Make move to center mouse mode keep pivot at center after translation

[leeyang+chimerax@…]

Yang wants the center of rotation to stay at the middle of the graphics window when using the "center" or "center or translate" mouse modes.

On Dec 9, 2023, at 5:14 AM, leeyang+chimerax--- via ChimeraX-users <chimerax-users@…> wrote:

Hi Tom,

Thank you for implementing those mousemodes. If I'm not mistaken, you had kindly implemented the 'center' function in that first daily build to address an earlier request that I had made.

Currently it centers and fixes cofr at the location. It does mean in subsequent manual drag-to-translate operations, the cofr does not track the viewport. I wonder if there may be some utility in having the option to either a) respect current cofr setting and merely center the location, or b) setting cofr to either frontcenter or centerofview in addition to centering the location? My current workaround is to click-to-center and then reset cofr with an alias.

RE: frames, Thanks for clueing me into the relevant section of code. Found the equivalent file on Windows. The idea would be to reduce the number of frames in order to speed up the animation slightly; but it only feels like it could do with a tweak when zoomed in and the movement spans a short "real" distance. This works.

Cheers,
Yang

[Tom Goddard]

I see, you want the center of rotation to stay in the middle of the graphics window. Currently if you use the "center" mouse mode it centers the clicked atom and makes the center of rotation at that new atom position which is in the middle of the window, but then if you translate the center of rotation stays at the atom position even though it is no longer in the center of the window. This center of rotation method is called "fixed" (fixed to the atom).

The trouble I see with your suggestions is that if the center of rotation method is centerofview or frontcenter then when the atom is centered the rotation will not be about the atom. I think that is very bad behavior. The first thing you are likely to do after centering is rotate to get a clearer view and if the rotation is not about the atom position then the atom moves away. The reason it won't be centered on the atom is because the z-depth of the rotation center will be wrong. For centerofview it will be midway between the front and back of the scene, and for frontcenter it will be on the front-most object at the center. ChimeraX doesn't have a center of rotation method that does what is needed here. I don't see a good way to do it using current ChimeraX modes.