Hi John, Luigi, I think you also will need to specify the region as a data_source to the add_projection call: pc.add_projection("Density", 0,source=reg) and I think PlotCollection sitll takes the pf, not the reg variable, as its argument. Here's an example script, if you wanted to do it with a sphere: http://paste.enzotools.org/show/441/ I'll add the scripts you sent with this modification to the cookbook. Thanks! -Matt On Tue, Apr 13, 2010 at 11:04 AM, John Wise wrote:

Hi,

I would think that you'd have to change the region boundaries, if you wanted to change the projection depth.  To have L=0.2, you have to change your region boundaries to reflect that change, i.e. [0.4,0.4,0.4] -> [0.6,0.6,0.6].

If I wanted to make projections with a given width and depth, I'd do the following.

L = 0.1  # projection width d = 0.2  # projection depth center = na.array([0.5]*3) dim = 0

left_edge = center - 0.5*L right_edge = center + 0.5*L

# For the projection dimension, use the depth left_edge[dim] = center - 0.5*d right_edge[dim] = center + 0.5*d

pf = get_pf() reg = pf.h.region(center, left_edge, right_edge)

pc = PlotCollection(reg, center=center) pc.add_projection("Density", 0, "Density") pc.set_width(L, '1') pc.save("test", override=True)

Cheers, John

On 04/13/2010 10:21 AM, Luigi Iapichino wrote:

...

Dear all,

by comparing volume renderings and projections in my simulation I realised that I don't have control on the projection depth, in the latter case. This is the relevant part of my script:

...

L = 0.1 # projection width

pf = load(data_dir+fn) # load data

reg = pf.h.region([0.5, 0.5, 0.5],[0.45,0.45,0.45],[0.55,0.55,0.55]) # a small region #reg = pf.h.region([0.5, 0.5, 0.5],[0,0,0],[1,1,1]) # the whole box

pc = PlotCollection(reg, center=[0.5,0.5,0.5])

p = pc.add_projection("Density", 0, weight_field="Density",fig_size=(7.5,6))

p.modify["contour"]("Density", ncont=5,take_log=True)

pc.set_width(L, '1')

pc.save("%s_test-proj" % pf,override=True)

If I set L = 0.2, I don't see in the central region of the projection more structures than with L=0.1 (only the plot width is, of course, larger). Moreover, the result do not depend on the region that I select (either third or fourth line). It seems to me that, in all cases, the projection is actually performed along the whole simulation box, and not on the region that I select. In other words, I can control the plot width, but not the projection depth. In the volume rendering, on the other hand, the plot width sets the depth, too. Is anything wrong or missing in my script?

If needed, I can send you a few explanatory plots and the whole sample script for reproducing my issue. Looking forward to your suggestions,

  Luigi

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

Hi Luigi, Here is one more option. The projection function accepts a keyword called "field_cuts", which allows you to only project cells that satisfy certain field based criteria. For example, you could do this: pc = PlotCollection(pf, center=[0.5,0.5,0.5]) pc.add_projection('Density', 0, weight_field='Density', field_cuts=['grid["x"] > 0.4, 'grid["x"] < 0.6']) The one advantage I see of doing it this way is that this will work in parallel, where supplying a source keyword will not. Britton On Tue, Apr 13, 2010 at 9:12 AM, John Wise wrote:

Hi Matt,

I was curious if a PlotCollection could accept an AMRregion, and it could, to my surprise! However, it mangles the imagenames, and thus you have to specify it manually. I've always used the source argument before, but kept to Luigi's style on this sample code :)

Thanks! John

On 04/13/2010 11:06 AM, Matthew Turk wrote:

...

Hi John, Luigi,

I think you also will need to specify the region as a data_source to the add_projection call:

pc.add_projection("Density", 0,source=reg)

and I think PlotCollection sitll takes the pf, not the reg variable, as its argument. Here's an example script, if you wanted to do it with a sphere:

http://paste.enzotools.org/show/441/

I'll add the scripts you sent with this modification to the cookbook. Thanks!

-Matt

On Tue, Apr 13, 2010 at 11:04 AM, John Wise wrote:

...

Hi,

I would think that you'd have to change the region boundaries, if you wanted to change the projection depth. To have L=0.2, you have to change your region boundaries to reflect that change, i.e. [0.4,0.4,0.4] -> [0.6,0.6,0.6].

If I wanted to make projections with a given width and depth, I'd do the following.

L = 0.1 # projection width d = 0.2 # projection depth center = na.array([0.5]*3) dim = 0

left_edge = center - 0.5*L right_edge = center + 0.5*L

# For the projection dimension, use the depth left_edge[dim] = center - 0.5*d right_edge[dim] = center + 0.5*d

pf = get_pf() reg = pf.h.region(center, left_edge, right_edge)

pc = PlotCollection(reg, center=center) pc.add_projection("Density", 0, "Density") pc.set_width(L, '1') pc.save("test", override=True)

Cheers, John

On 04/13/2010 10:21 AM, Luigi Iapichino wrote:

...

Dear all,

by comparing volume renderings and projections in my simulation I realised that I don't have control on the projection depth, in the latter case. This is the relevant part of my script:

...

L = 0.1 # projection width

pf = load(data_dir+fn) # load data

reg = pf.h.region([0.5, 0.5, 0.5],[0.45,0.45,0.45],[0.55,0.55,0.55]) # a small region #reg = pf.h.region([0.5, 0.5, 0.5],[0,0,0],[1,1,1]) # the whole box

pc = PlotCollection(reg, center=[0.5,0.5,0.5])

p = pc.add_projection("Density", 0, weight_field="Density",fig_size=(7.5,6))

p.modify["contour"]("Density", ncont=5,take_log=True)

pc.set_width(L, '1')

pc.save("%s_test-proj" % pf,override=True)

If I set L = 0.2, I don't see in the central region of the projection more structures than with L=0.1 (only the plot width is, of course, larger). Moreover, the result do not depend on the region that I select (either third or fourth line). It seems to me that, in all cases, the projection is actually performed along the whole simulation box, and not on the region that I select. In other words, I can control the plot width, but not the projection depth. In the volume rendering, on the other hand, the plot width sets the depth, too. Is anything wrong or missing in my script?

If needed, I can send you a few explanatory plots and the whole sample script for reproducing my issue. Looking forward to your suggestions,

Luigi

_______________________________________________

yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

Hi John, You know ... I have to confess I'm a bit surprised that it works, but thinking back, I don't think there are any methods that it calls on the parameter file that aren't present in the AMRegion. And, I think you're on to something: it makes more sense to have a PlotCollection affiliated with either the (implicit) "all data in a parameter file" or with a specific 3D data object. Maybe this is something that should definitely Work? -Matt On Tue, Apr 13, 2010 at 11:12 AM, John Wise wrote:

Hi Matt,

I was curious if a PlotCollection could accept an AMRregion, and it could, to my surprise!  However, it mangles the imagenames, and thus you have to specify it manually.  I've always used the source argument before, but kept to Luigi's style on this sample code :)

Thanks! John

On 04/13/2010 11:06 AM, Matthew Turk wrote:

...

Hi John, Luigi,

I think you also will need to specify the region as a data_source to the add_projection call:

pc.add_projection("Density", 0,source=reg)

and I think PlotCollection sitll takes the pf, not the reg variable, as its argument.  Here's an example script, if you wanted to do it with a sphere:

http://paste.enzotools.org/show/441/

I'll add the scripts you sent with this modification to the cookbook.  Thanks!

-Matt

On Tue, Apr 13, 2010 at 11:04 AM, John Wise  wrote:

...

Hi,

I would think that you'd have to change the region boundaries, if you wanted to change the projection depth.  To have L=0.2, you have to change your region boundaries to reflect that change, i.e. [0.4,0.4,0.4] -> [0.6,0.6,0.6].

If I wanted to make projections with a given width and depth, I'd do the following.

L = 0.1  # projection width d = 0.2  # projection depth center = na.array([0.5]*3) dim = 0

left_edge = center - 0.5*L right_edge = center + 0.5*L

# For the projection dimension, use the depth left_edge[dim] = center - 0.5*d right_edge[dim] = center + 0.5*d

pf = get_pf() reg = pf.h.region(center, left_edge, right_edge)

pc = PlotCollection(reg, center=center) pc.add_projection("Density", 0, "Density") pc.set_width(L, '1') pc.save("test", override=True)

Cheers, John

On 04/13/2010 10:21 AM, Luigi Iapichino wrote:

...

Dear all,

by comparing volume renderings and projections in my simulation I realised that I don't have control on the projection depth, in the latter case. This is the relevant part of my script:

...

L = 0.1 # projection width

pf = load(data_dir+fn) # load data

reg = pf.h.region([0.5, 0.5, 0.5],[0.45,0.45,0.45],[0.55,0.55,0.55]) # a small region #reg = pf.h.region([0.5, 0.5, 0.5],[0,0,0],[1,1,1]) # the whole box

pc = PlotCollection(reg, center=[0.5,0.5,0.5])

p = pc.add_projection("Density", 0, weight_field="Density",fig_size=(7.5,6))

p.modify["contour"]("Density", ncont=5,take_log=True)

pc.set_width(L, '1')

pc.save("%s_test-proj" % pf,override=True)

If I set L = 0.2, I don't see in the central region of the projection more structures than with L=0.1 (only the plot width is, of course, larger). Moreover, the result do not depend on the region that I select (either third or fourth line). It seems to me that, in all cases, the projection is actually performed along the whole simulation box, and not on the region that I select. In other words, I can control the plot width, but not the projection depth. In the volume rendering, on the other hand, the plot width sets the depth, too. Is anything wrong or missing in my script?

