I am looking for ideas on creating a "bumpy sphere" surface without the obvious symmetries and pole-singularity of standard spherical coordinates. Any ideas?

Looking for something qualitatively similar to this above.

Can we implement this? @Mr.Wizard @R.M. meta.mathematica.stackexchange.com/q/1741/12

@Szabolcs Is this too random?

@JasonB It might work. Is it based on a random point cloud?

That is using the code from Rahul's post here: mathematica.stackexchange.com/a/99761/9490

It uses a contourplot3d with a random function added in

@Szabolcs Of course you could also do something like this (included here only for its ugliness)

@JasonB Thanks for the pointer Jason!

@Szabolcs Which version? "wstp-and-mathlink"? Maybe "wstp-IS-mathlink" ? :^)

@Mr.Wizard Personally I'd be happy with any choice for as long as the two are made two be synonyms of the same tag. Currently there is absolutely no functional difference between MathLink and WSTP, and I'd be surprised if that changed in the near future.

Is there an easy way of converting:

to:

{{1, 2, a, b}, {3, 4, c, d}}

eg. given two matrix with the same number of rows, concatenate them, the first being on the left and second being on the right

@Tyilo Flatten /@ Transpose[{matrix1, matrix2}]

ArrayFlatten@{{m1, m2}} seems to work too

@JasonB thanks

@JasonB You're right, that's not what Image3D should look like.

@Szabolcs I'm tempted to blame it on the linux implementation being behind the others, since I would think I have a decent PC

I am using VectorAnalysis after a long time for 3-D cartesian problems (x,y,t) with NDSolve after 2012/2013. I notice that a lot has changed and mma code written in 2012 doesn't "work" any more because of the now prevailing syntax differences. So, I am unable to replicate this "non-linear evolution equation" anymore with mathematica 10.x. Is this problem worth a question on its own?