Wolfram Mathematica

A: How do I draw a pair of buttocks?

2:58 AM

Okay, while scanning the archives, I just saw this:

340

I have to confess that I see this as a proper challenge, as I am usually quite creative in finding/combining functions to provide a desired behavior. So I will give it another try. which is generated using box[x_, x1_, x2_, a_, b_] := Tanh[a (x - x1)] + Tanh[-b (x - x2)]; ex[z_, z0_, s_] := E...

I've been doing surface synthesis for quite a while (e.g. Chinese roof, nautilus), but I don't think I ever thought to do asscheeks.

(@belisarius, I fully agree with your comment there. ;) )

The Toad

@Guesswhoitis. More importantly, are you going to fix it? ;)

3:21 AM

@The Toad, I'll probably put this high up on my priority list when I get my computer back… :D

6 hours later…

kirma

9:20 AM

5 hours later…

bobthechemist

2:23 PM

Anyone around who has experience with CreateExecutable? I'm trying the example on the RPi and it returns $Failed. I can call CreateObjectFile on a "Hello World" example and I note that M dumps this file in a /home/pi/.WolframEngine/... subdirectory. Setting the WorkingDirectory and TargetDirectory options doesn't seem to help in this case.

Also, the obj file created with CreateObjectFile can be compiled manually from the shell.

Reading all these answers related to finite elements has me getting more excited about trying out version 10…

…though I have to ask this: has anyone worked on how to convert Comsol (Femlab) meshes into something Mathematica can deal with?

bobthechemist

@Guesswhoitis. let me know when you figure that out; I can't afford my COMSOL license any longer.

Murta

3:01 PM

@Pfon R is very interesting for some applications when compared to Mathematica. For example, it's very easy to distribute your code, and everybody can use it without restrictions. I'm just starting, it's soon to say something. As a beginner, I miss the elegancy of Mathematica notation and function names, as miss the rich Mathematica overload, compact anonymous functions and the simplicity of Map (and not 5 times of apply as in R).

@Pfon In the other way, R has OOP, and a rich IDE (RSTUDIO), with cool GIT integration, debug, and other facilities that I miss in Mathematica. DataFrame are cool. It's nice experience to study another language, and broke some prejudices. :)

1 hour later…

Guillochon

4:03 PM

Hey all, does anyone have much experience with the DiscreteVariables option in NDSolve?

Mr.Wizard

@Guillochon Sorry, I do not.

4:17 PM

I need to create a two-dimensional polymer path of equal length segments, where the orientation of each segment is changed from the orientation of the previous segment by an angle difference dtheta that is taken from a normal distribution with standard deviation sigma in radians.

To do this: For[i = 1, i < 5, i++, p = Append[p, {x, y}];
dtheta = Random[NormalDistribution[0, 0.3]];
x = a + x Cos[dtheta] - y Sin[dtheta];
y = a + x Sin[dtheta] + y Cos[dtheta]] ,where: a =0.01, x = y=0, p =List[]

I am not sure if I am doing the job correctly. I used the 2D rotation matrix to change the orientation of each segment. I found out that the length of each segment is unequal too. Could anyone please check my work?

4:56 PM

@psimeson You can check it by choosing a few dtheta manually and plotting the path

I plotted it but the length of each segments is not equal

@psimeson I think you're looking for something like this, yes? Graphics@Line@AnglePath@RandomVariate[NormalDistribution[0, 0.03], 5]

AnglePath is the important function, it only works if you have Mathematica 10.

I am mathematica 10 but on this machine.. I am 9 on this one so it didn't work

Is there way to do this is in mathematica 9?

5:14 PM

@psimeson I wanted to check my understanding of what you are trying to do. In your code the length of a segment it dependent on the position, and you have an offset a so that essentially you are generating a bunch of vectors centered on {a,a}. But what you really want to do is choose an angle, go in that direction a certain length, choose and new angle and go from the new point in that direction, right?

I'd use FoldList[] for random walks like this.

I am trying to build up a polymer path segment by segment, such that each segment has same length but different orientation. Using your code (which works on Mathematica 10), I kinda get a polymer chain but I don't understand how it's working.

@Pickett you are correct :" what you really want to do is choose an angle, go in that direction a certain length, choose and new angle and go from the new point in that direction, right?" that certain length is same

@Guesswhoitis. I will look into FoldList[]

The kicker is that the angle of the new segment is dependent on the angle of the previous one.

5:19 PM

Not so. The angle at each point will be picked randomly so it won't depend on the previous angle

I think this : dtheta = Random[NormalDistribution[0, 0.3] assure that angle is random

"where the orientation of each segment is changed from the orientation of the previous segment by an angle difference " says otherwise.

If you don't care about adjusting the angle dependent on the previous angle, you can simple change your code to

a = 1; x = y = 0;  p = {};

For[i = 1, i < 5, i++,
  p = Append[p, {x, y}];
  dtheta = Random[NormalDistribution[0, 0.3]];
  {x, y} = {x, y} + {Cos[dtheta], Sin[dtheta]};
  ];

I would also use FoldList normally, but it's just a matter of code style.

So in this case you are just offsetting x co-ordinate by Cos[dtheta] and y co-ordinate by Sin[dtheta], right ? How can I be sure that the length of each segment is same?

Hint: Cos[x]^2 + Sin[x]^2 == 1

@Guesswhoitis. I think ""where the orientation of each segment is changed from the orientation of the previous segment by an angle difference " is kinda ambiguous. I am trying to create a fake polymer chain with known persistence length. So, that I can test if my code for figuring out the persistence length of polymer chain is working or not.

In case of real polymer I think the angle dependence is not required.

5:32 PM

Looks clear to me; the intent is that your changes in direction are not as abrupt, since you only change from the previous angle by some amount.

true

5:57 PM

that's that reason for using sigma < 0.5 in Random[NormalDistribution[0, 0.3]]