HP Forum Archive 14

[ Return to Index | Top of Index ]

hp-SOLVE in the 33s Message #1 Posted by Eric Lundgren on 25 Feb 2005, 1:40 a.m. I am proudly the owner of another (exam redundancy) 33s. After entering the quadratic form a*sqre(x) + b*x = - c I am concerned to obtain only one root (and incorrect). Perhaps I need to think about this, yet I have an excel spreadsheet and XEQ on the 33s, both which arrive at same answer...which is likely because the same idiot programmed both..;) Any thoughts..? Eric

Re: hp-SOLVE in the 33s Message #2 Posted by Namir on 25 Feb 2005, 5:22 a.m., in response to message #1 by Eric Lundgren HP's calculator website has 4 learning modules about using the SOLVE feature. I encourage you to look at them and download them.

Re: hp-SOLVE in the 33s Message #3 Posted by bill platt on 25 Feb 2005, 8:54 a.m., in response to message #1 by Eric Lundgren Hi Eric, I think the following posts should be helpful for understanding how to get the 33s solver to work most effectively. Also be sure to read the manual and the guides on the hp website--the solver is a powerful tool, but slightly different from the 32sii in some ways (though just as good if not better) and it is helpful to understand the nuances. http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv014.cgi?read=62608 http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv014.cgi?read=61213 http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv014.cgi?read=61224 http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv014.cgi?read=61136 Best regards, Bill

Re: hp-SOLVE in the 33s Message #4 Posted by Karl Schneider on 25 Feb 2005, 11:51 p.m., in response to message #1 by Eric Lundgren Quote: After entering the quadratic form a*sq(x) + b*x = - c I am concerned to obtain only one root (and incorrect). As Namir and Bill said, there is osme supplemental info available at HP's website. The info describes how the 33S will use a "direct solution" method for equations of certain forms, solving algebraically instead of iteratively. This is not always desirable, because one must then know how to modify the equaiton so that the 33S will find the other roots. In any case, it should have returned at least one correct answer. What were the coefficients? -- KS

Re: hp-SOLVE in the 33s Message #5 Posted by eric lundgren on 26 Feb 2005, 3:24 p.m., in response to message #4 by Karl Schneider Karl, For the coefficients A = 30 B = -25 C = 1 SOLVE yields x= 4.84, which by a sanity check is way off. My short program yields Sol'ns +/- = 0.7912 0.0421 which is correct, any ideas? I'll look into the book to. Eric

Works fine on my 33S.... Message #6 Posted by Karl Schneider on 26 Feb 2005, 6:34 p.m., in response to message #5 by eric lundgren All I can suggest is to follow the instructions. The equation should look like this: AxSQ(X)+BxX=-C Display the equation in the x-register; press SOLVE; press "2" for X; enter the three coefficient values with "R/S" after each one. You can find the desired root by defining the search range between two values: Store one value in the variable X; put the other value in the x-register; then do as above. -- KS

Re: Works fine on my 33S.... Message #7 Posted by Eric Lundgren on 26 Feb 2005, 9:17 p.m., in response to message #6 by Karl Schneider Karl, I don't know what to say. I entered the eqn again in place of my original one, but using some parentheses for confidence, and it solved correctly, giving the first correct root. However, I really believe that I had it entered as you listed. Well, pc's don't lie, so I must have erred in the entry. Thanks for the help! Eric

Re: hp-SOLVE in the 33s Message #8 Posted by Veli-Pekka Nousiainen on 27 Feb 2005, 3:48 a.m., in response to message #5 by eric lundgren 0.79120341757 0.0421299157633 are the correct roots [VPN]

Re: hp-SOLVE in the 33s Message #9 Posted by Barnardo Vargas on 1 Mar 2005, 6:56 a.m., in response to message #1 by Eric Lundgren Don't forget that computer programs are not magical black boxes. They are based on algorithms, and all HP SOLVE programs are based on well known mathematical algorithms, such as Newton's, regula falsi, etc., which DO have limitations, i.e., they are not always applicable, they will not always converge to a root; that depends on some conditions the function must fulfill. Even the most powerful computer in the world programmed with the "best" root-finding algorithm can fail to return a correct answer in some cases, and that doesn't mean its programmers are incompetent or that the programs contain bugs. Get used to it.

[ Return to Index | Top of Index ]

Go back to the main exhibit hall