Opened 3 years ago

Closed 3 years ago

#7488 closed enhancement (limitation)

make BILD transparency work

Reported by: mlists@… Owned by: Tom Goddard
Priority: moderate Milestone:
Component: Graphics Version:
Keywords: Cc: Greg Couch, Elaine Meng
Blocked By: Blocking:
Notify when closed: Platform: all
Project: ChimeraX

Description

From Francois Berenger:

In the old Chimera (not X), if I put two transparent spheres in a .bild file:
one sphere can completely hide the other (e.g. a small one inside of a bigger one).
While, if there was true transparency, no sphere should completely hide another.
Is this better handled in ChimeraX?
I _love_ .bild files, because it is so easy to generate them in order to annotate
a 3D molecular system then view all of it (molecules + annotations) in Chimera.
Thanks a lot,
F.

example file francois.bild attached

Hi Francois,
You can open that same bild file in ChimeraX to see for yourself -- unfortunately, however, the problem with transparency is the same.
<https://rbvi.ucsf.edu/chimerax/docs/user/formats/bild.html>

ChimeraX also has the same problems with multiple transparent models as Chimera, see:
<https://rbvi.ucsf.edu/chimerax/docs/user/commands/colornames.html#transparency-limitations>
<https://www.rbvi.ucsf.edu/pipermail/chimerax-users/2018-July/000274.html>

This BILD problem is additional, however, as the spheres are all in the same model (not multiple models) if you open them from a single file. I'll make a ChimeraX ticket for the problem with your name as reporter, so that you would receive notifications if anything happens.

To see the problem with multiple models, you can create the spheres with the "shape" command -- the transparency may work in certain places but not in others, which is the gist of the problem. Some of the spheres will not look transparent at all, but might look transparent if you open some other structure and put it behind them.
<https://rbvi.ucsf.edu/chimerax/docs/user/commands/shape.html>

Here are the shape commands to make the same spheres as in your BILD file (also attached as shape.cxc):

color name tgreen 0,100,0,75
color name torange 100,65,0,75
color name tgrey 50,50,50,75
shape sphere radius 1.5 center -0.542700,0.122500,-0.047867 color tgreen
shape sphere radius 1.5 center 1.060560,-1.270600,-0.236620 color tgreen
shape sphere radius 1.25 center 0.611100,-2.324200,-0.346200 color torange
shape sphere radius 1.25 center 1.108100,2.092000,0.140700 color torange
shape sphere radius 1.25 center -3.097400,0.651000,0.062700 color torange
shape sphere radius 1.5 center -0.542700,0.122500,-0.047867 color tgrey
shape sphere radius 1.5 center 1.060560,-1.270600,-0.236620 color tgrey
shape sphere radius 1.5 center -1.453700,2.784800,0.269900 color tgrey
shape sphere radius 1.5 center 3.272900,0.026100,-0.134900 color tgrey
shape sphere radius 1.5 center -2.354000,-1.959600,-0.245900 color tgrey

Sorry about the limitations,
Elaine

Attachments (2)

francois.bild (564 bytes ) - added by Elaine Meng 3 years ago.
shape.cxc (817 bytes ) - added by Elaine Meng 3 years ago.

Download all attachments as: .zip

Change History (5)

by Elaine Meng, 3 years ago

Attachment: francois.bild added

by Elaine Meng, 3 years ago

Attachment: shape.cxc added

comment:1 by Elaine Meng, 3 years ago

Cc: Elaine Meng added

comment:2 by pett, 3 years ago

Reporter: changed from mlists@… to mlists@…

comment:3 by Tom Goddard, 3 years ago

Resolution: limitation
Status: assignedclosed
Type: defectenhancement

Ticket #3838 explains that we do not have the resources (funding / programmers) to handle multilayer transparent surfaces correctly

https://www.rbvi.ucsf.edu/trac/ChimeraX/ticket/3838

It is unlikely this going to change. The techniques to do this well (without degrading performance for everyone who does not use multi-layer transparency) are just too complex for us to implement.

Note: See TracTickets for help on using tickets.