Programs name : i don't understand?
#1

Hi,

I'm porbably missing something but i read this forum everyday and i can read the reviews of many HP programmable calculators like the 15C, 41C, 34C, 25C, 65, 67 and other classic one. These, i thinks, are the most love one here.

I tried learning programming all of these and i found that the only one that can have more than one program in memory at the same time is the the 41 series. I especially love the 15C but i need a calculator that can take more than one program at a time, like 10 programs or more! On the 15C, after using a program, i erase it and enter another one i need. I also tried to put a new program at the end of the last i entered as i enter it as new label in my original program... But this way, the "master" program is becomming very big and less ergonomic to use and modify all the labels that acts as new programs...

Is there a way to enter 10 INDIVIDUALS programs on all these calculators like it is possible on the 41C? I'm sure it is possible because all comments about the the 15C are as positive as the 41C so...

#2

The HP-42S have the capability to store several programs by name, and have all 15C's functions and many more, and is software-compatible with the 41C. I just love it! (BTW I have all three, 15C, 41CV, 42S).

Best regards,

Nelson

#3

I no longer have a 15C, but if I remember the manual correctly, you can have multiple programs in it if you use the A - F keys as the program labels. You will probably need to start at the end of the previous program and continue with the line numbers, but I remember doing that. You may have to setup a display that would alert you for your number entries at a prompt or load the stack before running the program.

Hope this helps.
Richard

#4

Shoot! Even the 34C could in this way theoretically have multiple programs (h LBL A, h LBL B, or h LBL <some single digit; I liked 0>), but I think the complaint is, unless I misread him, is that these machines, 34C, 15C, are quite limited in RAM. The 34C had to "steal" register memory to make more program lines (7 lines for one register), so if you had some data in such a register, it was gone, though.

I must be getting old; I rolled those keystrokes right off and it's been YEARS since I actually did it; I couldn't do that with the 32SII or 48G+ (and definitely not the 49G+).

Oh, even so, I think the 15C and 34C (and 32S series) might also run out of labels for subroutines after a while and some are invalidated for other reasons, but right now I can't remember what (so I might be wrong here, I hope). But these new 48G series and 49G series tricorders don't use labeled addressing of subroutine branches. I'm still trying to teach myself these, so I won't comment much more on this except to say that I think they look so far to me like you may not really ever run out of "labels" (incorrect term for these newer machines) for programs.

#5

Ed:

I think the problem that Frederic is alluding to is that, in every programmable calculator except the 41 and 42, the programs you write aren't truly independent. Putting it into 41 terms, the labels are all global instead of local. If the "program" at LBL A has a GTO 7 and LBL 7 (by oversight) is in the "program" at LBL B, running program A executes part of the code of program B, giving rise to all sorts of problems.

I loved the 65 and the 67, but they share this problem as well. The 41, on the other hand, has that all-important END instruction that creates hard boundaries between programs and guarantees that LBL 07 is truly local.

Did I get your question right, Fred?

-Ernie

#6

Ernie,

Thanks for reminding me about the local and global labels, I had forgotten about that. The program code would have to be slim-lined with little to no spaghetti and no duplication of subroutine labels between programs, or that would cause problems.

#7

Ah, I see. Yes, I did have problems in those days with exactly that. Local variables would have kept you from worrying about something as trivial as labels for a subroutine, something that I know the 48G series has, so I assume the 49G series must have, also.

Would it have been too (much more) cumbersome for a design EE to have built that into the hardwiring (or at least firmware) in circuits like in a 34C or 15C?

If not, that'd be a great feature to put into a "replacement" or "reissue" type scientific programmable. This would truly be a time saver, as there'd be one less small thing (easily overlooked) to worry about. Good thinking, gentlemen!

#8

I never seem individual progs. When Neumann Janos invented how the computers stores the progs and datas, the individuality of progs are stopped.

We have got only one memory - we have got only one program.

Labels, subroutines, program names are just humbug... :)

Csaba

ps.: Neumann Janos == John von Neumann, but I'm a hungarian, and Neumann Janos was a hungarian, so...

#9

Ernie --

I think that Ernie Malaga identified the crux of the issue -- END statements allowing separation of program units in the 41C* and 42S, with alphanumeric "global" and numeric or single-character "local" labels.

My contrast, the other calcs you mention lack the END statement, and all labels are effectively "global". This makes it more dificult to manage libraries of separate programs.

Here's how to get around this problem on the 11C, 15C, and 34C. (NOTE: These instructions will NOT work on the 10C or 12C -- they have no labels or line-delete key; programs cannot be edited and subroutine-calls can be emulated only with GTO to line numbers.)

The 15C offers 25 labels: A-E, 0-9, and .0-.9
The 34C offers 12 labels: A, B, 0-9.
(I'd have to look up the 11C on this website.)

