On 01/03/2014 02:19 PM, dwight wrote: How could they leave out the HP-35, HP-65, *and* HP-67?

It is funny, once I saw how clumsy an algebraiccalculator was to use for more complex problems,I found the RPN entry to make more sense.When you actually work the problems, you find thatit matches the order that you'd enter the values ona HP calculator.Doing complex problem used to be either a lotof memory buttons or (()) buttons just to match theequations.It goes something like this:Take these two numbers and add them together.Then take these three numbers and add them together.Now multiply the results together.As compared to:(3 +4 +5) * ( 2+7 )Dwight

---------------------------------------- Sorry about the format ( Outlook ) Order of operations becomes much more important as things become more abstract. It is hard to describe in simples math problems. Being able to tell the computer what to do and in what exact order you expect it to be done can make quite a difference when creating correct code. In most all case, when one abstracts operations, one tends to order them in the sequence they are expected to be done. Even in languages that have algebraic notation for math. Crossing between the two is clumsy. Dwight

On 01/05/2014 10:01 AM, Liam Proven wrote: I use an RPM calculator and have for decades. My big gripe is that my HP-16 doesn't tell me how deep the stack currently is and doesn't err out when I try to pop the stack beyond what's been pushed. Nor do they tell me when I've tried to push something beyond the T register. Of course, infix calculators usually don't tell you that the cell that you're recalling doesn't contain anything. Either deficiency could have been solved with just a simple flag that distinguishes between a register/cell that's empty or one that contains data. Has this been remedied in any of the newer RPN calculators? --Chuck

On 01/05/2014 12:40 PM, Tony Duell wrote: That's it--I've been tripped up more than once by either leaving something on that stack that I shouldn't, or taking too much off of it. I just tried this not only on my HP-16C, but also my wife's HP-15C. both Voyagers. Same result--stack underflow just repeats the last value "popped" from the stack. A shame that there weren't a few dots or a digit on the display that indicated what's on the stack (or not on it). While I use my calculator for a lot of quick-and-dirty tasks, I almost never use its programmable feature. There are computers all over the place for that. --Chuck

On Jan 5, 2014, at 10:01 , Liam Proven <lproven at gmail.com> wrote: They are also good for engineers, or anybody else who needs to calculate values of nontrivial nested equations. I've found it to take less mental "stack space" to enter a complex equation on an RPN calculator than an algebraic one, since there's no need to think ahead to count how many times I need to pound the parenthesis keys on an RPN calculator. Having used both varieties heavily (mostly HP and Casio), I strongly prefer the RPN type. My first calculator as a little kid was an RPN 4-banger, though I didn't appreciate the RPN vs. algebraic distinction until much later, and most of my calculators were algebraic until I got to college and started the heavy engineering courses. It was a Montgomery Ward model, and Google suggests it was probably the model P10 which was a rebranded Novus 650 "Mathbox". http://spyropoulos.net/calcs-other/WardsP10.jpg When I got my HP28S in college, I wasn't sure I was going to like that silly RPN stuff (and I probably didn't recall that my first calculator was RPN). I found it to be much easier to use than my previous algebraic calculators for the sorts of electrical engineering equations I needed to calculate. Of course, its more advanced features such as handling complex numbers and numerically solving for unknowns in systems of equations were also quite helpful, but RPN entry turned out to be a lot easier even for plug-n-chug calculations. I still have my old HP28S, but not long ago I finally upgraded to an HP48gx. I wasn't able to find any current=production RPN scientific calculators that had form and fit that appealed to me, so I use that vintage 48gx even though it's pretty slow by today's standards. I have a 48gx emulator that I use on my iPhone and laptop when my real calculator isn't handy, but I prefer the physical keyboard on a good dedicated calculator. One of the big plusses of my previous Casio calculator was its handling of fractions, but I still prefer RPN entry. I've written some programs on my HP48gx to handle fraction manipulation when dealing with silly imperial units like inches. They're not quite as clean as the fractional math on my old long-lost Casio. If you need to handle a lot of non-decimal fractions, such as when doing woodworking with imperial tools and measurements, and you rarely need to calculate equations with more than a couple levels of nested parentheses, then one of those old Casio calculators with a fraction key would be quite handy. When the parentheses get nested deeper, RPN really shines. I wonder if my early exposure to an RPN calculator contributed to my easy acceptance of RPN entry and stack-based languages like Postscript? Or maybe it's just a matter of how my brain is wired? I can imagine that some folks may have a strong preference for algebraic or RPN entry based on how their brains happen to be wired. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

On Sun, Jan 05, 2014 at 11:22:02AM -0800, Mark J. Blair wrote: [RPN calculators] This might just be Stockholm Syndrome affecting your judgement :) FWIW, if I want to do a nontrivial calculation, I will generally fire up Python's REPL and build up the expression piece by piece, inspecting the answer at each stage to make sure it's plausible, and editing it using the command history to produce the final answer.

I had almost the reverse experience. It took me a vouple of hours at most to become familiar with RPN on my HP41 all those years ago, and my first thought when U'd done so was 'why is normal algebraic notation so clumsy?; It's certainly not how I think. I do not think of the entire expresion and then evaluate it. I will eb thining 'I need to calcuate the votlage here, then this oen is 3/2 times as much, Sp the curent throug that rsistor is ... and so on. Something thati s much easier to do on an RPN machine than an infix one. Actually, I find it very unnatural to read an algebraic expression left to right. Isubconciously look at the highest priority term first -- look at hte deepest level of parentheses, etc. Whcih means I actually find it easier to enter such an expression in RPN than on an infix machine. -tony

Hmm... Everybody I know who has used an HP enver wants to use anything else. There are a couple of reasons for this : 1) RPN (which includes RPL, of course) does what you expect -- every time. You do now have to worry aobut precendence rules. An operator/function is executed when you press it. There is no ambiguity 2) HP calcualtors do nto do 'funny rounding'. Casios often do. What I mean by that is that some machiens roudn the answer to give what the machine thinks you want. A quick test of this. On a machine with no symbollic capabilites, select radians mode and calculate SIN(pi). If you get 0, reject the machine. The point is that the 3.1415926... number shown on the display is NOT pi. It's a truncated or rounded version of it. And SIN of that number is not zero. There is no justification fo the machien t oassume that this number continues as pi. Evey HP I've used gives a small non-zero value for this. FWIW, having used an RPN machine for many years I find infix calculators very hard ot use. I had an HP41 at school in the early 9180s. Such things were not common, but I was certainly not the only person at school to have one. There was the odd HP32E/33E/34C around too. There were rather fewer TI59s than HPs. -tony

On 3 Jan 2014, at 22:19, dwight <dkelvey at hotmail.com> wrote: The Register is a British news source. It's reporting on what was popular in the UK.