|
| Latest Headlines |
|
|
|
|
|
October 18, 2004 |
|
|
The latest version of APTL is available for download. This
version add support for Microsoft® Visual Studio® C++
compiler version 7.0 or higher.
Click here for more information.
|
|
|
|
June 28, 2004 |
|
|
The 0.9.5 beta version of SCTL is available for download
here. This version
adds support for win32 targets using the
MinGW framework with the GNU
gcc/g++ compilers. Updates also include extensive additions to
the multithreading classes.
|
|
|
|
May 29, 2004 |
|
|
The 0.9.4 beta version of SCTL is available for download
here. This version
adds support for block-format matricies and accompanying
expression templates for interoperability with existing SCTL
matrix and vector types.
|
|
|
|
May 5, 2004 |
|
|
The 0.9.2 beta version of SCTL is available for download
here. This version
adds support for sparse-column matricies (sparse column vectors)
and expression templates for interoperability with regular sparse
matricies (sparse row vectors). See the following
code sample for an
example of using sparse-column matricies.
|
|
|
|
April 30, 2004 |
|
|
The first beta release of SCTL is now available for download
here. See the following
example codes for a simple
introduction and general compilation instructions.
|
|
|
|
April 12, 2004 |
|
|
The Opstone benchmarks family now includes a singular value
decomposition (SVD) benchmark. SVD is a powerful matrix
decomposition technique that is commonly used in analyzing
near-singular linear systems as well as linear least-squares
minimization problems.
|
|
|
|
April 06, 2004 |
|
|
New '04q2' version of Opstone benchmarks are now available for
download here.
These new versions provide some minor performance tweaks as well
as a few bug fixes as reported by several testers.
|
|
|
|
Feb 24, 2004 |
|
|
Check out this sample code
to see how SCTL can boost your applications performance while
improving the general readability of your C/C++ source code.
Using a Cholesky matrix factorization example, by modifying only
6 lines of code results in a 35% increase in execution speed!
|
|
