Integral with unexpected result
|
05-30-2016, 04:42 AM
Post: #1
|
|||
|
|||
Integral with unexpected result
In home, int ((e^x)/(ln x)) from .01 to 1 produces ~58.9
In CAS, [~-23.55 ~-86.48] The integral should be infinite. This on hand held version 10077 |
|||
05-30-2016, 04:43 AM
Post: #2
|
|||
|
|||
RE: Integral with unexpected result
Correction...In home, the result is ~-58.92
|
|||
05-30-2016, 04:58 AM
Post: #3
|
|||
|
|||
RE: Integral with unexpected result
It's impossible to get infinity by a numerical method (it would require finding an equivalent at x=1). The CAS displays a big warning in the terminal before returning two values that are not close at all, a clear sign that the values are garbage.
|
|||
05-31-2016, 04:44 AM
Post: #4
|
|||
|
|||
RE: Integral with unexpected result
Hello,
In home, the numerical method does seem to converge to a solution One of the "good" thing about the method used in home is that it never evaluates the function on the 'edges', so it can deal with a function that is not defined there... but, for fast moving functions, this can lead to weird results. Cyrille Although I work for the HP calculator group, the views and opinions I post here are my own. I do not speak for HP. |
|||
« Next Oldest | Next Newest »
|
User(s) browsing this thread: 2 Guest(s)