[yt-users] Image resolution

David Collins dcollins at physics.ucsd.edu
Fri Dec 3 12:24:43 PST 2010


Thanks.  I think the second option is what I'm looking for, and 'dig
through MPL' is what I was afraid you were going to say.

Basically I want a plot like what comes out of pc.add_projection();
pc.save(), color bar and axes labels and whatnot, but with the
projection at a given resolution.  Some combination of fig_size and
dpi will do it, provided I know how things scale, which points, as you
mentioned, to the MPL docs.

d.

On Fri, Dec 3, 2010 at 11:23 AM, Matthew Turk <matthewturk at gmail.com> wrote:
> Hi Dave,
>
> On Fri, Dec 3, 2010 at 10:44 AM, david collins <antpuncher at gmail.com> wrote:
>> Hi--
>>
>> I've been asked to produce some figures with a specific resolution in
>> my projection plots.
>>
>> I know that the fig_size will change the size of the entire image, but
>> I don't know how the borders and colorbar sizes scale.  Is there an
>> obvious way to get this information, or a way to directly control the
>> size of the projection portion of the figure itself?
>
> It's not entirely clear to me what you're asking, and depending on the
> output image format, the answer will vary.  I've selected a couple
> options, but I'm not sure that they cover what you are looking for.
>
> If you are looking for a PNG image made from either a slice or a
> projection, wherein a floating-point 2D buffer is scaled to some
> min/max and then output completely filling the entire image, you can
> and should use the FixedResolutionBuffer and then the write_image
> command.  Both are provided in yt.mods in the current development tip.
>  The DPI used in write_image is set to be 100, but as you know this is
> not always relevant for PNG files.
>
> If you are looking for a PNG image with a colorbar, then you must use
> matplotlib, and you will have to supply either axes= and figure=
> arguments to the creation of a Plot object, or you can use the
> FixedResolutionBuffer and manually create your entire image.  Figures
> can have set DPIs as well as sizes, but as for the border and whatnot,
> you will have to read the matplotlib documentation for more help with
> that.
>
> If you are looking for an EPS image without a colorbar, I would
> suggest you create a PNG and then convert it by hand.
>
> If you are looking for an EPS image with a colorbar, you will have to
> follow the same steps for the plot collection.  Alternately, John Wise
> may have a solution to this using the PyX library that he has created
> an excellent wrapper for.
>
> Finally, if your question is less about the number of pixels in an
> image plane as opposed to the final *size* of the image including
> everything, you should investigate the judicious modification of the
> DPI in a given figure.
>
> Good luck,
>
> Matt
>
>>
>> Thanks,
>> d.
>>
>> --
>> Sent from my Stone Tablet and carried by my Pterodactyl.
>> _______________________________________________
>> yt-users mailing list
>> yt-users at lists.spacepope.org
>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>



-- 
Sent from my Stone Tablet and carried by my Pterodactyl.



More information about the yt-users mailing list