Working on HP's again



#51

Hello guys. I am new to the forums (fora?) I'm 50 yrs old and ditched my retail job to go back to grad school to get certified to teach math.

I bought an HP-15C about 1985 for help in getting my BA in math. RPN was new to me and I absolutely loved it!

Needing some money 1 day I listed the calculator and manual on eBay and got nearly 350$ for it. That was about 2001. I have missed it sorely.

About a month ago I was in a local flea market with my wife looking at decorations and junk (her kinda stuff) when lo and behold, I turned a corner and found, lying on a table, an HP-48GX in the box! I took it out, turned it on, it worked. Had everything in the box it was supposed to (all manuals, papers, case, pc cord)...it all looked mint, except for a name in ink on the cover. It was marked 25$...I didn't even blink...I bought it and went to sit in the car and rediscover the joys of RPN while my wife shopped.

Loved it so much, I bought a 50G about a week ago.

Just for fun I typed 999^999 in my nSpire CAS CX and got an error, and infinity. My 50G gave me all nearly 3000 digits within seconds. I love HP.

Sorry for the long intro. I'll go read some more archives. I expect to have a bunch of questions for you experts before long.

Cheers!


#52

Quote:
About a month ago I was in a local flea market with my wife looking at decorations and junk (her kinda stuff) when lo and behold, I turned a corner and found, lying on a table, an HP-48GX in the box!

Hi Les and welcome to the "forum"..., or to the "board" as I read many times in other places...complimenst for you purchase and for your enthusiasm from a same age guy

Tell me please where's that market, I'll take there immediately my wife for shopping.......................

PS: (Long life to Latin)

Cheers from Italy

Aurelio


#53

Quote:
Tell me please where's that market, I'll take there immediately my wife for shopping...
Hopefully being married is not demanding, I'm divorced and have no plans to change this status... but then again, the address????? (no fun, no game...)

#54

Quote:
Hopefully being married is not demanding, ...

It sure is, Luiz, but it's not mandatory ;-)
#55

I can tell you it is in the middle of Arkansas, USA. But I can also tell you, now I go there at LEAST twice a week! Better be early to beat me to the next one...


#56

Quote:
I can tell you it is in the middle of Arkansas, USA. But I can also tell you, now I go there at LEAST twice a week! Better be early to beat me to the next one...

OK no problems, even if quite a bit far from my home.......just it means that I will take vacations and I'll rent a room there, for a while, for whole a month, at LEAST!

Cheers
Aurelio

#57

Hey, I'm 50, too (writing these words let us two years younger than if we pronounce them too fast...)

And welcome! I'm sure you'll enjoy this bunch of guys, here.

Cheers.

Luiz (Brazil)


#58

Thanks Luiz, I appreciate the comments!

#59

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.


#60

Quote:
No.Stricto sensu, plural form of 'forum' is 'forii'.

Sorry, in which language?

Massimo

#61

Quote:
..and found, lying on a table, an HP-48GX in the box!...I bought it and went to sit in the car and rediscover the joys of RPN

What, no one is going to point out that Les discovered the joys of RPL with that 48, not RPN :-)

(Oh, I guess I just did that.)


#62

It depends on what your personal definition of "RPN" is...

There was an article about this by Richard in the latest HP Solve newsletter that is worth a read.

HP (as far as I can tell) has been working under the definition that RPN is basically "inputs first, then operation" for about 20 years now. Most people when starting to describe RPN to someone not familiar with it say something like "You put in the numbers first, and then push the operation to apply to it" or "it works just like on paper. Write down the numbers then add them."

Using that definition, both the "traditional" 4 line operation, the 30b way of doing it, and RPL both fall under the RPN umbrella and it is a more inclusive term describing a way of looking at and working a problem. This is why HP has been saying the 48, the 50 and the 30b all support RPN.

Like I said however, depending on one's personal view you may consider that completely bunk. :-)

TW

Edited: 18 June 2012, 11:37 a.m.


#63

On of the most striking differences is that
in RPN you enter 15.9 STO 4
while in RPL you enter 15.9 ENTER 4 STO
to store a value 15.9 in a hypothetical register 4. It's also one of the most annoying differences IMHO. Thus I wouldn't agree on RPN and RPL being the same, but RPL being RPN exaggerated until it hurts ;-)


#64

Yes, in this case in RPN you enter the operation and then the argument.


#65

In this case RPL is more RPN than RPN :D

#66

(italics are mine)

Quote:
Using that definition, both the "traditional" 4 line operation, the 30b way of doing it, and RPL both fall under the RPN umbrella and it is a more inclusive term describing a way of looking at and working a problem. This is why HP has been saying the 48, the 50 and the 30b all support RPN.

Like I said however, depending on one's personal view you may consider that completely bunk. :-)


It is.

Regards.

V.


#67

Quote:
It is.

The "correct" answer would have been "I do" ;)


#68

Yes, that's right. I said, "I do" when I married RPN, too ;-)

#69

Quote:
The "correct" answer would have been "I do" ;)

The correct answer is that the following affirmation:

    "(This is why HP has been saying) the 48, the 50 and the 30b all support RPN."

is completely bunk.

Regards.

V.


#70

Really? Are you saying that HP has never made such a statement?


#71

Hi,

Quote:
Really? Are you saying that HP has never made such a statement?

You do understand what underlining is used for, don't you ?

I underlined this affirmation:

    "the 48, the 50 and the 30b all support RPN."
so that's what I consider completely bunk. Saying (HP or whomever) that those RPL calculators "support" RPN is utterly misleading at best and plain deception at worst.

RPN in this context and place isn't meant literally as generic "Reverse Polish Notation" but as the very specific, awesome paradigm used in classic HP models all they way from the HP-35 to the HP42S. These wonderful machines did support RPN, the RPL models do not.

Else, if RPL is considered as "supporting" RPN, then FORTH should also be considered as supporting RPN, and so long and so forth (pun definitely intended).

In other words, stating such is completely bunk, nothing but a lame attempt at misleading. The 48, 50, and 30 b *DO NOT* support RPN, period. That is, unless they intend to redefine for whatever misleading purposes what RPN means and has always meant for us life-long HP RPN lovers.

Best regards.

V.

Edited: 19 June 2012, 4:45 a.m.


#72

I fullheartedly concur - see my statement posted a bit earlier :-)

#73

You do see the subject of your post, don't you?

I believe from the discussions on this thread that there seems to be some differences (or at least leeway) as to the exact definition of "RPN."

If you want to narrow it down to one specific paradigm, could you specify exactly how CHS interacts with stack lift? Not all machines from the 35 to 42S treat it the same way.

The same with RCL. It's a prefix operation on most but not all of those machines.

#74

Quote:
You do understand what underlining is used for, don't you ?


Kiyoshi asked a simple question; a rude answer is not helpful.

People variously use underline, boldface, italics, asterisks, and _this format_ for all kinds of different reasons, generally to draw attention to something. Most forum respondents are able to deal with this without being impolite.

The 30b is not a true RPN machine in the sense of traditional RPN machines (65, 32s, 12c, and many more). It's programming system is more a hybrid of RPN and RPL. But I doubt that anyone buys a 30b these days due to HP advertising it as an "RPN" machine. It's not.

#75

Quote:
Using that definition, both the "traditional" 4 line operation, the 30b way of doing it, and RPL both fall under the RPN umbrella and it is a more inclusive term describing a way of looking at and working a problem. This is why HP has been saying the 48, the 50 and the 30b all support RPN.

I tend to agree with this. I personally am much more productive with the "RPL" way (I grew up on a 48G) than with "pure" RPN; but I'd say they're two different "dialects" of the same language. Everytime the issue is raised here, I see comments like "In this particular case, RP(L/N) uses one keystroke less - or makes more sense - or whatever" but mostly, in a wider sense, they behave much the same. This is why, although preferring RPL, I can use RPN without big problems, but I always mess up when trying to use algebraic calculators.
Ymmv.

Edited: 18 June 2012, 1:06 p.m.

#76

Yes.

FWIW: I figure that RPN is the entry system, or logic. Whatever. Whether it's stack depth is 2, 3, 4, 5, 8, or limited only by memory. Last X or no. It's Postfix Notation.

Again, FWIW: RPL is a programming language as is keystroke programming which HP and others do using RPN.

I never got into programming in RPL. Are the versions on the 48, 49, & 50 as close as the RPN on say a 41 & 42?

#77

I definitely consider it "bunk" but at the same time yes, they are certainly related (as Walter says, "until it hurts.")

I would say that both RPN and RPL are fundamentally "postfix" contrasted to "infix".

That is as far as the entry logic goes.

For programming, they are completely different animals.
For memory useage, thing ey are completely different animals (and RPN is infix for that--and guess what--it was perfectly "logical" to us old-fart pre-RPL guys! :-)

For entry logic, they are similar, but that treatment of ENTER and DUP being different is a very significant one and the #1 cause of trouble found here in the forum when someone more familiar with one or the other is getting bad results...


#78

Although what is expected in this forum (DUP on enter) causes very large problems for anyone except those who have been trained by history to expect it.

TW


#79

Haha true.

#80

Yes, the DUP on enter is what mostly makes me feel uncomfortable when using RPN. That's why I tried to ask the WP34s team to add an option (and I repeat, an *option*, before you pull out the flame-throwers :) ) so that the user could choose to disable the DUP on Enter, to make it more like how the HP30b works. But they aren't going to add it, I suppose there's not enough request for it, or it would wreak havoc with user programs (but then, doesn't a user-selectable stack-size do that too?)
Anyway, I'm grateful for the latest HP calculators that still support postfix (the 30b, the 50g, and so on), and I hope that an eventual follower to the 50g will still support it, even if only as an option (as it is now for the recent machines).


#81

Quote:
Yes, the DUP on enter is what mostly makes me feel uncomfortable when using RPN. That's why I tried to ask the WP34s team to add an option ... so that the user could choose to disable the DUP on Enter, to make it more like how the HP30b works. But they aren't going to add it, I suppose there's not enough request for it, or it would wreak havoc with user programs (but then, doesn't a user-selectable stack-size do that too?)

Our project target for WP 34S was and is an RPN scientific. RPN meaning RPN like in vintage HP pocket calcs from HP-35 till HP-42S. Newer dialects are considered as aberrations ;-)

The larger stack became important with complex number support, since the space provided in the HP-30b is not sufficient for a variable type structure as provided by the HP-42S. With eight levels we could feature four complex levels - please see page 28f of the manual. I admit this has some very welcome side effects in real domain, but the mechanics there keep working exactly as with four levels.

Switching off the duplication of x in Y traditionally featured by ENTER would cost some nice tricks (like ENTER * for x^2) but give nothing equivalent IMHO. And yes, you're the only one requesting this IIRC.


#82

I see your point about the 8-level stack and complex numbers - I never thought about that, because I rarely use complex numbers, and when I do I use the 50g that can display them in a way I'm used to.
I use the 8 stack levels with reals because I need at least 5 levels for the (very simple) math I need to do at work - with only 4 I have to use a storage register and that slows things down.

Quote:
Switching off the duplication of x in Y traditionally featured by ENTER would cost some nice tricks (like ENTER * for x^2) but give nothing equivalent IMHO. And yes, you're the only one requesting this IIRC.

That's why I pointed out I was just asking if it could be added as an option, as I realise most people here are much more used to the "old" way of doing things! :) I know about those tricks and they are nice, but they don't make up (for me of course) for the confusion I sometimes still have with the RPN stack lift dynamics. And with this system you sometimes have to do "opposite tricks" - like having to press ENTER two times if you want to dup something, because otherwise the stack lift disabling "eats" the last level.

And by the way having to press "ENTER ENTER *" isn't really *that* much slower than "ENTER *"! :)

But please don't see this as a "pestering" on my part, I'm perfectly conscious that what you are doing, for free, is wonderful, and already more than we could have ever hoped for.

Cristian


#83

The eight level stack for reals is almost the same as an unlimited stack in my experience. I don't think I've ever encountered a problem where it ran out. Of course, it would be trivial to contrive such an example.

You could always submit a patch for the alternate ENTER behaviour and command to set & clear this. We currently have a single bit unused in the mode word which we could commit to this.......


- Pauli


#84

Quote:
The eight level stack for reals is almost the same as an unlimited stack in my experience. I don't think I've ever encountered a problem where it ran out. Of course, it would be trivial to contrive such an example.

I agree. For me (for the admittedly specific operations I do at work) 4 is too few, 5 is ok, anything more is extra.


Quote:
You could always submit a patch for the alternate ENTER behaviour and command to set & clear this. We currently have a single bit unused in the mode word which we could commit to this.......

I'm not sure if this is a "sarcastic" (tongue-in-cheek) comment or not, but if it's not... I really wish I had the ability! :) But my C programming ability doesn't go much beyond "hello world". I'm not even sure I could *understand* the entry code, let alone patch it... But I could decide to start studying the language just for the purpose! ;)


#85

It wasn't sarcastic. If someone submitted a patch to do this, we'd certainly consider it for inclusion.

It isn't the entry code that would need changing, it is more the stack lift code. The entry code is horrible, the stack management relatively clean.


Pauli

#86

Quote:
Although what is expected in this forum (DUP on enter) causes very large problems for anyone except those who have been trained by history to expect it.

You're ignoring all the millions(?) of HP-12C users for the past 30 years. It's not just history, it's what HP has had as its flagship calculator for all this time that keeps true RPN alive and well.

#87

Quote:
Although what is expected in this forum (DUP on enter) causes very large problems for anyone except those who have been trained by history to expect it.

TW


The one trap I always manage to fall into is "ENTER *" on an RPL calculator. As one of those people you are describing in your statement, could you please enlighten me which problems you run into with the DUP on enter? I honestly haven't managed to think of any so far.


#88

I believe I miswrote slightly. The "traditional" ENTER functionality on an HP RPN doesn't work well for me and never has due to my exposure to the 48 first. I always end up with two copies of whatever I am working on.

ENTER as most people understand means "accept the input you just gave" or else "move to the next line". Note that is not in reference to calculators, but general understanding and usage of most modern electronics.

With classic RPN machines you are writing directly to X as you input. This makes it very difficult for many (myself included) to switch between something like the 15 and the 50 for example. I feel extremely limited since I don't have the flexibility I grew up on and am forced into what I see as a more limited operation model.

When you have nothing but extremely simple possibility for inputs (real, complex number and maybe 1 or 2 more) this is easy to do. Once there is anything more then the basic objects or entry that starts to not make much sense anymore.

TW


Edited: 19 June 2012, 2:07 p.m.


#89

Quote:
ENTER as most people understand means "accept the input you just gave" or else "move to the next line". Note that is not in reference to calculators, but general understanding and usage of most modern electronics.

Exactly, this is just my problem. I end up doing i.e. 2 ENTER 3 ENTER + (which I intend as "Enter number 2, now enter number 3, now sum them") and getting 6 instead of 5 as a result. My brain has problems getting comfortable with the fact that the first argument must be ENTERed, while the second doesn't have to.
I especially get confused on one-line machines when I don't immediately see that X and Y are the same.

A specific usage case where the DUP on ENTER is unconfortable to me, happens often in my job. I often have to enter a number, before knowing if I have to apply an operation to it or if I have to enter another number.
So if I enter the number then I happen to not have to enter another number, I have to remember to press Rv before performing the operation, otherwise the operation is performed on the duplicated X values. I have actually lost money (not much luckily) after starting to use the WP34s, because I was so used to RPL.
The solution would be to enter the digits *without* pressing ENTER, and pressing it only if I have to add another number, before entering it. But this feels weird.

And anyway: I haven't seen many people that, having "grown" on RPL, and meeting RPN later, find the latter more efficient. And the other way around. This leads me to believe that there's no "best", but there's just "what we got used to".

Cristian


(Edit: wording)


Edited: 19 June 2012, 8:23 p.m.


#90

Interesting is that I leave out the second ENTER on RPL machines. 2 ENTER 3 + works just fine on my 28S.


- Pauli


#91

Yes, that is also true on the 48G and beyond. Any function call automatically pushes the command line into level one and operates on it (and level 2) as applicable. There has never been a need to push ENTER after every input. Obviously Bill Wickes was either not going to keystrokes or he was doing as much as possible to make the transition easy. Just think how pedantic it would be if you *had* to push ENTER after every input! 4 ENT 3 ENT + That would be stupid :-)

#92

Yes, perspective changes everything!

Quote:
e solution would be to enter the digits *without* pressing ENTER, and pressing it only if I have to add another number, before entering it. But this feels weird.

Nope, not weird. The correct way to use an RPN machine is as you suggest--don't press ENTER unless you are needing to lift to Y disable the stack lift for later. It is very bad practice to ENTER just because it says ENTER and indeed this is a problem not only with RPL people but also with some newbies including my own daughter.


#93

ENTER should be treated in your mind as a "numeric value separator", not as a "numeric value terminator", at least on the older flavors of RPN. It's not a numeric value terminator in RPL, either, which is what makes RPL so confusing for me.

/rant/Why didn't they just implement an APL interpreter?/\rant/


#94

Quote:
ENTER should be treated in your mind as a "numeric value separator", not as a "numeric value terminator", at least on the older flavors of RPN.

That's exactly how HP itself explained RPN to the user: see "RPN: A Logic Without Equal!" on the museum DVD :-)

#95

I see your point but I still don't find it consistent. If the entry line is actually the X register and ENTER is not needed just to place a value in the stack, then I will find the value in X on the left of the display or on the right depending on whether it was entered by hand or is the result of an operation, and this is an inconsistency. Or at least I see it as one.
And having to do something *before* entering a new number (in this case, pressing ENTER before the new entry, but mind you, only if the number is displayed on the left, because otherwise ENTER is not needed and it would just DUP X which is not the intended result) is another inconsistency. Again, IMO.

The bottom line is this: I think both systems have their advantages and their defects. But I really don't like when supporters of one system try to "bash" the other, or when people try to convince others that what they prefer is rubbish... To each their own, I say, and best of both worlds would be if calculators supported both at the user's choice.


#96

IMHO it's far simpler: How should the poor RPN calculator know that a sequence of digits put in represents one or two (or more) numbers? Well, let's be good boys and tell him by separating two subsequent numbers in input by an ENTER. Else, ENTER is simply not needed in RPN. That's all IMHO (hope I didn't miss anything important).

BTW, I don't really care if a number is adjusted to the left or right in display - that's just a visual aid which came somewhere on the time line.

Edited: 23 June 2012, 4:30 a.m.

#97

Quote:
I often have to enter a number, before knowing if I have to apply an operation to it or if I have to enter another number.
So if I enter the number then I happen to not have to enter another number, I have to remember to press Rv before performing the operation, otherwise the operation is performed on the duplicated X values. I have actually lost money (not much luckily) after starting to use the WP34s, because I was so used to RPL.
The solution would be to enter the digits *without* pressing ENTER, and pressing it only if I have to add another number, before entering it. But this feels weird.

Ok, I can see that being annoying. I just have never had such an application.

Quote:
And anyway: I haven't seen many people that, having "grown" on RPL, and meeting RPN later, find the latter more efficient. And the other way around. This leads me to believe that there's no "best", but there's just "what we got used to".

This is probably what the real issue is. I grew up on RPN. And when I solve a problem using my calculator, I think about the problem. And I don't (and don't want to) think about how to operate the calculator. And for me this means using an algebraic is *impossible* and an RPL hard.

#98

Quote:
It depends on what your personal definition of "RPN" is...
RPN, as an entry system, is apparently a de-facto-standard.

#99

I disagree. Depending on how you define it it can mean different things to different people. You can either be more inclusive or more restrictive.

example: having anything other then a 4 level stack can't be RPN since that is the traditional number used on the most machines. An infinite level stack is therefore disqualified even if it has the same ENTER behavior and stack lift.

You can argue that that is a meaningless detail that doesn't matter, but I that is just as distinct a difference in my mind as those arguing about a different ENTER behavior.

TW


Edited: 19 June 2012, 2:15 p.m.


The stack depth belong to this standard as much as stack lifting rules. Reason: If one of them (or both) changes, methods and programs might break. To ensure continuity and safe operation, we have such standards. RPN is one of them, RPL another one.

My *opinion* is different. I've always taken the acronym RPN literally, but learned from this board that most people actually see it as a standard.

To me, it doesn't matter too much, as long as the actual behaviour is well documented (and reasonable).

Edited: 20 June 2012, 1:30 a.m.


Possibly Related Threads...
Thread Author Replies Views Last Post
  HP Prime Save Power Turn Off Not Working Timothy Roche 12 568 10-27-2013, 01:41 PM
Last Post: Michael de Estrada
  Temporary User Mode Key Programs not working in RPN BruceTTT 7 371 10-14-2013, 01:46 PM
Last Post: BruceTTT
  HP-97 suddenly not working Brad Barton 10 342 02-25-2013, 08:15 PM
Last Post: Brad Barton
  [41CL] Problem with PowerCL ... not working Doug (NYC) 5 260 01-22-2013, 07:18 PM
Last Post: Doug (NYC)
  42S exit key not working dwsmith 4 224 11-05-2012, 10:59 AM
Last Post: dwsmith
  HP-46 not working Ignacio Sánchez 1 131 10-24-2012, 11:17 AM
Last Post: Donald Williams
  [WP34s] last QtGui version not working fhub 8 380 08-31-2012, 05:08 PM
Last Post: fhub
  Modfile not working with Win 7 x64 Richard Wagoner 6 279 07-04-2012, 10:32 PM
Last Post: Diego Diaz.
  held breath, found tapes & drive working Garth Wilson 9 372 06-12-2012, 12:52 AM
Last Post: Garth Wilson
  Working around broken Spice battery connections Les Wright 7 358 06-06-2012, 06:50 PM
Last Post: Mike T.

Forum Jump: