PDA

View Full Version : reliability of Mathematica


niko2000
Oct14-04, 07:35 AM
Hi,
Recently I was doing some calculations with Mathematica then I did it on paper and the results weren't the same. I am almost sure I didn't make any mistake. I study electrical engineering and we have to do a lot of calculations. I have used Matlab and Mathematica for calculations with symbols. Until now I haven't done much complex calculations with Mathematica. Does anyone know if Mathematica is 100% reliable for every equation or is it only a tool for simple equations?
Regards,
Niko

mewmew
Oct14-04, 12:06 PM
In short, yes, it is correct 100% of the time I would say. If it wasn't there would be no use for it. I use it all the time and it can out-math pretty much anything out there, it is a computer program after all, its job is to do math calculations.

shmoe
Oct14-04, 12:10 PM
I haven't used Mathematica very much but I've never had Maple screw up and I'd wager Mathematica is quite reliable.

There is a disclaimer though, these programs will be able to reliably perform any computation that you tell them to perform. Make sure you're actually telling it to do what you think you're telling it to do. They aren't free of user error.

chroot
Oct14-04, 01:12 PM
Mathematica, MATLAB, and other CA systems will sometimes disagree when doing really, really intensive tensor-calc problems with lots of ODEs. The ODE solvers are not always perfect, but are, well, virtually always right.

If you're doing some simple EE work, and your pencil-and-paper solution doesn't match Mathematica's solution, it's almost surely your error. Check your pencil-and-paper work, then check to make sure you accurately supplied the problem to Mathematica.

- Warren

Chrono
Oct14-04, 01:43 PM
They wouldn't release it if they knew that it wasn't 100% reliable. Even if it was 99%, I doubt they would.

I think you may have done something wrong when you did it on paper. Double check it.

niko2000
Oct14-04, 05:29 PM
Thank you.
Maybe the result isn't different but it isn't simplified. I am sure I haven't done any error with my calculations.
Something that also seems strange is that Mathematica gave the result in note like this:
Sqrt[a^2......] I have put FullSimplify command but it hasn't changed. Anyone knows why?

chroot
Oct14-04, 06:33 PM
You might want to use an ExpandAll followed by a FullSimplify.

- Warren

niko2000
Oct15-04, 03:39 AM
It still doesn't work.
I get this result:
Sqrt[a^2*Cosh[x]^2*Sin[y]^2]