FastROCS TK 1.9.0¶
General notices¶
FastROCS TK no longer supports Nvidia Tesla, GEForce, and Quadro cards with a compute capability of less than 3.0. This includes all GPUs with Fermi architecture (e.g., Tesla C2050 and GEForce GTX 430). If you are running one of these older GPUs and cannot upgrade your hardware, do not update to this version of FastROCS TK.
New features¶
FastROCS TK is now supported for Java and C++. See the new documentation pages for Java and C++ for examples of toolkit usage and syntax.
Note
Java and C++ FastROCS TK API points are currently intended to be used only for toolkit functionality and not client-server functionality. The Python API should be used if client-server utilities are needed.
A new
OEShapeDatabasePrep
function has been added to provide the same functionality as the shapedatabaseprep example, except as a multi-threaded function. This should improve performance when prepping databases on systems with multiple cores. This function is non-deterministic, so molecules will not appear in the same order as the input file.
Major bug fixes¶
A memory leak that was only detectable when running large numbers of queries in succession against the same database without breaking context has been fixed.
The
OEShapeDatabase.AddMol
method was previously missing the ability to load molecules forOEFastROCSOrientation_AsIs
alternative starting points.OEShapeDatabase.AddMol
now takes theOEFastROCSOrientation
constant as an additional argument. The default isOEFastROCSOrientation_Inertial
.
Documentation changes¶
The examples section of the documentation has been reorganized by type: Simple Examples, Alternative Starts, and Client-Server. All are viewable from the FastROCS Example Suite, which includes brief descriptions as well as links to the specific example pages. The example pages now provide more in-depth descriptions of usage and download links to the example code.
New examples have been added to the Alternative Starts section to demonstrate using FastROCS TK with alternative starts.
Documentation for both the Java and C++ versions of FastROCS TK has been added.