[chimerax-users] Wire representation for models?
Tom Goddard
goddard at sonic.net
Tue May 23 11:50:58 PDT 2017
Hi Oliver,
On Nvidia or AMD graphics stick display is faster than wire as can be seen on the Chimera benchmarks web page
http://plato.cgl.ucsf.edu/trac/chimera/wiki/benchmarks <http://plato.cgl.ucsf.edu/trac/chimera/wiki/benchmarks>
But as you say, Intel graphics which is typical low-end laptop graphics, has much slower stick rendering. This is because the Intel OpenGL drivers are not providing hardware accelerated instancing that makes the sticks (and spheres) fast. Modern graphics is making wire hard to do well — using modern OpenGL, which ChimeraX uses but not Chimera, the OpenGL standard does not support linewidths other than 1 pixel — a signicant problem on retina display where the lines become very thin. The modern solution to this is to use a small diameter cylinder. Chimera 1 has been held back from utilizing the latest graphics features (e.g. enabling ambient occlusion lighting) by all the effort we put into supporting the low-end systems (Intel graphics). Our emphasis in going from Chimera 1 to ChimeraX is to give use of modern graphics higher priority than supporting low-end computers. Ideally if we had enough money we could do both. All that said, we might add wire molecular display but it is a low priority. You may be better off using Chimera 1 on computers not suited for 3d graphics.
Tom
> On May 23, 2017, at 9:47 AM, Oliver Clarke wrote:
>
> Hi, would it be possible to add wire representation to ChimeraX?
>
> Two reasons:
>
> 1: In regular Chimera, wire representation is much, much faster for rotation/translation for large models when clipping is enabled, so I find it useful to get the right view (before switching to another representation). This is probably due to the fact that I’m running Chimera on a MacBook pro with integrated graphics, but I think this is probably not so uncommon as a use case.
>
> 2: Even although it is old school and basic, for some purposes (e.g. in conjunction with mesh maps), I prefer this representation, as it has clean, simple lines which are nice for emphasizing connectivity/angles.
>
> Cheers
> Oli
> _______________________________________________
> ChimeraX-users mailing list
> ChimeraX-users at cgl.ucsf.edu
> Manage subscription:
> http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimerax-users/attachments/20170523/6c402852/attachment.html>
More information about the ChimeraX-users
mailing list