1. Before assigning a label in a program (e.g., LBL A), exit program mode and try to go to that label in run mode(e.g., GTO A). If the display returns without error, the label is in use; try others one-by-one, until you get an error message, then use that one.

2. Put a "return" statement (RTN) at the end of every routine. That way, any routine will stop when completed, and individual routines can be executed.

3. You may delete an individual routine line-by-line on these three calcs by pressing the back arrow (11C, 15C) or the DEL instruction (34C), starting with the last instruction in the routine. However, CLEAR PROGRAM in PRGM mode will wipe out all user programming in one fell swoop.

#10

Ok, understand what you mean for the 15C or the 34C. I just try it on my 15C and it works; thanks!

I think it's a way to do multiple programs on a 15C but... it's not very "cool" in my opinion... HP-15C were created after the 41C no? why don't they include this important ergonomic feature in the 15C (the end function)?

And what about the "XEQ" button : i don't see it on any other programmable one!?!? And i think it's the master key on a 41C no? And the user key - I really like the fact that the calculator don't make the difference between my program and ROM program - especially when i can "program" a key to execute my own program like any other key on the 41.

I read about the 31SII here as a great calculator; is there a ROM capability of naming program independently?

I all stop programming HP, TI, casio and sharp calculators because of this lack except the 41C; which HP should i bought to find this feature (no graphing; i don't like these for my hobby, just for the job...)

If there is no other one, is it possible that the old 41C is the best programmable calculator ever produce so i can stop buying HP calculators because i already have the best one???


Edited: 29 Oct 2003, 8:50 a.m.

#11

Hi Frederic,


On the 41 series, and the 32s, 32sii (and maybe the 42--I don't know that one) the "XEQ" key calls any labelled routine and runs it. On the 15c and 11c, the GSB does the same thing. (In other words, in RUN mode, pushing GSB <label> will do the same thing as XEQ <label> does on the 41.


On the 15c and 11c, the normal way to start running a program is to push f <label>. This does exactly the same thing as pushing GSB <lbl>.


But the really cool thing on the 15c and 11c is the "user mode" which switches the "f" key and the unshifted position for keys A-E, so that to run a program labelled "E", in user mode, you merely push "E" (which is the 1/x key).

Note that there are some subtle differences between labelling with letters versus labelling with numbers---only the letters can be called in one keypress using user mode. But GSB <label> will run any routine from run mode.


Of course you understand the difference between GSB and GTO in program mode-right? GSB will return to the routine that called it, whereas GTO will return to line 000-or go wherever the second routine goes!


Another thing about the 15c and 11c--the machine will allow you to accidentally put two different routines in with the same label--in which case, if that label is called, the program will go to the next one----this is why Karl suggested that you go out to RUN mode and search for the intended label before using it--so that you don't find yourself with that mess!


In the 32sii, the machine will return an error message right in program mode, if you try to use a duplicate label.

Regards,


Bill

#12

The "feature" allowing multiple use of labels goes back to the 29C, if not beyond, and was designed in, not an accident. The idea is to allow "overloading" of labels because so few are available.

Thus, the main routine might be called "A", and it might repeatedly call subroutine "A" located further on in memory. This effectively provides a second label. The only hitch is that you have to make sure you start at line 000 to insure you invoke the main routine (the first "A" encountered) and not the subroutine from the keyboard.

With the 32s' 26 labels, it probably made sense to restrict each to a single use, despite that model also having relatively limited program memory.

#13

I'm sure this must have been discussed time after time, but in case someone hasn't...

The names of the letters in "XEQ" is "ex-e-cue", which is pretty damn close to "execute." I've always thought it was a very clever mnemonic, and wondered who thought it up.

-Ernie

#14

Quote:
The "feature" allowing multiple use of labels goes back to the 29C, if not beyond, and was designed in, not an accident. The idea is to allow "overloading" of labels because so few are available.

Actually that feature goes back all the way to the original HP-65. Amazing, isn't it?

-Ernie

#15

I have personally got stuck with the HP-42S as the "best calculator" for me and that I therefore can stop there. I am aware of that I might be influencing you to actually buy one at a quite high price from eBay (or similar source), but I would be surprised if you were disappointed. Especially if you are coming from the HP-41C(V/X) and the HP-15C. In a way the HP-42S combines some of the finest features of these two calculators (in fact I like to think of the HP-42S as a HP-41CV with the Advantage module, with much more memory, faster CPU, two-line LCD and a smaller form factor).

The only thing that could change my recommendation would be if HP decided to produce a successor to the HP-42S in its new calculator line. As I have pointed out in my (plentiful) previous postings in this forum, this is actually technically "within reach" for HP. They have the emulation of the Saturn CPU (the HP-49G+ uses it), the HP-42S supports 32K RAM (there are postings on this site telling how a standard HP-42S can be upgraded to 32K by just changing the 8K RAM chip to a 32K RAM chip) - and so on. Well, to put it shortly, I have personally not stopped hoping for such a calculator...

Best regards,
Erik Ehrling (Sweden)

#16

Ernie wrote:

"The names of the letters in "XEQ" is "ex-e-cue", which is pretty damn close to "execute." I've always thought it was a very clever mnemonic, and wondered who thought it up."


Command files in the (then) large HP3000 mainframe computer (circa 1972) had the .xeq extension (e.g.: cp.xeq, days.xeq, etc).

Best regards from V.

#17

Erik posted:

"I like to think of the HP-42S as a HP-41CV with the Advantage module, with much more
memory, faster CPU, two-line LCD and a smaller form factor)."


Yes, but you forgot " ... less I/O and any expandability at all". Don't get me wrong, I like the HP42S very much and think it's one of the finest HP calculators ever, but for me the lack of I/O for saving/loading programs and data is a tremendously big no-no, specially considering the much larger RAM.

That being so, I'd rather have an HP-41CX with Advantage ROM, PPC/HEPAX/ZEN ROM, a card reader and a number of card holders with plenty of programs recorded in magnetic cards, ready to load, plus a sizable number of blank cards for any programs or data I might develop anew. The enjoyment factor and actual possibilities would be far greater.

That said, an HP42S with card reader (or any other I/O allowing for easily recording/loading programs, say SD cards, or even a barcode reader of all things !) would be a real *killer* calc, if there ever was one. I would order five at once ! :-)

Best regards from V.

#18

Hm. I always wonder how many (large) programs people develop for their calc. The 42S has what, 6kb or so. Even a somewhat complex routine needs only a few hundred bytes. What the heck do you key in there that you need 32kb??

#19

Quote:
That said, an HP42S with card reader (or any other I/O allowing for easily recording/loading programs, say SD cards, or even a barcode reader of all things !) would be a real *killer* calc, if there ever was one. I would order five at once!

The problem with this is that you have just made the 42SII the same size as the 48GX. I think having the 32KB (or more) of RAM, read/write IR and a USB port would be enough in terms of hardware improvements. This would allow the calc to stay the same size that it is now. The only other improvement I would like to see is the 128 or so unit conversions that come with the 48G Series added to the 42SII. I work in the Planning Department for a city government and I am constantly having to convert Rods, Chains, Furlongs, Meters, Hectares, Acres, Square Feet, Miles, Kilometers, etc. from deeds, surveys, and subdivision plat. It would be nice to have the 48G unit conversions in something that would fit in my shirt pocket instead of hanging from my belt.

#20

Quote:
Yes, but you forgot " ... less I/O and any expandability at all". Don't get me wrong, I like the HP42S very much and think it's one of the finest HP calculators ever, but for me the lack of I/O for saving/loading programs and data is a tremendously big no-no, specially considering the much larger RAM.

Yes, the lack of serial I/O is definitely a serious limitation and I would say that deliberately crippling the HP-42S by not giving it I/O can actually be called one of the worst mistakes HP has ever made! By not giving I/O to the HP-42S they killed a long line of very fine keystroke programmable four level RPN calcs: The HP-67 -> The HP-41C -> The HP-42S. (Hopefully they realise this and are willing to rectify it almost 15 years later...)

Theoretically I/O would be possible to add to the HP-42S. The CPU is able to create the signals needed and there is room for an additional 32K ROM that even allows for MCODE to be called from UserRPN! The "only" thing that is needed is a huge amount of Saturn Assembly expertise and good insights into electronics (unfortunately I have neither of the two).

Quote:
That being so, I'd rather have an HP-41CX with Advantage ROM, PPC/HEPAX/ZEN ROM, a card reader and a number of card holders with plenty of programs recorded in magnetic cards, ready to load, plus a sizable number of blank cards for any programs or data I might develop anew.

Well, I would actually prefer a double speed HP-48GX running HP-42X (if you haven't tried it you should!) to a fully equipped HP-41C(V/X). This setup gives you the I/O, roughly four times the speed of the HP-41C and 96K of RAM...

To wrap things up...

Three things to dream about:

- HP-42SII (or HP-42S Platinum?) with 32K RAM + I/O

- Hrastprogrammer developing HP-42X for the new HP-49G+

- Serial I/O for the original HP-42S

Best regards,


Erik Ehrling (Sweden)


Homepage:http://w1.322.telia.com/~u32220482/index.html

#21

I think that we think about other users ;)

We need quite a few bytes to develop a simple routine for our own pourpouses, but our tendency is to keep the code more friendly, so we spend more and more bytes just to make a kindly of human and machine dialog.

My dream machine is a 49G+ in a size and keyboard layout of a 15C.

8^)

M_Pio

#22

Just get a 48/49 G series calculator. THey rock! I have over 50 large programs in my 49g and still have LOTS of memory left.

#23

I have a 48G but don'T learn it so much; i like keystroke programming? it it? RPL??? humm..


So the 41CX is the best for programmability and I/O capability and not so bad for direct calculation! Where can i find a "advantage module"? what it adds to my calc? I already have the stat/math module and it's great!

If the 42 cna't be plug on my 41CX printer it's not cool..

Just for fun, i have a CX with double memory module install : How can i convert "register" in KBytes? In another word a fully load 41CX have how much memory in KB?

Mine is full, no more momory so i use card reader. It should be great to add about 5000 registers of memory to the calc so i can have all my programs in the calculator.

#24

Yes, all true, perhaps, but you'd still miss the two most sensuous things about the HP-41CX: its "heft" and its keyboard. Count me among those who just love to hold it, touch it, look at it, and press its keys. I couldn't afford one when they were new, but now, with things a little better, I've managed to put together a brand new HP-41CX, brand-new printer, brand-new wand, and two rebuilt good-as-new card readers. My wife thinks I'm crazy. She might be right.

#25

Yes, the same thing occurred to me as I plugged my 256MEGAbyte SD card into my -49g+.

#26

Frederic --

If you want a well-equipped 41C*, you should get an Advantage Pac (at least the module and manual).

Try eBay or the MoHPC Classifieds; expect to pay at least US$50. A scan of the manual can be obtained on the CD-ROM or DVD-ROM set from MoHPC.

The 42S has an IR output for the 82240B or older 82240A printers. Try eBay.

The 41CX has 124 extended registers built-in; each X-Memory module provides 238 registers, for a total of 600 extended registers. At 7 bytes apiece, that's 4200 bytes of extended memory. You can store programs, data, and text in the X-memory, but you cannot run programs from it.

#27

Well, I would actually prefer a double speed HP-48GX running HP-42X (if you haven't tried it you should!) to a fully equipped HP-41C(V/X). This setup gives you the I/O, roughly four times the speed of the HP-41C and 96K of RAM...

Not to mention that you can both HP-41X and HP-42X installed on the same machine so you can have HP-41X with 28K RAM and CCD/HEPAX/Printer/PPC/ZENROM/ADVANTAGE/CARDREADER/etc together with 96K RAM version of HP-42X at the same time ... if you have enough RAM ports, of course. If this isn't enough, you can install HP-71X with 160K RAM and Math/Forth/Assembler/HPIL/etc modules.

Regarding the HP-41X/42X/71X on HP-49G+: I am affraid that there won't be any FULL/EXTENDED versions of my emulators for this calculator because HP had very STUPIDLY dropped the size of Port1 from 256K to 128K (only one bank).

Best regards to Erik and all of you ...

#28

If you're like me you'll still need to keep buying HP calculators as you can never have too many - afterall you need a backup 41 in case your 41 gets sick, and a backup for the backup and so on - plus if you get more than a couple of modules you need storage for them and the best module holder I've found is a 41! ;-)

And then, of course, you succumb to the seductive red glow of leds but I won't go there for the moment.

James



Possibly Related Threads…
Thread Author Replies Views Last Post
  Another Prime mystery - templates don't work anymore Michael de Estrada 2 1,306 11-02-2013, 11:34 AM
Last Post: Michael de Estrada
  Prime Updates (don't get excited yet) Tim Wessman 3 1,474 09-26-2013, 06:18 PM
Last Post: Han
  Warning - Don't update RPN-67 Pro on your iPad Michael de Estrada 9 2,669 05-02-2013, 03:21 PM
Last Post: BShoring
  wp34s crystal kits -- don't worry David Griffith 7 2,064 04-16-2012, 04:58 PM
Last Post: Marcus von Cube, Germany
  OT: To All Number Crunchers: Don't miss this VBA numeric library Namir 0 785 04-04-2012, 11:25 AM
Last Post: Namir
  WP 34s: Trying to understand SLV Miguel Toro 31 7,214 10-23-2011, 03:55 PM
Last Post: Miguel Toro
  (sigh) I don't understand (TAS revisited) Jim Johnson 10 2,661 09-22-2011, 07:22 PM
Last Post: Steve Leibson
  I do not understand :-) robertoataulfo 11 2,747 09-10-2011, 05:05 PM
Last Post: Håkan Thörngren
  OT: I would like to understand robertoataulfo 5 1,602 08-19-2011, 05:00 PM
Last Post: Walter B
  I don't understand why anyone would need this Michael de Estrada 3 1,280 06-21-2011, 02:59 PM
Last Post: Michael de Estrada

Forum Jump: