<?xml version="1.0" encoding="utf-8"?>
<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Recent posts to news</title><link href="https://sourceforge.net/p/viennacl/news/" rel="alternate"/><link href="https://sourceforge.net/p/viennacl/news/feed.atom" rel="self"/><id>https://sourceforge.net/p/viennacl/news/</id><updated>2015-08-01T09:04:50.797000Z</updated><subtitle>Recent posts to news</subtitle><entry><title>ViennaCL 1.7.0 released!</title><link href="https://sourceforge.net/p/viennacl/news/2015/08/viennacl-170-released/" rel="alternate"/><published>2015-08-01T09:04:50.797000Z</published><updated>2015-08-01T09:04:50.797000Z</updated><author><name>Karl Rupp</name><uri>https://sourceforge.net/u/koarl0815/</uri></author><id>https://sourceforge.net97a5f51d47a3376f1d37a7e861caa836ea989215</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The highlights of the 1.7.0 release are:&lt;br /&gt;
 * Fine-grained parallel incomplete LU factorization preconditioners&lt;br /&gt;
   (based on recent paper by Chow and Patel &lt;span&gt;[4]&lt;/span&gt;)&lt;br /&gt;
 * Fast sparse matrix-matrix multiplication&lt;br /&gt;
   (based on recent paper by Gremse et al. &lt;span&gt;[5]&lt;/span&gt;)&lt;br /&gt;
 * Improved performance of sparse matrix-vector products&lt;br /&gt;
 * Fine-grained parallel algebraic multigrid preconditioners for CUDA, OpenCL, and OpenMP.&lt;br /&gt;
 * Conversion between vectors and matrices of different numeric type&lt;br /&gt;
 * Lanczos eigenvalue solver now optionally also returns eigenvectors.&lt;br /&gt;
 * Interface to/from the Armadillo library &lt;span&gt;[4]&lt;/span&gt;&lt;/p&gt;
&lt;p&gt;A full list of changes is available here:&lt;br /&gt;
&lt;a href="http://viennacl.sourceforge.net/doc/changelog.html"&gt;http://viennacl.sourceforge.net/doc/changelog.html&lt;/a&gt; &lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL 1.3.1 released!</title><link href="https://sourceforge.net/p/viennacl/news/2012/08/viennacl-131-released/" rel="alternate"/><published>2012-08-09T21:00:06Z</published><updated>2012-08-09T21:00:06Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.net2d28ac50684f5e7c70a800cbad4e61d71cd87a80</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The following bugfixes and enhancements have been applied:&lt;br /&gt;
- Fixed a compilation problem with GCC 4.7 caused by the wrong order of function declarations. Also removed unnecessary indirections and unused variables.&lt;br /&gt;
- Improved out-of-source build in the src-version (for packagers).&lt;br /&gt;
- Added virtual destructor in the runtime_wrapper-class in the kernel generator.&lt;br /&gt;
- Extended flexibility of submatrix and subvector proxies (ranges, slices).&lt;br /&gt;
- Block-ILU for compressed_matrix is now applied on the GPU during the solver cycle phase. However, for the moment the implementation file in viennacl/linalg/detail/ilu/opencl block ilu.hpp needs to be included separately in order to avoid an OpenCL dependency for all ILU implementations.&lt;br /&gt;
- SVD now supports double precision.&lt;br /&gt;
- Slighly adjusted the interface for NMF. The approximation rank is now specified by the supplied matrices W and H.&lt;br /&gt;
- Fixed a problem with matrix-matrix products if the result matrix is not initialized properly (thanks to Laszlo Marak for finding the issue and a fix).&lt;br /&gt;
- The operations C += prod(A, B) and C −= prod(A, B) for matrices A, B, and C no longer introduce temporaries if the three matrices are distinct.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL 1.2.0 released!</title><link href="https://sourceforge.net/p/viennacl/news/2012/01/viennacl-120-released/" rel="alternate"/><published>2012-01-01T12:30:54Z</published><updated>2012-01-01T12:30:54Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.netc64692b8eaf2126f90e0f4fea1561067ba1ef6ae</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;ViennaCL 1.2.0 is now available for download! Highlights of the new release are the following features (all experimental):&lt;br /&gt;
- Several algebraic multigrid preconditioners&lt;br /&gt;
- Sparse approximate inverse preconditioners&lt;br /&gt;
- Fast Fourier transform&lt;br /&gt;
- Structured dense matrices (circulant, Hankel, Toeplitz, Vandermonde)&lt;br /&gt;
- Reordering algorithms (Cuthill-McKee, Gibbs-Poole-Stockmeyer)&lt;br /&gt;
- Proxies for manipulating subvectors and submatrices&lt;br /&gt;
The features are expected to reach maturity in the 1.2.x branch.&lt;/p&gt;
&lt;p&gt;Happy New Year!&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL 1.1.2 released!</title><link href="https://sourceforge.net/p/viennacl/news/2011/05/viennacl-112-released/" rel="alternate"/><published>2011-05-21T20:36:07Z</published><updated>2011-05-21T20:36:07Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.net21f4154e216b2438ec02ff4b2726cd0553577f55</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Version 1.1.2 is now available for download. The highlights of the new release are as follows:&lt;br /&gt;
- Improved CG and BiCGStab performance especially for smaller systems&lt;br /&gt;
- Improved performance of Jacobi and row scaling preconditioner setup&lt;br /&gt;
- Improved performance of coordinate_matrix&lt;br /&gt;
This is the final release of the 1.1.x family. The next version 1.2.0 will introduce a number of new features such as additional sparse matrix formats, support for integer vectors and matrices as well as better handling of multiple OpenCL platforms.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL within Google Summer of Code 2011</title><link href="https://sourceforge.net/p/viennacl/news/2011/03/viennacl-within-google-summer-of-code-2011/" rel="alternate"/><published>2011-03-19T17:24:18Z</published><updated>2011-03-19T17:24:18Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.net71168deba5807c5b3361c40317294ac517d9e7b0</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;ViennaCL is part of the "Computational Science and Engineering at TU Wien" project of the Google Summer of Code 2011. If you with to contribute, check out &lt;a href="http://www.iue.tuwien.ac.at/cse/ideas.html" rel="nofollow"&gt;http://www.iue.tuwien.ac.at/cse/ideas.html&lt;/a&gt; and &lt;br /&gt;
&lt;a href="http://www.google-melange.com/gsoc/program/accepted_orgs/google/gsoc2011" rel="nofollow"&gt;http://www.google-melange.com/gsoc/program/accepted_orgs/google/gsoc2011&lt;/a&gt;&lt;br /&gt;
Deadline for application is April 8th, 2011.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL 1.1.1 released</title><link href="https://sourceforge.net/p/viennacl/news/2011/03/viennacl-111-released/" rel="alternate"/><published>2011-03-17T13:56:52Z</published><updated>2011-03-17T13:56:52Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.net2ad5c2d835abb66d18b4eb6bd84007c28eb4146b</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;This new revision release has a focus on better interaction with other linear algebra libraries. The few known glitches with version 1.1.0 are now removed. Highlights:&lt;br /&gt;
- New out-of-the-box support for Eigen and MTL 4&lt;br /&gt;
- Improved default parameters for BLAS level 3 routines&lt;br /&gt;
- Fixed compilation problems with older OpenCL 1.0 headers&lt;br /&gt;
Please send us your wishes for future releases!&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL 1.1.0 released</title><link href="https://sourceforge.net/p/viennacl/news/2010/12/viennacl-110-released/" rel="alternate"/><published>2010-12-24T00:09:43Z</published><updated>2010-12-24T00:09:43Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.netb30de5b28dcd3dd06807a2984a9d3bd12e3f920d</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Even though not initially intended, the new version 1.1.0 is now just in time for Christmas! The highlights of this new version are as follows:&lt;br /&gt;
- Full BLAS level 3 support&lt;br /&gt;
- Multi-GPU support&lt;br /&gt;
- Two additional preconditioners&lt;br /&gt;
- Automated performance tuning environment&lt;br /&gt;
- ViennaCL can operate with existing, user-provided OpenCL contexts&lt;br /&gt;
More information can be found on the project homepage.&lt;/p&gt;
&lt;p&gt;Thanks for the patience and merry christmas to all users of ViennaCL!&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL 1.1.0 delayed to mid December</title><link href="https://sourceforge.net/p/viennacl/news/2010/12/viennacl-110-delayed-to-mid-december/" rel="alternate"/><published>2010-12-03T12:34:56Z</published><updated>2010-12-03T12:34:56Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.netcffcc4329312401c3e263d418a13500978aa8290</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The upcoming ViennaCL 1.1.0 is going to be a huge step forward. We are currently fine-tuning the interfaces for the new features (which also include support for multiple GPUs), hunt bugs and extend our automated testing environment. This has turned out to be rather time-consuming, therefore we decided to postpone the release to mid December. Please be patient, we are not going to disappoint you!&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Teaser for ViennaCL 1.1.0</title><link href="https://sourceforge.net/p/viennacl/news/2010/11/teaser-for-viennacl-110/" rel="alternate"/><published>2010-11-05T16:51:39Z</published><updated>2010-11-05T16:51:39Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.netd07fd3ea3bb292b28d4f937ad137708e02e704b2</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The highlights of the upcoming release are as follows:&lt;br /&gt;
* BLAS level 3 functionality&lt;br /&gt;
* A performance tuning facility&lt;br /&gt;
* Additional preconditioners&lt;br /&gt;
* Better out-of-the-box support for STL classes&lt;br /&gt;
* The ViennaCL logo :-)&lt;br /&gt;
More details at &lt;a href="http://viennacl.sourceforge.net/index.php%3Fid=news.html"&gt;http://viennacl.sourceforge.net/index.php%3Fid=news.html&lt;/a&gt;&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>ViennaCL 1.0.5 released!</title><link href="https://sourceforge.net/p/viennacl/news/2010/08/viennacl-105-released/" rel="alternate"/><published>2010-08-05T15:52:21Z</published><updated>2010-08-05T15:52:21Z</updated><author><name>Josef Weinbub</name><uri>https://sourceforge.net/u/billisbrother/</uri></author><id>https://sourceforge.netb60c17a63c8d392fc54d6c6d68aa062ee0c4bbe5</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;This is the last 1.0.x release. The main changes are as follows:&lt;br /&gt;
* Added a reader and writer for MatrixMarket files (thanks to Evan Bollig for suggesting that)&lt;br /&gt;
* Eliminated a bug that caused the upper triangular direct solver to fail on NVIDIA hardware for large matrices (thanks to Andrew Melfi for finding that)&lt;br /&gt;
* The number of iterations and the final estimated error can now be obtained from iterative solver tags.&lt;br /&gt;
* Improvements provided by Klaus Schnass are included in the developer converter script (OpenCL kernels to C++ header)&lt;br /&gt;
* Disabled the use of reference counting for OpenCL handles on Mac OS X (caused seg faults on program exit)&lt;/p&gt;&lt;/div&gt;</summary></entry></feed>