[VA] Short & Sweet Math Challenges #23: "May the 4th Be With You !" Special

[Valentin Albillo]

.
Hi, J-F Garnier and cortopar:

(05-12-2018 08:37 AM)J-F Garnier Wrote:  Valentin, your solution of the Step the Third is really great!

Thank you very much, I'm glad you liked it.

Quote:I didn't known that it was possible to evaluate long expressions in that way.
I learnt something today (yes, it's still possible to learn new things, after more than 30 years of HP71 usage), thanks again:

You're welcome. The fact that you learned something new is proof enough that you're stilll pretty young, you know what they say about old dogs not being able to learn new tricks. :-D

Quote:How did you come to discover this possibility? By chance or rational exploration?

I discovered that VAL can evaluate long expressions soon after I got me a pretty (and expensive!) HP-71B complete with HP-IL ROM, Forth/Assembler ROM and all the IDS volumes.

Afterwards, when I also got the Math ROM back inf 1985 (without its IDS, regrettably) I did test the limits of most of its functions and empirically discovered that the funny functions FNROOT and INTEGRAL were akin to VAL in that regard. I took notice of those facts at the time and then it was just a matter of finding the right occasion to share the knowledge.

Quote:Of course, it is only possible thanks to the VAL function of the HP71, that is actually an expression evaluation function, rather than the classic Basic VAL function that can only convert a number from its ASCII representation. For instance, the solution is not applicable to the HP75.

Indeed, I've found no other version of VAL (or its equivalent in other languages) which can do that. The HP-85/86/87 couldn't either, neither could the HP9816/26/36 or the various versions of Pascal or Visual This/Visual That, etc.

Quote:The HP71 design team did a great job at the time.

They did 95% great things and a few % not that great. For instance, I will never forgive the <expletive> who decided the inclusion of that abomination called "CALC mode", which utterly wasted 5 Kb of the 64 Kb ROM which could've been put to much, much better use.

Or the <milder expletive> who implemented string handling in such a way that the system must have the whole string on the stack to do anything with it: say you need to extract or change the Nth character of a long string, the operating system cannot simply do it using pointer/address manipulations, no, it must copy the whole string to the stack, using lots of valuable RAM and time and thus making using long strings inefficient and slow. And there are more (meager string functions, missing complex functions, poor PEEK/POKE implementation, missing matrix functions in the Math ROM, etc) ...

cortopar Wrote:I bet it doesn’t feel like it, but even if this thread only had two posts, your original challenge and your answers, it would be tremendously valuable to many of us who observe more than interact.

Thank you very much, Bob, you're far too kind and I'm really glad you appreciate my S&SMCs.

Quote:Your posts over the years are about 98% of the reason for my interest in the 71b. Thanks for continuing to carry the 71b torch, and thanks for all your posts on other models that have added to my RPN and math knowledge.

Thanks again. Though I tend to use the HP-71B more than other models, for obvious reasons, I do like and admire a lot the classic RPN ones and I have scores of programs written by myself for the HP-11C, HP-15C, HP-25, HP-34C, HP-67/97, HP-41C, HP42S and even the HP35s, most of them unpublished.

I intend to scan or type them anew as soon as I can find some way to put them online, which I'm actually looking into right now.

Best regards.
V.
.