I guess I'll be a voice of dissent. I personally think a fixed 2d matrix is too easy. It seems all you would have to do is either use a flat boost::array of Rows * Columns or the like, and copy boost::array's interface. This is from someone who is programming a game and already made a 2d matrix class. I do, however, think that a fixed-size n-dimensional matrix would be better for GSoC, and can see the value in that. It would be difficult, though, without varidic templates. On Sun, Mar 29, 2009 at 1:25 AM, Emil Dotchevski <emildotchevski@gmail.com> wrote:
On Sat, Mar 28, 2009 at 7:54 PM, Patrick Mihelich <patrick.mihelich@gmail.com> wrote:
I think a fixed-size matrix class would be a great GSoC project. Or base of a larger project for the really ambitious student :). Many extra points if it is able to piggy-back on Joel's SIMD library. Lack of good fixed-size support is one reason I abandoned uBLAS.
Who and why would use the fixed size matrix class?
This seems like a strange question coming from a game developer. Fixed-size matrix can be dramatically more efficient since you can stack-allocate it and make optimal use of SIMD instructions. I use Eigen2's fixed-size vectors/matrices all the time.
I think you misunderstood my question. It was in reply to something along the lines of "we can write it and even if it isn't good for game developers, that's fine." My point was that game developers are the primary audience for such a library. See my previous post: http://article.gmane.org/gmane.comp.lib.boost.devel/187988.
Emil Dotchevski Reverge Studios, Inc. http://www.revergestudios.com/reblog/index.php?n=ReCode _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost