfectly glad to try and help others by making available potentially useful programs for it, I definitely do not want to spend a single cent on it (doing that would be tantamount to me endorsing this HP product, which I absolutely don't), nor would I want other generous people like yourself spending their hard-earned money on my behalf.

Thanks anyway, perhaps some kind MoHP contributor will offer a suitable simulator/emulator.

Best regards from V.


#30

Valentin, I believe you should try the 33s. I find it is actually a very useful replacement to the 32sii. The more I use it the more I like it. I wouldn't swap my 33s for a 41C or CV, maybe CX and not sure...

Arnaud


#31

Valentin, making your programs available for the 33S would increase its relative value, thereby potentially increasing sales of the machine you so dislike.

Isn't that inconsistent? :-)

(And, unless you did huge amounts of very bad programming, I don't think there is a hope of doing a 7x7 on the 33S...and, wasn't a 5x5 the largest ever done on the HP67/97? anyone know?)


#32

Gene,

I agree with you. The number of registers in the HP33s limit the number of linear equations. Systems of 7x7 are not doable on the 33s.

Namir


#33

Hi, Namir:

Namir posted:
"The number of registers in the HP33s limit the number of linear equations. Systems of 7x7 are not doable on the 33s."

Why, thank you !!

Your clever advice has saved me from countless frustrating hours stubbornly struggling to achieve the impossible !!

Thanks, O sage one! ;-)

("Eppur si muove")

Best regards from V.


#34

Hi Valentin,

Quote:
Namir wrote in a previous post:

The number of registers in the HP33s limit the number of linear equations. Systems of 7x7 are not doable on the 33s.


A quick look in the manual (page 13-21) reveals what can be indirectly addressed using (i):

If i contains:      Then (i) will address:
---------------------------------------
+/- 1 variable A or label A
. . .
+/- 26 variable Z or label Z
+/- 27 variable i
+/- 28 n register
+/- 29 SIGMA x register
+/- 30 SIGMA y register
+/- 31 SIGMA x^2 register
+/- 32 SIGMA y^2 register
+/- 33 SIGMA xy register
<=-34, >=+C34 or 0 error: INVALID (i)
---------------------------------------

So there are 32 registers plus the index. That might be enough for a 5x5 system.

Edited: 26 Oct 2005, 3:54 a.m. after one or more responses were posted


#35

Excuse me, Marcus, but the way your message is posted, i.e., in direct reply to one of mine and then quoting some text, makes it seem as if you're quoting me and thus said text is mine, which it isn't, it's Namir's.

Please edit your posting to reflect that, I wouldn't appreciate your message being archived with that highly inaccurate line being misleadingly attributed to me, even if by mistake.

Thanks in advance and best regards from V.


#36

Hi Valentin,

you are right! I've corrected my post.

Marcus


#37

[NT]

#38

If it were possible, I'd be very interested in seeing the result.

I have to ask: why would anyone want to solve a 7 by 7 system of equations on a lowly HP33s? This type of task is far more suited to a PC or a graphical calculator. Entering over 50 numbers on a HP33s sounds painful.

Thanks,

.

#39

I just mused over the subject for a couple of seconds, and have thought that you don't really need 49 + 7 registers for a 7x7 equation, IF you solve it in real time as you enter the terms.

This is how I'd sketch it out.

First, you enter the first row (7 + 1 terms). You need to store all of these terms.

Say the first two rows are like this

[a1 a2 a3 a4 a5 a6 a7] = [b1]
[c1 c2 c3 c4 c5 c6 c7] = [b2]


You would not actually store the numbers c1, c2, etc. Instead, you store c2 -a2/c1, c3 - a3/c1, etc. In the process, you've also saved one register! C1 would not be needed to be stored at all; it has become zero.

So, you could use Gaussian elimination as you go. In each row, you save one additional register.

Ultimately, you'd only need to store 35 terms. And you if reduce the first term of each row to one, by dividing by the first term in each row, you'd only need to store 28 terms.

So, yes, it can be done.

Pivoting would be a tricky issue, I'd guess, without having thought about it much.

Edited: 25 Oct 2005, 11:55 a.m.

#40

Hi, Gene:

Gene posted:
"Valentin, making your programs available for the 33S would increase its relative value, thereby potentially increasing sales of the machine you so dislike. Isn't that inconsistent? :-)"

I'm not doing it for the 33S or for HP, I'm doing it for the people. It is my understanding that this 33S is the only RPN model allowed in certain very important US exams, so lots of people use it for that purpose and would welcome having good math & engineering software for it. That's why I wanted to contribute to the community by porting such potentially useful software of mine to this machine. My HP-71B, HP-15C, 41C, 34C, or whatever-C programs will be of no use to people taking said exams, right ?

"(And, unless you did huge amounts of very bad programming, I don't think there is a hope of doing a 7x7 on the 33S..."[and, wasn't a 5x5 the largest ever done on the HP67/97? anyone know?)"

Yes, I know. You're wrong. But who am I to discuss with such geniuses as you and Namir ("Systems of 7x7 are not doable on the 33s") ? ;-)

Thanks for your interest in my plea and

Best regards from V.


#41

Hi,

Quote:
I'm not doing it for the 33S or for HP, I'm doing it for the people. It is my understanding that this 33S is the only RPN model allowed in certain very important US exams, so lots of people use it for that purpose and would welcome having good math & engineering software for it.

I guess you are referring to the NCEES exams. No-one needs to solve a 7 by 7 system of equations on this exam. These type of questions are just not asked. When the new policy was introduced I remember reading many comments from test-takers, saying while a HP48 is handy, a normal scientific calculator is all that is required.


While it may be an interesting programming challenge, I doubt such a solver will help anyone.


#42

Hi, .

. posted:"I guess you are referring to the NCEES exams. No-one needs to solve a 7 by 7 system of equations on this exam."

C'mon, ., it's just an example of what's available, man ! Would you like me to post a comprehensive list when I can't even get a #@$&!# 33S simulator/emulator to begin with ? :-)

Anyway, perhaps it was a bad idea on my part, noone seems too interested so it may be the case I'm trying to cater for a nonexistent need. Seemed a fair use for good programs written for classic machines at the time, though.

Nah, forget it.

Best regards from V.


#43

Hi,

Please don't let me discourage you! I just picked on the the 7x7 equation solver as an example. I'm sure many other old programs could have a new life on the HP33s. I just don't think porting a 7x7 solver is worth the effort :-)

.

#44

Valentin,

I support your effort to provide/port useful programs for the 33s. While I agree with you that it should not have been released in the form it was, it is the only RPN calc that you can walk into a store and replace new. More importantly, it is, like you said the only programmable/RPN permitted on the NCEES exams.

Personally, I enjoy the challenge of a programming task that may/may not be possible! :) For example, I found a x-section moment of inertia for composite sections developed for the 49g+. By composite, I mean a cross-section composed of varying rectangular segments (imbalanced I-beam for example).

I really enjoyed writing my own version for my 33s. It provided the same ouput: X-sect I, and Ybar (y-centroid). And it fit in my pocket :)

It is too bad that I can go into Fry's Electronics and still find *NEW* HP 12C AND 12CPlatinum, yet they can't re-release a run of the classic 32sii w/ more memory. Especially after reading the solicitation for a 12C Anniversary wish-list on this board! How many versions of the 12C are required?

I have finally accepted using my 33s after my 32sii expired this summer. I have already suffered on a numerical methods exam due to poorly visible display and hitting the corner of buttons at the same time.

Regards,
ECL

#45

Gene: "[and, wasn't a 5x5 the largest ever done on the HP67/97? anyone know?)"

Valentin: Yes, I know. You're wrong.

Gene again: The program I know of for the HP67/97 that solved the largest matrix work was the "Mr. 5x5" from late 1978 in the PPC Journal.

I do not believe it is possible to solve a problem of higher order on that calculator, unless one puts restrictions on the size of the elements. There are only 26 registers available on the 67/97 and the 5x5 system takes up 25 of them of course.

Valentin, prove me wrong! :-) I'd love to see a link to a 67/97 program allowing the solution of a 7x7 matrix (and, I'll be amazed (once again) if it allows for 7-10 digit values).

Gene


#46

I explained this above. You do not need to store all 7 x 7 elements and use Cramer's rule (by solving with the determinate) once all the terms have been entered. You can store the problem as

[1 a1 a2 a3 a4 a5 a6 ] [x1]      [y1] 
[0 1 b1 b2 b3 b4 b5 ] [x2] [y2]
[0 0 1 c1 c2 c3 c4 ] [x3] [y3]
[0 0 0 1 d1 d2 d3 ] [x4] = [y4]
[0 0 0 0 1 e1 e2 ] [x5] [y5]
[0 0 0 0 0 1 f1 ] [x6] [y6]
[0 0 0 0 0 0 1 ] [x7] [y7]

(My best attempt to write a matrix expression with text)

This is the Gaussian elimation method, and it actually takes fewer computation steps than Cramer's rule, too. It also can be solved as you go (eg, the third line can be reduced to
[0 0 1 c1 c2 c3 c4 ]
before you even enter the fourth line), which saves on registers.

x1, x2, etc. is the solution column and doesn't need to be stored. You could write a program that leaves it up to the user to write those down when they're output. So, if you count, you'll see you only need to store 29 elements.

Depending on how you implement the method exactly, you might need a couple of other variables (eg., loop counters for entering the terms), and you'll have to store the first term in each row until all the terms have been entered and divided by that term.

But the HP33s's 32 registers should be enough wiggle room for that. (Plus, of course, you have the four elements of the stack, too, and towards the end you won't need to store variables - like in the above example, y7 - long before they're output to the user)

With 25 registers, could a 7x7 be done? Without working through the nitty-gritty details, I don't know, but it should be enough to do a 6 x 6 system, which only needs 21 stored terms.

If the code is general enough (if it uses loops and indirect addressing), I'd guess the code for a 6x6 system could be ported to a calculator with more registers to solve a 7x7 (or however large) system.


#47

It would be an interesting test. :-)

As I recall, the "Mr. 5x5" did a determinant and inverse.

And, I'm pretty certain the technique you describe was known to the PPC'ers at the time, so I'm not sure why certain methods were chosen.

Gene

#48

Hi again, Gene:

Gene posted:
"Prove me wrong"

Let's make this clear, Gene. I'm stating that I *do* have a 7x7-system solving program for the HP-67, which works, which is a real programming marvel to make it work in 224 bytes/26 registers, and which was written not by me but by my best friend Fernando del Rey, 20 or so years ago. It uses no magnetic card tricks or such, just very clever programming techniques.

So, unless you're plainly stating that I'm lying, I'd expect you to take my word for it, because I'm neither writing down the 200-step listing by hand for you to see, nor can provide a link to it, because web links still do not reach old paper listings.

Further, I'd expect a little more respect from you in the future, as I think I've never done or said anything to you that turned out to be intentionally false or misleading. In other words, if I say that something can be done, you can bet your house that it *can* be done, and more, I've done it or seen it with my very own eyes or else I wouldn't be claiming it's possible.

Thanks for your comments and best regards from V.

Edited: 26 Oct 2005, 4:27 a.m.


#49

Valentin, if you're going to get offended, perhaps you could at least do me the favor of reading my post more closely?

I quote: "Valentin, prove me wrong! :-) "

In internet circles, emoticons indicate a teasing or non-literal interpretation of something said.

Which, of course, you just ignored.

What I said was that the largest I know about is Mr. 5x5 from the PPC Journal.

You stated you had (or had seen) one that could do a 7x7 the HP67. I said "Show me :-)"

If you take offense at that, then methinks you were your feelings too close to your sleeves.

I'd still like to see such a program, of course.


#50

Gene posted:
"you were your feelings too close to your sleeves"

What !? Oh I see, you mean "wear" instead of "were", right ?

We non-native English speakers are sometimes quite bemused to try and understand native English speakers when they do not know their spelling ...

I guess such a person has no trouble with "its" versus "it's", "wear" versus "were", and a miriad other usual misspellings but for non-natives like myself, it can be rather puzzling at first, IMHO.

:_)

:=$

P:(/

Enough emoticons for you ?

Best regards from V.

Edited: 26 Oct 2005, 8:39 a.m.


#51

I don't recognize any of those.

Perhaps they stand for various forms of sneer? 8)

#52

I do find it sad that you chose not to acknowlege the misunderstanding of my original post.

Instead, you point out a word misspelled in haste. What's the point? Perhaps this is the most telling phrase in your reply "Further, I'd expect a little more respect from you in the future..." Are we interacting here or out to seek praise and adulation?

My post was not hostile by any means of ordinary internet usage. You misunderstood. Your reply points out a spelling mistake.

Guess the intent of your reply is lost in translation as well.

Let me restate my original point: Mr. 5x5 is the highest level determinant and inverse program I know of to be published for the HP67/97.

Gene

(P.S. If there happen to be any grammatical or spelling mistakes in this post, please allow me to find them on my own.)


#53

Is it just me, or does Valentin need professional help?


#54

Quote:
Is it just me, or does Valentin need professional help?

Careful readers, when they become posters, usually sign their thoughts with full name.


#55

No problem. My name is Jordi Hidalgo.

#56

Now, Valentin -
without keying in the 200 lines, you might just lift the tip of the veil and tell us *how* it is done?
Curiosity killed the cat, and it's killing me, too.
Cheers, Werner


#57

Hi, Werner:

Werner posted:
"Now, Valentin - without keying in the 200 lines, you might just lift the tip of the veil and tell us *how* it is done? Curiosity killed the cat, and it's killing me, too."


There's no mystery at all, Werner,
Crawl's post explained the proper way to do it. Fernando had the key idea at the time, which I think it's brilliant indeed, and cleverly managed to make it all work in as little as 224 steps or less for the program itself, and using just 26 registers. He didn't use magnetic cards to store intermediate values but simply diagonalized on the fly.

The whole process was suitably fast and smooth, and amazingly, there were still enough program steps available to include both an output routine and an error correction routine as well ! :-)

This program is ideal for the 33S, because of its extremely reduced number of addressable registers (but still larger than the 26 the HP-67 had), better instruction set (RCL arithmetic, for instance) and vastly superior speed when compared to the HP-67. So I thought it would make a worthwhile contribution to port it to this ungainly machine. But ... :-( Thanks for your interest and

Best regards from V.


#58

I just took the last NCEES exam offered this spring. And while I would have loved to use my Hp48G as I had earlier when I took the FE exam (mostly for the great units features, solver with long variable names, ease of overall use and for the real horsepower to tackls some of the advanced math ie Fourier analasys), nothing tougher than a 3x3 (or a 4x4 with an obvious reduction eq) was needed for any of the problems with 3 unknowns. If you needed a 4x4 or greater, you were racing away from the real solution.

I think the posters who gave you negative feedback where just trying to save you wasted effort on a needless exercise. That you can do this was not questioned by me if you triangalize the matrix. In reality though a circuit with complex arguements with a 3x3 is probably impossible to handle with the lowly Hp33s (unless you state otherwise B^O ).

Since the Hp33s is 3-5 times better than any other calculator allowed to use on these exams, they ain't going to be any super complex number crunching challenges. It just so happens to be my calculator at work (so I can leave my more valuable 42s at home, I suspect you would agree with THAT philosphy).

I appreciated your offer, and I do prefer to crunch my numbers with a calculator over a PC, but when I have to do any complex sophisticated number crunching, I do it with a PC for a record as well as the ease and save the results with the projects.


#59

Hi, Ron:

Ron posted:

"I think the posters who gave you negative feedback where just trying to save you wasted effort on a needless exercise."

However, let me insist in that solving 7x7 systems was but an example. The nucleus of the thing is, I have many, many quite decent programs for the HP-67/97, HP-34C, and HP-41C, among others, and I thought it would be a real contribution to port some of these to the 33S, where they'd run perfectly Ok, and which seems to be in need of them.

"In reality though a circuit with complex arguements with a 3x3 is probably impossible to handle with the lowly Hp33s (unless you state otherwise B^O )."

Would you believe that the 0.5 Kb HP-15C can manage circuits requiring a 4x4 complex system (as demonstrated in its HP-supplied manuals) ? It's somewhat laborious but can be done.

"It just so happens to be my calculator at work (so I can leave my more valuable 42s at home, I suspect you would agree with THAT philosphy)."

Certainly, certainly. Though for work I do use a SHARP PC-1350, suitably programmed. It's not that I don't care for it, but rather that I own six or seven, mostly mint, and so I don't mind (carefully) using one of the not-so-mint as a workhorse.

Thanks for your input and

Best regards from V.

#60

Hi again, Werner:

The 7x7 sytem solution's program for the HP-67/97 is still unpublished, as far as I recall, but it actually was submitted at the time (1980) to PPC Journal, for publication.

However, despite it being a real breakthrough in its genre (only 5x5 program were available at the time, without the recourse to multiple uses of magnetic cards to store intermediate results), Richard Nelson declined to publish it, because he was more keen in articles for the just recently introduced HP41C, which more or less meant the death of interest in 67/97-related materials.

That being so, Fernando and I decided to rewrite it for the new machine, taking advantage of its advanced features such as labeled prompts, multiple registers for indirect addressing and such, but essentially implementing the very same 'diagonalize on the fly' technique used in the 67/97 program.

We submitted the resulting article to PPC Journal and this time it got published, so you can find it there. Details:

   Linear Equations - Gaussian Elimination

by Valentin Albillo (4747) & Fernando del Rey (4995)

PPC Calculator Journal - V7 N5 Page 63 & 64 - June 1980

This 'on the fly' technique allowed for an enormous increase in the size of the systems which could be solved. For instance:
    HP-41C + 1 memory module:  up to 14x14
+ 2 id. : up to 20x20
+ 4 id. : up to 30x30
Frankly, being able to solve a 30x30 system of equations on a small, handheld calculator back in 1980 was absolutely amazing, and most people we showed it to wouldn't believe their eyes ! :-)

By the way, the unpublished 67/97 program is exactly 221 steps long, uses all 26 registers and all labels except D and E, and contrary to what I said, it doesn't include an output routine, you simply recall each unknown from its own register (R1-R7). For the record, the instructions look like this:

          ______________________________________________________________________
/ |
/ 7x7 System of linear equations |
/ |
| Start Restart /
| row row Reset /
|_____[A]________[B]___________[C]_____________________________________/

Instructions:

Start: [C]

Enter equations:

a11 [A] a12 [R/S], ..., b1 [R/S]
a21 [A] a22 [R/S], ..., b2 [R/S]
...
a71 [A] a72 [R/S], ..., b7 [R/S]

Recall solution:

RCL 1 -> x1
RCL 2 -> x2
...
RCL 7 -> x7

Example:

x1 + x2 + x3 + x4 + x5 + x6 + x7 = 7
2x1 - 3x2 + x3 - x4 + x5 - x6 + x7 = 0
5x1 + 5x2 - 4x3 - 6x4 + 2x5 - 3x6 + x7 = 0
4x1 + 3x2 + 2x3 + x4 - 5x5 - x6 - 4x7 = 0
3x1 + 3x4 - 3x5 - 2x6 - x7 = 0
2x1 - 6x2 -14x3 + 2x4 + 6x5 + 5x6 + 5x7 = 0
x1 - x2 - x3 - x4 - x5 - x6 + 4x7 = 0

Exact solution: x1 = x2 = x3 = x4 = x5 = x6 = x7 = 1

Computed solution: x1 = 1 +24e
x2 = 1 - e
x3 = 1 +10e
x4 = 1 + 8e where e = 1E-9
x5 = 1 - 6e
x6 = 1 +16e
x7 = 1 + e

That's all. Best regards from V.

#61

Hi, Arnaud:

Arnaud posted:

"Valentin, I believe you should try the 33s. I find it is actually a very useful replacement to the 32sii."

Thanks for the advice, Arnaud, but I own several HP32S and HP32SII, which offer approximately the same programming paradigm as the 33S, in a far better form factor and ergonomics. And, as for programming power and usability, any of my vintage SHARPs can run rings around the 33S, also with much better ergonomics.

My point is, I see this 33S being used a lot as it's the one RPN model allowed in some important US exams, and having a lot of software potentially useful to such users, which could be converted to this model, I was willing to help and contribute by doing so. But I'm not in the least interested in the machine, I dislike its physical incarnation profoundly (*not* it's programming paradigm, mind you: it's the same as the 32S/32SII, which I like), so that's why I'm aiming for an emulator/simulator.

Thanks again for your interest and

Best regards from V.

#62

Wow, I'd love to have my Windows calculator work like my 34C!!


#63

Quote:
I'd love to have my Windows calculator work like my 34C!!

You're in luck, there's a 34C simulator that runs on Windows. It's not very polished yet, and there isn't yet a user-friendly installer. (Someone contributed one, but I haven't yet got it merged into the distribution.) Anyhow, you can find it here.

I hope to make a new release available by the end of the year. In addition to the Windows installer, it should look nicer thanks to a lot of work by Maciej Bartiosak, who also had created a port for MacOS.


#64

Hey Mr. Smith!

Thanks for the tip!

My real 34C is now in a 25C case (don't ask... long story) zipped up and in my drawer. If I have my home PC on, chances are there'll be a 34C on the screen!

Wow. I thought it was just a program that works like a 34C; well it has the skin, too!

