Initial Condition 2: Cold (follow up)

Walter recommended that I check whether the r^-1/2 behavior might be recovered more robustly if I increased the particle number, so these 4 runs have 10k particles rather than 1000. Sorry about the r^-2, I was mixing up scaling with that in another project, that was simply a mistake on my part. I left it in the plots for comparison, however, because I'm not finding a clear power law behavior in the density. Here are the initial conditions. They are the same save the number of particles.


To answer Walter's question about how I found the middle, I assumed it was at zero. I checked this by plotting the density and checking that the spike lined up with zero, at least for the binning I chose (bins because I'm using a histogram tool to compute the density, density being the number of particles per unit length).

Here is the final state, after 400 ATUs.


I wanted to see if, in this run, the phase space plots all initially converged as in the previous case. It looks identical to the plot in the last post, I'll spare you, you can take my word for it.

However, in poking around this simulation I noticed something interesting, take a look at this plot form 100 ATUs:



What struck me is that I could see windings, but they seemed fat in the orange curves. Here's a zoom into the middle part, indeed the particles seem to be doing similar winding behavior, but as a clump. Maybe you guys have seen this before, but I thought it was interesting and pretty different from all the behavior I'd seen before.


I checked the other runs, the green and the blue do not seem to have this behavior, but the red had something that looked related:


The green and the blue had winding too, but in a single thin stream of particles, much as is seen in the Binney paper. Here, I'll just show you:


See, single winding, all the way in. I'm pretty sure this has to do with the fact that the red and the yellow both form two blobs in phase space that then merge together. We are probably seeing something related to the final merger. Anyhow, I did some spot checking to make sure that (e.g.) the blue never does anything like this, it seems like the answer is no it doesn't though I don't know for sure. Although there are some funky things going on too... for example:


Anyhow, it might be a matter of mild concern that we don't get to the r^-1/2 behavior. I've run stage 1 of the dynamical mass experiment nearly 2x as long, I thought maybe it was just a matter of time but it seems like no, the simulations have reached their final density distributions. Here is the run with 20k particles, out to 700 ATUs.
Same kind of shape.