HP Forum Archive 19

[ Return to Index | Top of Index ]

A tribute to the old manuals (again :) Message #1 Posted by Geir Isene on 14 June 2009, 4:07 a.m. Working on a really big statistical analysis program based on the Multiple Curve Fitting by Kolb, I am including analysis on moments, kurtosis and skewness. Having searched for a while on the net, I did'nt find the simple formulas I needed for my program. I remembered that I did indeed see the formulas presented somewhere. I checked the HP-41 stat pac manual - and there it was. This is but one such example of easy formula presentation in our old trusted manuals. The HP-65, HP-67/97 and the HP-41 manuals have helped me out like this many times. Just letting someone know my sentiments here :-)

Re: A tribute to the old manuals (again :) Message #2 Posted by Mad Dog ebaycalcnut on 14 June 2009, 6:21 a.m., in response to message #1 by Geir Isene People used to have to think before using computers/calculators and thus spend a little time simplifying formulae analytically before touching a keypad or keyboard due to RAM or CPU constraints. Nowadays, as powerful as computers are, they can be sloppy with code, and it doesn't matter. It's a shame. Also, people just used canned stuff and thus need to think even less.. Edited: 14 June 2009, 6:23 a.m.

Re: A tribute to the old manuals (again :) Message #3 Posted by Mark Edmonds on 14 June 2009, 9:09 a.m., in response to message #2 by Mad Dog ebaycalcnut I agree that losing the understanding of the basics is a sad and undesirable problem. Ultimately though, the consequence that people write sloppy code and get away with it due to sheer CPU/RAM horsepower is more of an academic problem than a functional problem these days. Programmers aren't paid to write beautiful code. They are paid to write code that gets the results even if it does it in the most ugly manner possible. The customer buys the result, not the method. Additionally, it can also be argued that with so many ready-made solutions available these days, people can tackle greater problems because they spend less time working out the smaller steps. So even if the tools are potentially more inefficient, even if the quality of engineering is reduced and even if the solution is an ugly behemoth, if the hardware power makes it work and keep prices down, it is difficult to argue that this is a "bad thing". It is not a situation I like at all but one that needs to be accepted because certain skills that might have been important twenty years ago simply don't matter any more. Mark

Re: A tribute to the old manuals (again :) Message #4 Posted by Ron Ross on 14 June 2009, 9:42 a.m., in response to message #3 by Mark Edmonds Except that it violates the basic principles of KISS. And any time someone violates basic rules, trouble can result. If there are two ways to do something, the simpler method is less prone to problems either in errors or later in having to correct errors. Logic flow is much easier to trace in 5 steps vs 50 steps. Admittedly, I have indulged in poor code on more than one occassion, but have found that my simpler more compact code was always easier to use and modify on future applications.

Re: A tribute to the old manuals (again :) Message #5 Posted by designnut on 14 June 2009, 10:34 a.m., in response to message #4 by Ron Ross When I first started using the HP-35 I realized the formulas were written for pen and pencil math and copied from book to book through the years. I rearranged and rethought all the formulas to be amenable to keystroke procedures. I discovered relationships between variables that had not been explored before as they did not have the tools I had. It made a real impact when I found people programming solutions that I could do with a few keystrokes and keep repeating with different values as I had used the t register to supply part of the equation, namely 2PiF. Sam

Re: A tribute to the old manuals (again :) Message #6 Posted by Don Shepherd on 14 June 2009, 11:33 a.m., in response to message #3 by Mark Edmonds Quote: skills that might have been important twenty years ago simply don't matter any more Mark, if you're talking about skills such as keeping the code as simple and as clear as possible--not to mention using plenty of good quality comments in the code--then I have to, respectfully, disagree. As most programmers know, code can get extremely complex very quickly. Without some good standards and quality review procedures, even the best programmer can generate code that has hidden flaws. I would maintain that the likelihood of catching those hidden flaws decreases as code becomes more complex. Sure, the code may work in 99% of the circumstances, but what about that other 1%? Does it really matter? You bet it matters. People can get killed due to hidden flaws in code. I worked many years in testing air traffic control software. If that code does not work correctly 100% of the time, in all different situations, targets may not show up at the right place on the controller's display, or may not show up at all, and lives are at stake. You just cannot take a chance on that code not working 100% of the time. Regarding the recent tragic Air France crash in the Atlantic, I hope they find the black boxes because that is probably the only way they will know what caused that crash.

Re: A tribute to the old manuals (again :) Message #7 Posted by Antonio Maschio (Italy) on 23 June 2009, 4:10 a.m., in response to message #6 by Don Shepherd I generally follow any post on this Forum regarding manuals and documentation, both because I love old HP manuals and because I always hope someone write here "Hp manuals are back!". But, no surprise, instead of good news, I read reports about inconsistencies, errors and even fun facts. What I like best, in old HP manuals, is that you find everything you need when you are a beginner and have to learn how to use your calculator, but years after, any manual turns into a book of mathematics, that can be read as it is, with no calculator aside, just to refresh some knowledge and/or for the pleasure of reading it (for instance the HP34C manuals, the HP15C manuals, the HP48g manuals and, I'm sure but have no experience in it, the HP41C and the HP42S manuals). My little contribute. -- Antonio

Re: A tribute to the old manuals (again :) Message #8 Posted by Mike Morrow on 23 June 2009, 9:55 a.m., in response to message #7 by Antonio Maschio (Italy) IMHO, the best manual set *ever* for any calculator is the HP-15C Owner's Handbook with the optional HP-15C Advanced Functions Handbook, both spiral bound. This set is perfection and has never been bettered, not even by HP-41 or HP48 manuals. It's too bad these manuals didn't set the pattern for all later high-end HP calculators. The HP42S Owner's Manual is acceptable...really not bad overall, especially the early spiral-bound version. Still, it's a very long way from matching the appeal of HP-15C manuals. In fairness, an HP42S manual in the style of the HP-15C manual and as comprehensive with examples and helpful appendices would be huge, due to the many additional features of the HP42S.

Re: A tribute to the old manuals (again :) Message #9 Posted by Antonio Maschio (Italy) on 23 June 2009, 11:01 a.m., in response to message #8 by Mike Morrow That's it. I have never read a HP41/42 manual. But I own the original HP15C manual and a printed copy of the advanced manual and I agree with you they are the best (as far as my experience can say). -- Antonio

Re: A tribute to the old manuals (again :) Message #10 Posted by Michael de Estrada on 23 June 2009, 11:06 a.m., in response to message #8 by Mike Morrow I certainly agree with you regarding the HP-15C manuals. The HP-11C and original HP-12C manuals are also excellent. I also very much like the spiral bound manuals for the HP-48SX, and found them very useful when I bought my HP-50g.

[ Return to Index | Top of Index ]

Go back to the main exhibit hall