From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Tue Sep 22 2009 - 13:22:47 CDT
On Tue, Sep 22, 2009 at 1:08 PM, Rick Calloway <rick_at_cnainc.net> wrote:
> Hi community,
rick,
> Can someone help me better understand the precision of the output coordinate
> data.
depends in part on the file format, some formats have very limited
accuracy (.pdb).
if you load .dcd files then you have IEEE754 style single precision
floating point,
i.e. about 7 digits precision. also, VMD internally stores coordinates
as floats.
> I am looking for statistically significant movement of the center of mass of
> a large molecule. Using VMD's measure center command returns angstroms out
> to 15 decimal places and looks quite tempting.
that is only because VMD's measure returns data to the Tcl interpreter
and in Tcl all floating point data is handled internally as double precision.
> If the last 12 or so decimal places had meaning I could significantly reduce
> the runtime for my system.
> How can I determine how many of these digits are significant?
write a test files and check it out? the raw data is a float, but i
have no idea how this error propagates to the individual measure
property.
cheers,
axel.
> Thank you in advance.
>
> Rick Calloway
-- Dr. Axel Kohlmeyer akohlmey_at_gmail.com Institute for Computational Molecular Science College of Science and Technology Temple University, Philadelphia PA, USA.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:18 CST