46,146 reputation
167185
bio website uoregon.edu/~noeckel
location
age
visits member for 2 years, 8 months
seen 41 mins ago

Sep
28
comment What is the recommended way to define numeric function with special cases?
@LeonidShifrin Do you really think If is inferior when it concerns purely numeric functions? I wonder why that would be the case. Of course it's not usually equally aesthetic, but is there another reason you're thinking about? I guess the question doesn't really define the functions as purely numeric, so you're right if one admits that symbolic simplifications could still be performed. Thinking about it, it would be best if the question were made more specific.
Sep
28
comment Smoothing a single rectangular pulse in NDSolve
Closely related: A: NDSolve for PDE with discontinuous initial/terminal condition
Sep
28
reviewed Leave Closed weiner process conditional expected maximum?
Sep
28
reviewed Close NDSolve::derarg error: requiring pure function in solving differential equations
Sep
27
answered Function to Expand exponentials
Sep
27
revised Orthogonal vector in two dimensions
Clarified hat vector
Sep
27
comment Orthogonal vector in two dimensions
I think it's OK to leave this as a valid question - but only after changing the wording to clarify what "hat vector" means here.
Sep
27
comment Fitting ellipse to 5 given points on the plane
@MarkMcClure Looks like your assumption is correct (of course I already upvoted your answer).
Sep
27
comment Fitting ellipse to 5 given points on the plane
@user153012 For more than five points, see my comment to the original question: it becomes a fitting problem that has been asked about before.
Sep
27
comment Fitting ellipse to 5 given points on the plane
This could be considered to be a special case of Fitting points to tilted, off-center ellipse. But it's different in that this question admits a solution that goes through all points.
Sep
27
answered Efficient way to obtain values of a function defined by an Integral
Sep
27
comment Discrepancy between Mathematica and Fortran expressions — Fortran gives wrong output (C,C++ programmer trying to use Fortran)
This question is not really about Mathematica, and it doesn't specify in what way the output of the Fortran code is different. Do you mean the data type is different, or are you making plots, or are there special values of kx, ky, kz that give different values of Om1... It's impossible to tell what the issue really is.
Sep
26
reviewed Close Enumerating all connected graphs with 9 vertices
Sep
26
reviewed Close Is there a way to logically test an optional color input inside a module?
Sep
26
awarded  Good Question
Sep
26
comment Solving Laplace's equation in rectangle domain by separation of variables
To make this question useful to other readers, you should definitely boil it down to the absolutely essential minimum. In the process, you may already be able to eliminate all of the little bugs in this code (mentioned by @xzczd). It's always a good idea to develop a solution to a complicated boundary condition by first getting a simpler case to work. As it is now, the code can't be copied and run without editing. You should look at the code formatting help of the input editor.
Sep
26
awarded  Nice Answer
Sep
25
comment Tube and CapForm
In version 8 the ends are flat in both examples, too. So definitely a bug.
Sep
25
comment Derivatives of Convolutions with Integration by Parts and Delta Functions
Can you define p[x] with some additional lines of code?
Sep
24
comment How to replace the first occurance in an expression instead of replacing all?
Yes, that works. But now you'd have to do the same if it were Times instead of Plus... anyway, you put your finger exactly on the reason why the original didn't work, and changing the attributes is certainly a way to make it work.