Feature Requests item #1743334, was opened at 2007-06-26 07:42
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=835080&aid=1743334&group_id=165310
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Rendering
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Benoit Jacob (benoitjacob)
Assigned to: Nobody/Anonymous (nobody)
Summary: speed is not what it used to be
Initial Comment:
Hi,
I have some protein file that I use for testing; in the default viewpoint with maximized window I used to get 8 FPS and since some time I only get 3.5 FPS. I think that change happened approximately at the time when threading was introducted, or maybe when renderOpaque/renderTransparent were introduced.
Moreover, frustum culling does no longer seem to help performance. With that same molecule, when using a viewpoint from which only a few atoms were visible, I used to get more than 40 FPS. Now I can’t get more than 4.5 FPS.
Some things to investigate:
– is blending enabled when it’s not needed (opaque objects)?
– is backface culling disabled when it doesn’t have to be disabled (opaque objects) ?
– is some operation being performed on all atoms/bonds before frustum culling gets a chance to cull them out? Currently, frustum culling is done in the Painter methods (see the check with the dot product). Maybe it needs to operate somewhere upstream instead?
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=835080&aid=1743334&group_id=165310