Clip broken + depth cueing question

[olibclarke@…]

The following bug report has been submitted:
Platform:        Darwin-18.2.0-x86_64-i386-64bit
ChimeraX Version: 0.9 (2019-01-26)
Description
Hi,

Clip doesn't seem to be working properly, with either surface or mesh models. When I run a command like `clip near 10 far -10`, the position of the clipping planes shown in the side view window is correct, but nothing is visible in the main window, unless I reset the view.

Also, I'm not sure if this is a bug, but it is different behavior to Chimera so I wanted to clarify - depth cueing no longer has any effect on mesh maps when meshlighting is set to false. Is this the intended behavior?

Cheers
Oli

Log:
> camera ortho

> cofr centerOfView

> lighting soft

> lighting depthCueColor white

UCSF ChimeraX version: 0.9 (2019-01-26)  
© 2016-2019 Regents of the University of California. All rights reserved.  
How to cite UCSF ChimeraX  

> open
/Users/olibclarke/Dropbox/Clarke_Lab_Shared/Tg/initial_map_phenix_autosharp.mrc

File not found:
/Users/olibclarke/Dropbox/Clarke_Lab_Shared/Tg/initial_map_phenix_autosharp.mrc  

File not found:
/Users/olibclarke/Dropbox/Clarke_Lab_Shared/Tg/initial_map_phenix_autosharp.mrc  

> open
/Users/olibclarke/Dropbox/Downloads/cryosparc_P27_J97_006_volume_map_sharp_local.mrc

File not found:
/Users/olibclarke/Dropbox/Downloads/cryosparc_P27_J97_006_volume_map_sharp_local.mrc  

File not found:
/Users/olibclarke/Dropbox/Downloads/cryosparc_P27_J97_006_volume_map_sharp_local.mrc  

> open /Users/olibclarke/Dropbox/GPU_sync/cryosparc_P27_J106_006_autosharp.mrc

File not found:
/Users/olibclarke/Dropbox/GPU_sync/cryosparc_P27_J106_006_autosharp.mrc  

File not found:
/Users/olibclarke/Dropbox/GPU_sync/cryosparc_P27_J106_006_autosharp.mrc  

> help help:quickstart

> open
/Users/olibclarke/Dropbox/Downloads/cryosparc_P27_J329_005_volume_map_sharp.mrc

Opened cryosparc_P27_J329_005_volume_map_sharp.mrc, grid size 360,360,360,
pixel 1.06, shown at level 0.338, step 2, values float32  

> help volume

> help alia

> help alias

> help volume

Expected a density maps specifieror a keyword  

Expected a density maps specifieror a keyword  

> volume #1 style mesh

> volume #1 style mesh smoothLines true

> volume #1 style mesh smoothLines true squareMesh false

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

Expected a collection of one of 'All', 'atoms', 'bonds', 'cartoons',
'pseudobonds', 'ribbons', 'rings', or 'surfaces'or a keyword  

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 blue

Unknown command: background solid black  

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 blue

Unknown command: background black  

> help background

No help found for 'background'  

> set bgColor black

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 medium blue

> help depthcue

No help found for 'depthcue'  

> help lighting

> lighting soft

> lighting soft

> lighting full

> lighting flat

> lighting full

> lighting flat

> set silhouettes false

> lighting soft

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 medium blue

> lighting depthCue true

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 medium blue

> lighting depthCue true

Unknown command: depthcue start 0; depthcue end 0.5  

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 medium blue

> lighting depthCue true

Unknown command: depthcuestart 0; depthcueend 0.5  

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 medium blue

> lighting depthCue true

Unknown command: depthCueStart 0; depthCueEnd 0.5  

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 medium blue

> lighting depthCue true depthCueStart 0 depthCueEnd 0.5

> volume #1 style mesh smoothLines true squareMesh false meshLighting false

> color #1 medium blue

> lighting depthCue true depthCueStart 0 depthCueEnd 0.1

> volume #1 style mesh smoothLines true squareMesh false meshLighting true

> color #1 medium blue

> lighting depthCue true depthCueStart 0 depthCueEnd 0.1

> volume #1 style mesh smoothLines true squareMesh false meshLighting true

> color #1 medium blue

> lighting depthCue true depthCueStart 0 depthCueEnd 1

> volume #1 style mesh smoothLines true squareMesh false meshLighting true

> color #1 blue

> lighting depthCue true depthCueStart 0 depthCueEnd 1

> lighting soft

> lighting simple

> lighting soft

> help clip

> clip near 10 far 10

Unknown command: reset  

> view cofr false

> clip near 10 far 10

> toolshed show "Side View"

> clip near 10 far -10

> view cofr false

> clip near 10 far -10

> clip near 100 far -100




OpenGL version: 4.1 INTEL-12.4.7
OpenGL renderer: Intel(R) Iris(TM) Graphics 6100
OpenGL vendor: Intel Inc.
File attachment: Screenshot 2017-05-26 10.23.18.png

Screenshot 2017-05-26 10.23.18.png​

[olibclarke@…]

Added by email2trac

[Tom Goddard]

The clip command appears to be working correctly. As the documentations says "Offsets are in physical distance units (generally Å), with positive values increasing away from the viewer for the near/far planes." So your command "clip near 10 far -10" tries to put the far plane in front of the near plane, so it is not surprising that nothing is shown. The side view does appear to show this incorrectly -- it does not show near and far planes at the same distance. So I'll reassign this to Greg so he can look at the side view problem.

The clip command is pretty horrible. After you use "clip near 10 far -10" it forces far to be 10 (as shown by clip list). Since clip planes are not yet shown these values are relative to the center of the bounding box as explained in the docs. But then the next clip command where the planes are show interprets the values as relative to their current positions -- quite confusing. We have planned to revise the clip command syntax but have not done so yet.

[Tom Goddard]

Depth cue is currently considered part of lighting so turning off mesh lighting does also turn of depth cue. This could be changed.

[Tom Goddard]

The graphics allows the far clip plane to be in front of the near clip plane, but treats this state as if the far plane is exactly at the near plane position. Side view is instead showing the far plane in front of the near plane. Side view should show the far plane coincident with the near plane if it is positioned in front of the near plane.

[olibclarke@…]

But when I try “clip near 10 far 10” I get the same result… but "clip near -10 far 10” seems to work. I guess that is the expected behavior, I just didn’t understand the syntax when I read the docs quickly, sorry.

I just want to clip in a slab either side of the center of rotation, which in Chimera used the fromCenter True option - would it be possible to add this option back to ChimeraX?

Cheers
Oli

[olibclarke@…]

Added by email2trac

[olibclarke@…]

I’m not a great fan of mesh lighting - I prefer in Chimera to turn mesh lighting off and just use depth cueing instead. I think it would be good if this was available in ChimeraX too.

Cheers
Oli

[Tom Goddard]

I've made the clip command warn if the far plane is placed in front of the near plane. Same with front/back planes.

[Greg Couch]

Let Tom decide if this ticket can be closed.

[Tom Goddard]

This problem was mostly solved by having the clip command warn when you put the near clip plane behind the far clip plane.

Three other issues have been made separate tickets, side view does not show correct clip plane positions #3106, depth cue with no mesh lighting #3107, and clip positions relative to center of rotation #3108.