[yt-users] Profile/phase misalignment

[Matthew Turk]

Hi Dave,

It's a bit tricky to figure this out precisely, but I just made a
script that tested something similar on the IsolatedGalaxy dataset.

http://i.imgur.com/CE4lI.png
http://paste.yt-project.org/show/3050/

The image I made looks fine to me, but yours does look a bit strange.
There are a few things I would ask:

1) How did you set xmin, xmax, ymin, ymax?
2) What are the field definitions of CellVolumeNorm and
Lambda_therm_neg?  By any chance, does Lambda_therm_neg have a ton of
zero values in it?
3) Can you try again with a colormap that doesn't have white at the bottom?

-Matt

On Fri, Jan 11, 2013 at 12:52 PM, david collins <antpuncher at gmail.com> wrote:
> Hi, all--
>
> We're trying to overplot a phase diagram and profile, and we're seeing
> a strange mis-alignment of the data.  I've attached the plot, and the
> arguments used to create it are as follows:
>
> phase_therm_neg = \
>         pc.add_phase_object(dd, ["Pressure", "Lambda_therm_neg",
> "CellVolumeNorm"], weight=None, x_bins=xnbins, y_bins=ynbins, \
>                                 x_bounds=[xmin,xmax],
> y_bounds=[ymin,ymax], cmap = "PuBuGn")
> profile_therm_neg = pc.add_profile_object(dd, ["Pressure",
> "Lambda_therm_neg"], weight='CellVolumeNorm', \
>
> x_bounds=[xmin,xmax], x_bins=xnbins)
> phase_therm_neg.modify["line"](profile_therm_neg.data["Pressure"],
> profile_therm_neg.data["Lambda_therm_neg"])
>
>
> The fact that the line is below the peak of the color in the
> P=[10,10^4] range could be real, since the peak of the color
> distribution will trace the median, but the line is the average,
> though the effect is more pronounced than I would expect.  But in
> 10^4-10^5, there simply aren't any points down there, which leads me
> to believe I'm doing something funny.  Is there any obvious error in
> this script to anyone else?  Or any funny statistical phenomenon that
> I might be missing?
>
> d.
>
> --
> Sent from my computer.
>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>