If needed, I can send you a few explanatory plots and the whole sample script for reproducing my issue. Looking forward to your suggestions,

  Luigi

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

Hi Luigi, Glad to hear it's working now. I think for projecting smallish regions of the box, specifying the source region is definitely better, because with the field_cuts method, you are still checking every cell to see whether it satisfies the requirement. The field_cuts method is more advantageous when cutting over non-position fields, like say only projecting gas within some temperature range. It's a little known feature that comes in handy every now and then. Britton On Tue, Apr 13, 2010 at 9:47 AM, Luigi Iapichino < luigi@ita.uni-heidelberg.de> wrote:

Hi John, Matt, Britton,

thanks for your suggestions! Both specifying the data_source or the field_cuts work as I needed. The former option is much faster (~4X) in my problem, but I imagine that this is rather setup-dependent, and the choice between working in a region or in parallel should be taken case by case. Best,

Luigi

...

Hi John,

You know ... I have to confess I'm a bit surprised that it works, but thinking back, I don't think there are any methods that it calls on the parameter file that aren't present in the AMRegion. And, I think you're on to something: it makes more sense to have a PlotCollection affiliated with either the (implicit) "all data in a parameter file" or with a specific 3D data object.

Maybe this is something that should definitely Work?

-Matt

On Tue, Apr 13, 2010 at 11:12 AM, John Wise wrote:

...

Hi Matt,

I was curious if a PlotCollection could accept an AMRregion, and it could, to my surprise! However, it mangles the imagenames, and thus you have to specify it manually. I've always used the source argument before, but kept to Luigi's style on this sample code :)

Thanks! John

On 04/13/2010 11:06 AM, Matthew Turk wrote:

...

Hi John, Luigi,

I think you also will need to specify the region as a data_source to the add_projection call:

pc.add_projection("Density", 0,source=reg)

and I think PlotCollection sitll takes the pf, not the reg variable, as its argument. Here's an example script, if you wanted to do it with a sphere:

http://paste.enzotools.org/show/441/

I'll add the scripts you sent with this modification to the cookbook. Thanks!

-Matt

On Tue, Apr 13, 2010 at 11:04 AM, John Wise

...

...

wrote:

...

Hi,

I would think that you'd have to change the region boundaries, if you wanted to change the projection depth. To have L=0.2, you have to change

your

...

...

region boundaries to reflect that change, i.e. [0.4,0.4,0.4] -> [0.6,0.6,0.6].

If I wanted to make projections with a given width and depth, I'd do the following.

L = 0.1 # projection width d = 0.2 # projection depth center = na.array([0.5]*3) dim = 0

left_edge = center - 0.5*L right_edge = center + 0.5*L

# For the projection dimension, use the depth left_edge[dim] = center - 0.5*d right_edge[dim] = center + 0.5*d

pf = get_pf() reg = pf.h.region(center, left_edge, right_edge)

pc = PlotCollection(reg, center=center) pc.add_projection("Density", 0, "Density") pc.set_width(L, '1') pc.save("test", override=True)

Cheers, John

On 04/13/2010 10:21 AM, Luigi Iapichino wrote:

...

Dear all,

by comparing volume renderings and projections in my simulation I realised that I don't have control on the projection depth, in the latter case. This is the relevant part of my script:

...

L = 0.1 # projection width

pf = load(data_dir+fn) # load data

reg = pf.h.region([0.5, 0.5, 0.5],[0.45,0.45,0.45],[0.55,0.55,0.55]) # a small region #reg = pf.h.region([0.5, 0.5, 0.5],[0,0,0],[1,1,1]) # the whole box

pc = PlotCollection(reg, center=[0.5,0.5,0.5])

p = pc.add_projection("Density", 0, weight_field="Density",fig_size=(7.5,6))

p.modify["contour"]("Density", ncont=5,take_log=True)

pc.set_width(L, '1')

pc.save("%s_test-proj" % pf,override=True)

If I set L = 0.2, I don't see in the central region of the

...

...

...

...

...

more structures than with L=0.1 (only the plot width is, of course, larger). Moreover, the result do not depend on the region that I select (either third or fourth line). It seems to me that, in all cases, the projection is actually performed along the whole simulation box, and not on the region that I select. In other words, I can control the plot width, but not the projection depth. In the volume rendering, on the other hand, the

On Tuesday 13 April 2010, Matthew Turk wrote: projection plot

...

...

...

...

...

width sets the depth, too. Is anything wrong or missing in my script?

If needed, I can send you a few explanatory plots and the whole sample script for reproducing my issue. Looking forward to your suggestions,

Luigi

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

--

---------------------------------------------------------------

Luigi Iapichino Zentrum fuer Astronomie der Universitaet Heidelberg Institut fuer Theoretische Astrophysik Albert-Ueberle-Str. 2, D-69120 Heidelberg, Germany Tel: +49 6221 548983, Fax: +49 6221 544221 e-mail: luigi@ita.uni-heidelberg.de URL: http://www.ita.uni-heidelberg.de/~luigi/http://www.ita.uni-heidelberg.de/%7Eluigi/ _______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org