HP Forum Archive 14

[ Return to Index | Top of Index ]

Why the 15c should not come back Message #1 Posted by bill platt on 21 July 2004, 3:19 p.m. OK, I got your attention! The 15c is a wonderful machine. But: after using a 32s, do you really think it makes sense to have keycodes instead of actual function names? after using a 32sii, where you can enter many algebraic expressions into the solver, do you really want to have to program, in RPN, line by line, for every function solve or integrate problem? after trying an editable Formula or true algebraic machine, such as a sharp pc 1500, (or even a Casio fx 115ms plus, or even HP30s), do you really really see an advantage to having a limited stack, intermediate operation machine, with no ablility to recall and edit and re-use an equation or expression, without turning it inside out in RPN? (The 17bii gives you the same feeling only better---editable, long variables, fully algebraic solver, returns the values to the stack...) Afer the 33s, with its 32 kb (8 kb really when efficiency isconsidered) do you really think there is a reason to spend real money on a scientific calculator with less than 500 bytes? After RPL, where you can evaluate an algebraic, DUP it, operate on it with stack logic, save it, edit it, merge it with other algebraics, re-use it, etc, is there really any reason for RPN? Afer you have tried all of these machines, seen what is possible, enjoyed the absolutely beautiful design of the 15c, savored its remarkable power, acclimated to and internalized its keycodes...... Now, think logically, clearly, and efficiently for the future....why would you spend real money on a new device (without true collector value both nostalgic and monetary) with the limitations of the 15c? The future is not RPN. But a stack could (and should even) be a part of the future. Algebraic objects should be normal standard fare---and they should be editable. Memory should be plentiful. Variables and registers, too. I really like the 15c. I enjoy it. I use it regularly. I greatly appreciate Valentin's challenges, Karl's tips---- because I have one, and know it, I will use it. I like the user mode---I put what I use frequently under labels a through E and have direct access--its great! But I can see that it is antiquated--a bit obtuse if you don't already know it......there is no reason to really produce it anymore in real nuimbers....you can do so much better! Why should a person learn math, and then have to go learn a new way of writing math? Why shouldn't the rules of precedence, the notation, etc be the same as it is for maths? Why should a user have to be molded by the arbitrary limits of the machine's architecture, when it is perfectly possible with available technology, to develop an interface which is consistent with mathematical notation? (I will note that most "algebraic" half-breeds (not good formula machines) are even worse when it comes to distortion--and the "4 function" types are abominable--as are old Ti's where clearing or overtyping after an incoorect operator will give unexpected results....at least RPN is consistent.) Rather than bringing back the 15c, I think we should push, through openRPN, to develop a truly powerful machine, which is capable of working in the new paradigm. Best regards, Bill

Re: Why the 15c should not come back Message #2 Posted by Ron Ross on 21 July 2004, 3:59 p.m., in response to message #1 by bill platt All good points, and I agree that the 15c unchanged and unimproved would suffer from the failings you mention, buuuut: Like the HP33s, toss in gobs of cheap RAM. Give it a dot matrix display and basically beef it up to an Hp42s in a 15c housing. Toss in a cheap USB interface and file transfer. Give it some units conversions. And most of all keep it at its present size. Price it just a bit below the Hp48Gii or same price as the Hp 17Bii+ (ie retail approx $99 US or even better, less). Now you have a real pocket calculator. I would buy a couple. Would it have shortcomings, yes. But they would be shortcomings due to choice and technology, not bean counting and gimickery to appease market share or to not rob from Hp's precious graphics calculator sales. Dot matrix LCD's aren't as nice and crisp to view as the segmented display, but gives you lots more options. Toss in an alpha toggle key and a two line display like the 42s with USB and a SDK to develop programs on the PC and you eliminate concern about the awkward keyboard. Many people complain about this on the 42s, but if there were a serial port, that would eliminate most complaints. Just me Ranting and Raving, Again. I want a pocket calculator, not a graphics monster.

