Hi Alex,
The graphical computational power of even the lowest-end graphic chips these
days will suffice the displaying of our models. To give you some idea: we
are still using a Quadro FX 1000 card and an old CRT on one of our stereo
systems; my 6-year old laptop has an ATI Radeon X1300, which can still drive
the stereo display on Zalman or LG D2342P. You can check the benchmarks of
these chips to see why dual card is guaranteed to be a waste.
If you are determined to go the passive 3D monitor(Zalman or LG D2342) path,
then you actually do not need to worry about the graphic card - even the
on-board one will do. Quadro cards are for the use with 120Hz monitors and
shutter glasses, whereas the passive 3D monitors are CPL-based and do not
need special graphic cards.
Having said all the above, UCSF Chimera might become a little demanding when
displaying electron densities, so a discrete mainstream Nvidia or ATI chip
would probably make life easier in a few cases.
Zhijie
--------------------------------------------------
From: "Alex Kavian" <alek6...@gmail.com>
Sent: Wednesday, February 13, 2013 2:47 PM
To: <CCP4BB@JISCMAIL.AC.UK>
Subject: [ccp4bb] GeForce Graphics cards
Hi there,
I have an off-toptic question about Graphics card. My searches on
pymolwiki and ccp4bb archives resulted in the following conclusion: Coot
and pymol are not compatible with the new GeForce graphics cards. Hoewver,
most of the posts I found were from 2009 and 2010. Does anyone here have
any experience with GeForce 660 or 680 for stereo applications of Coot,
Pymol or UCSF chimera? Will quadro cards work equally smooth in the stereo
mode of these programs? Do 3D applications put too much pressure on the
Graphics card which would justify installing dual Graphics card? (not sure
if is relevant, but just for the record, we want to buy passive 3D monitor
due to the budget restrictions).
Thanks,
Alex
