The Museum of HP Calculators

HP Forum Archive 13

[ Return to Index | Top of Index ]

41C & 15C
Message #1 Posted by Vincent Weber on 24 Aug 2003, 6:37 a.m.

Hi all,

I am just curious to know how 41C users welcomed the 15C in 1982. The 15C was,overall, not as an ambitious machine as the 41C: No alphanumerics (even in program listing), no expansion ports, less memory, no global labels, no user-defined keyboard, and poor performance (about half the speed of the 41C). Nevertheless, the 15C introduced new features that the 41C did not have: a clever complex stack (making full use of the 4-level stack), smooth matrix integration (matrix could even be recalled on the stack), native SOLVE and INTEG support, recall arithmetic, and all 12 conditionals. All this in a much smaller package, with every function accessible directly thanks to a well-organized keyboard with two shifted keys. At the time of the 15C introduction (1982), only the MATH module was available for the 41C. This module could do basic matrix operation, but no proper matrix arithmetics which were natively supported by the 15C. Only in 1985 was introduced the Advantage module. IMHO, this module (finally) outperformed the 15C in terms of matrix support, especially with complex matrixes. The 41C' alphanumeric capabilities made it easier to enter an review a matrix than the 15C, even if the matrixes could not be recalled on the stack. But the complex integration was not as good as the one of the 15C (the complex stack was limited to 2 levels). However, it was possible to use the CA application on the PPC ROM to work with an 'infinite' complex stack. Only the 42S would bring the best of both '41C and '15C worlds (without the expansion and I/O, though - nothing is perfect !) in 1988. Overall I think the 41C is a better machine than the 15C, but that 41C users must have been frustrated that a mid-range, cheaper model could outperform it in quite a few areas, at least before the Advantage module was released. That's why I think it today we have a 41 emulator for Palm (P41CX) while we still do not have a 15C emulator (only a simulator for PocketPC, Pocket15C by Lygea).

Thanks for sharing your thougts on the above :)

Best regards,

Vincent Weber

      
Re: 41C & 15C
Message #2 Posted by Patrick on 24 Aug 2003, 12:16 p.m.,
in response to message #1 by Vincent Weber

I bought a 41C/card reader in 1980, within a week of them first becoming available at my local college book store. I still have both. That machine replaced the HP-25 that I had been using since 1976.

The Spice models came out after the Woodstocks and I was somewhat unimpressed with them. It appeared that HP had taken a step backwards in build quality and I was disappointed with them for backing off of the quality heritage of the Classics and Woodstocks. Of course, SOLVE and INTEGRATE were quite fascinating on the 34C (there, Norm), but the looks and apparent quality still threw me off. Besides, my 25 showed no signs of deterioration so I had no appetite for another battery-hungry LED model (nor did I have enough money in those lean college days to splurge on a third calculator).

When the Voyagers came out, the disappointment of the Spices was still with me. I therefore gave the 11C just a quick perusal. I did like the looks of them, although you have to remember that they looked very, very odd in those days. People like me came to think of the Classics as what a calculator should look like (the 41C certainly didn't disappoint in that respect). So, I went into the bookstore to try out a Voyager, to measure the mettle of which they were made. The 11C sure felt good in the hand. The build quality had certainly returned, and the display was great. Then I pressed the ENTER key. Yikes! It doesn't click! Well, it does, but nowhere like it should! That lone low-tactile key really turned me off. It was bad enough that they had oriented it up and down (horrors!), but damn that weak key click!! I decided right then and there not to buy one and never even followed the Voyager series after that. Despite being an HP fanatic even then, it hardly made a difference to me when the 15C and 16C were announced.

Thinking back on this, it strikes me that I was looking to find fault in these devices. I was certainly very happy with the 41C and used it on a daily basis, either for useful stuff or for exploring the then new area of synthetic programming. Probably I realized I didn't have enough money to justify another purchase and found a reason not to indulge my passion.

People who remember my previous postings here know that I am now a huge fan of the Voyagers, and particularly of the 15C. I agree with you, Vincent, about the extremely well engineered keyboard and features. It is a wonderful little machine. I lament not giving them more of their due back when they were available. Every Voyager I own -- all 22 of them -- were purchased on eBay. However, to this day, I remain very unhappy with the click on the ENTER key, and probably always will.

      
Re: 41C & 15C
Message #3 Posted by Bill Smith on 24 Aug 2003, 1:08 p.m.,
in response to message #1 by Vincent Weber

actually, it was the other way around for me.

my first calc was the 25 in my freshman year of 1977. when the 34C arrived, i upgraded primarily for the increased program space and continuous memory. i couldn't really justify the 41C, but when the 15C arrived, i just couldn't resist. it is still the best calc ever made, IMHO. i finally got a 41CV during my first job in 1985; alas i still have the coupon for the free Advantage module, and can't quite stomach the ebay prices to obtain one now. the 41CV was my desktop PC through the remainder of the eighties, and i still occasionally use a couple of analysis programs too long, complex and infrequently used to port to another platform. They work on Calc41 on my 200LX without modification.

      
Have a look at these articles
Message #4 Posted by Valentin Albillo on 25 Aug 2003, 4:57 a.m.,
in response to message #1 by Vincent Weber

Hi, Vincent:

Have a look at these articles, recently published in Datafile, which discuss at length *exactly* the topics you raise in your message. Both articles form a matched pair, discussing and comparing the HP-41C vs. the HP-15C matrix capabilities, and including a thoroughly commented program implemented in *both* machines, with appropriate remarks about the relevant differences:

DatafileVA007.pdf

DatafileVA008.pdf

Hope you'll find them interesting, any comments welcome :-)

Best regards from V.

BTW, should you want to have a look at some of my former Datafile articles:

DatafileVA001.pdf

DatafileVA002.pdf

DatafileVA003.pdf

DatafileVA005.pdf

DatafileVA006.pdf

Edited: 25 Aug 2003, 6:32 a.m.

            
Re: Have a look at these articles
Message #5 Posted by Vincent Weber on 25 Aug 2003, 6:59 a.m.,
in response to message #4 by Valentin Albillo

Hi Valentin,

Thanks a lot ! I really appreciated your article. Very accurate and informative. I have just two comments:

1/While the matrix operations on the Advantage are clearly better than the ones on the 15C, for complex numbers handling the 15C has an edge (full complex stack); although one may argue that what is useful with complex numbers is matrix operations (much better done on the 41C as you pointed out) and basic operations on complex numbers are not that useful, especially when it is quite cumbersome to deal with polar form on both machines; I think the makers of the Advantage ROM should have included the CA implementation from the PPC ROM (infinite complex stack). 2/In the end, when you say that a 41C will never look at a 15C or 42S with envy again... this is true of the 15C, but I am not sure that it is true of the 42S, which has a really fantastic matrix integration ! 3/Advantage ROM, to be complete, should have included hyperbolics and hex/bin arithmetitics.

What do you think ? Cheers, Vincent

                  
Re: Have a look at these articles
Message #6 Posted by Valentin Albillo on 25 Aug 2003, 8:05 a.m.,
in response to message #5 by Vincent Weber

Vincent posted:

"While the matrix operations on the Advantage are clearly better than the ones on the 15C, for complex numbers handling the 15C has an edge (full complex stack)"

Agreed, no machine handles complex numbers better than the HP-15C (not even the HP42S), save the HP-71B/Math ROM bundle, but that's not RPN and lacks some functions that the HP-15C has defined for complex arguments (such as inverse trigonometrics). Anyway, as stated in my article, the intention was to "commemorate" the Advantage ROM so as it wasn't a full review, I focused on just one important topic: matrix handling and compatibility.

"I think the makers of the Advantage ROM should have included the CA implementation from the PPC ROM (infinite complex stack)."

Disagree. I'm not very fond of "infinite" stacks, a la RPL because I think the disadvantages (variable size which can result in less than 4 stack registers or even none, no automatic replication of the upper level, 'garbage collection' frequently necessary, lots of memory used by the stack, slower, cumbersome LAST functionality, etc) do not pay off for the advantages. To me, it's just a book case of 'diminishing returns', i.e.: 2 or 3 levels are too few, 4 levels are much better, even ideal, 5 levels would perhaps be marginally better, but 6, 7, or 'infinite' levels do not really improve things appreciably and can even detract from usability.

In the end, when you say that a 41C will never look at a 15C or 42S with envy again... this is true of the 15C, but I am not sure that it is true of the 42S, which has a really fantastic matrix integration !

Yes it does, but it completely lacks I/O ... and that's extremely important and that's the way it is. If you can point out some particular strength of the HP42S, such as its 'fantastic matrix integration', it's only fair that I should be allowed to point out and remind you of the greatest strenght of the HP-41C system, namely its I/O and expandability, right ?

You can bet that if I have my HP-41CX with a double X-MEM module attached (i.e: nearly same RAM as a 42S), with an Advantage ROM and a PPC ROM plugged in, and a beautiful card reader ready to download any program from a small card holder full of them, in seconds, and you're next to me with an HP42S, you're probably the one that's going to feel envious as soon as we began to do something serious with them, or simply show off each other our latest tricks.

In short, the HP42S may be a fine machine, etc, etc, but next to the HP-41C described above, the fact that you can't plug any ROM modules into it to expand its capabilities, and the fact that you have to key in every program you intend to use, and can never backup anything on external media, make it a non-contender, actually. No way any 41CX user is going to feel 'envious'. What wouldn't any HP42S user give for being able to use a card reader, uh ? :-)

Advantage ROM, to be complete, should have included hyperbolics and hex/bin arithmetics.

Agreed. However, I would have preferred that all its many functions were machine-coded, complex functions in particular. Also, a better TVM section, a la HP-12C, would have been nice as well.

Thanks for your interest and kind comments, and best regards from V.

                        
Re: Have a look at these articles
Message #7 Posted by Vincent Weber on 26 Aug 2003, 7:54 a.m.,
in response to message #6 by Valentin Albillo

Hi Valentino,

I agree with you on the I/O on the 42S. I just think that the 42S has a lot of valuable features: the complex and matrix integration is neat, the variables is a nice touch, the menus are nice, the solver makes it easy to solve for different variables (without the equation feature of the 32S, too), and the speed is great. Sadly the lack of I/O limits its interest, but a lot of features are alreayd on-board, and when I praised it I was more talking from an emulation point of view: a emulator/simulator for Palm of the 42S would be great, and would allow to enter programs. I am actually working on such a simulator, rather than emulator: I gives me more freedom to correct some flaws of the 42S (the worse alpha entry method ever, the lack of equations and fractions from the 32SII, the lack of direct polar/rect entry without changing the complex display mode); furthermore, the 42S ROM is not available from HP (unbelievable, when you know that the 41, 48 and 49 ROMs are available !); emulating the Saturn processor is complex, and does not bring benefits as the 42S had no ML/Synthetics available.

Would anybody would be interested in such a simulator, btw ? I am trying to see whether there is a "market" :)

I am actually quite happy with P41CX, the 41 emulator for Palm by Charles Lee. It is loaded with full extended memory (600 register) and leaves the 4 ports free, so I can have the Advantage ROM (with great matrix support), the PPC ROM (bringing part of the 32SII functionality, such as fractions and random numbers), the CCD (improving the catalogs: I never understood why you cannot XEQ or ASN a function directly from the catalog !!), and the Finance module. Actually I prefer the TVM program from the Advantage to the one of the Finance module (better presented, IMHO). Why don't you like it ?

Cheers, Vincent

                              
Re: Have a look at these articles
Message #8 Posted by Valentin Albillo on 26 Aug 2003, 11:05 a.m.,
in response to message #7 by Vincent Weber

Vincen posted:

"Hi Valentino,"

Hi, Vincento.

"Actually I prefer the TVM program from the Advantage to the one of the Finance module (better presented, IMHO). Why don't you like it ?"

I didn't say that I don't like it, only that it is very basic. It should have included more of the advanced financial functions present in the HP-12C. As it is, financially-aware HP-41C users can still be envious of the HP-12C's capabilities, as the Advantage ROM financial functionality is no match for it (unlike the matrix functionality or even the complex-numbers functionality, up to a point).

Best regards from V.

      
Re: 41C & 15C
Message #9 Posted by Ron Ross on 25 Aug 2003, 9:07 a.m.,
in response to message #1 by Vincent Weber

I had just that choice in 1985. The Hp15c was $50 ($100) cheaper than an Hp41cv (cx). And then I would need to fork over another $50-100 for a software package ROM to get the features that came standard with the 15c. I wanted to replace my trash Ti-55 (I liked the calculator and its functions and feature, just the keyboard was worthless after 6 months of use and I was on my second one). Well an Hp15c was definitely going to be less expensive and do everything I needed.

The Hp15c could easily have been my last HP (and I mean this in the very best way!) 8o)

Later I bought an Hp42s (which I feel is better than a 15c, but I still like my stylish 15c better). I also have a 41c and a math module. I use my 42s (now expanded to 32K), but without real input, I acknowledge, the 41c series is a better calculating machine. But that is also why I often use a 48 or 49 as they are superior to a 41c in this regard. But a 41c, 42s or 15c are all able to fit is a pocket or take up minimal space in a breifcase. That isn't true of a 48/49. Also, I don't like the idea of pulling out a cannon in a meeting to do a simple calculation (simple is a relative term) and would rather use a much less obtrusive calculator such as a 41, 42 or 15.

The 41c series certainly had their time and place and were never really replaced by any equilivent calculator. The 42s was meant as an upgrade to the 15c user (which it did very well). As such, no real I/O wasn't considered all that much of a hinderance. The 28/48/49 is the upgrade path of the 41 series, but the 41 user had to give up small one hand portability and the established I/O of the 41 series (which wasn't always acceptable). If the 42s had been released with 32K and real I/O, I do not think any 41c user could ridicule a 42s (I believe Hp had such an animal in the planning as an Hp42sx). But that never happened, probably since Hp wanted the 48 to do so well and feared robbing sales from the more estemed model.

In comparison MBA's and business types never had to give up their cherished toys, since Hp still makes the 12c (15c equavalent) and the Hp17b (Hp42s equavalent) in the lower end line of calculators. That doesn't seem fair. 8o(

I'll end my rant here.

            
Re: 41C & 15C
Message #10 Posted by Veli-Pekka Nousiainen on 26 Aug 2003, 7:28 a.m.,
in response to message #9 by Ron Ross

"The 42s was meant as an upgrade to the 15c user (which it did very well). As such, no real I/O wasn't considered all that much of a hinderance."

BULL. The 42S is code compatible with the 41 and even the HP model numbering refers to that. YOU may think it's an upgrade of your 15C, but that's not what HP made it for. VPN

                  
Re: Your Wrong!!
Message #11 Posted by Ron Ross on 26 Aug 2003, 7:52 a.m.,
in response to message #10 by Veli-Pekka Nousiainen

Refer to the literature of the time when the calculator was released. The Hp32s was the upgrade for the HP11c. The Hp42s was the upgrade for the 15c, and the 28c/s was the upgrade for the 41c (poor upgrade path for 41c users).

This is documented by HP's own sales literature (for what that is REALLY worth).

That the code was 41c compatiable was a nice bit of bait, no 41c user would go for a calculator with no I/O. And that is why the 28 was probably replaced with a 48 series after only a couple of years. A 42sx was on the drawing boards (from what I have heard), but was scrapped (and I am guessing here, because it would rob sales from the 48 line). 42sx specs will make quite a few somewhat irritated: 32K RAM and serial I/O like the 48G.

                        
Re: Your Wrong!!
Message #12 Posted by Vincent Weber on 26 Aug 2003, 8:05 a.m.,
in response to message #11 by Ron Ross

All what you say is perfectly correct. It is amazing to see that the 41C has been constantly under attack since its birth: - In 1980, by the first basic pocketcomputer (Sharp PC1211). - In 1982, by the Casio fx-602p (less impressive but faster), and from HP by the 15C (less powerful but with nice extra features as discussed in this thread). - In 1984 by the HP-71B that could simulate the 41C in a much faster way. - In 1985 by the first graphic calculator (Casio fx-7000G). - In 1987 by the first HP RPL calculator (HP-28C). I understand why 41C users did not take the 28C seriously: Its punky 2K memory did not allow to take advantage of its amazing features, and you could actually have much more elaborated programs on the '41 - let alone the major flaw of the 28, i.e the lack of I/O. - In 1988 by the 28S (which started to undermine it seriously because of its huge memory for the time and its revolutionnary features), and then the 42S (much improved in many ways, but the lack of I/O did not allow to take it a a serious rival). Interesting enough, the sales brochures of the time showed the 41 and 28 on equal grounds: the 41 for exandability, the 28S for build-in capabilities. Amazingly, HP was still pricing the 41CX more expensive than the 28S ! - Finally, in 1990 by the 48SX. It is interesting to note that *only then* after 11 years (1979-1990), HP decided to discontinue the '41 series, which is a acknowledgement that no calc until the 48 could actually replace it. And the '41 fans (among which your humble servant) will still say that they prefer the '41 to the (super-powerful, but bulky, and with less flexible I/O) '48 :)

It is impressive that the '41 survived all these treats without a problem. That shows how flexible and well design this calculator was - to be able to compete with supposely much more powerful machines.

My 2 cents. Cheers, Vincent


[ Return to Index | Top of Index ]

Go back to the main exhibit hall