▼
Posts: 2,247
Threads: 200
Joined: Jun 2005
Hi All,
Now that my collection of vintage HP calculator (as well as graphing HP, TI, and Casio machines) is growing again, I feel compelled to program.play with them to justify their presence. Of course using these old limited machines is no match with today's PCs (even the smallest ones). Yet I feel that working with the limited vintage machine challenges us to program in a lean way. The temptation today is to justify "wasteful" programming, such as extra arrays that store duplicate data that we justify for whatever reasons. And that's just the tip of the iceberg.
Do you feel that programming vintage HP, TI, Casio, Sharp, and other machines is a total, partial, nostalgic waste of time, or a nice reminder and exercise in lean programming?
Namir
▼
Posts: 325
Threads: 18
Joined: Jul 2006
Quote:
Do you feel that programming vintage HP, TI, Casio, Sharp, and other machines is a total, partial, nostalgic waste of time, or a nice reminder and exercise in lean programming?
Yes. :-)
I enjoy programming the vintage machines, but I also enjoy programming newer machines (and computers, for that matter). But if I need to get something done, I'll use the computer (or at least a 50g).
My first calculator had 49 steps of programming memory, and I thought it was wonderful. I also remember when a big computer had 64K of memory and we managed to get real work done on them. Now as back then, I enjoy the challenge of fitting programs into limited memory.
Anyone else?
Posts: 528
Threads: 40
Joined: Dec 2008
The calculator imposes three (or more) constraints. Your code most be small, your data must be small, and your algorithms must be fast. All three provide intellectual challenges that can help you to think creatively so in that sense they're all useful traits. But lets compare them to "programming in the real world."
Small code doesn't matter very much any more, unless you're dealing with embedded systems or other small devices (like calculators!). Or maybe I'm just showing my own bias - I program for a large distributed system.
Small data can be important if there is a lot of it.
Fast code is definitely important. A bad algorithm can eat up all the CPU power available today, tomorrow, and for thousands of years to come.
Dave
Posts: 472
Threads: 58
Joined: Apr 2008
I wondered this when I first started collecting. Why would anyone waste their time on old calculators? Well... why would anyone waste their time building furniture with hand tools rather than power tools? Why would anyone waste their time restoring old furniture? Why would anyone want to restore and drive an old car or an old airplane? Why restore mechanical watches or clocks? Why would anyone stick with an older spouse? (Better not go there...)
In addition to the already mentioned benefits, it's simply a love and a passion for some of us-- me, anyway. I know that even college-age programmers get a kick out of learning to use older equipment. And it does also help teach the value of concise and elegant solutions. And I still feel a bit strange to think about the thousands of years of scientists, up to and including Einstein, who would have given anything for a "lowly" HP-35...
▼
Posts: 69
Threads: 6
Joined: Aug 2010
Posts: 562
Threads: 27
Joined: Oct 2006
I think it's tremendously useful.. Just a few months ago I had to write a program to calculate averages and standard deviations. Rather than using a 200 element array and computing the stdev and average separately like the rest of the class, I just used a three element array and accumulated n,x,x^2 then at the very end calculated the stdev and avg. The idea was inspired by the way my first 32sii calculator could average 2 numbers or 50,000 using the same RAM.
i ♥ calculator programming.
Another example, Polynomial multiplication and Discrete Time convolution. Through programming, I realized these are the exact same processes. In that case I arrived at the same answer with different approaches, and learned a math lesson besides.
The skill of writing a 22-byte program is indispensable, and missing from modern computer science.
Posts: 1,248
Threads: 33
Joined: Aug 2007
While I don't have any "vintage" HP's (my earliest are Pioneers), the programs I write for them (I include Solver routines) are practical working tools in my profession, civil engineering. Not an academic exercise. No one disputes computers are more powerful. But having the convenience of a handheld on the desk/conf. room table for quick problem solving - can't be beat.
Posts: 1,792
Threads: 62
Joined: Jan 2005
Quote:
Yet I feel that working with the limited vintage machine challenges us to program in a lean way...
Do you feel that programming vintage HP, TI, Casio, Sharp, and other machines is a total, partial, nostalgic waste of time, or a nice reminder and exercise in lean programming?
Loaded question, Namir! ;^)
Seriously, a handheld calc is the wrong tool for a big job (see David Hayden's criteria above), but I still prefer to program a vintage HP for straightforward tasks instead of struggling with MS Excel on an PC.
-- KS
Posts: 556
Threads: 9
Joined: Jul 2007
total waste of time unless you do it for self satisfaction instead of solving sudoku or alike
▼
Posts: 528
Threads: 40
Joined: Dec 2008
Quote:
total waste of time unless you do it for self satisfaction instead of solving sudoku or alike
Why not do both ? :)
Posts: 2,247
Threads: 200
Joined: Jun 2005
If you want to experience the speed or slowness of an algorithm, then implement it on a programmable calculator or even pocket BASIC machine.
|