efinley%ug.utah.edu@cs.utah.edu (Elliott Finley) (02/10/90)
Has anyone written or does anyone know of a better integral solver than the 28s' own 'wimpy' polynomial integral solver? I'd like to see one that would digest ANYTHING, but I'm not sure if that is even possible.
mblakele@jarthur.Claremont.EDU (Tad Blakeley) (02/10/90)
In article <1990Feb9.095137.15021@hellgate.utah.edu> efinley%ug.utah.edu@cs.utah.edu (Elliott Finley) writes: >Has anyone written or does anyone know of a better integral solver than the >28s' own 'wimpy' polynomial integral solver? I'd like to see one that would >digest ANYTHING, but I'm not sure if that is even possible. It's not practical, especially not for a handheld. Programs like Mathematica and Milo (for the Mac) use integral lookup tables to try and solve certain _forms_ of indefinite integrals. If the integral you want to solve isn't in the table, which closely matches the CRC Math Tables, then the computer can't solve it. Also, many programs have a hard time recognizing alternative forms of the same integral. The CRC integral tables would take up more than 16k of the HP's memory, leaving little room for the program to read and interpret them. You might try writing a program, INTEGRATE, that would check an equation to see if it was in a library of forms, and pass it on to the HP polynomial solver if it wasn't. This way, you could keep the indefinite integrals tha you most often use in you own math tables. It's probably better to buy some math tables on paper, instead. -- tad -- ~ ...[physicists] who aren't actually dead are in some way very ill. S W -- D. Adams T I mblakele@jarthur.claremont.edu, but call me Tad R B West is Best, Carpe Potum ~