Hi Jason, Well, this is just polemic... But anyway. If I export the scene to povray, the camera will always be placed at the origin. If I then want to add something to the scene that is defined in the original coordinates of the pdb file, I have to apply the transformation encoded in the matrix obtained from get_view to the coordinates to get them in the view of the camera. That means the view acts on the coordinates, and not on the camera. Otherwise, the view recipe would have to be inverted. The main point is probably, but I'm guessing here, that OpenGL has the screen as reference, which can be regarded a fixed camera. The difference in speed probably arises, as for changing the view, the OpenGL engine can apply the operations on the whole scene, which is quick, while rotate (drag, whatever) changes coordinates relative to each other, which is a per particle operation and thus slow.
So much for the polemic part :) Related to this, it could be nice to have camera objects (location, focal point, field of view, roll) with paths between them :) Cheers, Tsjerk On Wed, May 12, 2010 at 4:06 PM, Jason Vertrees <jason.vertr...@schrodinger.com> wrote: > Hi all, > > PyMOL allows for both camera and object coordinate changes. When you > use the mouse to move stuff around, that's the camera moving*--just > like the 'turn' command. When you use "rotate", object dragging or > some of the other matrix transformation commands (fit, super, align, > matrix_copy, update, etc) you are affecting the coordinates of the > objects themselves. > > Try the following ton convince yourself: > > # get a protein > fetch 1oky, async=0 > > # the original view matrix > get_view > > # move the camera & get view > turn x, 90 > get_view > > # rotate 1oky & get view > rotate x, 90, 1oky > get_view > > Compare the values after turning vs rotating. > > This brings up another point, for large molecular assemblies, it's > much faster internally to use 'turn' rather than 'rotate'. 'Turn' is > simply a matrix multiplication, the other is that and storing the new > coordinates back into the object. > > Cheers, > > -- Jason > > * -- you can move objects themselves with the mouse, but only in > editing or dragging mode > > On Wed, May 12, 2010 at 4:27 AM, Tsjerk Wassenaar <tsje...@gmail.com> wrote: >> Hi Robert, >> >> PyMOL>help rotate >> >> DESCRIPTION >> >> "rotate" can be used to rotate the atomic coordinates of a >> molecular object. Behavior differs depending on whether or not the >> "object" parameter is specified. >> >> <snip> >> >> So much for rotate. The help for turn is misleading: >> >> PyMOL>help turn >> >> DESCRIPTION >> >> "turn" rotates the camera about one of the three primary axes, >> centered at the origin. >> >> <snip> >> >> It does not turn the camera, actually. It changes the orientation of >> the scene with respect to the camera. The effect is the same, and you >> can interconvert. That is the basis for the povray macros I wrote some >> while ago to combine things in object space with those in the pymol >> camera space in the exported scene. >> >> Cheers, >> >> Tsjerk >> >> On Wed, May 12, 2010 at 9:21 AM, Robert Fenwick >> <robert.fenw...@irbbarcelona.org> wrote: >>> >>> What is the difference between the turn and the rotate commands? >>> On 12 May 2010, at 06:21, Tsjerk Wassenaar wrote: >>> >>> Hi Mike, Tom, >>> >>> If I'm not mistaken, it is always the camera that is moving when you rotate >>> >>> a molecule. >>> >>> Unfortunately I have to inform you that you are mistaken... In Pymol >>> the camera is always fixed. That also means that it is not possible to >>> have the camera fly through the protein. >>> >>> But there isn't really any difference with the protein flying around >>> the camera. You can interconvert object oriented view and camera >>> oriented view. Have a look at http://pymolwiki.org/index.php/Get_view, >>> http://pymolwiki.org/index.php/Set_view and >>> http://pymolwiki.org/index.php/Scene for starters. >>> >>> Hope it helps, >>> >>> Tsjerk >>> >>> >>> Hello everyone, >>> >>> I am trying to do something that seems a little uncommon. I wish to have a >>> >>> moveable camera. So instead of rotating/translating the protein/molecule/DNA >>> >>> I move the camera point/location and the vector that the camera points in. I >>> >>> wish to be able to move into the molecule and view binding pockets from the >>> >>> perspective of the molecule. I have tried a few web searches on this and >>> >>> thought I would ask everyone here if they have heard of something similar? >>> >>> I am currently trying to find some 3d modelling software that will load a >>> >>> .PDB file so that I can export it to a game engine (Unreal development kit) >>> >>> as a back up method if all else fails. >>> >>> Thanks for your time, >>> >>> Tom >>> >>> ------------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> >>> PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net) >>> >>> Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users >>> >>> Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net >>> >>> >>> >>> -- >>> >>> Michael Zimmermann >>> >>> Ph.D. student in Bioinformatics and Computational Biology >>> >>> Department of Biochemistry, Biophysics and Molecular Biology >>> >>> Iowa State University >>> >>> ------------------------------------------------------------------------------ >>> >>> >>> _______________________________________________ >>> >>> PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net) >>> >>> Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users >>> >>> Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net >>> >>> >>> >>> >>> -- >>> Tsjerk A. Wassenaar, Ph.D. >>> >>> post-doctoral researcher >>> Molecular Dynamics Group >>> Groningen Institute for Biomolecular Research and Biotechnology >>> University of Groningen >>> The Netherlands >>> >>> ------------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net) >>> Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users >>> Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net >>> >>> R. Bryn Fenwick >>> robert.fenw...@irbbarcelona.org >>> Post-doctoral fellow >>> Chemistry and Molecular Pharmacology Programme >>> Institute for Research in Biomedicine (IRB Barcelona) >>> Parc Científic de Barcelona >>> Baldiri Reixac 10, 08028 Barcelona, SPAIN >>> Tel. (+34) 9340 20460 >>> >>> >>> >> >> >> >> -- >> Tsjerk A. Wassenaar, Ph.D. >> >> post-doctoral researcher >> Molecular Dynamics Group >> Groningen Institute for Biomolecular Research and Biotechnology >> University of Groningen >> The Netherlands >> >> ------------------------------------------------------------------------------ >> >> _______________________________________________ >> PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net) >> Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users >> Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net >> > > > > -- > Jason Vertrees, PhD > PyMOL Product Manager > Schrodinger, LLC > > (e) jason.vertr...@schrodinger.com > (o) +1 (603) 374-7120 > -- Tsjerk A. Wassenaar, Ph.D. post-doctoral researcher Molecular Dynamics Group Groningen Institute for Biomolecular Research and Biotechnology University of Groningen The Netherlands ------------------------------------------------------------------------------ _______________________________________________ PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net) Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net
