Hi Sean e.a.,
In reply to your messages, first an overview as I picked up from the
1. eMovie integration (Simon Kolstoe)
2. Powerpoint plugin (Roger Rowlett)
3. Enhanced morphing (ibid)
4. Foolproof download (David Hall)
5. DynoPlot integration (ibid)
6. Improved undo/redo (ibid)
7. Ray/png size/resolution (Georg Steinkeller)
8. Set mirrors for downloading (Adrian Schreyer)
9. WebGL output (ibid)
10. Vector graphics support (Sean Law)
11. PDF support (ibid)
12. Trajectory support (ibid)
13. CG object GUI (ibid)
As noted 2 and 3 are available in the incentive version. 4, 7 and 8
seem relatively straightforward to implement. For 5 steps should
probably be taken by the DynoPlot crew, or at least it should be done
6 is indeed problematic but highly desirable. Undoing operations on
states would require to store the differences between states, which
can be huge, e.g. in the case of splitting accidentally. On the other
hand, using wild cards makes it trivial to remove the splitted states
again, so I don't think that particular point is worth the trouble. On
the other hand, a view is not memory demanding, and it would be easy
to keep a stack of those to allow undo/redo.
9. Yea :)
10. Yea. Not by any means trivial though. But I'd be happy to
brainstorm on that one.
11. Would be cool.
12. Definite yea. I fear that this requires some serious changes
inside though to allow efficient viewing of a trajectory, such as VMD
and Yasara offer. The lack of trajectory support is a reason for many
working on MD to use VMD rather than Pymol. On the other hand, for the
open source/freeware version, the gromacs trajectory API could be
interesting to aid in realization of it.
In relation to this I'd really like to have support added for reading
topology information, and to use this to set bonds, atom types, masses
and charges, rather than the internal databases. Especially with the
momentum of coarse grained simulations these days, it would be
required to keep Pymol favoured over VMD.
13. Errm, yea, but... This would almost come down to integration of
Pymol with Blender. Oh, please do, yea, yea! :D
In relation to the last point, I would (14.) really like to see
improved POV-Ray output. The mesh that is output currently is a mess,
and highly inefficient. A correct mesh would require for each mesh
object (cartoon, surface, ... per object) to be written as a set of
nodes, with a set of pigments and a set of triangles, referring to the
node list. Now each triangle is its own mesh2 object, which is
terrible. In addition, I would also like to write POV-Ray include
files on a per object basis, maybe as a setting "split_povray_output".
That makes it much more easy to modify the scene. The master file
would only contain the global settings (camera, light, etc.) and a
number of include statements for the other objects. Actually, this is
what I'm now usually doing by hand, enabling/writing/disabling each
object and each representation... O, it's also almost trivial to
include all views in the master file, so that a complete scene can be
written out as a set of POV-Ray files. I'd be happy to streamline and
contribute my POV-Ray code for that.
Another thing on my mind is the desire to (15.) read in images as
objects, or, better, as pigment/texture on an object. That will make
it easy to overlay a structure on a graph for instance.
Finally, unless I think of more later, I would like to be able to
(16.) read and change coordinates quicker. Now, iterating over states
to fiddle with coordinates is painstakingly slow. I'm afraid though
that this is heavily tied in with the internal structure of Pymol,
regarding objects and states. But if that were streamlined, probably
together with 12, then probably Pymol would be able to become a true
platform for analysis of trajectories.
On Wed, Jan 20, 2010 at 5:26 AM, Jason Vertrees
> We need to move forward on PyMOL development, so I'm collecting ideas
> from all our users on what they most want in future versions of PyMOL.
> We already have a page on the PyMOL wiki for this
> (http://pymolwiki.org/index.php/Ideas) from 2008/2009. We can discuss
> the ideas here and on the wiki; this list is more for two-way
> communication and the wiki for one-way. If you have a specific
> request for an enhancement you can also email it directly to
> help@...; please put "PyMOL" in the subject so I can get
> them easily. Each request submitted will be considered for
> Last, you're particularly passionate about your feature request, email
> me directly and we can chat about it via phone/skype.
> -- Jason
> Jason Vertrees, PhD
> PyMOL Product Manager
> Schrodinger, LLC
> (e) Jason.Vertrees@...
> (o) +1 (603) 374-7120
> Throughout its 18-year history, RSA Conference consistently attracts the
> world's best and brightest in the field, creating opportunities for Conference
> attendees to learn about information security's most important issues through
> interactions with peers, luminaries and emerging and established companies.
> PyMOL-users mailing list (PyMOL-users@...)
> Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
> Archives: http://www.mail-archive.com/pymol-users@...>
Tsjerk A. Wassenaar, Ph.D.