But truth to tell, the 32SII and 33S both are "34C-like" enough that I've come to be fond of both, more the former than the latter, but the latter is gaining, mainly because of slightly larger programming space and the inclusion of a constants table. Boy; if the 34C had a constants table back then... it might have saved more program lines and even more time!


#65

Ah, I've installed HP's HP33S simulator/emulator on my PC.

It *DOES* work like a real 33S, except you can use mouse clicks to your keypressing...

... and no, the equation for the volume of a solid cylinder is not preloaded in this emulator! ;)

Now, I don't like using Powerpoint or PC based projections for lectures, but I will, for HP's sake, try to keep the 33S emulator on the computer screen especially when students are wont to come by.

I'll use it when I'm too lazy to actually reach over and whip out the real 33S, which is more often than I had admitted to myself.

Too bad it doesn't have a "right hand rule" preprogrammed key! LOL!!


#66

I have thought about trying out HP's 33s simulator, but I didn't think I could justify putting it in my syllabus that my students must purchase that specific calculator. Especially when they are being required to purchase a TI-84 for earlier courses, and the local Walmart isn't re-stocking any HPs.

Apparently HP wants to see that syllabus statement before they'll let you have the simulator. How strong of a statement did you have to word?

RZ


#67

I do not put calculator requirements on my syllabi, especially for advanced courses; for the introductory levels, I just say get *ANY* scientific, not necessarily a programmable calculator.

I merely posted for HP on their site a text copy of the real syllabus which dos not mention a calculator requirement. (At the level of that particular course, if they don't have the common sense to obtain a workable machine... well, there's the first F of the season... ) Some of the advanced students may even use Maple, Origin, etc. on a PC.

But as I said here and they did not ask, I do not like using prepared-in-advance Powerpoint files on a laptop with a projector, nor slides with an overhead.

What I do, however, is to let them watch me crank through a calculation with my 33S or sometimes, 48G. If they ask, and some do, I show them and let them try my 33S and try to explain RPN and its glories of elegance and ease (they don't get to touch the 48G). So far, in many years, only a few takers. The world is sold on algebraic TIs. The more budget conscious go for the no-names in drug stores or discount shops (in which places you can get a fairly fully featured scientific, even if not programmable, for around $10 USD).

But the few who have taken an interest in RPN HP calcs are among the better and more serious students.


#68

Quote:
But the few who have taken an interest in RPN HP calcs are among the better and more serious students.

There you go: RPN makes you smarter!

I'm guessing it's the exceptional student that is curious enough to get the 33s demo. And of of these, only the very exceptional ones who are willing to adopt what seems to most a contrarian approach to calculation. So the choice of RPN may reflect a "think outside the box" type of intellectual curiosity that would map well onto exceptional math ability.

But I'm also guessing you have other exceptional students that don't get bitten by the RPN bug. Otherwise my lead above would not only make a great ad slogan, it would also stand up to FTC scrutiny for truth in advertising!


#69

Quote:


There you go: RPN makes you smarter!...

... Otherwise my lead above would not only make a great ad slogan, it would also stand up to FTC scrutiny for truth in advertising!


Hee hee hee... or is it the smart ones in their smartness choose RPN?!

#70

Ed,

Are you a professor of engineering? I am a senior undergrad M.E., and I've been doing quite well with my 32sii/33s combo (well, just a 33s at this point).

It would be interesting (if you are in engineering) to hear your opinion on the programmable (no graph capability) vs. full-blown 49g+ in the upper-undergrad to grad work curriculum.

I am beginning to run out of variables on my 33s (no shortage of ram though :) ). I have been tempted to make an official switch to the 49g+ (I own one, and have done some basic things with it).

The two interests in converting are: backup capability, & multicharacter variables. However, the seemingly infinite opportunities for customization and program development are not so appealing to me as it could be a distraction that would be fun to entertain.

Thoughts? BTW, I do plan to continue with graduate work as well.
Thanks,
ECL


#71

You flatter me.

I am a professor of chemistry.

It turns out that the calcs, HP or otherwise, to most chemists are more useful in teaching than research, as the synthetic guys (total potboilers) need only rather basic calculations generally doable on the back of an envelope like G. N. Lewis did and the theorists (complete numbercrunchers) either go the back of the envelope route or rely heavily on PCs or mainframes.

To me, the involvement in actual research calculation has been of the light duty type (quick and dirty short programs in a 33S to convert eV to wavenumber and vice versa, etc.), but the heavier (for us) usage has been in teaching. A bulk of this is in the basic statistical handling of grades, but also in use as a light, small, convenient, and highly portable machine to execute handy programs like ones to help index x-ray patterns, etc. in a classroom or teaching lab. Yes, this is normally done very efficiently in a computer, but in teaching it, not all areas have computer access and some that do don't have the right software.

Now having said all this... and from a chemist's vantage point, I personally DON'T LIKE graphing calculators... well, let me take that back a bit... I don't like the graphing aspect:

I think a student, especially an undergrad, should go through the motions of plotting whatever functions or curves he needs ON PAPER, aided by the output of a trusty HP RPN scientific programmable, to truly get the total feel and significance of the plot.

I have found that, especially for high school aged kids, but still significantly for college students, having the instant gratification of a visual output on a screen on a calculator robs the student of full understanding of the (chemical, physical, biological, statistical) meaning of the curve. Of course, you guys are obviously bright folk and won't have this problem, so for you all, it's probably okay. But for your more average performer, I hate it.

The only time I use my 48G, 48G+, or 49G+ is as a glorified scientific programmable; i.e., more programming capability and bettery equation or constants library (not necessarily faster; I once mimicked some FORTRAN programs that calculated some figures given some chemical or algebraic restrictions for convergence and it took literally overnight). For instance, that homemade powder pattern indexing program I mentioned above just CAN'T be done in a 33S or 32SII or equivalent- there just isn't the program or variable memory space available in them.

But aside from applications that demand more power, I find myself always reaching back most often for my 33S (at work) or 32SII (at home) as they are truly, the most convenient powerful machines (the others are more powerful, but bulkier and thus harder to carry, put on a cluttered surface, etc.)

Ah, look what you made me do... a ten mile long post... :p


#72

Ed,

Thanks for the response, very nice insights indeed.

I agree on the ease of use / un-clutteredness of the 33/32 series. Also, I noticed today, (obviously) I can access any program via only two keystrokes (XEQ & LABEL), which is certainly not true for deeply hidden progs in an obscure directory on my 49g+ :)

ECL


#73

Still, just a bit more RAM for program lines and memory registers in the 33S would be fabulous.

#74

To (at least attempt to) answer Valentin's question about the availability of a simulator/emulator for the 33S. As far as I know, and I would welcome correction, there is no freely available simulator/emulator for the 33S. An emulator (their term) is available directly from HP, but it is not freely available to anyone. It is only available to "educators." Valentin, I don’t believe that your occupation is known to this community, but perhaps you may qualify as an educator (your posts to this Forum are certainly educational.) If you do qualify, information is available here.


#75

Thanks for the info, Jeff, but I don't qualify as an educator.

Well, at least I tried ... Let's forget about this.

Thanks and best regards from V.


#76

Valentin,

If HP cannot provide an HP-33S emulator to YOU giving all your activities in this forum and further (but can provide it to some middle school teacher) then they are making a terrible mistake (in the absence of stronger words).

Best regards.


#77

Hi, HrastProgrammer:

Hrastprogrammer wrote:

"If HP cannot provide an HP-33S emulator to YOU giving all your activities in this forum and further [...] then they are
making a terrible mistake [...]"

Thanks a lot for your kind, supportive words but I don't think anybody at HP gives a damn about what I may or may not do in this forum or elsewhere, and if someone actually does, I doubt they can pull the necessary strings to accomplish even such a minor feat as letting me have an HP33 simulator.

Nevertheless, it's probably the case that there's actually not much public interest in my proposal, as I know for sure there are a number of posters in this forum that do have said software, yet noone has said a word to me regarding this matter, either publicly or in private e-mail.

So it's best to forget about it; it would mean a lot of work for me and if there's no interest, I'd rather spend my time and efforts in more worthwhile endeavors.

Thanks again for your support and best regards from V.

Edited: 28 Oct 2005, 5:52 a.m.


#78

So it's best to forget about it; it would mean a lot of work for me and if there's no interest, I'd rather spend my time and efforts in more worthwhile endeavors.

I think you have a lot of MUCH (and I mean MUCH) better machines to deal with than HP-33S :-)

BTW, I enjoyed reading all your articles so I am keen on the next one ...


#79

Hi again, Hrastprogrammer !

Hrastprogrammer wrote:

"I think you have a lot of MUCH (and I mean MUCH) better machines to deal with than HP-33S :-)"

Certainly, I was doing it for the people, not the machine, HP, or my own enjoyment ...

"BTW, I enjoyed reading all your articles so I am keen on the next one ..."

Thank you very much. Which are the latest you read ? Are you missing any of these ? :

    HP-71B Math ROM Baker's Dozen Vol. 1
HP-71B Math ROM Baker's Dozen Vol. 2
HP-71B Fantastic Four
HP-71B Short & Sweet Sudoku Solver
HP-71B Sudoku Solver Sublime Sequel
Mean Matrices
HP-71B Minimax Polynomial Fit
If so, let me know and I'll send them to you in PDF format. "Mean Matrices", in particular, would probably interest you for sure !

Thanks again and best regards from V.


#80

It seems that I don't have the last three articles: "HP-71B Sudoku Solver Sublime Sequel", "Mean Matrices" and "HP-71B Minimax Polynomial Fit". So, you know my e-mail address ;-)


#81

I'd like them, too!
Werner


#82

This Message was deleted. This empty message preserves the threading when a post with followup(s) is deleted. If all followups have been removed, the original poster may delete this post again to make this placeholder disappear.

#83

Quote:
    HP-71B Math ROM Baker's Dozen Vol. 1
HP-71B Math ROM Baker's Dozen Vol. 2
HP-71B Fantastic Four
HP-71B Short & Sweet Sudoku Solver
HP-71B Sudoku Solver Sublime Sequel
Mean Matrices
HP-71B Minimax Polynomial Fit

I'd love to read them all. Were are they published? (I assume in Datafile, but I don't know were to find the publication.)

I would be glad, if you could point me to a source.

Marcus


#84

Hi, Marcus:

Marcus posted:

"I'd love to read them all. Were are they published? (I assume in Datafile, but I don't know were to find the publication.)
I would be glad, if you could point me to a source."

Thanks for your interest in my articles. They are indeed 'premiered' (published) in Datafile, and you can get this publication by following this link.

Once an article's been published in Datafile, I usually let six months or so pass by, then make it available at my web site, and it also usually spreads to other sites as well. After the six-months period has elapsed, you can easily find any available article by using Google, for instance:

Keywords: "math rom baker dozen"

immdiately returns a number of links, and the very first of them will already get you a PDF copy of that article.

Thanks again and best regards from V.

#85

Valentin,

Quote:
Nevertheless, it's probably the case that there's actually not much public interest in my proposal, as I know for sure there are a number of posters in this forum that do have said software, yet noone has said a word to me regarding this matter, either publicly or in private e-mail.

As the kind of collector who isn't interested in nice looking (probably broken) exhibits behind glass, I recently acquired a brand new 33s. My intend is to understand what are the specifics of this machine. I'm mainly interested in how calculators work and what can be done with them; this includes new machines as well as old ones.

If you were so kind to publish your software, either in their original form or in a revised version for the new calc, I would be delighted to play around with it, just out of curiosity! (I don't really need any of my calculators, for me they are just interesting toys.)


Possibly Related Threads...
Thread Author Replies Views Last Post
  HP-70 simulator updated Willy R. Kunz 3 271 11-26-2013, 08:20 PM
Last Post: BShoring
  where can I get the HP Prime simulator? John Ioannidis 4 257 09-27-2013, 12:28 PM
Last Post: John Ioannidis
  OT: EDSAC simulator Mike (Stgt) 2 189 08-20-2013, 07:27 AM
Last Post: Mike (Stgt)
  Yet another simulator Mike T. 3 206 08-05-2013, 07:10 AM
Last Post: Namir
  HP-16C simulator fhub 12 543 06-30-2013, 10:14 PM
Last Post: Robert Prosperi
  HP-15C Simulator / Release 3.2.00, Build 5319 Torsten 10 480 05-11-2013, 05:19 PM
Last Post: Thomas Klemm
  Christmas tree simulator on HP39gII Mic 0 124 12-08-2012, 07:19 AM
Last Post: Mic
  A Simulator for a Stack Machine Thomas Klemm 0 104 10-19-2011, 03:02 PM
Last Post: Thomas Klemm
  Anyone know anything about this HP41C RAM / EPROM ROM SIMULATOR? Bruce Larrabee 11 431 09-30-2011, 02:25 PM
Last Post: Ángel Martin
  Update of ILPer, the HP-IL peripheral simulator Christoph Giesselink 33 1,031 06-29-2011, 06:21 AM
Last Post: Frido Bohn

Forum Jump: