|All good things ...|
Message #21 Posted by Valentin Albillo on 29 Oct 2003, 10:21 a.m.,
in response to message #17 by Norm
... must come to and end.
This must be so as well for this very interesting and polite thread (a real example of people having nearly opposite points of view, but nevertheless managing to discuss them without going for each other's throats, a real rarity in forums but not so in this one).
Hi Norm & everyone following this thread.
I've read your long and interesting reply with all due attention and though the thread must finish now, I want to comment a few points I feel you may have misunderstood, probably because of me not explaining myself with total clarity (a typical problem we non-native English speakers frequently must deal with). Let's see:
"Your view that "RPN is obsolete" however is [...] unpopular with the types on this chat board"
I know, I know, but just the same ...
"there are an increasing number of posters here who are thinking like you...... who aren't loyal to RPN but like the latest "cereal box" trinkets from "KINPO"
Not me. My HP collection does not include anything but classic, worthwhile models. No China-made junk (save for an HP-12C that I promptly got rid of), and no RPL machines at all (save an HP-28S, for historical reasons). HP48s and 49s have no place at all next to my 15C's, 41CX's, 71B's, 32S's and 42S's, among others, not forgetting my beloved 24+ year old HP-34C. I wouldn't touch a KINPO model with a ten feet pole.
"HP (and some like Valentin) are enamored with the view that a calculator is now a "COMPUTER IN A CALCULATOR" (the graphing calculator craze). I don't agree."
Neither do I. I have not a single graphing calculator, from any brand name. The fact that many of my SHARPS do have large screens which are pixel-addressable (and thus can show graphics on them) doesn't mean they are 'graphing calculators', at all. If you want some function graphed on the display, you'd better write a program to do it, pixel by pixel. The large screen is nonetheless very useful to enter and edit programs, and for hand calculations and sometimes graphics come handy, such as for my Othello program, with the board always displayed on screen.
"Giving me a cheezy, terribly low-res graph on some 100 x 150 pixel (?) screen on a calculator is (a) useless
(b) interferes with me getting my work done"
For the typical, student-oriented graphing calculator, you're probably right. But for the scientific calculator which happens to have a large screen, as mentioned above, it's just the opposite. It greatly facilitates your work, as you can see previous calculations and results still in the display (till they scroll out), and you get to see more program lines at a time, and longer program lines, without scrolling. One of the worst flaws of the otherwise excellent HP-71B was that its very small, 1x22 character display made a chore of entering and editing any programs. A 4x24 display, on the other hand, is extremely convenient. Even the small, 2-line display of the HP42S is a great help over one line displays.
"Let me elaborate..... my chosen engineering skill is analog electronic design [ ... ] I also stand by that there is no reason that a 32Sii wont function 60 years from now [ ... ] I am the EE, I win you lose."
Nope. Though I work as a Senior Systems Engineer right now, I'm also an EE ... and anyway, time will pass by, and we will
eventually see who was right. My best advice would be: use your 34C and enjoy it as much as you can, even if that means it will wear out. Keeping it in a safe, avoiding using it to make it last is futile, and in the end it will die by itself and you'll be left with a nonworking machine that you never fully used like it deserved to.
"WELL, even if that was the original root cause of
RPN, it's NOT THE REASON FOR IT. EASE OF USE IS THE REASON"
That's the problem, Norm. Just stating your point in capital letters (as if shouting) won't make it right. As I see it, the problem is you seem to never have given any other machines a chance, at all. You just decided at some past date that RPN was it, and there you are. Never questioning it. Never really comparing it against more modern approaches. Never really taking some good non-RPN machine and invest the time to get to know it and see its good points.
That being so, seems to me you're just rooting in what you know, never trying out new things and gaining experience. Me, I am not a TI-junkie or otherwise, sadly ignorant of The-RPN-Revelation. On the contrary, I've been into RPN all my life, bought every classic RPN machine as they were being released, and have written thousands and thousands and more thousands of lines of highly-optimized RPN code, and still do. I have many articles (+programs) published in PPC, Technical Notes, Datafile, a couple books by HP (one for the 34C), etc. etc, so I know RPN pros and cons inside out.
But I'll never let myself get blinded or limited by a single point of view, however right it may seem, and I got to know and experience other machines, and of course, tested them against RPN ones.
You know the results. I still like very much RPN, still write RPN programs and articles, and will do so in the near future, but I know (because I've tested it) that RPN can't beat the best algebraic systems neither in ease of use, nor in speed, nor in convenience. It may do for you, but not for me. I can evaluate and reevaluate any cumbersome expression (not "formula", as you seem to understand this term as a symbolic formula)
in my SHARP PC-1350, say, long before you've had a chance to even decide on where to begin with.
But don't trust me, do the test yoursef, pitting two persons, one versed in RPN, the other versed in some powerful algebraic model, and give them some complex expression they've never seen before. Provided they're both proficient with their respective models, the non-RPN guy will win hands down. A reevaluation changing some value would be a killer for the RPN guy, as the other fellow would just bring the whole expression to the display with a single keystroke, then place the cursor over the value to change and press ENTER, and there you are, while the RPN fellow would have to reenter all numbers (even the ones not changed) and operators from scratch, as if he were evaluating it for the first time. Even using registers to store parameters wouldn't save the day for him, unless he would resort to try and write a program on the spot, which the non-RPN guy wouldn't need to. Try it !
"You advocate an "algebraic calculator" but this is so wrong it aint funny!! Why? Because you have to start an
entire new step..... type in the formula as an abstract!!!!!!!"
You misunderstood something here. Seems to me you're thinking I'm advocating graphing calculators, where you must enter symbolic expressions with some kind of equation editor and such. That's not correct. I completely dislike graphing calculators and symbolic equations editors, never use them and never would. The calculators I advocate are the ones where you can enter:
X=3.0013431 , (X/EXP 1)^X*SQRT(2*PI*X)*(1+1/(12*X)+1/(288*X^2)-139/(51840*X^3))
and press [ENTER] and you'll get the result instantly evaluated on screen (while the expression itself still remains on-screen, above, ready for re-evaluation). No tedium and no burden, absolutely straightforward.
"It's about whether you are going to "BELIEVE" in RPN or not."
For a lot of people, yourself probably included, this is indeed kind of an 'emotional' issue, but not for me. There are issues that do require some faith into the proceedings, but this isn't one of them. This is a matter that can be tested, and rational reasons can be given and discussed. The problem is lots of people *won't discuss it* (lest they might risk losing their 'faith' in RPN, which they dare not), *won't give other machines any chance whatsoever* (after all, not being RPN they can't be any good), and above all, *they won't perform any fair tests, at all* (why testing, when RPN's superiority is so obvious ?). This might be an acceptable attitude for most people, in this issue and in other more important real-life issues, but it's not an acceptable position for me.
"OK, well, there's no flames here, no war here, just talk, and fast typists. And smiley faces :o) [ ... ] and with that final thought to logically win out over your stance, I wish you well and hope the time to trade this stuff was not too burdensome."
I wish the same for you, Norm. I've found this exchange of opinions most interesting, still think I must thank you for your consideration and patience with my 'heretic' :-) points of view. If you post any reply, I'll read it with interest, but I won't add further to this thread. In any case, rest assured that I like RPN a lot, and you'll be hard pressed to find any other person who has devoted so much time (and money) to it and its HP hardware implementations. That's a whole lot more of a chance than most RPN people would give to any non-RPN systems.
Best regards from V.