Re: Why the 15c should not come back Message #3 Posted by marais on 21 July 2004, 4:40 p.m., in response to message #2 by Ron Ross That's exactly my point. What we like most about the 15C are its form factor, its highly usable keyboard and display, and most of all its conceptual integrity. But IT IS true that keycodes are somewhat outdated, and at the time when the 15C was issued, were my primary reason to go for a casio 602p instead - programmability is so much simpler with readable ASCII op-opcodes. Well, in the meantime I got two 15C (along with a 2nd 602p), and as a professional EE I also admit that programmability has become less important than built-in matrix and complex numbers cability, but anyway, my point is: An improved 15C with dot matrix display (with equal contrast of course) and somewhat more memory would be the perfect machine for any practicing engineer. Students wouldn't go for it, but nobody said we need one calc for everybody. BTW, I remember having seen a mockup of an improved 15C (with the golden faceplate taken from the 12c) somewhere on the internet, meant as an april fools joke - it actually came pretty close to what I imagine as my dream machine. Anybody remember that link?

Re: Why the 15c should not come back Message #4 Posted by bill platt on 21 July 2004, 5:01 p.m., in response to message #1 by bill platt There is no reason to assume that I am suggesting that the form factor should grow from the original 15c target. There is also no reason to assume that because I seem to favor RPL, that it should be graphing. In fact, I use my RPL 48Gx far less than my regular machines--because the functions are "buried". I like the RPL stack, the Algebraic power, but find the lack of convenient access to many functions a drag. RPL can be applied to a 2-line, non-graphing scientific. RPL could easily be modified to allow for stack depth control, or ability to load a constant, or even some block of standard memory "registers." I find myself using the 30s a lot now, though---and that is my point----if you bother to step outside our traditional RPN world, you will discover that there is so much more possible, right now, with available technology, that will make you never want to be pure RPN ever again. Until you have tried it, given it a chance, you really cannot appreciate (I didn't!) the power of being able to put in algebraic expressions, edit them, find them again, re-use them etc. Why shouldn't a modern, relatively expensive machine enable this? By way of comparison, I immediately saw the value of the equation list in my 32sii, when it replaced my 11c. I started to write equations instead of programs, for many tasks. It was easier to see what the "program" was for, when it is an equation. This was in 1995. And there was no editing allowed, and it still made sense! Then I found the 48 series--and was frustrated by too much horsepower and many simple things buried. But Algebraics a big plus---and if only my head were more pliable, the programming style might have been good, too! Then, I came across the 17bii---and the equation solver there blows the 32sii away! When I saw that the 30s allowed editing of an algebraic expression, and saved the last so many expressions, I became even more fascinated by the possibilities. Note that to a Sharp or Casio user, my discoveries are ridiculously old---like, all this was possible 15 years ago! For a good current screen, take a look at the 30s (minus the glary plastic over-screen). You have two lines, dot-matrix upper, segmented lower, big characters, adjustable display, very good contrast, and provision for alphanumerics. Much nicer than the 33s---better by far than the 48. What I am trying to do is to stop "copping out" with "I would be just fine with such and so." While that might work for us old folks, the fact is that the next generation should not be saddled with new products which are not as good as possible. (This is what bothers me most about the 33s---like, why not make the equation list editable!!!?). regards, Bill

Re: Why the 15c should not come back Message #5 Posted by Ron Ross on 21 July 2004, 5:12 p.m., in response to message #4 by bill platt And that is why I think the 27s is the 2nd best pocket calculator ever made. I prefer RPN, but the solver and the amount of memory available to the 27s is still unmatched by any other scientific pocket calculator except the 42s. If it had an RPN mode, it would have been a great favorite among the RPN crowd due to its simple to use solver (the very same as the 17Bii that you mention). It is just so simple to use, and that is what makes it both easy and powerful.

Re: Why the 15c should not come back Message #6 Posted by JDH on 21 July 2004, 7:42 p.m., in response to message #5 by Ron Ross I own the HP 48gx, 42s, 17bII, 32sII, 15c, 12c and I have to say the 15c is still my day to day favourite, actually purchased a backup for the day when my original stops working. As others have said before, I can't see the same model going back into production but something of similar form factor with extra memory and two line dot-matrix display would be good.

32S/42S/48SX "replacing" 11C/15C/41CX Message #7 Posted by Karl Schneider on 22 July 2004, 2:21 a.m., in response to message #4 by bill platt Bill -- Interesting "short essays"; I have a few points: I agree: Owing to their limitations, the 15C (and 11C) probably would be not very successful in today's marketplace. As a comparison, the 1966 Ford Mustang is a classic design that remains common and relatively popular even today. However, even if it met modern emissions and safety standards, it wouldn't sell well as a re-released car. It is antiquated -- more so than the 15C -- in many respects. Quote: When I saw that the 30s allowed editing of an algebraic expression, and saved the last so many expressions, I became even more fascinated by the possibilities. Note that to a Sharp or Casio user, my discoveries are ridiculously old---like, all this was possible 15 years ago! Ah, but the 27S and 17B with the solver of algebraic expressions are also 15 years old! Quote: For a good current screen, take a look at the 30s (minus the glary plastic over-screen). You have two lines, dot-matrix upper, segmented lower, big characters, adjustable display, very good contrast, and provision for alphanumerics. Much nicer than the 33s---better by far than the 48. No argument here with the displays of the 33S and the 48-series. I've toyed with the 30S (two collegues have them, one to replace his employer-issued 28S that finally broke), and I don't share your enthusiasm for it. I find the digits too narrow in proportion to height -- Voyagers, and all Pioneers without the two-line display are more legible. I also don't find its "pure algebraic" system (shared by Casio and TI) very much of a help: To perform a calculation on a previous result, you do [shift][Ans]...[=]. Heck, it was easier with RPN and old "hybrid algebraic". Proofreading a long, parenthesis-laden expression isn't simple, either -- you can't see the whole expression; braces and brackets can't be used to distinguish pairs.

Both formula AND control structures programming Message #8 Posted by Vincent Weber on 22 July 2004, 4:56 a.m., in response to message #7 by Karl Schneider Hi all, Just to add to these very good comments, one thing that strikes me is the very small amount of machines that can do both formula programming AND have control structures for looping and branching. To my knowledge, the only non-graphing machines that falls into this category were: 1) The fx-4000P from Casio, made in... 1985 ! 2) The 32SII from HP, only made in 1991. And yet, this machine does not have editing capabilities... 3) The 67 Galaxy, made in 1992 only. There were some descendants of these machines, e.g. the Casio fx-4500P or the HP-33S, but still, the offer was quite limited. It is a pity, as I feel that formula programming only is too limited, while keystroke programming only is a pain to deal with complex equations. Programmable calculators seem to be out of fashion... That makes me think that Casio is sometimes rated unfairly. Think that, back in 1981, the fx-602P was a very good alternative to the HP-41, cheaper and faster; than as early as 1985, Casio invented both the formula + structure programmable with the FX-4000P AND the graphing calculator with the FX-7000G. I was a high school studend in 1988 and Casio was at the time the vendor of choice for us (I had a FX-8000G which I still like today). Only latter in 1990 I discovered HP with the fabulous HP-48SX. The fx-4000P was a wonderful nice little machine, with equation editing and replay, 550 steps memory, 96 registers with indirect addressing, and was even smaller than the HP-15C... Those were the days. I do agree that formula programming is missed on some HP wonderful machines, such as the 41C or the 15C. The ACE ROM does some effort for the 41C with 'PROG', but no editing, no replay, no integration with solver... HP has yet to produce something like the 4000P, inegrating the best of Casio (equation editing) and HP (infinite number of alphanumeric labels, multi-variable solver...). That would be a hit ! Long live Casio ! :) Cheers, Vincent

Re: Both formula AND control structures programming Message #9 Posted by Ron Ross on 22 July 2004, 10:46 a.m., in response to message #8 by Vincent Weber Hp did indeed come out with that calculator. My wife had a Casio fx 4000 and really liked it. She later switched over to the Hp 27s. It is actually better than the Casio for everything except fitting in your pocket (the 27s fits in your pocket, just not as nicely).

Re: Both formula AND control structures programming Message #10 Posted by Vincent Weber on 22 July 2004, 6:10 p.m., in response to message #9 by Ron Ross May I respectfully disagree - the HP27S is a wonderful machine which I respect a lot, but is -not- a true programmable... IF and SOLVE are nice functions, but do not allow everything that the keystrokes programmable can do... My point is, there are a lot of keystroke programmable calculators out there, and there are tons of formula programmable too. But there are very few calcs than can do both ! Cheers, Vincent

Re: Both formula AND control structures programming Message #11 Posted by Holger S. on 23 July 2004, 4:18 a.m., in response to message #8 by Vincent Weber I have both the FX-4000P and the FX-4500P as well as a TI-67 Galaxy. I like the 4000P. The 4500 has a cramped keyboard, even worse than a 32SII. Pretty much impossible to find anything quickly, athough it looks interesting. The TI has a nice form factor, but IMO the process to store and recall equations is cumbersome. Also, its display is hard to read with a decimal point too small to notice. The disadvantage of all three calculators and the reason I don't use them regularly is the algebraic input. Typing [SQRT] Ans Exe just to calculate the root of what's in the display isn't very intuitive. As for programming, I think it depends much on what kind of program you like to write. Formula-like problems are easier to implement, although I think keystroke programming is easier to learn. However, more complex branching is easier with keystroke programming (however it's even easier with a real programming language like Basic). I like the Casios, my first programmable was a 602P, too. And the 4000P and 4500P are dirt cheap nowadays. I bought mine on ebay for 2 and 10 Euros, respectively. But the most usable calc I have is without doubt the 42S while the most beautiful and fascinating is the 41CX. Cheers, Holger

Re: Both formula AND control structures programming Message #12 Posted by Vincent Weber on 23 July 2004, 11:13 a.m., in response to message #11 by Holger S. Hi Holger, I largly agree with your comments. I am fascinated with the 41CX and I love the 42S too, but I still think that both lack an equation mode a la 32SII, with editing capabilities. Nothing is perfect: The 32SII really shines on some points (Equations, fraction mode, ease of use, speed) but is a real move backwards in some others (cramp memory, primitive complex numbers, lack of labels and registers). My dream would be a 42S + equation mode with editing + proper alphanumeric keyboard + franction mode + USB i/o. That would be my dream calc... I also think I forgot some interesting SHARP non-basic models in the very small club of non-graphing formula + control structure calculators. So, I think the complete list of such calcs are: - Casio: fx-3900P, fx-4000P, fx-4500P, fx-4800P - HP: 32SII, 33S - TI: 67 Galaxy - Sharp: EL-5050 (very nice dual keyboard, I would love to own one; the EL-5150 is horizontal version, but is a large a BASIC computer), EL-512H & EL-512II (more limited), and EL-5120 (too modern for my taste). That's a total of 12 machines in the whole calculator history, out of which only 4 (fx-4000P, 32SII, 67 Galaxy and EL-5050) are really interesting IMHO. Something the market should re-explore ! Cheers, Vincent

Re: Both formula AND control structures programming Message #13 Posted by John Smitherman on 23 July 2004, 2:28 p.m., in response to message #12 by Vincent Weber I am curious - posters to this forum discuss in detail calculators from HP and often mention calculators from TI, Sharp and Casio. Are there currently any European companies that produce calculators? John

EL-5020 Message #14 Posted by bill platt on 26 July 2004, 2:39 p.m., in response to message #12 by Vincent Weber Hi Holger, You mention an EL-5050, 5150, and 5120. Do you know anything about the EL-5020? Especially, where to find a (or copy of) manual? Best regards, Bill

Re: Why the 15c should not come back Message #15 Posted by Bill Smith on 22 July 2004, 12:05 p.m., in response to message #1 by bill platt I disagree. I don't have a 42S, but based on its features, it is the only calculator that might surpass the 15C as a pocket calcultor for this old engineer. I had a 32SII at work for a while, but never liked it as well. In the day of the PC, my use of a calculator is perfectly met by the limited programability of the 15C. Its size, form, quality, durability, efficiency, and RPN are for me simply what a calculator is all about. There are simply too many tools that are just plain better at most of the other aspects of data aquisition, analysis, number crunching, and presentation than a calculator. Certainly other folks have different needs than I, but if my 12C, 15C, 41CV, 41CX, (many 41 modules) and 71B all were stolen or destroyed, the only replacement I could justify on the basis of need would be the 15C. The future may not be RPN, but for the next 5 or 10 years, the best calc tool for my job is the 15C. Regards, Bill S.

Re: Why the 15c should not come back Message #16 Posted by OpenRPN on 22 July 2004, 11:32 p.m., in response to message #1 by bill platt I certainly believe OpenRPN will produce the great calculators of the future. When I set out to found the project my goal was to make the calculator that you take everywhere you go, that does everything *you* need of it to do. Beyond that it must be the most durable calculator ever (if I can pull it off I would love to design it to be run over by a truck!). Ultimately, I think everyone will be impressed as soon as the first prototypes start to appear. Thanks to all for your support. -Hugh

[ Return to Index | Top of Index ]

Go back to the main exhibit hall